Hepatotoxicity: From Genomics to In Vitro and In Vivo Models

Hepatotoxicity From Genomics to in vitro and in vivo Models Editor SAURA C. SAHU US Food and Drug Administration, Laur...

Author: Saura Sahu

17 downloads 1167 Views 9MB Size Report

This content was uploaded by our users and we assume good faith they have the permission to share this book. If you own the copyright to this book and it is wrongfully on our website, we offer a simple DMCA procedure to remove your content from our site. Start by pressing the button below!
Report copyright / DMCA form

DOWNLOAD PDF

Hepatotoxicity From Genomics to in vitro and in vivo Models

Editor

SAURA C. SAHU US Food and Drug Administration, Laurel, MD, USA

Hepatotoxicity

Hepatotoxicity From Genomics to in vitro and in vivo Models

Editor

SAURA C. SAHU US Food and Drug Administration, Laurel, MD, USA

C 2007 Copyright

John Wiley & Sons Ltd, The Atrium, Southern Gate, Chichester, West Sussex PO19 8SQ, England Telephone (+44) 1243 779777

Email (for orders and customer service enquiries): [email protected] Visit our Home Page on www.wileyeurope.com or www.wiley.com All Rights Reserved. No part of this publication may be reproduced, stored in a retrieval system or transmitted in any form or by any means, electronic, mechanical, photocopying, recording, scanning or otherwise, except under the terms of the Copyright, Designs and Patents Act 1988 or under the terms of a licence issued by the Copyright Licensing Agency Ltd, 90 Tottenham Court Road, London W1T 4LP, UK, without the permission in writing of the Publisher. Requests to the Publisher should be addressed to the Permissions Department, John Wiley & Sons Ltd, The Atrium, Southern Gate, Chichester, West Sussex PO19 8SQ, England, or emailed to [email protected], or faxed to (+44) 1243 770620. Designations used by companies to distinguish their products are often claimed as trademarks. All brand names and product names used in this book are trade names, service marks, trademarks or registered trademarks of their respective owners. The Publisher is not associated with any product or vendor mentioned in this book. This publication is designed to provide accurate and authoritative information in regard to the subject matter covered. It is sold on the understanding that the Publisher is not engaged in rendering professional services. If professional advice or other expert assistance is required, the services of a competent professional should be sought. The Publisher and the Author make no representations or warranties with respect to the accuracy or completeness of the contents of this work and specifically disclaim all warranties, including without limitation any implied warranties of fitness for a particular purpose. The advice and strategies contained herein may not be suitable for every situation. In view of ongoing research, equipment modifications, changes in governmental regulations, and the constant flow of information relating to the use of experimental reagents, equipment, and devices, the reader is urged to review and evaluate the information provided in the package insert or instructions for each chemical, piece of equipment, reagent, or device for, among other things, any changes in the instructions or indication of usage and for added warnings and precautions. The fact that an organization or Website is referred to in this work as a citation and/or a potential source of further information does not mean that the author or the publisher endorses the information the organization or Website may provide or recommendations it may make. Further, readers should be aware that Internet Websites listed in this work may have changed or disappeared between when this work was written and when it is read. No warranty may be created or extended by any promotional statements for this work. Neither the Publisher nor the Author shall be liable for any damages arising herefrom. Other Wiley Editorial Offices John Wiley & Sons Inc., 111 River Street, Hoboken, NJ 07030, USA Jossey-Bass, 989 Market Street, San Francisco, CA 94103-1741, USA Wiley-VCH Verlag GmbH, Boschstr. 12, D-69469 Weinheim, Germany John Wiley & Sons Australia Ltd, 42 McDougall Street, Milton, Queensland 4064, Australia John Wiley & Sons (Asia) Pte Ltd, 2 Clementi Loop #02-01, Jin Xing Distripark, Singapore 129809 John Wiley & Sons Canada Ltd, 6045 Freement Bld, Mississauga, Ontario, Canada 5R4J3 Wiley also publishes its books in a variety of electronic formats. Some content that appears in print may not be available in electronic books. Anniversary Logo Design: Richard J. Pacifico Library of Congress Cataloging-in-Publication Data Hepatotoxicity : from genomics to in vitro and in vivo models / editor, Saura C. Sahu. p. ; cm. Includes bibliographical references and index. ISBN 978-0-470-05716-2 (pbk. : alk. paper) 1. Hepatotoxicology. I. Sahu, Saura C. [DNLM: 1. Hepatitis, Toxic—etiology. 2. Drug Evaluation, Preclinical. 3. Liver—drug effects. 4. Models, Biological. 5. Toxicogenetics—methods. WI 715 H5295 2008] RC848.H48H43 2008 616.3 623071–dc22 2007038233 British Library Cataloguing in Publication Data A catalogue record for this book is available from the British Library ISBN 978-0-470-05716-2 Typeset in 10/12pt Times by Aptara Inc., India. Printed and bound in Great Britain by Antony Rowe Ltd, Chippenham, England. This book is printed on acid-free paper responsibly manufactured from sustainable forestry in which at least two trees are planted for each one used for paper production.

Dedicated to My parents, Gopinath and Inchhamani, for their gift of life, love and living examples My teachers for their gift of education My wife, Jharana, for her life-long companionship, love and support for following my dreams My children, Megha, Sudhir and Subir, for their love and care

Contents

Contributors Preface Acknowledgements SECTION 1

MODELS FOR HEPATOTOXICITY TESTING

1

Current in vitro Models to Study Drug-Induced Liver Injury Julio C. Davila, Jinghai J. Xu, Keith A. Hoffmaster, Peter J. O’Brien and Stephen C. Storm

2

Utilization of an in vitro Hepatotoxicity Test in the Early Stage of Drug Discovery Ikuo Horii, Hiroshi Yamada, Rie Kikkawa, Toshinori Yamamoto, Tamio Fukushima and Kaori Tomizawa

3

4

5

6

xi xvii xix

3

57

Use of Hepatocytes for Characterizing a Candidate Drug’s Metabolism and Drug Interaction Potential Srikanth C. Nallani, John M. Strong and Shiew Mei Huang

69

Human-Based in vitro Experimental Systems for the Evaluation of Human Drug Safety Albert P. Li

89

Hepatocytes as a Model for Screening Food-Related Hepatotoxins and Studying Mechanisms of their Toxicity Saura C. Sahu Some Experimental Models of Liver Damage Pablo Muriel

105 119

viii

Contents

SECTION 2 7

Application of Short- and Long-Term Hepatocyte Cultures to Predict Toxicities Gregor Tuschl, Jens Hrach, Philip G. Hewitt and Stefan O. Mueller

SECTION 3 8

141

BIOMARKERS OF HEPATOTOXICITY

Biomarkers of Mycotoxin Exposure in Liver Toxicity Angela J. Harris

SECTION 4 9

HEPATOCYTE CULTURES

177

MECHANISMS OF HEPATOTOXICITY

Mechanisms of Toxic Liver Injury Nora Anderson and Jürgen Borlak

191

10

A Role of Cytochrome P450 in Quinone-Induced Hepatotoxicity Yasuhiro Ishihara and Norio Shimamoto

287

11

A Mechanistic View of Troglitazone Hepatotoxicity Rawiwan Maniratanachote and Tsuyoshi Yokoi

299

12

Role of the Kupffer Cell in Hepatotoxicity and Hepatocarcinogenesis James E. Klaunig, Stacy M. Corthals, Lisa M. Kamendulis and Binu K. Philip

313

13

Sinusoidal Cells in Liver Injury and Repair Carol R. Gardner and Debra L. Laskin

341

14

Cytokines in Liver Diseases Pablo Muriel

371

15

Bile Acids as Modulators of Apoptosis Rui E. Castro, Susana Solá, Clifford J. Steer and Cec´ılia M.P. Rodrigues

391

16

Drug-Induced Intrahepatic Cholestasis by Interaction with the Hepatic Bile Salt Export Pump (BSEP) Christoph Funk, Johannes Noé, Renée Portmann, Ruben Alvarez-Sánchez, Florian Klammers, Christiane Lamy, Axel Paehler and Michael Pantze

SECTION 5 17

421

GENOMICS OF HEPATOTOXICITY

Application of Toxicogenomics in Predicting Hepatotoxicity – Potentials and Challenges Wen Lin, Guoxiang Shen, Tin Oo Khor and Ah-Ng Tony Kong

449

Contents

ix

18

Genomic Profiling of Liver Injury Kevin Gerrish and David E. Malarkey

465

19

Use of DNA Arrays in Understanding Hepatic Test Systems Angela J. Harris and Daniel A. Casciano

489

20

Prediction of Hepatotoxicity Based on the Toxicogenomics Database Tetsuro Urushidani

507

21

Relationship between N-acetyltransferase-2 Gene Polymorphism and Isoniazid-Induced Hepatotoxicity Yasuo Shimizu, Kunio Dobashi and Masatomo Mori

SECTION 6 22

GENDER DIFFERENCES IN HEPATOTOXICITY

Human and Animal-Based Differences in Hepatic Xenobiotic Metabolism and Toxicity Peter J. O’Brien, Katie Chan and Raymond J. Poon

SECTION 7

Hepatotoxicity in Oncology Drug Development Wei Chen, Kenneth Hastings and John K. Leighton

24

The Potent Rat Hepatocarcinogen Methapyrilene: An Hypothesis Regarding its Hepatotoxicology Daniel A. Casciano

25

26

Index

565

577

HEPATOTOXICITY AND BOTANICAL SUPPLEMENTS

Botanical Supplements and Hepatotoxicity Shabana Khan, Ikhlas Khan and Larry Walker

SECTION 9

539

HEPATOTOXICITY AND HEPATOCARCINOGENICITY

23

SECTION 8

531

591

RISK ANALYSIS OF HEPATOTOXINS

Physiologically Based Pharmokinetic Modeling and Risk Assessment of Hepatotoxicants Kannan Krishnan

609

635

Contributors

Ruben Alvarez-Sánchez F. Hoffman-La Roche Ltd, Non-Clinical Development – Drug Safety, PRBN-D 69/154, 4070 Basel, Switzerland Nora Anderson Medical School of Hannover, Center for Pharmacology and Toxicology, Carl-Neuberg-Strasse 1, D-30625 Hannover, Germany Jurgen ¨ Borlak Fraunhofer Institute of Toxicology and Experimental Medicine, NikolaiFuchs-Strasse 1, D-30625 Hannover, Germany and Medical School of Hannover, Center for Pharmacology and Toxicology, Carl-Neuberg-Strasse 1, D-30625 Hannover, Germany Daniel A. Casciano Department of Pharmacology and Toxicology, University of Arkansas for Medical Sciences, Little Rock, AR 72205, USA and Dan Casciano and Associates, 47 Marcella Drive, Little Rock, Little Rock, AR 72223, USA Rui E. Castro Centro de Patogénese Molecular, Faculty of Pharmacy, University of Lisbon, Av, das For¸cas Armadas, Lisbon 1600-083, Portugal Katie Chan Graduate Department of Pharmaceutical Sciences, Faculty of Pharmacy, University of Toronto, 144 College Street, Toronto, ON M5S 3M2, Canada Wei Chen Division of Drug Oncology Products, Office of Oncology Drug Products, US Food and Drugs Administration, FDA/CDER/OODP/DDOP, 10903 New Hampshire Avenue, Silver Spring, MD 20993, USA Stacy M. Corthals Center for Environmental Health, Department of Pharmacology and Toxicology, Indiana School of Medicine, 541 Clinical Drive, Indianapolis, IN 46202, USA

xii

Contributors

Julio C. Davila Pfizer Inc., PGRD, Saint Louis Laboratories, 700 Chesterfield Parkway North TA1, St Louis, MO 63017, USA Kunio Dobashi Gunma University School of Health Sciences, Gunma University, 3-39-15 Showa-machi, Maebashi, Gunma 371-8511, Japan Tamio Fukushima Worldwide Safety Sciences, Pfizer Global Research and Development, Nagoya Central Research Laboratories, SS-Nagoya, Pfizer Japan Inc., 5–2 Taketoyo, Aichi 470-2393, Japan Christoph Funk F. Hoffman-La Roche Ltd, Non-Clinical Development – Drug Safety, PRBN-D 69/154, 4070 Basel, Switzerland Carol R. Gardner Department of Pharmacology and Toxicology, Rutgers University, 160 Frelinghuysen Road, Piscataway, NJ 08854, USA Kevin Gerrish Laboratory of Molecular Toxicology – Maildrop D2-04, Environmental Stress and Cancer Group, National Institute of Environmental Health Sciences, 111 Alexander Drive, PO Box 12233, Research Triangle Park, NC 27709, USA Angela J. Harris Center for Toxicology and Environmental Health, LLC, 615 W. Markham, Little Rock, AR 72201, USA Kenneth Hastings Office of New Drugs, Center for Drug Evaluation and Research, US Food and Drug Administration, 5600 Fishers Lane, Rockville, MD 20857, USA Philip G. Hewitt Molecular Toxicology, Institute of Toxicology, Merck KGaA, 64271 Darmstadt, Germany Keith A. Hoffmaster Pfizer Inc., Research Training Center (RTC), 01/5R206 ADME Biology, DTC 308, 620 Memorial Drive, Cambridge, MA 02139, USA Ikuo Horii Worldwide Safety Sciences, Pfizer Global Research and Development, Nagoya Central Research Laboratories, SS-Nagoya, Pfizer Japan Inc., 5–2 Taketoyo, Aichi 470–2393, Japan Jens Hrach Molecular Toxicology, Institute of Toxicology, Merck KGaA, 64271 Darmstadt, Germany Shiew Mei Huang Office of Clinical Pharmacology and Biopharmaceutics, Center for Drug Evaluation and Research, Food and Drug Administration, WO21, HFD-850, 10903 New Hampshire Avenue, Silver Spring, MD 20993, USA Yasuhiro Ishihara Faculty of Pharmaceutical Sciences at Kagawa Campus, Tokushima Bunri University, 1314-1, Shido, Sanuki, Kagawa 769-2193, Japan

Contributors

xiii

Lisa Kamendulis Center for Environmental Health, Department of Pharmacology and Toxicology, Indiana School of Medicine, 541 Clinical Drive, Indianapolis, IN 46202, USA Ikhlas Khan National Center for Natural Products Research, Research Institute of Pharmaceutical Sciences and Department of Pharmacognosy, School of Pharmacy, University of Mississippi, PO Box 1848, University, MS 38677, USA Shabana Khan National Center for Natural Products Research, Research Institute of Pharmaceutical Sciences, School of Pharmacy, University of Mississippi, PO Box 1848, University, MS 38677, USA Tin Oo Khor Department of Pharmaceutics, Ernest Mario School of Pharmacy, Rutgers University, The State University of New Jersey, 160 Frelinghuysen Road, Piscataway, NJ 08854, USA Rie Kikkawa Worldwide Safety Sciences, Pfizer Global Research and Development, Nagoya Central Research Laboratories, SS-Nagoya, Pfizer Japan Inc., 5–2 Taketoyo, Aichi 470-2393, Japan Florian Klammers F. Hoffman-La Roche Ltd, Non-Clinical Development – Drug Safety, PRBN-D 69/154, 4070 Basel, Switzerland James E. Klaunig Center for Environmental Health, Department of Pharmacology and Toxicology, Indiana School of Medicine, 541 Clinical Drive, Indianapolis, IN 46202, USA Ah-Ng Tony Kong Department of Pharmaceutics, Ernest Mario School of Pharmacy, Rutgers University, The State University of New Jersey, 160 Frelinghuysen Road, Piscataway, NJ 08854, USA Kannan Krishnan Département de Santé Environmental/Santé an Travail, Université de Montréal, 2375 Cote Ste Catherine, Bureau 4105, Montréal, QC H3T 1A8, Canada Christiane Lamy F. Hoffman-La Roche Ltd, Non-Clinical Development – Drug Safety, PRBN-D 69/154, 4070 Basel, Switzerland Debra L. Laskin Department of Pharmacology and Toxicology, Rutgers University, 160 Frelinghuysen Road, Piscataway, NJ 08854, USA John K. Leighton Division of Drug Oncology Products, Office of Oncology Drug Products, US Food and Drugs Administration, FDA/CDER/OODP/DDOP, 10903 New Hampshire Avenue, Silver Spring, MD 20993, USA Albert P. Li The ADMET Group LLC, In vitro ADMET Laboratories LLC and Advanced Pharmaceutical Sciences Inc., 15235 Shady Grove Road, Suite 303, Rockville, MD 20850, USA

xiv

Contributors

Wen Lin Department of Pharmaceutics, Ernest Mario School of Pharmacy, Rutgers University, The State University of New Jersey, 160 Frelinghuysen Road, Piscataway, NJ 08854, USA David E. Malarkey National Toxicology Program Pathology Group, Laboratory of Experimental Pathology – Maildrop B3-06, National Institute of Environmental Health Sciences, 111 Alexander Drive, PO Box 12233, Research Triangle Park, NC 27709, USA Rawiwan Maniratanachote Drug Metabolism and Toxicology, Division of Pharmaceutical Sciences, Graduate School of Medical Sciences, Kanazawa University, Kakumamachi, Kanazawa 920-1192, Japan Masatomo Mori Department of Medicine and Molecular Science, Gunma University, 3-39-15 Showa-machi, Maebashi, Gunma 371-8511, Japan Stefan O. Mueller Molecular Toxicology, Institute of Toxicology, Merck KGaA, 64271 Darmstadt, Germany Pablo Muriel Sección Externa de Farmacologia, Cinvestav-IPN, Apdo, Postal 14-740, México 07000, DF, México Srikanth C. Nallani Office of Clinical Pharmacology, Center for Drug Evaluation and Research, Food and Drug Administration, WO21, HFD-870, 10903 New Hampshire Avenue, Silver Spring, MD 20993, USA Johannes Noé F. Hoffman-La Roche Ltd, Non-Clinical Development – Drug Safety, PRBN-D 69/154, 4070 Basel, Switzerland Peter J. O’Brien Pfizer Inc., Drug Safety Research and Development, Ramsgate Road, SSEU Europe 380/1.025, IPC 339, Sandwich, CT13 9NJ, UK Peter J. O’Brien Graduate Department of Pharmaceutical Sciences, Faculty of Pharmacy, University of Toronto, 144 College Street, Toronto, ON M5S 3M2, Canada Axel Paehler F. Hoffman-La Roche Ltd, Non-Clinical Development – Drug Safety, PRBN-D 69/154, 4070 Basel, Switzerland Michael Pantze F. Hoffman-La Roche Ltd, Non-Clinical Development – Drug Safety, PRBN-D 69/154, 4070 Basel, Switzerland Binu K. Philip Center for Environmental Health, Department of Pharmacology and Toxicology, Indiana School of Medicine, 541 Clinical Drive, Indianapolis, IN 46202, USA Raymond Poon Environmental and Occupational Toxicology Division, Health Canada, Ottawa, ON K1A 0L2, Canada

Contributors

xv

Renée Portmann F. Hoffman-La Roche Ltd, Non-Clinical Development – Drug Safety, PRBN-D 69/154, 4070 Basel, Switzerland Cec´ılia M. P. Rodrigues Centro de Patogénese Molecular, Faculty of Pharmacy, University of Lisbon, Av, das For¸cas Armadas, Lisbon 1600-083, Portugal Saura C. Sahu Division of Toxicology, Office of Toxicology, Office of Applied Research and Safety Assessment, Center for Food Safety and Applied Nutrition, US Food and Drug Administration, 8301 Muirkirk Road, Laurel, MD 20708, USA Guoxiang Shen Department of Pharmaceutics, Ernest Mario School of Pharmacy, Rutgers University, The State University of New Jersey, 160 Frelinghuysen Road, Piscataway, NJ 08854, USA Norio Shimamoto Faculty of Pharmaceutical Sciences at Kagawa Campus, Tokushima Bunri University, 1314-1, Shido, Sanuki, Kagawa 769-2193, Japan Yasuo Shimizu Department of Medicine and Molecular Science, Gunma University, 3-39-15 Showa-machi, Maebashi, Gunma 371-8511, Japan Susana Solá Centro de Patogénese Molecular, Faculty of Pharmacy, University of Lisbon, Av, das For¸cas Armadas, Lisbon 1600-083, Portugal Clifford J. Steer Departments of Medicine and Genetics, Cell Biology and Development, University of Minnesota Medical School, 6–160 Jackson Hall, 321 Church St. SE, Minneapolis, MN 55455, USA Stephen C. Strom Department of Pathology, University of Pittsburgh, 200 Lothrop Street 450 South BST, Pittsburgh, PA 15261, USA John M. Strong Laboratory of Clinical Pharmacology, Office of Testing and Research, Food and Drug Administration, WO64, 10903 New Hampshire Avenue, Silver Spring, MD 20993, USA Kaori Tomizawa Worldwide Safety Sciences, Pfizer Global Research and Development, Nagoya Central Research Laboratories, SS-Nagoya, Pfizer Japan Inc., 5–2 Taketoyo, Aichi 470-2393, Japan Gregor Tuschl Molecular Toxicology, Institute of Toxicology, Merck KGaA, 64271 Darmstadt, Germany Tetsuro Urushidani Department of Pathophysiology, Faculty of Pharmaceutical Sciences, Doshisha Women’s College of Liberal Arts, Kodo, Kyotanabe, Kyoto 610-0395, Japan and Toxicogenomics Project, National Institute of Biomedical Innovation, Saito-Asagi, Ibaraki, Osaka 567-0085, Japan

xvi

Contributors

Larry Walker National Center for Natural Products Research, Research Institute of Pharmaceutical Sciences and Department of Pharmacology, School of Pharmacy, University of Mississippi, PO Box 1848, University, MS 38677, USA Jinghai J. Xu Pfizer Inc., Research Training Center (RTC), 01/55564 Systems Biology, DTC 367, 620 Memorial Drive, Cambridge, MA 02139, USA Hiroshi Yamada Worldwide Safety Sciences, Pfizer Global Research and Development, Nagoya Central Research Laboratories, SS-Nagoya, Pfizer Japan Inc., 5 – 2 Taketoyo, Aichi 470-2393, Japan Toshinori Yamamoto Worldwide Safety Sciences, Pfizer Global Research and Development, Nagoya Central Research Laboratories, SS-Nagoya, Pfizer Japan Inc., 5 – 2 Taketoyo, Aichi 470-2393, Japan Tsuyoshi Yokoi Drug Metabolism and Toxicology, Division of Pharmaceutical Sciences, Graduate School of Medical Sciences, Kanazawa University, Kakuma-machi, Kanazawa 920-1192, Japan

Preface

The main purpose of this book is to assemble up-to-date information on hepatotoxicity in a single edition. There have been several books published during the last three decades on hepatotoxicity and drug-induced liver disease. However, this text will probably be the first to collate most of the genomic technologies, including the latest toxicogenomics. The book is designed primarily for those research scientists currently engaged in the field of liver toxicity. However, it will also be of interest to toxicologists in general and to students and teachers in toxicology in particular. In addition, it will be of interest to industry, as well as to government regulators and risk assessors of foods, drugs, environmental and agricultural products. Saura C. Sahu

Acknowledgements

I am indebted to my mentors from whom I have learned so much throughout my professional career and who have paved the way for this book. I owe special thanks to the following scientists who have contributed directly or indirectly to this text and I am deeply grateful to each one of them. I thank Dr Thomas A. Cebula for his inspiration, enthusiasm, leadership, endless energy and passion for science while I also thank Dr Joseph E. LeClerc and Dr Richard B. Raybourne for their encouragement and support. In addition, I wish to thank Dr Daniel A. Casciano for his professional advice and contribution and Dr Philip W. Harvey, Covance Laboratories, UK, for his encouragement and support. Finally I thank all of the scientists who have contributed to this book in their own areas of expertise.

Section 1 Models for Hepatotoxicity Testing

1 Current in vitro Models to Study Drug-Induced Liver Injury Julio C. Davila, Jinghai J. Xu, Keith A. Hoffmaster, Peter J. O’Brien and Stephen C. Strom

1.1

Introduction

In a published report on the root causes of failed drugs over a 10-year period, it was concluded that the pharmaceutical industry as a whole is still facing the same challenge of selecting drug candidates with high efficacy and low toxicity (Schuster et al., 2005). Among all human organ toxicities, hepatotoxicity and cardiovascular toxicity were the two most prominent causes, accounting for two out of three market withdrawals in the last decade (Schuster et al., 2005). In clinical drug development phases, about half of attrition were due to insufficient efficacy and one third were due to toxicity (Schuster et al., 2005). While regulatory animal toxicity testing was able to identify more than 70 % of human toxicities in a retrospective analysis, hepatotoxicity in humans had the poorest correlation with regulatory animal toxicity tests (Olson et al., 2000). In only half of new pharmaceuticals that produced hepatotoxicity in the clinical stage was there any concordant signals in animal toxicity studies (Olson et al., 2000). Obviously, identifying better models to predict human hepatotoxicity is a critical need for the pharmaceutical industry. Drug-induced liver injury (DILI) can be broadly classified into two categories, based on incidence, animal model predictability and dose-dependency. DILI type 1 (DILI-1) is characterized by relatively higher incidence, reproducible in at least one animal species and dose-dependent increase in incidence and severity of the observed injury. Acetaminophen is a classic case of DILI-1 and accounts for nearly half of acute liver failure in the United States (Larson et al., 2005). It can be modeled in more than one strain of rodents (Mehendale, 2005)

Hepatotoxicity: From Genomics to in vitro and in vivo Models Edited by S. C. Sahu C 2007 John Wiley & Sons, Ltd

4

Hepatotoxicity

and has a clear dose-dependency in both animals and humans. In humans, the reported ratio of median dose leading to acute liver failure to maximally recommended therapeutic dose is 2 for unintentional overdose to 6 for intentional overdose (Larson et al., 2005). DILI-2, on the other hand, has a relatively lower incidence, occurring at therapeutic dosesfrom 1 in every 1000 patients to 1 in every 100 000 patients (Lee, 2003). It is typically not predicted by the classic animal models of rodents, dogs and monkeys. Because of its rare incidence and large patient-to-patient variability, there is no clear indication of dose-dependency, although within the same susceptible patient, an underlying concentration–response may be present. DILI-2 is also referred to as idiosyncratic hepatotoxicity. The term ‘idiosyncratic’ derives from the Greek meaning ‘mixture of characteristics’ and in the context of drug toxicity refers to the combination of genetic and non-genetic factors that make a patient susceptible to drug injury. Numerous epidemiological studies have identified increasing age (Andrade et al., 2005), females (Ostapowicz et al., 2002) and certain disease-associations (Boelsterli, 2003) as the top demographic factors associated with higher incidence of DILI. While the reasons for such drug idiosyncracy are not entirely clear, most likely there are multiple factors that encompass both toxicokinetic, and toxicodynamic or adaptive reasons. In terms of toxicokinetic reasons, there have been numerous cases where factors such as age, gender, disease states, enzyme induction and inhibition, genetic polymorphism, food and gut micro-organisms have been implicated in causing variability in the pharmacokinetics of drugs. These factors can easily cause several-fold variations in drug concentrations among individuals given the drug at the same dose (Tam, 1993). In addition, the practice of polypharmacy (patients taking multiple drugs) tends to increase with age, which may surpass the predictability of well-controlled clinical trials on drug–drug interaction studies that are typically conducted on a limited number of drug pairs (FDA, 1999). In addition, the presence of disease states, including inflammation and infection, can affect both drug metabolism (Renton, 2004) and drug transport (Fernandez et al., 2004; Hartmann et al., 2002). In terms of adaptive reasons, the human liver has a remarkable ability to regenerate. Indeed, livers that underwent partial hepatectomy can regenerate and patients can go on to live normal lives. In addition, most of idiosyncratic hepatotoxicity was evident only after weeks of drug therapy, not upon a single dose of drug administration. The liver’s adaptive response after the initial drug ‘insult’ is probably crucial to the final manifestation of the full-blown liver injury. These adaptive responses almost certainly involve tissue repair (or lack thereof) (Mehendale, 2005), and the innate and adaptive immune systems (Holt and Ju, 2006). Numerous factors, including genetic polymorphism in tissue repair and immune responses, systemic and/or tissue inflammation, disease states, and continued challenge by polypharmacy, can affect a particular patient’s response to drug-induced liver injury. The mechanisms of such adaptive responses, including signal transduction pathways (Jaeschke and Bajt, 2006; Schwabe and Brenner, 2006), interactions between hepatocytes and cells of the host immune systems (Minagawa et al., 2004) and cellular decision-making (Malhi et al., 2006), will likely be an important and fruitful area of research in the post-genomic era. In vitro cellular models of drug toxicity have unique and important roles to play in order to understand both the mechanisms of initial drug injury and the many signal transduction pathways involved in tissue repair. This present chapter will focus primarily on the use of in vitro models of drug-induced liver injury. The better known mechanisms of drug-induced liver injury have been reviewed in previous publications (Begriche et al.,

Models to Study Drug-Induced Liver Injury

5

Table 1.1 Major mechanisms of drug-induced liver injury Mechanism

Prototypical drug

Reference

Altered lipid metabolism causing fatty liver

Amineptine, amiodarone, doxycycline, tetracycline, tianeptine, pirprofen Cyclosporine A, estradiol-17 beta-d-glucuronide, taurolithocholate, ethinyl estradiol Flucloxacillin, diclofenac, tienilic acid, sulfamethoxazole, halothane

Letteron et al., 2003, and references therein Crocenzi et al., 2003a,b; Micheline et al., 2002; Roman et al., 2003 Aithal et al., 2004; Carey and van Pelt, 2005; Robin et al., 1996; Sanderson et al., 2006 Amin and Hamza, 2005; Reid et al., 2005

Decreased bile salt clearance causing cholestasis Formation of protein adduct causing immune reaction Increased oxidative stress leading to cellular injury Decreased mitochondrial function leading to apoptosis or necrosis Cytotoxic T cell-mediated cell killing Incomplete or dysregulated tissue repair

Sodium diethyldithiocarbamate, diclofenac, ketoconazole, acetaminophen and drugs that cause steatohepatitis Nimesulide, amiodarone, tamoxifen, stavudine, zidovudine and drugs that cause steatohepatitis Sulfamethoxazole, lidocaine, carbamazepine, lamotrigine, phenindione Acetaminophen, carbon tetrachloride, chloroform, thioacetamide, trichloroethylene, allyl alcohol

Begriche et al., 2006; Tay et al., 2005 Sanderson et al., 2006, and references therein Mehendale, 2005, and references therein

2006; Jaeschke et al., 2002a; Kaplowitz, 2002; Lee, 2003). These are summarized in Table 1.1. A single small-molecule drug may invoke a multitude of such mechanisms (e.g. amiodarone, perhexiline, diclofenac and acetaminophen). This has led some researchers to postulate the ‘multi-hit’ hypothesis of drug hepatotoxicity (Begriche et al., 2006; Letteron et al., 2003; Pirmohamed et al., 2002). Cellular models based on human liver tissues have unique capabilities for studying such DILI mechanisms, as illustrated by the examples shown in Table 1.2. However, before describing the applications of in vitro models, we will first explore the various types of in vitro models that are available to researchers today.

1.2 1.2.1

In Vitro Models to Study DILI Primary Hepatocytes

Currently, the primary hepatocytes system is the in vitro model of choice for studying drug metabolism and hepatotoxicity of new drugs (Castell et al., 2006; Davila et al., 1998;

6

Hepatotoxicity

Table 1.2 injury

Representative applications of cell-based assays to study drug-induced liver

Applications

Methodology

Multiparameter cytotoxicity

Combine several different readouts including multi-spectral cytometric analysis Neutral lipid stain (e.g. Oil Red O)

Steatosis Cholestasis Phospholipidosis Reactive metabolite Oxidative stress

Mitochondria damage Identify targets of toxicological importance

Uptake and efflux of taurocholate Phospholipid accumulation in cytoplasm or lysosomal stain GSH adduct formation; GSH depletion (e.g. monochlorobimane) Redox-sensitive dyes

Mitochondria membrane potential dyes (e.g. TMRM) RNAi technology and/or specific inhibitors

Representative reference O’Brien et al., 2006 Amacher and Martin, 1997 Kostrubsky et al., 2006 Gum et al., 2001; O’Brien et al., 2006 Lilius et al., 1996; Thompson et al., 1998) Lautraite et al., 2003; LeBel et al., 1992; Wang and Joseph, 1999 Haskins et al., 2001 Lee and Sinko, 2006; Pichler et al., 2005; Tan et al., 2005; Xu et al., 2005

Gebhardt et al., 2003; Gomez-Lechon et al., 2004; Guillouzo et al., 1977; LeCluyse, 2001; Sinz and Kim, 2006). Primary hepatocytes represent a unique system since they are able to retain, under a refined cultured condition, Phase I and II enzyme activities, as well as their inducibility by xenobiotics. However, primary hepatocytes that fail to preserve and/or re-establish cell polarity and the expression of liver-specific genes are less responsive to drugs and therefore do not accurately reflect the metabolic potential of the liver tissue (Berthiaume et al., 1996; Kocarek et al., 1992; LeCluyse et al., 1996; Tuschl and Mueller, 2006; Waring et al., 2003) It is well established that primary hepatocytes cultured on plastic surfaces and allowed to form an epithelial monolayer (2-dimensional configuration) lose not only up to 75 % of total CYP450 during the first 24-h period, but also liver-specific functions and differentiation processes (Davila and Morris, 1999; Farkas and Tannembaum, 2005; Gomez-Lechon et al., 2004; LeCluyse, 2001); therefore, it is critical that cell culture conditions be refined for studying liver metabolism and toxicity of drugs in humans. Several optimized culture conditions have been reported to maintain liver-specific functions and gene responsiveness (co-factors expression) to levels comparable with those in vivo; these include the use of r extracellular matrix or complex substrata (e.g. collagen and/or Matrigel , sandwich configuration) (Farkas and Tannembaum, 2005; LeCluyse, 2001; Schuetz et al., 1988; Sidhu et al., 1994), chemically defined culture conditions (e.g. the use of low concentrations of insulin and dexamethasone in a serum-free amino acid-rich culture medium) (Enat et al.,


7

1984; LeCluyse, 2001; Sidhu et al., 1994) and co-culture hepatocytes with other cell types (e.g. sinusoidal cells, kidney epithelial cells, kupffer cells) (Begue et al., 1984; Donato et al., 1994). r Primary rat hepatocytes cultured and overlaid with Matrigel in an optimized media enables the hepatocytes to retain and maintain longer-term viability and expression of more stable differentiated liver function, such as Phase I and II metabolizing enzymes, and responr siveness to drugs and chemicals (Davila and Morris, 1999; Davila et al., 1998). Matrigel is basically a solubilized basement membrane preparation extracted from the Engelbreth– Holm–Swarm (EHS) mouse sarcoma, a tumor rich in extracellular matrix (ECM) proteins. This complex matrix is mainly composed of laminin, collagen IV, heparan sulfate proteoglycans and entactin, which polymerizes at room temperature to produce biologically active matrix material resembling the mammalian cellular basement membrane. We have previously reported on the development of a modified sandwich configuration model initially described by Dunn (Dunn et al., 1991) and others (Kocarek et al., 1992; LeCluyse, 2001; Schuetz et al., 1988; Sidhu et al., 1994), where freshly isolated hepatocytes are resuspended, r r plated and cultured in a Matrigel –Matrigel system with serum-free media containing −7 insulin and hydrocortisone (10 M) (Davila and Morris, 1999). One advantage of using r this model is that the concentration of Matrigel can be easily monitored before and after the initial plating. The concentration of ECM used for culturing is known to affect the morphology and functionality of the hepatocytes. In preliminary studies, we found that rat r hepatocytes, cultured and overlaid with Matrigel at 0.35 mg/ml, acquire a 3-dimensional (3D) configuration and are reorganized as acinar structures; cells become more cuboidal with a distinctive canalicular network and the polarized phenotype and function of normal hepatocytes are well preserved (Davila and Morris, 1999) . Under these culture conditions, the basal levels of a variety of liver genes are maintained and can be induced/stabilized by xenobiotics to levels that are comparable to those achieved in vivo. The mRNA levels of most of the rat CYP450 enzymes (e.g. CYP1A, CYP2B, CYP3A and CYP4A) can be induced 2–4 h after the initial plating. This is an improvement from the traditional methods using primary hepatocytes where cells have been reported to be unresponsive to inducers during the first 24–48 h (Kocarek et al., 1992; Richert et al., 2002; Silva et al., 1998). Freshly isolated rat hepatocytes cultured under this refined culture condition are proven to be a valuable and important in vitro toxicological approach to assess the chemical-induced changes in expression of rat liver CYP450 and Phase II conjugating enzymes (Davila and Morris, 1999; Davila et al., 1998). However, the ultimate species of interest in predicting liver metabolism and toxicity of drugs is human. 1.2.1.1

Refinement of a Primary Human Hepatocyte System for Drug Metabolism and Toxicity

Primary human hepatocytes are a useful model system to potentially predict the metabolism and toxicity of compounds and allow a better understanding of the mechanism of actions of drugs in the human population. Caveats in using primary human hepatocytes are the limited supply of human liver tissues and the ability of such cultures to loose their phenotype after culturing. Despite their limitations, primary human hepatocytes are currently the system of choice to predict human risk before new chemical entities (NCEs) are first tested in Phase I clinical trials. In this section, we describe the refinement of a feasible, reproducible and simple cell culture system that could be used for routine toxicological applications early during drug discovery and development. This model has been specifically refined for assessing the

8

Hepatotoxicity

metabolism and toxicity of drugs. Validation studies supporting the use of this optimized human hepatocyte model for metabolism-mediated toxicity studies include: (1) microscopy analysis to identify hepatocytes in a 3D configuration with well define intracellular structures characteristic of normal and functional hepatocytes; (2) the measurement of several constitutive and inducible expressions of human Phase I (CYP1A1, CYP1A2, CYP2B6, CYP2C9, CYP2D6, CYP2E1, CYP3A4, CYP4A11 and fatty acyl CoA oxidase, FACO) and Phase II (uridine diphosphate-glucuronosyltransferase, UDPGT2B7; sulfotransferase, SULT2A1; glutathione-S-transferase, GSTYa) biotransformation enzymes at the mRNA and protein and activity level; (3) in vivo and in vitro comparison of relative Phase I enzyme mRNA levels; (4) the differential expression of CYP450 enzymes in human and rat hepatocytes following treatment with various inducing agents. 1.2.1.2

Procurement, Isolation and Culturing of Human Hepatocytes

Human donor livers used in these studies were received from the National Disease Research Interchange (NDRI, PA) within 16 h of patient procurement. Livers were screened at the NDRI for evidence of infectious agents and preserved with ViaSpan (Dupont/Pharma, Wilmington, DE) upon removal from donors. Livers were only accepted from donors with minimal fat content < 10 %, normal liver enzyme levels, low visible fibrosis, medications which would not be considered as potent liver enzyme inducers (particularly, CYP450s) and patients ranging from 14 to 60 years old. Freshly isolated human hepatocytes were obtained from The University of Pittsburgh, Department of Pathology (Stephen Strom’s laboratory) within 24 h after cell isolation and from our laboratory as indicated below. The isolation of human hepatocytes from the whole liver was performed by a three-step perfusion technique using liver sections in situ, as described by Strom (Strom et al., 1996). The culturing of freshly isolated human hepatocytes was then processed as indicated for rat hepatocytes (Davila and Morris, 1999) with some modification. Briefly, after isolation, liver cells were filtered through a sterile nylon mesh (100 μm) and collected in several 50-ml conical test tubes. Cells were centrifuged twice at 30 × g for 2 min and then the final pellet was resuspended with 30 ml Williams E solution containing hydrocortisone and insulin (10−7 M) and glutamine (4 mM). Ten ml of diluted Percoll (30 vol %) were placed in the bottom of each tube and centrifuged at 1100 × g for 5 min. The pellets were washed with Williams E at 30 × g for 2 min and then resuspended in 20 ml Williams E solution. Viability of the isolated hepatocytes used in these studies was typically greater than 90 %. r The final cell suspension was resuspended in diluted Matrigel solution (0.35 mg/ml) to 5 a final cell density of 4 × 10 cells/ml and plated in 12 (1 ml) or 6 (2 ml) well plates, as previously indicated for rat hepatocytes (Davila and Morris, 1999). 1.2.1.3

Cell Culture Treatment

Hepatocytes were treated for two consecutive days at 24 h and 48 h after initial plating with optimal concentrations of phenobarbital, PB (100 μM), rifampicin, RIF (10 μM), 3-methylcholanthrene, 3MC (1 μM), clofibrate, CLO (50 μM), tamoxifen (TAM, 10 μM), omeprazole (OMZ, 20 μM), pregnenolone-16α-carbonitrile (PCN, 1 μM) and hydrocortisone (HC, 20 μM). Compounds were dissolved in dimethyl sulfoxide (DMSO) and then diluted with Williams E prior to administration to the cultures (final DMSO concentration was 0.1 vol %). The concentrations of the inducers used in these studies were selected


9

based on preliminary experiments to determine the concentration of inducer producing the highest levels of enzyme expression in the absence of toxicity (data not shown). When the enzyme was not inducible, a maximum non-toxic concentration was used (50–100 μM). The concentrations of inducers used in this study were not toxic to the hepatocytes as indicated by morphological examination, MTT reduction assay and recovery of total RNA (data not shown). 1.2.1.4

PCR and Fluorogenic cDNA Probes

Messenger RNA was determined using real-time PCR. Primer sets specific for human cyclophylin (CYC, house-keeping gene), CYP450 1A1, 1A2, 2B6, 2C9, 2D6, 2E1, 3A4, 4A11, FACO, UDPGT2B7, GSTYa and SULT2A1, and TaqMan fluorogenic probes specific for each primer set are described in Tables 1.3 and 1.4, respectively. The TaqMan fluorogenic probes were designed in accordance to the guidelines issued by Perkin-Elmer. The cDNA samples were analyzed using an ABI Prism 7700 (ABI) sequence detection system (Perkin Elmer Biosynthesis, Foster City, CA). The results generated by the ABI system are expressed as cycle threshold (Ct), representing the PCR cycle at which an increase in TaqMan probes fluorescence is detected (specific signal) above the baseline signal for each amplicon (Table 1.5, A–F). The target gene Ct (Table 1.5, C) for a sample is then normalized by subtracting the target gene Ct from its specific cyclophilin Ct (Table 1.5, B), therefore adjusting all individual samples to the expression of a common housekeeping gene (Ct, Table 1.5, D). The resulting value for each sample is then further expressed as the difference in Ct for the sample compared to the difference in Ct for the vehicle control (Ct, Table 1.5, E) with the lowest Ct value across experiments. Ct provides a value that is equal to how many cycles each sample is from the experimental control. The final calculation yields a relative quatitation of mRNA expression for each sample by expressing the results as fold induction above the experimental level control. All studies were repeated a minimum of five times, unless otherwise indicated, and representative data are presented. 1.2.1.5

Statistical Analysis

Data are presented as ± SE of the mean and statistically analyzed, using Dunnett’s T-test (P < 0.05). Each determination was performed in triplicate (n = 3 wells) and the data presented are from one of five separate experiments, unless otherwise indicated, giving qualitatively similar results. 1.2.1.6

Results and Discussion

Microscopy analysis. Phase contrast and electron micrographs of three-day old primary r human hepatocytes cultured with Matrigel in a well defined serum-free culture medium r are orare presented in Figure 1.1. Primary human hepatocytes cultured with Matrigel ganized as acinar structures and acquire a 3-dimensional (3D) configuration, cells retain cell polarity and phenotype (Figure 1.1(a)) and viability for several weeks. This technique, where freshly isolated hepatocytes are suspended and plated in a culture media containing r diluted Matrigel (0.35 mg/ml) and then overlaid with a second layer of Matrigel, allows the development of a 3D configuration. Cells cultured in this fashion reorganize into structures that are characteristic of the tissue of origin, adopting an in vivo-like morphology (Berthiaume et al., 1996; Davila and Morris, 1999; Farkas and Tannembaum, 2005;

5 Sense primer (5 to 3 ) CTTgTCCATggCAAATgCTg gATgAgAACgCCAATgTCCAg ACAACgCTgAATggCTTC CACTCATCAgCTCTgTATTCg TgCTATgAgACTTgAgAgg TATggCCTTCACCACAACC ATgAAgCAACCCgAgACACCA AATggACATgAACAACCCTCA CgTgACATCgAAgTACAgg gTTCgAgCAAgTgAggCAC gCATCTACgAggCAATCTACCA gAgATTgATgggATgAAgC CTgAgTTATgAggAgCTg

Cyclophilin (CYC), #Y00052 CYP1A1, #K03191 CYP1A2, #NM 000761 CYP2B6, #X13494 CYP3A4, #D11131, #M18907 CYP4A11, #NM 000778 CYP2E1, #J02625 CYP2C9, #M61855 CYP2D6, #M33388 Fatty acyl-coA oxidase (FACO), #U07866 UDP-glucuronosyltransferase (UDPGT-PB), #J05428 Glutathione-S-transferase (GST)-Ya, # M14777 Hydroxysteroid sulfotransferase (ST), #X84816

gTgATCTTCTTgCTggTCTTgC CTgCCAATCACTgTgTCTAgC gCTgAACTCCAgTTgCTgT gTAgACTCTCTCTgCAACATgAg gCAAACCTCATgCCAATgCAg TCAACACAAgTCgTgCAATgg AACAACTCCATgCgAgCCAg CTCAgggTTgTgCTTgTCgT gAgAAgCTgAAgTgCTgCAgC CAAgCACAgAgCCAAgTgTC CACATCCAAAgAgTggTACTg AggTAgTCTTgTCCATggCTC CTCAgAAgTTgTgCTTTgTCC

3’ Anti-sense primer (5 to 3 )

Oligonucleotides for PCR primers for human CYC, CYP450s, FACO, UDPGT, GST and ST

Target gene, accession #

Table 1.3

179 175 266 355 236 243 234 158 261 294 343 268 196

Fragment size (bp)


11

Table 1.4 Fluorogenic probe sequences for human CYC, CYP450s, FACO, UDPGT, GST and ST. TaqMan probes were designed in accordance to the guidelines issued by Perkin-Elmer Target gene

Fluorogenic probe (TaqMan)

Cyclophilin (CYC) CYP1A1 CYP1A2 CYP2B6 CYP3A4 CYP4A11 CYP2E1 CYP2C9 CYP2D6 Fatty acyl-coA oxidase (FACO) UDP-glucuronosyltransferase (UDPGT-PB) Glutathione-S-transferase (GST)-Ya Hydroxysteroid sulfotransferase (ST)

CCACAATATTCATgCCTTCTTTCAC CTTggATCTTTCTCTgTACCCTgg CAgCATCATCTTCTCACTCAAgg CCAAggACCTCATCgACACCTACCT CCAAgCTATgCTCTTCACCgTgAC TCTgCTCAACACAgCCACgCTTTC ACAgTCgTAgTgCCAACTCTggACTC ACAAgTCAACTgCAgTgTTTTCCAAgCT CACTCATCACCAACCTgTCATCggT CAACCAAAgCAACAgCATCTgAgC CAATCCAgAAgACTgCTCgATCCA CAgACCAgAgCCATTCTCAACTAC CTCAAgAACAgCTCCTTTCAgAgC

LeCluyse et al., 2005) As indicated by electron microscopy (Figure 1.1(b)), these cells exhibit distinctive structural characteristics of in vivo hepatocytes, including numerous mitochondria, bile canaliculi, nuclei, glycogen, gap junctions and a rich smooth and rough endoplasmic reticulum. This suggests the presence in this cell system of the machinery needed for the cells to express a wide array of liver-specific functions

(A)

(B)

n

er

g bc

m

gj

Figure 1.1 Phase contrast (A, ×320) and electron micrographs (B, × 1200) of human hepaR tocytes plated and overlaid with Matrigel for three days. Freshly isolated human hepatocytes suspended, plated and cultured with Matrigel at (0.35 mg/ml) are reorganized as acinar strucutures and acquired a 3D configuration. The electron micrograph shows a well-defined nucleus (n), endoplasmic reticulum (er), mitochondria (m), bile canaliculi (bc), glycogen droplets (g) and gap junctions (gj)

B Cyclophilin-Ct

18.2 17.2

17.5

A Treatment

Control Sample A

Sample B

26.83

32.45 28.45

C Target gene-Ct

2−Ct (Control) = 1.00 2−Ct (Sample A) = 8.00 2−Ct (Sample B) = 30.27

Ct Control − Ct Control = 0.00 Ct Sample A − Ct Control = −3.00 Ct Sample B − Ct Control = −4.92

(Control)C − B = 14.25 (Sample A)C − B = 11.25 (Sample B)C − B = 9.33

F Relative quantitation

E Ct

D Ct

Table 1.5 Example of the results generated by the ABI 7700 sequence detector, showing TaqMan calculations to yield relative quantitation (mRNA expression)


13

Measurement of several Phase I and II biotransformation enzyme expressions in primary human hepatocytes. In this study, we found that both the constitutive and inducible expression of the major families and subfamilies of human mRNA CYP450 (CYP1A1, CYP1A2, CYP2B6, CYP2C9, CYP2D6, CYP3A4, CYP2E1, CYP4A11) and Phase II (UDPGT2B7, SULT2A1, GSTYa) enzymes, as well as the rate of protein synthesis, are maintained in culture for several days. The rank ordering for the average basal mRNA expression levels of human CYP450 and Phase II isoforms expressed in this primary cell culture system from the lowest to the highest was as follows: CYP4A11 < CYP2B6 < CYP1A1 < SULT2A1 < CYP3A4 \ < CYP2D6 < CYP1A2 < CYP2E1 < GSTYa < CYP2C9 < UDPGT2B7 < FACO. The expression of Phase I and II enzymes at the mRNA, protein and activity levels (e.g. CYP1A and CYP3A) in three-day old cell cultures are reported in Figures 1.2, 1.3 and 1.4, respectively. In agreement with other investigators (Bowen et al., 2000; LeCluyse, 2001; Strom et al., 2001), we found that CYP1A1, CYP1A2, CYP2B6 and CYP3A4 are inducible by 3MC, and PB and RIF, respectively, in human hepatocytes (Figures 1.2 and 1.3). Interestingly, the basal r r levels of CYP3A4 and CYP2B6 in human hepatocytes cultured by a Matrigel –Matrigel system did not disappear within the 48 h in culture as previously reported by others (Kocarek et al., 1992; Richert et al., 2006). One explanation would be the difference of the substrata and the culture media components used in these studies. However, in agreement with others (Donato et al., 1995; Silva et al., 1998), some variability in the induction of CYP3A and CYP2B among the donors was observed. Seven out of ten liver donors responded to RIF at 10 μM, while two out of ten cell cultures responded to PB at 100 μM. Furthermore, we have also found that RIF induces CYP1A1 mRNA in all of the human hepatocytes tested. CYP2C9, CYP2D6 and CYP2E1 were found to be unaffected by the inducing agents used in this study. This lack of responsiveness of these enzymes has been previously documented and may be explained by genetic polymorphism, differences in substrate specificity or the high constitutive levels of the enzyme (Elkins and Wrighton, 1999; Guengerich, 1997; Mimura et al., 1993; Waxman et al., 1991). In addition, the disruption of circulating levels of growth hormone patterns in vitro may have altered the constitutive levels and inducibility of some enzymes, e.g. of CYP2D6 and CYP2C9 (Gonzalez, 1996; Neuman et al., 1993; Prough et al., 1996; Waxman et al., 1991). These secondary contributors to enzyme induction were not studied as part of this investigation. Moreover, we found that CLO was not a good inducer of CYP4A11 and FACO; however, we have found that WY14,643 and cyprofibrate at 20 μM are good inducers of CYP4A11, 20- and 7-fold higher than controls, respectively (data not shown). To further characterize this in vitro human hepatocyte system, we studied the effects of liver enzyme inducers on the expression of several Phase II enzymes, UDPGT2B7, GSTYa, and SULT2A1 at the mRNA level (Figure 1.2(g–l)). In contrast to the expression of CYP450 enzymes, Phase II enzyme expression and activity have been scarcely studied in human liver samples. One reason is the lack of specific substrate probes for the various isoenzymes and the lack of sensitive analytical methodology (Burchell and Coughtrie, 1989; Eaton and Bammler, 1999). However, RT–PCR technology allows us to assess the ability of a drug to modulate hepatic Phase II gene expression, thus offering a specific and rapid alternative to quantification of gene induction by immunodetection or substrate metabolism. UDPGT2B7, GSTYa and SULT2A1 were found to be expressed at a high basal

14

Hepatotoxicity

Relative mRNA expression (arbitrary units)

Human CYP1A1

Human CYP 1A2 mRNA

(a) 120

120

90

90

90

*

60

60

30

PB

RIF


PB

RIF

CLO

CON 3MC

(f)

(e)

40 30

60

20

20

10

10

0

0

50

PB

RIF

CLO

Human CYP4A11 mRNA (g)

0 CON 3MC

50

PB

RIF

CLO

Human FACO mRNA

CON 3MC

50

(h)

40

40

40

30

30

30

20

20

20

10

10

10

0

0 CON 3MC

50

PB

RIF

CLO

Human UDPGT2B7 mRNA (j)

*

PB

RIF

Human ST2A1 mRNA

50

(k) 40

30

30

20

20

20

10

10

10

30

*

*

0

0 CON 3MC

PB

RIF

CLO

RIF

CLO

(i)

CON 3MC

CLO

40

40

PB

Human CYP2E1 mRNA

0 CON 3MC

50

CLO

50

30

*

* RIF

Human CYP2D6 mRNA

Human CYP2C9 mRNA

50

90

30

* PB

0

40

CON 3MC


30

* CON 3MC

*

(d)

60

0

CLO

Human CYP3A4 mRNA 150

*

30

*

0

120

(c)

(b)

120

CON 3MC


Human CYP2B6 mRNA 150

150

150

* CON 3MC

* PB

RIF

PB

RIF

CLO

Human GSTYa mRNA (l)

*

0 CLO

CON 3MC

PB

RIF

CLO

Figure 1.2 Analysis of mRNA (TaqMan) levels for the analysis of CY450 expression in human hepatocytes. Data represent the results obtained from the analysis of Phase I and II biotransforR mation enzyme expression in human hepatocytes cultured with diluted Matrigel following exposure to prototypical CYP450 inducing agents (3MC, 1 μM; PB, 100 μM; RIF, 10 μM; CLO, 50 μM). The effects of CYP450 enzyme inducers on the expression of CYP1A1 (a) and CYP1A2 (b), CYP2B6 (c), CYP3A4 (d), CYP2C9 (e), CYP2D6 (f), CYP4A11 (g), FACO (h) CYP2E1 (i), UDPGT2B7 (j), ST2A1 (k) and GST-Ya (l) are presented. Data presented are expressed as the mean (± SE) relative mRNA expression levels for each enzyme (fold induction). Data are from one of five separated experiments (N = 3 dishes) using five different donors giving qualitatively similar results

Models to Study Drug-Induced Liver Injury CON

3MC

PB

15

CLO

RIF

CYP1A1/A2

CYP2B6 CYP3A4 CYP4A11 CYP2E1

Figure 1.3 Analysis of CYP450 apoprotein expression in human hepatocytes cultured in the R presence of diluted Matrigel using Western immunoblotting. Microsomes were prepared from cells exposed to either vehicle (CON, DMEM/0.1 % DMSO) or prototypical CYP450 enzyme inducers (3MC, 1 μM; PB, 100 μM; RIF, 10 μM; CLO, 50 μM) for 48 h. Cells were lyzed and microsomal proteins analyzed for expression of CYP1A1/CYP1A2, CYP2B6, CYP3A4, CYP4A11 and CYP2E1 protein by Western immunoblotting. Data are from one of three separated experiments using three different donors giving qualitatively similar results

CYP1A activity

24

(b) Testosterone metabolism (nmol/min/mg)

EROD metabolism (nmol/min/mg)

(a) 28

20 16 12 8 4

1

CYP3A activity

0.8

0.6

0.4

0.2

0

0 Control

3MC 10 µM

Control

RIF 10 µM

Figure 1.4 Analysis of CYP1A (a) and CYP3A (b) activity levels by HPLC and fluorometric R analysis in primary human hepatocytes cultured with Matrigel after treatment with RIF and 3MC, respectively

16

Hepatotoxicity

level in this culture system but not markedly induced by chemical treatment. We found that UDPGT2B7 was inducible by 3MC, PB and RIF, and that SULT2A1 was inducible by 3MC and RIF, and that GSTYa was inducible by only 3MC (Figure 1.2(g–l)). 3MC and PB are known inducing agents of Phase II enzymes in the liver (Eaton and Bammler, 1999; Wormhoudt et al., 1999). Data presented here support the potential utility of this in vitro system and RT-PCR for assessing the induction of Phase II conjugating enzymes in human hepatocytes by xenobiotics. In vivo and in vitro comparison of Phase I and II enzyme mRNA levels. As part of the validation study, in vivo and in vitro comparison of Phase I and II enzyme mRNA levels were performed. In these studies, mRNA levels were measured from one hundred human liver tissues and from fourteen separated human hepatocyte cultures (3 day old cells) using real-time PCR. The mean of the target genes was calculated and reported as Ct values (see Table 1.5, D). As indicated in Table 1.6, the mRNA basal levels of CYP1A1, CYP1A2, CYP2C9, CYP2D6, CYP2E1, GSTYa and UDPGT2B7 enzymes in primary hepatocytes were highly maintained for three days and comparable to those of liver tissues. However, CYP2B6, CYP3A4, CYP4A11, FACO and ST2A1 mRNA basal levels were found to be decreased but returned back to levels comparable to liver tissues when CYP450 inducers such as rifampicin (RIF, 10 μM), phenobarbital (PB 100 μM) and WY-14,643 (20 μM) were administrated to the cell cultures. The establishment and validation of a cell culture system which mimics drug biotransformation in vivo is particularly important when considering the elimination of NCEs that generate toxic or reactive intermediate metabolites early in the development process. Differential expression of CYP3A by RIF, OMZ, TAM, PCN and HC in rat and human hepatocytes. The last set of validation studies were designed to examine the differential expression of CYP450 enzymes (CYP1A, CYP2B, CYP3A, and CYP4A) in human and Table 1.6 In vivo and in vitro comparison of relative CYP450 enzyme (mRNA) levels in human liver (data are presented as Ct values) Enzyme phase I–II 1A1 1A2 2B6 2C9 2D6 2E1 3A4 4A11 FACO ST2A1 GSTYa UDPGT2B7

-Ct in vivo

-Ct in vitro

8.9 4.5 6.4 0.9 2.9 −2.3 0.9 4.2 1 1 1.8 −0.2

8.5 2.5 8.7 1.0 3.4 2.0 5.2 10 4.6 4 1.1 0.11

-Ct in vitro induced cells — — 6.2 (Phenobarbital) — — — −0.4 (rifampicin) 3.8 (WY-14,643) 1.7 (WY-14,643) 3 (Phenobarbital)

Models to Study Drug-Induced Liver Injury (a)

17

180


Rat CYP3A1 mRNA 150 120 90

* 60

*

30

*

0 CON (b)

RIF

OMZ

Human CYP3A4 mRNA Relative mRNA expression (arbitrary units)

TAM

PCN

HC

PCN

HC

180

*

150 120

*

90 60 30

*

0 CON

RIF

OMZ

TAM

Figure 1.5 Comparison of CYP3A induction by rifampicin (RIF, 10 μM), omeprazole (OMZ, 20 μM), tamoxifen (TAM, 10 μM), pregnenolone-16α-carbonitrile (PCN, 1 μM) and hydrocortisone (HC, 20 μM) in rat (a) and human (b) hepatocytes using RT-PCR and fluorogenic cDNA probes (TaqMan). Data presented are expressed as the mean (± SE) relative mRNA expression levels for the enzyme (fold induction). Data are from one of three separate experiments using three different donors giving qualitatively similar results

rat hepatocytes. Results from our studies demonstrate a clear species difference in the induction of these enzymes in rat and human hepatocytes. In particular, CYP3A expression was found to be differentially induced by a number of known CYP3A inducers (Figure 1.5). RIF and TAM were found to markedly induce CYP3A expression in human hepatocytes, whereas OMZ, PC and HC were found to induce CYP3A expression in rat hepatocytes but not in human hepatocytes. These differences in expression of CYP3A, CYP2B and CYP4A enzymes following exposure to xenobiotics have been attributed in part to the

18

Hepatotoxicity

ability of xenobiotics to differentially activate several members of the nuclear hormone receptor (NHR) superfamily such as PXR/SXR, CAR and PPAR, respectively (Giguere, 1999; Peraza et al., 2006; Roberts, 1999; Schuetz et al., 2000; Xie et al., 2000). The quantity or the quality of these nuclear hormone receptors and the presence of a truncated, inactive form of the receptor, such as in the case of PPARα (Gonzalez and Lee, 1996; Peraza et al., 2006; Peters, 2005), have been suggested as potential factors responsible for such species differences in responses to chemicals. In addition, it has been reported that the species origin of the receptor, rather than the promoter structure of the gene, e.g. CYP3A, dictates the species-specific pattern of CYP3A inducibility (Luo et al., 2004; Xie et al., 2000). The ability of these primary rat and human hepatocytes to respond to a variety of chemicals and to detect differences in CYP450s induction suggests the presence in this cell system of multiple regulatory pathways and of transcription factors associated with functional receptors (Li and Chiang, 2006). Variations in drug metabolism by different species represent a challenge to early metabolism and safety screening. Differential drug-induced expression of liver cytochrome P450 (CYP450) enzymes has been reported to be responsible in part for variations in metabolism and toxicity among species (Hengstler et al., 1999; Kocarek et al., 1995; Lewis et al., 1998). These striking differences in drug-induced CYP450 expression among animal species make it difficult to rely solely on the use of laboratory animals as surrogate species for predicting drug metabolism and toxicity in humans. Therefore, human cellbased systems are important tools in assessing and predicting human drug metabolism and toxicity. r Primary human hepatocytes cultured in a refined cultured condition using Matrigel provides an attractive screening strategy for testing the qualitative and quantitative induction of human CYP450 enzymes by NCEs and to extend our understanding of the role of these NHRs in regulating CYP450 expression by xenobiotics. Results from these studies also highlight the importance of assessing enzyme induction across species as it relates to the potential for drug–drug interactions and metabolism-mediated toxicity. In conclusion, r r we have established a 3D-human cell culture system using a Matrigel –Matrigel configuration and a quantitative RT-PCR method using fluorogenic cDNA technology (TaqMan) to examine the effects of xenobiotics on the expression of a number of human CYP450 and Phase II conjugating enzymes. We have demonstrated that several Phase I and II liver biotransformation enzymes can be induced by xenobiotics in this system. In addition, we have further demonstrated the need for using human cell culture induction systems to assess the metabolism and toxicity of xenobiotics in the human population. 1.2.2

Stem Cell-Derived Hepatocytes

New platforms for discovery, target validation and predictive toxicity screening in therapeutic areas such as arthritis and inflammation and cardiovascular are needed for a positive identification of new therapeutic agents in preclinical development. Included in the testing scheme is the evaluation of compounds or recombinant gene products on in vitro cell systems, such as primary cell cultures, which best reflect the tissue of origin. While primary hepatocytes represent the desired model system for drug metabolism and toxicity, the limited supply of human tissues and the variable quality and interindividual differences hinder its routine use.


19

Stem cell technology provides unprecedented opportunities not only for investigating new ways to prevent and treat a vast arrays of diseases but also for changing the way we identify new molecular targets, discover and develop new drugs, as well as test them for safety. Because stem cells are self-renewing population of cells, they can be continuously cultured in an undifferentiated state and give rise to more specialized cells of the human body, such as heart, liver, bone marrow, blood vessels, pancreatic islet and nerve cells. Therefore, it offers an important new tool to develop unique in vitro model systems for testing drugs and chemicals and potentially predict or anticipate toxicity in humans (Davila et al., 2004; Sinz and Kim, 2006). Recent advances in the isolation and culture expansion of multi-potent stem cells from embryonic or adult tissue offer these cell systems as alternative sources of progenitor cells for wide number of tissue-specific cell types. Although human embryonic stem cells (HESCs) retain the potential to differentiate into all of the major cell types of the body, controversy surrounding their research use has limited their application to date. However, adult stem cells, which can be isolated from non-human embryonic tissue, retain their multipotency in culture, and therefore offer a useful cell culture system for the evaluation of compounds during drug discovery and development. Significant and continuing advances in stem cell research to generate an reliable supply of fully functional human hepatocytes has the potential to provide a consistent source of normal and functional human cells that would more closely predict the impact of a new drug on human livers in the body. The ability to obtain an unlimited supply of hepatocytes that retain the expression and activity of drug metabolism enzymes would revolutionize toxicity testing, address the largest bottleneck in new drug research and accelerate the drug development process. While the biological qualities of an optimal hepatocyte source could be debated, there are at least four properties that would seem essential to satisfy most of the requirements of stem cell-derived hepatocytes. The stem cell-derived hepatocyte must: (1) (2) (3) (4)

be routinely available; be easily cryopreserved and restored to culture; provide cells with a metabolic profile observed in adult hepatocytes; provide for a diverse genetic background of donors.

Although there are reports of stem cell-derived hepatocyte-like cells, it is safe to say that there are currently no published reports of cells that fulfill all of the above mentioned requirements. The properties of the ideal stem cell must be kept in mind, as certain cell types such as bone marrow, peripheral blood or amnion, would be expected to be routinely available and in sufficient numbers to provide for a wide range of genetic backgrounds; cell types such as embryonic stem cells (ES) may only be available from a far more limited number of individuals. These considerations could be critically important to eventual drug metabolism and toxicology studies with hepatocytes derived from these stem cell sources. The number of published reports concerning the differentiation of specific cell types into hepatocyte-like cells is too numerous to cite in the context of this limited review. The sources of stem cells reported to differentiate along a hepatic lineage is considerably smaller. By far the most common reports of non-hepatic cells becoming hepatocytes are from the bone marrow, peripheral or placental cord blood or embryonic stem cells of mouse or human origin.

20

Hepatotoxicity

1.2.2.1

Bone Marrow-Derived Hepatocyte-Like Cells

Petersen et al. (Petersen et al., 1999) were the first to suggest that bone marrow cells could differentiate to hepatocytes. Because putative hepatic stem or progenitor cells express surface marker antigens such as Thy-1 or c-kit which are also found on hematopoietic cells, it was logical to assume that they might have a common source. This initial report was soon followed by several others confirming and extending the initial observations demonstrating in vitro and/or in vivo differentiation of bone marrow-derived cells to hepatocytes (Alison et al., 2000; Lagasse et al., 2000; Krause et al., 2001). The initial exuberance was soon followed by stunning reports that most of the hepatic characteristics expressed in transplanted cells were the result of cell fusion with recipient hepatocytes (Wang et al., 2003; Vassilopoulos et al., 2003). Conflicting reports still remain in the literature. A consensus position is that there may be some low level differentiation of bone marrow-derived hematopoietic cells to hepatocytes although it is most likely quite limited (Wagers et al., 2002; Cantz et al., 2004; Yamaguchi et al., 2006). In vitro differentiation of bone marrow cells to hepatocytes would seem to be devoid of the ‘fusion artifact’ and might prove a useful method to derive hepatocyte-like cells from bone marrow. There are now several reports of the generation of hepatocyte-like cells from bone marrow cells propagated with growth factors such as hepatocyte growth factor (HGF) and/or basic fibroblast growth factors (bFGF) (Avital et al., 2001, 2002; Fiegel et al., 2003; Miyazaki et al., 2002, 2004; Kawasaki et al., 2005; Saji et al., 2004). While several hepatic characteristics such as the expression of albumin, alpha-1-antitrypsin, tyrosine aminotransferase, and urea production, were cited as evidence of hepatic differentiation, there are no reports of the production of cells with a complete set of adult hepatocyte characteristics. These initial results are encouraging, and if full hepatic differentiation could be completed from hematopoietic cells in vitro, they could be an extremely useful future source of hepatocytes. It is interesting that the cell type thought to be responsible for fusion with recipient hepatocytes following bone marrow cell transplants are the myelomonocytic cells in the bone marrow (Willenbring et al., 2004). Investigators have tried to induce hepatic differentiation in cultured monocytes (Ruhnke et al., 2005a,b). These authors report albumin secretion and urea production at rates lower than, but similar to, human hepatocytes and ethoxycoumarin-O-deethylase activity which was inducible with 3-methylcholanthrene and low but measurable CYP3A4 mediated drug metabolism. Tissue such as bone marrow is complex and contains many cell types. In addition to the hematopoietic cells, there are other cell types in bone marrow. In general, hematopoietic cells from bone show limited proliferation in culture and grow as a non-adherent cell type. A second cell type found in bone marrow grows as adherent cells and shows much greater potential to proliferate. This adherent fraction can be isolated from bone marrow, umbilical cord blood and even peripheral blood. Although not necessarily the same cell type, the adherent fraction is alternately referred to as stromal cells, fibroblasts or mesenchymal stem cells (MSCs). There are several reports of the differentiation of mesenchymal cells to hepatocyte-like cells, both in vitro and in vivo (Beerheide et al., 2002; Kakinuma et al., 2003; Kogler et al., 2004; Newsome et al., 2003; Schwartz et al., 2002; Nonome et al., 2005). Markers identified were mainly markers of early hepatic differentiation. In a careful study of MSC-derived hepatocyte-like cells, Ott and coworkers concluded that although the cells expressed some hepatic markers following transplantation into the liver, the MSC-derived


21

cells did not express several crucial hepatic markers such as cytokeratin expression (Sharma et al., 2005). The lack of cytokeratin expression would suggest that the cells remained mesenchymal and did not fully differentiate into an epithelial cell type such as a parenchymal hepatocyte. However, more recent publications keep the idea of mesenchymal stem cellderived hepatocytes alive (Aurich et al., 2006). More information on the expression mature liver genes such as the CYP450 genes and adult hepatocyte metabolic activities are needed to determine the actual state of differentiation of these hepatocyte-like cells (Hengstler et al., 2005; Teramoto et al., 2005) 1.2.2.2

Embryonic Stem Cell-Derived Hepatocytes

It might seem that the derivation of hepatocytes from embryonic stem (ES) cells might be the major focus of researchers trying to generate hepatocytes from stem cells. However, this is not the case. More publications report hepatic differentiation of stem cells from sources other then ES cells. There is good preliminary data to suggest that hepatocyte-like cells can be derived from ES cells (reviewed in Teramoto et al., 2005, and Lavon and Benvenisty, 2005). While most reports concern mouse ES cells, there are reports of human ES cell-derived hepatocyte-like cells (Lavon et al., 2004; Rambhatla et al., 2003; Shirahashi et al., 2004; Soto-Gutierrez et al., 2006). Each of the reports show the expression of early markers of hepatic differentiation as well as some more differentiated functions such as some limited drug ammonia metabolism or urea production. There are no reports of robust drug metabolism by any ES-derived hepatocyte-like cells. While quite preliminary, these initial results suggest that at least some of the ES cells are able to differentiate along a hepatic lineage. It is not entirely clear from the published work if differentiation is being directed by the experimental design or if spontaneous differentiation of ES cells is occurring in the cultures. More work will be needed to fully characterize the ES-derived hepatocytes and to optimize and direct the differentiation process. Results reported with pluripotent cells like ESs is complicated by the possibility of differentiation along other pathways. Hepatic marker genes such as albumin, alpha fetoprotein and transthyretin are not only expressed in developing liver, they are also expressed in yolk sac. Yolk sac performs many liver-like functions during fetal life. Thus, it is difficult to determine if the expression of hepatic marker genes is evidence of hepatic differentiation or simply differentiation to visceral endoderm of the yolk sac. A recent report by Asahina et al. (Asahina et al., 2004) suggests that the expression of CYP enzymes such as CYP7A1 was the only definitive marker of actual hepatic differentiation and could distinguish between differentiation of ES cells to visceral endoderm (yolk sac) and the definitive endoderm from which the liver and pancreas eventually are derived. By this measure, the expression of CYP7A1 in stem cell-derived hepatocyte-like cells indicated authentic hepatic differentiation. 1.2.2.3

Hepatocyte-Like Cells Derived from other Sources

There are a number of reports to suggest that hepatocyte-like cells can be derived from other tissue sources. Hepatocyte-like cells were generated from pancreatic tumor cells following long-term exposure to dexamethasone (Tosh et al., 2002). A number of liver-enriched genes we detected included glucose-6 phosphatase, carbamoylphosphate synthetase (CPS I) and glutamine synthetase (GS), as well as some CYP450 and Phase II enzymes. If high levels of mature liver enzymes could be induced, the transdifferentiation of other cell types into

22

Hepatotoxicity

hepatocytes might prove to be a useful cell source. Miki et al. (Miki et al., 2005) reported that cells with characteristics of pluripotent stem cells could be identified and isolated from human amnion. The amnion-derived stem cells could be induced to differentiate into cell types from all three germ layers, including cells with hepatocyte characteristics. The expression of many CYP450 genes and metabolic activity consistent with the expression and induction of CYP1A family proteins was reported (Davila et al., 2004). A significant observation was that CYP7A1 was expressed in amnion-derived hepatocyte-like cells, suggesting that the culture conditions used in the experiments induced differentiation of the stem cells to definitive endoderm and along an authentic hepatic lineage. 1.2.2.4

Hepatic Stem or Progenitor Cells

It is far beyond the scope of this discussion to thoroughly review the topic of hepatic stem cells. Recent reviews have updated this area (Forbes et al., 2002; Walkup and Gerber, 2006). Recent reports suggest the identification and isolation of putative hepatic stem or progenitor cells (Dan et al., 2006; Nowak et al., 2005; Schmelzer et al., 2006). The expression of some mature liver functions and the ability to expand the cells in culture suggest that the hepatic stem or progenitor cells identified in these studies could be a useful source of human hepatocytes (Dan et al., 2006; Nowak et al., 2005). 1.2.2.5

Summary and Suggestions for Future Studies

In here are a number of sources of cells which all give rise to cells with characteristics of hepatocytes. However, to date none of the stem cell sources, including those derived from the liver, have been shown to produce cells with full mature liver functions. Thus, additional effort is warranted with each stem cell type with the goal to improve hepatic differentiation, in vitro. A number of hepatic genes have been used in the studies cited above as markers of hepatic differentiation. In most cases, the reports identified markers of early liver development, such as the expression albumin and alpha 1-antitrypsin, cytokeratins or human hepatocyte antigen. While these markers are useful to show differentiation along a hepatic lineage, the hepatocyte-like cells acquire the expression of these gene products early in the differentiation process. The expression of later markers of mature liver function, such as CYP450 gene expression and metabolic activity, are rarely reported. In addition, in most studies when mature liver genes were reported they were merely detected by qualitative methods such as immunohistochemistry or RT-PCR. For the field to move forward it will be necessary to get quantitative estimates of the expression of a number of mature liver functions in the stem cell-derived hepatocytes which can be compared to authentic adult human hepatocytes. Although not hepatocyte-specific, the CYP450 genes are expressed at higher levels in liver than most other tissues. In addition, because the CYP450 gene products are critically involved in drug metabolism and toxicology, the expression of these genes will be central to the utilization of the stem cell-derived hepatocytes for basic research. The important drug metabolizing genes of the cytochrome P450 (CYP) families are expressed relatively late in hepatic development and, as such, we propose that they are useful markers of hepatic maturation. There are marked differences in the expression of the CYP enzymes between immature and mature hepatocytes. These differences in the expression of specific CYP450 genes can be used as a gage of hepatic maturation. It is believed that the CYP enzymes expressed early in hepatic development are needed and are responsible for the metabolism and elimination


23

of many endogenous compounds, such as steroids prostaglandins and retinoic acid. Both fetal and adult liver express genes of the CYP450 1A, 2C, 3A and 7A families; however, the expression of these genes even in fetal liver is generally in the range of a few percentage of normal adult levels. For most CYP enzymes there is a dramatic increase in expression in the postnatal period. In addition, the expression of specific family members differs in characteristic ways between fetal and adult liver which makes the expression of these genes diagnostic for the degree of hepatic differentiation. For example, Fetal liver expresses little of the adult form CYP1A2 (Maenpaa et al., 1993) but expresses high levels of CYP1A1 (Hines and McCarver, 2002; Omiecinski et al., 1990; Shimada et al., 1996; Yang et al., 1994). Adult liver expresses little CYP1A1 as its expression is slowly extinguished in the postnatal period. The expression of CYP1A2 increases after birth, however; it is one of the last of the CYP enzymes to develop in the postnatal period. Recent studies suggest that CYP1A2 levels may only reach 10 % of adult levels in the first postnatal month and may remain at or below 50 % of the adult levels out to one year (Sonnier and Cresteil, 1998; Tateishi et al., 1997). We propose that the expression of CYP1A2, is an indication of ‘true’ hepatic maturation and the ratio of expression of CYP1A1 to CYP1A2 provide a measure of the maturation process. A similar expression profile exists for members of the CYP3A family. Fetal liver expresses high levels of CYP3A7 (> 98 %) and little CYP3A4 (< 2% of total 3A) (Schuetz et al., 1994; Wrighton et al., 1988; Yang et al., 1994) while CYP3A4 is the major CYP enzyme expressed in adult liver. The expression of CYP3A7 is extinguished in most individuals over the first 6 postnatal months (Hines and McCarver, 2002; Stevens et al., 2003). Some CYP3A7 expression remains in 10–20 % of individuals but even in these CYP3A7 would only be expected to account for < 20 % of total CYP3A expression in that individual. Even after birth, CYP3A7 is not immediately extinguished such that the ratio of CYP3A4/3A7 may not equal 1 until 2–5 years of age. There is a change in the ratio of expression of CYP3A4 to 3A7 during development with the ratio of 3A4/3A7 extremely low during fetal development and high in adults. We suggest that the ratios of CYPs 3A4 to 3A7 and 1A2 to 1A1 will provide an estimate of the degree of hepatic differentiation of the stem cells. In addition to mRNA measurements, the metabolic activity of the individual CYP enzymes should be evaluated with probe drugs, and Western Blots should be run for individual CYP proteins. In this manner, direct comparisons can be made between stem cell-derived hepatocytes and authentic human hepatocytes for each endpoint studied. If implemented, suggestions like these will add quantitative analysis to this area of stem cell research. The development of a stem cell-derived hepatocyte system to examine the potential of new chemical entities (NCEs) to cause hepatotoxicity by a metabolism-mediated process early during the discovery process will provide a competitive advantage within the Pharmaceutical Industry. This novel cell-based system will provide an efficient means of aiding rational drug design and selection, selecting species differences in their responses to xenobiotics and, most importantly, for reducing the probabilities of causing unexpected adverse reaction in the liver when the compound reaches the market place. In addition to primary hepatocytes cultured in a sandwich configuration and stem cellderived hepatocytes, several recent developments in biological engineering and cellular biology have enabled a variety of new models for studying hepatotoxicity in vitro. Genetically engineered cells, as well as so called ‘three-dimensional (3-D) hepatocyte bioreactors’, have been established recently and are just now being applied to further our understanding of

24

Hepatotoxicity

toxic events in the liver. In the case of engineered cells, these systems seek to provide a basis for more mechanistic insight into our current understanding of drug-induced hepatotoxicity, whereas the 3-D perfused models attempt to re-establish the in vivo microenvironment as well as build and improve upon some of the limitations of existing primary cell models. 1.2.3

Cell Lines and Genetically Engineered Cells

Advances in molecular biology have revolutionized our ability to investigate and dissect complex mechanisms of hepatotoxicity. Although primary hepatocytes remain the ‘gold standard’ for investigating the potential of a novel therapeutic agent to elicit a toxic event, several cellular models can be applied to address similar predictions of hepatotoxicity. Cell lines are available for studying various mechanisms of toxicity and these cells and cellderived systems are becoming better defined and understood as new data emerge from the laboratories and into the literature. Some systems have gained application in the field of hepatotoxicity as they derived from liver-based, cancer cell systems (e.g. HepG2, WIF-B9 cells), whereas some systems have been engineered/created to study specific mechanisms of hepatotoxicity where aspects of a given mechanism are understood (e.g. CYP-engineered cells, BSEP-expressing cells or vesicles). The utility and predictability of these systems relies heavily upon understanding how each system is related to the intact liver in vivo. Although each system may not express the full complement of proteins responsible for overall hepatic function, we are learning now that each system may have merit for predicting hepatotoxic events very early on in the drug-discovery process. 1.2.3.1

WIF-B9 Cells

A sub-clone of the WIF 12-1 cell line, WIF-B9 cells are derived from Fao rat hepatoma cells fused with WI38 human fibroblasts. As a result of their multi-species origin, these cells co-express many rat and human liver-specific proteins (Decaens et al., 1996; Ihrke et al., 1993; Shanks et al., 1994). These cells are capable of synthesizing albumin and were shown recently to express an array of rat and human P450 isoforms, some of which are inducible (Biagini et al., 2006a; Decaens et al., 1996). In addition to metabolic capacity, WIF-B9 cells are polarized, such that they establish functional canalicular spaces into which the cells are capable of transporting bile acid-like molecules (Bravo et al., 1998). Interestingly, these cells do not express the major human protein responsible for the uptake of bile acids into the cell (NTCP; sodium-taurocholate co-transporting polypeptide). However, they may express functional rat Ntcp and/or human OATPs (organic anion transporting polypeptides) that also have been shown to uptake bile acids (Konieczko et al., 1998). As in the case of NTCP, WIF-B9 cells do not express the human isoform of BSEP (bile salt export pump), the transport protein responsible for canalicular excretion of bile acids across the canalicular membrane. Therefore, due to the functional expression of rat Ntcp and rat Bsep (demonstrated by the vectorial excretion of a fluorescently labeled bile acid in these cells), the ability of this model to predict transporter-mediated hepatotoxicity (e.g. druginduced cholestasis) would be limited to predictions based upon the rodent genes (Bravo et al., 1998; Gradilone et al., 2005). Although, this system may better reflect transportermediated toxicity relative to that in the rat, WIF-B9 cells have shown promise in the prediction of hepatic toxicity, as measured by general cytotoxicity measurements (Biagini et al., 2006b). Specific evaluation of WIF-B9 cells for determination of alcohol-induced


25

liver injury indicated that CYP2E1 was functional and that alcohol dehydrogenase (ADH) and aldehyde dehydrogenase (ALDH) activity was intact and similar to that observed in isolated hepatocytes (Schaffert et al., 2004). Moreover, the recent demonstration of aquaporin expression and localization suggests that this cell line may have utility in the prediction of toxic events that result in retarded bile formation in the liver (Gradilone et al., 2005). 1.2.3.2

HEPG2 Cells

Probably the most studied cell line with respect to hepatotoxic endpoints is the human hepatoma HEPG2 cell line. HEPG2 cells were first identified and isolated from a Caucasian adolescent over two decades ago (Aden et al., 1979; Knowles et al., 1980). Since their discovery in 1980, these cells have been utilized to examine various mechanisms of hepatotoxicity from ethanol-induced liver damage to most recently a high-content-screening approach for predicting hepatotoxicity due to drug administration (Neuman et al., 1993; O’Brien et al., 2006). One characteristic of HEPG2 cells that makes them attractive for use in hepatotoxicity studies is the expression and inducibility of relevant drug metabolizing enzymes. Expression and function of the key Phase I enzymes CYP1A, CYP2B, CYP3A, and CYP2E as well as several Phase II enzymes (e.g. UDPGT and GST) has been demonstrated and several of these enzymes can be induced by prototypical enzyme inducers (Knasmuller et al., 1998, 2004). However, it should be noted that not all drug metabolizing enzymes are present in HEPG2 cells; those that are may not be expressed at levels similar to those observed in vivo and may not respond to inducers similar to that in primary cells (Smith et al., 2005). Nevertheless, some investigators have transfected metabolic enzymes into this cellular system to establish a more realistic model for hepatotoxicity. HEPG2 cell transfection with CYP3A4, followed by administration of compounds known to form toxic metabolites, resulted in P450-specific formation of reactive species, leading to measured cytotoxicity (Vignati et al., 2005). Furthermore, following treatment with acetaminophen, HEPG2CYP2E1 cells generated reactive metabolites that formed protein adducts and damaged mitochondria, hence suggesting that introduction of mechanism-specific enzymes known to cause toxicity into HEPG2 cells may yield a predictive model for evaluating novel molecular entities (Bai and Cederbaum, 2004). In addition to drug metabolizing enzymes, expression of nuclear transcription factors in HEPG2 cells has provided insight into mechanisms of hepatotoxicity. ‘Upregulation’ of adipose differentiation-related protein through PPARγ was demonstrated in HEPG2 cells, offering insight into the mechanism of hepatic steatosis (Motomura et al., 2006). Furthermore, TCDD was able to upregulate CYP1A1 (as an indicator of oncogenesis) in HEPG2 cells through the aryl hydrocarbon receptor (Zhang et al., 2006). However, one must interpret these data carefully as CYP1A1 induction in HEPG2 cells is much more sensitive to that observed in primary hepatocytes and may not translate to the in vivo response. Much like the P450 enzymes, nuclear receptor transfection into HEPG2 cells also has been utilized to understand the involvement of these molecular factors in various mechanisms of hepatotoxicity (McCarthy et al., 2004; Vignati et al., 2004). HEPG2 cells also show promise for prediction of phospholipidosis, mitochondrial toxicity and oxidative stress (Bova et al., 2005; Kessova and Cederbaum, 2005; Sawada et al., 2005). Although these experimental systems have been optimized to predict specific hepatotoxic endpoints, the idea that HEPG2 cells can predict overall hepatotoxicity from a

26

Hepatotoxicity

mechanism-independent perspective is now becoming accepted. Evaluation of multiple endpoints through the use of high content screening (see Section 1.3 below) in HEPG2 cells suggests that this cellular system can detect human hepatotoxicity with 80% sensitivity and 90% specificity (O’Brien et al., 2006). These results show promise for this cell line to predict overall hepatotoxicity and underscore that even after more than 25 years of research with this cell line, we continue to uncover the underlying biology of the system and develop means by which to exploit the cells to predict in vivo adverse events. 1.2.3.3

BSEP-Expressing Cell Lines

Modulation of normal bile salt export pump (BSEP) activity (e.g. through inhibition) has been identified as a major molecular mechanism responsible for drug-induced cholestasis observed in the clinic (Fattinger et al., 2001; Funk et al., 2001). Several methodologies have been created to assess the potential for novel therapeutic agents to interact with the BSEP and cause a clinical manifestation of cholestasis. Of the systems available, BSEPtransfected cells and vesicles derived from these cells show a high degree of concordance with clinical observations of cholestasis. One of the first and simplest experimental models involved the expression of a rodent BSEP using an Sf9 insect cell-based system, followed by formation of vesicles from the BSEP-expressing cells (Stieger et al., 2000). The addition of the bile acid taurocholate, as well as ATP, to the vesicle system resulted in functional BSEPmediated transport into vesicles. In addition, several known inhibitors (e.g. cyclosporine A, glibenclamide and rifampicin) were able to inhibit the transport of taurocholate by the BSEP. Although the cell lines expressing the rodent isoforms of the BSEP were generated very soon after the full-length clone was isolated, due to the presence of a cryptic bacterial promoter in the BSEP cDNA, the human isoform of the BSEP was much more difficult to clone and express. However, two groups were able to clone and express functional human BSEP into Sf9 cells and demonstrate a high degree of functional similarity of this protein to the rodent isoform (Byrne et al., 2002; Noe et al., 2002). Following the functional expression of the human BSEP, several groups have utilized these Sf9-derived vesicles to explain the observed hepatotoxicity of therapeutic agents such as nefazadone, ritonavir, and saquinavir (Kostrubsky et al., 2006; McRae et al., 2006). Most recently, the transfection of the human BSEP (coupled with human NTCP) into LLC-PK1-polarized cells has enabled the study of vectorial/polarized transport of bile acids across a cell monolayer (Mita et al., 2006). Undoubtedly, as our understanding of the role of specific transporter proteins in various mechanisms of hepatotoxicity continues to expand, the more predictive these systems will be in assessing the potential for a toxic event resulting from the administration of a novel therapeutic agent. 1.2.3.4

CYP-Engineered Cells

In some instances, hepatotoxicity may not be related to the original molecular entity administered to the patient. Rather, the formation of a highly reactive metabolite by CYP450 may contribute to an observed clinical adverse event. To predict the potential for CYP450 involvement in toxicity, cells have been engineered to express each of the major P450 isozymes responsible for the metabolism of xenobiotics. Immortalization of liver epithelial cells with an SV40 T antigen virus followed by transfection with human CYP450 genes yielded a THLE-CYP cell line that has been applied to address metabolism-mediated toxicity (Pfeifer et al., 1993). These transfected cells have been shown to express upwards of


27

20-fold higher levels of CYP450 and have low basal expression levels of P450 enzymes, making them an ideal system to study isolated mechanisms of metabolism. The most studied compound in this system, Aflatoxin B1, is able to form DNA adducts that are modulated by the presence of the CYP enzymes (Mace et al., 1997). Moreover, these cells were able to identify the oxidative pathways for diclofenac and derive the contribution of each P450mediated process to the overall metabolism in vivo. (Bort et al., 1999a). Specific CYP isozymes have been expressed in other cell lines. When exposed to 1,3-dichloropropanol or cyclophosphamide, NIH-3T3 cells or V79 cells engineered to express individual CYP isoforms displayed a higher degree of toxicity than the control cells (Bull et al., 2001). This increased sensitivity was hypothesized to be due to the ability of the CYP-expressing cells to form reactive metabolites of these two compounds, thereby enhancing the cytotoxicity. Although these cells have promise for predicting the impact of Phase I metabolism on the hepatotoxic endpoint, these reagents often lack Phase II enzymes, overexpress the P450 of interest and do not respond to inducers of metabolism pathways. As a result, ultimate in vitro/in vivo correlation of these systems remains a challenge for their widespread use and adoption for assessing hepatotoxicity. 1.2.3.5

Other Cell Lines and Engineered Cellular Systems

In addition to the cellular systems mentioned above, several other cell lines have been used, although not as widespread, to address mechanisms of hepatotoxicity. These reagents include cell lines such as HUH-7, HepaRG and HEP3B cells (Aninat et al., 2006; Le Vee et al., 2006; Manov et al., 2002; Xu et al., 1997). Continued evaluation and testing of these cell lines will be a critical step for the acceptance of these as viable reagents to predict toxic events in the liver. 1.2.4

Three Dimensional (3D) Hepatocyte Bioreactors

Outside of using the intact organ ex vivo, one of the major limitations of current in vitro models of liver function has been the static nature of most in vitro systems. Due to the static nature of these experimental models, it is often difficult to re-establish the microenvironment that cells experience under normal in vivo conditions, and thus mimic in vivo liver physiology. It has been well established that flow through the liver sinusoids imparts shear stress onto liver cells that dictates cellular gene expression and function (Hara et al., 2003; Sato et al., 1997, 1999). Moreover, extensive cell-to-cell contact, establishment of cell polarity and the impact of non-parenchymal cells all can be important for re-establishing in vivo hepatic function (Hamilton et al., 2001; Hoebe et al., 2001; Hoffmaster et al., 2004; LeCluyse et al., 2005; Saad et al., 1993). Static systems have been used extensively, and albeit to some degree of success have shown promise for predicting some incidences of hepatotoxicity in the clinic. In contrast to the more frequent use of static cultures of primary hepatocytes and hepatic cell lines to predict hepatotoxicity as mentioned above, little measure of hepatoxic endpoints in these complex, perfused, 3D models of liver function have been performed. Given the in vivo phenotype observed in many of these novel systems, one can only imagine that the predictive nature of each of these models may far exceed our current ability to assess hepatotoxic events before entering the clinic with a promising new chemical entity. Although the application to the field of hepatotoxicity has not yet been

28

Hepatotoxicity

exploited with these emerging technologies, the background and rationale for several of the leading 3D perfused models is described briefly below. 1.2.4.1

MIT Liverchip

One of the most promising 3D models for recapitulating liver function has been developed by fabricating a series of small channels into a scaffold where primary hepatocytes can be seeded and subsequently perfused with a user-defined cell culture medium. Developed at the Massachusetts Institute of Technology (MIT), this unique microfabricated bioreactor design enables morphogenesis of 3D tissue structures, optimal oxygen transfer to the established tissue and physiologic shear stresses to be imparted upon the cells cultured in the device (Powers et al., 2002a,b). As a result, these in vitro cultures of cells yield morphology similar to that observed in the intact liver, expression of liver-specific proteins and mRNA near physiologic levels and metabolic function for several key Phase I enzymes in line with formation rates and metabolic profiles observed in freshly isolated hepatocytes or native, intact tissue (Sivaraman et al., 2005). For example, whereas cytochrome P450 levels decline rapidly in primary hepatocytes shortly after isolation, the expression of Cyp2c and Cyp3a was maintained in the ‘Liverchip’, and further supported by formation of specific testosterone metabolites of these CYP isoforms. Testosterone hydroxylation rates in these studies also were similar to rates observed in vivo (Sivaraman et al., 2005). One key advantage of this approach is the ability to scale the system and adapt the system to a high-throughput configuration. Since the functional unit of the reactor is a single channel, the mass of tissue in the device can be controlled directly through fabrication of larger or smaller scaffolds containing more or less individual channels. In addition, the device has been recently adapted to a multi-well configuration amenable to higherthroughput and automated liquid handling – ideal design criteria for screening new chemical entities in a drug discovery environment (unpublished data). Some challenges remain before this technology gains wide application in the field of toxicology. As the majority of data generated to date reflects experiments conducted with rat hepatocytes, the introduction and success of human hepatocytes in the ‘Liverchip’ will be very important for translating data from this system to understand the clinical manifestations of hepatotoxicity. The further characterization of cellular function in the device (e.g. bile acid transport, mitochondrial function, etc.), coupled with specific experiments designed to assess different mechanisms of hepatotoxicity will help demonstrate the overall utility of this system. Nevertheless, the results and data generated to date suggest strongly that this system offers superiority over currently available static culture systems for evaluating liver-specific function. 1.2.4.2

Hollow Fiber Reactor

Originally designed as an extracorporeal bridging device for liver transplant patients, the multi-compartment woven fiber bioreactor has shown that hepatocytes spontaneously form 3D structures once implanted into the device (Gerlach et al., 2003). These structures display morphology similar to that observed in vivo, and cells can be maintained upwards of seven weeks in culture, without undergoing major histological changes (Gerlach et al., 1994; Gerlach et al., 1995). The fundamental principal behind this technology is the incorporation of artificial capillaries providing perfustate flow that promote the formation of 3D structures. When cultured for two weeks, freshly isolated cells were capable of maintaining liverspecific function such as urea synthesis, albumin production, P450 function and glucose


29

metabolism (Zeilinger et al., 2000, 2002). Recent data also suggests that non-parenchymal cells are able to organize into functional structures complementary to the hepatocytes, and cells may have some proliferative capacity when cultured in the device (Zeilinger et al., 2004). 1.2.4.3

Zonation-Based Bioreactor

A flat-plate reactor that incorporates both flow and the concept of oxygen gradients over the length of the reactor chamber has been fabricated (Allen and Bhatia, 2003; Allen et al., 2005). These researchers constructed a model based upon the concept of zonation of the liver in vivo, and have recreated this environment by varying the degree of oxygen tension over a flat plate of hepatocytes with or without co-culture with fibroblasts (Allen and Bhatia, 2003; Jungermann, 1986; Pang et al., 1994). Once this gradient is established in culture, regional heterogeneity of drug metabolizing enzymes (e.g. CYP3A, CYP2B) can be seen, similar to the patterns observed in intact liver. Cell viability can be maintained in this device for at least five days and P450 enzymes can be induced by prototypical inducing agents such as dexamethasone. The zonation-based bioreactor is one of the few novel perfusion based system where specific toxicity studies have been designed and conducted. Treatment of the bioreactor with a range of acetaminophen concentrations resulted in zonal toxicity of the hepatocytes in the device; toxicity was higher in the low oxygen portion of the culture (Allen et al., 2005). These data are supported by the hypothesis that low oxygen concentrations promote glutathione depletion, an important detoxification pathway for the toxic NAPQI acetaminophen mediated metabolite. However, enzymes other than CYP3A responsible for acetaminophen metabolism (e.g. CYP2E, UGT, SULT, etc.) have yet to be characterized in the reactor system. 1.2.4.4

Microvascular-Based MEMS Bioreactor

The highly branched vasculature of the liver represents a challenge for reconstruction in an in vitro system. Nevertheless, using principles based on photolithography and silicon micromachining, a group of scientists has machined successfully a vascular network onto a silicon wafer using micro-electromechanical system (MEMS) techniques (Fidkowski et al., 2005; Kaihara et al., 2000). Transfer of this network onto a biodegradable polymer (the wafer acts as a micro-mold) and subsequent removal results in the formation of a two-dimensional polymer scaffold similar to the vascular and capillary networks of the liver. The resultant vascular channels can be seeded with endothelial cells; hepatocytes are seeded into an adjacent parenchymal compartment. These compartments are separated by the presence of a highly permeable membrane that allows the movement of small molecules but isolates the cell types within the device. Oxygen exchange can also happen across this membrane and although the cells remain physically separated, inter-cellular communication through soluble factors is accommodated in this design. When multiple monolayers of hepatocytes are layered and adhered to each other, these ‘sheets’ of hepatocytes recreate a three-dimensional tissue structure that is highly vascularized (Kulig and Vacanti, 2004). The cells remain viable for several days in culture and are able to maintain liver-specific function (e.g. albumin synthesis, urea production, etc.). As the initial application of this technology focused on the fabrication of an implantable human liver, the incorporation of biodegradable and biocompatible polymers has been the recent focus for the development of the technology. However, researchers have realized quickly that this model system shows

30

Hepatotoxicity

significant promise for the evaluation and screening of potential hepatotoxicants in a human cell-based system prior to moving into the clinic. Experiments are just now being designed to address and exploit this opportunity to better understand the potential for hepatotoxicity in this more complex, more physiologic model system.

1.3

Applications of in Vitro Models

For in vitro models to be considered relevant to in vivo situations, several criteria have to be satisfied: (1) the models are well-characterized and understood; (2) the concentrations of the test articles used are considered relevant in vivo; (3) the measured outcomes are considered relevant in vivo; (4) the underlying mechanisms are considered relevant in vivo. Representative applications of cell-based assays currently used to study DILI are summarized in Table 1.2. Although enzyme leakages (ALT, AST, LDH leakages) can be easily measured both in vivo and in vitro, they do not provide any insight into the underlying mechanisms of damage. Recently, high-content analysis (HCA) has emerged as a versatile technology platform with several distinctive advantages and applications. HCA is essentially a multi-parametric analysis of data that often include both biochemical and cytometric measurements. The HCA readouts may be kinetic on live cells, or single timepoint readouts on fixed cells. High-resolution analysis of sub-population of cells in each well, as well as the morphometric measurements of subcellular organelles, is often part of the HCA readouts. These image cytometric measurements compliment the traditional biochemical readouts such as enzyme leakage, ATP content, albumin synthesis or urea synthesis. High-content screening (HCS) is an abbreviated version of HCA, by selecting a few parameters from the HCA that were deemed to be most informative, predictive and robust. By selecting only a subset of the most informative parameters to measure, HCS can significantly increase the throughput of HCA to meet today’s drug discovery needs. 1.3.1

High-Content Screening for Human Hepatotoxicity Potential

Recent studies have demonstrated that the clinical occurrence of drug-induced human hepatotoxicity is highly concordant with in vitro cytotoxicity when assessed in a cell-based model with a novel combination of critical features and using high content screening (HCS; (O’Brien et al., 2006)). In contrast, concordance was low for previous cytotoxicity assays applied to drugs that produce human hepatotoxicity, because of poor assay sensitivity (Xu et al., 2004). Whereas these previous assays were only effective with severely hepatotoxic chemicals and drugs, they had high specificity, with positive test results being more than 90 % indicative of hepatotoxicity potential. HCS technology applies epifluorescence microscopy and image analysis to live cells incubated in microtiter plates under physiological conditions. It rapidly and automatically photomicrographs and quantitates fluorescence of multiple dyes at different subcellular locations (Haskins et al., 2001). This enables sensitive detection and following of the progression of key cytomorphologic and intracellular biochemistry effects of hepatotoxic drugs, such as on cell proliferation, nuclear and mitochondrial DNA, mitochondrial mass and activity, intracellular calcium, lysosomal mass, oxidative stress, and cell structure (Figures 1.6 and 1.7).


31

Figure 1.6 Detection of human hepatotoxicity potential with high-content screening. Composite images of hepatocytotoxic changes produced by incubation of HepG2 cells with cerivastatin (25 μM) for three days. 1, increased mitochondrial potential; 2. increased ionized calcium; 3, permeabilized plasma membrane with increased Ca and decreased mitochondrial membrane potential; 4, ruptured cell

There are several critical features of the cell model used with HCS that produce high assay sensitivity and concordance with human toxicity, in contrast to past cytotoxicity models (O’Brien et al., 2006; O’Brien and Haskins, 2006). These include duration of treatment, multiparametric monitoring of individual and live cells, drug metabolic competency and testing at concentrations relevant to efficacious concentrations. Most critical was the need for sufficient time to allow expression of the cytotoxicity (O’Brien et al., 2006; Slaughter et al., 2002; Xu et al., 2004). Three days of incubation of cells was effective for more than 95 % human hepatotoxic drugs, whereas a single day of incubation was frequently ineffective (O’Brien et al., 2006) or produced cytotoxicity at a much higher concentration (O’Brien et al., 2006; Slaughter et al., 2002; Xu et al., 2004). Next most important for predictivity was the number and choice of parameters measured. Early, sublethal effects on cell proliferation, cell morphology and mitochondria occurred consistently and ubiquitously with toxicity and when used collectively were most diagnostic. The occurrence and timing of effects on intracellular calcium concentration, lysosomal mass, oxidative stress or plasma membrane permeability frequently provided additional information indicative of mechanism of toxicity. For example, in Figure 1.7, phospholipidosis from cationic amphiphilic drugs, mitochondrial DNA depletion by nucleoside reverse transcriptase inhibitors that also inhibit mitochondrial DNA polymerase gamma and redox cyclers that produce reactive oxygen species are demonstrated. In addition, in Figure 1.6, the complex mechanism of statin-induced toxicity is demonstrated with early sublethal effects on apoptosis, mitochondrial function and calcium homeostasis (Diaz and O’Brien, 2006). Use of human hepatocytes with potential for drug metabolism is also important for assessment of hepatotoxicity as numerous drugs produce this effect by their hepatic metabolites. Drugs producing idiosyncratic hepatotoxicity and/or toxicity by their metabolites

32

Hepatotoxicity (a)

(b)

(c)

Figure 1.7 Mechanisms of human hepatotoxicity identified in HepG2 cells by high-content screening. (a) Phospholipidosis was produced by incubating cells with 12.5 μM imipramine for three days and then staining them with Hoechst 33342 for DNA, Mitotracker Far Red for mitochondrial mass and Lysotracker Green for lysosomal mass, as indicated. (b) Mitochondrial DNA depletion by nucleoside reverse transcriptase inhibitors is demonstrated by incubation of HepG2 cells for three days with 1.6 μM zalcitabine and staining nuclear and mitochondrial DNA (arrows) with Picogreen. (c) Oxidative stress produced by menadione. HepG2 cells were treated with 100 μM menadione and stained with Hoechst 3342 and dihydroethidium. Selection of objects for assessment is indicated by the circles and is based on Hoechst 3342 staining of nuclear DNA. Ethidium forms as a result of oxidative product of dihydroethidium, binds nuclear DNA and fluoresce with a different wavelength than the Hoechst. This is seen in the photomicrograph as increased whiteness of the encircled areas (right)


33

(Kalgutkar et al., 2005; Kaplowitz, 2005) were detected with HepG2 cells as effectively as drugs producing toxicity directly (Table 1.7). This high concordance contrasted remarkably with the 15 % concordance of seven other conventional assays in which a 50 % effect at 30 μM was considered positive for changes in any of the seven readouts: DNA synthesis, protein synthesis, glutathione depletion, superoxide secretion, caspase-3 activity, membrane integrity and mitochondrial reductive activity. The basis for this difference in sensitivity was not determined but may relate in part to induction of metabolic competence over the three days of exposure in the HCS assay. Finally, assay effectiveness also depended on measurement at the single-cell level in order to allow exclusion of extracellular staining or artifact or dead cells. Determination of the concentration producing cytotoxicity in the HCS sublethal cytotoxicity assay was assessed by 12-point dose response curves (Figure 1.8). A toxic effect was defined as the point when values for the parameter departed from the baseline and negative controls by more than two coefficients of variation. In Figure 1.8, typical dose – responses are illustrated in which all eight parameters measured were affected. The pattern and sequence of changes in the different parameters frequently reflected the mechanism of toxicity. For fenofibrate (Figure 1.8), there was nuclear swelling and inhibition of cell proliferation, followed by mild increases in intracellular calcium with some loss of mitochondrial membrane potential and an increase in membrane permeability, followed by overt oxidative stress with mitochondrial biogenesis. This pattern contrasted with that of cerivastatin (Figure 1.6), where there was first nuclear shrinkage and increased mitochondrial membrane potential, followed by increased intracellular ionized calcium. At higher concentrations, the calcium progressively increased, the mitochondrial potential progressively fell and membrane permeability increased. Virtually all drugs and chemicals cause toxicity at high enough concentrations. Thus, it is critical to assess toxicity at concentrations relevant to those that are used for drug efficacy. Efficacious concentration, as defined as the maximal serum concentration of drug used for treatment (Cmax ), is highly variable, ranging 10 000 000-fold in one study of 187 marketed human drugs from 100 pM to 2 mM, with 90 % values less than 100 μM, 60 % less than 10 μM, 37 % less than 1 μM and 12 % less than 100 nM (Figure 1.8). Most human hepatotoxic drugs (94 % of 102 tested) are cytotoxic in the sublethal HSC cytotoxicity assay at concentrations less than 100-fold Cmax , whereas most non-toxic drugs (96 % of 23 tested) are cytotoxic in this assay at concentrations more than 100-fold Cmax (O’Brien et al., 2006) (Figure 1.9). Hormesis, in which compensatory adaptive changes precede and occur at lower doses than degenerative changes, was detected for half of the toxic drugs for cell proliferation, cell morphology and mitochondria (O’Brien et al., 2006). It is demonstrated for cerivastatin in Figure 1.6 for mitochondrial membrane potential and for fenofibrate in Figure 1.8 for mitochondrial mass and nuclear area. Hormesis could not be assessed for parameters that normally have low values, such as intracellular calcium measured by fluo4 or membrane permeability measured by toto-3, because assay methods were not sufficiently sensitive. However, for calcium, more sensitive dyes, with calcium dissociation constants closer to the physiologic concentration of ionized calcium, have detected biphasic effects on resting calcium (O’Brien et al., 1990). In the sublethal HCS assay, sensitivity and specificity for identification of human hepatotoxicant drugs were 94 % and 96 %, respectively, when testing only hepatotoxicant drugs

1 2 3 4 5 6 7 8 9 10 11 12 13

Acetaminophen Chloramphenicol r Danazol i,r Diclofenac i,r Felbamate r Flutamide i,r Hydralazine r Ibuprofen r Indomethacin r Imipramine r Isoniazide i,r Leflunomide i,r Methyldopa

r

i,r

Druga

Cell Numberc 500↑4000↓ 260 13 16↑126↓ 315 50 67 50 190 0.8↑50↓ — 25 330

Conventional Testsb

− − − − − − ND − − − − + ND 4000 — 50↑100↓ 250 160↑ 50 135↑ — 47 100 50↓ — 330

Mitochondrial potentialc — — 25 — — 50 270 — — 50 — — 330

Ca — — 50 126 — 50 — — — 50 — — 330

Membrane Pote ntial 7800 — 13 126↑ 315 — 75 — 2↑ 50 — — 330

Nuclear area

130 57 0.16 4.2 42 6 5 250 6 0.6 40 340 11

Cmax (uM)

4 5 81 4 4 8 13.5 0.2 0.3 1 1 0.07 30

TI

8 53 — 99 24 90 87 99 90 100 0 99.3 15

PPB (%)

OS OS — OP, Ap IM M IM IM, M IM, OS IM, M IM, OS IM —

Mechanismd

Table 1.7 Concordance of HCS assay results with idiosyncratic or drug metabolite-implicated human hepatotoxicity (based on O’Brien et al., 2006). Drugs are compared with how they tested in seven conventional cytotoxicity assays (column 2) and how they tested in the HCS assay (columns 3–7). Results for the HCS assay are tabulated for lowest observed adverse effect level (LOAEL) concentrations (in μM) causing effects on cell number, mitochondrial membrane potential, intracellular ionized calcium concentration, membrane permeability and nuclear area, respectively. The maximal plasma total concentrations of drugs associated with efficacy (C max ) are indicated in column 8. The ratio of the lowest cytotoxic concentrations from columns 3–7 to Cmax from column 8 is indicated in column 9 (TI). The percentage of drug that is bound to plasma proteins is also indicated (column 10) and where known the cellular mechanism of cytotoxicity (column 11). The sensitivities of the conventional assays and the HCS assay are indicated in the bottom row. The sensitivity of the HCS assay was determined using a TI cutoff of 100 in column 9. Tacrine is the only drug in this panel that falls just outside of the 100 cutoff. It is of interest to note that tacrine only causes asymptomatic elevations of liver enzymes that do not progress to frank liver injury despite continued use of the drug

Minocycline Nitrofurantoin r Piroxicam r Procainamide i Rifampicin i Sulindac r Sulfamethoxazole r Tacrine r Tamoxifen i,r Terbinafine i,r Valproate

i

i

15 %

− − − ND + − − − + ND −

13 12 5 630 50 285 1100↑ 25 100↑ 3 1000

25 50 160 630↑ — 140 — 50↑100↓ 25 6 4000↑

c

b

Superscript ‘i’, idiosyncratic heptatotoxicity; Superscript ‘r’, reactive metabolite. ND, not determined. ↑signal increased;↓ signal decreased. d OS, oxidative stress; Ap, apoptosis; IM, imune-mediated; M, mitochondrial.

a

Sensitivity

14 15 16 17 18 19 20 21 22 23 24

— 50 — 630 100 — — 50 — 6 —

— 50 — — 100 — — 25 25 6 4000 96 %

50↑ 0.7↑ 5↑ 320 13↑ 570↑ — 13↑50↓ 0.4↑25↓ 13 8000

8 6 5 12 9 19 217 0.1 0.4 4 540

2 0.1 1 27 1 8 5 130 1 0.8 2

76 62 99 18 81 94 62 55 98 99 93

IM, M IM, OS, M — IM — IM, OS IM, M OS, M M — OS, M

36

Hepatotoxicity Nuclear area

Nuclear area

30

30

25

25

20

20

15

15

10

10

lonized calcium (Fluo 4)

Mitochondrial DNA (Picogreen)

200

3500

150

3000 2500

Cytotoxicity parameter value

100

2000

50

1500

0

1000

TMRM

Dihydroethidium oxidation rate 0.18

500 400

0.12

300 0.06

200 100

0.00

0

Plasma membrane integrity (Toto-3)

MitoTracker Deep Red 1200

500 400 300

800

200 100 400

0

Cell count (10× objective, 10 fields)

Cell count (20× objective, 6 fields)

5000

300

4000

250 200

3000

150

2000

2000

500

1000

250

62.5

125.0

31.3

7.8

15.6

3.9

0

2000

500

1000

250

125

62.50

31.25

7.81

15.63

3.91

0

0 1.95

50

0

2.0

100

1000

Fenofibrate concentration (µM)

Figure 1.8 Use of high-content screening for quantitative determination of dose–response relationships for drug-induced, human hepatocellular toxicity (based on O’Brien et al., 2006). Fenofibrate (25 μM) was assayed in HepG2 cells at doubling concentrations from 2 to 2000 using two different combinations (columns 1 and 2) of fluorescent probes. For both assays, Hoechst 33342 was used to determine nuclear area and cell count. Ionized calcium was assayed using Fluo4, mitochondrial membrane potential using TMRM, plasma membrane integrity using TOTO-3, mitochondrial DNA using picogreen, mitochondrial mass using MitoTracker Deep Red and reactive oxygen species using dihydroethidium. The circles in each dose–response curve indicate the lowest observed adverse effect level (LOAEL) values which were used to tabulate the LOAEL concentrations shown in Table 1.7

37

Cumulative frequency


Human hepatoxic drugs

Non-toxic drugs

10 000

Cumulative frequency

In vivo efficacious concentration (C max)

Safety margin (In vitro cytotoxicity concentration/in vivo C max)

Figure 1.9 Safety margins for HCS sublethal cytotoxicity assay concordant with human hepatotoxicity. A frequency histogram (a) for the in vivo efficacious concentration in humans (Cmax ) of 187 marketed drugs demonstrates a wide range and indicates the need to assess cytotoxicity in the context of efficacious concentration. A frequency histogram (b) of the safety margin (in vitro HCS cytotoxicity concentration/Cmax ) for 102 human hepatotoxicant drugs and the 23 drugs for which this ratio was determined indicates that a safety margin of 100 is the most effective discriminator of human hepatotoxicity potential (based on O’Brien et al., 2006)

and non-toxic drugs (O’Brien et al., 2006). However, when other drugs that produce other organ toxicities (e.g. kidney, heart, bone marrow, muscle and pancreas) were tested their cytotoxic effects were not distinguishable from those of drugs causing hepatotoxicity. Thus, cytotoxicity in the HCS assay was concordant with human toxicity but not specific for liver toxicity. An additional caveat in the use of the assay is that it did not detect cholestatic effects of drugs. Nor did it detect other organ toxicities produced by drugs’ effects on proteins not found in hepatocytes. 1.3.2

Primary Hepatocytes and Liver-Specific Toxicities

While hepatic cell lines have several advantages over primary cells, it is still prudent to take a selected subset of compounds into primary hepatocytes. The reasons are mainly the following. primary hepatocytes are more differentiated cells compared to cell lines. They maintain higher levels of metabolic activity, normal p53 status and cell cycle regulation and more normal levels of transporter protein expression (Le Vee et al., 2006; Wilkening et al., 2003). For example, primary hepatocytes are more suitable than HepG2 to study metabolismmediated liver toxicants, such as benzo[a]pyrene, dimethylnitrosamine (DMN),

38

Hepatotoxicity

2-amino-1-methyl-6-phenylimidazo[4,5-b]pyridine (PhIP) (Wilkening et al., 2003), acetaminophen, diclofenac, cyclophosphamide, disulfiram (Bort et al., 1999b; Wang et al., 2002) and nimesulide (Rainsford et al., 2001). High-content analysis of primary hepatocytes has also been used to study more liver-specific toxicities, such as steatosis, steatohepatitis, cholestasis and reactive metabolites. Steatosis, or accumulation of fatty acids (i.e. ‘fatty liver’), can be caused by a variety of drugs including amiodarone, perhexiline maleate, 4,4 -diethylaminoethoxyhexestrol (DEAH), tetracycline, valproic acid and several antiviral nucleoside analogues, such as fialuridine (Fromenty and Pessayre, 1997). One of the major mechanisms involved in steatosis is the inhibition of beta-oxidation of long-chain fatty acids, either by direct inhibition or indirect inhibition, such as CoA sequestration or mitochondria DNA damage (Fromenty and Pessayre, 1995; Jaeschke et al., 2002b). The resulting fatty acid accumulation can be detected and quantified by staining primary hepatocytes with neutral lipid stains such as Oil red O and performing HCA under the microscope (Amacher and Martin, 1997; McMillian et al., 2001). This is different from using phospholipids stains such as NBD-PC or NBD-PE to study phospholipids accumulation in the cytoplasm (i.e. phospholipidosis) (Gum et al., 2001). In addition, the aldehyde products of lipid peroxidation, 4-hydroxynonenal and malondialdehyde (MDA), can be measured by biochemical readouts (Berson et al., 1998a,b). MDA is a known stimulator of inflammatory responses. Therefore, repeated challenge by steatotic drugs and the resulting lipid peroxidation represent an important mechanism of drug-induced steatohepatitis (Ivanov et al., 1992). While simple steatosis was regarded as benign, research in the past decade suggested that mitochondria dysfunction and reactive oxygen species generation are important factors to differentiate simple steatosis from nonalcoholic steatohepatitis (NASH) (Fromenty et al., 2004). Using primary cultured hepatocytes, it was found that amiodarone, perhexiline maleate and DEAH increased the mitochondrial formation of reactive oxygen species and caused lipid peroxidation (Berson et al., 1998b). Based on these findings, it would be important for high content assays to measure mitochondria function and oxidative stress in addition to simple neutral lipid accumulation. Intrahepatic cholestasis, defined as impairment in bile formation and/or bile flow, is another common manifestation of drug-induced liver disease. In humans, intrahepatic cholestasis most often occurs in the elderly (Feuer and Di Fonzo, 1992). As the population ages and the occurrence of multiple drug therapy in geriatric patients increases, it is expected that jaundice and/or drug-induced intrahepatic cholestasis will become even more prevalent. Bile formation is dependant on the specific transporter proteins in hepatocytes. The expression and the appropriate membrane sorting of these transporters are highly dependent upon the differentiated phenotype of the cells (Hoffmaster et al., 2004; LeCluyse et al., 1994). Sandwich-cultured primary hepatocytes are currently the most well-characterized in vitro model to study the disposition of taurocholate (Liu et al., 1999), bilirubin and its glucuronide (Lengyel et al., 2005) and drug-induced cholestasis (Kostrubsky et al., 2003). As expected, inhibition of important hepatobiliary transporters, such as the bile salt export pump (BSEP), can result in drug-induced cholestasis (Kostrubsky et al., 2006). Although in the cases of rifampicin (Li and Chiang, 2006), it was found to induce bile acid and bilirubin detoxification (Marschall et al., 2005) via PXR activation and this drug has been successfully used to treat cholestatic liver disease. This example highlights the importance of using longer-term primary hepatocyte cultures (> 3 days) that express differentiated and


39

polarized liver functions as opposed to the short-term BSEP inhibition experiments (< 1 h), to provide a more complete picture of in vivo outcome. It is well known that many hepatotoxic agents can be metabolized to reactive metabolites that can either be detoxified or react with glutathione, enzymes, nucleic acids, lipids or proteins (Knowles et al., 2000; Pessayre, 1995). These reactive intermediates are electrophilic metabolites or free radicals that are generated during the metabolism of a broad range of functional groups. Reactive metabolite formation and/or oxidative stress induced by drugs are also considered important factors in immune-mediated idiosyncratic drug hypersensitivity (Park et al., 2001). There are several rapid in vitro methods to detect and measure the generation of such reactive intermediates. For example, high-throughput assays for identifying pharmaceutical compounds that produce reactive metabolites have been developed. These methods involve incubating drug candidates with a liver microsomal drug metabolizing enzyme system in the presence of glutathione and detecting glutathione conjugates via tandem mass spectrometry (Chen et al., 2001; Haroldsen et al., 1988; Pearson et al., 1988). However, it is now recognized that reactive metabolite formation per se is not sufficient to cause tissue injury (Park et al., 2005; Sanderson et al., 2006). Hence, cell-based models to evaluate the toxicological consequences of the formation of reactive intermediates (or lack thereof) are becoming increasingly important. In cells, the reduced form of glutathione (GSH) is known to interact with electrophilic compounds/metabolites and free radicals to play a key role in the detoxification of such reactive molecules. Depletion of the reduced form of glutathione was reported to be a marker of hepatotoxicity (e.g. Fouin-Fortunet et al., 1984), suggesting its evaluation is important in toxicological studies. For example, monochlorobimane and chloromethylfluorescein diacetate (CMFDA) have been successfully used to monitor cellular GSH levels by epifluorescence in isolated hepatocytes (Lilius et al., 1996; Thompson et al., 1998). The fluorometric dye, 2 ,7 -dichlorodihydrofluorescein (H2 DCF) can be used to detect oxidative stress induced by various oxidative radicals in human cells including hepatocytes (Lautraite et al., 2003; LeBel et al., 1992; Wang and Joseph, 1999). The diacetate form of H2 DCF freely enters the cell and is hydrolyzed by intracellular esterases to liberate H2 DCF, which then reacts with oxidizing species to produce the highly fluorescent DCF. These cell-based assays are amenable to high-throughput evaluation of chemical compounds and their reactive metabolites and/or oxidative species in drug discovery (Lautraite et al., 2003). 1.3.3

RNAi Technology

Cell-based systems are currently being coupled with new technologies such as RNA interference (RNAi) as a potential tool to identify genes with predictive functions significant to drug development. RNAi represents an evolutionary conserved mechanism developed by nature to protect the genome against endogenous and exogenous stress insults, and regulate gene expression. There are numerous excellent reviews on the use of RNAi as a research tool and as therapeutic agents (Cejka et al., 2006; Cullen, 2006; Dillon et al., 2005; Dykxhoorn and Lieberman, 2005; Dykxhoorn et al., 2006; Hannon and Rossi, 2004; Lee and Sinko, 2006; Shankar et al., 2005; Sledz and Williams, 2005). Briefly, endogenous mammalian RNAi is mediated by small-interfering RNAs (siRNAs) produced from long double-stranded hair pin RNA known as micro RNA (miRNA) through the action of an endonuclease of the ribonuclease-III type, called ‘drosha’ and ‘dicer’. These siRNAs

40

Hepatotoxicity

or small double-stranded RNAs are incorporated into the RNA-inducing silencing complex (RISC), which contains RNAase activity, to become small single-stranded RNAs. The single-stranded RNA guides RISC to mRNA that has a complementary sequence and mediates gene silencing by targeting RNA. This silencing is caused by either translational repression if there is a mismatch in the sequence of the siRNA and target mRNA or by RNA cleavage if there is an exact match between the siRNA sequence and the target mRNA. Thus, the synthesis of the protein encoded by the mRNA targeted by the siRNAs is prevented, and that protein is selectively depleted from the cell. The enzymatic machinery required to process siRNA is ubiquitously expressed in most eukaryotic cells and can be co-opted by exogenous RNAs to direct the sequence-specific gene silencing. Therefore, siRNA can be chemically synthesized and transfected into cells, tissues and animals after being integrated into specialized delivery systems, such as liposomes. Alternatively, plasmids or viral vectors can be used to express shRNA (short hairpin RNA that structurally mimics a miRNA precursor). Successful ‘knockdowns’ of specific genes allow the investigation of potential drug targets, as well as the genetic basis of physiological and disease processes in mammalian systems. RNA interference (RNAi) technology is widely used in vitro for modulating specific targets and assessing biologic pathways; in addition, this novel technology has the potential to be used for addressing drug-induced liver injury (DILI). Toxicological effects observed upon treatment with a given drug in the presence or absence of a specific RNAi molecule can be evaluated in cell-based systems to determine whether the toxicity associated with chemical entities is mediated by interaction with that specific target. For example, Hep2 cells stably expressing specific small interfering RNA directed against the activating signal cointegrator-2 (ASC-2) have been successfully used to demonstrate that ASC-2 protein is likely to participate in acetaminophen-mediated hepatotoxicity as a transcriptional coactivator of the xenobiotic nuclear receptor CAR in vivo (Choi et al., 2005); mouse hepatocytes silenced against Fas expression were found to be protected from cytotoxicity caused by actinomycin D (Song et al., 2003); Moreover, RNAi technology will further demonstrate participation of the specific cytochrome P450 (CYP450) enzyme in the DILI process. This was illustrated when CHL-3A4 cells with 3A4 siRNAs significantly diminished the cytotoxicity of cyclophosphamide and ifosfamide (Chen et al., 2006). RNAi technology has also been utilized to understand the role of transport proteins such as Mrp1-3, Mdr1 and Bcrp (Lee and Sinko, 2006; Sahi, 2005; Tian et al., 2004; Xu et al., 2005), and the functional analysis of nuclear receptors such as the Farnesoid X receptor and androgen receptor (Plass et al., 2002; Wright et al., 2003) and their co-regulators, such as SRCs, p300, NcoR and SMRT (Debes et al., 2002; Yoon et al., 2003; Zhou et al., 2003) in DILI. At present, the prediction of metabolism-dependent hepatotoxicity is difficult or even impossible because there are no suitable experimental (in vivo/in vitro) models and we do not understand the basic mechanism involved in the toxicity when it does occur in man. Cell-based models coupled with RNAi technology could be useful systems for predicting metabolic activity of drugs in vivo. In vivo delivery is still the greatest challenge to the effective use of this technology in target characterization and investigating adverse pharmacology and chemically mediated toxicity. However, recent advances in in vivo delivery and vector technology make it probable that RNAi can be effectively used for in vivo hypothesis-driven research (Aligner, 2006; Behlke, 2006; Lu et al., 2006).


1.4

41

Conclusions

We have reviewed various in vitro models currently used or under development to better enable the evaluation of drug-induced liver injury (DILI). Continuing advances in stem cells research, new immortalized cell lines and genetically engineered cells, 3D-hepatocyte bioreactors and our enhanced ability to predict outcomes from primary cells have allowed a better understanding of the mechanisms of toxicity, metabolism of drugs and the species differences in expression of toxicity. Each model described in this review has significant advantages and it is possible to find an appropriate system for any particular toxicological question asked. Further research should be directed towards the refinement of existing methodologies and the development of new alternatives and testing paradigms for human relevancy. This approach coupled with the establishment of new technologies (e.g. HCS, RNAi and ‘omics’) will improve our ability to assess DILI earlier in the drug discovery process and would allow us to identify compounds with decreased risk of hepatotoxicity at later stages of development for eventual human health risk assessment. To fully establish the credibility and relevance of in vitro toxicity evaluation of drugs, it is essential that in vitro/in vivo correlations of toxicity of NCEs or drugs be determined by toxicologists and cell culture scientists.

References Aden D. P., Fogel A., Plotkin S., Damjanov I. and Knowles B. B. (1979). Controlled synthesis of HBsAg in a differentiated human liver carcinoma-derived cell line. Nature 282: 615–616. Aithal G. P., Ramsay L., Daly A. K., Sonchit N., Leathart J. B. S., Alexander G., Kenna J. G., Caldwell J. and Day C. P. (2004). Hepatic adducts, circulating antibodies and cytokine polymorphisms in patients with diclofenac hepatotoxicity. Hepatology 39: 1430–1440. Aligner A. (2006). Gene silencing through RNA interference (RNAi) in vivo: strategies based on the direct application of siRNAs. J Biotech 124: 12–25. Alison M. R., Poulsom R., Jeffery R., Dhillon A. P., Quaglia A., Jacob J., Novelli M., Prentice G., Williamson J. and Wright N. A. (2000). Hepatocytes from non-hepatic adult stem cells. Nature 406: 257. Allen J. W. and Bhatia S. N. (2003). Formation of steady-state oxygen gradients in vitro: application to liver zonation. Biotechnol Bioeng 82: 253–262. Allen J. W., Khetani S. R. and Bhatia S. N. (2005). In vitro zonation and toxicity in a hepatocyte bioreactor. Toxicol Sci 84: 110–119. Amacher D. E. and Martin B.-A. (1997). Tetracycline-Induced Steatosis in Primary Canine Hepatocyte Cultures. Fund Appl Toxicol 40: 256–263. Amin A. and Hanza A. A. (2005). Oxidative stress mediates drug-induced hepatotoxicity in rats: a possible role of DNA fragmentation. Toxicology 208: 367–375. Andrade R. J., Lucena M. I., Fernandez M. C., Pelaez G., Pachkoria K., Garcia-Ruiz E., GarciaMunoz B., Gonzalez-Grande R., Pizarro A., Duran J. A., Jimenez M., Rodrigo L., Romero-Gomez M., Navarro J. M., Planas R., Costa J., Borras A., Soler A., Salmeron J. and Martin-Vivaldi R. (2005). Drug-induced liver injury: an analysis of 461 incidences submitted to the Spanish registry over a 10-year period. Gastroenterology 129: 512–521. Aninat C., Piton A., Glaise D., Le Charpentier T., Langouet S., Morel F., Guguen-Guillouzo C. and Guillouzo A. (2006). Expression of cytochromes P450, conjugating enzymes and nuclear receptors in human hepatoma HepaRG cells. Drug Metab Dispos 34: 75–83.

42

Hepatotoxicity

Asahina K., Fujimori H., Shimizu-Saito K., Kumashiro Y., Okamura K., Tanaka Y., Teramoto K., Arii S. and Teraoka H. (2004). Expression of the liver-specific gene Cyp7a1 reveals hepatic differentiation in embryoid bodies derived from mouse embryonic stem cells. Genes Cells 9: 1297–1308. Aurich I., Mueller L. P., Aurich H., Luetzkendorf J., Tisljar K., Dollinger M., Schormann W., Walldorf J., Hengstler J., Fleig W. E. and Christ B. (2006). Functional integration of human mesenchymal stem cell-derived hepatocytes into mouse livers. Gut 56: 405–415. Avital I., Feraresso C., Aoki T., Hui T., Rozga J., Demetriou A. and Muraca M. (2002). Bone marrowderived liver stem cell and mature hepatocyte engraftment in livers undergoing rejection. Surgery 132: 384–390. Avital I., Inderbitzin D., Aoki T., Tyan D. B., Cohen A. H., Ferraresso C., Rozga J., Arnaout W. S. and Demetriou A. A. (2001). Isolation, characterization, and transplantation of bone marrow-derived hepatocyte stem cells. Biochem Biophys Res Commun 288: 156–164. Bai J. and Cederbaum A. I. (2004). Adenovirus mediated overexpression of CYP2E1 increases sensitivity of HepG2 cells to acetaminophen induced cytotoxicity. Mol Cell Biochem 262: 165– 176. Beerheide W., von Mach M. A., Ringel M., Fleckenstein C., Schumann S., Renzing N., Hildebrandt A., Brenner W., Jensen O., Gebhard S., Reifenberg K., Bender J., Oesch F. and Hengstler J. G. (2002). Downregulation of beta2-microglobulin in human cord blood somatic stem cells after transplantation into livers of SCID-mice: an escape mechanism of stem cells? Biochem Biophys Res Commun 294: 1052–1063. Begriche K., Igoudjil A., Pessayre D. and Fromenty B. (2006). Mitochondrial dysfunction in NASH: causes, consequences and possible means to prevent it. Mitochondrion 6: 1–28. Begue J. M., Guguen-Guillouzo C., Pasdeloup N. and Guillouzo A. (1984). Prolonged maintenance of active CYP450 in adult rat hepatocytes co-cultured with another liver cell type. Hepatology 4: 839–842. Behlke M. A. (2006). Progress towards in vivo use of siRNAs. Mol Ther 13: 644–670. Berson A., Beco V. D., Letteron P., Robin M. A., Moreau C., Kahwaji J. E., Verthier N., Feldmann G., Fromenty B. and Pessayre D. (1998a). Steatohepatitis-Inducing Drugs Cause Mitochondrial Dysfuntion and Lipid Peroxidation in Rat Hepatocytes. Gastroenterology 114: 764–774. Berson A., De Beco V., Letteron P., Robin M. A., Moreau C., El Kahwaji J., Verthier N., Feldmann G., Fromenty B. and Pessayre D. (1998b). Steatohepatitis-inducing drugs cause mitochondrial dysfunction and lipid peroxidation in rat hepatocytes. Gastroenterology 114: 764–774. Berthiaume F., Moghe P. V., Toner M. and Yarmush M. L. (1996). Effect of extracellular matrix topology on cell structure, function and physiological responsiveness: hepatocytes cultured in a sandwich configuration. FASEB J 10: 1471–1484. Biagini C., Bender V., Borde F., Boissel E., Bonnet M. C., Masson M. T., Cassio D. and Chevalier S. (2006a). Cytochrome P450 expression-induction profile and chemically mediated alterations of the WIF-B9 cell line. Biol Cell 98: 23–32. Biagini C. P., Boissel E., Borde F., Bender V. E., Bouskila M., Blazy F., Nicaise L., Mignot A., Cassio D. and Chevalier S. (2006b). Investigation of the hepatotoxicity profile of chemical entities using Liverbeads (R) and WIF-B9 in vitro models. Toxicol in Vitro 20: 1051–1059. Boelsterli U. A. (2003). Disease-related determinants of susceptibility to drug-induced idiosyncratic hepatotoxicity. Curr Opin Drug Discov Devel 6: 81–91. Bort R., Mace K., Boobis A., Gomez-Lechon M. J., Pfeifer A. and Castell J. (1999a). Hepatic metabolism of diclofenac: role of human CYP in the minor oxidative pathways. Biochem Pharmacol 58: 787–796. Bort R., Ponsoda X., Jover R., Gomez-Lechon M. J. and Castell J. V. (1999b). Diclofenac toxicity to hepatocytes: a role for drug metabolism in cell toxicity. J Pharmacol Exp Ther 288: 65–72. Bova M. P., Tam D., McMahon G. and Mattson M. N. (2005). Troglitazone induces a rapid drop of mitochondrial membrane potential in liver HepG2 cells. Toxicol Lett 155: 41–50.


43

Bowen W. P., Carey J., Miah A., McMurray H., Munday P., James R., Coleman R. and Brown A. M. (2000). Measurement of cytochrome P450 gene induction in human hepatocytes using quantitative real-time reverse transcriptase-polimerase chain reaction. Drug Metabol Dispos 28: 781–788. Bravo P., Bender V. and Cassio D. (1998). Efficient in vitro vectorial transport of a fluorescent conjugated bile acid analogue by polarized hepatic hybrid WIF-B and WIF-B9 cells. Hepatology 27: 576–583. Bull S., Langezaal I., Clothier R. and Coecke S. (2001). A Genetically engineered cell-based system for detecting metabolism-mediated toxicity. Altern Lab Anim 29: 703–716. Burchell B. and Coughtrie W. H. (1989). UDP-Glucuronosyltransferases. Pharmacol Ther 43: 261– 289. Byrne J. A., Strautnieks S. S., Mieli-Vergani G., Higgins C. F., Linton K. J. and Thompson R. J. (2002). The human bile salt export pump: characterization of substrate specificity and identification of inhibitors. Gastroenterology 123: 1649–1658. Cantz T., Sharma A. D., Jochheim-Richter A., Arseniev L., Klein C., Manns M. P. and Ott M. (2004). Reevaluation of bone marrow-derived cells as a source for hepatocyte regeneration. Cell Transplant 13: 659–666. Carey M. A. and van Pelt F. N. (2005). Immunochemical detection of flucloxacillin adduct formation in livers of treated rats. Toxicology 216: 41–48. Castell J. V., Jover R., Martinez-Jimenez C. P. and Gomez-Lechon M. J. (2006). Hepatocyte cell lines: their use, scope and limitations in drug metabolism studies. Expert Opin Drug Metabol Toxicol 2: 183–212. Cejka D., Losert D. and Wacheck V. (2006). Short interfering RNA (siRNA): tool or therapeutic? Clin Sci 110: 47–58. Chen W. G., Zhang C., Avery M. J. and Fouda H. G. (2001). Reactive metabolite screen for reducing candidate attrition in drug discovery. Adv Exp Med Biol 500: 521–524. Chen J., Yang X. X., Min H., Hu Z. P., He M., Duan W., Chan E., Cheu F. S., Chen X. and Zhou S. F. (2006). Small interfering RNA-mediated silencing of cytochrome P450 3A4 gene. Drug Metabol Dispos 34, 1650–1657. Choi E., Lee S., Yeom S., Kim G. H., Lee J. W. and Kim S.-W. (2005). Characterization of activating signal cointegrator-2 as a novel transcriptional coactivator of the xenobiotic nuclear receptor constitutive androstane receptor. Mol Endocrinol 19: 1711–1719. Crocenzi F. A., Mottino A. D., Sanchez Pozzi E. J., Pellegrino J. M., Rodriguez Garay E. A., Milkiewicz P., Vore M., Coleman R. and Roma M. G. (2003a). Impaired localisation and transport function of canalicular Bsep in taurolithocholate induced cholestasis in the rat. Gut 52: 1170–1177. Crocenzi F. A., Mottino A. D., Cao J., Veggi L. M., Pozzi E. J., Vore M., Coleman R. and Roma M. G. (2003b). Estradiol-17beta-d-glucuronide induces endocytic internalization of Bsep in rats. Am J Physiol Gastrointest Liver Physiol 285: G449–G459. Cullen B. R. (2006). Induction of stable TNA interference in mammalian cells. Gene Ther 13: 503– 508. Dan Y. Y., Riehle K. J., Lazaro C., Teoh N., Haque J., Campbell J. S. and Fausto N. (2006). Isolation of multipotent progenitor cells from human fetal liver capable of differentiating into liver and mesenchymal lineages. Proc Natl Acad Sci USA 103: 9912–9917. Davila J. C. and Morris D. L. (1999). Analysis of cytochrome P450 and Phase II conjugating enzyme expression in adult male rat hepatocytes. In vitro Cell Dev Biol 35: 120–130. Davila J. C., Rodrigues R. J., Melchert R. B. and Acosta D. (1998). Predicitive value of in vitro model systems in toxicology. Ann. Rev. Pharmacol. Toxicol 38: 63–96. Davila J. C., Cezar G. G., Thiede M., Strom S., Miki T. and Trosko J. (2004). Use and application of stem cells in toxicology. Toxicol Sci 79: 214–223. Debes J. D., Schmidt L. J., Huang H. and Tindall D. J. (2002). p300 mediates androgen-independent transactivation of the androgen receptor by interleukin 6. Cancer Res 62: 5632–5636.

44

Hepatotoxicity

Decaens C., Rodriguez P., Bouchaud C. and Cassio D. (1996). Establishment of hepatic cell polarity in the rat hepatoma-human fibroblast hybrid WIF-B9. A biphasic phenomenon going from a simple epithelial polarized phenotype to an hepatic polarized one. J Cell Sci 109: 1623–1635. Diaz D. and O’Brien P. J. (2006). Defining the sequence of events in cerivastatin toxicity using a high content multi-parametric cytotoxicity assay. Eur Pharmaceut Rev 11: 38–45. Dillon C. P., Sandy P., Nencioni A., Kissler S., Rubinson D. A. and Van Parijs L. (2005). RNAi as an experimental and therapeutic tool to study and regulate physiological and disease processes. Annu Rev Physiol 67: 147–173. Donato M. T., Castell J., V. and Gomez-Lechon J. M. (1994). Cytochrome P450 activities in pure and co-cultured rat hepatocytes. Effects of model inducers. In vitro Cell Dev Biol 7: 237–245. Donato M. T., Castell J. V. and Gomez-Lechon J. M. (1995). Effect of model inducers on cytochrome P450 activities of human hepatocytes in primary cultures. Drug Metabol Dispos 23: 553–558. Dunn J. C. Y., Tompkins R. G. and Yarmush M. L. (1991). Long term in vitro function of adult hepatocytes in a collagen sandwich configuration. Biotechnol Prog 7: 237–245. Dykxhoorn D. M. and Lieberman J. (2005). The silent revolution: RNA interference as basic biology, research tool and therapeutic. Annu Rev Med 56: 401–423. Dykxhoorn D. M., Palliser D. and Lieberman J. (2006). The silent treatment: siRNAs as small molecule drugs. Gene Ther 13: 541–552. Eaton D. L. and Bammler T. K. (1999). Consise review of the glutathione S-transferases and their significance to toxicity. Toxicol Sci 49: 156–164. Elkins S. and Wrighton S. A. (1999). The role of CYP2B6 in human xenobiotic metabolism. Drug Metabol Rev 31: 719–754. Enat R., Jefferson D. M. and Ruiz-Opazo N. (1984). Hepatocyte proliferation in vitro; its dependence on the use of serum-free hormonally defined medium and substrata of extrascellular matrix. Proc Natl Acad Sci USA 81: 1411–1415. Farkas D. and Tannembaum S. R. (2005). Characterization of chemically induced hepatotoxicity in collagen sandwiches of rat hepatocytes. Toxicol Sci 85: 927–934. Fattinger K., Funk C., Pantze M., Weber C., Reichen J., Stieger B. and Meier P. J. (2001). The endothelin antagonist bosentan inhibits the canalicular bile salt export pump: a potential mechanism for hepatic adverse reactions. Clin Pharmacol Ther 69: 223–231. FDA (1999). In vivo Drug Metabolism/Drug Interaction Studies – Study Design, Data Analysis and Recommendations for Dosing and Labeling. [http://www.fda.gov/cder/guidance/2635fnl.pdf]. Fernandez C., Buyse M., German-Fattal M. and Gimenez F. (2004). Influence of the pro-inflammatory cytokines on P-glycoprotein expression and functionality. J Pharm Pharm Sci 7: 359–371. Feuer G. and Di Fonzo C. J. (1992). Intrahepatic Cholestasis: A Review of Biochemical–Pathological Mechanisms. Drug Metab Drug Interact 10: 1–161. Fidkowski C., Kaazempur-Mofrad M. R., Borenstein J., Vacanti J. P., Langer R. and Wang Y. (2005). Endothelialized microvasculature based on a biodegradable elastomer. Tissue Eng 11: 302–309. Fiegel H. C., Lioznov M. V., Cortes-Dericks L., Lange C., Kluth D., Fehse B. and Zander A. R. (2003). Liver-specific gene expression in cultured human hematopoietic stem cells. Stem Cells 21: 98–104. Forbes S., Vig P., Poulsom R., Thomas H. and Alison M. (2002). Hepatic stem cells. J Pathol 197: 510–518. Fouin-Fortunet H., Letteron P., Tinel M., Degott C., Fish C. and Pessayre D. (1984). Mechanism for isaxonine hepatitis. II. Protective role of glutathione and toxicological studies in mice. J Pharmacol Exp Ther 229: 851–858. Fromenty B. and Pessayre D. (1995). Inhibition of mitochondrial beta-oxidation as a mechanism of hepatotoxicity. Pharmacol Therapeut 67: 101–154. Fromenty B. and Pessayre D. (1997). Impaired mitochondrial function in microvesicular steatosis. Effects of drugs, ethanol, hormones and cytokines. J Hepatol 26(Suppl 2): 43–53.


45

Fromenty B., Robin M. A., Igoudjil A., Mansouri A. and Pessayre D. (2004). The ins and outs of mitochondrial dysfunction in NASH. Diabetes Metab 30: 121–138. Funk C., Ponelle C., Scheuermann G. and Pantze M. (2001). Cholestatic potential of troglitazone as a possible factor contributing to troglitazone-induced hepatotoxicity: in vivo and in vitro interaction at the canalicular bile salt export pump (BSEP) in the rat. Mol Pharmacol 59: 627– 635. Gebhardt R., Hengstler J. G., Muller D., Glockner R., Buenning P., Laube B., Schmelz. E, Ullrich M., Utesch D., Hewitt N., Ringel M., Hilz B. R., Bader A., Langsch A., Koose T., Burger H., Mass J. and Oesch F. (2003). New hepatocyte in vitro systems for drug metabolism: metabolic capacity and recommendations for application in basic research and drug development, standard operation procedures. Drug Metabol Rev 35: 145–213. Gerlach J. C., Encke J., Hole O., Muller C., Ryan C. J. and Neuhaus P. (1994). Bioreactor for a larger scale hepatocyte in vitro perfusion. Transplantation 58: 984–988. Gerlach J. C., Schnoy N., Encke J., Smith M. D., Muller C. and Neuhaus P. (1995). Improved hepatocyte in vitro maintenance in a culture model with woven multicompartment capillary systems: electron microscopy studies. Hepatology 22: 546–552. Gerlach J. C., Mutig K., Sauer I. M., Schrade P., Efimova E., Mieder T., Naumann G., Grunwald A., Pless G., Mas A., Bachmann S., Neuhaus P. and Zeilinger K. (2003). Use of primary human liver cells originating from discarded grafts in a bioreactor for liver support therapy and the prospects of culturing adult liver stem cells in bioreactors: a morphologic study. Transplantation 76: 781– 786. Giguere V. (1999). Orphan nuclear receptors: from gene to funciton. Endocrine Rev 20: 689–725. Gomez-Lechon M. J., Donato M. T., Castell J. V. and Jover R. (2004). Human hepatocytes in primary culture: the choice to investigate drug metabolism in man. Curr Drug Metabol 5: 433–462. Gonzalez F. (1996). The CYP2D6 Subfamily, Ioannides C, New York, NY, USA. Gonzalez F. J. and Lee Y. H. (1996). Constitutive expression of hepatic cytochrome P450 genes. FASEB J 10: 51–57. Gradilone S. A., Tietz P. S., Splinter P. L., Marinelli R. A. and LaRusso N. F. (2005). Expression and subcellular localization of aquaporin water channels in the polarized hepatocyte cell line, WIF-B. BMC Physiol 5: 13. Guengerich F. (1997). Comparisons of catalytic selectivity of cytochrome P450 subfamily enzymes from different species. Chem Biol Interact 106: 161–182. Guillouzo A., Morel F., Langouet S., Maheo K. and Rissel. M. (1977). Use of hepatocytes for the study of hepatotoxic compounds. J Hepatol 26: 73–80. Gum R. J., Hickman D., Fagerland J. A., Heindel M. A., Gagne G. D., Schmidt J. M., Michaelides M. R., Davidsen S. K. and Ulrich R. G. (2001). Analysis of Two Matrix Metalloproteinase Inhibitors and Their Metabolites for Induction of Phospholipidosis in Rat and Human Hepatocytes. Biochem Pharmacol 62: 1661–1673. Hamilton G. A., Jolley S. L., Gilbert D., Coon D. J., Barros S. and LeCluyse E. L. (2001). Regulation of cell morphology and cytochrome P450 expression in human hepatocytes by extracellular matrix and cell-cell interactions. Cell Tissue Res 306: 85–99. Hannon G. and Rossi J. (2004). Unlocking the potential of the human genome with RNAi interference Nature 341: 371–378. Hara Y., Sato Y., Yamamoto S., Nakatsuka H., Takeishi T., Hirano K., Kobayashi T., Watanabe T. and Hatakeyama K. (2003). Dispersion method of excessive portal hypertension (shear stress) and changes of portal pressure and flow after living-related liver transplantation with a splenorenal shunt: a case report. Transplant Proc 35: 414–415. Haroldsen P. E., Reilly M. H., Hughes H., Gaskell S. J. and Porter C. J. (1988). Characterization of glutathione conjugates by fast atom bombardment/tandem mass spectrometry. Biomed Environ Mass Spectrom 15: 615–621.

46

Hepatotoxicity

Hartmann G., Cheung A. K. and Piquette-Miller M. (2002). Inflammatory cytokines, but not bile acids, regulate expression of murine hepatic anion transporters in endotoxemia. J Pharmacol Exp Ther 303: 273–281. Haskins J. R., Rowse P., Rahbari R. and de la Iglesia F. A. (2001). Thiazolidinedione toxicity to isolated hepatocytes revealed by coherent multiprobe fluorescence microscopy and correlated with multiparameter flow cytometry of peripheral leukocytes. Arch Toxicol 75: 425–438. Hengstler J. G., Van Der Burg B., Steinberg P. and Oesch J. (1999). Interspecies differences in cancer susceptibility and toxicity. Drug Metabol Rev 31: 917–970. Hengstler J. G., Brulport M., Schormann W., Bauer A., Hermes M., Nussler A. K., Fandrich F., Ruhnke M., Ungefroren H., Griffin L., Bockamp E., Oesch F. and von Mach M. A. (2005). Generation of human hepatocytes by stem cell technology: definition of the hepatocyte. Expert Opin Drug Metab Toxicol 1: 61–74. Hines R. N. and McCarver D. G. (2002). The ontogeny of human drug-metabolizing enzymes: Phase I oxidative enzymes. J Pharmacol Exp Ther 300: 355–360. Hoebe K. H., Witkamp R. F., Fink-Gremmels J., Van Miert A. S. and Monshouwer M. (2001). Direct cell-to-cell contact between Kupffer cells and hepatocytes augments endotoxin-induced hepatic injury. Am J Physiol Gastrointest Liver Physiol 280: G720–G728. Hoffmaster K. A., Turncliff R. Z., LeCluyse E. L., Kim R. B., Meier P. J. and Brouwer K. L. (2004). P-glycoprotein expression, localization and function in sandwich-cultured primary rat and human hepatocytes: relevance to the hepatobiliary disposition of a model opioid peptide. Pharm Res 21: 1294–1302. Holt M. P. and Ju C. (2006). Mechanisms of drug-induced liver injury. AAPS J 8: E48–E54. Ihrke G., Neufeld E. B., Meads T., Shanks M. R., Cassio D., Laurent M., Schroer T. A., Pagano R. E. and Hubbard A. L. (1993). WIF-B cells: an in vitro model for studies of hepatocyte polarity. J Cell Biol 123: 1761–1775. Ivanov M. A., Heuillet E., Vintezou P., Melcion C. and Cordier A. (1992). Primary Culture of Hepatocytes in the Investigation of Drug-Induced Steatosis, in G. Jolles and A. Cordier (eds), In Vitro Methods in Toxicology, Academic Press, New York, NY, USA, 165–187. Jaeschke H. and Bajt M. L. (2006). Intracellular signaling mechanisms of acetaminophen-induced liver cell death. Toxicol Sci 89: 31–41. Jaeschke H., Gores G. J., Cederbaum A. I., Hinson J. A., Pessayre D. and Lemasters J. J. (2002a). Mechanisms of hepatotoxicity. Toxicol Sci 65: 166–176. Jungermann K. (1986). Functional heterogeneity of periportal and perivenous hepatocytes. Enzyme 35: 161–180. Kaihara S., Borenstein J., Koka R., Lalan S., Ochoa E. R., Ravens M., Pien H., Cunningham B. and Vacanti J. P. (2000). Silicon micromachining to tissue engineer branched vascular channels for liver fabrication. Tissue Eng 6: 105–117. Kakinuma S., Tanaka Y., Chinzei R., Watanabe M., Shimizu-Saito K., Hara Y., Teramoto K., Arii S., Sato C., Takase K., Yasumizu T. and Teraoka H. (2003). Human umbilical cord blood as a source of transplantable hepatic progenitor cells. Stem Cells 21: 217–227. Kalgutkar A. S., Gardner I., Obach R. S., Shaffer C. L., Callegari E., Henne K. R., Mutlib A. E., Dalvie D. K., Lee J. S., Nakai Y., O’Donnell J. P., Boer J. and Harriman S. P. (2005). A comprehensive listing of bioactivation pathways of organic functional groups. Curr Drug Metab 6: 161–225. Kaplowitz N. (2002). Biochemical and cellular mechanisms of toxic liver injury. Semin Liver Dis 22: 137–144. Kaplowitz N. (2005). Idiosyncratic drug hepatotoxicity. Nat Rev Drug Discov 4: 489–499. Kawasaki T., Tamura S., Kiso S., Doi Y., Yoshida Y., Kamada Y., Saeki A., Saji Y. and Matsuzawa Y. (2005). Effects of growth factors on the growth and differentiation of mouse fetal liver epithelial cells in primary cultures. J Gastroenterol Hepatol 20: 857–864.


47

Kessova I. G. and Cederbaum A. I. (2005). The effect of CYP2E1-dependent oxidant stress on activity of proteasomes in HepG2 cells. J Pharmacol Exp Ther 315: 304–312. Knasmuller S., Parzefall W., Sanyal R., Ecker S., Schwab C., Uhl M., Mersch-Sundermann V., Williamson G., Hietsch G., Langer T., Darroudi F. and Natarajan A. T. (1998). Use of metabolically competent human hepatoma cells for the detection of mutagens and antimutagens. Mutat Res 402: 185–202. Knasmuller S., Mersch-Sundermann V., Kevekordes S., Darroudi F., Huber W. W., Hoelzl C., Bichler J. and Majer B. J. (2004). Use of human-derived liver cell lines for the detection of environmental and dietary genotoxicants; current state of knowledge. Toxicology 198: 315–328. Knowles B. B., Howe C. C. and Aden D. P. (1980). Human hepatocellular carcinoma cell lines secrete the major plasma proteins and hepatitis B surface antigen. Science 209: 497–499. Knowles S. R., Uetrecht J. and Shear N. H. (2000). Idiosyncratic Drug Reactions: the Reactive Metabolite Syndromes. Lancet 356: 1587–1591. Kocarek T. A., Schuetz E. G. and Guzelian P. S. (1992). Expression of multiple forms of cytochrome P450 mRNA in primary cultures of rat hepatocytes maintained on Matrigel. Mol Pharmacol 43: 328–334. Kocarek T. A., Schuetz E. G., Strom S. C., Fisher R. A. and Guzelian P. S. (1995). Comparative analysis of cytochrome CYP3A induction in primary cultures of rat, rabbit, and human hepatocytes. Drug Metabol Dispos 23: 415–421. Kogler G., Sensken S., Airey J. A., Trapp T., Muschen M., Feldhahn N., Liedtke S., Sorg R. V., Fischer J., Rosenbaum C., Greschat S., Knipper A., Bender J., Degistirici O., Gao J., Caplan A. I., Colletti E. J., Almeida-Porada G., Muller H. W., Zanjani E. and Wernet P. (2004). A new human somatic stem cell from placental cord blood with intrinsic pluripotent differentiation potential. J Exp Med 200: 123–135. Konieczko E. M., Ralston A. K., Crawford A. R., Karpen S. J. and Crawford J. M. (1998). Enhanced Na+ -dependent bile salt uptake by WIF-B cells, a rat hepatoma hybrid cell line, following growth in the presence of a physiological bile salt. Hepatology 27: 191–199. Kostrubsky V. E., Strom S. C., Hanson J., Urda E., Rose K., Burliegh J., Zocharski P., Cai H., Sinclair J. F. and Sahi J. (2003). Evaluation of hepatotoxic potential of drugs by inhibition of bile-acid transport in cultured primary human hepatocytes and intact rats. Toxicol Sci 76: 220–228. Kostrubsky S. E., Strom S. C., Kalgutkar A. S., Kulkarni S., Atherton J., Mireles R., Feng B., Kubik R., Hanson J., Urda E. and Mutlib A. E. (2006). Inhibition of hepatobiliary transport as a predictive method for clinical hepatotoxicity of nefazodone. Toxicol Sci 90: 451–459. Krause D. S., Theise N. D., Collector M. I., Henegariu O., Hwang S., Gardner R., Neutzel S. and Sharkis S. J. (2001). Multi-organ, multi-lineage engraftment by a single bone marrow-derived stem cell. Cell 105: 369–377. Kulig K. M. and Vacanti J. P. (2004). Hepatic tissue engineering. Transpl Immunol 12: 303–310. Lagasse E., Connors H., Al-Dhalimy M., Reitsma M., Dohse M., Osborne L., Wang X., Finegold M., Weissman I. L. and Grompe M. (2000). Purified hematopoietic stem cells can differentiate into hepatocytes in vivo. Nat Med 6: 1229–1234. Larson A. M., Polson J., Fontana R. J., Davern T. J., Lalani E., Hynan L. S., Reisch J. S., Schiodt F. V., Ostapowicz G., Shakil A. O. and Lee W. M. (2005). Acetaminophen-induced acute liver failure: results of a United States multicenter, prospective study. Hepatology 42: 1364–1372. Lautraite S., Bigot-Lasserre D., Bars R. and Carmichael N. (2003). Optimisation of cell-based assays for medium throughput screening of oxidative stress. Toxicol in Vitro 17: 207–220. Lavon N. and Benvenisty N. (2005). Study of hepatocyte differentiation using embryonic stem cells. J Cell Biochem 96: 1193–1202. Lavon N., Yanuka O. and Benvenisty N. (2004). Differentiation and isolation of hepatic-like cells from human embryonic stem cells. Differentiation 72: 230–238.

48

Hepatotoxicity

Le Vee M., Jigorel E., Glaise D., Gripon P., Guguen-Guillouzo C. and Fardel O. (2006). Functional expression of sinusoidal and canalicular hepatic drug transporters in the differentiated human hepatoma HepaRG cell line. Eur J Pharm Sci 28: 109–117. LeBel C. P., Ischiropoulos H. and Bondy S. C. (1992). Evaluation of the probe 2 ,7 -dichlorofluorescin as an indicator of reactive oxygen species formation and oxidative stress. Chem Res Toxicol 5: 227– 231. LeCluyse E. (2001). Human hepatocyte culture systems for the in vitro evaluation of cytochrome P450 expression and regulation. Eur. J. Pharm. Sci. 13: 343–368. LeCluyse E. L., Audus K. L. and Hochman J. H. (1994). Formation of extensive canalicular networks by rat hepatocytes cultured in collagen–sandwich configuration. Am J Physiol 266: C1764-C1774. LeCluyse E. L., Bullock P. L. and Parkinson A. (1996). Strategies for restoration and maintenance of normal hepatic structure and function in long-term cultures of rat hepatocytes. Adv Drug Deliv Rev 22: 133–186. LeCluyse E. L., Alexandre E., Hamilton G. A., Viollon-Abadie C., Coon D. J., Jolley S. and Richert L. (2005). Isolation and culture of primary human hepatocytes. Meth Mol Biol 290: 207– 229. Lee S. H. and Sinko P. J. (2006). siRNA – getting the message out. Eur J Pharm Sci 27: 401–410. Lee W. M. (2003). Drug-induced hepatotoxicity. New Engl J Med 349: 474–485. Lengyel G., Veres Z., Szabo P., Vereczkey L. and Jemnitz K. (2005). Canalicular and sinusoidal disposition of bilirubin mono- and diglucuronides in sandwich-cultured human and rat primary hepatocytes. Drug Metab Dispos 33: 1355–1360. Letteron P., Sutton A., Mansouri A., Fromenty B. and Pessayre D. (2003). Inhibition of microsomal triglyceride transfer protein: another mechanism for drug-induced steatosis in mice. Hepatology 38: 133–140. Lewis D. F. V., Ioannides C. and Park D. V. (1998). Cytochrome P450 and species differences in xenobiotic metabolism and activation of carcinogens. Environ Health Perspect 106: 633– 641. Li T. and Chiang J. Y. (2006). Rifampicin induction of CYP3A4 requires pregnane X receptor cross talk with hepatocyte nuclear factor 4alpha and coactivators, and suppression of small heterodimer partner gene expression. Drug Metabol Dispos 34: 756–764. Lilius H., Haestbacka T. and Isomaa B. (1996). A combination of fluorescent probes for evaluation of cytotoxicity and toxic mechanisms in isolated rainbow trout hepatocytes. Toxicol in Vitro 10: 341–348. Liu X., Chism J. P., LeCluyse E. L., Brouwer K. R. and Brouwer K. L. (1999). Correlation of biliary excretion in sandwich-cultured rat hepatocytes and in vivo in rats. Drug Metab Dispos 27: 637–644. Lu P. Y., Xie F. and Woodle M. C. (2006). In vivo application of RNA interference: from functional genomics to therapeutics. Adv Genet 54: 117–142. Luo G., Guenthner T., Gan L. S. and Humphreys W. G. (2004). CYP3A4 induction by xenobiotics: biochemistry, experimental methods and impact on drug driscovery an develeopment. Curr Drug Metabol 5: 483–505. Mace K., Aguilar F., Wang J. S., Vautravers P., Gomez-Lechon M., Gonzalez F. J., Groopman J., Harris C. C. and Pfeifer A. M. (1997). Aflatoxin B1-induced DNA adduct formation and p53 mutations in CYP450-expressing human liver cell lines. Carcinogenesis 18: 1291–1297. Maenpaa J., Rane A., Raunio H., Honkakoski P. and Pelkonen O. (1993). Cytochrome P450 isoforms in human fetal tissues related to phenobarbital-inducible forms in the mouse. Biochem Pharmacol 45: 899–907. Malhi H., Gores G. J. and Lemasters J. J. (2006). Apoptosis and necrosis in the liver: a tale of two deaths? Hepatology 43: S31–S44. Manov I., Hirsh M. and Iancu T. C. (2002). Acetaminophen hepatotoxicity and mechanisms of its protection by N -acetylcysteine: a study of Hep3B cells. Exp Toxicol Pathol 53: 489–500.


49

Marschall H. U., Wagner M., Zollner G., Fickert P., Diczfalusy U., Gumhold J., Silbert D., Fuchsbichler A., Benthin L., Grundstrom R., Gustafsson U., Sahlin S., Einarsson C. and Trauner M. (2005). Complementary stimulation of hepatobiliary transport and detoxification systems by rifampicin and ursodeoxycholic acid in humans. Gastroenterology 129: 476–485. McCarthy T. C., Pollak P. T., Hanniman E. A. and Sinal C. J. (2004). Disruption of hepatic lipid homeostasis in mice after amiodarone treatment is associated with peroxisome proliferator-activated receptor-alpha target gene activation. J Pharmacol Exp Ther 311: 864–873. McMillian M. K., Grant E. R., Zhong Z., Parker J. B., Li L., Zivin R. A., Burczynski M. E. and Johnson M. D. (2001). Nile Red binding to HepG2 cells: an improved assay for in vitro studies of hepatosteatosis. In vitro Mol Toxicol 14: 177–190. McRae M. P., Lowe C. M., Tian X., Bourdet D. L., Ho R. H., Leake B. F., Kim R. B., Brouwer K. L. and Kashuba A. D. (2006). Ritonavir, Saquinavir and Efavirenz, but not Nevirapine, Inhibit Bile Acid Transport in Human and Rat Hepatocytes. J Pharmacol Exp Ther 318: 1068–1075. Mehendale H. M. (2005). Tissue repair: an important determinant of final outcome of toxicant-induced injury. Toxicol Pathol 33: 41–51. Micheline D., Emmanuel J. and Serge E. (2002). Effect of Ursodeoxycholic Acid on the Expression of the Hepatocellular Bile Acid Transporters (Ntcp and bsep) in Rats with Estrogen-Induced Cholestasis. J Pediatr Gastroenterol Nutr 35: 185–191. Miki T., Lehmann T., Cai H., Stolz D. B. and Strom S. C. (2005). Stem Cell Characteristics of Amniotic Epithelial Cells. Stem Cells 23: 1549–1559. Mimura M, Baba T, Yamazaki H, Ohmori S, Inui Y, Gonzalez F. J, Guengerich F.P. and Shimada T. (1993). Characterization of cytochrome P450 2B6 in human liver microsomes. Drug Metabol Dispos. 2: 1048–1056. Minagawa M., Deng Q., Liu Z. X., Tsukamoto H. and Dennert G. (2004). Activated natural killer T cells induce liver injury by Fas and tumor necrosis factor-alpha during alcohol consumption. Gastroenterology 126: 1387–1399. Mita S., Suzuki H., Akita H., Hayashi H., Onuki R., Hofmann A. F. and Sugiyama Y. (2006). Inhibition of bile acid transport across NTCP and BSEP co-expressing LLC-PK1 cells by cholestasis-inducing drugs. Drug Metab Dispos 34: 1575–1581. Miyazaki M., Akiyama I., Sakaguchi M., Nakashima E., Okada M., Kataoka K. and Huh N. H. (2002). Improved conditions to induce hepatocytes from rat bone marrow cells in culture. Biochem Biophys Res Commun 298: 24–30. Miyazaki M., Masaka T., Akiyama I., Nakashima E., Sakaguchi M. and Huh N. H. (2004). Propagation of adult rat bone marrow-derived hepatocyte-like cells by serial passages in vitro. Cell Transplant 13: 385–391. Motomura W., Inoue M., Ohtake T., Takahashi N., Nagamine M., Tanno S., Kohgo Y. and Okumura T. (2006). Up-regulation of ADRP in fatty liver in human and liver steatosis in mice fed with high fat diet. Biochem Biophys Res Commun 340: 1111–1118. Neuman M. G., Koren G. and Tiribelli C. (1993). In vitro assessment of the ethanol-induced hepatotoxicity on HepG2 cell line. Biochem Biophys Res Commun 197: 932–941. Newsome P. N., Johannessen I., Boyle S., Dalakas E., McAulay K. A., Samuel K., Rae F., Forrester L., Turner M. L., Hayes P. C., Harrison D. J., Bickmore W. A. and Plevris J. N. (2003). Human cord blood-derived cells can differentiate into hepatocytes in the mouse liver with no evidence of cellular fusion. Gastroenterology 124: 1891–1900. Noe J., Stieger B. and Meier P. J. (2002). Functional expression of the canalicular bile salt export pump of human liver. Gastroenterology 123: 1659–1666. Nonome K., Li X. K., Takahara T., Kitazawa Y., Funeshima N., Yata Y., Xue F., Kanayama M., Shinno E., Kuwae C., Saito S., Watanabe A. and Sugiyama T. (2005). Human umbilical cord blood-derived cells differentiate into hepatocyte-like cells in the Fas-mediated liver injury model. Am J Physiol Gastrointest Liver Physiol 289: G1091–G1099.

50

Hepatotoxicity

Nowak G., Ericzon B. G., Nava S., Jaksch M., Westgren M. and Sumitran-Holgersson S. (2005). Identification of expandable human hepatic progenitors which differentiate into mature hepatic cells in vivo. Gut 54: 972–979. O’Brien P. J. and Haskins J. R. (2006). In vitro cytotoxicity assessment, in K. Guiliano, D. Lansing Taylor and J. R. Haskins (eds), High-Content Screening: A Powerful Approach to Systems Cell Biology and Drug Discovery, Human a Press, Totowa, NJ, USA, 415–425. O’Brien P. J., Kalow B. I., Ali N., Lassaline L. A. and Lumsden J. H. (1990). Compensatory increase in calcium extrusion activity of untreated lymphocytes from swine susceptible to malignant hyperthermia. Am J Vet Res 51: 1038–1043. O’Brien P. J., Irwin W., Diaz D., Howard-Cofield E., Krejsa C. M., Slaughter M. R., Gao B., Kaludercic N., Angeline A., Bernardi P., Brain P. and Hougham C. (2006). High concordance of drug-induced human hepatotoxicity with in vitro cytotoxicity measured in a novel cell-based model using high content screening. Arch Toxicol 80: 580–604. Olson H., Betton G., Robinson D., Thomas K., Monro A., Kolaja G., Lilly P., Sanders J., Sipes G., Bracken W., Dorato M., Van Deun K., Smith P., Berger B. and Heller A. (2000). Concordance of the toxicity of pharmaceuticals in humans and in animals. Regulat Toxicol Pharmacol 32: 56–67. Omiecinski C. J., Redlich C. A. and Costa P. (1990). Induction and developmental expression of cytochrome P450IA1 messenger RNA in rat and human tissues: detection by the polymerase chain reaction. Cancer Res 50: 4315–4321. Ostapowicz G., Fontana R. J., Schiodt F. V., Larson A., Davern T. J., Han S. H., McCashland T. M., Shakil A. O., Hay J. E., Hynan L., Crippin J. S., Blei A. T., Samuel G., Reisch J. and Lee W. M. (2002). Results of a prospective study of acute liver failure at 17 tertiary care centers in the United States. Ann Intern Med 137: 947–954. Pang K. S., Schwab A. J., Goresky C. A. and Chiba M. (1994). Transport, binding and metabolism of sulfate conjugates in the liver. Chem Biol Interact 92: 179–207. Park B. K., Naisbitt D. J., Gordon S. F., Kitteringham N. R. and Pirmohamed M. (2001). Metabolic activation in drug allergies. Toxicology 158: 11–23. Park K., Williams D. P., Naisbitt D. J., Kitteringham N. R. and Pirmohamed M. (2005). Investigation of toxic metabolites during drug development. Toxicol Appl Pharmacol 207: 425–434. Pearson P. G., Threadgill M. D., Howald W. N. and Baillie T. A. (1988). Applications of tandem mass spectrometry to the characterization of derivatized glutathione conjugates. Studies with S-(N -methylcarbamoyl)glutathione, a metabolite of the antineoplastic agent N -methylformamide. Biomed Environ Mass Spectrom 16: 51–56. Peraza M. A., Burdick A. D., Marin H. E., Gonzalez F. J. and Peters J. M. (2006). The toxicology of ligands for peroxisome proliferators-activated receptors (PPAR). Toxicol Sci 90: 269–295. Pessayre D. (1995). Role of reactive metabolites in drug-induced hepatitis. J Hepatol 23: S16–S24. Peters T. (2005). Do preclinical testing strategies help predict human hepatotoxic potentials? Toxicol Pathol 33: 146–154. Petersen B. E., Bowen W. C., Patrene K. D., Mars W. M., Sullivan A. K., Murase N., Boggs S. S., Greenberger J. S. and Goff J. P. (1999). Bone marrow as a potential source of hepatic oval cells. Science 284: 1168–1170. Pfeifer A. M., Cole K. E., Smoot D. T., Weston A., Groopman J. D., Shields P. G., Vignaud J. M., Juillerat M., Lipsky M. M., Trump B. F., Lechner J. F. and Harris C. C. (1993). Simian virus 40 large tumor antigen-immortalized normal human liver epithelial cells express hepatocyte characteristics and metabolize chemical carcinogens. Proc Natl Acad Sci USA 90: 5123–5127. Pichler A., Zelcer N., Prior J. L., Kuil A. J. and Piwnica-Worms D. (2005). In vivo RNA interferencemediated ablation of MDR1 P-glycoprotein. Clin Cancer Res 11: 4487–4494. Pirmohamed M., Naisbitt D. J., Gordon F. and Park B. K. (2002). The danger hypothesis – potential role in idiosyncratic drug reactions. Toxicology 181/182: 55–63.


51

Plass J. R. M., Mol O., Heegsma J., Geuken M., Faber K. N., Jansen P. L. M. and Muller M. (2002). Farsenoid X receptor and bile salts are involved in transcriptional regulation of the gene encoding the human bile salt export pump. Hepatology 35: 589–596. Powers M. J., Domansky K., Kaazempur-Mofrad M. R., Kalezi A., Capitano A., Upadhyaya A., Kurzawski P., Wack K. E., Stolz D. B., Kamm R. and Griffith L. G. (2002a). A microfabricated array bioreactor for perfused 3D liver culture. Biotechnol Bioeng 78: 257–269. Powers M. J., Janigian D. M., Wack K. E., Baker C. S., Beer Stolz D. and Griffith L. G. (2002b). Functional behavior of primary rat liver cells in a three-dimensional perfused microarray bioreactor. Tissue Eng 8: 499–513. Prough R. A., Xiao G. H. and Falkner K. C. (1996). Hormonal regulation of hepatic enzymes involved in foreign compound metabolism. FASEB J 10: 1369–1377. Rainsford K. D., Seabrook R. W., Spencer S. and Hewson A. T. (2001). Effects of nimesulide and its metabolites or manufacturing intermediates on the viability and growth of the human hepatoma HepG2 cell line. Life Sci 69: 2965–2973. Rambhatla L., Chiu C. P., Kundu P., Peng Y. and Carpenter M. K. (2003). Generation of hepatocytelike cells from human embryonic stem cells. Cell Transplant 12: 1–11. Reid A. B., Kurten R. C., McCullough S. S., Brock R. W. and Hinson J. A. (2005). Mechanisms of acetaminophen-induced hepatotoxicity: role of oxidative stress and mitochondrial permeability transition in freshly isolated mouse hepatocytes. J Pharmacol Exp Ther 312: 509–516. Renton K. W. (2004). Cytochrome P450 regulation and drug biotransformation during inflammation and infection. Curr Drug Metab 5: 235–243. Richert L., Binda D., Hamilton G. A., Viollon-Abadie A. E., Bigot-Lasserre D., Bars R., Coasssolo P. and LeCluyse E. (2002). Evaluation of the effect of culture configuration on morphology, survival time, antioxidant status and metabolic capabilities of cultured rat hepatocytes. Tox in Vitro 16: 89–99. Richert L., Liguori M., Abadie C., Heyd B., Mantion G., Khalkic N. and Waring J. F. (2006). Gene expression in human hepatocytes in suspension after isolation is similar to the liver in origin, is not affected by hepatocyte cold storage and cryopreservation, but is strongly changed after hepatocyte plating. Drug Metabol Dispos 34: 870–879. Roberts R. (1999). Peroxisome proliferators: mechanisms of adverse effects in rodents and molecular basis for species differences. Arch Toxicol 73: 413–418. Robin M. A., Maratrat M., Le Roy M., Le Breton F. P., Bonierbale E., Dansette P., Ballet F., Mansuy D. and Pessayre D. (1996). Antigenic targets in tienilic acid hepatitis. Both cytochrome P450 2C11 and 2C11–tienilic acid adducts are transported to the plasma membrane of rat hepatocytes and recognized by human sera. J Clin Invest 98: 1471–1480. Roman I. D., Fernandez-Moreno M. D., Fueyo J. A., Roma M. G. and Coleman R. (2003). Cyclosporin A induced internalization of the bile salt export pump in isolated rat hepatocyte couplets. Toxicol Sci 71: 276–281. Ruhnke M., Nussler A. K., Ungefroren H., Hengstler J. G., Kremer B., Hoeckh W., Gottwald T., Heeckt P. and Fandrich F. (2005a). Human monocyte-derived neohepatocytes: a promising alternative to primary human hepatocytes for autologous cell therapy. Transplantation 79: 1097–1103. Ruhnke M., Ungefroren H., Nussler A., Martin F., Brulport M., Schormann W., Hengstler J. G., Klapper W., Ulrichs K., Hutchinson J. A., Soria B., Parwaresch R. M., Heeckt P., Kremer B. and Fandrich F. (2005b). Differentiation of in vitro-modified human peripheral blood monocytes into hepatocyte-like and pancreatic islet-like cells. Gastroenterology 128: 1774–1786. Saad B., Scholl F. A., Thomas H., Schawalder H., Streit V., Waechter F. and Maier P. (1993). Crude liver membrane fractions and extracellular matrix components as substrata regulate differentially the preservation and inducibility of cytochrome P-450 isoenzymes in cultured rat hepatocytes. Eur J Biochem 213: 805–814.

52

Hepatotoxicity

Sahi J. (2005). Use of in vitro transporter assays to understand hepatic and renal disposition of new drug candidates. Exp Opin 1: 409–427. Saji Y., Tamura S., Yoshida Y., Kiso S., Iizuka A. S., Matsumoto H., Kawasaki T., Kamada Y., Matsuzawa Y. and Shinomura Y. (2004). Basic fibroblast growth factor promotes the transdifferentiation of mouse bone marrow cells into hepatic lineage cells via multiple liver-enriched transcription factors. J Hepatol 41: 545–550. Sanderson J. P., Naisbitt D. J. and Park B. K. (2006). Role of bioactivation in drug-induced hypersensitivity reactions. AAPS J 8: E55–E64. Sato Y., Koyama S., Tsukada K. and Hatakeyama K. (1997). Acute portal hypertension reflecting shear stress as a trigger of liver regeneration following partial hepatectomy. Surg Today 27: 518–526. Sato Y., Tsukada K. and Hatakeyama K. (1999). Role of shear stress and immune responses in liver regeneration after a partial hepatectomy. Surg Today 29: 1–9. Sawada H., Takami K. and Asahi S. (2005). A toxicogenomic approach to drug-induced phospholipidosis: analysis of its induction mechanism and establishment of a novel in vitro screening system. Toxicol Sci 83: 282–292. Schaffert C. S., Todero S. L., McVicker B. L., Tuma P. L., Sorrell M. F. and Tuma D. J. (2004). WIF-B cells as a model for alcohol-induced hepatocyte injury. Biochem Pharmacol 67: 2167–2174. Schmelzer E., Wauthier E. and Reid L. M. (2006). The phenotypes of pluripotent human hepatic progenitors. Stem Cells 24: 1852–1858. Schuetz E. G., Curtiss D. L., Omiecinski C. J., Mueller-Eberhard U., Kleinman H. K., Elswick B. and Guzelian P. S. (1988). Regulation of gene expression in adult rat hepatocytes cultured on a basement membrane matrix. J Cell Physiol 134: 309–323. Schuetz E. G., Schmid W., Schutz G., Brimer C., Yasuda K., Kamataki T., Bornhrim L., Myles K. and Cole T. J. (2000). The glucocorticoid receptor is essential for induction of cytochrome P450B by steroids but not for drug or steroid induction of CYP3A or P450 reductase in mouse liver. Drug Metabol Dispos 28: 268–366. Schuetz J. D., Beach D. L. and Guzelian P. S. (1994). Selective expression of cytochrome P450 CYP3A mRNAs in embryonic and adult human liver. Pharmacogenetics 4: 11–20. Schuster D., Laggner C. and Langer T. (2005). Why drugs fail – a study on side effects in new chemical entities. Curr Pharm Des 11: 3545–3559. Schwabe R. F. and Brenner D. A. (2006). Mechanisms of Liver Injury. I. TNF-alpha-induced liver injury: role of IKK, JNK and ROS pathways. Am J Physiol Gastrointest Liver Physiol 290: G583– G589. Schwartz R. E., Reyes M., Koodie L., Jiang Y., Blackstad M., Lund T., Lenvik T., Johnson S., Hu W. S. and Verfaillie C. M. (2002). Multipotent adult progenitor cells from bone marrow differentiate into functional hepatocyte-like cells. J Clin Invest 109: 1291–1302. Shankar P., Manjunath N. and Lieberman J. (2005). The prospect of silencing disease using RNA interference. JAMA 293: 1367–1373. Shanks M. R., Cassio D., Lecoq O. and Hubbard A. L. (1994). An improved polarized rat hepatoma hybrid cell line. Generation and comparison with its hepatoma relatives and hepatocytes in vivo. J Cell Sci 107: 813–825. Sharma A. D., Cantz T., Richter R., Eckert K., Henschler R., Wilkens L., Jochheim-Richter A., Arseniev L. and Ott M. (2005). Human cord blood stem cells generate human cytokeratin 18negative hepatocyte-like cells in injured mouse liver. Am J Pathol 167: 555–564. Shimada T., Yamazaki H., Mimura M., Wakamiya N., Ueng Y. F., Guengerich F. P. and Inui Y. (1996). Characterization of microsomal cytochrome P450 enzymes involved in the oxidation of xenobiotic chemicals in human fetal liver and adult lungs. Drug Metab Dispos 24: 515–522. Shirahashi H., Wu J., Yamamoto N., Catana A., Wege H., Wager B., Okita K. and Zern M. A. (2004). Differentiation of human and mouse embryonic stem cells along a hepatocyte lineage. Cell Transpl 13: 197–211.


53

Sidhu J. S., Farin F. M. and Kavanagh T. J. (1994). Effect of tissue-culture substratum and extracellular matrix overlay on liver-selective and xenobiotic inducible gene expression in primary rat hepatocytes In vitro Toxicol 7: 225–242. Silva J. M., Morin P. E., Day S. H., Kennedy B. P., Payetter P., Rushmore T., Yergey J. A. and Nicoll-Griffith D. A. (1998). Refinement of an in vitro cell model for cytochrome P450 induction. Drug Metabol Dispos 26: 490–496. Sinz M. W. and Kim S. (2006). Stem cells, immortalized cells and primary cells in ADMET assays. Drug Discov Technol 3: 79–85. Sivaraman A., Leach J. K., Townsend S., Iida T., Hogan B. J., Stolz D. B., Fry R., Samson L. D., Tannenbaum S. R. and Griffith L. G. (2005). A microscale in vitro physiological model of the liver: predictive screens for drug metabolism and enzyme induction. Curr Drug Metab 6: 569– 591. Slaughter M. R., Thakkar H. and O’Brien P. J. (2002). Effect of diquat on the antioxidant system and cell growth in human neuroblastoma cells. Toxicol Appl Pharmacol 178: 63–70. Sledz C. A. and Williams B. R. (2005). RNA interference in biology and disease. Blood 106: 787–794. Smith C. M., Graham R. A., Krol W. L., Silver I. S., Negishi M., Wang H. and Lecluyse E. L. (2005). Differential UGT1A1 induction by chrysin in primary human hepatocytes and HepG2 Cells. J Pharmacol Exp Ther 315: 1256–1264. Song E., Lee S.-K., Wang J., Ince N., Ouyang N., Min J., Chen J., Shankar P. and Lieberman J. (2003). RNA interference targeting Fas protects mice from fulminant hepatitits. Nature Med 9: 347–351. Sonnier M. and Cresteil T. (1998). Delayed ontogenesis of CYP1A2 in the human liver. Eur J Biochem 251: 893–898. Soto-Gutierrez A., Navarro-Alvarez N., Rivas-Carrillo J. D., Chen Y., Yamatsuji T., Tanaka N. and Kobayashi N. (2006). Differentiation of human embryonic stem cells to hepatocytes using deleted variant of HGF and poly-amino-urethane-coated nonwoven polytetrafluoroethylene fabric. Cell Transpl 15: 335–341. Stevens J. C., Hines R. N., Gu C., Koukouritaki S. B., Manro J. R., Tandler P. J. and Zaya M. J. (2003). Developmental expression of the major human hepatic CYP3A enzymes. J Pharmacol Exp Ther 307: 573–582. Stieger B., Fattinger K., Madon J., Kullak-Ublick G. A. and Meier P. J. (2000). Drug- and estrogeninduced cholestasis through inhibition of the hepatocellular bile salt export pump (BSEP) of rat liver. Gastroenterology 118: 422–430. Strom S. C., Pisarov L. A., Dorko K., Thompson M. T., Schuetz J. D. and Schuetz E. G. (1996). Use of human hepatocytes to study P450 gene induction. Meth Enzymol 272: 388–401. Tam Y. K. (1993). Individual variation in first-pass metabolism. Clin Pharmacokinet 25: 300–328. Tan N. S., Michalik L., Desvergne B. and Wahli W. (2005). Multiple expression control mechanisms of peroxisome proliferator-activated receptors and their target genes. J Steroid Biochem Mol Biol 93: 99–105. Tateishi T., Nakura H., Asoh M., Watanabe M., Tanaka M., Kumai T., Takashima S., Imaoka S., Funae Y., Yabusaki Y., Kamataki T. and Kobayashi S. (1997). A comparison of hepatic cytochrome P450 protein expression between infancy and postinfancy. Life Sci 61: 2567–2574. Tay V. K., Wang A. S., Leow K. Y., Ong M. M., Wong K. P. and Boelsterli U. (2005). Mitochondrial permeability transition as a source of superoxide anion induced by the nitroaromatic drug nimesulide in vitro. Free Radic Biol Med 39: 949–959. Teramoto K., Asahina K., Kumashiro Y., Kakinuma S., Chinzei R., Shimizu-Saito K., Tanaka Y., Teraoka H. and Arii S. (2005). Hepatocyte differentiation from embryonic stem cells and umbilical cord blood cells. J Hepatobil Pancreat Surg 12: 196–202. Thompson D. C., Barhoumi R. and Burghard R. C. (1998). Comparative toxicity of eugenol and its quinone methide metabolite in cultured liver cells using kinetic fluorescence bioassays. Toxicol Appl Pharmacol 149: 55–63.

54

Hepatotoxicity

Tian X., Zamek-Gliszczynski J., Zhang P. and Brower K. L. R. (2004). Modulation of multidrug resistance-associated protein 2 (Mrp2) and Mrp3 expression and function with small interfering RNA in sandwich-cultured rat hepatocytes. Mol Pharmacol 66: 1004–1010. Tosh D., Shen C. N. and Slack J. M. (2002). Differentiated properties of hepatocytes induced from pancreatic cells. Hepatology 36: 534–543. Tuschl G. and Mueller S. O. (2006). Effects of cell culture conditions on primary rat hepatocytes-cell morphology an differential gene expression. Toxicology 218: 205–215. Vassilopoulos G., Wang P. R. and Russell D. W. (2003). Transplanted bone marrow regenerates liver by cell fusion. Nature 422: 901–904. Vignati L. A., Bogni A., Grossi P. and Monshouwer M. (2004). A human and mouse pregnane X receptor reporter gene assay in combination with cytotoxicity measurements as a tool to evaluate species-specific CYP3A induction. Toxicology 199: 23–33. Vignati L., Turlizzi E., Monaci S., Grossi P., Kanter R. and Monshouwer M. (2005). An in vitro approach to detect metabolite toxicity due to CYP3A4-dependent bioactivation of xenobiotics. Toxicology 216: 154–167. Wagers A. J., Sherwood R. I., Christensen J. L. and Weissman I. L. (2002). Little evidence for developmental plasticity of adult hematopoietic stem cells. Science 297: 2256–2259. Walkup M. H. and Gerber D. A. (2006). Hepatic stem cells: in search of. Stem Cells 24: 1833–1840. Wang H. and Joseph J. A. (1999). Quantifying cellular oxidative stress by dichlorofluorescein assay using microplate reader. Free Radic Biol Med 27: 612–616. Wang K., Shindoh H., Inoue T. and Horii I. (2002). Advantages of in vitro cytotoxicity testing by using primary rat hepatocytes in comparison with established cell lines. J Toxicol Sci 27: 229– 237. Wang X., Willenbring H., Akkari Y., Torimaru Y., Foster M., Al-Dhalimy M., Lagasse E., Finegold M., Olson S. and Grompe M. (2003). Cell fusion is the principal source of bone-marrow-derived hepatocytes. Nature 422: 897–901. Waring J. F., Ciurlionis R., Jolly R. A., Heindel M., Gagne G., Fagerland J. A. and Ulrich R. G. (2003). Isolated human hepatocytes in culture display markedly different gene expression patterns depending on atachment status. Tox in Vitro 17: 693–701. Waxman D. J., Pampori N. A. and Ram P. (1991). Interpulsive interval in circulating growth hormone patterns regulates sexually dimorphic expression of hepatic cytochrome P450. Biochem J 88: 6868–6872. Wilkening S., Stahl F. and Bader A. (2003). Comparison of primary human hepatocytes and hepatoma cell line HepG2 with regard to their biotransformation properties. Drug Metab Dispos 31: 1035– 1042. Willenbring H., Bailey A. S., Foster M., Akkari Y., Dorrell C., Olson S., Finegold M., Fleming W. H. and Grompe M. (2004). Myelomonocytic cells are sufficient for therapeutic cell fusion in liver. Nature Med 10: 744–748. Wormhoudt L. W., Commandeur J. N. M. and Vermeulen N. P. E. (1999). Genetic polymorphisms of human N -acetyltransferase, cytochrome P450, glutathione-s-tranferase, and epoxide hydrolase enzymes: Relevance to xenobiotic metabolism and toxicity. Crit Rev Toxicol 29: 59– 124. Wright M. E., M.J. T. and Aebersold R. (2003). Androgen receptor represses the neuroendocrine transdifferentiation process in prostate cancer cells. Mol Endocrinol 17: 1726–1737. Wrighton S. A., Molowa D. T. and Guzelian P. S. (1988). Identification of a cytochrome P-450 in human fetal liver related to glucocorticoid-inducible cytochrome P-450HLp in the adult. Biochem Pharmacol 37: 3053–3055. Xie W., Barwick J. L., Downes M., Blumberg B., Simon C. M., Nelson M. C., Neuschwander-Tetri B. A., Brunt E. M., Guzelian P. S. and Evans R. M. (2000). Humanized xenobiotic respnse in mice expressing nuclear receptor SXR. Nature 406: 435–439.


55

Xu Y., Bradham C., Brenner D. A. and Czaja M. J. (1997). Hydrogen peroxide-induced liver cell necrosis is dependent on AP-1 activation. Am J Physiol 273: G795–G803. Xu J. J., Diaz D. and O’Brien P. J. (2004). Applications of cytotoxicity assays and pre-lethal mechanistic assays for assessment of human hepatotoxicity potential. Chem Biol Interact 150: 115–128. Xu D., McCarty D., Fernandes A., Fisher M., Samulski R. J. and Juliano R. L. (2005). Delivery of MDR1 small interfering RNA by self-complementary recombinant adeno-associated virus vector. Mol Ther 11: 523–530. Yamaguchi K., Itoh K., Masuda T., Umemura A., Baum C., Itoh Y., Okanoue T. and Fujita J. (2006). In vivo selection of transduced hematopoietic stem cells and little evidence of their conversion into hepatocytes in vivo. J Hepatol 45: 681–687. Yang H. Y., Lee Q. P., Rettie A. E. and Juchau M. R. (1994). Functional cytochrome P4503A isoforms in human embryonic tissues: expression during organogenesis. Mol Pharmacol 46: 922–928. Yoon H. G., Chan D. W., Huang Z. Q., Li J., Fondell J. D., Qin J. and Wong J. (2003). Purification and functional characterization of the human N-CoR complex: the roles of HDAC3, TBL1 and TBLR1. Embo J 22: 1336–1346. Zeilinger K., Auth S., Unger J., Grebe A., Mao L., Petrik M., Holland G., Appel K., Nussler A., Neuhaus P. and Gerlach J. (2000). Liver cell culture in bioreactors for in vitro drug studies as an alternative to animal testing. Altex 17: 3–10. Zeilinger K., Sauer I. M., Pless G., Strobel C., Rudzitis J., Wang A., Nussler A. K., Grebe A., Mao L., Auth S. H., Unger J., Neuhaus P. and Gerlach J. C. (2002). Three-dimensional co-culture of primary human liver cells in bioreactors for in vitro drug studies: effects of the initial cell quality on the long-term maintenance of hepatocyte-specific functions. Altern Lab Anim 30: 525–538. Zeilinger K., Holland G., Sauer I. M., Efimova E., Kardassis D., Obermayer N., Liu M., Neuhaus P. and Gerlach J. C. (2004). Time course of primary liver cell reorganization in three-dimensional highdensity bioreactors for extracorporeal liver support: an immunohistochemical and ultrastructural study. Tissue Eng 10: 1113–1124. Zhang Z. Y., Pelletier R. D., Wong Y. N., Sugawara M., Zhao N. and Littlefield B. A. (2006). Preferential inducibility of CYP1A1 and CYP1A2 by TCDD: differential regulation in primary human hepatocytes versus transformed human cells. Biochem Biophys Res Commun 341: 399–407. Zhou G., Hashimoto Y., Kwak I., Tsai S. Y. and Tsai M. J. (2003). Role of the steroid receptor coactivator SRC-3 in cell growth. Mol Cell Biol 23: 7742–7755.

2 Utilization of an in vitro Hepatotoxicity Test in the Early Stage of Drug Discovery Ikuo Horii, Hiroshi Yamada, Rie Kikkawa, Toshinori Yamamoto, Tamio Fukushima, and Kaori Tomizawa

2.1

General Introduction

Safety evaluation of pharmaceutical products has been conducted in the form of risk assessment and the study of prediction and prevention of adverse effects on the basis of an extensive scientific area over toxicology, pathology, pharmacology, biochemistry, physiology, etc. With the rapid progress of genomic science, diseases and their causes have become understood at the genetic level. With remarkable advancements both in drug discovery, R and D activities and in concomitant technical supports, drug discovery strategies have become focused on drug safety screening in their early stages. Today, drug discovery efforts start from ‘molecular targeting’ based on pharmacogenomics, probably because the introduction of drug discovery research centered on gene targeting, as well as combinatorial chemistry, has realized the synthesis of many compounds over a short period. The phenomenon suggests the necessity of evaluating the toxicities of various compounds with small amounts of their samples as quickly as possible. This series of toxicological strategies is now commonly called ‘High-Throughput Toxicology’ (HTP-Tox), an essential part of toxicological study in the early phase of drug discovery. Among others, in vitro evaluation systems, particularly as screening systems, play an important role in the early phase of pharmaceutical development. They are also important to clarify the mechanisms of toxicity observed during development. On the other hand,


58

Hepatotoxicity

toxicopanomics technologies (a collective designation for the ‘-omics’, such as toxicogenomics, toxicoproteomics and metabonomics) is expected to be applicable to predictive toxicology and mechanism-based risk assessment in the area of toxicology (Stubberfield and page, 1999, Pennie et al., 2000, Suter et al., 2004). Currently, toxicopanomics technologies are being applied to the development of new in vitro safety evaluation systems.

2.2

Introduction of in vitro Screening Tests

One of the advantages of in vitro tests is that direct effects on cells can be evaluated under experimental conditions, controlled artificially and strictly. Closely designed in vitro tests provide useful information to clarify toxicological mechanisms. Since the quick, efficient measurement of multiple compounds is available simultaneously under the same conditions, in vitro tests are useful for the screening or ranking of compounds. Generally, since most compounds are metabolized in the liver, using hepatotoxicity as the indicator for a screening system can trigger safety evaluation in the early phase of drug discovery. On the other hand, cells never function independently in an organism. Instead, they form close and complicated networks with each other or with the matrix of the organism to realize their functions in a three-dimensional structure. Therefore, in vitro test data should be interpreted in view of the existence form of cells in the organism and the surrounding environment, and the results obtained must be reviewed in an in vivo system. Particularly, how much the cells to be used in an in vitro test maintain normal properties significantly affects the accuracy of extrapolation into the subsequent relevant in vivo system. Although primary cells in culture and tissue slices maintain relatively normal properties, it is difficult to maintain their properties over a long period. Established cell lines keep properties stable for a long time, but it must be noted that they are inconsistent in many aspects with the properties of normal cells. There are many considerations in conducting an in vitro test, including how to allow cells to maintain normal functions and how to integrate the in vivo metabolic factors of compounds into the in vitro system environment. Anyway, this basic approach is applicable to hepatotoxicity evaluation systems as well.

2.3

Correlations between in vitro Systems and in vivo Systems

When an in vitro system is used to predict the hepatotoxicity of compounds, it is necessary to set endpoints carefully and evaluate correlations with the relevant in vivo system. While some screening systems have been reported for in vitro hepatotoxicity evaluation (Groneberg et al., 2002, Farkas et al., 2005) we report prediction of in vivo hepatotoxicity using primary rat liver cell cultures (Kikkawa et al., 2005, 2006). Four compounds known as hepatotoxicity agents were investigated, namely acetaminophen (APAP), amiodarone (AMD), tetracycline (TC) and carbon tetrachloride (CTC). To evaluate hepatotoxicity in rats, 300 or 1000 mg/kg of APAP, 300 or 1000 mg/kg of AMD, 600 or 2000 mg/kg of TC and 0.3 or 1 ml/kg of CTC were orally administered once to rats, and changes in blood biochemical parameters as well as histopathological changes were investigated 6 and 24 h after administration. For the livers of APAP-administered rats, changes in protein expression were investigated by proteomics. It was found, 24 h after

Utilization of an in vitro Hepatotoxicity Test

59

administration, that all of the compounds had caused histopathological changes, such as inflammatory ones. Immunohistological examinations revealed the expression of oxidative stress-related proteins 6 h after administration. The analysis of changes in protein expression by proteomics detected changes in the oxidative stress-related proteins and mitochondrial metabolism-related proteins, suggesting their usefulness as hepatotoxicity evaluation markers in in vivo systems. Then, using the same APAP, AMD, TC and CTC, toxicity in primary liver cell cultures was evaluated. In investigation, morphological changes, LDH release (as the indicator of cytotoxicity) and changes in WST-1 (as the indicator of effects on mitochondrial respiration) were evaluated. Furthermore, changes in protein expression were analyzed by proteomics. It was found that all of the compounds induced morphological changes and caused a dose-dependent increase in LDH release and a decrease in WST-1. Changes in WST-1 occurred in shorter exposure times than those in LDH, suggesting that mitochondrial respiration ability would be a useful parameter for cytotoxicity in in vitro hepatotoxicity screening. The analysis of changes in protein expression by proteomics in APAP-exposed samples showed changes in the expression of oxidative stress-related proteins and mitochondrial regulation-related proteins (Figure 2.1). Thus, hepatotoxicity evaluation using primary liver cell cultures was found to be useful to predict hepatotoxicity in in vivo systems, and that the oxidative stress-related proteins

Acetaminophen

in vivo

Amiodarone

Tetracycline

Carbon tetrachloride

24 h post-dose

1000 mg/kg

Control

1000 mg/kg

Acetaminophen

in vitro

2000 mg/kg

Amiodarone

Tetracycline

10 µM

500 µM

1 ml/kg Carbon tetrachloride

24 h post-dose

10 000 µM

Control

1000 µM

(% of total)

LDH

LGH leakage

Acetaminophen (in vitro 24 h incubation) 100

(% of control)h

WST-1 levels

LDH (EC50)

WST-1 (EC50)

Acetaminophen

3.8

4.2

Amiodarone

1.6

1.5

Tetracycline

1.0

1.6


2.4

2.5

0

150

WST-1

Log E50 of toxicants-exposure (in vitro 24 h incubation)

50

0 1 2 3 4 log x (concentration)

5

100 50 0

EC50 : Log EC50 (µM) 0 1 2 3 4 5 log x (concentration)

Figure 2.1 Morphological changes of rat liver and primary rat liver cell cultures treated with APAP, AD, TC and CTC. Log EC50 values calculated from the concentration–response curves from the LDH release data and the WST-1 data after exposure to APAP, AD, TC and CTC

60

Hepatotoxicity

and mitochondrial regulation-related proteins are useful as hepatotoxicity markers was also suggested. It is considered that future development of new biomarkers will help develop more predicable in vitro hepatotoxicity evaluation systems.

2.4 2.4.1

Application of Toxicopanomics Technologies to in vitro Safety Evaluation Systems Selection and Optimization of Evaluation Systems

Selection of cells (primary cells in culture, cell strains, tissue slices, derived tissues, etc.) and setting of culture conditions (monolayer culture, spheroid, co-culture, culture period, etc.) are important issues to be discussed in establishing in vitro evaluation systems. Panomics provide useful data to select and optimize the evaluation systems. For example, rat liver cells vary in gene expression, as shown below, depending, on cell types (primary cultured hepatocyte, BRL3A and NRLclone9 cell lines, or liver slices) and culture time (Boess et al., 2003). Among these cell types, liver slices showed the gene expression profiles closest to those of the rat liver, and primary liver cell cultures (monolayer culture or sandwich culture) also showed gene expression profiles relatively close to those of rat liver cells. However, the profiles of the cell lines used differed greatly from those of the rat liver (Figure 2.2).

Rat liver cell line (NRL) Rat liver cell line (BRL3A)

Gene expression profiles of the cell lines differed greatly from those of the rat liver

Rat liver Rat liver slice Rat primary liver cell (Monolayer) Rat primary liver cell (Sandwich)

Liver slices showed the gene expression profiles closest to those of the rat liver

Primary liver cell cultures (monolayer culture or sandwich culture) showed gene expression profiles relatively close to those of rat liver cells

Figure 2.2 Cluster analysis results for the similarity of gene expression. Modified from Boess F., Kamber M., Romer S., Gasser R., Muller D., Alberini S. and Suter L., ‘Gene expression in two hepatic cell lines, cultured primary hepatocytes and liver slices, compared to the in vivo gene expression in rats: possible implications for toxicogenomics use of in vitro systems’, Toxicological Sciences, 2003, 73(2), 386–402, by permission of Oxford University Press on behalf of the Society of Toxicology


61

The gene expression profiles of liver slices and primary cell cultures changed with time, more greatly in the early phase of culture. The expression of P450-related genes decreased with time. In addition, the expression of metabolic enzyme-related genes in primary cell cultures is reported that many of the Phase-I enzyme genes decrease with time while the Phase-II enzyme genes show various changes (Baker et al., 2001). Characteristically, liver slices showed an increase of the expression of inflammation-related genes, whereas primary cells in culture showed an increase of the expression of adaptation-related genes. Thus, it has been demonstrated at the gene level that the properties of cells differ between cell types, and change depending on culture period or culture conditions. It is also known that gene expression in cells exposed to compounds varies depending on the exposure period (Harries et al., 2001). Therefore, this information on gene expression, and information on protein expression as well, are considered to be useful to select an appropriate evaluation system and optimize the system. Of course, a single in vitro system is insufficient to reproduce or predict complicated phenomena occurring through networks in the relevant in vivo system. Using panomics data in establishing evaluation system to find the features of individual systems and combine them appropriately, is expected to lead to the establishment of a high-accuracy evaluation system. Furthermore, panomics data-based accurate estimation of adaptive changes or stress-induced changes intrinsic to individual systems is considered to lead to reliable toxicity evaluation. 2.4.2

Clarification of Toxicological Mechanisms and Development of Screening Systems

Even the same hepatotoxic substance affects cells differently, depending on the compounds. For example, when exposed to a known hepatotoxic substance, HepG2 cells or primary liver cell cultures show gene expression specific to the compound exposed, suggesting its association with toxicological mechanisms (Harries et al., 2001, Burczynski et al., 2000, Waring et al., 2001). Toxicopanomics provides comprehensive information on gene expression or protein expression, thus significantly contributing to the clarification of toxicological mechanisms triggered by compounds. Moreover, establishing a database which integrates various data on the existing toxic substances and performing bioinformatics-driven analysis can help target markers dependent on toxicological mechanisms. However, this process requires an enormous amount of toxicopanomics information, collected under comparable and reproducible conditions. Therefore, collecting in vitro evaluation data obtained according to a unified protocol serves as an effective measure to establish a toxicity prediction database. Then an in vitro screening system using newly identified markers as endpoints will be developed. 2.4.3

Improvement of Extrapolation by Developing Bridging Biomarkers

Accurate extrapolation between test systems or species is a challenge in the safety evaluation of compounds. In many cases, there are quantitative differences in dose–response relationship between humans and the experimental models, and in extreme cases, even qualitative differences in biological response are expected. This reminds us of the importance of developing ‘bridging biomarkers,’ which are used to correlate toxic reactions between

62

Hepatotoxicity Bridging biomarkers

Animal

Human

Bridging biomarkers

Bridging biomarkers

Human cells in culture

Animal cells in culture Bridging biomarkers

Figure 2.3

Use of bridging biomarkers to extrapolate from laboratory models to humans

test systems or species (Aardema and MacGregor, 2002) (Figure 2.3), and toxicopanomics is considered to be effective in developing new bridging biomarkers. With reliable bridging biomarkers, it is expected that the development of in vitro screening systems capable of accurately evaluating toxicity in humans or animals is feasible.

2.4.4

Improvement of Evaluation Sensitivity

Toxicity-related changes in gene expression are known to be often acuter than morphological changes (Corvi, 2002). To reduce the size and improve the throughput of an in vitro test system, new sensitive parameters must be specified in addition to the existing ones. Introduction of new sensitive parameters is also required to predict toxicity previously considered to be impossible to detect in in vitro systems. Toxicogenomics is expected to be an effective measure to meet these requirements. It is important to demonstrate marker genes and protein functions from the standpoint of toxicoproteomics, and gene functions must be established to clarify toxicological mechanisms based on genes targeted by toxicogenomics. One of the measures is to use siRNA (small interfering RNA), the effector for RNAi (RNA interference), which recently attracts attention for its medical applications. It is known that gene silencing, eventually control of protein biosynthesis, can be achieved by introducing siRNA to mammalian cells (e.g. HeLa cells) (Caplen et al., 2001, Elbashir et al., 2001). Using siRNA to ‘knock down’ specific genes paves the way for gene function analysis. Genes and proteins do not independently function, but form networks to closely cooperate in vital activities. Therefore, it is required to comprehensively detect and analyze the intracellular activities of genes and proteins. As indicated by nonlinear biological responses not always dependent on quantitative changes of proteins, not only the quantitative changes but the intracellular localization of proteins must be examined. Networks with different types of cells are often significant in the organism, and it is necessary to evaluate the development of functions via network formation. For this purpose, it is recommendable to use cell-based assay technologies (Beske and Goldbarg, 2002) or cell transfection array technologies (Ziauddin and Sabatini, 2001).


2.5 2.5.1

63

Introduction of Evaluation Systems Using in vitro Hepatotoxicity Tests in the Early Phase of Drug Discovery Examples of Phospholipidosis Evaluation

Phospholipidosis, a pathological condition with phospholipids excessively accumulated in cells, often impedes development in drug discovery. Examples of predicting induction of phospholipidosis based on the evaluation on an in vitro system and the physical properties are presented below. The in vitro experiment was performed based on a method reported by Ulrich et al. (1991), with modification. Primary rat cell cultures were added with compounds and a fluorescent-labeled lipid vesicle, and 24 h later, intake of the fluorescent dye into the cells was observed under the fluorescence microscope. Evaluation with ten compounds known for their ability to induce phospholipidosis and six ‘negative’ compounds yielded results correlating with their in vivo induction – thus, the usefulness of the in vitro system was confirmed (Figure 2.4). Development of ‘in-silico’ prediction systems is also given priority in studies to discover new drugs. A system using two physical property values (ClogP and net charge of a given molecule) is introduced for the prediction system of phospholipidosis (Tomizawa et al., 2006). It is proved that with the two parameters, the system can predict in vitro results with accuracy. 2.5.2

Examples of Biomarker Development

In the hepatotoxicity evaluation of compounds on an in vitro screening system, it is essential to set biomarkers well reflecting in vivo toxicity. Examples of developing hepatotoxicity biomarkers by means of proteomics (Yamamoto et al., 2005, 2006) and metabonomics technologies (Yamamoto et al., 2006) are explained below. The development of hepatotoxicity biomarkers by proteomics technologies used four compounds which induce hepatotoxicity via different mechanisms: acetaminophen (APAP), amiodarone (AMD), tetracycline (TC) and carbon tetrachloride (CTC). To investigate changes in protein expression in the rat liver, 1000 mg/kg of APAP, 1000 mg/kg of AMD, 2000 mg/kg of TC and 1 ml/kg of CTC were orally administered once to rats, and the liver was extracted 24 h later. To investigate changes in protein expression in primary rat liver cell cultures, 10 mM of APAP, 50μM of AMD and 500μM of TC were exposed to cells for 24 h. These samples were analyzed for changes in protein expression by 2-dimensional electrophoresis and mass spectrometry. In the rat liver, eight proteins whose expression commonly changed in response to the administration of the four compounds were identified (2-oxoisovalerate dehydrogenase, 60k Da ‘heat-shock’ protein, adenylate dehydrogenase-4, carbonic anhydrase-3, glutamate dehydrogenase, NADP-dependent malic enzyme, NDRG1-related protein NDRG2b1 and serotransferrin) (Figure 2.5(b)). In the primary rat liver cell cultures, three proteins whose expression commonly changed in response to the exposure to the three compounds were identified (glutathione peroxidase, peroxiredoxin 1, and peroxiredoxin) (Figure 2.5(a)). These proteins are expected to be used as hepatotoxicity biomarkers in in vivo or in vitro evaluation systems. Furthermore, the expression of three proteins (triosephosphate, glutamate dehydrogenase and beta-actin) in APAP, three proteins (fuructose-bisphosphate aldolase B, serotransferrin and short-chain

64 (a)

Hepatotoxicity DMSO (1 %, vehicle control)

Toluidine blue

EM, 1800×

EM, 1800×

Toluidine blue (b)

in vivo results Positive

Negative

Positive

True positive 10/10

False positive 0/10

Negative

False negative 0/10

True negative 6/6

in vitro results

Figure 2.4 (a) Fluorescence microscopy images, toluidine blue staining and transmission electron microscopy (TEM) images of primary rat liver cell cultures treated with DMSO (1 %, vehicle control) and amiodarone (2 μg/ml). The fluorescence-labeled lipid appears as fluorescent droplets surrounding the nuclei. Accumulation of dense inclusion bodies were induced by amiodarone (toluene blue staining, ‘black arrow’), while the inclusion bodies formed distinct lamellar structures (EM, ‘white arrow’: ×1800). (b) Correlations between in vitro and in vivo evaluation systems for phospholipidosis

3-hydroxyacyl-CoA dehydrogenase) in AMD, three proteins (plasma retinol-binding protein, serotransferrin and HMG-CoA synthase) in TC and thirteen proteins (elongation factor 1-γ , catalase, senescence marker protein-30, liver carboxylesterase 4, peroxiredoxin 2, d-dopachrome tautomerase, acyl-CoA dehydrogenase (long-chain specific), glutamate dehydrogenase, phenylalanine-4-hydroxylase, glycogen phosphorylase (liver form), pyruvate kinase isoenzymes R/L, hydroxymethylglutaryl-CoA synthase and glutathione peroxidase) in CTC changed commonly between in vivo and in vitro (Figure 2.5(c)). As reliable markers for in vivo hepatotoxicity, these proteins may be used in in vitro hepatotoxicity evaluation systems. In the development of hepatotoxicity biomarkers by metabonomics technologies, 1400 mg/kg of APAP was orally administered once to human hepatocyte transplanted chimeric

Utilization of an in vitro Hepatotoxicity Test (a)

APAP

(b)

APAP

11

12 CCl 4

1

3

21

6

65

4

10

1

0

3

8

0

8

2

1

44 27 AMD

TC

AMD

8

6 TC

(c)

APAP

TC

In vivo

In vitro 3

34

In vivo

17

In vitro 23

AMD

In vitro 3

24

CTC

In vivo 23

3

35

In vivo

In vitro 61

13

35

Figure 2.5 Venn diagrams showing the number of differentially expressed spots after treatment with APAP, AMD, TC and CTC: (a) protein expression changes in primary rat liver cell cultures; (b) protein expression changes in rat liver; (c) protein expression changes in primary rat liver cell culture and rat liver

mice, and intravesical urine and serum, collected 1, 4 and 24 h later, were used as samples. Intrinsic metabolites in the urine and serum samples were measured using 600 MHz NMR spectroscopy and metabolite changes were analyzed by spectral principal component analysis. The results showed changes in three metabolites in the urine (Figure 2.6) and serum. These metabolites had been produced by APAP administration and therefore may be used as biomarkers for hepatotoxicity induced by APAP. They may also be used as biomarkers for APAP-induced hepatotoxicity in humans because the animals used in the experiment were a ‘human model’.

2.6

Future Prospects

Focused on reduction in size and improvement of throughput, development of in vitro screening systems has progressed with the aim of collecting single endpoint information for many compounds, efficiently over a short period of time. Such a method can no doubt evaluate an aspect of toxicological profiles of a particular compound group highly efficiently. For high-quality evaluation in view of the complexity of toxicological mechanisms, however, multilateral evaluation tests must be performed for multiple intracellular parameters to be detected simultaneously. From this standpoint, cell-based assays, whose recent technological advancement is remarkable, will be more demanded to be applied to

66

Hepatotoxicity 0 mg/kg, 24 h 1400 mg/kg, 1 h 1400 mg/kg, 4 h

PCA score plot (3D)

1400 mg/kg, 24 h

Control PTE153-14 100 0 PTE153-4 CTE117-10

60

−100

−200

−300

200 150 60 100 50 0 −50 − −00 40

40

20

4h 20 PTE122-10 0 −20

CTE117-11

PTE120-11

PTE153-3 0

PTE120-3 PTE154-7 PTE154-13

PTE122-9 −20

−40

CTE101-4 24 h

−60 −100 −50 0 50 100 150 200 100

Figure 2.6

−40 PTE153-7 −60 0

−100

−200

−300

Urinary metabolite changes in APAP-administered mice

screening systems, as well as the toxicological mechanism clarification systems mentioned above. To clarify toxicological mechanisms in an in vitro system, technologies which can freely control the networks surrounding target cells, such as the cell–cell or cell–matrix communication environment, and technologies to comprehensively detect intracellular changes are required. For this purpose, development of new technologies by integrating cell-based assay systems with microfabrication techniques and even with nanotechnologies is desired. ‘On-chip’ single-cell analysis systems are considered to be an example of technologies representing this approach (Inoue et al., 2001). These systems are designed to control cell–cell communications under certain conditions and analyze the functions and relations of cells, which are contained in small culture spaces with 20–50 μm in diameter and 5–10 μm in depth on the chip surface. As comprehensive analyses of genes or proteins became available, bioinformatics (information-processing technologies) has long been expected to grow to interpret floods of data created by those analyses. Bioinformatics is currently under energetic study and development as technologies to select and compile data characteristic of specific life phenomena. However, trends toward the next generation have already been found, i.e. system biology,


67

a novel study area aimed to understand life phenomena as a system (Kitano, 2001), is attracting attention. With the progress of development and implementation of new technologies represented by panomics, understanding of components ‘constructing’ life, such as genes and proteins has rapidly advanced. System biology is intended to comprehend such information in the dynamics of life phenomena. Results from studies on system biology will assume important roles in developing simulation models (e.g. E-cell (Tomita et al., 1999)) and analyzing life behaviors (including pathological conditions). The area of toxicology will also greatly benefit from system biology, which is expected to grow into ‘system toxicology’. In studies on system biology, in vitro experiments are still an essential process. Cellbased assay systems capable of comprehensive analysis seem to be needed to demonstrate life phenomena (hypotheses) simulated by different methods. As described above, while development of in vitro safety evaluation systems is dramatically advancing, information has become ‘batch-processed’ with the advent of highthroughput and comprehensive analysis systems. In the future, even ‘in-silico systems’ simulating life phenomena and automatically analyzing life behaviors will be developed. Toxicologists can benefit from such novel technologies. Now that various new tools are available to researchers, their ability and sensitivity to achieve extensive, higher-quality toxicity evaluation utilizing those tools will be tested.

References Aardema M. J. and MacGregor J. T. (2002). Toxicology and genetic toxicology in the new era of ‘toxicogenomics’: impact of ‘-omics’ technologies. Mutat Res 499: 13–25. Baker T. K., Carfagna M. A., Gao H., Dow E. R., Li Q., Searfoss G. H. and Ryan T. P. (2001). Temporal gene expression analysis of monolayer cultured rat hepatocytes. Chem Res Toxicol 14: 1218–1231. Beske O. E. and Goldbard S. (2002). High-throughput cell analysis using multiplexed array technologies. Drug Discov Today 7: S131–S135. Boess F., Kamber M., Romer S., Gasser R., Muller D., Alberini S. and Suter L. (2003). Gene expression in two hepatic cell lines, cultured primary hepatocytes and liver slices compared to the in vivo liver gene expression in rats: possible implications for toxicogenomics use of in vitro systems. Toxicol Sci 73: 386–402. Burczynski M. E., McMillian M., Ciervo J., Li L., Parker J. B., Dunn II R. T., Hicken S., Farr S. and Johnson M. D. (2000). Toxicogenomics-based discrimination of toxic mechanism in HepG2 human hepatoma cells. Toxicol Sci 58: 399–415. Caplen N. J., Parrish S., Imani F., Fire A. and Morgan R. A. (2001). Specific inhibition of gene expression by small double-stranded RNAs in invertebrate and vertebtate systems. Proc Natl Acad Sci USA 98: 9742–9747. Corvi R. (2002). Genomics: an in vitro toxicology point of view. ATLA 30: 129–131. Elbashir S. M., Harborth J., Lendeckel W., Yalcin A., Weber K. and Tuschl T. (2001). Duplexes of 21-nucleotide RNAs mediate RNA interference in cultured mammalian cells. Nature 411: 494–498. Farkas D. and Tannenbaum S. R. (2005). In vitro methods to study chemically induced hepatotoxicoty: a literature review. Curr Drug Metab 6: 111–125. Groneberg D. A., Grosse-Siestrup C. and Fischer A. (2002). In Vitro Models to Study Hepatotoxicity. Toxicol Pathol 30: 394–399. Harries H. M., Fletcher S. T., Duggan C. M. and Baker VA. (2001). The use of genomics technology to investigate gene expression changes in cultured human liver cells. Toxicol in vitro 15: 399–405.

68

Hepatotoxicity

Inoue I., Wakamoto Y., Moriguchi H., Okano K. and Yasuda K. (2001). On-chip culture system for observation of isolated individual cells. Lab Chip 1: 50–55. Kikkawa R., Yamamoto T., Fukushima T., Yamada H. and Horii I. (2005). Investigation of a hepatotoxicity screening system in primary cell cultures – ‘What biomarkers would need to be addressed to estimate toxicity in conventional and new approaches?’. J Toxicol Sci 30: 61–72. Kikkawa R., Fujikawa M., Yamamoto T., Hamada Y., Yamada H. and Horii I. (2006). In vivo hepatotoxicity study of rats in comparison with in vitro hepatotoxicity screening system. J Toxicol Sci 31: 23–34. Kitano H. (2002). System biology: a brief overview. Science 295: 1662–1664. Pennie W. D., Tugwood J. D., Oliver G. J. A. and Kimber I. (2000). The principles and practice of toxicogenomics: Application and opportunities. Toxicol Sci 54: 277–283. Stubberfield C. R. and Page M. J. (1999). Applying proteomics to drug discovery. Exp Opin Invest Drugs 8: 65–70. Suter L., Babiss L. E. and Wheeldon E. B. (2004). Toxigenomics in predictive toxicology in drug development. Chem Biol 11: 161–171. Tateno C., Yoshizane Y., Saito N., Kataoka M., Utoh R., Yamasaki C., Tachibana A., Soeno Y., Asahina K., Hino H., Asahara T., Yokoi T. and Furukawa T. and Yoshizato K. (2004). Near Completely Humanized Liver in Mice Shows Human-Type Metabolic Responses to Drugs. Am J Pathol 165: 901–912. Tomita M., Hashimoto K., Takahashi K., Shimizu T. S., Matsuzaki Y., Miyoshi F., Saito K., Tanida S., Yugi K., Venter J. C. and Hutchison III C. A. (1999). E-CELL: software environment for whole-cell simulation. Bioinformatics 15: 72–84. Tomizawa K., Sugano K., Yamada H. and Horii I. (2006). Physicochemical and cell-based approach for early screening of phospholipidosis inducing potential. J Toxicol Sci 31 r r r– r r r r. Ulrich R. G., Kilgore K. S., Sun E. L., Cramer C. T. and Ginsberg L. C. 1991. An in vitro fluorescence assay for the detection of drug-induced cytoplasmic lamellar bodies Toxicol Meth 1: 89–105. Waring J. F., Ciurlionis R., Jolly R. A., Heindel M. and Ulrich R. G. (2001). Microarray analysis of hepatotoxins in vitro reveals a correlation between gene expression profiles and mechanisms of toxicity. Toxicol Lett 120: 359–368. Yamamoto T., Kikkawa R., Yamada H. and Horii I. (2005). Identification of oxidative stress-related proteins for predictive screening of hepatotoxicity using a proteomic approach. J Toxicol Sci 30: 213–227. Yamamoto T., Kikkawa R., Yamada H. and Horii I. (2006). Investigation of proteomic biomarkers in vivo hepatotoxicity study of rat liver: Toxicityy differentiation in hepatotoxicants. J Toxicol Sci 31: 49–60. Ziauddin J. and Sabatini D. M. (2001). Microarrays of cells expressing defined cDNAs. Nature 411: 107–110.

3 Use of Hepatocytes for Characterizing a Candidate Drug’s Metabolism and Drug Interaction Potential Srikanth C. Nallani, John M. Strong and Shiew Mei Huang

3.1

Introduction

Drug-induced liver injury (DILI) is a major health problem which can lead to liver failure resulting in liver transplants or death. It is the most frequent reason for withdrawal of drugs from the pharmaceutical market (Temple and Himmel, 2002) and for labeling drugs with a ‘black box warning’ (Fung et al., 2001). Based on the seriousness of this problem, the FDA recently formed a working group along with representatives from the Pharmaceutical and Research Manufacturers of America (PhRMA) the American Association for the Study of Liver Disease (AASLD) and the National Institute of Diabetes and Digestive and Kidney Diseases to provide a forum to discuss pre-clinical, clinical and post-marketing approaches related to minimizing the risk of DILI. The working group initially sponsored a two-day Heptatoxicity Workshop in February 2001. Since that time the working group has met annually to discuss progress towards realizing its goal of minimizing DILI. The FDA working group maintains a website at www.fda.gov/cder/livertox/, which includes preclinical, clinical, and post-marketing white papers and presentations by participants at the working group meetings. Although a few drugs are excreted unchanged into the bile or urine, most drugs are cleared from the body partially or extensively by drug metabolizing enzymes found mainly in the liver or gut mucosa. Drug metabolism can proceed through detoxification pathways leading to elimination of the drug from the body, or in some cases through intoxification pathways resulting in toxicity (Figure 3.1). Hepatotoxicity: From Genomics to in vitro and in vivo Models Edited by S. C. Sahu C 2007 John Wiley & Sons, Ltd

70

Hepatotoxicity Drug Detoxification pathways

Elimination

Figure 3.1

Intoxification pathways

Toxicity

Detoxification pathways

Metabolic pathways leading to drug elimination from the body or to hepatotoxicity

The balance between these two pathways in humans can be affected by numerous factors, such as genetic polymorphisms of drug-metabolizing enzymes, environmental, disease state, sex, age and drug–drug interactions. Although some of these factors can be identified and are relatively stable with time, others such as concomitant medications can alter metabolism abruptly through drug-interactions such as inhibition or induction of drugmetabolizing enzymes and are of particular concern. Drug–drug interactions can lead to changed systemic exposure, to the co-administered drugs or their metabolites, resulting in variations in drug response, including adverse drug reactions and efficacy. The relationship between drug metabolism and adverse events, including hepatotoxicity, cannot be overemphasized. This is illustrated by the fact that the FDA has recently published a number of Guidances for Industry addressing this issue (e.g. ‘Guidance for Industry: Drug Interaction Studies – Study Design, Data Analysis and Imapct for Dosing and Labeling’ (FDA, 2006) and ‘Guidance for Industry: Safety Testing of Drug Metabolites’ (FDA, 2005) and a website containing information on drug–drug interactions (http://internet-dev/ cder/drug/drugInteractions/). It is obviously most desirable that the hepatotoxic potential of drugs be discovered as early as possible during pre-clinical screening and specific human testing. When developing strategies to deal with metabolites and their potential role in hepatotoxicity, Guenguerich (2006), Smith and Obach (2005) and Liebler and Guengerich (2005), respectively, have proposed similar definitions of mechanistic categories of toxicity that could be caused by the parent drug or metabolites which should be considered. Their general classifications include: r parent drug or slightly modified metabolite that elicits its effect on the intended target; r parent drug or metabolite that binds to a receptor or enzyme that is not the primary target; r parent drug or reactive metabolite covalently binds to protein that lead to autoimmune responses; r drug bioactivated to reactive compounds that can cause oxidative stress through redox cycling or reactive metabolites, usually electrophiles, that can covalently bind to critical cellular proteins leading to cell death. Traditionally, drug metabolites in general have not been routinely evaluated in crossspecies safety assessments because their specific contribution to the overall toxicological potential of the parent drug has been unknown. In the past, measurements of circulating concentrations of a parent drug in animals were used as an index of systemic exposure in

Hepatocytes and Metabolism and Drug Interactions

71

humans. Animal toxicology studies are an important component of assessing drug safety for subsequent human exposure; however, recently there has been a clear trend towards decreased reliance totally upon animal studies and increased emphasis upon experiments with human-derived tissues.

3.2

Metabolic Pathway Identification in Animals and Humans in vitro

Studies in laboratory animals do not always reflect human drug toxicity because of interspecies differences. It has been reported that hepatotoxicity has one of the poorest correlation with regulatory animal toxicity tests (Olson et al., 2000) and that only approximately half of the new pharmaceuticals that produced hepatotoxicity in clinical drug development were detected during animal toxicity studies (Olson et al., 2000, 1998). Metabolites may not be adequately assessed during standard nonclinical studies because they occur only in humans (unique metabolite) or they are produced at much higher levels than in animals. The Agency recommends that attempts be made to identify as early as possible during the drugdevelopment process differences in drug metabolism in animals used in nonclinical safety assessments compared to humans. (FDA, 2005). Studies using human and animal liver tissue, including freshly isolated liver slices or hepatocytes and cryopreserved hepatocytes, to identify a drug’s metabolic pathways in both species are generally conducted before initiation of clinical trials. These intact cellular models contain the complete complement of liver enzymes and cofactors as will as the natural orientation for linked enzymes. This type of study can provide information to prospectively design experiments so that the animal species most similar to humans can be chosen, on a case-by-case basis, for each drug and help in the interpretation of data obtained from animal toxicology experiments. The FDA recommends that toxicological testing in animals should be performed for unique human metabolites and those that represent >10 % of the dose (FDA, 2005).

3.3

Characterizing a Candidate Drug’s Metabolic Pathways Using Human Hepatocytes

Identification of molecules with acceptable metabolic profiles early in the drug-development process is crucial for the success of drug-candidate development. Prediction of in vivo pharmacokinetics before human drug administration based on various in vitro systems is a common practice in the pharmaceutical industry. These systems include human liver slices, freshly isolated, cryopreserved hepatocytes and hepatocytes in culture, and subcellular liver tissue, such as microsomes or post-mitochondrial fractions (S9). Freshly prepared or cryopreserved hepatocytes in suspension are the most commonly used tissue for characterizing a drug’s metabolic pathways; however, due to the loss of cell viability, these experiments are limited to 4–6 h (Gomez-Lechon et al., 2004). Hepatocyte cultures can be beneficial for investigating the metabolic stability of drugs over longer periods of time to investigate secondary, tertiary, etc. metabolites which may not be detected in short-term incubations (Zhao et al., 2005). Hepatocytes in culture are also the method of choice for CYP enzyme induction studies.

72

Hepatotoxicity

It can be reasonably considered that human hepatocytes reflect the heterogeneity of CYP (CYPs 1A2, CY2A6, 2B6, 2C8, 2C9, 2C19, 2D6, 2E1 and 3A4) expression in human liver. The relative contribution of specific CYP activities in human hepatocytes and human liver microsomes was discussed previously (Gomez-Lechon et al., 2004). The good reflection of hepatic CYP enzyme pattern of hepatocytes as compared to human microsomes constitutes an invaluable tool for preclinical metabolic and toxicity testing of candidate drugs. Approaches for studying other non-CYP enzymes, such as flavin-containing monooxygenase (FMOs), monamine oxidases (MAOs), molybdenum-containing oxidases (Mo–CO), and transferase reactions such as UDP-glucuronosyl-transferases (UGTS), and sulfotransferases (STs), methyltransferases, acetyltransferases and glutathione transferases are evolving. For example, Bjorhnsson et al. (2003) discusses methods to differentiate between CYP and non-CYP oxidative metabolites. Metabolites formed from transferase reactions can be identified by their molecular structure. 3.3.1

Hepatocyte Characteristics and Considerations for Experimental Conditions

Hepatocyte cultures in vitro are known for cell ‘de-differentiation’ which is accompanied by the loss of some specific enzymes, especially cytochrome P450 monooxygenases, affecting the metabolic capacity of the system (Morel et al., 1990; Skett et al., 1995). The functional relevance of an in vivo-like extracellular matrix geometry for the biotransformation of primary hepatocytes in vitro was studied by using a variety of drugs (Bader et al., 1995). Tissue architecture and parenchymal cell morphology reflect the functional differentiation of the liver. Hepatocytes are arranged as monolayer plates in vivo, an organization which allows the cells to be highly vascularized. Sandwiching hepatocytes between two layers of collagen has shown to provide the cells with an in vivo-like environment that enables the cells to preserve their specific functionality for several days (Berthiaume et al., 1996). This observation was further substantiated by Kern et al. (1997) who showed that hepatocyte sandwich cultures not only retained their long-term viability but also their drug-metabolizing capacity. Hepatocytes in suspension offer a simple and cost-effective method to study a drug’s metabolic stability and to identify its overall metabolic pathways and can be automated for screening purposes during early drug development They can be procured from several commercial sources which include cell viability data and instructions for performing drug incubations with their product. Cryopreserved hepatocytes are particularly attractive because they are readily available as ‘off the shelf items’ and one does not have to wait for a donor liver to perform their study; however, one should be aware that drug-metabolizing CYP, glutathione-S-transferase and sulfotransferase enzyme activity has been reported to be less when compared to freshly isolated hepatocytes. On the other hand, UDP-glucoronosyl transferase activity is unaltered by cryopreservation (Hengstler et al., 2000). Damage to the cell membrane while thawing cryopreserved hepatocytes and the resulting reduction in the levels of cofactors such as NADPH might partially explain the loss of CYP enzyme activity. Some commonly used experimental conditions for these types of hepatocyte short term experiments are listed in Table 3.1. Following incubation, the samples can be analyzed for metabolites in the medium and cells separately or following disruption of the cells as the total metabolic content in the incubation. The most preferred method of analysis is by liquid chromatography combined with mass spectrometry/mass spectrometry (LC–MS/MS), a highly sensitive and specific technique which can provide both qualitative structural information and quantification


73

Table 3.1 Typical experimental conditions for metabolic characterization of drug candidates in hepatocytes Experimental condition Incubation time and medium Hepatocyte Concentration Incubation vial

Incubation volume Incubation conditions

Current practice and rationale For ease of operation, reactions for ≤ 3 h are conducted in balanced salt solution; ≤ 6 h in cell culture medium 1–5 × 106 viable cells/ml provide optimal cell viability and metabolic capacity, low cellular aggregation For qualitative and quantitative metabolic stability, reactions can be carried out in varied capacity: –centrifuge (glass) tubes ( 25 % of a drug’s clearance, studies to identify drug-metabolizing CYP enzyme in vitro are recommended. These latter studies investigate the individual CYP enzymes responsible for each of the metabolites formed and identified using human hepatocytes. These studies use a combination of techniques including human liver tissues, such as post-mitochondrial fraction (S9), cytosolic fraction and microsomes. A detailed discussion of the methods and techniques used for the latter studies are beyond the scope of this chapter and the reader is referred to the following reference for the further information (Bjornsson et al., 2003).

3.4

Prediction of in vivo Clearance of Candidate Drugs

The metabolic stability data obtained from the human hepatocyte studies described in the previous section provides important information if performed in the early stages of drug

1–14 52–94

Midazolam 1-hydroxylation

Testosterone 6β-hydroxylation

3A4/5

a

P-Nitrophenol 3-hydroxylation Lauric acid 11-hydroxylation Aniline 4-hydroxylation Erythromycin N-demethylation Dextromethorphan N-demethylation Triazolam 4-hydroxylation Terfenadine C-hydroxylation Nifedipine oxidation

Omeprazole 5-hydroxylation Fluoxetine O-dealkylation Debrisoquine 4-hydroxylation

7-ethoxyresorufin-O-deethylation theophylline-N-demethylation caffeine-3-N-demethylation tacrine 1-hydroxylation — — Propofol hydroxylation S-mephenytoin-N-demethylation Amodiaquine N-deethylation Rosiglitazone para-hydroxylation Flurbiprofen 4’-hydroxylation Phenytoin-4-hydroxylation

Note that this is not an exhaustive list. For an updated list, see the following link: http://www.fda.gov/cder/drug/drugInteractions/default.htm

2E1

9–15 0.44–8.5 39–157

(±)-Bufuralol 1’-hydroxylation Dextromethorphan O-demethylation Chlorzoxazone 6-hydroxylation

67–838 1.5–4.5 3.4–52 13–35

0.30–2.3 13–162 17–23 67–168 5.4–19

1.7–152

2D6

2C19

2C9

2C8

Tolbutamide methyl-hydroxylation S-warfarin 7-hydroxylation Diclofenac 4’-hydroxylation S-mephenytoin 4’-hydroxylation

Coumarin-7-hydroxylation Nicotine C-oxidation Efavirenz hydroxylase Bupropion-hydroxylation Taxol 6-hydroxylation

2A6

2B6

Phenacetin-O-dethylation

1A2

Substrate acceptable

Substrate preferred

CYP

Km (μM)

Preferred and acceptable chemical substrates for in vitro experimentsa

Table 3.2

3.3 130 6.3–24 33–88 133–710 234 15 5.1–47

17–26 3.7–104 5.6

0.18–0.21 280–1230 220–1565 2.8, 16 — — 3.7–94 1910 2.4 4.3–7.7 6–42 11.5–117

Km (μM)

74 Hepatotoxicity


75

development. The ability to classify the metabolic clearance of a drug as low, intermediate or high is useful when prioritizing further development of candidate drugs. Once the individual metabolites formed by human hepatocytes are identified and the contribution of each metabolite to the overall in vitro clearance of the drug is determined, the total intrinsic drug clearance in vitro, CL,i, is obtained by using the following equation: CL, i =

j n Vmax j

j=1

Km

(3.1)

where Vmax is the maximal velocity of the metabolic reaction, K m is the Michaelis–Menten constant and j represents the number of metabolites quantified in the study. Assuming that the clearance of the drug is independent of the blood flow rate through the clearing organ, in this case the liver, one can predict the in vivo intrinsic clearance of the drug using the simple well-stirred model (Pang and Rowland, 1977) by employing the following equation: CLin vivo =

fuCL, iQ H fuCL, i + Q H

(3.2)

where fu = fraction unbound in plasma and Q H = hepatic blood flow (Davies and Morris, 1993). Scaling of the in vitro intrinsic clearances, CL, i, for calculation of CLin vivo can be accomplished by assuming that approximately 120 million cells are found in 1 g of liver and that the total liver weight is about 25 gm/kg of body weight with an average blood flow of about 21 ml/min/kg (Davies and Morris, 1993). In the past, human liver microsomes or the S9 fractions have been used for this purpose of metabolic stability evaluation. Recently the use of fresh or cryopreserved hepatocytes is gaining acceptance. For example, clearance of substrates by fresh or cryopreserved human hepatocytes via CYP enzyme and UGT transferase reactions in vitro have been used to predict the CLin vivo for these enzymes (Soars et al., 2002). The extent of ‘underprediction’ of CLin vivo was shown to be less that those determined using human liver microsomes. In another study designed to establish the effect of product inhibition of CYP enzymes by a candidate drug’s metabolites, the results indicated that microsomes predicted a lower C L in vivo compared to that predicted using heptatocytes (Naritomi et al., 2003).

3.5

Characterizing Drug Metabolizing Enzyme Iinduction-Potential of Candidate Drug

The average percentage of individual CYP enzymes have been characterized in human liver and hepatocytes by quantifying CYP mRNA expression, protein and enzyme activity. The basal expression and activity of CYP and non-CYP drug-metabolizing enzymes is highly variable between batches of hepatocytes isolated from different donors. In addition, the expression of individual enzymes decreases with time at variable rates during incubation of primary cultures of hepatocytes. 3.5.1

Regulation of Drug Metabolizing Enzyme Expression in Hepatocytes

Many of the drug-metabolizing enzymes and drug transporters can be induced after exposure to specific xenobiotics. Over the past decade, our understanding of mechanisms that

76

Hepatotoxicity

regulate the gene expression of these drug-metabolizing enzymes and transporters in hepatocytes has greatly expanded. Nuclear receptors, including the aryl hydrocarbon receptor (AhR), pregnane-X-receptor (PXR) or steroid and xenobiotic receptor (SXR), the constitutive androstane receptor (CAR) and peroxisome proliferator activated receptor (PPAR), play a key role in this process (Kliewer et al., 1999). Table 3.3 presents a partial list of drugs (and chemicals) that interact with nuclear receptors expressed in different organs

Table 3.3 A partial list of drugs (and chemicals) that interact with nuclear receptors expressed in different organs known to regulate the expression of target drug metabolizing enzyme and drug transporter genes (Tirona and Kim, 2005; Klaassen and Slitt, 2005; Chang and Waxman, 2006) Nuclear receptor (NR)

Tissuerxpressing NR

Target gene of the NR

β-Naphtoflavone, 2,3,7,8tetrachlorodibenzo-pdioxin

Aryl hydrocarbon receptor (AhR)

Acetaminophen, Phenobarbital, phenytoin, TCPOBOP

Constitutive androstane receptor (CAR)

Lung Thymus Kidney Liver Liver

Clofibrate, ciprofibrate

Peroxisome proliferator-activated receptor α (PPARα)

Liver Heart Muscle Kidney

Clotrimazole, dexamethasone, paclitaxel, phenobarbital, ritonavir, rifampicin

Pregnane X receptor (PXR)

Liver Small intestine Colon Lymphocytes

CYP1A1 CYP1A2 UGT1A1 UGT1A6 CYP2A6 CYP2B6 CYP2C9 CY2C19 CYP3A4 Sult1a1 Sult2a1 Sult2a9 UGT1A1 Mrp2 Mrp3 Mrp5 Mdr1 CYP4A UGT1A9 UGT2B4 Mdr2 Oatp1 CYP1A2 CYP2B6 CYP2C9 CYP2C19 CYP3A4 CYP3A7 SULT2A1 UGT1A1 UGT1A3 UGT1A4 Mdr1 Mrp2 Mrp3

Drug/chemical


77

known to regulate the expression of target drug metabolizing enzyme and drug transporter genes. It is important to note that in some cases ‘cross-reactivity’ exists between individual nuclear receptors. For example, phenobarbital induces CYP2B and CYP3A mainly through activation of CAR, whereas dexamethasone and rifampin induce CYP2C and CYP3A enzymes by means of their interaction with PXR (Moore et al., 2000). UDP-glucuronosyl transferases and sulfotransferases are involved in the detoxification and ultimate excretion of a wide variety of structurally diverse endo- and xenobiotics, including many therapeutic agents and endogenous steroids. Emerging evidence strongly suggests that the nuclear receptors PXR, CAR, and PPAR can also regulate these transferases (Sugatani et al., 2005; Runge-Morris and Kocarek, 2005). In general, drug transporters such as the ‘organic anion transporting polypeptide family’ (OATPs) along with ‘organic cation transporter 1’ (OCT1) and ‘organic anion transporter 2’ (OAT2) mediate uptake of a large number of xenoblotics from blood into liver. Conversely, ‘multi-drug resistance proteins’ (MDRs), ‘multi-drug resistance associated proteins’ (MRPs) and the ‘breast cancer resistance protein’ (BCRP) mediate efflux of xenobiotics from liver into bile or blood. The nuclear receptors PXR and CAR have been shown to be involved in the regulation of these transporters (Tirona and Kim, 2005). Coordinate regulation of Phase I and Phase II enzyme induction, along with drug transporter induction, has been reported with the treatment of prototypical enzyme inducers. Taken together, induction of the drug metabolism and drug transporting enzymes in a coordinated fashion provides an important means to protect the body from xenobiotics ‘insults’. A drug that induces a specific drug-metabolizing enzyme can increase the rate of metabolic clearance of a co-administered drug that is a substrate of the induced pathway, resulting in sub-therapeutic blood concentrations. Alternatively, the induced metabolic pathway could lead to increased formation of a bioactivated compound with the potential to produce liver toxicity, e.g. acetaminophen conversion to the putative toxin, N-acetyl-pbenzoquinoneimine. 3.5.2

Experimental Considerations for Evaluating CYP Induction in Human Hepatocytes

Presently, the most reliable method to study a drug’s induction potential is accomplished by using freshly isolated human hepatocytes in culture or cryopreserved hepatocytes that are capable of attachment and incubation in culture. An overview of a typical procedure for conducting an induction experiment is described below; however, more detailed information can be found in a review article (LeCluyse, 2001). Initially, hepatocytes at a density of 100 000–200 000 cells/cm2 are placed on multiwell plates coated with a matrix such as r collagen or Matrigel and allowed to attach for 2–4 h. It is recommended that monolayer cultures or cells with medium are replaced and overlayed (sandwiched) with collagen or r Matrigel to provide a 3-dimensional environment and to minimize variability in cellular integrity. Hepatocytes are then incubated for 2–3 days with medium replacement every 24 h in order to stabilize the hepatocytes. The incubation medium is not critical and several products, such as Chee’s Medium, William’s medium E, Ham’s and F-12/Lebovitz L-15 (1:1), are commercially available. Customized media that include the following supplements, 6 μg/ml insulin, 6 μg/ml transferrin, 6 ng/ml selenium, 2 mg/ml BSA/linoleic acid complex and a glucocorticoid such as dexamethasone (25–100 ng/ml excessive dexamethasone

78

Hepatotoxicity

Table 3.4

Chemical inducers for in vitro experimentsa

CYP Inducer—preferred 1A2

Omeprazole ß-Naphthoflavone 3-Methylcholanthrene

2A6

Dexamethasone

2B6

Phenobarbital

2C8 2C9

Inducer concentrations Fold Inducer – (μM) induction acceptable 25–100 33–50 1, 2

14–24 4–23 6–26

Lansoprazole

Pyrazole

Inducer concentration Fold (μM) induction 10

10

50

9.4

1000

7.7

500–1000

5–10

Phenytoin

50

5–10

Rifampin

10

2–4

Phenobarbital

500

2–3

Rifampin

10

3.7

Phenobarbital

100

2.6

2C19 Rifampin

10

20

—

—

—

2D6 None identified

—

—

—

—

—

2E1

None identified

—

—

—

—

—

3A4

Rifampin

10–50

4–31

phenobarbital Phenytoin Rifapentine Troglitazone Taxol Dexamethasone

100–2000 50 50 10–75 4 33–250

3–31 12.5 9.3 7 5.2 2.9–6.9

a

Note that this is not an exhaustive list. For an updated list, see the following link: [http://www.fda.gov/cder/drug/ drugInteractions/default.htm].

can induce enzymes), should be sufficient to ensure maintenance of monolayer integrity and hepatocyte morphology for at least 1 week and give reliable results with positive controls. Additional additives, specific for sugar metabolism, amino acids, antioxidants, hormones and heavy metals, are also acceptable in an attempt to improve cellular maintenance. After the hepatocytes have stabilized, the wells containing hepatocytes are divided into three sets and incubated for 2–3 days with medium replacement every 24 h in the presence of (set 1) test drug, (set 2) a known inducer drug listed in Table 3.4 as a positive control and (set 3) no drug and used as a negative control. Inducer-drug concentrations should be utilized, based on the expected human plasma drug concentrations. At least three test drug concentrations spanning the therapeutic range should be studied. If this information is not available, concentrations ranging over at least three orders of magnitude should be studied since the induction response may not be linear with concentration. To assess the integrity of the hepatocyte culture following drug treatment, the viability of hepatocytes should be assessed by visual inspection or by cell-viability assays such as MTT or MTS assays, LDH assay, ATP assay, etc. Following treatment of the three sets of hepatocytes, the medium is replaced and a test probe substrate (Table 3.2), is incubated with the hepatocytes or microsomes prepared from the individual hepatocytes incubations to determine the relative enzyme activity of untreated and treated cells. When conducting the experiments to determine enzyme activity, the following experimental conditions are relevant: microsomal protein concentrations, if used, usually range from 0.01 to 2.0 mg/ml. Typically, the incubation time ranges from 5 to


79

30 min. Preferably, pioglitazone doses. 3.7.1

Screening for Drug-Induced Hepatotoxicity in Cells in vitro

There is considerable evidence that reactive metabolites are formed from drugs known to cause liver toxicity (Walgren et al., 2005) and tools are now readily available to screen for reactive metabolites in vitro and in humans in vivo; however, the mechanism by which these toxic species initiate and propagate cellular injury is poorly understood (Park et al., 2005; Zhou et al., 2005). Recently, there has been a major focus on developing cellular models in vitro and sensitive and specific biomarkers in vivo that can predict the risk for drug-induced liver toxicity. Hepatoxicity can potentially result from cellular interactions of the parent drug or one of its metabolic products and drug–drug interactions can modify human exposure to these molecular entities. Therefore, any cellular model in vitro used to screen for drug-induced hepatoxicity should express the full complement of metabolizing enzymes. Several investigators have used immortalized cells including a human hepatoma cell line (HepG2) (Wilkening et al., 2003; Vignati et al., 2005) and a human hepatocyte derived cell line (Fa2N-4) (Ripp et al., 2006) to screen for hepatotoxins; however, the activity and expression, especially CYP enzymes, have been demonstrated to be extremely low (Rodriguez-Antona et al., 2002; Jover et al., 2001).

84

Hepatotoxicity

Some investigators have used various cell models in vitro to overcome this deficiency, for example, Vignati et al. (2005) incubated drugs in combination with HepG2 cells in combination with microsomes containing cDNA-expressed CYP 3A, while Dai and Cederbaum (1995) demonstrated the cytotoxicity of acetaminophen in human CYP2E1-transfected HepG2 cells.

3.8

Conclusions

Screening drug candidates early during the drug-development process for their potential to interact with other drugs and/or to produce bioactivation toxic metabolites resulting in hepatotoxicity is of high priority for the pharmaceutical industry. Further research efforts are underway to develop reliable models in vitro to minimize the risk for drug-induced hepatotoxicity. Currently, human hepatocytes are the preferred in vitro model to investigate the total biotransformation and interaction potential of a drug in human liver and to identify a drug’s potential to produce toxicity in vitro.

References Alvarez-Sanchez R., Montavon F., Hartung T. and Pahler A. (2006). Thiazolidinedione bioactivation: a comparison of the bioactivation potentials of troglitazone, rosiglitazone and pioglitazone using stable isotope-labeled analogues and liquid chromatography tandem mass spectrometry. Chem Res Toxicol 19: 1106–1116. Bader A., Knop E., Boker K., Fruhauf N., Schuttler W., Oldhafer K., Burkhard R., Pichlmayr R. and Sewing K. F. (1995). A novel bioreactor design for in vitro reconstruction of in vivo liver characteristics. Artif Organs 19: 368–374. Baughman T. M., Graham R. A., Wells-Knecht K., Silver I. S., Tyler L. O., Wells-Knecht M. and Zhao Z. (2005). Metabolic activation of pioglitazone identified from rat and human liver microsomes and freshly isolated hepatocytes. Drug Metab Dispos 33: 733–738. Berthiaume F., Moghe P. V., Toner M. and Yarmush M. L. (1996). Effect of extracellular matrix topology on cell structure, function, and physiological responsiveness: hepatocytes cultured in a sandwich configuration. FASEB J 10: 1471–1484. Bjornsson T. D., Callaghan J. T., Einolf H. J., Fischer V., Gan L., Grimm S., Kao J., King S. P., Miwa G., Ni L., Kumar G., McLeod J., Obach R. S., Roberts S., Roe A., Shah A., Snikeris F., Sullivan J. T., Tweedie D., Vega J. M., Walsh J. and Wrighton S. A. (2003). The conduct of in vitro and in vivo drug-drug interaction studies: a Pharmaceutical Research and Manufacturers of America (PhRMA) perspective. Drug Metab Dispos 31: 815–832. Chang T. K. and Waxman D. J. (2006). Synthetic drugs and natural products as modulators of constitutive androstane receptor (CAR) and pregnane X receptor (PXR). Drug Metab Rev 38: 51–73. Dai Y. and Cederbaum A. I. (1995). Cytotoxicity of acetaminophen in human cytochrome P4502E1transfected HepG2 cells. J Pharmacol Exp Ther 273: 1497–1505. Davern T. J., James L. P., Hinson J. A., Polson J., Larson A. M., Fontana R. J., Lalani E., Munoz S., Shakil A. O. and Lee W. M. (2006). Measurement of serum acetaminophen–protein adducts in patients with acute liver failure. Gastroenterology 130: 687–694. David J. P. (2005). The molecular toxicology of acetaminophen. Drug Metab Rev 37: 581–594.


85

Davies B. and Morris T. (1993). Physiological parameters in laboratory animals and humans. Pharm Res 10: 1093–1095. Dieckhaus C. M., Thompson C. D., Roller S. G. and Macdonald T. L. (2002). Mechanisms of idiosyncratic drug reactions: the case of felbamate. Chem Biol Interact 142: 99–117. Dieckhaus C. M., Fernandez-Metzler C. L., King R., Krolikowski P. H., and Baillie T. A. (2005). Negative ion tandem mass spectrometry for the detection of glutathione conjugates. Chem Res Toxicol 18: 630–638. Evans D. C., Watt A. P., Nicoll-Griffith D. A. and Baillie T. A. (2004). Drug–protein adducts: an industry perspective on minimizing the potential for drug bioactivation in drug discovery and development. Chem Res Toxicol 17: 3–16. FDA (2000). ‘Medwatch’, Food and Drugs Administration (FDA), Silver Spring, MD, USA [http://www.fda.gov/medwatch/safety/2000/safety00.htm#rezuli]. FDA (2005). ‘Guidance for Industry: Safety Testing of Drug Metabolites’ Food and Drug Administration (FDA), Silver Spring, MD, USA [http://www.fda.gov/cder/guidance/6366dft.pdf]. FDA (2006). FDA Guidance for Industry: Drug Interaction Studies – Study Design, Data Analysis and Impact on Dosing and Labeling’. Food and Drug Administration (FDA), Silver Spring, MD, USA [http://www.fda.gov/cder/guidance/6695dft.pdf]. Fung M., Thornton A., Mybeck K., Wu J. H., Hornbuckle H. and Muniz E. (2001). Evaluation of the characterisitics of safety withdrawal of prescription drugs from worldwide pharmaceutical markets – 1960 to 1999. Drug Inform J 35: 293–317. Gomez-Lechon M. J., Donato M. T., Castell J. V. and Jover R. (2004). Human hepatocytes in primary culture: the choice to investigate drug metabolism in man. Curr Drug Metab 5: 443–462. Guengerich F. P. (2006). Cytochrome P450s and other enzymes in drug metabolism and toxicity. AAPS J 8: E101–E111. Hengstler J. G., Utesch D., Steinberg P., Platt K. L., Diener B., Ringel M., Swales N., Fischer T., Biefang K., Gerl M., Bottger T., Oesch F. (2000). Cryopreserved primary hepatocytes as a constantly available in vitro model for the evaluation of human and animal drug metabolism and enzyme induction. Drug Metab Rev 32: 81–118. Isley W. L. (2003). Hepatotoxicity of thiazolidinediones. Expert Opin Drug Saf 2: 581–586. Jover R., Bort R., Gomez-Lechon M. J. and Castell J. V. (2001). Cytochrome P450 regulation by hepatocyte nuclear factor 4 in human hepatocytes: a study using adenovirus-mediated antisense targeting. Hepatology 33: 668–675. Kapetanovic I. M., Torchin C. D., Strong J. M., Yonekawa W. D., Lu C., Li A. P., Dieckhaus C. M., Santos W. L., Macdonald T. L., Sofia R. D. and Kupferberg H. J. (2002). Reactivity of atropaldehyde, a felbamate metabolite in human liver tissue in vitro. Chem Biol Interact 142: 119–134. Kassahun K., Pearson P. G., Tang W., McIntosh I., Leung K., Elmore C., Dean D., Wang R., Doss G. and Baillie T. A. (2001). Studies on the metabolism of troglitazone to reactive intermediates in vitro and in vivo. Evidence for novel biotransformation pathways involving quinone methide formation and thiazolidinedione ring scission. Chem Res Toxicol 14: 62–70. Kern A., Bader A., Pichlmayr R. and Sewing K. F. (1997). Drug metabolism in hepatocyte sandwich cultures of rats and humans. Biochem Pharmacol 54: 761–772. Klaassen C. D. and Slitt A. L. (2005). Regulation of hepatic transporters by xenobiotic receptors. Curr Drug Metab 6: 309–328. Kliewer S. A., Lehmann, J. M., Milburn, M. V. and Willson, T. M. (1999). The PPARs and PXRs: nuclear xenobiotic receptors that define novel hormone signaling pathways: 2. Recent Prog Horm Res 54: 345–367. LeCluyse E. L. (2001). Human hepatocyte culture systems for the in vitro evaluation of cytochrome P450 expression and regulation. Eur J Pharm Sci 13: 343–368. Liebler D. C. and Guengerich F. P. (2005). Elucidating mechanisms of drug-induced toxicity. Nat Rev Drug Discov 4: 410–420.

86

Hepatotoxicity

Lloyd S., Hayden M. J., Sakai Y., Fackett A., Silber P. M., Hewitt N. J. and Li A. P. (2002). Differential in vitro hepatotoxicity of troglitazone and rosiglitazone among cryopreserved human hepatocytes from 37 donors. Chem Biol Interact 142: 57–71. Moore L. B., Parks D. J., Jones S. A., Bledsoe R. K., Consler T. G., Stimmel J. B., Goodwin B., Liddle C., Blanchard S. G., Willson T. M., Collins J. L. and Kliewer S. A. (2000). Orphan nuclear receptors constitutive androstane receptor and pregnane X receptor share xenobiotic and steroid ligands: 8. J Biol Chem 275: 15122–15127. Morel F., Beaune P. H., Ratanasavanh D., Flinois J. P., Yang C. S., Guengerich F. P. and Guillouzo A. (1990). Expression of cytochrome P-450 enzymes in cultured human hepatocytes. Eur J Biochem 191: 437–444. Mutlib A., Lam W., Atherton J., Chen H., Galatsis P. and Stolle W. (2005). Application of stable isotope labeled glutathione and rapid scanning mass spectrometers in detecting and characterizing reactive metabolites. Rapid Commun Mass Spectrom 19: 3482–3492. Naritomi Y., Terashita S., Kagayama A. and Sugiyama Y. (2003). Utility of hepatocytes in predicting drug metabolism: comparison of hepatic intrinsic clearance in rats and humans in vivo and in vitro. Drug Metab Dispos 31: 580–588. Olson H., Betton G., Stritar J. and Robinson D. (1998). The predictivity of the toxicity of pharmaceuticals in humans from animal data – an interim assessment. Toxicol Lett 102–103: 535–538. Olson H., Betton G., Robinson D., Thomas K., Monro A., Kolaja G., Lilly P., Sanders J., Sipes G., Bracken W., Dorato M., Van D. K., Smith P., Berger B. and Heller A. (2000). Concordance of the toxicity of pharmaceuticals in humans and in animals. Regul Toxicol Pharmacol 32: 56–67. Pang K. S. and Rowland M. (1977). Hepatic clearance of drugs. I. Theoretical considerations of a ‘well-stirred’ model and a ‘parallel tube’ model. Influence of hepatic blood flow, plasma and blood cell binding, and the hepatocellular enzymatic activity on hepatic drug clearance. J Pharmacokinet Biopharm 5: 625–653. Park B. K., Kitteringham N. R., Maggs J. L., Pirmohamed M. and Williams D. P. (2005). The role of metabolic activation in drug-induced hepatotoxicity. Annu Rev Pharmacol Toxicol 45: 177–202. Parker R. J., Hartman N. R., Roecklein B. A., Mortko H., Kupferberg H. J., Stables J. and Strong J. M. (2005). Stability and comparative metabolism of selected felbamate metabolites and postulated fluorofelbamate metabolites by postmitochondrial suspensions. Chem Res Toxicol 18: 1842–1848. Ripp S. L., Mills J. B., Fahmi O. A., Trevena K. A., Liras J. L., Maurer T. S. and de Morais S. M. (2006). Use of Immortalized Human Hepatocytes to Predict the Magnitude of Clinical Drug–Drug Interactions Caused by CYP3A4 Induction. Drug Metab Dispos 34: 1742–1748. Rodriguez-Antona C., Donato M. T., Boobis A., Edwards R. J., Watts P. S., Castell J. V. and GomezLechon M. J. (2002). Cytochrome P450 expression in human hepatocytes and hepatoma cell lines: molecular mechanisms that determine lower expression in cultured cells. Xenobiotica 32: 505–520. Rollins D. E., von B. C., Glaumann H., Moldeus P. and Rane A. (1979). Acetaminophen: potentially toxic metabolite formed by human fetal and adult liver microsomes and isolated fetal liver cells. Science 205: 1414–1416. Rumack B. H. (2002). Acetaminophen hepatotoxicity: the first 35 years. J Toxicol Clin Toxicol 40: 3–20. Runge-Morris M. and Kocarek T. A. (2005). Regulation of sulfotransferases by xenobiotic receptors: 1. Curr Drug Metab 6: 299–307. Skett P., Tyson C., Guillouzo A. and Maier P. (1995). Report on the international workshop on the use of human in vitro liver preparations to study drug metabolism in drug development (held in Utrecht, The Netherlands, 6–8 September 1994). Biochem Pharmacol 50: 280–285. Smith D. A. and Obach R. S. (2005). Seeing through the mist: abundance versus percentage. Commentary on metabolites in safety testing. Drug Metab Dispos 33: 1409–1417. Soars M. G., Burchell B. and Riley R. J. (2002). In vitro analysis of human drug glucuronidation and prediction of in vivo metabolic clearance. J Pharmacol Exp Ther 301: 382–390.


87

Sugatani J., Nishitani S., Yamakawa K., Yoshinari K., Sueyoshi T., Negishi M. and Miwa M. (2005). Transcriptional regulation of human UGT1A1 gene expression: activated glucocorticoid receptor enhances constitutive androstane receptor/pregnane X receptor-mediated UDPglucuronosyltransferase 1A1 regulation with glucocorticoid receptor-interacting protein 1: 3. Mol Pharmacol 67: 845–855. Temple R. J. and Himmel M. H. (2002). Safety of newly approved drugs: implications for prescribing. JAMA 287: 2273–2275. Tirona R. G. and Kim R. B. (2005). Nuclear receptors and drug disposition gene regulation: 1. J Pharm Sci 94: 1169–1186. Vignati L., Turlizzi E., Monaci S., Grossi P., Kanter R. and Monshouwer M. (2005). An in vitro approach to detect metabolite toxicity due to CYP3A4-dependent bioactivation of xenobiotics. Toxicology 216: 154–167. Walgren J. L., Mitchell M. D. and Thompson D. C. (2005). Role of metabolism in drug-induced idiosyncratic hepatotoxicity. Crit Rev Toxicol 35: 325–361. Watkins P. B. and Seeff L. B. (2006). Drug-induced liver injury: summary of a single topic clinical research conference: 1. Hepatology 43: 618–631. Wilkening S., Stahl F. and Bader A. (2003). Comparison of primary human hepatocytes and hepatoma cell line Hepg2 with regard to their biotransformation properties. Drug Metab Dispos 31: 1035– 1042. Zhao P., Kunze K. L. and Lee C. A. (2005). Evaluation of time-dependent inactivation of CYP3A in cryopreserved human hepatocytes. Drug Metab Dispos 33: 853–861. Zhou S., Chan E., Duan W., Huang M. and Chen Y. Z. (2005). Drug bioactivation, covalent binding to target proteins and toxicity relevance. Drug Metab Rev 37: 41–213.

4 Human-Based in vitro Experimental Systems for the Evaluation of Human Drug Safety Albert P. Li

4.1

Accurate Assessment of Drug Safety as a Major Challenge to the Pharmaceutical Industry

Drug development is a costly and time-consuming process. DiMassi et al. (2003) have estimated that for R&D initiated in 2001 with approval twelve years later (based on the average time required for approval), the out-of-pocket cost for a single approved drug is estimated to be US $970 million, equivalent to a capitalized cost of US $1.9 billion. A major challenge in drug development is the low clinical phase success rate (estimated to be 23 % by DiMassi et al. (2003)). Lack of efficacy and the occurrence of unacceptable adverse effects are the two principle reasons of failure. Furthermore, marketed drugs continue to be withdrawn or have their use limited due to adverse effects. Although the number of drugs required to be withdrawn is relatively small compared to all marketed drugs, the economical impact on the company with the withdrawn drug can be astronomical. Efficacy and safety are therefore two co-dependent requirements for successful drug development – clinical failure will result if the drug candidate possesses only one of these two properties. Drug discovery should not only emphasize on pharmacological effects, but should also take into account drug safety to allow the selection of drug candidates with the highest probability of success.


90

4.2

Hepatotoxicity

Human-Specific Drug Toxicity as a Major Reason for the Failure of Prediction of Human Drug Toxicity with Preclinical Nonhuman Animal Models

The fact that drug candidates would fail in clinical trials due to unacceptable safety illustrates that the old paradigm of drug safety testing, namely, the use of laboratory animals to predict human safety, is not adequate. Safety testing in nonhuman laboratory animals can be used to remove chemical entities which are toxic to both laboratory animals and humans, but would not be useful in the detection of toxicities that are human-specific.

4.2.1

Species–Species Differences in Drug Metabolism

Species-differences in drug metabolism have been extremely well-established. An example of rat–human differences in drug metabolism can be illustrated by a study in our laboratory using rat and human liver microsomes. In this study (Easterbrook et al., 2001), we evaluated the formation of human metabolites using rat and human liver microsomes from substrates of the major human P450 isoforms. Aroclor 1254-induced and uninduced rat liver microsomes were compared with human liver microsomes in phenytoin O-deethylation, coumarin 7-hydroxylation, tolbutamide 4-hydroxylation, s-mephenytoin 4 -hydroxylation, chloroxazone 6-hydroxylation and testosterone 6beta-hydroxylation. We found that both induced and uninduced rat liver microsomes were active in all of the pathways studied with the exception of coumarin 7-hydroxylation. Coumarin 7-hydroxylation was observed with human liver microsomes but not the rat liver microsomes. The results of this study illustrate that species–species differences in metabolism exist, and that a major human metabolite may not be formed in a nonhuman animal. The ramification of this finding is that if the toxicity of a drug is the result of the formation of a toxic metabolite, and if the formation of the toxic metabolite is specific to humans, then the safety data from nonhuman animal species will lead to an underestimation of human toxicity. Similarly, nonhuman animal data may overestimate human drug toxicity and may draw a wrong conclusion on the safety of the drug in humans, if the toxic metabolite is unique to the nonhuman animals and is not formed in humans.

4.2.2

Species–Species Differences in Target Cell Sensitivity to Toxicant Effects

Another possible explanation for human–nonhuman animal differences in drug toxicity is that a toxicant may have species-specific toxicity towards target cells. An example of this is the species differences of the toxicity of various anticancer agents on myeloid progenitor cells. Erickson-Miller et al. (1997) evaluated the toxicity of ercamptothecin (CAM), topotecan (TPT) and 9-aminocamptothecin (AC) to human, canine, and murine myeloid progenitors (CFU-GM) in vitro and found that the murine IC90 values were 2.6-, 2.3-, 10-, 21-, 5.9- and 11-fold higher than human values for CAM lactone (NSC-94600) and sodium salt (NSC-100880), TPT (NSC-609699), and racemic (NSC-629971), semisynthetic and synthetic preparations (NSC-603071) of 9AC, respectively. In contrast, canine IC90 values were the same as, or lower than, the human IC90 values for all six compounds.

Human-Based in vitro Experimental Systems

4.3

91

Overcoming Species–Species Differences as an Approach to Better Predict Human Drug Toxicity

Experimental systems that can provide information towards human-specific toxicity therefore should be valuable in the prediction of human drug toxicity in vivo. As in vivo experimentation with humans in vivo during preclinical phases is neither practical nor ethical, surrogates for humans in vivo need to be applied. Recently, human-based in vitro experimental systems have been developed and applied routinely for the evaluation of drug metabolism and drug–drug interactions (Huang et al., 1999; Davit et al., 1999; Li, 2004a). This ready acceptance of the in vitro systems can be attributed to the well-defined disciplines of drug metabolism. As the liver is the major organ for xenobiotic metabolism, in vitro hepatic systems such as liver microsomes and hepatocytes are found to readily reproduce in vivo hepatic metabolism. Regulatory agencies such as the US FDA (FDA, 1997, 1999, 2006) has endorsed the use of in vitro human-based experimental systems in the definition of human metabolic fate and drug–drug interaction potential of drugs. In fact, the US FDA has recommended that in vitro results with human-based systems such as liver microsomes and hepatocytes can be used to define human drug–drug interaction potential, without further clinical trials (clinical trials are recommended only if the results suggest significant drug–drug interaction potential). It is interesting to note that the application of in vitro drug metabolism technologies using human-based experimental systems has been attributed to the virtual elimination of pharmacokinetics as a reason for clinical trial failure. The success in the application of in vitro drug metabolism systems, in combination with data from relevant in vivo animal models, in the prediction of human metabolism suggest that the same approach will also be successful for safety evaluation (MacGregor et al., 2001). 4.3.1

In vitro–in vivo Strategy (IVIVS) for the Prediction of Human Drug Toxicity

Based on the premise that the inability to accurately predict human drug toxicity is due to species–species differences, i. e. there are human-specific drug properties that cannot be revealed by nonhuman animal studies, a safety evaluation strategy is proposed here for the preclinical evaluation of human drug toxicity: (1) Apply of human-based in vitro systems to provide human-specific toxicity data. (2) Select a relevant animal species to develop in vivo parameters. (3) Predict human in vivo drug toxicity via a combination of human-specific information obtained in vitro and in vivo parameters obtained from nonhuman animals in vivo. This IVIVS will require the development of in vitro experimental systems with humanspecific properties to cover the key toxic drug effects in humans and a vigorous set of parameters defining the relevant nonhuman animal species. 4.3.2

Selection of Human in vitro Experimental Systems

For the in vitro–in vivo paradigm to serve the intended purpose of accurate prediction of human drug toxicity, one has to careful in the choice of experimental systems. The following

92

Hepatotoxicity

are guidelines for the choice of in vitro systems: (1) Human-specific properties. As the purpose is to derive human-specific information, the system used should have the desired human properties. Examples of human-specific properties include human drug metabolism enzymes, transporters and molecular targets relevant to drug toxicity. (2) Physiologically relevant experimental conditions. In vitro systems, when used under experimental conditions that would not be achieved in vivo, may yield information that is difficult to extrapolate to the in vivo situation. Examples of irrelevant experimental conditions include the use of too high a drug concentration, interaction of the test substance with media ingredients that are not present in human body fluids and the use of cells with properties irrelevant to the target cells in vivo. (3) Appropriate endpoints. The endpoints used in conjunction with a chosen experimental system must be scientifically proven to be representative of the in vivo toxic events Examples of a relevant and a less relevant experimental system for the evaluation of human hepatotoxicity are presented below: Relevant: primary human hepatocytes: r Relevant species: human. r Relevant organ: liver. r Relevant target cells: hepatocytes or liver parenchymal cells are often the cells initially damaged by toxic drugs and reactive metabolites. r Relevant human-specific metabolism: human hepatocytes as freshly isolated cells or cryopreserved freshly isolated cells are well-established to contain full complements of human drug metabolizing enzymes (Li et al., 1999), including Phase I oxidation (e.g. P450 isoforms, monoamine oxidase, flavin-dependent monooxygenases), Phase II conjugation (UDP-dependent glucuronyl transferase, sulfotransferases, gluthathione transferase, N -acetylaminotransferases) and transporter activities (Shitara et al., 2003; Jigorel et al., 2005; Li, 2007). Less relevant: HepG-2 cells (human hepatoadenocarcinoma cell line): r Relevant species: human. r Relevant organ: liver. r Irrelevant target cells: adenocarcinoma cells are not the target of hepatotoxicity in vivo. r Irrelevant human-specific metabolism: Hep-G2 cells are known to have extremely low activities of human metabolizing enzymes, with the embryonic rather the adult phenotypes being expressed (e.g. CYP1A2 and CYP3A7 rather than CYP1A2 and CYP3A4/5) (Rodriguez-Antona et al., 2002). The preferred human in vitro systems are therefore primary cells derived from human organs, used within a period that the cells would retain differentiated functions, thereby serving as surrogates of the similar cells in vivo. The primary cell culture systems that are currently available and the respective organspecific toxicity that can be evaluated include the following: r hepatocytes for liver toxicity; r renal proximal tubule epithelial cells for nephrotoxicity;


93

r vascular endothelial cells for vascular toxicity; r neuronal and glial cells for neurotoxicity; r cardiomyocytes for cardiotoxicity. 4.3.3

Application of Human-Based in vitro Systems in Drug Development

In vitro toxicity assays can be applied in various phases of drug development (Li, 2000; 2004a): (1) Early screening of intrinsic toxicity. Cell-based systems are used for rapid screening of drug candidates, especially structural analogs, to allow the selection of less toxic structures for further development. The screening assay can allow logical evaluation of structures responsible for toxicity (toxicophore) which, hopefully, can be separated from structures for pharmacological activity (pharmacophores). Toxicity screening with in vitro systems require only limited amount of test articles, and is rapid and quantitative. Toxicity is most effective when one has an indication for in vivo toxicity (e.g. hepatotoxicity or nephrotoxicity) for a lead molecule, therefore allowing the selection of the most appropriate in vitro system for screening (e.g. hepatocytes for hepatotoxicity and renal proximal tubule cells for nephrotoxicity). (2) Mechanistic evaluations. Besides screening, in vitro experimental systems can be used to define toxic mechanisms. The defined experimental conditions and the availability of reagents and approaches for multiple endpoints allow one to define the key pathways involved in a toxicology phenomenon. Mechanistic understanding is critical to drug development. It allows a better understanding of human health risks, defines potential risk factors and evaluates the relationship between efficacy and adverse effects.

4.4

Overcoming the Major Deficiencies of an in vitro System

An argument routinely raised against the application of in vitro systems in safety evaluation is that toxicity is a complex phenomenon and therefore cannot be adequately modeled by simple in vitro systems such as cell culture assays. The major deficiencies of in vitro experimental systems can be defined as follows: (1) Lack of systemic effects. in vitro experimental systems in general consist of single cell types. Toxic effects are evaluated in the absence of influences from systemic effects that may be critical to drug toxicity. An example is the participation of the immune system in organ toxicity. One hypothesis for idiosyncratic hepatotoxicity, for instance, is the hapten hypothesis which postulates that liver failure arises from the cytotoxicity of antibodies towards antigens developed between the idiosyncratic drug (or its metabolites) on the plasma membrane of the hepatocytes. (2) Absence of chronic dosing. It is generally believe that drug toxicity due to acute cytotoxic events can be studied effectively with in vitro systems. However, toxic effects due to chronic, low-dose treatments may require multiple events that may or may not be obtained with in vitro studies, with cells treated for a relatively short time period (e.g. 24 h). Long-term treatments (e.g. months to years) of cells in culture is

94

Hepatotoxicity

theoretically possible but in practice near impossible. Furthermore, it is extremely difficult to maintain primary cells, the preferred cell system, in a differentiated state for a long time period. For in vitro systems to be useful, one needs to develop experimental approaches to overcome these deficiencies. 4.4.1

Target Cell Initiation Theory for Drug-Induced Organ Toxicity (TACIT)

It is well-established that the final manifestation of drug-induced organ toxicity may require systemic, multiple organ interactions. For instance, damage to a cell population in an organ may lead to the ‘recruitment’ of inflammatory cells which in turn may secrete cytotoxic cytokines, leading to further damage to the organ. The involvement of inflammation in liver toxicity has been suggested by one line of evidence that hepatotoxicity is enhanced in animals treated with pro-inflammatory agents (Ganey et al., 2002). However, it can be visualized that organ toxicity may be due to a cascade of events, started with a specific cell populations that initiate the cascade. The Target Cell Initiation Theory of Organ Toxicity (TACIT) is proposed here: Drug-induced organ failure is caused by changes in a target cell population which have the potential to initiate a cascade of events, and may or may not involve different cell types within the organ, or mechanisms involving target organs, leading ultimately to extensive damage to the organ in question.

A corollary to the TACIT is that although organ failure may involve multiple cell types and multiple organs, toxicants can be identified in the following are known: r the initiating target cells; r the critical initiation events which would cause the organ-toxicity cascade in the target cells. 4.4.2

Practical Application of TACIT

TACIT states that one can use the target cells in isolation in combination with critical endpoints reflecting initiating events to detect agents that can cause organ failure, regardless of the subsequent pathways which may involve nontarget cells and organs. An example of an application of the hypothesis is the evaluation of hepatotoxic potential using primary hepatocytes: (1) Target cell selection. For most toxic drugs, liver parenchymal cells are the cells that are initially damaged. The parenchymal cells represent the major cell type of the liver, with all of the drug metabolizing enzymes of the liver. The cells can be damaged by the drugs directly (e.g. high, bolus concentration after oral administration) or the metabolites (e.g. cytotoxic or highly reactive metabolites). Primary hepatocytes, as fresh or cryopreserved isolates, represent the most appropriate cell system for the evaluation of hepatotoxicity. (2) Endpoint selection. Hepatotoxicity is known to be initiated for both cytotoxic and noncytotoxic events (Table 4.1). Endpoints based on these known events can be selected to be used with hepatocytes for the evaluation of hepatotoxicity.


95

Table 4.1 Mechanisms of action of various types of hepatotoxicants. These mechanisms can be used as endpoints in combination with hepatocytes in vitro as early screening assays for hepatotoxicity Hepatotoxic mechanism

Toxicants

Direct cytotoxicity Toxic/reactive metabolite formation Toxic antibodies towards neoantigens

Cadmium chloride; cis-platinum Acetaminophen; cocaine; cyclophosphamide; ethanol

Cholestasis Steatosis Phospholipidosis

Amineptine; erythromycin derivatives; halothane imipramine; isaxonine; alpha-methyldopa; tienilic acid Rifampin; glibenclamide; cyclosporin A Carbamazepine; cycloheximide; tetracycline Amiodarone; diethylaminoethoxyhexestrol; perhexiline

Direct cytotoxicity. Liver toxicity is known to be preceded with cytotoxicity to the hepatocytes. Cytotoxicity can be using endpoints such as leakage of cytoplasmic enzymes, cellular ATP content, and mitochondrial metabolic activities (Li, 2004a). Correlation of cytotoxicity with human donor properties may lead to the understanding of key factors for hepatotoxicity of a toxicant (Lloyd et al., 2002; Hewitt et al., 2002). (b) Indirect cytotoxic effects. Toxicity of a substance to the liver may result from effects that may not lead to cytotoxicity directly, but will precipitate subsequent events which will lead to severe hepatotoxicity. These prelethal effects include the following: (i) Formation of reactive metabolites. Reactive metabolites formed in the hepatocytes react with cellular macromolecules which may lead to direct toxicity due to inactivation of key biochemical pathways. The reactive metabolites may also react with cellular proteins to form neoantigens, leading to cytotoxic inflammatory events. Reactive metabolites can be identified as glutathione (GSH) conjugates (e.g. using LC/MS) (Hartman et al., 2002; Prabhu et al., 2002; Zang et al., 2005; Cysyk et al., 2006). An indirect indication of reactive metabolite formation is the depletion of cellular GSH in the absence of cytotoxicity. (ii) Steatosis. Microvesicular steatosis, abnormal accumulation of intracellular lipid, is known to occur with exposure to hepatotoxic drugs. Accumulation of lipid vesicles and subsequent rupture of the lipid-ladden hepatocytes is believed to be a key step towards severe liver damage. Steatosis can be measured in cultured hepatocytes via the quantification of intracellular triglyceride accumulation and light microscopical examination of cells with lipid-specific stains such as Nile red (Amacher and Martin, 1997). (iii) Cholestasis. Bile acid retention in hepatocytes as a result of compromised efflux is known to be related to liver damage. The basolateral and apical transporters for bile acids are thought to be targets of drugs that induce cholestasis. Inhibition of bile acid transporter therefore is a relevant endpoint for the screening of drugs with cholestatic potential (Sahi et al., 2006). (a)

96

Hepatotoxicity

(iv) Phospholipidosis. Impaired phospholipid metabolism is known to be related to inflammatory events and histopathological changes. Some hepatotoxic drugs are known to cause phospholipidosis in hepatocytes which can be detected morphologically using electron microscopy as cytoplasmic lamella bodies. Screening for lipidosis can be performed by also using a dye for intracellular lipids such as Nile Red (Kasahara et al., 2006). Unlike steatosis, the relationship between phospholipidosis and hepatotoxicity is not yet established. However, this endpoint is a frequently used histopathological endpoint used to suggest hepatotoxic potential (Reasor et al., 2006). 4.4.3

High-Content Endpoint Assays

High-content assays, such as toxicogenomics for gene expression, proteomics for protein expression, metabonomics for natural metabolite synthesis and cellomics for cell morphology changes, are relatively novel technologies with potential application in the definition of drug toxicity. It is anticipated that these high-content assays, which allow one to evaluate the effect of a test substance on myriad events (e.g. expression of 20 000 human genes), one may achieve the following: (1) (2) (3) (4) (5)

Discovery of response patterns (response profiles) that can be used to define toxicity. Discovery of new biomarkers for specific types of toxicity. Discovery of early indicators of chronic toxicity. Discovery of indicators of idiosyncratic toxicity. Elucidation of toxic mechanisms.

The high-content assays may be extremely valuable to the TACIT approach with in vitro systems. For instance, prediction of chronic toxicity using in vitro systems may be possible, provided that the key events for chronic toxicity can be defined in the target cells after treatment with the toxicant in vitro. High-content assays performed on target cells with chronic toxicants versus nontoxic analogs may aid the discovery of these key initiating events. Another potential valuable application of high-content assays is to the definition of idiosyncratic drug toxicity. Efforts to use a single theory (e.g. reactive metabolite formation) to universally defined idiosyncratic drug toxicity have proven not yet to be useful. One possible explanation for the illusiveness of the definition of idiosyncratic toxicants is that there may be multiple factors involved (Li, 2002). These factors are proposed to be the chemical properties of the toxicant and the genetic make up of the human victim, as well as the complex environmental factors surrounding the victim. It is projected that the factors that can be readily controlled are the ‘chemical properties’. Therefore, a more practical approach to eliminate an idiosyncratic drug may be the definition of the common properties of the toxicants (Li, 2002). Towards this goal of defining the properties of drugs that would cause idiosyncratic hepatotoxiciy, we have applied toxicogenomics to evaluate troglitazone, a drug that was withdrawn from the market due to its association with idiosyncratic hepatotoxicity. Trogltiazone was compared to the relatively nontoxic structural analogs, rosiglitazone and pioglitazone (Kier et al., 2004). Human hepatocytes from three donors were treated in vitro with the three chemicals, followed by extraction of mRNA and hybridization with cDNA microarrays. The gene expression heat map (using various ‘grades of shading’ to classify the magnitude of


97

response) is presented in Figure 4.1. Dramatic differences were observed in gene expression profiles between the toxic troglitazone and the nontoxic structural analogues. Troglitazone was found to selectively ‘down-regulate’ genes with protective functions and ‘up-regulate’ genes with toxification functions. Similar toxicogenomics results with troglitazone were also obtained from the laboratory of Waring and coworkers (Liguori et al., 2005). The results illustrate the application of toxicogenomics in the development of an hypothesis to allow subsequent mechanistic investigations. Toxicogenomics studies with troglitazone and its less toxic analogs, rosiglitazone and pioglitazone, provided clues for the differential hepatotoxicity of these structurally related chemicals. All three chemicals appear to induce genes for oxidative metabolism (e.g. P450 isoform 3A4) which may be responsible for the generation of reactive metabolites (Kassahun et al., 2000). Troglitazone, however, uniquely induced gene expression of the toxification pathways such as apoptosis and inflammation, and suppressed gene expressions of the gene responsible for detoxification pathways such as acute phase proteins, Phase II conjugation, and stress proteins (Kier et al., 2004; Liguori et al., 2005). It is projected that the genes that are affected differentially by troglitazone and its less toxic structural analogs may be used to predict the human hepatotoxic potential of new troglitazone structural analogs. The combination of toxicgonomics and human cells in vitro may be a powerful combination to aid the accurate prediction of human drug toxicity. The key is to develop an in vitro system to incorporate in vivo factors that are key to toxicity. One such system is the Integrated Discrete Multiple Organ Co-culture (IdMOC) system. 4.4.4

IdMOC for Multiple Organ Interactions

One major drawback of an in vitro system is that each cell type is studied in isolation, while in the human body there may be multiple organ interactions that are critical to drug toxicity. The multiple organ interaction is not covered by the TACIT approach, as the initiating events may include effects of a toxicant on a target cell and a nontarget cell. A simple example is a drug which is first metabolized by the liver to form metabolites which may cause toxicity in a ‘distant’ organ, such as the heart. The TACIT approach using the target cell with the parent drug is therefore unlikely to produce the initiating events for the prediction of the ultimate organ toxicity. To overcome this deficiency, we have developed the Integrated Discrete Multiple Organ Co-culture (IdMoc) system (Li et al., 2004). The IdMOC allows the co-culturing of cells from different organs as physically separated cultures that are interconnected by an overlying medium, akin to the blood circulation connecting the multiple organs in the human body. The IdMOC models the multiple organ interaction in the whole organism in vivo, thus allowing the evaluation of organ-specific effects of a drug and its metabolites (Figure 4.2). The IdMOC represents an improved in vitro experimental system for routine screening of ADMET drug properties. The IdMOC involves the ‘wells-in-a-well’ concept. The typical IdMOC plate consists of an outer well containing within which are several wells (Figure 4.3). Cells of different origins (e.g. from different organs) are initially cultured, each in its own specific medium, in the wells. When the cells are established, the inner wells are flooded with an overlying medium, thereby connecting the multiple cell types. The multiple-cell types now can interact via the overlying medium, akin to the multiple organs in a human body interacting via the systemic circulation.

Translation

Oxidative stress

PHASE II metabolism

Mitochondrial function

Acute phase

Suppressed categories

−1.2 −1.3 −1.9 −1.3 −1.2 −1.8 −1.9

−2.1

−1.8 −1.7 −2.2 −1.7 −1.6 −1.7 −1.6 −1.7 −1.7 −2.0 −1.6 −1.8 −1.8 −2.0 −2.0

−2.1 −1.7 −1.9 −1.8 −1.4 −1.8 −1.9

−2.1

−1.9 −1.6 −1.7 −1.8 −1.5 −1.6 −1.6 −2.2 −1.7 −2.4 −1.5 −1.8 −1.8 −2.0 −1.9

10 µM −1.6 −1.4 −1.7 −1.4 −1.2 −2.0 −2.0 −2.6 −1.9 −1.6 −2.1 −2.0 −1.8 −1.5 −1.5 −1.9 −1.4 −1.7 −1.5 −1.6 −1.6 −1.8 −1.7

−1.4 −1.3 −1.5 −1.2 −1.0 −1.5 −1.6 −2.5 −1.7 −1.6 −2.2 −1.5 −1.4 −1.2 −1.3 −2.2 −1.7 −1.5 −1.4 −1.5 −1.4 −1.5 −1.9

20 µM

6h

−2.2 −1.4 −2.1 −2.0 −1.9 −1.3 −1.4 −2.4 −1.0 −1.3 −1.7 −1.5 −1.4 −1.2 −1.8

−3.7

−1.9 −1.6 −1.5 −1.4 −1.3 −1.0 −1.7

−1.8 −1.2 −1.5 −1.6 −1.3 1.0 −1.3 −1.5 −1.4 −1.3 −1.5 −1.5 −1.4 −1.4 −1.7

−3.2

−1.6 −1.4 −1.3 −1.2 −1.1 −1.4 −1.3

50 µM −1.6 −1.6 −1.7 −1.5 −1.6 −1.4 −1.7 −2.4 −1.6 −1.4 −1.7 −1.6 −1.4 −1.5 −1.3 −2.8 −1.4 −1.2 −1.7 −1.6 −1.4 −1.5 −1.7

−1.6 −1.6 −1.8 −1.5 −1.5 −1.6 −1.7 −2.4 −1.7 −1.2 −1.5 −1.6 −1.4 −1.5 −1.4 1.1 −1.6 −1.3 −1.8 −1.6 −1.6 −1.6 −1.8

10 µM

I: Troglitazone

−1.7 −1.5 −1.7 −1.6 −1.3 −1.4 −1.5 −1.7 −1.6 −1.3 −2.0 −1.6 −1.5 −1.6 −1.9

−2.5

−1.2 −1.5 −1.4 −1.3 −1.6 −1.7 −1.5

−1.9 −1.7 −2.2 −1.9 −1.6 −1.4 −1.8 −1.9 −1.8 −1.3 −2.1 −1.8 −1.5 −1.9 −2.2

−2.7

−2.4 −2.0 −1.6 −1.6 −2.1 −1.9 −1.7

20 µM

24 h

−1.4 −1.1 −1.2 1.0 1.1 −1.6 −1.1 1.0 −1.1 −1.6 −1.1 −1.1 −1.2 −1.2 −1.1

−1.5

−1.2 −1.0 −1.1 1.0 1.1 −1.4 −1.1 −1.0 −1.1 −1.5 −1.0 −1.1 −1.1 −1.1 −1.1

−1.4

Pio −1.3 −1.4 −1.2 −1.6 −1.3 −1.2 −1.2 −1.0 −1.2 −1.1 −1.6 −1.2 −1.3 −1.4

−1.3 −1.1 −1.2 −1.0 1.1 −1.4 −1.0 1.0 1.0 −1.4 −1.1 −1.1 −1.2 −1.2 −1.2

−1.4

−1.2 −1.0 −1.0 1.0 1.0 −1.2 −1.1 1.1 −1.1 −1.3 −1.1 −1.1 −1.1 −1.1 −1.0

−1.4

II: 25 µM, 24 h Rosi −1.2 −1.3 −1.1 −1.4 −1.3 −1.2 1.0 −1.2 −1.2 −1.0 −1.4 −1.1 −1.2 −1.4

−2.0 −2.4 −3.2 −2.3 −1.8 −4.7 −2.4 −2.8 −2.8 −5.0 −1.1 −1.7 −1.7 −1.5 −1.6

−1.8

−1.7 −1.8 −1.5 −2.7 −1.9 −2.6 −2.6 1.2 −2.2 −4.3 −1.0 −1.6 −1.5 −1.4 −1.5

−1.3

Tro −1.9 −4.3 −1.6 −2.1 −2.0 −1.7 −1.6 −2.1 −4.4 −2.5 −3.0 −1.9 −2.2 −2.2

Figure 4.1 Heat map summarizing the toxicogenomics results of the treatment of primary human hepatocytes with the hepatotoxic drug, troglitazone(tro), and its less toxic analogs, i.e. rosiglitazone(rosi), and pioglitazone(pio). The results of genes that are differentially suppressed and induced by troglitazone are shown in (a) and (b), respectively. Genes are grouped by physiological mechanisms of action and each line represents the results of a single gene. Significant response is defined as >50 % induction or suppression over solvent control. The microarray used was the TOX-chip from Phase I Molecular Toxicology, Inc.

(a)

Immediate–early

Apoptosis

Induced Genes

Cell growth

Inflammation

DNA replication and repair

Phase-1 metabolism

(b)

1.9 1.4 1.2 1.6 1.7 1.2 2.0 1.9 2.2 6.7 3.3 4.2 3.9 3.3 3.1 3.4 2.7 3.2 3.8 3.9 2.8 2.0 1.6 2.0 1.7 1.7 1.6 1.4 2.4 2.9 1.4 1.4 1.3 1.4 1.7

1.8 1.4 1.5 1.2 2.0 3.2 2.8 2.2 4.2 10.6 5.8 6.5 7.4 10.0 5.1 8.6 2.4 4.0 3.4 3.7 3.3 3.1 1.8 2.0 2.4 2.3 2.4 1.7 1.4 2.5 1.5 1.8 2.6 1.7 1.6

10 µM 1.8 1.2 −1.0 1.4 3.2 1.8 2.8 1.8 2.8 12.4 4.1 7.3 11.4 7.1 5.1 7.1 4.0 3.7 4.0 4.3 3.0 2.4 2.0 2.8 1.8 1.7 2.5 2.1 1.7 2.9 1.8 1.7 2.0 1.6 2.8 2.2 1.9 1.8 1.3 1.8 1.3 2.5 1.4 2.8 9.7 4.4 5.4 4.3 4.5 4.1 5.5 3.2 1.9 3.6 2.6 2.6 2.3 1.6 1.7 2.3 2.1 2.4 1.9 2.5 2.7 1.7 1.7 2.9 1.8 1.5

20 µM 2.2 1.2 1.2 −1.1 2.5 1.2 2.2 1.6 2.7 7.6 3.6 4.4 4.0 4.5 3.2 4.3 4.5 3.7 3.0 3.5 2.8 2.1 2.5 1.5 2.0 2.1 2.9 1.5 2.4 2.5 1.4 −1.1 2.5 1.9 2.1

6h 1.1 1.0 1.4 1.0 1.5 2.3 2.6 1.2 2.9 5.9 3.6 6.1 5.2 5.3 5.2 5.1 2.1 3.4 2.9 3.7 2.4 2.3 4.4 2.5 2.2 2.7 2.8 2.0 2.4 2.6 1.9 2.6 2.4 3.2 1.5

Figure 4.1

1.3 1.1 −1.2 1.1 1.4 1.6 2.2 1.1 2.7 7.3 3.2 4.7 2.7 4.6 4.0 5.6 2.4 3.2 2.8 3.0 1.8 2.0 3.7 1.4 1.4 1.7 2.5 2.1 2.1 2.2 1.6 1.2 1.8 1.1 1.3

50 µM

10 µM

24 h

(Continued)

1.9 1.5 −1.1 1.0 1.3 2.1 1.3 1.3 2.4 2.1 1.3 1.2 2.5 3.1 1.5 1.7 1.7 2.4 10.4 10.9 4.5 6.2 5.3 6.8 2.5 4.7 5.1 5.2 3.3 4.7 4.2 6.8 5.4 6.2 3.2 3.1 5.2 5.7 3.7 3.9 3.0 3.1 2.2 2.4 1.1 −1.1 1.7 1.6 1.8 2.0 1.9 2.3 2.1 1.9 2.7 1.8 1.6 1.6 2.2 2.5 1.6 2.1 −1.0 1.5 2.1 1.8 1.9 3.1 2.6 1.7

I: Troglitazone

1.1 1.8 −1.4 1.2 3.0 1.6 1.4 1.0 4.1 2.7 −1.4 −1.1 2.3 3.7 2.1 1.6 1.5 2.6 10.1 11.3 4.2 4.2 6.6 7.7 3.7 3.0 4.6 6.6 5.5 6.1 4.9 5.6 4.2 4.1 1.5 1.7 4.2 2.9 3.9 3.5 3.1 2.6 2.7 2.6 1.3 2.1 1.2 1.6 1.1 2.5 1.7 2.2 1.7 1.4 1.8 1.8 1.0 1.6 2.4 1.8 1.9 1.5 1.3 2.0 1.4 1.8 2.3 2.7 1.7 1.9

20 µM

II: 25 µM Pio Rosi 1.1 1.1 1.2 −1.1 1.2 −1.0 1.3 1.0 1.2 1.0 1.1 1.2 1.7 1.3 1.8 1.3 −1.0 1.0 −1.1 −1.0 1.6 1.2 1.4 1.3 1.0 1.6 1.1 1.4 −1.1 −1.1 1.6 −1.1 1.9 −1.0 1.7 1.5 3.2 1.8 1.6 3.6 1.9 1.2 1.5 2.1 2.0 1.4 1.6 1.6 1.8 1.1 2.0 1.2 2.2 1.4 1.8 2.2 −1.2 −1.5 1.2 −1.1 3.3 1.7 1.5 5.0 1.1 1.0 −1.0 1.0 −1.1 −1.3 −1.0 −1.1 1.1 −1.1 1.1 −1.0 1.2 1.6 1.2 1.8 −1.3 −1.0 −1.4 −1.5 1.2 1.3 1.3 1.3 −1.4 −1.4 −1.5 −1.2 1.1 1.3 1.1 1.1 1.1 1.1 1.2 1.1 1.0 1.2 –1.1 1.1 1.2 1.0 1.2 1.0 1.1 1.0 1.1 1.1 1.1 1.1 1.0 −1.1 1.2 1.2 −1.0 −1.2 −1.0 −1.2 −1.1 −1.1 1.1 1.3 −1.0 −1.2 1.1 1.0 1.0 1.0 1.0 −1.1 1.0 −1.0 1.2 1.6 1.3 1.7 Tro 1.3 1.8 1.3 1.5 1.7 1.1 1.5 1.1 2.4 2.2 1.5 1.8 1.2 1.7 −1.0 -1.1 1.5 3.9 −2.5 1.8 1.2 3.5 −1.2 2.6 −1.5 3.3 2.0 2.4 −1.1 1.1 1.0 2.0 1.2 1.6 −1.4 1.4 1.6 1.2 2.5 1.2 3.5 1.5 2.6 1.6 2.0 2.3 2.1 1.3 2.6 3.0 1.6 1.2 1.7 1.2 1.5 1.5 3.2 2.3 2.5 1.3 2.8 1.8 3.1 1.6 2.3 1.5 1.7 1.2 1.7 1.2

100

Hepatotoxicity

(a)

(b)

Overlying medium connecting the multiple organs

Organ A cells (e.g. hepatocytes)

Organ B cells (e.g. kidney proximal tubule cells)

Organ C cells (e.g. vascular endothelial cells)

Figure 4.2 Schematic presentation of the Integrated Discrete Multiple Organ Co-culture (IdMOC) system. The IdMOC system is based on the concept that in the human body there are multiple organs that are physically separated but interconnected by the systemic circulation (a). A toxicant entering the systemic circulation will have interactions with all of the interconnected organs. Results of the interactions, which can be the formation of metabolites or the induction of reactive biomolecules (e.g. cytokines), will have the potential to interact with the multiple organs. The concept is reduced to practice by using a ‘wells-in-a-chamber’ concept, with multiple wells in a containing chamber, allowing the culturing of cells from different organs as physically discrete wells interconnected by an overlying medium (b)

The IdMOC system can be used for the following: (1) Differential cytotoxicity. Evaluation of the toxicity of a substance on different cell types (e.g. cells from different organs) under virtually identical experimental conditions with multiple cell-type interactions. (2) Differential distribution. Evaluation of the differential accumulation/distribution of a substance among multiple cell types. (3) Multiple organ metabolism. Evaluate the ultimate metabolic fate of a substance upon metabolism by cells representing multiple organs with metabolic functions (e.g. liver, kidney, lung).


101

Figure 4.3 Photograph of an IdMOC plate based on the format of a standard 96-well plate (IdMOC-96). This plate consists of six individual wells inside a containing chamber. The wells are filled with colored liquids (left) to show that they are physically separated. The chambers are filled (right) to show the connection of the wells. Each chamber therefore represents a single experimental unit, thus allowing each plate to be used for multiple-treatment conditions. For instance, one IdMOC-96 plate can be used to evaluate the effect on six cell types for fifteen test articles (using one chamber as control) at a single concentration, two test articles at four concentrations in duplicates or a single test article at four concentrations in quadruplicates

We have recently evaluated the differential cytotoxicity of tamoxifen using the IdMOC system (Li et al., 2004). Further validation of the system is now in progress in our laboratory.

4.5

Conclusions

Accurate prediction of human drug toxicity represents a major challenge for drug development. The high rate of clinical failure of drug candidates that have been carefully selected from preclinical studies illustrates clearly that the routine, ‘classical’ approach of preclinical safety evaluation is inadequate. It is argued here that species–species differences in drug toxicity is a major reason – human-specific toxicity, by definition, cannot be predicted with nonhuman laboratory animals . It is proposed here that human in vivo drug toxicity can be predicted by using a combination of human-based in vitro experimental systems and in vivo laboratory animals. It is fully recognized that regulatory toxicology has evolved to be a rigid discipline, with efficiency as a major objective, and from the drug manufacturers’ viewpoint, regulatory

102

Hepatotoxicity

acceptance as the ultimate goal. Investigational approaches will be difficult to implement for a new drug candidate at the final stages of development. A practical approach will be to apply investigational approaches early in drug development, using experimental systems such as the in vitro human-based experimental systems described here, to remove drug candidates with human safety concerns as early as possible. Efforts towards designing in vitro experimental approaches to maximize the strengths, applying within their limitations, and incorporating relevant in vivo factors whenever possible, may be rewarded with a better prediction of human drug toxicity before the drugs are exposed to the human population. The use of primary cells, high content assays and IdMOC are described here as potentially important approaches. The challenge of the accurate prediction of human drug toxicity can only be met if one realizes that classical toxicological approaches as practiced now, although important, are not adequate. This realization, plus the willingness to accept, evaluate and apply new approaches, are keys to our future success. It also needs to be recognized that drug safety evaluation and the ultimate development of safe drugs requires scientific insight and expertise from multiple scientific disciplines (Li, 2004b,c).

References Amacher D. E. and Martin B. A. (1997) Tetracycline-induced steatosis in primary canine hepatocyte cultures. Fundam Appl Toxicol 40:256–263. Cysyk R. L., Parker R. J., Barchi J. J. Jr., Steeg P. S., Hartman N. R. and Strong J. M. (2006). Reaction of geldanamycin and C17-substituted analogues with glutathione: product identifications and pharmacological implications. Chem Res Toxicol 19:376–381. Davit B., Reynolds K., Yuan R., Ajayi F., Conner D., Fadiran E., Gillespie B., Sahajwalla C., Huang S. M. and Lesko L. J. (1999). FDA evaluations using in vitro metabolism to predict and interpret in vivo metabolic drug–drug interactions: impact on labeling. J Clin Pharmacol 39:899–910. DiMassi J. A. Hansen R. W. and Grabowski H. G. (2003.) The price of innovation: new estimates of drug development costs. J Health Econ 22:151–185. Easterbrook J., Fackett D. and Li A. P. (2001). A comparison of aroclor 1254-induced and uninduced rat liver microsomes to human liver microsomes in phenytoin O-deethylation, coumarin 7-hydroxylation, tolbutamide 4-hydroxylation, S-mephenytoin 4 -hydroxylation, chloroxazone 6-hydroxylation and testosterone 6beta-hydroxylation. Chem Biol Interact 134:243–249. Erickson-Miller C. L., May R. D., Tomaszewski J., Osborn B., Murphy M. J., Page J. G. and Parchment R. E. (1997). Differential toxicity of camptothecin, topotecan and 9-aminocamptothecin to human, canine and murine myeloid progenitors (CFU-GM) in vitro. Cancer Chemother Pharmacol 39:467– 472. FDA (1997). ‘Drug Metabolism/Drug Interaction Studies in the Drug Development Process: Studies in Vitro’ (issued 4/1997, posted 4/8/1997), Food and Drug Administration (FDA), r r r, r r r [http://www.fda.gov/cder/guidance/clin3.pdf]. FDA (1999). ‘In vivo Drug Metabolism/Drug Interaction Studies – Study Design, Data Analysis and Recommendations for Dosing and Labeling’ (issued 11/24/1999, posted 11/24/1999), Food and Drug Administration (FDA), [http://www.fda.gov/cder/guidance/2635fnl.pdf]. FDA (2006). Guidance for Industry (Draft Guidance). Drug Interaction Studies – design, data analysis, and implications for dosing and labelling. www.fda.gov/cber/gdlns/interactstud.htm. Ganey PE, Luyendyk JP, Maddox JF, Roth RA (2002). Adverse drug reactions: inflammatory episodes as consequence and contributor. Chem Biol Interact 150:35–51.


103

Hartman N. R., Cysyk R. L., Bruneau-Wack C., Thenot J. P., Parker R. J. and Strong J. M. (2002). Production of intracellular 35S-glutathione by rat and human hepatocytes for the quantification of xenobiotic reactive intermediates. Chem Biol Interact 142:43–55. Hewitt N. J., Lloyd S., Hayden M., Butler R., Sakai Y., Springer R., Fackett A. and Li A. P. (2002). Correlation between troglitazone cytotoxicity and drug metabolic enzyme activities in cryopreserved human hepatocytes. Chem Biol Interact 142:73–82. Huang S. M., Lesko L. J. and Williams R. L. (1999) Assessment of the quality and quantity of drug– drug interaction studies in recent NDA submissions: study design and data analysis issues. J Clin Pharmacol 39:1006–1014. Jigorel E., Le Vee M., Boursier-Neyret C., Bertrand M. and Fardel O. (2005). Functional expression of sinusoidal drug transporters in primary human and rat hepatocytes. Drug Metab Dispos 33:1418– 1422. Kasahara T., Tomita K., Murano H., Harada T., Tsubakimoto K., Ogihara T., Ohnishi S. and Kakinuma C. (2006). Establishment of an in vitro high-throughput screening assay for detecting phospholipidosis-inducing potential. Toxicol Sci 90:133–141. Kassahun K., Pearson P. G., Tang W., McIntosh I., Leung K., Elmore C., Dean D., Wang R., Doss G. and Baillie T. A. (2000). Future of toxicology–metabolic activation and drug design: challenges and opportunities in chemical toxicology. Chem Res Toxicol 19:889–893. Kier L. D., Neft R., Tang L., Suizu R., Cook T., Onsurez K., Tiegler K., Sakai Y., Ortiz M., Nolan T., Sankar U. and Li A. P. (2004). Applications of microarrays with toxicologically relevant genes (tox genes) for the evaluation of chemical toxicants in Sprague Dawley rats in vivo and human hepatocytes in vitro. Mutat Res 549:101–113. Li A. P. (2002). A review of the common properties of drugs with idiosyncratic hepatotoxicity and the ‘multiple determinant hypothesis’ for the manifestation of idiosyncratic drug toxicity. Chem Biol Interact 142:7–23. Li A. P. (2004a). In vitro approaches to evaluate ADMET drug properties. Curr Top Med Chem 4:701–706. Li A. P. (2004b). An integrated, multidisciplinary approach for drug safety assessment. Drug Discov Today 9:687–693. Li A. P. (2004c). A comprehensive approach for drug safety assessment. Chem Biol Interact 150:27– 33. Li A. P. (2007). Human hepatocytes: isolation, cryopreservation and applications in drug development. Chem Biol Interact 168:16–29. Li A. P., Lu C., Brent J. A., Pham C., Fackett A., Ruegg C. E. and Silber P. M. (1999). Cryopreserved human hepatocytes: characterization of drug-metabolizing enzyme activities and applications in higher throughput screening assays for hepatotoxicity, metabolic stability and drug–drug interaction potential. Chem Biol Interact 121:17–35. Li A. P., Bode C. and Sakai Y. (2004). A novel in vitro system, the integrated discrete multiple organ cell culture (IdMOC) system, for the evaluation of human drug toxicity: comparative cytotoxicity of tamoxifen towards normal human cells from five major organs and MCF-7 adenocarcinoma breast cancer cells. Chem Biol Interact 150:129–136. Liguori M. J., Anderson M. G., Bukofzer S., McKim J., Pregenzer J. F., Retief J., Spear B. B. and Waring J. F. (2005). Microarray analysis in human hepatocytes suggests a mechanism for hepatotoxicity induced by trovafloxacin. Hepatology 41:177–86. Lloyd S., Hayden M. J., Sakai Y., Fackett A., Silber P. M., Hewitt N. J. and Li A. P. (2002). Differential in vitro hepatotoxicity of troglitazone and rosiglitazone among cryopreserved human hepatocytes from 37 donors. Chem Biol Interact 142:57–71. MacGregor J. T., Collins J. M., Sugiyama Y., Tyson C. A., Dean J., Smith L., Andersen M., Curren R. D., Houston J. B., Kadlubar F. F., Kedderis G. L., Krishnan K., Li A. P., Parchment R. E., Thummel K., Tomaszewski J. E., Ulrich R., Vickers A. E. and Wrighton S. A. (2001). In vitro

104

Hepatotoxicity

human tissue models in risk assessment: report of a consensus-building workshop. Toxicol Sci 59:17–36. Prabhu S., Fackett A., Lloyd S., McClellan H. A., Terrell C. M., Silber P. M. and Li A. P. (2002). Identification of glutathione conjugates of troglitazone in human hepatocytes. Chem Biol Interact 142:83–97. Reasor M. J., Hastings K. L. and Ulrich R. G. (2006). Drug-induced phospholipidosis: issues and future directions. Expert Opin Drug Saf 5:567–583. Rodriguez-Antona C., Donato M. T., Boobis A., Edwards R. J., Watts P. S., Castell J. V. and GomezLechon M. J. (2002). Cytochrome P450 expression in human hepatocytes and hepatoma cell lines: molecular mechanisms that determine lower expression in cultured cells. Xenobiotica 32:505–520. Sahi J., Sinz M. W., Campbell S., Mireles R., Zheng X., Rose K. A., Raeissi S., Hashim M. F., Ye Y., de Morais S. M., Black C., Tugnait M. and Keller L. H. (2006). Metabolism and transportermediated drug–drug interactions of the endothelin-A receptor antagonist CI-1034. Chem Biol Interact 159:156–168. Shitara Y., Itoh T., Sato H., Li A. P. and Sugiyama Y. (2003). Inhibition of transporter-mediated hepatic uptake as a mechanism for drug–drug interaction between cerivastatin and cyclosporin A. J Pharmacol Exp Ther 304:610–616. Zhang Z., Chen Q., Li Y., Doss G. A., Dean B. J., Ngui J. S., Silva Elipe M., Kim S., Wu J. Y., Dininno F., Hammond M. L., Stearns R.A., Evans D. C., Baillie T. A. and Tang W. (2005). In vitro bioactivation of dihydrobenzoxathiin selective estrogen receptor modulators by cytochrome P450 3A4 in human liver microsomes: formation of reactive iminium and quinone type metabolites. Chem Res Toxicol 18:675–685.

5 Hepatocytes as a Model for Screening Food-Related Hepatotoxins and Studying Mechanisms of their Toxicity Saura C. Sahu

5.1

Introduction

Liver is an important organ, which plays a central role in the metabolism of toxins entering the body through the digestive tract and transported via the portal vein. It is the primary organ involved in xenobiotic metabolism and is a major target organ for chemicals, drugs and microbial pathogens (Treinen-Molsen, 2001). Hepatotoxicity is also a common adverse health effect of food-related toxins (Miller and Miller, 1986; Willett et al., 2004; Stickel et al., 2005), and it is thus one of the most serious safety concerns for food additives, food contaminants, dietary supplements, food-borne microbial pathogens and other food-related products regulated by the US Food and Drug Administration (FDA). In vivo animal studies for assessing the hepatotoxic potential of food-related toxins are expensive and time-consuming. For these reasons, alternate in vitro models are important to complement and/or supplement the animal studies for toxicological risk assessment (Green et al., 2001; MacGregor et al., 2001). The in vitro models are relatively inexpensive and they provide well-defined and reproducible experimental conditions for toxicity testing. They are excellent systems to study mechanisms of toxicity and structure–activity relationships at the cellular and molecular level. They are also very useful tools for rapid screening of potential toxins.


106

Hepatotoxicity

The US Food and Drug Administration (FDA) is mandated by the NIH Revitalization Act of 1993 (P.L. No. 103-43) in concert with the ICCVAM Authorization Act of 2000 (P.L. No. 106-545) for federal regulatory agencies to: ‘(a) establish criteria for the validation and regulatory acceptance of alternative testing methods and (b) recommend a process through which scientifically validated alternative methods can be accepted for regulatory use’. Criteria for validation of alternative toxicological test methods have been published (ICCVAM, 1997). The use of in vitro cytotoxicity results to predict starting doses for in vivo acute toxicity studies reducing the number of animals for safety evaluation by the regulatory agencies is allowed (Federal Register, 2001). Primary hepatocytes and some hepatocyte cell lines retain many of the metabolic enzymes characteristic of the intact liver in vivo (Hengstler et al., 2000, 2002; Runge et al., 2001). Therefore, they represent an excellent in vitro model for studying liver function, xenobiotic metabolism, pharmacology and toxicology. They are widely used for the evaluation of liver functions and toxicity of chemicals, drugs and microbes (Barsig et al., 1998; Castell et al., 1997; Li et al., 1997; Michalopoulos, 1999; Sahu et al., 2001, 2006).

5.2 5.2.1

Hepatocyte Cultures Hepatocytes in Primary Culture

Hepatocytes for primary culture are isolated by the two-step in situ collagenase perfusion of liver following the method of Seglen (1976). Isolated hepatocytes survive for a few hours, but they can be cryopreserved for a couple of years with somewhat reduced viability and metabolizing enzyme activities (Chesne et al., 1993). Long-term culture of hepatocytes has been achieved using co-culture with other cells (Donata et al., 1994), in dimethyl sulfoxide (DMSO)-supplemented medium (Isom et al., 1985), collagen sandwich (Kono et al., 1995), collagen-coated dishes and serum-free medium supplemented with hormones and growth factors (Ferrini et al., 1997), cell aggregation (Tateno and Yoshizato, 1996), gel entrapment (Guyomard et al., 1996) and spheroid cultures (Walker et al., 2000; Hamilton et al., 2001; Khalil et al., 2001). Rat hepatocytes in primary culture are the most frequently used and best characterized in vitro model for testing liver toxicity of chemicals and drugs. They have a long history of use by many laboratories throughout the world (Carfagna et al., 1996; DiPetrillo et al., 2002; Hill and Roth, 1998; Hussain and Frazier, 2003; Sahu et al., 2001; Susa et al., 1997; Tirmenstein et al., 2000; Tseng et al., 1997). Hepatotoxicity data obtained from the primary rat hepatocytes are consistent with the data obtained in vivo from rats (Kefalas and Stacey, 1993), and thus, the in vitro and in vivo data obtained from rat studies appear to be complementary. In addition, the in vitro data obtained from the primary rat hepatocytes can be validated with data obtained from in vivo studies. Human hepatocytes in primary culture are the ideal in vitro system for hepatotoxicity assessment of regulated products. However, their availability is limited. Cryopreserved primary human hepatocytes are commercially available, but optimal conditions for maintaining human hepatocytes in culture have not been perfected yet. Unlike the primary rat hepatocytes, which have been extensively used in research for almost three decades, the use of primary human hepatocytes is relatively recent. High inter-individual variability in

Screening Hepatotoxins and Mechanism of Toxicity

107

phenotypes and genotypes of human liver, as well as large variations in individual consumption of various diets, are important factors for consideration. For example, donated human liver tissues from nine males and nine females were used to evaluate gender differences in the gene expression of human livers (Delongchamp et al., 2005). This study, which tested 31 100 genes, concluded that the gene expression of 224 genes differed between sexes. The observed gender differences in expression were small. False discovery rates exceeded 80 % for every set of genes selected, making it impossible to identify specific genes with gender differences (Delongchamp et al., 2005). In addition, the in vitro data obtained from cryopreserved human hepatocytes would be difficult to validate in vivo. These are examples of some of the unresolved issues and challenges associated with the use of cryopreserved human hepatocytes for obtaining experimental data that can be used for risk assessment by regulatory agencies. However, as continued research clarifies these issues there will undoubtedly be increasing use of this important experimental system for safety evaluation and risk management of regulated products. 5.2.2

Hepatocyte Cell Lines

Hepatocyte cell lines derived from animal and human liver have been used for hepatotoxicity testing. Rat clone-9 hepatocytes originally isolated from the normal liver of a 4-week-old male Sprague–Dawley rat were used as a model to test chemical hepatotoxicity (Barhoumi et al., 2002; Grunne et al., 2001; Neerman et al., 2004; Reeves et al., 2001; Sahu et al., 2006; Thompson et al., 1998). Mouse BNL CL.2 hepatocytes originally isolated from an embryonic liver of a normal BALB/c mouse were used as a model for testing virulence and intracellular growth of Listeria monocytogenes (Haponsaph and Czuprynski, 1996). Human WRL-68 hepatocytes originally isolated from the normal embryonic liver of a human fetus were used to test hepatotoxicity of metals (Lopez-Ortal et al., 1999; Olivares et al., 1997; Ramirez et al., 2000). These hepatocytes are a useful tool for studying hepatitis virus replication in vitro (Tagawa et al., 1995). Human hepatocyte Hep G2 cells originally isolated from the hepatocellular carcinoma of a 15-year-old Caucasian male are widely used in hepatotoxicity testing. Scheers et al. (2001) found a good correlation between the long-term cytotoxicity results of 27 chemicals in Hep G2 cells and their acute human toxicity (Ekwall et al., 1999) in a multiple evaluation of an in vitro cytotoxicity study in Sweden. They concluded that this in vitro model is a good predictor of long-term human toxicity.

5.3

Limitations of the Hepatocyte Culture Model

The major drawbacks with the hepatocyte culture model are the loss of liver architecture and the degenerative changes following the cell isolation which make this in vitro tool unsuitable for long-term studies. However, with the improved culture conditions, hepatocytes can be maintained in culture with cell integrity and high metabolizing capacities for several weeks (Tateno and Yoshizato, 1996; Michalopoulos, 1999). Another limitation for this popular model is the absence of non-parenchymal cells, which makes it unsuitable for assessing hepatotoxicity mediated by non-parenchymal cells. This limitation is overcome by cocultures of hepatocytes with other cells (Guguen-Guillouzo et al., 1983; Donato et al.,

108

Hepatotoxicity

1994). Marked phenotypic stabilization of hepatocytes has been observed by co-culturing them with liver epithelial cells (Guguen-Guillouzo et al., 1983). An important aspect of an in vitro screening system is its validation by establishing a correlation with results obtained from in vivo studies. This can be done using appropriate animal models. With proper validation the hepatocyte model is a promising in vitro system for use in quantitative risk assessment (Walum et al., 1992).

5.4

Endpoints of Hepatotoxicity

Hepatotoxicity is not induced by any particular single mechanism; it may involve many different changes in cellular function, the detection of which provides sensitive indices of hepatotoxicity (Lee, 2003) as described below. 5.4.1

Cytotoxicity

Injury to the cell membrane leads to leakage of cytoplasmic enzymes. Therefore, cytotoxicity is a common index of this facet of hepatotoxicity. The in vitro cytotoxicity is a good predictor of acute in vivo hepatotoxicity for safety assessment (Federal Register, 2001). It can be used for screening hepatotoxins (Sahu, 2003; Dambach et al., 2005). 5.4.2

Cell Death

Cell death caused by necrosis or apoptosis is the ultimate endpoint of toxicity. Hepatocyte cultures are used to study apoptosis and cell death. Hepatocyte apoptosis is associated with retention of bile constituents in hepatocytes during cholestasis (Patel et al., 1998). The foodborne pathogen Listeria monocytogenes induces apoptosis in infected mouse hepatocytes (Rogers et al., 1996). Oxidative stress induces apoptosis in rat hepatocyte cultures (Haidara et al., 2002). The dietary carcinogen 3-amino-1,4-dimethyl-5H-pyridole[4,3-b]indole induces apoptosis in rat hepatocytes (Ashida et al., 1998). 5.4.3

Oxidative Stress

Reactive oxygen species induce oxidative stress which plays an important role in many human diseases (Sahu, 1991; McCord, 1995; De Maria et al., 1996), in the toxicity of many xenobiotics (Kehrer, 1993), and in viral (Schwarz, 1996) as well as bacterial infections of the liver (Inoue et al., 1995). Aflatoxin B1 (AFB1 ) induces oxidative stress in cultured rat hepatocytes (Shen et al., 1995,1996). AFB1 cytotoxicity in rat hepatocytes is associated with cellular lipid peroxidation (Shen et al., 1995). The flavonoid myricetin induces oxidative stress leading to lipid peroxidation and DNA strand breaks in cultured rat hepatocytes (Sahu et al., 2001). Oxidative stress leading to lipid and protein oxidation occurs in chronic hepatitis C (De Maria et al., 1996). Nitric oxide plays an important role in hepatotoxicity (Muriel, 2000). Hepatocytes express inducible nitric oxide synthatase (Muriel, 2000). The hepatotoxin acetaminophen induces nitric oxide synthatase in rat hepatocytes (Gardner et al., 1998). The human hepatoblastoma HepG2 cell cultures have been used to evaluate the protective effects of antioxidants on oxidative liver injury (Yang et al., 1999). Lipid peroxides (Stacey and Klaassen, 1981), 8-hydroxydeoxyguanosin (Shigenaga et al., 1990)


109

and dichlorofluorescein (Yerushalmi et al., 2001) produced by cellular reactive oxygen species are commonly used endpoints of oxidative stress. 5.4.4

Mitochondrial Dysfunction

Mitochondrial dysfunction is an early marker of cell stress. Mitochondrial permeability leads to release of cytochrome-C from mitochondria and induction of apopotic cell death. Hepatocyte cultures are used to screen toxins that induce mitochondrial dysfunction (Rat et al., 1994). 5.4.5

Cytochrome P450 Induction

Cytochrome P450 enzymes catalyze the oxidation of endogenous substrates and xenobiotics. Induction of these enzymes alters cellular metabolism leading to hepatotoxicity (Guengerich, 1991). This has been used to screen hepatotoxins in hepatocyte cultures (Dambach et al., 2005). 5.4.6

Hepatosteatosis

Accumulation of lipids in hepatocytes leads to ‘fatty liver’ or hepatosteatosis. Lipid accumulation can result from altered mitochondrial lipid oxidation or from altered secretion of lipoproteins. The human hepatoma cell line Hep G2 has been used for studying hepatosteatosis (McMillian et al., 2001).

5.5 5.5.1

Application of Hepatocyte Cultures Screening of Food-Related Products for Potential Hepatotoxicity

Hepatocyte cultures have been widely used for screening chemicals, drugs, carcinogens, mutagens, mycotoxins, microbial pathogens and viruses for their hepatotoxic potential (Sahu, 2003; Dambach et al., 2005). Hepatotoxicity is routinely assessed in the safety evaluation of food additives and is a common adverse health effect of food contaminants and dietary supplements. Many of the botanical herbal products labeled as dietary supplements show hepatotoxicity (Willett et al., 2004; Stickel et al., 2005). Dietary supplements such as Chaparral (Sheikh et al., 1997), Ephedra (Bajaj et al., 2003), Germander (Lekehal et al., 1996), Kava (Teschke et al., 2003), LipoKinetix (Favreau et al., 2002; Novak and Lewis, 2003) and pyrrolizidine alkaloids (Stedman, 2002) are known hepatotoxins. Rat hepatocytes have been shown to be a good in vitro model for testing hepatotoxicity of herbal products (Lekehal et al., 1996). The human hepatocellular carcinoma cell line Hep G2 was found to be an excellent model for assessing cytotoxicity of medicinal plant extracts (Ruffa et al., 2002). 5.5.2

Toxicogenomics Studies

Hepatotoxicity involves changes in gene expression, which can be detected by DNA microarrays (Minami et al., 2006). Such studies are used to identify mechanisms of toxicity. Hepatocyte culture has been used for toxicogenomics studies. Harris et al. (2001) have used the cultured human hepatoma cell line HepG2 to study the gene expression changes

110

Hepatotoxicity

induced by carbon tetrachloride and ethanol. De Longueville et al. (2003) used primary rat hepatocyte culture to study gene expression patterns in eleven different hepatotoxins. Delongchamp et al. (2005) used a primary human hepatocyte culture to evaluate gender differences in hepatotoxicity. This technology has been used to study acute hepatitis C virus infection in liver (Bigger et al., 2001) and to investigate host–microbe interactions (Cummings and Relman, 2000). 5.5.3

Interactive Hepatotoxicity Studies

Humans are exposed to many different potential toxins. Their interaction is a matter of concern. Synergistic enhancement of toxicity is not uncommon. For example, lipopolysaccharide (LPS) has been shown to influence toxicity of xenobiotics (Cebula et al., 1984; Zhou et al., 2000; Wiesenfeld et al., 2006). Rat hepatocytes have been used as an experimental model to study the interactive toxicities of carbon tetrachloride and trichloroethylene (Kefalas and Stacey, 1993). Primary rat hepatocytes have been used to study the interactive hepatotoxicity of flunitrazepam and ethanol (Assaf and Abdel-Rahman, 1999) as well as LPS and allyl alcohol (Sneed et al., 2000). Therefore, hepatocyte cultures can be used as a model to predict interactive hepatotoxicity of food-related products. 5.5.4

Studies on Species Differences in Hepatotoxicity

Hepatocyte cultures are predictive models for metabolism and pharmakokinetics in different species. Good correlation between in vitro and in vivo drug metabolism has been established using animal and human hepatocyte cultures (Sandker et al., 1994; Pathernik et al., 1995). Human, dog and rat hepatocyte cultures have been shown to be valuable in vitro tools for study of species differences in drug metabolism and pharmacokinetics (Bayliss et al., 1999). Differences in sensitivity towards chemical toxicity have been demonstrated in rat and human hepatocyte cultures (Merrill et al., 1995). Similar sensitivity towards chloroforminduced cytolethality is reported in mouse and rat hepatocytes (Amman et al., 1998). Species-dependent differences in glutathione conjugation of perchloroethylene has been studied in rat and mouse hepatocyte cultures (Lash et al., 1998). A similar metabolism of thiabendazole in rat, rabbit, calf, pig and sheep hepatocytes has been reported (Coulet et al., 1998). 5.5.5

Studies on Gender Differences in Hepatotoxicity

Gender difference plays an important role in the toxic responses of the liver (TreinenMoslen, 2001). Females are predisposed to hepatotoxicity and concomitant agents that induce cytochrome P450 enzymes also increase individual susceptibility (Stedman, 2002). Epidemiological evidence shows that women have greater susceptibility to alcohol-induced liver damage than men (Jensen, 1996; Schenker, 1997). Drug-induced acute liver failure occurs more frequently in women than men (Miller, 2001). Adverse reactions to therapeutic drugs are more common in women than men (Kando et al., 1995). The female mouse is more susceptible to fumonisin B1 -induced hepatocellular neoplasm than the male mouse (NTP, 1999). However, cocaine causes more liver damage in male mice than female mice (Visalli et al., 2004). There exists a distinct sexual dimorphism


111

in rat liver cytochrome P450 (CYP) activity (Lewis et al., 1998). Intrinsic sexual differences in hepatocytes of males and females appear to result in different levels of responses to CYP. This intrinsic sexual difference in hepatocytes is more pronounced in rats than humans. Successful use of the primary rat hepatocytes in culture as an in vitro model for evaluating the sex differences in hepatotoxicity has been demonstrated (Carfagna et al., 1996). This model has been used to show the gender difference in hepatotoxicity of safingol (Carfagna et al., 1996).

5.6

Conclusions

The hepatocyte model is an established, well-characterized and widely used popular in vitro tool for hepatotoxicity testing (Groneberg et al., 2002). The standardized methods of hepatocyte isolation from a wide range of species, including humans, and their maintenance in culture with high metabolizing capacity for several days under closely controlled and easily manipulated conditions have made this model the most widely used in vitro tool for studies on the metabolism, toxicity and mechanisms of action of chemicals and drugs (El-Sakka et al., 2002; Yamamoto et al., 2002; Dvorak et al., 2002; Shiota et al., 2002; Haidara et al., 2002; Frenzel et al., 2002; Qu et al., 2001; Muriel, 2000), for screening of mutagens, carcinogens and mycotoxins (Uhl et al., 2000; Knasmuller et al., 1997), for virulence assessment of microbial pathogens and viruses (Wing and Gregory, 2002; Galle et al., 1994), for qualitative and quantitative inter-species comparison (Bayliss et al., 1999; Ammann et al., 1998; Coulet et al., 1998; Lash et al., 1998; Merril et al., 1995; Sandker et al., 1994) and for toxicogenomics studies (Bigger et al., 2001; Harris et al., 2001; Cummings and Relman, 2000; De Longueville et al., 2003; Delongchamp et al., 2005; Minami et al., 2006). The hepatocyte model provides a homogeneous preparation of a single cell type, which can be cryopreserved without loss of significant functional activity (Chesne et al., 1993). Numerous studies have reported good correlation between in vitro hepatocytes and in vivo situations (Li, 1997; Li et al., 1997; Pathernick et al., 1995; Sandker et al., 1994). The use of in vitro models in hazard assessment requires careful consideration. At the present time, there is insufficient evidence for their use in quantitative risk assessment. However, they are suitable for use in qualitative hazard assessment, information from which can be used for quantitative risk analysis (Walum et al., 1992). In summary, the hepatocyte cell culture is a good in vitro tool for evaluating hepatotoxic potential of herbal products, dietary supplements, food additives, food-borne toxicants and microbial pathogens. The results obtained from such in vitro screening can be used to guide the development of in vivo studies to assess the safety of test materials of interest.

Acknowledgments I wish to thank Dr Thomas J. Sobotka for his critical review of this manuscript.

112

Hepatotoxicity

References Amman P., Laethem C. and Kedderis G. (1998). Choloroform-induced cytolethality in mouse and rat hepatocytes. Toxicol Appl Pharmacol 149: 217–225. Ashida H., Shiotani B., Adachi H., Kanazawa K. and Danno G. (1998). Tryptophan pyrolysis products, Trp-P-1 and Trp-P-2 induce apoptosis in rat hepatocytes. Biosci Biotech Biochem 62: 2283–2287. Assaf M. S. and Abdel-Rahman M. (1999). Hepatotoxicity of flunitrapezam and alcohol in vitro. Toxicol in vitro 13: 393–401. Bajaj J., Knox J., Komorowski R. and Saeian K. (2003). The irony of herbal hepatitis: Ma-Huanginduced hepatotoxicity. Dig Dis Sci 48: 1925–1928. Barhoumi R., Mouneimne Y., Awooda I., Safe S., Donnelly K. and Burghardt R. (2002). Characterization of calcium oscillations in normal and benzo[a]pyrene-treated clone-9 cells. Toxicol Sci 68: 444–450. Barsig J., Flesch I. and Kaufmann S. (1998). Hepatic cells as chemokine producers in murine listeriosis. Immunobiology 199: 87–104. Bayliss M. K., Bell J., Jenner W., Park G. and Wilson K. (1999). Utility of hepatocytes to model species difffrences in the metabolism of loxtidine and to predict pharmacokinetic parameters in rat, dog and man. Xenobiotica 29: 253–268. Bigger C. B., Brasky K. and Lanford R. (2001). DNA microarray analysis of chimpanzee liver during acute resolving hepatitis C virus infection. J Virol 75: 7059–7066. Carfagna M. A., Young K. and Susick R. (1996). Sex differences in rat hepatic cytolethality of the protein kinase C inhibitor safingol: Role of biotransformation. Toxicol Appl Pharmacol 137: 173–181. Castell J. V., Gomez-Lechon M. J., Ponsoda X. and Bort R. (1997). The use of cultured hepatocytes to investigate mechanisms of drug hepatotoxicity. Cell Biol Toxicol 13: 331–338. Cebula T. A., El-Hage A. N. and Ferrans V. J. (1984). Toxic interactions of benzyl alcohol with bacterial endotoxin. Infect Immun 44: 91–96. Chesne C., Guyomard C., Fautrel A., Poullain M. Fremond B., De Jong H. and Guillouzo A. (1993). Viability and function in primary cultures of adult hepatocytes from various animal species and human beings after cryopreservation. Hepatology 18: 406–414. Coulet M., Eeckhoutte C., Sutra J., Hoogenboom A., Kuiper H., Castell J., Alvinerie M. and Galtier P. (1998). Comparative metabolism of thiabendazole in cultured hepatocytes from rats, rabbots, calves, pigs and sheep. J. Agric Food Chem 46: 742–748. Cummings C. A. and Relman D. A. (2000). Using DNA microarrays to study host-microbe interactions. Emerg Infect Dis 6: 513–523. Dambach D. M., Andrews B. and Moulin F. (2005). New technologies and screening strategies for hepatotoxicity: Use of in vitro models. Toxicol Pathol 33: 17–26. Delongchamp R. R., Velasco C., Dial S. and Harris A. J. (2005). Genome-wide estimation of gender differences in the gene expression of human livers: statistical design and analysis. BMC Bioinform 6 (Suppl. 2): S13–S26. De Longueville F., Atienzar F., Marcq L., Eavard S., Leroux F., Arnould T., Remacle J. and Canning M. (2003). Use of low-density microarray for studying gene expression patterns induced by hepatotoxins on primary cultures of rat hepatocytes. Toxicol Sci 75: 378–392. De Maria N., Colantoni A., Liu G., Rogers B., Thiel D. and Floyed R. (1996). Association between reactive oxygen species and disease activity in chronic hepatitis C. Free Radic Biol Med 21: 291–295. DiPetrilllo K., Wood S., Chatfield K., Bement J., Jeffery E., Sinclair P. and Sinclair J. (2002). Effects of caffeine on acetaminophen hepatotoxicity in cultured hepatocytes treated with ethanol and isopentanol. Toxicol Appl Pharmacol 185: 91–97.


113

Donata M. T., Castell J. and Gomez-Lechon M. (1994). Cytochrome P450 activities in pure and co-cultured rat hepatocytes: Effects of model inducers. In vitro Cell Dev Biol 30A: 825–832. Dvorak Z, Ulrichova J., Pichard-Garcia L., Modriansky M. and Maurel P. (2002). Comparative effect of colchicine and colchiceine on cytotoxicity and CYP gene expression in primary human hepatocytes. Toxicol In Vitro 16: 219–227. Ekwall B., Clemendson C., Ring P. and Romert L. (1999). EDIT: a new multicenter to develop and evaluate batteries of in vitro tests for acute and chronic systemic toxicity. Altern Lab Animals 27: 339–349. El-Sakka S., Salem E. and Abdel-Rahman M. (2002). In vitro hepatotoxicity of alachlor and its byproducts. J Appl Toxicol 22: 31–35. Favreau J. T., Ryu M., Braunstein G., Park S., Love L. and Fong T. (2002). Severe hepatotoxicity associated with the dietary supplement LipoKinetix. Ann Internal Med 136: 590–595. Federal Register (2001). In vitro methods for assessing acute systemic toxicity. 66(189): 49686– 49687 (September 28). Ferrini J., Picard L., Domergue J. and Maurel P. (1997). Long-term primary cultures of adult human hepatocytes. Chem Biol Interact 107: 31–45. Frenzel J., Richter J. and Eschrich K. (2002). Fructose inhibits apaoptosis induced by reoxygenation in rat hepatocytes. Biochim Biophys Acta 1542: 82–94. Galle P. R., Hagelstein J., Kommerell B., Volkmann M., Schranz P. and Zentgraf H. (1994). In vitro experimental infection of primary human hepatocytes with hepatitis B virus. Gastroenterology 106: 664–673. Gardner C. R., Heck D., Yang C., Thomas P., Laskin J. and Laskin D. (1998). Role of nitric oxide in acetaminophen-induced hepatotoxicity in rat. Hepatology 27: 748–754. Green S., Goldberg A. and Zurlo J. (2001). Test smart – high production volume chemicals: an approach to implementing alternatives into regulatory toxicology. Toxicol Sci 63: 6–14. Groneberg D. A., Grosse-Siestrup C. and Fischer A. (2002). In vitro models to study hepatotoxicity. Toxicol Pathol 30: 394–399. Grunne T., Klotz L., Geiche J., Rudeck M. and Sies H. (2001). Protein oxidation and proteolysis by nonradical oxidants singlet oxygen or peroxynitrite. Free Rad Biol Med 30: 1243–1253. Guengerich F. P. (1991) Oxidation of toxic and carcinogenic chemicals by human cytochrome P450 enzymes. Chem Res Toxicol 4: 391–407. Guguen-Guillouzo C., Clement B., Baffet G., Beaumont C., Glaise D. and Guillouzo A. (1983). Maintenance and reversibility of active albumin secretion by adult rat hepatocytes co-cultured with other liver epithelial cell type. Exp Cell Res 143: 47–54. Guyomard C., Rialland L., Fremond B., Chesne C. and Guillouzo A. (1996). Influence of alginate gel entrapment and cryopreservation on survival and xenobiotic metabolism capacity of rat hepatocytes. Toxicol Appl Pharmacol 141: 349–356. Haidara K., Morel I., Abalea V., Gascon M. and Denizeau F. (2002). Mechanisms of tertbutylhydroperoxide-induced apoptosis in rat hepatocytes. Biochim Biophys Acta 1542: 173–185. Hamilton G. A., Westmoreland C. and George E. (2001). Effects of medium composition on morphology and function of rat hepatocytes cultured as spheroids and monolayers. In vitro Cell Dev Biol Animal 37: 656–667. Haponsaph R. and Czuprynski C. J. (1996). Inhibition of multiplication of Listeria monocytogenes in a murine hepatocyte cell line TIB-73 by IFN-γ and IFN-α Microb Pathogen 20: 287–295. Harris H. M., Fletcher S., Duggan C. and Baker V. (2001). The use of genomics technology to invetsigate gene expression changes in cultured human liver cells. Toxicol in vitro 15: 399–405. Hengstler J. G., Ringel M. J., Biefang K., Hammell S., Klebach M. and Oesch F. (2000). Cultures with cryopreserved hepatocytes: applicability for studies of enzyme induction. Chemico-Biol Interact 125: 51–73.

114

Hepatotoxicity

Hengstler J. G., Utesch D., Steinberg P., Milbert U., Gerl M., Diener B., Platt K. and Bottger T. (2002). Cryopreserved primary hepatocytes as a constantly available in vitro model for the evaluation of human and animal drug metabolism and enzyme induction. Drug Metab Rev 32: 81–118. Hill D. A. and Roth R. A. (1998) Alpha-napthylisothiocyanate causes neutrophils to release factors that are cytotoxic to hepatocytes. Toxicol Appl Pharmacol 148: 169–175. Hussain S. M. and Frazier J. M. (2003). Involvement of apoptosis in hydrazine induced toxicity in rat primary hepatocytes. Toxicol in vitro 17: 343–355. ICCVAM (1997). Validation and Regulatory Acceptance of Toxicological Test Methods – A Report of the ad hoc Interagency Co-ordinating Committee on the Validation of Alternative Methods. NIH Publication No. 97-3981, National Institute of Environmental Health Sciences, Research Triangle Park, NC, USA. Inoue S., Itagaki S. and Amano F. (1995). Intracellular killing of Listeria monocytogenes in J774.1 macrophage-like cell line and lipopolysaccaride(LPS)-resistant mutant LPS1916 cell line defective in generation of reactive oxygen intermediate after LPS treatment. Infect Immun 63: 1876–1886. Isom H. C., Secott T., Georgoff I., Woodworth C. and Mummaw J. (1985). Maintenance of differentiated rat hepatocytes in primary culture. Proc Natl Acad Sci USA 82: 3252–3256. Jensen G. (1996). Prediction of risk of liver diseases by alcohol intake, sex and age: a prospective population study. Hepatology 23: 1025–1029. Kando J. C., Yonkers K. and Cole J. (1995). Gender as risk factors for adverse events to medication. Drugs 50: 1–6. Kefalas V. and Stacey N. H. (1993). Use of primary cultures of rat hepatocytes to study interactive toxicity. Toxicol in vitro 7: 235–240. Kehrer J. P. (1993). Free radicals as mediators of tissue injury and disease. Crit Rev Toxicol 23: 21–48. Khalil M., Shariat-Panahi A., Toole R., Ryder T., Roberts E., Hodgson H. and Selden C. (2001). Human hepatocyte cell lines proliferating as cohesive spheroid colonies in alginate markedly up regulate both synthetic and detoxificatory liver function. J Hepatol 297: 68–77. Knasmuller S., Bresgen N., Gelderblom W., Zohrer E. and Eckl P. (1997). Genotoxic effects of three Fusarium mycotoxins in bacteria and in primary culture of rat hepatocytes. Mut Res 391: 39–48. Kono Y., Yang S., Letarte M. and Roberts E. (1995). Establishment of a human hepatocyte line derived from primary culture in a collagen gel sandwich culture system. Exp Cell Res 221: 478–485. Lash L. H., Qian D., Desai K., Elfarra A., Sicuri A. and Parker J. (1998). Glutathione conjugation of perchloroethylene in rat and mice in vitro. Toxicol Appl Pharmacol 150: 49–57. Lee W. M. (2003). Drug-induced hepatotoxicity. New Engl J Med 349: 474–485. Lekehal M., Pessayer D., Lereau J., Moulis C., Foureste I. and Fau D. (1996). Hepatotoxicity is the herbal medicine germander. Hepatology 24: 212–218. Lewis D. F. V., Ioannides C. and Parke D. V. (1998). Cytochromes P450 and species differences in xenobiotic metabolism and activation of carcinogens. Environ Health Perspect 106: 633–641. Li A. P. (1997). Primary hepatocyte cultures as an in vitro experimental model for the evaluation of pharmacokinetic drug–drug interactions. Adv Pharmacol 43: 103–130. Li A. P., Reith M., Rasmussen A., Hall S., Xu L., Kaminski D. and Cheng L. (1997). Primary human hepatocytes as a tool for the evaluation of structure–activity relationship in cytochrome P450 induction potential of xenobiotics. Chem Biol Interact 107: 17–30. Lopez-Ortal P., Souza V., Bucio L., Gonzalez E. and Gutierrez-Ruitz M. (1999). DNA damage produced by cadmium in a human fetal hepatic cell line. Mutat Res 439: 301–306. MacGregor J. T., Collins J. M., Sugiyama Y., Tyson C. A., Dean J., Smith L., Andersen M., Curren R., Houston J., Kadlubar F., Kedderis G. L., Parchment R., Thummel K., Ulrich R., Vickers E. and Wrighton S. A. (2001). In vitro human tissue models in risk assessment: report of a consensusbuilding workshop. Toxicol Sci 59: 17–36. McCord J. M. (1995). Superoxide radical. Proc Soc Exp Biol Med 209: 112–117.


115

McMillian M. K., Grant E., Shong Z., Parker J. Li L. and Johnson M. (2001). Nile red binding to Hep G2 cells: an improved assay for in vitro studies of hepatosteatosis. In vitro Mol Toxicol 14: 177–190. Merrill J. C., Beck D. J., Kaminski D. A. and Li A. P. (1995). Polybrominated biphenyl induction of cytochrome P450 mixed function oxidase activity in primary rat and human hepatocytes. Toxicology 99: 147–152. Michalopoulos G. K. (1999). Hepatocyte primary cultures: Currently used systems and their applications for studies of hepatocyte growth and differentiation. Tissue Eng Meth Protocol 18: 227–243. Miller M. A. (2001). Gender-based differences in toxicity of pharmaceuticals – the Food and Drug Administration’s perspective. Int J Toxicol 20: 149–152. Miller E. C. and Miller J. A. (1986). Carcinogens and mutagens that may occur in foods. Cancer 58: 1795–1803. Minami K., Maniratanachote R., Katoh M., Nakajima M. and Yokoi T. (2006). Simultaneous measurement of gene expression for heptotoxicity in thioacetamide-administered rats by DNA microarrays. Mutat Res Genet Toxicol Environ Mutag 603: 64–73. Muriel P. (2000). Regulation of ntiric oxide synthesis in the liver. J Appl Toxicol 20: 189–195. Neerman M. F., Zhang W., Parrish A. and Simanek E. (2004). In vitro and in vivo evaluation of a melamin dendrimer as a vehicle for drug delivery. Int J Pharmaceut 281: 129–132. Novak D. and Lewis J. (2003). Drug-induced liver disease. Curr Opin Gastroenterol 19: 203–215. NTP (1999). Toxicology and carcinogenesis studies of fumonisin B1 in rats and mice. National Toxicology Program, National Institutes of Health, Summary Report TR-496, Bethesda, MD, USA, 7–9. Olivares I. P., Bucio L., Souza V., Carabez A. and Gutierrez-Ruiz M. (1997). Comparative study of the damage produced by acute ethanol and acetaldehyde treatment in a human fetal hepatic cell line. Toxicology 120: 133–144. Patel T., Roberts L., Jones B. and Gores G. (1998). Dysregulation of apoptosis as a mechanism of liver disease: An overview. Semin Liver Dis 18: 105–114. Pathernik S., Smihmid J., Sauter T., Schidberg F. and Koebe H. (1995). Metabolism of pimobendan in long-term human hepatocyte culture: in vivo–in vitro comparison. Xenobiotica 25: 811–823. Qu B., Li L., Wong K., Tan T. and Halliwell B. (2001). Mechanism of clofibrate hepatotoxicyt. Free Rad Biol Med 31: 659–669. Ramirez P., Razo L., Gutierrez-Ruiz M. and Gonsebatt M. (2000) Arsenite induces DNA-protein crosslinks and cytokeratin expression in the WRL-68 human hepatic cell line. Carcinogenesis 21: 701–706. Rat P., Korwin-Zmijowska C., Warnet J. and Adolphe M. (1994). New in vitro fluorometric microtitration assays for toxicological screening. Cell Biol Toxicol 10: 329–337. Reeves W. R., Barhoumi R., Burghardt R., Lemke S., Mayura K., McDonald T., Philips T. and Donnelly K. (2001). Evaluation of methods for predicting toxicity of polycyclic aromatic hydrocarbon mixtures. Environ Sci Technol 35: 1630–1636. Rogers H. W., Callery M. P., Deck B. and Unanue E. R. (1996). Listeria monocytogenes induces aptosis of infected hepatocytes. J Immunol 156: 679–684. Ruffa M. J., Ferraro G., Wagner M., Calcagno M., Campos R. and Cavallaro L. (2002). Cytotoxic effects of Argentine medicinal plant extracts on human hepatocellular carcinoma cell line. J Ethnopharmacol 79: 335–339. Runge D., Michalopoulos G., Strom S. and Runge D. M. (2001). Recent advances in human hepatocyte culture systems. Biochem Biophys Res Commun 274: 1–3. Sahu S. C. (1991). Role of oxygen free radicals in the molecular mechanisms of carcinogenesis: A review. J Environ Sci Health C 9: 83–112. Sahu S. C. (2003). Hepatocyte culture as an in vitro model for evaluating the hepatotoxicity of food-borne toxicants and microbial pathogens: a review. Toxicol Mechan Meth 13: 111–119.

116

Hepatotoxicity

Sahu S. C., Flynn T., Bradlaw J., Roth W., Barton C. and Yates J. (2001). Pro-oxidant effects of the flavonoid myricetin on rat hepatocytes in culture. Toxicol Meth 11: 277–283. Sahu S. C., Ruggles D. I. and O’Donnell M. W. (2006). Pro-oxidant activity and toxicity of nordihydroguaiaretic acid in clone-9 rat hepatocyte cultures. Food Chem Toxicol 44: 1751–1757. Sandker G., Vos R., Delbressine L., Sloof M., Meijer D. and Groothuis G. (1994). Metabolism of three pharmacologically active drugs in isolated human and rat hepatocytes. Xenobiotica 24: 143–155. Scheers E.M., Ekwall B. and Dierickx P. J. (2001). In vitro long-term cytotoxicity testing of 27 MEIC chemicals on HepG2 cells and comparison with acute human toxicity data. Toxicol in vitro 15: 153–161. Schenker S. (1997). Medical consequences of alcohol abuse: is gender a factor? Alcohol Chem Exp Res 21: 179–181. Schwarz K. B. (1996). Oxidative stress during viral infection: A review. Free Rad Biol Med 21: 641–649. Seglen P. O. (1976). Preparation of isolated rat liver cells. Meth Cell Biol 13: 29–83. Sheikh N. M., Philen R. and Love L. (1997). Chaparral-associated hepatotoxicity. Arch Int Med 157: 913–919. Shen H. M., Ong C. N. and Shi C. Y. (1995). Involvement of reactive oxygen species in aflatoxin B1 -induced cell injury in cultured rat hepatocytes. Toxicology 99: 115–123. Shen H. M., Shi C. Y., Shen Y. and Ong C. N. (1996). Detection of elevated reactive oxygen species level in cultured rat hepatocytes treated with aflatoxin B1 . Free Radic Biol Med 21: 139–146. Shigenaga M. K., Park J., Cundy K., Gimeno A. and Ames B. (1990). In vivo oxidative DNA damage: measurement of 8-hydroxydeoxyguanosis. Meth Enzymol 186: 521–530. Shiota G., Maeta Y., Mukpyama T. Oyama K., Yashima K., Ito H., Murawaki Y. and Kawasaki H. (2002). Effects of Sho-saiko-to on hepatocarinogenosis and 8-hydroxy-21 -deoxyguanosine formation. Hepatology 35: 1125–1133. Sneed R. A., Buchweitz J., Jean P. and Geney P. (2000). Pentoxifylline attenuates bacterial polysaccharide–induced enhancement of allyl alcohol hepatotoxicity. Toxicol Sci 56: 203–210. Stacey N. H. and Klaassen C. D. (1981). Inhibition of lipid peroxidation without prevention of cellular injury in isolated rat hepatocytes. Toxicol Appl Pharmacol 58: 8–18. Stedman C. (2002). Herbal hepatotoxicity. Seminars Liver Dis 22: 195–206. Stickel F., Patsenker E. and Schuppan D. (2005). Herbal hepatotoxicity. J Hepatol 43: 901–910. Susa N., Ueno S., Furukawa Y., Ueda J. and Sugiyama M. (1997). Potent protective effect of melatonin on chromium (vi)-induced DNA strand breaks in cultures of rat hepatocytes. Toxicol Appl Pharmacol 144: 377–384. Tagawa M., Kato N., Yokosuka O., Ohto M. and Omata M. (1995). Infection of human hepatocyte cell line with hepatitis c virus. J Gastro Hepato 10: 523–527. Tateno C. and Yoshizato K. (1996). Long-term cultivation of adult rat hepatocytes that undergo multiple cell division. Am J Pathol 148: 383–392. Teschke R., Gaus W. and Loew D. (2003). Kava extracts: safety and risks including rare hepatotoxicity. Phytomedicine 10: 440–446. Thompson D. C., Barhoumi R. and Burghardt R. (1998). Comparative toxicity of eugenol and its quinine methide metabolites in cultured liver cells. Toxicol Appl Pharmacol 149: 55–63. Tirmenstein M. A., Nicholls-Grzemski F., Zhang J. and Farris M. (2000). Glutathione depletion and production of reactive oxygen species in isolated rat hepatocytes. Chem Biol Interact 127: 201–217. Treinen-Moslen M. (2001). Toxic responses of the liver, in C. D. Klaassen (ed.), Casarett and Doull’s Toxicology: The Basic Sciences of Poison, Fourth Edition, McGraw-Hill Professional, New York, NY, USA, 471–490. Tseng T., Kao E., Chu C., Chou F., Wu H. and Wang C. (1997). Protective effects of dried flower extracts of Hibiscus sabdariffa against oxidative stress in rat hepatocytes. Food Chem Toxicol 35: 1159–1164.


117

Uhl M., Helma C. and Knasmuller S. (2000). Evaluation of the single cell electrophoresis assay with HepG2 cells. Mutat Res 468: 213–225. Visalli T., Turkall R. and Abdel-Rahman M. (2004). Cocaine hepatotoxicity. Int J Toxicol 23: 163–170. Walker T. M., Rhodes P. and Westmoreland C. (2000). The differential cytotoxicity of methotrexate in rat hepatocyte monolayer and spheroid cultures. Toxicol in vitro 14: 475–485. Walum E., Balls M, Bianchi V., Blauboer B., Bolcsfoldi G., Guillouzo A., Moore G., Oldand L., Reinhardt C. and Spielman H. (1992). ECITTS: an integral approach to the application of in vitro test systems to the hazard assessment of chemicals. Alt Labor Animals 20: 406–428. Wiesenfeld P. W., Garthoff L. H., Sobotka T. J., Suagee J. and Barton C. N. (2007). Acute oral toxicity of colchicine in rats: effects of gender, vehicle matrix and pre-exposure to lipopolysaccharide. J Appl Toxicol 27: 421–433. Willett K. L., Roth R. A. and Walker L. (2004). Workshop overview: hepatotoxicity assessment of Botanical Dietary Supplements. Toxicol Sci 79: 4–9. Wing E. J. and Gregory S. H. (2002). Listeria monocytogenes: clinical and experimental update. J Infect Dis 185: S18–S24. Yamamoto Y., Yamazaki H., Ikeda T., Watanabe T., Iwabuchi H., Nakajima M. and Yokoi T. (2002). Formation of novel quinone epoxide metabolites of troglitazone with cytotoxicity to HepG2 cells. Drug Metab Dispos 30: 155–160. Yang C., Shen H. and Ong C. (1999). Protective effect of Ebselen against hydrogen peroxide-induced cytotoxicity and DNA damage in HepG2 cells. Biochem. Pharmacol 57: 273–279. Yerushalmi B., Dahl R., Devereaux M., Gumpricht E. and Sokol R. (2001). Bile acid-induced rat hepatocyte apoptosis in inhibited by antioxidants. Hepatology 33: 616–626. Zhou H., Harkem J., Hotchkiss J., Yan D., Roth R. and Pestka J. (2000). Lipopolysacchardie and trichothecene vomitoxin (deoxynivalenol) synergistically induce apoptosis in murine lymphoid organs. Toxicol Sci 53: 252–263.

6 Some Experimental Models of Liver Damage Pablo Muriel

6.1

Introduction

The liver is the largest gland in the human body and accounts for approximately 2.5 % of total body weight. The liver weighs almost 1500 kg in the adult. The liver is divided into four lobes. The right and left lobes are separated by the falciform ligament. These lobes are supplied by the right and left branches of the hepatic artery and the portal vein. Bile is drained from the liver by the left and right hepatic ducts. The right lobe is further divided into two smaller lobes: the quadrate and the caudate. The liver is a unique organ anatomically located to serve its dual role as a metabolic and biochemical transformation ‘factory’. The liver receives blood containing substances absorbed or secreted by the gastrointestinal organs including the spleen, intestine, stomach and pancreas. It uses these substances as raw materials and modifies them or synthesizes new chemicals. These are then returned to the blood stream or to bile for excretion. The vulnerability of the liver to injury is as much a function of its anatomical proximity to the blood supply and digestive tract as to its ability to concentrate and biotransform xenobiotics. Xenobiotics in the blood pass through the hepatic artery and portal vein and then drain through the central and the hepatic veins into the vena cava. The principal liver duct joins the cystic duct from the gall bladder to form the common bile duct, which drains into the duodenum. The literature on experimental hepatotoxicity has become enormous (Zimmerman, 1999). A fascinating area in its own right, it includes many facets relevant to human and veterinary medicine. Experimentally induced liver damage allows studies of human liver diseases,


120

Hepatotoxicity

accidental and industrial toxicity, screening of medicinal agents for potential hepatotoxic effects, studies of hepatic physiology, pathology, and regeneration, and the development of diagnostic tools. Models of liver damage can provide useful tools for the study of teaching of hepatic histopathology. Almost all of the known acute and chronic lesions of the liver can be induced experimentally. These include necrosis (zonal, massive or diffuse), steatosis, hepatic venular lesions and cirrhosis of several morphological types. Specific lesions such as sinusoidal ‘free’ acidophilic bodies (apoptotic bodies), nucleolar and nuclear abnormalities and megalocytes can be produced by several experimental models of hepatic injury. Cholestasis can also be produced experimentally. Models of liver damage have used whole animals or various in vitro preparations. Use of whole animals is essential to demonstrate that an agent has an adverse effect on the liver in a setting of physiologic significance. Whole animals also must be employed if the effects of various factors and manipulations on disease are to have meaning for the mechanism of injury and for the pathophysiologic impact of the hepatic injury. In this review, we will focus on in vivo experimental models of hepatic injury. Studies have used a variety of animal species. Most popular have been rats because of their size and relatively low cost. Most of the accumulated information related to liver damage and on modifiers of susceptibility, such as age, sex and diet, applies to the rat (Zimmerman, 1999). The general use of a relatively uniform experimental model permits comparison of results from different laboratories. Mice, guinea pigs, hamsters, rabbits, gerbils, cats, dogs, horses, sheep, etc., have been employed to various degrees. Primates have also been used because of the greater similarity to disease in humans. In the following section, the most common models of liver injury (Figure 6.1) will be described.

Figure 6.1 The most common models of experimental liver damage. Note that in some cases the causative agent is the same as in human diseases while in others it is not: BDL, bile duct ligation

Experimental Models of Liver Damage

6.2

121

Carbon Tetrachloride

Carbon tetrachloride (CCl4 ) once was used widely as a solvent, cleaner and degreaser, both for industrial and home use. Today, the scientific database on the effects of haloalkanes is so vast that it is no longer employed for such purposes although it is used as a model of experimental liver injury (Weber et al., 2003). Today, CCl4 proves highly useful as an experimental model for the study of certain hepatotoxic effects (Muriel, 1997; Muriel et al., 2003; Moreno and Muriel, 2006). It consistently produces liver injury in many species, including non-human primates (Kumar et al., 1972; Yoshida et al., 1999). CCl4 -induced toxicity, depending on dose and duration of exposure, covers a variety of effects. At low doses, transient effects prevail, such as loss of Ca2+ homeostasis (Muriel and Mourelle, 1990), lipid peroxidation (Muriel, 1997), release of noxious or beneficial cytokines (Kyung-Hyun et al., 2006; Muriel, 2007) and apoptotic events followed by regeneration. Other effects, with higher doses or longer exposure, are more serious and develop over a longer period of time, such as fatty degeneration, fibrosis, cirrhosis and even cancer (Weber et al., 2003). In addition, acute intoxication with CCl4 at high doses, when the hepatocelular necrosis exceeds the regenerative capacity of the liver, fatal liver failure will ensue. Extreme doses of CCl4 result in nonspecific solvent toxicity, including central nervous system depression and respiratory failure and death (Berger et al., 1986). r CCl4 -metabolism begins with the formation of the trichloromethyl free radical, CCl3 (McCay et al., 1984) through the action of the mixed function cytochrome P450 oxygenase system of the endoplasmic reticulum (Recknagel et al., 1989). This process involves reductive cleavage of a carbon–chlorine bond. Free radical activation of CCl4 in mitochondria has also been observed (Tomasi et al., 1987) and may contribute significantly to its toxicity. The major cytochrome isozyme to execute biotransformation of CCl4 is CYP2E1, but CYP2B1 and CYP2B2 are also able to attack the CCl4 molecule (Gruebele et al., 1996). In humans, CYP2E1 dominates CCl4 metabolism at environmentally relevant concentrations, but at higher concentrations other cytochromes, particularly CYP3A, also contribute importantly r (Zangar et al., 2000). The resulting CCl3 radical is reactive enough to bind covalently to CYP2E1, either to the active site of the enzyme (Manno et al., 1992) or to the heme group (Fernandez et al., 1982), thus causing its own inactivation (Fuji, 1997). r The CCl3 radical reacts with several important biological substances, like fatty acids r proteins, lipids, nucleic acids and amino acids (Weber et al., 2003). CCl3 also acts by abstracting hydrogen from unsaturated fatty acids to form chloroform. DNA adducts is a mechanism for CCl4 -induced carcinogenesis (DiRenzo et al., 1982). 6.2.1

Lipid Peroxidation r The CCl3 , radical in the presence of oxygen, is transformed into the trichloromethyl peroxy r r radical, CCl3 OO . This radical is more reactive and thus shorter lived than the CCl3 radical r (Mico and Pohl, 1983). The lifetime of the CCl3 OO radical is in the millisecond range and disappears from the tissue by reacting with suitable substrates to complete its electron r r pair. CCl3 OO is more likely than CCl3 to abstract hydrogen from polyunsaturated fatty acids (PUFAs) (Forni et al., 1983) leading to lipid peroxidation (Comporti, 1985; Trible et al., 1987). The abstraction of hydrogen from PUFAs starts sequential reactions that finish in the complete disintegration of the fatty acid molecules with the consequent formation

122

Hepatotoxicity

of aldehydes, other carbonyls and alkanes in a process called lipid peroxidation. It is important to point out that lipid peroxidation may produce injury by compromising the integrity of membranes and by covalent binding of reactive intermediates to important biological molecules. 6.2.2

Aldehyde Toxicity

The products of lipid peroxidation may produce further liver damage (Esterbauer et al., 1982,1991). They may inhibit enzymes (Hruszkewycz et al., 1978), damage DNA (Ueda et al., 1985), and block lipoprotein secretion. Reactive aldehydes first appear in liver about 6 h after CCl4 intoxication, show a maximum at 24 h and disappear by 36 to 72 h latter r r (Hartley et al., 1999). Aldehydes are less reactive than CCl3 or CCl3 OO , and thus they may reach other organs. However, they do not usually reach toxic concentrations. This is because of their unspecific action because they react typically with amino or sulfhydryl groups, and so a large number of groups need to be inactivated before measurable damage occurs (Sundari and Ramakrishna, 1997). The activities of aldehydes displaying a 4-hydroxy-2,3-trans-unsaturated configuration and especially of 4-OH-2,3-trans-nonenal, is favored, and they are formed in high concentrations after exposure of cells to CCl4 (Benedetti et al., 1980, 1984). These compounds inhibit several enzymes, including adenylate cyclase, at pathophysiological concentrations (Poli et al., 1989). There is also evidence that 4-hydroxynonenal decreases cytochrome P450 (Poli et al., 1989) and inhibits the activities of Ca2+ (Parola et al., 1990) and Na+ , K+ (Morel et al., 1998) –ATPases, and phosphatase and protein kinase activities (Omura et al., 1999). The effect of CCl4 -derived haloalkanes on these enzyme activities may contribute to CCl4 -induced toxicity

6.3

Hypomethylation

It has been reported that in connection with CCl4 -induced inhibition of protein synthesis an important deficit in ribosomal RNA methylation occurs (Clawson et al., 1987). The synthesis of polyamines is also reduced by CCl4 intoxication (Rao et al., 1989). Remethylation of CCl4 intoxicated ribosomes recuperates their normal function (Clawson et al., 1987). These researchers suggest that disturbance of cytosolic Ca2+ levels are to blame for the effect. Carbon tetrachloride inhibits DNA methylation (Ruchirawat et al., 1983) and treatment of rats for three weeks leads to global hypomethylation of liver DNA; this deficiency can be corrected by external supplementation of the methyl donor S-adenosylmethionine (SAM) (Varela-Moreiras et al., 1995). Moreover, several alterations observed in CCl4 -induced cirrhosis, mainly on lipids, can be completely prevented by SAM treatment (Muriel and Mourelle, 1992a). 6.3.1

Calcium Homeostasis

It is very well known that CCl4 increases intracellular Ca2+ levels in hepatocytes (Weber et al., 2003). This effect compromises the viability of the cell because the intracellular concentration of Ca2+ is maintained about four orders of magnitude lower than the extracellular space.


123

An early effect is observed only 15 min after the administration of CCl4 to hepatocytes from mitochondria (Albano et al., 1985) and endoplasmic reticulum because the mitochondrial uniport system and the Ca2+ -ATPase of the endoplasmic reticulum become inactivated (Tsokos-Kuhn et al., 1985). This results in increased cytosolic Ca2+ (Long and Moore, 1986). However, mitochondria and endoplasmic reticulum may be responsible for determining the level of intracellular calcium in the short term, but these organelles cannot continue to accumulate Ca2+ within themselves indefinitely; then, in the long term, the plasma membrane Ca2+ -ATPase must be responsible for maintaining the ultimate level of Ca2+ within the cell and CCl4 strongly inhibits this ATPase activity (Muriel and Mourelle, 1992b). Increased levels of Ca2+ in the cell destroy cytoskeletal structures and activate several catabolic enzymes such as endonucleases, proteases and phospholipases, whose continued action results in cell death via apoptosis or necrosis (Nicotera et al., 1992; Zawaski et al., 1993). 6.3.2

Acute CCl4 Intoxication

Acute intoxication in rats or mice are frequently used to produce toxic hepatitis in order to test possible hepatoprotective agents, mainly those with antioxidant properties, and to study liver regeneration after a chemical hepatectomy (Muriel and Mourelle, 1992c; Sotelo-Félix et al., 2002; Agarwal et al., 2006). 6.3.3

CCl4 -Induced Cirrhosis

Lamson and Wing (1926) were the first to report that CCl4 intoxication produced cirrhosis, while later Cameron and Karunaratne (1932) published a systematic study that established the morphology and show the standard experimental conditions of the model. Nowadays, CCl4 -induced cirrhosis is probably the most widely used model to reproduce cirrhosis in rats and mice (Moreno and Muriel, 2006). To produce experimental cirrhosis, it is necessary to give repeated doses of CCl4 , regardless of the animal and route of administration. The time interval between each ‘insult’ should not be too long, since damaged liver can recover, rendering the effect of the toxin noncumulative (Cameron and Karunaratne, 1932). In our laboratory, cirrhosis is induced in male Wistar rats weighing initially around 100 g – CCl4 is dissolved in liquid petroleum and administered three times per week intraperitoneally at a dose of 0.4 g/kg. Cirrhosis is observed after two months and severity of the disease increased after three months of treatment; after four months, mortality is too high (over 80 %) (Muriel et al., 2005). Well-established CCl4 -cirrhosis is characterized by changes involving several abdominal organs. The liver is frequently enlarged; however, in advanced states of the disease, it may be smaller than normal with gross nodularity. Splenomegaly and ascites are usually present. Increased mesenteric fat lobulation and lymph node hyperplasia frequently occur. Two main problems with CCl4 -cirrhosis are that the response of individual animals to the toxin is variable and that mortality is high (30–60 %); interestingly, these factors also occur in human cirrhosis. Experimental CCl4 -induced cirrhosis seems to reproduce the main characteristics of human cirrhosis: the liver is grossly nodular, there is portal hypertension in most of the animals and the normal architecture is replaced by nodules of regenerating liver surrounded by fibrotic tissue septa with proliferated bile ducts. Portocabal anastomosis develops within

124

Hepatotoxicity

the connective tissue septa (Daniel et al., 1952), as occurs in man. However, a deeper analysis shows important differences between this experimental model and its human counterpart (Pérez-Tamayo, 1983). Alcoholic cirrhosis is frequently preceded or accompanied by centrilobular sclerosing hyaline necrosis and pericelular and interlaminar fibrosis; these characteristics are not observed in the CCl4 model of cirrhosis. In summary, this model of experimental cirrhosis shares several features with alcoholic human cirrhosis but also shows various differences (Pérez-Tamayo, 1983). The CCl4 -induced cirrhosis is the most widely model used to reproduce human cirrhosis in experimental animals to study pathophysiological mechanisms of the disease (Muriel, 1997, 2006, 2007; Muriel and Escobar, 2003; Moreno and Muriel, 2006; Chavez et al., 2006) and many possible antifibrotic drugs are tested in this model before using them in humans (Cedillo et al., 1996; Muriel et al., 2003; Kyung-Hyun et al., 2006).

6.4

Acetaminophen (Paracetamol)

Acetaminophen (paracetamol; APAP) in normal therapeutic doses is generally considered one of the safest aspirin-like analgesics. However, large overdoses can cause severe hepatic damage and even liver failure. Overdose of APAP is the most frequent cause of druginduced hepatic failure in USA and UK (Lee, 2004). In addition to its clinical relevance, APAP experimental intoxication is widely used as a model of liver damage. Recently, two excellent reviews have appeared on APAP-induced hepatic injury; one of them is concerned with the role of inflammation (Jaeschke, 2005a) while the other deals with intracellular signaling (Jaeschke and Bajt, 2006) in the mechanism of APAP-induced liver damage. Therefore, here I am only going to summarize the most prominent features of APAP overdosing. Since APAP toxicity is associated with depletion of GSH (Jollow et al., 1973; Mitchell et al., 1973a,b), the GSH precursor N -acetylcysteine (NAC), introduced in the 1970s, is an effective antidote against APAP hepatic damage. However, no other effective drug is available in the clinic today. The inflammatory response is dependent on the initial damage to the hepatic cells; therefore, intracellular events may offer interesting targets for therapeutic interventions. When an overdose occurs, a fraction of APAP is activated metabolically to the reactive metabolite N -acethyl- p-benzoquinone imine (NAPQI) that first consumes cellular glutathione and then binds covalently to proteins. This leads to increases in the intracellular Ca2+ concentration, ‘Bax’ and ‘Bid’ translocation to the mitochondria and oxidant stress and peroxynitrite formation in the mitochondria. Reactive oxygen species and peroxynitrite induce the membrane permeability transition that causes the collapse of the mitochondrial membrane potential and ‘abolishes’ ATP synthesis. In fact, some free radical scavengers prevent experimental APAP induced liver injury (Muriel et al., 1992; Muriel, 1997). The decreasing of ATP levels appears to prevent caspase activation by the release of cytochrome c and Smac. The apoptosis-inducing factor (AIF) (Susin et al., 2000) and endonuclease G (van Loo et al., 2001) induce chromatin condensation and nuclear DNA fragmentation, respectively. The extensive DNA damage and the rapid elimination of mitochondria, together with activation of intracellular proteases (calpains), conduce to plasmatic membrane breakdown and oncotic necrosis of the hepatic cells. The massive cell death and liver failure after APAP intoxication can be explained by these intracellular events. Many aspects,


125

however, require further investigation. Furthermore, APAP-induced hepatic injury can be modulated by changes in the activity of cytochromes P450 and Phase II detoxification enzymes, modulation in the GSH/GSSG levels and redox state of the cell. Experimental APAP intoxication is a useful model to study the pathophysiology and pharmacology of an overdose of the drug in humans. In addition, it provides a useful tool to study necrosis and apoptosis mechanisms and the effects of hepatoprotective drugs. For example, recently, the receptor antagonist (BN52021) to platelet activating factor (PAF) prevented acute APAP toxicity in rats (Grypioti et al., 2006). This was explained because PAF is an endogenous mediator of various proinflammatory processes in the liver.

6.5

Cholestasis

Bile is the exocrine secretion of the liver. Like most other secretions, it is an aqueous solution of organic and inorganic compounds. Bile acids, bile pigments, cholesterol and phospholipids are the major organic compounds. Bile also contains small amounts of protein. Bile is normally secreted from the hepatocyte to the duodenum, a process that normally depends on the function of several membrane transport systems in hepatocytes and cholangiocytes, and on the functional and structural integrity of the bile-secretory apparatus. Cholestasis comes from the Greek ‘chole’ which means bile and ‘stasis’, standing still; thus meaning bile flow stagnation that can be the result of a failure in the transports of hepatocytes or ductular cells or from a physical blockade of the external bile ducts. The former is considered intrahepatic, and the latter extrahepatic cholestasis. 6.5.1

Human Cholestasis

The main etiologic agents of adult intrahepatic cholestasis are drugs and pregnancy. Antibiotics are the most frequent medications that cause intrahepatic cholestasis. In addition, estrogens, cyclosporine A, rifamycin SV, rifampicin, glibenclamide, chlorpromazine, erythromycin and oxypenicillins may produce intrahepatic cholestasis (Rodr´ıguez-Garay, 2003). Obstructive cholestasis, extrahepatic biliary atresia (EBA), primary sclerosing cholestasis (PSC) and primary biliary cirrhosis (PBC) are other causes of human cholestasis. PBC, an autoimmune cholestatic liver disease, is accompanied by reduced levels of anion exchanger 2 immunoreactivity at the bile canaliculus and bile ducts (Medina et al., 1997). A stone or a tumor at the level of the extrahepatic bile ducts usually leads to extrahepatic cholestasis. The use of animals to reproduce cholestatic disease in the human is a useful tool to study the pathophysiology of the disease and also to test potential liver drugs with anticholestatic properties. 6.5.2

Experimental Prolonged Bile Duct Ligation (Biliary Cirrhosis)

The effects of conventional bile duct ligation (BDL) on hepatic morphology have been examined extensively since publication of the earliest histological studies approximately 70 years ago (Cameron and Oakley, 1932; Cameron and Hasan, 1958; Trams and Symeonidis, 1957). Kountouras et al. (1984) reported a systematic study of prolonged BDL as a model for cirrhosis in the rat. They subjected rats to double ligation of the common bile duct,

126

Hepatotoxicity

with section between the two ligatures. Animals were studied after five to fifty two days of BDL. After five days of BDL, there was bile duct proliferation, mild edema and an acute inflammatory reaction in portal areas. Fibrosis was slight and limited to the regions of new bile duct formation. A light infiltrate of neutrophil leucocytes around the bile ducts was observed. Hepatocytes contained numerous mitotic figures and necrosis was also observed. BDL for ten days or more was accompanied by fibrosis and bile duct proliferation in portal areas. Bile infarcts were occasionally noted, mainly in portal areas. Biliary obstruction for fifteen days induced the production of fibrous septa bridged portal areas extended into the lobules. After twenty eight days of bile flow obstruction, most of the animals showed cirrhosis; concentric fibrosis was observed around the interlobular and septal bile ducts. All rats ligated for thirty or more days showed cirrhosis with ascites (Kountouras et al., 1984). As mentioned before, the most commonly used method to induce experimental cirrhosis is chronic administration with CCl4 which shares several characteristics with human alcoholic cirrhosis but also shows some important differences (Pérez-Tamayo, 1983). In addition, the individual response of animals is variable, mortality is high and the method takes a relatively long time, 8–12 weeks, to produce cirrhosis (Muriel et al., 2005). In addition, nowadays CCl4 intoxication is extremely rare in man. In the model of prolonged bile duct obstruction, the morphological changes are comparable to those in human biliary cirrhosis. In addition, in the BDL model no bioactivation of any external toxin is required as it occurs in the CCl4 model. This obstructive model seems to have some advantages over the CCl4 and may prove to be a useful tool for studying human cirrhosis (Fernández-Mart´ınez et al., 2006) and to prove possible antifibrotic drugs (Fernández-Mart´ınez et al., 2001). However, both models possess advantages and disadvantages, for example, administration of CCl4 is easier to perform than to carry out a surgery. Reversion studies are easier in the CCl4 model because it is easier to stop giving the noxious stimuli. Morphologically, the CCl4 model resembles alcoholic cirrhosis, while the BDL model is similar to biliary cirrhosis. In CCl4 cirrhosis, oxidant stress in the main mechanism of liver damage (Comporti, 1985; Muriel, 1997), while in the BDL model oxidant stress occurs as a consequence and not as a cause of hepatic injury (Muriel and Suárez, 1994; Muriel, 1997; Barón and Muriel, 1999). Therefore, both models are useful tools to study cirrhosis, for example, if an antifibrotic drug functions in both models it has more possibilities to work in man. 6.5.3

Estrogen-Induced Cholestasis

Estrogens are known to cause intrahepatic cholestasis during pregnancy in susceptible women, who are using oral contraceptives or who are on postmenopausal hormone replacement therapy (Schreiber and Simon, 1983). Intrahepatic cholestasis experimentally produced by intoxication with 17α-ethynylestradiol (EE) in rodents is a widely used model to investigate the pathophysiological mechanisms of intrahepatic cholestasis. Treatment with EE diminishes the ATP-dependent taurocholate transport in the canalicular membrane of the hepatocyte that is thought to be due to impaired expression of the canalicular bile salt export pump (BSEP) (Lee et al., 2000). In addition, EE treatment decreases uptake of bile acids in the sinusoidal membrane by ‘down-regulating’ the expression of the Na+ /taurocholate cotransporting polypeptide (NTCP) (Simon et al., 1996). The mentioned


127

evidence indicates that EE produce cholestasis by decreasing the efflux and influx of bile acids in hepatic cells, leading to a decreased bile flow. The homeostasis of bile acids is regulated by several nuclear receptors, including farnesoid X receptor (FXR), pregnane X receptor (PXR), constitutive/androstane receptor (CAR) and estrogen receptor α (ERα) (Francis et al., 2003; Guo et al., 2003; Yamamoto et al., 2006). There is evidence that bile acids decrease bile acid biosynthesis by ‘downregulating’ transcription of the rate-limiting CYP7A1 through the FXR-SHP (short heterodimer partner)-LRHI (liver receptor homolog 1) cascade (Sinal et al., 2000). Responsible for bile acid uptake into the hepatocytes, NTCP is repressed by FXR activation (Sinal et al., 2000). At the same, time FXR ‘up-regulates’ the expression of BSEP that increases bile acid efflux from the liver to the bile (Sinal et al., 2000). Thus, FXR regulates transport of bile acids and avoids accumulation in hepatocytes. Recently, Yamamoto et al. (2006) provided direct evidence that ERα can become a risk factor for the development of hepatotoxicity following estrogen exposures. The administration of EE to rats decreases bile flow and produces impairment of transport mechanisms in canalicular and basolateral hepatocyte membranes (Rodr´ıguez-Garay, 2003). After EE administration, the excretion of bile salts, bilirubin, phospholipids, cholesterol, HCO3 and sulfobromophtalein is reduced (Bosard et al., 1993). EE treatment to rats results in a decreased sinusoidal membrane surface density (Hornstein et al., 1992). This was associated with impairment of sinusoidal transport molecules involved in the uptake of cholephilic substances (Alvaro et al., 1997). The EE effect has been attributed to the endogenous estrogen metabolite estradiol-17β-Dglucuronide (Meyers et al., 1980). This metabolite reduces bile flow and bile acid secretion in rats in a dose-dependent and reversible manner. Both EE and 17β-d-glucuronide increase tight juntional permeability in rat liver, allowing paracellular regurgitation of bile constituents into the blood (Kan et al., 1989). In conclusion, animal models of estrogen-induced cholestasis are necessary for the progressive understanding of EE acquired human cholestasis, as well as for the development of effective therapeutic interventions.

6.6

Ischemia/Reperfusion Liver Injury

Several excellent reviews have appeared recently dealing with this problem (Glantzounis et al., 2005; Pascher and Klupp, 2005; Casillas-Ram´ırez et al., 2006; Boros and Bromberg, 2006; Jaeschke, 2005b; Nieuwenhuijs et al., 2006; Spiegel and Bahde, 2006). Therefore, the reader is referred to those references and only a summary is presented here. Hepatic ischemia/reperfusion (I/R) damage is well known as an important cause of morbidity and mortality in liver resections and transplantations where anoxic or ischemic hepatic damage occurs. In addition, it also occurs as a consequence of systemic hypoxia or with conditions that produce reduction in the blood supply to the liver that results in low perfusion with the resulting hypoxia. This frequently occurs after systemic hypoxia or with conditions that cause low blood flow to the liver. Models of liver I/R damage represent an experimental simplification of the clinical situation. The choice of an appropriate model will depend on the clinical question to be

128

Hepatotoxicity

answered. Detailed descriptions of experimental models can be obtained in a recent review (Spiegel and Bahde, 2006). Probably one of the best models is the in vivo global liver ischemia and porto-caval decompression. Ex vivo models are mainly suitable for toxicity evaluation and for answering special questions.

6.7

Hepatic Amebiasis

Amebiasis may have been first recognized as a deadly disease by Hippocrates (460 to 377 BC), who described a patient with fever and dysentery. A review of E. histolytica and its laboratory diagnosis appeared recently (Tanyuksel and Petri, 2003). Amebiasis, associated with high morbidity and mortality, continues to be a major public health problem throughout the world. Poverty, ignorance, overcrowding, poor sanitation and malnutrition favor transmission and increases ‘disease burden’. 6.7.1

Models of Experimental Liver Amebiasis

An excellent recent review on this topic has recently been published (Tsutsumi and Shibayama, 2006); therefore, only a summary of the most important features are presented herein. Reinertson and Thompson (1951) were the first to report an animal model for amebiasis – the hamster. They injected trophozoites of E. histolytica either through the portal vein or directly into the hepatic lobe that produced large amebic hepatic abscesses. Tsutsumi et al. (1984) suggested the role of host inflammatory cells in the process of liver injury after inoculation with E. histolytica trophozoites in hamsters. The results were confirmed with electron microscopic (Tsutsumi and Mart´ınez-Palomo, 1988) and immunocytochemical (Ventura-Juárez et al., 2002) studies. These studies indicated that liver abscesses are not produced by direct lysis of the trophozoites but by lysosomal enzymes released from disintegrating inflammatory cells that accumulate around the amebas and are killed by this parasite. Hepatic amebiasis can also be reproduced in gerbils, although trophozoites are less virulent in gerbils than in hamsters. Thus, progression of hepatic injury and death of the infected animals takes a longer time, therefore, allowing sequential studies of the hepatic lesions (Shibayama-Salas et al., 1997). In the case of mice in liver amebiasis, the introduction of genetically modified animals, like the severe combined immunodeficient (SCID) mice, has generated important information. When SCID mice were injected intrahepatically with E. histolytica trophozoites, liver abscesses were developed in all animals; however, only one of seven control animals developed an abscess (Cieslak et al., 1992). This study suggests that SCID mice constitute a powerful model for studying the protective immunity in invasive amebiasis. E. histolytica infects only humans and non-human primates; thus, reproducible animal models of amebiasis are limited. Nevertheless, information obtained from different models of hepatic amebiasis has permitted the better understanding of the mechanisms that take place in the damaged hepatic cell. Despite the fact that we still do not have an animal model that mimics the complete cycle of E. histolytica, as in the human disease, several investigations performed with normal or genetically or surgically altered laboratory animals have led to better understanding of the pathogenesis, immunology and pharmacology of the amebic lesion.


6.8

129

Concanavalin A

Fulminant hepatic failure due to viral infection, intoxication or other causes is an important medical problem. Mechanisms that lead to apoptosis of hepatocytes frequently involved immune responses. Thus, the development of animal models that mimic the immunedependent parenchymal cell death are essential to understand the action mechanisms that conduce to hepatic failure and to develop pharmacological tools to prevent the human disease. Tiegs et al. (1992) developed a T cell-dependent experimental liver injury in mice induced by concanavalin A (Con A). Con A administration to mice produced dose-dependent apoptosis of hepatocyte (Gantner et al., 1995). Con A has high affinity toward the hepatic sinus producing activation of T cells (Gantner et al., 1995). A subset of T lymphocytes, the CD4-positive cells, is activated since anti-CD4 antibodies block activation, and consequently apoptosis of hepatocytes occurs (Tiegs et al., 1992). Polymorphonuclear is also recruited in the hepatic parenchyma probably by CD4-positive cells (Mizuhara et al., 1994). Con A-induced hepatitis in mice is characterized by increased serum levels of transaminases and infiltration, at the same time, of Kupffer cells, T cells and eosinophils into the liver (Jaruga et al., 2003). Natural killer T (NKT) hepatic cells play important roles in the Con A-induced hepatic damage by releasing several cytokines, including interferon gamma (IFN-γ ), tumor necrosis factor alpha (TNF-α), interleukin 4 (IL-4) and IL-5 (Takeda et al., 2000). Several other cytokines are involved in Con A-induced hepatic damage; some of them, like IL-6 (Mizuhara et al., 1994), IL-10 (Louis et al., 1997), IL-11 (Bozza et al., 1999), IL-22 (Radaeva et al., 2004) and IL-15 (Li et al., 2006), depict beneficial properties, whereas IL-12 (Nicoletti et al., 2000) and IL-18 (Faggioni et al., 2000) aggravate Con A-induced hepatitis. Therefore, hepatitis induced by Con A in mice may serve as a good model of viral hepatitis in the human. Obviously, viruses are not involved in this model. In fact, there is no model of viral hepatitis in animals.

6.9

Alcoholic Liver Damage in Rats

The rat constitutes one the most-used animals in the laboratory and alcohol abuse constitutes one of the most common causes of human liver diseases. Therefore, it seems logical to assume that administration of alcohol to rats may be the best model of alcoholic liver damage. However, it is important to consider that there are several characteristics in the rat that make it difficult to establish such a model: (a) natural aversion to ethanol, (b) higher ethanol metabolic rate and (c) greater tolerance to ethanol (Tsukamoto et al., 1990). To overcome these differences a long-term gastric catheter was designed and implanted in male adult rats (Tsukamoto et al., 1984). Currently, a single gastric catheter with a large bore size is used for infusion of ethanol and a diet. This method has been demonstrated to be the first ‘rat-model’ that induced progressive alcoholic liver injury including fatty liver, liver necrosis, inflammation and fibrosis (Tsukamoto et al., 1986; French et al., 1986). The exact control of ethanol and nutrient intake leads to a high reproducibility and low variability. Unfortunately, this model does not produce cirrhosis. However, the model allows investigations for the pathogenetic mechanisms of alcoholic liver necrosis

130

Hepatotoxicity

and fibrosis. These mechanisms include the importance of sustained blood alcohol level in alcoholic liver disease (ALD), the role of the amount and the type of dietary fat in ALD, association of hypoxia with necrosis involvement of hepatic stellate and Kupffer cells with alcoholic fibrogenesis, as reviewed by Tsukamoto et al. (1990). More recently, the NK-κB factor (inhibited with allopurinol) was discovered to participate in ALD in this model (Kono et al., 2000). The natural aversion to ethanol, higher ethanol metabolic rate and greater tolerance to ethanol of rodents has led scientists to choose other models of liver damage where a catheter is not needed and cirrhosis is easily produced, as in CCl4 or BDL. However, this model has the advantage that the causative agent is very commonly used in humans.

6.10

Concluding Remarks

Liver diseases constitute a major medical problem of worldwide proportions (Williams, 2006). In Africa and Asia, their main causes are viral and parasitical infections. In many countries, alcohol abuse is the major cause of liver diseases. The vulnerability of the liver to chemical injury is as much a function of its anatomical proximity to the blood supply and digestive tract as to its ability to biotransform and concentrate xenobiotics. Xenobiotics in the blood pass through the portal vein and hepatic artery, and then drain through the central and the hepatic veins into the vena cava. Liver diseases are among the first ten causes of death in the world. In sharp contrast, effective liver drugs are very limited. With the exceptions of antiviral drugs effective in the treatment of viral hepatitis, there are very few drugs, if any, able to protect or ‘cure’ the damaged liver in a clinical situation. In this context, fundamental liver pathophysiology and pharmacology is needed; therefore, animal models of hepatic injury are the best tools to investigate liver disease processes and to test drugs with the ability to interfere with the hepatic injury process. Research with this and other models has provided a lot of knowledge in basic pathology which is leading to the discovery of new therapeutical approaches against liver damage and the recent advances promises prompt new discoveries.

Acknowledgements The author expresses his gratitude to Dr Mario G. Moreno for carefully reviewing the manuscript.

References Agarwal M., Srivastava V. K., Saxena K. K., and Kumar A. (2006). Hepatoprotective activity of Beta vulgaris against CCl4 -induced hepatic injury in rats. Fitoterapia 77: 91–93. Albano E., Bellomo G., Carini R., Biasi E., Poli G., and Dianzani M. U. (1985). Mechanisms responsible for carbon tetrachloride-induced perturbation of mitochondrial calcium homeostasis. FEBS Lett 192: 184–188.


131

Alvaro D., Gigliozzi A., Piat C., Carli L., Fraioli F., Romeo R., Francia C., Attili A. F. and Capocaccia L. (1997). Inhibition of biliary bicarbonate secretion in ethinyl estradiol-induced cholestasis is not associated with impaired activity of the Cl− /CO− 3 exchanger in rat. J Hepatol 26: 146–157. Barón V. and Muriel P. (1999). Role of glutathione, lipid peroxidation and antioxidants on acute bile-duct obstruction in the rat. Biochim Biophys Acta 1472: 173–180. Benedetti A., Comporti M. and Esterbauer H. (1980). Identification of 4-hydroxy nonenal as a cytotoxic product originating from the peroxidation of liver microsomal lipids. Biochim Biophys Acta 620: 281–296. Benedetti A., Comporti M., Fulceri R. and Esterbauer H. (1984). Cytotoxic aldehydes originating from the peroxidation of liver microsomal lipids: identification of 4,5-dihydroxydecenal. Biochim Biophys Acta 792: 172–181. Berger L. M., Bhatt H., Combes B. and Estabrock R. W. (1986). CCl4 -induced toxicity in isolated hepatocytes: the importance of direct solvent injury. Hepatology 6: 36–45. Boros P. and Bromberg J. S. (2006). New cellular and molecular immune pathways in ischemia/reperfusion injury. Am J Transplan 6: 652–658. Bosard R., Stieger B., O’Neill B., Fricker G. and Meier P. J. (1993). Ethynylestradiol treatment induces multiple canalicular membrane transport alterations in rat liver. J Clin Invest 91: 2714–2720. Bozza M., Bliss J. L., Maylor R., Erickson J., Donnelly L., Bouchard P., Dorner A. J. and Trepiccho W. L. (1999). Interleukin-11 reduces T-cell-dependent experimental liver injury in mice. Hepatology 30: 1441–1447. Cameron G. R. and Hasan S. M. (1958). Disturbances of structure and function in the liver as the result of biliary obstruction. J Pathol Bacteriol 75: 333–349. Cameron G. R. and Karunaratne W. A. E. (1932). Carbon tetrachloride cirrhosis in relation to liver regeneration. J Pathol Bacteriol 42: 1–21. Cameron G. R. and Oakley C. L. (1932). Ligation of the common bile duct. J Pathol Bacteriol 35: 769–798. Casillas-Ram´ırez A., Mosbah I. B., Ramalho F., Roselló-Catafau J. and Peralta C. (2006). Past and future approaches to ischemia-reperfusion lesion associated with liver transplantation. Life Sci 79: 1881–1894. Cedillo A., Mourelle M. and Muriel P. (1996). Effect of colchicine and trimethylcolchicinic acid on CCl4 -induced cirrhosis in the rat. Pharmacol Toxicol 79: 241–246. Chavez E., Alcantar L. K., Moreno M. G. and Muriel P. (2006). Gadolinium chloride inhibits the spontaneous resolution of fibrosis in CCl4 -induced cirrhosis. Toxicol Mechan Meths 16: 1–7. Cieslak P. R., Virgin IV H. W. and Stanley S. L. (1992). A severe combined immunodeficient (SCID) mouse model for infection with Entamoeba histolytica. J Exp Med 176: 1605–1609. Clawson G. A., MacDonald J. and Woo C. (1987). Early hypomethylation of 2 -O-ribose moieties in hepatocyte cytoplasmic ribosomal RNA underlies the protein synthetic defect produced by CCl4 . J Cell Biol 105: 705–711. Comporti M. (1985). Lipid peroxidation and cellular damage in toxic liver injury. Lab. Invest. 53: 599–623. Daniel P. M., Prichardt M. L. and Reynoll P. C. (1952). The portal circulation in experimental cirrhosis of the liver. J Pathol Bacteriol 64: 53–60. DiRenzo A. B., Gandolfi A. J. and Sipes I. G. (1982). Microsomal bioactivation and covalent binding of aliphatic halides to DNA. Toxicol Lett 11: 243–252. Esterbauer H., Cheeseman K. H., Dianzani M. U., Poli G. and Slater T. F. (1982). Separation and characterization of the aldehydic products of lipid peroxidation stimulated by ADP-Fe2+ in rat liver microsomes. Biochem J 208: 129–140.

132

Hepatotoxicity

Esterbauer H., Schaur R. J. and Zollner H. (1991). Chemistry and biochemistry of 4-hydroxynonenal, malonaldehyde and related aldehydes. Free Radic Biol Med 11: 81–128. Faggioni R., Jones-Carson J., Reed D. A., Dinarello C. A., Feingold K. R., Grunfeld C. and Fantussi G. (2000). Leptin-deficient (ob/ob) mice are protected from T cell-mediated hepatotoxicity: role of tumor necrosis factor alpha and IL-18. Proc Natl Acad Sci USA 97: 2367–2372. Fernandez G., Villaruel M. C., De Toranzo E. G. D. and Castro J. A. (1982). Covalent binding of carbon tetrachloride metabolites to the heme moiety of cytochrome P450 and its degradation products. Res Commun Chem Pathol Pharmacol 35: 283–290. ´ V. and Muriel P. (2001). Effects of thalidoFernández-Mart´ınez E., Morales-R´ıos M. S., Pérez-Alvarez mide and 3-pthalimido-3-(3,4-dimethoxyphenyl)-propanamide on bile duct obstruction-induced cirrhosis in the rat. Drug Dev Res 54: 209–218. Fernandez-Martinez E., Perez-Alvarez V., Tsutsumi V., Shibayama M. and Muriel P. (2006). Chronic bile duct obstruction induces changes in plasma and hepatic levels of cytokines and nitric oxide in the rat. Exp Toxicol Pathol 58: 49–58. Francis G. A, Fayard E., Picard F. and Auwerx J. (2003). Nuclear receptors and the control of metabolism. Annu Rev Physiol 65: 261–311. French S. W., Miyamoto K. and Tsukamoto H. (1986). Ethanol-induced hepatic fibrosis in the rat: role of the amount of dietary fat. Alcohol Clin Exp Res 10: 135–195. Forni L. G., Packer J. E., Slater T. F. and Willson R. L. (1983). Reaction of the trichloromethyl and halothane-derived peroxy radicals with unsaturated fatty acids: a pulse radiolysis study. Chem Biol Interact 45: 171–177. Fuji K. (1997). Preventive effect of isofluran on destruction of cytochrome P450 during dehalogenation of carbon tetrachloride in guinea pig microsomes. Drug Metabol Drug Interact 14: 99–107. Gantner F., Leist M., Lohse A. W., Germann P. G. and Tiegs A. (1995). Concanavalin A-induced T-cell-mediated hepatic injury in mice: the role of tumor necrosis factor. Hepatology 21: 190–198. Glantzounis G. K., Salacinski H. J., Yang W., Davidson B. R. and Seifalian M. (2005). The contemporary role of antioxidant therapy in attenuating liver ischemia-reperfusion injury: a review. Liver Transplant 11: 1031–1047. Gruebele A., Zawaski K., Kaplan D. and Novak R. F. (1996). Cytochrome P450 2E1- and cytochrome P450 2B1/2B2-catalyzed carbon tetrachloride metabolism: effects on signal transduction as demonstrated by altered immediate–early (c-Fos and c-Jun) gene expression and nuclear AP-1 and NF-κB transcription factor levels. Drug Metabol Disp 24: 15–22. Grypioti A. D., Theocharis S. E., Demopoulos C. A., Papadopoulou-Daifoti Z., Basayiannis A. C. and Mykoniatis M. G. (2006). Effect of platelet-activating factor (PAF) receptor antagonist (BN52021) on acetaminophen-induced acute liver injury and regeneration in rats. Liver Int 26: 97–105. Guo G. L., Lambert G., Negishi M., Ward J. M., Brewer H. B. Jr, Kliewer S. A., González F. J. and Sinal C. J. (2003). Complementary roles of farnesoid X receptor, pregnane X receptor, and constitutive androstane receptor in protection against bile acid toxicity. J Biol Chem 278: 45062–45071. Hartley D. P., Kolaja K. L., Reichard J. and Petersen D. R. (1999). 4-Hydroxynonenal and malondialdehyde hepatic protein adducts in rats treated with carbon tetrachloride: immunochemical detection and lobular localization. Toxicol Appl Pharmacol 161: 23–33. Hornstein B., Stamler L., Bianchi L. and Landmann L. (1992). Ethinylestradiol increases volume and decreased sinusoidal membrane surface in the rat liver: a stereological analysis. Hepatology 16: 217–223. Hruszkewycz A. M., Glende E. A. and Recknagel R. O. (1978). Destruction of microsomal cytochrome P450 and glucose-6-phosphatase by lipids extracted from peroxidized microsomes. Toxicol Appl Pharmacol 46: 695–702.


133

Jaeschke H. (2005a). Role of inflammation in the mechanism of acetaminophen-induced hepatotoxicity. Expert Opin Drug Metab Toxicol 1: 389–397. Jaeschke H. (2005b). Mechanisms of liver injury. II. Mechanisms of neutrophil-induced liver cell injury during hepatic ischemia-reperfusion and other acute inflammatory conditions. Am J Physiol Gastrointest Liver Physiol 290: 1083–1088. Jaeschke H. and Bajt M. L. (2006). Intracellular signaling mechanisms of acetaminophen-induced liver cell death. Toxicol Sci 89: 31–41. Jaruga B., Hong F., Sun S., Radaeva S. and Gao B. (2003). Crucial role of IL-4/STAT6 in T cellmediated hepatitis: up-regulating eotaxins and IL-5 and recruiting leukocytes. J Immunol 171: 3233–3244. Jollow D. J., Mitchell J. R., Potter W. Z., Davis D. C., Gillette J. R. and Brodie B. B. (1973). Acetaminophen-induced hepatic necrosis. II Role of covalent binding in vivo. J Pharmacol Exp Ther 187: 195–202. Kan K. S., Monte M. J., Parslow R. A. and Coleman R. (1989). Oestradiol 17 beta-glucuronide increases tight-juntional permeability in rat liver. Biochem J 261: 297–300. Kono H., Rusyn I., Bradford B. U., Connor H. D., Mason R. P. and Thurman R. G. (2000). Allopurinol prevents early alcohol-induced liver injury in rats. J Pharmacol Exp Ther 293: 296–303. Kountouras J., Barbara H., Billing H. and Scheuer P. J. (1984). Prolonged bile duct obstruction: a new experimental model for cirrhosis in the rat. Br J Exp Pathol 65: 305–311. Kumar V., Deo M. G. and Ramalingaswami V. (1972). Mechanism of fatty liver in protein deficiency. An experimental study in the rhesus monkey. Gastroenterology 62: 445–451. Kyung-Hyun K., Hyun-Chul K., Mee-Yul H., Hoon-Kyu O., Tae Sung L., Young Chae C., Ho-Jung S., Nam-Hee W. and Kwan-Kyu P. (2006). The antifibrotic effect of TGF-β1 siRNAs in murine model of liver cirrhosis. Biochem Biophys Res Common 343: 1072–1078. Lamson P. D. and Wing R. (1926). Early cirrhosis of the liver produced in dogs by carbon tetrachloride. J Pharmacol Exp Therap 29: 191–202. Lee W. M. (2004). Acetaminophen and the US Acute Liver Study Group: lowering the risks of hepatic failure. Hepatology 40: 6–9. Lee J. M., Trauner M., Soroka C. J., Stieger B., Meier P. J. and Boyer J. L. (2000). Expression of the bile salt export pump is maintained after chronic cholestasis in the rat. Gastroenterology 118: 163–172. Li B., Sun R., Wei H., Gao B. and Tian Z. (2006). Interleukin-15 prevents concanavalin A-induced liver injury in mice via NKT cell-dependent mechanism. Hepatology 43: 1211–1219. Long R. L. and Moore L. (1986). Elevated cytosolic calcium in rat hepatocytes exposed to carbon tetrachloride. J Pharmacol Exp Ther 238: 186–191. Louis H., Le Moine O., Peny M. O., Quertinmont E., Fokan D., Goldman M. and Deviere J. (1997). Production and role of interleukin-10 in concanavalin A-induced hepatitis in mice. Hepatology 25: 1382–1389. Manno M., Ferrara R., Cazzaro S., Rigotti P. and Ancona E. (1992). Suicidal inactivation of human cytochrome P-450 by carbon tetrachloride and halotane in vitro. Pharmacol Toxicol 70: 13–18. McCay P. B., Lay E. K., Poyer J. L., DuBose C. M. and Jansen E. G. (1984). Oxygen and carboncentered free radical formation during carbon tetrachloride metabolism. J Biol Chem 259: 2135– 2143. Medina J. F., Martinez-Anso E., Vazquez J. J. and Prieto J. (1997). Decreased anion exchanger 2 immunoreactivity in the liver of patients with primary biliary cirrhosis. Hepatology 25: 12–17. Meyers M., Slikker W. and Vore M. (1980). Characterization of cholestasis induced by estradiol17beta-d-glucuronide in the rat. J Pharmacol Exp Ther 214: 87–93.

134

Hepatotoxicity

Mico B. A. and Pohl L. R. (1983). Reductive oxygenation of carbon tetrachloride: trichloromethyl peroxy radical as a possible intermediate in the conversion of carbon tetrachloride to electrophilic chlorine. Arch Biochem Biophys 225: 596–609. Mitchell J. R., Jollow D. J., Potter W. Z., Davis D. C., Gillette J. R. and Brodie B. B. (1973a). Acetaminophen-induced hepatic necrosis. I. Role of drug metabolism. J Pharmacol Exp Ther 187: 185–194. Mitchell J. R., Jollow D. J., Potter W. Z., Davis D. C., Gillette J. R. and Brodie B. B. (1973b). Acetaminophen-induced hepatic necrosis. IV. Protective role of glutathione. J Pharmacol Exp Ther 187: 211–217. Mizuhara H., O’Neill E., Seki N., Ogawa T., Kusunoki C., Otsuka K., Sato S., Niwa M., Senoh H. and Fujiwara H. (1994). T-cell activation-associated hepatic-injury: mediation by tumor necrosis factor and protection by interleukin 6. J Exp Med 179: 1529–1537. Morel P., Tallineau C., Pontcharraud R., Piriou A. and Huguet F. (1998). Effects of 4-hydroxynonenal, a lipid peroxidation product, on dopamine transport and Na+ , K+ –ATPase activity in rat striatal synaptosomes. Neurochem Int 33: 531–540. Moreno M. G. and Muriel P. (2006). Inducible nitric oxide synthase is not essential for the development of fibrosis and liver damage induced by CCl4 in mice. J Appl Toxicol 26: 326–332. Muriel P. (1997). Peroxidation of lipids and liver damage, in S. I. Baskin and H. Salem (eds), Oxidants and Free Radicals, Taylor & Francis, Washington, DC, USA, 237–257. Muriel P. (2006). Role of nitric oxide in liver disorders, in S. Ali, D. Mann and S. L. Friedman (eds), Liver Diseases: Biochemical Mechanisms and New Therapeutic Insights, Science Publishers, Enfield, NH, USA, 115–131. Muriel P. (2007). Cytokines in liver diseases, S. C. Sahu (eds), in Hepatotoxicity: from Genomics to in vitro and in vivo Models, John Wiley & Sons Ltd, Chichester, UK, r r r– r r r r Muriel P. and Escobar Y. (2003). Kupffer cells are responsible for liver cirrhosis induced by carbon tetrachloride. J Appl Toxicol 23: 103–108. Muriel P. and Mourelle M. (1990). The role of membrane composition in ATPases activities of cirrhotic rats. Effect of silymarin. J Appl Toxicol 10: 281–284. Muriel P. and Mourelle M. (1992a). Characterization of membrane fraction lipid composition and function of cirrhotic rat liver. Role of S-adenosyl-L-methionine. J Hepatol 14: 16–21. Muriel P. and Mourelle M. (1992b). The role of membrane composition in ATPase activities of cirrhotic rat liver: Effect of silymarin. J Appl Toxicol 10: 281–284. Muriel P. and Mourelle M. (1992c). Prevention by silymarin of membrane alterations in acute CCl4 liver damage. J Appl Toxicol 10: 275–279. Muriel P. and Suárez O. R. (1994). Role of lipid peroxidation on biliary obstruction in the rat. J Appl Toxicol 14: 423–426. Muriel P., Garciapiña T., Pérez-Alvarez V. and Mourelle M. (1992). Silymarin protects against paracetamol-induced lipid peroxidation and liver damage. J Appl Toxicol 12: 439–442. ´ Muriel P., Fernández-Mart´ınez E., Pérez-Alvarez V., Lara-Ochoa F., Ponce S., Garc´ıa J., Shibayama M. and Tsutsumi V. (2003). Thalidomide ameliorates CCl4 -cirrhosis in the rat. Eur J Hepatol Gastroenterol 15: 951–957. Muriel P., Moreno M. G., Hernández M. C., Chávez E. and Alcazar L. K. (2005). Resolution of liver fibrosis in chronic CCl4 administration in the rat after discontinuation of the treatment: effect of silymarin, silibinin, colchicine and trimethylcolchinic acid. Basic Clin Pharmacol & Toxicol 96: 275–280. Nicoletti F., Di Marco R., Zaccone P., Salvaggio A., Magro G., Bendtzen K., and Moroni P. (2000). Murine concanavalin A-induced hepatitis is prevented by interleukin 12 (IL-12) antibody and exacerbated by exogenous IL-12 through an interferon-gamma-dependent mechanism. Hepatology 32: 728–733.


135

Nicotera P., Bellomo G. and Orrenius S. (1992). Calcium mediated mechanisms in chemically induced cell death. Annu Rev Pharmacol Toxicol 32: 449–470. Nieuwenhuijs V. D., De Bruijn M. T., Padbury R. T., and Barritt G. T. (2006). Hepatic ischemiareperfusion injury: roles of Ca2+ and other intracellular mediators of impaired bile flow and hepatocyte damage. Dig Dis Sci 51: 1087–1102. Omura M., Katsumata T., Misawa H. and Yamaguchi M. (1999). Decrease in protein kinase and phosphatase activities in the liver nuclei of rats exposed to carbon tetrachloride. Toxicol Appl Pharmacol 160: 192–197. Parola M., Albano E., Autelli R., Barrera G., Biocca M. E., Paradisi L. and Dianzani M. U. (1990). Inhibition of the high affinity Ca2+ –ATPase activity in rat liver plasma membranes following carbon tetrachloride intoxication. Chem Biol Interact 73: 103–119. Pascher A. and Klupp J. (2005). Biologics in the treatment of transplant rejection and ischemia/reperfusion injury: new applications for TNFalpha inhibitors? BioDrugs 19: 211–231. Pérez-Tamayo R. (1983). Is cirrhosis of the liver experimentally produced by CCl4 an adequate model of human cirrhosis? Hepatology 3: 112–120. Poli G., Chiarpotto E., Biasi F., Leonarduzzi G., Comoglio A., Albano E. and Dianzani M. U. (1989). Role of aldehydes in the propagation of lipid peroxidation-induced liver toxicity. Adv Biosci 76: 73–81. Radaeva S., Sun R., Pan H. N., Hong F. and Gao B. (2004). Interleukin 22 (IL-22) plays a protective role in T cell-mediated murine hepatitis: IL-22 is a survival factor for hepatocytes via STAT3 activation. Hepatology 39: 1332–1342. Rao S. B., Young R. A. and Mehendale H. M. (1989). Hepatic polyamines and related enzymes following chlordecone-potentiated carbon tetrachloride toxicity in rats. J Biochem Toxicol 4: 55– 63. Recknagel R. O., Glende E. A. Jr, Dolak J. A. and Waller R. L. (1989). Mechanism of carbon tetrachloride toxicity. Pharmacol Ther 43: 139–154. Reinertson J. W. and Thompson P. E. (1951). Experimental amebic hepatitis in hamsters. Proc Soc Exp Biol Med 76: 518–520. Rodr´ıguez-Garay E. A. (2003). Cholestasis: human disease and experimental animal models. Ann Hepatol 2: 150–158. Ruchirawat M., Mostafa M. H., Shank R. C., and Weisburger E. K. (1983). Inhibitory effects of carbon tetrachloride on diethylnitrosamine metabolism and DNA alkylation. Carcinogenesis 4: 537–539. Schreiber A. J. and Simon F. R. (1983). Estrogen-induced cholestasis: clues to pathogenesis and treatment. Hepatology 3: 607–613. Shibayama-Salas M., Navarro-Garc´ıa F., López-Revilla R., Mart´ınez-Palomo A. and Tsutsumi V. (1997). In vivo and in vitro intestinal amebiasis in gerbils. Parasitol Res 83: 170–176. Simon F. R., Fortune J., Iwahashi M., Gartung C., Wolkoff A. and Sutherland E. (1996). Ethinyl estradiol cholestasis involves alterations in expression of liver sinusoidal transporters. Am J Physiol 271: G1043–G1052. Sinal C. J., Tohkin M., Miyata M., Ward J. M., Lambert G. and Gonzalez F. J. (2000). Targeted disruption of the nuclear receptor FXR/BAR impairs bile acid and lipid homeostasis. Cell 102: 731–744 Sotelo-Félix J. I., Mart´ınez-Fong D. and Muriel P. (2002). Protective effect of carnosol on CCl4 induced acute liver damage in rats. Eur J Gastronterol Hepatol 14: 1001–1006. Spiegel H. U. and Bahde R. (2006). Experimental models of temporary normothermic liver ischemia. J Invest Surg 19: 113–123. Sundari P. N. and Ramakrishna B. (1997). Does oxidative protein damage play a role in the pathogenesis of carbon tetrachloride-induced liver injury in the rat? Biochim Biophys Acta 1362: 169–176.

136

Hepatotoxicity

Susin S. A., Daugas E., Ravagnan L., Samejima K., Zamzami N., Loeffler M., Constantini P., Ferri K. F., Irinopoulou T., Prevost M. C., Brothers G., Mak T. W., Penninger J., Earnshaw W. C. and Kroemer G. (2000). Two distinct pathways leading to nuclear apoptosis. J Exp Med 192: 571–580. Takeda K., Hayakawa Y., Van Kaer L., Matsuda H., Yagita H. and Okumura K. (2000). Critical contribution of liver natural killer T cells to a murine model of hepatitis. Proc Natl Acad Sci USA 97: 5498–5503. Tanyuksel M. and Petri W. A. (2003). Laboratory diagnosis of amebiasis. Clin Microbiol Rev 16: 713–729. Tiegs G., Hentschel J. and Wendel A. (1992). A T cell-dependent experimental liver injury in mice inducible by concanavalin A. J Clin Invest 90: 196–203. Tomasi A., Albano E., Bani S., Botti B., Corongiu F., Dessi M. A., Lannone A., Vannini V. and Dianzani M. U. (1987). Free radical metabolism of carbon tetrachloride in rat liver mitochondria. Biochem J 246: 313–317. Trams E. G. and Symeonidis S. (1957). Morphologic and functional changes in the livers of rats after ligation or excision of the common bile duct. Am J Pathol 33: 13–27. Trible D. L., Aw T. Y., and Jones D. P. (1987). The pathophysiological significance of lipid peroxidation in oxidative cell injury. Hepatology 7: 377–386. Tsokos-Kuhn J. O., Todd E. L., McMillin-Wood J. B. and Mitchell J. R. (1985). ATP-dependent calcium uptake by rat liver plasma membrane vesicles. Effect of alkylating hepatotoxins in vivo. Mol Pharmacol 28: 56–61. Tsukamoto H. Reidelberger R. D., French S. W. and Largman C. (1984). Long-term cannulation model for blood sampling and intragastric infusion in the rat. Am. J. Physiol 247: 595–599. Tsukamoto H., Towner S. J., Ciofalo L. M. and French S. W. (1986). Ethanol-induced liver fibrosis in rats fed high-fat diet. Hepatology 6: 814–822. Tsukamoto H., Gaal K. and French S. W. (1990). Insights into the pathogenesis of alcoholic liver necrosis and fibrosis: status report. Hepatology 12: 599–608. Tsutsumi V. and Mart´ınez-Palomo A. (1988). Inflammatory reaction in experimental hepatic amebiasis. An ultrastructural study. Am J Pathol 130: 112–119. Tsutsumi V. and Shimayama M. (2006). Experimental amebiasis: a selected review of some in vivo models. Arch Med Res 37: 210–220. Tsutsumi V., Mena-López R., Anaya-Velázquez F. and Mart´ınez-Palomo A. (1984). Cellular bases of experimental amebic liver abscess formation. Am J Pathol 117: 81–91. Ueda K., Kobayashi S., Morita J. and Komano T. (1985). Site-specific damage caused by lipid oxidative products. Biochim Biophys Acta 824: 341–348. van Loo G., Schotte P., van Gurp M., Demol H., Hoorelbeke B., Gevaert K., Rodriguez I., Ruiz-Carrillo A., Vandekerckhove J., Declercq W., Beyaert R., Vandenabeele P. (2001). Endonuclease G: a mitochondrial protein released in apoptosis and involved in caspase-independent DNA degradation. Cell Death Differ 8: 1136–1142. Varela-Moreiras G., Alonso-Aperte E., Rubio M., Gasso M., Deulofem R., Alvarez L., Caballeria J., Rodes J. and Mato J. M. (1995). Carbon tetrachloride-induced hepatic injury is associated with global DNA hypomethylation. Hepatology 22: 1310–1315. Ventura-Juárez J., Campos-Rodr´ıguez R. and Tsutsumi V. (2002). Early interactions of Entamoeba histolytica trophozoites with parenchimal and inflammatory cells in the hamster liver: an immunocytochemical study. Can J Microbiol 48: 123–131. Weber L. W. D., Boll M. and Stampfl A. (2003). Hepatotoxicity and mechanism of action of haloalkanes: carbon tetrachloride as a toxicological model. Crit Rev Toxicol 33: 105–136. Williams R. (2006). Global challenges in liver disease. Hepatology 44: 521–526. Yamamoto Y., Moore R., Hess H. A., Guo G. L., González F. J., Korach K. S., Maronpot R. R. and Negishi M. (2006). Estrogen receptor α mediates 17α-ethynylestradiol causing hepatotoxicity. J Biol Chem 281: 16625–16631.


137

Yoshida T., Adachi E., Nigi H., Fujii S. and Yanagi M. (1999). Changes of sinusoidal basement membrane collagens in early hepatic fibrosis induced with CCl4 in cynomolgus monkeys. Pathology 31: 29–35. Zangar R. C., Benson J. M., Burnett V. L. and Springer D. L. (2000). Cytochrome P450 2E1 is the primary enzyme responsible for low-dose carbon tetrachloride metabolism in human liver microsomes. Chem Biol Interact 15: 233–243. Zawaski K., Gruebele A., Kaplan D., Reddy S., Mortensen A. and Novak R. F. (1993). Evidence for enhanced expression of c-fos, c-jun and the Ca(2+)-activated neutral protease in rat liver following carbon tetrachloride administration. Biochem Biophys Res Commun 197: 585–590. Zimmerman H. J. (1999). Hepatotoxicity. The Adverse Effects of Drugs and Other Chemicals on the Liver, Lippincott Williams & Wilkins, Philadelphia, PA, USA.

Section 2 Hepatocyte Cultures

7 Application of Short- and Long-Term Hepatocyte Cultures to Predict Toxicities Gregor Tuschl, Jens Hrach, Philip G. Hewitt and Stefan O. Mueller

7.1

Introduction

Hepatotoxicity leads to a broad variety of liver pathophysiologies, including, for example, steatosis (fatty liver), cholestasis (obstruction of bile secretion), fibrosis (increased production and deposition of extracellular matrix components), hepatitis (inflammation), necrosis (cell death) or the formation of liver tumours. These pathological findings may arise from diseases affecting the liver, but also from xenobiotics, alcohol abuse or undesired drug– drug interactions. Hepatotoxicity is a major issue in pharmaceutical drug development [1] and drug-induced liver injury (DILI) is the major reason for attrition in clinical studies, as well as post-marketing [2]. In addition, 30 to 50 % of acute liver failures and 15 % of liver transplantations are related to chemical-induced hepatotoxicity [3–6]. Unfavourable hepatic reactions are often idiosyncratic, occurring on a background of transient liver injuries in less than 1 in 10 000 patients [7]. This may add to the fact that certain human hepatotoxicities have not been predictable from standard in vitro cytotoxicity assay results [8] or regulatory animal studies [9, 10]. This holds true despite the fact that there is reasonable understanding of the general pathophysiological mechanisms of most drug-induced hepatoxicities [11–14]: inhibition of mitochondrial function, disruption of intracellular calcium homeostasis, activation of apoptosis, oxidative stress, inhibition of specific enzymes or transporters and formation of reactive metabolites that cause direct toxicity or immunogenic response, potentially leading to idiosyncratic effects. Although there are ways to analyse


142

Hepatotoxicity

most of these parameters, prediction of human hepatotoxicity is poor due to strong interspecies variations in data obtained from in vivo studies and the lack of true physiological conditions, especially the complexity of a whole organism, in in vitro experiments. Along with hypersensitivity and skin reactions, hepatotoxicity in humans has the poorest correlation with regulatory animal toxicity tests [9, 10]. On the other hand, conventional in vitro cytotoxicity assays also have low predictive value for the detection of human hepatotoxicity. However, it is worth mentioning that if these assays identified a compound as a liver-toxicant, there is more than 80 % specific correlation to the corresponding findings in humans [8]. The poor sensitivity of standard cytotoxicity assays is due to several reasons [8]. They usually measure lethal events in late stages of toxicity, but toxicity may not be lethal per se. In vitro cytotoxicity may take several days to appear [8, 15–18], making repeated administration of the investigated drug necessary. This, of course, raises the need for long-term in vitro models that facilitate extended exposure times. In addition, standard tests in general investigate one parameter of cell damage, whereas hepatotoxicity has many different mechanisms and is considered a multi-step process. In order to improve sensitivity, it will be necessary to analyse several morphological, biochemical and functional endpoints in parallel. Finally, tests should be performed not only with high concentrations, causing acute toxicity, but also with in vivo pharmacological concentrations. To address the limitations of many in vitro applications, we will present here long-term hepatocyte cultures as a model for the investigation of chronic and low-dose effects and also high-content screening (HCS), as a powerful tool which can be used to study several parameters in parallel at the single cell level. We will give an overview of the main in vitro test systems used for the assessment of hepatotoxicity and describe examples of the analysis of morphological, functional and gene expression endpoints in hepatocyte cultures.

7.2

In vitro Liver Models

The drug development process comprises a variety of steps to assess whether a test compound has adequate bioreactivity, appropriate physicochemical properties, metabolic stability, safety and efficacy in humans. While in vivo models are used to investigate hepatic drug effects in the context of toxicokinetics and systemic influences, cell culture models provide test systems for the investigation of specific mechanisms in a precisely controlled environment [19]. Hepatotoxicity occurs by the direct or indirect interaction of the toxicants with basic hepatocyte constituents such as proteins, lipids, RNA or DNA which can be analysed at the molecular, cellular or organ level. In vivo studies, limited by animal welfare/ethical concerns and difficulties in distinguishing between primary and secondary toxic effects, can be supplemented and partly replaced by in vitro models to accomplish a more detailed view of the mechanism of toxicity. There are many in vivo and in vitro test systems currently in use to predict hepatotoxicity in humans. However, their application is limited due to the frequent idiosyncratic nature of liver toxicity and the inherent differences between the metabolic activity in human and non-human species. The simplicity of some in vitro systems provides the ability to specifically manipulate and analyse a small number of parameters. The most commonly used test systems of the past few decades include, for example, the isolated perfused liver, liver slices, primary hepatocytes in suspension or culture, cell lines, transgenic cells and sub-cellular fractions like S9-mix, microsomes,

Application of Hepatocyte Cultures to Predict Toxicities

143

supersomes or cytosol (Table 7.1). The reduction of the complexity of the system and the increase in throughput offers the ability to study specific parameters more closely but creates inherent constraints for each model (Table 7.1). This limits their widespread use and acceptance by the regulatory authorities as an alternative for in vivo screening [20], although studies have shown that in vitro cytotoxicity data can be used to identify appropriate doses for in vivo studies [25]. One major obstacle of some in vitro models is the limitation in metabolic activation of xenobiotics, mainly due to the downregulation of cytochrome P450 (CYP) enzymes over time [26, 27]. This is especially important since Phase I and Phase II metabolic conversion of chemicals has a great influence on their toxicity and can lead to both detoxification and toxification of xenobiotics [28–30]. To overcome these problems, new and innovative strategies are being developed in order to find reliable markers that are involved not only in early toxic responses but also in chronic toxicities, both occurring at sub-lethal doses of a test compound. Furthermore, there is a strong need for a robust longterm in vitro screening system that allows the characterization of drug/chemical-induced toxicities and helps to reduce the use of animals in toxicity testing. 7.2.1

Isolated Perfused Liver

Ideally, an in vitro test system should adequately represent the in vivo situation of drug metabolism and biotransformation in the liver, but there is still a long way to go to accomplish this goal. Most liver specific features are preserved in whole isolated and perfused livers, first developed in 1972 by Gordon and colleagues [31]. Especially, the threedimensional architecture of the liver, with cell–cell, cell–matrix interactions and functional bile canaliculi are maintained, leading to in vivo-like metabolism. Additionally, all liver cell types are present and the communication between them can play an important role in mediating toxicity. Perfused livers from different kinds of laboratory animals have been used for studies of hepatotoxicity, especially for compounds with the potential to affect bile flow. Furthermore, whole isolated liver is the only in vitro model that allows the measurement of haematodynamics. Despite all of these advantages, the isolated perfused liver model is difficult to handle and retains its functional integrity for only a few hours. For that reason, it can only be used for toxicants that are expected to have toxic effects at very early ‘time-points’. Moreover, reproducibility is low, the use of animals is not significantly reduced and human organs are rarely available. 7.2.2

Liver Slices

First used in 1923 by Otto Warburg [32] and improved in the following decades [33– 35], precision-cut liver slices are another in vitro model, with partly conserved liver cytoarchitecture, cell–cell, cell–matrix contacts and the presence of different cell types, which are frequently used in in vitro toxicology [36]. The preparation of slices from different parts of the liver facilitates lobe- and zone-specific analysis of metabolism and toxicity. In addition, since many slices can be prepared from the same human or animal donor liver, reproducibility and throughput can be increased significantly. Another major advantage is the possibility to conduct histopathologic examinations, as well as biochemical and molecular-biological studies from the same tissue. Although the thickness of liver slices is 200–250 μm, resembling 10–20 cell layers, the adequate supply of nutrients and oxygen from the incubation medium is only maintained for the outer cell layers. Therefore, liver

144

Hepatotoxicity

Table 7.1

Overview of common liver in vitro systemsa

Model

Advantages

Disadvantages

Isolated perfused liver

– functions closest to in vivo – in vivo-like expression of drug metabolising enzymes and transporters – three dimensional cytoarchitecture – functional bile canaliculi – lobular structure preserved – collection of bile possible – short-term kinetic studies – in vivo cytoarchitecture preserved – reasonably high throughput – functional drug metabolizing enzymes, transporters and bile canaliculi – zone-specific metabolism and toxicity may be studied – lobular structure preserved, selective effects detectable – human tissue slices more easily available than whole organs – long-term use possible – re-establishment of 3D cytoarchitecture – continuous perfusion with medium – hepatocyte specific gene expression closer to in vivo than in hepatocyte cultures

– not a high throughput system – hepatic function preserved for only a short time (2–3 h) – complicated to use – study of human liver difficult/impossible – best suited for livers of small animals – no significant reduction in the number of animals used – hepatic function preserved for no more than 6 to 48 h – bile cannot be collected and analysed – necrotic cells/scar tissue at edges of the slice – presence of necrotic cells might affect active transport of drugs through the outer cells

Liver tissue slices

3D-Bioreactors (bioartificial liver systems)

spheroids

– re-establishment of 3D cytoarchitecture – presence of non-parenchymal cells on outer layer and extra-cellular matrix throughout the spheroids

Primary hepatocyte cultures

– reasonably high throughput – differentiated function maintained in many short-term and long-term cultures – viability and differentiation preserved for 2 weeks, depending on culture conditions – potential for use of long-term cultures in chronic toxicity, drug metabolism and drug–drug interaction studies

very low throughput

– necrotic, hypoxic cells in centre of spheroids – accumulation of bile in centre of spheroids possible – not usable for long-term investigations (disaggregation and dedifferentiation) – culture may need special supplements in media – survival, differentiation status and function depends on culture conditions – no single system has yet been able to preserve all the different liver specific functions in vitro – difficult to regain cells for FACS analysis


145

Table 7.1 Overview of common liver in vitro systemsa (Continued) Model

Advantages

Disadvantages

– relatively easy to use – analysis of human samples possible – functional drug metabolizing enzymes, transporters and bile canaliculi, depending on culture conditions – co-culture with other liver cells possible Hepatocytes in suspension

Liver cell lines

S9-mix Microsomes

Supersomes/ baculosomes

– reasonably high throughput – most drug metabolizing enzymes well-preserved at in vivo levels – easy to use – zone-specific metabolism and toxicity may be studied depending on the method of isolation – cryopreservation possible – analysis of human samples possible – unlimited availability – some liver-specific functions have been shown to be maintained – easy to use – reasonably high throughput – contains microsomal and cytosolic fractions – Phase I and Phase II activities – high throughput system – maintain expression of Phase I enzymes – can be recovered from frozen tissue – production of metabolites for structural analysis possible – use for drug inhibition, covalent binding and clearance studies – available from several species (including human) – high throughput – one or more human enzymes (CYPs, UGTs) can be specifically expressed

– lack of cell polarity limits use for drug transporter studies – lack of functional bile canaliculi – short-term viability (2–4 h) – lack of cell–cell and cell–matrix contacts – variations in samples from different human donors

– lacks in vivo phenotype – only a small set of hepatic functions expressed, at levels different from liver – genotypic instability – cofactors required for activity – lower enzyme activity compared to microsomes or cytosol – lacks Phase II and other cytosolic enzymes – short-term studies – cofactors required for activity – inadequate representation of the diversity of hepatic functions

– cofactors required for activity – UGT-reaction partly impaired

(Continued )

146

Hepatotoxicity

Table 7.1

Overview of common liver in vitro systemsa (Continued)

Model

Advantages

Disadvantages

Mitochondria

– high throughput – analysis of the effect of drugs on respiration, ATP synthesis and fatty acid oxidation

– only very short-term studies

Cytosol

– soluble Phase II enzymes (GST, ST, NAT) can be studied separately, depending on added cofactors – high throughput – one or more human enzymes can be specifically expressed – unlimited cell number

– cofactors required for activity – no CYPs, UGTs

Expression systems (cDNA clones in transient and stable expression systems) The virtual hepatocyte

– mathematical modelling of cellular events – prediction of unknown interactions may be possible

– studies may lack in vivo relevance – no physiologic levels of expressed enzymes – only single (some) enzymes can be analysed – limited computational power – still in experimental stage

a

Adapted from Brandon et al. (2003) [20], Guillouzo (1998) [21], Groneberg et al. (2002) [22], Farkas and Tannenbaum (2005) [23] and Sivaraman et al. (2005) [24].

slices are only useful for short-term toxicology studies due to the limited viability and rapid decline of liver-specific functions. The metabolic activity of tissue slices can be preserved for 1–2 days in culture [37, 38]. Another weakness is the inability to measure bile flow and other functional parameters, such as the portal flow. 7.2.3

Cultures of Isolated Liver Cells

Isolated hepatocytes or whole liver cell suspensions are used as starting material for a variety of in vitro models of different complexity and throughput. These include suspension cultures, adherent cultures on coated surfaces or embedded in extracellular matrix, cocultures with other cell types and ‘3D-Bioreactor’ setups. Fresh liver cells can be obtained relatively easy by different perfusion techniques [39, 40]. All of these procedures involve perfusion of the liver with Ca2+ -free buffers combined with enzymes/proteases, which disintegrate the extra cellular matrix, leading to the separation of the cells from each other. The isolation of liver cells is routinely performed for many species used in toxicity testing, but also with organs from human donors [41, 42]. Partial liver resections and nontransplantable whole livers are the main source of human tissue [43]. Hepatocytes, which make up the largest liver cell population, are highly polarized and differentiated cells, but these features are lost during liver perfusion. Therefore, the most challenging task is to re-establish cell polarity, cell–cell and cell–matrix interactions and to maintain the differentiated state as long as possible in culture [44, 45]. The quality of cultures not only depends on the conditions applied after seeding, but also on cell preparation


147

and handling before culturing. Variations in the quality of hepatocyte preparations are particularly distinctive for human cells [43, 46]. For example, the type and concentration of the enzyme used for perfusion, along with its specific protease activity, affect yield and viability of the cell suspension obtained. Low protease activity may result in incomplete separation of cells, resulting in residual tissue ‘clumps’. High protease activity will decrease viability and increase the quantity of dead cells and cell debris in the culture but also lead to a higher number of cells undergoing necrosis during the first 24 h of culture. Moreover, the cell suspension, containing all parenchymal and nonparenchymal liver cells, should be protected from light and stored on ice for as short a time as possible before seeding. The desired cell populations (hepatocytes, endothelial, Kupffer or stellate cells) can be separated by subsequent washing steps and gradient centrifugation [47, 48]. A combined digitonincollagenase perfusion allows the isolation and separation of periportal and perivenous hepatocytes from a single liver [49]. Cryopreservation of hepatocytes offers the opportunity to store the cells for months or even years, making it possible to repeat experiments with cells from the same donor and hence to increase reproducibility [50]. However, cell viability, attachment, liver-specific metabolism and protein neogenesis can be reduced or impaired after thawing [51]. There is no standard protocol for the different culture techniques and each laboratory uses its own standards. The main factors influencing the quality of the culture include media composition, extra cellular matrix environment, cell density and the properties of other cells in co-cultures. There is partial agreement on some of the parameters and we will describe the effects of several different culture methods on hepatocyte characteristics later in this chapter. 7.2.4

Hepatocyte Suspension Cultures

Hepatocyte preparations usually contain about 40–70 × 106 cells/g liver, with 85–95 % viability and less than 5 % contaminating nonparenchymal cells after purification. In suspension, the survival of cells is short and strongly depends on the medium composition, but is normally not longer than a few hours [52]. Although the system is fairly high throughput, easy to use and preserves most of the metabolising enzymes at in vivo levels for a short time, it is only useable for acute toxicology or metabolism studies. By capturing the cells into beads of alginate, the survival-time can be prolonged to 24 h [53]. However, the lack of functional bile canaliculi, cell polarity, cell–cell and cell–matrix contacts limits the use of alginate-embedded cells for drug transporter studies. 7.2.5

Short- and Long-Term Hepatocyte Cultures

Survival time in culture can be increased if hepatocytes are cultured on adhesive surfaces, for example, tissue culture vessels coated with extracellular matrix components. The most commonly used models are the monolayer culture, where hepatocytes are usually attached r and the sandwich culture system, where cells to dried films of collagen I or Matrigel are embedded between two layers of gelled extracellular matrix proteins. Intracellular signalling is closely connected to the interaction between extracellular matrix, cell-adhesion molecules and the cytoskeleton and therefore has major impact on gene expression and the metabolic capacity of cells [54, 55]. Therefore, the application of extracellular matrix to hepatocyte cultures is one step towards the creation of a more in vivo-like environment.

148

Hepatotoxicity

Unfortunately, hepatocytes still lose many of their specific functions during isolation [56, 57] or during the first few days if maintained in monolayer culture [58]. Today, cultures of primary rat and human hepatocytes are used in a variety of pharmacological and toxicological experiments. For example, the evaluation of hepatic drug uptake and metabolism, drug–drug interactions and hepatotoxicity. The development of long-term primary hepatocyte cultures would also be a step towards the study of chronic effects in vitro. Culturing hepatocytes in a sandwich configuration between two layers of gelled extracelr lular matrix proteins (e.g. collagen I or Matrigel ), has prolonged the time of cultures r displaying hepatocyte-specific functions dramatically [59–62]. Matrigel is a laminin-rich preparation from the Engelbreth–Holm–Swarm mouse sarcoma [63]. The composition of r Matrigel , with extracellular matrix components and growth factors is variable [64], while collagen I, mostly prepared from rat tail tendons, provides more standardised conditions. Another factor with great influence on morphological development and cell survival of hepatocytes in culture is basal medium formulation and the addition/omission of serum, specified hormone mixtures or other supplements [65–67]. Among the most frequently used basal media, Dulbecco’s modified Eagle medium (DMEM), modified Chee’s medium (MCM) and Williams’ medium E (WME), the DMEM/F12 mix seems most appropriate to maintain liver-specific function and rebuild bile canaliculi in cell culture [68]. Another report identified WME and DM-160 medium to be the most suitable with regard to cell attachment, survival and albumin secretion among ten media tested under serum-free conditions [69]. Once the cells are in culture, the addition of the glucocorticoid dexamethasone (DEX) at nanomolar concentrations, is essential for the long-term preservation of hepatocytespecific functions, polygonal hepatocyte morphology, the structural integrity of cytoplasmic membranes and bile canaliculi-like structures [70, 71]. If DEX is added at micromolar concentrations or cultures are supplemented with Rifampicine, Phenobarbital (PB) or 3methylcholantrene, CYP expression is induced and by this means the metabolic activity can be increased [72]. Apart from maintaining CYP expression, PB is also able to suppress hepatocyte growth [73]. Insulin enhances the attachment of hepatocytes and alters their morphology [74]. Selenium, a structural component of the enzyme glutathione peroxidase, which plays an essential role in the neutralization of metabolically generated peroxides, has also been shown to be beneficial when added to the medium [75]. The addition of mitogenic compounds, for example, epidermal growth factor, leads to hepatocyte proliferation, allowing longer culturing, but on the other hand leads to the ‘down-regulation’ of metabolic enzymes and thereby evokes dedifferentiation. Since primary cell cultures are inherently non-sterile, antibiotics have to be added to the culture. Therefore, it has to be taken into account that the antimycotic agent Fungizone (Amphotericin) stresses the cells to a considerable extent and should not be added routinely. Apart from extracellular matrix application and media formulation, cell density also influences the morphology and function of hepatocytes in culture [55]. The formation of bile canaliculi-like structures declines along with the expression and function of MRP2 and MDR1a/b as cell density is reduced [76]. If seeded at very high densities (almost 100 % confluency) cells do not tend to spread out that much in monolayer cultures and display hepatocyte-like morphology for 1–2 additional days, but then they also start to detach from the surface. It has to be taken into account that the same cell density (cells/cm2 ) will result in unequal confluency of cultures seeded on either collagen film or gel, due to the fact

Application of Hepatocyte Cultures to Predict Toxicities Collagen monolayer

149

Collagen sandwich time after seeding

4h

+ serum

– serum

+ serum

– serum

1d

3d

7d

14 d

Figure 7.1 Effect of extracellular matrix environment and media formulation on morphological development and structural integrity in primary rat hepatocyte cultures. Hepatocytes were isolated from male Wistar rats using a modification of the two-step perfusion method described by Seglen [39]. Cells were cultured for the indicated times on collagen monolayer or in a collagen gel sandwich with serum-free or serum-containing medium; media were changed daily. Arrows indicate bile canaliculi-like structures. The white scale bar in the bottom right of each image corresponds to 200 μm

that hepatocytes in a three-dimensional extracellular matrix environment require less area. Since polar differentiation of hepatocytes in culture is only visible in cell aggregates, cell density should always be close to confluency (≥ 90 %). Generally, after being in culture, cells should not be removed from the incubator longer than necessary and not be submitted to direct sunlight. Figure 7.1 gives an overview of the morphological changes occurring over time under different cell culture conditions. Since it is well known that serum enhances the surface attachment ability of hepatocytes [77], cells are generally seeded in medium containing fetal calf serum, regardless of the subsequent culture conditions.

150

Hepatotoxicity

A morphological distinction between hepatocytes seeded on collagen film (monolayer) or collagen gel (sandwich) is already visible after 4 h. Monolayer-cultured cells adopt their polygonal shape and establish extensive cell-to-cell contacts; whereas in sandwich culture this takes markedly longer, being still mostly spherical and singular after 4 h, probably resulting from the cells’ immersion in the collagen gel (Figure 7.1). In general, monolayercultured cells appear more flattened than sandwich-cultured cells, because of the lack of a three-dimensional extracellular matrix environment. All cells display clear cytoplasm and well-delineated plasma membranes. After overnight incubation, sandwich-cultured hepatocytes also form aggregates and the typical cuboidal cell shape is seen (Figure 7.1). In both culture systems, the formation of structures resembling bile canaliculi (light areas between cells, Figure 7.1, arrows) starts and increased during the following 24 h, but to a greater extent in the serum-free cultures (not shown). This is consistent with the findings of Terry and Gallin [78] who reported that serum is capable of inhibiting the re-establishment of bile canaliculi in hepatocyte cultures. After 72 h of culture on collagen film, there is a strong perturbation of cell morphology in both serum-free and serum-containing cultures (Figure 7.1). Monolayer-cultured hepatocytes spread out and formed fibroblast-like protrusions, the nuclei volume was increased and the cytoplasm appeared granulated. There are no longer well-delineated plasma membrane borders and bile canaliculi-like structures have disappeared almost entirely. The morphological stability of hepatocytes seeded on collagen film can be increased considerably if these cultures are overlaid with collagen gel. However, the time of overlay is crucial for the phenotypic outcome. Since hepatocyte monolayer cultures do not have a steady appearance over time, even in the first days of culture, they should be overlaid within the first 24 h. The best results are obtained if the monolayer/sandwich hybrid is prepared at the initial medium change 4 h after seeding. In this configuration, the morphology at the time of overlay can be preserved for more than 1 week. If the hepatotypic morphology is lost after several days on collagen film, it will not be restored after overlay – only the state at the time of overlay will be preserved (data not shown). By overlaying monolayer cultures of primary hepatocytes with collagen gel, it was possible to obtain cultures displaying some features of collagen sandwich cultures over extended periods of time (e.g. bile canaliculilike structures), although the overall appearance is still different and these hybrid cultures do not last as long as sandwich cultures. For cells cultured in a collagen sandwich, there is no spreading visible; polygonal cell shape and plasma membrane boundaries are still stable after 72 h of culture (Figure 7.1). However, for hepatocytes incubated in serum-containing medium, there is a noticeable deterioration in cytoplasmic integrity and the stability of canaliculi-like structures. In contrast, primary rat hepatocytes in a collagen sandwich and incubated with serum-free media show steady aggregates of polygonally shaped cells with mostly clear cytoplasm and a stable bile canaliculi-like network (Figure 7.1). This is in agreement with the previously reported beneficial effects of an extracellular matrix overlay on cell viability and the preservation of normal morphology [60, 79, 80]. The fact that collagen-sandwich cultures have successfully been used for metabolism and induction studies [81–83] indicates that the collagen overlay does not interfere with the interaction of the cells with the test compounds. In conclusion, only serum-free collagen sandwich cultures of hepatocytes show no signs of deterioration when incubated for more than one week, maintaining polygonal shape, clear cytoplasm and stable bile canaliculi-like structures. The fact that bile canaliculi-like structures are stable


151

Figure 7.2 (a) Fibroblast growth in serum-supplemented primary rat hepatocyte monolayer cultures 10 days after seeding. (b) Co-culture of primary rat hepatocytes and Kupffer cells 2 days after seeding. Hepatocytes were isolated from male Wistar rats using a modification of the two-step perfusion method described by Seglen [39]. Kupffer cells were subsequently separated using a modified protocol originally described by Smedsrod and Pertoft [47]. Cells were cultured on collagen monolayer with serum-free medium. Arrows indicate Kupffer cells. The white scale bar in the bottom right of each image corresponds to 200 μm

over time, make these cultures especially valuable for transport studies [84, 85]. Furtherr do not more, hepatocytes seeded on gelled collagen and overlaid with a layer of Matrigel demonstrate superior survival in comparison to cells overlaid with another layer of gelled collagen, exemplified by their lack of stable bile canaliculi-like structures (unpublished data). Hence, using gelled collagen I as the extracellular matrix component in hepatocyte sandwich cultures seems most appropriate for long-term use. The morphological stability of DEX-supplemented serum-free collagen sandwich cultures of primary rat hepatocytes can endure for as long as 4 weeks. DEX can also improve and prolong the differentiation status of serum-containing cultures considerably. However, culturing hepatocytes with serum-supplemented medium may lead to fibroblast overgrowth after several days regardless of the fact whether the cells are cultured in monolayer or sandwich configurations (Figure 7.2(a)). 7.2.6

Co-Cultures

Hepatocytes compose about 60–70 % of the cells in the intact organ; however, liver toxicity may not always originate from these cells. Therefore, co-cultures of hepatocytes with other non-parenchymal liver cells, such as endothelial, Kupffer or stellate cells and also stable cell lines or fibroblasts can be applied to reflect a more physiological situation. Kupffer cells represent the second largest cell population of the liver and they are very important for cytokine signalling within the organ. The non-perenchymal Kupffer cells, which are resident macrophages, are located in the hepatic sinusoids, in between or on top

152

Hepatotoxicity

of endothelial cells, but also make contact to the parenchymal hepatocytes through their cytoplasmic extensions [86]. Thus, Kupffer cells can act as biosensors affecting hepatocyte and ultimately the livers’ fate through cytokine signalling. For example, the excretion of TNFα or nitric oxide can lead to inflammatory reactions or apoptosis [87, 88]. This may be especially relevant for the development of biologically derived pharmaceuticals. In order to investigate the ‘crosstalk’ between Kupffer cells and hepatocytes and its influence on cytotoxicity in vitro, co-cultures of the two can be used (Figure 7.2(b)). Different studies have shown the improvement of some hepatocyte-specific functions in co-culture with other cell types [89, 90] but also when using non-parenchymal cell conditioned medium for hepatocyte cultures [91]. Co-cultured hepatocytes can survive for several weeks and liver-specific metabolism and expression of Phase I and Phase II enzymes are maintained at higher levels than in pure hepatocyte cultures [92, 93]. 7.2.7

Spheroid Cultures

If a crude liver cell suspension is seeded into cell culture vessels and surface attachment is prevented by continuous shaking, spherical multicellular aggregates (spheroids) will form. Cell-to-cell contacts are re-established, hepatocytes are located on the inside, nonparenchymal cells on the outside and the deposition of extracellular matrix is seen throughout the spheroids. Sometimes, alginate or other materials are added to make up the internal structure of the spheres. Several studies have showed the positive effect of this culture method on the expression of hepatotypic genes and the maintenance of metabolic capacity [94, 95]. However, the fusion of smaller spheres may lead to hypoxic and necrotic cells dying at the centre of larger spheroids (>100 μm). Another problem could arise from the accumulation of bile in the centre of spheroids. The maintenance of prolonged functional activity of spheroids has been related to the restoration and stability of cell polarity and close cell-to-cell contacts [96]. 7.2.8

3D Bioreactor Cultures

Another skilful attempt to mimic a liver-like environment in vitro are bioartificial liver systems (3D-bioreactors). Their major advantage is the re-establishment of the 3D liver cyto-architecture with cell–cell contacts and a three-dimensional extracellular matrix environment, combined with continuous medium perfusion, providing a constant supply of oxygen and nutrients. The first systems established were hollow fiber bioreactors [97]. Today, hepatocyte monolayer and collagen sandwich cultures are being used for bioreactor setups [98]. In addition, spheroids, for example, seeded into channels that are etched into a silicon scaffold, have been reported [99]. It has been shown that cell viability, metabolic activity, including CYP and Phase II reactions, albumin and urea secretion, as well as general hepatotypic gene expression can be sustained over extended culture duration in 3D bioreactors [24]. However, considerable challenges arise from the construction of the bioreactors. For example, composition of the basement structure, flow rate and liquid pressure has to be taken into account. In fact, the shearing power generated by the correctly adjusted flow rate of the medium can improve the metabolic capacity of the culture [100]. Although 3D bioreactors provide high-quality metabolic data and allow the quantitative analysis of pharmacokinetic parameters from key animal species and humans, their use as a tool for high throughput screening is extremely limited due to the large operational effort.


7.2.9

153

Cell Lines

To overcome the problem of limited survival and availability of primary cultures, different cell lines are used in a variety of toxicological applications. Since most hepatic cell lines originate from tumours, they have lost the high degree of differentiation of a hepatocyte and their gene expression pattern is distinctively different from normal liver cells. In addition, many cell lines display genetic instability. For example, the frequently used human hepatoma cell line, HepG2, lacks expression of CYP2E1 along with other CYP isoforms and Phase II enzymes, making them insensitive to secondary toxic compounds [101]. To complicate matters, different sources of HepG2 cells can have very different enzyme profiles [102]. Several transfected variants of HepG2 have been constructed which express increased levels of drug-metabolizing enzymes, including CYP1A1, CYP1A2, CYP2E1 and glutathioneS-transferases [101]. These new clones might be particularly useful for the investigation of specific classes of genotoxic compounds and also for mechanistic studies. In vivo relevance may not always be given because expression of the cloned enzymes is not at physiological levels and only single enzyme functions can be analysed. Other cell lines, such as the immortalized but untransformed transgenic hepatocyte cell lines (e.g. MMH-GH [103]), have been used with promising results. Recently, the human hepatoma cell line HepaRG was described [104]. After application of a differentiation protocol, HepaRG cells display hepatocyte morphology and expression of drug-metabolizing enzymes at near in vivo levels [105, 106]. However, these novel cell lines have to be confirmed as a reasonable alternative cell-based assay for use in toxicological studies [107]. 7.2.10

Subcellular Fractions

The simplest liver in vitro models are sub-cellular fractions, such as, organ homogenates, microsomes, mitochondria or nuclei. Most sub-cellular fractions can be prepared and separated relatively easy by homogenization of the tissue and sequential centrifugation. They are available from a large number of species, including humans. Another major advantage is that they can be frozen until needed. Nevertheless, they are only suitable for short-term studies with specific questions such as drug inhibition, covalent binding or clearance studies. For example, liver supernatants (‘S9’) are used as an activation system for xenobiotics in in vitro genotoxicity assays (e.g. Ames-assay [108]). Mitochondria are added for the analysis of drug effects on respiration, ATP-synthesis and fatty acid oxidation. To acquire increased metabolic activity, CYP expression is often induced by treatment with Aroclor 1254 or a Phenobarbital/beta-naphthoflavone (BNF) mixture prior to preparation [109], leading to elevated and unphysiological expression. Apart from the limited viability over time, most systems have to be supplemented with cofactors to preserve enzymatic activity. Other disadvantages include the absence of complete enzyme systems, for example, the lack of Phase II enzymes in microsomes.

7.3

Endpoints for the Analysis and Characterization of Hepatocyte Cultures

The list of tests used by in vitro toxicologists to gain insight into the effect of a test substance on cells is extensive, some of which will be described in detail in this chapter. These range

154

Hepatotoxicity

from standard tests, e.g. cell viability measurements, to hepatocyte specific activity tests, such as bile production, CYP activity or drug transport, to mention only a few. In addition, the analysis of single genes or proteins with established molecular methods like real-time PCR or Western-blotting is commonly applied. In recent years ‘-omics’ methods have been introduced to generate a more complete understanding of the complexity of toxic events and to integrate this knowledge into Systems Biology approaches. 7.3.1

Morphology – High-Content Imaging

As seen in Figure 7.1, hepatocytes cultured in serum-free collagen sandwich cultures stay morphologically unchanged for a few weeks, offering the ability to investigate alterations in cellular structures induced by chemical treatment with classic light microscopy but also with the use of high-content imaging (HCI), which can be integrated into high-content screening (HCS). HCI is the automated multiparametric analysis of spatial and temporal changes of sub-cellular structures at the single cell level with visible or fluorescent light microscopy techniques [110–113]. The parameters that can be measured include vacuolization, formation of lipid droplets, nuclear morphology, plasma membrane integrity (blebs), cell number (proliferation), size or shape (e.g. neurone outgrowth), but also the analysis of molecular targets like nuclear translocation of transcription factors. By using multi-channel fluorescence detection, alternatively labelled molecular targets can be followed at the same time [114, 115]. Data acquisition is automated and all parameters can be recorded in parallel. Cells can be cultured in several standard tissue cultures formats, e.g. 96- or 384-well plates and many cells are analysed in each well simultaneously. In contrast to standard cytotoxicity assays, where single parameters are averaged over different populations of cells, HCI can create a wealth of data for the detailed analysis of toxic events [113]. This technology is particularly important for the analysis of sub-lethal toxicities. HCS data can also be implemented into Systems Biology approaches and may benefit drug discovery [110, 111]. O’Brien and colleagues demonstrated the ability to detect human hepatotoxicity with high sensitivity and specificity with the use of HCS [114]. 7.3.2

Functional Endpoints

There are large numbers of assays and kits to assess functional and biochemical parameters of cultured cells. In vitro toxicity in hepatocyte cultures is either investigated by nonspecific cytotoxicity or by the measurement of compound-specific endpoints. Chemicalinduced changes in cellular functions may be irreversible, ultimately leading to cell death, where others may be transient. Irreversible endpoints include the induction of apoptosis, which can be investigated by the measurement of increased caspase activity or the loss of plasma membrane integrity. Plasma membrane damage can be analysed by the detection of cytoplasmic enzyme release (e.g. lactate dehydrogenase, LDH) or the uptake of specific dyes such as Neutral Red and Trypan Blue into the cytoplasm. In addition, alterations in general hepatocyte functions like albumin, urea or bile secretion provide information on the impairment of cellular processes. The energy status of the cell is often used to determine cytotoxicity. This is achieved by studying the ATP content of the cells or the mitochondrial or enzymatic capacity to reduce tetrazolium salts (XTT, MTT, WST) [116]. Several compounds may cause oxidative stress, leading to glutathione (GSH) depletion. The induction of general stress proteins can be measured with ELISA (enzyme-linked


155

immunosorbent assay)-based assays. Another important hepatocyte specific feature is the uptake of xenobiotics in order to be metabolized and the export of metabolites. Drug transport can be studied by fluorescent dyes or analysis of bile acid transport with high performance liquid chromatography (HPLC) [84, 117]. 7.3.3

Gene Expression

Toxic events, leading to morphological and functional impairment, can also be investigated by analysing the underlying alterations that occur on the gene expression level in hepatocyte cultures. However, recent work has implied that the cell isolation procedure itself leads to gene-expression changes in hepatocytes, especially for specific CYPs [118]. This is caused by changes in the shape of isolated hepatocytes and the induction of ribonuclease activity during perfusion and the first few hours of culture [119]. In addition, one needs to take into account the gene expression changes that occur over time in untreated hepatocyte cultures. Figure 7.3 displays a visualization of the expression changes of 550 hepatotypic genes that occur over time in monolayer and sandwich cultures of primary rat hepatocytes. It is obvious that basal gene expression in liver tissue and freshly isolated hepatocytes is distinctly different from other cultures, regardless of the culture conditions (Figure 7.3 and

(a)

(b)

Comp. 3

– serum

late

Comp. 3 liver, fresh

Comp. 1

liver, fresh

Comp. 1

+ serum

Comp. 2 monolayer + serum sandwich + serum

Comp. 2 monolayer – serum sandwich – serum

early liver freshly isolated hepatocytes

Figure 7.3 Principal Components Analysis (PCA) of gene-expression alterations occurring over time in different primary rat hepatocyte cultures, as well as in intact liver (‘liver’) and fresh hepatocyte suspension (‘fresh’). Cells were cultured as monolayer and sandwich, with (+ serum) or without (− serum). Messenger RNA levels were determined with custom ‘Sentrix BeadChip Arrays’ from Illumina, containing 550 genes. Data from three biological replicates were analysed with the ‘Genedata Expressionist Analyst’ software. Culture samples were taken 2 h, 1 d, 2 d, 3 d, 5 d, 7 d and 9 d after the designated time t0 = 24 h after seeding. The PCA displays a non-supervised dimensionality reduction of the data to show the highest variance in the dataset. (a) Comparison of all four cell culture conditions or (b) only serum-free cultures with freshly isolated hepatocytes and intact liver. Darker shades of grey represent sandwich cultures or intact liver, while lighter shades of grey represent the respective monolayer cultures or freshly isolated hepatocytes

156

Hepatotoxicity – serum 9d

fresh

9d

Sandwich

fresh

+ serum

9d

9d

9d

fresh

9d

Monolayer

fresh

9d

9d

Figure 7.4 Heat maps, displaying gene-expression alterations occurring over time in different primary rat hepatocyte cultures, as well as in intact liver and fresh hepatocyte suspension. Cells were cultured as monolayer and sandwich with (+ serum) or without (− serum). Messenger RNA levels were determined with custom ‘Sentrix BeadChip Arrays’ from Illumina, containing 550 genes. Data from three biological replicates were analysed with the ‘Genedata Expressionist Analyst’ software. Culture samples were taken 2 h, 1 d, 2 d, 3 d, 5 d, 7 d and 9 d after the designated time, t0 = 24 h after seeding. Each square in a column or line represents the correlation of a given sample relative to another. Darker shades of grey indicate very high correlation, whereas lighter squares mean low correlation. Overall, there is a high similarity between liver and fresh cells (upper left corner, columns left of the arrows). Columns to the right of the arrows display cell culture samples with increasing incubation time

Figure 7.4, arrows). This has also been previously reported for human hepatocytes [120]. In addition, serum-containing cultures are clearly separated from serum-free cultures in the principal components analysis shown in Figure 7.3. Strikingly, serum supplementation is a major modulator of hepatic gene expression. Another important observation is the fact that the strongest gene expression changes were observed during the first days of culture, independent of the culture conditions [120, 121]. However, after an initial adaptation period


157

mRNA levels stabilized over time in serum-free cultures, but not in serum-containing cultures (Figure 7.4). These expression levels remained constant for a minimum of four weeks in sandwich cultures without serum. By comparing serum-free collagen sandwich and monolayer cultures ten days after seeding, gene expression in cells cultured between two layers of gelled collagen I appeared to be closer to the in vivo samples or freshly isolated hepatocytes (Figure 7.4). Thus, for stable long-term cultures of primary rat hepatocytes, the serum-free collagen sandwich culture seems most suitable. As well as the gene expression alterations occurring over time in the differentially cultured primary rat hepatocytes, there are differences in mRNA levels between the specific cell cultures at a given time-point [122]. Figure 7.5 shows the gene expression changes of 22 hepatotoxicity-related genes, occurring in dimethyl sulfoxide (DMSO)-treated primary rat hepatocyte monolayer or sandwich cultures, after 72 h in culture (relative to 24 h cultures). DMSO is a very commonly used solvent for test materials designated for the treatment of cells in culture. Su and Waxman [123] reported that the addition of 2 vol % DMSO did not affect CYP1A1 mRNA levels but had great influence on the expression patterns of other CYPs and several liver transcription factors. Furthermore, DMSO was capable of restoring normal CYP2B1, CYP3A1, CYP4A1, HNF4 and CEBPα levels. It also decreased initially elevated HNF6 expression and maintained constant levels of HNF3α, HNF3β and CEBPβ [123]. Moreover, DMSO is used for the hepatocyte differentiation of the hepatoma derived cell line HepaRG [104–106]. However, organic solvents impact the metabolic activity of hepatocytes and therefore the concentration used for dissolving test agents must be used with caution [124–126]. As seen in Figure 7.5, there is a distinct difference in the observed expression patterns for the four cell cultures tested. There is a marked ‘downregulation’ for most of the assayed genes in the serum-containing monolayer culture whereas in the three other cultures almost all genes reached higher abundance levels at 72 h. The observed rise of MDR1 expression over time in primary hepatocytes was also reported by Fardel et al. [127] and Chieli et al. [128]. The increase in MDR1 expression over time is less pronounced in the respective sandwich cultures, which is in agreement with Lee [129]. Cyclins D1 and G1 were elevated over time in the serum-containing sandwich culture (Figure 7.5) as well as in the respective monolayer relative to every other culture condition at 72 h (not shown). This indicates a disturbed cell cycle regulation in normally non-dividing hepatocytes, especially in the serum-containing cultures. Another noticeable feature is the increase in the heat shock factor HSP70 over time in the serum-containing cultures (Figure 7.5). When comparing the cultures at 24 h, there was an increase under serum-free conditions, which was still present at 72 h in the respective collagen monolayer culture but had disappeared in the collagen sandwich. GADD45α expression, although altered only weakly, displayed lower levels in the serum-free monolayer and both sandwich cultures compared to the serum-containing monolayer culture. Since GADD45α is associated with stress signalling, this indicates that the level of cellular stress was highest in the serum-containing monolayer culture, making this culture less appropriate for the detection of toxic effects. Since CYP enzymes are the major monoxygenases involved in liver Phase I metabolism, the expression of these molecules in hepatocyte cultures is especially important. Metabolic conversion of chemicals may lead to altered drug–drug interaction patterns and subsequently cause possible adverse effects. Therefore, the CYP inducing potential of a drug is of great interest. Cultures of primary rat hepatocytes show significant alterations in CYP expression over time (Figure 7.5) but also in relation to each other. In particular, the CYP2C

158

Hepatotoxicity

+/–

1

Fold regulation

A B C D

–

–

Figure 7.5 Gene-expression analysis of 22 hepatotoxicity related genes in primary rat hepatocyte cultures. Cells were cultured as monolayer with (A) or without serum (B) and as collagen sandwich with (C) or without serum (D). Shown are values of fold regulation for the temporal gene expression changes occurring at 72 h compared to the values at 24 h (set as ‘1’) in the respective DMSO-treated cell culture. Messenger RNA levels were determined using quantitative real-time PCR with the ‘TaqMan Low Density Arrays’ (Applied Biosystems). Values were calculated by the efficiency-corrected comparative CT method with 18S rRNA serving as calibrator. Bars illustrate mean values from quadruplicate measurements with standard deviations. Elevated levels of gene expression are indicated by positive figures > + 1 and ‘down-regulation’ is indicated by negative figures < − 1. Please note that the positive, as well as the negative, y-axes are scaled logarithmically (data taken from Tuschl and Mueller (2006) [122]). MRP3, multidrug resistance protein 3; AFP, alpha-fetoprotein; AP, alkaline phosphatase, tissue-nonspecific; ApoE, apolipoprotein E; FABP2, fatty acid binding protein 1; FAS, fatty acid synthase; GADD45α, growth arrest and DNA-damage-inducible 45 alpha; GSTα1, glutathione S-transferase alpha 1; HNF4α, hepatocyte nuclear factor 4 alpha; HSP70, heat shock 70kD protein 1A; MDR1, multidrug resistance transporter 1; COX2, prostaglandin-endoperoxide synthase 2; ACOX, acyl-coenzyme A oxidase 1, palmitoyl; OAT1, organic anion transporter 1; SOD2, superoxide dismutase 2, mitochondrial; SulT1c2, sulfotransferase family, cytosolic, 1C, member 2; Txn2, thioredoxin 2

and CYP4A1 genes are considerably decreased in the serum-containing monolayer culture after 3 days (Figure 7.5). CYP1A1 levels are strongly increased in the serum-free monolayer cultures and to a lesser extent in the sandwich cultures. After 72 h in culture, CYP mRNAs are more abundant in the serum-free monolayer cultures and both sandwich cultures in comparison to the serum-containing monolayer. Indeed, there is evidence in the literature


159

that the presence of DEX [130] and a three-dimensional extracellular matrix environment [85] helps to maintain constant levels of CYP expression. The low abundance of the examined CYP mRNAs at 72 h indicates a limited metabolic capacity in serum-containing collagen monolayer cultures, making only short-term applications acceptable [58]. Analysis of CYP expression in primary rat hepatocytes, cultured for 10 days under various conditions, shows distinct patterns for specific CYPs (Figure 7.6). After being relatively constant in the first days of culture, there was a strong reduction of CYP1A2 and CYP2E1 expression after 3 days in all cultures, although the reduction was lowest in the serum-free sandwich. The major Phase I enzymes in rat, CYP2C and CYP3A1, were strongly repressed in the serum-containing cultures over time, displaying stable values close to in vivo in the serum-free cultures until 10 days after seeding. In general, variations in gene expression over time were least pronounced in the serum-free collagen sandwich cultures. Therefore, this culture seems most adequate for the study of gene expression alterations triggered by drugs or chemicals. 7.3.4

The ‘-omics’ Challenge – Systems Biology

The detection of gene expression changes in differentially cultured cells or induced by chemical treatment is no longer limited to several genes at a time, since whole genome arrays became available. The sequencing of the entire chromosomal DNA sequences of human as well as many ‘laboratory’ animal species enabled the synthesis of whole genome arrays. This novel technique created a new field of research, called genomics, often termed transcriptomics, since the molecules that are detected are the transcribed mRNAs. To be able to analyse the whole cellular population of specific molecules, there is now also proteomics, lipidomics, metabolomics, etc. and the development of new methods in the ‘-omics’ fields is rapidly growing. The combination of different ‘-omics’ techniques, called ‘cytomics’, creates an enormous amount of data that could generate a more detailed and complete insight into cellular physiology [131]. However, one of the most challenging tasks of the new ‘-omics’ era techniques is data storage, handling and analysis. Large computational efforts and bioinformatics methods are applied to create algorithms in order to elucidate statistically and biologically significant results. Therefore, rather than looking at the single gene or protein level, profile signatures are determined, with the goal to identify, for example, specific gene patterns that are predictive for toxicity. There is a lot of activity to create databases with compound signatures in order to classify chemicals [132, 133]. This is important for the pharmaceutical industry, especially in the field of toxicology, for hazard identification and characterization of candidate molecules early in development. In addition, toxicogenomics data are valuable for the elucidation of the underlying mechanism of toxicity [134–136]. Many of these data are put together in the Comparative Toxicogenomics Database (CTB) to achieve a better understanding of the etiology of human diseases [137]. Apart from identifying different types of hepatotoxicitiy, there have been attempts to find biomarkers for the early identification of hepatocarcinogenesis with the use of toxicogenomics and proteomics methods [138–140]. There is hope that these new methods will make it possible to detect changes in the molecular pattern of the cell that are indicative of the pathological endpoint, at subtoxic doses, before the endpoint becomes histopathologically detectable [141]. This could facilitate a considerable reduction in the time needed to obtain results and the number of animals used in toxicity testing [142]. In addition, there are several suppliers of

Figure 7.6 Long-term CYP expression in primary rat hepatocyte cultures. Cells were cultured as monolayers with (M+) or without serum (M−) and as collagen sandwiches with (S+) or without serum (S−). Messenger RNA levels of CYP1A2 (a), CYP2E1 (b), CYP2C (c) and CYP3A1 (d) were determined with custom ‘Sentrix BeadChip Arrays’ from Illumina, containing 550 genes. Data from three biological replicates were analysed with the ‘Genedata Expressionist Analyst’ software. Culture samples were taken 1 d (A), 2 d (B), 3 d (C), 5 d (D), 7 d (E) and 9 d (F) after the designated time t0 = 24 h after seeding. Bars illustrate gene expression changes relative to samples from respective cultures at t0. Non-significant changes are not shown (missing bars)


161

commercial databases, with data from in vivo but also in vitro experiments (e.g. GENELOGIC (USA), ICONIX Biosciences (USA)). Although most data are generated from in vivo liver samples, there are efforts to build databases for the screening of hepatotoxicity based on primary hepatocyte cell culture experiments, making it necessary to characterize the cell culture model used, by genomic and proteomic approaches [121, 122, 143]. One step further is the Systems Biology approach, the mathematical modelling of biological processes based on the knowledge of cellular physiology, ultimately leading to in silico biology or the ‘virtual cell’. Although there are various kinds of software for the description and modelling of signalling pathways, the complexity of biological processes limits the application of systems biology, even though there may be an impact on drug discovery in the future [144–146].

7.4 7.4.1

Application of Short and Long-Term Hepatocyte Cultures Cytotoxicity Measurements in Long-Term Hepatocyte Sandwich Cultures

Successive applications are not feasible for cytotoxicity measurements in monolayer cultures. However, repeated dosing of long-term cultures of primary hepatocytes offers the possibility to detect ‘chronic’ effects. This is especially interesting since most in vitro models have very limited functionality over extended periods of time. Rather than generating acute toxic effects with high concentrations of a test compound, the utilization of substance levels close to pharmacologically active plasma concentrations increases the in vivo relevance of the data generated. Figure 7.7 gives an example of long-term cytotoxicity measurement in collagen sandwich cultures of primary human hepatocytes. Cells were treated daily for eight days and the cellular energy status was analysed by measuring the ATP content. The different substances used in this experiment lead to distinct reactions in culture. In comparison to the results obtained after two days, the non-toxic negative control, Metformin, did not show increased cytotoxicity after eight days, which supports the stability of the cultures over time. In contrast, troglitazone displayed increasing toxicity with increasing incubation times. Another feature was displayed by compound ‘X’. After showing considerable toxicity after two days, the IC50 values increased with incubation time, indicating adaptation of the cells to the toxic properties of this substance. This example clearly shows that with long-term cytotoxicity measurements more detailed information on the toxic potential of substances can be obtained. 7.4.2

Cytochrome P450 Induction in Long-Term Hepatocyte Sandwich Cultures

Short-term cultures of primary hepatocytes are frequently used to assess the CYP inducing capability of candidates in preclinical drug development. CYP induction may result in unfavourable drug–drug interactions, perhaps leading to the accelerated/reduced clearance of a co-administered drug. With regard to the more stable expression of CYPs, investigating CYP induction in serum-free collagen sandwich cultures of primary hepatocytes may give more reproducible results. Figure 7.8 shows an example of the analysis for CYP1A1/2 induction, on the mRNA level, in collagen sandwich cultures treated with the model inducers 3-methylcholanthrene (3-MC) and BNF after 3 days in culture. 3-MC and BNF induced

162

Hepatotoxicity

Figure 7.7 Long-term cytotoxicity measurement in serum-free collagen sandwich cultures of primary human hepatocytes. Cells were treated with the indicated concentrations of the negative control Metformin (A), the test substance compund X (B) and the positive control troglitazone (C) for 8 days with daily medium changes. Cytotoxicity was determined as the cells’ ATP contents relative to the vehicle-treated control and is displayed after 2 days (a) and after 8 days (b) of treatment. The boxes highlight the increased cytotoxicity of the indicated troglitazone concentrations after 8-day treatment while the circles show the decreased sensitivity to compund X after 8 days

CYP1A, where CYP1A1 was more strongly induced in comparison to CYP1A2. These results support the applicability of long-term hepatocyte cultures for CYP-induction studies after extended periods of time. In addition, if treatment is stopped but culturing is continued, this offers the ability to define whether CYP induction can be resolved after a recovery period. It has even been suggested that serum-free collagen sandwich cultures can be used to examine CYP induction of several test compounds consecutively in one culture with recovery phases between treatment stages (PRIMACYT Cell Culture Technology GmbH, personal communication). This would be a step towards higher throughput and also help to further reduce animal usage in preclinical drug development. 7.4.3

Gene Expression Profiling in Short-Term Hepatocyte Cultures – Comparison with in vivo

Although hepatocyte sandwich cultures are clearly superior to monolayer cultures with respect to hepatocyte-like morphology and expression of hepatotypic genes, valuable results can also be obtained from short-term experiments with monolayer cultures. We elucidated whether the drug development candidate and mixed PPARα/γ agonist EMD 392949 (EMD) showed a species-specific induction of the major drug metabolizing enzymes in rats and


163

Figure 7.8 CYP induction in serum-free primary rat hepatocyte sandwich cultures. Cells were treated 3 days after seeding with 10 μM 3-MC (A), 10 μM BNF (B) and 20 μM BNF (C). After 24 h treatment, CYP1A1 and CYP1A2 mRNA levels were determined using quantitative ‘TaqMan’ real-time PCR. Values were calculated by the efficiency-corrected comparative CT method with 18S rRNA serving as calibrator. Bars illustrate mean values of fold induction from triplicate measurements of a single PCR run with standard deviations relative to the vehicle control

humans. Furthermore, EMD was compared to fenofibrate, a PPARα agonist, Muraglitazar, a mixed PPARα/γ agonist and pioglitazone, a PPARγ agonist. Compound-specific gene expression changes were examined in primary rat and human hepatocytes, cultured on collagen I coated plates and treated for 24 h and 72 h. Strikingly, alterations in the expression of selected PPARα and PPARγ target genes that were observed in liver samples from a subchronic rat toxicity study were reproduced in primary rat hepatocyte cultures (Figure 7.9). In addition, species-specific reactions could also be detected. For example, the induction of inter-individual and inter-species variability of PPARα- or PPARγ -dependant reactions in rodents or primates, and the CYP-inducing potential of several compounds that was observed in human but not in rats (Figure 7.9). Interestingly, in rat and human hepatocyte cultures some compound-induced alterations were more prominent at the 24 h time-point, where others were more distinct after 72 h. Another feature worth mentioning was the difference in donor-dependent response to the pharmaceuticals in human hepatocyte cultures (data not shown). This clearly shows the necessity to analyse cells from several human donors in order to obtain reliable results.

164

Hepatotoxicity

Fold regulation

(a)

+/– 1

–10

–100

Figure 7.9 Gene expression analysis in rat liver samples (a) and serum-free rat (b) and human (c) hepatocyte monolayer cultures. Messenger RNA levels were determined using quantitative real-time PCR with ‘TaqMan Low Density Arrays’ (Applied Biosystems). Values were calculated by the efficiency-corrected comparative CT method with 18S rRNA serving as calibrator. Bars illustrate mean values from three biological replicates/donors with standard deviations. Elevated levels of gene expression are indicated by positive figures > + 1, while ‘down-regulation’ is indicated by negative figures < − 1. Please note that the positive and the negative y-axes are scaled logarithmically in (a) and (b) and linear in (c). Shown are values of fold-regulation for the treatment-induced gene expression changes relative to the vehicle-treated control (set as 1). (a) Treatment of rats with 3 mg/kg (lo) and 100 mg/kg (hi) EMD 392949 for 17 weeks. (b) Rat and (c) human hepatocyte monolayer cultures treated for 72 h with 3 μM (A), 30 μM (B) and 100 μM (C) EMD 392949, 100 μM fenofibrate (D), 3 μM pioglitazone (E) and 1 μM (F) or 30 μM (G) muraglitazar. CPT1A, carnitine-palmitoyltransferase 1A; G6Pc, glucose-6-phosphatase, catalytic subunit; GCK, glucokinase; HNF4α, hepatocyte nuclear factor 4 alpha; PXR, pregnane-X-receptor; CAR, constitutive androstane receptor; ACOX, acyl-coenzyme A oxidase 1, palmitoyl; MDR1, multidrug resistance transporter 1; ADPGK, ATP-dependent glucokinase; PEPCK, phosphoenolpyruvate carboxykinase 1, soluble; OATP2, organic anion transporter 2; OATP8, organic anion transporter 8


Fold regulation

(b)

+/– 1

–10

Fold regulation

(c)

+/– 1.0 – –

Figure 7.9

(Continued)

165

166

Hepatotoxicity

Overall, rat hepatocytes in vitro perfectly reflected the observed effects in vivo for selected deregulated PPAR target genes. These results showed that primary hepatocyte cultures could serve as an appropriate model to predict effects observed in vivo.

7.5

Summary

Many different in vitro models are currently in use with the goal to predict human hepatotoxicity. Each system has specific limitations with regards to availability, throughput, viability of the cells over time and the possibility to analyse a variety of parameters. We have presented here long-term hepatocyte sandwich cultures as a valuable in vitro model for the study of hepatotoxicity. Hepatocyte preparations are available from many species, including humans, allowing the study of species-specific effects. The survival of hepatocytes in sandwich cultures and the maintenance of hepatotypic functions are stably preserved for a few weeks. This offers the opportunity to increase incubation time in order to study long-term effects and also to apply pharmacologically relevant concentrations of the test compound. Since the number of cells needed for the analysis of a specific parameter is usually low, depending on the sensitivity of the method, and experiments can be conducted in various cell culture formats, cultures of primary hepatocytes can be considered a medium-to-high throughput system. Another important feature is the ability to analyse many different methods in parallel, making it possible to obtain various data from the same sample. Optical clarity and stable morphology allow the use of the full range of optical methods, including histopathological methods and HCI. Classic in vitro cytotoxicity measurements, as well as many biochemical assays, can be performed since most hepatotypic functions are preserved. In addition, cell culture samples are compatible with most ‘-omics’ analyses. Hence, a diversity of results can be created from hepatocyte cultures and integrated into HCS and Systems Biology approaches to bring them into biological context. Therefore, we consider long-term hepatocyte sandwich cultures a valuable tool for the assessment of hepatoxicity in vitro, in order to reduce time and the number of animals used in preclinical drug development, and also to provide mechanistic data for the understanding of toxic mechanisms.

Acknowledgements We thank Dr P.-J. Kramer, Merck KGaA for supporting this work and Professor L. Richert for providing human hepatocytes.

References [1] F. Ballet, Hepatotoxicity in drug development: detection, significance and solutions, J. Hepatol., 26(Suppl. 2), 26–36 (1997). [2] M. Fung, A. Thornton, K. Mybeck, W. Hsiao-Hui, K. Hornbuckle and E. Muniz, Evaluation of the characteristics of safety withdrawal of prescription drugs from worldwide pharmaceutical markets 1960–1999, Drug Inf. J., 35, 293–317 (2001).


167

[3] R.J. Andrade, R. Carmargo, M.I. Lucena and R. Ganzalez-Grande, Causality assessment in drug-induced hepatotoxicity, Expert Opin. Drug Saf., 3, 329–344 (2004). [4] N. Kaplowitz, Drug-induced liver disorders. Implications for drug development and regulation, Drug Saf., 24, 483–490 (2001). [5] J.H. Lewis, Drug-induced liver disease, Curr. Opin. Gastroenterol., 18, 307–313 (2002). [6] M.W. Russo, J.A. Galanko, R. Shrestha, M.W. Fried and P. Watkins, Liver transplantation for acute liver failure from drug-induced liver injury in the United States, Liver Transpl., 10, 1018–1023 (2004). [7] N. Kaplowitz, Idiosyncratic drug hepatotoxicity, Nat. Rev., 4, 489–499 (2005). [8] J.J. Xu, D. Diaz and P.J. O’Brien, Applications of cytotoxicity assays and pre-lethal mechanistic assays for assessment of human hepatotoxicity potential, Chem. Biol. Interact., 150, 115–128 (2004). [9] H. Olson, G. Betton, J. Stritar and D. Robinson, The predictivity of the toxicity of pharmaceuticals in humans from animal data – an interim assessment, Toxicol. Lett., 102–103, 535–538 (1998). [10] H. Olson, G. Betton, D. Robinson, K. Thomas, A. Monro, G. Kolaja, P. Lilly, J. Sanders, G. Sipes, W. Bracken, M. Dorato, K. Van Deun, P. Smith, B. Berger and A. Heller, Concordance of the toxicity of pharmaceuticals in humans and in animals, Regul. Toxicol. Pharmacol., 32, 56–67 (2000). [11] U.A. Boelsterli, Disease-related determinants of susceptibility to drug-induced idiosyncratic hepatotoxicity, Curr. Opin. Drug Discov. Devel., 6, 81–91 (2003). [12] U.A. Boelsterli, Idiosyncratic drug hepatotoxicity revisited: new insights from mechanistic toxicology, Toxicol. Mech. Methods, 13, 3–20 (2003). [13] H. Jaeschke, G.J. Gores, A.I. Cedeerbau, J.A. Hinson, D. Pessayre and J.J. Lemasters, Mechanisms of hepatotoxicity, Toxicol. Sci., 65, 166–176 (2002). [14] W.M. Lee, Drug-induced hepatotoxicity, New Engl. J. Med., 349, 474–85 (2003). [15] M.R. Slaughter, H. Thakkar and P.J. O’Brien, Effect of diquat on the antioxidant system and cell growth in human neuroblastoma cells, Toxicol. Appl. Pharmacol., 178, 63–70 (2002). [16] W. Lewis, B.J. Day and W.C. Copeland, Mitochondrial toxicity of NRTI antiviral drugs: an integrated cellular perspective, Nat. Rev. Drug. Discov., 2, 812–822 (2003). [17] W.G. Schoonen, W.M. Westerink, J.A. de Roos and E. Debiton, Cytotoxic effects of 100 reference compounds on HepG2 and HeLa cells and of 60 compounds on ECC-1 and CHO cells. I. Mechanistic assays on ROS, glutathione depletion and calcein uptake, Toxicol. in vitro, 19, 505–516 (2005). [18] W.G. Schoonen, W.M. Westerink, J.A. de Roos and E. Debiton, Cytotoxic effects of 110 reference compounds on Hep G2 and for 60 compounds on HeLa, ECC-1 and CHO cells. II. Mechanistic assays on NAD(P)H, ATP and DNA contents. Toxicol. in vitro, 19, 491–503 (2005). [19] R.G. Ulrich, J.A. Bacon, C.T. Cramer, G.W. Peng, D.K. Petrella, R.P. Stryd and E.L. Sun, Cultured hepatocytes as investigational models for hepatic toxicity: practical applications in drug discovery and development, Toxicol. Lett., 82-83, 107–115 (1995). [20] E.F. Brandon, C.D. Raap, I. Meijerman, J.H. Beijnen and J.H. Schellens, An update on in vitro test methods in human hepatic drug biotransformation research: pros and cons, Toxicol. Appl. Pharmacol., 189, 233–246 (2003). [21] A. Guillouzo, Liver cell models in in vitro toxicology, Environ. Health Perspect., 106(Suppl. 2), 511–32 (1998). [22] D.A. Groneberg, C. Grosse-Siestrup and A. Fischer, In vitro models to study hepatotoxicity, Toxicol, Pathol., 30, 394–399 (2002). [23] D. Farkas and S.R. Tannenbaum, In vitro methods to study chemically-induced hepatotoxicity: a literature review, Curr. Drug Metab., 6, 111–125 (2005).

168

Hepatotoxicity

[24] A. Sivaraman, J.K. Leach, S. Townsend, T. Iida, B.J. Hogan, D.B. Stolz, R. Fry, L.D. Samson, S.R. Tannenbaum and L.G. Griffith, A microscale in vitro physiological model of the liver: predictive screens for drug metabolism and enzyme induction, Curr. Drug Metab., 6, 569–591 (2005). [25] G. Scholz, I. Pohl, E. Genschow, M. Klemm and H. Spielmann, Embryotoxicity screening using embryonic stem cells in vitro: correlation to in vivo teratogenicity, Cells Tissues Organs, 165, 203–211 (1999). [26] K. De Smet, S. Beken, M. Depreter, F. Roels, A. Vercruysse and V. Rogiers, Effect of epidermal growth factor in collagen gel cultures of rat hepatocytes. Toxicol. in vitro, 13, 579–585 (1999). [27] K.Z. Ching, K.A. Tenney, J. Chen and E.T. Morgan, Suppression of constitutive cytochrome P450 gene expression by epidermal growth factor receptor ligands in cultured rat hepatocytes, Drug Metab. Dispos., 24, 542–546 (1996). [28] D.C. Dahlin, G.T. Miwa, A.Y. Lu and S.D. Nelson, N -acetyl- p-benzoquinone imine: a cytochrome P-450-mediated oxidation product of acetaminophen, Proc. Natl. Acad. Sci. USA, 81, 1327–1331 (1984). [29] J.A. Holme, D.C. Dahlin, S.D. Nelson and E. Dybing, Cytotoxic effects of N -acetyl- pbenzoquinone imine, a common arylating intermediate of paracetamol and N -hydroxyparacetamol, Biochem. Pharmacol., 33, 401–406 (1984). [30] D.J. Miner and P.T. Kissinger, Evidence for the involvement of N -acetyl- p-quinoneimine in acetaminophen metabolism, Biochem. Pharmacol., 28, 3285–3290 (1979). [31] E.M Gordon, M.C. Douglas, P. Jablonski, J.A. Owen, A. Sali and J.M. Watts, Gastroduodenal hormones and bile-secretion studies in the isolated perfused pig liver, Surgery, 72, 708–721 (1972). [32] O. Warburg, Versuche an u¨ berlebendem Karzinomgewebe, Biochem. Z., 142, 317–333 (1923). [33] K. Brendel, R.L. McKee, V.J. Hruby, D.G. Johnson, A.J. Gandolfi and C.L. Krumdieck, Precision cut tissue slices in culture: a new tool in pharmacology, Proc. West Pharmacol. Soc., 30, 291–293 (1987). [34] C.L. Krumdieck, J.E. Dos Santos and K.-J. Ho, A new instrument for the rapid preparation of tissue slices. Anal. Biochem., 104, 118–123 (1980). [35] P.F. Smith, A.J. Gandolfi, C.L. Krumdieck, C.W. Putnan, C.F. Zukowski, W.M. Davis and K. Brendel, Dynamic organ culture of precision liver slices for in vitro toxicology, Life Sci., 36, 1367–1375 (1985). [36] C. Lerche-Langrand and H.J. Toutain, Precision-cut liver slices: characteristics and use for in vitro pharmaco-toxicology, Toxicology, 153, 221–253 (2000). [37] J.A. Beamand, R.J. Price, M.E. Cunninghame and B.G. Lake, Culture of precision-cut liver slices: effect of some peroxisome proliferators, Food Chem. Toxicol., 31, 137–147 (1993). [38] S. Ekins, O.I. Murray, J.A. Williams, M.D. Burke, N.C. Marchant and G.M. Hawksworth. Quantitative differences in phase I and II enzyme activities between rat precision-cut liver slices and isolated hepatocytes, Drug Metab. Dispos., 23, 1274–1279 (1995). [39] O. Seglen, Preparation of Isolated Rat Liver Cells, Meth. Cell Biol., 13, 29–83 (1976). [40] R.B. Howard, A.K. Christensen, F.A. Gibbs and L.A. Pesch, The enzymatic preparation of isolated intact parenchymal cells from rat liver, J. Cell Biol., 35, 675–684 (1967). [41] G. Fabre, R. Rahmani, M. Placidi, J. Combalbert, J. Covo, J.P. Cano, C. Coulange, M. Ducros and M. Rampal, Characterization of imidazolam metabolism using human hepatic microsomal fractions and hepatocytes in suspension obtained by perfusing whole human livers, Biochem. Pharmacol., 37, 4389–4397 (1988). [42] E.L. LeCluyse, E. Alexandre, G.A. Hamilton, C. Viollon-Abadie, D.J. Coon, S. Jolley and L. Richert, Isolation and culture of primary human hepatocytes, Meth. Mol. Biol., 290, 207–229 (2005).


169

[43] L. Richert, E. Alexandre, T. Lloyd, S. Orr, C. Viollon-Abadie, R. Patel, S. Kingston, D. Berry, A. Dennison, B. Heyd, G. Mantion and D. Jaeck, Tissue collection, transport and isolation procedures required to optimize human hepatocyte isolation from waste liver surgical resections. A multilaboratory study, Liver Int., 24, 371–378 (2004). [44] E.L. LeCluyse, P.L. Bullock and A. Parkinson, Strategies for restoration and maintenance of normal hepatic structure and function in long-term cultures of rat hepatocytes, Adv. Drug Deliv. Rev., 22, 133–186 (1996). [45] E.L. LeCluyse, P. Bullock, A. Parkinson and J.H. Hochman, Cultured rat hepatocytes, in R.T. Borchardt, G. Wilson and P. Smith (eds.), Model Systems for Biopharmaceutical Assessment of Drug Absorption and Metabolism, Plenum Press, New York, 1996, 121–160. [46] P. Olinga, M. Merema, I.H. Hof, K.P. de Jong, M.J. Slooff, D.K. Meijer and G.M. Groothuis, Effect of human liver source on the functionality of isolated hepatocytes and liver slices, Drug Metab. Dispos., 26, 5–11 (1998). [47] B. Smedsrod and H. Pertoft, Preparation of pure hepatocytes and reticuloendothelial cells in high yield from a single rat liver by means of Percoll centrifugation and selective adherence, J. Leukoc. Biol., 38, 213–30 (1985). [48] R. Weiskirchen amd A.M. Gressner, Isolation and culture of hepatic stellate cells, Meth. Mol. Med., 117, 99–113 (2005). [49] T. Tordjmann, B. Berthon, B. Lardeux, A. Moreau, E. Jacquemin, L. Combettes, G. Feldmann and M. Claret, An improved digitonin–collagenase perfusion technique for the isolation of periportal and perivenous hepatocytes from a single rat liver: physiological implications for lobular heterogeneity, Hepatology, 26, 1592–1599 (1997). [50] J.N. Lawrence and D.J. Benford, Development of an optimal method for the cryopreservation of hepatocytes and their subsequent monolayer culture, Toxicol in vitro, 5, 39–50 (1991). [51] A. Guillouzo, L. Rialland, A. Fautrel and C. Guyomard, Survival and function of isolated hepatocytes after cryopreservation, Chem. Biol. Interact., 121, 7–16 (1999). [52] R. Gebhardt, J.G. Hengstler, D. Muller, R. Glockner, P. Buenning, B. Laube, E. Schmelzer, M. Ullrich, D. Utesch, N. Hewitt, M. Ringel, B.R. Hilz, A. Bader, A. Langsch, T. Koose, H.I. Burger, J. Maas and F. Oesch, New hepatocyte in vitro systems for drug metabolism: metabolic capacity and recommendations for application in basic research and drug development, standard operation procedures, Drug Metab. Rev., 35, 145–213 (2003). [53] L. Rialland, C. Guyomard, M. Scotte, C. Chesné and A. Guillouzo, Viability and drug metabolism capacity of alginate-entrapped hepatocytes after cryopreservation, Cell Biol. Toxicol., 16, 105–116 (2000). [54] C. Guguen-Guillouzo and A. Guillouzo, Modulation of functional activities in cultured rat hepatocytes, Mol. Cell Biochem., 53–54, 35–56 (1983). [55] G.A. Hamilton, S.L. Jolley, D. Gilbert, D.J. Coon, S. Barros and E.L. LeCluyse, Regulation of cell morphology and cytochrome P450 expression in human hepatocytes by extracellular matrix and cell–cell interactions, Cell Tissue Res., 306, 85–99 (2001). [56] D.M. Bissell and P.S. Guzelian, Phenotypic stability of adult rat hepatocytes in primary monolayer culture, Ann. NY. Acad. Sci., 349, 85–98 (1980). [57] L.M. Reid and D.M. Jefferson, Culturing hepatocytes and other differentiated cells, Hepatology, 4, 548–559 (1984). [58] M.H. Grant, M.A. Melvin, P. Shaw, W.T. Melvin and M.D. Burke, Studies on the maintenance of cytochromes P-450 and b5, monooxygenases and cytochrome reductases in primary cultures of rat hepatocytes, FEBS Lett., 190, 99–103 (1985). [59] J.C. Dunn, M.L. Yarmush, H.G. Koebe and R.G. Tompkins, Hepatocyte function and extracellular matrix geometry: long-term culture in a sandwich configuration, FASEB J., 3, 174–177 (1989).

170

Hepatotoxicity

[60] J.C. Dunn, R.G. Tompkins and M.L. Yarmush, Long-term in vitro function of adult hepatocytes in a collagen sandwich configuration, Biotechnol. Prog., 7, 237–245 (1991). [61] L. Richert, D. Binda, G. Hamilton, C. Viollon-Abadie, E. Alexandre, D. Bigot-Lasserre, R. Bars, P. Coassolo and E. LeCluyse, Evaluation of the effect of culture configuration on morphology, survival time, antioxidant status and metabolic capacities of cultured rat hepatocytes, Toxicol in vitro, 16, 89–99 (2002). [62] E. LeCluyse, A. Madan, G. Hamilton, K. Carroll, R. DeHaan and A. Parkinson, Expression and regulation of cytochrome P450 enzymes in primary cultures of human hepatocytes, J. Biochem. Mol. Toxicol., 14, 177–188 (2000). [63] F. Berthiaume, P.V. Moghe, M. Toner and M.L. Yarmush, Effect of extracellular matrix topology on cell structure, function and physiological responsiveness: hepatocytes cultured in a sandwich configuration, FASEB J., 10, 1471–1484 (1996). [64] S. Vukicevic, H.K. Kleinman, F.P. Luyten, A.B. Roberts, N.S. Roche and A.H. Reddi, Identification of multiple active growth factors in basement membrane Matrigel suggests caution in interpretation of cellular activity related to extracellular matrix components, Exp. Cell Res., 202, 1–8 (1992). [65] J.M. Pascussi, S. Gerbal-Chaloin, J.M. Fabre, P. Maurel and M.J. Vilarem, Dexamethasone enhances constitutive androstane receptor expression in human hepatocytes: consequences on cytochrome P450 gene regulation, Mol. Pharmacol., 58, 1441–1450 (2000). [66] R.Z. Turncliff, P.J. Meier and K.L. Brouwer, Effect of dexamethasone treatment on the expression and function of transport proteins in sandwich-cultured rat hepatocytes, Drug Metab. Dispos., 32, 834–839 (2004). [67] J.S. Sidhu, F. Liu and C.J. Omiecinski, Phenobarbital responsiveness as a uniquely sensitive indicator of hepatocyte differentiation status: requirement of dexamethasone and extracellular matrix in establishing the functional integrity of cultured primary rat hepatocytes, Exp. Cell Res., 292, 252–264 (2004). [68] Z. Turncliff, X. Tian and K.L. Brouwer, Effect of culture conditions on the expression and function of Bsep, Mrp2, and Mdr1a/b in sandwich-cultured rat hepatocytes, Biochem. Pharmacol., 71, 1520–1529 (2006). [69] M. Miyazaki, L.Y. Bai and J. Sato, Selection of medium for serum-free primary culture of adult rat hepatocytes, Acta Med. Okayama, 44, 9–12 (1990). [70] S. Yamada, P.S. Otto, D.L. Kennedy and T.F. Whayne Jr., The effects of dexamethasone on metabolic activity of hepatocytes in primary monolayer culture, in vitro, 16, 559–570 (1980). [71] B.A. Laishes and G.M. Williams, Conditions affecting primary cell cultures of functional adult rat hepatocytes. II. Dexamethasone enhanced longevity and maintenance of morphology, in vitro, 12, 821–832 (1976). [72] J.G. Hengstler, M. Ringel, K. Biefang, S. Hammel, U. Milbert, M. Gerl, M. Klebach, B. Diener, K.L. Platt, T. Bottger, P. Steinberg and F. Oesch, Cultures with cryopreserved hepatocytes: applicability for studies of enzyme induction, Chem. Biol. Interact., 125, 51–73 (2000). [73] M. Miyazaki, W.M. Mars, D. Runge, T.H. Kim, W.C. Bowen and G.K. Michalopoulos, Phenobarbital suppresses growth and accelerates restoration of differentiation markers of primary culture rat hepatocytes in the chemically defined hepatocyte growth medium containing hepatocyte growth factor and epidermal growth factor, Exp. Cell. Res., 241, 445–457 (1998). [74] B.A. Laishes and G.M. Williams, Conditions affecting primary cell cultures of functional adult rat hepatocytes. I. The effect of insulin, In Vitro, 12, 521–532 (1976). [75] A.S. Muller and J. Pallauf, Effect of increasing selenite concentrations, vitamin E supplementation and different fetal calf serum content on GPx1 activity in primary cultured rabbit hepatocytes, J. Trace Elem. Med. Biol., 17, 183–192, (2003).


171

[76] B.M. Johnson, P. Zhang, J.D. Schuetz and K.L. Brouwer, Characterization of transport protein expression in multidrug resistance-associated protein (mrp) 2-deficient rats, Drug Metab. Dispos., 34, 556–562 (2006). [77] G.M. Williams, E. Bermudez and D. Scaramuzzino, Rat hepatocyte primary cell cultures. III. Improved dissociation and attachment techniques and the enhancement of survival by culture medium, In Vitro, 13, 809–817 (1977). [78] T.L. Terry and W.J. Gallin, Effects of fetal calf serum and disruption of cadherin function on the formation of bile canaliculi between hepatocytes, Exp. Cell Res., 214, 642–653 (1994). [79] A.I. Musat, C. Sattler, G.L. Sattler and H.C. Pitot, Re-establishment of cell polarity of hepatocytes in primary culture, Hepatology, 18, 198–205 (1993). [80] E.L. LeCluyse, K.L. Audus and J.H. Hochman, Formation of extensive canalicular networks by rat hepatocytes cultured in collagen-sandwich configuration, Am. J. Physiol., 266 (Cell Physiol., 35), C1764–C1774 (1994). [81] A. Kern, A. Bader, R. Pichlmayr and K.F. Sewing, Drug metabolism in hepatocyte sandwich cultures of rats and humans, Biochem. Pharmacol., 54, 761–772 (1997). [82] E.L. LeCluyse, P. Bullock, A. Madan, K. Carroll and A. Parkinson, Influence of extracellular matrix overlay and medium formulation on the induction of cytochrome P-450 2B enzymes in primary cultures of rat hepatocytes, Drug Metab. Dispos., 27, 909–915 (1999). [83] J.S. Sidhu, F.M. Farin and C.J. Omiecinski, Influence of extracellular matrix overlay on phenobarbital-mediated induction of CYP2B1, 2B2, and 3A1 genes in primary adult rat hepatocyte culture, Arch. Biochem. Biophys., 301, 103-113 (1993). [84] V.E. Kostrubsky, S.C. Strom, J. Hanson, E. Urda, K. Rose, J. Burliegh, P. Zocharski, H. Cai, J.F. Sinclair and J. Sahi, Evaluation of hepatotoxic potential of drugs by inhibition of bile-acid transport in cultured primary human hepatocytes and intact rats, Toxicol. Sci., 76, 220–228 (2003). [85] P.P. Annaert and K.L. Brouwer, Assessment of drug interactions in hepatobiliary transport using rhodamine 123 in sandwich-cultured rat hepatocytes, Drug Metab.Dispos., 33, 388–394 (2005). [86] E. Wisse, Observations on the fine structure and peroxidase cytochemistry of normal rat liver Kupffer cells, J. Ultrastruct. Res., 46, 393–426 (1974). [87] C. El-Bahay, E. Gerber, M. Horbach, Q.H. Tran-Thi, E. Rohrdanz and R. Kahl, Influence of tumor necrosis factor-alpha and silibin on the cytotoxic action of alphaamanitin in rat hepatocyte culture, Toxicol. Appl. Pharmacol., 158, 253–260 (1999). [88] Z. Kmiec, Cooperation of liver cells in health and disease, Adv. Anat. Embryol. Cell Biol., 161, 1–151 (2001). [89] M. Hirose, M. Yamato, O.H. Kwon, M. Harimoto, A. Kushida, T. Shimizu, A. Kikuchi and T. Okano, Temperature-responsive surface for novel co-culture systems of hepatocytes with endothelial cells: 2-D patterned and double layered co-cultures, Yonsei Med. J., 41, 803–813 (2000). [90] N. Uyama, Y. Shimahara, N. Kawada, S. Seki, H. Okuyama, Y. Iimuro and Y.J. Yamaoka, Regulation of cultured rat hepatocyte proliferation by stellate cells, Hepatology, 36, 590–599 (2002). [91] Y.H. Kang, F. Berthiaume, B.D. Nath and M.L. Yarmush, Growth factors and nonparenchymal cell conditioned media induce mitogenic responses in stable long-term adult rat hepatocyte cultures, Exp. Cell Res., 293, 239–247 (2004). [92] M.K. Auth, D. Woitaschek, M. Beste, T. Schreiter, H.S. Kim, E. Oppermann, R.E. Joplin, U. Baumann, P. Hilgard, S. Nadalin, B.H. Markus and R.A. Blaheta, Preservation of the synthetic and metabolic capacity of isolated human hepatocytes by coculture with human biliary epithelial cells, Liver Transpl., 11, 410–419 (2005).

172

Hepatotoxicity

[93] M. Mesnil, J.M. Fraslin, C. Piccoli, H. Yamasaki and C. Guguen-Guillouzo, Cell contact but not junctional communication (dye coupling) with biliary epithelial cells is required for hepatocytes to maintain differentiated functions, Exp. Cell Res., 173, 524–533 (1987). [94] Y. Guigoz, P. Werffeli, D. Favre, M. Juillerat, R. Wellinger and P. Honegger, Aggregate cultures of foetal rat liver cells: development and maintenance of liver gene expression, Biol. Cell, 60, 163–171 (1987). [95] J. Landry, D. Bernier, C. Ouellet, R. Goyette and N. Marceau, Spheroidal aggregate culture of rat liver cells: histotypic reorganization, biomatrix deposition, and maintenance of functional activities, J. Cell Biol., 101, 914–923 (1985). [96] H.F. Lu, K.N. Chua, P.C. Zhang, W.S. Lim, S. Ramakrishna, K.W. Leong and H.Q. Mao, Three-dimensional co-culture of rat hepatocyte spheroids and NIH/3T3 fibroblasts enhances hepatocyte functional maintenance, Acta Biomater., 1, 399–410 (2005). [97] C. Planchamp, T.L. Vu, J.M. Mayer, M. Reist and B. Testa, Hepatocyte hollow-fibre bioreactors: design, set-up, validation and applications, J. Pharm. Pharmacol., 55, 1181–1198, (2003). [98] A. Bader, N. Fruhauf, K. Zech, A. Haverich and J.T. Borlak, Development of a small-scale bioreactor for drug metabolism studies maintaining hepatospecific functions, Xenobiotica, 28, 815–825 (1998). [99] M.J. Powers, K. Domansky, M.R. Kaazempur-Mofrad, A. Kalezi, A. Capitano, A. Upadhyaya, P. Kurzawski, K.E. Wack, D.B. Stolz, R. Kamm and L.G. Griffith, A microfabricated array bioreactor for perfused 3D liver culture, Biotechnol. Bioeng., 78, 257–269 (2002). [100] N.A. Mufti, N.A. Bleckwenn, J.G. Babish and M.L. Shuler, Possible involvement of the Ah receptor in the induction of cytochrome P-450IA1 under conditions of hydrodynamic shear in microcarrier-attached hepatoma cell lines, Biochem. Biophys. Res. Commun., 208, 144–152 (1995). [101] S. Knasmüller, V. Mersch-Sundermann, S. Kevekordes, F. Darroudi, W.W. Huber, C. Hoelzl, J. Bichler and B.J. Majer, Use of human-derived liver cell lines for the detection of environmental and dietary genotoxicants; current state of knowledge, Toxicology, 198, 315–328 (2004). [102] N.J. Hewitt and P. Hewitt, Phase I and II enzyme characterization of two sources of HepG2 cell lines, Xenobiotica, 34, 243–256 (2004). [103] M.G. Sacco , L. Amicone, E.M. Catò, D. Filippini, P. Vezzoni and M. Tripodi, Cell-based assay for the detection of chemically induced cellular stress by immortalized untransformed transgenic hepatocytes, BMC Biotechnol., 4, 5 (2004). [104] R. Parent, M.J. Marion, L. Furio, C. Trepo and M.A. Petit, Origin and characterization of a human bipotent liver progenitor cell line, Gastroenterology, 126, 1147–1156 (2004). [105] P. Gripon, S. Rumin, S. Urban, J. Le Seyec, D. Glaise, I. Cannie, C. Guyomard, J. Lucas, C. Trepo and C. Guguen-Guillouzo, Infection of a human hepatoma cell line by hepatitis B virus, Proc. Natl. Acad. Sci. USA, 99, 15655–15660 (2002). [106] C. Aninat, A. Piton, D. Glaise, T. Le Charpentier, S. Langouet, F. Morel, C. Guguen-Guillouzo and A. Guillouzo, Expression of cytochromes P450, conjugating enzymes and nuclear receptors in human hepatoma HepaRG cells. Drug. Metab. Dispos., 34, 75–83 (2006). [107] T. Hartung and L. Gribaldo, New hepatocytes for toxicology?, Trends Biotechnol., 22, 613–615 (2004). [108] B.N. Ames, F.D. Lee and W.E. Durston, An Improved Bacterial Test System for the Detection and Classification of Mutagens and Carcinogens, Proc. Natl. Acad. Sci. USA, 70, 782–786 (1973). [109] R.D. Callander, J.M. Mackay, P. Clay, C.R. Elcombe and B.M. Elliott, Evaluation of phenobarbital/beta-naphthoflavone as an alternative S9-induction regime to Aroclor 1254 in the rat for use in in vitro genotoxicity assays, Mutagenesis, 10, 517–522 (1995). [110] K.A. Giuliano, J.R. Haskins and D.L. Taylor, Advances in high content screening for drug discovery, Assay Drug Dev. Technol., 1, 565–577 (2003).


173

[111] K.A. Giuliano, W.S. Cheung, D.P. Curran, B.W. Day, A.J. Kassick, J.S. Lazo, S.G. Nelson, Y. Shin and D.L. Taylor, Systems cell biology knowledge created from high content screening, Assay Drug Dev. Technol., 3, 501–514 (2005). [112] D.L. Taylor, E.S. Woo and K.A. Giuliano, Real-time molecular and cellular analysis: the new frontier of drug discovery, Curr. Opin. Biotechnol., 12, 75–81 (2001). [113] V.C. Abraham, D.L. Taylor and J.R. Haskins, High content screening applied to large-scale cell biology, Trends Biotechnol., 22, 15–22 (2004). [114] P. J. O’Brien, W. Irwin, D. Diaz, E. Howard-Cofield, C. M. Krejsa, M. R. Slaughter, B. Gao, N. Kaludercic, A. Angeline, P. Bernardi, P. Brain and C. Hougham, High concordance of druginduced human hepatotoxicity with in vitro cytotoxicity measured in a novel cell-based model using high content screening, Arch. Toxicol., 80, 580–604 (2006). [115] J.R. Haskins, P. Rowse, R. Rahbari and F.A. de la Iglesia, Thiazolidinedione toxicity to isolated hepatocytes revealed by coherent multiprobe fluorescence microscopy and correlated with multiparameter flow cytometry of peripheral leukocytes. Arch . Toxicol., 75, 425–438 (2001). [116] M.V. Berridge, A.S. Tan, K.D. McCoy and R. Wang, The Biochemical and Cellular Basis of Cell Proliferation Assays that Use Tetrazolium Salts, Biochemica, Features 4, 14–19 (1996). [117] X. Liu, J.P. Chism, E.L. LeCluyse, K.R. Brouwer and K.L. Brouwer, Correlation of biliary excretion in sandwich-cultured rat hepatocytes and in vivo in rats. Drug Metab. Dispos., 27, 637–644 (1999). [118] C.R. Padgham and A.J. Paine, Altered expression of cytochrome P-450 mRNAs, and potentially of other transcripts encoding key hepatic functions, are triggered during the isolation of rat hepatocytes, Biochem. J., 289, 621–624 (1993). [119] C.P. Hodgkinson, M.C. Wright and A.J. Paine, Fibronectin-mediated hepatocyte shape change reprograms cytochrome P450 2C11 gene expression via an integrin-signaled induction of ribonuclease activity. Mol. Pharmacol., 58, 976–981 (2000). [120] L. Richert, M.J. Liguori, C. Abadie, B. Heyd, G. Mantion, N. Halkic and J.F. Waring, Gene expression in human hepatocytes in suspension after isolation is similar to the liver of origin, is not affected by hepatocyte cold storage and cryopreservation, but is strongly changed after hepatocyte plating, Drug Metab. Dispos., 34, 870–879 (2006). [121] F. Boess, M. Kamber, S. Romer, R. Gasser, D. Muller, S. Albertini and L. Suter, Gene expression in two hepatic cell lines, cultured primary hepatocytes and liver slices compared to the in vivo liver gene expression in rats: possible implications for toxicogenomics use of in vitro systems, Toxicol. Sci., 73, 386–402 (2003). [122] G. Tuschl and S.O. Mueller, Effects of cell culture conditions on primary rat hepatocytes–cell morphology and differential gene expression, Toxicology, 218, 205–215 (2006). [123] T. Su and D.J. Waxman, Impact of dimethyl sulfoxide on expression of nuclear receptors and drug-inducible cytochromes P450 in primary rat hepatocytes, Arch. Biochem. Biophys., 424, 226–234 (2004). [124] J. Easterbrook, C. Lu, Y. Sakai and A.P. Li, Effects of organic solvents on the activities of cytochrome P450 isoforms, UDP-dependent glucuronyl transferase and phenol sulfotransferase in human hepatocytes, Drug Metab. Dispos., 29, 141–144 (2001). [125] N. Chauret, A. Gauthier and D.A. Nicoll-Griffith, Effect of common organic solvents on in vitro cytochrome P450-mediated metabolic activities in human liver microsomes, Drug Metab. Dispos., 26, 1–4 (1998). [126] D. Hickman, J.P. Wang, Y. Wang and J.D. Unadkat, Evaluation of the selectivity of in vitro probes and suitability of organic solvents for the measurement of human cytochrome P450 monooxygenase activities, Drug Metab. Dispos., 26, 207–215 (1998). [127] O. Fardel, V. Lecureur and A. Guillouzo, Regulation by dexamethasone of P-glycoprotein expression in cultured rat hepatocytes, FEBS Lett., 327, 189–193 (1993).

174

Hepatotoxicity

[128] E. Chieli, E. Santoni-Rugiu, F. Cervelli, A. Sabbatini, M. Petrini, N. Romiti, A. Paolicchi and R. Tongiani, Differential modulation of P-glycoprotein expression by dexamethasone and 3-methylcholanthrene in rat hepatocyte primary cultures, Carcinogenesis, 15, 335–341 (1994). [129] C.H. Lee, Differential regulation of P-glycoprotein genes in primary rat hepatocytes by collagen sandwich and drugs, J. Cell. Biochem., 86, 12–20 (2002). [130] J.S. Sidhu and C.J. Omiecinski, Modulation of xenobiotic-inducible cytochrome P450 gene expression by dexamethasone in primary rat hepatocytes, Pharmacogenetics, 5, 24–36 (1995). [131] G. Valet, Cytomics as a new potential for drug discovery, Drug Discov. Today, 11, 785–791 (2006). [132] W.B. Mattes, S.D. Pettit, S.A. Sansone, P.R. Bushel and M.D. Waters, Database development in toxicogenomics: issues and efforts, Environ. Health Perspect., 112, 495–505 (2004). [133] C.J. Stoeckert Jr, H.C. Causton and C. A. Ball, Microarray databases: standards and ontologies, Nat. Genet., 32(Suppl), 469–473 (2002). [134] J.F. Waring, R. Ciurlionis, R.A. Jolly, M. Heindel and R.G. Ulrich, Microarray analysis of hepatotoxins in vitro reveals a correlation between gene expression profiles and mechanisms of toxicity, Toxicol. Lett., 120, 359–368 (2001). [135] J.F. Waring, R. Gum, D. Morfitt, R.A. Jolly, R. Ciurlionis, M. Heindel, L. Gallenberg, B. Buratto and R.G. Ulrich, Identifying toxic mechanisms using DNA microarrays: evidence that an experimental inhibitor of cell adhesion molecule expression signals through the aryl hydrocarbon nuclear receptor, Toxicology, 181–182, 537–550 (2002). [136] M.E. Burczynski, M. McMillian, J. Ciervo, L. Li, J.B. Parker, R.T. Dunn II, S. Hicken, S. Farr and M.D. Johnson, Toxicogenomics-based discrimination of toxic mechanism in HepG2 human hepatoma cells, Toxicol. Sci., 58, 399–415 (2000). [137] C.J. Mattingly, M.C. Rosenstein, G.T. Colby, J.N. Forrest Jr and J.L. Boyer, The Comparative Toxicogenomics Database (CTD): a resource for comparative toxicological studies, J. Exp. Zoolog. A Comp. Exp. Biol., 305, 689–692 (2006). [138] H. Ellinger-Ziegelbauer, B. Stuart, B. Wahle, W. Bomann and H.J. Ahr, Characteristic expression profiles induced by genotoxic carcinogens in rat liver, Toxicol. Science, 77, 19–34 (2004). [139] H. Ellinger-Ziegelbauer, B. Stuart, B. Wahle, W. Bomann and H.J. Ahr, Comparison of the expression profiles induced by genotoxic and nongenotoxic carcinogens in rat liver, Mutat. Res., 575, 61–84 (2005). [140] K. Fella, M. Gluckmann, J. Hellmann, M. Karas, P.J. Kramer and M. Kroger, Use of twodimensional gel electrophoresis in predictive toxicology: identification of potential early protein biomarkers in chemically induced hepatocarcinogenesis, Proteomics, 5, 1914–1927 (2005). [141] A.N. Heinloth, R.D. Irwin, G.A. Boorman, P. Nettesheim, R.D. Fannin, S.O. Sieber, M.L. Snell, C.J. Tucker, L. Li, G.S. Travlos, G. Vansant, P.E. Blackshear, R.W. Tennant, M.L. Cunningham and R.S. Paules, Gene expression profiling of rat livers reveals indicators of potential adverse effects, Toxicol. Sci., 80, 193–202 (2004). [142] M. Kroeger, How omics technologies can contribute to the ‘3R’ principles by introducing new strategies in animal testing, Trends Biotechnol., 24, 343–346 (2006). [143] F.A. Witzmann, J.W. Clack, K. Geiss, S. Hussain, M.J. Juhl, C.M. Rice and C. Wang, Proteomic evaluation of cell preparation methods in primary hepatocyte cell culture, Electrophoresis, 23, 2223–2232 (2002). [144] L. Hood and R.M. Perlmutter, The impact of systems approaches on biological problems in drug discovery, Nat. Biotechnol., 22, 1215–1217 (2004). [145] H.P. Fischer, Towards quantitative biology: integration of biological information to elucidate disease pathways and to guide drug discovery, Biotechnol. Annu. Rev., 11, 1–68 (2005). [146] C.L. Barrett, T.Y. Kim, H.U. Kim, B.O. Palsson and S.Y. Lee, Systems biology as a foundation for genome-scale synthetic biology, Curr. Opin. Biotechnol., 17, 488–492 (2006).

Section 3 Biomarkers of Hepatotoxicity

8 Biomarkers of Mycotoxin Exposure in Liver Toxicity Angela J. Harris

8.1

Introducion

Humans are exposed daily to a multitude of environmental and dietary agents capable of causing adverse health effects if absorbed at sufficient doses. Technical advances in the identification and quantification of chemicals and chemical metabolites in biological tissues, as well as recognition that individual chemicals or chemical classes affect biological responses in specific ways, has led to an upsurge in research dedicated to the discovery of biomarkers of chemical exposure. Obviously, the first step in this process is identification of putative biomarkers. The relatively new fields of genomics, proteomics and molecular epidemiology have provided particularly powerful new tools for this step. Both genomics and proteomics technologies allow investigators to determine the effects of chemical or biological agents on expression of hundreds to thousands of genes and proteins simultaneously at any given point in time. By using these tools, not only can an investigator detect altered expression of a single gene or protein of interest as has traditionally been measured using Northern and Western blot techniques, respectively, but can detect changes in global patterns of expression. Thereby, patterns of expression may also be discerned which could serve as a biomarker pattern indicative of any number of biological effects, including exposure, toxicity, genotoxicity, drug efficacy or status of tumor progression. Identification of putative biomarkers is followed by a validation process that is used to verify its predictive value. Biomarkers have been defined as A physical sign or laboratory measurement that occurs in association with pathological process and that has putative diagnostic and/or prognostic utility (Lesko and Atkinson, 2001). There are three basic


178

Hepatotoxicity

types of biomarkers: biomarkers of dose, biomarkers of a biological effect and biomarkers of individual susceptibility. At the present time, there are very few validated biomarkers of exposure to environmental chemical agents. Validation requires that the presence of the biomarker in humans be linked to exposure to the agent in question. This has been useful in occupational monitoring programs for exposure to certain chemicals. Some chemicals or their metabolites can be monitored in the blood and/or urine of workers that are potentially exposed in the workplace. These biomarkers, biological exposure indices (BEIs), are used primarily to quantify or estimate the absorbed dose of a chemical in an occupational setting. However, in many cases, the presence of a biomarker at any concentration in a biological tissue cannot be correlated with the probability that any adverse health effect will occur. This is due to many factors which are related to the half-life of the biomarker in the biological tissues, lack of sufficient information to correlate qualitative or quantitative measure of the biomarker with adverse health effects and a lack of understanding concerning the specificity or mechanism of toxicity. Validated biomarkers are those which have been specifically associated with exposure to the agent, are linked quantitatively to a biological effect and/or can be used to predict the susceptibility of the exposed individual to that agent.

8.2

Mycotoxin and Liver Toxicity

Mycotoxins are a diverse group of naturally occurring toxins produced by various fungal species. Humans are most commonly exposed to mycotoxins through ingestion of foods that have been contaminated with mycotoxin-producing strains and many of these toxins have been shown to be hepatotoxic, hepatocarcinogenic or both. Aflatoxins are produced by Aspergillus flavus and Aspergillus parasiticus and are a worldwide problem in areas where hot, humid conditions promote fungal growth on crops. Aflatoxins cause liver toxicity and are genotoxic. Aflatoxin B1 (AFB1 ) is classified by the International Agency for Research on Cancer (IARC) as a human liver carcinogen. Biomarkers for aflatoxins, particularly AFB1 have been extensively studied and are the most well validated of any mycotoxin biomarker. Fumonisin B1 (FB1 ), a mycotoxin produced by Fusarium verticillioides has also been reported to have toxic or carcinogenic effects on the liver. AFB1 and FB1 are structurally diverse and have different mechanisms of action. Therefore, identification of biomarkers will necessarily be mycotoxin- or mycotoxin class-specific. Biomarker identification for mycotoxin exposure is a growing area of research, in part due to increasing interest in reported health effects in occupants of buildings contaminated with molds capable of mycotoxin production. A summary of crops and food occasionally contaminated with AFB1 or FB1 is found in Table 8.1. Except for AFB1 , there are no well validated biomarkers of exposure to specific mycotoxins. Biomarkers of AFB1 exposure in animals and humans are an example of well validated biomarkers of environmental exposure and should be considered a model for further development of validated mycotoxin biomarkers. 8.2.1

Aflatoxin B1

AFB1 is a potent hepatotoxin and hepatocarcinogen in many animal species and in man. Ingestion of foods heavily contaminated by AFB1 causes aflatoxicosis in both animals

Biomarkers of Mycotoxin Exposure in Liver Toxicity

179

Table 8.1 Summary of foods potentially affected by mycotoxins Mycotoxin

Primary affected crops

Mycotoxin producer

Aflatoxin

Corn, peanuts, sorghum, milk, cottonseed Mill, wheat, nuts Sorghum, rice, beer, mung beans

Aspergillus flavus Aspergillus parasiticus Fusarium verticillioides

Fumonisin

and humans. Severe aflatoxicosis is characterized by hemorrhagic necrosis of the liver, edema, lethargy and death, particularly in children (Cullen and Newberne, 1993). AFB1 is classified as a known human carcinogen and causes primary liver cancer in several animals, including rainbow trout (Sinnhuber et al., 1968), rats (Busby and Wogan, 1984) and tree shrews (Reddy et al., 1976). There is also substantial epidemiological evidence of increased risk for primary liver cancer in individuals living in areas where high concentrations of AFB1 are present in food, especially in areas where hepatitis B infection is also common (Chen et al., 1996; Groopman et al., 1996). Several different aflatoxins are produced by Aspergillus; however, AFB1 is by far the most potent. AFB1 is metabolized by cytochromes P450, primarily CYP1A2 and CYP3A4, to the reactive metabolite AFB1 -8,9 epoxide which readily binds to nucleophilic sites on DNA (Ueng et al., 1995; Gallagher et al., 1996). AFB1 -8, 9 epoxide preferentially binds at the N-7 position of guanine due to the physical characteristics of the metabolite and DNA (Gopalakrishnan et al., 1990). Thus, the predominant DNA lesion formed is the 8,9dihydro-8- (N7-guanyl)-9-hydroxy-AFB1 (AFB1 -N7-Gua) adduct (Essigmann et al., 1977; Lin et al., 1977). It is this lesion that is responsible for the genotoxicity and carcinogenicity of AFB1 . Bulky AFB1 -N7-Gua adducts that are not repaired prior to DNA replication and subsequent cell division may result in mutations at the site of the lesion. The most commonly reported mutations are G → T transversions (Foster et al., 1983; Cariello et al., 1994) although G → C transversions, as well as G → A transitions have also been noted (Levy et al., 1992). The accumulation of mutations in genes playing a critical role in tumor progression or suppression increases the risk of cancer. Since metabolism of AFB1 occurs predominantly in the liver, the highest level of the reactive metabolite, AFB1 -8, 9 epoxide, will also be present in the liver; hence, the liver as the primary target organ for AFB1 induced toxicity and cancer.

8.3

Biomarkers of AFB1 Exposure

The AFB1 -N7-Gua adduct is rapidly removed from DNA by nucleotide excision repair pathways (Bedard and Massey, 2006) and excreted in the urine from which it can be isolated using immunoaffinity chromatography and then quantified by high performance liquid chromatography (HPLC) (Groopman, 1985, 1992a). AFB1 dose after ingestion is well correlated with the concentration of AFB1 -N7-Gua adduct present in the urine of exposed animals (Croy and Wogan, 1981; Groopman et al., 1992a) and humans (Groopman et al., 1992b). The level of AFB1 -N7-Gua adduct in urine is also well correlated with the formation of DNA adducts in the liver (Groopman et al., 1992a) and so is a biomarker

180

Hepatotoxicity

both of the severity of biological effects and of dose. Since the bulky AFB1 -N7-Gua DNA adduct is rapidly repaired, the urinary AFB1 -N7-Gua biomarker is considered an indicator of recent exposure. AFB1 also forms protein adducts. The AFB1 -8,9 epoxide can be further modified to AFB1 -8,9 dihydrodiol which may be slowly converted to a dialdehyde phenolate ion capable of forming Schiff bases with lysine to form protein adducts (Wild et al., 1986; Sabbioni et al., 1987). The primary serum protein adduct formed is aflatoxin–albumin (serum AF– alb). Unlike DNA adducts, serum protein adducts are not repaired. Therefore, serum AF-alb is a better indicator than urinary AFB1 -7-Gua of cumulative AFB1 exposure. A positive correlation between serum AF-alb and AFB1 dose in animals and humans exposed to AFB1 has been established (Wild et al., 1990a,b, 1992). Epidemiological studies have also evaluated the relationship between AFB1 adduct biomarkers and the presence of primary liver cancer in geographic areas with high dietary intake of AFB1 -contaminated foods. A nested case control study conducted in Shanghai showed that the relative risk (RR) of primary liver cancer (PLC) in men who were positive for AFB1 -N7-Gua was 9.1 (95 % CI = 2.9–29.2) (Qian et al., 1994). Interestingly, the RR was 7.3 (95 % CI = 2.2–24.4) for individuals that were positive for hepatitis B antigen (HBsAg) but negative for aflatoxin urinary biomarkers. However, the RR for PLC was 59.4 (95 % CI = 16.6–212) for individuals who were positive for both biomarkers, indicating a strong synergy between HBV status and dietary AFB1 exposure. Another nested control study in Taiwan showed that the odds ratio (OR) for development of PLC in individuals that were positive for urine aflatoxin biomarkers was 1.7 (95 % CI = 0.3–10.8), compared to an OR of 111.9 (95 % CI = 13.8–905) in individuals positive for both HBsAg and AFB1 -N7-Gua (Wang et al., 1996). These data indicate the effectiveness of agent-specific biomarkers as monitors of exposure in epidemiological studies. Urine and serum biomarkers of AFB1 exposure have also been used to evaluate putative biomarkers of susceptibility in humans. For instance, serum AF-alb and urine AFB1 -N7Gua levels in humans exposed to AFB1 have been compared in individuals with polymorphisms in various Phase I and Phase II metabolic genes (Wild et al., 2000; Sun et al., 2002) and DNA repair genes (Lunn et al., 1999) in order to determine the effects of certain polymorphisms on metabolism of AFB1 and subsequent formation of DNA and protein adducts.

8.4

Genomic Biomarker Identification

AFB1 induced changes in gene expression after short-term exposure in vitro have been reported in cultured primary rat hepatocytes (Harris et al., 1998), Saccharomyces cerevisiae engineered to express human CYP1A2 (Guo et al., 2006), cultured primary human hepatocytes and HepG2 cells (Harris et al., 2004). Investigators have also evaluated altered gene expression in AFB1 -induced tumors in tree shrews (Duan et al., 2005; Li et al., 2004) and rainbow trout (Tilton et al., 2005). At the present time, there are insufficient data to identify any particular gene or pattern of gene expression that is specifically altered in AFB1 -induced toxicity or tumor development. This may be partially due to the selection of the gene array, different dosing regimes, comparison of in vitro versus in vivo effects, different species, endpoint selection and other variables in experimental design.


8.4.1

181

Fumonisin

Fumonisin (FB1 ), produced by Fusarium verticillioides (= monoliforme) (Sheldon, 1904) is hepatotoxic in several species (Gelderblom et al., 1988, 1991, 2001; Marasas et al., 1988a) and a hepatocarcinogen in rats and mice (Wilson et al., 1985; Voss et al., 1995; Howard et al., 2001). Fusarium strains affect corn crops worldwide (Table 8.1) and have been linked to increased risk for esophageal cancer in areas with high dietary consumption of FB1 -contaminated foods (Marasas et al., 1979, 1988b). FB1 has been shown to act as a cancer promoter in rats treated with diethylnitrosamine (DEN), a known initiator (Gelderblom et al., 1988). Gamma-glutamyl-tranpeptidase-positive (GGT+ ) foci appeared in the liver of rats after four weeks on a diet containing 0.1 % FB1 . The International Agency for Research on Cancer (IARC) has classified FB1 as possibly carcinogenic to human, class 2B (IARC, 1993). Unlike AFB1 , FB1 is not genotoxic and the mechanism of carcinogenesis is likely epigenetic. At least one outbreak of acute human and animal illness has been attributed to FB1 contaminated sorghum and/or bread (Bhat et al., 1997). In 1995, individuals in 27 villages in India became ill after unseasonal rains caused unusually high mold growth on the sorghum and maize crops. Many villagers who ate unleaved bread prepared from moldy maize and/or sorghum developed abdominal pain, diarrhea and borborygmi, typically within an hour after ingestion. The symptoms resolved once exposure ceased. There were no human fatalities although some domestic animals fed the moldy foods died. There have also been reports of equine leukoencephalomalacia and mild hepatosis in horses that ate Fusariumcontaminated feed (Marasas et al., 1988a). Fumonisins have been shown to disrupt ceramide synthase (Merrill et al., 2001; Riley et al., 2001). As a result, sphingolipid metabolism is disrupted, increasing the levels of ceramide precursors sphinganine (Sa) and sphingosine (So) in tissues, including the liver (Norred et al., 1992) and serum of treated animals (Wang et al., 1992). Disruption of sphingolipid metabolism has significant effects on normal cellular function which is dependent on the regulated synthesis and turnover of complex sphingolipids to maintain a number of signaling pathways. It is hypothesized that disruption of sphingolipid metabolism plays a role in hepatotoxicity; however, the specific pathways by which these events cause toxic and carcinogenic effects in the liver of rodents have not been determined.

8.5

Biomarkers of FB1 Exposure

Unlike AFB1 , there are no well-validated biomarkers for FB1 exposure; however, serum Sa and So levels have been proposed as candidates of exposure and biological effect. The most sensitive candidate biomarker of FB1 exposure reported thus far is serum Sa relative to So, the Sa/So ratio (Riley et al., 1994; Tran et al., 2006). There are a number of studies that have evaluated the use of serum or urine Sa or Sa/So ratios as biomarkers of exposure. A few studies have also attempted to correlate these potential biomarkers with known serum biomarkers of liver toxicity to evaluate their usefulness as biomarkers of biological effects. Some of these data are summarized below.

182

8.5.1

Hepatotoxicity

Serum Biomarkers in Animals

Serum Sa and So concentrations increased in ponies fed diets containing 44 mg/kg FB1 , even before increased levels of serum biomarkers of liver damage, such as aspartate aminotransaminase or gamma glutamyl transaminase were apparent (Wang et al., 1992). Serum Sa concentration peaked seven days after exposure at serum levels 22-fold higher than before dosing. The serum Sa/So ratio was also increased 6.4-fold. Increased serum biomarkers of liver toxicity typically occurred at about day 10. Tran et al. (2006) reported that an increase in serum Sa could be detected in ducks fed feed containing FB1 as low as 2 mg/kg. A dose-related increase in serum Sa/So was observed when the cumulative FB1 dose in feed was 100–200 mg; however, there was no apparent increase in serum Sa/So above that dose. Tran et al. (2006) also found that effects on serum Sa/So were more apparent in ducks exposed over a short period of time (7–21 days) versus a long period of exposure (42–77 days). A statistically significant dose-related increase in serum Sa/So ratio compared to controls was also reported in vervet monkeys fed a diet with culture material of Fusarium verticilloidides containing FB1 (control mean Sa/So = 0.43; low mean Sa/So = 1.72; high mean Sa/So = 2.57) (Shephard et al., 1996; Gelderblom et al., 2001). Blood serum Sa/So levels were assessed over a 51-day period in vervet monkeys fed a single dose of 1 mg/kg or 10 mg/kg of purified FB1 . Serum Sa/So ratios significantly elevated in the high-treatment group over controls ( p < 0.05) were observed three days after exposure, continued to increase for 27 days and then declined to control values by day 51 (van der Westhuizen et al., 2001a). As reported above for ponies and ducks, the initial rise in serum Sa/So ratios corresponded in general with an increase in serum levels of biomarkers of liver toxicity (GGT and ALT). Vervet monkeys dosed three times a week with 1 mg/kg of FB1 for 51 days had a maximum mean plasma Sa/So ratio of 4.3 after 30 days of treatment (van der Westhuizen et al., 2001b). This level was approximately 3-fold higher than that of non-treated animals. The plasma Sa/So ratio declined by about one third over the course of the remaining 21 days of treatment. These data indicate that the serum Sa/So ratios may be indicative of recent acute exposure to FB1 ; however, a dose–response relationship has not been clearly defined nor has it been clearly correlated with the severity of biological effects.

8.5.2

Urine Biomarkers in Animals

Interestingly, only modest or no differences have been reported by many authors in urine Sa/So ratios observed in animals treated with FB1 (Wang et al., 1992; Shephard et al., 1996; Castegnaro et al., 1998; Garren et al., 2001; van der Westhuizen et al., 2001a,b). However, some studies have reported that dietary FB1 can elevate Sa/So ratios in urine. For instance, rats fed 1–5 mg/kg FB1 on a daily basis had urine Sa/So ratios that ranged from 1.2 to 10 compared to a range of 0.1 to 0.7 in untreated rats (Castegnaro et al., 1996). Some of these seeming inconsistencies may be temporally related to length of treatment and the time of testing relative to treatment. For instance, although an initial rapid increase in the urine Sa/So ratio was seen in vervet monkeys treated with FB1 , it was not sustained and dropped to pre-treatment values very quickly (van der Westhuizen et al., 2001b). In addition, significant interindividual differences in urine Sa/So ratios may obscure subtle effects of FB1 on Sa/So ratio in urine (Garren et al., 2001; Gelderblom et al., 2001).


8.5.3

183

Serum Biomarkers in Humans

Methods have been developed and tested for analysis of serum Sa and So in humans (Shephard and van der Westhuizen, 1998; Castegnaro et al., 1998) although only a limited number of studies have evaluated serum Sa/So ratios in humans. No significant differences were seen in serum Sa/So ratios of African males and females, or in individuals living in different geographic areas in Africa (van der Westhuizen et al., 1999). Recently, serum Sa/So ratios have been evaluated as a biomarker for increased risk of neural tube defects in Mexican–American women exposed to FB1 in corn tortillas (Missmer et al., 2006). A dose-related increase was seen in the OR for neural tube defects (NTDs) with increased serum Sa/So ratio and FB1 intake, except at the highest levels. The OR for NTDs was 4.4 (95 % CI 1.2–15.5) in individuals with a serum Sa/So ratio of 0.31–0.35. Control values were ≤ 0.10. The OR for NTDs was 2.3 (95 % CI 1.1–5.1) for pregnant women with a daily dietary FB1 intake of 150.1–650 ng/kg/day.

8.5.4

Urine Biomarkers in Humans

Methods have been determined to analyze the concentration of Sa and So in the urine of humans potentially exposed to dietary FB1 (Castegnaro et al., 1996; Solfrizzo et al., 1997; Qiu and Liu, 2001). These studies report that Sa and So concentrations in the urine of females are higher than in males. It is not known what causes this difference but it is hypothesized to be due to higher number of sloughed cells in the urine of females. Qiu and Liu (2001) evaluated the Sa/So ratio in the urine of human volunteers who consumed a corn diet containing 0.08 to 41.1 mg/kg for one month. These data indicate that ingestion of FB1 affects sphingolipid metabolism in humans since there was a trend toward higher Sa/So ratios in individuals eating foods with the highest contamination by FB1. Males appeared to be more sensitive to this effect than females. A second study compared urinary Sa/So ratios in humans living in areas known to ingest more FB1 in their diets (North Argentina and Brazil) to humans living in areas where maize consumption was very low (Central Argentina) (Solfrizzo et al., 2004). There was a statistically significant difference ( p < 0.001) in the Sa/So ratios from the FB1 -containing diets (mean Sa/So = 1.27) compared to controls (mean Sa/So = 0.36); however, there was also a statistically significant difference in the Sa/So urinary ratios between the two high FB1 areas (South Brazil mean = 1.57 vs. North Argentina mean = 0.69, p < 0.05) even though the mean FB1 intake was similar between the two groups (0.35 mg/kg). These data suggest that in humans the use of the Sa/So ratios may not be useful as a biomarker of exposure except in geographic areas where foods are highly contaminated with FB1 . Similar conclusions were reached in a study of male and female volunteers in rural areas in Africa (van der Westhuizen et al., 1999). Clearly, more research needs to be done in this area to validate the usefulness of this biomarker in the human population for exposure and/or biological effect after exposure to dietary levels likely to occur in humans eating FB1 -contaminated foods. Although the Sa and Sa/So ratios may ultimately be useful only in the case of relatively high, acute exposures, there are some data to suggest that actual measurements of fumonisins in the hair of experimental animals (Sewram et al., 2001) and humans (Sewram et al., 2003) may be useful for monitoring chronic exposure to dietary levels of fumonisins.

184

Hepatotoxicity

Table 8.2

Gene expression changes in FB1 -exposed mice

Gene description

Accession number

Glutathione S-transferase, mu3 (Gstm3) Ras homolog gene family, member u (Arhu) Cyclin-dependent kinase inhibitor 1A (Cdkn1a) Cyclin D1 (Ccnd1) Ras homolog gene family, member Q Glutathione S-transferase, pi 2 (Gstp2) Flavin-containing monooxygenase 5 (Fmo5) Cytochrome P450, family 3 (Cyp3a16) Sulfotransferase-related protein SULT-X2

J03953 AV246963 AW048937 AI849928 AW060401 X53451 U90535 D26137 Sult-x2

8.6

Fold change 15.42 8.27 5.8 5.22 4.72 4.68 4.24 3.16 −73.3

Genomic Biomarker Identification

Limited studies are available that have evaluated the effects of FB1 on gene expression in the liver. Voss et al. (2006) reported that wild type mice fed culture material from F. verticillioides (CM) or purified FB1 for 7 days had increased expression of 180 genes and decreased expression of 193 genes. A group of 48 genes were affected by FB1 treatment only, and not by CM. The expression of 138 genes was altered by both FB1 and CM. Genes with the highest changes in gene expression after FB1 treatment are summarized in Table 8.2. These genes include those associated with cellular proliferation (cyclins), signal transduction (ras) and Phase I and Phase II metabolism. At the present time, there are insufficient data to evaluate their usefulness as potential biomarkers of exposure.

8.7

Summary

Human exposure to mycotoxins is a worldwide problem. Acute toxicity due to overexposure to mycotoxins in the diet can cause symptoms that often include nausea, vomiting, diarrhea and abdominal pain. Hepatotoxicity may also occur. Chronic exposure to some mycotoxins, such as AFB1 and FB1 , is also associated with primary liver cancer. Identification of biomarkers of mycotoxin exposure allows investigators to determine the extent and effects of exposure. They are also invaluable in epidemiological studies designed to determine the effects of exposure in humans. The use of genomics has also been applied in a limited fashion to identify individual genes or patterns of gene expression that may be used as biomarkers of exposure to specific mycotoxins or biological effects (liver toxicity). Although these studies may be eventually used to identify genes that are markers of exposure and/or toxicity to mycotoxins, they will have limited utility in evaluating human exposure or biological effect in the exposed population. Useful biomarkers are those which can be monitored in easily accessible fluids, such as urine, blood or saliva. Therefore, applying proteomic and metabonomics technologies to identify putative biomarkers of mycotoxin exposure is likely to be the most effective strategy for future work in mycotoxin research.


185

References Bedard LL and Massey TE. (2006). Aflatoxin B1 induced DNA damage and its repair. Cancer Lett 241: 174–183. Bhat RV, Shetty PH, Amruth RP and Sudershan RV. (1997). A foodborne disease outbreak due to the consumption of moldy sorghum and maize containing fumonisin mycotoxins. Clin Toxicol 35: 249–255. Busby WF and Wogan GN. (1984). Aflatoxins, in CE Searle (ed.), Chemical Carcinogens, American Chemical Society, Washington, DC, USA, 945–1136. Cariello NF, Cui L and Skopek TR. (1994). In vitro mutational spectrum of aflatoxin B1 in the human hypoxanthine guanine phosphoribosyltransferase gene. Cancer Res 54: 4436–4441. Castegnaro M, Garren L, Gaucher I and Wild CP. (1996). Development of a new method for the analysis of sphinganine and sphingosine in urine and tissues. Nat Toxins 4: 284–290. Castegnaro M, Garren L, Galendo D, Gelderblom WC, Chelule P, Dutton MF and Wild CP. (1998). Analytical method for the determination for sphinganine and sphingosine in serum as a potential biomarker for fumonisin exposure. J Chromatogr B Biomed Sci Appl 720: 15–24. Chen CJ, Wang LY, Lu SN, Wu MH, Lou SL, Zhang YJ, Wang LW and Santella RM. (1996). Elevated aflatoxin exposure and increased risk of hepatocellular carcinoma. Hepatology 24: 38–42. Croy RG and Wogan GN. (1981). Temporal patterns of covalent DNA adducts in rat liver after single and multiple doses of aflatoxin B1 . Cancer Res 41: 197–203. Cullen JM and Newberne PM. (1993). Acute hepatoxicity of aflatoxins, in DL Eaton and JD Groopman (eds), The Toxicology of Aflatoxins: Human Health, Veterinary and Agricultural-Significant, Academic Press, London, UK, 1–23. Duan X-X, Ou J-S, Li Y, Su J-J, Ou C, Yang C, Yue H-F and Ban K-C. (2005). Dynamic expression of apoptosis-related genes during development of laboratory hepatocellular carcinoma and its relation to apoptosis. World J Gastroenterol 11: 4740–4744. Essigmann JM, Croy RG, Nadzan AM, Busby WF Jr, Reinhold VN, Buchi G and Wogan GN. (1977). Structural identification of the major DNA adduct formed by aflatoxin B1 in vitro. Proc Natl Acad Sci USA 74: 1870–1874. Foster PL, Eisenstadt E and Miller JH. (1983). Base substitution mutations induced by metabolically activated aflatoxin B1. Proc Natl Acad Sci USA 80: 2695–2698. Gallagher EP, Kunze KL, Stapleton PL and Eaton DL. (1996). The kinetics of aflatoxin B1 oxidation by human cDNA-expressed and human liver microsomal cytochromes P450 1A2 and 3A4. Toxicol Appl Pharmacol 141: 595–606. Garren L, Galendo D, Wild CP and Castegnaro M. (2001). The induction and persistence of altered sphingolipid biosynthesis in rats treated with fumonisin B1 . Food Addit Contam 18: 850–856. Gelderblom WC, Jaskiewicz K, Marasas WF, Thiel PG, Horak RM, Vleggaar R and Kriek NP. (1988). Fumonisins – novel mycotoxins with cancer-promoting activity produced by Fusarium moniliforme. Appl Environ Microbiol 54: 1806–1811. Gelderblom WC, Kriek NP, Marasas WF and Thiel PG. (1991). Toxicity and carcinogenicity of Fusarium moniliforme metabolite, fumonisin B1 in rats. Carcinogenesis 12: 1247–1251. Gelderblom WC, Seier JV, Snijman PW, Van Schalkwyk DJ, Shephard GS and Marasas WF. (2001). Toxicity of culture material of Fusarium verticillioides strain MRC 826 to nonhuman primates. Environ Health Perspect 109: 267–276. Gopalakrishnan S, Harris TM and Stone MP. (1990). Intercalation of aflatoxin B1 in two oligodeoxynucleotide adducts: comparative 1H NMR analysis of d(ATCAFBGAT).d(ATCGAT) and d(ATAFBGCAT)2. Biochemistry 29: 10438–10448. Groopman JD, Donahue RR, Zhu JQ, Chen JS and Wogan GN. (1985). Aflatoxin metabolism in humans: detection of metabolites and nucleic acid adducts in urine by affinity chromatography. Proc Natl Acad Sci USA 82: 6492–6496.

186

Hepatotoxicity

Groopman JD, Hasler JA, Trudel LJ, Pikul A, Donahue PR and Wogan GN. (1992a). Molecular dosimetry in rat urine of aflatoxin-N7-guanine and other aflatoxin metabolites by multiple monoclonal antibody affinity chromatography and immunoaffinity/high performance liquid chromatography. Cancer Res 52: 267–274. Groopman JD, Hall AJ, Whittle H, Hudson GJ, Wogan GN, Montesano R and Wild CP. (1992b). Molecular dosimetry of aflatoxin-N7-guanine in human urine obtained in the Gambia, West Africa. Cancer Epidemiol Biomark Prev 1: 221–227. Groopman JD, Scholl P and Wang JS. (1996). Epidemiology of human aflatoxin exposures and their relationship to human liver cancer. Prog Clin Biol Res 395: 211–222. Guo Y, Breeden LL, Fan W, Zhao LP, Eaton DL and Zarbl H. (2006). Analysis of cellular responses to aflatoxin B1 in yeast expressing human cytochrome P450 1A2 using cDNA microarrays. Mutat Res 593: 121–142. Harris AJ, Shaddock JG, Manjanatha MG, Lisenbey JA and Casciano DA. (1998). Identification of differentially expressed genes in aflatoxin B1 -treated cultured primary rat hepatocytes and Fischer 344 rats. Carcinogenesis 19: 101–109. Harris AJ, Dial SL and Casciano DA. (2004). Comparison of basal gene expression profiles and effects of hepatocarcinogens on gene expression in cultured primary human hepatocytes and HepG2 cells. Mutat Res 549: 79–99. Howard PC, Eppley RM, Stack ME, Warbritton A, Voss KA, Lorentzen RJ, Kovach RM and Bucci TJ. (2001). Fumonisin B1 carcinogenicity in a two-year feeding study using F344 rats and B6C3F1 mice. Environ Health Perspect 109 (Suppl. 2): 277–282. IARC (1993). International Agency for Research on Cancer, 1972–1999, Monographs on the Evaluation of Carcinogenic Risks to Humans, Vols 1–96, Lyon, France. Lesko LJ and Atkinson AJ Jr. (2001). Use of biomarkers and surrogate endpoints in drug development and regulatory decision making: criteria, validation, strategies. Annu Rev Pharmacol Toxicol 41: 347–366. Levy DD, Groopman JD, Lim SE, Seidman MM and Kraemer KH. (1992). Sequence specificity of aflatoxin B1-induced mutations in a plasmid replicated in xeroderma pigmentosum and DNA repair proficient human cells. Cancer Res 52: 5668–5673. Li Y, Wan D-F, Su JJ, Cao J, Ou C, Qiu XK, Ban KC, Yang C, Qin LL, Luo D, Yue HF, Zhang LS and Gu JR. (2004). Differential expression of genes during aflatoxin B1 -induced hepatocarcinogenesis in tree shrews. World J Gastroenterol 10: 497–504. Lin JK, Miller JA and Miller EC. (1977). 2,3-Dihydro-2-(guan-7-yl)-3-hydroxy-aflatoxin B1 , a major acid hydrolysis product of aflatoxin-B1 -DNA or -ribosomal RNA adducts formed in hepatic microsome-mediated reactions and in rat liver in vivo. Cancer Res 37: 4430–4438. Lunn RM, Langlois RG, Hsieh LL, Thompson CL and Bell DA. (1999). XRCC1 polymorphisms: effects on aflatoxin B1 -adducts and glycophorin A variant frequency. Cancer Res 59: 2557– 2561. Marasas WF, van Rensburg SJ and Mirocha CJ. (1979). Incidence of Fusarium species and the mycotoxins, deoxynivalenol and zearalenone in corn-produced esophageal cancer areas in Transkei. J Agric Food Chem 27: 1108–1112. Marasas WF, Kellerman TS, Gelderblom WC, Coetzer JA, Thiel PG and van der Lugt JJ. (1988a). Leukoencephalomalacia in a horse induced by fumonisin B1 isolated from Fusarium moniliforme. Onderstepoort J Vet Res 55: 197–203 Marasas WF, Jaskiewicz K, Venter FS and Van Schalkwyk DJ. (1988b). Fusarium moniliforme contamination of maize in esophageal cancer areas in Transkei. S Afr Med J 74: 110– 114. Merrill AH Jr, Sullards MC, Wang E, Voss KA and Riley RT. (2001). Sphingolipid metabolism: roles in signal transduction and disruption by fumonisins. Environ Health Perspect 109 (Suppl 2): 283–289.


187

Missmer SA, Suare L, Felkner M, Wang E, Merrill AH Jr, Rothman KJ and Hendricks KA. (2006). Exposure to fumonisins and the occurrence of neural tube defects along the Texas–Mexico Border. Environ Health Perspect 114: 237–241. Norred WP, Plattner RD, Vesonder RF, Bacon CW and Voss KA. (1992). Effects of selected secondary metabolites of Fusarium moniliforme on unscheduled synthesis of DNA by rat primary hepatocytes. Food Chem Toxicol 30: 233–237. Qian GS, Ross RK, Yu MC, Yuan JM, Gao YT, Henderson BE, Wogan GN and Groopman JD. (1994). A follow-up study of urinary markers of aflatoxin exposure and liver cancer risk in Shanghai, People’s Republic of China. Cancer Epidemiol Biomark Prev 3: 3–10. Qiu M and Liu X. (2001). Determination of sphinganine, sphingosine and Sa/So ratio in urine of humans exposed to dietary fumonisin B1 . Food Addit Contam 3: 263–269. Reddy JK, Svododa DJ and Rao MS. (1976). Induction of liver tumors by aflatoxin B1 in the tree shrew (Tupaia glis), a nonhuman primate. Cancer Res 36: 151–160. Riley RT, Voss KA, Yoo HS, Gelderblom WC and Merril AH Jr. (1994). Mechanism of fumonisin toxicity and carcinogenesis. J Food Prot 57: 638–645. Riley RT, Enongene E, Voss KA, Norred WP, Meredith FI, Sharma RP, Spitsbergen J, Williams DE, Carlson DB and Merrill AH Jr. (2001). Sphingolipid perturbations as mechanisms for fumonisin carcinogenesis. Environ Health Perspect 109(Suppl. 2): 301–308. Sabbioni G, Skipper P, Buchi G and Tannenbaum SR. (1987). Isolation and characterization of the major serum albumin adduct formed by aflatoxin B1 in vivo in rats. Carcinogenesis 8: 819–824. Sewram V, Nair JJ, Nieuwoudt TW, Gelderblom WC, Marasas WF and Shephard GS. (2001). Assessing chronic exposure to fumonisin mycotoxins: the use of hair as a suitable noninvasive matrix. J Anal Toxicol 25: 450–455. Sewram V, Mshicileli N, Shephard GS and Marasas WF. (2003). Fumonisin mycotoxins in human hair. Biomarkers 8: 110–118. Sheldon JL. (1904). A corn mold (Fusarium moniliforme sp.), in Agricultural Experimental Station of Nebraska 17th Annual Report, University of Nebraska, Lincoln, NE, USA, pp. 23–32. Shephard GS and van der Westhuizen L. (1998). Liquid chromatographic determination of the sphinganine/sphingosine ratio in serum. J Chromatogr 710: 219–222. Shephard GS, van der Westhuizen L, Thiel PG, Gelderblom WC, Marasas WF and van Schalkwyk DJ. (1996). Disruption of sphingolipid metabolism in non-human primates consuming diets of fumonisin-containing Fusarium moniliforme culture material. Toxicon 34: 527–534. Sinnhuber RO, Wales JH, Ayres JL, Engebrecht RH and Amend DL. (1968). Dietary factors and hepatoma in rainbow trout (Salmo gairdneri). 1. Aflatoxins in vegetable protein feedstuffs. J Natl Cancer Inst 41: 711–718. Solfrizzo M, Avantaggiato G and Visconti A. (1997). Rapid method to determine sphinganine/sphingosine in human and animal urine as a biomarker for fumonisin exposure. J Chromatogr 692: 87–93. Solfrizzo M, Chulze SN, Mallmann C, Visconti A, De Girolamo A, Rojo F and Torres A. (2004). Comparison of urinary sphingolipids in human populations with high and low maize consumption as a possible biomarker of fumonisin dietary exposure. Food Addit Contam 21: 1090–1095. Sun CA, Wu DM, Wang LY, Chen CJ, You SL and Santella RM. (2002). Determinants of formation of aflatoxin-albumin adducts: a seven-township study in Taiwan. Br J Cancer 87: 966–970. Tilton SC, Gerwick LG, Hendricks JD, Rosato CS, Corley-Smith G, Givan SA, Bailey GS, Bayne CJ and Williams DE. (2005). Use of a rainbow trout oligonucleotide microarray to determine transcriptional patterns in aflatoxin B1 -induced hepatocellular carcinoma compared to adjacent liver. Toxicol Sci 88: 319–330. Tran ST, Tardieu D, Auvergne A, Bailly JD, Babile R, Durand S, Benard G and Guerre P. 2006. Serum sphinganine and the sphinganine to sphingosine ratio as a biomarker of dietary fumonisins during chronic exposure in ducks. Chem Biol Interact 10: 41–50.

188

Hepatotoxicity

Ueng YF, Shimada T, Yamazaki H and Guengerich FP. (1995). Oxidation of aflatoxin B1 by bacterial recombinant human cytochrome P450 enzymes. Chem Res Toxicol 8: 218–225. van der Westhuizen L, Brown NL, Marasas WF, Swanevelder S and Shephard GS. (1999). Sphinganine/sphingosine ratio in plasma and urine as a possible biomarker for fumonisin exposure in humans in rural areas of Africa. Food Chem Toxicol 37: 1153–1158. van der Westhuizen L, Shephard GS and van Schalkwyk DJ. (2001a). The effect of a single gavage dose of fumonisin B1 on the sphinganine and sphingosine levels in vervet monkeys. Toxicon 39: 273–281. van der Westhuizen L, Shephard GS and van Schalkwyk DJ. (2001b). The effect of repeated gavage doses of fumonisin B1 on the sphinganine and sphingosine levels in vervet monkeys. Toxicon 39: 969–972. Voss KA, Chamberlain WJ, Bacon CW, Herbert RA, Walters DB and Norred WP. (1995). Subchronic feeding study of the mycotoxin fumonisin B1 in B6C3F1 mice and Fischer 344 rats. Fundam Appl Toxicol 24: 102–110. Voss KA, Liu J, Anderson SP, Dunn C, Miller JD, Owen JR, Riley RT, Bacon CW and Corton JC. (2006). Toxic effects of fumonisin in mouse liver are independent of the peroxisome proliferatoractivated receptor alpha. Toxicol Sci 89: 108–119. Wang E, Ross PF, Wilson TM, Riley RT and Merrill AH Jr. (1992). Increases in serum sphingosine and sphinganine and decreases in complex sphingolipids in ponies given feed containing fumonisins, mycotoxins produced by Fusarium moniliforme. J Nutr 122: 1706–1716. Wang LY, Hatch M, Chen CJ, Levin B, You SL, Lu SN, Wu MH, Wu WP, Wang LW, Wang Q, Huang GT, Yang PM, Lee HS and Santella RM. (1996). Aflatoxin exposure and risk of hepatocellular carcinoma in Taiwan. Int J Cancer 67: 620–625. Wild CP, Garner RC, Montesano R and Tursi F. (1986). Aflatoxin B1 binding to plasma albumin and liver DNA upon chronic administration to rats. Carcinogenesis 7: 853–858. Wild CP, Jiang YZ, Allen SJ, Jansen LA, Hall AJ and Montesano R. (1990a). Aflatoxin–albumin adducts in human sera from different regions of the world. Carcinogenesis 11: 2271–2274. Wild CP, Jiang YZ, Sabbioni G, Chapot B and Montesano R. (1990b). Evaluation of methods for quantitation of aflatoxin–albumin adducts and their application to human exposure assessment. Cancer Res 50: 245–251. Wild CP, Hudson G, Sabbioni G, Chabot B, Hall AJ, Wogan GN, Whittle H, Montesano R and Groopman JD. (1992). Dietary intake of aflatoxins with the level of albumin-bound aflatoxin in peripheral blood in The Gambia, West Africa. Cancer Epidemiol Biomark Prev 1: 229–234. Wild CP, Yin F, Turner PC, Chemin I, Chapot B, Mendy M, Whittle H, Kirk GD and Hall A. (2000). Environmental and genetic determinants of aflatoxin–albumin adducts in The Gambia. Int J Cancer 86: 1–7. Wilson TM, Nelson PE and Knepp CR. (1985). Hepatic neoplastic nodules, adenofibrosis and cholangiocarcinomas in male Fischer 344 rats fed corn naturally contaminated with Fusarium moniliforme. Carcinogenesis 6: 1155–1160.

Section 4 Mechanisms of Hepatotoxicity

9 Mechanisms of Toxic Liver Injury Nora Anderson and Jürgen Borlak

9.1

Introduction

Toxic responses may arise from acute or chronic exposure to chemicals and drugs. Unlike the intentional exposure to therapeutic drugs, exposure to industrial chemicals is accidental. This requires different strategies to investigate toxicity, which always coincides with metabolic dysfunctions and induction of stress-triggered programs. Whether or not an organ will overcome metabolic dysfunction and resume its physiologic function is highly dependent on the circumstances prevailing. In many ways, the liver, as the central organ for detoxification and clearance of exogenous and endogenous substances, constitutes a primary target for toxic reactions, but nonetheless has the remarkable ability to undergo regenerative growth. Several factors can be named that contribute to a high incidence of toxicity in the liver. These include remarkable exposure to high drug concentrations, enormous metabolic activity and the presence of several enzymes held responsible for generation of reactive metabolites, most notably ROS. Possible targets of toxic substances are macromolecular structures or individual molecules such as the bile acid transporters (i) or members of the nuclear receptor family (ii), as well as intracellular lipids (iii), proteins (iiii) or nucleic acids (iiiii) (see Figure 9.1). These targeted molecules become dysfunctional units and activate secondary pathways to result in programmed events, such as apoptosis, necrosis and autophagy, mitochondrial failure and immunological reactions (Kaplowitz, 2002; Lee and Ferguson, 2003). In addition, the functional integrity of a cell can also be disturbed by direct cytolytic reactions, membrane disruption and distortion of trans-membrane transport mechanisms. Even though several different mechanisms might be involved in the onset and progression of hepatotoxicity of a single substance, there are likely to be only a few central mechanisms that are activated within the general toxic response of the liver. Hepatotoxicity: From Genomics to in vitro and in vivo Models Edited by S. C. Sahu C 2007 John Wiley & Sons, Ltd

192

Hepatotoxicity

Figure 9.1

Cellular sites of drug-related toxicity

The actual sensitivity to certain toxins, however, is related to various determinants, which are also subject to high interindividual variability. Therefore, it is difficult to predict whether hepatotoxicity might occur upon treatment with a new substance, even though when studied in large patient cohorts. The complexity of different interacting pathways and mechanisms that are involved in the toxic responses complicate the distinct assignment of observed phenomena to either adverse effects or, at least in part, to therapeutic drug effects. This might explain why complications due to drug-related hepatotoxicity have such a high incidence, as for example in the USA, where incidences of 1 in 10 000 and 1 in 100 000 patients have been reported (Larrey, 2002). Despite knowledge gained over the past decade, no real progress has been made to face liver toxicity at least from the clinical point of view. Here we wish to discuss hepatotoxicity in a broad sense and follow several lines of evidence. We will particularly focus on the role of oxidative stress and inflammation for liver injury, but also summarize the vast amount of literature on toxicity induced by carbon tetrachloride, amantadine, iron, acetaminophen and halothane, among others. Whenever possible, the molecular pathology of a specific toxicant will be discussed in order to provide insight into mechanisms of the various stages of toxic liver injury. Finally, the needs to direct future research will be emphasized. 9.1.1

Basic Facts about Toxic Liver Injury

Notably, the development of toxic liver injury follows a two-staged course, the first phase of which is characterized by initiation of the injury and may involve direct interaction with a toxicant. This phase may exhibit dose-dependency (Mehendale, 2005). In contrast, the second phase is characterized by progress of the injury in a toxicant-independent fashion that is dominated by secondary events. Among the mechanisms contributing to phase two of toxic liver injury, the three leading mechanisms proposed are contribution of inflammatory cells (i) (Czaja et al., 1994; Laskin and Pendino, 1995; Piguet et al., 1990), oxidative stress and lipid peroxidation

Mechanisms of Toxic Liver Injury

193

(ii) (Poli, 1993; Slater, 1984) and leakage of degrading enzymes (iii) (Mehendale and Limaye, 2005; Poli et al., 1987). Furthermore, tissue repair has been determined to be one of the major factors that influence the fate of the liver after toxic injury, which may either be directed towards regeneration or loss of function and necrosis. Therefore, the toxicological response to different toxicants to a significant part obeys similar principles and may be characterized by the same mechanisms, although the acute and first phase of liver injury may be initiated by mechanisms specific to the toxicant. This may also explain why the morphology of toxic or drug-related liver injury usually differentiates into one of the four following phenotypes: hepatocellular injury, often associated with elevated liver enzymes (e.g. isoniazid) (i), cholestatic injury, which due to damage of the bile duct epithelia displays predominantly elevated alkaline phosphatase (AP) (e.g. amoxixillin – clavulanic acid) (ii), mitochondrial injury involving microvesicular steatosis (e.g. valproic acid) (iii) and the often delayed immunologic injury that is associated with fever, rash or eosinophilia (e.g. phenytoin) (iiii) (Navarro and Senior, 2006). Some drugs, as indicated by the examples given in parentheses, typically cause certain morphologies, but do not have a specific correlation to them. Thus, they can induce a mixed injury that is characterized by more than one feature of the different morphologic phenotypes. It is not known why certain drugs induce certain morphologies of liver injury but it is likely that these drugs share similar mechanisms of initiating toxicity in the liver. In the following, mechanisms recognized in different types of liver injury will be discussed, and general features of toxicity will be portrayed in detail.

9.2

Cholestatic Liver Injury

Cholestasis may result from an impairment of bile secretion or an obstruction of bile flow. The mechanistics underlying cholestatic liver injury has been linked to an impairment of bile salt transport by inhibition or downregulation of ATP-dependent bile salt transporters and also to alterations of actin resulting in disruption of the cytoskeleton and an impaired transport of bile along the canalicular system and into the bile ducts (Cullen, 2005). Interaction with bile acid transporters is a mechanism that has been observed under administration of several drugs and toxicants leading to toxic liver injury and is believed to be a central event in the development of cholestatic liver injury (Lewis, 2000). The connection between bile acid transporters and liver injury has first been drawn from observations made after experimental partial hepatectomy, where the acute liver injury was accompanied by regulation of hepatobiliary organic anion transporters (Geier et al., 2002; Trauner et al., 2005). Bile production is important for the clearance of endogenous and exogenous metabolites through the liver, and such is its detergent effect for adsorption and digestion of lipids. The primary bile salts, cholic acid and chenodeoxycholic acid, are derived from cholesterol and released in conjugation with taurine and glycine conjugates (Russell and Setchell, 1992). The uptake of bile salts that underwent enterohepatic circulation is located at the basolateral membrane of the hepatocyte and is mediated by a Na+ -dependent transporter system that communicates with portal vein blood via the space of Disse. This transporter system is

194

Hepatotoxicity

termed NTCP/Ntcp and belongs to a family of multi-specific organic anionic transporters (OATP/Oatp). Within the hepatocyte, bile acids are transported via diffusion or intracellular membrane vesicles. The canalicular transporters at the apical site of the hepatocyte belong to the family of ATP-dependent ABC transporters, such as the bile salt export pump (BSEP), multidrug transporter 2 (MDR2) and ABCG5/8 transporter (Stieger et al., 1992). Increased extracellular bile concentrations were shown to induce the apoptotic program of cell death in a death receptor-dependent fashion (Higuchi and Gores, 2003). Therefore, deregulation of bile transporters that leads to increased exposure to bile acids may cause serious hepatocellular damage and was furthermore demonstrated to trigger changes in intracellular signal transducers that affect cholangiocyte secretion, proliferation and survival (Xia et al., 2006). Defects of bile acid transporters have been demonstrated to be responsible for development of cholestasis in two of four types of progressive familial intrahepatic cholestasis (PFIC) (the Byler Syndrome (PFIC-2) and PFIC-3) (Jansen et al., 2001). These diseases arise from an impairment of the bile salt export pump (BSEP) and multidrug resistance protein 2 (MRP2) at the apical site of the hepatocytes. Another hereditary disease affecting bile acid transport is the Dubin–Johnson syndrome, in which mutations in MRP2 give rise to high serum bilirubin as a result of the liver’s inability to excrete this metabolite into bile (Paulusma et al., 1997). Carbon tetrachloride (CCl4 ) causes cholestasis and fibrosis and is a toxin frequently used to probe for mechanisms of hepatotoxicity. Indeed, carbon tetrachloride was found to interfere with the regulation of bile acid transporters, causing downregulation of sodium taurocholate co-transporting polypeptide (Ntcp) and organic anion transporting polypeptides 1 (Oatp1) and 2 (Oatp2) (Geier et al., 2002). The mechanisms responsible for this action are still not fully understood; however, pro-inflammatory mediators such as tumour necrosis factor alpha (TNF-α) were suggested to play a central role in CCl4 -induced liver injury and have been demonstrated to exert effects on bile acid transporters (Geier et al., 2002). Such were TNF-α and interleukin 6 (IL-6) shown to inhibit the sodium-dependent uptake of taurocholate by hepatocytes (Green et al., 1994; Whiting et al., 1995). Further in vitro studies have demonstrated that interleukin-1 beta (IL-1β) decreases basolateral Ntcp and Mrp2 promotor activity via reduction of the nuclear levels of transactivating retinoid X receptor/retinoic acid receptor (RXRα/RARXα) (Denson et al., 2000; Li et al., 2002). Moreover, TNF-α and IL-1β were shown to downregulate Oatp2 and MRP2, which are responsible for transport of bilirubin and other organic ions (Zollner et al., 2001). In the case of carbon tetrachloride, the upregulation of pro-inflammatory cytokines, which is frequently observed in cholestatic liver disease, was related to early activation of the early growth response factor-1 (Egr-1) transcription factor. This was supported by studies in Egr-1 knockout mice, where neutrophil accumulation, upregulation of macrophage inflammatory protein-2 (MIP-2) and intercellular adhesion molecule-1 (ICAM-1) in the liver were significantly reduced as compared to wild-type mice (Kim et al., 2006). Moreover, bile acids themselves revealed a stimulating effect on the expression of Egr-1 protein in hepatocytes in an animal model of cholestasis. Thus, elevated bile acids concentrations during cholestasis might cause a secondary downregulation of bile acid transporters by mediation of inflammatory cytokines (Kim et al., 2006).


195

Investigations in animal models of hepatic ischemia-reperfusion indicated a crucial role of Kupffer cells for the development of cholestatic liver injury (Kudo et al., 2004). Hepatic ischemia-reperfusion is often followed by cholestasis and, in fact, was associated with downregulation of several bile acid transporting proteins, accompanied by increased levels of TNF-α, IL-1β and IL-6 and decreased levels of the nuclear transcription factor HNF-1α (Tanaka et al., 2006). Interestingly, depletion of Kupffer cells with gadolinium chloride prevented increased expression of TNF-α and effects on HNF-1α, suggesting that alterations in the expression of hepatic transporters in hepatic ischemia-reperfusion injury may be a Kupffer cell-mediated event (Tanaka et al., 2006). By contrast, Kupffer cell involvement in ischemia-reperfusion has been suggested to exhibit protection against liver injury in particular by upregulation of heme degradation and bilirubin excretion. This was implicated by results of a study in which Kupffer cell-depleted rats had displayed significantly elevated the levels of GOT and GPT, as well as serum endotoxin concentrations after ischemia-reperfusion compared to animals with an intact Kupffer cell response (Kobayashi et al., 2002). Direct interaction of the toxicant with bile salt exporter proteins or with genes coding for these is likely to be a primary event of cholestatic liver injury (Yamamoto et al., 2006). The latter has been implied in oestrogen-induced hepatotoxicity that is observed in women during pregnancy or upon contraceptive and postmenopausal hormone therapy (Schreiber and Simon, 1983). The synthetic oestrogen derivative 17α-ethinylestradiol (EE2) causes a decrease of the taurocholate transport at the hepatic canalicular membrane (Lee et al., 2000) and the sodium taurocholate co-transporting polypeptide protein (NTCP) and consequently reduces bile flow (Simon et al., 1996). Oestrogens are ligands of cytosolic receptors that belong to the family of steroid hormone receptors. Once activated these receptors are translocated into the nucleus, where they operate as transcription factors for specific genes. Hence, direct regulation of genes coding for bile transporter proteins would be one anticipated mechanism causing cholestatic liver injury as a result of EE2 administration. Indeed, recent studies in various transcription factor knockout models have recognized oestrogen receptor alpha to be responsible for repression of multiple transporters for hepatic bile and cholesterol, possibly accounting for rising extracellular bile concentrations (Yamamoto et al., 2006). In livers of wild-type mice, the oestrogen receptor alpha (ERα) mediated downregulation of bile acid and cholesterol transporters expression in response to EE2, subsequently decreasing the secretion of bile acids and cholesterol. Moreover, ERα induced a shift of bile acid synthesis towards the acidic pathway by upregulating the expression of CYP7B1 and downregulating the expression of CYP7A1 and CYP8B1. In contrast, Erα null mice were resistant to oestrogen 17α-ethinylestradiol (EE2)-induced hepatotoxicity (Yamamoto et al., 2006). Another possible mechanism in drug-induced cholestasis was found in liver toxicity induced by the immunosuppressive drug cyclosporin A, the antibiotic rifampicin and the antidiabetic drug glibenclamide. These potential liver toxicants cause cholestasis by direct interaction with bile acid transporters and were shown to competitively inhibit the BSEP taurocholate transport (Byrne et al., 2002). In the case of ANIT-induced liver injury, the pharmacodynamics of a drug-transporting bile acid transporter have been emphasized to essentially add to the mechanistics of liver toxicity. Notably, ANIT is conjugated to glutathione in the hepatocyte and then transported across the canalicular membrane by MRP2, where it damages the cholangiocytes lining

196

Hepatotoxicity

the bile ducts (Liu et al., 2003). Secretion of ANIT and also arsenite by the Mrp2 transporter was demonstrated to induce a cycling of both toxins that was held accountable for a concentration-dependent depletion of GSH. By contrast, Mrp2-deficient rats displayed no GSH excretion and were protected from ANIT-induced cholestasis (Dietrich et al., 2001). 9.2.1

Cholestasis and Nuclear Receptors

Physiologically, bile acid homeostasis is maintained by the enterohepatic nuclear factors farnesol X receptor (FXR), liver X receptor (LXR), pregnane X receptor (PXR) and CAR (Francis et al., 2003; Guo et al., 2003; Karpen, 2002; Xie et al., 2001). Bile salts are natural ligands for several nuclear hormone receptors that regulate transcription of genes involved in bile acid homeostasis. Specifically, the nuclear transcription receptor FXR is an essential regulatory component of cholesterol homeostasis, which is activated by bile acids (Forman et al., 1995a). FXR is responsible for transcription of genes coding for bile acid biosynthesis and transport. Activation of FXR, for example, was shown to induce repression of the basolateral NTCP bile uptake transporters and to mediate upregulation of the bile salt export pump (BSEP), resulting in increased uptake of bile by the hepatocyte and increased export of bile into the bile canaliculi (Ananthanarayanan et al., 2001; Sinal et al., 2000). Moreover, FXR induces expression of the multidrug resistancerelated protein 2 (MRP2; ABCC2), controls expression of the MDR3 (ABCB4) gene, encoding for the canalicular phospholipid flippase (Lambert et al., 2003; Liu et al., 2003), and represses key genes involved in bile acid biosynthesis, such as sterol 12α-hydroxylase (CYP8B1) and cholesterol 7 alpha-hydroxylase (CYP7A1) (Repa and Mangelsdorf, 2000). Knockout models have supported the vital activity of these transcription factors for liver health and have suggested that they play a central role in the development of cholestatic liver injury. FXR-null mice, for instance, were found to exhibit a phenotype similar to that of the inherited cholestatic liver disorder Bylers disease, and indeed further investigations revealed that progressive familial intrahepatic cholestasis type 1 is associated with a decreased activity of farnesoid X receptor (Chen et al., 2004a). The activities of both the human FXR and BSEP promoters were found to be reduced in patients diagnosed with Bylers disease. While type 2 of this disease arises from mutations in BSEP (Strautnieks et al., 1998), as mentioned before, progressive familial intrahepatic cholestasis type 1 (PFIC1) is caused by mutation of the gene encoding a putative aminophospholipid transferase that is required for post-translational modifications enabling the nuclear translocation of FXR (Suchy and Ananthanarayanan, 2006). Therefore, the phenotype of Bylers disease has been suggested to arise from an enhanced ileal uptake of bile salts via upregulation of the apical sodiumdependent bile acid transporter and diminished canalicular secretion of bile salts secondary to downregulation of the bile salt excretory pump due to an altered function of FXR (Chen et al., 2004a). FXR activity is closely related to processes involved in the development of cholestasis. Experimental treatment with the synthetic FXR agonist GW4064 revealed the protective effect of FXR activity against cholestatic liver damage in rat models of extrahepatic and intrahepatic cholestasis (Fiorucci et al., 2005; Rizzo et al., 2005). In ANIT-induced toxic injury, treatment with the FXR agonist resulted in significant reductions in serum alanine aminotransferase aspartate aminotransferase and lactate dehydrogenase. In addition, the extent of inflammatory reaction and necrosis was reduced compared to untreated mice (Liu et al., 2003).


197

Besides deregulation of bile acid transport, dysfunctional FXR might also be responsible for interruption of the physiological downregulation of bile acid synthesis. Since bile acids are known to repress the bile acid synthesis by a negative feedback mechanism via downregulation of CYP7A1 transcription (the rate-limiting enzyme of bile synthesis) through the FXR-SHP-LRH1 cascade (SHP = short heterodimer partner) (Goodwin et al., 2000; Lu et al., 2000; Sinal et al., 2000). Despite decreased expression of the bile salt export pump in FXR-null mice, which was suggested to be responsible for their increased sensitivity to bile acid-mediated toxicity (Miyata et al., 2005), it was indicated that FXR-null mice were relatively protected from secondary cholestasis in a bile duct ligation (BDL) model (Stedman et al., 2005, 2006). It was reported that FXR-knockout mice displayed reduced serum and liver bile acid concentrations and less extensive liver injury in response to BDL in comparison to wild-type mice. This protective effect was attributed to increased hepatic expression of the multidrug resistance associated protein (mrp) 4, resulting in an increased transport of bile acids into the blood and consecutively in enhanced urinary excretion (Marschall et al., 2006; Stedman et al., 2006). Thus, it has been hypothesized that the pharmacologically antagonising of FXR may provide a therapeutic benefit in obstructive cholestasis (Stedman et al., 2006). Hence dysregulation of FXR may substantially contribute to the processes involved in development of cholestasis due to toxic liver injury. This is particularly interesting, since repression of FXR activity is not related to direct binding to a xenobiotic ligand, such as seen in other nuclear receptors, including CAR or PXR, implicating a secondary mechanism that may affect the concentrations of the endogenous FXR ligands, as for instance secondary bile acids. 9.2.2

Bile Acid Synthesis

Another possible mechanism that might cause elevation of the extracellular bile acids is altered expression of enzymes involved in bile acid synthesis. This, for instance, has been implicated in liver toxicity induced by ANIT, where cholestatic liver injury was found to be associated with upregulation of NTCP and CYP7A1 expression, thereby increasing bile uptake from the blood in addition to enhanced intracellular bile acid production. Several mechanisms, however, may contribute to the morphology of cholestatic injury, and a wide spectrum of various aspects is available to explain the development of cholestatic liver injury. Notably, toxicant-induced cholestasis is associated with bile duct obstruction that eventually will lead to a decrease of the bile flow. In ANIT-induced cholestasis this has been related to an induction of biliary epithelial cell (BEC) hyperplasia in rats, which was assigned to enhanced BEC proliferation and consecutive bile duct obstruction. The BEC proliferation upon ANIT treatment was correlated to BEC necrosis and bile duct obstruction in a time- and dose-dependent way (Kossor et al., 1995). Another animal model in which cholestasis was induced by feeding lithocholic acid (LCA) has demonstrated that increased concentration of toxic bile acids cause bile duct obstruction and destructive cholangitis accompanied by periductal fibrosis (Fickert et al., 2006). Taken together, these findings suggest that bile duct obstruction in toxic cholestasis may result from bile duct proliferation mediated by increased extracellular bile acid

198

Hepatotoxicity Toxicant, chemokines

Bile acid import

Bile acid export Bile

Blood Apoptosis

NTCP

Hepatocyte OATP

Bile acid synthesis

Proliferation

BSEP

Bile acids

Cholangiocyte

ABCG

CYP7A1

Figure 9.2 Possible mechanisms of toxic disruption of bile acid homeostasis, mediated via deregulation of bile acid transporters. An example is the transcriptional downregulation of NTCP and OATP along with an increase of bile acid export by upregulation of BSEP or ABCG. Subsequently increased extracellular bile acids trigger metabolic dysfunctions in hepatocytes and cholangiocytes, causing cell death and bile duct obstruction. Reproduced by permission of the American Society for Biochemistry and Molecular Biology from Yamamoto et al. (2006), Journal of Biological Chemistry, 281, 16625–16631

concentrations which result from toxic disruption of bile acid homeostasis in hepatocytes, as depicted in Figure 9.2. It can be concluded that mechanisms causing cholestasis can be directed either through drugs that interfere with the expression of bile transporters via activation of nuclear transcription factors or through extracellular activation triggered by enhanced cytokine release. The resulting effects, however, affect different types of liver cells, such as hepatocytes, cholangiocytes or hepatic stellate cells. Chronic cholestasis of the liver induces onset of fibrotic processes, which sooner or later will lead to the development of liver fibrosis. In this context, besides the severe metabolic derangements found in hepatocytes, notably the influence of bile acids upon other liver cells has become an important field of research. Specifically, the activation of hepatic stellate cells (HSCs) and portal fibroblasts to myofibroblastic cells is an early and major event of chronic cholestatic liver injury (Hines et al., 1993; Tang et al., 1994; Tuchweber et al., 1996). Mechanisms of cholestatic liver injury in summary: – Obstruction of bile flow (proliferation of bile epithelia) – Inhibition of bile salt transporters→impaired secretion into bile canaliculi (direct interaction or transcriptional downregulation) – Inhibition of physiological downregulation of bile acid synthesis (inhibition of negative feedback mechanisms) – Induction of genes in bile acid synthesis (CYP8B1 and CYP7A1)


9.3

199

Steatosis

The term ‘hepatic steatosis’ refers to an intracellular accumulation of lipid droplets in the cytoplasm. ‘Primary steatosis’ is often observed in patients displaying symptoms of the metabolic syndrome including obesity, diabetes, hypertriglycerinaemia and insulin resistance. Secondary hepatic steatosis is extrinsically induced by alcohol, several drugs, copper accumulation in Wilson’s disease and other factors (Pessayre et al., 2001). Hepatic steatosis may also be accompanied by hepatocellular necrosis, inflammation and fibrosis, in which case it is termed as ‘non-alcoholic steatohepatitis’ (NASH). The morphology of steatosis has been linked to primary mitochondrial failure and has been implicated for microvesicular steatosis, non-alcoholic steatohepatitis (NASH) and cytolytic hepatitis (Fromenty and Pessayre, 1995). This was concluded from observations that these pathologies were associated with early perturbations of the mitochondrial metabolism and associated with apoptosis induced by mitochondrial pathways. Several events have been demonstrated to cause failure of the mitochondrial metabolism, which eventually will end in cellular necrosis and apoptosis (Kim et al., 2003a).The two main mechanisms by which toxicants cause primary mitochondrial failure are uncoupling of the oxidative phosphorylation and initiation of cell death. Different toxicants, however, have their different ways of reaching these endpoints. For example, uncoupling of the respiratory chain can result from direct interference with mitochondrial enzymes or from dysfunctions caused by mutations of the mitochondrial DNA (mtDNA) (both, for instance, observed effects of iron) (Ramm and Ruddell, 2005). Uncoupling of the mitochondrial oxidative phosphorylation is associated with energy depletion and extensive production of reactive oxygen species (ROS), leading to a vicious circle of cellular stress. In fact, the presence of ROS has been associated with the pathology of fatty liver diseases (Yang et al., 2004b). Increases in intracellular ROS have been observed in alcohol-related NASH and steatohepatitis in Wilson’s disease, accounting for enhanced oxidative degeneration of mitochondrial DNA (Mansouri et al., 1997, 1999). A combination of modified mitochondrial DNA and oxidation of mitochondrial proteins and lipids has also been implicated in increased ROS formation in alcoholic patients (Pessayre et al., 2001). Ethanol-induced hepatotoxicity progresses through different phases, beginning with accumulation of lipids and hepatic steatosis, passing via fibrosis and ending in cirrhosis in a worst-case scenario. Mitochondrial toxicity is often associated with ultrastructural lesions, including crystalline inclusions and formation of oversized megamitochondria, as have been observed in livers from NASH patients (Sanyal et al., 2001). Besides impairment of oxidative phosphorylation, especially alterations of fatty acid ß-oxidation have been suggested to significantly contribute to the development of the steatotic phenotype (Fromenty and Pessayre, 1995). Physiologically, fatty acids are metabolized via mitochondrial and peroxisomal ßoxidation. Disturbance of these pathways through, for instance, dysfunction of mitochondrial key enzymes causes accumulation of the non-esterified fatty acids (NEFAs). This has been demonstrated in patients with NASH and ethanol-induced liver disease, where levels of circulating free fatty acids were found to be elevated and were positively correlated with the severity of the disease (Nehra et al., 2001). On the one hand, fatty acids have been demonstrated to exert toxic effects on the cell and were associated with induction of apoptosis by a mitochondrial and a lysosomal pathway (Martins et al., 2006).

200

Hepatotoxicity

On the other hand, accumulation of fatty acids results in further perturbations of the lipid metabolism and subsequently non-metabolized fatty acids are esterified to triglycerides. Importantly, cholesterol accumulation in the inner mitochondrial membrane due to chronic ethanol treatment is believed to cause a decrease of GSH transporter activity and thereby selectively depletes mitochondrial GSH. This was demonstrated to be important not only for activation of apoptotic cascades, but also for the redox status of the entire cell, since the mitochondrion is also the most significant cellular organelle for production of H2 O2 (Fernandez-Checa and Kaplowitz, 2005). Alterations of the ß-oxidation cause a gradual remodelling of the normal liver, progressing towards fatty liver. If the lipid fraction constitutes more than 5–10 % by weight of the organ, the morphology is referred to as ‘steatosis’. Impairment of the ß-oxidation may result from the obligatory connection of ß-oxidation and oxidative phosphorylation. Here the generation of reduced NAD+ during ß-oxidation is essential for the operation of the mitochondrial respiration (as demonstrated in the diagram below). Diminished co-substrate NAD may result from interference with the mitochondrial respiration and will consecutively lead to a decreased rate of ß-oxidation. β-oxidation NAD+

NADH mitochondrial respiration

Accordingly, perturbations of the mitochondrial ß-oxidation can derive from dysfunction of several mitochondrial enzymes of ß-oxidation or oxidative respiration. This can be linked to mutations of mitochondrial DNA (mtDNA) or defect-transport mechanisms of nuclear proteins required for the mitochondrial function. Recent investigations revealed increased rates of apoptosis in NASH hepatocytes that were associated with activation of NF-κB, which correlated with the severity of the disease (Feldstein et al., 2003a; Ribeiro et al., 2004). Concomitantly, increased expression of death receptors Fas and TNFR1 and activation of caspases 3 and 7 were observed (Feldstein et al., 2003a; Ribeiro et al., 2004). Further investigations revealed that fatty acids induced apoptosis by two different pathways. One is based on translocation of the antiapoptotic protein Bax to the lysosome, which causes lysosomal destabilization, accompanied by release of the lysosomal cysteine protease cathepsin B (Feldstein et al., 2004). The other pathway is initiated at the site of the mitochondrion and causes cell death by activation of JNK and release of cytochrome c (Malhi et al., 2006a). Release of cathepsin B was furthermore associated with enhanced expression of TNF-α, while inhibition of this lysosomal protease was found to be protective against liver toxicity of free fatty acid, which otherwise resulted in development of hepatic steatosis, liver injury, and insulin resistance (Feldstein et al., 2004, 2006). Studies in steatotic livers of wild-type and cathepsin B-knockout (Ctsb(−/−)) mice have supported the protective effect of cathepsin inhibition against stress-induced apoptosis in a model of cold ischemia–warm reperfusion injury (Baskin-Bey et al., 2005). Mitochondrial dysfunction is a general theme in liver toxicology, and several mechanisms triggered by liver toxicants have been identified, which will be discussed later in detail.


201

Mechanisms of steatotic liver injury in summary: – Inhibition of β-oxidation→inhibition of mitochondrial respiration – Accumulation of possibly toxic fatty acids→metabolic dysfunction and activation of lysosomal proteases – Induction of mitochondrial dysfunction and ROS→lipid peroxidation

9.4

Hepatitis

Hepatitis is characterized by the presence of inflammatory cells. In a toxicological context, there are at least two important types of hepatitis. Primary hepatitis results directly from damages induced by the toxicant, while secondary hepatitis has a delayed onset and is mediated by secondary events. Primary hepatitis can be induced by direct and acute reaction with the toxicant, while secondary hepatitis is mediated by intermediate events leading to a delayed form of inflammation. 9.4.1

Halothane-Induced Immune Hepatitis and Bioactivation

Drug-induced immune hepatitis is mostly characterized by a delayed onset and has been described for several substances, such as the anaesthetic halothane, the antidiuretic tienilic acid, the antibiotic sulphamethoxazole (Farrell et al., 2003; Gruchalla and Sullivan, 1991), the neuroleptic carbamazepine (Madden et al., 1996; Naisbitt et al., 2003) and the proteinase inhibitor nevirapine (Park et al., 2005). Intriguingly, all of these drugs share one property – they all undergo bioactivation leading to formation of their hepatotoxic metabolites, which are conjugated to intracellular proteins and recognized by immunocompetent cells, thereby initiating an immunologic response in the liver. The resulting immunoallergic hepatitis has been best described in patients treated with halothane. Halothane-induced hepatitis occurs after the second exposure to the drug, usually on the 5th day after the anaesthesia, and is accompanied by general inflammatory symptoms, such as fever, rash, and eosinophilia (Pohl, 1990). The incidence is rare (1 in 35 000 patients) but the mortality associated with the development of an acute insufficiency of the liver is high (up to 50 %) (Bourdi et al., 1996). In patients diagnosed with halothane hepatitis, the presence of unique antibodies against microsomal proteins is characteristic. Among these are, for instance, antibodies targeting the E2 subunit of pyruvate carboxylase, the 2-oxoglutarate dehydrogenase complex (Frey et al., 1995) and native and isoforms of protein disulfide isomerase (Martin et al., 1993) in humans as well as others (Bourdi et al., 1996). Responsible for the modification of cellular proteins (lysine residues in particular) and formation of neoantigens is the main metabolite of halothane, namely trifluoroacetylchloride (Kenna et al., 1988). Formation of antibodies results from processing of trifluoroacetylated proteins and presentation of neoantigens together with class MHC II by Kupffer or B-cells. Subsequent activation of T-helper cells triggers the transformation of B-cells to plasma cells to produce specific antibodies, although activation of T-killer cells leading to cytolytic reactions might be another possible pathway.

202

Hepatotoxicity

Note that the exact immunologic mechanism leading to the hepatocellular injury by halothane is not completely understood, even though evidence for a major role of the T-cell response has been provided (Lohse et al., 1990, 1992). Intriguingly, in immunoallergic hepatitis induced by halothane, tienilic acid and antihypertensive dihydralazine antibodies against specific isoforms of the cytochrome enzymes that metabolize these particular drugs, have been identified. These antibodies interact with the CYPs in a way that leads to inhibition of their activity – a feature that has been emphasized to contribute to the toxicity observed under administration of these drugs to a major extent (Park et al., 2005). The reactive metabolite of tienilic acid (ticynafen), for instance, binds covalently to the CYP2C9, by which it is generated, and thereby leads to formation of unique antibodies, termed anti-liver/kidney microsome antibodies type 2 (LKM2) (Bonierbale et al., 1999; Lecoeur et al., 1996; Neuberger and Williams, 1989). In dihydralazine-induced hepatitis, antibodies recognizing CYP1A2 (Belloc et al., 1997; Bourdi et al., 1990) have been identified, while in case of halothane extensive generation of autoantibodies reacting with CYP 2E1 has been detected (Bourdi et al., 1996). Mechanism of autoimmune hepatitis in summary: – Formation of auto antibodies (autoimmune hepatitis)→inhibition of metabolising enzymes and inflammatory reaction

9.5

Liver Fibrosis

Liver fibrosis is characterized by an excessive accumulation of connective tissue, leading to a disturbed liver architecture that is associated with severe pathophysiological consequences. Formation of so-called pseudo-septa and regeneration nodules leads to a diversion of the blood flow and impairs supply with nutrients and oxygen. Furthermore, toxic liver fibrosis is accompanied by both bile ductular proliferation and inflammation under various conditions (Muller et al., 1996). The dynamic processes that are involved in the advance of fibrosis are comparable to those activated during wound healing, including differentiation of local immunocompetent cells and increased production of extracellular matrix (ECM). Hepatocyte layers, sinusoids, portal and central vein tracts determine the distinct architecture of the normal liver. In regions exposed to high levels of shear stress, the extracellular matrix in capsule, septal, perivascular and periductal areas is built of strong collagens I, II, V and fibronectin (Schuppan, 1990). This composition of extracellular matrix is called intestinal matrix and is mainly generated by local fibroblasts. In the perisinoidal space of Disse, most of the intercellular communication between liver and blood cells takes place. Here the extracellular matrix, which is mainly generated by stellate cells, is more delicate and resembles the composition of a basement membrane, containing collagen IV, VI, laminin and fibronectin (Arenson et al., 1988; Hahn et al., 1980). The remodelling within the fibrotic process leads to dramatic metabolic derangements as a result of excessive ECM production and deposition of collagens I and II in the space of Disse. The production and composition of several ECM compounds results from a


203

balanced equilibrium of regulatory mechanisms. These involve production of fibrogenic cytokines by Kupffer cells and hepatocytes, including interleukins 1 and 6 (IL-1, IL-6), and participation of pro-fibrogenic growth factors, namely fibroblastic growth factors (FGFs), platelet-derived growth factors (PDGFs), and macrophages colony stimulating growth factor (M-CSF), as well as tumour growth factor (TGF-β) (Poli, 2000). Prevalence of one of the opposing mechanisms regeneration and inflammation within the liver decides the fate of functional liver parenchyma to either regenerate or be inescapably lost (Mehendale, 2005). In particular, the interaction between different liver cell types and between cells and interaction with the extracellular matrix have been recognized to account for both activation of fibrotic processes and regeneration.

9.5.1

Cellular Interactions in Fibrosis due to Toxic Liver Injury

The orchestration of signalling between different types of cells is essential for the secretory, synthetic and metabolic function of the liver. More than 15 different cell types are found in the liver (Malarkey et al., 2005). Hepatocytes are the centre of metabolic and synthetic actions and account for 60 % of the liver cells. The sinusoidal endothelial cells (SECs) are the primary barrier between blood and the hepatocytes filtering blood and the space of Disse (Smedsrod, 2004). These types of cells are particularly susceptible to oxidative stress and energy depletion and have been suggested to play a central role in toxic events (Dobbs et al., 1994; McCuskey, 2006). Kupffer cells are derived from monocytic cells and play a role in immune response and inflammation. Their ability to produce cytokines, such as TNF-α, and radical oxygen species has been implicated to play a crucial role in promoting the progression of toxic liver injury (Videla et al., 2003; Vrba and Modriansky, 2002). Stellate cells (Ito cells) produce extracellular matrix, store and metabolize vitamin A and lipids and are involved in processes that lead to regeneration, fibrosis and cirrhosis (Mabuchi et al., 2004). The interplay between the different cell types of the liver is well balanced. Non-parenchymal cells can influence both positively and negatively the proliferation of hepatocytes (Malik et al., 2002). A deregulation of their interaction due to ischemic or toxic stress can cause severe injury. The liver’s susceptibility to toxic stimuli is variable and shows regional differences. It was demonstrated that the architecture of the liver itself has influence on the changes that occur within physiologic as well as pathologic conditions. Morphologic arrangement of portal vein vessels, bile ducts and arteries, which form the functional units of the acini and the Rappaport fields, located between the branches of the central liver veins, form a 3-dimensional structure that seems to influence the metabolic competence of the liver. Indeed, gradients in protein concentrations, enzyme production and oxygen supply contribute to the variable susceptibility of different regions of the liver to stress and nutrient depletion. These factors seem to play a role for the development of liver injury, but also for the mechanisms of regeneration. Varying regional susceptibility has been observed during toxic responses (Heinloth et al., 2004; Irwin et al., 2004), chemical carcinogenesis (Elba et al., 2002; Garcia-Torres et al., 2003), as well as during development and progression of cirrhosis (Matsuzaki et al., 1997).

204

Hepatotoxicity

In the past decade, Kupffer cells and stellate cells were found to play a key role in inflammatory signalling and progression of fibrosis. Activation of Kupffer cells and stellate cells is believed to account for profibrotic and inflammatory events observed in development of liver fibrosis. Besides, there is evidence for an involvement of reactive oxygen species (ROS) and oxidative stress in several steps of fibrosis. Specifically, ROS were reported to exert stimulatory effects on cytokine generation by Kupffer cells and biosynthesis of extracellular matrix precursors by stellate cells (Pinzani and Marra, 2001; Refik et al., 2004; Thakur et al., 2006). It is assumed that any kind of liver injury triggers a signal cascade in which macrophages release cytokines and thereby recruit stellate cells, portal fibroblasts and others to initiate differentiation into myofibroblasts, characterized by expression of desmin or smooth muscle alpha actin (SMA), eventually resulting in increased ECM production (Knittel et al., 1999). Moreover, hepatic stellate cells (HSCs) have a central role in regeneration processes of the liver. Activated upon liver injury, HSCs display increased proliferation, motility and extracellular matrix (ECM) production (Marra et al., 1999b). This transformation into the myofibroblastic phenotype is a central event during liver fibrosis and has been related to the activity of extracellular signal-regulated kinase (ERK), which is a member of the protein–serine/threonine kinases known as mitogen-activated protein kinases (MAPKs) (Marra et al., 1999). In response to exposure to the platelet-derived growth factor (PDGF) and ECM, the ERK pathway is activated, induces the ability to increase cell migration (Klemke et al., 1997) and upregulates the expression of procollagen genes (Davis et al., 1996; Svegliati-Baroni et al., 1999). Furthermore, activated hepatic stellate cells produce cytokines, such as TGF-β, which induces immigration of inflammatory cells by chemoattraction and is also able to induce apoptosis of hepatocytes by a paracrine mechanism, notably activating apoptosis by binding to TNFR1 and consecutive caspase activation (Gressner et al., 1996). 9.5.2

Stellate Cells Abide Oxidative Stress in Liver Fibrosis

Generation of ROS has been closely associated to enhanced production of extracellular matrix in ethanol-induced fibrosis in stellate cells. Specifically, CYP2E1 that is induced upon ethanol treatment is a potent generator of reactive oxygen species. The superoxide anion resulting from this reaction forms a hydrogen peroxide which leads to formation of the highly reactive 1-hydroxyethyl radical that is probably the main source of oxidative stress in ethanol-induced liver injury. The strongest indicator for extensive oxidative stress in ethanol-induced liver damage is lipid peroxidation. Indeed, end-products of lipid peroxidation, such as 4-hydroxy-2,3nonenal (HNE) or malonaldehyde (MDA), in particular, were demonstrated to induce upregulation of procollagen I synthesis in stellate cells (Greenwel et al., 2000; Parola et al., 1993). The causal relation of lipid peroxidation and fibrogenesis was demonstrated in experiments in which application of the antioxidant vitamin E ameliorated liver toxicity and fibrogenesis (Parola et al., 1992). In addition, the main metabolite of ethanol itself, namely acetaldehyde, is accountable for H2 O2 production in stellate cells. It was demonstrated that hydrogen peroxide directly induces transcription of procollagen α 1 (I) (Greenwel et al., 2000), although ethanol was shown to account for the major fraction of enhanced α 1 (I) procollagen mRNA expression (Fontana et al., 1997).


Lipid peroxidation

Lipid aldehydes

ROS Redox sensors

Upregulation of ECM production

Liver cell

Inflammation

205

Stellate cells and fibroblasts

Stellate cell

IL 1, 6 growth factors

Figure 9.3 Central role of lipid peroxidation: toxic stimuli result in production of ROS and lipid peroxidation, which activate enhanced ECM production in fibroblasts and stellate cells

Acetaldehyde was also reported to induce upregulation of the α 2 (I) collagen and fibronectin gene expression, which is believed to involve activation of the protein kinase C and down-stream activation of signal transduction pathways, including phosphatidylinositol 3 -kinase (PI3K) and extracellular-regulated kinase (1/2) (ERK1/2) (Svegliati-Baroni et al., 2001). The importance of acetaldehyde as a profibrotic factor may further arise from its ability to form adducts with the carboxyl-terminal propeptides of procollagen in hepatic stellate cells. This has been demonstrated to result in a reduced feedback of collagen synthesis and hence may be responsible for collagen overproduction in ethanol-induced fibrosis (Ma et al., 1997). In conclusion, upregulation of ECM production is a major finding in fibrosis and is triggered in stellate cells via ROS-dependent signalling. The transcriptional activation of ECM precursors can be induced by several mechanisms, including direct stimulation via ROS, production of cytokines and generation of aldehydic products from lipid peroxidation (see Figure 9.3). Even though, the full extent of pathological processes causing fibrosis is not fully understood, induction of TNF-ß1 and activation of the redox-sensitive transcription factor activator protein-1 (AP-1) are most likely events contributing to the progression and inflammatory reaction of fibrosis (Leonarduzzi et al., 2000). Furthermore, the role of the extracellular matrix composition needs to be emphasized as a mechanism by which activation of hepatic stellate cells may contribute to severe metabolic derailment in liver injury and development of fibrosis. Growth and differentiation of hepatocytes is vitally dependent on interaction with the extracellular matrix (ECM). Lack of ECM was demonstrated to induce a loss of hepatocyte differentiation, which was associated with activation of several stress-signalling pathways, including the MAPK, SAPK/JNK and c-Jun signalling pathways, that caused recruitment of the activated protein-1 (AP-1) complex (Otsu et al., 2001; Sidhu et al., 2004). The effects of ECM on the metabolic competence of hepatocytes has been impressingly demonstrated in in vitro experiments, where the effects of adult ECM, fibroblast ECM and foetal ECM were compared in regard to activity of liver-specific enzymes, proteins and transcription factors, such as hepatic nuclear factor-4 (HNF-4). It was shown that adult hepatocytic ECM supports the expression of adult genes, while foetal hepatocytic ECM induced expression of foetal

206

Hepatotoxicity

genes. In contrast, fibroblastic ECM inhibited hepatocyte proliferation and tissue-specific gene expression in both foetal and adult hepatocytes (Brill et al., 2002). Moreover, the composition of ECM was shown to modulate other essential functions of liver cells. Exposure to increased collagen concentrations within the ECM has been demonstrated to severely influence the sensitivity to harmful stimuli. Thus, it was shown that in hepatocytes grown on a collagen matrix activation of the heat shock response failed to protect against TNF-α-induced apoptosis (Gosnell et al., 2000). Therefore, activation of hepatic stellate cells may in turn disrupt the hepatocyte’s metabolism in two different ways – first by production of inflammatory cytokines and secondly by an enhanced production of extracellular matrix proteins. Furthermore, apoptosis itself has been proposed to have a central role in activation of this profibrotic cascade. This was, for instance, supported by studies in Fas-deficient mice, where extrahepatic cholestasis was induced by performing bile duct ligation. Three weeks after the intervention, the Fas-deficient mice displayed a significantly reduced amounts of actin and collagen as compared to wild-type mice (Canbay et al., 2002). Moreover, engulfment of apoptotic bodies was demonstrated to induce stimulation of Kupffer cellular generation of death ligands Fas and TNF-α, while caspase inhibition abrogated the latter (Canbay et al., 2003). Therefore, deregulation of apoptotic pathways may contribute to promoting liver fibrosis by an impairment of regeneration processes. This was supported by the finding that fibrosis following chronic hepatocytic apoptosis was associated with inactivation of the pro-apoptotic protein bcl-xl (Takehara et al., 2004). It can be concluded that liver fibrosis arises from a ‘potpourri’ of several mechanisms and profibrotic effects, in which the hepatocellular function and regeneration is altered by inflammation, proliferation of cells of the myofibroblastic phenotype and enhanced production of extracellular matrix. Mechanisms of toxic fibrosis in summary: – Activation of stellate cells→induction of chemoattraction and apoptosis and ROS – Generation of ROS→up-regulation of profibrotic genes – Disturbance of extracellular matrix composition→affecting differentiation and cell metabolism – Presence of apoptosis→enhanced production of death receptors

9.6 9.6.1

General Mechanisms and Features of Toxic Liver Injury Nuclear Receptors, Recognition of Ligands and Xenobiotic Response

The cellular response to xenobiotics and toxicants involves several general pathways that lead to adaptive responses, the first step of which is recognition of an external component. There are different alarming systems to accomplish the latter. Even though some substances are recognized and effective by mediation of extracellular receptors, such as ßblockers or K+ -channel inhibitors, most drugs and toxicants enter the cell unrecognized


207

and passively by diffusion, giving rise to wanted and unwanted drug effects. In addition, responses to xenobiotics comprise adaptive reactions of the cellular metabolism, such as enhanced rates of biotransformation, drug transport, proliferation and/or regeneration. Within the past decade, an enormous gain in information regarding cellular xenobiotic sensing systems (the so-called xenosensors) has been achieved. An impressive and highly relevant example for cellular response to drugs and toxicants is the induction of CYP450 enzymes, which play a major role in Phase I reactions of biotransformation. The nuclear receptor family of ligand-activated transcription factors has been discovered to be substantially involved in recognition of drugs and toxicants and regulation of CYP enzyme expression. Nuclear receptors (NRs) are regulators of gene transcription and bind to specific DNA response elements implemented in the flanking regions of genes which they control. The human nuclear receptor superfamily consists of four groups: steroid hormone receptors, dimeric orphan receptors, RXR heterodimers and monomeric receptors. Activation of the RXR heterodimers requires the retinoid X receptor (RXR) as a binding partner. The ligands of the orphan receptors are, as yet, unidentified. The family of ligand-activated nuclear receptors is substantially involved in the xenobiotic response and their impact goes far beyond a simple enhancement of biotransformation. In the following, the role of the constitutive androstane receptor (CAR), the aryl hydrocarbon receptor (AHR), the PXR, FXR, vitamin D receptor and the oestrogen receptor for induction of bioactivation and cellular response to xenobiotics in the liver will be elucidated. 9.6.1.1

The Aryl Hydrocarbon Receptor (AHR)

The aryl hydrocarbon receptor (AHR) is a transcription factor that induces expression of human CYP1A1, CYP1A2 and CYP1B1 (Quattrochi et al., 1994; Whitlock, 1999). In addition, the AHR was determined to be responsible for regulation of genes encoding for enzymes of part II of the biotransformation, such as NADPH:quinone oxidoreductase, GST, and UDP-glucuronosyltransferase (UGT) (Mimura and Fujii-Kuriyama, 2003). Furthermore, there is evidence that the AHR plays an important role in cell cycle regulation, apoptosis and in early embryonic development (Ma and Whitlock, 1996; Nebert et al., 2000; Peters and Wiley, 1995). The inactive form of the AHR is located in the cytoplasm, where it occurs in a complex with heat-shock proteins (hsp90) and a small co-chaperon protein termed p23 (Kazlauskas et al., 1999, 2001). Upon ligand binding, the complex dissociates and AHR translocates into the nucleus, where it forms a heterodimer with the AHR nuclear translocator (ARNT), also known as the hypoxia inducible factor (HIF-1ß) (Hankinson et al., 1991). This heterodimer binds to a DNA region called the xenobiotic response element (XRE, also the AHR-mediated aromatic hydrocarbon response element (AHRE) or dioxin response element (DRE)), which is located in close proximity to genes encoding for the cytochrome enzymes and other enzymes of the biotransformation. Furthermore, the AHR–AHRE complex is responsible for transcriptional regulation of the aryl hydrocarbon repressor (AhRR), which inactivates the complex and therefore serves as a negative feedback mechanism (Mimura and Fujii-Kuriyama, 2003). Ligands for AHR are environmental toxins, including polycyclic aromatic hydrocarbons (PAHs), polyhalogenated biphenyls and dioxins, such as the carcinogen 2,3,7,8-tetrachlorodibenzo-p-dioxin (TCDD) (Whitlock, 1999).

208

Hepatotoxicity

Due to observations made in AHR-knockout mice that happen to be resistant to TCDD toxicity, it is generally accepted that the AHR is responsible for mediation of toxic, carcinogenic and teratogenic effects of its ligands (Brunnberg et al., 2006; Fernandez-Salguero et al., 1996; Gonzalez and Fernandez-Salguero, 1998; Swanson and Bradfield, 1993). The mechanisms by which TCDD toxicity is further advanced are not completely understood. There is evidence, however, that expression of CYP1A1 is a major event within the development of TCDD toxicity. This cytochrome is not only responsible for biotransformation of AHR ligands, but also exerts its biological effects via metabolism of endogenous substrates, such as arachidonic acid, oestrogens and bilirubin. It also significantly participates in the production of reactive oxygen and has effects on K+ and Ca2+ channels (Rifkind, 2006). In addition, elevated levels of the TGFß1 and TGF-ß3, accompanied by increased apoptosis rates and fibrosis, were found in the portal areas of livers from AHRnull mice (Zaher et al., 1998b), suggesting a role for AHR in the regulation of cytokine signalling and inflammation. Moreover, interaction with other nuclear transcription factors may be relevant for AHRmediated toxicity. Specifically, the AHR was shown to interact with nuclear factor kappa B (NF-κB), which is involved in several physiological and pathological responses, including immune modulation, inflammatory responses and apoptosis. It was demonstrated, for example, that TCDD triggers expression of several apoptotic genes, such as Fas ligand (FasL). This regulation is apparently induced by AHR and subsequent NF-κB activation. Studies in AHR knockout mice had revealed that AHR activation was required for regulation of the FasL promoter activity through NF-κB in thymic stromal cells (Camacho et al., 2005). In addition, crosstalk of AHR with the retinoid acid receptor (RAR), vitamin D receptor (VDR) and peroxisome proliferator-activated receptors (PPARs) has been reported (Fallone et al., 2004). Even though the complete complex of AHR-related actions needs to be unravelled, AHR is an excellent example as to how drug interactions with the cellular metabolism can be induced. 9.6.1.2

The Pregnane X Receptor (PXR)

The pregnane X receptor (PXR, also called the steroid and xenobiotic receptor (SXR) or pregnane-activated receptor (PAR)) is a member of the NR1I subfamily of nuclear receptors, which also includes the vitamin D receptor (VDR), constitutive androstane receptor (CAR) and the frog benzoate X receptors (BXRs) (Kliewer et al., 2002). The pregnane X receptor binds to elements in CYP enzyme drug-responsive enhancers and thereby induces enhanced transcription of CYP3A4 (Bertilsson et al., 1998; Blumberg et al., 1998; Lehmann et al., 1998), CYP3A7 (Pascussi et al., 1999), and the two major isoforms of human liver microsomal carboxylesterases (HC1 and HC2) (Zhu et al., 2000). Furthermore, there is evidence suggesting PXR regulation of the genes coding for CYP2C8, 2C9 and CYP3A11 (Pascussi et al., 2000c; Staudinger et al., 2001a). Interestingly, many target genes of PXR are substantially involved in xenobiotic metabolism and bile acid metabolism (see also Table 9.1). Furthermore, PXR has been implicated in the pathogenesis of cholestatic liver injury and the development of drug-induced hepatomegaly due to its regulatory function in bile acid biosynthesis (cholesterol 7α-hydroxylase) as well as its effect on bile acid and xenobiotic-metabolizing (CYP3A) and -transporting (Oatp2) proteins in vivo (Staudinger et al., 2001b).


209

Table 9.1 Genes regulated by PXR Feature

Inducer

Biotransformation Rifampicin, phenobarbital (Phase I) (PB), clotrimazole, 5betapregnane-3,20-dione, organochlorine, organophosphate, pyrethroid pesticides Nonylphenola

a b

References

CYP2B6

Lemaire et al., 2004; Xie et al., 2000

CYP2B9

Hernandez et al., 2006 Ferguson et al., 2005 Chen et al., 2004b; Gerbal-Chaloin et al., 2002 Chen et al., 2003; Gerbal-Chaloin et al., 2001 Durr et al., 2000; Pascussi et al., 2001 Pascussi et al., 1999 Falkner et al., 2001

Dexamethasone Rifampicin, hyperforin, phenobarbital

CYP2C8 CYP2C9

Rifampicin

CYP2C19

Dexamethason, hyperforin

CYP3A4

Rifampicin Biotransformation RU38486b (Phase II) Pregnenolone-16 alpha-carbonitrile (PCN) Flavonoids, chrysin, rifampicin

Hepatic transporters

Gene

CYP3A7 GluthathioneS-tranferase Sulfotransferase

PCN

UDPglucoronosyltransferase Carboxylesterases

PCN, spironolactone

Oatp2

PCN, spironolactone 2-Acetylaminofluorene, rifamipicin, PCN, spironolacton

Mrp3 MRP2

Runge-Morris et al., 1999 Madhu and Klaassen, 1991; Sugatani et al., 2004 Hosokawa et al., 1993 Cheng et al., 2005; Staudinger et al., 2001a Maher et al., 2005 Anapolsky et al., 2006; Johnson and Klaassen, 2002; Kast et al., 2002

Environmental oestrogen. Glucocorticoid receptor (GR) antagonist.

Several substances are known to activate PXR, such as steroid hormones and their metabolites, including progesteron, oestrogen and corticosterone, 5ß-pregnane and androsterol, as well as a variety of structurally varying drugs, such as calcium channel inhibitors, statins, antidiabetic drugs, human immunodefiency virus protease inhibitors and anticancer drugs (Blumberg et al., 1998; Drocourt et al., 2001; Dussault et al., 2001; Goodwin et al., 2002; Jones et al., 2000; Kliewer et al., 1998; Kliewer and Willson, 2002; Liddle and Goodwin, 2002; Synold et al., 2001). One of the most potent inhibitors of PXR is hyperforin, the active component of St. John’s wort (Moore et al., 2000; Wentworth et al., 2000).

210

Hepatotoxicity

In contrast to AHR and CAR, PXR is located exclusively in the nucleus, where it forms a heterodimer with the 9-cis retinoic acid receptors (RXR; NR2B) after ligand activation (Sueyoshi and Negishi, 2001). Notably, PXR is activated by pregnenolone 16 α-carbonitrile (PCN), a glucocorticoid receptor antagonist that induces expression of the CYP3A family of steroid hydroxylases and modulates sterol and bile acid biosynthesis in vivo. Treatment with PCN caused hepatomegaly due to induction of hyperplasia and hypertrophy in wildtype mice. This effect was completely missing in PXR-null mice (Staudinger et al., 2001a), suggesting an involvement of PXR in the development of drug-induced hepatomegaly. Expression of PXR itself was shown to be stimulated by the glucocorticoid receptor (GR) and is repressed during the acute phase reaction by interleukin-6 (Beigneux et al., 2002; Jover et al., 2002; Pascussi et al., 2000c, 2001). Moreover, PXR is known to interact with other nuclear receptors, such as the hepatic nuclear factor 4 α (HNF-4α). This liver-specific transcription factor is essential for the embryonic development and differentiation of the liver (Watt et al., 2003). By binding to the PXR promoter, HNF-4α regulates transcription of PXR and seems to be involved in a general response of CYP450 (Kamiya et al., 2003; Tirona et al., 2003). In addition, it has recently been demonstrated that PXR is activated by different bile acids, such as secondary bile acid lithocholic acid (LCA) and its 3-keto metabolite. Based on this finding, it has been proposed that cytochrome induction via PXR may protect the liver from bile acid toxicity by enhancing degradation through CYP3A4 (Staudinger et al., 2001b). In fact, this was supported by elevated levels of hydroxylated bile acid products of CYP3A4 found in patients suffering from cholestasis (Araya and Wikvall, 1999; Bremmelgaard and Sjovall, 1979). This information may help to explain why treatment with rifampicin and ursodesoxycholic acid had relieved patients from pruritus associated with intrahepatic cholestasis. In this case, these drugs had caused an enhanced clearance of bile acids by therapeutic induction of CYP3A4 (Bachs et al., 1989; Cancado et al., 1998; Gillespie and Vickers, 1993). 9.6.1.3

The Constitutive Androstane Receptor (CAR)

The constitutive androstane receptor (CAR) is located in the cytoplasm and forms a heterodimer by binding with the retinoid X receptor (RXR) upon activation. This nuclear receptor controls genes involved in Phase I (CYP2B, CYP2C, CYP3A) and Phase II (UGT1A1) of biotransformation and transporters for xenobiotics and bile acids (SLC21A6, MRP2) (Pascussi et al., 2003b). Accordingly, CAR is activated by xenobiotics such as phenobarbital and by endogenous compounds such as bilirubin metabolite(s). Activation of CAR is mediated directly by ligands, such as the phenobarbital-like CYP inducer 1,4-bis [2-(3,5-dichloropyridyloxy)]benzene (TCPONOP) (Tzameli et al., 2000). Upon stimulation, CAR undergoes translocation into the nucleus, where its activity is additionally controlled by protein phosphorylation events (Handshin and Meyer, 2003). The transcriptional regulation of CAR itself was shown to be mediated by the glucocorticoid receptor (GR), which acts via the glucocorticoid responsive element (Pascussi et al., 2000b, 2003a). Thus, it was demonstrated that pro-inflammatory cytokines, such as IL-6 or IL1β, by inducing AP-1 or NF-κB activation and thereby leading to GR inhibition, would inhibit CAR gene expression and phenobarbital-mediated CYP gene expression in human hepatocytes (Pascussi et al., 2003b). Taken together, these findings reveal a fundamental


211

link between processes that activate the glucocorticoid receptor, such as inflammatory reactions, and the detoxifying capacity of hepatocytes. CAR has been implicated in liver toxicity induced by acetaminophen (APAP). Here, pretreatment with the CYP-inducers ethanol (isoforms of CYP2E and 3A) (Kostrubsky et al., 1997; Sinclair et al., 1998) and phenobarbital (PB) (isoforms of CYP2B and 3A) (Burk et al., 1990; Pirotte, 1984) had increased toxicity of this antipyretic and pain-relieving drug. The influence of CAR, which is known to mediate PB effects, was assessed by application of toxic APAP doses to CAR knockout mice. These mice were found to be resistant against APAP toxicity, which was indicated by the absence of hepatocellular damage or GSH depletion that was found in wild-type mice (Zhang et al., 2002). 9.6.1.4

The Farnesoid X Receptor (FXR)

The farnesoid X receptor (FXR) belongs to the superfamily of RXR (retinoid X receptor) heterodimers, such as CAR, LXR, PXR and the PPARs. The FXR nuclear transcription factor is activated by bile acids and is located in the liver, kidney, adrenal glands and intestine (Forman et al., 1995). Together with the Liver X receptor (LXR), FXR is an essential regulatory component of cholesterol homeostasis, in which bile acids serve as ligands. The role of FXR has been emphasized particularly in processes involved in the development of cholestasis. Abrogation of alterations and restoring FXR activity by synthetic antagonists was demonstrated to be a promising approach to encountering liver toxicity by ANIT (Liu et al., 2003) and may have introduced a new paradigm for the treatment of hepatotoxicity. 9.6.1.5

The Peroxisome Proliferator-Activated Receptors (PPARs)

The peroxisome proliferator-activated receptors (PPARs) were named for their ability to induce hepatic peroxisome proliferation in response to xenobiotic stimuli. The three human PPAR isoforms (PPARalpha, PPARdelta and PPARgamma) are believed to play a central role in regulating carbohydrate and lipid metabolism. In addition, the PPARs are assumed to possess anti-inflammatory activity. These are exerted by inhibiting the induction of pro-inflammatory cytokines, as well as by stimulating the production of anti-inflammatory molecules (Kostadinova et al., 2005). Xenobiotics, fatty acids and eicosanoids are ligands for this nuclear receptor family (Mehendale, 2000). After binding to the retinoid X receptor, PPARs enfold their transcriptional regulation by binding to specific DNA regions, termed peroxisome proliferator response elements (PPREs), which are found in flanking regions upstream of the target gene sequences (Kane et al., 2006). Among the PPAR target genes are carnitine palimitoyl transferase I, HMG coA synthase 2, fatty acid ω-hydroxylase and apoA-1, all of which are key enzymes of the lipid metabolism. In addition, PPARs participate in the regulation of important genes in cell growth, proliferation and tumour cell aberration, such as c-myc, c-Ha-ras, fos, jun and egr-1 (Corton et al., 2000; Vanden Heuvel, 1999). In the response to xenobiotics, PPARs are believed to play a central role: specifically, they are involved in detoxification, survival and regeneration processes that are activated by xenobiotics. For instance, PPARα induces increased expression of several detoxifying enzymes, such as 7β-hydroxysteroid dehydrogenase type 11 and glutathione S-transferase l1 (GST’1) (Motojima, 2004). This has been connected to its ability to metabolize possible toxic plant compounds. For example, it was demonstrated that feeding PPARα knockout

212

Hepatotoxicity

mice with sesame caused lethal hepatotoxicity, while in wild-type mice sesame strongly induced Cyp2c29, 2c38 and 2b9 in the intestine and liver (Motojima and Hirai, 2006). The human CYPs corresponding to these mouse CYPs, however, which are CYP2C9, 2B6 and 3A4, are regulated by the constitutive androstane receptor (CAR) (Pascussi et al., 2003c). This finding suggests a possible crosstalk and overlapping of transcriptional programs between PPARα and CAR, such as suggested for the axis of PPARa-RXR-LXR and demonstrated for nuclear vitamin D(3) receptor (VDR), which represses the transcriptional activity of PPARα (Anderson et al., 2004; Motojima and Hirai, 2006; Sakuma et al., 2003). Another member of the PPAR family is the lipid-activated transcription factor PPARγ , which was shown to directly regulate the functional expression of drug transporters, such as the ABCG2 transporter. This member of the ATP-binding cassette transporters is known to perform clearance of endogenous and exogenous toxic agents. Activation of PPARγ and consecutively increased amounts of the ABCG2 transporter protein were shown to significantly increase efflux of xenobiotics in human dendritic cells (Szatmari et al., 2006). In the liver, PPARα has been found to be highly expressed (Motojima and Hirai, 2006). In addition to its function within the lipid metabolism, PPARα was suggested to strongly influence the response to xenobiotics. Specifically, this nuclear receptor has been recognized as the primary site of action for fibrates (Kane et al., 2006), which represent a class of lipid-lowering drugs that have been linked to cases of clinically severe rhabdomyolysis occurring under co-administration with HMG-CoA reductase inhibitors (the statins). High LDL-cholesterol and triglyceride levels represent a major risk factor for cardiovascular diseases. For this reason, this drug–drug interaction has been especially unfortunate, since a combination of these two classes of lipid-lowering drugs had promised to expand the efficacy of lipid-lowering therapy to a new level. There is evidence that these interactions might arise from increasing plasma levels of statins, observed under co-administration with fibrates, in particular gemfibrozil (Kyrklund et al., 2003). Note that clofibrate itself had been connected to clinical cases of liver toxicity (Keller et al., 1990, 1992). It has been proposed that peroxisomal proliferators might cause local toxicity due to inhibition of mitochondrial oxygen uptake. Specifically, this has been demonstrated for the peroxisomal proliferators clofibrate, aspirin, valproate, ethylhexanol, ciprofibrate and perfluorooctanoate (Keller et al., 1992). Interestingly, clofibrate administration was also found to exert a protective effect on liver toxicity induced by acetaminophen (APAP) (Nicholls-Grzemski et al., 1992). Initially, investigations revealed that clofibrate increased the hepatic glutathione (GSH) content and led to the suggestion that enhanced detoxification of APAP by covalent binding to GSH may diminish toxic effects of APAP (Nicholls-Grzemski et al., 2000). New evidence, however, implies that an activation at PPARα might be responsible for this protective effect of peroxisome proliferators, such as gemfibrozil (Mehendale, 2000; Nguyen et al., 1999). The importance of PPARα activation for the protection of peroxisome proliferators from liver injury through APAP have been supported by the finding that gemfibrozil administration in PPARα knockout mice did not result in any protection from liver injury (Chen et al., 2000). Several possible mechanisms by which PPARα activation might protect from liver toxicity have been discussed. These include increased protection against oxidative cell damage by increasing antioxidant levels or induction of oxyradical-degrading enzymes, preventing cell – death through induction of resistance or induction of cell-death inhibitors, as well as stimulation of cell proliferation through responsive mitogenic genes


213

(Roberts et al., 1998). The latter might facilitate an enhanced recovery and regeneration from early and reversible liver injury (Soni and Mehendale, 1998). Moreover, it was demonstrated that activation of NF-κB, which is important for induction of inflammation, was reduced by activation of PPARα (Huang et al., 2006). The anti-inflammatory effect of PPARα seems to be dependent on suppression of the DNA-binding activity of NF-κB and consecutive downregulation of TNF-α expression (Ye et al., 2004). Thus, overexpression and activation of PPARα could be one of the first steps leading to a survival pathway that is activated by certain xenobiotics. PPARγ has been implicated in liver toxicity mediated by the insulin-senstizing drug troglitazone, which was approved in 1997 for treatment of Type II diabetes. After initial success, this treatment course was observed to be associated with frequent cases of severe hepatotoxicity that finally led to withdrawal of troglitazone from the market in 2000 (Gale, 2001). No significant elevation of transaminases in pre-clinical trials had predicted the toxicity associated with an extraordinarily increased risk of acute liver failure later observed in patients (Graham et al., 2003). Troglitazone and other glitazones have been recognized ligands of PPARγ and were also shown to selectively induce PPARγ in the liver of diabetic rodents. Furthermore, upon treatment with troglitazone, diabetic mice developed severe microvesicular periacinar steatosis (Bedoucha et al., 2001; Boelsterli and Bedoucha, 2002). The therapeutic effect of glitazones and other PPAR agonists, such as the lipid-lowering drug fenofibrate, however, have been clearly related to activity at the PPARs (Harano et al., 2006; Memon et al., 2000). Specifically, PPARγ activates a number of genes that lead to an enhanced uptake of glucose and lipids, increase glucose oxidation and decrease both free fatty acid concentrations and insulin resistance (Dumasia et al., 2005; Way et al., 2001). Troglitazone was demonstrated to induce a coordinated stimulation of fatty acid uptake, oxidation and oxidative phosphorylation in adipose tissue and muscle of diabetic patients (Boden et al., 2005). In the liver, PPARγ activation was shown to coordinately decrease the expression of genes involved in gluconeogenesis and increase levels of genes involved in fatty acid turnover (Kersten, 2002). It has been proposed that the therapeutic effects of PPAR agonists were due to a reversion of a dysfunction in the lipid metabolism found in diabetes and obesity (Cha et al., 2005). Indeed, obese and diabetic mice inherently display upregulation of PPARy in the liver, which was hypothesized to be responsible for an increased susceptibility towards PPARγ -activating drugs, such as troglitazone (Bedoucha et al., 2001; Memon et al., 2000). There is evidence, however, for involvement of several different mechanisms of liver toxicity induced by troglitazone. For instance, biological activation and generation of electrophilic metabolites associated with consecutive GSH depletion were observed upon treatment with this drug (Smith, 2003). The key role of peroxisome proliferator-activated receptors (PPARs) in the regulation of the lipid and glucose metabolisms, however, has motivated to target different metabolic alterations by pharmacologically activating these receptors. Dys- and hyperlipidaemia are common risk factors for cardiovascular diseases and also characterize the metabolic syndrome, which carries a high cardiovascular risk and destines the majority of patients to develop Type 2 diabetes, along with all its related effects and malignant secondary diseases.

214

Hepatotoxicity

Notably, even though it displays a relatively low intrinsic activity, the pan PPAR receptor agonist bezafibrate significantly reduces triglycerides and improves insulin resistance, thereby reducing the cardiovascular risk in patients (Tenenbaum et al., 2005). The dual PPAR agonists for PPAR α and δ isoforms, the glitazars, have been tested in advanced phases of clinical studies. Targeting of dyslipidaemia via activation of PPAR α and simultaneously improving the insulin resistance via PPAR δ is expected to considerably advance the life expectancy of Type 2 diabetes patients and patients with metabolic syndrome by reducing their cardiovascular risk (Fievet et al., 2006). As revealed in recent studies, however, dual PPAR agonists are related to toxicity and currently unacceptable adverse reactions (Fievet et al., 2006; Nissen et al., 2005). Therefore, further research regarding the exact functions of PPARs in different organs is a necessity to explore and realize the potential of PPAR agonists to provide benefit to this particular patient cohort. In general, it can be concluded that PPARs play a central role in the xenobiotic response of the liver, where they accomplish distinctive regulatory tasks in different cell types. In chronic liver injury induced by carbon tetrachloride, for instance, a downregulation of PPARα and PPARγ expression could be observed in hepatocytes, while increased levels of these transcription factors were found in Kupffer cells associated with inverse correlation to levels of activated NF-κB (Orfila et al., 2005). Furthermore, activation of PPARα has been strongly correlated with the induction of chemically induced carcinogenesis, as for example by di(2-ethylhexyl)phthalate (DEHP) (Corton et al., 2000). The hypothesis put forward states that mitogenic stimulation and suppression of apoptosis in preneoplastic cells due to undue activation of PPARα might contribute to carcinogenic processes (Melnick, 2001). 9.6.1.6

Liver-Enriched Transcription Factors

Regulation of the expression of liver-specific genes has been primarily attributed to the transcriptional regulation that is mediated by interaction of trans-acting liver-enriched transcription factors with cis-regulatory elements of DNA that share liver-specific motifs. These liver-enriched transcription factors (LETFs) are responsible for maintenance of the tissue-specific phenotype, differentiation and are important for liver regeneration in toxic liver injury (Schrem et al., 2002, 2004). Six families of liver-enriched transcription factors have been characterized so far, including the hepatocyte nuclear factors -1 (HNF-1), -3 (HNF-3, now termed FOXA − (forkhead box subclass A) proteins), -4 (HNF-4), -6 (HNF-6), the basic region leucine zipper (bZIP) family members CCAAT/enhancer-binding proteins (C/EBPα, β, δ and γ ) and the D-binding protein. The hepatocyte nuclear factors have been identified to be essential for regulation processes of differentiation and metabolism (Duncan, 1998). HNF-1 is formed by the heterodimer HNF-1 α and the homodimer HNF-1 β and constitutes a major transactivator of the albumin transcription. It mediates regulation of the glucose-6-phosphatase gene transcription by insulin and glucocorticoids (Streeper et al., 1998). Moreover, HNF-1 is believed to be involved in the downregulation of protein expression upon increased oncotic pressure (Pietrangelo and Shafritz, 1994). The HNF-3 or FOXA subfamily consists of three proteins, namely HNF-3 α, β, and γ , now FOXA3, which play a role in developmental differentiation and were found to decrease


215

dramatically in acute phase livers (Jacob et al., 1994). Like HNF-3, the HNF-4 subfamily is composed of three different proteins (α, β, γ ) that bind to fatty acyl-CoA thioesters as ligands (Hertz, 1998). Their potential to transactivate liver-specific genes is dependent on phosphorylation of tyrosine residues. The hepatic nuclear factor-6 (HNF-6) is expressed in liver, pancreas and neurones during early stages of differentiation, implicating a central role in the developmental programs of these tissues (Landry et al., 1997). Expression of HNF-6 can be regulated by growth hormone (GH) mediated by transcriptional regulation of the HNF-6 gene by STAT5 (signal transducer and activator of transcription 5) and HNF-4 (Lahuna et al., 1997, 2000; Rastegar et al., 2000). The hepatic nuclear transcription factors were shown to essentially contribute to the metabolic competence that is specific and unique to the liver. Transfection with HNF-4 cDNA, for instance, was demonstrated to induce activation of liver-specific genes in a dedifferentiated rat hepatoma cell line that displayed a loss of function of HNF-1 and HNF-4 (Deschatrette and Weiss, 1974; Faust et al., 1994; Spath and Weiss, 1997). The CCAAT/enhancer binding proteins or C/EBP transcription factors belong to the family of bZip transcription factors, such as c-jun and c-fos (AP-1). The target-specific DNA enhancer elements are characterized by a complex network of interactions with different transcription factors. C/EBPs have regulatory activities in cell cycle control and development and are also involved in the regulation of biotransformation, as e.g. together with HNF-1 α in control of the human cholesterol 7α-hydroxylase CYP7A1 gene (Antes et al., 2000). Depression of nuclear transcription factors has been implicated to represent an important mechanism in toxicant-induced liver injury. For example, reduced expression of HNF-3 γ was observed upon exposure to carbon tetrachloride (Nakamura et al., 1999). Moreover, the liver-enriched transcription factor HNF-3 β and the C/EBP family are modulated by cytokines. IL-1, IL-6 and IFN-γ were found to exert different effects on the LETFs (Samadani et al., 1995). This is a mechanism that could explain the metabolic response of the liver to inflammatory reaction, such as the protein synthesis switch with pronounced synthesis of acute phase proteins and decreased albumin expression observed during the acute phase response of the liver (Whalen et al., 2004). Indeed, TNF-α treatment in primary mouse hepatocytes as well as TNF-α overexpression was found to induce translocation of C/EBP-β by phosphorylation, which resulted in abrogated transcription of the albumin gene (Buck et al., 2001a). A lack of TNF-α signalling via TNFR-1 in carbon tetrachloride toxicity, as investigated in transgenic TNFR-1-knockout mice, was associated with a decreased DNA binding activity of NF-κB and STAT3, reduced plasma levels of (IL)-6 and liver IL-6 mRNA and significantly reduced DNA binding activity of C/EBP, resulting in a reduced regeneration ability of the liver in these TNFR-1-knockout mice as compared to the wild-type controls (Yamada and Fausto, 1998). Furthermore, activity of C/EBP-β was found to be responsible for the stellate cell proliferation observed in response to administration of the hepatotoxic substance carbon tetrachloride. This was supported by the finding that transgenic animal models presenting a lack in C/EBP-β activity ((C/EBP-β −/−) knockout and nonphosphorylatable mutants) displayed stellate cell apoptosis instead of proliferation upon carbon tetrachloride administration.

216

Hepatotoxicity

The antiapoptotic effect of C/EBP-β activity was furthermore assigned to inhibition of procaspases 1 and 8 (Buck et al., 2001b). A more recent report in which C/EBP-β transfection of cultured hepatocytes resulted in increased DNA synthesis rates and prolonged cell viability in culture had emphasized the role of the C/EBP-β activity for hepatocellular survival and liver cell regeneration in toxic liver injury. This was documented, in addition, by introduction of the C/EBP-β gene into liver-damaged mice, which hitherto had displayed significantly suppressed serum AST and ALT activities (Isoda et al., 2005). Taken these findings together, it can be concluded that activation of liver-enriched nuclear transcription factors by inflammatory cytokines and probably also by suitable liver toxins may be an important mechanism in liver cell defence and survival in toxic injury. It should be noted that an inflammatory environment is characterized by increased generation of reactive nitrogen and oxygen species and thereby exerts enhanced cellular stress and involves a higher risk of mutagenesis (Iwai et al., 2002). Hence, activation of C/EBP-β and its antiapoptotic effect seem to be predestined to contribute to CCl4 -induced carcinogenesis as well. The transcription factor C/EBP-ξ (also known as GADD153 or CHOP-10) acts as a regulatory control of the C/EBPs. It encodes a C/EBP protein without a functional DNA binding domain which, by dimerization with other C/EBPs, is able to deactivate their DNA-binding activity (Carlson et al., 1993). C/EBP-ξ is induced by mediation of AP-1 and p38 mitogen activated kinase (MAPkinase) upon exposure to DNA-damaging agents, such as methyl methanesulfonate, hydrogen peroxide and UV irradiation (Guyton et al., 1996; Luethy et al., 1990; Wang and Ron, 1996). 9.6.1.7

Vitamin D Receptor

Even though the vitamin D receptor (VDR) is found only rarely in the liver under physiological conditions, effects on liver physiology and drug metabolism exerted by vitamin D3 have been frequently observed. Specifically, VDR has been discovered to be an important sensor of bile acids and is able to prevent toxicity by regulating bile acid metabolism (Makishima et al., 2002). VDR is activated by the hepatotoxic secondary bile acid lithocholic acid (LCA) which is also known to be a potential carcinogen (Nagengast et al., 1995; Ogawa et al., 1998). In association with RXR, the VDR–RXR heterodimer interacts with a response element of the human major cytochrome enzyme CYP3A4 gene. This leads to transcriptional upregulation, enabling CYP3A4 to catabolize LCA (Araya and Wikvall, 1999; Xie et al., 2001). Under this aspect, VDR resembles the function of the evolutionarily related nuclear receptors PXR, FXR and LXR, which are also representatives of the cellular bile acid sensors. In addition, VDR has been proposed to be an active member of the nuclear receptor crosstalk with constitutive androstane receptor and peroxisome proliferator-activated receptor α (Makinen et al., 2002; Sakuma et al., 2003). Moreover, the metabolisms of phenobarbital and vitamin D3 were long known to be connected in some way, indicating other significant functions of VDR for the xenobiotic metabolism. This has been discovered through observations made on the protective effect of phenobarbital against intoxication with vitamin D3 (Gascon-Barre and Cote, 1978).


217

As of today, it has been demonstrated that normal nonparenchymal liver cells, such as sinusoidal endothelial, Kupffer and stellate cells, very well express significant amounts of VDR (Gascon-Barre et al., 2003). Nuclear receptors (NR) in liver toxicity in summary: – – – – –

Induction and repression (directly, secondary, e.g. due to inflammation) Induction of apoptosis Disturbance of cholesterol and bile acid homeostasis Reduced activity – reduced biotransformation and accumulation of possible toxins Interference with NR or NR crosstalk in vital cellular functions→reduced metabolic competence

9.6.2

Biotransformation and CYP Induction

Inhibition or induction of enzymes involved in detoxification have a high impact on drug effects and the toxicity of certain substances. Biologic detoxification is located in the liver and is responsible for inactivation and successful elimination of endogenous and exogenous metabolites and toxicants. The metabolic reactions responsible for detoxification comprise Phase I and Phase II reactions. Phase I reactions increase the polarity of the xenobiotics by inserting new functional groups on the drug molecule. Major enzymes performing Phase I reactions are hydrolases and oxidoreductases, such as the cytochrome P450-dependent monooxygenases (CYPs) or the cyclooxygenases. Conjugation to endogenous hydrophilic molecules in Phase II reactions result in potent increases in polarity and water solubility, thereby achieving proper drug elimination. Such reactions are glucuronidation, sulfatation and conjugation to amino acids mediated by enzymes of Phase II of the biotransformation, such as UDP-glucose dehydrogenase (UGDH) and members of the UDP glucuronosyl transferase (UGT) family. Phase III describes transport processes mediated by transmembranous transporter proteins, such as the organic anion transporting polypeptide family (Oatps) along with the organic cation transporter 1 (Oct1) and organic anion transporter 2, which mediate uptake of a large number of xenobiotics from blood into liver, and the multidrug resistance proteins (MDRs), including p-glycoprotein (P-gp) or the multidrug resistance-associated protein (MRP), which mediate the efflux of xenobiotics from the liver into bile or blood (Haimeur et al., 2004; Smith et al., 2005; Wakabayashi et al., 2006). There are major ways in which idiosyncratic toxicity arises from interaction at metabolizing enzymes: one possibility is enzyme inhibition while the other one is enzyme induction. Drug–drug interactions at the site of biotransformation I enzymes are commonly recognized to cause severe clinical complications, such as the ventricular arrhythmia observed upon co-administration of CYP3A4 inhibitors, including terfenadine, astemizole, cisapride or pimozide. Another severe complication is rhabdomyolysis, which has been associated with the co-administration of 3-hydroxy-3-methylglutaryl-coenzyme A reductase inhibitors (‘statins’) and CYP3A4 inhibitors (Dresser et al., 2000). The toxicity of these interactions can be attributed to increasing concentrations of the drug or its activated metabolites due

218

Hepatotoxicity

to inhibition of their metabolizing enzymes. On the contrary, upregulation of metabolizing enzymes has been associated with organ toxicity, in particular affecting the liver. Enhanced transcription and expression of detoxifying enzymes in the presence of suitable substrates can be considered as a natural adaptive response. Thus, increased substrate supply provokes an increased metabolism rate and results in an enhanced clearance of the particular substance. In regard to CYP monooxygenase reactions, five different types of xenobiotic inducers can be distinguished, named after drugs that induce cytochrome monooxygenases of a typical pattern. These are the phenobarbital-type-inducing CYP2B1, CYP2Cs and CYP3As, the rifampicin-dexamethasone type (CYP3As, CYP2Cs, CYP2B1), the ethanol, isoniazide type (CYP2E1), the clofibrate type (CYP4As) and the polycyclic aromatic hydrocarbons (TCDD) type (CYP1A1, CYP1A2, CYP1B1) (Handschin and Meyer, 2003). Clinically important are, in particular, the inhibitors of cytochrome CYP3A4, such as the fungicides itraconazole and ketoconazole, the antibiotics clarithromycin and erythromycin, the 5HT2A serotonin receptor antagonist and antidepressive drug nefazodone, the proteinase inhibitor ritonavir and grapefruit juice (Dresser et al., 2000). The transcriptional regulation and hence induction of detoxifying Phase I, Phase II and Phase III enzymes has been attributed to the activation of ligand-activated nuclear receptors (Klaassen and Slitt, 2005; Xie et al., 2004). Regulation of the different isoforms of cytochrome-dependent monoxygenases, for example, has been attributed to activation of the nuclear aryl hydrocarbon receptor (AHR: CYP1A1, CYP1A2, CYP1B1) (Quattrochi et al., 1994; Tang et al., 1999; Whitlock, 1999), the pregnane X receptor (PXR: CYP3A4) (Bertilsson et al., 1998; Blumberg et al., 1998; Lehmann et al., 1998), CYP3A7 (Pascussi et al., 1999), the constitutive androstane receptor (CAR: CYP2B6) (Sueyoshi et al., 1999), CYP3A4 (Savkur et al., 2003), CYP2C8 (Ferguson et al., 2005), 2C9 (Al Dosari et al., 2006) and the glucocorticoid receptor (GR: CYP3A5) (Schuetz et al., 1996). Recently, the nuclear PPARα receptor has been found to be involved in the negative regulation of normal and rifampicin-mediated UGDH gene expression (Vatsyayan et al., 2005). UDP-glucose dehydrogenase (UGDH) (EC1.1.1.22) catalyses the NADPH-dependent oxidation of UDP–glucose to UDP-glucuronic acid and is therefore essential for Phase II reactions of the biotransformation. Overexpression of Phase III enzymes, a common resistance strategy found in cancer cells, has recently been demonstrated to be part of the response to chemicals that also increase expression of Phases I and II biotransformation enzymes in the liver, resulting in an enhanced hepatic uptake and biliary excretion due to increased expression of various transporters, e.g. Oatps, MRPs and MDRs via activation of nuclear receptors (Klasssen and Slitt, 2005). The nuclear receptor CAR, for example, was demonstrated to be involved in the regulation of multidrug resistance transporter type 3 (MDR3) (Cherrington et al., 2002), while expression of MDR1 (ABCB1), MRP1 (ABCC1), MRP2 (ABCC2) and the breast cancer resistance protein BCRP (ABCG2) was positively correlated with pregnane x receptor activity (Albermann et al., 2005). Even though the central role of the nuclear receptor superfamily in transcriptional regulation of genes coding for detoxifying enzymes has become apparent, the way in which this response is induced by xenobiotics is not sufficiently elucidated. While the pregnane X receptor (PXR) and constitutive androstane receptor (CAR) have been demonstrated to directly interact with their xenobiotic ligands, other nuclear receptors strongly involved in the induction of detoxifying enzymes are activated by ligands of endogenous origin, such


219

as fatty acids and eicanosoids serving as ligands for the peroxisome proliferator-activated receptors (PPARs). In addition, there is evidence for an overlap of transcriptional programmes of different nuclear receptors, such as between PPARα and CAR, LXR or the nuclear vitamin D(3) receptor (VDR) (Anderson et al., 2004; Motojima and Hirai, 2006; Sakuma et al., 2003). The current understanding of the transcriptional regulation of enzymes involved in biotransformation is based on a concept of transcription factor networks that regulate basal and xenobiotic-modulated expression (Pascussi et al., 2003c). This crosstalk between nuclear receptors has been demonstrated for several members of these ligand-activated transcription factors, e.g. for PPARα, RXR and LXR (Anderson et al., 2004). 9.6.2.1

How Does Upregulation of Metabolizing Enzymes Cause Liver Toxicity?

Two major hypotheses have been put forward to explain the mechanisms by which upregulation of detoxifying enzymes can cause hepatotoxicity. The first hypothesis involves generation of bioactive metabolites by certain detoxifying enzymes, such as observed in the oxidative group transfer mediated by cytochrome P450dependent monooxygenases to the narcotic halothane. In this case, activation to the active metabolite trifluoroacetylchloride can cause severe immune hepatitis. Another important example described later in detail is the activation of acetaminophen to the hepatotoxic metabolite N-acetylbenzoquinoneimine (NAPQI) by dehydrogenation of cytochrome P450dependent monooxygenases. In addition, biotransformation can lead to formation of toxic species, such as, for instance, observed in dechlorination of chloroform to phosgene. This toxic metabolite forms a specific adduct with a mitochondrial phospholipids (Di Consiglio et al., 2001) that has been proposed to cause a progressing mitochondrial damage with swelling of the organelles and formation of megamitochondria (Guastadisegni et al., 1999), presumed to induce cell death by activation of apoptosis (Hartig et al., 2005). The second hypothesis is based on the undue generation of reactive oxygen species during biotransformation. This, for instance, takes place during the one-electron quinone reduction via flavoprotein P450-reductase with production of superoxide ions, as is involved in the metabolic activation of paraquat or doxorubicin (Kukielka and Cederbaum, 1990; Ravi and Das, 2004). In addition, dehalogenation can exert reactive radicals that cause toxic reactions, as is observed during biotransformation of carbon tetrachloride formation (Connor et al., 1986). 9.6.2.2

Production of Toxic Metabolites in Hepatotoxicity

An important example of liver toxicity induced by an active metabolite generated in the biotransformation process is acetaminophen. Acetaminophen (APAP) is an analgetic and antipyretic drug that is ubiquitously administered. Overdosing of acetaminophen causes severe hepatotoxicity that is associated with centrilobular necrosis (James et al., 2003a). The active metabolite of APAP is N-acetyl-p-benzoquinone imine (NAPQI), generated in step I of the biotransformation by CYP2E1, CYP1A2 and CYP3A4 (Raucy et al., 1989; Thummel et al., 1993). This bioactive metabolite can interact at different levels and sites of the cellular metabolism and has been associated with affection of transcription factors and gene expression, protein expression and activity.

220

Hepatotoxicity

The presence of NAPQI was shown to be substantially involved in ubiquitous inhibition of several cellular enzymes, some of which are essential for physiologic metabolic processes. Among these is, for instance, glyceraldehyde-3-phosphate dehydrogenase (GAPDH), which transforms the NAD+ -dependent phosphorylation of glyceraldehyde-3-phosphate to 1,3 bisphoglycerate within glycolysis (Andersson et al., 1990; Burcham and Harman, 1991), or γ -glutamylcysteinyl synthetase, which is the rate-limiting step of glutathione (GSH) synthesis, essential for detoxification of oxidative radicals and toxins such as APAP (Kitteringham et al., 2000). Moreover, NAPQI has been demonstrated to be an effective inducer of cytochrome enzymes, e.g. CYP2E1, which has been associated with increased production of reactive oxygen species (ROS) such as superoxide radicals and H2 O2 (Johansson et al., 1988; Nordmann et al., 1992; Rashba-Step et al., 1993; Wu and Cederbaum, 1994). ROS can activate caspases by direct stimulation of the mitochondrial membrane transition pore and cytochrome c release and also controls the stability of p53, which acts in surveillance of cell integrity (Jaeschke et al., 2003). In addition, oxidative stress was shown to be associated with degradation of actin filaments and dysregulation of Ca2+ homeostasis, which is observed early in APAP-induced hepatotoxicity (Tsokos-Kuhn et al., 1988). Accordingly, it has been proposed that the hepatotoxicity of NAPQI is at least in part due to production of oxidative stress resulting from a futile cycling of P450 leading to reduction of molecular oxygen (Goeptar et al., 1995). ROS is believed to be harmful in particular to APAP-treated hepatocytes, since APAP and its major metabolite NAPQI are both detoxified by conjugation with glutathione (GSH), resulting in depletion of GSH that is normally able to dispose of the oxygen radical metabolites. The connection between induction of cytochrome enzymes by NAPQI and hepatotoxicity of APAP had been confirmed by studies with knockout mice. Specifically, CYP450 1A2and 2E1-deficient mice showed a relative resistance to APAP-induced toxicity compared to wild-type mice. In these mice, a higher dose of APAP was needed to induce hepatotoxicity and an increased overall survival has been observed (Zaher et al., 1998a). The absence of the APAP-metabolizing enzymes P450 1A2 and 3A11 (corresponding to human 2A4) in mice due to a deficiency of the nuclear receptor CAR (constitutive androstane receptor) also resulted in resistance to APAP toxicity (Zhang et al., 2002). It can be presumed that the presence of the bioactive metabolite NAPQI is strongly associated with the striking features observed in APAP-induced hepatotoxicity. For example, formation of NAPQI is involved in the depletion of glutathione (GSH), formation of protein adducts and free oxygen radicals (Gibson et al., 1996; James et al., 2003a), which altogether, to an unknown weighting, contribute to the liver injury observed under overdosage of acetaminophen. The resulting metabolic derangements involve in any case the initiation of an immediate hepatocellular defence response that includes translocation of redox-sensitive transcription factors, such as Nrf-2 and NF-κB. These mediate the orchestration of the adaptive cellular response by enhanced transcription of genes encoding for antioxidant proteins and Phase II drug-metabolizing enzymes (Park et al., 2005). As of today, the mechanism of APAP-induced hepatotoxicity is certainly not completely understood. In particular, the role of the interaction between hepatocytes and other liver cell populations, such as Kupffer cells, has been addressed only in part by now. Evidence is available that the consecutive release of proinflammtory cytokines, such as TNF-α and IL-1 alpha, is associated with certain pathological manifestations of APAP-induced


221

hepatotoxicity, but studies with TNF-α-knockout mice are inconclusive, e.g. mice display a liver phenotype similar to wild-type mice treated with APAP, but also exhibit exaggerated hepatotoxicity, delayed recovery of GSH levels and less increase in Akt expression, a downstream target of PI-3K, compared to wild-type mice in response to acetaminophen (Boess et al., 1998; Chiu et al., 2003a,b). Thus, it has been concluded that TNF-α release by macrophages has no central role in acetaminophen-induced liver injury but may be responsible for activation of protective signalling pathways in hepatocytes. The antituberculostatic drug isoniazid (INH) has long been known to cause a spectrum of liver injuries, starting from an asymptomatic elevation of liver transaminases and reaching up to hepatitis or massive necrosis associated with liver failure (Mitchell et al., 1976). Bioactivation of INH and its acetyl radical is assumed to be a significant determining factor for the toxicity caused by this drug. Generation of this reactive electrophile metabolite occurs by acetylation and subsequent hydrolysation (Nelson et al., 1976). In addition, it was demonstrated that isoniazid significantly inhibited CYP2C19 and CYP3A (Desta et al., 2001). The induction of CYP2E1 observed for hydrazine, which is the major metabolite of INH, was suggested to be strongly involved in the hepatotoxicity caused by INH (Yue et al., 2004). Formation of adducts with macromolecules has been put forward as one explanation for the mechanism leading to liver injury induced by isoniazid. The previously discussed involvement of CYP2E1 and enhanced production of oxidative stress upon isoniazid treatment, however, may provide another explanation that is worth being investigated further (Attri et al., 2000; Sodhi et al., 1997). The anti-oestrogen tamoxifen is used in the second-line therapy of breast cancer. During its biotransformation, tamoxifen undergoes α-hydroxylation and sulfonation. The resulting sulfate ester is instable and breaks down to form a genotoxic metabolite, a reactive carbocation. The subsequent formation of DNA adducts in the liver is associated with an increased risk to develop liver cancer. Interestingly, this risk is excessive in rats when treated with tamoxifen, as in rats bioactivation through sulfonation is a major route as opposed to the human liver, where glucuronylation is preferential. Indeed, the glucuronide of αhydroxytamoxifen was found to have a much higher stability compared to the sulfonide (Park et al., 2005). Other clinically relevant cases in which bioactivation leads to formation of hepatotoxic metabolites have been described before, as in the case of halothane-induced immune hepatitis, where formation of antibodies causes inhibition of drug-metabolizing monooxygenases. 9.6.2.3

Knockout Models and Biotransformation in Liver Toxicity

Drug-metabolizing enzymes, in particular cytochrome monooxygenases, show a high rate of polymorphisms associated with marked differences in expression and catalytic activity throughout the population. This is of special interest, since bioactivation of hepatotoxic substances may play an important role in idiosyncratic liver toxicity. Trying to provide evidence of the role of drug-metabolizing enzymes in liver toxicity, the application of knockout and transgenic animal models has exceedingly contributed to the current knowledge of the functional spectrum of CYP450s and Phase II metabolizing enzymes in the physiology and development of toxic liver injury (Gonzalez, 2002). For the interpretation of data derived from experimental studies with knockout or transgenic animals, it is of considerable importance that knockout animals derived in different laboratories display important heterogeneity and that furthermore unintended

222

Hepatotoxicity

gene compensation may occur during the generation process. For example, generation of CYP1A2-null mice in one laboratory resulted in generation of a mouse line in which the majority of nonviable animals had died after birth displaying an immature respiratory system, while breeding the surviving animals resulted in healthy descendants, suggesting involvement of gene compensation (Liang et al., 1996; Pineau et al., 1995). Table 9.2 summarizes the recent biotransformation knockout models in liver toxicty (see Appendix). 9.6.3

Hepatotoxicity at the Organelle Level

Several toxicants have been reported to cause primary damage of certain cellular organelles and thereby develop their toxic effects. The mitochondrion is a common target of toxicity due to its proximity to ROS production at the oxidative phosphorylation complex. Mitochondrial dysfunction has been observed in APAP- and ethanol-related hepatotoxicity as well as in toxicity induced by antiretroviral nucleoside reverse transcriptase inhibitors (NRTIs) and fatty acid oxidation inhibitors (Szewczyk and Wojtczak, 2002; Vickers et al., 2006; White, 2001). Direct toxic effects on cell organelles, however, may significantly contribute to the specific phenotype and pathology of toxic liver injury. Next to mitochondria, peroxisomes and lysosomes have been identified as primary targets of liver toxicity. Thus, destabilization of lysosomes by fatty acids may contribute to the pathology of steatosis and NASH (Feldstein et al., 2004). Release of lysosomal hydrolases, in particular of cathepsin B, causes severe cellular damage and induces activation of apoptosis. The underlying mechanisms of lysosomal permeability due to fatty acid exposure result from translocation of the pro-apoptotic Bcl-2 family member Bax to the lysosome. In addition, the Bax antagonist Bcl-XL was reported to be suppressed in response to palmitate treatment of hepatocytes. This was confirmed by investigations in which inhibition of Bax activity diminished lysosomal permeability (Feldstein et al., 2006). 9.6.3.1

Phospholipidosis

Drug-induced phospholipidosis (DIPL) is characterized by intracellular accumulation of phospholipids and appearance of so-called lamellar bodies (also: myeloid bodies, dense bodies). This condition can occur in several organs and is induced primarily by cationic amphiphilic drugs. In the liver, DIPL has been associated with fibrotic and inflammatory changes. The assumed target of drugs inducing phospholipidosis is the lysosome. Basically, two hypotheses have been put forward to explain the underlying mechanisms of drug-induced phospholipidosis (Reasor and Kacew, 2001). The first hypothesis assumes that binding of CADs to phospholipids results in indigestible drug–lipid complexes which accumulate and are stored in the form of lysosomal lamellar bodies (Halliwell, 1997), while the second hypothesis is based on the observation that production of lamellar bodies was associated with inhibition of phospholipase activity (Reasor and Kacew, 2001). Furthermore, other lysosomal enzyme activities have been modulated by amphiphilic drugs, notably the activity of cathepsin B protease (Gerbaux et al., 1996). Although some of the drugs that cause DIPL, such as amiodarone and propanolol, also cause hepatotoxicity, it is still unclear whether this is a result of independent adverse effects or due to the phenotype, which has been termed ‘phospholipidosis’.

Acetaminophen

CYP1A2 (−/−)

Flutamide

TCDD, 4-PeCDF, PCB 153

4-Aminobiphenyl

TCDD

Hexachlorobenzene and iron Phenacetin

Toxicanta

– increased drug plasma levels and alternative metabolite in urine of KO – nutritional GSH depletion and repetitive drug administration caused increased aminotransferase activity and moderate hepatitis in KO not WT→potential toxicity of alternative metabolite

– higher mortality in KO mice (less metabolite – more parent substance) – pronounced hepato- and splenomegaly in KO mice – less hepatocellular damage or uroporphyria in KO→KO mice protected from uroporphyria and hepatic injury – KO mice were not resistant to TCDD-induced immune suppression – small differences in malignant foci and involvement of eosinophils – similar extent of ROS stress in KO and WT – no sequestration of TCDD and 4-PeCDF in the liver in KO mice

– less extensive liver injury and kidney damage in KO – similar olfactory mucosal damages in KO and WT – doses of 250 mg/kg lethal only in KO and HT – no difference in liver enzyme and metabolite excretion between WT and KO – no significant differences in liver non-protein thiol concentrations – no differences in DNA adduct formation between KO and WT→no significant role in APAP hepatotoxicity for CYP1A2 – increased resistance in CYP1A2 and CYP2E1 KO – no hepatic uroporphyrin accumulation in KO

Resultsa

Transgenic animal models of hepatotoxicity

Transgenic model

Table 9.2

(Continued)

Matsuzaki et al., 2006

Diliberto et al., 1997

Smialowicz et al., 2004 Kimura et al., 1999

Smith et al., 2001

Sinclair et al., 2000 Sinclair et al., 1998; Sinclair et al., 2000 Peters et al., 1999

Tonge et al., 1998

Genter et al., 1998

References

CYP2E1 (−/−)

Benzo[a]pyrene

CYP1A1 (−/−)

Ethanol

Acetone

Chloroform Acrylonitrile, acrylamide


Acetaminophen Benzene

TCDD

Toxicanta

(Continued)

Transgenic model

Table 9.2

– higher doses needed to induce toxicity in KO – reduced metabolite excretion in KO mice – KO were more resistant to benzene-induced cyto- and genotoxicity,→benzene metabolism of the liver 95 % dependent on CYP2E1 – no abnormality in liver histology found in KO→KO highly resistant to CCl4 toxicity – protection from toxicity in the liver, kidney and nose in KO – WT excrete epoxides and GSH conjugation metabolites from acrylonitrile (AN) and acrylamide (AM) – urine after cytochrome inhibition and in KO displayed only GSH/AM or AN conjugates→CYP2E1 possibly the only cytochrome P450 involved in AM and AN metabolism – blood acetone levels similar in KO (null) and WT – fasting revealed differences: significantly increased acetone levels in KO compared to WT – both WT and KO showed similar signs of toxicity due to ethanol – no difference in fatty acid metabolism between KO and WT mice (fed choline- and methionine-deficient diet)

– higher sensitivity of KO compared with WT and HT, with dramatically higher BaP–DNA adduct levels in KO – KO mice were protected from TCDD toxicity (displayed less hepatocyte hypertrophy and less intrahepatocytic lipid and total liver fat accumulation, no uroporphyria, showed reduced lethality and no wasting syndrome such as seen in WT) – thymic atrophy, decreased spleen size and leukocytopenia occurred in both genotypes

Resultsa

Kono et al., 1999 Leclercq et al., 2000

Bondoc et al., 1999

Constan et al., 1999 Sumner et al., 1999

Lee et al., 1996 Sinclair et al., 2000 Valentine et al., 1996 Powley and Carlson, 2001 Wong et al., 1998

Uno et al., 2001, 2004a Uno et al., 2004b

References

(Continued)

NQO1 (−/−)

Ahr (−/−)

Cyp1A2/1B1(−/−)

Menadione

Benzo[a]pyrene

TCDD

Benzo[a]pyrene

Acetaminophen

CYP1A2 and 2E1 (−/−) Cyp1A1/1B1(−/−)

– KO more sensitive than HT and WT

– highly resistant DKO, less GSH depletion and APAP adducts compared to WT – Cyp1a1(−/−) knockout mice are protected, show prolonged survival – higher BaP-DNA adduct levels in Cyp1a1(−/−) than in Cyp1a1(+/−) mice – no induction of CYP1A1 in Ahr–KO mice – resistance against TCDD-induced liver injury in KO (10-fold higher concentrations than WT liver injury dose)→role of CYP1A1 induction – reduced embryotoxicity – reduced teratogenicity in Ahr–KO – no subcutaneous tumours in KO but in WT and HZ

– in 6 months, 70 % lymphomas in WT vs. only 7.5 % in KO – generally, increased number of tumours in WT not KO→role of CYP1B1 in DMBA carcinogenesis (bioactivation to DMBA-3,4-diol) – CYP1B1 in bone marrow stromal cells→required for DMBA-induced pre-B-cell apoptosis (role in lymphoblastoma formation)

– highly significant adduct formation between acrylamide–GSH product glycidamide and DNA as well as haemoglobin in KO mice—important role for CYP2E1 in adduct formation/toxicity in acrylamide-induced injury – KO resistant to DMBA toxicity

Acrylamide

Dimethylbenz[a] anthracene

Resultsa

Toxicanta

CYP1B1 (−/−)

Transgenic model

Table 9.2

(Continued)

Radjendirane et al., 1998

Shimizu et al., 2000

Mimura et al., 1997; Peters et al., 1999

Uno, 2004a

Zaher et al., 1998

Heidel et al., 1999

Ghanayem et al., 2005

References

Acetaminophen

Galactosamine/ endotoxin

SOD1 (−/−) and SOD1/GPX1 double (−/−)

Glutathione peroxidase-1 (Gpx1) (−/−)

Diquat

Acetaminophen

CCl4

GSTP1 GSTA4-4 (−/−)

Glutathione S-transferases (GST) P2

Toxicanta

(Continued)

Transgenic model

Table 9.2

– higher susceptibility to oxidant stress induced by neutrophils and liver injury in KO than in WT (induced significantly more genes coding for inflammatory response, oxidative stress, growth arrest and responses to DNA damage and/or its repair in KO) – both KO and WT developed apoptosis – increased plasma aminotransferase activities in KO→enhanced sensitivity to secondary cell injury in KO – low survival rates in KO vs. WT mice – increased liver injury, lower activities of liver thioredoxin reductase and catalase and higher liver carbonyl contents in KO

– In contrary to WT, KO and DKO were protected from lethality due to APAP overdosage (prolonged survival time, less cell injury, less hepatic GSH depletion than in WT) – liver injury protection associated with block of the APAP-mediated hepatic protein nitration and reduction of CYP2E1 in KO liver – KO higher resistance to APAP, but more sensitivity to NAPQI toxicity compared to WT – DKO showed lower CYP2E1 activity, but higher glutathione reductase and thioredoxin reductase activities than WT – less peroxynitrite-mediated protein nitration in KO and DKO

– hepatotoxicity in both KO and WT mice – KO show a higher grade of degenerative change (cellular swelling, disarray, vacuolization) surrounding the centrilobular necrotic lesions→role of GSTA-4 in secondary damage – KO highly resistant to APAP-induced hepatotoxicity (difference in recovery of GSH levels in KO, not WT mice)

Resultsa

Fu et al., 1999a Fu et al., 1999b

Jaeschke et al., 1999 Bajt et al., 2002 Li et al., 2003

Lei et al., 2006

Zhu and Lei, 2006

Henderson et al., 2000

Dwivedi et al., 2006

References

Methacrylonitrile (MAN) and acrylonitrile (AN) Acetaminophen

Acetaminophen

CCl4

Acetaminophen

Thioacetamide

Epoxide hydrolase (mEH) (−/−) Deoxyribonuclease 1(−/−)

Hsp70i (−/−)

CAR (−/−)

Retinoid X receptor alpha (−/−)

Metallothionein (MT)-I and MT-II (−/−)

cis-Diamminedichloroplatinum (cisplatin)

Acetaminophen

Arsenic

Toxicanta

(Continued)

Transgenic model

Table 9.2

Oliver et al., 2006

– downregulation of CYP1A2 and CYP3A11 in KO, while upregulated in WT mice – significantly less lipid peroxidation in WT mice than in KO – more GSH depletion and reduced liver regeneration in KO – kidney and liver lesions in KO more severe than in WT mice – higher levels of cytokines in KO, more sensitivity to GSH depletion – higher susceptibility to acetaminophen-induced lethality and hepatotoxicity in KO mice compared to WT – increased sensitivity to oxidative stress – higher sensitivity to and pronounced apoptosis in cisplatin-induced toxicity in KO mice

Wu et al., 2004

(Continued)

Liu et al., 1998

Liu et al., 1999

Liu et al., 2000

Yamazaki et al., 2005

Tolson et al., 2006

Napirei et al., 2006

El Hadri et al., 2005

References

– KO more sensitive to APAP toxicity – thermal upregulation of HSP decreased hepatotoxicity in both WT and KO, but toxicity was higher in KO – slightly less liver damage in KO – differences pronounced after stimulation with phenobarbital→CAR activation caused CCl4 hepatotoxicity, while CAR inhibition resulted in partial protection against CCl4 -induced hepatotoxicity – resistance against APAP-induced liver injury in KO

– significant reduction in blood cyanide levels in MAN-treated KO→abolished AN to cyanide metabolism in KO mice – necrosis but less affection by overdose of APAP in KO

Resultsa

a

– clofibrate treatment showed marked protection against APAP toxicity in WT but not in KO – no reduced hepatic glutathione depletion and arylation of cytosolic proteins in KO vs. WT mice −→ protective role of PPARalpha

– significantly increased expression of tumour necrosis factor-alpha (TNF-α), connective tissue growth factor (CTGF) and reduced IL10 in KO – significantly more severe liver injury and lethality in KO mice – untreated KO: lower UDP-GT expression and NPSH content – no significant differences between KO and WT mice

Chen et al., 2000

Boess et al., 1998

Enomoto et al., 2001

Gardner et al., 2002

Cover et al., 2005

Xie et al., 2004

Zhou et al., 2002

References

TCDD (2,3,7,8-tetrachlorodibenzo- p-dioxin); 4-PeCDF (2,3,4,7,8-pentachlorodibenzofuran – dioxin-like compound); PCB 153 (2,2 ,4,4 ,5,5 -hexachlorobiphenyl).

Acetaminophen + clofibrate

Acetaminophen

Acetaminophen

TNF/lymphotoxinalpha (−/−) PPARalpha (−/−)

– less sensitivity to APAP toxicity in KO

Acetaminophen

Acetaminophen

– no significant differences in toxicity between KO and WT mice

Acetaminophen

Poly(ADP-ribose) polymerases (PARP) (−/−) gp91 phox (−/−) subunit of NADPH oxidase Inducible nitric oxide synthase (NOS II)

nrf (−/−)

– greater susceptibility of KO to arsenic-induced pathological changes (elevated LPO levels increased glutathione S-transferase (GST) activity) – enhanced expression of potential oncogenic cyclin D1 only in KO – no difference in liver injuries between KO and WT mice

Arsenic

Mdr1a/1b(−/−)

– resistance to acute alcohol-induced hepatotoxicity in KO – reduced signs of oxidative stress in KO

Ethanol

Metallothionein overexpression

Resultsa

Toxicanta

(Continued)

Transgenic model

Table 9.2


229

Drug-induced phospholipidosis was recently reviewed by us (Anderson and Borlak, 2006). 9.6.3.2

The Role of Mitochondrial Dysfunction in Hepatotoxicity

Alterations of the mitochondrial metabolism have been demonstrated to play a central role in different kinds of toxic liver injury. Functional impairment of these organelles is associated with energy depletion and extensive production of ROS that lead to a vicious circle of cellular stress. Several events have been demonstrated to cause failure of the mitochondrial metabolism, which eventually will end in cellular necrosis and apoptosis (Kim et al., 2003). The three main mechanisms by which toxicants cause primary functional mitochondrial failure are uncoupling of the oxidative phosphorylation, acivation of the caspases cascade leading to apoptosis, and disturbance of the Ca2+ homeostasis that causes perturbations and dysfunctions of the cellular metabolism, such as a disruption of the wellarranged actin assembly, possibly resulting in disruption of the cell membrane integrity (observed under Amanita phalloides toxin) (Watanabe and Phillips, 1986). In addition, damage of the mitochondrial structure, disruption of the β-oxidation of lipids and alterations of mitochondrial DNA synthesis have been recognized mechanisms in liver toxicity (Cullen, 2005). Different toxicants, however, have their different ways of reaching these endpoints. For example, uncoupling of the respiratory chain can result from direct interference with mitochondrial enzymes or from dysfunctions caused by mutations of the mitochondrial DNA (mtDNA) (both, for instance, observed effects of iron) (Ramm and Ruddell, 2005). Mitochondrial dysfunction results in cell death by apoptosis and is mediated by cytochrome c(1)-dependent activation of the mitochondrial permeability transition pores (MPT-pores) via Ca2+ - or ROS-related signalling (Rauen et al., 2004). This mechanism, for instance, has been discussed to contribute to the hepatotoxic effects of the non-steroidal anti-inflammatory drug salicylic acid (aspirin), the antipyretic drug acetaminophen (Reid et al., 2005; Vendemiale et al., 1996), the narcoleptic valproic acid, toxicity of bile acids (Rolo et al., 2003) and p-hydroxybenzoic acid (parabens) and others (Nakagawa and Moore, 1999; Trost and Lemasters, 1996). In the following, some examples of mitochondrial dysfunction in relation to their morphologic appearance in toxic liver injury will be discussed. The herbicide paraquat is a quatenary nitrogen bipyridinium that causes toxic reactions in the lung, kidneys and liver. In isolated hepatocytes, paraquat caused ultrastructural mitochondrial alterations and cell death in a dose- and time-dependent fashion (Palmeira et al., 1994a). This has been linked to an impressive depletion of ATP and NADPH, accompanied by depletion of glutathione (GSH) and an increase in oxidised glutathione (GSSG) (Palmeira et al., 1994b). More evidence for a central mitochondrial dysfunction was provided by an uncoupling of the respiration chain and partial inhibition of ATPsynthase and the mitochondrial complexes II and IV (Palmeira et al., 1995a). The underlying mechanism in paraquat toxicity has been proposed to result from the formation of oxidative paraquat radicals. By oxidation of critical SH groups at the respiratory complex, these are believed to induce an increased permeability of the proton pore (Castilho et al., 1995). The subsequent decrease of the mitochondrial membrane potential could provide an explanation for an uncoupling of the respiratory complex (Wallace et al.,

230

Hepatotoxicity

1997). Accordingly, cell death of hepatocytes due to paraquat exposure has been linked to primary mitochondrial failure, resulting in a critical breakdown of the cellular energy supply. Besides, a decreased GSH/GSSG ratio has been suggested to contribute to the progression of irreversible cell injury. Certainly, the decreased GSH/GSSG ratio indicates an imbalance of the cellular redox system, as a result of enhanced intrinsic production of ROS. Investigations regarding the formation of oxygen radicals in the presence of paraquat revealed that an NADH-quinone oxidoreductase embedded in the mitochondrial outer membrane is responsible for an enhanced production of superoxide anions (Hirai et al., 1992). It is, therefore, not surprising that isolated hepatocytes treated with paraquat displayed significant amounts of lipid peroxides, since these are directly formed by reaction of oxidative radicals with intracellular and membranous lipids (Palmeira et al., 1995b). Lipid peroxidation has been demonstrated to cause an impairment of the cellular and metabolic integrity by modification of proteins (Britton, 1996; Khan et al., 2002). Taken together, these findings suggest that metabolic perturbations induced by lipid peroxidation might provide an alternate or additional mechanism contributing to paraquatinduced toxicity. Importantly, formation of undue ROS can be observed in several cases of hepatotoxicity induced by different agents. Even under physiological conditions, mitochondria release as much as 2–4 % of reducing equivalents from the respiratory chain, which produce superoxide anions and hydrogen peroxide from oxygen and water. In addition, radicals can be generated by formation of unstable intermediates at the respiratory chain that are produced when alternate agents serve as electron acceptors. These can carry electrons over to oxygen to produce superoxide radicals (Wallace et al., 1997). Cellular stress, as induced by various physiological and non-physiological conditions, including ageing, toxic agents or inflammation, can exacerbate mitochondrial ROS production (Gao et al., 2004). In addition, undue amounts of ROS can result from diminished clearance through the antioxidative defence system (Pessayre et al., 2001; Vendemiale et al., 1996), which will be discussed in detail later. Increased ROS production is known to induce damage in lipids, proteins and nucleic acids. In mitochondria, it can cause direct uncoupling of the oxidative phosphorylation by increasing the activity of the uncoupling protein UCP2 (Douette and Sluse, 2006). Furthermore, it was demonstrated that alterations of the mitochondrial respiration could be induced by an overload with metals, such as iron and copper, to result in increased generation of ROS and subsequently lipid peroxidation. Adduct formation of cytochrome c oxidase with the lipid peroxidation product 4-hydroxynonenal (HNE) resulted in consecutively decreased activity of cytochrome c oxidase (Chen et al., 1998). This enzyme is the terminal electron acceptor in the respiratory chain and holds an essential role in the production and regulation of energy (Ludwig et al., 2001). Oxidative attack of DNA results in the formation of oxidised bases, such as 8hydroxy-2 -deoxyguanosine (8-oxoG) (Bohr et al., 2002). This 8-OxoG can mispair with 2 -deoxycytidine 5 -triphosphate or with 2 -deoxyadenosine triphosphate during DNA replication, forming C∗ 8-oxoG and A∗ 8-oxoG mispairs. In humans, the MutY repair enzyme is responsible for recognition and repair of this defect. If the repair mechanism fails,


231

deletion of C:G and transversion to A:T can lead to further alterations (Parker and Eshleman, 2003). A combination of modified mtDNA and oxidation of mitochondrial proteins and lipids has also been implicated in increased ROS formation in alcoholic patients (Pascussi et al., 2001). Additional mechanisms have been identified to cause mitochondrial dysfunction. For example, an impairment of mitochondrial membrane integrity has been suggested to be part of the toxic response observed under intoxication with aflatoxin B. The major metabolite of aflatoxin was shown to damage the mitochondrial membrane and thereby cause depletion of the mitochondrial respiration (Sajan et al., 1995, 1996, 1997). Nucleoside reverse transcriptase inhibitors (NRTIs) are the cornerstones of the highly active antiretroviral therapy (HAART) against the human immunodefiency virus. The hepatotoxicity associated with these substances clinically appears to be highly variable and ranges from low-grade hepatotoxicity and asymptomatic lactacidaemia to severe liver insufficiency with massive steatosis and life-threatening lactic acidosis (Van Huyen et al., 2006). Investigations revealed that NRTIs can affect the function of mitochondrial polymerase gamma, which replicates mitochondrial DNA. This interaction has been suggested to cause depletion of the mitochondrial DNA or qualitative changes that might relate to the mitochondrial cytopathy found in NRTI-treated patients (White, 2001). The ultrastructural alterations found in mitochondria during NRTI-induced toxicity have recently been associated with a decreased expression of the cytochrome oxidase (COX) subunit I, which is encoded by mitochondrial as well as nuclear DNA (Van Huyen et al., 2006). The exact actions by which toxic effects on mitochondria may affect the cellular metabolism still need to be established. Induction of apoptosis and an impairment of the energy metabolism due to altered oxidative phosphorylation, however, are believed to be the major consequences of mitochondrial dysfunction (Szewczyk and Woitczak, 2002) and current research has focused on elucidating the possible sequences leading to these endpoints.

Toxic effects on mitochondria in summary: – – – – –

Interferences in the oxidative phosphorylation Inhibition of β-oxidation Mutagenic effects on mitochondrial DNA Inhibition of transport mechanisms (BSP) (substrate or cofactor depletion) Induction of apoptosis by MPT and cytochrome c release

9.6.4

Cellular Stress

Oxidative stress. Stress can be defined as any disturbance of the cellular metabolism, which can be induced by various conditions, such as temperature or pH fluctuations, starving or over-alimentation (supernutrition), radiation, ageing and several endogenous and exogenous agents and toxins.

232

Hepatotoxicity

Oxidative stress is defined as an imbalance between production and clearance of reactive oxygen/nitrogen species (ROS/RNS) (Sies, 1991). Accumulation of oxidative damage has been proposed to represent the ageing process per se. It is implicated in both acute and chronic cell injury and possibly plays a role in chemical carcinogenesis (Klaunig et al., 1998). Reactive oxygen species may be represented by the superoxide anion O− 2 , hydrogen r peroxide (H2 O2 ), the hydroxyl radical OH , organic peroxides or radicals (Fridovich, 1978). The natural abundance of ROS arises from reactions at the inner mitochondrial membrane. Superoxide radicals are produced in reaction with oxygen during electron transfer from the cofactors NADH and FADH2 to the first complexes of the respiratory chain. Thereafter, the mitochondrial manganese superoxide dismutase (MnSOD) transforms superoxides to H2 O2 (Nohl et al., 2005). Nitric oxide is a representative of the reactive nitrogen species (RNS) and provides the main source for formation of peroxynitrate by reaction with superoxide (Beckman, 1996). Peroxynitrate leads to nitration of protein tyrosine residues and is detoxified by GSH. It has recently been implicated to be involved in APAP-induced toxicity, where it was found to be located specifically in areas of necrotic centrilobular liver cells (Hinson et al., 1998). To keep the intracellular equilibrium between oxidative and reductive valences, the presence of antioxidants is of vital importance. The antioxidative defence is provided by nonenzymatic antioxidants, such as glutathione and vitamins E and C, and also by enzymatic antioxidants, including superoxide dismutase (SOD), glutathione peroxidase, glutathione reductase, catalase and hemoprotein peroxidases. Superoxide dismutase and catalase together, for instance, build a functional unit in which SOD captures superoxides and thereby forms H2 O2 , which is still highly reactive and can be transformed to water and oxgen by catalase (Scandalios et al., 2005). Additionally, drug-metabolizing Phase II enzymes such as glutathione-S-tranferase, glucuronosyl transferases and NAD(P)H:quinone reductase, provide direct and indirect protection against oxidative stress. Glutathione-S-transferase, for instance, is responsible for inactivation of electrophilic substances (by conjugation with nucleophilic glutathione) to enable clearance and excretion. An example of indirect antioxidative defence is the removal of possible inducers of ROS, such as is performed by NAD(P)H:chinon oxidoreductase (Sies, 1991). Glutathione (GSH), for instance, is an important intracellular antioxidant that scavenges free radicals and other radical species (e.g. the hydroxyl radical, lipid peroxyl radical, peroxynitrite and H2 O2 ) by direct and indirect enzymatic reactions (Lei, 2002). Two mechanisms allow glutathione to protect the cell from undue oxidative stress: first, it can protect protein thiols from ROS/RNS and secondly, it can reverse oxidative modifications, e.g. through removal of disulfide bonds and nitrosthiols (Han et al., 2006). In addition, glutathione is involved in Phase II reactions of biotransformation, thereby connecting antioxidant defence and the detoxifying system. Possible interferences between these two systems have been implicated in acetaminophen (APAP)-induced hepatotoxicity. Detoxification of the active APAP metabolite NAPQI is achieved by reaction with glutathione (GSH). Thus, it can be observed that toxic overdosage of APAP results in depletion of the GSH reservoir, which can also affect the cellular antioxidant system and causes


233

dramatic shifts in the cellular redox state (Davis et al., 1974; Mitchell et al., 1973). Indeed, there is specific evidence for the presence of a significant quantity of ROS in cells exposed to toxic concentrations of APAP. Specifically, lipid oxidation, mitochondrial damage and presence of peroxynitrate, which is formed by reaction of NO with superoxide, have been observed (Gao et al., 2004; Park et al. 2002; Sies, 1991). Glutathione depletion is a common feature in liver toxicity induced by various drugs or xenobiotics. This can result from either detoxification of a drug through conjugation with GSH (APAP) or from exceedingly high production of ROS that can no longer be compensated by GSH. In this context, GSH depletion has different effects depending on the compartment affected. Mitochondrial GSH depletion was found to induce necrosis in primary hepatocyte cultures, while cytoplasmic GSH depletion resulted in sensitisation of proteins to TNF-induced apoptosis by inhibition of NF-κB-dependent survival gene expression (Matsumaru et al., 2003; Nagai et al., 2002). The consequence, however, is an increased potential of oxidative stress and the damage resulting from this. The transcriptional regulation of genes coding for enzymes of the antioxidant defence is a central event within the cellular response to oxidative stress and is mediated by interaction of transcription factors at specific DNA segments to either enhance or silence the expression of a certain gene. The antioxidant response element (ARE) is such an important DNA segment that was discovered in rodents and is relevant for basal expression and monofunctional induction of drug-metabolizing enzymes, such as γ -glutamylcysteine synthetase and heme oxygenase-1 (Friling et al., 1990; Kong et al., 2001; Rushmore et al., 1991). The oxidative stress-sensitive transcription factors Nrf-1 and Nrf-2 were found to specifically interact with AREs and thereby provide a positive regulation of antioxidative genes (Moinova and Mulcahy, 1999). The ARE complex consists of Nrf-1/2 and Maf proteins and was found to be activated by depletion of GSH that involved downstream activation of the PI3 and p38 MAP kinases and activation of the transcription factors AP-1 and NF-κB (Kang et al., 2000; Yang et al., 2005). Besides transcriptional regulation, antioxidant gene regulation is influenced and regulated by phosphorylation and nuclear translocation of transcription factors. The existence of a redox-sensing system has been suggested to be responsible for the detection of oxidative stress and subsequent induction of the anti-oxidative defence. Among these, glutathione and the protein thioredoxin have been proposed to play a key role (Nordberg and Arner, 2001). The mechanism by which GSH acts as a cellular sensor of the intracellular redox situation is regulated by the ratio of unoxidised GSH to the oxidised glutathione disulfide (GSSG). This ratio is maintained by GSSG reductase, which performs an NADPH-dependent reconversion from GSSG to GSH. Hence the GSH ratio is not only dependent on glutathione synthesis (gluamate cysteine ligase and GSH synthetase) and activity of GSSG reductase, but via NADPH dependence is also essentially connected to the energy status of the cell (Han et al., 2006). A shift of GSH/GSSG redox towards an oxidized state was shown to activate important signal transduction pathways, including activation of protein kinase B, protein phosphatases 1 and 2 A, calcineurin, c-Jun N-terminal kinase, apoptosis signalregulated kinase 1 and mitogen-activated protein kinase, AP-1 and NF-κB (Pinkus et al., 1996; Sen, 2000).

234 9.6.4.1

Hepatotoxicity Nrf-2 and ARE Signalling and AP-1

The DNA binding activity of AP-1, NF-κB and Nrf-2 is dependent on the reduction of cysteine residue 506. In this way, the cellular redox can influence these transcription factors (Bloom et al., 2002). The glutathione and thioredoxin (TRX) systems share the feature of controlling the intracellular thiol/disulfide redox environment by thiol-dependent reductions. Even though they have been shown to regulate the cellular redox by different pathways, an overlapping of function has been reported for these two systems (Hansen et al., 2004). Both were demonstrated to function through Nrf-2-dependent signalling, although at different cellular sites. While GSH has recently been demonstrated to control the cytoplasmic dissociation of Nrf-2, TRX was shown to be responsible for the nuclear regulation of Nrf-2/DNA interactions (Hansen et al., 2004). Nrf-2 is a redox-sensitive transcription factor that regulates genes in response to oxidative stress. This is mediated via a cis-acting enhancer sequence: the antioxidative response element (ARE) (Nguyen et al., 2003). The antioxidative response element was shown to be responsible for transcriptional activation of genes involved in glutathione synthesis and regulation (Kwak et al., 2003; Moinova and Mulcahy, 1999; Rushmore and Pickett, 1993; Wild et al., 1998), genes coding for other thiols and other proteins with antioxidative properties (thioredoxin-1 (Kim et al., 2003b) & HO-1 = (Inamdar et al., 1996; Ishii et al., 2000)), as well as genes encoding drug-metabolizing enzymes (Favreau and Pickett, 1991; Jaiswal, 1991; McMahon et al., 2001; Rushmore and Pickett, 1990). Upon stimulation by oxidative stress, Nrf-2 dissociates from its inhibitory complex with the molecule Keap-1 and translocates from the cytoplasm to the nucleus. How activation of Nrf-2 occurs has not been fully elucidated – mechanisms that have been particularly implicated so far are involvement of phospho-dependent signalling cascades dominated by kinases, such as phosphoinositol-3-K (PI3K), MAP-kinase and phosphokinase C (PKC), and also direct phosphorylation (Nguyen et al., 2003). In addition to oxidative stress, ARE is also activated by several xenobiotics, including antioxidants (Rushmore et al., 1991), thiol-containing isocyanates (Bonnesen et al., 2001; Talalay et al., 1995), heavy metals (Prestera and Talalay, 1995) and heme complexes (Inamdar et al., 1996; Ren and Smith, 1995). AP-1. The activator protein-1 (Ap-1) complex is a transcription factor composed of several components from the family of dimeric basic region–leucine zipper proteins, including several proteins belonging to the Jun, Fos, Maf and ATF subfamilies (Shaulian and Karin, 2002). Ap-1 regulates vital cellular functions, such as cell proliferation and differentiation, survival and cell death, and has been implicated to be involved in the response to liver injury induced by toxicants and bile acids (Bernt et al., 2006). Several stimuli are known to induce Ap-1 signalling, however – in the context of toxic liver injury, induction by cytokines and cell matrix–interaction, as well as chemical stress, are of particular interest (Shaulian and Karin, 2002). Nuclear factor-κB. The oxidative stress-sensitive NF-κB has been attributed a central role in mediating inflammation, proliferation and cell death, and it is crucially involved in the regulation of cell survival (Hayden and Ghosh, 2004; Karin, 1999).


235

Specifically, involvement of NF-κB in regulation of apoptosis and ROS is of particular interest, since severity and progression of liver injury was shown to depend on the incidence of hepatocytic apoptosis and extent of oxidative stress. Activation of NF-κB represents the major event in the signalling pathway of TNFR1dependent apoptosis. The TNFR1-associated adaptor proteins TRADD, TRAF-2 and RIP are involved in activation of the IκB-kinases that initiate the release of NF-κB from its inhibitory proteins (IκB) (Hayden and Ghosh, 2004). In addition, activation of NF-κB can also prevent apoptosis by downregulation of apoptotic genes or upregulation of antiapoptotic genes, including the cellular inhibitors of apoptosis 1 (IAP-1 and IAP-2), FLICE-inhibitory protein (FLIP), TRAF-1 and TRAF-2, which, for instance counteract apoptotic signalling by caspase inhibition (Poppelmann et al., 2005; Shishodia and Aggarwal, 2002). There is evidence that depletion of GSH may be the possible catalyst for decreased NF–κB activity and contribute to enhanced sensitivity to TNF-induced apoptosis by sustained activation of c-Jun-N-terminal kinase (JNK) (Matsumaru et al., 2003). Changes of NF-κB activity are early events observed after acetaminophen (APAP) administration. It was demonstrated that APAP significantly reduces the DNA binding activity of NF-κB up to complete inhibition 4 h after dosing. The same finding was made for nuclear factor-IL6 (nf-IL-6), which regulates genes involved in the acute phase reaction of the liver (Blazka et al., 1995). It has been proposed that APAP toxicity is characterized by negative regulation of genes that are altered during the toxic response (such as those implicated in hepatotoxicity induced by metals (Cisternas et al., 2005), ethanol-induced liver injury (Yuan et al., 2006), carbon tetrachloride (Liu et al., 1995) and others). For example, APAP was found to prevent the nuclear transition of the NF-κB and NF-IL6, compromising activation of the so-called survival genes (Blazka et al., 1995; Manna et al., 1998). In addition to its antiapoptotic effect, activation of NF-κB has been implicated to promote the second phase of liver injury (in particular inflammation and fibrosis) by induction of cytotoxic cytokine synthesis and initiation of chemotaxis of mononuclear cells, indicating ambivalent functions of NF-κB for the toxic response (Liu et al., 1995). A finding that supported the crucial role of oxidative stress in APAP-induced hepatotoxicity was that administration of the antioxidant thiol agent N -acetylcysteine (NAC), a cysteine precursor, amended APAP-induced toxicity, which was presumably due to replenishment of GSH levels (Bessems and Vermeulen, 2001; Corcoran and Wong, 1986; James et al., 2003b). There is also evidence that NAC influences activation of NF-κB and expression of inflammatory mediators (Bellezzo et al., 1998; Fox and Leingang, 1998; Pahan et al., 1998). The connection between glutathione and NF-κB was suggested to be mediated by thioredoxin, which is reduced by GSH, promotes the reduced state of NF-κB and also provides cysteine which is required for DNA binding of NF-κB (Matsumaru et al., 2003; Matthews et al., 1992). The oxidative response is associated with activation of several pathways of mitogenactivated protein kinases (MAPKs) signalling that are involved in coordinating the response to formation of ROS. MAPKs include a large number of serine/threonine kinases that are involved in the regulation of a wide array of cellular processes, including proliferation, differentiation, stress adaptation and apoptosis (Kolch, 2000). Members of the MAPK subfamilies, such as

236

Hepatotoxicity

extracellular signal-regulated kinase (ERK), c-jun N-terminal kinases (JNK) and p38, have been shown to be substantially involved in the response to oxidant injury and, therefore, could potentially contribute to a cell’s fate between survival or apoptosis following exposure to oxidative stress (Cuda et al., 2002; Johnson and Lapadat, 2002; Martindale and Holbrook, 2002). Signalling of ROS has particularly been recognized to contribute to TNF-α-dependent activation of NF-κB and JNK, as well as downstream apoptotic pathways, notably, by enhancing the TNFR1 (tumour necrosis factor alpha-tumour necrosis factor receptor 1)dependent pathway and interacting with the redox regulatory protein thioredoxin that acts as regulator of apoptosis signal-regulating kinase 1 (ASK1) (Shen and Pervaiz, 2006). By contrast, opposing observations were made, in which superoxide had an inhibitory effect on death receptor-dependent and independent induction of cellular death (Clement et al., 2005; Clement and Stamenkovic, 1996). Based on these findings, it was proposed that a slight pro-oxidant condition might even provide a survival advantage (Shen and Pervaiz, 2006). 9.6.4.2

How do ROS Harm Cellular Integrity?

Nitric oxide is known to act both as an intra- and extracellular messenger molecule; likewise, the physiological role of ROS for cell development, differentiation and survival has been explored in the past decades (Dalton et al., 1999; Delaunay et al., 2000; Muriel, 2000). Increased amounts of intracellular reactive oxygen species, however, were shown to cause severe damage to lipids, proteins, and DNA (Scandalios, 2005). Protein damage that occurs under conditions of oxidative stress may be found as direct oxidation of proteins by ROS and/or RNS or adduction of secondary products of oxidation, such as sugars (glycooxidation) or polyunsaturated lipids (lipoxidation) (Dalle-Donne et al., 2005; Sayre et al., 2001; Stadtman, 1992). Direct protein damage due to ROS can occur via different mechanisms causing polypeptide chain cleavage, crosslinking, and modification of the side-chain of amino acids (Berlett and Stadtman, 1997; Dean et al., 1997). The modifications induced by oxidative/nitrosative stress may be permanent or reversible (Stadtman and Berlett, 1998). While reversible modifications (usually at Cys residues) modulate the protein function, they may also have a role in protecting the cell from irreversible damage. Irreversible modifications, such as protein–protein crosslinking, are generally associated with permanent loss of function and may lead to either degradation of damaged proteins (Berlett and Stadtman, 1997; Dean et al., 1997; Grune et al., 2003) or their progressive intracellular accumulation into cytoplasmic inclusions, as observed in age-related neurodegenerative disorders (Giasson et al., 2000, 2002; Butterfield and Kanski, 2001). The functional consequences of such reactions are diverse and include partly or complete inhibition of function, enhanced protein aggregation and proteolysis, changes in protein flux, altered immunogenicity or perturbations of cell signal transduction (Han et al., 2006). ROS are known to be involved in activation and inhibition of several signalling pathways, including c-Jun NH2 -terminal protein kinase (JNK), protein kinase C (PKC) and tyrosine kinase signalling. In addition, H2 O2 itself has been suggested to serve as a second messenger in any signalling pathway although no suitable receptor has, as yet, been identified (Han et al., 2006).


237

There is indeed cumulating evidence that lipid peroxidation and protein modification due to oxidative stress is involved in the pathogenesis of chronic liver injury and fibrogenesis. This was supported by a study which investigated in vitro effects of oxidized low-density lipoproteins (oxLDL) on collagen and fibronectin synthesis of cultured human and rat hepatic stellate cells (HSC). Specifically, it was demonstrated that oxLDL stimulated the synthesis of collagen Types I and III and fibronectin of cultured HSC in a dose-dependent fashion (Schneiderhan et al., 2001). Moreover, production of reactive oxygen species and of reactive aldehydes from lipid peroxidation is part of the toxic response associated with other liver toxicants known to cause fibrosis, as observed with CCl4 and iron. Iron, in particular, is a potent inducer of reactive oxygen from O2 through the Fenton reaction. This reaction is a commonly accepted mechanism by which free radicals are formed. It consists of two steps, in which ferric iron first is reduced to ferrous iron (Equation (9.1)) to later generate a hydroxyl radical by attacking hydrogen peroxide (Equation (9.2)) (Buda et al., 2003): r− Fe+ −−→ Fe+ (9.1) 3 + O2 − 2 + O2 r − Fe+ −−→ Fe+ 2 + H 2 O2 − 3 + OH + OH

(9.2)

In hereditary hemochromatosis, different types of genetic defects lead to excessive iron absorption, eventually promoting an early onset of liver fibrosis and cirrhosis (Franchini, 2006). Investigations in patients diagnosed with hemochromatosis revealed that blood iron levels were positively correlated with the extent of lipid peroxidation and hence indicated an increased amount of oxidative stress in these patients (Young et al., 1994). A correlation between iron levels and lipid peroxidation supports the central role of ROS and lipid peroxidation in the pathology of hereditary hemochromatosis. Enhanced production of extracellular matrix due to iron provides at least one possible mechanism for initiation of early liver fibrosis associated with this disease. ROS effects onto DNA may lead to formation of specific adducts. These so-called etheno (epsilon)-modified DNA bases are, for instance, generated by reactions of DNA with the major product of lipid peroxidation, namely trans-4-hydroxy-2-nonenal. Even though elimination by base excision repair can be performed via ethenobase-DNA glycosylases in vivo (Gros et al., 2003), a high level of these etheno(epsilon) adducts has been associated with an increased risk to develop cancer. Patients diagnosed with cancer-prone diseases, such as Wilson’s disease and primary hemochromatosis, displayed significantly increased urinary levels of ethenobases, while in patients diagnosed with chronic hepatitis, cirrhosis and hepatocellular carcinoma levels of urinary etheno-deoxyadenosine were measured up to 20–90-fold higher than compared to controls (Bartsch and Nair, 2004). An analogous product, 8-nitroguanine (8-NO(2)-G), is formed by the reactions of guanine, guanosine or 2-deoxyguanosine with nitrogen species (RNS) generated from peroxynitrite. This has also been linked to inflammatory reactions and promotion of carcinogenesis (Ohshima et al., 2006). 9.6.5

Oxidation of Lipids and Lipid Toxicity to the Liver

Lipid oxidation in the liver occurs in association with several liver toxicants, such as carbon tetrachloride, ethanol and the metals iron and copper.

238

Hepatotoxicity

Increased production of reactive oxygen species (ROS) is responsible for peroxidation of biomembranes and polyunsaturated fatty acids. There are several possibilities by which lipid peroxidation might interfere with important cellular functions. For example, membrane lipid peroxidation products may react with DNA and proteins to cause oxidative modifications (Park et al., 2002). In addition, it was demonstrated that lipid peroxidation destabilizes organelle compartments. Iron-induced lipid peroxidation, for instance, is associated with enhanced lysosomal fragility and inactivation of microsomal enzymes (Barreto et al., 2005; Britton et al., 1990, 1994; Lieber, 1993). Furthermore, mitochondrial dysfunction was associated with lipid peroxidation in iron- and copper-induced liver toxicity (Sokol et al., 1990). The extent of lipid peroxidation in vitro can be assessed by measurement of thiobarbituric acid reactants in the liver. In vivo, ethane in exhaled breath serves as an index for increased intracorporeal lipid peroxidation (Gutteridge and Quinlan, 1983; Riely et al., 1974). Treatment with several different steatogenic hepatotoxicants had linked enhanced lipid peroxidation with the presence of oxidizable fat in the liver. Due to the fact that steatosis is commonly associated with enhanced lipid peroxidation, it has been hypothesized that lipid peroxidation might be the major pathogenic factor in hepatocellular necrosis and mild inflammation observed in steatohepatitis (Letteron et al., 1996). The aldehydes malondialdehyde (MDA) and 4-hydroxynonenal (HNE) are endproducts of the peroxidation of membrane n-6 polyunsaturated fatty acids (Romero et al., 1998). Oxidized fatty acids have an important role in cellular signal transduction, such as in activation of Jun N -terminal kinase (Castello et al., 2005). Indeed, HNE is involved in the regulation of different genes and modulates various enzyme activities. It was demonstrated that increased HNE concentrations activated the protein kinase C-delta and thereby triggered apoptosis in isolated rat hepatocytes (Castello et al., 2005; Chiarpotto et al., 2005; Leonarduzzi et al., 2005). Moreover, HNE was shown to enhance DNA binding of AP-1 transcription factor and to activate macrophage and smooth muscle cells. This chemotactic effect of mediated by activation of the phosphoinositide-specific phospholipase C and leads to induction of the expression and synthesis of monocyte chemotactic protein 1 (MCP-1) and TGFβ1 (Chiarpotto et al., 2005; Leonarduzzi et al., 2005; Rossi et al., 1993). It was also proposed that lipid aldehydes may cause impairment of the cellular function and integrity by covalent binding with proteins (Britton, 1996; Khan et al., 2002). Redox alterations induced by ROS/RNA can induce apoptosis and necrosis in hepatocytes and other cells. Several regulators of cell survival and death are strongly influenced by the intracellular redox, including caspases, NF-κB, JNK and PKC, which have already been mentioned. ROS can activate caspases by direct stimulation of the mitochondrial membrane transition pore and cytochrome c release and also controls the stability of p53, which act in surveillance of cell integrity (Jaeschke et al., 2003). It has been observed that low levels of ROS, in particular, cause apoptosis, while high levels of ROS tend to induce necrosis. This has been explained by the different redox potentials (and therefore different sensitivities) of the diverse proteins that might be involved and also by caspases inhibition that was observed with high levels of ROS (Hampton and Orrenius, 1997). Moreover, oxidative stress-inducing and GSH-depleting agents were shown to influence the response to TNF-α, which is an important inflammatory cytokine and triggers


239

apoptosis as well as necrosis (Matsumaru et al., 2003). In addition, the mitochondrial GSH level was shown to be especially important for activation of apoptosis via initiation of MPT. In particular, this has been suggested to be an important mechanism of apoptotic liver cell injury in acetaminophen-, rifampicin- and INH-induced hepatotoxicity (Chowdhury et al., 2006; Masubuchi et al., 2005; Reid et al., 2005). Other investigators had proposed that oxidative stress may be a major determinate of progression of hepatocellular damage, but does not represent a relevant mechanism in APAP toxicity. This was concluded from the finding that co-administration of the antioxidant vitamin E with APAP did not abrogate APAP-induced injury in animals. In addition, massive lipid peroxidation was observed in APAP-induced hepatotoxicity only after the onset of injury, suggesting that LPO may be a consequence rather than a cause of cellular damage (Knight et al., 2003). Another interesting connection has been made between the occurrence of oxidative stress and the degradation of actin filaments observed in early APAP-induced hepatotoxicity (Tsokos-Kuhn et al., 1988). Oxidation-dependent cytoskeletal protein alterations have been observed and investigated in quinone-induced toxicity (Mirabelli et al., 1989). Quinones may induce toxicity by several mechanisms, including alkylation and oxidative stress following redox cycling. The toxicity of redox cycling quinones in isolated rat hepatocytes was accompanied by a dose-dependent increase in the relative abundance of F-actin and an oxidation-dependent actin crosslinking, as well as by dissociation of the α-actinin from the actin network (Bellomo et al., 1990; Thor et al., 1988). The mechanisms responsible for these alterations have been suggested to involve the depletion of cytoskeletal protein sulfhydryl groups and the increase in cytosolic Ca2+ -concentration following the alkylation or oxidation of free sulfhydryl groups in several Ca2+ -transport systems (Bellomo et al., 1990). Loss of intracellular F-actin-containing stress fibres has been linked to cytotoxicity of hepato- and nephrotoxins (Dartsch et al., 1998). Stress fibres are arrangements of myosin and actin filaments that are linked to the plasma membrane via association with focal adhesions (McCue et al., 2004). They mediate adhesion to the extracellular matrix, migration and are produced during shear stress. Stress fibres in fibroblasts have an essential role in wound healing and morphogenesis and also fibrosis. The transdifferentiation of the stellate cell to myofibroblasts is the key for the regeneration ability of the liver and is closely associated with cytoskeletal reorganization and an increased extracellular matrix production in these cells (Kato et al., 1999). Inhibition of stellate cell activation and disturbance of stress fibre formation have been suggested as possible targets for treatment of liver fibrosis and cirrhosis (Imanishi et al., 2004; Matsui et al., 2004; Matsui and Kawada, 2005). As specified above, the impairment that can arise from oxidative damage concerns a wide range of cellular functions. Liver toxicity, for instance, may, in part, be promoted by enhanced production of ROS. In addition, toxicants may cause damage by an impairment of the antioxidative defence. This, for instance, may result from a decreased expression of antioxidants or an impaired function due to oxidative modification. Likewise, ethanol was demonstrated to impair the antioxidant defence system by depleting GSH and promoting its leakage, thereby causing mitochondrial damage and apoptosis (Lieber, 1997; Seitz and Stickel, 2006).

240

Hepatotoxicity

Both acetaminophen and amiodarone (AD) were reported to induce significant decreases of glutathione peroxidase (GPX), which is normally responsible for scavenging oxidative radicals, although the same study had revealed that the potential hepatotoxic drugs APAP, AD and tetracycline induced upregulation of other antioxidative proteins, including peroxiredoxins 1 and 2 (PRX1/PRX2) in primary cultured rat hepatocytes (Yamamoto et al., 2005). The complexity of the response to oxidative stress has, as yet, not been elucidated, but the interaction with the response to toxic stimuli is most likely to cause exponentiation of the subsequent hepatocellular damage.

Oxidative stress in liver toxicity in in summary: – – – – – – –

Reducing antioxidative defence (e.g. GSH depletion, SOD repression) Alterations of the antioxidatie response Induction of e.g. CYP1E2→ROS ROS→lipid peroxidation, DNA and protein adducts→impaired metabolic function Direct activation of apoptotic programs and cascades by ROS ROS signalling in death and survival ROS→inflammation→fibrosis

9.6.6 9.6.6.1

Programs in the Hepatotoxic Response Apoptosis and Necrosis

Apoptosis and necrosis are assumed to represent two different sequences that arrive at a common endpoint – cellular death. While apoptosis is a highly organized and structured procedure that is crucial in embryonic development and regeneration activity, the common understanding of the term ‘necrosis’ comprehends an acute incidence that is caused by heavy alterations initiated by an overwhelming stimulus. Necrosis. Necrosis is characterized by ATP depletion that is caused by severe metabolic perturbations. Diminished levels of ATP cause a breakdown of the osmotic regulation and integrity of the cytoskeleton, which results in the characteristic morphology of cell swelling accompanied by membrane blebbing (Gores et al., 1990; Nishimura et al., 1998). A subsequent collapse of the residual cellular functions can be observed by loss of the mitochondrial potential and lysosomal stability, finally leading to an irreversible disturbance of plasma membrane permeability (Herman et al., 1988). Even though apoptosis and necrosis are distinct and independent pathways according to the traditional view, recent discoveries have implemented a link between these two events. Mitochondrial permeability transition (MPT), for instance, was demonstrated to play a causative role in induction of both necrosis and apoptosis in ischemia–reperfusion injury experiments in cultured hepatocytes (Kim et al., 2003a). The current view considers the possibility that necrosis and apoptosis share common pathways, for instance beginning at a common death signal, which depending on modifying factors, such as intracellular ATP concentrations, either entail cell lysis or programmed cell death (Lemasters, 1999). Taking


241

this into account, John J. Lemasters suggested to introduce a new term, e.g. ‘necrapoptosis’, to preserve the traditional distinction between apoptosis and necrosis (Lemasters, 1999). Apoptosis. Apoptosis is distinctively characterized by morphological alternations, such as chromatin condensation, nuclear fragmentation and DNA degradation, cell shrinking and membrane blebbing, as well as formation of apoptotic bodies (councilman bodies), which are membrane-surrounded vesicles of the cellular content. Due to this localized degradation of single cells, the apoptotic process exerts less effects onto the environment than necrosis, during which release of potassium attracts inflammatory cells, while release of intracellular enzymes, such as proteases and hydrolases, damage the extracellular matrix and surrounding cells. The apoptotic programme is activated by several stimuli, including both exogenous and endogenous sources, such as DNA damage and ER stress, radiation, ROS and deprivation of growth factors or nutrition. Correspondingly, the terms ‘extrinsic’ and ‘intrinsic pathways’ of apoptosis have been formed. The extrinsic pathway is induced by an external ligand that leads to activation of the so-called death receptors. The four best studied of these transmembrane proteins are the Fas receptor (CD95), which is activated by binding to the socalled Fas ligand, tumour necrosis factor receptor 1 (TNF-α receptor, TNFR1) and tumour necrosis factor-related apoptosis-inducing ligand receptors 1 (TRAIL-R1) and 2 (TRAILR2). In the liver, additionally, death receptor 3 (DR3/TRAMP) and death receptor 6 are of importance (Malhi et al., 2006b). In response to binding to one of their dedicated ligands (Fas, TNF-α and TRAIL), these receptors form oligomers, thereby inducing recruitment of adaptor proteins, such as Fas-associated protein with death domain (FADD) or TNFR1associated death domain protein (TRADD) (Faubion and Gores, 1999). A central event of apoptosis is the activation of specific apoptotic enzymes, the cysteine– aspartate-specific proteases (caspases). Binding to specific adaptors leads to activation of certain caspases (-2,-8,-9,-10), inducing a whole cascade that involves autoactivation and downstream activation of caspases that are capable of degrading cellular contents. Upstream caspases are activated by the so-called death-inducing signalling complex (DISC), which consists of oligomerized death receptors, their adaptor proteins, procaspase 8, and procaspase 10. In contrast, within the intrinsic pathway activation of caspases, for instance, results from damage of DNA and mitochondria and subsequent leakage of pro-apoptotic factors into the cytosol. These include, among others, cytochrome c, SMAC/DIABLO (second mitochondria derived activator of caspases/direct IAP (inhibitor of apoptosis) binding protein with low PI), HtrA2/Omi, apoptosis-inducing factor and endonuclease G (Du et al., 2000; Jiang and Wang, 2004; Verhagen et al., 2000; Widlak and Garrard, 2005). These pathways, however, operate independently. There has been evidence that in several cells and tumours the intrinsic and extrinsic apoptotic pathways are connected in a coordinated crosstalk (Khosravi-Far and Esposti, 2004). Furthermore, the complex regulation of apoptosis is mediated by an interplay of pro-apoptotic and apoptotic factors, including members of the bcl-2 family, such as the pro-apoptotic proteins Bak, Bid, Bad, Bim, Noxa, Puma and the antiapoptotic proteins Bcl-2, Bcl-XL and Mcl-1 (Cory et al., 2003). The proteins Bax and Bak cause changes of the mitochondrial membrane integrity and thereby trigger mitochondrial dysfunction (Wei et al., 2001). An illustration of the apoptotic cascade is presented in Figure 9.4.

242

Hepatotoxicity

FAS

TNF-R1

TRAIL-R1

PC 8

TRAIL-R2 PC 8

PC 8

PC 8

TRADD TRAF-2

Lysosome

Cas 8 Bid

Bid

Pc 3

Bid

Bax

Cas 3 Bak

MPT Bax

Bax

Cathepsin B

Cytochrome C NFkB

ROS

Fatty acids

Figure 9.4 The apoptotic cascade. In the liver, apoptosis can be induced by two distinct pathways. The type 1 pathway (Peter and Kramer, 2003; Scaffidi et al., 1998) is mediated via death receptors and adaptor proteins, such as TRADD and TRAF2. These induce activation of procaspase 8 to caspase 8, which is responsible for activation of the executioner caspase 3. The type 2 pathway involves cleavage of the pro-apoptotic protein Bid by active caspase 8, followed by further activation of Bcl-2 members, triggering mitochondrial dysfunction and release of cytochrome c (Hatano et al., 2000; Li et al., 2002b)

The inevitable death cascade continues with the release of antiapoptotic factors by the mitochondrion, including cytochrome c and SMAC/DIABLO, thus accelerating apoptosis by reducing caspase via inhibition of IAP (Chai et al., 2000). Notably, the mitochondrial transition pore (MPTpores) that provides one important mechanism of cytochrome c release, is regulated by Ca2+ and induces permeabilization of the inner mitochondrial membrane. In consequence, mitochondrial depolarization results in uncoupling of the oxidative phosphorylation and mitochondrial swelling (Zoratti and Szabo, 1995). Cytochrome c subsequently activates caspase 3 and associates with Apaf-1 to a formation called the apoptosome, which subsequently activates caspase 9 (Li et al., 1997). The executioner caspase 3 leads to activation of endonucleases and induces DNA fragmentation (Liu et al., 1997). Apoptosis in liver injury. Apoptosis has been proposed to be a major event in toxic liver injury. An increased incidence of hepatocyte apoptosis has been observed in patients and animal models of different types of toxic liver injury, including those induced by ethanol, copper and autoimmune-mediated mechanisms.


243

Elevated levels of the death receptor Fas and its ligand were demonstrated in ethanolinduced liver injury and found to be even more pronounced in non-alcoholic steatosis, where Fas levels correlated with the severity of the disease (Feldstein et al., 2003a; Pianko et al., 2001). In the liver, Fas receptors are expressed in hepatocytes, cholangiocytes, sinusoidal endothelial cells, stellate cells and Kupffer cells (Malhi et al., 2006b). They are mainly located at the Golgi complex and to a smaller extent are also found implemented in the plasma membrane (Bennett et al., 1998). Upon certain stimuli, a cascade is activated involving activation of sphingomyelinase, protein kinase C-zeta and NADPH oxidase. Consecutive superoxide signalling is responsible for activation of src-kinase, subsequently leading to phosphorylation cascades and activation of Fas. This causes the Fas receptor to translocate to the plasma membrane, where it is activated by binding to the Fas ligand, released by cytotoxic T lymphocytes and natural killer cells (Lee and Ferguson, 2003). The physiological role of the Fas receptor has recently been explored in a Fas-knockout mouse strain. Fas-deficient mice displayed substantial liver hyperplasia, accompanied by enlargement of hepatocytic nuclei and massive proliferation of lymphocytes (Adachi et al., 1995). The Fas system was also found to be involved in the pathogenesis of Wilson’s disease, where a hepatic overload of copper induces fulminant hepatic failure. Studies in patients diagnosed with this hereditary disease as well as in vitro investigations revealed that copper overload activates Fas-induced apoptosis and that this factor also contributes to the massive cell death during hepatic failure (Strand et al., 1998). In ethanol-induced liver injury, circulating levels of the Fas receptor and Fas ligand were found to be increased and notably correlated with the severity of the disease (Natori et al., 2001; Taieb et al., 1998; Ziol et al., 2001). Ethanol has been demonstrated to induce apoptosis by two different pathways: first, it induces the mitochondrial permeability transition (MPT) and secondly, it was shown to cause upregulation of the CD95-Fas ligand (Minana et al., 2002). Other studies, however, have revealed an increased expression of the death receptor TNFR1 in acute and chronic liver injury induced by alcohol (Deaciuc et al., 1995). An increased expression of death receptors may sensitize the liver to lymphocytemediated cytotoxicity (Natori et al., 2001; Tagami et al., 2003). An involvement of inflammatory cells has been concluded from findings where enhanced Fas receptor expression and infiltrating neutrophils colocalized in apoptotic liver areas (Jaeschke, 2002a,b). A possible mechanism by which induction of the Fas ligand by ethanol occurs, is a initiation of the corresponding signalling cascades as a result of undue generation of ROS from the mitochondrial acetaldehyde metabolism (Guicciardi and Gores, 2005; Hug et al., 1997). Acetaldehyde is toxic to the mitochondria and aggravates oxidative stress by binding to reduced glutathione and promoting its leakage (Lieber, 2004). The relevance of ROS for ethanol-related apoptosis was supported by investigations that demonstrated the antiapoptotic effect of antioxidants in ethanol-induced liver inury (Kurose et al., 1997). The toxic effects of ROS are believed to be triggered by induction of TNF-α expression and subsequent activation of NF-κB, which is also involved in the principal signalling pathway initiated by activation of TNFR1 (Chan et al., 1999). Moreover, Fas involvement has been proposed to play a role in the pathogenesis of non-alcoholic steatohepatitis (NASH). Similarly, increased apoptosis and expression of

244

Hepatotoxicity

Fas receptor and TNFR1 were observed in patients diagnosed with NASH (Ribeiro et al., 2004); this was associated with caspase 8-dependent cleavage of Bid and activation of the executioner caspases 3 and 7 (Feldstein et al., 2003a,b). It is still unclear which role the apoptosis-induced mitochondrial dysfunction has in the development of the steatosis. NASH can be induced by drugs or by nutritional factors. Notably, fatty acids occurring along with this condition, presumably as result of a mitochondrial dysfunction, were found to induce the lysosomal pathway of apoptosis and cause mitochondrial dysfunction alone (Feldstein et al., 2004). Hence accumulation of fatty acids could cause predisposition of the liver to cell death induced by death receptors (Feldstein et al., 2003b). In models of cholestatic liver disease, the Fas receptor was shown to play a central role in bile acid-mediated apoptosis. Hydrophobic bile acids and bile salts induce apoptosis in hepatocytes, as demonstrated in experiments with primary hepatocytes in vitro (Faubion et al., 1999; Guicciardi and Gores, 2002; Higuchi et al., 2001). In animal models of extrahepatic cholestasis, the toxicity of hydrophbic bile salts has been investigated intensively. Increased rates of apoptosis have been observed, which was related to a pathway involving the Fas receptor (Miyoshi et al., 1999). From studies with mutant Fas ligand-expressing and Fas-null hepatocytes, it was demonstrated that apoptosis due to deoxycholic acid occurred in a ligand-independent fashion (Qiao et al., 2001), although it was also demonstrated that presence of the Fas receptor was required in bile acid-induced apoptosis (Gupta et al., 2004). While this was not associated with increased expression levels of the Fas receptor, an enhanced trafficking of the latter to the plasma membrane has been demonstrated (Sodeman et al., 2000). Activation of the Fas receptor was suggested to result from autoactivation induced by the increased density of the Fas receptor at the cell surface (Guicciardi and Gores, 2005). Another study had demonstrated the involvement of other death receptor pathways in bile acid-mediated apoptosis, such as the finding that apoptosis in Fas receptor-deficient cells occurred in a TRAIL associated manner (Higuchi et al., 2001). Furthermore, the Fas receptor-initiated apoptosis was implicated in the pathogenesis of viral hepatitis, hepatic fibrosis, hereditary biliary cirrhosis and autoimmue hepatitis, as well as in the carcinogenesis of hepatocellular and cholangiocellular carcinomas (Ghavami et al., 2005). The central role of apoptosis in toxic and cholestatic liver injury, hepatic inflammatory and fibrosis was supported by the protective effect of caspase inhibition on bile acidmediated cytotoxicity, progression of fibrosis, and stellate cell activation (Canbay et al., 2004). This, in turn, had led to development of caspase inhibitors for the therapeutic intervention in liver damage, which was demonstrated to be effective in animal models of extrahepatic cholestasis (Guicciardi and Gores, 2002). In the meantime, the therapeutic concept of caspase inhibition, as for instance, implemented in the case of the pan caspase inhibitor IDN-6556, has been introduced in clinical trials, where it is currently being evaluated (Linton, 2005; Valentino et al., 2003). 9.6.6.2

Chemokines and Cytokines as Mediators of Hepatocellular Toxicity

Inflammatory reactions are mediated by chemokines and cytokines. These play an important role in the pathology of acute toxic hepatitis, immune hepatitis and chronic liver diseases, including fibrosis and cirrhosis.


245

In particular, TNF-α and the members of the C–X–C chemokine family, including interleukin-8 (IL-8) in humans and macrophage inflammatory protein-2 (MIP-2) in rodents, have been suggested to play an important role in damage and repair processes associated with various hepatotoxins (Dong et al., 1998). The most important inflammatory mediator, TNF-α, mediator of both local and systemic inflammation and is mainly produced by activated macrophages but also by other cell types, such as lymphoid cells, mast cells, endothelial cells and fibroblasts (Schwabe and Brenner, 2006; Tracey et al., 1987; Wajant et al., 2003). TNF-α is known to influence both apoptotic and survival mechanisms and is thus part of the difficult regulation between anti-apoptotic and pro-apoptotic pathways. Via Fas ligand activation of the Fas receptor, TNF-α induces apoptosis and, on the contrary, deploys a self-limiting action by inducing nuclear transcription factor NF-κB. NF-κB is responsible for the transcriptional regulation of genes that are termed survival genes and include antioxidant and anti-inflammatory proteins (Baichwal and Baeuerle, 1997; Bubici et al., 2006; Xiao and Ghosh, 2005). By stimulating the expression of the inducible NO synthetase (iNOS), NF-κB enhances the production of nitric oxide, which can inhibit the caspase cascade (Bradham et al., 1998). On the other hand, NF-κB upregulates expression of the Fas receptor and FasL genes and may thereby sensitise cells for the extrinsic pathway of apoptosis (Chan et al., 1999). A possible protective role of TNF-α has been implicated in APAP-induced hepatotoxicity. Here, studies in TNF-α receptor TNFR1-deficient mice had revealed that a lack of TNF response resulted in a delayed recovery of intracellular GSH levels, induction of hemeoxygenase-1 and decreased expression of CuZn superoxide dismutase and its catalase activity, both responsible for the reduction of superoxide anions and hydrogen peroxide levels (Chiu et al., 2003a). Next to TNF-α, hydrogen peroxide was demonstrated to induce production of IL-8 and MIP-2 in freshly isolated hepatocytes (Dong et al., 1998a). Thus, increased oxidative stress may directly induce cytokines that contribute to an inflammatory response. Specifically, this is the proposed mechanism by which metals are believed to induce expression of inflammatory cytokines. This was supported by the finding that administration of an antioxidative substance, such as N -acetylcysteine, abrogated metal-induced cytokine expression. A TNF-α receptor knockout mouse model revealed moreover that cadmium chloride (CdCl2 ) and vanadium pentoxide (V2 O5 ) induced expression of inflammatory cytokines independent of TNF-α production (Dong et al., 1998b). 9.6.6.3

Acute Phase Reaction

After acute liver injury, the liver metabolism switches to an acute phase response, which includes transcriptional activation of acute phase genes and enhanced synthesis of acute phase proteins, such as C-reactive protein (CRP) or alpha 1-acid glycoprotein (AGP) (Ezendam et al., 2004). In transgenic mice, high levels of the acute phase protein CRP were demonstrated to protect from lethal sepsis, assumingly by blocking cytokine stimulation induced by bacterial lipopolysaccharides (LPS) (Xia and Samols, 1997). The acute phase response is also activated in toxic liver injury. Within 24 h after exposure to toxicants, plasma AGP concentrations were found to be increased up to 2–3.5 times of the normal levels in toxic liver injury due to CCl4 , allyl alcohol, bromobenzene, acetaminophen or N -nitrosodimethylamine.

246

Hepatotoxicity

Although not all liver toxicants seem to induce this reaction, ethionine-induced liver injury, for instance, showed no effects on plasma AGP concentrations, whereas galactosamine even markedly decreased AGP levels (Sugihara et al., 1992). In CCl4 -induced liver injury, the rise in plasma AGP concentrations was positively correlated with increased rates of plasma protein binding, which can be explained by the weak basic properties of CCl4 and the acidic properties of AGP (Sugihara et al., 1991). Taken together, these findings suggest that the acute phase response can be considered to be part of a physiological defensive reaction. The acute phase protein CRP, for instance, is known to induce cytokine expression in monocytic cells and was also suggested to act as a mediator of the classical complement system activation (Heijnen et al., 2006). The same cascades that may protect from bacterial invasion, however, may substantially contribute to the development of the ischemia– reperfusion injury of the liver. In an animal model, the complement complex C5b9 and CRP were found to be co-localized in hepatocytes, and in addition it was demonstrated that inhibition of complement was able to protect rats from ischemia–reperfusion liver injury (Heijnen et al., 2006). Studies in animal models, where the acute phase response was induced by injection of Escherichia coli LPS, indicated that several CYP-dependent activities might be suppressed during the acute phase reaction. The study revealed that the maximum velocity (Vmax ) values and the amount of CYP1A1/1A2 and CYP2E1 apoproteins were decreased during the acute phase response (El Kadi, 1998; Saitoh et al., 1999). Further studies revealed that the relation between the nature of the inflammatory reaction has an influence on P450 activity, and that some cytochrome monoxygenases are more (CYP3A6) susceptible than others (CYP1A1 and 1A2) (Bleau et al., 2001). The impact of the acute phase response upon the metabolizing function of the liver was furthermore emphasized by the finding that the acute phase response resulted in decreased gene expression of albumin and glutathione S-transferase, suggesting an impairment of drug transport and neutralization (short-time inactivation), as well as alterations of the Phase II biotransformation (Whalen et al., 2004). 9.6.6.4

TNF-α Antibodies in Acute Liver Failure

Driven by the consideration that an exacerbating inflammatory reaction can cause severe alterations of the liver metabolism, antagonizing the inflammatory mediators was experimentally investigated as a possible therapeutic strategy. Decreasing pro-inflammatory cytokine levels by pentoxifylline, as a TNF-α synthesis inhibitor, immunotherapy with IL-4 or IL-1 receptor antagonists, as well as reactivation of PPAR-α by administration of long-chain n-3 fatty acids have been evaluated in different disease models in which the inflammatory reaction is a central and progressing factor (Prandota, 2005). In fact, antagonizing the effect of TNF-α, which is particularly involved in cell death during fulminant liver failure (Wang et al., 2003), achieved successful results in animal models of LPS-induced liver injury and ischemia-reperfusion injury (Sneed et al., 2000; Yang et al., 2004b; Yee et al., 2003). In contrast, abrogating TNF-α signalling was not able to prevent liver toxicity induced by the mycotoxin Fumonisin B1 (FB1), in which TNF-α was shown to have a central role (He and Sharma, 2005). Nevertheless, further investigation


247

may help to assess the potential that lies in antagonizing the inflammatory reaction in toxic liver injury. The inflammatory response is not only transmitted by monocytic cells: a variety of cytokine genes were found to be expressed by hepatocytes in response to inflammatory stimuli, such as LPS (Rowell et al., 1997) or interferon gamma (IF-γ ) (Ren et al., 2002) and includes among others MCP-1 (Wang et al., 1999), cytokine-induced neutrophil chemoattractant (CINC/ Interleukin-8) (Ohkubo et al., 1998) and the liver-specific cytokine CCL16 (Bauer et al., 1989; Nomiyama et al., 2001). The production of cytokines by hepatocytes was further implicated to be modulated by non-parenchymal cells, such as Kupffer cells (Mawet et al., 1996). Intriguingly, the major aldehydic product of lipid peroxidation, HNE, is chemotactic to macrophages and smooth muscle cells by inducing expression of MCP-1 TGF-β1 (Chiarpotto et al., 2005, Leonarduzzi et al, 2005). Thus, hepatocellular damage by oxidative stress is one possible mechanism by which the inflammatory response may be initiated. Furthermore, the inflammatory reaction in response to toxicants involves enhanced Kupffer cell infiltration and neutrophil infiltration that can be observed along with activation of NF-κB and enhanced expression of intercellular adhesion molecule-1 and cytokineinduced neutrophil chemoattractant-1 mRNAs in hepatocytes (Bauer et al., 2000). There is strong evidence for neutrophils to contribute to liver injury, causing cellular damage by burst of ROS and receptor mediated-neutrophil cytotoxicity. In particular, the release of TNF-α and IL-1 α has been suggested to be responsible for pathological manifestations of APAP-induced hepatotoxicity (Blazka et al., 1995). Indeed, inactivation of macrophages was shown to have a protective effect against APAPinduced hepatotoxicity (Goldin et al., 1996; Laskin and Pilaro, 1986). Even though the role of inflammation for the onset and progression of the acute toxic injury and chronic hepatic disease is not completely understood, inflammation greatly contributes to secondary (toxicant concentration-independent) liver injury. This may explain why inhibition of neutrophils protected hepatocytes from secondary toxicity and limited acute injury, but had no effect on prevention of APAP-induced liver injury (Lawson, 2000). 9.6.6.5

Activators of Survival and Repair

The activation of redox-sensitive transcription factors enables adaptive defence responses against toxic stimuli. In the case of APAP, the activation of these transcription factors has been observed to occur even at low and non-toxic concentrations (Goldring et al., 2004). The oxidative stress-sensitive transcription factor NF-κB plays an essential role in the regulation of genes that are involved in tissue damage and inflammation, such as the gene coding for cyclooxygenase-2 or antioxidative heme oxygenase-1 (HO-1). Heme oxygenase-1 is responsible for degradation of the prooxidant heme into carbon monoxide, iron and the antioxidant biliverdin. Induction of MO-1, for instance, has been observed to be part of the adaptive response upon treatment with carbon tetrachloride and has been suggested to be of critical importance for the recovery from hepatocellular injury. This was concluded from the observation that inhibition of heme oxygenase-1 resulted in a sustained increase in microsomal free heme concentrations associated with exacerbated liver injury (Nakahira et al., 2003).

248

Hepatotoxicity

In studies investigating the role of HO-1 in chemical carcinogenesis it was shown that HO-1 expression increased along with progressive exposure to the mutagenic pdimethylaminoazobenzene. Notably, an inverse relationship between the expression of HO-1 protein and the degree of tumour differentiation in hepatocellular carcinomas was observed, therefore supporting a significant role of this enzyme in tissue repair processes (Caballero et al., 2004). 9.6.6.6

Regeneration

The regenerative activity of the liver has been a mythology itself for thousands of years. Some important molecular insights have highlighted the role of apoptosis, autophagy and proliferation for the enormous ability of the liver to completely regenerate after partial hepatectomy and loss of parenchyma. Experimental partial hepatectomy was demonstrated to result in extensive changes of the liver’s gene expression, which was mainly attributed to modulation of liver-enriched transcription factors (Schrem et al., 2004). Liver regeneration is associated with increased expression of the hepatocyte growth factor (HGF), which was demonstrated to confer a significant mitogenic stimulus. This cytokine is produced by Kupffer cells and sinusoidal endothelial cells, but not by parenchymal hepatocytes, and exerts not only mitogenic, but also cytotoxic as well as growth inhibitory activities, which are presumably mediated by activation of the transcription factor NF-κB and consecutive induction of the liver-enriched transcription factor C/EBP-β (Shen et al., 1997). Physiologically, hepatocyte proliferation and expression of the C/EBP-β transcription factor are exclusive to the postnatal liver development (Buck et al., 1994). Activation of C/EBP-β is required for a normal proliferative response, where it causes cell cycle arrest through inhibition of cyclin-dependent kinases (cdks). During liver regeneration, however, hepatocytes regain these proliferative features by a mechanism that has not been elucidated. After partial hepatectomy, hepatocytes were demonstrated to acquire the ability to produce the antiapoptotic bcl-2 protein and also display a decreased transcription of the three pro-apoptotic members of the family, Bak, Bad, and Bax (Tzung et al., 1997). Similar experiments revealed enhanced expression of the cytokines TGF-β 1, 2 and 3 during liver regeneration (Bissell et al., 1995). Furthermore, evidence was provided that TNF-α expression and interaction at the TNFR1 receptor were required for liver regeneration (Yamada et al., 1997). A possible link between the mandatory involvement of TNF-α has been provided by the finding that TGF-α-dependent activation of C/EBP-β is essential for hepatocellular proliferation via p90 ribosomal S kinase (RSK) (Buck et al., 1999). The TNF cytokines in the liver were attributed the exert effects on cell growth control, inflammation, and cytotoxicity (Diehl et al., 1994). TNF-α induces cell death in cultured hepatocytes and mouse liver in vivo but, surprisingly, inhibition of TNF-α transcription resulted in extensive cell death, which suggested the activation of further antiapoptotic pathways by TNF-α (Bradham et al., 1998). This pro-apoptotic effect has been ascribed to TNF-α-dependent activation of NF-κB (Patel et al., 1998). Even though NF-κB might contribute a positive effect to liver generation,


249

recent studies have demonstrated progress of liver regeneration in presence of a NF-κB inhibitor (Laurent et al., 2005). The early growth response-1 transcription factor (Egr-1) is another gene that has been recognized to be part of the immediate-early gene expression response during early liver regeneration, notably by regulating hepatocellular mitotic progression through the spindle assembly cell cycle checkpoint (Liao et al., 2004). 9.6.6.7

MAPK P38 Inhibitors and Liver Toxicity

The p38 mitogen-activated protein kinase (MAP kinase) is a member of the well-studied MAP kinase family, which is known to be involved in the regulation of vital cellular functions, including cell cycle control, differentiation and apoptosis (Ashwell, 2006). Specifically, the JUN N -terminal kinases (JNKs) and p38 were shown to be activated during the cellular stress response and in reaction to pro-inflammatory stimuli, including lipopolysaccaride (LPS), interleukin-1 (IL-1), transforming growth factor beta (TGF-β) and tumour necrosis factor (TNF) (Freshney et al., 1994; Hannigan et al., 1998; Mendelson et al., 1996; Raingeaud et al., 1995; Rouse et al., 1994). Furthermore, p38 was found to be responsible for signal integration of inflammatory pathways. Activation and response of primary murine T-cells was closely related to activation of p38, and in turn inhibition of p38 was also shown to suppress T cell reactivity to certain stimuli (Zhang et al., 1999). Constitutive expression of the mitogen-activated protein kinase kinase 6 (MAPKK6), an important activator of p38, had suggested a role for p38 in thymocyte development as well as in CD8+ -mediated apoptosis (Pedraza-Alva et al., 2006; Rincon, 2001). In addition, it has been demonstrated that cytokine production of IL-1 beta, IL-2, IL-4, IL-5, IL-8, IL-13 and IFN-γ is mediated by activation of p38 (Clark et al., 2003; Dodeller et al., 2005; Pesu et al., 2002; Zhang and Kaplan, 2000). Targeting the p38 MAP kinase with specific inhibitors has contributed new insights into the physiological role of the MAP kinase pathways in apoptosis, proliferation, survival, stress response and inflammatory processes (Dambach, 2005). Such inhibition of the alpha and beta forms of p38 by BRIRB 796 exerted a suppressive effect on inflammatory processes due to administration of LPS, such as activation of peripheral white blood cells, cytokine production and release of acute phase protein C-reactive protein (CRP) (Branger et al., 2002). In contrast, information gained from studies in knockout mice had implicated a vital role of p38 in development and differentiation, for deletion of the P38alpha gene in knockout mice resulted in embryonic death due to defects in the cardiovascular system (Allen et al., 2000). Moreover, several studies had linked p38 activity to proliferation, differentiation and pathophysiological response of heart and skeleton muscle, neurones and blood cells (Behr et al., 2003; Liang and Molkentin, 2003; Piao et al., 2002; Platanias, 2003; Ravingerova et al., 2003; Verma et al., 2002; Zetser et al., 1999). Development of selective p38 MAP kinase inhibitors, however, has involved the hope for a molecular-targeted drug with the ability to prevent activation of cell death and disease progression in inflammatory diseases. Unfortunately, several substances that have been brought forward in pre-clinical and clinical studies have revealed severe adverse effects on different organ systems, as, for instance, VX-745 and BIRB 796, which exhibited severe toxic effects on the central nervous system (CNS) and the liver, respectively (Cirillo et al., 2002).

250

Hepatotoxicity

In the liver, p38 activity has been related to liver injury induced by ischemia–reperfusion, viral hepatitis and toxicants, such as ethanol and polycyclic aromatic hydrocarbons (Andrysik et al., 2006; Hsu et al., 2006; Thakur et al., 2006; Wang et al., 2004). Although p38 seems to be closely related to the inflammatory component and activation of Kupffer cells in liver injury, it has also been assigned an important role in regeneration processes after partial hepatectomy (Awad et al., 2000; Spector et al., 1997). The prospects to develop a molecular-targeted drug to interfere downstream in the TNF-α pathway are very promising, since progression of liver injury is often related to activation of the pro-inflammatory cascade; however, currently the lack of specificity of p38 inhibitors and an unacceptable safety profile call for further research to augment the knowledge regarding the necessity and function of p38 (Dambach, 2005). 9.6.6.8

Fluoroquinolones and Liver Toxicity

An example of drug withdrawal from the market is Trovafloxacin, a drug that displayed high antistaphylococcal activity in vitro and was a promising candidate for the therapy of oxacillin-resistant staphylococcal endocarditis (Bayer et al., 1998). After several cases of hepatic incidences, including severe hepatitis and fulminant liver failure (Lazarczyk et al., 2001; Lucena et al., 2000), Trovafloxacin in 1999 was withdrawn from the German market and restricted to special indications in the US (Blum et al., 1994). Another fluoroquinolone antibiotic, Temafloxacin, with similar chemical properties as Trovafloxacin, was withdrawn worldwide in 1992, due to several toxic effects including renal failure and hepatic dysfunction (Blum et al., 1994). It has been reasoned that the toxicity of these substances resulted from similar mechanisms based on their chemical structure. Both contain a unique difluorinated side-chain that is different from other quinolones and is responsible for the lipophility of these substances (Lucena et al., 2000). However, the mechanism by which Trovafloxacin and Temafloxacin cause liver toxicity remains unelucidated. A recent study, in which gene expression under Trovafloxacin treatment was investigated in a model that intended to stimulate idiosyncratic liver toxicity by co-application of LPS, revealed that GM-CSF was upregulated in treated rats. Additional upregulation of the pro-inflammatory cytokines IL-6, matrix metalloproteinase 9 and mitogen-activated protein kinase kinase-1 implicated that the toxicity of Trovafloxacin might result from an inflammatory reaction due to involvement of polymarphonuclear (PMN) cells (Waring et al., 2006). This hypothesis was supported by the finding that depletion of PMN was able to prevent hepatocellular injury in LPS/Trovafloxacin-treated rats (Waring et al., 2006). In vitro experiments demonstrated a toxic effect of pefloxacin, ciprofloxacin and ofloxacin in hepatocyte cultures and unscheduled DNA synthesis, as well as a significant dose-dependent increase in chromatid-type breakage were observed in in vivo investigations in rats that had been treated with doses of 20–200 mg/kg/day (Basaran et al., 1993; McQueen et al., 1991; Nordmann et al., 1989). Although clinically adverse effects (e.g. torsades de pointes, hepatotoxicity and dysglycaemias) are rarely observed under therapy of currently available quinolones, such as ciprofloxacin and ofloxacin, it has been suggested that adverse reactions may be confined to “susceptible” patient populations (Bertino and Fish, 2000; Owens and Ambrose, 2005). Therefore, although the safety profile of application of quinolones seems to be acceptable, it is necessary to be aware of the possible occurrence of toxic effects.


251

Biotransformation in liver toxicity in summary: – – – – –

Induction, repression and inhibition of metabolizing enzymes Induction of e.g. CYP1E2→ROS Repression of metabolizing enzymes→accumulation of toxins Bioactivation (generation of toxic metabolites during biotransformation) Interference with NR or NR crosstalk in vital cellular functions→reduced metabolic competence – Inhibition of drug transporters→drug accumulation 9.6.7

Genetic Polymorphisms and Drug-Related Toxicity

The different isoforms of drug-metabolizing cytochrome mono-oxygenases have long been recognized to underlie variable expression due to genetic polymorphisms (Pirmohamed and Park, 2003). These polymorphisms may be associated with reduced (poor metabolizer) as well as enhanced (extensive metabolizer) activity of the CYP enzymes (Pirmohamed and Park, 2003). In particular, frequent polymorphisms of the genes encoding for CYP1A2 and CYP2D6 were found responsible for causing a vast majority of adverse drug reactions (Wolf and Smith, 1999a,b). In the case of CYP2D6 polymorphisms, increased liver toxicity was observed after application of the calcium channel antagonist perhexilene to treat ischemic heart disease (Pirmohamed and Park, 2001). Beside CYP polymorphisms, variations of other drug-metabolizing enzymes have been associated with frequent adverse drug effects and individual susceptibility to specific toxic effects. Investigations in patients with Gilbert’s syndrome demonstrated that a reduced activity of uridine diphosphate glucuronosyl transferase (UGT1A1) was often associated with a reduced activity of N -acetylation, rendering these patients slow acetylators (Fengler et al., 1993). Gilbert’s syndrome is characterized by reduced activity of uridine diphosphate glucuronosyl transferase (UGT1A1) due to frequent gene polymorphisms, which leads to impaired glucuronidation with infrequent occurrence of hyperbilirubinaemia and jaundice that is usually harmless (Sampietro and Iolascon, 1999). Multiple alterations of drugmetabolizing enzymes and transporters, however, may be likely to account for a significant part of unexpected idiosyncratic drug reactions (Pirmohamed and Park, 2001). Drug withdrawal from the pharmaceutical market occurs frequently for retrospective safety issues after introduction to the market (Friedman et al., 1999). The common causes for withdrawal are adverse effects on the hepatic, haematological and cardiovascular systems. The major cause, however, is the hepatic reaction that may become noticeable by elevation of liver enzymes, jaundice, hepatitis or severe hepatocellular damage, necrosis and hepatic failure (Fung et al., 2001). 9.6.7.1

Thiopurine Methyltransferase and 6-Mercaptopurine and Azathioprine

The efficient immunosuppressive drug azathioprine is frequently used in therapy regimes after organ transplantation to prevent rejection and is also applied in autoimmune inflammatory diseases, such as bowels disease, autoimmune hepatitis and rheumatoid arthritis. Azathioprine is degraded to its metabolite, the imidazol derivative 6-mercaptopurine (6-MP), and is further metabolized by methylation via thiopurine methyltransferase

252

Hepatotoxicity

(TPMT), which is under control of common genetic polymorphisms resulting in a deficiency of 0.3 % and a reduced activity in 10 % of the Caucasian population (Sanderson et al., 2004). Azathioprine treatment in patients displaying a TPMT polymorphism may result in toxicity of the liver and severe myelosuppression (Kontorinis et al., 2004; Lennard et al., 1989). In vitro experiments testing azathioprine in rat hepatocytes had confirmed that even therapeutic concentrations of azathioprine and 6-MP caused liver toxicity by a mechanism which was completely reversible by application of xanthine oxidase inhibitor allopurinol and vitamin E analogue trolox, suggesting the involvement of oxidative stress, mitochondrial injury and ATP depletion in azathioprine and 6-MP toxicity (Lee and Farrell, 2001; Tapner et al., 2004). In particular, the incidence of hepatotoxicity in inflammatory bowel disease patients receiving thiopurines was found to be relevant, although it was mainly associated with corticosteroids co-treatment (Bastida et al., 2005). As a consequence, pre-emptive testing of the TPMT activity has been recommended prior to treatment with azathioprine, aiming to guide the drug’s dosage and to possibly prevent toxicity (Krynetski et al., 1996; Sies et al., 2005; Yates et al., 1997). On the other hand, only 15 % to 28 % of patients displaying adverse drug reactions to azathioprine show a TPMT-deficiency and it has been suggested that the remaining cases of adverse effects may be explained by a deficiency of the enzyme inosine triphosphate pyrophosphatase (ITPase). ITPase deficiency results in accumulation of inosine nucleotide, which was found to form the potentially toxic metabolite 6-thioITP in combination with 6-MP (Marinaki et al., 2004a,b; Sumi et al., 2002). A previous study, however, in which ITPase and TPMT genotypes and activity were correlated with the clinical response to azathioprine in 65 liver transplant recipients, had indicated that adverse drug reactions may not be predicted by preventive assessment of the TPMT activity in liver transplant patients (Breen et al., 2005). Even though there is evidence that the mechanism of action of azathioprine and 6MP is primarily associated with the induction of mitochondrial dysfunction, the exact circumstances by which these drugs cause liver toxicity and myelosuppression are unclear. The recent understanding, however, has led to the recognition that elevated 6-MP plasma metabolite concentrations are strongly associated with toxic incidences, whereas reduction and monitoring of these was clearly demonstrated to be beneficial for the patients (Seidman, 2003). Currently, the effect of antioxidant treatment in azathioprine and 6-MP toxicity is being assessed (Raza et al., 2003), signifying that a clinical benefit may result only from a deeper understanding of the underlying molecular processes. 9.6.7.2

Dihydropyrimidine Dehydrogenase (DPD) and Toxicity of 5-Fluorouracil

An ongoing example for the impact of genetic polymorphisms on toxic effects of therapy regimes is 5-fluorouracil (5-FU). 5-FU is a chemotherapeutic used to treat enterohepatic and gynaecological cancers. Adverse effects, such as mucositis, neutropenia, cardiac and neuronal toxicity, have been linked to deficiency of the liver enzyme dihydropyrimidine dehydrogenase (DPD), which is the rate-limiting step of 5-fluorouracil (5-FU) metabolism (Baek et al., 2006; van Kuilenburg et al., 2004). The clinical use of preventive screening for DPD mutations has recently been discussed for 5-FU (van Kuilenburg, 2006) and oral fluoropyrimidine capecitabine, which also is suspicious for severe toxicity in DPD-deficient cancer patients (Ciccolini et al., 2006; Saif and Diasio, 2006). A tragic example of the individual impact of genetic polymorphisms in drug metabolism is a case report about a pharmacogenetic syndrome, in which two genetic polymorphisms


253

collided in a patient treated for colorectal adenocarcinoma, where the combined therapy of 5-fluorouracil and irinotecan had a fatal outcome (Steiner et al., 2005). Lethal toxicity arose from heterozygosity and consecutive reduced activity of two drug-metabolizing enzymes, DPD and uridine diphosphate glucuronosyltransferase 1A1 (UGT1A1), which induced an impaired glucuronidation activity (Gilbert’s syndrome) and was linked to toxicity in the treatment with irinotecan (Steiner et al., 2005). Introduction of routine screening for currently known drug-metabolizing enzymes that are important for drug toxicity, e.g. TPMT, UGT1A1 and DPD, prior to treatment may help to prevent cases of severe toxicity under chemotherapeutic drug treatment. The knowledge that a specific defect may cause drug-related organ toxicity allows preventive strategies to be introduced that help to avoid adverse effects, for instance, by pre-selecting suitable patients for the treatment, but also enables the development of new approaches to come up against adverse effects, e.g. by enhancing gene expression of the enzyme in question (Shimizu et al., 2001; Yamaguchi et al., 2006).

9.7

Conclusions

In summary, several mechanisms are involved in the onset and progression of toxic liver injury. In particular, the initiating parenchymal damage is most likely to result from drugor toxin-specific alterations of the cellular homeostasis. These mechanisms depend on the concentration of the toxin and arise from direct interaction with cellular functions, e.g. by interference with functions of synthesis and integrity, xenobiotic metabolism, as well as transcription and transporter activities. It is one of the greatest challenges of the future to elucidate the true origin of the functional alteration induced by certain toxins. On the other hand, a deep understanding of the secondary events involved in toxic liver injury and the stress-related signalling pathways is also necessary for the development of new therapeutic approaches to limit the progression of liver injury and prevent acute liver failure. The key targets for future therapeutic approaches are namely stress-related JNK-p38, the TNF-α and NK-κB pathways, as well as the complement and apoptotic signalling cascades and their organelles of origin, such as the mitochondrion, the lysosome and the peroxisome. Furthermore, addressing the sources of extracellular and intracellular stress, such as the ROS-generating machinery and expression of pro-inflammatory cytokines, may bear a promising approach to encounter liver injury. Nevertheless, it is of never-ending importance that toxicity is as a function of toxin concentration and exposure time. Therefore, the best therapeutic strategy is prevention of toxic liver injury, for which it is necessary to gain further knowledge about mechanisms of drug interactions, the role of individual gene expression and liver regeneration in the susceptibility to liver toxicity.

References Adachi M, Suematsu S, Kondo T, et al. (1995). Targeted mutation in the Fas gene causes hyperplasia in peripheral lymphoid organs and liver. Nat Genet 11:294–300. Al Dosari MS, Knapp JE and Liu D (2006). Activation of human CYP2C9 promoter and regulation by CAR and PXR in mouse liver. Mol Pharm 3:322–328.

254

Hepatotoxicity

Albermann N, Schmitz-Winnenthal FH, Z’graggen K, et al. (2005). Expression of the drug transporters MDR1/ABCB1, MRP1/ABCC1, MRP2/ABCC2, BCRP/ABCG2 and PXR in peripheral blood mononuclear cells and their relationship with the expression in intestine and liver. Biochem Pharmacol 70:949–958. Allen M, Svensson L, Roach M, Hambor J, McNeish J and Gabel CA (2000). Deficiency of the stress kinase p38alpha results in embryonic lethality: characterization of the kinase dependence of stress responses of enzyme-deficient embryonic stem cells. J Exp Med 191:859–870. Ananthanarayanan M, Balasubramanian N, Makishima M, Mangelsdorf DJ and Suchy FJ (2001). Human bile salt export pump promoter is transactivated by the farnesoid X receptor/bile acid receptor. J Biol Chem 276:28857–28865. Anapolsky A, Teng S, Dixit S and Piquette-Miller M (2006). The role of pregnane X receptor in 2-acetylaminofluorene-mediated induction of drug transport and -metabolizing enzymes in mice. Drug Metab Dispos 34:405–409. Anderson N and Borlak J (2006). Drug-induced phospholipidosis. FEBS Lett 580:5533–5540. Anderson SP, Dunn C, Laughter A, et al. (2004). Overlapping transcriptional programs regulated by the nuclear receptors peroxisome proliferator-activated receptor alpha, retinoid X receptor and liver X receptor in mouse liver. Mol Pharmacol 66:1440–1452. Andersson BS, Rundgren M, Nelson SD and Harder S (1990). N -acetyl- p-benzoquinone imineinduced changes in the energy metabolism in hepatocytes. Chem Biol Interact 75:201–211. Andrysik Z, Machala M, Chramostova K, Hofmanova J, Kozubik A and Vondracek J (2006). Activation of ERK1/2 and p38 kinases by polycyclic aromatic hydrocarbons in rat liver epithelial cells is associated with induction of apoptosis. Toxicol Appl Pharmacol 211:198–208. Antes TJ, Chen J, Cooper AD and Levy-Wilson B (2000). The nuclear matrix protein CDP represses hepatic transcription of the human cholesterol-7alpha hydroxylase gene. J Biol Chem 275:26649– 26660. Araya Z and Wikvall K (1999). 6alpha-hydroxylation of taurochenodeoxycholic acid and lithocholic acid by CYP3A4 in human liver microsomes. Biochim Biophys Acta 1438:47–54. Arenson DM, Friedman SL and Bissell DM (1988). Formation of extracellular matrix in normal rat liver: lipocytes as a major source of proteoglycan. Gastroenterology 95:441–447. Ashwell JD (2006). The many paths to p38 mitogen-activated protein kinase activation in the immune system. Nat Rev Immunol 6:532–540. Attri S, Rana SV, Vaiphei K. et al. (2000). Isoniazid- and rifampicin-induced oxidative hepatic injury – protection by N -acetylcysteine. Hum Exp Toxicol 19:517–522. Awad MM, Enslen H, Boylan JM, Davis RJ and Gruppuso PA (2000). Growth regulation via p38 mitogen-activated protein kinase in developing liver. J Biol Chem 275:38716–38721. Bachs L, Pares A, Elena M, Piera C and Rodes J (1989). Comparison of rifampicin with phenobarbitone for treatment of pruritus in biliary cirrhosis. Lancet 1:574–576. Baek JH, Kim JG, Kim SN, et al. (2006). Unpredicted severe toxicity after 5-fluorouracil treatment due to dihydropyrimidine dehydrogenase deficiency. Kor J Intern Med 21:43–45. Baichwal VR and Baeuerle PA (1997). Activate NF-kappa B or die? Curr Biol 7:R94–R96. Bajt ML, Ho YS, Vonderfecht SL and Jaeschke H (2002). Reactive oxygen as modulator of TNF and fas receptor-mediated apoptosis in vivo: studies with glutathione peroxidase-deficient mice. Antioxid Redox Signal 4:733–740. Barreto R, Kawakita S, Tsuchiya J, et al. (2005). Metal-induced oxidative damage in cultured hepatocytes and hepatic lysosomal fraction: beneficial effect of a curcumin/absinthium compound. Chin J Dig Dis 6:31–36. Bartsch H and Nair J (2004). Oxidative stress and lipid peroxidation-derived DNA-lesions in inflammation driven carcinogenesis. Cancer Detect Prev 28:385–391. Basaran A, Erol K, Basaran N, et al. (1993). Effects of ciprofloxacin on chromosomes, and hepatic and renal functions in rats. Chemotherapy 39:182–188.


255

Baskin-Bey ES, Canbay A, Bronk SF, et al. (2005). Cathepsin B inactivation attenuates hepatocyte apoptosis and liver damage in steatotic livers after cold ischemia–warm reperfusion injury. Am J Physiol Gastrointest Liver Physiol 288:G396–G402. Bastida G, Nos P, Aguas M, et al. (2005). Incidence, risk factors and clinical course of thiopurineinduced liver injury in patients with inflammatory bowel disease. Aliment Pharmacol Ther 22:775– 782. Bauer I, Vollmar B, Jaeschke H, et al. (2000). Transcriptional activation of heme oxygenase-1 and its functional significance in acetaminophen-induced hepatitis and hepatocellular injury in the rat. J Hepatol 33:395–406. Bauer J, Bauer TM, Kalb T, et al. (1989). Regulation of interleukin 6 receptor expression in human monocytes and monocyte-derived macrophages. Comparison with the expression in human hepatocytes. J Exp Med 170:1537–1549. Bayer AS, Li C and Ing M (1998). Efficacy of trovafloxacin, a new quinolone antibiotic, in experimental staphylococcal endocarditis due to oxacillin-resistant strains. Antimicrob Agents Chemother 42:1837–1841. Beckman JS (1996). Oxidative damage and tyrosine nitration from peroxynitrite. Chem Res Toxicol 9:836–844. Bedoucha M, Atzpodien E and Boelsterli UA (2001). Diabetic KKAy mice exhibit increased hepatic PPARgamma1 gene expression and develop hepatic steatosis upon chronic treatment with antidiabetic thiazolidinediones. J Hepatol 35:17–23. Behr TM, Berova M, Doe CP, et al. (2003). p38 mitogen-activated protein kinase inhibitors for the treatment of chronic cardiovascular disease. Curr Opin Investig Drugs 4:1059–1064. Beigneux AP, Moser AH, Shigenaga JK, Grunfeld C and Feingold KR (2002). Reduction in cytochrome P-450 enzyme expression is associated with repression of CAR (constitutive androstane receptor) and PXR (pregnane X receptor) in mouse liver during the acute phase response. Biochem Biophys Res Commun 293:145–149. Bellezzo JM, Leingang KA, Bulla GA, Britton RS, Bacon BR and Fox ES (1998). Modulation of lipopolysaccharide-mediated activation in rat Kupffer cells by antioxidants. J Lab Clin Med 131:36–44. Belloc C, Gauffre A, Andre C and Beaune PH (1997). Epitope mapping of human CYP1A2 in dihydralazine-induced autoimmune hepatitis. Pharmacogenetics 7:181–186. Bellomo G, Mirabelli F, Richelmi P, Malorni W, Iosi F and Orrenius S (1990). The cytoskeleton as a target in quinone toxicity. Free Radic Res Commun 8:391–399. Bennett M, Macdonald K, Chan SW, Luzio JP, Simari R and Weissberg P (1998). Cell surface trafficking of Fas: a rapid mechanism of p53-mediated apoptosis. Science 282:290–293. Berlett BS and Stadtman ER (1997). Protein oxidation in aging, disease, and oxidative stress. J Biol Chem 272:20313–20316. Bernt C, Vennegeerts T, Beuers U and Rust C (2006). The human transcription factor AP-1 is a mediator of bile acid-induced liver cell apoptosis. Biochem Biophys Res Commun 340:800–806. Bertilsson G, Heidrich J, Svensson K, et al. (1998). Identification of a human nuclear receptor defines a new signaling pathway for CYP3A induction. Proc Natl Acad Sci USA 95:12208–12213. Bertino J Jr and Fish D (2000). The safety profile of the fluoroquinolones. Clin Ther 22:798–817. Bessems JG and Vermeulen NP (2001). Paracetamol (acetaminophen)-induced toxicity: molecular and biochemical mechanisms, analogues and protective approaches. Crit Rev Toxicol 31:55–138. Bissell DM, Wang SS, Jarnagin WR and Roll FJ (1995). Cell-specific expression of transforming growth factor-beta in rat liver. Evidence for autocrine regulation of hepatocyte proliferation. J Clin Invest 96:447–455. Blazka ME, Germolec DR, Simeonova P, Bruccoleri A, Pennypacker KR and Luster MI (1995). Acetaminophen-induced hepatotoxicity is associated with early changes in NF-kB and NF-IL6 DNA binding activity. J Inflamm 47:138–150.

256

Hepatotoxicity

Bleau AM, Fradette C, El Kadi AO and Cote MC du SP (2001). Cytochrome P450 down-regulation by serum from humans with a viral infection and from rabbits with an inflammatory reaction. Drug Metab Dispos 29:1007–1012. Bloom D, Dhakshinamoorthy S and Jaiswal AK (2002). Site-directed mutagenesis of cysteine to serine in the DNA binding region of Nrf2 decreases its capacity to upregulate antioxidant response element-mediated expression and antioxidant induction of NAD(P)H:quinone oxidoreductase1 gene. Oncogene 21:2191–2200. Blum MD, Graham DJ and McCloskey CA (1994). Temafloxacin syndrome: review of 95 cases. Clin Infect Dis 18:946–950. Blumberg B, Sabbagh W Jr, Juguilon H, et al. (1998). SXR, a novel steroid and xenobiotic-sensing nuclear receptor. Genes Dev 12:3195–3205. Boden G, Homko C, Mozzoli M, Showe LC, Nichols C and Cheung P (2005). Thiazolidinediones upregulate fatty acid uptake and oxidation in adipose tissue of diabetic patients. Diabetes 54:880– 885. Boelsterli UA and Bedoucha M (2002). Toxicological consequences of altered peroxisome proliferator-activated receptor gamma (PPARgamma) expression in the liver: insights from models of obesity and type 2 diabetes. Biochem Pharmacol 63:1–10. Boess F, Bopst M, Althaus R, et al. (1998). Acetaminophen hepatotoxicity in tumor necrosis factor/lymphotoxin-alpha gene knockout mice. Hepatology 27:1021–1029. Bohr VA, Stevnsner T and Souza-Pinto NC (2002). Mitochondrial DNA repair of oxidative damage in mammalian cells. Gene 286:127–134. Bondoc FY, Bao Z, Hu WY, et al. (1999). Acetone catabolism by cytochrome P450 2E1: studies with CYP2E1-null mice. Biochem Pharmacol 58:461–463. Bonierbale E, Valadon P, Pons C, Desfosses B, Dansette PM and Mansuy D (1999). Opposite behaviors of reactive metabolites of tienilic acid and its isomer toward liver proteins: use of specific antitienilic acid–protein adduct antibodies and the possible relationship with different hepatotoxic effects of the two compounds. Chem Res Toxicol 12:286–296. Bonnesen C, Eggleston IM and Hayes JD (2001). Dietary indoles and isothiocyanates that are generated from cruciferous vegetables can both stimulate apoptosis and confer protection against DNA damage in human colon cell lines. Cancer Res 61:6120–6130. Bourdi M, Larrey D, Nataf J, et al. (1990). Anti-liver endoplasmic reticulum autoantibodies are directed against human cytochrome P-450IA2. A specific marker of dihydralazine-induced hepatitis. J Clin Invest 85:1967–1973. Bourdi M, Chen W, Peter RM, et al. (1996). Human cytochrome P450 2E1 is a major autoantigen associated with halothane hepatitis. Chem Res Toxicol 9:1159–1166. Bradham CA, Plumpe J, Manns MP, Brenner DA and Trautwein C (1998). Mechanisms of hepatic toxicity. I. TNF-induced liver injury. Am J Physiol 275:G387–G392. Branger J, van den BB, Weijer S, et al. (2002). Anti-inflammatory effects of a p38 mitogen-activated protein kinase inhibitor during human endotoxemia. J Immunol 168:4070–4077. Breen DP, Marinaki AM, Arenas M and Hayes PC (2005). Pharmacogenetic association with adverse drug reactions to azathioprine immunosuppressive therapy following liver transplantation. Liver Transpl 11:826–833. Bremmelgaard A and Sjovall J (1979). Bile acid profiles in urine of patients with liver diseases. Eur J Clin Invest 9:341–348. Brill S, Zvibel I, Halpern Z and Oren R (2002). The role of fetal and adult hepatocyte extracellular matrix in the regulation of tissue-specific gene expression in fetal and adult hepatocytes. Eur J Cell Biol 81:43–50. Britton RS (1996). Metal-induced hepatotoxicity. Semin Liver Dis 16:3–12. Britton RS, O’Neill R and Bacon BR (1990). Hepatic mitochondrial malondialdehyde metabolism in rats with chronic iron overload. Hepatology 11:93–97.


257

Britton RS, Ramm GA, Olynyk J, Singh R, O’Neill R and Bacon BR (1994). Pathophysiology of iron toxicity. Adv Exp Med Biol 356:239–353. Brunnberg S, Andersson P, Lindstam M, Paulson I, Poellinger L and Hanberg A (2006). The constitutively active Ah receptor (CA-AhR) mouse as a potential model for dioxin exposure – Effects in vital organs. Toxicology 224: 191–201. Bubici C, Papa S, Pham CG, Zazzeroni F and Franzoso G (2006). The NF-kappaB-mediated control of ROS and JNK signaling. Histol Histopathol 21:69–80. Buck M, Turler H and Chojkier M (1994). LAP (NF-IL-6), a tissue-specific transcriptional activator, is an inhibitor of hepatoma cell proliferation. EMBO J 13:851–860. Buck M, Poli V, van der GP, Chojkier M and Hunter T (1999). Phosphorylation of rat serine 105 or mouse threonine 217 in C/EBP beta is required for hepatocyte proliferation induced by TGF alpha. Mol Cell 4:1087–1092. Buck M, Zhang L, Halasz NA, Hunter T and Chojkier M (2001a). Nuclear export of phosphorylated C/EBPbeta mediates the inhibition of albumin expression by TNF-alpha. EMBO J 20:6712–6723. Buck M, Poli V, Hunter T and Chojkier M (2001b). C/EBPbeta phosphorylation by RSK creates a functional XEXD caspase inhibitory box critical for cell survival. Mol Cell 8:807–816. Buda F, Ensing B, Gribnau MC and Baerends EJ (2003). O2 evolution in the Fenton reaction. Chemistry 9:3436–4444. Burcham PC and Harman AW (1991). Acetaminophen toxicity results in site-specific mitochondrial damage in isolated mouse hepatocytes. J Biol Chem 266:5049–5054. Burk RF, Hill KE, Hunt RW Jr and Martin AE (1990). Isoniazid potentiation of acetaminophen hepatotoxicity in the rat and 4-methylpyrazole inhibition of it. Res Commun Chem Pathol Pharmacol 69:115–118. Butterfield DA and Kanski J (2001). Brain protein oxidation in age-related neurodegenerative disorders that are associated with aggregated proteins. Mech Ageing Dev 122:945–962. Byrne JA, Strautnieks SS, Mieli-Vergani G, Higgins CF, Linton KJ and Thompson RJ (2002). The human bile salt export pump: characterization of substrate specificity and identification of inhibitors. Gastroenterology 123:1649–1658. Caballero F, Meiss R, Gimenez A, Batlle A and Vazquez E (2004). Immunohistochemical analysis of heme oxygenase-1 in preneoplastic and neoplastic lesions during chemical hepatocarcinogenesis. Int J Exp Pathol 85:213–222. Camacho IA, Singh N, Hegde VL, Nagarkatti M and Nagarkatti PS (2005). Treatment of mice with 2,3,7,8-tetrachlorodibenzo- p-dioxin leads to aryl hydrocarbon receptor-dependent nuclear translocation of NF-kappaB and expression of Fas ligand in thymic stromal cells and consequent apoptosis in T cells. J Immunol 175:90–103. Canbay A, Higuchi H, Bronk SF, Taniai M, Sebo TJ and Gores GJ (2002). Fas enhances fibrogenesis in the bile duct ligated mouse: a link between apoptosis and fibrosis. Gastroenterology 123:1323– 1330. Canbay A, Feldstein AE, Higuchi H, et al. (2003). Kupffer cell engulfment of apoptotic bodies stimulates death ligand and cytokine expression. Hepatology 38:1188–1198. Canbay A, Feldstein A, Baskin-Bey E, Bronk SF and Gores GJ (2004). The caspase inhibitor IDN6556 attenuates hepatic injury and fibrosis in the bile duct ligated mouse. J Pharmacol Exp Ther 308:1191–1196. Cancado EL, Leitao RM, Carrilho FJ and Laudanna AA (1998). Unexpected clinical remission of cholestasis after rifampicin therapy in patients with normal or slightly increased levels of gammaglutamyl transpeptidase. Am J Gastroenterol 93:1510–1517. Carlson SG, Fawcett TW, Bartlett JD, Bernier M and Holbrook NJ (1993). Regulation of the C/EBPrelated gene gadd153 by glucose deprivation. Mol Cell Biol 13:4736–4744. Castello L, Marengo B, Nitti M, et al. (2005). 4-Hydroxynonenal signalling to apoptosis in isolated rat hepatocytes: the role of PKC-delta. Biochim Biophys Acta 1737:83–93.

258

Hepatotoxicity

Castilho RF, Kowaltowski AJ, Meinicke AR, Bechara EJ and Vercesi AE (1995). Permeabilization of the inner mitochondrial membrane by Ca2+ ions is stimulated by t-butyl hydroperoxide and mediated by reactive oxygen species generated by mitochondria. Free Radic Biol Med 18: 479–486. Cha BS, Ciaraldi TP, Park KS, Carter L, Mudaliar SR and Henry RR (2005). Impaired fatty acid metabolism in type 2 diabetic skeletal muscle cells is reversed by PPARgamma agonists. Am J Physiol Endocrinol Metab 289:E151–E159. Chai J, Du C, Wu JW, Kyin S, Wang X and Shi Y (2000). Structural and biochemical basis of apoptotic activation by Smac/DIABLO. Nature 406:855–862. Chan H, Bartos DP and Owen-Schaub LB (1999). Activation-dependent transcriptional regulation of the human Fas promoter requires NF-kappaB p50-p65 recruitment. Mol Cell Biol 19:20982108. Chen J, Schenker S, Frosto TA and Henderson GI (1998). Inhibition of cytochrome c oxidase activity by 4-hydroxynonenal (HNE). Role of HNE adduct formation with the enzyme subunits. Biochim Biophys Acta 1380:336–344. Chen C, Hennig GE, Whiteley HE, Corton JC and Manautou JE (2000). Peroxisome proliferatoractivated receptor alpha-null mice lack resistance to acetaminophen hepatotoxicity following clofibrate exposure. Toxicol Sci 57:338–344. Chen Y, Ferguson SS, Negishi M and Goldstein JA (2003). Identification of constitutive androstane receptor and glucocorticoid receptor binding sites in the CYP2C19 promoter. Mol Pharmacol 64:316–324. Chen F, Ananthanarayanan M, Emre S, et al. (2004a). Progressive familial intrahepatic cholestasis, type 1, is associated with decreased farnesoid X receptor activity. Gastroenterology 126:756–764. Chen Y, Ferguson SS, Negishi M and Goldstein JA (2004b). Induction of human CYP2C9 by rifampicin, hyperforin, and phenobarbital is mediated by the pregnane X receptor. J Pharmacol Exp Ther 308:495–501. Cheng X, Maher J, Dieter MZ and Klaassen CD (2005). Regulation of mouse organic aniontransporting polypeptides (Oatps) in liver by prototypical microsomal enzyme inducers that activate distinct transcription factor pathways. Drug Metab Dispos 33:1276–1282. Cherrington NJ, Hartley DP, Li N, Johnson DR and Klaassen CD (2002). Organ distribution of multidrug resistance proteins 1, 2, and 3 (Mrp1, 2, and 3) mRNA and hepatic induction of Mrp3 by constitutive androstane receptor activators in rats. J Pharmacol Exp Ther 300:97–104. Chiarpotto E, Castello L, Leonarduzzi G, Biasi AF and Poli G (2005). Role of 4-hydroxy-2,3-nonenal in the pathogenesis of fibrosis. Biofactors 24:229–236. Chiu H, Gardner CR, Dambach DM, et al. (2003a). Role of p55 tumor necrosis factor receptor 1 in acetaminophen-induced antioxidant defense. Am J Physiol Gastrointest Liver Physiol 285:G959– G966. Chiu H, Gardner CR, Dambach DM, et al. (2003b). Role of tumor necrosis factor receptor 1 (p55) in hepatocyte proliferation during acetaminophen-induced toxicity in mice. Toxicol Appl Pharmacol 193:218–227. Chowdhury A, Santra A, Bhattacharjee K, Ghatak S, Saha DR and Dhali GK (2006). Mitochondrial oxidative stress and permeability transition in Isoniazid and Rifampicin induced liver injury in mice. J Hepatol 45:117–126. Ciccolini J, Mercier C, Dahan L, et al. (2006). Toxic death-case after capecitabine + oxaliplatin (XELOX) administration: probable implication of dihydropyrimidine deshydrogenase deficiency. Cancer Chemother Pharmacol 58:272–275. Cirillo PF, Pargellis C and Regan J (2002). The non-diaryl heterocycle classes of p38 MAP kinase inhibitors. Curr Top Med Chem 2:1021–1035. Cisternas FA, Tapia G, Arredondo M, et al. (2005). Early histological and functional effects of chronic copper exposure in rat liver. Biometals 18:541–551.


259

Clark AR, Dean JL and Saklatvala J (2003). Post-transcriptional regulation of gene expression by mitogen-activated protein kinase p38. FEBS Lett 546:37–44. Clement MV and Stamenkovic I (1996). Superoxide anion is a natural inhibitor of FAS-mediated cell death. EMBO J 15:216–225. Clement MV, Sivarajah S and Pervaiz S (2005). Production of intracellular superoxide mediates dithiothreitol-dependent inhibition of apoptotic cell death. Antioxid Redox Signal 7:456–464. Connor HD, Thurman RG, Galizi MD and Mason RP (1986). The formation of a novel free radical metabolite from CCl4 in the perfused rat liver and in vivo. J Biol Chem 261:4542–4548. Constan AA, Sprankle CS, Peters JM, et al. (1999). Metabolism of chloroform by cytochrome P450 2E1 is required for induction of toxicity in the liver, kidney, and nose of male mice. Toxicol Appl Pharmacol 160:120–126. Corcoran GB and Wong BK (1986). Role of glutathione in prevention of acetaminophen-induced hepatotoxicity by N -acetyl-l-cysteine in vivo: studies with N -acetyl-d-cysteine in mice. J Pharmacol Exp Ther 238:54–61. Corton JC, Lapinskas PJ and Gonzalez FJ (2000). Central role of PPARalpha in the mechanism of action of hepatocarcinogenic peroxisome proliferators. Mutat Res 448:139–151. Cover C, Fickert P, Knight TR, et al. (2005). Pathophysiological role of poly(ADP-ribose) polymerase (PARP) activation during acetaminophen-induced liver cell necrosis in mice. Toxicol Sci 84:201– 208. Cory S, Huang DC and Adams JM (2003). The Bcl-2 family: roles in cell survival and oncogenesis. Oncogene 22:8590–8607. Cuda G, Paterno R, Ceravolo R, et al. (2002). Protection of human endothelial cells from oxidative stress: role of Ras-ERK1/2 signaling. Circulation 105:968–974. Cullen JM (2005). Mechanistic classification of liver injury. Toxicol Pathol 33:6–8. Czaja MJ, Xu J, Ju Y, Alt E and Schmiedeberg P (1994). Lipopolysaccharide-neutralizing antibody reduces hepatocyte injury from acute hepatotoxin administration. Hepatology 19:1282– 1289. Dalle-Donne I, Scaloni A, Giustarini D, et al. (2005). Proteins as biomarkers of oxidative/nitrosative stress in diseases: the contribution of redox proteomics. Mass Spectrom Rev 24:55–99. Dalton TP, Shertzer HG and Puga A (1999). Regulation of gene expression by reactive oxygen. Annu Rev Pharmacol Toxicol 39:67–101. Dambach DM (2005). Potential adverse effects associated with inhibition of p38alpha/beta MAP kinases. Curr Top Med Chem 5:929–939. Dartsch PC, Hildenbrand S, Kimmel R and Schmahl FW (1998). Investigations on the nephrotoxicity and hepatotoxicity of trivalent and hexavalent chromium compounds. Int Arch Occup Environ Health 71(Suppl):S40–S45. Davis BH, Chen A and Beno DW (1996). Raf and mitogen-activated protein kinase regulate stellate cell collagen gene expression. J Biol Chem 271:11039–11042. Davis DC, Potter WZ, Jollow DJ and Mitchell JR (1974). Species differences in hepatic glutathione depletion, covalent binding and hepatic necrosis after acetaminophen. Life Sci 14:2099–2109. Deaciuc IV, D’Souza NB and Spitzer JJ (1995). Tumor necrosis factor-alpha cell-surface receptors of liver parenchymal and nonparenchymal cells during acute and chronic alcohol administration to rats. Alcohol Clin Exp Res 19:332–338. Dean RT, Fu S, Stocker R and Davies MJ (1997). Biochemistry and pathology of radical-mediated protein oxidation. Biochem J 324:1–18. Delaunay A, Isnard AD and Toledano MB (2000). H2 O2 sensing through oxidation of the Yap1 transcription factor. EMBO J 19:5157–5166. Denson LA, Auld KL, Schiek DS, McClure MH, Mangelsdorf DJ and Karpen SJ (2000). Interleukin1beta suppresses retinoid transactivation of two hepatic transporter genes involved in bile formation. J Biol Chem 275:8835–8843.

260

Hepatotoxicity

Deschatrette J and Weiss MC (1974). Characterization of differentiated and dedifferentiated clones from a rat hepatoma. Biochimie 56:1603–1611. Desta Z, Soukhova NV and Flockhart DA (2001). Inhibition of cytochrome P450 (CYP450) isoforms by isoniazid: potent inhibition of CYP2C19 and CYP3A. Antimicrob Agents Chemother 45:382– 392. Di Consiglio E, De Angelis G, Testai E and Vittozzi L (2001). Correlation of a specific mitochondrial phospholipid–phosgene adduct with chloroform acute toxicity. Toxicology 159:43–53. Diehl AM, Yin M, Fleckenstein J, et al. (1994). Tumor necrosis factor-alpha induces c-jun during the regenerative response to liver injury. Am J Physiol 267:G552–G561. Dietrich CG, Ottenhoff R, de Waart DR and Oude Elferink RP (2001). Role of MRP2 and GSH in intrahepatic cycling of toxins. Toxicology 167:73–81. Diliberto JJ, Burgin D and Birnbaum LS (1997). Role of CYP1A2 in hepatic sequestration of dioxin: studies using CYP1A2 knock-out mice. Biochem Biophys Res Commun 236:431–433. Dobbs BR, Rogers GW, Xing HY and Fraser R (1994). Endotoxin-induced defenestration of the hepatic sinusoidal endothelium: a factor in the pathogenesis of cirrhosis? Liver 14:230–233. Dodeller F, Skapenko A, Kalden JR, Lipsky PE and Schulze-Koops H (2005). The p38 mitogenactivated protein kinase regulates effector functions of primary human CD4 T cells. Eur J Immunol 35:3631–3642. Dong W, Simeonova PP, Gallucci R, et al. (1998a). Cytokine expression in hepatocytes: role of oxidant stress. J Interferon Cytokine Res 18:629–38. Dong W, Simeonova PP, Gallucci R, et al. (1998b). Toxic metals stimulate inflammatory cytokines in hepatocytes through oxidative stress mechanisms. Toxicol Appl Pharmacol 151:359–366. Douette P and Sluse FE (2006). Mitochondrial uncoupling proteins: new insights from functional and proteomic studies. Free Radic Biol Med 40:1097–1107. Dresser GK, Spence JD and Bailey DG (2000). Pharmacokinetic–pharmacodynamic consequences and clinical relevance of cytochrome P450 3A4 inhibition. Clin Pharmacokinet 38:41–57. Drocourt L, Pascussi JM, Assenat E, Fabre JM, Maurel P and Vilarem MJ (2001). Calcium channel modulators of the dihydropyridine family are human pregnane X receptor activators and inducers of CYP3A, CYP2B, and CYP2C in human hepatocytes. Drug Metab Dispos 29:1325–1331. Du C, Fang M, Li Y, Li L and Wang X (2000). Smac, a mitochondrial protein that promotes cytochrome c-dependent caspase activation by eliminating IAP inhibition. Cell 102:33–42. Dumasia R, Eagle KA, Kline-Rogers E, May N, Cho L and Mukherjee D (2005). Role of PPARgamma agonist thiazolidinediones in treatment of pre-diabetic and diabetic individuals: a cardiovascular perspective. Curr Drug Targets Cardiovasc Haematol Disord 5:377–386. Duncan SA, Navas MA, Dufort D, Rossant J and Stoffel M (1998). Regulation of a transcription factor network required for differentiation and metabolism. Science 281:692–695. Durr D, Stieger B, Kullak-Ublick GA, et al. (2000). St John’s Wort induces intestinal Pglycoprotein/MDR1 and intestinal and hepatic CYP3A4. Clin Pharmacol Ther 68:598–604. Dussault I, Lin M, Hollister K, Wang EH, Synold TW and Forman BM (2001). Peptide mimetic HIV protease inhibitors are ligands for the orphan receptor SXR. J Biol Chem 276:33309–33312. Dwivedi S, Sharma R, Sharma A, et al. (2006). The course of CCl4 induced hepatotoxicity is altered in mGSTA4-4 null (−/−) mice. Toxicology 218:58–66. El Hadri L, Chanas B and Ghanayem BI (2005). Comparative metabolism of methacrylonitrile and acrylonitrile to cyanide using cytochrome P4502E1 and microsomal epoxide hydrolase-null mice. Toxicol Appl Pharmacol 205:116–125. El Kadi AO du SP (1998). Depression of the hepatic cytochrome P450 by an acute inflammatory reaction: characterization of the nature of mediators in human and rabbit serum, and in the liver. Life Sci 63:1361–1370. Elba S, Buongiorno GP, Caruso ML, Noviello MR and Manghisi OG (2002). Main characteristics of hepatocellular carcinoma and cirrhosis and therapeutic approaches. Curr Pharm Des 8:1007–1011.


261

Enomoto A, Itoh K, Nagayoshi E, et al. (2001). High sensitivity of Nrf2 knockout mice to acetaminophen hepatotoxicity associated with decreased expression of ARE-regulated drug metabolizing enzymes and antioxidant genes. Toxicol Sci 59:169–177. Ezendam J, Staedtler F, Pennings J, et al. (2004). Toxicogenomics of subchronic hexachlorobenzene exposure in Brown Norway rats. Environ Health Perspect 112:782–791. Falkner KC, Pinaire JA, Xiao GH, Geoghegan TE and Prough RA (2001). Regulation of the rat glutathione S-transferase A2 gene by glucocorticoids: involvement of both the glucocorticoid and pregnane X receptors. Mol Pharmacol 60:611–619. Fallone F, Villard PH, Seree E, et al. (2004). Retinoids repress Ah receptor CYP1A1 induction pathway through the SMRT corepressor. Biochem Biophys Res Commun 322:551–556. Farrell J, Naisbitt DJ, Drummond NS, et al. (2003). Characterization of sulfamethoxazole and sulfamethoxazole metabolite-specific T-cell responses in animals and humans. J Pharmacol Exp Ther 306:229–237. Faubion WA and Gores GJ (1999). Death receptors in liver biology and pathobiology. Hepatology 29:1–4. Faubion WA, Guicciardi ME, Miyoshi H, et al. (1999). Toxic bile salts induce rodent hepatocyte apoptosis via direct activation of Fas. J Clin Invest 103:137–145. Faust DM, Boshart M, Imaizumi-Scherrer T, Schutz G and Weiss MC (1994). Constancy of expression of the protein kinase A regulatory subunit R1 alpha in hepatoma cell lines of different phenotypes. Cell Growth Differ 5:47–53. Favreau LV and Pickett CB (1991). Transcriptional regulation of the rat NAD(P)H:quinone reductase gene. Identification of regulatory elements controlling basal level expression and inducible expression by planar aromatic compounds and phenolic antioxidants. J Biol Chem 266:4556– 4561. Feldstein AE, Canbay A, Angulo P, et al. (2003a). Hepatocyte apoptosis and fas expression are prominent features of human nonalcoholic steatohepatitis. Gastroenterology 125:437–443. Feldstein AE, Canbay A, Guicciardi ME, Higuchi H, Bronk SF and Gores GJ (2003b). Diet associated hepatic steatosis sensitizes to Fas mediated liver injury in mice. J Hepatol 39:978–983. Feldstein AE, Werneburg NW, Canbay A, et al. (2004). Free fatty acids promote hepatic lipotoxicity by stimulating TNF-alpha expression via a lysosomal pathway. Hepatology 40:185–194. Feldstein AE, Werneburg NW, Li Z, Bronk SF and Gores GJ (2006). Bax inhibition protects against free fatty acid-induced lysosomal permeabilization. Am J Physiol Gastrointest Liver Physiol 290:G1339–G1346. Fengler JD, Baumgarten R, Eike E, et al. (1993). Genetic polymorphism in Gilbert-Meulengracht syndrome (GMS). Z Gastroenterol 31(Suppl 2):81–82. Ferguson SS, Chen Y, LeCluyse EL, Negishi M and Goldstein JA (2005). Human CYP2C8 is transcriptionally regulated by the nuclear receptors constitutive androstane receptor, pregnane X receptor, glucocorticoid receptor, and hepatic nuclear factor 4alpha. Mol Pharmacol 68:747–757. Fernandez-Checa JC and Kaplowitz N (2005). Hepatic mitochondrial glutathione: transport and role in disease and toxicity. Toxicol Appl Pharmacol 204:263–273. Fernandez-Salguero PM, Hilbert DM, Rudikoff S, Ward JM and Gonzalez FJ (1996). Arylhydrocarbon receptor-deficient mice are resistant to 2,3,7,8-tetrachlorodibenzo- p-dioxin-induced toxicity. Toxicol Appl Pharmacol 140:173–179. Fickert P, Fuchsbichler A, Marschall HU, et al. (2006). Lithocholic acid feeding induces segmental bile duct obstruction and destructive cholangitis in mice. Am J Pathol 168:410–422. Fievet C, Fruchart JC and Staels B (2006). PPARalpha and PPARgamma dual agonists for the treatment of type 2 diabetes and the metabolic syndrome. Curr Opin Pharmacol 6:606–614. Fiorucci S, Clerici C, Antonelli E, et al. (2005). Protective effects of 6-ethyl chenodeoxycholic acid, a farnesoid X receptor ligand, in estrogen-induced cholestasis. J Pharmacol Exp Ther 313:604– 612.

262

Hepatotoxicity

Fontana L, Jerez D, Rojas-Valencia L, Solis-Herruzo JA, Greenwel P and Rojkind M (1997). Ethanol induces the expression of alpha 1(I) procollagen mRNA in a co-culture system containing a liver stellate cell-line and freshly isolated hepatocytes. Biochim Biophys Acta 1362:135–144. Forman BM, Goode E, Chen J, et al. (1995). Identification of a nuclear receptor that is activated by farnesol metabolites. Cell 81:687–693. Fox ES and Leingang KA (1998). Inhibition of LPS-mediated activation in rat Kupffer cells by N -acetylcysteine occurs subsequent to NF-kappaB translocation and requires protein synthesis. J Leukoc Biol 63:509–514. Franchini M (2006). Hereditary iron overload: update on pathophysiology, diagnosis and treatment. Am J Hematol 81:202–209. Francis GA, Fayard E, Picard F and Auwerx J (2003). Nuclear receptors and the control of metabolism. Annu Rev Physiol 65:261–311. Freshney NW, Rawlinson L, Guesdon F, et al. (1994). Interleukin-1 activates a novel protein kinase cascade that results in the phosphorylation of Hsp27. Cell 78:1039–1049. Frey N, Christen U, Jeno P, et al. (1995). The lipoic acid containing components of the 2-oxoacid dehydrogenase complexes mimic trifluoroacetylated proteins and are autoantigens associated with halothane hepatitis. Chem Res Toxicol 8:736–746. Fridovich I (1978). The biology of oxygen radicals. Science 201:875–880. Friedman MA, Woodcock J, Lumpkin MM, Shuren JE, Hass AE and Thompson LJ (1999). The safety of newly approved medicines: do recent market removals mean there is a problem? JAMA 281:1728–1734. Friling RS, Bensimon A, Tichauer Y and Daniel V (1990). Xenobiotic-inducible expression of murine glutathione S-transferase Ya subunit gene is controlled by an electrophile-responsive element. Proc Natl Acad Sci USA 87:6258–6262. Fromenty B and Pessayre D (1995). Inhibition of mitochondrial beta-oxidation as a mechanism of hepatotoxicity. Pharmacol Ther 67:101–154. Fu Y, Cheng WH, Porres JM, Ross DA and Lei XG (1999a). Knockout of cellular glutathione peroxidase gene renders mice susceptible to diquat-induced oxidative stress. Free Radic Biol Med 27: 605–611. Fu Y, Cheng WH, Ross DA and Lei X (1999b). Cellular glutathione peroxidase protects mice against lethal oxidative stress induced by various doses of diquat. Proc Soc Exp Biol Med 222:164– 169. Fung M, Thornten A, Mybeck K and Hsiao-Hui J (2001). Evaluation of the characteristics of safety withdrawal of prescription drugs from worldwide pharmaceutical markets-1960 to 1999. Drug Information Journal 35: 293–317. Gale EA (2001). Lessons from the glitazones: a story of drug development. Lancet 357:1870–1875. Gao D, Wei C, Chen L, Huang J, Yang S and Diehl AM (2004). Oxidative DNA damage and DNA repair enzyme expression are inversely related in murine models of fatty liver disease. Am J Physiol Gastrointest Liver Physiol 287:G1070–G1077. Garcia-Torres ML, Zaragoza A, Giner R, Primo J and del Olmo JA (2003). Incidence and epidemiological factors of hepatocellular carcinoma in Valencia during the year 2000. Rev Esp Enferm Dig 95:385–394. Gardner CR, Laskin JD, Dambach DM, et al. (2002). Reduced hepatotoxicity of acetaminophen in mice lacking inducible nitric oxide synthase: potential role of tumor necrosis factor-alpha and interleukin-10. Toxicol Appl Pharmacol 184:27–36. Gascon-Barre M and Cote MG (1978). Effects of phenobarbital and diphenylhydantoin on acute vitamin D3 toxicity in the rat. Toxicol Appl Pharmacol 43:125–135. Gascon-Barre M, Demers C, Mirshahi A, Neron S, Zalzal S and Nanci A (2003). The normal liver harbors the vitamin D nuclear receptor in nonparenchymal and biliary epithelial cells. Hepatology 37:1034–1042.


263

Geier A, Kim SK, Gerloff T, et al. (2002). Hepatobiliary organic anion transporters are differentially regulated in acute toxic liver injury induced by carbon tetrachloride. J Hepatol 37:198–205. Genter MB, Liang HC, Gu J, et al. (1998). Role of CYP2A5 and 2G1 in acetaminophen metabolism and toxicity in the olfactory mucosa of the Cyp1a2(−/−) mouse. Biochem Pharmacol 55: 1819–1826. Gerbal-Chaloin S, Daujat M, Pascussi JM, Pichard-Garcia L, Vilarem MJ and Maurel P (2002). Transcriptional regulation of CYP2C9 gene. Role of glucocorticoid receptor and constitutive androstane receptor. J Biol Chem 277:209–217. Gerbal-Chaloin S, Pascussi JM, Pichard-Garcia L, et al. (2001). Induction of CYP2C genes in human hepatocytes in primary culture. Drug Metab Dispos 29:242–251. Gerbaux C, Van Bambeke F, Montenez JP, Piret J, Morlighem G and Tulkens PM (1996). Hyperactivity of cathepsin B and other lysosomal enzymes in fibroblasts exposed to azithromycin, a dicationic macrolide antibiotic with exceptional tissue accumulation. FEBS Lett 394:307–310. Ghavami S, Hashemi M, Kadkhoda K, Alavian SM, Bay GH and Los M (2005). Apoptosis in liver diseases – detection and therapeutic applications. Med Sci Monit 11:RA337–RA345. Giasson BI, Duda JE, Murray IV, et al. (2000). Oxidative damage linked to neurodegeneration by selective alpha-synuclein nitration in synucleinopathy lesions. Science 290:985–989. Giasson BI, Ischiropoulos H, Lee VM and Trojanowski JQ (2002). The relationship between oxidative/nitrative stress and pathological inclusions in Alzheimer’s and Parkinson’s diseases. Free Radic Biol Med 32:1264–1275. Gibson JD, Pumford NR, Samokyszyn VM and Hinson JA (1996). Mechanism of acetaminopheninduced hepatotoxicity: covalent binding versus oxidative stress. Chem Res Toxicol 9:580–585. Gillespie DA and Vickers CR (1993). Pruritus and cholestasis: therapeutic options. J Gastroenterol Hepatol 8:168–173. Goeptar AR, Scheerens H and Vermeulen NP (1995). Oxygen and xenobiotic reductase activities of cytochrome P450. Crit Rev Toxicol 25:25–65. Goldin RD, Ratnayaka ID, Breach CS, Brown IN and Wickramasinghe SN (1996). Role of macrophages in acetaminophen (paracetamol)-induced hepatotoxicity. J Pathol 179:432–435. Goldring CE, Kitteringham NR, Elsby R, et al. (2004). Activation of hepatic Nrf2 in vivo by acetaminophen in CD-1 mice. Hepatology 39:1267–1276. Gonzalez FJ (2002). Transgenic models in xenobiotic metabolism and toxicology. Toxicology 181– 182:237–239. Gonzalez FJ and Fernandez-Salguero P (1998). The aryl hydrocarbon receptor: studies using the AHR-null mice. Drug Metab Dispos 26:1194–1198. Goodwin B, Jones SA, Price RR, et al. (2000). A regulatory cascade of the nuclear receptors FXR, SHP-1 and LRH-1 represses bile acid biosynthesis. Mol Cell 6:517–26. Goodwin B, Redinbo MR and Kliewer SA (2002). Regulation of cyp3a gene transcription by the pregnane x receptor. Annu Rev Pharmacol Toxicol 42:1–23. Gores GJ, Herman B and Lemasters JJ (1990). Plasma membrane bleb formation and rupture: a common feature of hepatocellular injury. Hepatology 11:690–698. Gosnell JE, Wong CB, Kumwenda ZL, Welch WJ and Harris HW (2000). Extracellular matrix regulates the hepatocellular heat shock response. J Surg Res 91:43–49. Graham DJ, Drinkard CR and Shatin D (2003). Incidence of idiopathic acute liver failure and hospitalized liver injury in patients treated with troglitazone. Am J Gastroenterol 98:175–179. Green RM, Whiting JF, Rosenbluth AB, Beier D and Gollan JL (1994). Interleukin-6 inhibits hepatocyte taurocholate uptake and sodium-potassium-adenosinetriphosphatase activity. Am J Physiol 267:G1094–G1100. Greenwel P, Dominguez-Rosales JA, Mavi G, Rivas-Estilla AM and Rojkind M (2000). Hydrogen peroxide: a link between acetaldehyde-elicited alpha1(I) collagen gene up-regulation and oxidative stress in mouse hepatic stellate cells. Hepatology 31:109–116.

264

Hepatotoxicity

Gressner AM, Polzar B, Lahme B and Mannherz HG (1996). Induction of rat liver parenchymal cell apoptosis by hepatic myofibroblasts via transforming growth factor beta. Hepatology 23:571– 581. Gros L, Ishchenko AA and Saparbaev M (2003). Enzymology of repair of etheno-adducts. Mutat Res 531:219–229. Gruchalla RS and Sullivan TJ (1991). Detection of human IgE to sulfamethoxazole by skin testing with sulfamethoxazoyl-poly-l-tyrosine. J Allergy Clin Immunol 88:784–792. Grune T, Merker K, Sandig G and Davies KJ (2003). Selective degradation of oxidatively modified protein substrates by the proteasome. Biochem Biophys Res Commun 305:709–718. Guastadisegni C, Balduzzi M, Mancuso MT and Di Consiglio E (1999). Liver mitochondria alterations in chloroform-treated Sprague–Dawley rats. J Toxicol Environ Health A 57:415–429. Guicciardi ME and Gores GJ (2002). Bile acid-mediated hepatocyte apoptosis and cholestatic liver disease. Dig Liver Dis 34:387–392. Guicciardi ME and Gores GJ (2005). Apoptosis: a mechanism of acute and chronic liver injury. Gut 54:1024–1033. Guo GL, Lambert G, Negishi M, et al. (2003). Complementary roles of farnesoid X receptor, pregnane X receptor and constitutive androstane receptor in protection against bile acid toxicity. J Biol Chem 278:45062–45071. Gupta S, Natarajan R, Payne SG, et al. (2004). Deoxycholic acid activates the c-Jun N -terminal kinase pathway via FAS receptor activation in primary hepatocytes. Role of acidic sphingomyelinasemediated ceramide generation in FAS receptor activation. J Biol Chem 279:5821–5828. Gutteridge JM and Quinlan GJ (1983). Malondialdehyde formation from lipid peroxides in the thiobarbituric acid test: the role of lipid radicals iron salts and metal chelators. J Appl Biochem 5:293–299. Guyton KZ, Xu Q and Holbrook NJ (1996). Induction of the mammalian stress response gene GADD153 by oxidative stress: role of AP-1 element. Biochem J 314:547–554. Hahn E, Wick G, Pencev D and Timpl R (1980). Distribution of basement membrane proteins in normal and fibrotic human liver: collagen type IV, laminin and fibronectin. Gut 21:63–71. Haimeur A, Conseil G, Deeley RG and Cole SP (2004). The MRP-related and BCRP/ABCG2 multidrug resistance proteins: biology, substrate specificity and regulation. Curr Drug Metab 5:21–53. Halliwell WH (1997). Cationic amphiphilic drug-induced phospholipidosis. Toxicol Pathol 25:53–60. Hampton MB and Orrenius S (1997). Dual regulation of caspase activity by hydrogen peroxide: implications for apoptosis. FEBS Lett 414:552–556. Han D, Hanawa N, Saberi B and Kaplowitz N (2006). Mechanisms of liver injury. III. Role of glutathione redox status in liver injury. Am J Physiol Gastrointest Liver Physiol 291:G1–G7. Handschin C and Meyer UA (2003). Induction of drug metabolism: the role of nuclear receptors. Pharmacol Rev 55:649–673. Hankinson O, Brooks BA, Weir-Brown KI, et al. (1991). Genetic and molecular analysis of the Ah receptor and of Cyp1a1 gene expression. Biochimie 73:61–66. Hannigan M, Zhan L, Ai Y and Huang CK (1998). The role of p38 MAP kinase in TGF-beta1-induced signal transduction in human neutrophils. Biochem Biophys Res Commun 246:55–58. Hansen JM, Watson WH and Jones DP (2004). Compartmentation of Nrf-2 redox control: regulation of cytoplasmic activation by glutathione and DNA binding by thioredoxin-1. Toxicol Sci 82:308– 317. Harano Y, Yasui K, Toyama T, et al. (2006). Fenofibrate, a peroxisome proliferator-activated receptor alpha agonist, reduces hepatic steatosis and lipid peroxidation in fatty liver Shionogi mice with hereditary fatty liver. Liver Int 26:613–620. Hartig S, Fries S and Balcarcel RR (2005). Reduced mitochondrial membrane potential and metabolism correspond to acute chloroform toxicity of in vitro hepatocytes. J Appl Toxicol 25:310– 317.


265

Hatano E, Bradham CA, Stark A, Iimuro Y, Lemasters JJ and Brenner DA (2000). The mitochondrial permeability transition augments Fas-induced apoptosis in mouse hepatocytes. J Biol Chem 275:11814–11823. Hayden MS and Ghosh S (2004). Signaling to NF-kappaB. Genes Dev 18:2195–224. He Q and Sharma RP (2005). Inhibition of tumor necrosis factor alpha signaling by anti-tumor necrosis factor alpha antibodies and pentoxifylline is unable to prevent fumonisin hepatotoxicity in mice. Toxicon 46:404–413. Heidel SM, Holston K, Buters JT, Gonzalez FJ, Jefcoate CR and Czupyrynski CJ (1999). Bone marrow stromal cell cytochrome P4501B1 is required for pre-B cell apoptosis induced by 7,12dimethylbenz[a]anthracene. Mol Pharmacol 56:1317–1323. Heijnen BH, Straatsburg IH, Padilla ND, Van Mierlo GJ, Hack CE and Van Gulik TM (2006). Inhibition of classical complement activation attenuates liver ischaemia and reperfusion injury in a rat model. Clin Exp Immunol 143:15–23. Heinloth AN, Irwin RD, Boorman GA, et al. (2004). Gene expression profiling of rat livers reveals indicators of potential adverse effects. Toxicol Sci 80:193–202. Henderson CJ, Wolf CR, Kitteringham N, Powell H, Otto D and Park BK (2000). Increased resistance to acetaminophen hepatotoxicity in mice lacking glutathione S-transferase Pi. Proc Natl Acad Sci USA 97:12741–12745. Herman B, Nieminen AL, Gores GJ and Lemasters JJ (1988). Irreversible injury in anoxic hepatocytes precipitated by an abrupt increase in plasma membrane permeability. FASEB J 2:146–151. Hernandez JP, Chapman LM, Kretschmer XC and Baldwin WS (2006). Gender-specific induction of cytochrome P450s in nonylphenol-treated FVB/NJ mice. Toxicol Appl Pharmacol 216:186–196. Hertz R, Magenheim J, Berman I and Bar-Tana J (1998). Fatty acyl-CoA thioesters are ligands of hepatic nuclear factor-4alpha. Nature 392:512–516. Higuchi H, Bronk SF, Takikawa Y, et al. (2001). The bile acid glycochenodeoxycholate induces trail-receptor 2/DR5 expression and apoptosis. J Biol Chem 276:38610–38618. Higuchi H and Gores GJ (2003). Bile acid regulation of hepatic physiology: IV. Bile acids and death receptors. Am J Physiol Gastrointest Liver Physiol 284:G734–G738. Hines JE, Johnson SJ and Burt AD (1993). In vivo responses of macrophages and perisinusoidal cells to cholestatic liver injury. Am J Pathol 142:511–518. Hinson JA, Pike SL, Pumford NR and Mayeux PR (1998). Nitrotyrosine–protein adducts in hepatic centrilobular areas following toxic doses of acetaminophen in mice. Chem Res Toxicol 11:604–607. Hirai K, Ikeda K and Wang GY (1992). Paraquat damage of rat liver mitochondria by superoxide production depends on extramitochondrial NADH. Toxicology 72:1–16. Hosokawa M, Hattori K and Satoh T (1993). Differential responses of rat hepatic microsomal carboxylesterase isozymes to glucocorticoids and pregnenolone 16 alpha-carbonitrile. Biochem Pharmacol 45:2317–2322. Hsu MK, Qiao L, Ho V, et al. (2006). Ethanol reduces p38 kinase activation and cyclin D1 protein expression after partial hepatectomy in rats. J Hepatol 44:375–382. Huang TH, Li Y, Razmovski-Naumovski V, et al. (2006). Gypenoside XLIX isolated from Gynostemma pentaphyllum inhibits nuclear factor-kappaB activation via a PPAR-alpha-dependent pathway. J Biomed Sci 13:535–548. Hug H, Strand S, Grambihler A, et al. (1997). Reactive oxygen intermediates are involved in the induction of CD95 ligand mRNA expression by cytostatic drugs in hepatoma cells. J Biol Chem 272:28191–28193. Imanishi Y, Maeda N, Matsui H, et al. (2004). Suppression of Rat Stellate Cell Activation and Liver Fibrosis by a Japanese Herbal Medicine, Inchinko-to (TJ135). Comp Hepatol (3 Suppl 1):S11. Inamdar NM, Ahn YI and Alam J (1996). The heme-responsive element of the mouse heme oxygenase-1 gene is an extended AP-1 binding site that resembles the recognition sequences for MAF and NF-E2 transcription factors. Biochem Biophys Res Commun 221:570–576.

266

Hepatotoxicity

Irwin RD, Boorman GA, Cunningham ML, Heinloth AN, Malarkey DE and Paules RS (2004). Application of toxicogenomics to toxicology: basic concepts in the analysis of microarray data. Toxicol Pathol (32 Suppl 1):72–83. Ishii T, Itoh K, Takahashi S, et al. (2000). Transcription factor Nrf2 coordinately regulates a group of oxidative stress-inducible genes in macrophages. J Biol Chem 275:16023–16029. Isoda K, Koide H, Kojima M, et al. (2005). Stimulation of hepatocyte survival and suppression of CCl4 -induced liver injury by the adenovirally introduced C/EBPbeta gene. Biochem Biophys Res Commun 329:182–187. Iwai S, Karim R, Kitano M, et al. (2002). Role of oxidative DNA damage caused by carbon tetrachloride-induced liver injury – enhancement of MeIQ-induced glutathione S-transferase placental form-positive foci in rats. Cancer Lett 179:15–24. Jacob A, Budhiraja S, Qian X, Clevidence D, Costa RH and Reichel RR (1994). Retinoic acidmediated activation of HNF-3 alpha during EC stem cell differentiation. Nucleic Acids Res 22:2126–2133. Jaeschke H (2002a). Inflammation in response to hepatocellular apoptosis. Hepatology 35:964–966. Jaeschke H (2002b). Neutrophil-mediated tissue injury in alcoholic hepatitis. Alcohol 27:23–27. Jaeschke H, Ho YS, Fisher MA, Lawson JA, Farhood A (1999). Glutathione peroxidase-deficient mice are more susceptible to neutrophil-mediated hepatic parenchymal cell injury during endotoxemia: importance of an intracellular oxidant stress. Hepatology 29:443–450. Jaeschke H, Knight TR and Bajt ML (2003). The role of oxidant stress and reactive nitrogen species in acetaminophen hepatotoxicity. Toxicol Lett 144:279–288. Jaiswal AK (1991). Human NAD(P)H:quinone oxidoreductase (NQO1) gene structure and induction by dioxin. Biochemistry 30:10647–10653. James LP, Mayeux PR and Hinson JA (2003a). Acetaminophen-induced hepatotoxicity. Drug Metab Dispos 31:1499–1506. James LP, McCullough SS, Lamps LW and Hinson JA (2003b). Effect of N -acetylcysteine on acetaminophen toxicity in mice: relationship to reactive nitrogen and cytokine formation. Toxicol Sci 75:458–467. Jansen PL, Muller M and Sturm E (2001). Genes and cholestasis. Hepatology 34:1067–1074. Jiang X and Wang X (2004). Cytochrome C-mediated apoptosis. Annu Rev Biochem 73:87–106. Johansson I, Ekstrom G, Scholte B, Puzycki D, Jornvall H and Ingelman-Sundberg M (1988). Ethanol, fasting- and acetone-inducible cytochromes P-450 in rat liver: regulation and characteristics of enzymes belonging to the IIB and IIE gene subfamilies. Biochemistry 27:1925–1934. Johnson DR and Klaassen CD (2002). Regulation of rat multidrug resistance protein 2 by classes of prototypical microsomal enzyme inducers that activate distinct transcription pathways. Toxicol Sci 67:182–189. Johnson GL and Lapadat R (2002). Mitogen-activated protein kinase pathways mediated by ERK, JNK and p38 protein kinases. Science 298:1911–1912. Jones SA, Moore LB, Shenk JL, et al. (2000). The pregnane X receptor: a promiscuous xenobiotic receptor that has diverged during evolution. Mol Endocrinol 14:27–39. Jover R, Bort R, Gomez-Lechon MJ and Castell JV (2002). Down-regulation of human CYP3A4 by the inflammatory signal interleukin-6: molecular mechanism and transcription factors involved. FASEB J 16:1799–1801. Kamiya A, Inoue Y and Gonzalez FJ (2003). Role of the hepatocyte nuclear factor 4alpha in control of the pregnane X receptor during fetal liver development. Hepatology 37:1375–1384. Kane CD, Francone OL and Stevens KA (2006). Differential regulation of the cynomolgus, human and rat acyl-CoA oxidase promoters by PPARalpha. Gene 380:84–94. Kang KW, Ryu JH and Kim SG (2000). The essential role of phosphatidylinositol 3-kinase and of p38 mitogen-activated protein kinase activation in the antioxidant response element-mediated rGSTA2 induction by decreased glutathione in H4IIE hepatoma cells. Mol Pharmacol 58:1017–1025.


267

Kaplowitz N (2002). Biochemical and cellular mechanisms of toxic liver injury. Semin Liver Dis 22:137–144. Karin M (1999). The beginning of the end: IkappaB kinase (IKK) and NF-kappaB activation. J Biol Chem 274:27339–27342. Karpen SJ (2002). Nuclear receptor regulation of hepatic function. J Hepatol 36:832–850. Kast HR, Goodwin B, Tarr PT, et al. (2002). Regulation of multidrug resistance-associated protein 2 (ABCC2) by the nuclear receptors pregnane X receptor, farnesoid X-activated receptor and constitutive androstane receptor. J Biol Chem 277:2908–2915. Kato M, Iwamoto H, Higashi N, et al. (1999). Role of Rho small GTP binding protein in the regulation of actin cytoskeleton in hepatic stellate cells. J Hepatol 31:91–99. Kazlauskas A, Poellinger L and Pongratz I (1999). Evidence that the co-chaperone p23 regulates ligand responsiveness of the dioxin (Aryl hydrocarbon) receptor. J Biol Chem 274:13519– 13524. Kazlauskas A, Sundstrom S, Poellinger L and Pongratz I (2001). The hsp90 chaperone complex regulates intracellular localization of the dioxin receptor. Mol Cell Biol 21:2594–2607. Keller BJ, Yamanaka H, Liang DC and Thurman RG (1990). Hepatotoxicity due to clofibrate is oxygen-dependent in the perfused rat liver. Toxicol Appl Pharmacol 104:259–266. Keller BJ, Yamanaka H and Thurman RG (1992). Inhibition of mitochondrial respiration and oxygendependent hepatotoxicity by six structurally dissimilar peroxisomal proliferating agents. Toxicology 71:49–61. Kenna JG, Neuberger J and Williams R (1988). Evidence for expression in human liver of halothaneinduced neoantigens recognized by antibodies in sera from patients with halothane hepatitis. Hepatology 8:1635–1641. Kersten S (2002). Peroxisome proliferator activated receptors and obesity. Eur J Pharmacol 440:223– 234. Khan MF, Wu X, Tipnis UR, Ansari GA and Boor PJ (2002). Protein adducts of malondialdehyde and 4-hydroxynonenal in livers of iron loaded rats: quantitation and localization. Toxicology 173:193– 201. Khosravi-Far R and Esposti MD (2004). Death receptor signals to mitochondria. Cancer Biol Ther 3:1051–1057. Kim JS, Qian T, Lemasters JJ (2003a). Mitochondrial permeability transition in the switch from necrotic to apoptotic cell death in ischemic rat hepatocytes. Gastroenterology 124:494–503. Kim YC, Yamaguchi Y, Kondo N, Masutani H and Yodoi J (2003b). Thioredoxin-deoendent redox regulation of the antioxidant responsive element (ARE) in electrophile response. Oncogene 22:1860–1865. Kim ND, Moon JO, Slitt AL and Copple BL (2006). Early growth response factor-1 is critical for cholestatic liver injury. Toxicol Sci 90:586–595. Kimura S, Kawabe M, Ward JM, et al. (1999). CYP1A2 is not the primary enzyme responsible for 4-aminobiphenyl-induced hepatocarcinogenesis in mice. Carcinogenesis 20:1825–1830. Kitteringham NR, Powell H, Clement YN, et al. (2000). Hepatocellular response to chemical stress in CD-1 mice: induction of early genes and gamma-glutamylcysteine synthetase. Hepatology 32:321–333. Klaassen CD and Slitt AL (2005). Regulation of hepatic transporters by xenobiotic receptors. Curr Drug Metab 6:309–328. Klaunig JE, Xu Y, Isenberg JS, et al. (1998). The role of oxidative stress in chemical carcinogenesis. Environ Health Perspect 106 (Suppl 1):289–295. Klemke RL, Cai S, Giannini AL, Gallagher PJ, de Lanerolle P and Cheresh DA (1997). Regulation of cell motility by mitogen-activated protein kinase. J Cell Biol 137:481–492. Kliewer SA and Willson TM (2002). Regulation of xenobiotic and bile acid metabolism by the nuclear pregnane X receptor. J Lipid Res 43:359–364.

268

Hepatotoxicity

Kliewer SA, Moore JT, Wade L, et al. (1998). An orphan nuclear receptor activated by pregnanes defines a novel steroid signaling pathway. Cell 92:73–82. Kliewer SA, Goodwin B and Willson TM (2002). The nuclear pregnane X receptor: a key regulator of xenobiotic metabolism. Endocr Rev 23:687–702. Knight TR, Fariss MW, Farhood A and Jaeschke H (2003). Role of lipid peroxidation as a mechanism of liver injury after acetaminophen overdose in mice. Toxicol Sci 76:229–236. Knittel T, Kobold D, Piscaglia F, et al. (1999). Localization of liver myofibroblasts and hepatic stellate cells in normal and diseased rat livers: distinct roles of (myo-)fibroblast subpopulations in hepatic tissue repair. Histochem Cell Biol 112:387–401. Kobayashi T, Hirano K, Yamamoto T, et al. (2002). The protective role of Kupffer cells in the ischemia–reperfused rat liver. Arch Histol Cytol 65:251–261. Kolch W (2000). Meaningful relationships: the regulation of the Ras/Raf/MEK/ERK pathway by protein interactions. Biochem J 351:289–305. Kong AN, Owuor E, Yu R, et al. (2001). Induction of xenobiotic enzymes by the MAP kinase pathway and the antioxidant or electrophile response element (ARE/EpRE). Drug Metab Rev 33:255– 271. Kono H, Bradford BU, Yin M, et al. (1999). CYP2E1 is not involved in early alcohol-induced liver injury. Am J Physiol 277:G1259–G1267. Kontorinis N, Agarwal K, Gondolesi G, Fiel MI, O’Rourke M and Schiano TD (2004). Diagnosis of 6 mercaptopurine hepatotoxicity post liver transplantation utilizing metabolite assays. Am J Transplant 4:1539–1542. Kossor DC, Goldstein RS, Ngo W, et al. (1995). Biliary epithelial cell proliferation following alphanaphthylisothiocyanate (ANIT) treatment: relationship to bile duct obstruction. Fundam Appl Toxicol 26:51–62. Kostadinova R, Wahli W and Michalik L (2005). PPARs in diseases: control mechanisms of inflammation. Curr Med Chem 12:2995–3009. Kostrubsky VE, Szakacs JG, Jeffery EH, et al. (1997). Role of CYP3A in ethanol-mediated increases in acetaminophen hepatotoxicity. Toxicol Appl Pharmacol 143:315–323. Krynetski EY, Tai HL, Yates CR, et al. (1996). Genetic polymorphism of thiopurine Smethyltransferase: clinical importance and molecular mechanisms. Pharmacogenetics 6:279–290. Kudo A, Kashiwagi S, Kajimura M, et al. (2004). Kupffer cells alter organic anion transport through multidrug resistance protein 2 in the post-cold ischemic rat liver. Hepatology 39:1099–1109. Kukielka E and Cederbaum AI (1990). NADPH- and NADH-dependent oxygen radical generation by rat liver nuclei in the presence of redox cycling agents and iron. Arch Biochem Biophys 283:326– 333. Kurose I, Higuchi H, Miura S, et al. (1997). Oxidative stress-mediated apoptosis of hepatocytes exposed to acute ethanol intoxication. Hepatology 25:368–378. Kwak MK, Kensler TW and Casero RA Jr (2003). Induction of phase 2 enzymes by serum oxidized polyamines through activation of Nrf2: effect of the polyamine metabolite acrolein. Biochem Biophys Res Commun 305:662–670. Kyrklund C, Backman JT, Neuvonen M and Neuvonen PJ (2003). Gemfibrozil increases plasma pravastatin concentrations and reduces pravastatin renal clearance. Clin Pharmacol Ther 73:538– 544. Lahuna O, Fernandez L, Karlsson H, et al. (1997). Expression of hepatocyte nuclear factor 6 in rat liver is sex-dependent and regulated by growth hormone. Proc Natl Acad Sci USA 94:12309–12313. Lahuna O, Rastegar M, Maiter D, Thissen JP, Lemaigre FP and Rousseau GG (2000). Involvement of STAT5 (signal transducer and activator of transcription 5) and HNF-4 (hepatocyte nuclear factor 4) in the transcriptional control of the hnf6 gene by growth hormone. Mol Endocrinol 14:285–294. Lambert G, Amar MJ, Guo G, Brewer HB Jr, Gonzalez FJ and Sinal CJ (2003). The farnesoid X-receptor is an essential regulator of cholesterol homeostasis. J Biol Chem 278:2563–2570.


269

Landry C, Clotman F, Hioki T, et al. (1997). HNF-6 is expressed in endoderm derivatives and nervous system of the mouse embryo and participates to the cross-regulatory network of liver-enriched transcription factors. Dev Biol 192:247–257. Larrey D (2002). Epidemiology and individual susceptibility to adverse drug reactions affecting the liver. Semin Liver Dis 22:145–155. Laskin DL and Pendino KJ (1995). Macrophages and inflammatory mediators in tissue injury. Annu Rev Pharmacol Toxicol 35:655–677. Laskin DL and Pilaro AM (1986). Potential role of activated macrophages in acetaminophen hepatotoxicity. I. Isolation and characterization of activated macrophages from rat liver. Toxicol Appl Pharmacol 86:204–215. Laurent S, Horsmans Y, Starkel P, Leclercq I, Sempoux C and Lambotte L (2005). Disrupted NFkappa B activation after partial hepatectomy does not impair hepatocyte proliferation in rats. World J Gastroenterol 11:7345–7350. Lawson JA, Farhood A, Hopper RD, Bajt ML and Jaeschke H (2000). The hepatic inflammatory response after acetaminophen overdose: role of neutrophils. Toxicol Sci 54:509–516. Lazarczyk DA, Goldstein NS and Gordon SC (2001). Trovafloxacin hepatotoxicity. Dig Dis Sci 46:925–926. Lecoeur S, Andre C and Beaune PH (1996). Tienilic acid-induced autoimmune hepatitis: anti-liver and-kidney microsomal type 2 autoantibodies recognize a three-site conformational epitope on cytochrome P4502C9. Mol Pharmacol 50:326–533. Lee AU and Farrell GC (2001). Mechanism of azathioprine-induced injury to hepatocytes: roles of glutathione depletion and mitochondrial injury. J Hepatol 35:756–764. Lee HO and Ferguson TA (2003). Biology of FasL. Cytokine Growth Factor Rev 14:325–335. Lee SS, Buters JT, Pineau T, Fernandez-Salguero P and Gonzalez FJ (1996). Role of CYP2E1 in the hepatotoxicity of acetaminophen. J Biol Chem 271:12063–12067. Lee JM, Trauner M, Soroka CJ, Stieger B, Meier PJ and Boyer JL (2000). Expression of the bile salt export pump is maintained after chronic cholestasis in the rat. Gastroenterology 118:163–172. Lehmann JM, McKee DD, Watson MA, Willson TM, Moore JT and Kliewer SA (1998). The human orphan nuclear receptor PXR is activated by compounds that regulate CYP3A4 gene expression and cause drug interactions. J Clin Invest 102:1016–1023. Lei XG (2002). In vivo antioxidant role of glutathione peroxidase: evidence from knockout mice. Methods Enzymol 347:213–225. Lei XG, Zhu JH, McClung JP, Aregullin M and Roneker CA (2006). Mice deficient in Cu,Znsuperoxide dismutase are resistant to acetaminophen toxicity. Biochem J. Lemaire G, de Sousa G and Rahmani R (2004). A PXR reporter gene assay in a stable cell culture system: CYP3A4 and CYP2B6 induction by pesticides. Biochem Pharmacol 68:2347–2358. Lemasters JJ (1999). V. Necrapoptosis and the mitochondrial permeability transition: shared pathways to necrosis and apoptosis. Am J Physiol 276:G1–G6. Lennard L, Van Loon JA and Weinshilboum RM (1989). Pharmacogenetics of acute azathioprine toxicity: relationship to thiopurine methyltransferase genetic polymorphism. Clin Pharmacol Ther 46:149–154. Leonarduzzi G, Arkan MC, Basaga H, Chiarpotto E, Sevanian A and Poli G (2000). Lipid oxidation products in cell signaling. Free Radic Biol Med 28:1370–1378. Leonarduzzi G, Chiarpotto E, Biasi F and Poli G (2005). 4-Hydroxynonenal and cholesterol oxidation products in atherosclerosis. Mol Nutr Food Res 49:1044–1049. Leclercq IA, Farrell GC, Field J, Bell DR, Gonzalez FJ, Robertson GR (2000). CYP2E1 and CYP4A as microsomal catalysts of lipid peroxides in murine nonalcoholic steatohepatitis. J Clin Invest 105:1067–1075. Letteron P, Fromenty B, Terris B, Degott C and Pessayre D (1996). Acute and chronic hepatic steatosis lead to in vivo lipid peroxidation in mice. J Hepatol 24:200–208.

270

Hepatotoxicity

Lewis JH (2000). Drug-induced liver disease. Med Clin North Am 84:1275–1311. Li P, Nijhawan D, Budihardjo I, et al. (1997). Cytochrome c and dATP-dependent formation of Apaf-1/caspase-9 complex initiates an apoptotic protease cascade. Cell 91:479–489. Li D, Zimmerman TL, Thevananther S, Lee HY, Kurie JM and Karpen SJ (2002a). Interleukin-1 beta-mediated suppression of RXR:RAR transactivation of the Ntcp promoter is JNK-dependent. J Biol Chem 277:31416–31422. Li S, Zhao Y, He X, et al. (2002b). Relief of extrinsic pathway inhibition by the Bid-dependent mitochondrial release of Smac in Fas-mediated hepatocyte apoptosis. J Biol Chem 277:26912– 26920. Li C, Liu J, Waalkes MP, Jaeschke H (2003). Gene array analysis of the hepatic response to endotoxin in glutathione peroxidase-deficient mice. Toxicol Lett 144:397–406. Liang Q and Molkentin JD (2003). Redefining the roles of p38 and JNK signaling in cardiac hypertrophy: dichotomy between cultured myocytes and animal models. J Mol Cell Cardiol 35:1385–1394. Liang HC, Li H, McKinnon RA, et al. (1996). Cyp1a2(−/−) null mutant mice develop normally but show deficient drug metabolism. Proc Natl Acad Sci USA 93:1671–1676. Liao Y, Shikapwashya ON, Shteyer E, Dieckgraefe BK, Hruz PW and Rudnick DA (2004). Delayed hepatocellular mitotic progression and impaired liver regeneration in early growth response-1deficient mice. J Biol Chem 279:43107–43116. Liddle C and Goodwin B (2002). Regulation of hepatic drug metabolism: role of the nuclear receptors PXR and CAR. Semin Liver Dis 22:115–122. Lieber CS (1993). Biochemical factors in alcoholic liver disease. Semin Liver Dis 13:136–153. Lieber CS (1997). Role of oxidative stress and antioxidant therapy in alcoholic and nonalcoholic liver diseases. Adv Pharmacol 38:601–628. Lieber CS (2004). Alcoholic fatty liver: its pathogenesis and mechanism of progression to inflammation and fibrosis. Alcohol 34:9–19. Linton SD (2005). Caspase inhibitors: a pharmaceutical industry perspective. Curr Top Med Chem 5:1697–1717. Liu SL, Degli ES, Yao T, Diehl AM and Zern MA (1995). Vitamin E therapy of acute CCl4 -induced hepatic injury in mice is associated with inhibition of nuclear factor kappa B binding. Hepatology 22:1474–1481. Liu X, Zou H, Slaughter C and Wang X (1997). DFF, a heterodimeric protein that functions downstream of caspase-3 to trigger DNA fragmentation during apoptosis. Cell 89:175–184. Liu J, Liu Y, Habeebu SS and Klaassen CD (1998). Metallothionein (MT)-null mice are sensitive to cisplatin-induced hepatotoxicity. Toxicol Appl Pharmacol 149:24–31. Liu J, Liu Y, Hartley D, et al. (1999). Metallothionein-I/II knockout mice are sensitive to acetaminophen-induced hepatotoxicity. J Pharmacol Exp Ther 289:580–586. Liu J, Liu Y, Goyer RA, Achanzar W and Waalkes MP (2000). Metallothionein-I/II null mice are more sensitive than wild-type mice to the hepatotoxic and nephrotoxic effects of chronic oral or injected inorganic arsenicals. Toxicol Sci 55:460–467. Liu Y, Binz J, Numerick MJ, et al. (2003). Hepatoprotection by the farnesoid X receptor agonist GW4064 in rat models of intra- and extrahepatic cholestasis. J Clin Invest 112:1678–1687. Lohse AW, Manns M, Dienes HP, Meyer zum Buschenfelde KH and Cohen IR (1990). Experimental autoimmune hepatitis: disease induction, time course and T-cell reactivity. Hepatology 11:24– 30. Lohse AW, Brunner S, Kyriatsoulis A, Manns M and Meyer zum Buschenfelde KH (1992). Autoantibodies in experimental autoimmune hepatitis. J Hepatol 14:48–53. Lu TT, Makishima M, Repa JJ, et al. (2000). Molecular basis for feedback regulation of bile acid synthesis by nuclear receptors. Mol Cell 6:507–515. Lucena MI, Andrade RJ, Rodrigo L, et al. (2000). Trovafloxacin-induced acute hepatitis. Clin Infect Dis 30:400–401.


271

Ludwig B, Bender E, Arnold S, Huttemann M, Lee I and Kadenbach B (2001). Cytochrome C oxidase and the regulation of oxidative phosphorylation. Chembiochem 2:392–403. Luethy JD, Fargnoli J, Park JS, Fornace AJ Jr and Holbrook NJ (1990). Isolation and characterization of the hamster gadd153 gene. Activation of promoter activity by agents that damage DNA. J Biol Chem 265:16521–16526. Ma Q and Whitlock JP Jr (1996). The aromatic hydrocarbon receptor modulates the Hepa 1c1c7 cell cycle and differentiated state independently of dioxin. Mol Cell Biol 16:2144–2150. Ma X, Svegliati-Baroni G, Poniachik J, Baraona E and Lieber CS (1997). Collagen synthesis by liver stellate cells is released from its normal feedback regulation by acetaldehyde-induced modification of the carboxyl-terminal propeptide of procollagen. Alcohol Clin Exp Res 21:1204–1211. Mabuchi A, Mullaney I, Sheard PW, et al. (2004). Role of hepatic stellate cell/hepatocyte interaction and activation of hepatic stellate cells in the early phase of liver regeneration in the rat. J Hepatol 40:910–916. Madden S, Maggs JL and Park BK (1996). Bioactivation of carbamazepine in the rat in vivo. Evidence for the formation of reactive arene oxide(s). Drug Metab Dispos 24:469–479. Madhu C and Klaassen CD (1991). Protective effect of pregnenolone-16 alpha-carbonitrile on acetaminophen-induced hepatotoxicity in hamsters. Toxicol Appl Pharmacol 109:305–313. Maher JM, Cheng X, Slitt AL, Dieter MZ and Klaassen CD (2005). Induction of the multidrug resistance-associated protein family of transporters by chemical activators of receptor-mediated pathways in mouse liver. Drug Metab Dispos 33:956–962. Makinen J, Frank C, Jyrkkarinne J, Gynther J, Carlberg C and Honkakoski P (2002). Modulation of mouse and human phenobarbital-responsive enhancer module by nuclear receptors. Mol Pharmacol 62:366–378. Makishima M, Lu TT, Xie W, et al. (2002). Vitamin D receptor as an intestinal bile acid sensor. Science 296:1313–1316. Malarkey DE, Johnson K, Ryan L, Boorman G and Maronpot RR (2005). New insights into functional aspects of liver morphology. Toxicol Pathol 33:27–34. Malhi H, Bronk SF, Werneburg NW and Gores GJ (2006a). Free fatty acids induce JNK-dependent hepatocyte lipoapoptosis. J Biol Chem 281:12093–12101. Malhi H, Gores GJ and Lemasters JJ (2006b). Apoptosis and necrosis in the liver: a tale of two deaths? Hepatology 43:S31–S44. Malik R, Selden C and Hodgson H (2002). The role of non-parenchymal cells in liver growth. Semin Cell Dev Biol 13:425–431. Manna SK, Zhang HJ, Yan T, Oberley LW and Aggarwal BB (1998). Overexpression of manganese superoxide dismutase suppresses tumor necrosis factor-induced apoptosis and activation of nuclear transcription factor-kappaB and activated protein-1. J Biol Chem 273:13245–13254. Mansouri A, Gaou I, Fromenty B, et al. (1997). Premature oxidative aging of hepatic mitochondrial DNA in Wilson’s disease. Gastroenterology 113:599–605. Mansouri A, Gaou I, De Kerguenec C, et al. (1999). An alcoholic binge causes massive degradation of hepatic mitochondrial DNA in mice. Gastroenterology 117:181–190. Marinaki AM, Ansari A, Duley JA, et al. (2004a). Adverse drug reactions to azathioprine therapy are associated with polymorphism in the gene encoding inosine triphosphate pyrophosphatase (ITPase). Pharmacogenetics 14:181–187. Marinaki AM, Duley JA, Arenas M, et al. (2004b). Mutation in the ITPA gene predicts intolerance to azathioprine. Nucleosides Nucleotides Nucleic Acids 23:1393–1397. Marra F, Arrighi MC, Fazi M, et al. (1999). Extracellular signal-regulated kinase activation differentially regulates platelet-derived growth factor’s actions in hepatic stellate cells, and is induced by in vivo liver injury in the rat. Hepatology 30:951–958. Marschall HU, Wagner M, Bodin K, et al. (2006). Fxr(−/−) mice adapt to biliary obstruction by enhanced Phase I detoxification and renal elimination of bile acids. J Lipid Res 47:582–592.

272

Hepatotoxicity

Martin JL, Kenna JG, Martin BM, Thomassen D, Reed GF and Pohl LR (1993). Halothane hepatitis patients have serum antibodies that react with protein disulfide isomerase. Hepatology 18:858–863. Martindale JL and Holbrook NJ (2002). Cellular response to oxidative stress: signaling for suicide and survival. J Cell Physiol 192:1–15. Martins DL, Cury-Boaventura MF, Giannocco G, Nunes MT and Curi R (2006). Comparative toxicity of fatty acids on a macrophage cell line (J774). Clin Sci (London) 111:307–317. Masubuchi Y, Suda C and Horie T (2005). Involvement of mitochondrial permeability transition in acetaminophen-induced liver injury in mice. J Hepatol 42:110–116. Matsui H, Ikeda K, Nakajima Y, Horikawa S, Imanishi Y and Kawada N (2004). Sulfur-containing amino acids attenuate the development of liver fibrosis in rats through down-regulation of stellate cell activation. J Hepatol 40:917–925. Matsui H and Kawada N (2005). Effect of S-adenosyl-L-methionine on the activation, proliferation and contraction of hepatic stellate cells. Eur J Pharmacol 509:31–36. Matsumaru K, Ji C and Kaplowitz N (2003). Mechanisms for sensitization to TNF-induced apoptosis by acute glutathione depletion in murine hepatocytes. Hepatology 37:1425–1434. Matsuzaki S, Onda M, Tajiri T and Kim DY (1997). Hepatic lobar differences in progression of chronic liver disease: correlation of asialoglycoprotein scintigraphy and hepatic functional reserve. Hepatology 25:828–832. Matsuzaki Y, Nagai D, Ichimura E, et al. (2006). Metabolism and hepatic toxicity of flutamide in cytochrome P450 1A2 knockout SV129 mice. J Gastroenterol 41:231–239. Matthews JR, Wakasugi N, Virelizier JL, Yodoi J and Hay RT (1992). Thioredoxin regulates the DNA binding activity of NF-kappa B by reduction of a disulphide bond involving cysteine 62. Nucleic Acids Res 20:3821–3830. Mawet E, Shiratori Y, Hikiba Y, et al. (1996). Cytokine-induced neutrophil chemoattractant release from hepatocytes is modulated by Kupffer cells. Hepatology 23:353–358. McCue S, Noria S, Langille BL (2004). Shear-induced reorganization of endothelial cell cytoskeleton and adhesion complexes. Trends Cardiovasc Med 14:143–151. McCuskey RS (2006). Sinusoidal endothelial cells as an early target for hepatic toxicants. Clin Hemorheol Microcirc 34:5–10. McMahon M, Itoh K, Yamamoto M, et al. (2001). The Cap’n’Collar basic leucine zipper transcription factor Nrf2 (NF-E2 p45-related factor 2) controls both constitutive and inducible expression of intestinal detoxification and glutathione biosynthetic enzymes. Cancer Res 61:3299–3307. McQueen CA, Way BM, Queener SM, Schluter G and Williams GM (1991). Study of potential in vitro and in vivo genotoxicity in hepatocytes of quinolone antibiotics. Toxicol Appl Pharmacol 111:255–262. Mehendale HM (2000). PPAR-alpha: a key to the mechanism of hepatoprotection by clofibrate. Toxicol Sci 57:187–190. Mehendale HM (2005). Tissue repair: an important determinant of final outcome of toxicant-induced injury. Toxicol Pathol 33:41–51. Mehendale HM and Limaye PB (2005). Calpain: a death protein that mediates progression of liver injury. Trends Pharmacol Sci 26:232–236. Melnick RL (2001). Is peroxisome proliferation an obligatory precursor step in the carcinogenicity of di(2-ethylhexyl)phthalate (DEHP)? Environ Health Perspect 109:437–442. Memon RA, Tecott LH, Nonogaki K, et al. (2000). Up-regulation of peroxisome proliferator-activated receptors (PPAR-alpha) and PPAR-gamma messenger ribonucleic acid expression in the liver in murine obesity: troglitazone induces expression of PPAR-gamma-responsive adipose tissuespecific genes in the liver of obese diabetic mice. Endocrinology 141:4021–4031. Mendelson KG, Contois LR, Tevosian SG, Davis RJ and Paulson KE (1996). Independent regulation of JNK/p38 mitogen-activated protein kinases by metabolic oxidative stress in the liver. Proc Natl Acad Sci USA 93:12908–12913.


273

Mimura J and Fujii-Kuriyama Y (2003). Functional role of AhR in the expression of toxic effects by TCDD. Biochim Biophys Acta 1619:263–268. Mimura J, Yamashita K, Nakamura K, et al. (1997). Loss of teratogenic response to 2,3,7,8tetrachlorodibenzo-p-dioxin (TCDD) in mice lacking the Ah (dioxin) receptor. Genes Cells 2:645– 654. Minana JB, Gomez-Cambronero L, Lloret A, et al. (2002). Mitochondrial oxidative stress and CD95 ligand: a dual mechanism for hepatocyte apoptosis in chronic alcoholism. Hepatology 35:1205– 1214. Mirabelli F, Salis A, Vairetti M, Bellomo G, Thor H and Orrenius S (1989). Cytoskeletal alterations in human platelets exposed to oxidative stress are mediated by oxidative and Ca2+ -dependent mechanisms. Arch Biochem Biophys 270:478–488. Mitchell JR, Jollow DJ, Potter WZ, Gillette JR and Brodie BB (1973). Acetaminophen-induced hepatic necrosis. IV. Protective role of glutathione. J Pharmacol Exp Ther 187:211–217. Mitchell JR, Zimmerman HJ, Ishak KG, et al. (1976). Isoniazid liver injury: clinical spectrum, pathology and probable pathogenesis. Ann Intern Med 84:181–192. Miyata M, Tozawa A, Otsuka H, et al. (2005). Role of farnesoid X receptor in the enhancement of canalicular bile acid output and excretion of unconjugated bile acids: a mechanism for protection against cholic acid-induced liver toxicity. J Pharmacol Exp Ther 312:759–766. Miyoshi H, Rust C, Roberts PJ, Burgart LJ and Gores GJ (1999). Hepatocyte apoptosis after bile duct ligation in the mouse involves Fas. Gastroenterology 117:669–677. Moinova HR and Mulcahy RT (1999). Up-regulation of the human gamma-glutamylcysteine synthetase regulatory subunit gene involves binding of Nrf-2 to an electrophile responsive element. Biochem Biophys Res Commun 261:661–668. Moore LB, Goodwin B, Jones SA, et al. (2000). St. John’s wort induces hepatic drug metabolism through activation of the pregnane X receptor. Proc Natl Acad Sci USA 97:7500–7502. Motojima K (2004). 17beta-hydroxysteroid dehydrogenase type 11 is a major peroxisome proliferator-activated receptor alpha-regulated gene in mouse intestine. Eur J Biochem 271:4141– 4146. Motojima K and Hirai T (2006). Peroxisome proliferator-activated receptor alpha plays a vital role in inducing a detoxification system against plant compounds with crosstalk with other xenobiotic nuclear receptors. FEBS J 273:292–300. Muller D, Enderle GJ, Low O, Dietze E and Krell H (1996). Bile ductular proliferation and altered leukotriene elimination in thioacetamide-induced fibrosis of rat liver. J Hepatol 25:547–553. Muriel P (2000). Regulation of nitric oxide synthesis in the liver. J Appl Toxicol 20:189–195. Nagai H, Matsumaru K, Feng G and Kaplowitz N (2002). Reduced glutathione depletion causes necrosis and sensitization to tumor necrosis factor-alpha-induced apoptosis in cultured mouse hepatocytes. Hepatology 36:55–64. Nagengast FM, Grubben MJ and van Munster IP (1995). Role of bile acids in colorectal carcinogenesis. Eur J Cancer 31A:1067–1070. Naisbitt DJ, Britschgi M, Wong G, et al. (2003). Hypersensitivity reactions to carbamazepine: characterization of the specificity, phenotype and cytokine profile of drug-specific T cell clones. Mol Pharmacol 63:732–741. Nakagawa Y and Moore G (1999). Role of mitochondrial membrane permeability transition in phydroxybenzoate ester-induced cytotoxicity in rat hepatocytes. Biochem Pharmacol 58:811–816. Nakahira K, Takahashi T, Shimizu H, et al. (2003). Protective role of heme oxygenase-1 induction in carbon tetrachloride-induced hepatotoxicity. Biochem Pharmacol 66:1091–1105. Nakamura T, Akiyoshi H, Shiota G, et al. (1999). Hepatoprotective action of adenovirus-transferred HNF-3gamma gene in acute liver injury caused by CCl4 . FEBS Lett 459:1–4. Napirei M, Basnakian AG, Apostolov EO and Mannherz HG (2006). Deoxyribonuclease 1 aggravates acetaminophen-induced liver necrosis in male CD-1 mice. Hepatology 43:297–305.

274

Hepatotoxicity

Natori S, Rust C, Stadheim LM, Srinivasan A, Burgart LJ and Gores GJ (2001). Hepatocyte apoptosis is a pathologic feature of human alcoholic hepatitis. J Hepatol 34:248–253. Navarro VJ and Senior JR (2006). Drug-related hepatotoxicity. N Engl J Med 354:731–739. Nebert DW, Roe AL, Dieter MZ, Solis WA, Yang Y and Dalton TP (2000). Role of the aromatic hydrocarbon receptor and [Ah] gene battery in the oxidative stress response, cell cycle control and apoptosis. Biochem Pharmacol 59:65–85. Nehra V, Angulo P, Buchman AL and Lindor KD (2001). Nutritional and metabolic considerations in the etiology of nonalcoholic steatohepatitis. Dig Dis Sci 46:2347–2352. Nelson SD, Mitchell JR, Timbrell JA, Snodgrass WR and Corcoran GB III (1976). Isoniazid and iproniazid: activation of metabolites to toxic intermediates in man and rat. Science 193:901– 903. Neuberger J and Williams R (1989). Immune mechanisms in tienilic acid associated hepatotoxicity. Gut 30:515–519. Nguyen KA, Carbone JM, Silva VM, et al. (1999). The PPAR activator docosahexaenoic acid prevents acetaminophen hepatotoxicity in male CD-1 mice. J Toxicol Environ Health A 58:171– 186. Nguyen T, Sherratt PJ and Pickett CB (2003). Regulatory mechanisms controlling gene expression mediated by the antioxidant response element. Annu Rev Pharmacol Toxicol 43:233–260. Nicholls-Grzemski FA, Calder IC and Priestly BG (1992). Peroxisome proliferators protect against paracetamol hepatotoxicity in mice. Biochem Pharmacol 43:1395–1396. Nicholls-Grzemski FA, Calder IC, Priestly BG and Burcham PC (2000). Clofibrate-induced in vitro hepatoprotection against acetaminophen is not due to altered glutathione homeostasis. Toxicol Sci 56:220–228. Nishimura Y, Romer LH and Lemasters JJ (1998). Mitochondrial dysfunction and cytoskeletal disruption during chemical hypoxia to cultured rat hepatic sinusoidal endothelial cells: the pH paradox and cytoprotection by glucose, acidotic pH and glycine. Hepatology 27:1039–1049. Nissen SE, Wolski K and Topol EJ (2005). Effect of muraglitazar on death and major adverse cardiovascular events in patients with type 2 diabetes mellitus. JAMA 294:2581–2586. Nohl H, Gille L and Staniek K (2005). Intracellular generation of reactive oxygen species by mitochondria. Biochem Pharmacol 69:719–723. Nomiyama H, Hieshima K, Nakayama T, et al. (2001). Human CC chemokine liver-expressed chemokine/CCL16 is a functional ligand for CCR1, CCR2 and CCR5, and constitutively expressed by hepatocytes. Int Immunol 13:1021–1029. Nordberg J and Arner ES (2001). Reactive oxygen species, antioxidants and the mammalian thioredoxin system. Free Radic Biol Med 31:1287–1312. Nordmann P, Pechinot A and Kazmierczak A (1989). Cytotoxicity and uptake of pefloxacin, ciprofloxacin and ofloxacin in primary cultures of rat hepatocytes. J Antimicrob Chemother 24: 355–363. Nordmann R, Ribiere C and Rouach H (1992). Implication of free radical mechanisms in ethanolinduced cellular injury. Free Radic Biol Med 12:219–240. Ogawa A, Murate T, Suzuki M, Nimura Y and Yoshida S (1998). Lithocholic acid, a putative tumor promoter, inhibits mammalian DNA polymerase beta. Jpn J Cancer Res 89:1154–1159. Ohkubo K, Masumoto T, Horiike N and Onji M (1998). Induction of CINC (interleukin-8) production in rat liver by non-parenchymal cells. J Gastroenterol Hepatol 13:696–702. Ohshima H, Sawa T and Akaike T (2006). 8-nitroguanine, a product of nitrative DNA damage caused by reactive nitrogen species: formation, occurrence, and implications in inflammation and carcinogenesis. Antioxid Redox Signal 8:1033–1045. Oliver JR, Jiang S and Cherian MG (2006). Augmented hepatic injury followed by impaired regeneration in metallothionein-I/II knockout mice after treatment with thioacetamide. Toxicol Appl Pharmacol 210:190–199.


275

Orfila C, Lepert JC, Alric L, Carrera G, Beraud M and Pipy B (2005). Immunohistochemical distribution of activated nuclear factor kappaB and peroxisome proliferator-activated receptors in carbon tetrachloride-induced chronic liver injury in rats. Histochem Cell Biol 123:585–593. Otsu K, Ito K, Kuzumaki T and Iuchi Y (2001). Differential regulation of liver-specific and ubiquitously-expressed genes in primary rat hepatocytes by the extracellular matrix. Cell Physiol Biochem 11:33–40. Owens RC Jr and Ambrose PG (2005). Antimicrobial safety: focus on fluoroquinolones. Clin Infect Dis (41 Suppl 2):S144–S157. Pahan K, Sheikh FG, Namboodiri AM and Singh I (1998). N -acetyl cysteine inhibits induction of no production by endotoxin or cytokine stimulated rat peritoneal macrophages, C6 glial cells and astrocytes. Free Radic Biol Med 24:39–48. Palmeira CM, Moreno AJ and Madeira VM (1994a). Metabolic alterations in hepatocytes promoted by the herbicides paraquat, dinoseb and 2,4-D. Arch Toxicol 68:24–31. Palmeira CM, Moreno AJ and Madeira VM (1994b). Interactions of herbicides 2,4-D and dinoseb with liver mitochondrial bioenergetics. Toxicol Appl Pharmacol 127:50–57. Palmeira CM, Moreno AJ and Madeira VM (1995a). Mitochondrial bioenergetics is affected by the herbicide paraquat. Biochim Biophys Acta 1229:187–192. Palmeira CM, Moreno AJ and Madeira VM (1995b). Thiols metabolism is altered by the herbicides paraquat, dinoseb and 2,4-D: a study in isolated hepatocytes. Toxicol Lett 81:115–123. Park BK, Kitteringham NR, Maggs JL, Pirmohamed M and Williams DP (2005). The role of metabolic activation in drug-induced hepatotoxicity. Annu Rev Pharmacol Toxicol 45:177–202. Park JE, Yang JH, Yoon SJ, Lee JH, Yang ES and Park JW (2002). Lipid peroxidation-mediated cytotoxicity and DNA damage in U937 cells. Biochimie 84:1199–1205. Parker AR and Eshleman JR (2003). Human MutY: gene structure, protein functions and interactions, and role in carcinogenesis. Cell Mol Life Sci 60:2064–2083. Parola M, Muraca R, Dianzani I, et al. (1992). Vitamin E dietary supplementation inhibits transforming growth factor beta 1 gene expression in the rat liver. FEBS Lett 308:267–270. Parola M, Pinzani M, Casini A, et al. (1993). Stimulation of lipid peroxidation or 4-hydroxynonenal treatment increases procollagen alpha 1 (I) gene expression in human liver fat-storing cells. Biochem Biophys Res Commun 194:1044–1050. Pascussi JM, Jounaidi Y, Drocourt L, et al. (1999). Evidence for the presence of a functional pregnane X receptor response element in the CYP3A7 promoter gene. Biochem Biophys Res Commun 260:377–381. Pascussi JM, Drocourt L, Fabre JM, Maurel P and Vilarem MJ (2000a). Dexamethasone induces pregnane X receptor and retinoid X receptor-alpha expression in human hepatocytes: synergistic increase of CYP3A4 induction by pregnane X receptor activators. Mol Pharmacol 58:361–372. Pascussi JM, Gerbal-Chaloin S, Fabre JM, Maurel P and Vilarem MJ (2000b). Dexamethasone enhances constitutive androstane receptor expression in human hepatocytes: consequences on cytochrome P450 gene regulation. Mol Pharmacol 58:1441–1450. Pascussi JM, Gerbal-Chaloin S, Pichard-Garcia L, et al. (2000c). Interleukin-6 negatively regulates the expression of pregnane X receptor and constitutively activated receptor in primary human hepatocytes. Biochem Biophys Res Commun 274:707–713. Pascussi JM, Drocourt L, Gerbal-Chaloin S, Fabre JM, Maurel P and Vilarem MJ (2001). Dual effect of dexamethasone on CYP3A4 gene expression in human hepatocytes. Sequential role of glucocorticoid receptor and pregnane X receptor. Eur J Biochem 268:6346–6358. Pascussi JM, Busson-Le Coniat M, Maurel P and Vilarem MJ (2003a). Transcriptional analysis of the orphan nuclear receptor constitutive androstane receptor (NR1I3) gene promoter: identification of a distal glucocorticoid response element. Mol Endocrinol 17:42–55. Pascussi JM, Dvorak Z, Gerbal-Chaloin S, Assenat E, Maurel P and Vilarem MJ (2003b). Pathophysiological factors affecting CAR gene expression. Drug Metab Rev 35:255–268.

276

Hepatotoxicity

Pascussi JM, Gerbal-Chaloin S, Drocourt L, Maurel P and Vilarem MJ (2003c). The expression of CYP2B6, CYP2C9 and CYP3A4 genes: a tangle of networks of nuclear and steroid receptors. Biochim Biophys Acta 1619:243–253. Patel T, Roberts LR, Jones BA and Gores GJ (1998). Dysregulation of apoptosis as a mechanism of liver disease: an overview. Semin Liver Dis 18:105–114. Paulusma CC, Kool M, Bosma PJ, et al. (1997). A mutation in the human canalicular multispecific organic anion transporter gene causes the Dubin–Johnson syndrome. Hepatology 25:1539–1542. Pedraza-Alva G, Koulnis M, Charland C, et al. (2006). Activation of p38 MAP kinase by DNA double-strand breaks in V(D)J recombination induces a G2/M cell cycle checkpoint. EMBO J 25:763–773. Pessayre D, Berson A, Fromenty B and Mansouri A (2001). Mitochondria in steatohepatitis. Semin Liver Dis 21:57–69. Pesu M, Aittomaki S, Takaluoma K, Lagerstedt A and Silvennoinen O (2002). p38 Mitogen-activated protein kinase regulates interleukin-4-induced gene expression by stimulating STAT6-mediated transcription. J Biol Chem 277:38254–38261. Peter ME and Krammer PH (2003). The CD95(APO-1/Fas) DISC and beyond. Cell Death Differ 10:26–35. Peters JM and Wiley LM (1995). Evidence that murine preimplantation embryos express aryl hydrocarbon receptor. Toxicol Appl Pharmacol 134:214–221. Peters JM, Morishima H, Ward JM, Coakley CJ, Kimura S and Gonzalez FJ (1999). Role of CYP1A2 in the toxicity of long-term phenacetin feeding in mice. Toxicol Sci 50:82–89. Pianko S, Patella S, Ostapowicz G, Desmond P and Sievert W (2001). Fas-mediated hepatocyte apoptosis is increased by hepatitis C virus infection and alcohol consumption, and may be associated with hepatic fibrosis: mechanisms of liver cell injury in chronic hepatitis C virus infection. J Viral Hepat 8:406–413. Piao CS, Che Y, Han PL and Lee JK (2002). Delayed and differential induction of p38 MAPK isoforms in microglia and astrocytes in the brain after transient global ischemia. Brain Res Mol Brain Res 107:137–144. Pietrangelo A and Shafritz DA (1994). Homeostatic regulation of hepatocyte nuclear transcription factor 1 expression in cultured hepatoma cells. Proc Natl Acad Sci USA 91:182–186. Piguet PF, Grau GE and Vassalli P (1990). Subcutaneous perfusion of tumor necrosis factor induces local proliferation of fibroblasts, capillaries, and epidermal cells, or massive tissue necrosis. Am J Pathol 136:103–110. Pineau T, Fernandez-Salguero P, Lee SS, McPhail T, Ward JM and Gonzalez FJ (1995). Neonatal lethality associated with respiratory distress in mice lacking cytochrome P450 1A2. Proc Natl Acad Sci USA 92:5134–5138. Pinkus R, Weiner LM and Daniel V (1996). Role of oxidants and antioxidants in the induction of AP-1, NF-kappaB and glutathione S-transferase gene expression. J Biol Chem 271:13422–13429. Pinzani M and Marra F (2001). Cytokine receptors and signaling in hepatic stellate cells. Semin Liver Dis 21:397–416. Pirmohamed M and Park BK (2001). Genetic susceptibility to adverse drug reactions. Trends Pharmacol Sci 22:298–305. Pirmohamed M and Park BK (2003). Cytochrome P450 enzyme polymorphisms and adverse drug reactions. Toxicology 192:23–32. Pirotte JH (1984). Apparent potentiation of hepatotoxicity from small doses of acetaminophen by phenobarbital. Ann Intern Med 101:403. Platanias LC (2003). The p38 mitogen-activated protein kinase pathway and its role in interferon signaling. Pharmacol Ther 98:129–142. Pohl LR (1990). Drug-induced allergic hepatitis. Semin Liver Dis 10:305–315. Poli G (1993). Liver damage due to free radicals. Br Med Bull 49:604–620.


277

Poli G (2000). Pathogenesis of liver fibrosis: role of oxidative stress. Mol Aspects Med 21:49–98. Poli G, Albano E and Dianzani MU (1987). The role of lipid peroxidation in liver damage. Chem Phys Lipids 45:117–142. Poppelmann B, Klimmek K, Strozyk E, Voss R, Schwarz T and Kulms D (2005). NF{kappa}Bdependent down-regulation of tumor necrosis factor receptor-associated proteins contributes to interleukin-1-mediated enhancement of ultraviolet B-induced apoptosis. J Biol Chem 280:15635– 15643. Powley MW and Carlson GP (2001). Hepatic and pulmonary microsomal benzene metabolism in CYP2E1 knockout mice. Toxicology 169:187–194. Prandota J (2005). Important role of proinflammatory cytokines/other endogenous substances in druginduced hepatotoxicity: depression of drug metabolism during infections/inflammation states, and genetic polymorphisms of drug-metabolizing enzymes/cytokines may markedly contribute to this pathology. Am J Ther 12:254–261. Prestera T and Talalay P (1995). Electrophile and antioxidant regulation of enzymes that detoxify carcinogens. Proc Natl Acad Sci USA 92:8965–8969. Qiao L, Studer E, Leach K, et al. (2001). Deoxycholic acid (DCA) causes ligand-independent activation of epidermal growth factor receptor (EGFR) and FAS receptor in primary hepatocytes: inhibition of EGFR/mitogen-activated protein kinase-signaling module enhances DCA-induced apoptosis. Mol Biol Cell 12:2629–2645. Quattrochi LC, Vu T and Tukey RH (1994). The human CYP1A2 gene and induction by 3methylcholanthrene. A region of DNA that supports AH-receptor binding and promoter-specific induction. J Biol Chem 269:6949–6954. Radjendirane V, Joseph P, Lee YH, et al. (1998). Disruption of the DT diaphorase (NQO1) gene in mice leads to increased menadione toxicity. J Biol Chem 273:7382–7389. Raingeaud J, Gupta S, Rogers JS, et al. (1995). Pro-inflammatory cytokines and environmental stress cause p38 mitogen-activated protein kinase activation by dual phosphorylation on tyrosine and threonine. J Biol Chem 270:7420–7426. Ramm GA and Ruddell RG (2005). Hepatotoxicity of iron overload: mechanisms of iron-induced hepatic fibrogenesis. Semin Liver Dis 25:433–449. Rashba-Step J, Turro NJ and Cederbaum AI (1993). ESR studies on the production of reactive oxygen intermediates by rat liver microsomes in the presence of NADPH or NADH. Arch Biochem Biophys 300:391–400. Rastegar M, Rousseau GG and Lemaigre FP (2000). CCAAT/enhancer-binding protein-alpha is a component of the growth hormone-regulated network of liver transcription factors. Endocrinology 141:1686–1692. Raucy JL, Lasker JM, Lieber CS and Black M (1989). Acetaminophen activation by human liver cytochromes P450IIE1 and P450IA2. Arch Biochem Biophys 271:270–283. Rauen U, Petrat F, Sustmann R and de Groot H (2004). Iron-induced mitochondrial permeability transition in cultured hepatocytes. J Hepatol 40:607–615. Ravi D and Das KC (2004). Redox-cycling of anthracyclines by thioredoxin system: increased superoxide generation and DNA damage. Cancer Chemother Pharmacol 54:449–458. Ravingerova T, Barancik M and Strniskova M (2003). Mitogen-activated protein kinases: a new therapeutic target in cardiac pathology. Mol Cell Biochem 247:127–138. Raza M, Ahmad M, Gado A and Al Shabanah OA (2003). A comparison of hepatoprotective activities of aminoguanidine and N -acetylcysteine in rat against the toxic damage induced by azathioprine. Comp Biochem Physiol C Toxicol Pharmacol 134:451–456. Reasor MJ and Kacew S (2001). Drug-induced phospholipidosis: are there functional consequences? Exp Biol Med (Maywood) 226:825–830. Refik MM, Comert B, Oncu K, et al. (2004). The effect of taurine treatment on oxidative stress in experimental liver fibrosis. Hepatol Res 28:207–215.

278

Hepatotoxicity

Reid AB, Kurten RC, McCullough SS, Brock RW and Hinson JA (2005). Mechanisms of acetaminophen-induced hepatotoxicity: role of oxidative stress and mitochondrial permeability transition in freshly isolated mouse hepatocytes. J Pharmacol Exp Ther 312:509–516. Ren Y and Smith A (1995). Mechanism of metallothionein gene regulation by heme-hemopexin. Roles of protein kinase C, reactive oxygen species, and cis-acting elements. J Biol Chem 270:23988– 23995. Ren X, Kennedy A and Colletti LM (2002). CXC chemokine expression after stimulation with interferon-gamma in primary rat hepatocytes in culture. Shock 17:513–520. Repa JJ and Mangelsdorf DJ (2000). The role of orphan nuclear receptors in the regulation of cholesterol homeostasis. Annu Rev Cell Dev Biol 16:459–481. Ribeiro PS, Cortez-Pinto H, Sola S, et al. (2004). Hepatocyte apoptosis, expression of death receptors, and activation of NF-kappaB in the liver of nonalcoholic and alcoholic steatohepatitis patients. Am J Gastroenterol 99:1708–1717. Riely CA, Cohen G and Lieberman M (1974). Ethane evolution: a new index of lipid peroxidation. Science 183:208–210. Rifkind AB (2006). CYP1A in TCDD toxicity and in physiology – with particular reference to CYP dependent arachidonic acid metabolism and other endogenous substrates. Drug Metab Rev 38:291–335. Rincon M (2001). MAP-kinase signaling pathways in T cells. Curr Opin Immunol 13:339–345. Rizzo G, Renga B, Mencarelli A, Pellicciari R and Fiorucci S (2005). Role of FXR in regulating bile acid homeostasis and relevance for human diseases. Curr Drug Targets Immune Endocr Metabol Disord 5:289–303. Roberts RA, James NH, Woodyatt NJ, Macdonald N and Tugwood JD (1998). Evidence for the suppression of apoptosis by the peroxisome proliferator activated receptor alpha (PPAR alpha). Carcinogenesis 19:43–48. Rolo AP, Palmeira CM and Wallace KB (2003). Mitochondrially mediated synergistic cell killing by bile acids. Biochim Biophys Acta 1637:127–132. Romero FJ, Bosch-Morell F, Romero MJ, et al. (1998). Lipid peroxidation products and antioxidants in human disease. Environ Health Perspect 106(Suppl 5):1229–1234. Rossi MA, Di Mauro C and Dianzani MU (1993). Action of lipid peroxidation products on phosphoinositide specific phospholipase C. Mol Aspects Med 14:273–279. Rouse J, Cohen P, Trigon S, et al. (1994). A novel kinase cascade triggered by stress and heat shock that stimulates MAPKAP kinase-2 and phosphorylation of the small heat shock proteins. Cell 78:1027–1037. Rowell DL, Eckmann L, Dwinell MB, et al. (1997). Human hepatocytes express an array of proinflammatory cytokines after agonist stimulation or bacterial invasion. Am J Physiol 273:G322–G332. Runge-Morris M, Wu W and Kocarek TA (1999). Regulation of rat hepatic hydroxysteroid sulfotransferase (SULT2-40/41) gene expression by glucocorticoids: evidence for a dual mechanism of transcriptional control. Mol Pharmacol 56:1198–1206. Rushmore TH and Pickett CB (1990). Transcriptional regulation of the rat glutathione S-transferase Ya subunit gene. Characterization of a xenobiotic-responsive element controlling inducible expression by phenolic antioxidants. J Biol Chem 265:14648–14653. Rushmore TH and Pickett CB (1993). Glutathione S-transferases, structure, regulation and therapeutic implications. J Biol Chem 268:11475–11478. Rushmore TH, Morton MR and Pickett CB (1991). The antioxidant responsive element. Activation by oxidative stress and identification of the DNA consensus sequence required for functional activity. J Biol Chem 266:11632–11639. Russell DW and Setchell KD (1992). Bile acid biosynthesis. Biochemistry 31:4737–4749. Saif MW and Diasio R (2006). Is capecitabine safe in patients with gastrointestinal cancer and dihydropyrimidine dehydrogenase deficiency? Clin Colorectal Cancer 5:359–362.


279

Saitoh T, Kokue E and Shimoda M (1999). The suppressive effects of lipopolysaccharide-induced acute phase response on hepatic cytochrome P450-dependent drug metabolism in rabbits. J Vet Pharmacol Ther 22:87–95. Sajan MP, Satav JG and Bhattacharya RK (1995). Activity of some respiratory enzymes and cytochrome contents in rat hepatic mitochondria following aflatoxin B1 administration. Toxicol Lett 80:55–60. Sajan MP, Satav JG and Bhattacharya RK (1996). Alteration of energy-linked functions in rat hepatic mitochondria following aflatoxin B1 administration. J Biochem Toxicol 11:235–241. Sajan MP, Satav JG and Bhattacharya RK (1997). Effect of aflatoxin B in vitro on rat liver mitochondrial respiratory functions. Ind J Exp Biol 35:1187–1190. Sakuma T, Miyamoto T, Jiang W, et al. (2003). Inhibition of peroxisome proliferator-activated receptor alpha signaling by vitamin D receptor. Biochem Biophys Res Commun 312:513–519. Samadani U, Porcella A, Pani L, et al. (1995). Cytokine regulation of the liver transcription factor hepatocyte nuclear factor-3 beta is mediated by the C/EBP family and interferon regulatory factor 1. Cell Growth Differ 6:879–890. Sampietro M and Iolascon A (1999). Molecular pathology of Crigler–Najjar type I and II and Gilbert’s syndromes. Haematologica 84:150–157. Sanderson J, Ansari A, Marinaki T and Duley J (2004). Thiopurine methyltransferase: should it be measured before commencing thiopurine drug therapy? Ann Clin Biochem 41:294–302. Sanyal AJ, Campbell-Sargent C, Mirshahi F, et al. (2001). Nonalcoholic steatohepatitis: association of insulin resistance and mitochondrial abnormalities. Gastroenterology 120:1183–1192. Savkur RS, Wu Y, Bramlett KS, et al. (2003). Alternative splicing within the ligand binding domain of the human constitutive androstane receptor. Mol Genet Metab 80:216–226. Sayre LM, Smith MA and Perry G (2001). Chemistry and biochemistry of oxidative stress in neurodegenerative disease. Curr Med Chem 8:721–838. Scaffidi C, Fulda S, Srinivasan A, et al. (1998). Two CD95 (APO-1/Fas) signaling pathways. EMBO J 17:1675–1687. Scandalios JG (2005). Oxidative stress: molecular perception and transduction of signals triggering antioxidant gene defenses. Braz J Med Biol Res 38:995–1014. Schneiderhan W, Schmid-Kotsas A, Zhao J, et al. (2001). Oxidized low-density lipoproteins bind to the scavenger receptor, CD36, of hepatic stellate cells and stimulate extracellular matrix synthesis. Hepatology 34:729–737. Schreiber AJ and Simon FR (1983). Estrogen-induced cholestasis: clues to pathogenesis and treatment. Hepatology 3:607–813. Schrem H, Klempnauer J and Borlak J (2002). Liver-enriched transcription factors in liver function and development. Part I: the hepatocyte nuclear factor network and liver-specific gene expression. Pharmacol Rev 54:129–158. Schrem H, Klempnauer J and Borlak J (2004). Liver-enriched transcription factors in liver function and development. Part II: the C/EBPs and D site-binding protein in cell cycle control, carcinogenesis, circadian gene regulation, liver regeneration, apoptosis and liver-specific gene regulation. Pharmacol Rev 56:291–330. Schuetz JD, Schuetz EG, Thottassery JV, Guzelian PS, Strom S and Sun D (1996). Identification of a novel dexamethasone responsive enhancer in the human CYP3A5 gene and its activation in human and rat liver cells. Mol Pharmacol 49:63–72. Schuppan D (1990). Structure of the extracellular matrix in normal and fibrotic liver: collagens and glycoproteins. Semin Liver Dis 10:1–10. Schwabe RF and Brenner DA (2006). Mechanisms of Liver Injury. I. TNF-alpha-induced liver injury: role of IKK, JNK and ROS pathways. Am J Physiol Gastrointest Liver Physiol 290:G583–G589. Seidman EG (2003). Clinical use and practical application of TPMT enzyme and 6-mercaptopurine metabolite monitoring in IBD. Rev Gastroenterol Disord 3(Suppl 1):S30–S38.

280

Hepatotoxicity

Seitz HK and Stickel F (2006). Risk factors and mechanisms of hepatocarcinogenesis with special emphasis on alcohol and oxidative stress. Biol Chem 387:349–360. Sen CK (2000). Cellular thiols and redox-regulated signal transduction. Curr Top Cell Regul 36:1–30. Shaulian E and Karin M (2002). AP-1 as a regulator of cell life and death. Nat Cell Biol 4:E131–E136. Shen HM and Pervaiz S (2006). TNF receptor superfamily-induced cell death: redox-dependent execution. FASEB J 20:1589–1598. Shen BJ, Chang CJ, Lee HS, Tsai WH, Miau LH and Lee SC (1997). Transcriptional induction of the agp/ebp (c/ebp beta) gene by hepatocyte growth factor. DNA Cell Biol 16:703–711. Shimizu Y, Nakatsuru Y, Ichinose M, et al. (2000). Benzo[a]pyrene carcinogenicity is lost in mice lacking the aryl hydrocarbon receptor. Proc Natl Acad Sci USA 97:779–782. Shimizu M, Tamura T, Yamada Y, Akiyama Y, Saijo N and Nishio K (2001). CPT-11 alters the circadian rhythm of dihydropyrimidine dehydrogenase mRNA in mouse liver. Jpn J Cancer Res 92:554–561. Shishodia S and Aggarwal BB (2002). Nuclear factor-kappaB activation: a question of life or death. J Biochem Mol Biol 35:28–40. Sidhu JS, Liu F and Omiecinski CJ (2004). Phenobarbital responsiveness as a uniquely sensitive indicator of hepatocyte differentiation status: requirement of dexamethasone and extracellular matrix in establishing the functional integrity of cultured primary rat hepatocytes. Exp Cell Res 292:252–264. Sies H (1991). Role of reactive oxygen species in biological processes. Klin Wochenschr 69:965–968. Sies C, Florkowski C, George P, et al. (2005). Measurement of thiopurine methyl transferase activity guides dose-initiation and prevents toxicity from azathioprine. NZ Med J 118:U1324. Simon FR, Fortune J, Iwahashi M, Gartung C, Wolkoff A and Sutherland E (1996). Ethinyl estradiol cholestasis involves alterations in expression of liver sinusoidal transporters. Am J Physiol 271:G1043–G1052. Sinal CJ, Tohkin M, Miyata M, Ward JM, Lambert G and Gonzalez FJ (2000). Targeted disruption of the nuclear receptor FXR/BAR impairs bile acid and lipid homeostasis. Cell 102:731–744. Sinclair J, Jeffery E, Wrighton S, et al. (1998). Alcohol-mediated increases in acetaminophen hepatotoxicity: role of CYP2E and CYP3A. Biochem Pharmacol 55:1557–1565. Sinclair PR, Gorman N, Walton HS, et al. (2000). CYP1A2 is essential in murine uroporphyria caused by hexachlorobenzene and iron. Toxicol Appl Pharmacol 162:60–67. Slater TF (1984). Free-radical mechanisms in tissue injury. Biochem J 222:1–15. Smedsrod B (2004). Clearance function of scavenger endothelial cells. Comp Hepatol 3(Suppl 1):S22. Smialowicz RJ, Burgin DE, Williams WC, Diliberto JJ, Setzer RW and Birnbaum LS (2004). CYP1A2 is not required for 2,3,7,8-tetrachlorodibenzo- p-dioxin-induced immunosuppression. Toxicology 197:15–22. Smith MT (2003). Mechanisms of troglitazone hepatotoxicity. Chem Res Toxicol 16:679–687. Smith AG, Clothier B, Carthew P, et al. (2001). Protection of the Cyp1a2(−/−) null mouse against uroporphyria and hepatic injury following exposure to 2,3,7,8-tetrachlorodibenzo- p-dioxin. Toxicol Appl Pharmacol 173:89–98. Smith NF, Figg WD and Sparreboom A (2005). Role of the liver-specific transporters OATP1B1 and OATP1B3 in governing drug elimination. Expert Opin Drug Metab Toxicol 1:429–445. Sneed RA, Buchweitz JP, Jean PA and Ganey PE (2000). Pentoxifylline attenuates bacterial lipopolysaccharide-induced enhancement of allyl alcohol hepatotoxicity. Toxicol Sci 56:203– 210. Sodeman T, Bronk SF, Roberts PJ, Miyoshi H and Gores GJ (2000). Bile salts mediate hepatocyte apoptosis by increasing cell surface trafficking of Fas. Am J Physiol Gastrointest Liver Physiol 278:G992–G999. Sodhi CP, Rana SV, Mehta SK, Vaiphei K, Attari S and Mehta S (1997). Study of oxidative-stress in isoniazid–rifampicin induced hepatic injury in young rats. Drug Chem Toxicol 20:255–269.


281

Sokol RJ, Devereaux M, Mierau GW, Hambidge KM and Shikes RH (1990). Oxidant injury to hepatic mitochondrial lipids in rats with dietary copper overload. Modification by vitamin E deficiency. Gastroenterology 99:1061–1071. Soni MG and Mehendale HM (1998). Role of tissue repair in toxicologic interactions among hepatotoxic organics. Environ Health Perspect 106 (Suppl 6):1307–1317. Spath GF and Weiss MC (1997). Hepatocyte nuclear factor 4 expression overcomes repression of the hepatic phenotype in dedifferentiated hepatoma cells. Mol Cell Biol 17:1913–1922. Spector MS, Auer KL, Jarvis WD, et al. (1997). Differential regulation of the mitogen-activated protein and stress-activated protein kinase cascades by adrenergic agonists in quiescent and regenerating adult rat hepatocytes. Mol Cell Biol 17:3556–3565. Stadtman ER (1992). Protein oxidation and aging. Science 257:1220–1224. Stadtman ER and Berlett BS (1998). Reactive oxygen-mediated protein oxidation in aging and disease. Drug Metab Rev 30:225–243. Staudinger J, Liu Y, Madan A, Habeebu S and Klaassen CD (2001a). Coordinate regulation of xenobiotic and bile acid homeostasis by pregnane X receptor. Drug Metab Dispos 29:1467–1472. Staudinger JL, Goodwin B, Jones SA, et al. (2001b). The nuclear receptor PXR is a lithocholic acid sensor that protects against liver toxicity. Proc Natl Acad Sci USA 98:3369–3374. Stedman CA, Liddle C, Coulter SA, et al. (2005). Nuclear receptors constitutive androstane receptor and pregnane X receptor ameliorate cholestatic liver injury. Proc Natl Acad Sci USA 102:2063– 2068. Stedman C, Liddle C, Coulter S, et al. (2006). Benefit of farnesoid X receptor inhibition in obstructive cholestasis. Proc Natl Acad Sci USA 103:11323–11328. Steiner M, Seule M, Steiner B, et al. (2005). 5-Fluorouracil/irinotecan induced lethal toxicity as a result of a combined pharmacogenetic syndrome: report of a case. J Clin Pathol 58:553–555. Stieger B, O’Neill B and Meier PJ (1992). ATP-dependent bile-salt transport in canalicular rat liver plasma–membrane vesicles. Biochem J 284:67–74. Strand S, Hofmann WJ, Grambihler A, et al. (1998). Hepatic failure and liver cell damage in acute Wilson’s disease involve CD95 (APO-1/Fas) mediated apoptosis. Nat Med 4:588–593. Strautnieks SS, Bull LN, Knisely AS, et al. (1998). A gene encoding a liver-specific ABC transporter is mutated in progressive familial intrahepatic cholestasis. Nat Genet 20:233–238. Streeper RS, Eaton EM, Ebert DH, Chapman SC, Svitek CA and O’Brien RM (1998). Hepatocyte nuclear factor-1 acts as an accessory factor to enhance the inhibitory action of insulin on mouse glucose-6-phosphatase gene transcription. Proc Natl Acad Sci USA 95:9208–9213. Suchy FJ and Ananthanarayanan M (2006). Bile salt excretory pump: biology and pathobiology. J Pediatr Gastroenterol Nutr 43:S10–S16. Sueyoshi T, Kawamoto T, Zelko I, Honkakoski P and Negishi M (1999). The repressed nuclear receptor CAR responds to phenobarbital in activating the human CYP2B6 gene. J Biol Chem 274:6043–6046. Sueyoshi T and Negishi M (2001). Phenobarbital response elements of cytochrome P450 genes and nuclear receptors. Annu Rev Pharmacol Toxicol 41:123–143. Sugatani J, Yamakawa K, Tonda E, et al. (2004). The induction of human UDPglucuronosyltransferase 1A1 mediated through a distal enhancer module by flavonoids and xenobiotics. Biochem Pharmacol 67:989–1000. Sugihara N, Furuno K, Kita N, Murakami T and Yata N (1991). Increase in the plasma protein binding of weakly basic drugs in carbon tetrachloride-intoxicated rats. Chem Pharm Bull (Tokyo) 39:2671–2673. Sugihara N, Furuno K, Kita N, Murakami T and Yata N (1992). Plasma alpha 1-acid glycoprotein concentration in rats with chemical liver injury. Chem Pharm Bull (Tokyo) 40:2516–2519. Sumi S, Marinaki AM, Arenas M, et al. (2002). Genetic basis of inosine triphosphate pyrophosphohydrolase deficiency. Hum Genet 111:360–367.

282

Hepatotoxicity

Sumner SC, Fennell TR, Moore TA, Chanas B, Gonzalez F and Ghanayem BI (1999). Role of cytochrome P450 2E1 in the metabolism of acrylamide and acrylonitrile in mice. Chem Res Toxicol 12:1110–1116. Svegliati-Baroni G, Ridolfi F, Di Sario A, et al. (1999). Insulin and insulin-like growth factor-1 stimulate proliferation and type I collagen accumulation by human hepatic stellate cells: differential effects on signal transduction pathways. Hepatology 29:1743–1751. Svegliati-Baroni G, Ridolfi F, Di Sario A, et al. (2001). Intracellular signaling pathways involved in acetaldehyde-induced collagen and fibronectin gene expression in human hepatic stellate cells. Hepatology 33:1130–1140. Swanson HI and Bradfield CA (1993). The AH-receptor: genetics, structure and function. Pharmacogenetics 3:213–230. Synold TW, Dussault I and Forman BM (2001). The orphan nuclear receptor SXR coordinately regulates drug metabolism and efflux. Nat Med 7:584–590. Szatmari I, Vamosi G, Brazda P, et al. (2006). PPAR gamma-regulated ABCG2 expression confers cytoprotection to human dendritic cells. J Biol Chem 281:23812–23823. Szewczyk A and Wojtczak L (2002). Mitochondria as a pharmacological target. Pharmacol Rev 54:101–127. Tagami A, Ohnishi H, Moriwaki H, Phillips M and Hughes RD (2003). Fas-mediated apoptosis in acute alcoholic hepatitis. Hepatogastroenterology 50:443–448. Taieb J, Mathurin P, Poynard T, Gougerot-Pocidalo MA and Chollet-Martin S (1998). Raised plasma soluble Fas and Fas-ligand in alcoholic liver disease. Lancet 351:1930–1931. Takehara T, Tatsumi T, Suzuki T, et al. (2004). Hepatocyte-specific disruption of Bcl-xL leads to continuous hepatocyte apoptosis and liver fibrotic responses. Gastroenterology 127:1189–1197. Talalay P, Fahey JW, Holtzclaw WD, Prestera T and Zhang Y (1995). Chemoprotection against cancer by phase 2 enzyme induction. Toxicol Lett 82–83:173–179. Tanaka Y, Chen C, Maher JM and Klaassen CD (2006). Kupffer cell-mediated downregulation of hepatic transporter expression in rat hepatic ischemia-reperfusion. Transplantation 82:258–266. Tang L, Tanaka Y, Marumo F and Sato C (1994). Phenotypic change in portal fibroblasts in biliary fibrosis. Liver 14:76–82. Tang YM, Chen GF, Thompson PA, Lin DX, Lang NP and Kadlubar FF (1999). Development of an antipeptide antibody that binds to the C-terminal region of human CYP1B1. Drug Metab Dispos 27:274–280. Tapner MJ, Jones BE, Wu WM and Farrell GC (2004). Toxicity of low dose azathioprine and 6mercaptopurine in rat hepatocytes. Roles of xanthine oxidase and mitochondrial injury. J Hepatol 40:454–463. Tenenbaum A, Motro M and Fisman EZ (2005). Dual and pan-peroxisome proliferator-activated receptors (PPAR) co-agonism: the bezafibrate lessons. Cardiovasc Diabetol 4:14. Thakur V, Pritchard MT, McMullen MR, Wang Q and Nagy LE (2006). Chronic ethanol feeding increases activation of NADPH oxidase by lipopolysaccharide in rat Kupffer cells: role of increased reactive oxygen in LPS-stimulated ERK1/2 activation and TNF-alpha production. J Leukoc Biol 79:1348–1356. Thor H, Mirabelli F, Salis A, Cohen GM, Bellomo G and Orrenius S (1988). Alterations in hepatocyte cytoskeleton caused by redox cycling and alkylating quinones. Arch Biochem Biophys 266:397– 407. Thummel KE, Lee CA, Kunze KL, Nelson SD and Slattery JT (1993). Oxidation of acetaminophen to N -acetyl- p-aminobenzoquinone imine by human CYP3A4. Biochem Pharmacol 45:1563–1569. Tirona RG, Lee W, Leake BF, et al. (2003). The orphan nuclear receptor HNF4alpha determines PXR- and CAR-mediated xenobiotic induction of CYP3A4. Nat Med 9:220–224. Tolson JK, Dix DJ, Voellmy RW and Roberts SM (2006). Increased hepatotoxicity of acetaminophen in Hsp70i knockout mice. Toxicol Appl Pharmacol 210:157–162.


283

Tonge RP, Kelly EJ, Bruschi SA, et al. (1998). Role of CYP1A2 in the hepatotoxicity of acetaminophen: investigations using Cyp1a2 null mice. Toxicol Appl Pharmacol 153:102–108. Tracey KJ, Fong Y, Hesse DG, et al. (1987). Anti-cachectin/TNF monoclonal antibodies prevent septic shock during lethal bacteraemia. Nature 330:662–664. Trauner M, Wagner M, Fickert P and Zollner G (2005). Molecular regulation of hepatobiliary transport systems: clinical implications for understanding and treating cholestasis. J Clin Gastroenterol 39:S111–S124. Trost LC and Lemasters JJ (1996). The mitochondrial permeability transition: a new pathophysiological mechanism for Reye’s syndrome and toxic liver injury. J Pharmacol Exp Ther 278:1000– 1005. Tsokos-Kuhn JO, Hughes H, Smith CV and Mitchell JR (1988). Alkylation of the liver plasma membrane and inhibition of the Ca2+ ATPase by acetaminophen. Biochem Pharmacol 37:2125– 2131. Tuchweber B, Desmouliere A, Bochaton-Piallat ML, Rubbia-Brandt L and Gabbiani G (1996). Proliferation and phenotypic modulation of portal fibroblasts in the early stages of cholestatic fibrosis in the rat. Lab Invest 74:265–278. Tzameli I, Pissios P, Schuetz EG and Moore DD (2000). The xenobiotic compound 1,4-bis[2-(3,5dichloropyridyloxy)]benzene is an agonist ligand for the nuclear receptor CAR. Mol Cell Biol 20:2951–2958. Tzung SP, Fausto N and Hockenbery DM (1997). Expression of Bcl-2 family during liver regeneration and identification of Bcl-x as a delayed early response gene. Am J Pathol 150:1985–1995. Uno S, Dalton TP, Shertzer HG, et al. (2001). Benzo[a]pyrene-induced toxicity: paradoxical protection in Cyp1a1(−/−) knockout mice having increased hepatic BaP-DNA adduct levels. Biochem Biophys Res Commun 289:1049–1056. Uno S, Dalton TP, Derkenne S, Curran CP, Miller ML, Shertzer HG, Nebert DW (2004a). Oral exposure to benzo[a]pyrene in the mouse: detoxication by inducible cytochrome P450 is more important than metabolic activation. Mol Pharmacol 65:1225–1237. Uno S, Dalton TP, Sinclair PR, et al. (2004b). Cyp1a1(−/−) male mice: protection against high-dose TCDD-induced lethality and wasting syndrome, and resistance to intrahepatocyte lipid accumulation and uroporphyria. Toxicol Appl Pharmacol 196:410–421. Valentine JL, Lee SS, Seaton MJ, et al. (1996). Reduction of benzene metabolism and toxicity in mice that lack CYP2E1 expression. Toxicol Appl Pharmacol 141:205–213. Valentino KL, Gutierrez M, Sanchez R, Winship MJ and Shapiro DA (2003). First clinical trial of a novel caspase inhibitor: anti-apoptotic caspase inhibitor, IDN-6556, improves liver enzymes. Int J Clin Pharmacol Ther 41:441–449. Van Huyen JP, Batisse D, Heudes D, et al. (2006). Alteration of cytochrome oxidase subunit I labeling is associated with severe mitochondriopathy in NRTI-related hepatotoxicity in HIV patients. Mod Pathol 19:1277–1288. van Kuilenburg AB (2006). Screening for dihydropyrimidine dehydrogenase deficiency: to do or not to do, that’s the question. Cancer Invest 24:215–217. van Kuilenburg AB, Meinsma R and van Gennip AH (2004). Pyrimidine degradation defects and severe 5-fluorouracil toxicity. Nucleosides Nucleotides Nucleic Acids 23:1371–1375. Vanden Heuvel JP (1999). Peroxisome proliferator-activated receptors (PPARS) and carcinogenesis. Toxicol Sci 47:1–8. Vatsyayan J, Lee SJ and Chang HY (2005). Effects of xenobiotics and peroxisome proliferatoractivated receptor-alpha on the human UDPglucose dehydrogenase gene expression. J Biochem Mol Toxicol 19:279–288. Vendemiale G, Grattagliano I, Altomare E, Turturro N and Guerrieri F (1996). Effect of acetaminophen administration on hepatic glutathione compartmentation and mitochondrial energy metabolism in the rat. Biochem Pharmacol 52:1147–1154.

284

Hepatotoxicity

Verhagen AM, Ekert PG, Pakusch M, et al. (2000). Identification of DIABLO, a mammalian protein that promotes apoptosis by binding to and antagonizing IAP proteins. Cell 102:43–53. Verma A, Deb DK, Sassano A, et al. (2002). Activation of the p38 mitogen-activated protein kinase mediates the suppressive effects of type I interferons and transforming growth factor-beta on normal hematopoiesis. J Biol Chem 277:7726–7735. Vickers AE, Bentley P and Fisher RL (2006). Consequences of mitochondrial injury induced by pharmaceutical fatty acid oxidation inhibitors is characterized in human and rat liver slices. Toxicol in vitro 20:1173–1182. Videla LA, Fernandez V, Tapia G and Varela P (2003). Oxidative stress-mediated hepatotoxicity of iron and copper: role of Kupffer cells. Biometals 16:103–111. Vrba J and Modriansky M (2002). Oxidative burst of Kupffer cells: target for liver injury treatment. Biomed Pap Med Fac Univ Palacky Olomouc Czech Repub 146:15–20. Wajant H, Pfizenmaier K and Scheurich P (2003). Tumor necrosis factor signaling. Cell Death Differ 10:45–65. Wakabayashi K, Tamura A, Saito H, Onishi Y and Ishikawa T (2006). Human ABC transporter ABCG2 in xenobiotic protection and redox biology. Drug Metab Rev 38:371–391. Wallace KB, Eells JT, Madeira VM, Cortopassi G and Jones DP (1997). Mitochondria-mediated cell injury. Symposium overview. Fundam Appl Toxicol 38:23–37. Wang XZ and Ron D (1996). Stress-induced phosphorylation and activation of the transcription factor CHOP (GADD153) by p38 MAP kinase. Science 272:1347–1349. Wang H, Gao X, Fukumoto S, Tademoto S, Sato K and Hirai K (1999). Differential expression and regulation of chemokines JE, KC and IP-10 gene in primary cultured murine hepatocytes. J Cell Physiol 181:361–370. Wang YM, Feng GH, Huang F, Li Y and Zhao GZ (2003). Tumor necrosis factor-alpha, caspase-3 expression and hepatocyte apoptosis in fulminanting hepatic failure. Zhonghua Nei Ke Za Zhi 42:566–570. Wang WH, Gregori G, Hullinger RL and Andrisani OM (2004). Sustained activation of p38 mitogenactivated protein kinase and c-Jun N -terminal kinase pathways by hepatitis B virus X protein mediates apoptosis via induction of Fas/FasL and tumor necrosis factor (TNF) receptor 1/TNFalpha expression. Mol Cell Biol 24:10352–10365. Waring JF, Liguori MJ, Luyendyk JP, et al. (2006). Microarray analysis of lipopolysaccharide potentiation of trovafloxacin-induced liver injury in rats suggests a role for proinflammatory chemokines and neutrophils. J Pharmacol Exp Ther 316:1080–1087. Watanabe S and Phillips MJ (1986). Acute phalloidin toxicity in living hepatocytes. Evidence for a possible disturbance in membrane flow and for multiple functions for actin in the liver cell. Am J Pathol 122:101–111. Watt AJ, Garrison WD and Duncan SA (2003). HNF4: a central regulator of hepatocyte differentiation and function. Hepatology 37:1249–1253. Way JM, Harrington WW, Brown KK, et al. (2001). Comprehensive messenger ribonucleic acid profiling reveals that peroxisome proliferator-activated receptor gamma activation has coordinate effects on gene expression in multiple insulin-sensitive tissues. Endocrinology 142:1269– 1277. Wei MC, Zong WX, Cheng EH, et al. (2001). Proapoptotic BAX and BAK: a requisite gateway to mitochondrial dysfunction and death. Science 292:727–730. Wentworth JM, Agostini M, Love J, Schwabe JW and Chatterjee VK (2000). St John’s wort, a herbal antidepressant, activates the steroid X receptor. J Endocrinol 166:R11–R16. Whalen R, Voss SH and Boyer TD (2004). Decreased expression levels of rat liver glutathione S-transferase A2 and albumin during the acute phase response are mediated by HNF1 (hepatic nuclear factor 1) and IL6DEX-NP. Biochem J 377:763–768. White AJ (2001). Mitochondrial toxicity and HIV therapy. Sex Transm Infect 77:158–173.


285

Whiting JF, Green RM, Rosenbluth AB and Gollan JL (1995). Tumor necrosis factor-alpha decreases hepatocyte bile salt uptake and mediates endotoxin-induced cholestasis. Hepatology 22:1273– 1278. Whitlock JP Jr (1999). Induction of cytochrome P4501A1. Annu Rev Pharmacol Toxicol 39:103– 125. Widlak P and Garrard WT (2005). Discovery, regulation and action of the major apoptotic nucleases DFF40/CAD and endonuclease G. J Cell Biochem 94:1078–1087. Wild AC, Gipp JJ and Mulcahy T (1998). Overlapping antioxidant response element and PMA response element sequences mediate basal and beta-naphthoflavone-induced expression of the human gamma-glutamylcysteine synthetase catalytic subunit gene. Biochem J 332:373–381. Wolf CR and Smith G (1999a). Cytochrome P450 CYP2D6, IARC Science Publications, 148: 209– 229. Wolf CR and Smith G (1999b). Pharmacogenetics. Br Med Bull 55:366–386. Wong FW, Chan WY and Lee SS (1998). Resistance to carbon tetrachloride-induced hepatotoxicity in mice which lack CYP2E1 expression. Toxicol Appl Pharmacol 153:109–118. Wu D and Cederbaum AI (1994). Characterization of pyrazole and 4-methylpyrazole induction of cytochrome P4502E1 in rat kidney. J Pharmacol Exp Ther 270:407–413. Wu Y, Zhang X, Bardag-Gorce F, et al. (2004). Retinoid X receptor alpha regulates glutathione homeostasis and xenobiotic detoxification processes in mouse liver. Mol Pharmacol 65:550– 557. Xia D and Samols D (1997). Transgenic mice expressing rabbit C-reactive protein are resistant to endotoxemia. Proc Natl Acad Sci USA 94:2575–2580. Xia X, Francis H, Glaser S, Alpini G and LeSage G (2006). Bile acid interactions with cholangiocytes. World J Gastroenterol 12:3553–3563. Xiao C and Ghosh S (2005). NF-kappaB, an evolutionarily conserved mediator of immune and inflammatory responses. Adv Exp Med Biol 560:41–45. Xie W, Barwick JL, Simon CM, et al. (2000). Reciprocal activation of xenobiotic response genes by nuclear receptors SXR/PXR and CAR. Genes Dev 14:3014–3023. Xie W, Radominska-Pandya A, Shi Y, et al. (2001). An essential role for nuclear receptors SXR/PXR in detoxification of cholestatic bile acids. Proc Natl Acad Sci USA 98:3375–3380. Xie W, Uppal H, Saini SP, et al. (2004). Orphan nuclear receptor-mediated xenobiotic regulation in drug metabolism. Drug Discov Today 9:442–449. Yamada Y and Fausto N (1998). Deficient liver regeneration after carbon tetrachloride injury in mice lacking type 1 but not type 2 tumor necrosis factor receptor. Am J Pathol 152:1577–1589. Yamada Y, Kirillova I, Peschon JJ and Fausto N (1997). Initiation of liver growth by tumor necrosis factor: deficient liver regeneration in mice lacking type I tumor necrosis factor receptor. Proc Natl Acad Sci USA 94:1441–1446. Yamaguchi Y, Minami K, Kawabuchi Y and Emi M (2006). A Phase I trial of CPT-11 in combination with 5-fluorouracil plus leucovorin chemotherapy for patients with metastatic colorectal cancer. Hepatogastroenterology 53:201–205. Yamamoto T, Kikkawa R, Yamada H and Horii I (2005). Identification of oxidative stress-related proteins for predictive screening of hepatotoxicity using a proteomic approach. J Toxicol Sci 30:213–227. Yamamoto Y, Moore R, Hess HA, et al. (2006). Estrogen receptor alpha mediates 17alphaethynylestradiol causing hepatotoxicity. J Biol Chem 281:16625–16631. Yamazaki Y, Kakizaki S, Horiguchi N, Takagi H, Mori M and Negishi M (2005). Role of nuclear receptor CAR in carbon tetrachloride-induced hepatotoxicity. World J Gastroenterol 11:5966– 5972. Yang S, Koteish A, Lin H, et al. (2004a). Oval cells compensate for damage and replicative senescence of mature hepatocytes in mice with fatty liver disease. Hepatology 39:403–411.

286

Hepatotoxicity

Yang YL, Li JP, Xu XP, Dou KF, Yue SQ and Li KZ (2004b). Protective effects of tumor necrosis factor alpha antibody and ulinastatin on liver ischemic reperfusion in rats. World J Gastroenterol 10:3161–3164. Yang H, Magilnick N, Lee C, et al. (2005). Nrf1 and Nrf2 regulate rat glutamate–cysteine ligase catalytic subunit transcription indirectly via NF-kappaB and AP-1. Mol Cell Biol 25:5933–5946. Yates CR, Krynetski EY, Loennechen T, et al. (1997). Molecular diagnosis of thiopurine Smethyltransferase deficiency: genetic basis for azathioprine and mercaptopurine intolerance. Ann Intern Med 126:608–614. Ye P, Fang H, Zhou X, He YL and Liu YX (2004). Effect of peroxisome proliferator-activated receptor activators on tumor necrosis factor-alpha expression in neonatal rat cardiac myocytes. Chin Med Sci J 19:243–247. Yee SB, Ganey PE and Roth RA (2003). The role of Kupffer cells and TNF-alpha in monocrotaline and bacterial lipopolysaccharide-induced liver injury. Toxicol Sci 71:124–132. Young IS, Trouton TG, Torney JJ, McMaster D, Callender ME and Trimble ER (1994). Antioxidant status and lipid peroxidation in hereditary haemochromatosis. Free Radic Biol Med 16:393–397. Yuan GJ, Zhou XR, Gong ZJ, Zhang P, Sun XM and Zheng SH (2006). Expression and activity of inducible nitric oxide synthase and endothelial nitric oxide synthase correlate with ethanol-induced liver injury. World J Gastroenterol 12:2375–2381. Yue J, Peng RX, Yang J, Kong R and Liu J (2004). CYP2E1 mediated isoniazid-induced hepatotoxicity in rats. Acta Pharmacol Sin 25:699–704. Zaher H, Buters JT, Ward JM, et al. (1998a). Protection against acetaminophen toxicity in CYP1A2 and CYP2E1 double-null mice. Toxicol Appl Pharmacol 152:193–199. Zaher H, Fernandez-Salguero PM, Letterio J, et al. (1998b). The involvement of aryl hydrocarbon receptor in the activation of transforming growth factor-beta and apoptosis. Mol Pharmacol 54:313– 321. Zetser A, Gredinger E and Bengal E (1999). p38 mitogen-activated protein kinase pathway promotes skeletal muscle differentiation. Participation of the Mef2c transcription factor. J Biol Chem 274:5193–5200. Zhang J, Salojin KV, Gao JX, Cameron MJ, Bergerot I and Delovitch TL (1999). p38 mitogenactivated protein kinase mediates signal integration of TCR/CD28 costimulation in primary murine T cells. J Immunol 162:3819–3829. Zhang S and Kaplan MH (2000). The p38 mitogen-activated protein kinase is required for IL-12induced IFN-gamma expression. J Immunol 165:1374–1380. Zhang J, Huang W, Chua SS, Wei P and Moore DD (2002). Modulation of acetaminophen-induced hepatotoxicity by the xenobiotic receptor CAR. Science 298:422–424. Zhu JH and Lei XG (2006). Double null of selenium-glutathione peroxidase-1 and copper, zincsuperoxide dismutase enhances resistance of mouse primary hepatocytes to acetaminophen toxicity. Exp Biol Med (Maywood ) 231:545–552. Zhu W, Song L, Zhang H, Matoney L, LeCluyse E and Yan B (2000). Dexamethasone differentially regulates expression of carboxylesterase genes in humans and rats. Drug Metab Dispos 28:186– 191. Zhou Z, Sun X, Lambert JC, Saari JT and Kang YJ (2002). Metallothionein-independent zinc protection from alcoholic liver injury. Am J Pathol 160:2267–2274. Ziol M, Tepper M, Lohez M, et al. (2001). Clinical and biological relevance of hepatocyte apoptosis in alcoholic hepatitis. J Hepatol 34:254–260. Zollner G, Fickert P, Zenz R, et al. (2001). Hepatobiliary transporter expression in percutaneous liver biopsies of patients with cholestatic liver diseases. Hepatology 33:633–646. Zoratti M and Szabo I (1995). The mitochondrial permeability transition. Biochim Biophys Acta 1241:139–176.

10 A Role of Cytochrome P450 in Quinone-Induced Hepatotoxicity Yasuhiro Ishihara and Norio Shimamoto

10.1

Introduction

Quinones are widely distributed in nature, and are found in almost all respiring animal and plant cells. They form an important group of substrates for flavoenzymes and have vital physiological functions in a number of processes, such as photosynthesis and the respiratory chain in mitochondria. Some quinones are used as anticancer, antimalarial or antibacterial drugs (Asche, 2005; Lown, 1983; Vennerstrom and Eaton, 1988). However, their therapeutic use is limited in some cases because the use of most quinones is accompanied by adverse effects derived from their cytotoxicity. Therefore, understanding the toxic mechanisms of quinones would help us to use quinone drugs more appropriately and to develop novel quinone-based drugs. The toxic mechanisms of quinones are still under debate. The majority of xenobiotics, including food additives, cosmetic products, agrochemicals and drugs, are metabolized by xenobiotic-metabolizing enzymes, which are classified into two groups – Phase 1 and Phase 2. Of the Phase 1 group, cytochrome P450 plays a crucial role for the monooxygenation of xenobiotics as well as some endogenous substrates, which is localized largely in the endoplasmic reticulum in hepatocytes. Phase 2 enzymes conjugate substrates that are usually Phase 1 metabolites. The structure of quinone is not modified by Phase 1 enzymes and thus quinone compounds are thought to be metabolized mainly via a conjugation process. In the liver, quinone compounds are conjugated with glucuronic acid, sulfuric acid and reduced glutathione to form soluble conjugates, followed by release into the central vein or bile ducts. In that sense, quinone hepatotoxicity does not seem to be related


288

Hepatotoxicity

directly to xenobiotic enzymes. Very recently, however, we have shown that quinone hepatotoxicity is enhanced under conditions of cytochrome P450 inhibition, indicating clearly the involvement of cytochrome P450 in quinone hepatotoxicity (Ishihara et al., 2006). In this chapter, we revisit the generally accepted mechanisms of quinone hepatotoxicity and propose the importance of cytochrome P450 systems in quinone-induced hepatotoxicity on the basis of our recent work.

10.2

Mechanism of Quinone-Induced Hepatotoxicity

The toxicity of quinones is induced by two principal mechanisms. They are: (i) arylationl/alkylation, reaction with nucleophiles among important biological constituents, and (ii) redox cycle, superoxide anion generation via quinone–semiquinone interconversion, resulting in oxidative stress, as shown in Figure 10.1. 10.2.1

Arylationl/Alkylation

There are many thiols that play important roles in regulating cellular functions. These include low-molecular-mass thiol compounds and protein thiols. Reduced glutathione (GSH) as a representative low-molecular-mass thiol compound is present at millimolar concentrations in many types of cells, especially hepatocytes, and plays a fundamental role in managing cellular redox homeostasis by protecting cells from exogenous or endogenous oxidative stress (Wu et al., 2004). Protein thiols, which usually indicate the presence of cysteine in

(ii) Redox Cycle

(i) Arylation/Alkylation R–SH

– 2

O

Q

NAD(P)+

NADPH–cytochrome P450 reductase NADH–cytochrome b5 reductase

Q

Q–SR

O2

NAD(P)H

Cytochrome P450

GSH depletion Enzymatic inactivation Q: quinone Q : semiquinone radical

Figure 10.1 Mechanisms of quinone hepatotoxicity. Quinone cytotoxicity is elicited by two mechanisms: (i) the arylation/alkylation of nucleophiles among important biological constituents, for example, quinones react covalently with thiols, such as GSH or the cysteine residues of proteins, to form arylation/alkylation products that eventually cause cellular damage, and (ii) oxidative stress via redox cycling. Redox-cycling quinones accept one electron from intracellular reductases to form semiquinone radicals, which are auto-oxidized by donating one electron to molecular oxygen. In these processes, massive superoxide anion radicals are produced, leading to oxidative stress

Cytochrome P450 in Quinone-Induced Hepatotoxicity

289

proteins, are often active sites of proteins such as proteases (Thornberry et al., 1992), phosphatases (Barrett et al., 1999) and antioxidative enzymes (Arner and Holmgren, 2000; Fujii and Ikeda, 2002), and the functional loss of protein thiols induces dysfunction of these proteins. Electrophilic moieties of quinones (e.g. 2, 3, 5 and 6 positions of 1,4-benzoquinone (BQ)) react covalently with these functional thiols to form arylation/alkylation products. This reaction causes protein dysfunction, leading to serious cellular damage, and often cell death (necrosis). Some groups have reported that when primary hepatocytes were treated with arylating quinone, BQ, intracellular levels of GSH and protein thiol decreased in a time-dependent manner, followed by cell death (Rossi et al., 1986; Tapper et al., 2000). However, intracellular levels of glutathione disulfide (GSSG) did not change, suggesting that the decreased levels of GSH might not be due to oxidative stress but is the result of chemical reaction of quinones with GSH (Rossi et al., 1986). The relationship between quinone toxicity and oxidative stress is described in Sections 10.2.3 and 10.2.4. 10.2.2

Reactive Oxygen Species and their Scavenging Systems

Reactive oxygen species (ROS) are the general term for highly reactive molecules derived from oxygen, such as the superoxide anion, hydrogen peroxide and hydroxyl radical. ROS are generated constitutively from cellular organelles, especially liver endoplasmic reticulum and mitochondria, and a variety of xenobiotics such as quinones induce ROS production. Cells have defense systems to cope with routinely generated ROS. The superoxide anion is dismutated to hydrogen peroxide and molecular oxygen by superoxide dismutase (SOD). Mitochondrial Mn–SOD efficiently eliminates the superoxide anion that arises from molecular oxygen in the respiratory chain. The intracellular Cu/Zn–SOD is present in the cytoplasm, nucleus and peroxisomes of all mammalian cells, where it scavenges superoxide anion (Crapo et al., 1992). Hydrogen peroxide produced by the dismutation of the superoxide anion radical is degraded to molecular oxygen and water by catalase, glutathione peroxidase (GPx) and peroxiredoxin (Prx). Catalases are present ubiquitously in aerobic organisms, and the highest level of enzyme activity is in the liver and erythrocytes. Within cells, catalases are located mostly in peroxisomes because of the presence of many hydrogen peroxide-producing enzymes. There are four types of GPx, i.e. classical GPx (GPx1), gastrointestinal GPx (GPx2), plasma GPx (GPx3) and phospholipid hydroperoxide GPx (GPx4) (Imai and Nakagawa, 2003). The former three GPxs reduce hydrogen peroxide and organic alkyl hydroperoxides to water and corresponding alcohols, at the expense of oxidation of GSH to GSSG. The last type of peroxidase, phospholipid hydroperoxide GPx, is different from the other three GPxs with regard to its substrate specificity and localization. This enzyme has the unique ability to reduce membrane lipid hydroperoxides such as phospholipid and cholesterol hydroperoxides directly, and is located mostly in the testis (Thomas et al. 1990). Prx was discovered more recently than the above-mentioned catalase and GPx, and is a hydrogen peroxide-scavenging enzyme. Prx enzymes contain conserved cysteine residues among Prx enzymes that undergo peroxide-dependent oxidation and thiol (thioredoxin)-dependent reduction during a catalytic cycle. Mammalian cells express six isoforms of Prx (Prx I to VI), which are classified into three subgroups (2-Cys, atypical 2-Cys and 1-Cys) on the basis of the number and position of cysteine residues as active sites. The relatively high abundance of Prx enzymes in mammalian cells appears to play a role for removing the low levels of peroxides produced during normal cellular metabolism

290

Hepatotoxicity

(Rhee et al., 2005). Besides antioxidative enzymes as described above, cells have small molecules with antioxidative capacity, such as ascorbic acid, α-tocopherol and GSH (Mates, 2000). These molecules are able to scavenge ROS efficiently and specifically at relatively low concentrations. When ROS is not eliminated sufficiently by these anti-oxidative enzymes or small molecules, cells are damaged by oxidative ‘insults’, leading to cell death. 10.2.3

Redox Cycle

Quinones undergo one-electron reduction by various intracellular reductases to yield semiquinone radicals. The semiquinone radicals are then re-oxidized easily in co-operation with molecular oxygen to form the superoxide anion. This oxidation–reduction process between quinone and the semiquinone radical is called the quinone redox cycle. The superoxide anion generated via the redox cycle is dismutated to hydrogen peroxide and oxygen by superoxide dismutase. Then, hydrogen peroxide is converted into hydroxyl radicals in the presence of metal cations such as Fe2+ or Cu+ (Me(n−1)+ + H2 O2 → Men+ +OH− +HO· ). The hydroxyl radical is highly unstable and the most reactive oxygen radical among ROS, causing lipid peroxidation, protein oxidation and DNA damage, followed by cellular toxicity (Droge, 2002; Galli et al., 2005). There are some reports of quinone toxicity via redox cycling. Treatment with menadione, which has a redox cycling activity, increased oxygen consumption and superoxide anion production inside cells and decreased intracellular levels of GSH, leading to hepatotoxicity (Thor et al., 1982). When rat primary hepatocytes were treated with the redox cycling quinone, 2,3-dimethoxy-1,4-naphthoquinone (DMNQ), intracellular levels of GSH were decreased and levels of GSSG were increased, indicating oxidative stress induction, followed by cell death (Gant et al., 1988). In the liver, one-electron reduction between quinone and semiquinone radical is catalyzed mainly by flavoenzymes such as NADPH–cytochrome P450 reductase and NADH– cytochrome b5 reductase (Berlin and Haseltine, 1981; Kappus and Sies, 1981; Powis and Appel, 1980). These reductases use NAD(P)H as an electron donor to reduce quinones. The rate of one-electron reduction of quinones by these reductases is not considered to be associated with their structural profiles or lipid solubility but with the one-electron redox potential of these reductases. Two-electron reduction of quinones, which is mediated by NAD(P)H:quinone oxidoreductase 1 (DT-diaphorase), leads to hydroquinone formation (Danson et al., 2004; Ross and Siegel, 2004). Reduction of quinones by NAD(P)H:quinone oxidoreductase 1 may constitute a detoxification process since rapid reduction of the quinone to the hydroquinone would circumvent one-electron processes. Furthermore, hydroquinones may be conjugated with glucuronic acid or sulfate to form unreactive and water-soluble substances that are readily eliminated from the body. In fact, inhibition of NAD(P)H:quinone oxidoreductase 1 by dicoumarol enhanced hepatotoxicity induced by treatment with quinones (Keyes et al., 1985; Tsuda, 1990). 10.2.4

The Relative Contribution of Quinone-Induced Arylation/Alkylation or Redox Cycle to Hepatotoxicity

Quinone toxicity depends on two mechanisms, i.e. their arylating/alkylating or redoxcycling abilities. The contribution of arylating/alkylating and/or redox-cycling abilities


291

of quinones to cytotoxicity was reported to be a function of the chemical structure or the redox potential of the quinone (Watanabe et al., 2004), and the contribution could be examined experimentally. The idea is as follows: arylation/alkylation products of thiols were the results of the reaction with quinones, and they could not be recovered to the original protein thiols by treatment with a reductant. ROS that resulted from redox cycling, especially hydrogen peroxide, react with the thiolate anion, which is in equilibrium with protein thiols to initially produce the sulfenic acid and then the disulfide. These oxidized products are rereduced to the original thiols in the presence of a reductant such as dithiothreitol. Therefore, we can discriminate decreases in protein thiols due to direct arylationl/alkylation from that due to oxidative insults via the redox cycle by measuring the total loss of protein thiols and recoverable protein thiols upon addition of a reductant. Thus, loss of total protein thiols minus recoverable protein thiols equals the loss of protein thiols by arylating/alkylating (Bellomo et al., 1990; d’Arcy Doherty et al., 1987). However, it is difficult to evaluate the extent to which the reaction with protein thiols contributes to the cytotoxicity of quinones. Efforts were made to develop model quinones that have either arylation/alkylation ability or redox cycling ability to induce cytotoxicity. One of the representative compounds of the former is considered to be BQ or 1,4-naphthoquinone, and that of the latter is thought to be DMNQ. Henry and Wallace showed that cytotoxicity of arylating quinones BQ and 1,4-naphthoquinone was stronger than that of the redox cycling quinone DMNQ in primary rat hepatocytes (Henry and Wallace, 1996). Similar results were obtained by other groups (Miller et al.,1986; Schmieder et al., 2003; Seung et al., 1998; Toxopeus et al., 1993), indicating that hepatotoxicity of quinone compounds depends mainly on their arylating activities, and that the redox cycle is less involved in hepatotoxicity. Our results suggest that the primary toxicity of BQ is stronger than that of DMNQ in rat primary hepatocytes in a short-term exposure (Ishihara et al., 2006). However, a long-term exposure of hepatocytes to DMNQ induced oxidative stress sufficient to damage cells, leading to necrosis. In this regard, the redox cycle is considered to contribute a great deal to hepatotoxicity (Ishihara et al., 2006). More importantly, under cytochrome P450 inhibition, the hepatotoxicity of DMNQ, a redox-cycling quinone, became stronger than that of BQ, an arylating quinone. This is described in the next section.

10.3

A Role of Cytochrome P450 in Quinone-Induced Hepatotoxicity

The membrane-bound microsomal monooxygenase system is localized in the endoplasmic reticulum of most animal tissues. Although the highest content of microsomal monooxygenase components is found in hepatocytes, the system is present also in the lung, kidney, brain, lymphocytes, vascular smooth muscle, olfactory and intestinal epithelium (McKinnon and McManus 1996; Kapitulnik and Strobel 1999). Cytochrome P450, which is a heme-thiolate protein, is a major enzyme in the microsomal monooxygenase system. The predominant function of cytochrome P450 appears to be the oxygenation of exogenous compounds as well as some endogenous substrates (Gonzalez, 1988). A variety of substrates can be oxidized because of the existence of multiple cytochrome P450 isozymes with broad substrate selectivity. Most drugs are metabolized by cytochrome P450, resulting in production of water-soluble metabolites, which are then excreted from the body. Mono-oxygenation reactions typically require the input of two electrons that are transferred to cytochrome P450

292

Hepatotoxicity

by the flavoprotein NADPH–cytochrome P450 reductase and/or NADH–cytochrome b5 reductase. Because cytochrome P450 is closely related to the side-effects of drugs, interaction of cytochrome P450 with various drugs is an important subject in toxicology (Wienkers and Heath, 2005; Wilkinson, 2005). However, quinone compounds are metabolized via hydroquinone production by NAD(P)H:quinone oxidoreductase 1, followed by forming conjugates. Therefore, cytochrome P450 is not directly involved in quinone metabolism. In fact, the relationship between cytochrome P450 and metabolism of the quinone-based anticancer agent doxorubicin (adriamycin) was examined in NADPH–cytochrome P450 reductase-deleted mice in which cytochrome P450 cannot function. When doxorubicin was administered to both wild-type and NADPH–cytochrome P450 reductase-deleted mice, no change in the concentration of doxorubicin in the blood was observed between the two types of mice. Therefore, this result supports the idea that cytochrome P450 is not involved in quinone metabolism (Henderson et al., 2006). However, by considering two points – (1) cytochrome P450 accepts an electron from intracellular reductases, and (2) redox-cycling quinones also receive an electron from them – cytochrome P450 might be involved in the hepatotoxicity of redox-cycling quinone. Therefore, we examined the role of cytochrome P450 in the toxicity of the redox-cycling quinone DMNQ using rat primary hepatocytes. When rat primary hepatocytes were treated with 60 μM DMNQ and incubated for 5 h, cell viability was decreased to 65 %. Treatment with DMNQ in combination with cytochrome P450 inhibitors SKF or ketoconazole decreased the viability to 30 or 48 %, respectively, indicating that DMNQ-induced hepatotoxicity was enhanced under cytochrome P450 inhibition. On the other hand, the arylating quinone BQ-induced hepatotoxicity was not affected by pretreatment with a cytochrome P450 inhibitor (Ishihara et al., 2006). Therefore, enhancement of hepatotoxicity by cytochrome P450 inhibition occurs only when a redox-cycling quinone, not an arylating quinone, is treated. More importantly, both superoxide anion production and NADPH oxidation induced by treatment with DMNQ were potentiated by pretreatment with a P450 inhibitor (Table 10.1), indicating strongly that the DMNQ-induced redox cycle is enhanced under P450 inhibition. On the basis of these findings, we offer a putative mechanism for the enhancement of DMNQ-induced hepatotoxicity under cytochrome P450 inhibition, as shown in Figure 10.2. In the liver, an electron from NADPH–cytochrome P450 reductase is transferred to cytochrome P450, which metabolizes many endogenous substrates or xenobiotics (Gonzalez, 1988), and NADPH–cytochrome P450 reductase catalyzes the one-electron reduction of quinones in redox cycles (Powis et al., 1980; Berlin et al., 1981). Thus, when redox-cycling quinones exist in hepatocytes, the electron flows from NADPH–cytochrome P450 reductase are considered to be additionally transferred to redox-cycling quinones; this leads to the production of semiquinone radicals, which are re-oxidized by giving electrons to molecular oxygen to yield the superoxide anion. This explains how DMNQ increased the NADPH oxidation and production of the superoxide anion. More importantly, the inhibition of cytochrome P450 by SKF or ketoconazole enhanced NADPH oxidation and superoxide production, and NADPH-cytochrome P450 reductase inhibition suppressed DMNQinduced and DMNQ + SKF-induced or DMNQ + ketoconazole-induced NADPH oxidation (Table 10.1). Our scenario for these findings is as follows. The inhibition of cytochrome P450 suppresses the electron transfer from the reductase to cytochrome P450; thus, the reductase donates electrons only to quinones to produce semiquinone radicals, resulting in


293

Table 10.1 Enhancement of DMNQ-induced redox cycling and cell death under cytochrome P450 inhibition (details of this procedure are described else where (Ishihara et al., 2006)). Conditionsa,e Untreated DMNQ DMNQ + SKF DMNQ + ketoconazole

Cell viabilityb,e Superoxide production NADPH oxidation (%) (nmol/min/106 cells)c,e (nmol/min/mg protein)d,e 91.2 ± 1.2 64.6 ± 2.7 30.3 ± 4.2 f 47.6 ± 4.9g

14.4 ± 0.4 16.4 ± 0.5 18.3 ± 0.5g 18.1 ± 0.4g

6.8 ± 1.2 46.6 ± 3.2 77.4 ± 5.8 f 66.9 ± 4.2 f

a

Primary rat hepatocytes were pretreated with SKF (30 μM) or ketoconazole (30 μM) for 20 min and then incubated in the presence or in the absence of DMNQ (60 μ M). Cell viability was estimated as the percentage of LDH activity in the medium after incubation for 5 h. c The amount of superoxide anion radical produced by hepatocytes was quantified by a method using acetylated cytochrome c after 10 min incubation. d NADPH oxidation in rat liver microsomes was measured by changes in absorbance at 340 nm when DMNQ or DMNQ + cytochrome P450 inhibitors was added. e Values are given as the mean ± SE of five separate experiments. Data were analyzed by using the Student’s t-test, with Holm’s corrections for multiple comparisons. P values less than 0.05 were considered statistically significant. f P < 0.01 versus DMNQ-treated hepatocytes. g P < 0.05 versus DMNQ-treated hepatocytes. b

further increases in NADPH oxidation and production of superoxide anion (Figure 10.2). Namely, the quinone redox cycle is enhanced by cytochrome P450 inhibition. This scenario is supported by data showing that an NADPH–cytochrome P450 reductase inhibitor, diphenyleneiodonium chloride, suppressed NADPH oxidation induced by treatment with both DMNQ and DMNQ + SKF or DMNQ + ketoconazole (Ishihara et al., 2006). Many anticancer, antimalarial or antibacterial agents have quinone structures. These drugs have high therapeutic efficacy, but have the ability to induce cytotoxicity. To date, general considerations of quinone-induced cytotoxicity are as follows: (1) cytochrome P450 contributes less to quinone metabolism (detoxification), and (2) the toxicity of arylating quinones is greater than that of redox-cycling quinones. On the basis of these considerations, it is natural that quinone compounds having less arylating activity are required for developing less-toxic drugs. However, our results indicate that the toxicity of redox-cycling quinones is enhanced under cytochrome P450 inhibition. In addition, the hepatotoxicity of DMNQ under cytochrome P450 inhibition was much higher than that of BQ at the same concentration. In this regard, careful consideration of the redox-cycling ability in terms of hepatotoxicity must be taken when quinone-based drugs are used or developed.

10.4

Concluding Remarks

The hepatotoxicity of quinone compounds could be induced by two mechanisms, i.e. arylation/alkylation and the redox cycle. To understand the mechanisms of hepatotoxicity of quinones is to avoid or alleviate their hepatotoxicity. The hepatotoxicities of arylating quinones are generally recognized to be higher than that of redox-cycling quinones. However, the hepatotoxicity of redox-cycling quinones is potentially higher than that of arylating quinones under cytochrome P450 inhibition. Our proposal is that we should

294

Hepatotoxicity

(i) Normal state Redox-cycling quinone NAD(P)H


Cytochrome P450

(ii) Under cytochrome P450 inhibition Redox-cycling quinone NAD(P)H


Cytochrome P450 SKF Ketoconzole Electron flow

Figure 10.2 Enhancement of quinone hepatotoxicity under cytochrome P450 inhibition: a putative mechanism. In the liver, cytochrome P450 accepts an electron from NADPH– cytochrome P450 reductase or NADH–cytochrome b 5 reductase. When redox-cycling quinones are used under conditions where cytochrome P450 is operating functionally, NADPH–cytochrome P450 reductase or NADH-cytochrome b 5 reductase supplies electrons to both cytochrome P450 and redox-cycling quinones. In other words, quinones compete with cytochrome P450 for electron flow from the reductases. However, under cytochrome P450 inhibition, reductases are not able to transfer an electron to cytochrome P450 and, thus, they donate an electron only to redox-cycling quinones, forming semiquinone radicals. As a result, the quinone redox cycle can be potentiated by cytochrome P450 inhibition

not underestimate the hepatotoxicity of redox cycling quinones if we use them together with cytochrome P450 inhibitors, regardless of whether they are drugs. Therefore, careful consideration for quinone-based drugs will help to avoid hepatotoxicity. In addition, the strong toxicity of quinone-based drugs might be alleviated by therapy in combination with other drugs if the toxic mechanism is known. One example of this is the cardiotoxicity of doxorubicin, which is an anticancer agent with a quinone structure, being attenuated by pretreatment with antioxidants (Lenzhofer et al., 1983). We stress the mechanism of the redox-cycling quinone under the condition of P450 inhibition, which is thought to be readily attainable, and thus arouses concern in the case of managing quinone compounds.

10.5

Acknowledgements

We thank Dr M. Iwasaki for critical reading of this manuscript.


10.6

295

Abbreviations

BQ – 1,4-benzoquinone DMNQ – 2,3-dimethoxy-1,4-naphthoquinone GPx – glutathione peroxidase GSH – reduced glutathione GSSG – glutathione disulfide Prx – peroxiredoxin ROS – reactive oxygen species SOD – superoxide dismutase

References Arner ES and Holmgren A (2000). Physiological functions of thioredoxin and thioredoxin reductase. Eur J Biochem 267: 6102–6109. Asche C (2005). Antitumour quinones. Mini Rev Med Chem 5: 449–467. Barrett WC, DeGnore JP, Konig S, Fales HM, Keng YF, Zhang ZY, Yim MB and Chock PB (1999). Regulation of PTP1B via glutathionylation of the active site cysteine 215. Biochemistry 38: 6699– 6705. Bellomo G, Thor H and Orrenius S (1990). Modulation of cellular glutathione and protein thiol status during quinone metabolism. Methods Enzymol 186: 627–635. Berlin V and Haseltine WA (1981). Reduction of adriamycin to a semiquinone-free radical by NADPH cytochrome P-450 reductase produces DNA cleavage in a reaction mediated by molecular oxygen. J Biol Chem 256: 4747–4756. Crapo JD, Oury T, Rabouille C, Slot JW and Chang LY (1992). Copper, zinc superoxide dismutase is primarily a cytosolic protein in human cells. Proc Natl Acad Sci USA 89: 10405–10409. Danson S, Ward TH, Butler J and Ranson M (2004). DT-diaphorase: a target for new anticancer drugs. Cancer Treat Rev 230: 437–449. d’Arcy Doherty M, Rodgers A and Cohen GM (1987). Mechanisms of toxicity of 2- and 5-hydroxy-1,4-naphthoquinone; absence of a role for redox cycling in the toxicity of 2-hydroxy1,4-naphthoquinone to isolated hepatocytes. J Appl Toxicol 7: 123–129. Droge W (2002). Free radicals in the physiological control of cell function. Physiol Rev 82: 47–95. Fujii J and Ikeda Y (2002). Advances in our understanding of peroxiredoxin, a multifunctional, mammalian redox protein. Redox Rep 7: 123–130. Galli F, Piroddi M, Annetti C, Aisa C, Floridi E and Floridi A (2005). Oxidative stress and reactive oxygen species. Contrib Nephrol 149: 240–260. Gant TW, Rao DN, Mason RP and Cohen GM (1988). Redox cycling and sulphydryl arylation; their relative importance in the mechanism of quinone cytotoxicity to isolated hepatocytes. Chem Biol Interact 65: 157–173. Gonzalez FJ (1988). The molecular biology of cytochrome P450s. Pharmacol Rev 40: 243–288. Henderson CJ, Pass GJ and Wolf CR (2006). The hepatic cytochrome P450 reductase null mouse as a tool to identify a successful candidate entity. Toxicol Lett 162: 111–117. Henry TR and Wallace KB (1996). Differential mechanisms of cell killing by redox cycling and arylating quinones. Arch Toxicol 70: 482–489. Ishihara Y, Shiba D and Shimamoto N (2006). Enhancement of DMNQ-induced hepatocyte toxicity by cytochrome P450 inhibition. Toxicol Appl Pharmacol 214: 109–117. Imai H and Nakagawa Y (2003). Biological significance of phospholipid hydroperoxide glutathione peroxidase (PHGPx, GPx4) in mammalian cells. Free Radic Biol Med 34: 145–169.

296

Hepatotoxicity

Kapitulnik J and Strobel HW (1999). Extrahepatic drug metabolizing enzymes. J Biochem Mol Toxicol 13: 227–230. Kappus H and Sies H (1981). Toxic drug effects associated with oxygen metabolism: redox cycling and lipid peroxidation. Experientia 37: 1233–1241. Keyes SR, Rockwell S and Sartorelli AC (1985). Enhancement of mitomycin C cytotoxicity to hypoxic tumor cells by dicoumarol in vivo and in vitro. Cancer Res 45: 213–216. Lenzhofer R, Magometschnigg D, Dudczak R, Cerni C, Bolebruch C and Moser K (1983). Indication of reduced doxorubicin-induced cardiac toxicity by additional treatment with antioxidative substances. Experientia 39: 62–64. Lown JW (1983). The mechanism of action of quinone antibiotics. Mol Cell Biochem 55: 17–40. McKinnon RA and McManus ME (1996). Localization of cytochromes P450 in human tissues: implications for chemical toxicity. Pathology 28: 148–155. Mates JM (2000). Effects of antioxidant enzymes in the molecular control of reactive oxygen species toxicology. Toxicology 153: 83–104. Miller MG, Rodgers A and Cohen GM (1986). Mechanisms of toxicity of naphthoquinones to isolated hepatocytes. Biochem Pharmacol 35: 1177–1184. Powis G and Appel PL (1980). Relationship of the single-electron reduction potential of quinones to their reduction by flavoproteins. Biochem Pharmacol 29: 2567–2572. Rhee SG, Chae HZ and Kim K (2005). Peroxiredoxins: a historical overview and speculative preview of novel mechanisms and emerging concepts in cell signaling. Free Radic Biol Med 38: 1543–1552. Ross D and Siegel D (2004). NAD(P)H:quinone oxidoreductase 1 (NQO1, DT-diaphorase), functions and pharmacogenetics. Methods Enzymol 382: 115–144. Rossi L, Moore GA, Orrenius S and O’Brien PJ (1986). Quinone toxicity in hepatocytes without oxidative stress. Arch Biochem Biophys 251: 25–35. Schmieder PK, Tapper MA, Kolanczyk RC, Hammermeister DE, Sheedy BR and Denny JS (2003). Discriminating redox cycling and arylation pathways of reactive chemical toxicity in trout hepatocytes. Toxicol Sci 72: 66–76. Seung SA, Lee JY, Lee MY, Park JS and Chung JH (1998). The relative importance of oxidative stress versus arylation in the mechanism of quinone-induced cytotoxicity to platelets. Chem Biol Interact 113: 133–144. Tapper MA, Sheedy BR, Hammermeister DE and Schmieder PK (2000). Depletion of cellular protein thiols as an indicator of arylation in isolated trout hepatocytes exposed to 1,4-benzoquinone. Toxicol Sci 55: 327–334. Toxopeus C, van Holsteijn I, Thuring JW, Blaauboer BJ and Noordhoek J (1993). Cytotoxicity of menadione and related quinones in freshly isolated rat hepatocytes: effects on thiol homeostasis and energy charge. Arch Toxicol 67: 674–679. Thomas JP, Maiorino M, Ursini F and Girotti AW (1990). Protective action of phospholipid hydroperoxide glutathione peroxidase against membrane-damaging lipid peroxidation. In situ reduction of phospholipid and cholesterol hydroperoxides. J Biol Chem 265: 454–461. Thornberry NA, Bull HG, Calaycay JR, Chapman KT, Howard AD, Kostura MJ, Miller DK, Molineaux SM, Weidner JR, Aunins J, Elliston KO, Ayala JM, Casano FJ, Chin J Ding, G.J.F., Egger LA, Gaffney EP, Limjuco G, Palyha OC, Raju SM, Rolando AM, Salley JP, Yamin T, Lee TD, Shively JE, MacCross M, Mumford RA, Schmidt JA and Tocci MJ (1992). A novel heterodimeric cysteine protease is required for interleukin-1 beta processing in monocytes. Nature 356: 768–774. Thor H, Smith MT, Hartzell P, Bellomo G, Jewell SA and Orrenius S (1982). The metabolism of menadione (2-methyl-1,4-naphthoquinone) by isolated hepatocytes. A study of the implications of oxidative stress in intact cells. J Biol Chem 257: 12419–12425. Tsuda H (1990). Role of DT diaphorase the cytotoxicity of menadione and 4-nitroquinoline-1-oxide in cultured mammalian fibroblastic cells. Cancer Lett 55: 195–199.


297

Vennerstrom JL and Eaton JW (1988). Oxidants, oxidant drugs and malaria. J Med Chem 31: 1269– 1277. Watanabe N, Dickinson DA, Liu RM and Forman HJ (2004). Quinones and glutathione metabolism. Methods Enzymol 378: 319–340. Wienkers LC and Heath TG (2005). Predicting in vivo drug interactions from in vitro drug discovery data. Nat Rev Drug Discov 4: 825–833. Wilkinson GR (2005). Drug metabolism and variability among patients in drug response. N Engl J Med 352: 2211–2221. Wu G, Fang YZ, Yang S, Lupton JR and Turner ND (2004). Glutathione metabolism and its implications for health. J Nutr 134: 489–492.

11 A Mechanistic View of Troglitazone Hepatotoxicity Rawiwan Maniratanachote and Tsuyoshi Yokoi

11.1

Introduction

Thiazolidinediones (Figure 11.1) are a class of oral antidiabetic agents and are the synthetic ligands for the peroxisome proliferator-activated receptor γ (PPAR γ ) (Lehmann et al., r , (±)-5-[[4-[(3,4-dihydro-6-hydroxy-2,5,7,8-tretramethyl1995). Troglitazone (Rezulin 2H-1-benzopyran-2-yl)methoxy]phenyl]methyl]-2,4-thiazolidinedione) was the first thiazolidinedione antidiabetic agent approved for clinical use by the US Food and Drug Administration in 1997. Troglitazone lowers the blood glucose levels through increased glucose uptake by skeletal muscle, decreased hepatic glucose production and increased insulin sensitivity of the target tissue in animal models of metabolic impairment (Ciaraldi et al., 1990; Fujiwara et al., 1988, 1995). These pharmacological effects are exerted through PPAR γ -dependent transcription of genes involved in glucose and lipid metabolism and energy homeostasis (Lehmann et al., 1995; Saltiel and Olefsky, 1996; Spiegelman, 1998). Based on the pharmacological advantages and the apparent absence of severe toxic effects, troglitazone was thought likely to become a promising treatment for type II diabetes mellitus in patients with insulin resistance. However, in the combined North American clinical trials, elevations of serum alanine aminotransferase more than three times the upper limit of normal were observed in 48 out of 2510 patients (1.9 %) treated with troglitazone. Liver biopsies from two patients confirmed the hepatocellular nature of the injury as an idiosyncratic drug reaction (Watkins and Whitcomb, 1998). Meanwhile, troglitazone had been concomitantly reported to be associated with idiosyncratic hepatotoxicity, with some patients showing severe or fatal


300

Hepatotoxicity CH3 O

H3C

CH3 O

H3C

O S NH

O S NH

HO3SO

ST1A3

CH3

HO CH3

CH3

CH3

Troglitazone sulfate (M1)

UGT

Troglitazone

CH3 O

H3C

CYP3A4 CYP2C8

N O N

O S NH

S NH CH3

HO HO

Troglitazone glucuronide (M2)

O

Rosiglitazone N

O

O O OH

CH3

CH3

COOH

CH3

O

H3C

O

O

O OH

NH H 3C

H3C

S

S

NH

O O

Pioglitazone

Figure 11.1

CH3

Troglitazone quinone (M3)

Structures of thiazolidinediones and pathways of troglitazone metabolism

liver damage (Gitlin et al., 1998; Neuschwander-Tetri et al., 1998; Shibuya et al., 1998). Consequently, it was withdrawn from the market in the USA and Japan in March 2000. The hepatotoxic effects of troglitazone were not predicted from conventional animal models (Watanabe et al., 1999) or in cynomolgus monkeys (300–1200 mg/kg/day for 52 weeks), a primate model having similar metabolic profiles to humans (Rothwell et al., 2002). Two other thiazolidinediones, which are now on the market, rosiglitazone and pioglitazone, have been introduced in 1999 and are unlikely to share the hepatotoxic effects of troglitazone (Freid et al., 2000; Isley and Oki, 2000; Lebovitz et al., 2002). It should also be noted that the clinical dosage regimen for improvement of fasting glucose is distinguishable among these thiazolidinediones (Table 11.1). The recommended dose for troglitazone was 200 to 600 mg/day, for rosiglitazone 4 to 8 mg/day and for pioglitazone 15 to 45 mg/day (Hanefeld, 2001; Loi et al., 1999; PDR, 1999, 2005a,b). The dosage requirement for their efficacy might have reflected their hepatotoxic potential. This review summarizes the molecular mechanism of troglitazone hepatotoxicity from studies both in vivo and in vitro. Even though, there is no direct evidence to indicate the precise mechanism of the toxicity so far. Many points of view, however, have been proposed to contribute to the toxic effects of troglitazone.

11.2

Potential of Troglitazone Metabolites for Hepatotoxicity

In humans, troglitazone is predominantly metabolized by three pathways: sulfation, glucuronidation and oxidation to form a sulfate conjugate (M1), a glucuronide conjugate (M2) and a quinone metabolite (M3), respectively (Figure 11.1). M1 and M3 are the major metabolites in plasma, while M2 is a minor metabolite (Izumi et al., 1997a,b; Kawai et al.,

Mechanistic View of Troglitazone Hepatotoxicity

301

Table 11.1 Pharmacokinetic parameters of thiazolidinediones Parameter

Troglitazone (1997–2000)a

Rosiglitazone (1999–present)a

Pioglitazone (1999–present)a

Oral dosage (mg/day) Plasma protein binding (%) Absolute bioavailability (%) Cmax (μg/ml) AUC (μg-h/ml) tmax (h) Plasma elimination half-life (h) Biliary excretion (%) Urinary excretion (%) Effects of food

200–600 >99 40–50 0.90–2.82 7.4–22.1 99 83 1.4 ± 0.2b 11.6 ± 2.2b 500

1.4

0

0

>500

>500

Liver injuries (cases per 100 000 prescriptionsa )

Bsep inhibition potential IC50 (μM)

80–320

500

500–1000

Typical dose range (mg per day)

Table 16.2 Cholestatic effects of several sulfonylurea compounds. Cases of in vivo liver injury (as number of liver injury cases per 100 000 prescriptions) and in vitro inhibition potential on Bsep using taurocholate transport interference in liver plasma membrane vesicles

(Continued)

a

O

Cl

N

O

N

H N

O

H N

O

Data on incidence of liver injury in diabetic patients (Huerta et al., 2002).

Glibenclamide

Glipizide

Table 16.2

S O O

H N

H N

O S O O N H N H

2.7 ± 0.4

115 ± 20

3.4

4.3

10

5–60

Cholestatic Potential through Inhibition of BSEP

429

cholestatic effects were seen in rats, serum bile acid levels were increased by both, glibenclamide and bosentan, with an additive effect if both compounds were co-administered. In vitro, bosentan inhibited the Bsep-mediated taurocholate transport (K I , ∼ 12 μM; Fattinger et al., 2001). The three major metabolites of bosentan showed a Bsep inhibition similar to bosentan, suggesting that they contribute to the cholestatic effect in vivo (Fattinger et al., 2001). The in vitro and in vivo cholestatic potential measured in rats supported the relevance of this mechanism for the liver injury observed in man. 16.2.4

Cholestasis in Troglitazone-Induced Hepatotoxicity

Troglitazone is an insulin sensitizer of the thiazolidinedione class, compounds used for the treatment of type 2 non-insulin-dependent diabetes mellitus (Chen, 1998). Elevations in liver enzyme levels observed during clinical trials and several cases of fulminant hepatic failure were reported, leading to the withdrawal of this compound from the market (Gitlin et al., 1998; Shibuya et al., 1998; Herrine and Choudhary, 1999). Different hypotheses towards the mechanism(s) underlying the troglitazone-associated hepatotoxicity have been published. Several reports suggested a cholestatic mechanism to be involved in this liver injury (Gitlin et al., 1998) and a strong reduction of the bile flow has been observed in isolated perfused rat liver (Preininger et al., 1999). Several cases of liver injury following concurrent troglitazone and glibenclamide treatment lead to the speculation that this latter drug, known to induce cholestasis in some patients (Krivoy et al., 1996), might enhance the probability for troglitazone-induced liver injury (Shibuya et al., 1998; Fukano et al., 2000). This cholestatic potential of troglitazone is further supported by in vitro and in vivo mechanistic studies (Funk et al., 2001b). Alternatively, reactive metabolite formation, associated with covalent protein binding (haptenization) and induction of an immune-mediated idiosyncratic response, has also been shown to occur for troglitazone (Kassahun et al., 2001; Tettey et al., 2001; Alvarez-Sánchez et al., 2006). Cellular toxicity studies pointed towards a direct toxicity of troglitazone or troglitazone sulfate, while the metabolization by cytochrome P450 3A4 (which is also responsible for the formation of the reactive intermediate metabolites) rather represents a detoxification pathway (Hewitt et al., 2002; Masubuchi et al., 2006). In other studies, the conjugated metabolites troglitazone sulfate and glucuronide showed a lower potential for direct cellular toxicity (Kostrubsky et al., 2000; Honma et al., 2002). A combination of multiple mechanisms, namely cholestasis leading to increased drug levels in hepatocytes, mitochondrial toxicity induced by troglitazone and/or bile acids, reactive metabolite formation and covalent protein binding associated with antigen presentation and stimulation of an immune-response might be triggered by troglitazone. This could fit with the proposal of haptenization-associated low-grade toxicity (danger signal) needed for a significant immune reaction (Uetrecht, 1999; Kaplowitz, 2005). The different mechanisms of troglitazone hepatotoxicity have been recently reviewed (Smith, 2003). 16.2.4.1

The cholestatic Potential of Troglitazone

Troglitazone and to a greater extent its main metabolite, troglitazone sulfate, inhibited the canalicular Bsep (Figure 16.2) in a competitive manner with apparent K I values of 1.3 and 0.23 μM, respectively (Table 16.1; Funk et al., 2001b). The higher inhibition potential

430

Hepatotoxicity

of troglitazone sulfate in these rat liver membrane vesicles indicated the importance of this metabolite in contributing to the overall cholestatic potential of troglitazone. This was supported by studies in the isolated perfused rat liver (IPRL), where troglitazone triggered a strong decline in the bile flow in livers from male rats (Preininger et al., 1999). In this model, troglitazone was completely cleared from the perfusate, metabolized in the liver mainly to troglitazone sulfate and glucuronide and eliminated to a great extent into bile (Funk et al., 2001b). Troglitazone elicited a rapid decrease of both bile flow and biliary excretion of a radio-labeled taurocholic acid tracer added to the perfusion medium, while more of the labeled bile acid was recovered from the perfusion medium (Figure 16.4). In equilibration 10 µM troglitazone --------------------------------- ++++++++++++++++++++++++++++++++++ 110 male rat liver 100

(a)

Taurocholate excreted (%)

120

90 80 70

Excretion into perfusate

60 50 40 30

Excretion into bile

20 10 0

10

20

30

10

20

30

40

50

60

Time (min) (b)

equilibration 10 µM troglitazone --------------------------------- ++++++++++++++++++++++++++++++++++ female rat liver Excretion into perfusate 100 Taurocholate excreted (%)

120

80 60 40 Excretion into bile 20 0

10

20

30

10

20

30

40

50

60

Time (min)

Figure 16.4 Cholestatic effect of troglitazone in the isolated perfused liver from (a) male and (b) female rats. Interference with the biliary export of a radio-labeled taurocholate tracer (3 nM) is shown as the mean of four-individual liver perfusions.


431

female rat livers, the effect was milder and delayed as compared to the livers from male rats. This gender difference could be attributed to quantitative differences in the metabolism of troglitazone, mainly to a rapid formation of troglitazone sulfate in male liver due to the presence of a high affinity male-specific sulfotransferase (Funk et al., 2001a). In vivo, in rats, troglitazone similarly interfered with the hepatic elimination of bile acids and induced a transient increase in plasma bile acid levels (Figure 16.3.), while an accumulation of an intravenously administered radio-labeled taurocholate tracer was observed in liver tissue (Funk et al., 2001b). These results support the cholestatic mechanism of troglitazone and its metabolite troglitazonesulfate at the canalicular pole of the hepatocytes, where Bsep is localized, as the site of interference with the hepatobiliary excretion of bile acids. 16.2.4.2

Interrelation of Drug Metabolism and Hepatobiliary Export Processes

Several lines of evidence suggested that the hepatobiliary export of troglitazone and related metabolites might represent a rate-limiting step in the overall elimination of troglitazone. In patients with hepatic impairment, troglitazone sulfate was found to accumulate about 4-fold in plasma, with a 3-fold increased half-life (Ott et al., 1998). In rats, troglitazone sulfate accumulated as the major drug-related metabolite in liver tissue, reaching high intracellular concentrations (Funk et al., 2001b). The higher cholestatic potential of troglitazone in isolated perfused male rat livers as compared to the livers from female rats (Figure 16.4) can be mainly attributed to differences in metabolism rather than differences in bile excretion or in Bsep inhibition. Liver plasma membrane vesicles isolated from either male or female rat liver tissue exhibited a very similar capacity for ATP-dependent taurocholate transport, based on the intrinsic clearance calculated from the respective apparent K M and Vmax parameters (Table 16.3 (a)). In addition, the potential of troglitazone (Table 16.3(b)) and troglitazone sulfate (Table 16.3(c)) to inhibit this ATP-dependent transport was very similar among the two genders. The conversion of troglitazone to troglitazone sulfate was studied using liver cytosolic fractions prepared from male and female rat liver tissue. The respective kinetic evaluation indicated the presence of a high-affinity sulfotransferase in male rat liver in addition to a low-affinity isoenzyme similarly present in both genders (Funk et al., 2001a). The calculated intrinsic clearance values are more than 5-fold higher in male rats (Table 16.3(d)) and consistently explain the higher levels of the troglitazone-sulfate metabolite in male rats (Table 16.3(e)). These higher liver tissue levels, together with the 5-fold-lower IC50 value for Bsep of troglitazone sulfate, can explain the higher sensitivity of male animals towards troglitazone as observed in the in vivo acute rat cholestasis model. At a dose of 6.25 mg/kg, no effect was observed in female animals, while in male rats a marked increase in the total plasma bile acid level of more than 40 μM was observed (Table 16.3(f)). These in vitro and in vivo studies clearly underline the importance of both drug metabolism and drug export processes in interference of a drug molecule with the hepatic bile formation. 16.2.4.3

Reactive Troglitazone Metabolite Formation as an Additional Trigger for Hepatotoxicity

In addition to the formation of conjugated phase II metabolites, troglitazone has been shown to undergo bioactivation processes in rat and human liver microsomes, leading to the formation of reactive intermediates capable of covalent binding to protein thiol groups.

In vitro transport capacity of taurocholate into cLPMV In vitro inhibition of Bsep by TGZ in rat cLPMV In vitro inhibition of Bsep by TGZ-S in rat cLPMV Metabolism of TGZ by rat cytosolic sulfotransferases Liver tissue levels of TGZ-S (30 min after 6.25 mg/kg iv dose) In vivo cholestatic effect of intravenous TGZ, increase in plasma bile acid levels 10 min after an intravenous dose (6.25 mg/kg)

(a) (b) (c) (d) (e)

(f)

Endpoint measured in vitro or in vivo

Case

34.9

34 25 90 159 30

μl min−1 mg−1 % inhibition at 1 μM % inhibition at 1 μM μl min−1 · mg−1 nmol g−1 μM

Male

Unit

>−2.5

∼ 24.5 ∼ 20 ∼ 88 >35 > 8.2

Female

Rat gender studied

Table 16.3 Gender differences in metabolism and cholestatic potential of troglitazone in rats. The metabolism of troglitazone (TGZ) to the main metabolite troglitazone sulfate (TGZ-S), liver tissue levels of TGZ-S and in vitro and in vivo cholestatic effects of TGZ and TGZ-S are shown. Data from Funk et al., 2001a,b


R'

P450 P450, O

all TZDs

O

P450 P450, TGZ, RGZ

C * * N* *

S

* * * NH S * RO

O

R'

O

O

* * XH

GS S

OH

X=O, 15NH

Non TZD-related GSH conjugates

O

N R=

433

N

N

PGZ

RGZ

HO TGZ

H N R' =

N desmethyl-RGZ

Figure 16.5 Bioactivation of thiazolidinedione (TZD) derivatives troglitazone (TGZ), rosiglitazone (RGZ) and pioglitazone (PGZ). For TGZ, the non-TZD-related glutathione adduct appears to be the predominant adduct formed in human liver microsomes (∗ indicates the positions of 13 C or 15 N).

Glutathione adducts of these reactive intermediates have been characterized to elucidate the structure of the reactive intermediates formed during phase I metabolism of troglitazone predominantly catalyzed by P450 3A4 (Figure 16.5). P450-mediated S-oxidation is suggested to lead to an unstable thiazolidinedione (TZD)-sulfoxide which undergoes spontaneous ring opening to form a reactive sulfenic acid-α-keto-isocyanate (Kassahun et al., 2001). This TZD-dependent bioactivation was seen for all three TZD drugs in rat and human liver microsomes by a characteristic change of the isotopic difference between the non-labeled and stable isotope-labeled analogues of the TZD-drugs (Alvarez-Sánchez et al., 2006). In addition, troglitazone, rosiglitazone and pioglitazone are also prone to form reactive intermediates at higher substrate concentrations, suggesting that covalent binding to liver proteins may also be expected for these two compounds. In human liver microsomes, the major glutathione adduct derived from troglitazone showed, however, the conservation of all four labeled positions of the TZD moiety. This was the predominant adduct detected and appears to be more abundant than the GSH adducts derived from TZD ring scission. Its structure was consistent with the previously reported adduct derived from the quinone methide activated troglitazone metabolite (Kassahun et al., 2001; Alvarez-Sánchez et al., 2006). This observation suggests that TGZ bioactivation via TZD ring scission does not contribute significantly to the extent of GSH adduct formation. Although the relative contribution of less-abundant reactive intermediates to hepatotoxic events remains unknown,

434

Hepatotoxicity

these findings suggest that the formation of these reactive intermediates not related to the TZD-moiety might be associated with the hepatotoxic potential of TGZ. Formation of quinone and quinone methide-type reactive intermediates has been linked to the prooxidant activity of TGZ in rat primary hepatocytes on cumene hydroperoxide-induced lipid peroxidation and cytotoxicity. This effect was significantly higher for TGZ as compared to other vitamin E analogues (Tafazoli et al., 2005). Cytotoxicity in hepatic cells and oxidative stress-inducing properties for thiazolidinedione derivatives were dependent on the presence of the 6-hydroxychromane moiety, suggesting a link to this specific molecular structure of TGZ. In addition, N1S1 rat hepatoma cells were more sensitive to TGZ and TGZ quinone than to other TZD derivatives lacking the 6-hydroxychromane moiety (Narayanan et al., 2003). Reactive metabolite formation and covalent binding to proteins might represent one mechanism contributing to the hepatic toxicity observed with troglitazone in some patients, in which drug-induced intrahepatic cholestasis might increase the individual susceptibility.

16.2.4.4

Active Hepatobiliary Export of Troglitazone Sulfate

Only recently, the human canalicular bile salt export pump, BSEP, has been cloned and therefore has enabled more mechanistic in vitro studies (Byrne et al., 2002; Noé et al., 2002). For the rat Bsep, the available data indicate that this transporter is involved in the excretion of taurocholate and related non-glucuronidated (except acyl glucuronide) nonsulfated bile salts (Keppler et al., 1992). It is mainly responsible for the bile salt-dependent portion of the overall bile flow. An involvement of Bsep in the transport of xenobiotics was suggested for non-conjugated organic anions (Hofmann, 1992). Based on its expression pattern, a broader substrate specificity of Bsep has been postulated in analogy to other multidrug resistance proteins (Török et al., 1999). Recently, pravastatin was reported as being a substrate for both the rat Bsep and the human BSEP transport proteins (Hirano et al., 2005). In addition, for troglitazone and troglitazone sulfate we have studied the ATP-dependent transport properties using liver plasma membrane vesicles from Sprague–Dawley rats (wt) and Mrp2-deficient rats (TR−), respectively (Figure 16.6). The model substrates, taurocholate and LTC4 , for the two ABC transporters, Bsep and Mrp2, were transported as expected, the former one in both vesicle preparations, while the latter one was transported only in wt vesicles. The radio-labeled troglitazone did not show any detectable ATP-dependent transport in this experimental in vitro system (data not shown), in spite of its Bsep inhibition potential. However, troglitazone sulfate showed ATP-dependent transport in both vesicle preparations, although considerable unspecific binding was seen, indicating the involvement of Bsep in the hepatobiliary transport of this compound. This was further verified using membrane vesicles recombinantly expressing either one of the two human canalicular ABC transporters, MRP2 or BSEP. Both transporters were able to mediate an ATP-dependent uptake of troglitazone sulfate into the respective vesicles (Figure 16.7). The functional activity of both transporters was verified by transport of the corresponding model substrate, LTC4 or taurocholate, respectively. Biliary excretion of troglitazone sulfate has also been described as one step of an enterohepatic circulation of both troglitazone sulfate and glucuronide metabolites (Kawai et al., 2000).

Cholestatic Potential through Inhibition of BSEP – ATP

435

+ ATP

40 35 30 25 15 0.6

LTC4

–1

20

–1

Transport rate (pmol min mg )

Taurocholate

45

wt

TR–

wt

TR–

wt

TR–

0.4

0.2

Troglitazone sulfate

0.0 200 150 100 50 0

cLPMV source

Figure 16.6 Hepatobiliary transport of troglitazonesulfate using rat liver plasma membrane vesicles (cLPMV) prepared from Sprague–Dawley rats (wt) and Mrp2-deficient rats (TR− ), respectively. Taurocholate and LTC4 were used as model substrates testing Bsep and Mrp2 functional activities.

16.2.4.5

Dose and Co-mediations as Important Factors Contributing to Troglitazone-Induced Cholestasis

Typical daily troglitazone doses in man are in the range of 400 to 600 mg, producing troglitazone plasma concentrations of 1 to 2 μg/ml (Table 16.1; Loi et al., 1999). A quantitative extrapolation of the cholestatic effect observed in rats to man is not possible, due to species-specific differences in the process of bile formation and in the overall disposition of troglitazone. However, the processes involved in the hepatobiliary metabolism and export of the absorbed fraction of troglitazone represent potential targets for drug interactions and regulative changes. Therefore, troglitazone might induce intrahepatic cholestasis in combination with other cholestatic drugs, other diseases or pharmacogenetic liabilities, thereby contributing to the hepatotoxicity observed in some patients treated with troglitazone. Simvastatin (Herrine and Choudhary, 1999) and lisinopril (Gitlin et al., 1998), both drugs known to induce cholestatic side effects, were co-medicated with troglitazone in patients developing signs of liver toxicity. Another report described three out of four cases of

Hepatotoxicity 90 80

–1

–1

Troglitazone sulfate transport (pmol mg min )

436

70 60 50 40 30 20 10 0 mock

MRP2

BSEP

Membrane vesicles used Figure 16.7 ATP-dependent transport of troglitazone sulfate into vesicles prepared from insect cells expressing either MRP2 or BSEP, or mock transfected as control, respectively. ATPdependent transport is shown as the mean of three independent experiments.

troglitazone-induced fulminant hepatitis where glibenclamide has been co-medicated with troglitazone. An interaction of the two drugs was suspected without further elaborating on a possible mechanism (Shibuya et al., 1998). Co-medication of potentially cholestatic drugs with troglitazone might increase the incidence in patients to develop liver toxicity. 16.2.5

Cholestatic Potential of Nefazodone

The antidepressant nefazodone (serzone) was withdrawn from the market due to drugmediated hepatic injury. Based on the clinical symptoms of the idiosyncratic hepatotoxicity, including jaundice, ALT and AST increase, high biliary elimination of nefazodone seen in animals and the relative high therapeutic doses of nefazodone in the range 200–400 mg/day, the potential for a cholestatic mechanism was studied in more details (Kostrubsky et al., 2006). In vitro, nefazodone caused a strong inhibition of the human bile salt transporter BSEP with an IC50 value of 9 μM, using membrane vesicles expressing BSEP. Nefazodone was furthermore found to interfere with the bile acid elimination in sandwich-cultured human hepatocytes. In addition, cytotoxicity was observed in human hepatocytes at nefazodone concentrations in the range of 10 μM. By inhibition of the metabolism, this could be associated with the concentrations of intact nefazodone rather than its metabolites. In rats, a transient increase in the serum bile acid level was observed at 1 h after oral drug administration. Based on these data, it is likely that the clinical hepatotoxicity of nefazodone is linked to the ability of the drug to interfere with the bile acid transport, primarily by an inhibition


437

of the canalicular bile acid export protein BSEP. Interestingly, also for nefazodone the formation of reactive metabolites and covalent binding to liver proteins, including P450 3A4, have been suggested as a factor contributing to the hepatotoxicity of this compound (Kalgutkar et al., 2005). 16.2.6

Estrogen and Progesterone Steroid Hormones

Intrahepatic cholestasis of pregnancy (ICP) appears as transient cholestasis in pregnant women. The mechanism underlying this cholestasis has been investigated in detail (Vallejo et al., 2005). Certain progesterone metabolites, such as 5α-pregnan-3α-ol-20-one (PM4), are elevated in serum of women with ICP. PM4 is conjugated with sulfate in the liver and then excreted into bile and this PM4–sulfate metabolite was found to have a strong transinhibition effect on BSEP in an experimental system where BSEP was expressed in oocytes preloaded with bile acids. This trans-inhibition of BSEP by the PM4–sulfate metabolite is likely involved in the formation of ICP and a genetic deficiency in BSEP might be a contributing factor (Pauli-Magnus and Meier, 2005). Natural and synthetic estrogenic compounds and their metabolites are well known to interfere with the hepatic bile formation (Bossard et al., 1993). For 17α-ethinylestradiol, a decreased bile excretion capacity, due to multiple transporter defects in the canalicular membrane, has been shown. After a five-day treatment with 17α-ethinylestradiol rats showed a marked increase in liver weight, plasma bile acid levels (over 10-fold) and alkaline phosphatase (Bossard et al., 1993). In parallel, a decrease of both the ATP-dependent taurocholate transport and dinitrophenyl glutathione transport ratios were found in the canalicular membranes, indicating a down-regulation of Bsep and Mrp2. For a physiological estrogen metabolite, estradiol-17β-glucuronide (E2 17G), which can induce cholestasis at higher concentrations in rats, a dose-dependent inhibition of the ATPdependent taurocholate transport has been observed using rat liver membrane vesicles (Stieger et al., 2000). The effect was not seen in Bsep-expressing membrane vesicles, indicating a trans-inhibition of Bsep after export by Mrp2, which is only present in the vesicles, prepared from rat liver tissue. This in vitro finding was consistent with the fact that E2 17G induced a cholestasis in normal rats but not in TR− rats deficient of the canalicular Mrp2 exporter (Stieger et al., 2000). In conclusion, for several estrogenic compounds and progesterones, a trans-inhibition of Bsep/BSEP by a (conjugated) metabolite after being excreted into bile has been shown. This mechanism is likely to be involved in the induction of intrahepatic cholestasis of these steroid hormones by the accumulation of bile acids within the hepatocyte. 16.2.7

HIV Protease Inhibitors

Hepatotoxicity as a significant adverse effect in antiretroviral therapy of HIV-infected patients often forces withdrawal of the respective therapy (McRae et al., 2006). The in vitro cholestatic potential was studied for four compounds, ritonavir, saquinavir, nevirapine and efavirenz (McRae et al., 2006). Both the interference with hepatic bile acid uptake and biliary excretion were addressed in different experimental systems. The NTCP- and OATPmediated bile acid uptake into hepatocytes was studied in sandwich-cultured rat and human

438

Hepatotoxicity

hepatocytes, while the BSEP-mediated biliary excretion of bile acids was studied using BSEP-expressing membrane vesicles. Ritonavir and saquinavir inhibited both NTCP- and OATP-mediated bile acid uptake and the BSEP-mediated biliary excretion of bile acids. Efavirenz only modestly inhibited the two processes, while nevirapine did not alter bile acid transport in any of the experimental systems. This rank order was not in agreement with the reported clinical hepatotoxic potential. Ritonavir-associated hepatotoxicity has been reported in up to 9 % of patients while hepatic toxicity is not generally associated with saquinavir or efavirenz, whereas nevirapine is commonly associated with the development of cholestatic and hepatocellular hepatotoxicity (McRae et al., 2006). The data suggest that inhibition of BSEP by the selected antiretroviral drugs is unlikely to be cause of antiretroviral-associated hepatotoxicity. The BSEP inhibition observed for some compounds appeared to be countered by an inhibition of the bile acid uptake transport systems, NTCP and OATP, potentially reducing the liver bile acid level. To the probably multifactorial hepatotoxicity different/other mechanism(s), such as interplay with metabolism, co-medications, genetic predisposition to name just a few, might contribute. 16.2.8

Fusidate

The steroid antibiotic fusidate has been shown to induce conjugated hyperbilirubinemia as early as two days after initiating the treatment in up to 48 % of the patients given the compound intravenously (Bode et al., 2002). The molecular mechanism leading to this conjugated hyperbilirubinemia was investigated in rats, in vitro using liver membrane vesicles and in vivo by investigating the effects on the bile flow and biliary elimination of different tracers (Bode et al., 2002). Fusidate inhibited both the ATP-dependent transport of the Mrp2 substrates estradiol17β-glucuronide and leukotriene C4 with K I values of 7.6 and 5.5 μM, as well as the ATP-dependent transport of the Bsep substrate cholyltaurine with a K I value of 2.2 μM. In vivo, in rats, fusidate reduced the bile flow and the excretion of substrates of both transporters (estradiol-17β-glucuronide and cholyltaurine), infused as tracers by 75 and 80 %, respectively. Based on these in vitro and in vivo data, fusidate competitively interacts with both the Bsep and Mrp2 canalicular ABC transporters, both involved in the elimination of biliary constituents. 16.2.9

Sulindac

In patients, hepatotoxicity, often of cholestatic nature, is an adverse effect described for long-term treatment with the nonsteroidal anti-inflammatory drug sulindac (Bolder et al., 1999). The effects of sulindac on the bile flow and bicarbonate secretion were studied in bile fistula rats (Bolder et al., 1999). A marked choleretic effect was observed, along with elimination of the unconjugated compound into bile. Studies with isolated perfused rat livers indicated a prolonged choleretic effect indicative of cholehepatic recirculation of sulindac. In the same model, an interference with the hepatic elimination of a cholyltaurine tracer was observed and upon co-infusion of cholyltaurine an acute cholestasis was induced. A competitive inhibition of bile acid transport was observed at both the basolateral and the canalicular pole of the hepatocyte. Overall, in the lower dose range sulindac is thought


439

to undergo biliary excretion and cholehepatic recirculation inducing choleresis, while at higher concentrations sulindac is likely to competitively interfere with the canalicular bile salt transporter, thereby contributing to the cholestatic effect observed in patients (Bolder et al., 1999). 16.2.10

Rifampicin and Rifamycin SV

Rifampicin and rifamycin SV, macrolide antibiotic drugs, frequently induce cholestasis as a side-effect (Stieger et al., 2000; Mita et al., 2006). Both compounds interfered with the ATP-dependent taurocholate transport into liver membrane vesicles and membrane vesicles expressing rat Bsep, with K I values in the range of 1 to 12 μM (Stieger et al., 2000). In addition, in the polarized LLC-PK1 cells, expressing both human NTCP and BSEP, rifamycin SV and rifampicin interfered with the bidirectional transport of fluorescently labeled bile acids (Mita et al., 2006). Rifampicin interfered mainly with the bile acid exporter BSEP, leading to an increased intracellular bile acid concentration, while rifamycin SV interfered with both the bile acid uptake carrier NTCP and the exporter BSEP, resulting in a reduction of the intracellular bile acid level and a reduced overall bidirectional bile acid transport (Mita et al., 2006). These studies clearly indicate the complexity of in vitro interaction studies with the active transport processes involved in hepatic bile acid elimination.

16.3

Species Differences in Bsep/BSEP Inhibition

Orthologues of human BSEP have been identified in mouse, rat and rabbit, sharing 81, 82 and 87 % amino acid identity, respectively, with the human protein (Noé et al., 2002). A similar transport of individual bile acids was found for the mouse, rat and human Bsep/BSEP (taurochenodeoxycholate > taurocholate > tauroursodeoxycholate ≥ glycocholate) (Noé et al., 2002). In patients with BSEP deficiency, bile salts are nearly absent in bile, while mice appear to have alternative bile secretory pathways (e.g. Mdr1 and, to a lesser extent, Mdr2, Mrp2 and Mrp3), based on studies with Bsep knockout mice (Lam et al., 2005). This might explain the fact that for many cholestatic compounds no signals of impaired bile secretion and associated hepatotoxicity were seen in rodent toxicity studies. Species differences were also observed in the affinity of rat Bsep and human BSEP for cholestatic compounds. While the inhibition of Bsep/BSEP by midecamycin and cyclosporin A was similar between rat and human cLPMV, cloxacillin and glibenclamide showed a more marked inhibition of the human BSEP as compared to the rat Bsep (Horikawa et al., 2003). The inhibition potential of glibenclamide and three thiazolidinedione antidiabetics, troglitazone, rosiglitazone and pioglitazone, has been analyzed for the rat (Bsep) and human bile salt (BSEP) transporters expressed in membrane vesicles (Table 16. 4). The rank-order of inhibition was similar for the two transporters studied. However, the human BSEP appeared to be more sensitive for the compounds tested, especially for rosiglitazone, for which a 10-fold lower IC50 was observed when using the human BSEP. In addition, the low apparent inhibition of pioglitazone on the human BSEP is not in agreement with the clinical observations, as this compound is not associated with significant liver enzyme elevations. The dose, tissue distribution, metabolism and other factors have to be considered to explain this apparent discrepancy.

440

Hepatotoxicity

Table 16.4 Species differences in the inhibitory potential of different glitazones on the rat and human bile salt transporters (Bsep, BSEP). The inhibitory effects were determined on the ATP-dependent taurocholate transporters expressed in membrane vesicles

Inhibitor Rosiglitazone Glibenclamide Troglitazone Troglitazone sulfate Pioglitazone

16.4

Inhibition of ATP-dependent taurocholate transport in membrane vesicles, IC50 (μM)

Typical dose range used in clinics (mg/day)

Rat Bsep

Human BSEP

8 10 400–600 — 45

54.4 9.6 10.3 9.3 5.5

6.5 6.3 2.3 4.2 0.4

Regulation of BSEP Expression

ABC transporters, which are involved in hepatic bile formation (BSEP, MRP2, MDR3), are regulated by different transcriptional cascades: ligand-activated nuclear receptors FXR, PXR and CAR, ligand-independent nuclear receptors and transcriptional cofactors (Eloranta and Kullak-Ublick, 2005). For bile acid homeostasis, the nuclear farnesoid X receptor (FXR) plays a pivotal role regulating key enzymes involved in bile acid synthesis and metabolism (CYP7A1, UGT2B4), as well as transport (NTCP, BSEP) (Cai and Boyer, 2006). Mrp3, an ABC transporter located at the basolateral membrane exporting compounds from the hepatocyte into plasma, is significantly up-regulated under cholestatic conditions (Donner and Keppler, 2001), reducing the intracellular bile acid concentration and potentially protecting the liver from the toxic effects of high bile acid levels. A similar regulation can be observed in primary cultured hepatocytes, indicating that these cells acquire a cholestatic phenotype (Rippin et al., 2001). While the bile acid uptake carriers Ntcp and Oatp1 were down-regulated, the exporters involved in bile acid excretion, Bsep, Mrp2 and Mrp1 (into bile or plasma) were up-regulated. Regulation of transporters in disease states and by nutrients and drugs might significantly alter the bile acid transport capacity in the hepatocytes and contribute to the inter-individual variability and sensitivity towards potentially cholestatic drugs.

16.5

Genetic Origin of Intrahepatic Cholestasis

One form of progressive familial intrahepatic cholestasis (PFIC) is characterized by high serum and low biliary bile acid levels and normal serum γ -GT levels (Trauner et al., 1998; Jansen et al., 1999). In one form, a mutation in FIC1, a P-type adenosine triphosphatase, is responsible for the cholestatic disease called ‘Byler disease’ according to the Amish population descending from Jacob Byler where this form was discovered. Linked with a slightly different form of PFIC, PFIC2, mutations in BSEP were found to be responsible for this progressive liver disease (Strautnieks et al., 1998). An impaired activity of the


441

hepatobiliary bile salt export pump leads to increased intracellular bile acid levels and intrahepatic cholestasis. Numerous mutations have been described in BSEP, leading to different degrees of intrahepatic cholestasis. In PFIC2 patients, a functional expression of BSEP is typically missing, while in milder forms (benign recurrent intrahepatic cholestasis, BRIC) different mutations in BSEP lead to proteins with lower functional activities. Such a mutation was described in a patient with BRIC and the BSEP protein bearing the respective mutation, showing in vitro a significantly reduced taurocholate transport activity (Noé et al., 2005). Such Bsep transport proteins with reduced activity, as found in BRIC, might render a patient more sensitive and in combination with cholestatic drugs lead to a significant intrahepatic cholestasis.

16.6

Addressing BSEP Inhibition in Drug Discovery and Development to Reduce the Cholestatic Properties of New Molecular Entities

The potential of certain drugs to induce intrahepatic cholestasis has been outlined above and the mechanism by which these compounds interfere with the hepatobiliary elimination of bile acids, mainly by inhibition of BSEP by the drug or major drug metabolites, is well documented for many of these cholestatic compounds. Drugs, which are used at relatively high doses and for which biliary excretion of parent drug or metabolites represents a major elimination pathway, might be particularly prone to interfere with bile acid elimination. The in vitro inhibition of Bsep/BSEP can be easily determined using appropriate in vitro systems, such as assessing the ATP-dependent taurocholate transport in liver plasma membrane vesicles or membrane vesicles expressing the rat/human transporter (Stieger et al., 2000; Funk et al., 2001b; Kostrubsky et al., 2006). Special attention has to be given to include the relevant major drug metabolites in these studies. A high-speed screening system has been reported recently; ATP-dependent uptake was measured using BSEP expressing membrane vesicles and 96-well filtration plates (MultiScreen, Millipore) (Hirano et al., 2006). The vectorial transport of taurocholate was also assessed using double-transfected cells (Mita et al., 2006). The polarized LLC-PK1 cells expressed NTCP and BSEP at the respective membrane and were grown as monolayers on membrane inserts and fluorescent bile acids were used as substrates. The inhibition of the bidirectional transport by a group of cholestatic model compounds was assessed. The drawback of this system is that for the evaluation of the quantitative inhibition potential of the exporter BSEP, the intracellular concentration of the test compound is unknown. For the evaluation of the cholestatic potential of metabolized compounds, the relevant metabolites have to be assessed separately, as these cells are metabolically not competent (Mita et al., 2006). In addition to the in vitro inhibition studies with Bsep/BSEP, it is important to consider the in vivo effect, including tissue (liver) distribution kinetics and metabolism of the test drug. Animal models are used for this purpose with the obvious limitations of species differences in transport and metabolism. An acute cholestasis model in rats was used to characterize cholestatic compounds assessing the effects of intravenously administered compounds on the plasma bile acid level (Funk et al., 2001a,b). A transient increase in bile acids can be observed for cholestatic compounds, allowing a ranking of the in vivo

Hepatotoxicity

Rel. in vivo effect on plasma bile acids –1 (ED50· (Δ10+ Δ30) )

442

10

Glibenclamide

1

0.1 CsA

0

2

4

6

8

10

12

In vitro IC50 ( M) Figure 16.8 In vitro and in vivo cholestatic potential of reference compounds (cyclosporin and glibenclamide) and representatives of one chemical series. The in vitro cholestatic potential was determined as the IC50 of Bsep inhibition in cLPMV, while the in vivo effect represents the increment of plasma bile acids increase in rats. Compounds with both low in vitro (high IC50 values) and in vivo cholestatic potentials (*) were selected for further characterization.

cholestatic potential. Results of the in vitro Bsep inhibition are shown in correlation to the in vivo cholestatic potential in rats for a series of molecules within one chemical class along with the corresponding results of the cholestatic reference compounds cyclosporin A and glibenclamide (Figure 16.8). This approach allows for the selection of compounds with a low cholestatic potential and can be systematically applied early on in the drug research and development process.

Abbreviations ALT Alanine transaminase AST Aspartate transaminase Bsep (Abb11) Rat canalicular bile salt export pump BSEP (ABCB11) Human canalicular bile salt export pump cLPMV Canalicular liver plasma membrane vesicles LTC4 Leukotriene C4 , a leukotriene–glutathione conjugate Mrp2 (Abcc2) Rat multidrug resistance related protein 2 MRP2 (ABCC2) Human multidrug resistance related protein 2 MV Membrane vesicles isolated from insect cells (Sf9) recombinantly expressing selected ABC transporters TGZ Troglitazone TZD(s) Thiazolidinedione(s) wt Wild type


443

References Alvarez-Sánchez R, Montavon F, Hartung T and Paehler A (2006). Thiazolidinedione bioactivation: A comparison of the bioactivation potentials of troglitazone, rosiglitazone and pioglitazone using stable isotope-labeled analogues and liquid chromatography tandem mass spectrometry. Chem. Res. Toxicol. 19: 1106–1116. Arrese M, Ananthananarayanan M and Suchy FJ (1998). Hepatobiliary transport: molecular mechanisms of development and cholestasis. Pediatr. Res. 44: 141–147. Bode KA, Donner MG, Leier I and Keppler D (2002). Inhibition of transport across the hepatocyte canalicular membrane by the antibiotic fusidate. Biochem. Pharmacol. 64: 151–158. Boehme M, Jedlitschky G, Leier I, Buechler M and Keppler D (1994). ATP-dependent export pumps and their inhibition by cyclosporins. Advan. Enzyme Regul. 34: 371–380. Bolder U, Trang NV, Hagey LR, Schteingart CD, Ton-Nu HT, Cerrè C, Elferink RP and Hofmann AF (1999). Sulindac is excreted into bile by a canalicular bile salt pump and undergoes a cholehepatic circulation in rats. Gastroenterology 117: 962–971. Bossard R, Stieger B, O’Neill B, Fricker G and Meier PJ (1993). Ethinylestradiol treatment induces multiple canalicular membrane transport alterations in rat liver. J. Clin. Invest. 91: 2714–2720. Byrne JA, Strautnieks SS, Mieli-Vergani G, Higgins CF, Linton KJ and Thompson RJ (2002). The human bile salt export pump: characterization of substrate specificity and identification of inhibitors. Gastroenterology 123: 1649–1658. Cai SY and Boyer JL (2006). FXR: a target for cholestatic syndromes? Expert Opin. Ther. Targets 10: 409–421. Chen C (1998). Troglitazone: An antidiabetic agent. Am. J. Health – Syst. Pharm. 55: 905–925. Donner MG and Keppler D (2001). Up-regulation of basolateral multidrug resistance protein 3 (Mrp3) in cholestatic rat liver. Hepatology 34: 351–359. Eloranta JJ and Kullak-Ublick GA (2005). Coordinate transcriptional regulation of bile acid homeostasis and drug metabolism. Arch. Biochem. Biophys. 433: 397–412. Fattinger K, Funk C, Pantze M, Weber C, Reichen J, Stieger B and Meier PJ (2001). The endothelin antagonist bosentan inhibits the canalicular bile salt export pump: a potential mechanism for hepatic adverse reactions. Clin. Pharmacol. Ther. 69: 223–231. Fukano M, Amano S, Sato J, Yamamoto K, Adachi H, Okabe H, Fujiyama Y and Bamba T (2000). Subacute hepatic failure associated with a new antidiabetic agent, troglitazone: a case report with autopsy examination. Hum. Pathol 31: 250–253. Funk C, Pantze M, Scheuermann G, Jehle L, Ponelle C, Lazendic M and Gasser R (2001a). Troglitazone-induced intrahepatic cholestasis by an interference with the hepatobiliary export of bile acids in male and female rats. Correlation with the gender difference in troglitazone sulfate formation and the inhibition of the canalicular bile salt export pump (Bsep) by troglitazone and troglitazone sulfate. Toxicology 167: 83–98. Funk C, Ponelle C, Scheuermann G and Pantze M (2001b). Cholestatic potential of troglitazone as a possible factor contributing to troglitazone-induced hepatotoxicity. In vivo and in vitro interaction at the canalicular bile salt export pump (Bsep) in the rat. Mol. Pharmacol. 59: 627–635. Gerloff T, Stieger B, Hagenbuch B, Landmann L and Meier PJ (1997). The sister P-glycoprotein mediates ATP-dependent taurocholate (TCA) transport. Hepatology 26: 358A. Gitlin N, Julie NL, Spurr CL, Lim KN and Juarbe HM (1998). Two cases of severe clinical and histologic hepatotoxicity associated with troglitazone. Ann. Inter. Med. 129: 36–38. Gores GJ, Miyoshi H, Botla R, Aguilar HI and Bronk SF (1998). Induction of mitochondrial permeability transition as a mechanism of liver injury during cholestasis: a potential role for mitochondrial proteases. Biochim. Biophys. Acta 1366: 167–175. Herrine SK and Choudhary C (1999). Severe hepatotoxicity associated with troglitazone. Ann. Inter. Med. 130: 163–164.

444

Hepatotoxicity

Hewitt NJ, Lloyd S, Hayden M, Butler R, Sakai Y, Springer R, Fackett A and Li AP (2002. Correlation between troglitazone cytotoxicity and drug metabolic enzyme activities in cryopreserved human hepatocytes. Chem. Biol. Interact. 142: 73–82. Hirano M, Maeda K, Hayashi H, Kusuhara H and Sugiyama Y (2005). Bile salt export pump (BSEP/ABCB11) can transport a nonbile acid substrate, pravastatin. J. Pharmacol. Exp. Ther. 314: 876–882. Hirano H, Kurata A, Onishi Y, Sakurai A, Saito H, Nakagawa H, Nagakura M, Tarui S, Kanamori Y, Kitajima M and Ishikawa T (2006). High-speed screening and QSAR analysis of human ATPbinding cassette transporter ABCB11 (bile salt export pump) to predict drug-induced intrahepatic cholestasis. Mol. Pharm. 3: 252–265. Hofmann AF (1992). The cholehaptic circulation of unconjugated bile acids: an update, in Bile Acids and the Hepatobiliary System. From Basic Science to Clinical Practice, G. Paumgartner, A. Stiehl and W. Gerok (Eds), Kluwer Academic Publishers, Dordrecht, The Netherlands, pp. 143– 160. Honma W, Shimada M, Sasano H, Ozawa S, Miyata M, Nagata K, Ikeda T and Yamazoe Y (2002). Phenol Sulfotransferase, ST1A3, as the Main Enzyme Catalyzing Sulfation of Troglitazone in Human Liver. Drug Metab. Dispos. 30: 944–952. Horikawa M, Kato Y, Tyson CA and Sugiyama Y (2003). Potential cholestatic activity of various therapeutic agents assessed by bile canalicular membrane vesicles isolated from rats and humans. Drug Metab. Pharmacokinet. 18: 16–22. Huerta C, Zhao SZ and Garcia Rodriguez LA (2002). Risk of acute liver injury in patients with diabetes. Pharmacotherapy 22: 1091–1096. Jansen PL, Strautnieks SS, Jacquemin E, Hadchouel M, Sokal EM, Hooiveld GJ, Koning JH, De Jager Krikken A, Kuipers F, Stellaard F, Bijleveld CM, Gouw A, Van Goor H, Thompson RJ and Muller M (1999). Hepatocanalicular bile salt export pump deficiency in patients with progressive familial intrahepatic cholestasis (see comments). Gastroenterology 117: 1370–1379. Kadmon M, Klünemann C, Böhme M, Ishikawa T, Gorgas K, Otto G, Herfarth C and Keppler D (1993). Inhibition by cyclosporin A of adenosin triphosphate-dependent transport from the hepatocyte into bile. Gastroenterology 104: 1507–1514. Kalgutkar AS, Vaz AD, Lame ME, Henne KR, Soglia J, Zhao SX, Abramov YA, Lombardo F, Collin C, Hendsch ZS and Hop CE (2005). Bioactivation of the nontricyclic antidepressant nefazodone to a reactive quinone–imine species in human liver microsomes and recombinant cytochrome P450 3A4. Drug Metab. Dispos. 33: 243–253. Kaplowitz N (2005). Idiosyncratic drug hepatotoxicity. Nat. Rev. Drug Discov. 4: 489–499. Kassahun K, Pearson PG, Tang W, McIntosh I, Leung K, Elmore C, Dean D, Wang R, Doss G and Baillie TA (2001). Studies on the Metabolism of Troglitazone to Reactive Intermediates in vitro and in vivo. Evidence for Novel Biotransformation Pathways Involving Quinone Methide Formation and Thiazolidinedione Ring Scission. Chem. Res. Toxicol. 14: 62–70. Kawai K, Hirota T, Muramatsu S, Tsuruta F, Ikeda T, Kobashi K and Nakamura KI (2000). Intestinal absorption and excretion of troglitazone sulphate, a major biliary metabolite of troglitazone. Xenobiotica 30: 707–715. Keppler D, Mueller M, Boehme M, Mansur-Garza E and Berger U (1992). ATP dependent transport of taurocholate across the hepatocyte canalicular membrane mediated by a glycoprotein binding ATP and bile salt, in Bile Acids and the Hepatobiliary System. From Basic Science to Clinical Practice, G. Paumgartner, A. Stiehl and W. Gerok (Eds), Kluwer Academic Publishers, Dordrecht; The Netherlands, pp. 106–111 Kostrubsky VE, Sinclair JF, Ramachandran V, Venkataramanan R, Wen YH, Kindt E, Galchev V, Rose K, Sinz M and Strom SC (2000). The role of conjugation in hepatotoxicity of troglitazone in human and porcine hepatocyte cultures. Drug Metab. Dispos. 28: 1192–1197.


445

Kostrubsky SE, Strom SC, Kalgutkar AS, Kulkarni S, Atherton J, Mireles R, Feng B, Kubik R, Hanson J, Urda E and Mutlib AE (2006). Inhibition of hepatobiliary transport as a predictive method for clinical hepatotoxicity of nefazodone. Toxicol. Sci. 90: 451–459. Krivoy N, Zaher A, Yaacov B and Alroy G (1996). Fatal toxic intrahepatic cholestasis secondary to glibenclamide. Diabetes Care 19: 385–386. Kullak-Ublick GA and Meier PJ (2000). Mechanisms of cholestasis. Clin. Liver Dis. 4: 357–385. Lam P, Wang R and Ling V (2005). Bile acid transport in sister of P-glycoprotein (ABCB11) knockout mice. Biochemistry 44: 12598–12605. Loi CM, Young M, Randinitis E, Vassos A and Koup JR (1999). Clinical pharmacokinetics of troglitazone. Clin. Pharmacokinet. 37: 91–104. Masubuchi Y, Kano S and Horie T (2006). Mitochondrial permeability transition as a potential determinant of hepatotoxicity of antidiabetic thiazolidinediones. Toxicology 222: 233–239. McRae MP, Lowe CM, Tian X, Bourdet DL, Ho RH, Leake BF, Kim RB, Brouwer KL and Kashuba AD (2006). Ritonavir, Saquinavir and Efavirenz, but not Nevirapine, Inhibit Bile Acid Transport in Human and Rat Hepatocytes. J. Pharmacol. Exp. Ther. 318: 1068–1075. Mita S, Suzuki H, Akita H, Hayashi H, Onuki R, Hofmann AF and Sugiyama Y (2006). Inhibition of bile acid transport across NTCP and BSEP co-expressing LLC-PK1 cells by cholestasis-inducing drugs. Drug Metab. Dispos. 34: 1575–1581. Narayanan PK, Hart T, Elcock F, Zhang C, Hahn L, McFarland D, Schwartz L, Morgan DG and Bugelski P (2003). Troglitazone-induced intracellular oxidative stress in rat hepatoma cells: a flow cytometric assessment. Cytometry A 52: 28–35. Noé J, Stieger B and Meier PJ (2002). Functional expression of the canalicular bile salt export pump of human liver. Gastroenterology 123: 1659–1666. Noé J, Kullak-Ublick GA, Jochum W, Stieger B, Kerb R, Haberl M, Mullhaupt B, Meier PJ and Pauli-Magnus C (2005). Impaired expression and function of the bile salt export pump due to three novel ABCB11 mutations in intrahepatic cholestasis. J. Hepatol. 43: 536–543. Ott P, Ranek L and Young M (1998). Pharmacokinetics of troglitazone, PPAR-gamma antagonist, in patients with hepatic insufficiency. Eur. J. Clin. Pharmacol. 54: 567–571. Pauli-Magnus C and Meier PJ (2005). Hepatocellular transporters and cholestasis. J. Clin. Gastroenterol. 39: S103–S110. Preininger K, Stingl H, Englisch R, Fürnsinn C, Graf J, Waldhäusl W and Roden M (1999). Acute troglitazone action in isolated perfused rat liver. Br. J. Pharmacol. 126: 372–378. Rippin SJ, Hagenbuch B, Meier PJ and Stieger B (2001). Cholestatic expression pattern of sinusoidal and canalicular organic anion transport systems in primary cultured rat hepatocytes. Hepatology 33: 776–782. Roman ID, Monte MJ, Esteller A and Jimenez R (1989). Cholestasis in the rat by means of intravenous administration of cyclosporine vehicle, Cremophor EL. Transplantation 48: 554–558. Shibuya A, Watanabe M, Fujita Y, Saigenji K, Kuwao S, Takahashi H and Takeuchi H (1998). An autopsy case of troglitazone-induced fulminant hepatitis. Diabetes Care 21: 2140–2143. Smith MT (2003). Mechanisms of Troglitazone Hepatotoxicity. Chem. Res. Toxicol. 16: 679–687. Spreckelsen U, Kirchhoff R and Haacke H (1991). Cholestatic jaundice during lovastatin medication. Dtsch. Med. Wochenschr. 116: 739–740. Stieger B, Fattinger K, Madon J, Kullak-Ublick GA and Meier PJ (2000). Drug- and estrogeninduced cholestasis through inhibition of the hepatocellular bile salt export pump (Bsep) of rat liver. Gastroenterology 118: 422–430. Strautnieks SS, Bull LN, Knisely AS, Kocoshis SA, Dahl N, Arnell H, Sokal E, Dahan K, Childs S, Ling V, Tanner MS, Kagalwalla AF, Nemeth A, Pawlowska J, Baker A, Mieli-Vergani G, Freimer NB, Gardiner RM and Thompson RJ (1998). A gene encoding a liver-specific ABC transporter is mutated in progressive familial intrahepatic cholestasis. Nat. Genet. 20: 233–238.

446

Hepatotoxicity

Tafazoli S, Spehar DD and O’Brien PJ (2005). Oxidative stress mediated idiosyncratic drug toxicity. Drug Metab. Rev. 37: 311–325. Tettey JN, Maggs JL, Rapeport WG, Pirmohamed M and Park BK (2001). Enzyme-induction dependent bioactivation of troglitazone and troglitazone quinone in vivo. Chem. Res. Toxicol. 14: 965–974. Török M, Gutmann H, Fricker G and Drewe J (1999). Sister of P-glycoprotein expression in different tissues. Biochem. Pharmacol. 57: 833–835. Trauner M, Meier PJ and Boyer JL (1998). Molecular pathogenesis of cholestasis. New Engl. J. Med. 339: 1217–1227. Uetrecht JP (1999). New concepts in immunology relevant to idiosyncratic drug reactions: The ‘danger hypothesis’ and innate immune system. Chem. Res. Toxicol. 12: 387–395. Vallejo M, Briz O, Serrano MA, Monte MJ and Marin JJ (2005). Potential role of trans-inhibition of the bile salt export pump by progesterone metabolites in the etiopathogenesis of intrahepatic cholestasis of pregnancy. J. Hepatol. 44: 1150–1157. Velayudham LS and Farrell GC (2003). Drug-induced cholestasis. Expert Opin. Drug. Saf. 2: 287– 304. Zhang L, Strong JM, Qiu W, Lesko LJ and Huang SM (2005). Scientific Perspectives on Drug Transporters and Their Role in Drug Interactions. Mol. Pharm. 3: 62–69. Zimniak P, Pikula S, Bandorowicz Pikula J and Awasthi YC (1999). Mechanisms for xenobiotic transport in biological membranes. Toxicol. Lett. 106: 107–118.

Section 5 Genomics of Hepatotoxicity

17 Application of Toxicogenomics in Predicting Hepatotoxicity – Potentials and Challenges Wen Lin, Guoxiang Shen, Tin Oo Khor and Ah-Ng Tony Kong

17.1

Introduction

The liver is an important organ not only because it is the largest organ in the body, but also because it is the major organ to metabolize numerous xenobiotics and facilitates the detoxification of toxicants. Therefore, the liver itself is the main target of many potential adverse events generated from its exposure to xenobiotics. Drug-induced hepatotoxicity is a common cause of liver injury which mimics all forms of acute and chronic liver diseases and accounts for a large percentage of the cases of acute liver failure in drug therapy. It is not surprising that drug-induced idiosyncratic hepatotoxicity is the most frequent cause of drug withdrawal post-marketing. Hence, it is critical to be able to identify potential signals or ‘tags’ of hepatotoxicity in pre-clinical or early clinical trials during drug development processes. Traditional toxicological studies usually focus on the short-term 14-week and long-term 2-year rodent (such as rats) animal experiments to evaluate toxicological endpoints, including body weight, organ weight, death rate, tumor incidence and level of serum toxicity biomarkers. Therefore, traditional toxicology studies have the obvious disadvantages, such as being time-consuming and the generation of phenomenological results, plus at times, it’s almost impossible to predict the toxicity of new drug candidates based on currently existing traditional toxicology databases with these phenomenological endpoints. All of these shortcomings form the bottleneck in drug discovery and development processes in the pharmaceutical industry. The rapid development of genomic science provides


450

Hepatotoxicity

many opportunities in drug research and development. One of these is the application of genomics in toxicology, or toxicogenomics. Toxicogenomic studies integrate knowledge of genomics, bioinformatics and toxicology to measure the gene expression changes provoked by exposure to a known toxicant or unknown drug candidates (Nuwaysir et al., 1999; Chin and Kong, 2002; Khor et al., 2006). It is believed that toxicants could elicit unique gene expression profiles which could be used to classify them from other non-toxicants. Therefore, databases constructed based on gene expression profiles elicited by known classes of toxicants could be used to predict phenotypes of toxicity of unknown drug candidates, such as potential hepatotoxicity. However, the application of toxicogenomics to hepatotoxicity studies requires an awareness of the limitation of gene expression profiling in the context of mechanistic and predictive toxicology. For example, many known toxicants induce toxicity by directly binding to proteins and DNA, not by directly altering genes involved in toxicity. Gene expression profiles also depend on time, dose and duration of the exposure of the chemical. For many genes, their expression changes reflect secondary outcomes due to primary upstream events rather than a direct response to toxicant exposure. Nevertheless, microarray-based toxicogenomics is still a very young and promising tool to address the mechanisms of hepatotoxicity of known toxicants or predict potential liver toxicity of new drug candidates (Lord et al., 2006).

17.2

Methodology

Because of the availability of genome-scale DNA sequences and functional information, microarray-based gene expression profiling has become an unprecedented powerful tool in toxicogenomic studies. During the past decades, numerous systems were developed inhouse or commercially, to construct large-scale DNA arrays. However, in general, all of these platforms are either cDNA- or oligonucleotide-based arrays (Butte, 2002). For all microarray analysis, the total RNA from samples is amplified to produce antisense cRNA (Figure 17.1). For spotted cDNA microarrays, different fluorescence-labeled cDNAs are prepared from cRNA amplified from treated and control samples; the labeled cDNAs are then purified and hybridized to the cDNA arrays. A cDNA microarray uses a single probe to detect a target gene, and the two biological (control vs. treated) samples are labeled with different colors and mixed at 1:1 ratio. After hybridization, the two colors are scanned separately using a laser and the relative expression level of each gene is determined by comparing the intensities after overlapping the two images. For oligonucleotide-based genomic arrays such as ‘Affymetrix GeneChips’, the cRNAs amplified from biological samples are further subjected to biotin-labeling and fragmentation. Each biological sample is then hybridized with an individual microarray (Figure 17.1). Affymetrix arrays use 11 to 20 pairs of oligonuleotides (containing a pair of ‘mis-matched’ oligonucleotides) to probe a target gene; after hybridization and laser scanning, the intensity differences between ‘perfect-match’ and ‘mis-match’ probes are calculated by appropriate image analysis software to give both quantitative (signal intensity) and qualitative (statistical significance of signals) measurements. The relative expression levels of each gene between different biological samples are obtained after normalization and comparison using software such as ‘Affymetrix GCOS’ and ‘GeneSpring’. Whole genome-based microarrays contain probes representing expressed sequence tags (ESTs). Many ESTs represent transcribed genes with

Application of Toxicogenomics in Predicting Hepatotoxicity

451

Figure 17.1 Schematic illustration of using a microarray-based toxicogenomics approach to classify toxicants or predict the classes of unknown compounds

unknown functions. Therefore, genome-scale microarrays are not ideal platforms for toxicity molecular mechanisms studies due to the lack of functional information of these ESTs. However, gene expression profiles generated from these platforms after toxicants exposure have been successfully used to identify genes differentially regulated by different classes of toxicants or unknown compounds. The identification of genes, whose expression changes, are either directly linked to toxicity or related to the perturbation of cellular signaling pathways related to toxicity, appears to be the most important step to classify toxicants or predict toxicity. On the other hand, genes whose expression changes are not related to toxicity may interfere or confound the above applications in genomic array assays. For arrays designed specifically for toxicogenomic studies, especially these spotted cDNA arrays, sets of genes related to toxicity are selected to construct the array, such as the ‘NIEHS’ in-house printed ‘Human ToxChip V1.0’ cDNA array. The advantages of using these specific toxicologyfocused microarray platforms include the generation of less data points for further analysis and interpretation, and almost all the alternations of genes on the array could be related to toxicity induced by toxicants exposure. With respect to the technical aspect, each type of microarray has employed appropriate data normalization techniques. For spotted cDNA microarrays, simple normalization could be done by adjusting the overall brightness of each scanned microarray image, assuming

452

Hepatotoxicity

that the quantities of RNA used are equal, or by normalizing the expression levels of housekeeping genes, assuming their expression levels are constant across the experimental conditions. More advanced normalization methods include the use of a locally weighted scatterplot-smoothing regression algorithm, followed by standard deviation regularization between array sub-grids. Normalization to the expression levels of housekeeping genes, and RMA and GCRMA normalization methods, are used in ‘Affymetrix GeneChip’ array analysis. These approaches have substantially improved the reliability and reproducibility of microarray data, even across different platforms or laboratories (Larkin et al., 2005). Therefore, under the same experiment design and similar RNA handling protocols, biological differences between samples have more significant effects on gene expression profiles than platform-specific effects, suggesting that gene expression profiles generated from different microarray platforms are comparable to each other. However, the gene expression levels of both platforms should be transformed to the log2 ratio of gene expression in response to treatments, to the mean value of control, which appear to be more biologically relevant values than the intensity measurements among different platforms (Larkin et al., 2005). To further improve the reliability of microarray data in toxicogenomic applications, a proper design of experiments should be at least equally emphasized. For example, the transcription of many genes is regulated by hormones-related signaling pathways; therefore, the gender of animal or time of sampling should be consistent cross all the biological samples. Another important factor needing to be considered is the heterogenecity of the organ or tissue to be used for RNA isolation, since different parts of an organ may have different subpopulations of cells which might respond differently to the same toxicants exposure. Failure to minimize these variations during experimental design or RNA preparation may generate biological differences that could lead to the misinterpretation of microarray data.

17.3

Data Analysis

The successful application of toxicogenomics in predicting hepatotoxicity relies on the appropriate analysis of microarray data. Experiments using high-density array platforms could easily generate thousands to millions of data points in one experiment, which represents a big challenge for further downstream data analysis. Currently, there are at least two types of mathematical methods to identify genes or expression patterns of interest from microarray data based on the different mathematical algorithms used. The supervised methods are used to identify genes that fit a pre-defined pattern, while the unsupervised methods are used to characterize the signature of a data set without a priori input or knowledge of a training set (Butte, 2002; Maggioli et al., 2006). Supervised methods use statistical approaches such as the ANOVA F-test to identify genes whose expression levels are significantly different between different biological samples; more importantly, supervised methods could also detect a discriminatory gene set among expression profiles generated from studies of toxicants with known toxicological properties. Therefore, supervised methods not only can detect genes related to toxicantsinduced toxicity, but could also perform class comparisons based on gene expression profiles generated from different classes of toxicants. Overall, these methods could be used for discriminant analysis, class prediction and supervised pattern recognition. All of these


453

functions are important for toxicogenomic studies to predict the potential hepatotoxicity of new drug candidates or their toxicity classification, using the training databases constructed from gene expression profiles of known liver toxicants. The most popular supervised technique is the nearest-neighbor analysis. Nearest-neighbor methods are based on measuring the distance between observations, such as the Euclidean distance between two gene expression profiles. This method is commonly used to identify a gene expression pattern that best matches a designated query pattern. In toxicogenomics, a query pattern or training set can be generated by microarray analysis of tissues exposed to various toxicants with known types of toxicity compared with that in untreated control tissues. New compounds or drug candidates can then be tested in the same tissues under the same experimental conditions and the distance of their expression patterns from the training set can be calculated. Then, their potential toxicity could be classified. Another widely used supervised technique is support vector machines (SVMs). When given a set of training examples, SVMs will be able to recognize informative patterns in input data and make generations on previously unseen samples. Like other supervised methods, SVMs require prior knowledge of the classification system. Other supervised techniques that can be used in toxicogenomics include neural networks and learning vector quantization – however, an in-depth discussion of these methods is beyond the scope of this current chapter. While supervised methods require that the genes or conditions are associated with external information such as gene function or classes of toxicants to provide pre-existing classifications, unsupervised methods require no additional information besides the microarray data itself. These methods are used to find patterns or relationships in a data set instead of trying to achieve classification. There are three major classes of techniques in unsupervised learning: principal-component analysis (PCA), clustering and network determination (Butte, 2002). PCA is an approach to reduce the dimensionality of the complex data set. The first principle component represents more of the variations among samples than the second and third principle components. The combination of the three components can be used to visualize the genes or samples in a three-dimensional space. Therefore, clustering methods are mostly used to visualize the similarities between different biological samples such as gene expression profiles after treatment with different classes of toxicants. The most common type of clustering algorithms is hierarchical algorithms. Hierarchical clustering calculates the distance between samples and visually cluster the data set in a dendrogram tree. Data sets closer to each other in the dendrogram tree are more similar to each other than those in separated branches. This method is very useful to characterize the number of classes of genes represented by the treatments in toxicogenomics (class discovery), but not suitable for class prediction since it is a subjective technique and highly influenced by selection of similarity measure used to calculate the Euclidian distance. Another unsupervised method, relevance networks, analyze microarray data at the genome or network level. It first compares all of the genes or features in a pair-wise manner and then calculates a correlation coefficient and determines a threshold value. Pairs of features with a measure greater than the threshold value are kept and displayed as nodes with different degree of cross-connectivity. Using relevance networks, a gene could be directly or indirectly linked to other genes as well as phenotypic measurements such as measurement of liver toxicity. Overall, the combination of supervised and unsupervised analysis of high quality microarray data could successfully identify specific gene expression patterns between known

454

Hepatotoxicity

classes of toxicants, or classify unknown compounds, as well as potentially addressing molecular mechanisms related to liver toxicity (Figure 17.1).

17.4

Applications

Toxigenomics is a relatively new discipline in toxicology with promising applications in predicting hepatotoxicity. One of the first papers using the term ‘toxigenomics’ was published in 1999 by Nuwaysir et al. (1999). In this paper, the scope and practice of toxigenomics were described as the identification of potential human and environmental toxicants, and their putative mechanisms of toxicity. In the past few years, scientists have developed four distinctive, but nevertheless not mutually exclusive, applications of toxigenomics in predicting hepatotoxicity. First, gene expression profiles are promoted as an approach of elucidating the mechanisms of toxicity and adding knowledge to the body of toxicology. This is referred to as ‘mechanistic toxicology’. Secondly, researchers are exploring the possibility that gene expression profiles may provide a basis for a new molecular rationale for the classification of toxicants. Thirdly, scientists are actively pursuing the potential of gene expression profiles to enable the prediction of the toxicities of unknown compounds and thereby provide a basis for their classification. This is referred to as ‘predictive toxicology’. The fourth application of toxicogenomic is identification and characterization of genes whose expression can be used as predictive biomarkers for specific types of organ toxicity. 17.4.1

Mechanistic Toxicology

Toxicogenomics is a new tool for mechanistic studies that may one day exceed the conventional approaches in terms of sensitivity and speed. This is referred to as ‘mechanistic toxicology’ (Pennie, 2000; Fielden and Zacharewski, 2001). In response to an insult, the changes at the protein and gene expression level precede the response at the physiological level, in the liver or the entire organism. Thus, mechanisms leading to toxic endpoints at the physiological level may be restructured by examination of changes in expression of genes and proteins (Pennie, 2000; Burchiel et al., 2001; Fielden and Zacharewski, 2001; Hamadeh et al., 2001). The early toxigenomics studies (Burczynski et al., 2000; Bulera et al., 2001; Hamadeh et al., 2002) demonstrated that it is a valuable approach for the illumination of mechanism of response to hepatotoxicants. For instance, Burcyzynski et al. (2000) distinguished two mechanistically unrelated classes (cytotoxic anti-inflammatory drugs and DNA-damaging agents) of toxicants based on a cluster-type analysis of gene expression profiles in HepG2 cells exposed to various compounds. These studies are an early step towards demonstrating toxicogenomics-based discrimination of toxic mechanisms. Likewise, Elliger-Ziegebauer and coworkers (Ellinger-Ziegelbauer et al., 2004) addressed the issue whether genotoxic carcinogens lead to deregulation of genes involved in common pathways at an early time point during exposure, which could allow greater insight into the mechanisms leading to tumor formation. They gave rats four genotoxic hepatocarcinogens: dimethylnitrosamine, 2-nitrofluorene, aflatoxin B1 and 4-(methylnitrosamino)1-(3-pyridyl)-1-butanone. Overall, a DNA damage response, a detoxification response and survival/proliferation pathways were commonly affected by all four carcinogens, whereas


455

a necrotic inflammatory response at the gene expression level correlated with the histologically detectable necrosis and subsequent inflammation in the case of dimethylnitrosamine and aflatoxin B1. Thus, they identified a particular combination of mechanisms characterizing the early response to this class of compounds soon after treatment. McMillian and coworkers (McMillian et al., 2005) have performed a series of in-depth studies in drug-induced oxidative stress in rat liver through the methodology of toxigenomics. First, they established gene transcriptional signature patterns of macrophage activator (MA) using training set agents. Subsequently, using these gene sets, macrophage activator-like compounds were identified, including carbon tetrachloride, thioacetamide, allyl alchol, dimethylnitrosamine, d-galactosamine, concanvalin A, gadolinium, coumarin, etoposide and several NSAIDS. Using the same strategy, they constructed gene transcriptional signature patterns of peroxisome proliferators (PPs). To construct gene signatures for oxidative stressors/reactive metabolites (OSs/RMs) hepatotoxicants, a large set of compounds as training sets representing many forms of oxidative stress (redox cycling, reactive metabolites, glutathione and antioxidant depletors, and phase II enzyme inducers) were chosen to acquire the gene transcriptional signature patterns of oxidative stressors/reactive metabolites (OSs/RMs). Similarly, they identified some paradigm compounds using the gene signature. In addition to predicting and classifying toxicants based on gene signature sets, they have managed to look into possible mechanisms of toxicity in drug-induced oxidative stress. The most remarkable finding is that many of the genes in the OSs/RMs transcriptional signature gene appear to be regulated by only one transcription factor, Nrf2, which binds to the antioxidant response element (ARE) (also known as the electrophilic response element, EpRE) and known to be important in phase II drug metabolism enzyme induction and in handling oxidative stress (Kong et al., 2001). Using the OSs/RMs transcriptional signature gene set, the MA-like compounds clustered tightly together, but away from OSs/RMs and other compounds. Correspondingly, using a set of six Nrf2-inducible genes, NAD(P)H:menadione oxidoreductase/diaphorase, microsomal epoxide hydrolase, glutathione transferase (GST) mu type 2, GST alpha type, HSP90 alpha, and glutamate-cysteine ligase (gama-glutamylcysteine synthetase), OSs/RMs samples were separated from most other samples, whereas MA samples displayed different gene expression profiles and PP samples overlapped the control sample. They also found that the macrophages activators could active Stat-3 and NF-κB, whereas peroxisome proliferators could activate PPARα. Using the transcriptional signature gene set of macrophage activator, a number of peroxisome proliferators (PPs)-treated samples were found to cluster together but away from the macrophage activator-treated samples, largely due to opposite regulation of genes involved in fatty acid beta oxidation by peroxisome proliferators (PPs) (induced) and macrophage activator (MA) (repressed). They found a number of compounds weakly co-clustered with the PP-training set samples: benzafibrate, benzbromarone, dichloroacetate, valproic acid and the nonsteroidal anti-inflammatory drugs (NASIDs) diflunisal, fenbufen and ibuprofen. Most of these compounds have previously been reported to have mild PPARα agonist activity, which is responsible for much of the potential hepatotoxicity in the rodents of this class of compounds. Taken together, toxicogenomics provide important insights into the possible mechanisms involved in the toxicities of hepatotoxicants. Additional studies that examine multiple endpoints at the molecular, cellular, tissue and physiological levels would need to be integrated to better define the mechanisms of hepatotoxicity.

456

17.4.2

Hepatotoxicity

Classification of Hepatotoxicants

The field of toxicogenomics, through the use of DNA microarrays, has the potential to advance our understanding of how multiple genes are regulated when biological model systems are exposed to chemicals. Structurally unrelated compounds may belong to the same class of chemicals because of the similarity in the pharmacological or toxicological endpoints. Based on the premise that classes of chemicals and chemical-specific toxicity can be detected by using gene expression profiles, Hamadeh and coworkers (Hamadeh et al., 2002a,b) used a ‘rat model’ to investigate the hypothesis that treatment with classspecific xenobiotics results in chemical-specific profiles of altered gene expression. Two classes of chemical, an enzyme inducer phenobarbital and three peroxisome proliferators (PPs) (clofibrate-, Wyeth 14,643-, gemfibrozil) were included in their study. Gene expression profiles corresponding to three peroxisome proliferators and phenobarbital were then generated. Cluster analysis, along with other computational approaches, demonstrated close proximity in the gene expression pattern between clofibrate-, Wyeth 14,643- and gemfibrozil-exposed animals, but indicated a distinct partition between these compounds and phenobarbital-exposed animals. Comparison of gene expression profiles of two different animals exposed to the same compound resulted in a relatively high correlation as compared to animals exposed to different compounds belonging to the same class (Amin et al., 2002; Hamadeh et al., 2002a,b,c). This study, in part, suggests that it might be possible to implement this approach for classifying compounds. As stated previously, McMillian and co-workers (McMillian et al., 2004) used the same approach and successfully classified three classes of hepatotoxicants through gene signatures sets. However, these approaches are not yet ready to be utilized as stand-alone tools. Subsequently, Hamadeh and coworkers conducted a blinded study using rat liver samples from chemically treated Sprague–Dawley rats, as they intended to address the question whether it is possible to classify unknown compounds to be similar or dissimilar to either of the two chemical classes that have been previously characterized (Hamadeh et al., 2002a,b). In order to make a prediction on the properties of the blinded samples, they used the gene profile data set (Hamadeh et al., 2002a,b) corresponding to livers from rats exposed to four known compounds (clofibrate-, Wyeth 14,643-, gemfibrozil and phenobarbital) as a training set. Multiple approaches were used to find highly discriminatory or informative genes whose expression patterns could distinguish RNA samples derived from lives exposed to different chemicals. Genetic algorithm/K-nearest neighbor (GA/KNN) and linear discriminant analysis (LDA) were used in revealing single genes or group of genes that could separate known samples based on the class of chemicals involved in the exposure. They listed twenty two highly informative genes that clearly exhibited different patterns of expression between the two pharmacological/toxicological classes of compounds, peroxisome proliferators and enzyme inducers. For example, the tripepidylpeptidase II gene was identified as a highly discriminating gene between peroxisome proliferators and enzyme inducers, based on LDA and G/KNN. They performed set correlation analysis, which compared two sets of multiple variables by pairing each blinded sample with every known sample. Relative higher values of the correlation coefficient (r ) indicated strong correlation and a potential similarity between the compared samples. The results were very encouraging since they successfully made a positive prediction regarding the classes of twelve out of thirteen of the blinded samples. They also noted that ten other blinded samples did


457

not belong to the class of peroxisome proliferators. The results showed that they had a 92.3 % accuracy rate in class prediction. This work illustrates the successful prediction of properties of blinded samples using gene expression profiling. Overall, the results of the studies above indicate that animals treated with the same compound cluster together and compounds from the same xenobiotic class cluster closer when compared to compounds from a different class. This demonstrates that large gene expression profile databases will be able to help classify unknown compounds from exposed tissues. It also highlights the importance of developing analysis models for these types of data. 17.4.3

Predictive Toxicology

The predictive toxicology (Pennie et al., 2000; Fielden and Zacharewski, 2001) approach assumes that a similar treatment leading to the same toxic endpoint will share comparable changes in gene expression. The possibility that a specific group or class of compounds (grouped by toxic endpoint, mechanism, structure, target organ, etc.) may induce characterized gene expression profiles is the basis for the application of toxicogenomics to predictive toxicology. Two early papers demonstrated that gene expression data are useful for classification of hepatotoxins from animals treated with chemical (Bulera et al., 2001; Waring et al., 2001). Burlera and colleagues identified a blinded sample treated with toxicants using clustering analysis. Waring and coworkers demonstrated the strong correlation between classical toxicity endpoints and gene expression changes. Similarly, gene expression changes in the livers were evaluated following exposure of rats to methapyrilene, the rodent hepatic carcinogen, by Hamadeh and coworkers (Hamadeh et al., 2002a,b,c). Data from cDNA microarray analysis corroborated observed histopathological alterations such as hepatocellular necrosis, bile duct hyperplasia, microvesicular vacuolization and portal inflammation in a dose-dependent manner, demonstrating that gene expression analysis was more sensitive than histopathology in detecting early, low-dose hepatotoxicity. Other earlier studies using gene expression measurements corresponding to the in vitro response of rat hepatocytes treated with five known compounds revealed that the profiles of chemicals with similar toxic mechanisms clustered together. Similarly, other studies demonstrated the successful discrimination of gene expression fingerprints from chemically exposed rats (Bartosiewicz et al., 2000, 2001a,b; Harries et al., 2001). All of these papers illustrated the promise of predictive toxicology. Ruepp et al. (2005) constructed a large reference database with liver gene expression profiles from vehicle and compound treated samples. Subsequently, a supervised learning method SVM (support vector machine) was used to generate classification rules. Using models built with gene expression data, toxicity can be predicted at an early phase before classical toxicological changes occur. Based on histopathology and clinical chemistry results in conjunction with published data, individual gene expression profiles of rat livers treated with a multitude of compounds were allocated to the five training classes, controls, direct acting, cholestasis, steatosis and peroxisomal proliferators [Direct acting – bromobenzene, CCl4 , hydrazine, thioacetamide, 1,2-dichlorobenzene, coumarin, acetaminophen; steatosis – amineptine, amiodarone, four proprietary compounds; cholestasis – chlorpromazine, cyclosporin A, glibenclamide, lithocholic acid, methylene dianiline; peroxisomal proliferation – WY-14V643, five proprietary compounds; controls – 163 time-matched vehicle control rats. Each treatment group usually consisted of 5 animals.] This predictive model

458

Hepatotoxicity

was subsequently used to assess compounds with regard to a potential liability of hepatotoxicity. They found that two 5-HT6 receptor antagonists sharing the same pharmacological target displayed different toxicity profiles. The expression profiles of one compound were of no difference from control gene expression profiles; the gene expression profiles of another compound (Rx65) were identified as steatotic. Histopathology confirmed a dose-related increase in changes of fatty acid in the liver observed after repeated dosing with Rx65. Another set of studies was carried out on male rats with three different compounds of the same chemical class of antidiabetic compounds in development (Rx08, Rx09 and Rx10). The results suggest that all three compounds have similar transcriptional effects and have a steatotic liability in the rat liver. Histopathological assessment provided discordant results but in vitro experiments on primary hepatocytes verified the SVMs-based prediction on toxicogenomic data with regard to steatotic liability for all three investigated compounds. Through this predictive model, coumarin, a presumably non–hepatotoxicant, was predicted to be hepatotoxic after 6 and 24 h treatment on rats by the SVMs, which was confirmed by histopathological evaluation. Moreover, the prediction of hepatotoxicity of tacrine is quite interesting, as tacrine is viewed as a human-specific toxicant because it induces elevation of liver transaminase levels in humans with no evidence of liver toxicity in mice, rats or dogs during preclinical development. Using this model, Ruppe and coworkers identified a potential hepatotoxic (cholestasis) response after 6 h treatment from gene expression profiles. However, the pathological changes were not observed at this time point. The increased levels of bile acids and γ -glutamyltransferase (GGT) in serum after 24 h treatment supported the toxicogenomic assessment of a potential cholestatic liability of tacrine, but not exclusively as elevated levels of aspartate aminotransferase (AST) and alanine aminotransferase (ALT) in serum were observed. The toxicogenomic classification detected liver toxicity after 6 h, but failed to do so after 24 h. This indicates that gene expression changes can precede clinical chemistry changes and implies the need for time-course studies for a thorough assessment of toxicogenomics. The model used in Ruepp’s laboratory assigns all transcript profiles to one class. However, in reality, toxicants will often cause mixed toxicities. Another constraint in predictive toxicology is that the available compound database is still limited in size. A further strategy to augment the predictive power of gene expression profiling is to develop microarrays containing fewer genes that are specialized for monitoring gene changes relevant to toxicology. Kier et al. (2004) developed arrays with ‘toxicologically relevant’ genes (‘tox genes’), genes that are known or suspected to be affected by toxicants. The use of microarrays that focus on toxicologically relevant genes can have interpretative advantages and allow practical applications of gene expression evaluation for toxicological evaluations. A database containing gene expression, histopathology and clinical chemistry findings has been generated for 89 compounds using a microarray with rat-specific ‘tox genes’. Gene expression profiles of toxic compounds in this study displayed dose- and time-dependent–response relationships. Gene expression at 24 h was found to correlate well with organ toxicity observed at 72 h. Subsequently, they used the ‘Human-600’ array containing sequences from almost 600 human genes with known responsiveness to toxicity or human homologues of discovered rat ‘tox genes’ to test human hepatocytes samples treated with troglitazone, rosiglitazone and pioglitazone, respectively. Preliminary results demonstrated that the toxic drug, troglitazone, can be clearly distinguished from the less toxic analogues, rosiglitazone and pioglitazone, based on their effect on ‘tox gene’ expression in human hepatocytes.


459

These initial results suggest that evaluation of drugs in human hepatocytes, using toxigenomics as endpoints, may be useful in the prediction of hepatotoxicity. There are some commercial companies, such as ‘GeneLogic ToxExpress’ and ‘CuraGen’, which provide commercially available toxicogenomic databases, aiming to improve the prediction of hepatotoxicants and human-specific toxicants through gene expression profile, histopathology and biochemistry changes. ‘CuraGen’ appears to have identified expression profiles of marker genes on primary rat hepatocytes exposed to hepatotoxicants. This could potentially lead to the development of a predictive toxicogenomics screening system based on a predicted safety profile for the liver (Mansfield and Vincent, 2002). In summary, predictive toxigenomics appears to be a valuable approach to identify compounds with toxic liabilities at early time points. A comprehensive gene expression reference database and robust software for data analysis could play an important role in the interpretation of toxicogenomic data. While these studies have shown great promise, many challenges still remain. Among these challenges are the cost-intensive process of building relevant databases of gene expression profiles of known toxins (van Delft et al., 2005) and the questionable reproducibility of microarray data across different platforms (Hayes and Bradfield, 2005; Hayes et al., 2005a) and the issue of making useful predictions of hepatotoxicity across species or from in vitro systems to in vivo systems. As microarrays only profile gene expression at the mRNA level, this technology alone cannot identify corresponding changes in the level of functional protein. Additionally, protein modifications such as phosphorylation, ubiquitination, oxidation, acetylation, methylation, and hydroxylation, which may be critical for the function of many proteins, cannot be detected directly using microarray analysis. The development and maturation of technology of proteomics may help resolve issues around such differential protein modifications, leading to changes in the activity of gene products. 17.4.4

Identification of New Molecular Biomarkers

Drug-induced hepatotoxicity is a frequent cause of liver injury and biomarkers specific to these changes are useful to monitor their possible occurrence. Various biomarkers already exist; however, there are some limitations of current markers of toxicity. For example, the pathological examination of hepatotoxicity is limited to the sensitivity at an early stage and upon exposure to low doses, clinical chemistry endpoints are typically reflecting secondary effects of toxicity in the late stage and the biochemical or molecular biological endpoints are mostly based on single biomolecules. Toxicogenomics could facilitate the identification and characterization of toxicity through identification of sensitive and specific markers at early stages, and after exposure to low dose levels, and thus could provide the possibility to discover new markers of toxicity. Recent investigation (Heijne et al., 2005b) on bromobenzene-induced acute hepatic centrilobular necrosis showed changes in levels of genes and metabolites were related to the degree of necrosis, potentially providing putative novel markers of hepatotoxicity. This integrated analysis of hepatic toxigenomics and plasma metabolomics was able to more sensitively detect changes related to hepatotoxicity and discover novel markers. Heijne et al. (2005b) identified various genes and proteins that were perturbed in relation to hepatotoxicity upon exposure to bromobenzene. Some effects at the hepatic gene expression level were identified early (6 h) after treatment, when no other toxicity markers were

460

Hepatotoxicity

observed. Many of the bromobenzene-induced effects were seen at a dose level 2.5-fold to 10-fold below the level that induced effects observed using conventional methods. These effects could be early and more sensitive markers of hepatotoxicity. As toxicogenomic measurements display a multitude of effects, it is crucial to define their significance and specificity. Effects in various biological pathways in the cell cannot necessarily be directly associated with toxicity. To investigate the specificity of gene expression changes in liver, Heijne et al. (2005a) compared the gene expression changes induced by bromobenzene with reported responses of model hepatotoxicants, acetaminophen (APAP) and bromobenzene (BB), at high doses. The benzene-induced gene expression changes were related to pathways of biotransformation, glutathione synthesis, fatty acid and cholesterol metabolism, as well as others, which shared considerable overlap with gene expression profiles of BB and APAP (e.g. Ephx1, Nqo1, Gsta, Ugt, Cyp2c12, G6pc). Benzene and BB induced the biotransformation enzymes, Ephx1, Gsta, Afar and aldehyde dehydrogenases. The hepatic response to APAP was less pronounced. The pattern of gene induction appears to suggest that the electrophile response element (EpRE or ARE) could be involved in the transcriptional regulation. However, the use of Nrf2 −/− mice would be needed to confirm this (Shen et al., 2005). The common gene expression changes elicited by benzene, BB and APAP are possibly not a direct cause or result of hepatic necrosis. Thorough validation studies would need to be established regarding specificity, selectivity and accuracy. Several studies have addressed the question as to whether toxigenomics might be used to identify characteristic gene expression profiles induced by carcinogens. While direct genotoxic agents induce carcinogenesis by direct interactions with DNA, gene expression profiles might not necessarily be involved in these mechanisms of action. However, nongenotoxic carcinogens may act through various mechanisms in the initiation and progression of carcinogenesis, and gene expression changes could be involved in these mechanisms of action. Therefore, most toxicogenomics studies analyzed the effects of different classes of non-genotoxic carcinogens. van Delft et al. (van Delft et al., 2004) demonstrated that several supervised clustering methods using gene expression profiles from HepG2 cells treated with chemical carcinogens could potentially discriminate genotoxic from non-genotoxic carcinogens. Another study (Ellinger-Ziegelbauer et al., 2004) analyzed changes of gene expression induced by four model genotoxic carcinogens (dimethylnitrosamine, nitrofluorene, 4-(methylnitrosamino)-1-(3-pyridyl)-1-butanone and aflatoxin B1) in rat livers. Genes and pathways commonly deregulated by these genotoxic carcinogens were identified in these short-term in vivo studies. They have concluded that a set of gene expression markers attributed to the action of genotoxicants was acquired. In summary, toxigenomics could be able to provide many putative molecular-biomarkers molecules in hepatotoxicity. Specific biomarkers could be discovered based on mechanisms. However, so far, only a few examples, with the use of toxicogenomics, have been described. The power of toxicogenomics to discover new biomarkers will be augmented when proteomics and metabolomics are combined with gene expression profiles in the future.

17.5

Summary of Potentials and Pitfalls

Toxicogenomics is a relatively new scientific discipline to study xenobiotic-mediated gene expression changes, especially at higher or subtoxic dosage levels administered in vivo. It


461

can enable a better understanding of mechanisms of toxicities, help the prediction of toxicities of unknown compounds, define mechanism-based biomarkers of toxicity and classify compounds based on gene expression signatures. In addition, it could help to circumvent the limitations of conventional methods in toxicology. The challenges of toxigenomics reside now more in the proper incorporation and implementation of this technology. As with the larger field of gene expression analysis, toxicogenomics faces the problems of probe annotation and data comparison across different array platforms, and correlation with appropriate pathological/toxicological biomarkers, as well as the in vivo pathophysiological relevance. Nevertheless, toxigenomics methods will be of great benefit in toxicology when used in conjunction with the current development of proteomics and metabolomics and histopathological endpoints.

Abbreviations ALT APAP ARE AST BB EpRE ESTs GA/KNN GGT GST LDA MA NASIDs OSs/RMs PP SVMs

alanine aminotransferase acetaminophen antioxidant response element aspartate aminotransferase bromobenzene electrophilic response element expressed sequence tags genetic algorithm/K-nearest neighbor γ -glutamyltransferase glutathione transferase linear discriminant analysis macrophage activator nonsteroidal anti-inflammatory drugs oxidative stressors/reactive metabolites peroxisome proliferators support vector machines

Acknowledgements This work was supported in part by the National Institutes of Health grants R01-CA73674, R01-CA94828, R01-CA92515 and R01-CA118947 (to A.-N. T. Kong). The authors thank Dr McMillian for very helpful discussions. However, the authors are responsible for any errors made in this manuscript.

References Amin, R. P., H. K. Hamadeh, et al. (2002). Genomic interrogation of mechanism(s) underlying cellular responses to toxicants. Toxicology 181–182: 555–563. Bartosiewicz, M., M. Trounstine, et al. (2000). Development of a toxicological gene array and quantitative assessment of this technology. Arch Biochem Biophys 376: 66–73.

462

Hepatotoxicity

Bartosiewicz, M., S. Penn, et al. (2001a). Applications of gene arrays in environmental toxicology: fingerprints of gene regulation associated with cadmium chloride, benzo(a)pyrene and trichloroethylene. Environ Health Perspect 109: 71–74. Bartosiewicz, M. J., D. Jenkins, et al. (2001a). Unique gene expression patterns in liver and kidney associated with exposure to chemical toxicants. J Pharmacol Exp Ther 297: 895–905. Bulera, S. J., S. M. Eddy, et al. (2001). RNA expression in the early characterization of hepatotoxicants in Wistar rats by high-density DNA microarrays. Hepatology 33: 1239–1258. Burchiel, S. W., C. M. Knall, et al. (2001). Analysis of genetic and epigenetic mechanisms of toxicity: potential roles of toxicogenomics and proteomics in toxicology. Toxicol Sci 59: 193–195. Burczynski, M. E., M. McMillian, et al. (2000). Toxicogenomics-based discrimination of toxic mechanism in HepG2 human hepatoma cells. Toxicol Sci 58: 399–415. Butte, A. (2002). The use and analysis of microarray data. Nat Rev Drug Discov 1: 951–960. Chin, K. V. and A. N. Kong (2002). Application of DNA microarrays in pharmacogenomics and toxicogenomics. Pharm Res 19: 1773–1778. Ellinger-Ziegelbauer, H., B. Stuart, et al. (2004). Characteristic expression profiles induced by genotoxic carcinogens in rat liver. Toxicol Sci 77: 19–34. Fielden, M. R. and T. R. Zacharewski (2001). Challenges and limitations of gene expression profiling in mechanistic and predictive toxicology. Toxicol Sci 60: 6–10. Hamadeh, H. K., P. R. Bushel, et al. (2002a). Prediction of compound signature using high density gene expression profiling. Toxicol Sci 67: 232–240. Hamadeh, H. K., P. R. Bushel, et al. (2002b). Gene expression analysis reveals chemical-specific profiles. Toxicol Sci 67: 219–231. Hamadeh, H. K., R. P. Amin, et al. (2002c). An overview of toxicogenomics. Curr Issues Mol Biol 4: 45–56. Harries, H. M., S. T. Fletcher, et al. (2001). The use of genomics technology to investigate gene expression changes in cultured human liver cells. Toxicol In Vitro 15: 399–405. Hayes, K. R. and C. A. Bradfield (2005). Advances in toxicogenomics. Chem Res Toxicol 18: 403–414. Hayes, K. R., A. L. Vollrath, et al. (2005). EDGE: a centralized resource for the comparison, analysis and distribution of toxicogenomic information. Mol Pharmacol 67: 1360–1368. Heijne, W. H., D. Jonker, et al. (2005a). Toxicogenomic analysis of gene expression changes in rat liver after a 28-day oral benzene exposure. Mutat Res 575: 85–101. Heijne, W. H., R. J. Lamers, et al. (2005b). Profiles of metabolites and gene expression in rats with chemically induced hepatic necrosis. Toxicol Pathol 33: 425–433. Khor, T. O., S. Ibrahim and A. N. Kong (2006). Toxicogenomics in Drug Discovery and Drug Development: Potential Applications and Future Challenges. Pharm Res 23: 1659–1664. Kier, L. D., R. Neft, et al. (2004). Applications of microarrays with toxicologically relevant genes (tox genes) for the evaluation of chemical toxicants in Sprague – Dawley rats in vivo and human hepatocytes in vitro. Mutat Res 549: 101–113. Kong, A. N., E. Owuor, et al. (2001). Induction of xenobiotic enzymes by the MAP kinase pathway and the antioxidant or electrophile response element (ARE/EpRE). Drug Metab Rev 33: 255–271. Larkin, J. E., B. C. Frank, et al. (2005). Independence and reproducibility across microarray platforms. Nat Meth 2: 337–344. Lord, P. G., A. Nie, et al. (2006). Application of genomics in preclinical drug safety evaluation. Basic Clin Pharmacol Toxicol 98: 537–546. Maggioli, J., A. Hoover, et al. (2006). Toxicogenomic analysis methods for predictive toxicology. J Pharmacol Toxicol Meth 53: 31–37. Mansfield, T. and M. R. Vincent (2002). CuraGen Corporation. Pharmacogenomics 3: 551–554. McMillian, M., A. Y. Nie, et al. (2004). Inverse gene expression patterns for macrophage activating hepatotoxicants and peroxisome proliferators in rat liver. Biochem Pharmacol 67: 2141– 2165.


463

McMillian, M., A. Nie, et al. (2005). Drug-induced oxidative stress in rat liver from a toxicogenomics perspective. Toxicol Appl Pharmacol 207(Suppl 2): 171–178. Nuwaysir, E. F., M. Bittner, et al. (1999). Microarrays and toxicology: the advent of toxicogenomics. Mol Carcinog 24: 153–159. Pennie, W. D. (2000). Use of cDNA microarrays to probe and understand the toxicological consequences of altered gene expression. Toxicol Lett 112-1-13: 473–477. Pennie, W. D., J. D. Tugwood, et al. (2000). The principles and practice of toxigenomics: applications and opportunities. Toxicol Sci 54: 277–283. Ruepp, S., F. Boess, et al. (2005). Assessment of hepatotoxic liabilities by transcript profiling. Toxicol Appl Pharmacol 207(Suppl 2): 161–170. Shen, G., C. Xu, et al. (2005). Comparison of (−)-epigallocatechin-3-gallate elicited liver and small intestine gene expression profiles between C57BL/6J mice and C57BL/6J/Nrf2 (−/−) mice. Pharm Res 22: 1805–1820. van Delft, J. H., E. van Agen, et al. (2004). Discrimination of genotoxic from non-genotoxic carcinogens by gene expression profiling. Carcinogenesis 25: 1265–1276. van Delft, J. H., E. van Agen, et al. (2005). Comparison of supervised clustering methods to discriminate genotoxic from non-genotoxic carcinogens by gene expression profiling. Mutat Res 575: 17–33. Waring, J. F., R. A. Jolly, et al. (2001). Clustering of hepatotoxins based on mechanism of toxicity using gene expression profiles. Toxicol Appl Pharmacol 175: 28–42.

18 Genomic Profiling of Liver Injury Kevin Gerrish and David E. Malarkey

18.1

Introduction

Functional genomics, also known as transcriptomics, utilizes a wide range of novel highthroughput technologies, including DNA microarray analysis in an attempt to evaluate gene expression patterns and correlate them with specific biological processes. Scientific discovery and translational research in biomedical science have been facilitated by the emergence and advancement of microarray technology along with the availability of standardized commercial microarrays (Knudsen, 2004; Waters and Fostel, 2004; Hoheisel, 2006). In addition, the availability of the complete genome of several organisms has allowed for the application of microarray technology to perform expression analysis with several model organisms (from bacteria to human) (Lettieri, 2006). Therefore, functional genomics technologies provide a powerful tool in understanding the coordinate expression of genes in the normal or diseased state within multiple organisms. Disease conditions are thought to represent a disruption of organ homeostasis through either genetic perturbations and/or xenobiotic exposure (Hood et al., 2004). The advance of genomic technologies has revolutionized the approaches taken towards characterizing the complex molecular mechanisms that drive these disease processes. Microarray-based expression profiling has been used in efforts to analyze tissue responses to xenobiotic exposure as well as the molecular basis of complex diseases such as obesity and cancer (Afshari et al., 1999; Nuwaysir et al., 1999; Bartosiewicz et al., 2001; Bulera et al., 2001; Burchiel et al., 2001; Fielden and Zacharewski, 2001; Smith, 2001; Waring et al., 2001a,b Hamadeh et al., 2002a,b,c; Simmons and Portier, 2002; Tennant, 2002; Ulrich and Friend, 2002; Baranova et al., 2005; Thorgeirsson et al., 2006a,b). Many aspects of the technique make it a useful tool for the assessment of pathobiological states. Microarrays provide the


466

Hepatotoxicity

researcher with the means to assay thousands of genes for differential expression after various treatments under various conditions from a variety of tissue origins (Heller et al., 1997; Alon et al., 1999; Debouck and Goodfellow, 1999; Perou et al., 1999; Kaminski et al., 2000; McCaffrey et al., 2000; Perou et al., 2000; Chuaqui et al., 2002; Gerhold et al., 2002). The availability of species-specific platforms from commercial vendors allows for comparison of expression profiles among multiple species providing a more comprehensive analysis of ‘conserved’ responses (Segal et al., 2004; Kim et al., 2006; Thorgeirsson et al., 2006a; Zender et al., 2006; Lam et al., 2006a,b). Moreover, recent analyses have demonstrated that adequate reproducibility can be achieved across laboratories and platforms (Guo et al., 2006). The primary factors that appear to influence variation are the biological samples and human error, rather than technical diversity (Irizarry et al., 2005; Larkin et al., 2005; Lu et al., 2005; Petersen et al., 2005). Thus, microarray-based expression profiling appears to be well suited in the efforts devoted to discovering the molecular basis of disease processes. Analysis of distinct and coordinate patterns of gene expression provides a powerful tool for understanding the transcriptome in both homeostasis and disease. This information provides mechanistic insight into the pathophysiological processes that cause or drive disease (Heijne et al., 2003; Heinloth et al., 2004; McMillian et al., 2004; Baranova et al., 2005; Chung et al., 2005a,b; Fletcher et al., 2005; Hebbar et al., 2005; Heijne et al., 2005b; Lelliott et al., 2005; Lettieri, 2006; Tugendreich et al., 2006). In addition, these approaches may provide insight into biomarkers that will provide for a rapid and accurate diagnosis of the disease (Hamadeh et al., 2002a,b; Kier et al., 2004; Steiner et al., 2004; Tsai et al., 2005; Fielden and Kolaja, 2006; Maggioli et al., 2006). Furthermore, analyses of these data may allow clinicians to ascertain more precise prognostic indicators and devise new therapeutic interventions (Yeatman, 2003; Mount and Pandey, 2005; Fan and Ren, 2006; Lin et al., 2006; Thorgeirsson et al., 2006b). The liver transcriptome is very complex and widespread adoption of functional genomic technologies has led to the identification of key issues in liver research. These include but are not limited to the roles that diverse cellular interactions and signaling pathways play in regulating liver injury and disease (Shackel et al., 2002a,b; Baranova et al., 2005). We will focus on how the application of an integrative, functional genomic approach can enrich our understanding of the molecular pathogenesis of liver injury. To demonstrate how genomic tools can be useful we will provide an overview of published data and future approaches that will help to provide a deeper understanding of liver disease.

18.2

The Liver and Toxicogenomic Studies

Toxicogenomics combines the tools of traditional toxicology with genomics, bioinformatics, and high throughput experimentation (Waters and Fostel, 2004; Hayes and Bradfield, 2005; Lettieri, 2006). The prevailing notion in performing these experiments is that gene expression data can be used as an early indicator of toxicity because xenobiotic-mediated gene expression changes are often detectable before clinical chemistry, histopathology or clinical observations (Ulrich and Friend, 2002). Many toxicogenomic studies have involved the liver since it is the major source of xenobiotic metabolism and detoxification and much is known about its response to xenobiotic exposures (Bartosiewicz et al., 2001; Bulera et al., 2001; Lu et al., 2001; Thomas et al., 2001; Waring et al., 2001a,b; Donald et al., 2002;

Genomic Profiling of Liver Injury

467

Fountoulakis et al., 2002; Hamadeh et al., 2002a,b,c; Ruepp et al., 2002; Waring et al., 2002; Iida et al., 2003; Kramer et al., 2003; Mortuza et al., 2003; Heinloth et al., 2004; Kramer et al., 2004a; Peterson et al., 2004; Ulrich et al., 2004; Waring et al., 2004; Baranova et al., 2005; Currie et al., 2005; Ellinger-Ziegelbauer et al., 2005; Hebbar et al., 2005; Heijne et al., 2005a; Minami et al., 2005; Tugendreich et al., 2006). In addition, liver injury is one of the leading reasons for withdrawal of new drugs from the market (Lee, 2003; Gunawan and Kaplowitz, 2004; Goldkind and Laine, 2006). Although studies have addressed toxicity in other organs such as brain, heart, kidney and lung (Bartosiewicz et al., 2001; Huang et al., 2001; Hu et al., 2002; Xie et al., 2002; Dam et al., 2003; Mortuza et al., 2003; Amin et al., 2004; Kramer et al., 2004b; Thompson et al., 2004; Wagenaar et al., 2004) the principal focus of this review will be on how genomic approaches are being utilized in the study of liver injury. Included within this discussion will be how variables, including the structural and functional complexity of the liver, circadian rhythms and feeding state of the animals, may impact these studies. We will also describe the approaches taken to identify biomarkers of disease and toxic substance exposure, as well as elucidation of the mechanisms underlying these processes (Waters and Fostel, 2004). Moreover, we will also discuss how new genomics technology can influence hepatology research. The liver is structurally and functionally heterogeneous and is considered to be second only to the brain in transcriptome size (Shackel et al., 2002a). Insight into the morphological structures and functions of the liver are integral to interpretation of genomics and proteomics data that may further our understanding of liver disease. The liver performs many vital functions, including uptake and metabolism of amino acids, carbohydrates, bile acids, cholesterol, proteins, lipids and vitamins subsequent to release into bile and/or the bloodstream (Bloom and Fawcett, 1975; Jones and Spring-Mills, 1983; LaBrecque, 1994; Burt and Day, 2002). In vertebrates, the liver has two distinct blood supplies, regulates blood volume and is the major site of biotransformation and defense against foreign macromolecules and xenobiotics (Malarkey et al., 2005a). Adding to the complexity is that the liver is divided into lobes and lobules that appear to have the same histologic appearance, i.e. multiple units with a central vein surrounded by about 4–6 portal areas (Matsumoto and Kawakami, 1982). However, these functional units also contain up to 15 distinct cell types with different functional properties dependent on their localization within the lobule (Bloom and Fawcett, 1975; Jones and Spring-Mills, 1983; Eustis et al., 1990; Harada et al., 1999; Kogure et al., 1999; Kmiec, 2001; Burt and Day, 2002; MacSween et al., 2002; Malik et al., 2002; Malarkey et al., 2005a). Awareness of these fundamental properties is essential for understanding of liver function and disease processes that may lead to toxicity, cancer and other diseases. Although genomic approaches are prevalent in hepatology research, the reconstruction of the molecular mechanisms that trigger disease processes in the liver is a difficult challenge. Unlike model systems such as yeast, mammalian genomes are more complex and gene expression is influenced by a higher degree of combinatorial regulation and multiple signaling pathways. The liver transcriptome is poorly understood and very complex (includes approximately 25–40 % of the approximately 50,000 mammalian genes) (Shackel et al., 2002a). Many interactions may be context-specific, with different components of the molecular network active in different cellular states or phenotypes. During disease states, the liver transcriptome may double or triple due not only to differential gene expression but also the contributions from the various cell populations in the liver. Moreover, many exogenous

468

Hepatotoxicity

and endogenous factors, including circadian rhythm, age, sex, feeding state and diet, can influence hepatic gene expression (Cao et al., 2001; Akhtar et al., 2002; Endo et al., 2002; Iqbal et al., 2002; Kita et al., 2002; Panda et al., 2002; Takahashi et al., 2002; Boorman et al., 2005a,b; Morgan et al., 2005). These types of challenges can confound biological and pathological analyses of liver diseases. We and others have begun to address these issues and how they may affect study design and data interpretation in toxicogenomic studies.

18.3 18.3.1

Liver Morphology and Toxicogenomic Studies Lobular Structure

Although the liver has a uniform gross appearance there is increasing evidence of functional heterogeneity within and among the individual liver lobes. Mice and rats, the primary species used in liver toxicity studies, have four liver lobes: median, left, right and caudate and all, except the left, are further subdivided into two or more parts (Eustis et al., 1990; Harada et al., 1999; Kogure et al., 1999). The hepatic lobes of the rat appear to have similar vascular systems as that of human liver (Kogure et al., 1999). The portal vein and hepatic artery represent the two main vascular systems that supply blood to the liver. Blood flow to or from the lobe can exhibit individual variations in both rats and humans (Kogure et al., 1999; MacSween et al., 2002). The primary source of blood flow to the liver is different for these vessels and such differences may influence the liver’s response to xenobiotic exposure (Burt and Day, 2002). For example, any localized or generalized variation of blood flow coming through the vascular system may alter delivery of nutrients, toxins and other elements to the liver lobes that can potentially lead to lobe variation in liver disease (Stuart and Wheatley, 1995; MacSween et al., 2002; Thein et al., 2003; Daniel et al., 2004). Examples of lobe variation in liver function include lobe-specific heavy metal accumulation in the liver during liver development and disease (Faa et al., 1987, 1994, 1995; Ambu et al., 1995). The progression of cirrhosis also shows lobe specific differences progressing more rapidly in the right lobe compared to the left (Matsuzaki et al., 1997). In addition, response to carcinogens such as diethyl-nitrosamine also exhibits lobe specificity (Richardson et al., 1986). Perhaps most relevant to this review is that transcriptional profiling has revealed clear lobe variation in liver gene expression in control rats and following acetaminophen exposure (Heinloth et al., 2004; Irwin et al., 2005). Lobe variation in the degree of hepatocellular necrosis has been demonstrated in rats treated with acetaminophen (Malarkey et al., 2005a). Thus, it appears that lobular sampling can have a potential impact on the interpretation of the liver’s response to xenobiotic exposure. 18.3.2

Liver Cell Types

Increasing knowledge about liver cell types and their specific gene expression profiles would help to unravel the complex data sets generated in genomic experiments. This is of particular importance in that the expression of genes in each cellular compartment of the liver are different and these differences must be accounted for to accurately analyze the data. Therefore, in disease states the transcriptome would increase in complexity, not only due to differential gene expression but also to the contributions from the heterogeneous cellular composition of the liver.


469

At least fifteen different cell types are found in the normal liver (Malarkey et al., 2005a). The cell types are divided into two groups, i.e. parenchymal and non-parenchymal cells. Parenchymal cells, better known as hepatocytes, are the most numerous and compromise 60 % of the total cells of the liver whereas non-parenchymal cells, such as sinusoidal endothelial cells (SECs), Kupffer cells (KCs), hepatic stellate cells (HSCs) and biliary epithelium make up a large part of the remaining cell types (Malarkey et al., 2005a). Thus, hepatocytes would contribute a great deal to the transcriptome complexity of the normal and disease states. However, in disease states the cellular makeup of the liver changes and non-parenchymal cells could make greater contributions to the mRNA pool (Kmiec, 2001; Schieferdecker et al., 2001; Malik et al., 2002; Liu et al., 2004b; Breitkopf et al., 2005; Diehl, 2005). For example, the activation of hepatic stellate cells (HSCs) is a key step in the development of liver fibrosis. Several factors are upregulated in activated HSCs, which are thought to contribute to the development of liver fibrosis (Eng and Friedman, 2000; Lee et al., 2004). Further analysis of the differences in gene expression from quiescent and activated HSCs will provide profound insight into the cell activation mechanism. It has been proposed that cultured cell populations or microdissection techniques can be used to investigate the role of cellular heterogeneity in organ transcriptome complexity (Liotta and Petricoin, 2000). Recently, many groups have used a combination of in vivo and in vitro approaches to study the responses of specific liver cell types to xenobiotic exposure, as well as their role in regulating the liver’s response to disease (Harries et al., 2001; Mann and Smart, 2002; Michel et al., 2003; Khetani et al., 2004; Lee et al., 2004; Nonaka et al., 2004; Kofman et al., 2005; Sawada et al., 2005; Walisser et al., 2005; Richert et al., 2006; Tan et al., 2006; Tian et al., 2006). In addition, the use of laser capture microdissection in combination with transcriptional profiling has increased the understanding of molecular events regulating liver diseases such as cholestasis and cirrhosis (Honda et al., 2005a; Baba et al., 2006; Gehring et al., 2006). The combination of these approaches should provide a significant level of insight into regional and cell-specific gene expression profiles within the liver. Obtaining these expression signatures may allow for a more accurate assessment of the disease/injury-specific changes in gene expression. 18.3.3

Functional Gradients

There are considerable gradient differences in cellular and matrix composition throughout the liver that must be accounted for when evaluating genomic data (Germain et al., 1987; Gebhardt, 1992; Teutsch et al., 1999; Burt and Day, 2002; MacSween et al., 2002). The apparent differences in functional properties (e.g. enzyme levels or activities) between periportal and centrilobular cells may help to explain the regional distribution of lesions and susceptibility of cells to certain xenobiotics. Functional gradients of hepatocytes have been intensely studied. Many xenobiotics need to be metabolized to an active form in order to provide a therapeutic or toxic effect. The cytochrome P450 (Cyps) enzymes are the primary drug metabolism enzymes in the liver. Localization studies using immunohistochemical and in situ hybridization analyses have determined that expression of the Cyps (e.g. Cyp2e1) is higher in centrilobular hepatocytes (Jungermann and Katz, 1989). This difference may explain why some hepatotoxins (e.g. acetaminophen) appear to specifically affect the centrilobular region of the liver.

470

Hepatotoxicity

Differences in the levels of oxygen saturation, metabolic activity and glutathione content in periportal and centrilobular hepatocytes have also been reported (Gebhardt, 1992; MacSween et al., 2002). Foreknowledge of these differences would prove useful in downstream analyses of microarray data. For example, periportal hepatocytes have higher levels of urea cycle and fatty acid oxidation activity, suggesting that any change in the expression pattern of genes in these pathways may be indicative of periportal damage in the liver (Jungermann and Katz, 1989). In addition, the number, size and metabolic properties of non-parenchymal cells also appear to be regionally defined (Ratziu and Friedman, 1997; Braet and Wisse, 2002; MacSween et al., 2002; Bykov et al., 2004). For example, HSC properties vary along the periportal, midzonal and centrilobular regions of the liver (Ratziu and Friedman, 1997). In addition to size, there are differences in lipid and vitamin A content as well as differences in extracellular matrix components (Malarkey et al., 2005a). Therefore, any change in the expression of the components within these pathways may indicate not only HSC activation but localization of the cell population as well. 18.3.4

Other Influences

Gene expression profiling can also be affected by experimental conditions, including circadian rhythms, age, gender and feeding state of the animals. It has been proposed that 7–9 % of the transcripts in rodent liver fluctuates with circadian cycling (Akhtar et al., 2002; Kita et al., 2002; Panda et al., 2002; Storch et al., 2002; Boorman et al., 2005a). These include genes that control key metabolic pathways (e.g. glucose metabolism and steroid synthesis) as well as important xenobiotic detoxification enzymes (Gachon et al., 2006). The feeding state of the test subject also appears to have similar influences on the liver transcriptome (Stokkan et al., 2001; Endo et al., 2002; Iqbal et al., 2002; Takahashi et al., 2002; Murray, 2006). Moreover, age and gender have also been shown to affect liver gene expression (Sanz et al., 2002; Mehendale, 2005; Mulas et al., 2005; Sanguino et al., 2005; Boorman and Irwin, personal communication). Study design should take these variables into account since they may influence the response to the test agent which may impact data interpretation.

18.4 18.4.1

Phenotypic Anchoring and Study Design Phenotypic Anchoring

Since the liver is a very heterogeneous tissue, it has been proposed that there is a need for better characterization of the liver tissue from which the mRNA is extracted, a concept now referred to as ‘phenotypic anchoring’. This term refers to the relation of transcript profiles to specific adverse effects defined by conventional measures of tissue damage such as histopathology and/or clinical chemistry. The establishment of this linkage, particularly to histological alterations, has helped to define causal effects of compounds and delineation of disease pathogenesis, including specific cell types that contribute to gene expression profiles (Waring et al., 2001a,b, 2003; Hamadeh et al., 2002c; Huang et al., 2003; Michel et al., 2003; Orphanides, 2003; Minami et al., 2005, 2006; Craig et al., 2006). Documentation of the dose and time of xenobiotic exposure are often insufficient to fully define the response


471

of an individual test subject and so another measure is needed for full interpretation of the data obtained during a toxicogenomics study. Moreover, the phenotype may not properly anchor the molecular profile, because pathological lesions or altered clinical chemistry may be observed both before and after peak response to a compound. Therefore, phenotypic anchoring, as well as documentation of dose and time, can provide information regarding key events in the responses elicited by the xenobiotic. 18.4.2

Study Design

Clearly, there are many factors that can affect the interpretation of transcriptional profiling data from the liver. Although we can eliminate some of the impact of biological variation by designing our studies with appropriate numbers of test and control subjects, we must also consider many of the properties listed in the previous section. The National Center for Toxicogenomics at NIEHS has implemented a standard experimental design to account for and minimize the impact that these factors may have on data interpretation. The careful collection, management and integration of these data, in the context of an experimental protocol, are essential for interpreting the results of toxicogenomics studies. This is of particular importance since it will provide a standard frame of reference through which data can be provided to databases that now serve as general repositories of transcriptional profiling experiments (Edgar et al., 2002; Brazma et al., 2003; Ikeo et al., 2003; Waters et al., 2003; Mattes et al., 2004). Thus, standardization of study design may reduce the difficulty in comparison of transcriptional profiling data across laboratories and with data in the literature. The relative importance of these issues have also been reviewed in more detail by other groups (Waters and Fostel, 2004; Hayes and Bradfield, 2005; Hayes et al., 2005; Morgan et al., 2005; Corvi et al., 2006).

18.5

Mechanistic and Predictive Toxicogenomics

The advent of genomic technology, such as microarray-based expression profiling, has helped to expand the approaches to translational research and scientific discovery. Highthroughput gene expression studies allow for the investigation of thousands of genes at one time, providing a ‘snapshot’ of the transcription state of a diseased tissue. It is important to note that the underlying premise of these experiments is to globally assess the biological mechanisms regulating disease processes. For example, a typical microarray experiment from a tissue such as the liver generates a list of significantly differentially expressed genes for each biological sample. Data from toxicogenomics experiments can be ‘mined’ to determine which genes are differentially expressed in response to a xenobiotic treatment, providing important clues regarding both the mechanism of its toxicity as well as the molecular basis of the liver’s response to the exposure (Lettieri, 2006). It is also possible to identify gene expression changes associated with intrahepatic disease states, such as cirrhosis and non-alcoholic fatty liver disease (Baranova et al., 2005). In addition, a growing number of studies demonstrate that gene expression data are also useful for predictive toxicology studies in which gene expression ‘signatures’ from known xenobiotics (e.g. chemical or drug) are used to predict the toxicological class of unknowns (Maggioli et al., 2006).

472

Hepatotoxicity

Many initial studies have helped to illustrate the practical use of gene expression profiling to investigate the liver’s response to xenobiotic treatment (Bartosiewicz et al., 2001; Bulera et al., 2001; Waring et al., 2001a,b; Hamadeh et al., 2002a,b,c; Kelley-Loughnane et al., 2002; Ruepp et al., 2002). In many of these studies, approaches such as clustering and identification of gene signatures were successfully used to identify responses to classes of compounds. However, it soon became apparent that listing and clustering genes associated with a chemical treatment was far from identifying the biological processes and identifying the causal molecular mechanisms that regulate them (Waring et al., 2001a,b; Hamadeh et al., 2002a,b,c). Since the differentially expressed genes in transcriptional profiles from a chemically challenged or diseased liver represent key components of cellular processes that are associated with toxic (causative genes) and/or adaptive (reactive genes) responses it has become imperative to apply multiple tools to facilitate analyses of the data. The integration of transcriptional profiling, ‘phenotypic anchoring’, gene function annotation, statistical methodologies and informatics has proven to be critical in the attempts to identify robust signatures of disease mechanisms. Therefore, many recent toxicogenomics studies in the liver have utilized a combination of these methods to explore the modular organization and function of networks that regulate its response to xenobiotic exposure (Heijne et al., 2003, 2005a,b; Heinloth et al., 2004; Walsh et al., 2003; McMillian et al., 2004, 2005; Chung et al., 2005a,b; Fletcher et al., 2005; Hebbar et al., 2005; Lelliott et al., 2005; Tugendreich et al., 2006). In addition, these methods have been applied to studies of liver diseases, including hepatitis, cirrhosis and non-alcoholic steatohepatitis (Boutin et al., 2004; Campbell et al., 2004; Baranova et al., 2005; Honda et al., 2005a,b; Baba et al., 2006; Gehring et al., 2006). Ultimately, a combination of proteomics, metabonomics and genomics, otherwise known as a ‘systems-biology’ approach, should advance our understanding of the molecular, cellular and physiological events that regulate the development and progression of liver disease, allowing researchers to relate mechanistic cellular events to clinically relevant endpoints. Preliminary studies, as well as a proposed conceptual framework for other studies, have been reported recently (Craig et al., 2006). Ideally, standardized analysis methods could be uniformly applied to multiple-data sources from different xenobiotic/disease classes, thereby uncovering commonalities and differences among and within toxicogenomic studies. A growing number of studies demonstrate that gene expression data may be useful for predictive toxicology (Thomas et al., 2001; Hamadeh et al., 2002a,b; Kier et al., 2004; Steiner et al., 2004; Tsai et al., 2005; Maggioli et al., 2006). Most of the published studies have used class-prediction methods in which gene expression signatures from known toxins are used to predict the toxicological class of an unknown xenobiotic. A common feature of the published methods is the use of class comparison and class discovery steps. In the class comparison step, data are prepared and analyzed to define a set of genes that allow for differentiation among the toxicological classes represented in the data set. These genes are often referred to as the discriminatory or training gene set. In the class discovery step, the similarity among treatments in the training set is visualized by using techniques like clustering that groups treatments only on the similarities in their gene signatures. The methods are evaluated as a predictive tool by estimating the success rates in predicting the toxicological class of unknowns. These methods seem to hold promise for increasing the rate at which compounds can be evaluated for toxicity, reducing the length and costs of toxicological studies. However, the evolution of microarray-based expression profiling as a predictive tool in which the knowledge of


473

the biological effects of disease in one species is used to predict the effects in another species requires the results of multiple studies to be incorporated into a multi-domain, multi-genome knowledgebase. Ideally, this knowledgebase would be searchable by multiple, key terms (e.g. gene expression signature) to find results that are analogous to those observed with the experimental unknown (Waters and Fostel, 2004; Baranova et al., 2005; Hayes and Bradfield, 2005; Hayes et al., 2005; Fielden and Kolaja, 2006; Lettieri, 2006). This will allow the field of predictive toxicology to evolve using knowledge that is systemically generated through database and literature ‘mining’ of gene expression datasets over time. A continual refinement in the application and evaluation of these approaches will undoubtedly provide in-depth information for use in the public health and risk assessment fields. Thus, in addition to exploring mechanistic insights, these efforts may significantly aid the analytical search for predictive biomarkers of liver disease.

18.6

Applications in Toxicogenomics of Hepatotoxicant Injury

In the past five years, gene transcription studies of hundreds of hepatotoxicants (Table 18.1) have led to the accumulation of vast amounts of data. For example, the Environment, Drugs and Gene Expression (EDGE) database resource, maintained at the University of Wisconsin, has toxicogenomic data for over 20 different chemicals with over 100 different doses and time points for analysis (http://edge.oncology.wisc.edu/edge.php). The literature is also replete with similar data. The main efforts are focused on identifying transcription profiling of compound classes: toxin-specific biologic pathways, gene expression biomarkers and/or new genes of pathways involved in carcinogenesis. Most studies are conducted

Table 18.1 Partial list of chemicals from the literature used in the generation of toxicogenomics data in rat liver Acetaminophen Carbon tetrachloride Chloroform Methapyrilene Peroxisome proliferators Clofibrate Clofibric acid WY 14 643 Diethylhexylphthalate Diisonylphthalate Perfluoroheptanoioc acid Perfluorooctanooic acid Gemfibrozil Troglitizone Rosiglitizone pioglitizone Phenobarbital Phenytoin Endotoxin/Lipopolysaccharide

NSAIDS Aspirin Diflunisal Fenbufen Ibuprofen Allyl alcohol Bromobenzene Conazole fungicides Triadimefon Propiconazole Myclobutanil Cyclohexamide Thioacetamide Dieldrin D-galactosamine Dimethylnitrosamine Doxorubicin Coumarin Benzene

474

Hepatotoxicity Table 18.2 Functional categories of rat hepatoxicants commonly identified in literature analyses DNA repair Acute phase response Transcription Cytoskeleton Tissue repair/regeneration DNA damage Cell cycle Translation Oxidative stress

Apoptosis Cell proliferation Inflammation Biotransformation/drug metabolism Metabolism (glucose/glycogen, protein, lipids, carbohydrate, steroids) Glutathione synthesis Signal transduction Energy loss (ATP)

in rats (Fisher 344, Spraque-Dawley, and Wistar) with fewer studies in the mouse fish (Tilton et al., 2006) and other species. It is not surprising that hundreds to thousands of hepatic genes and proteins are up- or down-regulated after hepatotoxicant injury considering the liver has thousands of vital functions and 25–45 % of all ∼ 40 000 genes are expressed normally in the liver (Shackel et al., 2002a; Malarkey et al., 2005a). The vital functions include uptake and metabolism of amino acids, carbohydrates, bile acids, cholesterol, proteins, lipids and vitamins (Malarkey et al., 2005a). The liver also contains at least a dozen different cell types and is the major site for biotransformation and defense against foreign macromolecules and xenobiotics. Certain biological pathways and functional categories, indicated by gene transcription regulation, are commonly altered by hepatotoxicants whether the mechanism(s) of action is/are dissimilar or not. Frequently, altered pathways include acute phase response, oxidative stress, apoptosis, cell proliferation, metabolism (i.e. glucose, glycogen, protein, lipid and carbohydrate), inflammation and repair among many others (Table 18.2). Most are non-specific toxic responses of liver injury and repair that offer no new mechanistic information but do confirm validity of toxicogenomics assays. The key goal of toxicogenomics is to find the specific changes of injury, preferably before irreparable damage, and predict outcomes for agents with unknown in vivo toxicity/carcinogenicity. Many genomics studies performed with ‘model’ hepatotoxicants, such as acetominophen (APAP), phenobarbital, carbon tetrachloride and various peroxisome proliferators, have identified the strengths and weaknesses of genomics and proteomics technologies. For example, gene expression patterns in rats treated with phenobarbital reflect the causally related serum chemistry findings of impaired glycolysis and stimulated lipolysis and cholesterogenesis (Kiyosawa et al., 2004b). Most of the hepatotoxicants target hepatocytes but some may primarily or simultaneously target hepatic macrophages (Kupffer cells), endothelium (sinusoidal) or intestinal epithelium leading to endotoxin (lipopolysaccharide or LPS) release into the portal blood and indirect activation of Kupffer cells. Most studies are conducted with acute duration, exposures of multiple doses and sampling of liver at various times up to 72 h or more. In general, altered gene expression levels precede histological or clinical chemistry alterations and can be detected at lower doses. There are often increased numbers of genes differentially expressed as the damage increases with time and dose and during resolution of hepatic injury distinct pathways (i.e. regeneration) are upor down-regulated with an overall decrease in differentially expressed genes. For most


475

Table 18.3 Selected differentially expressed genes found in mouse and human liver cancer by microarray analysisa,b Mousec

Humand

IGF-1 and 2 IGFBP1 Intestinal trefoil factor 3 CD63 Osteopontin MIG Cathepsins E, S and L Lipocalin 2 CRG-L1 Major urinary protein

IGF-1 and 2 — Intestinal trefoil factor 3 CD63 Osteopontin MIG Cathepsin L Lipocalin 2 —

c-fos H19 Ly-6D Cyp4502c29 Cyp450a1 DNA primase GST m AFP

— — — — — — GST AFT

p53 Syndecan-1 GADD45 c-myc cyclin D1 Bad

P53 — — c-myc — — TGF-α

a

Genes were selected based on having a potential role in the carcinogenic process. Abbreviations: GST, glutathione-S-transferase; AFP, alpha fetoprotein; IGFBP1, insulin-like growth factor binding protein 1; IGF, insulin-like growth factor; MIG, monokine induced by γ interferon; CRG-L1, cancer-related gene liver 1; Ig, immunoglobulin. c Graveel et al., 2001; 2003; Meyer et al., 2003; Liu et al., 2004a; Malarkey et al., 2005b. d Thorgeirsson and Grisham, 2002; Suriawinata and Xu, 2004; Lee et al., 2005; Thorgeirsson et al., 2006a. b

hepatotoxicants there are many pathways in common, many of which are non-specific (Table 18.2) and specific toxicities are beginning to be distinguished from coincidentally expressed genes. Gene expression of liver neoplasms in mouse and humans have revealed genes in common and identified new biological insights and potential therapeutic targets (see reviews – Thorgeirsson and Grisham, 2002; Suriawinata and Xu, 2004; Malarkey et al., 2005b; Thorgeirsson et al., 2006a) (Table 18.3). There has been some success at identifying classes of liver toxicants by their gene expression profiles. McMillian et al. (2005) successfully classified about 100 hepatotoxicants into categories of macrophage activators (MAs), peroxisome proliferators (PPs) or oxidative stressors/reactive metabolites (OSs/RMs). All of the compounds produce oxidative stress responses but the three classes could be distinguished by upregulation of specific transcription factors (Stat3 and NFkB for MAs, PPAR alpha for PP and Nrf2 for OS/RM)

476

Hepatotoxicity

and repression/induction of certain cytochrome p450s (CYP2C23 for MA and OS/RM and CYP4A1 for PP). Some MAs, such as NSAIDs, flufenamic acid, diclofenac and flurbiprofen, are believed to indirectly activate liver macrophages after gut damage, leading to systemic release of bacterial endotoxin (LPS) into portal circulation. Gene-expression profiling successfully distinguished two conazole fungicides that are rodent hepatocarcinogens, triadimefon and propiconazole, from the non-carcinogen, myclobutanil (Ward, et al., 2006) (Hester et al., 2006). Profiles were distinguished by pathways including those involved with apoptosis, cell cycle, calcium signaling, cholesterol biosynthesis and oxidative stress. The number of altered metabolism, signaling and growth pathways generally increased with time and dose, and were highest with propiconazole. The findings were consistent with the findings of hepatomegaly, increased cell proliferation and increase of cytochrome p450s. Tugendreich et al. (2006) recently reported specific profile clustering of 55 compounds comprising 4 major drug classes with over three hundred compound–dose–time combinations. The classes were NSAIDs, HMG-CoA reductase inhibitors, PPARs and sex steroids. The NSAIDs induced a pattern similar to LPS control, including acute phase response, that indicated a possible role of gastrointestinal damage and release of bacterial endotoxin (LPS). The acute phase response occurred at 6 h post-treatment, before any clinical pathology alterations occurred. Transcriptome profiles for hepatic glutathione deficiency have also been described (Kiyosawa et al., 2004a, 2006). Acetaminophen and carbon tetrachloride are model compounds in which hepatotoxicosis is preceded by glutathione depletion in the rat. Glutathione acts to detoxify oxidative electrophilic molecules and free radicals and thus protects cells from oxidative stress. Sixty nine genes correlating with glutathione deficiency were identified in a glutathione-deficient rat model that served to distinguish the glutathione deficiency preceding toxicosis with APAP and not phenobarbital. Unexpectedly, clofibrate, known to deplete glutathione, did not affect the expression of the identified glutathione-deficiency genes. CCl4 is known to be metabolized to the highly reactive trichloromethyl free radical that causes lipid peroxidation and reacts with cellular proteins and DNA. Secondarily, inflammatory processes are thought to be caused by activated Kupffer cells. It is not surprising that metabolism is downregulated, DNA repair and protein damage pathways are elevated and glutamine synthetase (reflecting centrilobular necrosis) is altered (Fountoulakis et al., 2002). Minami et al. (2005) identified 17 toxicity markers not previously described in a study involving rats treated with APAP, bromobenzene, carbon tetrachloride, DMNA and thioacetamide. Peroxisome proliferators (PPs), some of which are rodent carcinogens, tend to affect pathways of cell proliferation, oxidative stress and suppression of apoptosis, pathways proposed to be involved in non-genotoxic carcinogenesis (Kramer et al., 2003). Most PPs also have marked effects on lipid, triglyceride and carbohydrate metabolism, beta-oxidation, and cytochrome p450 induction (Kramer et al., 2003; Michel et al., 2003; Yadetie et al., 2003). The human and rodent hepatotoxicant, troglitizone, caused numerous gene changes while the related PPs which are not hepatotoxic, rosiglitazone and pioglitazone, had few gene changes (Kier et al., 2004). Gene alterations following hepatotoxicant exposure most likely reflect the reparative responses of the liver rather than resulting in cell damage. APAP causes cell energy (ATP)


477

loss, mitochondrial damage and induction of reparative pathways, indicating the toxicity is a direct cell injury rather than a result of gene regulation (Ruepp et al., 2002; Heinloth et al., 2004). In a study using necrogenic doses of APAP, turpentine oil and d-galactosamine, there was a preferentially high expression of genes for liver damage repair with concomitant decrease in functional genes (Tygstrup et al., 2002). The liver appears to be intent to repair first, at all costs, to minimize loss of functions.

18.7

Conclusions

The field of toxicogenomics has progressed rapidly over the past seven years and appears to have validated the concept of gene-expression profiles as signatures of toxicant classes, disease subtypes or other biological endpoints. In particular, these studies have expanded our understanding of the molecular mechanisms of liver disease (Waters and Fostel, 2004; Baranova et al., 2005; Lettieri, 2006). Although microarray-based expression profiling appears to be well suited for exploring the molecular basis of disease processes in the liver we still face the major challenge of identifying the correct context and functional importance of the different events and mechanisms. Therefore, we must continue to consider these data within clear biochemical or physiological frameworks to allow independent experimental confirmation of their relevance to the pathogenesis of liver diseases.

Acknowledgements The authors thank Drs Bhanu Singh and Gary Boorman for their critical review of this chapter and insightful feedback.

References Afshari, C. A., Nuwaysir, E. F. and Barrett, J. C. (1999). Application of complementary DNA microarray technology to carcinogen identification, toxicology, and drug safety evaluation. Cancer Res 59: 4759–4760. Akhtar, R. A., Reddy, A. B., Maywood, E. S., Clayton, J. D., King, V. M., Smith, A. G., Gant, T. W., Hastings, M. H. and Kyriacou, C. P. (2002). Circadian cycling of the mouse liver transcriptome, as revealed by cDNA microarray, is driven by the suprachiasmatic nucleus. Curr Biol 12: 540– 550. Alon, U., Barkai, N., Notterman, D. A., Gish, K., Ybarra, S., Mack, D. and Levine, A. J. (1999). Broad patterns of gene expression revealed by clustering analysis of tumor and normal colon tissues probed by oligonucleotide arrays. Proc Natl Acad Sci USA 96: 6745–6750. Ambu, R., Crisponi, G., Sciot, R., Van Eyken, P., Parodo, G., Iannelli, S., Marongiu, F., Silvagni, R., Nurchi, V., Costa, V., Faa, G. and Desmet, V.J. (1995). Uneven hepatic iron and phosphorus distribution in beta-thalassemia. J Hepatol 23: 544–549. Amin, R. P., Vickers, A. E., Sistare, F., Thompson, K. L., Roman, R. J., Lawton, M., Kramer, J., Hamadeh, H. K., Collins, J., Grissom, S., Bennett, L., Tucker, C. J., Wild, S., Kind, C., Oreffo, V., Davis, J. W., 2nd, Curtiss, S., Naciff, J. M., Cunningham, M., Tennant, R., Stevens, J., Car, B.,

478

Hepatotoxicity

Bertram, T. A. and Afshari, C. A. (2004). Identification of putative gene based markers of renal toxicity. Environ Health Perspect 112: 465–479. Baba, N., Kobashi, H., Yamamoto, K., Terada, R., Suzuki, T., Hakoda, T., Okano, N., Shimada, N., Fujioka, S., Iwasaki, Y. and Shiratori, Y. (2006). Gene expression profiling in biliary epithelial cells of primary biliary cirrhosis using laser capture microdissection and cDNA microarray. Transl Res 148: 103–113. Baranova, A., Schlauch, K., Gowder, S., Collantes, R., Chandhoke, V. and Younossi, Z. M. (2005). Microarray technology in the study of obesity and non-alcoholic fatty liver disease. Liver Int 25: 1091–1096. Bartosiewicz, M. J., Jenkins, D., Penn, S., Emery, J. and Buckpitt, A. (2001). Unique gene expression patterns in liver and kidney associated with exposure to chemical toxicants. J Pharmacol Exp Ther 297: 895–905. Bloom, W. and Fawcett, D.W. (1975). Liver and gall bladder, In A Textbook of Histology, W. Bloom and D. W. Fawcett (Eds), W B Saunders, Philadelphia, PA, USA, pp. 688-718. Boorman, G. A., Blackshear, P. E., Parker, J. S., Lobenhofer, E. K., Malarkey, D. E., Vallant, M. K., Gerken, D. K. and Irwin, R. D. (2005a). Hepatic gene expression changes throughout the day in the Fischer rat: implications for toxicogenomic experiments. Toxicol Sci 86: 185– 193. Boorman, G. A., Irwin, R. D., Vallant, M. K., Gerken, D. K., Lobenhofer, E. K., Hejtmancik, M. R., Hurban, P., Brys, A. M., Travlos, G. S., Parker, J. S. and Portier, C. J. (2005b). Variation in the hepatic gene expression in individual male Fischer rats. Toxicol Pathol 33: 102–110. Boutin, S. R., Rogers, A. B., Shen, Z., Fry, R. C., Love, J. A., Nambiar, P. R., Suerbaum, S., and Fox, J. G. (2004). Hepatic temporal gene expression profiling in Helicobacter hepaticus-infected A/JCr mice. Toxicol Pathol 32: 678–693. Braet, F. and Wisse, E. (2002). Structural and functional aspects of liver sinusoidal endothelial cell fenestrae: a review. Comp Hepatol 1: 1. Brazma, A., Parkinson, H., Sarkans, U., Shojatalab, M., Vilo, J., Abeygunawardena, N., Holloway, E., Kapushesky, M., Kemmeren, P., Lara, G. G., Oezcimen, A., Rocca-Serra, P., and Sansone, S. A. (2003). ArrayExpress – a public repository for microarray gene expression data at the EBI. Nucleic Acids Res 31: 68–71. Breitkopf, K., Sawitza, I., Westhoff, J. H., Wickert, L., Dooley, S. and Gressner, A. M. (2005). Thrombospondin 1 acts as a strong promoter of transforming growth factor beta effects via two distinct mechanisms in hepatic stellate cells. Gut 54: 673–681. Bulera, S. J., Eddy, S. M., Ferguson, E., Jatkoe, T. A., Reindel, J. F., Bleavins, M. R. and De La Iglesia, F. A. (2001). RNA expression in the early characterization of hepatotoxicants in Wistar rats by high-density DNA microarrays. Hepatology 33: 1239–1258. Burchiel, S. W., Knall, C. M., Davis, J. W. II, Paules, R. S., Boggs, S. E. and Afshari, C. A. (2001). Analysis of genetic and epigenetic mechanisms of toxicity: potential roles of toxicogenomics and proteomics in toxicology. Toxicol Sci 59: 193–195. Burt, A. D. and Day, C.P. (2002). Pathophysiology of the liver, in Pathology of the Liver, A. D. Burt R. N. M. MacSween, B. C. Portmann, K. G. Ishak, P. J. Scheuer and P. P. Anthony (Eds), Churchill Livingstone, New York, NY, USA, pp. 67–105. Bykov, I., Ylipaasto, P., Eerola, L. and Lindros, K. O. (2004). Functional Differences between Periportal and Perivenous Kupffer Cells Isolated by Digitonin-Collagenase Perfusion. Comp Hepatol 3 (Suppl 1): S34. Campbell, K. M., Sabla, G. E. and Bezerra, J. A. (2004). Transcriptional reprogramming in murine liver defines the physiologic consequences of biliary obstruction. J Hepatol 40: 14–23. Cao, S. X., Dhahbi, J. M., Mote, P. L. and Spindler, S. R. (2001). Genomic profiling of shortand long-term caloric restriction effects in the liver of aging mice. Proc Natl Acad Sci USA 98: 10630–10635.


479

Chuaqui, R. F., Bonner, R. F., Best, C. J., Gillespie, J. W., Flaig, M. J., Hewitt, S. M., Phillips, J. L., Krizman, D. B., Tangrea, M. A., Ahram, M., Linehan, W. M., Knezevic, V. and Emmert-Buck, M. R. (2002). Post-analysis follow-up and validation of microarray experiments. Nat Genet 32 (Suppl) 509–514. Chung, H., Hong, D. P., Jung, J. Y., Kim, H. J., Jang, K. S., Sheen, Y. Y., Ahn, J. I., Lee, Y. S. and Kong, G. (2005a). Comprehensive analysis of differential gene expression profiles on carbon tetrachloride-induced rat liver injury and regeneration. Toxicol Appl Pharmacol 206: 27–42. Chung, H., Hong, D. P., Kim, H. J., Jang, K. S., Shin, D. M., Ahn, J. I., Lee, Y. S. and Kong, G. (2005b). Differential gene expression profiles in the steatosis/fibrosis model of rat liver by chronic administration of carbon tetrachloride. Toxicol Appl Pharmacol 208: 242–254. Corvi, R., Ahr, H. J., Albertini, S., Blakey, D. H., Clerici, L., Coecke, S., Douglas, G. R., Gribaldo, L., Groten, J. P., Haase, B., Hamernik, K., Hartung, T., Inoue, T., Indans, I., Maurici, D., Orphanides, G., Rembges, D., Sansone, S. A., Snape, J. R., Toda, E., Tong, W., van Delft, J. H., Weis, B. and Schechtman, L. M. (2006). Meeting report: Validation of toxicogenomics-based test systems: ECVAM-ICCVAM/NICEATM considerations for regulatory use. Environ Health Perspect 114: 420–429. Craig, A., Sidaway, J., Holmes, E., Orton, T., Jackson, D., Rowlinson, R., Nickson, J., Tonge, R., Wilson, I. and Nicholson, J. (2006). Systems toxicology: integrated genomic, proteomic and metabonomic analysis of methapyrilene induced hepatotoxicity in the rat. J Proteome Res 5: 1586–1601. Currie, R. A., Bombail, V., Oliver, J. D., Moore, D. J., Lim, F. L., Gwilliam, V., Kimber, I., Chipman, K., Moggs, J. G. and Orphanides, G. (2005). Gene ontology mapping as an unbiased method for identifying molecular pathways and processes affected by toxicant exposure: application to acute effects caused by the rodent non-genotoxic carcinogen diethylhexylphthalate. Toxicol Sci 86: 453–469. Dam, K., Seidler, F. J. and Slotkin, T. A. (2003). Transcriptional biomarkers distinguish between vulnerable periods for developmental neurotoxicity of chlorpyrifos: Implications for toxicogenomics. Brain Res Bull 59: 261–265. Daniel, G. B., DeNovo, R. C., Sharp, D. S., Tobias, K. and Berry, C. (2004). Portal streamlining as a cause of nonuniform hepatic distribution of sodium pertechnetate during per-rectal portal scintigraphy in the dog. Vet Radiol Ultrasound 45: 78–84. Debouck, C. and Goodfellow, P. N. (1999). DNA microarrays in drug discovery and development. Nat Genet 21: 48–50. Diehl, A. M. (2005). Lessons from animal models of NASH. Hepatol Res 33: 138–144. Donald, S., Verschoyle, R. D., Edwards, R., Judah, D. J., Davies, R., Riley, J., Dinsdale, D., Lopez Lazaro, L., Smith, A. G., Gant, T. W., Greaves, P. and Gescher, A. J. (2002). Hepatobiliary damage and changes in hepatic gene expression caused by the antitumor drug ecteinascidin-743 (ET-743) in the female rat. Cancer Res 62: 4256–4262. Edgar, R., Domrachev, M. and Lash, A. E. (2002). Gene Expression Omnibus: NCBI gene expression and hybridization array data repository. Nucleic Acids Res 30: 207–210. Ellinger-Ziegelbauer, H., Stuart, B., Wahle, B., Bomann, W. and Ahr, H. J. (2005). Comparison of the expression profiles induced by genotoxic and nongenotoxic carcinogens in rat liver. Mutat Res 575: 61–84. Endo, Y., Fu, Z., Abe, K., Arai, S. and Kato, H. (2002). Dietary protein quantity and quality affect rat hepatic gene expression. J Nutr 132: 3632–3637. Eng, F. J. and Friedman, S. L. (2000). Fibrogenesis I. New insights into hepatic stellate cell activation: the simple becomes complex. Am J Physiol Gastrointest Liver Physiol 279: G7–G11. Eustis, S. L., Boorman, G.A., Harada, T. and Popp, J.A. (1990). Liver, in Pathology of the Fischer Rat, S. L. Eustis, G.A. Boorman, M.R. Elwell, C.A. Montgomery and W.F. Mackenzie (Eds), Academic Press, San Diego, CA, USA, pp. 71–74.

480

Hepatotoxicity

Faa, G., Liguori, C., Columbano, A. and Diaz, G. (1987). Uneven copper distribution in the human newborn liver. Hepatology 7: 838–842. Faa, G., Sciot, R., Farci, A. M., Callea, F., Ambu, R., Congiu, T., van Eyken, P., Cappai, G., Marras, A., Costa, V. and Desmet, V.J. (1994). Iron concentration and distribution in the newborn liver. Liver 14: 193–199. Faa, G., Nurchi, V., Demelia, L., Ambu, R., Parodo, G., Congiu, T., Sciot, R., Van Eyken, P., Silvagni, R. and Crisponi, G. (1995). Uneven hepatic copper distribution in Wilson’s disease. J Hepatol 22: 303–308. Fan, J. and Ren, Y. (2006). Statistical Analysis of DNA Microarray Data in Cancer Research. Clin Cancer Res 12: 4469–4473. Fielden, M. R. and Kolaja, K. L. (2006). The state-of-the-art in predictive toxicogenomics. Curr Opin Drug Discov Devel 9: 84–91. Fielden, M. R. and Zacharewski, T. R. (2001). Challenges and limitations of gene expression profiling in mechanistic and predictive toxicology. Toxicol Sci 60: 6–10. Fletcher, N., Wahlstrom, D., Lundberg, R., Nilsson, C. B., Nilsson, K. C., Stockling, K., Hellmold, H. and Hakansson, H. (2005). 2,3,7,8-Tetrachlorodibenzo- p-dioxin (TCDD) alters the mRNA expression of critical genes associated with cholesterol metabolism, bile acid biosynthesis and bile transport in rat liver: a microarray study. Toxicol Appl Pharmacol 207: 1–24. Fountoulakis, M., de Vera, M. C., Crameri, F., Boess, F., Gasser, R., Albertini, S. and Suter, L. (2002). Modulation of gene and protein expression by carbon tetrachloride in the rat liver. Toxicol Appl Pharmacol 183: 71–80. Gachon, F., Olela, F. F., Schaad, O., Descombes, P. and Schibler, U. (2006). The circadian PARdomain basic leucine zipper transcription factors DBP, TEF and HLF modulate basal and inducible xenobiotic detoxification. Cell Metab 4: 25–36. Gebhardt, R. (1992). Metabolic zonation of the liver: regulation and implications for liver function. Pharmacol Ther 53: 275–354. Gehring, S., Dickson, E. M., San Martin, M. E., van Rooijen, N., Papa, E. F., Harty, M. W., Tracy, T. F., Jr. and Gregory, S. H. (2006). Kupffer cells abrogate cholestatic liver injury in mice. Gastroenterology 130: 810–822. Gerhold, D. L., Jensen, R. V. and Gullans, S. R. (2002). Better therapeutics through microarrays. Nat Genet 32 (Suppl) 547–551. Germain, A., Garrido, A., Canas, P., Llanos, A. and Valenzuela, A. (1987). Differences in the lipid peroxidative status, cytochrome P-450 content and microsomal oxygen consumption between right and left lobes of the liver in fetal sheep. Comparison with maternal liver. Biochem Int 15: 571–577. Goldkind, L. and Laine, L. (2006). A systematic review of NSAIDs withdrawn from the market due to hepatotoxicity: lessons learned from the bromfenac experience. Pharmacoepidemiol Drug Saf 15: 213–220. Graveel, C. R., Jatkoe, T., Madore, S. J., Holt, A. L. and Farnham, P. J. (2001). Expression profiling and identification of novel genes in hepatocellular carcinomas. Oncogene 20: 2704–2712. Graveel, C. R., Harkins-Perry, S. R., Acevedo, L. G. and Farnham, P. J. (2003). Identification and characterization of CRG-L2, a new marker for liver tumor development. Oncogene 22: 1730–1736. Gunawan, B. and Kaplowitz, N. (2004). Clinical perspectives on xenobiotic-induced hepatotoxicity. Drug Metab Rev 36: 301–312. Guo, L., Lobenhofer, E. K., Wang, C., Shippy, R., Harris, S. C., Zhang, L., Mei, N., Chen, T., Herman, D., Goodsaid, F. M., Hurban, P., Phillips, K. L., Xu, J., Deng, X., Sun, Y. A., Tong, W., Dragan, Y. P. and Shi, L. (2006). Rat toxicogenomic study reveals analytical consistency across microarray platforms. Nat Biotechnol 24: 1162–1169. Hamadeh, H. K., Bushel, P. R., Jayadev, S., DiSorbo, O., Bennett, L., Li, L., Tennant, R., Stoll, R., Barrett, J. C., Paules, R. S., Blanchard, K. and Afshari, C. A. (2002a). Prediction of compound signature using high density gene expression profiling. Toxicol Sci 67: 232–240.


481

Hamadeh, H. K., Bushel, P. R., Jayadev, S., Martin, K., DiSorbo, O., Sieber, S., Bennett, L., Tennant, R., Stoll, R., Barrett, J. C., Blanchard, K., Paules, R. S. and Afshari, C. A. (2002b). Gene expression analysis reveals chemical-specific profiles. Toxicol Sci 67: 219–231. Hamadeh, H. K., Knight, B. L., Haugen, A. C., Sieber, S., Amin, R. P., Bushel, P. R., Stoll, R., Blanchard, K., Jayadev, S., Tennant, R. W., Cunningham, M. L., Afshari, C. A. and Paules, R. S. (2002c). Methapyrilene toxicity: anchorage of pathologic observations to gene expression alterations. Toxicol Pathol 30: 470–482. Harada, T., Enomoto, A., Boorman, G.A. and Maronpot, R.R. (1999). Liver and gallbladder, in Pathology of the mouse, R. R. Maronpot (Ed.), Cache River Press, Vienna, Austria, pp. 119–126. Harries, H. M., Fletcher, S. T., Duggan, C. M. and Baker, V. A. (2001). The use of genomics technology to investigate gene expression changes in cultured human liver cells. Toxicol In Vitro 15: 399–405. Hayes, K. R. and Bradfield, C. A. (2005). Advances in toxicogenomics. Chem Res Toxicol 18: 403– 414. Hayes, K. R., Vollrath, A. L., Zastrow, G. M., McMillan, B. J., Craven, M., Jovanovich, S., Rank, D. R., Penn, S., Walisser, J. A., Reddy, J. K., Thomas, R. S. and Bradfield, C. A. (2005). EDGE: a centralized resource for the comparison, analysis, and distribution of toxicogenomic information. Mol Pharmacol 67: 1360–1368. Hebbar, V., Shen, G., Hu, R., Kim, B. R., Chen, C., Korytko, P. J., Crowell, J. A., Levine, B. S. and Kong, A. N. (2005). Toxicogenomics of resveratrol in rat liver. Life Sci 76: 2299–2314. Heijne, W. H., Stierum, R. H., Slijper, M., van Bladeren, P. J. and van Ommen, B. (2003). Toxicogenomics of bromobenzene hepatotoxicity: a combined transcriptomics and proteomics approach. Biochem Pharmacol 65: 857–875. Heijne, W. H., Jonker, D., Stierum, R. H., van Ommen, B. and Groten, J. P. (2005a). Toxicogenomic analysis of gene expression changes in rat liver after a 28-day oral benzene exposure. Mutat Res 575: 85–101. Heijne, W. H., Kienhuis, A. S., van Ommen, B., Stierum, R. H. and Groten, J. P. (2005b). Systems toxicology: applications of toxicogenomics, transcriptomics, proteomics and metabolomics in toxicology. Expert Rev Proteomics 2: 767–780. Heinloth, A. N., Irwin, R. D., Boorman, G. A., Nettesheim, P., Fannin, R. D., Sieber, S. O., Snell, M. L., Tucker, C. J., Li, L., Travlos, G. S., Vansant, G., Blackshear, P. E., Tennant, R. W., Cunningham, M. L. and Paules, R. S. (2004). Gene expression profiling of rat livers reveals indicators of potential adverse effects. Toxicol Sci 80: 193–202. Heller, R. A., Schena, M., Chai, A., Shalon, D., Bedilion, T., Gilmore, J., Woolley, D. E. and Davis, R. W. (1997). Discovery and analysis of inflammatory disease-related genes using cDNA microarrays. Proc Natl Acad Sci USA 94: 2150–2155. Hester SD, W. D., Nesnow S. and Thai S-F (2006). Transcriptional Profiles in Liver from Rats Treated with Triazole Conazole Fungicides: Propiconazole, Triadimefon and Myclobutanil. Toxicologic Pathology 34: 879–894. Hoheisel, J. D. (2006). Microarray technology: beyond transcript profiling and genotype analysis. Nat Rev Genet 7: 200–210. Honda, M., Kawai, H., Shirota, Y., Yamashita, T. and Kaneko, S. (2005a). Differential gene expression profiles in stage I primary biliary cirrhosis. Am J Gastroenterol 100: 2019–2030. Honda, M., Kawai, H., Shirota, Y., Yamashita, T., Takamura, T. and Kaneko, S. (2005b). cDNA microarray analysis of autoimmune hepatitis, primary biliary cirrhosis and consecutive disease manifestation. J Autoimmun 25: 133–140. Hood, L., Heath, J. R., Phelps, M. E. and Lin, B. (2004). Systems biology and new technologies enable predictive and preventative medicine. Science 306: 640–643. Hu, D., Cao, K., Peterson-Wakeman, R. and Wang, R. (2002). Altered profile of gene expression in rat hearts induced by chronic nicotine consumption. Biochem Biophys Res Commun 297: 729– 736.

482

Hepatotoxicity

Huang, Q., Dunn, R. T., II Jayadev, S., DiSorbo, O., Pack, F. D., Farr, S. B., Stoll, R. E. and Blanchard, K. T. (2001). Assessment of cisplatin-induced nephrotoxicity by microarray technology. Toxicol Sci 63: 196–207. Huang, E., Ishida, S., Pittman, J., Dressman, H., Bild, A., Kloos, M., D’Amico, M., Pestell, R. G., West, M. and Nevins, J. R. (2003). Gene expression phenotypic models that predict the activity of oncogenic pathways. Nat Genet 34: 226–230. Iida, M., Anna, C. H., Hartis, J., Bruno, M., Wetmore, B., Dubin, J. R., Sieber, S., Bennett, L., Cunningham, M. L., Paules, R. S., Tomer, K. B., Houle, C. D., Merrick, A. B., Sills, R. C. and Devereux, T. R. (2003). Changes in global gene and protein expression during early mouse liver carcinogenesis induced by non-genotoxic model carcinogens oxazepam and Wyeth-14,643. Carcinogenesis 24: 757–770. Ikeo, K., Ishi-i, J., Tamura, T., Gojobori, T. and Tateno, Y. (2003). CIBEX: center for information biology gene expression database. C R Biol 326: 1079–1082. Iqbal, M. J., Yaegashi, S., Ahsan, R., Lightfoot, D. A. and Banz, W. J. (2002). Differentially abundant mRNAs in rat liver in response to diets containing soy protein isolate. Physiol Genomics 11: 219–226. Irizarry, R. A., Warren, D., Spencer, F., Kim, I. F., Biswal, S., Frank, B. C., Gabrielson, E., Garcia, J. G., Geoghegan, J., Germino, G., Griffin, C., Hilmer, S. C., Hoffman, E., Jedlicka, A. E., Kawasaki, E., Martinez-Murillo, F., Morsberger, L., Lee, H., Petersen, D., Quackenbush, J., Scott, A., Wilson, M., Yang, Y., Ye, S. Q. and Yu, W. (2005). Multiple-laboratory comparison of microarray platforms. Nat Methods 2: 345–350. Irwin, R. D., Parker, J. S., Lobenhofer, E. K., Burka, L. T., Blackshear, P. E., Vallant, M. K., Lebetkin, E. H., Gerken, D. F. and Boorman, G. A. (2005). Transcriptional profiling of the left and median liver lobes of male f344/n rats following exposure to acetaminophen. Toxicol Pathol 33: 111–117. Jones, A. L. and Spring-Mills, E. (1983). The liver and gallbladder, in Histology, L. Weiss (Ed.), Elsevier Biomedical, New York, NY, USA, pp. 707–735. Jungermann, K. and Katz, N. (1989). Functional specialization of different hepatocyte populations. Physiol Rev 69: 708–764. Kaminski, N., Allard, J. D., Pittet, J. F., Zuo, F., Griffiths, M. J., Morris, D., Huang, X., Sheppard, D. and Heller, R. A. (2000). Global analysis of gene expression in pulmonary fibrosis reveals distinct programs regulating lung inflammation and fibrosis. Proc Natl Acad Sci USA 97: 1778–1783. Kelley-Loughnane, N., Sabla, G. E., Ley-Ebert, C., Aronow, B. J. and Bezerra, J. A. (2002). Independent and overlapping transcriptional activation during liver development and regeneration in mice. Hepatology 35: 525–534. Khetani, S. R., Szulgit, G., Del Rio, J. A., Barlow, C. and Bhatia, S. N. (2004). Exploring interactions between rat hepatocytes and nonparenchymal cells using gene expression profiling. Hepatology 40: 545–554. Kier, L. D., Neft, R., Tang, L., Suizu, R., Cook, T., Onsurez, K., Tiegler, K., Sakai, Y., Ortiz, M., Nolan, T., Sankar, U. and Li, A. P. (2004). Applications of microarrays with toxicologically relevant genes (tox genes) for the evaluation of chemical toxicants in Sprague–Dawley rats in vivo and human hepatocytes in vitro. Mutat Res 549: 101–113. Kim, M., Gans, J. D., Nogueira, C., Wang, A., Paik, J. H., Feng, B., Brennan, C., Hahn, W. C., Cordon-Cardo, C., Wagner, S. N., Flotte, T. J., Duncan, L. M., Granter, S. R. and Chin, L. (2006). Comparative oncogenomics identifies NEDD9 as a melanoma metastasis gene. Cell 125: 1269– 1281. Kita, Y., Shiozawa, M., Jin, W., Majewski, R. R., Besharse, J. C., Greene, A. S. and Jacob, H. J. (2002). Implications of circadian gene expression in kidney, liver and the effects of fasting on pharmacogenomic studies. Pharmacogenetics 12: 55–65. Kiyosawa, N., Ito, K., Sakuma, K., Niino, N., Kanbori, M., Yamoto, T., Manabe, S. and Matsunuma, N. (2004a). Evaluation of glutathione deficiency in rat livers by microarray analysis. Biochem Pharmacol 68: 1465–1475.


483

Kiyosawa, N., Tanaka, K., Hirao, J., Ito, K., Niino, N., Sakuma, K., Kanbori, M., Yamoto, T., Manabe, S. and Matsunuma, N. (2004b). Molecular mechanism investigation of phenobarbital-induced serum cholesterol elevation in rat livers by microarray analysis. Arch Toxicol 78: 435–442. Kiyosawa, N., Ito, K., Watanabe, K., Kanbori, M., Niino, N., Manabe, S. and Yamoto, T. (2006). Utilization of a toxicogenomic biomarker for evaluation of chemical-induced glutathione deficiency in rat livers across the GeneChip data of different generations. Toxicol Lett 163: 161– 169. Kmiec, Z. (2001). Cooperation of liver cells in health and disease. Adv Anat Embryol Cell Biol 161 (III–XIII): 1–151. Knudsen, S. (2004). Guide to Analysis of DNA Microarray Data, Wiley-Liss, Hoboken, NJ, USA. Kofman, A. V., Morgan, G., Kirschenbaum, A., Osbeck, J., Hussain, M., Swenson, S. and Theise, N. D. (2005). Dose- and time-dependent oval cell reaction in acetaminophen-induced murine liver injury. Hepatology 41: 1252–1261. Kogure, K., Ishizaki, M., Nemoto, M., Kuwano, H. and Makuuchi, M. (1999). A comparative study of the anatomy of rat and human livers. J Hepatobiliary Pancreat Surg 6: 171–175. Kramer, J. A., Blomme, E. A., Bunch, R. T., Davila, J. C., Jackson, C. J., Jones, P. F., Kolaja, K. L. and Curtiss, S. W. (2003). Transcription profiling distinguishes dose-dependent effects in the livers of rats treated with clofibrate. Toxicol Pathol 31: 417–431. Kramer, J. A., Curtiss, S. W., Kolaja, K. L., Alden, C. L., Blomme, E. A., Curtiss, W. C., Davila, J. C., Jackson, C. J. and Bunch, R. T. (2004a). Acute molecular markers of rodent hepatic carcinogenesis identified by transcription profiling. Chem Res Toxicol 17: 463–470. Kramer, J. A., Pettit, S. D., Amin, R. P., Bertram, T. A., Car, B., Cunningham, M., Curtiss, S. W., Davis, J. W., Kind, C., Lawton, M., Naciff, J. M., Oreffo, V., Roman, R. J., Sistare, F. D., Stevens, J., Thompson, K., Vickers, A. E., Wild, S. and Afshari, C. A. (2004b). Overview on the application of transcription profiling using selected nephrotoxicants for toxicology assessment. Environ Health Perspect 112: 460–464. LaBrecque, D. (1994). Liver regeneration: a picture emerges from the puzzle. Am J Gastroenterol 89: S86–S96. Lam, S. H., Winata, C. L., Tong, Y., Korzh, S., Lim, W. S., Korzh, V., Spitsbergen, J., Mathavan, S., Miller, L. D., Liu, E. T. and Gong, Z. (2006a). Trancriptome Kinetics of Arsenic-induced Adaptive Response in Zebrafish Liver. Physiol Genomics 27: 351–361. Lam, S. H., Wu, Y. L., Vega, V. B., Miller, L. D., Spitsbergen, J., Tong, Y., Zhan, H., Govindarajan, K. R., Lee, S., Mathavan, S., Murthy, K. R., Buhler, D. R., Liu, E. T. and Gong, Z. (2006b). Conservation of gene expression signatures between zebrafish and human liver tumors and tumor progression. Nat Biotechnol 24: 73–75. Larkin, J. E., Frank, B. C., Gavras, H., Sultana, R. and Quackenbush, J. (2005). Independence and reproducibility across microarray platforms. Nat Methods 2: 337–344. Lee, W. M. (2003). Drug-induced hepatotoxicity. N Engl J Med 349: 474–485. Lee, S. H., Seo, G. S., Park, Y. N., Yoo, T. M. and Sohn, D. H. (2004). Effects and regulation of osteopontin in rat hepatic stellate cells. Biochem Pharmacol 68: 2367–2378. Lee, J. S., Grisham, J. W. and Thorgeirsson, S. S. (2005). Comparative functional genomics for identifying models of human cancer. Carcinogenesis 26: 1013–1020. Lelliott, C. J., Lopez, M., Curtis, R. K., Parker, N., Laudes, M., Yeo, G., Jimenez-Linan, M., Grosse, J., Saha, A. K., Wiggins, D., Hauton, D., Brand, M. D., O’Rahilly, S., Griffin, J. L., Gibbons, G. F. and Vidal-Puig, A. (2005). Transcript and metabolite analysis of the effects of tamoxifen in rat liver reveals inhibition of fatty acid synthesis in the presence of hepatic steatosis. Faseb J 19: 1108–1119. Lettieri, T. (2006). Recent applications of DNA microarray technology to toxicology and ecotoxicology. Environ Health Perspect 114: 4–9. Lin, D. W., Coleman, I. M., Hawley, S., Dumpit, R., Gifford, D., Kezele, P., Hung, H., Knudsen, B. S., Kristal, A. R. and Nelson, P. S. (2006). Influence of surgical manipulation on prostate gene

484

Hepatotoxicity

expression: implications for molecular correlates of treatment effects and disease prognosis. J Clin Oncol 24: 3763–3770. Liotta, L. A. and Petricoin, E. F. (2000). Beyond the genome to tissue proteomics. Breast Cancer Res 2: 13–14. Liu, J., Xie, Y., Ward, J. M., Diwan, B. A. and Waalkes, M. P. (2004a). Toxicogenomic analysis of aberrant gene expression in liver tumors and nontumorous livers of adult mice exposed in utero to inorganic arsenic. Toxicol Sci 77: 249–257. Liu, Z. X., Govindarajan, S. and Kaplowitz, N. (2004b). Innate immune system plays a critical role in determining the progression and severity of acetaminophen hepatotoxicity. Gastroenterology 127: 1760–1774. Lu, T., Liu, J., LeCluyse, E. L., Zhou, Y. S., Cheng, M. L. and Waalkes, M. P. (2001). Application of cDNA microarray to the study of arsenic-induced liver diseases in the population of Guizhou, China. Toxicol Sci 59: 185–192. Lu, J., Getz, G., Miska, E. A., Alvarez-Saavedra, E., Lamb, J., Peck, D., Sweet-Cordero, A., Ebert, B. L., Mak, R. H., Ferrando, A. A., Downing, J. R., Jacks, T., Horvitz, H. R. and Golub, T. R. (2005). MicroRNA expression profiles classify human cancers. Nature 435: 834–838. MacSween, R. N. M., Desmet, V. J., Roskams, T. and Scothorne, R. J. (2002). Developmental anatomy and normal structure, in Pathology of the liver, A. D. Burt, R. N. M., MacSween, B. C. Portmann, K. G. Ishak, P. J. Scheuer and P. P. Anthony (Eds), Churchill Livingstone, New York, NY, USA, pp. 1–66. Maggioli, J., Hoover, A. and Weng, L. (2006). Toxicogenomic analysis methods for predictive toxicology. J Pharmacol Toxicol Methods 53: 31–37. Malarkey, D. E., Johnson, K., Ryan, L., Boorman, G. and Maronpot, R. R. (2005a). New insights into functional aspects of liver morphology. Toxicol Pathol 33: 27–34. Malarkey, D. E., Parker, J. S., Turman, C. A., Scott, A. M., Paules, R. S., Collins, J. and Maronpot, R. R. (2005b). Microarray data analysis of mouse neoplasia. Toxicol Pathol 33: 127–135. Malik, R., Selden, C. and Hodgson, H. (2002). The role of non-parenchymal cells in liver growth. Semin Cell Dev Biol 13: 425–431. Mann, D. A. and Smart, D. E. (2002). Transcriptional regulation of hepatic stellate cell activation. Gut 50: 891–896. Matsumoto, T. and Kawakami, M. (1982). The unit-concept of hepatic parenchyma – a re-examination based on angioarchitectural studies. Acta Pathol Jpn 32 (Suppl 2) 285–314. Matsuzaki, S., Onda, M., Tajiri, T. and Kim, D. Y. (1997). Hepatic lobar differences in progression of chronic liver disease: correlation of asialoglycoprotein scintigraphy and hepatic functional reserve. Hepatology 25: 828–832. Mattes, W. B., Pettit, S. D., Sansone, S. A., Bushel, P. R. and Waters, M. D. (2004). Database development in toxicogenomics: issues and efforts. Environ Health Perspect 112: 495–505. McCaffrey, T. A., Fu, C., Du, B., Eksinar, S., Kent, K. C., Bush, H., Jr., Kreiger, K., Rosengart, T., Cybulsky, M. I., Silverman, E. S. and Collins, T. (2000). High-level expression of Egr-1 and Egr-1-inducible genes in mouse and human atherosclerosis. J Clin Invest 105: 653– 662. McMillian, M., Nie, A. Y., Parker, J. B., Leone, A., Kemmerer, M., Bryant, S., Herlich, J., Yieh, L., Bittner, A., Liu, X., Wan, J. and Johnson, M. D. (2004). Inverse gene expression patterns for macrophage activating hepatotoxicants and peroxisome proliferators in rat liver. Biochem Pharmacol 67: 2141–2165. McMillian, M., Nie, A., Parker, J. B., Leone, A., Kemmerer, M., Bryant, S., Herlich, J., Yieh, L., Bittner, A., Liu, X., Wan, J., Johnson, M. D. and Lord, P. (2005). Drug-induced oxidative stress in rat liver from a toxicogenomics perspective. Toxicol Appl Pharmacol 207: 171–178. Mehendale, H. M. (2005). Tissue repair: an important determinant of final outcome of toxicantinduced injury. Toxicol Pathol 33: 41–51.


485

Meyer, K., Lee, J. S., Dyck, P. A., Cao, W. Q., Rao, M. S., Thorgeirsson, S. S. and Reddy, J. K. (2003). Molecular profiling of hepatocellular carcinomas developing spontaneously in acyl-CoA oxidase deficient mice: comparison with liver tumors induced in wild-type mice by a peroxisome proliferator and a genotoxic carcinogen. Carcinogenesis 24: 975–984. Michel, C., Desdouets, C., Sacre-Salem, B., Gautier, J. C., Roberts, R. and Boitier, E. (2003). Liver gene expression profiles of rats treated with clofibric acid: comparison of whole liver and laser capture microdissected liver. Am J Pathol 163: 2191–2199. Minami, K., Saito, T., Narahara, M., Tomita, H., Kato, H., Sugiyama, H., Katoh, M., Nakajima, M. and Yokoi, T. (2005). Relationship between hepatic gene expression profiles and hepatotoxicity in five typical hepatotoxicant-administered rats. Toxicol Sci 87: 296–305. Minami, K., Maniratanachote, R., Katoh, M., Nakajima, M. and Yokoi, T. (2006). Simultaneous measurement of gene expression for hepatotoxicity in thioacetamide-administered rats by DNA microarrays. Mutat Res 603: 64–73. Morgan, K. T., Jayyosi, Z., Hower, M. A., Pino, M. V., Connolly, T. M., Kotlenga, K., Lin, J., Wang, M., Schmidts, H. L., Bonnefoi, M. S., Elston, T. C. and Boorman, G. A. (2005). The hepatic transcriptome as a window on whole-body physiology and pathophysiology. Toxicol Pathol 33: 136–145. Mortuza, G. B., Neville, W. A., Delaney, J., Waterfield, C. J. and Camilleri, P. (2003). Characterisation of a potential biomarker of phospholipidosis from amiodarone-treated rats. Biochim Biophys Acta 1631: 136–146. Mount, D. W. and Pandey, R. (2005). Using bioinformatics and genome analysis for new therapeutic interventions. Mol Cancer Ther 4: 1636–1643. Mulas, M. F., Demuro, G., Mulas, C., Putzolu, M., Cavallini, G., Donati, A., Bergamini, E. and Dessi, S. (2005). Dietary restriction counteracts age-related changes in cholesterol metabolism in the rat. Mech Ageing Dev 126: 648–654. Murray, M. (2006). Altered CYP expression and function in response to dietary factors: potential roles in disease pathogenesis. Curr Drug Metab 7: 67–81. Nonaka, H., Sugano, S. and Miyajima, A. (2004). Serial analysis of gene expression in sinusoidal endothelial cells from normal and injured mouse liver. Biochem Biophys Res Commun 324: 15–24. Nuwaysir, E. F., Bittner, M., Trent, J., Barrett, J. C. and Afshari, C. A. (1999). Microarrays and toxicology: the advent of toxicogenomics. Mol Carcinog 24: 153–159. Orphanides, G. (2003). Toxicogenomics: challenges and opportunities. Toxicol Lett 140–141: 145– 148. Panda, S., Antoch, M. P., Miller, B. H., Su, A. I., Schook, A. B., Straume, M., Schultz, P. G., Kay, S. A., Takahashi, J. S. and Hogenesch, J. B. (2002). Coordinated transcription of key pathways in the mouse by the circadian clock. Cell 109: 307–320. Perou, C. M., Jeffrey, S. S., van de Rijn, M., Rees, C. A., Eisen, M. B., Ross, D. T., Pergamenschikov, A., Williams, C. F., Zhu, S. X., Lee, J. C., Lashkari, D., Shalon, D., Brown, P. O. and Botstein, D. (1999). Distinctive gene expression patterns in human mammary epithelial cells and breast cancers. Proc Natl Acad Sci USA 96: 9212–9217. Perou, C. M., Sorlie, T., Eisen, M. B., van de Rijn, M., Jeffrey, S. S., Rees, C. A., Pollack, J. R., Ross, D. T., Johnsen, H., Akslen, L. A., Fluge, O., Pergamenschikov, A., Williams, C., Zhu, S. X., Lonning, P. E., Borresen-Dale, A. L., Brown, P. O. and Botstein, D. (2000). Molecular portraits of human breast tumours. Nature 406: 747–752. Peterson, R. L., Casciotti, L., Block, L., Goad, M. E., Tong, Z., Meehan, J. T., Jordan, R. A., Vinlove, M. P., Markiewicz, V. R., Weed, C. A. and Dorner, A. J. (2004). Mechanistic toxicogenomic analysis of WAY-144122 administration in Sprague–Dawley rats. Toxicol Appl Pharmacol 196: 80–94. Petersen, D., Chandramouli, G. V., Geoghegan, J., Hilburn, J., Paarlberg, J., Kim, C. H., Munroe, D., Gangi, L., Han, J., Puri, R., Staudt, L., Weinstein, J., Barrett, J. C., Green, J., and Kawasaki, E. S.

486

Hepatotoxicity

(2005). Three microarray platforms: an analysis of their concordance in profiling gene expression. BMC Genomics 6: 63. Ratziu, V. and Friedman, S.L. (1997). Pathology of hepatic stellate cells, in Functional Heterogeneity of Liver Tissue:From Cell Lineage Diversity to Sublobular Compartment-Specific Pathogenesis, F. Vidal-Vanaclocha (Ed.), R.G. Landes Company, Austin, TX, USA, pp. 133–160. Richardson, F. C., Boucheron, J. A., Dyroff, M. C., Popp, J. A. and Swenberg, J. A. (1986). Biochemical and morphologic studies of heterogeneous lobe responses in hepatocarcinogenesis. Carcinogenesis 7: 247–251. Richert, L., Liguori, M. J., Abadie, C., Heyd, B., Mantion, G., Halkic, N. and Waring, J. F. (2006). Gene expression in human hepatocytes in suspension after isolation is similar to the liver of origin, is not affected by hepatocyte cold storage and cryopreservation, but is strongly changed after hepatocyte plating. Drug Metab Dispos 34: 870–879. Ruepp, S. U., Tonge, R. P., Shaw, J., Wallis, N. and Pognan, F. (2002). Genomics and proteomics analysis of acetaminophen toxicity in mouse liver. Toxicol Sci 65: 135–150. Sanguino, E., Roglans, N., Alegret, M., Sanchez, R. M., Vazquez-Carrera, M. and Laguna, J. C. (2005). Atorvastatin reverses age-related reduction in rat hepatic PPARalpha and HNF-4. Br J Pharmacol 145: 853–861. Sanz, N., Diez-Fernandez, C., Andres, D. and Cascales, M. (2002). Hepatotoxicity and aging: endogenous antioxidant systems in hepatocytes from 2-, 6-, 12-, 18- and 30-month-old rats following a necrogenic dose of thioacetamide. Biochim Biophys Acta 1587: 12–20. Sawada, H., Takami, K. and Asahi, S. (2005). A toxicogenomic approach to drug-induced phospholipidosis: analysis of its induction mechanism and establishment of a novel in vitro screening system. Toxicol Sci 83: 282–292. Schieferdecker, H. L., Schlaf, G., Jungermann, K. and Gotze, O. (2001). Functions of anaphylatoxin C5a in rat liver: direct and indirect actions on nonparenchymal and parenchymal cells. Int Immunopharmacol 1: 469–481. Segal, E., Friedman, N., Koller, D. and Regev, A. (2004). A module map showing conditional activity of expression modules in cancer. Nat Genet 36: 1090–1098. Shackel, N. A., Gorrell, M. D. and McCaughan, G. W. (2002a). Gene array analysis and the liver. Hepatology 36: 1313–1325. Shackel, N. A., McGuinness, P. H., Abbott, C. A., Gorrell, M. D. and McCaughan, G. W. (2002b). Insights into the pathobiology of hepatitis C virus-associated cirrhosis: analysis of intrahepatic differential gene expression. Am J Pathol 160: 641–654. Simmons, P. T. and Portier, C. J. (2002). Toxicogenomics: the new frontier in risk analysis. Carcinogenesis 23: 903–905. Smith, L. L. (2001). Key challenges for toxicologists in the 21st century. Trends Pharmacol Sci 22: 281–285. Steiner, G., Suter, L., Boess, F., Gasser, R., de Vera, M. C., Albertini, S. and Ruepp, S. (2004). Discriminating different classes of toxicants by transcript profiling. Environ Health Perspect 112: 1236–1248. Stokkan, K. A., Yamazaki, S., Tei, H., Sakaki, Y. and Menaker, M. (2001). Entrainment of the circadian clock in the liver by feeding. Science 291: 490–493. Storch, K. F., Lipan, O., Leykin, I., Viswanathan, N., Davis, F. C., Wong, W. H. and Weitz, C. J. (2002). Extensive and divergent circadian gene expression in liver and heart. Nature 417: 78–83. Stuart, E. T. and Wheatley, A. M. (1995). Redistribution of portal venous but not hepatic arterial flow is induced by hepatic nerve stimulation in the perfused rat liver. Arch Physiol Biochem 103: 99–108. Suriawinata, A. and Xu, R. (2004). An update on the molecular genetics of hepatocellular carcinoma. Semin Liver Dis 24: 77–88.


487

Takahashi, M., Tsuboyama-Kasaoka, N., Nakatani, T., Ishii, M., Tsutsumi, S., Aburatani, H. and Ezaki, O. (2002). Fish oil feeding alters liver gene expressions to defend against PPARalpha activation and ROS production. Am J Physiol Gastrointest Liver Physiol 282: G338–G348. Tan, Y., Shi, L., Hussain, S. M., Xu, J., Tong, W., Frazier, J. M. and Wang, C. (2006). Integrating time-course microarray gene expression profiles with cytotoxicity for identification of biomarkers in primary rat hepatocytes exposed to cadmium. Bioinformatics 22: 77–87. Tennant, R. W. (2002). The National Center for Toxicogenomics: using new technologies to inform mechanistic toxicology. Environ Health Perspect 110: A8–A10. Teutsch, H. F., Schuerfeld, D. and Groezinger, E. (1999). Three-dimensional reconstruction of parenchymal units in the liver of the rat. Hepatology 29: 494–505. Thein, E., Becker, M., Anetzberger, H., Hammer, C. and Messmer, K. (2003). Direct assessment and distribution of regional portal blood flow in the pig by means of fluorescent microspheres. J Appl Physiol 95: 1808–1816. Thomas, R. S., Rank, D. R., Penn, S. G., Zastrow, G. M., Hayes, K. R., Pande, K., Glover, E., Silander, T., Craven, M. W., Reddy, J. K., Jovanovich, S. B. and Bradfield, C. A. (2001). Identification of toxicologically predictive gene sets using cDNA microarrays. Mol Pharmacol 60: 1189– 1194. Thompson, K. L., Afshari, C. A., Amin, R. P., Bertram, T. A., Car, B., Cunningham, M., Kind, C., Kramer, J. A., Lawton, M., Mirsky, M., Naciff, J. M., Oreffo, V., Pine, P. S. and Sistare, F. D. (2004). Identification of platform-independent gene expression markers of cisplatin nephrotoxicity. Environ Health Perspect 112, 488–494. Thorgeirsson, S. S. and Grisham, J. W. (2002). Molecular pathogenesis of human hepatocellular carcinoma. Nat Genet 31: 339–346. Thorgeirsson, S. S., Lee, J. S. and Grisham, J. W. (2006a). Functional genomics of hepatocellular carcinoma. Hepatology 43: S145–S150. Thorgeirsson, S. S., Lee, J. S. and Grisham, J. W. (2006b). Molecular prognostication of liver cancer: end of the beginning. J Hepatol 44: 798–805. Tian, Y., Jochum, W., Georgiev, P., Moritz, W., Graf, R. and Clavien, P. A. (2006). Kupffer celldependent TNF-alpha signaling mediates injury in the arterialized small-for-size liver transplantation in the mouse. Proc Natl Acad Sci USA 103: 4598–4603. Tilton, S. C., Givan, S. A., Pereira, C. B., Bailey, G. S. and Williams, D. E. (2006). Toxicogenomic profiling of the hepatic tumor promoters indole-3-carbinol, 17beta-estradiol and beta-naphthoflavone in rainbow trout. Toxicol Sci 90: 61–72. Tsai, C. A., Lee, T. C., Ho, I. C., Yang, U. C., Chen, C. H. and Chen, J. J. (2005). Multi-class clustering and prediction in the analysis of microarray data. Math Biosci 193: 79–100. Tugendreich, S., Pearson, C. I., Sagartz, J., Jarnagin, K. and Kolaja, K. (2006). NSAID-induced acute phase response is due to increased intestinal permeability and characterized by early and consistent alterations in hepatic gene expression. Toxicol Pathol 34: 168–179. Tygstrup, N., Bangert, K., Ott, P. and Bisgaard, H. C. (2002). Messenger RNA profiles in liver injury and stress: a comparison of lethal and nonlethal rat models. Biochem Biophys Res Commun 290: 518–525. Ulrich, R. and Friend, S. H. (2002). Toxicogenomics and drug discovery: will new technologies help us produce better drugs? Nat Rev Drug Discov 1: 84–88. Ulrich, R. G., Rockett, J. C., Gibson, G. G. and Pettit, S. D. (2004). Overview of an interlaboratory collaboration on evaluating the effects of model hepatotoxicants on hepatic gene expression. Environ Health Perspect 112: 423–427. Wagenaar, G. T., ter Horst, S. A., van Gastelen, M. A., Leijser, L. M., Mauad, T., van der Velden, P. A., de Heer, E., Hiemstra, P. S., Poorthuis, B. J. and Walther, F. J. (2004). Gene expression profile and histopathology of experimental bronchopulmonary dysplasia induced by prolonged oxidative stress. Free Radic Biol Med 36: 782–801.

488

Hepatotoxicity

Walisser, J. A., Glover, E., Pande, K., Liss, A. L. and Bradfield, C. A. (2005). Aryl hydrocarbon receptor-dependent liver development and hepatotoxicity are mediated by different cell types. Proc Natl Acad Sci USA 102: 17858–17863. Walsh, P. J., Bookman, R. J., Zaias, J., Mayer, G. D., Abraham, W., Bourdelais, A. J. and Baden, D. G. (2003). Toxicogenomic effects of marine brevetoxins in liver and brain of mouse. Comp Biochem Physiol B Biochem Mol Biol 136: 173–182. Ward, W., Delker, D., Hester, S., Thai, S., Wolf, D., Allen, J. and Nesnow, S. (2006). Transcriptional Profiles in Liver from Mice Treated with Hepatotumorigenic and Non-Hepatotumorigenic Triazole Conazole Fungicides: Propiconazole, Triadimefon and Myclobutanil. Toxicol Pathol 34: 863–878. Waring, J. F., Ciurlionis, R., Jolly, R. A., Heindel, M. and Ulrich, R. G. (2001a). Microarray analysis of hepatotoxins in vitro reveals a correlation between gene expression profiles and mechanisms of toxicity. Toxicol Lett 120: 359–368. Waring, J. F., Jolly, R. A., Ciurlionis, R., Lum, P. Y., Praestgaard, J. T., Morfitt, D. C., Buratto, B., Roberts, C., Schadt, E. and Ulrich, R. G. (2001b). Clustering of hepatotoxins based on mechanism of toxicity using gene expression profiles. Toxicol Appl Pharmacol 175: 28–42. Waring, J. F., Gum, R., Morfitt, D., Jolly, R. A., Ciurlionis, R., Heindel, M., Gallenberg, L., Buratto, B. and Ulrich, R. G. (2002). Identifying toxic mechanisms using DNA microarrays: evidence that an experimental inhibitor of cell adhesion molecule expression signals through the aryl hydrocarbon nuclear receptor. Toxicology 181–182: 537–550. Waring, J. F., Cavet, G., Jolly, R. A., McDowell, J., Dai, H., Ciurlionis, R., Zhang, C., Stoughton, R., Lum, P., Ferguson, A., Roberts, C. J. and Ulrich, R. G. (2003). Development of a DNA microarray for toxicology based on hepatotoxin-regulated sequences. EHP Toxicogenomics 111: 53–60. Waring, J. F., Ulrich, R. G., Flint, N., Morfitt, D., Kalkuhl, A., Staedtler, F., Lawton, M., Beekman, J. M. and Suter, L. (2004). Interlaboratory evaluation of rat hepatic gene expression changes induced by methapyrilene. Environ Health Perspect 112: 439–448. Waters, M. D. and Fostel, J. M. (2004). Toxicogenomics and systems toxicology: aims and prospects. Nat Rev Genet 5: 936–948. Waters, M., Boorman, G., Bushel, P., Cunningham, M., Irwin, R., Merrick, A., Olden, K., Paules, R., Selkirk, J., Stasiewicz, S., Weis, B., Van Houten, B., Walker, N. and Tennant, R. (2003). Systems toxicology and the Chemical Effects in Biological Systems (CEBS) knowledge base. EHP Toxicogenomics 111: 15–28. Xie, T., Tong, L., Barrett, T., Yuan, J., Hatzidimitriou, G., McCann, U. D., Becker, K. G., Donovan, D. M. and Ricaurte, G. A. (2002). Changes in gene expression linked to methamphetamine-induced dopaminergic neurotoxicity. J Neurosci 22: 274–283. Yadetie, F., Laegreid, A., Bakke, I., Kusnierczyk, W., Komorowski, J., Waldum, H. L. and Sandvik, A. K. (2003). Liver gene expression in rats in response to the peroxisome proliferator-activated receptor-alpha agonist ciprofibrate. Physiol Genomics 15: 9–19. Yeatman, T. J. (2003). The future of cancer management: translating the genome, transcriptome and proteome. Ann Surg Oncol 10: 7–14. Zender, L., Spector, M. S., Xue, W., Flemming, P., Cordon-Cardo, C., Silke, J., Fan, S. T., Luk, J. M., Wigler, M., Hannon, G. J., Mu, D., Lucito, R., Powers, S., and Lowe, S. W. (2006). Identification and validation of oncogenes in liver cancer using an integrative oncogenomic approach. Cell 125: 1253–1267.

19 Use of DNA Arrays in Understanding Hepatic Test Systems Angela J. Harris and Daniel A. Casciano

19.1

Introduction

In order to understand how hepatic toxicants and toxins affect gene expression, it is first necessary to have an understanding of the basal gene expression in the hepatic test system (i.e. liver, primary hepatocytes, liver slices, cell lines, etc.) being used. This includes identification of genes that are normally expressed, as well as determination of the inherent variability in expression of those genes. Because data obtained from toxicological testing conducted in rodents are used to predict human responses, it is also critical to know the differences in basal gene expression between various animals and humans. An additional component to be considered is the extent of variability in hepatic tissues in rodents compared to tissues obtained from human donors. Significant differences in gene expression are expected among tissues from different human donors due to the substantial heterogeneity in the human population compared to the more genetically homogenous rodents commonly used in toxicological testing; however, even commonly used laboratory animals will have a certain degree of fluctuation in basal gene expression. In order to evaluate the toxicity of a chemical or drug, animals treated with the test agent are compared to untreated animals in order to determine the effect of the chemical agent on the treated animals. Although experiments involving animals are carefully designed to ensure that all parameters except for the tested agent are kept to a minimum, it is still essential that researchers have an understanding of the extent of variable gene expression in the animal system in which they work in order to be able to properly interpret gene expression obtained from test animals.


490

Hepatotoxicity

Not all toxicological testing is performed in whole animals. Animal testing is very expensive and therefore not appropriate for all toxicological analyses. Various in vitro hepatic testing systems have been developed in an attempt to minimize whole animal studies; however, data obtained from in vitro systems are not useful unless there is an understanding of how these data correlate with the corresponding in vivo response. In addition, changes in culture conditions and time in culture for all in vitro systems have effects on gene expression that alter toxicological responses in some way. Such changes need to be well understood in order to select the most appropriate in vitro system for the toxicological endpoint being evaluated. The use of cDNA arrays is a particularly powerful way to simultaneously analyze the expression of thousands of genes and multiple investigators have used these data to evaluate various hepatic test systems. For instance, gene array data have been used in a limited fashion thus far to identify genes that are normally expressed in whole liver from rodent and humans, estimate the extent of variability of expression in different hepatic testing systems, compare the basal gene expression of some in vitro systems to whole liver, and identify the effects of changes in culture conditions on gene expression in vitro . Specific information about the DNA array used for the experiments discussed in this chapter is summarized in Table 19.1. Contact or website information where additional gene array data for some cited references can be obtained is found in Table 19.2.

19.2

Commonly Used Hepatic Systems in Toxicological Testing

There are several options for hepatic testing in toxicological studies, each with its own advantages and disadvantages. Animal testing in rodents is the most informative of those available for toxicological studies; however, several in vitro systems are also commonly used, including liver slices, primary cultured hepatocytes and various hepatic cell lines. Toxicological studies in animals are most commonly conducted in rodent species such as mice and rats; however, animal testing is expensive and therefore not suitable for primary screening of toxicological compounds. In addition, the data obtained from toxicity studies involving rodents are not always concordant with toxicity subsequently detected in humans, and test animals do not reflect the genetic diversity of humans. Ideally, the optimal test system for predicting human liver toxicity would be studies involving humans; however, in vivo toxicological testing in humans is not done for obvious reasons. In addition to ethical considerations, a very high sample size would be required to evaluate toxicity in the human population due to the high degree of interindividual differences among humans. Therefore, even use of a human hepatic in vitro system from human donors would need to be relatively inexpensive to allow an investigator to evaluate responses from sufficient numbers of individuals to be predictive of the most common response. Or alternatively, sufficient data need to be obtained about the degree of heterogeneity present in the general population to develop an in vivo system (or systems) that covers a broad spectrum of predictive responses, the most common responses and identification of factors that are associated with idiosyncratic responses. In vitro testing using animal and/or human hepatic tissues are attractive alternatives to in vivo testing since they are less expensive and have more flexibility in design and utilization of resources. Human liver can be obtained through donors and used for short-term

DNA Arrays in Understanding Hepatic Test Systems Table 19.1

491

DNA arrays used by investigators Number of sequences on array

Citation

Manufacturer/Type

Bono et al., 2003 Novak et al., 2002

Custom/cDNA Affymetrix/Mu11kSuA and Mu11kSubB/ Oligonucleotides Custom/cDNA Custom/cDNA

57 931 11 000

Mouse Mouse

5285 417

Mouse Mouse

Affymetrix/RGU34A Afymetrix/RTU34 Affymetrix/RGU34A Agilent Technologies/Oligo Clontech/Stress Array Affymetrix RG&34A/Oligo Affymetrix/U34

8700 900 8700 20 000

Rat Rat Rat

207 8700

Rat Rat

970

Rat

Affymetrix/U95A v.1 and v.2 Invitrogen/cDNA

∼10 000

Human

31 110

Human

588

Human

∼31 000 234

Human Human

1185

Human

6800 ∼12 600

Human Human

6794

Human

4043

Human

1281

Human

Pritchard et al., 2001 Tomascik-Cheeseman et al., 2004 Baker et al., 2001 Boess et al., 2003 Boorman et al., 2005 Harris et al., 2003 Jessen et al., 2003 Schuppe-Koistinen et al., 2002 Butura et al., 2004 Delongchamp et al., 2005 Liu et al., 2003 Harris et al., 2003 Morgan et al., 2002 Morgan et al., 2003 Novak et al., 2002 Sonna et al., 2003 Waring et al., 2003 Yano et al., 2001 Yamashita et al., 2004

Clontech Cancer Array/cDNA Invitrogen/cDNA Clontech/Atlas Human Toxicology /cDNA Clontech Atlas Human Toxicology 1.2 Affymetric/HuGene FL Affymetrix HG-U95Av2/cDNA Incyte Pharmaceutical/cDNA Research Genetics/cDNA Custon/cDNA

Class

in vitro studies. One advantage of in vitro testing is that it removes some of the inherent heterogeneity in animal studies. Unlike animal testing, untreated liver tissue (or cells) is compared to treated tissue (or cells) from the same animal or human donor, rather than comparing data from an untreated animal to a different animal that has been given the treatment. The most commonly used in vitro methods are liver slices, primary hepatocytes isolated from whole liver, or one of several hepatoma cell lines derived from human (HepG2) or animal (BRL3A) liver. Liver slices offer the advantage of maintenance of the functional acinar structure and the multicellular environment of whole liver; however, they have not

492

Hepatotoxicity

Table 19.2 Sources of supplemental data and/or information Citation

Class

Location of data

Baker et al., 2001

Rat

Boess et al., 2003 Bono et al., 2003

Rat Mouse

Butura et al., 2004 Harris et al., 2004 Morgan et al., 2002, 2003

Human Human Human

Novak et al., 2002 Pritchard et al., 2001 Schuppe-Koistinen et al., 2002 Tomascik-Cheeseman et al., 2004

Mouse Mouse Rat Mouse

http://pubs.acs.org/subscribe/journals/crtoec/ supmat/index.html. www.roche.com/science-download.htm http://www.genome.org/cgi/content/full/13/ 6b/1318/DC1 http://www.imm.ki.se/butura.htm [email protected] http://taylorandfrancis.metapress.com/open url.asp?genre=journal&issn=0192-6233 http://www.idealibrary.com http://www.pedb.org www.imm.ki.se/ina http://microarray.llnl.gov

been used as extensively as some other in vitro systems due to technical difficulties that include producing slices of uniform thickness, cellular damage at edges, limitations in diffusion to core of slice, etc. Primary hepatocytes isolated from whole liver by various perfusion methods (Oldham et al., 1979; Shaddock et al., 1995) are used extensively in toxicological studies and as such, are considered the ‘gold standard’ for in vitro hepatic toxicology. Primary hepatocytes have been isolated from many animal species, including humans, mice and rats. However, when hepatocytes are isolated and cultured, liver-specific metabolic functions gradually decrease over time. Liver-specific functions in vivo are supported by some combination of liver architecture, multi-cellular interactions and a complex hormonal milieu which is not easily duplicated in vitro. Plating on a culture substratum is often used to help maintain liver specific function. Multiple substrata have been tested; however, the most commonly used culture conditions are monolayer culture on matrigel or collagen, and sandwich configuration between two layers of collagen (George et al., 1997; LeCluyse et al., 2000).

19.3

Basal Gene Expression in Whole Liver and In Vitro

The ability to simultaneously evaluate hundreds or thousands of genes is a powerful way to both identify expressed genes and to quantify the variability in gene expression in any test system. DNA arrays have been used to determine the basal gene expression in the liver of untreated mice (Bono et al., 2003; Novak et al., 2002; Pritchard et al., 2001) and the liver of rats treated with a commonly used test vehicle (Boorman et al., 2005). Expression data have also been obtained from the liver of individual human donors (Yano et al., 2001; Delongchamp et al., 2005). The basal gene expression pattern and the variability of that pattern has been evaluated in freshly isolated rat hepatocyte preparations (Harris et al., 2003), in cultured primary human hepatocytes from multiple donors (Harris et al., 2004)

DNA Arrays in Understanding Hepatic Test Systems

493

and in multiple replicates of a commonly used human hepatoma cell line, HepG2 (Sonna et al., 2003; Morgan et al., 2002). 19.3.1

Whole Liver – Rodent

Bono et al. (2003) interrogated 57 931 mouse cDNAs for tissue-specific expression of genes in 20 mouse tissues, including liver in C57BL/6J mice. Specific data were not reported but have been made available by the authors (Table 19.2). Liver-specific genes expressed in male C57BL/6J mice were identified by analysis of the expression data and then each gene was assigned a functional description using Gene Ontology (GO) Slim terms (http://www.ebi.ac.uk/proteome/goslimterms.html). Genes expressed in liver, but not other tissue included those that code for transporter proteins (13), chaperone proteins (2), enzymes (48), regulators of enzyme function (5), receptor proteins (27), signal transduction (2) and structural proteins (2). In addition, 81 liver-specific genes with no assigned function were also identified. It is interesting to note that almost as many genes with no known function were found to be expressed solely in the liver, as those with an identified function. This finding illustrates one of the strengths of using cDNA arrays since it is possible to detect an effect on gene expression in unknown genes, or in those genes which might not be hypothesized to be affected by treatment. There were no reported data on the variability of gene expression in tissues from different animals in this study. Data from studies that examined gene expression in mouse and rat liver showed that interindividual variability of expression was low. In mice, the hepatic expression of 5285 mouse genes from six male B57BL6 mice was compared (Pritchard et al., 2001). Of the interrogated genes, 2514 had detectable expression and the expression was surprisingly stable among the six animals. Significant variability (P < 0.05) was seen in only 0.8 % of expressed genes. In fact, the liver had the lowest variability of the three tissues evaluated (kidney, testes, liver). The expressed genes with the highest variability were primarily stressresponse and immune-modulated genes, including BCL-6, CisH and Gadd45; although one metabolic gene, Cyp4A12, varied about 3-fold. The authors suggest that the variability in gene expression was non-random since some of the effects in gene expression were noted to be related to the sequence in which animals were killed, and might reflect hormone and/or stress responses. Novak et al. (2002) used cDNA arrays to examine the variability of hepatic gene expression in three male B57BL6 mice. They used the coefficient of proportionality from pairwise comparisons of each of the three mice to estimate the ‘dispersion’ of expression. Pairwise comparisons with one particular mouse resulted in a consistently higher coefficient of proportionality (0.190 and 0.185) than comparisons between the other two mice (0.108 and 0.135). As noted in the Pritchard study, genes with the highest degree of variability were primarily stress-responses genes. Several of these stress-related genes code for heat shock proteins, including HSP25, HSP27, HSP84, HSP86 and HSP105. There was no overlap in the identity of specific genes with the most variable expression in the Pritchard and Novak studies. This could be due to selection of reported data, the gene set screened or other experimental differences. It is noteworthy that in both studies the genes with the greatest degree of variability were stress-response genes. It has been reported that the genes with the greatest variability of expression in many tissues, including the liver of B6C3F1 mice, are the stress-response genes and that expression of stress-response genes is highest in the liver (Tomascik-Cheeseman et al., 2004).

494

Hepatotoxicity

Basal gene expression in the liver of rats treated with a commonly used control vehicle was evaluated by Boorman et al., (2005). It is unusual for control animals to have had no treatment at all. In order to minimize experimentally induced differences between control and treated animals, control animals undergo dosing procedures and regimes that are identical with the test animals, except they do not receive the chemical agent under investigation. The most commonly administered compound is the vehicle in which the test compound is dissolved, usually saline or a solvent. Solvents commonly used to prepare solutions of toxicants have been shown to have effects on gene expression. For instance, DMSO increases expression of acyl-Co-oxidase, a gene involved in peroxisome proliferation and decreases expression of the housekeeping gene hypoxanthine guanine phosphoribosyl transferase (HGPT) in primary rat hepatocytes (P < 0.01) (Longueville et al., 2003). Two genes that play a role in stress response, GSH reductase and MDR-1b (multi-drug resistance), were decreased in rat hepatocytes treated with DMF (P < 0.01) (Longueville et al., 2003). Boorman et al. (2005) evaluated the variability of hepatic gene expression in 24 male Fischer 344 rats given 0.5 % methylcellulose, a commonly used solvent vehicle. The 24 rats were sub-grouped into two independent experiments of 12 rats each which were kept on a different 12 h light cycle. The gene expression of individual rats was compared to that of pooled matched controls from rats with similar experimental parameters. A combined total of 8333 genes with altered expression in at least one of the 24 rats were identified. Of these, 95 % of differentially expressed genes had differences of less than 1.5-fold. Expression differences greater than 2-fold were seen in less than 1 % of differentially expressed genes. None of these genes was specifically identified. Supervised cluster analysis of the data suggested that the majority of the differences were random changes. This conclusion is the reverse of that reached by Pritchard et al., (2001) regarding the nature of gene changes in basal gene expression. One of the strengths of the Boorman study is the simultaneous evaluation of various hepatic clinical chemistry values with toxicogenomic analysis. Although it was found that ALT levels differed between the two light cycle experimental groups of rats, there were no significant differences in gene expression between the two groups. It was noted, however, that the single rat with the highest ALT level did have some individual gene expression changes not seen in five other rats with the same treatment regime. Hepatic glutathione levels did not correlate with changes in gene expression and no consistent changes were found in basal gene expression due to weight of the rats. 19.3.2

Liver – Human

Gene expression in human liver is expected to be much more variable since humans are more genetically diverse than are the inbred animal strains of mice and rats commonly used in toxicological testing. Differences in hormonal status, nutritional status, use of pharmacological and/or recreational drugs, alcohol use, smoking, health status and other uncharacterized or unknown factors affect the basal expression of hepatic genes. Several questions remain about hepatic gene expression in humans. What genes are normally expressed in the liver of most humans? Which genes have the most variability of expression in the human population? What is the range of expression of any individual gene in the human population? Which genes are liver-specific genes? Microarrays are a powerful and relatively inexpensive way to obtain basic information concerning hepatic gene expression.

DNA Arrays in Understanding Hepatic Test Systems Table 19.3

495

Human Donor Information

Citation

Sex

Health Age

status

Social/other Smoker Non-smoker Smoker/heavy alcohol — — — — — — — — — Diazepam — — — — Citalopram, diazepam — — Metoprolol, diazepam

Harris et al., 2004 Harris et al., 2004 Harris et al., 2004

M M M

62 59 54

NIDDM, stroke, MI NIDDM, hypertension Head trauma

Waring et al., 2003 Waring et al., 2003 Waring et al., 2003 Waring et al., 2003 Yano et al., 2001 Yano et al., 2001 Yano et al., 2001 Yano et al., 2001 Yano et al., 2001 Butura et al., 2004; Westlind et al., 1999

M F M M M F M F F A

Butura et al., 2004; Westlind et al., 1999

B

M M F M M F

66 62 36 66 59 54 76 55 38 41 49 69 62 67 61

Inner cranial hemorrhage Inner cranial hemorrhage Inner cranial hemorrhage Inner cranial hemorrhage Colorectal liver metastases Colorectal liver metastases Colorectal liver metastases Colorectal liver metastases Cholangiocarcinoma — — — — —

M M F

NA 61 78

— — —

Butura et al., 2004; Westlind et al., 1999

C

As expected, gene expression in human liver has been shown to be highly variable. Yano et al. (2001) reported that of 4043 genes screened using RNA isolated from the liver of five individual donors (Table 19.3), a total of 2418 different genes was expressed. However, only 1212 genes or about half of detected genes were expressed in four of the five liver samples. Presumably a far lower number were expressed in all liver tissues, although these data were not reported. An average of 1635 ± 274 (range 1278–1995) genes was expressed in each of the livers tested. Delongchamp et al. (2005) reported that a total of 2800 genes (of 31 110 screened genes) was expressed in liver of nine male and nine female human donors. This number is remarkably similar to the 2418 expressed genes reported by Yano et al. (2001), even though far more genes were interrogated in the Delongchamp study. 19.3.3

Primary Hepatocytes – Rodent

The liver is composed of multiple cell types, including Kupfer cells, Ito cells, hepatocytes and endothelial cells, all of which contribute to basal hepatic gene expression. However, hepatocytes are the primary functional cell type responsible for most metabolic functions that occur in the liver. Isolated and/or cultured primary hepatocytes will have the same variability of basal gene expression that is seen in whole liver due to interindividual differences. In addition, there will be changes in gene expression in isolated primary hepatocytes due

496

Hepatotoxicity

to experimental variables, including the removal of hepatocytes from liver tissue, culturing conditions, time in culture, etc. In order to estimate variability in gene expression in hepatocytes that had been isolated from rat liver, the basal gene expression of three independent isolates of hepatocytes from male Fischer 344 rats was determined using cDNA analysis (Harris et al., 2003). The data from each individual isolate were compared to the geometric mean of the data from all three rats to assess the effects of interindividual differences and experimental variability of the three rat hepatocyte preparations. The correlation coefficient was determined for each isolate. Although the correlation coefficients indicate there is some scatter in the basal gene expression patterns of freshly isolated hepatocytes from different animals, the interindividual variability was fairly consistent (R = 0.82, 0.88, 0.80). 19.3.4

Primary Hepatocytes – Human

There is always some concern about how applicable rodent data are in predicting human responses. Because in vivo studies involving humans are not possible, primary hepatocytes isolated from donated human liver are used for many toxicological research purposes. Some limited data are available for gene expression comparisons in cultured primary human hepatocytes. The expression of about 31 000 genes was interrogated using mRNA isolated from primary human hepatocytes obtained from three individual donors (Harris et al., 2004) (Table 19.3). The number of detectable genes after 16 h of culture on matrigel was 1690, 1939 and 951, depending on the donor. Expression of 867 genes was detected in hepatocytes from all three donors; however 61, 160 and 410 genes were detected in hepatocytes from one donor that were not detected in the other two, suggesting a significant number of genes with variable expression. Although these data also reflect experimental variance, they most likely reflect the significant variability in expression expected from the heterogeneous human population. This degree of variability of expression is not surprising given the data reported for whole liver from different individuals. It should be noted that the range of detected genes in primary human hepatocytes (690, 1939 and 951) was much wider than that reported for whole human liver (1278–1995) (Yano et al., 2001). 19.3.5

HepG2 Cells

Cell lines developed from hepatocellular hepatomas are commonly used by many investigators in toxicological studies. One of the most frequently used is HepG2. This cell line was derived from the liver carcinoma of a Caucasian male (Knowles et al., 1980) and has been shown to exhibit limited phase I and phase II metabolic function (Dawson et al., 1985; Sassa et al., 1987). There have been two studies reporting comparisons of basal gene expression patterns in multiple replicates of HepG2 cultures. Sonna et al. (2003) found that a mean of 5493 ± 130 transcripts was classified as ‘present’ or ‘marginal’ in five replicates of HepG2 grown to about 80 % confluence. Detected sequences appeared to be somewhat variable due to culture conditions since only 3944 genes were detected in all five replicates. It is clear that many more genes are expressed in HepG2 (5493 ± 130) than in human liver (2418) (Yano et al., 2001) (2800) (Delongchamp et al., 2005). Expression data on 234 stress and toxicology genes were obtained from six replicates of HepG2 cells (Morgan et al., 2002). Expression of many genes varied among replicates 5–12 fold,


497

including CYP4A11, extracellular superoxide dismutase (SOD3), multi-drug resistance gene (MDR3), microsomal UDP-glucuronylsyltransferase (UDP-GT), heme oxygenase 2 (HO2) and flavin-containing monooxygenase 4 (FMO4).

19.4

Comparisons of in vivo and in vitro

An important component of evaluating in vitro data is how predictive it is of toxic responses in vivo. To a large extent, this will be dependent on similarities and differences in basal gene expression. Since the advent of cDNA arrays, many investigators have used them very effectively to compare and contrast the similarities and differences in gene expression among various in vitro systems as well as to whole liver. 19.4.1

Comparisons in Rodents

A comprehensive comparison of gene expression has been conducted for whole liver, liver slices and primary hepatocytes in both monolayer and sandwich culture from male HanBrl:WIST rats (Boess et al., 2003). Gene expression in liver slices after 6 h in culture was found to be highly correlated with that in whole liver (r 2 = 0.95). Of all of the tested systems, basal gene expression in cultured liver slices was most similar to whole liver, particularly earlier in culture (r 2 = 0.95 at 6 h vs. r 2 = 0.79 at 24 h). Jessen et al., (2003) also found that gene expression in liver slices from male Sprague–Dawley rats was highly correlated with that in whole liver (r 2 = 0.79 at 24 h). Of the 190 genes expressed in the liver, but not in the heart, kidney, testes or lung, 60 were expressed in the liver, but not in liver slices. Among these were CYP1A1, CYP1A2, CYP3A1, steroid 5 α-reductase and insulin-like growth factor binding protein. None of the interrogated liver-specific genes was expressed in liver slices but not in whole liver. There were a total of 3305 genes detected in whole liver, 2986 in primary hepatocytes and 2795 in liver slices (Jessen et al., 2003). Boess et al. (2003) also compared the basal gene expression of primary hepatocytes isolated from male HanBrl:WIST rats when cultured on collagen in monolayer or sandwich configuration 6 or 24 h after an overnight preculture to that of whole liver. As with the liver slices, primary hepatocytes cultured on collagen in monolayers or in sandwich configuration were more similar to whole liver earlier (r 2 = 0.78 and r 2 = 0.73, respectively, at 6 h) in culture than later (r 2 = 0.68 and r 2 = 0.68, respectively, at 24 h), although neither configuration was superior in simulating whole liver gene expression. Jessen et al. (2003) reported a slightly higher correlation coefficient in primary hepatocytes from Sprague– Dawley rats cultured on collagen for 24 h (r 2 = 0.80) with that in whole liver. Genes with increased expression in cultured hepatocytes, compared to whole liver and liver slices, included alpha-2-macroglobulin, guanylate cyclase (enterotoxin receptor), dig-1 (leukotriene B4 inactivation) and p21/waf (Boess et al., 2003). P4501A1 was expressed in cultured hepatocytes, but not in whole liver or liver slices (Boess et al., 2003; Jessen et al., 2003). Decreased expression of several isoforms of sulfotransferase, as well as apolipoprotein B, and insulin-like growth factors were found in cultured hepatocytes, but not in liver slices or whole liver (Boess et al., 2003; Jessen et al., 2003). In another study, the basal gene expression of primary hepatocytes from Sprague– Dawley rats cultured on collagen for about 48 h was compared to that of whole liver

498

Hepatotoxicity

(Schuppe-Koistinen et al., 2002). Unlike the data reported by Jessen et al., (2003), more of the screened genes were expressed in primary hepatocytes (535 genes) than in whole liver (432 genes). This may reflect differences due to time in culture (6 and 24 hours compared to 48 h). Of the 535 genes detected in primary hepatocytes, 133 had decreased expression and 224 genes had increased expression compared to whole liver. These data indicate that about 67 % of the genes expressed in primary hepatocytes following two days in culture have expression levels that differ from that found in whole liver. The majority have increased expression. The inappropriate expression of non-liver genes in cultured hepatocytes is common, more so than loss of expression of liver specific genes. Boess et al., (2003) compared gene expression in two rat hepatic cell lines BRL3A and NRL clone 9 to whole rat liver. Unlike liver slices and primary hepatocytes, the correlation of expression between BRL3A and NRL clone 9, and whole liver is very low (r 2 = 0.09 BRL3A, r 2 = 0.10 NRL, respectively). Many liver-specific transcripts are substantially decreased in rat hepatoma cell lines, including phase I and phase II metabolic genes as well as genes involved in the production of plasma proteins. The expression of over 1400 transcripts is altered more than 10-fold compared to expression in whole liver.

19.4.2

Comparisons in Human(s)

As described previously for rodents, primary human hepatocytes are routinely plated on collagen-coated plates for use in toxicological testing. Unlike rodent cells, human hepatocytes do not plate efficiently. Unattached cells are commonly found suspended in the media 24 h after culturing. The question has been, what are the differences, if any, between these two populations of cells? Are the suspended hepatocytes a subpopulation of the attached hepatocytes? Are suspended hepatocytes a mixture of other cells types present in whole liver which do not attach as well? Are there significant differences in the gene expression profiles of suspended hepatocytes and those that attach? Preliminary data have been reported that address some of these questions. In one study, primary human hepatocytes were isolated from the liver of four human donors (Table 19.3) and plated on culture dishes coated with collagen I (Waring et al., 2003). After 24 h, the medium containing unattached hepatocytes was removed. The viability of the suspended hepatocytes was over 95 % using the trypan blue method and therefore lack of attachment did not appear to be due to cell death. The basal gene expression of the attached and the corresponding suspended hepatocytes isolated from three individual donors was compared. Gene expression was also compared to that of whole liver in one donor. The expression of 29 genes was increased in attached human hepatocytes compared to hepatocytes in suspension from each of three donors compared. These genes include those involved in cell cycle regulation, DNA synthesis, metabolism, protease inhibitors, protein synthesis and cellular structure. The expression of only one gene was increased in suspended cells compared to attached cells. Interestingly, this gene, macrophage scavenger receptor 1, has been previously reported to be only expressed in Kupffer cells and sinusoidal endothelial cells (Hughes et al., 1995). The authors suggest there may be hepatic cells other than hepatocytes in the suspended population. Gene expression was more highly variable in the suspended hepatocytes from the three human donors than in attached hepatocytes from the same donors, suggesting that the suspended population was more heterogeneous


499

than the attached hepatocytes. Specific information about the level of variability was not provided. Increased expression of 33 genes, ranging 2–7 fold, was found for both attached and suspended hepatocytes. Fourteen of these genes had unknown function; however, expression of macrophage scavenger receptor 1 was reported to be increased in both attached and suspended hepatocytes compared to whole liver. The expression of 132 additional genes was increased in primary hepatocytes cultured on collagen and of 51 genes in suspended hepatocytes compared to whole liver. Decreased expression of 56 genes was found in both attached and suspended hepatocytes compared to whole liver. These genes included many liver-specific genes, such as those involved in immune response (complement proteins) and metabolism (alcohol dehydrogenase, apoplipoproteins, cytochrome P450s, serum amyloid, etc.). These data are from one donor only. However, it is evident from these data, that the population of hepatocytes that do not attach during plating of hepatocytes on collagen I, are a particularly heterogeneous group of hepatocytes that are not appropriate for routine toxicological testing. As noted in previous sections, HepG2 is a commonly used cell line in toxicological studies. The transcriptome from this human hepatoma line has been compared to that of primary human hepatocytes cultured on matrigel (Harris et al., 2004). Hepatocytes from three male donors were used in the comparison (Table 19.3). The expression of 2974 genes was detected in HepG2 cells. This number is substantially lower than the 5493 ± 130 genes reported by other authors (Sonna et al., 2003) and may reflect differences in culture conditions and experimental differences in cDNA array analysis. Of the detected genes, 867 (or about 30 %) were also expressed in cultured hepatocytes from all three human donors; however, 920 genes were expressed in HepG2 cells that were not detected in the hepatocytes from any donor. These data are very similar to that seen in comparison of rat hepatoma lines to rat whole liver (Boess et al., 2003). Although some proportion of the genes expressed in HepG2 that are not detectable in primary hepatocytes may be expressed in human liver but downregulated during culture, it is likely that many are unique to HepG2 and reflect the non-differentiated state of the cell line. The human hepatoma cell line B16A2 has been shown to express higher levels of some phase I enzymes, including CYP2E1 and CYP3A, than are found in HepG2. B16A2 cells have also been reported to express liver-enriched transcription factors (LETFs) which are necessary for maintenance of the normal hepatic expression of many important metabolic genes (Gomez-Lechon et al., 2001; Glaise et al., 1998; Le Jossic et al., 1996). It has also been shown that B16A2 cells become more differentiated or more hepatocyte-like once the cells reach confluence (Gomez-Lechon et al., 2001; Le Jossic et al., 1996). In order to evaluate this more fully, cDNA arrays were used to simultaneously compare the basal gene expression of B16A2 cells at zero, two and five weeks after reaching confluence in culture to HepG2 cells and to human liver (Butura et al., 2004). Three sets of pooled mRNA from three donors each (pooled human liver A–C) were used for analysis (Table 19.3). Of about 10 000 evaluated genes, there was detectable expression of 2365 genes in B16A2 cells five weeks after confluence, 2236 genes in HepG2 cells, and 2332 genes in mRNA from a pool of three human livers. The expression of 2025 genes was common to all three systems. There were 152 genes expressed in B16A2 cells that were not expressed in either HepG2 or pooled human liver, 162 genes expressed in pooled human liver not seen in HepG2 or B16A2, and only 42 genes expressed in HepG2 not expressed in B16A2 and pooled human

500

Hepatotoxicity

liver. Interestingly, expression of LETFs was not significantly different among HepG2, B16A2 and human liver. Three different gene clusters were identified in cultured B16A2 cells using hierarchical clustering. Cluster one contained 47 genes that had increased expression in B16A2 over time in culture, but were still expressed at much higher levels in pooled human liver. This cluster included serum albumin, alcohol dehydrogenase (alpha, beta and gamma), CYP3A4 and CYP2E1. Even though expression levels were increased significantly (discussed more fully below), the mRNA was consistently less than 10 % of that seen in human liver. Cluster two contained 33 genes with temporal increase in mRNA at confluence for two and five weeks. The expression of genes in cluster two was higher in B16A2 cells than in pooled human liver. Included in this gene set were ribosomal proteins, proteases and the multidrug resistance protein 1. Cluster three contained 72 genes that had expression levels that decreased over time in B16A2 cells in culture to levels similar to those seen in pooled human liver. Basal gene expression of normal liver tissue has also been compared to that of the human hepatoma cell line BEL-7402 (Liu et al., 2003). This cell line was created in 1974 from a Chinese patient with hepatocellular carcinoma. Altered gene expression was defined as genes with ≥ 1.5-fold change or a difference in expression of ≥ 5000. Using those criteria, 48 of the 588 interrogated cancer genes had expression levels that were altered from that in whole human liver. In general, up-regulation of cancer progression genes and downregulation of anti-cancer progression genes was seen. The relatively small differences seen in whole liver and the BEL-7402 cell line compared to that seen in other cell lines (Boess et al., 2003) may be partially due to the gene set selected for screening.

19.5

Effects of Culture Conditions

Gene expression in vitro is highly dependent on culture conditions and the length of time in culture. DNA arrays have been used effectively to identify parameters in culture that enhance expression of liver-specific genes in vitro and/or sustain hepatic gene expression most closely to that seen in vivo. Considerable research has been conducted to determine culture conditions that mimic in vivo conditions in vitro . The primary focus of these efforts has been to identify an appropriate matrix on which to culture primary hepatocyte and to develop culture media which support and enhance normal hepatocyte function. Some investigators have used DNA arrays to identify optimal conditions for culture of primary hepatocytes and for use of hepatoma cell lines, and to investigate temporal effects of hepatic cells in culture. 19.5.1

Matrix and Temporal Comparisons

The gene expression of primary rat hepatocytes from male Fischer 344 rats cultured as monolayers on matrigel or collagen was compared to gene expression in freshly isolated hepatocytes (Harris et al., 2003). Gene expression was assessed both 16 h and 72 h after plating. In general, after 16 h there was no substantial difference in the use of matrigel or collagen (R = 0.69 and R = 0.72, respectively) when compared to freshly isolated hepatocytes; however, after 72 h the use of matrigel appeared to suppress expression of genes


501

associated with a de-differentiated phenotype. Several genes with altered gene expression in hepatocytes cultured on collagen compared to freshly isolated hepatocytes were the same genes identified by other investigators when comparing genes expressed in primary rat hepatocytes cultured on collagen (Boess et al., 2003) and in primary rat hepatocytes cultured on the plastic Petri plate matrix (Baker et al., 2001). Baker et al., (2001) compared the expression of primary rats hepatocytes isolated from male Fisher 344 rats 4, 12, 24, 48 and 72 h post-plating. The hepatocytes were cultured on plastic dishes with no added culturing matrix. Detectable expression of about 28 % of the approximately 8700 evaluated genes (27.7 ± 2.3 % or about 2410 genes) was found in the cultured rat hepatocytes. Altered expression was defined as a greater than 2-fold change in expression compared to the freshly isolated hepatocytes (pellet). In general, as time in culture increased, the number of genes with altered expression also increased. For example, the mRNA levels of 170 genes were higher in cells plated for 4 h when compared to expression in the hepatocyte pellet; however, there were 468 genes with increased expression after 72 h in culture. Expression of phase I metabolic genes was generally decreased over time, however; there were gene-specific differences in the temporal changes. Expression of CYP1 family genes was decreased 24 to 48 h post-plating while mRNA from the CYP2 and CYP4 genes families was demonstrably less after only 12 h. More heterogeneity was seen in expression of phase II metabolic genes. For example, six genes involved in glucuronidation of xenobiotics had increased expression ranging 4–17.8 fold, with only two showing modest decreases in expression levels. Other phase II conjugation pathways had a similar pattern. One gene involved in sulfation of xenobiotics had increased expression after plating (rat mRNA for ST1B1); however, eight had decreased expression. Two glutathione-related genes had increased expression (GST P subunit pi, GST yb subunit mu) while seven were decreased. In addition to phase I and phase II metabolic pathways, there was a decrease in expression of genes involved in gluconeogenesis. These changes, as well as those in the cytoskeletal and extracellular matrix pathways, may be due to culture conditions to which the hepatocytes must adapt. Hierarchical cluster analysis was used to group the genes with altered expression into 64 different clusters. The expression of genes involving several metabolic and structural pathways was significantly affected. For instance, there was an increase in expression of more than 40 genes coding for cytoskeletal and extracellular matrix components. The genes with the most substantial and consistent increase in expression over time were annexin II (20 – fold higher at 72 h), vascular a-actin (12.7 - fold higher at 72 h) and cytokeratin 8 (11.1 - fold higher at 72 h). A summary of genes which have been found to have altered expression in primary hepatocytes cultured on collagen or plastic compared to expression in whole liver in four independent studies is shown in Tables 19.4 and 19.5 (Boess et al., 2003; Harris et al., 2003; Jessen et al., 2003; Baker et al., 2001). Temporal comparisons and matrix comparisons have also been made in hepatoma cell lines. As discussed in a previous section, the B16A2 hepatoma line has been reported to acquire a more differentiated state once the cells become confluent (Gomez-Lechon et al., 2001; Le Jossic et al., 1996). In order to determine what changes in gene expression might be responsible for the more highly differentiated phenotype once the cells are confluent, the gene expression profiles of B16A2 cells at zero, two or five weeks after confluence

502

Hepatotoxicity Table 19.4 Genes with increased expression in primary rat hepatocytes cultured on collagen or plastic Gene

Citation

Vimentin

Baker et al., 2001; Harris et al., 2003 Boess et al., 2003; Harris et al., 2003 Baker et al., 2001; Harris et al., 2003 Baker et al., 2001; Harris et al., 2003 Baker et al., 2001; Harris et al., 2003 Baker et al., 2001; Harris et al., 2003

Heme oxygenase 1 Bcl-x Mn superoxide dismutase G1/S-specific cyclin D3 Erk1 Glutathione S-transferase P subunit (72 h) Actin

Baker et al., 2001; Harris et al., 2003 Baker et al., 2001; Harris et al., 2003

were compared (Butura et al., 2004). About 10 000 genes were assessed by cDNA array analysis. After five weeks at confluence, there was detectable expression of 2365 genes. Of the interrogated genes, 152 genes were noted to have altered expression at different time points during culture (P = 0.001). Hierarchical cluster analysis of the gene expression data was used to group genes with altered expression. Increased expression of 80 genes was identified after two and five weeks Table 19.5 Genes with decreased expression in primary rat hepatocytes cultured on collagen or plastic Gene G1/S-specific cyclin D1 Catalase Sodium/bile acid cotransporter Glutathione S-transferase Ya subunit Glutathione peroxidase Insulin-like growth factor binding protein Sulfotransferases

Citation Baker et al., 2001; Boess et al., 2003; Harris et al., 2003 Baker et al., 2001; Harris et al., 2003 Boess et al., 2003; Harris et al., 2003 Baker et al., 2001; Harris et al., 2003 Baker et al., 2001; Harris et al., 2003 Boess et al., 2003; Jessen et al., 2003 Boess et al., 2003; Jessen et al., 2003


503

of confluent culture. Genes with higher levels of expression included those involved in lipid and protein synthesis (i.e. hyaluronin binding protein 2, sphingomyelin phosphodiesterase), inflammatory response (i.e. complement C5 precursor, complement C6 precursor), integral membrane proteins (i.e. ERBB-3 receptor), serine proteases (i.e. complement factor 1 precursor, serine protease HTRA1), ribosomal proteins (i.e. 40S and 60S), various proteases (i.e. aspartic proteases and metalloproteases) and multidrug resistance protein I. In addition, increased expression of many liver-specific or liver-enhanced genes was seen over time. The expression of albumin increased roughly 33 – fold. Expression of the mRNA for three different isoforms of alcohol dehydrogenase increased 20-30 fold, CYP2E1 increased 7-fold and CYP3A4 increased 8-fold after five weeks in confluent culture. Significant changes in gene expression will ultimately result in significant changes in important metabolic function. Analysis of these data as defined by KEGG (http://www.genome.ad.jp/kegg/) indicate that these changes would result in increased fatty acid metabolism (i.e. alcohol dehydrogenase 1A class 1), tyrosine metabolism (i.e. alcohol dehydrogenase 1C class 1 and monoamine oxidase B) and bile acid biosynthesis (i.e. alcohol dehydrogenase 1B class I, bile acid coenzyme A amino acid N -acyltransferase), all of which represent important liver metabolic functions. In addition to enhanced gene expression, there was a decrease in expression of 72 genes that are normally expressed at relatively high levels in B16A2. Not surprisingly, the types of genes with decreased expression represented different functional pathways than those that were increased. Metabolic pathways with a predicted decrease in function were the Wnt-signaling pathways (i.e. Frizzled-2, increase in inhibitor Dickkopf protein-1), sterol biosynthesis (i.e. geranyltransferase, lanosterol synthase) and integrin-mediated cell adhesion (i.e. calpain 2, p21 activated kinase). Lower expression of genes coding for membrane transport proteins (i.e. monocarboxylate transporter-5, l-type amino acid transporter 1), signal transduction proteins (i.e. regulator of G-protein signaling 20), unnamed proteins involved in secretion/trafficking of proteins, annexins and cytoskeletal proteins (i.e. actinbinding LIM protein 1, tropomyosin-1 fibroblast isoforms TM3), sugar-alcohol metabolism (i.e. transaldolase, aldose reductase) and unnamed cancer antigens was found. Gene expression changes associated with culture of the human liver cancer cell line, Huh 7 as spheroids on polyurethane foam have also been analyzed using cDNA array techniques (Yamashita et al., 2004). An additional comparison was made of the Huh 7 cell line in monolayer culture in medium containing the histone deacetylase inhibitor, trichostatin A (TSA). Both of these techniques had been previously shown to increase liverspecific functions, including synthesis of albumin and maintenance of ammonia removal by cultured cells. The gene expression data from both culture modifications after three days in culture were compared to that of Huh cells grown as monolayers. Altered gene expression was found in 1.72 % and 2.89 % of the 1281 screened cDNAs in the Huh 7 cells grown as spheroid culture and in monolayer with TSA, respectively, when compared to Huh 7 cells grown in monolayer. The expression of 7 genes was increased and of 15 genes was decreased in Huh 7 cells grown in spheroid formation compared to increased expression of 10 genes and decreased expression of 27 genes in Huh 7 cells grown in monolayer with TSA. Gene expression changes were defined as those with at least a 2-fold change in expression. Although both culture alterations were used to increase the differentiated state of the cell line, there were only a limited number of common genes affected. The expression of POU5F1, which codes for a transcription factor, was increased slightly over 2-fold in both cultures. The expression of SCYA17, which codes for a small chemokine,

504

Hepatotoxicity

and EGR1, which codes for a transcription factor, was decreased in both cultures. Several phase I and phase II metabolic genes had decreased expression in one or the other cultures. These data do not indicate that either culture condition increases the liver differentiation in Huh 7 cells; however, they do demonstrate the utility of gene arrays to quickly compare the effects of small alterations in culture conditions on the expression of hundreds of genes. 19.5.2

Media Changes

An interesting experiment in media replacement during culture of HepG2 cells indicated that there is a surprising level of apparently coordinated changes in expression of certain genes over 6 h in culture (Morgan et al., 2003). Sampling was performed at 5 min intervals for 6 h after media replacement in HepG2 cells seeded at a constant density in collagen-coated dishes. Three basic patterns were seen: quiescent genes were ‘turned on’ and expressed at high levels (CYP1A1), some genes were expressed at regular or ‘clockwork’ intervals (asparagine synthetase) or genes suddenly exhibited a ‘burst’ of expression (TIM-timeless homolog). Even more interesting was that groups of genes appeared to have apparently synchronous patterns of expression for periods of time. For instance, for about 350 min, the clockwork up- and down-expression of asparagine synthetase was almost exactly concordant with that of cyclohydrolase, even though their expression levels were not the same. The mRNA cycle for these genes was estimated to be about 36.8 min. Seven genes were identified that had coordinated ‘bursts’ of expression over a 350 min period. These genes were TIM, ICSB, IRF2, hDMP1, CDH2, Homer 3 and PMS2L1. These data highlight the importance of understanding the effects of culture conditions on in vitro systems and also powerfully demonstrate the strength of using DNA arrays to rapidly assess the changing dynamics of global gene expression within a cell culture.

19.6

Summary

A substantial amount of gene expression data has been generated by a number of investigators to evaluate and compare the numerous hepatic testing systems available for toxicological testing. Although it is beyond the scope of this chapter to describe the data in detail, it is apparent that these reports can be used to determine what system is appropriate for the toxicological endpoints of interest. For instance, although there are significant differences in gene expression between hepatoma cell lines and whole liver, use of the B16A2 hepatoma line could be useful for evaluation of the metabolites of chemicals or drugs metabolized by specific P450s like CYP2E1 or CYP3A4, once the cells become confluent. Although the levels of these metabolizing enzymes are still well below those found in vivo, they are still a useful ‘quick and inexpensive’ screen for possible metabolites. Gene array data can be reanalyzed in numerous ways in order to assist an investigator in making the proper decisions regarding experimental design.

References Baker TK, Carfagna MA, Gao H, Dow ER, Qingqin L, Searfoss GH and Ryan RP (2001). Temporal gene expression analysis of monolayer cultured rat hepatocytes. Chem Res Toxicol 14:1218–1231.


505

Boess F, Kamber M, Romer S, Gasser R, Muller D, Albertini S and Suter L (2003). Gene expression in two hepatic cell lines, cultured primary hepatocytes and liver slices compared to the in vivo liver gene expression in rats: Possible implications for toxicogenomics use of in vitro systems. Toxicol Sci 73: 386–402. Bono H, Yagi K, Kasukawa T, Nikaido I, Tominaga N, Miki R, Mizuno Y, Tomaru Y, Goto H, Nitanda H, Shimizu D, Makino H, Morita T, Fujiyama J, Sakai T, Shimoji T, Hume DA, Hayashizaki Y and Okazaki Y (2003). Systematic expression profiling of the mouse transcriptome using RIKEN cDNA microarrays. Genome Res 13:1318–1323. Boorman GA, Irwin RD, Vallant MK, Gerken DK, Lobenhoffer EK, Hejtmancik MR, Hurban P, Brys AM, Travlos GS, Parker JS and Portier CJ (2005). Variation in the hepatic gene expression in individual male Fischer rats. Toxicol Pathol 33:102–110. Butura A, Johansson I, Nilsson K, Wärng˚ard L, Ingelman-Sundberg M and Schuppe-Koistinen I (2004). Differentiation of human hepatoma cells during confluence as revealed by gene expression profiling. Biochem Pharmacol 67:1249–1258. Dawson JR, Adams DJ and Wolf CR (1985). Induction of drug metabolizing enzymes in human liver cell line HepG2. FEBS Lett. 183:219–222. Delongchamp RR, Velasco C, Dial S and Harris AJ (2005). Genome-wide estimation of gender differences in the gene expression of human livers: Statistical design and analysis. BMC Bioinformatics 6(Suppl 2): S13. George J, Goodwin B, Liddle C, Tapner M, Farrell GC (1997). Time-dependent expression of cytochrome P450 genes in primary cultures of well-differentiated human hepatocytes. J Lab Clin Med 129: 638–648. Glaise D, Ilyin GP, Loyer P, Cariou S, Bilodeau M, Lucas J, Puisieux A, Ozturk M and GuguenGuillouzo C (1998). Cell cycle gene regulation in reversibly differentiated new human hepatoma cell lines. Cell Growth Differ 9:165–176. Gomez-Lechon MJ, Donato T, Jover R, Rodriguez C, Ponsoda X, Glaise D, Castell JV and GuguenGuillouzo C (2001). Expression and induction of a large set of drug-metabolizing enzymes by the highly differentiated human hepatoma cell line BC2. Eur J Biochem 268:1448–1459. Harris AJ, Dial S, Shaddock J, Dragan Y and Casciano D (2003). The effects of substratum on gene expression in cultured primary rat hepatocytes. Toxicol Mechs Meth 14:257–270. Harris AJ, Dial SL and Casciano DA (2004). Comparison of basal gene expression profiles and effects of hepatocarcinogens on gene expression in cultured primary human hepatocytes and HepG2 cells. Mutat Res 549:79–99. Hughes DA, Fraser IP and Gordon S (1995). Murine macrophage scavenger receptor: in vivo expression and function as receptor for macrophage adhesion in lymphoid and non-lymphoid organs. Eur J Immunol 25:466–473. Jessen BA, Mullins JS, de Peyster A and Stevens GJ (2003). Assessment of hepatocytes and liver slices as in vitro test systems to predict in vivo gene expression. Toxicol Sci 75:208–222. Knowles BB, Howe CC and Aden DP (1980). Human hepatocellular carcinoma cell lines secrete the major plasma proteins and hepatitis B surface antigen. Science 209:497–499. LeCluyse EL, Fix JA, Audus KL and Hochman JH (2000). Regeneration and maintenance of bile canicular networks in collagen-sandwiched hepatocytes. Toxicol In Vitro 14:117–132. Le Jossic C, Glaise D, Corcos L, Diot C, Dezier JF, Fautrel A and Guguen-Guillouzo C (1996). transActing factors, detoxication enzymes and hepatitis B virus replication in a novel set of human hepatoma cells Lines. Eur J Biochem 238:400–409. Liu L-X, Jiang H-C, Liu Z-H, Zhu A-L, Zhou J, Zhang W-H, Wang X-Q and Wu M (2003). Gene expression profiles of hepatoma cell line BEL-7402. Hepato-Gastroenterology 50:1496– 1501. Longueville F, Atienzar FA, Marcq L, Dufrane S, Evrard S, Wouters L, Leroux F, Bertholet V, Gerin B, Whomsley R, Arnould T, Remacle J and Canning M (2003). Use of a low-density

506

Hepatotoxicity

microarray for studying gene expression patterns induced by hepatotoxicants on primary cultures of rat hepatocytes. Toxicol Sci 75:378–392. Morgan KT, Ni H, Brown HR, Yoon L, Qualls Jr CW, Crosby LM, Reynolds R, Gaskill B, Anderson SP, Kepler TB, Brainard T, Liv N, Easton M, Merrill C, Creech D, Sprenger D, Conner G, Johnson PR, Fox T, Sartor M, Richard E, Kuruvilla S, Casey W and Benavides G (2002). Application of cDNA microarray technology to in vitro toxicology and the selection of genes for a real-time RT-PCR based screen for oxidative stress in HepG2 cells. Toxicol Pathol 30: 435–451. Morgan KT, Casey W, Easton M, Creech D, Ni H, Yoon L, Anderson S, Qualls CW, Crosby LM, MacPherson A, Bloomfield P and Elston TC (2003). Frequent sampling reveals dynamic responses by the transcriptome to routine media replacement in HepG2 cells. Toxicol Pathol 31:448–461. Novak JP, Sladek R and Hudson TJ (2002). Characterization of variability in large-scale gene expression data: Implications for study design. Genomics 79:104–113. Oldham JW, Casciano DA and Farr JA (1979). The isolation and primary culture of viable, nonproliferating rat hepatocytes. TCA Man 5:1047–1050. Pritchard CC, Hsu L, Delrow J and Nelson PS (2001). Project normal: Defining normal variance in mouse gene expression. PNAS 98:13266–13271. Sassa S, Sugita O, Galbraith RA and Kappas A (1987). Drug metabolism by the human hepatoma cell, HepG2. Biochem Biphys Res Commun 14:52–57. Schuppe-Koistinen I, Frisk AL and Janzon L (2002). Molecular profiling of hepatotoxicity induced by an aminoguanidine carboxylate in the rat: gene expression profiling. Toxicology 179:197–219. Shaddock JG, Feuers RJ, Chou MW, Swenson DH and Casciano DA (1995). Genotoxicity of Tacrine in primary isolated hepatocytes from B6C3F1 mice and aged ad libitum and calorie restricted Fisher 344 rats. Mutat Res 344:79–88. Sonna LA, Cullivan ML, Sheldon HK, Pratt RE and Lilly CM (2003). Effect of hypoxia on gene expression by human hepatocytes (HepG2). Physiol Genomics 12:195–207. Tomascik-Cheeseman LM, Coleman MA, Marchetti F, Nelson DO, Kegelmeyer LM, Nath J and Wyrobek AJ (2004). Differential basal expression of genes associated with stress response, damage control, and DNA repair among mouse tissues. Mutat Res 561:1–14. Yamashita Y, Shimada M, Harimoto N, Tanaka S, Shirabe K, Ijima H, Nakazawa K, Fukuda J, Funatsu K and Maehara Y (2004). cDNA microarray analysis in hepatocyte differentiation in Huh 7 cells. Cell Transplant 13:793–799. Yano N, Habib NA, Fadden KJ, Yamashita H, Mitry R, Jauregui H, Kane A, Endoh M and Rifai A (2001). Profiling the adult human liver transcriptome: analysis by cDNA array hybridization. J Hepatol 35:178–186. Waring JF, Ciurlionis R, Jolly RA, Heindel M, Gagne G, Fagerland JA and Ulrich RG (2003). Isolated human hepatocytes in culture display markedly different gene expression patterns depending on attachment status. Toxicol In Vitro 17:693–701. Westlind A, Löfberg L, Tindberg N, Andersson TB and Ingelman-Sundberg M (1999). Interindividual differences in hepatic expression of CYP3A4: Relationship to genetic polymorphism in the 5’upstream regulatory region. Biochem Biophys Res Commun 259:201–205.

20 Prediction of Hepatotoxicity Based on the Toxicogenomics Database Tetsuro Urushidani

20.1

Introduction

Today, in the post-genomic era, there have been remarkable advances in the technology of searching for seeds of new drugs. However, the success rate of drug development is nevertheless decreasing not only in Japan but also worldwide. According to the investigation performed by the Japan Pharmaceutical Manufacturers Association (http://www.jpma.or.jp/12english/publications/index.html), 422 653 seeds for new drugs were synthesized or extracted from 1996 to 2000 in Japan, whereas only 63 were approved. Moreover, half of those approved were chemicals created by and purchased from foreign countries. It is obvious that the opportunity to create an original drug is now extremely low. Drug development in the previous century was usually based on screening by measuring effects of the chemicals in model animals with artificially created diseases, and subsequently it sometimes happened that a quite excellent drug was produced not for humans but for rats. In recent years, however, it has been possible to start the development by targeting disease-related genes whose molecular functions are well elucidated, and indeed, human-type genes are always available. Therefore, it is now easy to select a chemical, which is effective on the human-type molecule for at least the in vitro level. Even with this advantage, many candidate drugs have been eliminated largely because of toxicity, which could not be found in pre-clinical tests but appeared in clinical trials (Ismail and Landis, 2003). This not only brings about an increment in the cost for drug development that will be shifted to the medical expense, but also causes a serious ethical problem that the toxicity of the chemical is proven by using humans. From another


508

Hepatotoxicity

point of view, the low predictivity of present toxicological tests for clinical toxicity also means that many drug candidates might have been abandoned although they were in fact excellent. The current toxicological tests are much improved and controlled to assure reliability of the data in comparison to past adverse cases, such as thalidomide. However, reliability of the pre-clinical data does not necessarily lead to clinical safety. When the drug possesses an obvious organ or cellular toxicity as part of its pharmacological effects, its toxicity can be easily predicted from its dose–response relationship and thus controlled. However, many of the practical side effects in the clinical field are scarcely related to the pharmacological target and the origin of the toxicity is usually not the drug itself but its metabolite(s). If a serious adverse effect happened in the clinical dose range at a rate of 1/1000 (an impractically large number), such a drug should be withdrawn from the market at once. However, it is theoretically impossible to detect the effect with such a low incidence using 10 000 rats or more, even though there is no species difference, and this is, of course, still an impractical number for safety tests. Needless to say, the barrier of species difference is quite high and we sometimes encounter a case that there is a difficulty to find out an animal species accurately reproducing a clinically evident toxicological profile. In practical toxicological tests, we inevitably employ the hopeless assumption that any pathophysiological changes in the experimental animals associated with the over-dosage of the test drug can be extrapolated to the corresponding adverse effects with low incidence in humans at low dose. Many conscientious toxicologists feel that such a preclinical safety evaluation is not a certification but an excuse. In order to ‘predict the unpredictable toxicity’, the most realistic strategy would be ‘toxicogenomics’ which enables us to analyze all of the gene expression changes caused by an external stimulus in an organism. Even when an ‘unexpected’ hazard is elicited by drug administration at a very low incidence, its causal events should have occurred and some of them should be detectable as a subtle change. If all of the events were observed and their relationships were analyzed, a perfect prediction would be realized. Observation of all of the physiological processes had been thought to be practically impossible, but advances in biotechnology have made it possible. Since the most sensitive change among biological processes is that of mRNA, and many of the physiological responses are expected to be associated with changes in the expression level of mRNA, the most promising measure is the amount of mRNA. The development of microarray technology has made it possible to substantially quantify the expression level of all of the genes at once. Under such a circumstance, the Ministry of Health, Labour and Welfare, National Institute of Health Sciences (NIHS) and the working group of the Japan Pharmaceutical Manufacturers Association planned ‘The Toxicogenomics Project (TGP)’, a collaborative project of the government and private companies (Urushidani and Nagao, 2005). This is a 5-year project from 2002 to 2007, and after the rearrangements of the organization and mergers of the companies, the members are the NIHS, 15 pharmaceutical companies (Astellas, Chugai, Daiichi, Dainippon-Sumitomo, Eisai, Kissei, Mitsubishi, Mochida, Ono, Otsuka, Sankyo, Sanwa, Shionogi, Takeda and Tanabe) and the National Institute for Biomedical Innovation (NIBIO) as the core institute where the actual work is being performed. About half of the budget is from a grant of the Ministry of Health, Labour and Welfare, and the remaining is from the companies.

Prediction of Hepatotoxicity Based on the TGP Database

20.2

509

The Features of The Toxicogenomics Project (TGP)

The plan of the project at the beginning was as follows. The main goal is to create a gene expression database of 150 chemicals, mainly medical drugs. The target organ is mainly the liver. The reason was that most of the clinically serious adverse effects have occurred in this organ, and that the composition of the cell type is relatively homogenous in this organ and thus the expected variation based on differences of sampling would be minimal. Subsequently, the liver was considered to be favorable to accumulate the know-how of the toxicogenomics technique. Nephrotoxicity was of course considered to be important and the sampling and pathological examination of the kidney in addition to liver was to be performed in all of the animals. Among the animals showing nephrotoxic phenotypes, up to 20 % of the total compounds were subjected to transcriptome analysis of the kidney. For the test animal, 6 week- old male Spraque–Dawley rats were employed. At the starting point, the rat genome was not fully analyzed and there was a firm opinion that the mouse should be used because of the enrichment of the genome information. However, all of the classical toxicological tests had been done with rats and there was a vast amount of knowledge accumulated. When we were venturing into a totally new field, we believed that the accumulated knowledge should be used as much as possible. Another advantage was that more data, such as biochemistry and hematology, could be obtained from the rat rather than from the mouse. The biggest problem at the start of the TGP was the fact that we were behind the large pharmaceutical companies and bioventures in Europe and America who had already started similar projects to create databases of toxicogenomics (Porter et al., 2003; Boverhof and Zacharewski, 2006). Therefore, the following strategy was employed to catch up with the preceding projects. (1) Establishment of quantitativity and reproducibility. Data with excellent quantitativity are acquired using the ‘Affymetrix GeneChip’ and a new normalization method based on the externally added spike RNAs proportional to the sample DNA contents (Kanno et al., 2006) was employed. (2) Selection of test compounds. The chemicals, mainly medicinal, contain drug candidates withdrawn from development because of their hepato- or nephrotoxicity, that are supplied from the member companies. (3) Bridging between species. In addition to the in vivo experiments, exposure to the primary culture of rat liver and to human hepatocyte culture was performed. (4) Enrichment of protocol. Gene expression data possess multi-dose, multi-time points that link to various toxicological measures. These points are discussed in this order below. 20.2.1

Establishment of Quantitativity and Reproducibility

The ‘Affymetrix GeneChip’, an oligonucleotide array, is known to be superior to other arrays, such as the ‘Stanford’ type (Wildsmith and Spence, 2003). In the TGP, data with excellent quantitativity are being accumulated under thoroughly optimized and controlled conditions in order to draw their full ability. During the first two years (up to ca. 35 chemicals), the ‘Rat Expression Array 230A’ containing 15 923 probe sets was used while it was then shifted to the ‘Rat Genome 230 2.0 Array’ containing 31 099 probe sets after the version upgraded. The TGP started with employing a new normalization method, ‘percellome’, using an externally added spike RNA (Kanno et al., 2006). This method makes it possible to express

510

Hepatotoxicity

each gene expression as copy numbers per cell (per DNA) by adding external Bacillus subtilis mRNA proportional to the DNA contents in the homogenate. In our database, raw data, per-chip normalized data as well as spike-normalized data, are usable. Per-chip normalization (global normalization) is fundamentally based on the assumption that the total amount of mRNA is constant, and thus the change of each mRNA cannot be precisely estimated when the total transcription is drastically changed (Kanno et al., 2006). In the TGP, the version of ‘GeneChip’ was changed and this made it impossible to make a comparison between different chips based on values with global normalization. Moreover, in the in vitro experiments where extremely high concentrations of the chemicals were applied, drastic changes of total mRNA were often observed. Especially, when a chemical having direct cytotoxicity to hepatocyte was applied, total mRNA was obviously decreased, and many genes, which were actually down-regulated, were estimated as apparently upregulated because of normalization by the reduced value. In analysis of the kidney, the cell composition varies with the particular part of the organ. It was found that not only the members of expressed genes but also the total mRNA amounts differed among cortex, medulla and papilla. When region-specific genes were extracted, it was revealed that global normalization led to a biased conclusion because of the large difference in the total mRNA (Tamura et al., 2006a). Contrary to the above facts, we concluded that there is no problem in the usual analysis based on global normalization, at least in vivo liver data, since total RNA contents were almost unaffected, even at the toxic dose. When the ratio to control value is employed, global normalization is rather superior to spike normalization because the extra procedure introducing an additional error is avoided. In this present paper, analyses are based on global normalization.

20.2.2

Selection of Test Compounds

The TGP started with five representative hepatotoxicants, i.e. acetaminophen, isoniazid, carbon tetrachloride, phenobarbital and valproic acid. The 150 chemicals were selected and their exposure to rats have been completed – these are categorized in Figure 20.1. Although they are somewhat biased, they cover most of the therapeutic categories. At present, the drugs in the Japanese market number about 1500, including those not suitable for transcriptome analysis of the liver or kidney, e.g. drugs for dermatology or ophthalmology, drugs with almost identical structures and anti-cancer drugs whose toxicity to these organs is not a primary problem. Our aim is to create a database of representative medical compounds. In this sense, the number of 150, 10 % of total drugs, is considered to be enough for a ‘textbook of toxicology’. It would be interesting to extend the project toward a structure– toxicity relationship based on the present database. One unique feature of our project is the supply of chemicals from the member companies. These are drug candidates that were withdrawn from the various stages of the development process because of the emergence of toxicity in the liver or kidney. These are usually impossible to obtain, and are quite valuable and interesting samples. The candidates that were withdrawn mean that they were once considered to be hopeful candidates; in other words, their potential toxicity was underestimated in the early stage. Thus, they could be useful for the evaluation of our database as good model cases after our system is established.


511

Number of Chemicals 0

5

10

15

20

25

CNS Anti-inflammatory Peripheral Nerve Cardiovascular Respiratory Digestive Hormone Metabolic Anti-tumor Allergy Antibiotics Toxicants Preclinical

Figure 20.1 Chemicals selected (150 total). The chemicals selected for the Toxicogenomics Project are classified by their category: CNS, drugs for central nervous system; toxicants, not medical drugs but representative chemicals with hepato- or nephrotoxicity, such as carbon tetrachloride, hexachlorobenzene, aryl alcohol, etc; preclinical, drug candidates supplied from the member companies, which were withdrawn in various stages of drug development because of hepato- or nephrotoxicity

20.2.3

Bridging between Species

Even if the perfect prediction of hepatotoxicity is attained in the rat, this does not necessarily mean the improvement of prediction of toxicity in humans. The final goal should be the prediction in humans for drug development. There are too many problems in extrapolation of phenotype from rodent to human. However, if general toxic mechanisms or toxicological pathways are conserved over species, they would obviously be useful for bridging between animal models and clinical events. In the preliminary studies performed before the project, it was found that the gap between hepatocyte cell lines and normal liver were too wide to bridge even if they were both human, and thus normal human hepatocyte culture was employed as the only choice. At present, data collection has been completed for rat primary hepatocytes, whereas that of human hepatocytes has not been done yet, and their comparative analysis is a future subject. In the system of analysis for the TGP database described later, the object is to overview the responsiveness of the test compounds to the biomarker gene lists and make it easy to compare them between rat and human by an automatic ortholog conversion of the marker genes. However, there have been many difficult problems pointed out in the bridging between in vivo and in vitro experiments before solving the species difference (Boess et al., 2003). We also recognized this problem as the in vitro data accumulated. Although our final goal should be the prediction of clinical toxicity, we are now focusing on the increment of the efficacy of preclinical study as a more realistic goal. Namely, the increment of

512

Hepatotoxicity

predictivity of hepatotoxicity in rats is the first important step and thereafter the bridging between species difference is envisioned as the next step. Therefore, this topic is omitted from this present paper. 20.2.4

Enrichment of the Protocol

This point is the largest merit of the TGP and is thus discussed in detail. 20.2.4.1

Standard Protocol and Problems in Statistical Analysis of Gene Expression Data

The protocol employed in the TGP is summarized in Table 20.1. In vivo experiments consist of single and repeated oral administrations to rats (N = 5 for each group; 3 doses + vehicle control) and autopsy is done 3, 6, 9 and 24 h after the single dose and 24 h after repeated dose for 3, 7, 14 and 28 days. Data of blood biochemistry, hematology and histopathology (both liver and kidney) are obtained from all animals and the gene expression analysis in liver (also kidney in some cases) is performed in 3 out of 5 rats. In vitro experiments consist of rat and human hepatocytes (N = 2 for each group; 3 concentrations + vehicle control) and cell harvest is done at 2, 8 and 24 h after exposure. Table 20.1 Standard protocols employed in the Toxicogenomics Project In vivo Animal Vehicle Dose Route Sacrifice Sampling GeneChip analysis Items examined

Sprague–Dawley rat 6 week old N = 5 for each group 0.5 % methylcellulose or corn oil Low, middle, high (1:3:10) Oral (intravenous in a few cases) 3, 6, 9 and 24 h after a single administration 24 h after the last dose of repeated administration for 3, 7, 14 and 28 days Liver, kidney, plasma N=3 Histopathology: liver and kidney Body weight, organ weight (liver and kidney), food consumption Biochemistry, hematology: 37 standard items

In vitro: rat Animal Cell Vehicle Concentration Treatment GeneChip analysis Items examined

Sprague–Dawley rat 6 week old Hepatocyte isolated by collagenase digestion Culture medium or DMSO Low, middle, high (1:5:25) 2, 8 and 24 h Duplicate Cell biability (LDH release and DNA contents)

In vitro: human Cell Vehicle Concentration Treatment GeneChip analysis Items examined

Human frozen hepatocytes Culture medium or DMSO Low, middle, high (1:5:25, low is omitted in some cases) 2, 8 and 24 h Duplicate Cell biability (LDH release and DNA contents)


513

We believe that no other database contains data with such an enriched protocol in the world. Especially the fact that dose-dependency with N = 3 for each time point can be estimated is very powerful in erasing ‘noises’ inevitably associated with the statistical analysis of vast numbers of measurements. Although the price of ‘GeneChip’ is decreasing as it becomes more popular, the cost of the experiments using this device is still so high that either the sample numbers in a group, time points, or dose levels need to be reduced. However, the reduction of the dose levels ruins the analysis. The analysis of microarray data is usually performed by multivariate statistical techniques such as hierarchical clustering, k-means clustering, self-organizing map or principal component analysis (Kaminski and Friedman, 2002; Draghici, 2003). When the data sets are divided into ‘positive’ and ‘negative’ by any definition, discriminant analysis such as PAM (Tibshirani et al., 2002) or SVM (Brown et al., 2000) can be used. However, in any case, no confident results are obtained unless the size of the gene list is reduced to a reasonable level. Suppose only one ‘toxic’ dose was tested and compared with its control. When using the A chip, 15 923 pairs should be made and 31 099 pairs for the v.2.0 chip. Even when the p value is set to a very low value, too many ‘significant’ differences without biological significance are obtained. On the other hand, when one does not want to miss a certain gene with high significance in terms of biology or toxicology, the p value cannot be lowered in the experiments with such small N values. It has been pointed out that application of the usual biostatistics is difficult when the evaluation is done with very small N values against vast numbers of measurements such as whole genes. Especially when the data contain biologically based variations, no improvement is expected by the use of any sophisticated statistical technique. When data with multiple doses or time point are available, however, the extraction of biologically meaningful changes is easier. 20.2.4.2

Example 1: The Case of Omeprazole

Let’s refer to the actual data. The first one is omeprazole, which was administered to rat at 100, 300, and 1000mg/kg and analyzed with the A chip (15 923 probe sets). In the TGP database, there are three dose levels with vehicle control for each of four time points in single and repeated doses, respectively. When comparison between control and treated rats is made in each point, it will be 380 740 pairs. When an uncorrected Student’s t-test is applied, 36 883 pairs of ‘significant at p < 0.05’ are obtained. No one would like to perform a precise analysis of these probe sets and it is practically impossible. It is common sense in statistics that this number is overestimated and some correction is needed. It is necessary to erase the accidental difference not related to the drug effect, but the usual statistics only tells you to reduce the p value based on some criteria. A simple reduction of the p value to 0.001 results in the reduction of the ‘significant’ pair numbers to 1639, which is still a quite large number, but the main question here is whether this has elucidated the really biological meaningful differences. Regrettably, the answer is no, in most cases. Suppose ‘genes significantly changed by 100 mg/kg omeprazole at any time point’ were extracted by the t-test without correction. Figure 20.2(a) shows an arbitrarily selected probe set, 13 938 56 ×. At p < 0.05, this gene is considered to be up-regulated at 3 h whereas it is down-regulated at 9 h after administration. However, when all of the data are reviewed (Figure 20.2(b)), it is quite easy to conclude that no biologically meaningful change is caused by the drug, since there is no dose- or time-dependency. This type of extraction only creates a ‘heap’ of ‘junky’ genes.

(a) 3h *

50

Signal mean

9h

60

60

50

40

40

30

30

20

20

10

10

*

0

0 Control

(b)

Control

100mg/kg

100mg/kg

Repeated 50

40

40

Signal mean

Signal mean

Single 50

30 20

30 20

10

10

0

0 1000

24 300

mg/kg

1000

9 6

100 0

3

300

h

mg/kg

29

15 100

8

days

6

9 24

4

0

50 45

Signal mean

40 35 30 25 20 15 10 5 0 0

3

6

9

100

24 4

8 15 29

h

3

6

days

9 24

h

300

4

8 15 29

3

days

6

9 24

8 15 29

4

h

8 15 29

days

Repeated

2500

2500

2000

2000

Signal mean

Signal mean

3

days

Single

(c)

mg/kg

1000

4

h

1500 1000 500

1500 1000 500

0 1000

0

24 300

mg/kg

100

6 0

1000

9

h

3

300

mg/kg

15 100

8 0

4

29

days

***

2500

Signal mean

2000

*

1500

*

1000

*

500

*

0 0

3

6

h

9

100

24 4

8 15 29

days

3

6

h

9 24

300

4

8 15 29

days

3

6

9 24

h

1000

4

8 15 29

days

3

6

h

9 24

4

8 15 29

days

Figure 20.2 Effects of omeprazole on gene expression. (a) Effects of a single oral dose of omeprazole (100 mg/kg) on the expression of an arbitrarily selected probe set, 1 393 856× at 3 and 9 h after dosing. (b) Effects of single and repeated oral doses of omeprazole (100, 300, 1000 mg/kg) on the expression of 1 393 856 ×. The upper panels show the 3dimensional graphs of single (left) and repeated (right) dosing, while the lower panel shows the 2-dimensional graph with error bars. The arrows indicate where ‘statistical significance’ was noted in Figure 20.2(a). (c) Effects of single and repeated oral dose of omeprazole (100, 300, 1000 mg/kg) on the expression of heme oxygenase-1 (1370080 at). The upper panels show the 3-dimensional graphs of single (left) and repeated (right) dosing. The lower panel shows the 2-dimensional graph with error bars. Expression data were normalized by mean value and multiplied by 500, and expressed as the mean of N = 3 with SD where indicated. The asterisks indicate ‘statistical significance’ by the uncorrected Student’s t-test at * p < 0.05 and *** p < 0.001


515

Next, attention is paid to heme oxygenase-1, which is known to play important roles in cellular stress response (Figure 20.2(c)). These data obviously indicate that omeprazole potently and dose-dependently induces this gene at 3 h after dosing. Here, if gene extraction is performed by using only the highest dose (1000 mg/kg), only the change at 6 h should be noted. An even worse point is that one would conclude that this important gene is not responding to 1000 mg/kg omeprazole if extraction is performed without using the data of 6 h. Looking at Figure 20.2(c), most researchers would agree that omeprazole induces the expression of heme oxygenase-1. However, it is highly possible that omeprazole is regarded as a negative inducer of this gene by experiments with reduced data points and restricted statistics.

20.2.4.3

Example 2: Age-Related Difference in Toxicity

Before the start of the TGP, there was an argument regarding the age of rats. One asserted that 6-week old rats should be used because this age is recommended in the standard toxicity tests, whereas the other claimed that confident data with small variations would be only obtained from mature animals not younger than 10 weeks. Before reaching a conclusion to use 6-week old rats, a study to compare 6- and 12-week rats was performed regarding the sensitivity of hepatotoxicity. Acetaminophen, isoniazid and carbon tetrachloride were selected as representative hepatotoxicants and the sensitivity to those were compared between 6- and 12-week old rats by a single administration protocol. Although the latter two showed no age-related difference, acetaminophen was found to be more toxic in 12-week old rats than in 6-week old rats. The causal factors were suggested to be the higher expression of CYP3A13 that produces an active metabolite and/or the lower expression of a subtype of glutathione transferase (Morishita et al., 2006). The most interesting findings in this study were the following points. In order to compare gene expression changes at 24 h after dosing where pathological changes emerged, differentially expressed genes between 6- and 12-week old rats were extracted by statistics. In the usual course of investigation, one would try to attribute the difference in toxicity to the expression level of these genes by using correlation or discriminant analysis. However, a precise review of these genes revealed other features. Among the stress-responsive genes, many genes showed age-related difference not in the extent but in the time course. Figure 20.3 shows again a representative stress responsive gene, heme oxygenase-1. In 6-week old rats, the peak of induction by acetaminophen appeared at 9 h or earlier and the expression returned to basal level at 24 h, whereas the peak was later than 9 h in 12-week old rats. If the observation were done at 24 h only, the expression of heme oxygenase-1 would have been judged as ‘yes’ in 12-week old rats and ‘no’ in 6-week old rats by an ‘allor-none’ manner. However, the actual response was ‘yes’ in both cases and the difference was present in the response time. Other information from Figure 20.3 is that a threshold dose of acetaminophen exists in the induction of at least this gene. This is reasonable, considering the widely accepted toxic mechanism of this drug, i.e. the hepatocyte damage does not occur when glutathione is not depleted and the detoxification system for active metabolite is active (James et al., 2003). In the present case, the induction of this gene would be undetectable unless the test with the highest dose was performed.

516

Hepatotoxicity

12-week

5000

5000

4000

4000

Signal mean

Signal mean

6-week

3000 2000 1000 0 1000

300

mg/kg

50

0

3

6

9

24

3000 2000 1000 0

h

1000

300

mg/kg

50

0

3

6

9

24

h

Figure 20.3 Effects of a single oral dose of acetaminophen (50, 300, 1000 mg/kg) on the expression of heme oxygenase-1 (1370080 at). The left and right panels show the data of 6-week old and 12-week old rats, respectively. Expression data were normalized by mean value and multiplied by 500, and expressed as the mean of N = 3

20.2.4.4

Advantage of Multi-Time, Multi-Dose Protocol

One would expect that the sensitivity of transcriptome analysis should be always higher than any traditional toxicological technique. Who wants to do additional, quite expensive and tedious tests when obvious toxicological phenotypes are already obtained? However, the previous example tells us that this hope is too optimistic, especially when a toxicological threshold may exist in a certain step of a sequence of gene expression, as in phenotypes. There would be two alternative answers. One is that prediction of toxicity by toxicogenomics technology is only possible in a chronological manner where the administration of toxic doses is essential. In this case, the prediction would be ‘Keeping this dosage will cause that phenotype, etc’. The other is that the toxicity of overdosage with apparent threshold can be indirectly predicted even at the low dose by gene expression changes somewhere in the toxicological pathway if threshold does not exist in that step. This point will be discussed later. Figure 20.4 depicts a schematic expression of time- and dose-dependencies of gene expression changes. Needless to say, the conclusions drawn from using multiple components with different time- and dose-dependencies must differ among the cases, where a single observation each is made at time point A or B, or at dose level X or Y. In the case of the usual biology test focusing on a particular target, it would be possible to set an appropriate time and dose in a preliminary study. However, toxicogenomics is designed to observe any gene expression changes reflecting ‘any toxic phenomena in the future’ one chip at a time, and so it is practically impossible. In order to ‘mine’ the data with biological significance (this does not always coincide with statistical significance), data with multiple time and dose are considered to be essential. Any sophisticated statistical procedure cannot create anything from no data.

Prediction of Hepatotoxicity Based on the TGP Database (b) Gene A

Gene E

Gene B

Gene F

Gene C

Gene expression

Gene expression

(a)

517

Gene G

Gene H

Gene D A

B

Time

X

Y

Dose

Drug

Figure 20.4 Schematic expressions of (a) time- and (b) dose-dependencies of gene expression changes. (a) Some of the genes might be up-regulated by the administration of a drug (indicated by arrows) either transiently at an early time point (A) or at a later time point (B), or continuously with early (C) or later onsets (D). If observations are made either at the time point A or B, the results would be quite different, i.e. up-regulation is observed in A and C at time point A, whereas up-regulation occurs in B and C at point B and nothing is detectable in D. (b) Similar to the above, some of the genes might be up-regulated by the administration of a drug with either bell-shaped (E, F) or sigmoidal (G, H) dose-dependency. If observations are made at the concentrations of either X or Y, the results would be quite different, i.e. up-regulation in E and G at concentration X, whereas up-regulation in F and G at Y with nothing detectable in H

Even with such enriched data stored in our database, the procedure as to how to efficiently withdraw significant genes is immature. A statistical test covering dose-dependency, Williams test, for example, is available, but it does not always work efficiently. In general, three for each group is too small for any statistical analysis. However, the cost effectiveness is still questionable when the number is increased to, say, five. In order to make the statistical analysis applied to over 10 000 measures meaningful, N should be increased to a similar order, which is impractical in the biological data. In our experience, too strict statistics should not be applied for extracting significantly mobilized genes, in order not to overlook important genes with biological variance. We use various properly applied approaches case-by-case, e.g. ANOVA with a relatively large p, followed by proper filtering (elimination of the genes with significance in low dose only, or whose expression were inversely correlated with dose, etc.) or selection without statistics, based on the value of ratio to control. In any case, all of the data of extracted genes down to a reasonable number are stored in the database and so at anytime can refer their dose- or time-dependencies. 20.2.4.5

New Knowledge from Accumulated Data

When vast amounts of data are accumulated in the database, an interesting thing emerges by simple alignment of the data. Figure 20. 5(a) presents a summary of the vehicle control

518

Hepatotoxicity Single

(a)

Time

Repeated 9.00–11.00 am

18 9 12 15

0

5

10 Chemical 15 20 number

25

30

9 6 3 h

24

1400 1200 1000 800 600 400 200 0

5

10

15

Chemical number

Single

(b) Time

12 15

20

25

1400 1200 1000 800 600 400 200 0

30 4

8

29 15 days

Repeated

18 9

9.00–11.00 am

600

600

500

500

400

400

300 5

300

200 10

15

Chemical number

100 20

25

30

24 6 9 3 h

0

5

200 10

15

Chemical number

100 20

25

30

4

29 8 15 days

0

Figure 20.5 Circadian changes in the expression of (a) D the site albumin promoter binding protein, 1387874 and (b) the aryl hydrocarbon receptor nuclear translocater-like, 1370510 in vehicle controls for the early 35 chemicals in the database tested by using the RAE 230A chip. Left panels – circadian changes of the expression within a day. Sacrifice of the animals in single-dose experiments is done 3, 6, 9 and 24 h after oral administrations in the morning. They thus correspond to about 12.00 midday, 15.00 pm, 18.00 pm and 9.00 am the next day, respectively. Right panels – in repeated administration, sacrifice is done in the morning at around 9:00 am to 11:00 am, and thus the expression level is constant to the value in the morning through all of the chemicals and days. It is also obvious that expression does not change with age

group of single- and repeated-dose experiments. The left panel shows the expression level of a representative circadian gene, D site albumin promoter binding protein, for 3, 6, 9 and 24 h data of the vehicle controls from the first 35 chemicals tested with an old A chip. As the administration is done in the morning, this pattern shows the reproducible increase of expression in the afternoon. The right panel shows its expression level in the repeated dosing for 3, 7, 14 and 28 days. As the autopsy was done 24 h after the last dose, i.e. in the morning, the expression level appears uniformly low. On the other hand, there are genes


519

(a)

7000

5000 4000 5

3000

10

2000

15

1000

20 25

Chemical number

Signal intensity

6000

30 35

1

4

29 15 8 days

0

(b)

3500 3000

1500 5

1000 10

15

Chemical number

Signal intensity

2500 2000

500 20

25 30

1

4

8

29 15 days

0

Figure 20.6 Expression levels of (a) steroid delta-isomerase 3 and (b) the hemoglobin beta chain complex in vehicle controls for the early 35 chemicals in the database tested by using the RAE 230A chip

with reversed circadian rhythm such as ARNT-like (Figure 20.5(b)), i.e. high expression level at noon and it goes down towards the night. The field of circadian gene expression has recently attracted attention (Ueda et al., 2004). Using our TGP database, many genes with circadian rhythm, as well as potential drugs that affect the rhythm, can be easily extracted. Reviewing the vehicle control data of repeated dosing experiments, we noticed a group of genes whose expression level changed with age. Figures 20.6(a) and 20.6(b) show the expression levels of steroid delta-isomerase 3 and hemoglobin beta chain complex, respectively. The former increased, while the latter decreased with age, possibly reflecting

520

Hepatotoxicity

a physiological increase in steroid hormone synthesis and a decrease in extramedullary hematopoiesis with age. In TGP, either 0.5 % methylcellulose or corn oil works as vehicle, according to the solubility of the test drug. As a lot of data of both vehicles were being accumulated, the extraction of vehicle effects became possible. As we have already published (Takashima et al., 2006), it was revealed that the extent and/or time course of expression of genes related to lipid metabolism was affected by corn oil, since rats received orally a high caloric intake in the morning when they usually do not eat. These genes with reproducible expression patterns are now utilized for quality control of each expression data in the TGP. This could be applied for evaluation of data obtained in a different platform in the future. There are also many probe sets with quite poor reproducibility or large variance. If the cause of their variation is attributed to certain factors, like animal treatment or laboratory circumstances, such genes are in turn useful tools to evaluate inter-laboratory differences. It might be convenient to make a list of ‘useless’ genes that are excluded at the beginning in order to facilitate the analysis. However, care should be taken to exclude any genes that are absent in samples in the database even after a vast amount of data is accumulated, at least insofar as ‘toxicity’ is concerned. It is always possible that expression of such a gene is uniquely induced by a new drug. Figure 20.7 shows the overview of the expression of the TNFRSF16 associated protein 1 in the first 35 chemicals analyzed with the A chip. Only thioacetamide and methapyrilene induced its expression after repeated administrations. It would be interesting to investigate the toxicological mechanism of these non-genotoxic carcinogens making this observation as a clue. The point here is that the simple accumulation of the vast amount of data is scientifically valuable and it is important to collect any observable changes as precisely as possible.

20.3

Construction of a Toxicity Prediction System Based on the TGP Database

The TGP system consists of basically three parts, i.e. the database itself that stores gene expression data with related pathology (scoring and the photo of HE staining), hematology, blood biochemistry and chemical information, the data analysis system that consists of the tools for up- and download of data, clustering, discriminant analysis, principal component analysis, gene- or compound list manager, etc. and the prediction system that is used when the expression data are uploaded. In the summer of 2006, the final form of these systems has become operational. In the TGP database, in vivo data of 24 000 rats, expression data of about 24 000 ‘GeneChips’ corresponding to ca. 700 000 000 probe sets, 2 880 000 measured test items, the data of 48 000 pathology specimens and various related information and reports are to be stored in their final form. In order to pick up useful data for toxicologically meaningful analysis from such a large scale of data, an efficient, toxicologist-friendly system is essential. 20.3.1

Analysis System

Let’s see the previous omeprazole case again. The first one is pathology. In the single dose experiment, the significant and dose-dependent change was periportal eosinophility

Prediction of Hepatotoxicity Based on the TGP Database Thioacetamide

521

Methapyrilene

1600 1400

Signal intensity

1200 1000 800 600 400

4 24

200 0

8

29 15 Repeated (days)

9 5

10

15

20

6 25

Chemical number

30

35

3

Single (h)

Figure 20.7 Expression level of the TNFRSF16 associated protein 1 in the first 35 chemicals analyzed by using the RAE 230A chip. This gene was specifically up-regulated by repeated administration of thioacetamide (indicated by the arrow around the chemical number 17) and methapyrilene (indicated by the arrow around the chemical number 25). Expression data were normalized by mean value and multiplied by 500, and expressed as the mean of N = 3. On the time axis, sets of the data of control, low, middle and high doses are aligned in the order of 3, 6, 9 and 24 h after single doses, and 3, 7, 14 and 28 days of repeated doses

observed at 6 and 9 h. In the repeated ones, centrilobular hypertrophy was evident with a peak at 2 weeks. Simultaneously measured test items were 43 in total, and 22 items showed some changes. As it is quite tedious to check each individually, the extent of the change to control value in each measure was converted into a semi-quantitative heat-map and depicted as Figure 20.8(a). It can be seen that omeprazole did not induce serious toxic changes in a single dose, whereas obvious hepatic hypertrophy, anemia and some mobilization in plasma lipid emerged by repeated administration. The next one is the gene expression change in the liver. For each gene (probe sets), a graph of 4 × 4 matrix with N = 3 for each lattice can be drawn for single and repeated experiments. As it is again quite difficult to interpret the results consisting of more than 15 000 probe sets, checking one by one, a similar heat-map was depicted by a semi-quantitative conversion of the dose–response at each time point (Figure 20.8(b)). This is for the example of glutathione reductase. It is obvious that this gene was dose- and time-dependently up-regulated toward 24 h in the single dose, whereas the extent kept decreasing as the administration continued. Since glutathione reductase is known to involve oxidative stress responses, other genes

(a) Pathology

3h

6h

9h

24 h

4 days

8 days

15 days

29 days

Periportal/eosinophilic hypertrophy Cellular infiltration Total body weight Relative liver weight RBC Hb Ht MCV MCH Ret Plat WBC APTT Fbg ALP TC TG PL Glu BUN AST ALT LDH γ –GTP

(b)

3h

6h

9h

24 h

4 days 8 days 15 days

Repeated

2000

2000

1500

1500

Signal mean

Signal mean

Single

1000 500 0 1000

300 100 mg/kg 0 3

29 days

6

9 h

24

1000 500 0 1000

300 100 mg/kg 0

4

15 8 days

29

Figure 20.8 Schematic representation of an image of data stored in the TGP database. (a) Heat-map of phenotype induced by omeprazole administered to rats as an example. The upper panel shows the pathological findings while the lower panel shows the changes in organ/body weight, plasma biochemistry and hematology. The actual data table is much larger than this, but the items without significant change are omitted and the data are converted to a semiquantitative heat-map. (b) Heat-map conversion of the expression changes of glutathione reductase, 1369061 at. This way makes it convenient to overview time- and dose-dependent changes at a glance. (c) Heat-map expression of the gene expression changes together with the phenotype induced by omeprazole. Genes were categorized by their function and aligned by the order of the time when the first change appeared. This panel continues far down below. (d) The alignment of the panels of the chemicals prepared by the way described above. Practically, it continues in both directions of horizontal and vertical

(c)

3h

6h

9h

24h

4d

8d

15d

29d

Necrosis Periportal / eosinophilic hypertrophy Cellular Infiltration

Phenotype

Total Body Weight Relative Liver Weight RBC Hb Ht MCV MCH Ret

Pathology Hematology Biochemistry

Plat WBC APTT Fbg ALP TC TG PL Glu BUN

16 000 genes

AST(GOT) ALTÅiGPTÅj LDH É¡Å| GTP HOX1

Stress Response

v-myc

Stress Response

Trx REDase Stress Response GSHRD

Stress Response

GSTa2

Stress Response

HSP86

Stress Response

HSPÇPÉø

Stress Response

MMIF

Stress Response

Trx L

Stress Response

HSP 60

Stress Response

GSTa1

Stress Response

Smase

Stress Response

CD36

Stress Response

MAT2

Stress Response

TRAF

Stress Response

MAT1A

Stress Response

GSHPOX

Stress Response

C/EBP

Apoptosis

TNFaR1

Apoptosis

GTPCH

Apoptosis

G6PDH

Apoptosis

BID

Apoptosis

FasAF

Apoptosis

DAKL

Apoptosis

CYP1A1

Drug Metabolism

CYP2B19

Drug Metabolism

CYP1A2

Drug Metabolism

CYP26

Drug Metabolism

UDPGT

Drug Metabolism

CYP3A3

Drug Metabolism

CYP51

Drug Metabolism

CYP3A9

Drug Metabolism

HSP86 ARNT2

Apoptosis

Apoptosis

CASP3

AHR

Stress response

CYP

AhR AhR AhR

ARNT

AhR

PTPase

Phosphotyrosine

PTK2

Phosphotyrosine

AhR, PTK

Ppase/Tensi nPhosphotyrosine TYRO3

Phosphotyrosine

PTPaseNR2 Phosphotyrosine PTKR HDGF TGFb v-raf

Phosphotyrosine Cell cycle/Growth Cell cycle/Growth Cell cycle/Growth

v-Akt

Cell cycle/Growth

CDC10

Cell cycle/Growth

erv-1

Cell cycle/Growth

SCDF

Cell cycle/Growth

SCDFR1

Cell cycle/Growth

Cell cycle/growth 150 chemicals 1

(d)

2

Thioacetamide

Phenotype Pathology Hematology Biochemistry

Stress response

Apoptosis

CYP

3h 6h 9h 24h 4d Total Body Weight Relative Liver Weight RBC Hb Ht MCV MCH Re t Pla t WBC APTT Fb g ALP TC TG PL Glu BUN AST(GOT) ALTÅiGPTÅj LDH É¡Å| GTP HOX1 St r e ss v- m yc St r e ss Tr x REDa se St r e ss GSHRD St r e ss GSTa 2 St r e ss ÇfÇÉÇåÇÉ St r e ss HSP27 St r e ss MMIF St r e ss Tr x L St r e ss HSP 60 St r e ss GSTa 1 St r e ss Sm a se St r e ss CD36 St r e ss MAT2 St r e ss TRAF St r e ss MAT1A St r e ss ÇfÇr ÇsÇêÇQ St r e ss ÇfÇr ÇsÇî ÇQ St r e ss ÇfÇr ÇsÇxÇÇÇS St r e ss No x4 St r e ss NQO1 St r e ss Pr d x2 St r e ss GSHPOX St r e ss C/ EBP Apoptosis TNFa R12a Apoptosis GTPCH Apoptosis G6PDH Apoptosis BID Apoptosis CASP3 Apoptosis CASP8 Apoptosis Fa sAF Apoptosis DAKL Apoptosis CYP1A1 Dr u g Me t a b . CYP7A1 Dr u g Me t a b . CYP1A2 Dr u g Me t a b . CYP27 Dr u g Me t a b . CYP4F4 Dr u g Me t a b . CYP3A3 Dr u g Me t a b . CYP51 Dr u g Me t a b . CYP3A9 Dr u g Me t a b . AHR Ah R

8d 15d 29d 3h

Figure 20.8

3

Omeprazole 6h 9h 24h 4d

4

Me t h a p yr ile n e 8d 15d 29d 3h

(Continued)

6h 9h 24h 4d

8d 15d 29d 3h

5

In d o m e t h a c in 6h 9h 24h 4d

Et h io n in e

8d 15d 29d 3h 6h 9h 24h 4d 8d

524

Hepatotoxicity

belonging to this category were examined and aligned in the order of their responding time to overview the sequence of chronological expression of the functional genes. Continuing this step for other categories, it produces a long heat-map as shown in Figure 20.8(c), which actually continues far down below. It might be possible to hypothesize a cascade of gene expression related to the effects of omeprazole on the liver, but the whole body of data is too large to reach an acceptable conclusion without the aid of computerized pathway tools. Moreover, 150 in such a heat-map are lining horizontally (Figure 20.8(d)). The main problem here is how to extract the genes whose expression changes are correlated with a certain phenotype or potential toxicity by using the bioinformatics technique. In the case where a target or phenotype is clear, such as PPAR agonists, the extraction of significant genes is relatively easy (Tamura et al., 2006b). However, it is quite difficult to extract genes related to ‘toxicity’ with unknown mechanisms. For example, suppose one would try to extract genes related to above ‘hypertrophy’ caused by omeprazole. It would be unsuccessful if one extracts the commonly changed genes among the compounds that caused hypertrophy. This is not a matter of statistical technique but of pathology. First, it is quite rare that a certain pathological phenotype is attributed to a single toxicological mechanism. Secondly, the quantitativity of pathology scoring is poor. Thirdly, when the severity of a phenotype increases or decreases, its grading score does not always increase or decrease, but the name of the diagnosis changes. One cannot avoid this problem by including the diagnosis expressing the upper or lower grade, since the name does not necessarily express the severity of the phenotype alone. ‘Phenotype anchoring’ is a challenging issue in the toxicogenomics field (Moggs et al., 2004) and it is also under investigation in our TGP. It is also expected that progress in ‘toxicogenomics-oriented histopathology’ should be an important field in toxicology. We have therefore started with anchoring to clarify indicators in parallel. The TGP database contains representative drugs and hepatotoxicants, but it does not necessarily mean that their toxicological mechanisms are representative with respect to molecular biology. We then decided to perform additional single-dose experiments using compounds that possess clear molecular targets in order to investigate the relationship between molecular toxicity and gene expression changes. They included protein synthesis inhibitors, compounds related to infection/inflammation (lipopolysaccharide, TNFα, galactosamine), ER-stressor, and cytoskeleton disruptors. The results of phorone, a glutathione depletor, are presented. As described above, hepatotoxicity of acetaminophen is due to excess production of active metabolite over the detoxification capacity of intracellular glutathione (James et al., 2003). Therefore, any drugs that have a potential to deplete hepatocyte glutathione risk causing acetaminophen-type hepatotoxicity with overdosage. In a previous report, a list of marker genes for glutathione depletion was extracted using BSO, a glutathione biosynthesis inhibitor (Kiyosawa et al., 2004). Phorone was considered to be superior to BSO as a model system, since the mechanism of glutathione depletion is similar to that of acetaminophentype hepatotoxicants, i.e. it covalently binds to glutathione and is excreted from the cell. Phorone, at 40, 120 or 400 mg/kg was administered according to the same protocol as the regular single-dose experiments and the measurement of glutathione contents was performed in addition to the regular tests. Phorone caused a marked but transient depletion of glutathione with maximal depletion occurring at 3 h; then it recovered and showed


525

an increase of glutathione at 24 h as a rebound. A significant increase in plasma AST was observed at 400 mg/kg, indicating hepatotoxicity. Expression data at this point were excluded from the work-up in order not to extract the expression changes secondary to hepatic injury. In the next step, genes whose expression were inversely correlated with hepatic glutathione contents for each rat were statistically extracted and filtered to get 130 probe sets. Principal component analysis of the chemicals stored in the database using these sets revealed that chemicals with a risk of glutathione depletion, such as bromobenzene and coumarin, in addition to acetamineophen, were clearly separated from other chemicals or controls toward the direction of PC1, suggesting that the list was a useful as ‘marker gene list for risk assessment of glutathione depletion’ (Kiyosawa et al., in press). Our present strategy is to prepare biomarker gene lists that are related to certain toxicological phenotypes, pathways, or any biologically meaningful factors, as many as possible using various procedures. 20.3.2

Toxicity Prediction System Based on Biomarker Gene Lists

In the general terminology, ‘biomarker’ is designated as one or a few biological measures quantitatively reflecting a certain biological change. However, this does not fit for the ‘biomarker gene’ in toxicogenomics. As discussed above, toxicology based on the transcriptome has various problems, such as poor statistics because of small N s compared with genes, the requirement to make the beta-error small, toxicity with uncertain timeor dose-dependency, etc. Although the quantitativity of microarrays has greatly improved in recent years, its quantitativity and reproducibility cannot be superior to the enzymes in serum or the metabolites in urine when a labile object, mRNA, is measured. It is thus dangerous to make a decision based on one or few marker genes, and it would be effective to use relatively large numbers of genes as a whole in order to make the assessment robust. A comprehensible example is shown in Figures 20.2(c) and 20.7(c). If heme oxygenase-1 alone is employed as a ‘stress marker’, the potential to overlook omeprazole must be quite high, but it should go down if a ‘stress-responsive gene list’ is employed. In this connection, the TGP is now trying to substantialize such gene lists. When an assessment or prediction of toxicity is made by a list of multiple measures, it becomes necessary to summarize or quantify these measurements. Ideally, the quantification process should be optimized for each marker gene list. However this is practically difficult and thus a uniform system has to be created. In the TGP, a new scoring system was developed in one trial (Kiyosawa et al., 2006). The score is calculated based on the ratio to control value (log 2) for each gene in the marker list and expressed as a heat-map. This scoring system has made it easy to overview the assessments of a target compound against many marker lists, or to overview the assessments of many compounds against a particular marker list. However, there are some problems in this system, i.e. the score is biased when the list contains a gene whose expression change is extremely large (e.g. CYP1A1) and changes are canceled when up- and down-regulated genes co-exist in the list. Therefore, another scoring system, e.g. effect size (the absolute value of the difference between means divided by the standard deviation) is also available in the TGP system. Principal component analysis is a quite convenient tool to make a qualitative classification of compounds against a list of genes. As a prediction system, however, some quantitative

526

Hepatotoxicity

data would be favorable for the final output. Thus in our system, the following functions are added, i.e. when the user specifies a principal component with high contribution, the compounds are sorted by the value and the genes with large Eigenvector value are easily obtained. This gives an idea where the relative position of the test drug locates among the ones in the database, and suggests a candidate gene list for further investigation. When a phenotype that can be judged as positive/negative is available, discriminant analysis is known to be powerful (Porter et al., 2003), and prediction analysis of microarray (PAM) (Tibshirani et al., 2002) has been firstly employed in the TGP. By a semi-automatic system of training and validation, the efficiency improves for the creation of discriminators. As above, the system exhibiting the prediction by PAM as quantitative scores (to show the relative position of a test drug among chemicals in the database) is under consideration. The TGP also includes the support vector machine (SVM) (Brown et al., 2000) in the system. Although the present system has not come to completion, the following picture of TGP use has emerged (Figure 20.9). You have a candidate drug, X, which was administered to rats and ‘GeneChip’ data of the liver were obtained 24 h after dosing. The data are up-loaded to the TGP system and the marker viewer is activated to overview all the biomarker gene lists stored in the database. Alarms are noted for several markers (Figure 20.9(a)). Among them, one marker, M1, is selected, as this is highly related to the toxicological phenotype F1, if repeatedly administered. When PCA is performed using these marker genes, X is clearly separated in the direction of PC1 (Figure 20.9(b)). As the contribution of PC1 is found to be high, compounds are sorted by PC1 and their order is Y (high dose) > X > Z ( high dose) > Y (middle dose) > . . . ., etc. From the analysis, it is predicted that X causes phenotype F1 when repeatedly dosed, and it requires a higher dose than Y but lower than Z. The gene list with a high Eigenvector is harvested for further analysis of the toxicological pathway. There is also another point. It is known that X is pharmacologically similar to compounds P and Q that are stored in the database. It is also known that P and Q cause phenotype F2 and actually that a marker gene list M2, predicting F2, exists in the database, whereas no alarm was noticed in the first survey. Then all the compounds are overviewed against M2. It is obvious that not only X but also P and Q show low scores because most of the marker genes in the list show transient expression changes and thus F2 is non-predictable using M2 at 24 h (Figure 20.9(d)). Therefore, it is suggested that an additional early time point is necessary to judge whether X has a similar property to P and Q. This is just a simulation. The point here is that the prediction system of the TGP is not a simple output of a probability like a ‘weather-forecasting system’, but a supply of knowledge with suggestions for further investigation. As everybody knows, there is no drug without side effects. If a prediction like ‘this drug is safe’ appears, this must be a lie. Toxicologists do not want such a prediction, but want the information such as, ‘What kind of phenotype would appear in what dose level and what are the toxicological mechanisms involved?’.

20.4

The Image of Toxicity Testing after the TGP Database is Established

At present, we say that the usage of the TGP database/prediction system is in the following condition. In the quite early stage of drug development, the database is used to select a lead compound among candidates. As the full-scale toxicity test is quite costly, safety assessment

Prediction of Hepatotoxicity Based on the TGP Database X

(a) Marker check

527

(b) PCA analysis using M1 PC2

Marker M1 M2 M3 M4 M5

Phenotyp F1 F2 F3 F4 F5

3h

6h

9h

24 h X

PC1 (c) Sort by PC1 PC1

Y: high dose X Z: high dose Y: middle dose

A list of genes with high Eigenvectors is acquired for further analysis

(d) Scores vs. marker M2 Drugs X P Q R S

3h

6h

9h

24 h ? Suggestions for further tests

Figure 20.9 Schematic representation of an image of toxicity prediction process in the TGP system. (a) The data of new drug X that are up-loaded to the TGP system are checked against several markers useful to predict phenotypes in repeated administration from single-dose experiments. One of the markers (M1) shows an alert. (b) PCA is performed using these marker genes, M1. X is clearly separated in the direction of PC1 with high contribution value. Then, the compounds are sorted by PC1, showing the order of Y (high dose) > X > Z (high dose) > Y (middle dose). The gene list with high Eigenvectors can be harvested for further analysis. (c) Drugs P and Q cause phenotype F2. The marker M2, predicting future phenotype F2, is available in the database. When compounds, including X, are overviewed against M2, not only X but also P and Q show low scores at 24 h while the latter two show high scores in this marker gene list at earlier periods. This indicates that prediction of F2 is inconclusive without the expression data in earlier stages

of candidate drugs are usually done just before the clinical trial. If serious toxicity emerges at this stage, it might be necessary to return to the seeds, because toxicity is often inherent to the basic structure and thus never eliminated by minor modification. If the potential phenotype (when repeatedly dosed) is predictable in the early stage by gene expression data of a few numbers of experimental animals, it would effectively cut out time and cost for drug development. From another point of view, this also contributes to animal welfare by reducing the number of sacrifices. Application of toxicogenomics to the toxicity test in the final candidate just before clinical trials seems to be promising to improve the predictivity of clinical side effects. In fact there is a trend to employ toxicogenomics and pharmacogenomics technology for regulatory science (Petit, 2004). In such a case, various issues regarding validation and

528

Hepatotoxicity

standardization of data acquisition and analysis in addition to the species difference will be problems that should be solved urgently. The third field for toxicogenomics is post-marketing surveillance. One promising strategy against species difference would be the connection of the clinical data to the TGP database. As human in vivo experiments are impossible and the barrier between in vivo and in vitro is too high, the only way is to accumulate clinical phenomena that are related to the contents in the database. There is a movement to create databases of disease-related genotypes or SNPs, and it would be promising to make a functional network between these databases and the TGP to establish an integrated toxico/pharmacogenomics database. In any event, the completion of the TGP database is not a goal, but a beginning for toxicogenomics study. Compared with the enrichment of the data, the analysis/prediction procedures are in the developing stage. It is our desire that toxicology together with systems biology rapidly advances by efficient use of this database and that it contributes in accelerating the development of more effective and safer drugs.

Acknowledgments The following researchers were involved in the project until this paper was written. Taku Nagao (Project Leader), Toshikazu Miyagishima (Sub-Leader), Ken-ichi Aisaki, Takanobu Hanada, Satoru Hashimoto, Masayuki Heishi, Mituhiro Hirode, Akihiko Hirose, Takehiko Hosoiri, Katsuhide Igarashi, Shoichiro Ide, Dai Kakiuchi, Jun Kanno, Toshihiko Kasahara, Naoki Kiyosawa, Masanobu Komiya, Tomochika Matushita, Toshiko Miyazaki, Yumiko Mizukawa, Katsumi Morishita, Takamichi Muramatsu, Hiroyuki Nitta, Yoshie Ohno, Manabu Okuyama, Ko Omura, Atsushi Ono, Yoshiyuki Saeki, Atsushi Shibutani, Toshinori Shimizu, Tunefumi Shiwaku, Takamasa Suzuki, Kayoko Takashima, Kotaro Tamura, Hiroyuki Tomita, Naoki Torizuka, Hirohiko Totsuka, Soh Tsunezuka, Hiroyuki Ueda, Takeki Uehara and Tomoya Yamashita. Half of the project was supported by a grant from the Ministry of Health, Labour and Welfare, H14-toxico-001.

References Boess, F., Kamber, M., Romer, S., Gasser, R., Muller, D., Albertini, S. and Suter, L. (2003). Gene expression in two hepatic cell lines, cultured primary hepatocytes, and liver slices compared to the in vivo liver gene expression in rats: possible implications for toxicogenomics use of in vitro systems. Toxicol Sci 73, 386–402. Boverhof, D. R. and Zacharewski, T. R. (2006). Toxicogenomics in risk assessment: applications and needs. Toxicol Sci 89, 352–360. Brown, M. P., Grundy, W. N., Lin, D., Cristianini, N., Sugnet, C. W., Furey, T. S., Ares, M. Jr and Haussler, D. (2000). Knowledge-based analysis of microarray gene expression data by using support vector machines. Proc Natl Acad Sci USA 97, 262–267. Draghici, S. (2003). Analysis and visualization tools, in Data Analysis Tools for DNA Microarrays; A. M. Etheridge, L. J. Gross, S. Lenhart, P. K. Maini, H. M. Safer and E. O. Voit (Eds), CRC Press, London, UK, pp. 231–261.


529

Ismail, K. and Landis, J. (2003). Can the pharmaceutical industry reduce attrition rates? Nat Rev Drug Discov 3, 711–715. James, L. P., Mayeux, P. R. and Hinson, J. A. (2003). Acetaminophen-induced hepatotoxicity. Drug Metab Dispos 31, 1499–1506. Kaminski, N. and Friedman, N. (2002). Practical approaches to analyzing results of microarray experiments. Am J Respir Cell Mol Biol 27, 125–132. Kanno, J., Aisaki, K., Igarashi, K., Nakatsu, N., Ono, A., Kodama, Y. and Nagao, T. (2006). ‘Per cell’ normalization method for mRNA measurement by quantitative PCR and microarrays. BMC Genom, 7, 64. Kiyosawa, N., Ito, K., Sakuma, K., Niino, N., Kanbori, M., Yamoto, T., Manabe, S. and Matsunuma, N. (2004). Evaluation of glutathione deficiency in rat livers by microarray analysis. Biochem Pharmacol 68, 1465–1475. Kiyosawa, N., Shiwaku, K., Hirode, M., Omura, K., Uehara, T., Shimizu, T., Mizukawa, Y., Miyagishima, T., Ono, A., Nagao, T. and Urushidani, T. (2006). Utilization of a one-dimensional score for surveying the chemical-induced changes in expression levels of multiple biomarker gene sets using a large-scale toxicogenomics database. J Tox Sci 31, 433–448. Kiyosawa, N., Uehara, T., Omura, K., Hirode, M., Shimizu, T., Mizukawa, Y., Gao, W., Ono, A., Miyagishima, T., Nagao, T. and Urushidani, T. (2007). Identification of glutathione depletionresponsive genes using phorone-treated rat liver. J Tox Sci in press. Moggs, J. G., Tinwell, H., Spurway, T., Chang, H.-S., Pate, I., Lim, F. Le, Moore, D. J., Soames, A., Stuckey, R., Currie, R., Zhu, T., Kimber, I., Ashby, J. and Orphanides, G. (2004). Phenotypic Anchoring of Gene Expression Changes during Estrogen-Induced Uterine Growth. Environ Health Perspect 112, 1589–1606. Morishita, K., Mizukawa, Y., Kasahara, T., Okuyama, M., Takashima, K., Toritsuka, N. Miyagishima, T., Nagao, T. and Urushidani, T. (2006). Gene Expression Profile in Liver of Differing Ages of Rats after Single Oral Administration of Acetaminophen. J. Tox Sci 31, 491–508. Petit, S. (2004). Toxicogenomics in Risk Assessment: Communicating the Challenges. Environ Health Perspect 112, A662. Porter, M. W., Castle, A. L., Orr, M. S. and Mendrick, D. L. (2003). Predictive Toxicogenomics, in An Introduction to Toxicogenomics, M. E. Burczynski (Ed.), CRC Press, Boca Raton, FL, USA, pp. 225–260. Takashima, K., Mizukawa, Y., Morishita, K., Okuyama M., Kasahara, T., Toritsuka, N., Miyagishima, T., Nagao, T. and Urushidani, T. (2006). Effect of the difference in vehicles on gene expression in the rat liver – analysis of the control data in the Toxicogenomics Project Database. Life Sci 78, 2787–2796. Tamura, K., Ono, A., Miyagishima, T., Nagao, T., and Urushidani, T. (2006a). Comparison of gene expression profiles among papilla, medulla and cortex in rat kidney. J Tox Sci 31, 449–470. Tamura, K., Ono, A., Miyagishima, T., Nagao, T. and Urushidani, T. (2006b). Profiling of gene expression in rat liver and rat primary cultured hepatocytes treated with peroxisome proliferators. J Tox Sci 31, 471–490. Tibshirani, R., Hastie, T., Narasimhan, B. and Chu, G. (2002). Diagnosis of multiple cancer types by shrunken centroids of gene expression. Proc Natl Acad Sci USA 99, 6567–6572. Ueda, H. R, Chen, W., Minami, Y., Honma, S., Honma, K., Iino, M. and Hashimoto, S. (2004). Molecular-timetable methods for detection of body time and rhythm disorders from single-timepoint genome-wide expression profiles. Proc Natl Acad Sci USA 101, 11227–11232. Urushidani, T. and Nagao, T. (2005) Toxicogenomics: the Japanese initiative, in Handbook of Toxicogenomics – Strategies and Applications, J. Borlak (Ed.), Wiley-VCH, Weinhaim, Germany, pp. 623–631. Wildsmith S. and Spence, F. (2003). Preparation and Utilization of Microarrays, in An Introduction to Toxicogenomics, M. E. Burczynski (Ed.), CRC Press, Boca Raton, FL, USA, pp. 3–16.

21 Relationship between N-acetyltransferase-2 Gene Polymorphism and Isoniazid-Induced Hepatotoxicity Yasuo Shimizu, Kunio Dobashi and Masatomo Mori

21.1

Introduction

In 2004, there were an estimated nine million new cases of Mycobacterium tuberculosis (M. tuberculosis) infection and approximately two million deaths from tuberculosis [1]. Thus, tuberculosis remains the leading cause of mortality among the infectious diseases. Its treatment requires at least two anti-tuberculous drugs, such as isoniazid (INH) and rifampicin (RFP), and a combination of four drugs is commonly used [2]. In the recommended regimen for treatment of tuberculosis, at least six to nine months of therapy is required. However, there are patients who cannot continue treatment due to adverse effects [1, 2]. The major adverse effect of INH is hepatitis, associated with mild to severe elevation of ALT [3]. Three to four anti-tuberculous drugs, including INH and RFP, are recommended for treatment. However, RFP induces cytochrome P450 and amidase, which affect INH metabolism and so the concomitant use of INH and RFP may increase the risk of hepatotoxicity [2, 4]. When treatment for tuberculosis is given to patients with acquired immunodeficiency syndrome (AIDS), use of INH increases the occurrence of hepatitis due to concomitant administration of nevirapine, which is part of the regimen for highly active antiretroviral therapy for AIDS [5].


532

Hepatotoxicity

Genotype NAT2*4 NAT2*5B NAT2*6A NAT2*6B NAT2*7A NAT2*7B NAT2*12A NAT2*12C

Figure 21.1 types

21.2

Nucleotide change None 341T>C, 481C>T, 803A>G 282C>T, 590G>A 590G>A 857G>A 282C>T, 857G>A 803A>G 481C>T, 803A>G

NAT2 allele genotype NAT2*4/*4 NAT2*4/*12A NAT2*4/*12C

Phenotype RA RA

NAT2*4/*6A NAT2*4/*6B NAT2*4/*7A NAT2*4/*7B NAT2*6A/*12A NAT2*6A/*12C

IA IA IA IA IA IA

NAT2*5B/*6A NAT2*5B/*7B NAT2*6A/*6A NAT2*6A/*7B NAT2*7A/*7A NAT2*7B/*7B

SA SA SA SA SA SA

Comparison of major NAT2 genotypes, nucleotide changes, alleles and pheno-

N-acetyltransferase-2 Genotype and INH-Induced Hepatotoxicity

N-acetyltransferase-2 (NAT2) is the key enzyme involved in the acetylation of INH and sulfasalazine, and the rate of INH acetylation is affected by single nucleotide polymorphisms (SNPs) of NAT2, which produce rapid acetylators (RAs), intermediate acetylators (IAs) and slow acetylators (SAs) [4, 6]. The gene encoding NAT2 is located on chromosome 8 (8p22) and there are nine known SNPs (191G/A, 282C/T, 341T/C, 434A/C, 481C/T, 590G/A, 803A/G, 845A/C and 875G/A). The relationships between the major NAT2 genotypes, nucleotide changes, alleles, and phenotypes are shown in Figure 21.1. It was reported that 22 NAT2 genotypes were found in Koreans and 25 NAT2 genotypes were detected in a mixed population of Lhoi-San, Malays, Europeans, and Africans, while SAs were reported to account for about 50 % of Russians and Caucasians versus 10 % of Asians [7–9]. The SA genotype (NAT2 *6/*6, NAT2 *6/*7, NAT2 *7/*7) has been reported to be a high-risk genotype for INH/sulfasalazine-induced hepatotoxicity [10, 11].

21.3

NAT2 Genotyping

Although a relationship between INH-induced hepatotoxicity and NAT2 gene polymorphism has been reported, NAT2 genotyping is not done in daily medical practice. Difficulties include the complicated genotyping methods and the cost of such testing. The available methods for genotyping are as follows: (A) Polymerase chain reaction-restriction fragment length polymorphism (PCR-RFLP) analysis (B) Sequencing

N-acetyltransferase-2 Gene Polymorphism and Isoniazid-Induced Hepatotoxicity

533

(C) TaqMan probe assay (D) DNA microarray Methods (A) to (C) are already used in laboratories (12), and many modifications have been reported. Each method has its strengths and weaknesses, and the details are given elsewhere. In this section, the differences between NAT2 genotyping using DNA microarrays and other methods are explained. Method (A) is an established method that needs restriction enzymes. The amplified PCR products are digested by seven restriction enzymes, and the digested fragments are run on a gel for separation. Then the genotype is determined from the pattern of the fragments. However, the procedure is complicated. In addition, if digestion by the restriction enzyme is not adequate, it is possible to misjudge the genotype. The 341T/C SNP is included in NAT2*5B and so a limitation of the PCR-RFLP method is the lack of a suitable restriction enzyme to recognize this mutation in the NAT2 gene. To detect 341T/C, it is necessary to perform an additional mismatch PCR using PCR products that contain 341T/C to create the recognition site. Method (B) can simultaneously amplify nine SNPs of the NAT2 gene by using one pair of primers. However, it would be difficult in some cases to distinguish wild-type, heterozygous or mutant NAT2 alleles from the base sequences that are obtained. This method also needs an expensive sequencer. Method (C) requires expensive TaqMan probes for each SNP and a reaction apparatus. The reactions are performed in separate tubes and so the detection of all SNPs cannot be done simultaneously. This method is frequently used by laboratories with modifications to overcome its limitations and is effective at the laboratory level. Method (D), the DNA microarray, was developed in order to overcome the limitations of methods (A) to (C) [13]. Three different PCR primer sets ((x) for 191G/A, 282C/T and 341T/C; (y) for 434A/C, 481C/T and 590G/A; (z) for 803A/G, 845A/C and 857G/A) are used to obtain nine NAT2 SNPs. The PCR products obtained with the (x), (y) and (z) primers are mixed and are hybridized simultaneously to the DNA microarray. Signals are visualized by the peroxidase method, and thus are detectable using an office scanner. NAT2 genotypes can be obtained directly by hybridization. The strengths of method (D) (DNA microarray) compared with methods (A) to (C) are that it can detect 341T/C, does not need many restriction enzymes, does not need an expensive sequencer or a scanner to detect fluorescence, and is completed by hybridization of the DNA microarray after a one-step PCR, with the various NAT2 alleles being easy to detect. The limitation of method (D) is that glass slides with attached oligomers for NAT2 are required. However, glass slides are cheaper than a sequencer or scanner. An explanation of the method of NAT2 genotyping using a DNA microarray is shown in Figure 21.2. Three different samples are displayed in panels 1, 2, and 3. In lane 1 of all three panels, position markers are shown as black squares for the various SNPs (191G/A, 282C/T, 341T/C, 434A/C, 481C/T, 590G/A, 803A/G, 845A/C and 875G/A). Panel 1: NAT2*4/ ∗ 4 wild-type. In lane 2, only wild-type spots are shown as black squares. Panel 2: NAT2*4/ ∗ 6A. In lane 2, spots for 282C/C and 282C/T, as well as spots for 590G/A and 590G/G, are colored. These results indicate heterozygosity for SNPs 282 and 590. According to the combination of SNPs 282 and 590, the NAT2 phenotype was determined to be NAT2*4/*6A. Panel 3: NAT2*6A/ ∗ 6A. In lane 2, there is a spot for 282C/T, but that for 282C/C is not detected. This result indicates the presence of SNP 282. In addition, the spot for 590G/A can be seen, but not a 590G/G spot, which indicates the existence of SNP 590. From the combination of SNPs 282 and 590, the NAT2 phenotype was determined to be NAT2*6A/*6A.

534

Hepatotoxicity

NAT2*4/*4 Wild type C

NAT2*4/*6A Hetero type

NAT2*6A/*6A Mutant type

191G (W ) 191A (M )

C C

282C (W )

282 C/C

282T (M )

282 C/T

282 C/T

341T (W ) 341C (M )

C

434A (W ) 434C (M )

C

481C (W ) 481T (M )

C C

590G (W )

590 G/G

590A (M )

590 G/A

590 G/A

803A (W ) 803G (M )

C

845A (W ) 845C (M )

C

857G (W ) 857A (M )

1 Figure 21.2

21.4

2

3

An example of NAT2 genotyping using a DNA microarray

NAT2 Genotyping in Daily Medicine

SAs have a high risk of INH-induced hepatotoxicity [9]. Our study of treatment for tuberculosis and INH-induced hepatotoxicity in 42 Asian patients showed that 9.1 % of RAs, 26.7 % of IAs and 80 % of SAs had aminotransferase levels that were more than twice the upper limit of normal [13] (Figure 21.3). It was also reported that aminotransferases reached up to five times the upper limit of normal in 10–20 % of patients receiving INH alone [2]. Thus, to prevent the failure of anti-tuberculous therapy due to adverse drug reactions linked to hepatotoxiciy, NAT2 genotyping prior to INH administration should be useful for identifying SA patients. However, it would be difficult to perform NAT2 genotyping on all of the patients being treated with INH. Hepatotoxicity with an increase of aminotransferases occurs in about 20 % of patients taking INH, but the frequency of INH-induced hepatitis is lower than was previously thought. The risk of hepatitis that requires treatment as an adverse effect of INH is 2.7 % when INH is administered together with rifampicin [2]. It

N-acetyltransferase-2 Gene Polymorphism and Isoniazid-Induced Hepatotoxicity

*

100

Frequency of hepatotoxicity (%)

535

80 60 40 20 0 Total n = 42

RAs n = 22

IAs n = 15

SAs n=5

Figure 21.3 NAT2 phenotype and the frequencies of hepatotoxicity on patients receivign tuberculosis treatment, including INH. Closed bars show the frequency of hepatotoxicity of each NAT2 phenotype: total patients, 10/42 (23.8 %); RA, 2/22 (9.1 %); IA, 4/15 (26.7 %); SA, 4/5 (80 %). Statistically significance P∗ < 0.05 was evaluated by Fishrer’s exact test [13]

would be preferable to perform NAT2 genotyping on patients with a high risk of hepatitis, which include those with liver disease, alcohol intake and potential drug-interactions (for example, patients receiving anti-HIV drugs) [2, 5]. Although NAT2 genotyping is useful to predict the adverse effects of INH, there is no recommended regimen with a low dose of INH for SA patients. Pharmacokinetic analysis performed after a single dose of INH 1 has indicated that its t /2 is three times longer in SAs than in RAs [6]. On the basis of this result, the INH dose should be reduced to 1/3 in SAs. It is also necessary to study the pharmacokinetics of INH during long-term administration and compare them with theNAT2 phenotype. In conclusion, INH is usually part of first-line anti-tuberculous therapy. If NAT2 genotyping could be done in daily practice, it would be useful to determine a suitable INH dose based on the NAT2 genotype, so as to achieve efficacy with less risk of inducing the-resistance of M. tuberculosis to INH and so as to avoid hepatotoxicity. In addition, it seems to be necessary to establish the cost-effectiveness of NAT2 genotyping, as well as the optimum frequency of testing for hepatotoxicity according to NAT2 phenotype (RAs, IAs, or SAs). Future research that combines genomics, pharmokinetics, and bacteriology will contribute to preventing the failure of anti-tuberculous therapy.

References [1] World Health Organization. Global tuberculosis control – surveillance, planning, financing. WHO Report 2006 (2006). [2] American Thoracic Society/Centers for Disease Control and Prevention/Infectious Diseases Society of America. Treatment of Tuberculosis. Am. J. Respir. Crit. Care Med., 167, 603–662 (2003).

536

Hepatotoxicity

[3] Frieden TR, Sterling TR, Munsiff SS, Watt CJ and Dye C. Tuberculosis. Lancet, 362, 887–899 (2003). [4] Kita T, Tanigawara Y, Chikazawa S, Hatanaka H, Sakeda T, Komada F, Iwakata S and Okumura K. N-acetyltransferase-2 genotype correlated with isoniazid acetylation in Japanese tuberculous patients. Biol. Pharm. Bull., 24, 544–549 (2001). [5] Harries AD, Chimzizi R and Zacharah R. Safety, effectiveness and outcomes of concomitant use of highly active antiretroviral therapy with drugs for tuberculosis in resource-poor settings. Lancet, 367, 944–945 (2006). [6] Parkin DP, Vandenplas S, Botha FJ, Vandenplas ML, Seifart HI, van Helden PD, van der Walt BJ, Donald PR and van Jaarsveld PP. Trimodality of isoniazid elimination: phenotype and genotype in patients with tuberculosis. Am. J. Respir. Crit. Care. Med., 155, 1717–1722 (1997). [7] Gaikovitch EA, Cascorbi I, Mrozikiewicz PM, Brockmoller J, Frotschl R, Kopke K, Gerloff T, Chernov JN and Roots I. Polymorphisms of drug-metabolizing enzymes CYP2C9, CYP2C19, CYP2D6, CYP1A1, NAT2 and of P-glycoprotein in a Russian population. Eur. J. Clin. Pharmacol., 59, 303–312 (2003). [8] Kalow W. Interethnic variation of drug metabolism. Trends Pharmacol. Sci., 12, 102–107 (1991). [9] Ohno M, Yamaguchi I, Yamamoto I, Fukuda T, Yokota S, Maekura R, Ito M, Yamamoto Y, Ogura T, Maeda K, Komuta K, Igarashi T and Azuma J. Slow N-acetyltransferase-2 genotype affects the incidence of isoniazid and rifampicin-induced hepatotoxicity. Int. J. Tuberc. Lung. Dis., March, 256–261 (2000). [10] Huang YS, Chern HD, Su WJ, Wu JC, Lai SL, Yang SY, Chang FY and Lee SD. Polymorphism of the N-acetyltransferase-2 gene as a susceptibility risk factor for antituberculosis drug-induced hepatitis. Hepatology, 35, 883–889 (2002). [11] Kumagai S, Komada F, Kita T, Morinobu A, Ozaki S, Ishida H, Sano H, Matsubara T and Okumura K. N-acetyltransferase-2 genotype-related efficacy of sulfasalazine in patients with rheumatoid arthritis. Pharm. Res., 21, 324–329 (2004). [12] Chen X and Sullivan PF. Single nucleotide polymorphism genotyping: biochemistry, protocol, cost and throughput. Pharmacogenomics J., 3, 77–96 (2003). [13] Shimizu Y, Dobashi K, Mita Y, Endou K, Moriya S, Osano K, Koike Y, Higuchi S and Yabe S, Utsugi M, Ishizuka T, Hisada T, Nakazawa T and Mori M. DNA microarray genotyping of Nacetyltransferase-2 polymorphism using carbodiimide as the linker for assessment of isoniazid hepatotoxicity. Tuberculosis (Edinburgh), 86, 374–381 (2006).

Section 6 Gender Differences in Hepatotoxicity

22 Human and Animal-Based Differences in Hepatic Xenobiotic Metabolism and Toxicity Peter J. O’Brien, Katie Chan and Raymond Poon

22.1

Introduction

The risk of developing a particular disease is often different for women and men. Women are more likely to develop lupus, immune thrombocytopenic purpura, osteoporosis, fibromyalgia, thyroid cancer, irritable bowel syndrome, migraine and depression, whereas men are more likely to develop strokes, gout, alcoholic and cancer of the lungs, kidneys, bladder and pancreas. Women are also more likely to develop drug-induced adverse reaction diseases. While gender-specificity of rodent and pig hepatic enzymes has been well studied, the gender effects in human drug pharmacokinetics and pharmacodynamics have only been researched recently. This was partly because following the thalidomide and diethylstilbestrol tragedies, the US FDA in 1977 issued a guideline recommending exclusion of women of childbearing potential in clinical phase I and early phase II trials, except for life-threatening diseases. In 1993, the US FDA implemented new guidelines for drug trials which included both women and men in the entire drug development and clinical evaluation. The gender differences found have generally been thought to be rare. They were usually small and were attributed to the higher body fat percentage and lower average body weight of women that limited the volume of distribution of hydrophilic drugs, e.g. fluoroquinolones, ethanol and theophylline. Normalizing pharmacokinetic parameters for lean or total body weight corrected for this. Women also have a higher average organ blood flow and a smaller average plasma volume. Endogenous estrogens were also thought to slightly decrease drug binding


540

Hepatotoxicity

to plasma alpha-1-acid glycoprotein (Gandhi et al., 2004). In the following, this review will focus on drug metabolizing enzyme activity.

22.2

Gender-Specific Animal Hepatic Enzymes

The pharmacological activity, metabolism and toxicity of many drugs or xenobiotics often depend on the gender of all strains of rats and some strains of mice. In 1932, Nicholas and Barron were the first to show that amobarbital anesthetized female rats at half the dose required for male rats (Nicholas and Barron, 1932). Amobarbital, hexobarbital, amytal and pentobarbital, but not barbital, phenobarbital or chloral hydrate, also markedly extended the duration of anesthesia in female rats compared to male rats (Holck et al., 1937). Soon after this, it was shown that there were also gender differences in the hepatic steroid metabolism (Hubener and Amelung, 1953).These gender differences in rats were not found until 30 days of age and were later attributed to gender-specific P450 catalysed drug or steroid metabolism. Continuous neonatal androgen exposure was also required for the development of male-specific steroid metabolism. Female rats were also more susceptible to nicotine and strychnine (Axelrod, 1956). Female rats demethylated morphine at one-tenth the activity of males. Treating female rats with testosterone increased the enzyme activity, while administration of estradiol to male rats markedly lowered their enzyme activity. Thus, sex hormones are likely involved in the regulation of drug metabolizing enzyme systems (Axelrod, 1956). An intact pituitary gland was also necessary for the masculinizing action of androgens and for the feminizing action of estrogens as these actions were prevented by hypophysectomy (Mode and Gustafsson, 2006). Gender differences in P450 expression are also present in mice and pigs. In particular, CYP2A expression is very low in male compared to female mini- and micropigs (Gillberg et al., 2006). Gender differences in drug metabolism occur in numerous species besides mammals – this includes fish, reptiles and birds. 22.2.1

Hepatic Metabolism by Male versus Female Animal CYP Enzymes

Even the activity of P450 reductase can be gender-dependent, as female CBA mice have 2.5-fold more activity than that of male mice (Lofgren et al., 2004). NADPH:P450 reductase functions by reducing P450. This enables P450 to reductively activate oxygen to form a ferryl hydroxylating species that hydroxylates drugs, steroids and xenobiotics. 22.2.1.1

CYP1A1, 1A2 and 1B1 Enzymes

Both CYP1As catalyze the bioactivation of polyaromatic hydrocarbons and heterocyclic amines to mutagenic and toxic metabolites. The activity of pig CYP1A1 was four times higher in female than in male minipigs (Skaanild and Friis, 1999). However, mouse CYP1A2 was twice as active for males than females (Lofgren et al., 2004). 22.2.1.2

CYP2A Enzymes

CYP2As catalyze the metabolism of numerous toxic xenobiotics, including coumarin, aflatoxin B1, N-nitrosodiethylamine, 4-(methylnitrosamine)-1-(3-pyridyl)-1-butanone (NNK), nicotine, cotinine, 1,3-butadiene, acetaminophen and 2,6-dichlorobenzonitrile. Human

Differences in Hepatic Xenobiotic Metabolism and Toxicity

541

CYP2A6 polymorphisms and nicotine consumption may determine the degree of addiction. In rats, CYP2A1 and CYP2A2 are predominately expressed in the liver. CYP2A1 is female-predominant as it is suppressed at puberty in male rats. However, CYP2A2 is activated and predominates in males. CYP2A5 is female-predominant in mice and CYP2A is female-predominant in minipigs. Coumarin 7-hydroxylase is the marker enzyme activity for CYP2A and the levels of hepatic microsomal activities found were human > horse > dog, with an approximately 2-fold higher activity for females than males (Chauret et al., 1997). This gender-dependence is under the regulation of the nuclear receptor, constitutive androstane receptor (CAR), which has been shown in mice to be activated by oestrogens and deactivated by androgens Coumarin is not hepatotoxic in humans and is predominately metabolized by CYP 2A6 which catalyzes coumarin oxidation to 7 hydroxy-coumarin and is then glucuronidated. Rats, unlike humans, are highly susceptible to coumarin-induced hepatotoxicity. This is possibly because of their low coumarin 7-hydroxylase activity and their CYP1A2 and CYP3A catalyzed coumarin metabolism to form the 3,4-epoxide which can rearrange to the toxic reactive intermediate o-hydroxyphenyl acetaldehyde (Born et al., 1999). 22.2.1.3

CYP2C Enzymes

The levels of CYP2C11 and CYP2C13 are low in immature rats but increase in male rats, but not female rats during puberty. In mature rats, males have much higher levels of CYP2C11 and CYP2C13 than female rats. The gender difference in steroid metabolism was steroid 16α-hydroxylase for male rats and steroid sulfate 15β-hydroxylase for female rats. Higher drug metabolic activity found for male rats over female rats were the O-depropylation of 7-propoxycoumarin, the hydroxylation and the N -demethylation of aminopyrine or benzphetamine, but not the N -demethylation of ethylmorphine. These activity differences were attributed to sex-specific male constitutively expressed CYP2C11 or CYP2C13 versus female constitutively expressed CYP2C12 (Kato and Kamataki, 1982). These CYP2 genes and other steroid hydroxylase P450s are regulated by growth hormone. Plasma growth hormone secretion by the pituitary is pulsatile which in male rats has peaks of large amplitudes having a periodicity of approximately 3.3h with undetectable levels inbetween. However, in female rats growth hormone is secreted in more frequent pulses of lower amplitude. Hypophysectomy of female rats leads to a decreased CYP2C12 whereas hypohysectomy of male rats increased CYP2C11 and 13 activities. This was found to be due to the gender different effects of growth hormone which stimulated the expression of CYP2C11 but suppressed CYP2C12 expression (Mugford and Kedderis, 1998). Treatments with testosterone or estradiol failed to have an effect. The use of NSAIDs are associated with GI toxicity in a large number of patients with hepatitis developing in a few patients. CYP2C is the P450 responsible for catalyzing the metabolism of NSAIDs. Rat or human hepatocyte toxicity induced by diclofenac seems to involve activation by CYP2C and CYP3A to form reactive quinoneimines as cytotoxicity increased when rat hepatocyte CYP3A was induced by phenobarbital (Wang et al., 2004). Human and rat hepatocyte diclofenac cytotoxicity was prevented by the CYP2C, 3A inhibitors ketoconazole 25uM and proadifen 10uM but not the CYP1A1/2 inhibitor α-naphthoflavone (Bort et al., 1999). Microsomes prepared from male rat liver catalyzed the formation of a 51 kDa adduct, identified as an adduct of a diclofenac reactive metabolite and was gender-specific for male CYP2C11 activity. Microsomes from female liver, on the

542

Hepatotoxicity

other hand, did not form an adduct (Shen et al., 1997). CYP2C11 but not CYP2C12 also catalyzed the metabolic activation of limonene to the 8,9 epoxide which could contribute to the nephropathy induced in male but not female rats (Miyazawa et al., 2002). Tributyltin undergoes metabolic dealkylation catalyzed by CYP2C11 and CYP 2C6 in male rat liver whereas the dealkylation in female rats is less as it is only dealkylated by CYP2C6 (Ohhira et al., 2006). Tributyltin caused nephro-, hepato- and neurotoxicity in mammals and are also known as endocrine disrupters causing developmental and reproductive toxicities in mammals. The hepatotoxic mechanism seems to involve impaired oxidative phosphorylation caused by the accumulation of tributyltin in the mitochondria (Yoshizuka et al., 1992). Whether the gender difference in dealkylation increases liver susceptibility to tributyltin in female rats is not known. 22.2.1.4

CYP2E1 Enzymes

The basal activity of hepatic CYP2E1 was 28% higher in female than male SD rats (Morel et al., 1999). However this basal activity was four times higher in female than male minipigs (Skaanild and Friis, 1999). 22.2.1.5

CYP3A Enzymes

CYP3A2 is a P450 isoform which declines in female rats but not in male rats as they go through puberty. CYP3A exhibited only minor gender differences in male and female pigs (Skaanild and Friis, 1999). Aflatoxin B1 is metabolically activated by CYP3A2 in male rats. Male rats were also more susceptible to both aflatoxin metabolic activation and carcinoma of the liver. Carcinoma of the liver develops after a longer period in the female rat (Prince and Campbell, 1982). The acute LD50 for aflatoxin B1 was also much lower in the male rat compared to the female rat (Zimmerman, 1999). Male rodents were also more susceptible to chemical-induced skin cancer and estrogen-induced kidney tumors (Gurtoo and Parker, 1977). CYP3A2 accounts for the 2–5-fold higher ethylmorphine deethylase activity and acetylaminofluorene ring hydroxylation in male rats than female rats. CYP3A2 also accounts for the lower N-oxidation activities in male rats than female rats (Kobliakov et al., 1991). Male rats also had a much higher benzopyrene arylhydroxylase activity to form phenols. How this gender difference affects the carcinogenesis susceptibility of male rats to acetylaminofluorene or benzopyrene has not been reported. Rat CYP3A9 is expressed 28 fold higher in female rat compared with males. Murine CYP3A44 and CYP3A41 are expressed mostly in female liver with little in the male liver and no expression in brain or kidney (Anakk et al., 2007). This has been attributed to the regulatory influences of growth hormone and estrogen and testosterone. 22.2.2

Gender-Based CYP Inhibitors

Metyrapone (1 mM), cimetidine (1 mM) and chloramphenicol are broad-spectrum CYP450 inhibitors, e.g. chloramphenicol inhibits human CYP2C9, CYP3A4 and rat CYP2B1/ 2,2C11 (Park et al., 2003; Poet et al., 1996). Furthermore ketoconazole, the potent and selective human CYP3A4 inhibitor needed a 100-fold higher concentration (5 uM) to inhibit rat CYP3A1/2 activity and now inhibited CYP1A2, 2C6, 2C11activities as well (Eagling et al., 1998).


22.2.3

543

Gender-Based Hepatic CYP Activators and Enhancers

Gender differences exist in the 2-10 fold activation of hepatocyte CYP3A2 by caffeine, flavone or theophylline. This occurred for hepatocytes isolated from male rats but not for hepatocytes isolated from female rats. The first demonstration that oxidative drug metabolism could be increased in vivo with a CYP activator was shown when caffeine administration to rats in vivo caused an increase in drug metabolising activity (Mitoma et al., 1968). At first, this was attributed to CYP induction, but drug metabolism catalyzed by rat liver microsomes in vitro was also enhanced by caffeine addition. Caffeine or theophylline also potentiated acetaminophen hepatotoxicity in uninduced or phenobarbitalinduced male rats and prevented hepatotoxicity in 3MC-induced male rats (Kalhorn et al., 1990; Sato et al., 1985). Acetaminophen hepatotoxicity in rats has been shown to be catalyzed by CYP1A2, 2E1 and 3A2. Caffeine enhancement of acetaminophen hepatotoxicity in male rats was attributed to activation of CYP3A2 (a rat male constitutive P450). This was because caffeine or flavone increased the male rat liver microsomal-catalyzed activation of acetaminophen to the quinoneimine more than 4-fold, but did not affect female rat liver microsomal catalytic activity. 22.2.4 22.2.4.1

Gender-Dependent CYP Induction AHR-Regulated CYP1A1/2 Enzyme Induction

The aromatic hydrocarbon receptor is a cytosolic protein that binds planar ligands, e.g. tetrachlorodibenzo- p-dioxin (TCDD) or methylcholanthrene. It also binds endogenous indole derivatives, e.g. tryptophan, bilirubin, heterocyclic amines and dietary constituents, e.g. resveratrol. The ligand binding triggers the transformation of this receptor into the nuclear form, a heterodimer consisting of the AHR and the AHR nuclear translocator which in the nucleus acts as a ligand-activated transcription factor. The heterodimer interacts with dioxin responsive elements in the CYP1A1 gene, leading to mRNA synthesis which leaves the nucleus and initiates CYP1A1 protein synthesis. Induction of CYP1A2 and CYP1B1 synthesis is also AHR-mediated. CYP 1A is sexually dimorphic in the Sprague–Dawley (SD) rat, as hepatic CYP1A1 and CYP1A2 are more readily induced in the female rat than the male rat by heterocyclic amines pyridine or nicotine, which are major constituents of tobacco smoke (Iba et al., 1999). More men than women develop lung cancer, but the gap is narrowing in recent years. There is a report that women may be more susceptible to environmental tobacco smoke when 4-aminobiphenyl-hemoglobin adducts were measured (Airoldi et al., 2005). Tetrachlorodibenzo- p-dioxin (TCDD) is the most effective agent for inducing CYP1A1,2 and 1B1. TCDD is also a liver carcinogen in female but not male rats which could suggest that ovarian hormones play a role. 22.2.4.2

Orphan Nuclear Receptor-Regulated CYP2A, CYP2B and CYP3A Enzyme Induction

The nuclear orphan receptors CAR (constitutive androstane receptor) and PXR (pregnane X receptor) are abundantly expressed in liver and activate gene transcription in a constitutive way. Both receptors bind to various structurally unrelated ligands which regulate genes involved in the humoral response to both endobiotic and xenobiotic stress. The administration of phenobarbital (PB) translocates cytoplasmic CAR into the nucleus of liver or primary

544

Hepatotoxicity

hepatocytes. CAR in the nucleus forms a heterodimer with retinoid X receptor and binds to and activates NR1 enhancer within the conserved PB response element in CYP2B genes. Other CAR ligands include pesticides, chlorpromazine, polychlorinated biphenyls and organic solvents. The administration of pregnenolone 16α-carbonitrile (PCN) or rifampicin (a macrolide antibiotic) selectively activate PXR, resulting in increased transcription of the CYP3A gene. The steroid dexamethasone also activates both the glucocorticoid receptor (GR) and PXR which induces UGT1A1 and CYP3A2. Both CYP2B and CYP3A genes are also regulated by phenobarbital and clotrimazole. Recently a selective ligand for CAR has been discovered (Maglich et al., 2003). CYP2A, SULT1A1, UGT1A1 may also be induced depending on the CAR ligand and the biological species including human (Maglich et al., 2003). CAR and PXR thus regulate an overlapping set of genes that encode xenobiotic and endobiotic detoxifying enzymes. The steroids androstanol and androstenol are odorous steroids produced in the testis that are secreted into apocrine glands and urine. They and progesterone act as naturally occurring inverse agonists that reverse transcriptional activation by CAR and thus inhibit CAR constitutive activity (Forman et al., 1998). Estradiol and estrone formed by the ovary on the other hand reactivate CAR. NR1 was activated by estradiol but was repressed by progesterone or androgens in mouse primary hepatocytes. Furthermore, the latter repression was fully restored by estradiol (Kawamoto et al., 2000). These effects could contribute to gender-based drug metabolism differences. Minipig liver microsomes have similar CYP3A levels, although they have lower CYP1A, 2A, 2C and 2E levels than humans (Anzenbacher et al., 1998). However, CYP3A and CYP1A2, 2A6, 2E1 levels in female minipig liver microsomes are similar to human liver microsomes. Gender did not significantly affect microsomal CYP3A levels (Skaanild and Friis, 1999). Furthermore, castration of minipigs increased liver microsomal CYP2A activity (coumarin 7-hydroxylase ) by 12-fold and CYP3A by 2.5-fold, indicating that testicular products repressed NR1. Phenobarbital or CITCO (6-(4-chlorophenyl)imidazo[2,1-b] (1,3)thiazole-5-carbaldehyde O-3,4-dichlorobenzyl)oxime) incubated with female minipig hepatocytes induced CYP2A activity 14-fold, whereas androstenol was ineffective (Gillberg et al., 2006). This indicates that activation of NR1 by CAR ligands induced CYP2A levels. Because female rats have much lower CYP2B levels and testosterone 16β-hydroxylase activities than male rats, CYP2B induction by phenobarbital is much greater in female rats than in male rats. For similar reasons, CYP3A (testosterone 6β-hydroxylase) induction by pregnenolone-16α-carbonitrile was also much greater in female rats than in male rats. The low CYP3A activity in females could be attributed to their very low CYP3A2 levels in uninduced rat liver (Parkinson et al., 2004). Perhaps because of this the Ministry of Health, Labor and Welfare in Japan has recommended that induction studies in rats be conducted in female rats rather than male rats (www.nihs.go.jp/drug/DrugDiv-E.html). Interestingly, dexamethasone which induces CYP3A, decreased male-specific CYP2C11 and attenuated the sex differences in metabolism (Kato and Yamazoe, 1992). The expression of CYP3A2 was also suppressed by methylcholanthrene. 22.2.5

Gender Differences In the Induction of Other P450s

CYP2C11 is male specific and its expression is regulated by pulsatile growth hormone secretion at the transcriptional level. However, interestingly methylcholanthrene while inducing CYP1A1 can decrease male-specific CYP2C11 protein levels and attenuate the


545

sex differences in metabolism (Kato and Yamazoe, 1992). This arylhydrocarbon-downregulation of CYP2C11 occurred by a different negative transcriptional mechanism from CYP1A1 induction (Bhathena et al., 2002). Dexamethasone also decreased CYP2C11 protein levels (Kato and Yamazoe, 1992). Carbon tetrachloride was more hepatotoxic in male rats than female rats which was attributed to CYP2E1 being more readily induced by carbon tetrachloride in male rats (Morel et al., 1999; Zimmerman, 1999). 22.2.6

Examples of P450 Gender-Based Animal Hepatotoxins

1,4 dichlorobenzene is carcinogenic in male rats but not female rats. It is metabolised two fold faster by female liver microsomes but covalently binds to microsomes half as much as that catalysed by male liver microsomes. This is likely because the female forms much less of the 2,5-dichlorohydroquinone metabolite, which is readily oxidized to a quinone electrophile that binds to DNA. The P450s that metabolically activate 1,4 dichlorobenzene have not been identified but rat CYP2E1 or human CYP3A have been implicated (Nedelcheva et al., 1998). Benzene induces lipid peroxidation in rat liver and bone marrow. As well, benzene induces lipid peroxidation in bone marrow and leukemia in humans. Oxidative stress toxicity is induced in the liver, kidney and lungs of female rats following benzene administration more readily than with male rats. This has been attributed to metabolic activation to hydroquinone by the higher levels of hepatic CYP2E1 found in female rats than male rats (Verma and Rana, 2004). Higher GST levels in male rats may also contribute (Verma and Rana, 2003). Toluene is believed to be the toxic agent in glue sniffing and the short-term abuse coma is believed to result from the anesthetic properties of toluene. However, the long-term exposure resulting in CNS depression, organ failure and bone marrow toxicity has been attributed to toluene oxidation catalyzed by CYP2E1 and CYP2C11 to form benzoic acid which caused acidosis and coma. Male rat liver microsomes oxidatively metabolize toluene more rapidly than female rats. If toluene is the cause of toxicity, then female rats would be more susceptible, whereas if benzoic acid acidosis is the cause of toxicity then male rats would be more susceptible. If this extrapolates to humans, then the toxic effects of toluene to adolescent females would last longer (Shimamoto et al., 1999). The acute LD50 of Aflatoxin B1 in female rats is much higher than in the male rat. In addition, carcinoma of the liver develops after a longer period in the female. This has been attributed to higher P450 levels (CYP3A2) and lower epoxide hydrolase and GST levels in male rats (Zimmerman, 1999). Orotic acid, a precursor of pyrimidine nucleotide biosynthesis, induced a more fatty liver in female rats than male rats (Negishi and Aizawa, 1975). This is probably because the protective effects of androgens (Zimmerman, 1999). Phalloidin, a hepatotoxin from poisonous mushrooms, induced liver injury in female more than male rats while androgenic steroids enhanced the phallodin resistance of female rats (Zimmerman, 1999). Cocaine induced liver damage in male mice much more readily than in female mice and was associated with increased oxidative stress caused by lower glutathione and catalase but higher glutathione reductase activity (Visalli et al., 2005). 22.2.7

Gender-Based Phase II Enzymes

Phase II enzymes are also regulated by the hypothalamic–pituitary axis interactions with the sex hormones (Mugford and Kedderis, 1998).

546

Hepatotoxicity

22.2.7.1

Sulfotransferases (SULTs)

There are five major cytosolic sulfotransferases (phenol, glucocorticoid, hydroxysteroid, bile acid and aryl) that catalyze the detoxification of xenobiotics to form water-soluble sulfates. The SULT activities have high sexual dimorphism in which the male SULTs are activated at puberty but have low activity in female liver throughout their life span. Male rat livers have 2–3 higher aryltransferases and p-nitrophenol sulfotransferase activity (Mulder, 1986). Their SULT activities are catalyzed by three male-dominant SULTs, i.e. SULT1C1 (N-hydroxy-2-acetylaminofluorene), SULT1E2 (estrogen, estradiol, estrone and estriol) and SULT1A1 (phenol). Rat hepatic SULT1C1 was also greater in male relative to female rats (Kester et al., 2003). The carcinogenesis induced by N-hydroxy-2-acetylaminofluorene in male rats was also attributed to the gender difference in SULT1C1 (DeBaun et al., 1970). By contrast, female rats have higher levels of expression of hydroxysteroid, glucocorticoid and bile acid sulfotransferase activities (Mulder, 1986) that have been attributed to female-dominant hydroxysteroid SULTs, i.e. SULT-20/21, SULT-40/40 and SULT-60, which were responsible for the higher hydroxysteroid mRNA levels found in the livers of female rats (Runge-Morris and Wilusz, 1991; Runge-Morris, 1994). Sex steroid hormones may contribute to SULT activity as SULT activity in adult males was decreased 80 % by castration and was restored by testosterone administration. However, ovariectomy of female rats increased the SULT activity by 28 % which was increased 3-fold by testosterone. Hypophysectomy decreased the SULT activity for N-hydroxy-2acetylaminofluorene, glucocorticoids, hydroxy-steroids and bile acid in female rats more than male rats and was restored by growth hormone infusion in a pulsatile pattern whereas testosterone had no effect. Estrogen administration to male rats also decreased the SULT activity. SULT1E2 expression in rat liver is regulated by sex steroids and growth hormone, whereas SULT1C1 expression is mostly determined by the male GH secretory pattern (Runge-Morris, 1994).

22.2.7.2

Glucuronosyl Transferases (UGTs)

In male rats, glucuronosyl transferases have 50–65 % higher protein levels and expression levels than female rats, e.g. p-nitrophenol (Catania et al., 1995). However, estradiol and estrone glucuronidation (UGT2B1) was higher in female rats than male rats (Zhu et al., 1996). Bisphenol-A is an estrogenic endocrine disrupting-chemical, widely used in food packaging. The higher serum bisphenol A levels in female compared to male rats was attributed to poorer glucuronidation by hepatic UGT2B1 (Takeuchi et al., 2004). Ovariectomy of female rats decreased estrone glucuronidation by 43 %, which was restored by estradiol (Rao et al., 1977). Glucuronidation is the major metabolic pathway for thyroid metabolism. Fasting or food restriction induced triiodothyronine UGT i.e. bilirubin UGT in male rats than in females. However triiodothyronine UGT i.e. androsterone UGT was markedly decreased by food restriction in female rats but not male rats (Visser et al., 1996). UGT1A1 and UGT2B3 is repressed by testosterone in male rats whereas UGT1A1 is induced by progestins in female rats (Strasser et al., 1997). Male predominant expression in mice was observed for liver UGT2b1 whereas female predominant expression was observed for liver UGT1a1 and 1a5. Gender differences were also found for male and female GI tract, kidney, lung and nasal epithelia UGTs (Buckley and Klaassen, 2007).

Differences in Hepatic Xenobiotic Metabolism and Toxicity 22.2.7.3

547

Glutathione Transferases (GSTs)

There are four major cytosolic glutathione transferases, alpha, mu, theta and pi, which mostly detoxify xenobiotics to form water-soluble glutathione conjugates. Glutathione transferases also usually have a 2–3 higher activity in male rats than female rats (Mulder, 1986; Srivastava and Waxman, 1993). However, there are some substrates which are glutathionylated more rapidly in female rats (Mugford and Kedderis, 1998) Hypophysectomy of female rats increased the expression of male-specific GSTs, whereas hypophysectomy and continuous growth hormone infusion of male rats suppressed the expression of male specific GSTs (Srivastava and Waxman, 1993). 22.2.8 22.2.8.1

Gender-Based Oxygen Metabolism Glutathione (GSH) Homeostasis

Liver GSH levels in rats at 12 weeks of age but not at 7 weeks of age were higher in males than females. GSSG reductase and GST activities were also higher in males. However, liver GSH peroxidase and gamma-glutamyl transpeptidase activities were 20–30 % higher in female than male rats (Igarashi et al., 1983).

22.3

Human Liver

Of 300 drug applications submitted to the FDA between 1994 and 2000, only 20 % of drugs showed different pharmacokinetics in women to men (http://www.fda.gov/fdac/features/ 2005/405 sex.html#gender). In the following, some of the gender-based drug metabolizing enzyme activities reported in humans are reviewed. The hypothalamic–pituitary axis is also involved in CYP-catalyzed metabolism as it is in rats but is much less marked and is expressed differently. Although women have higher mean-growth hormone serum concentrations than men, the growth hormone is secreted in a pulsatile, Circadian pattern in men, resulting in widely fluctuating GH peaks, whereas in women there is less of an amplitude difference in the GH levels (Winer et al., 1990). 22.3.1

Gender-Based Hepatic CYP Activities

In contrast to the rat liver microsomes and hepatocytes, it is generally agreed that there is such widespread variability in CYP enzyme activity in human liver microsomes and hepatocytes from both male and female donors that it is difficult to statistically establish significant gender differences. The variation of hepatic CYP levels in the adult due to genetic predisposition and the individuals CYP induction status due to their diet, lifestyle, age or drug intake, can be greater than 50-fold for a particular CYP. These factors would easily outweigh gender differences. CYP down-regulation by inflammatory diseases suffered by patients can also contribute to differences in hepatic CYP activities. 22.3.1.1

CYP1A2 Enzymes

CYP1A2 accounts for nearly 15 % of the cytochrome P450s in human liver and catalyzes the metabolism of caffeine, theophylline, imipramine, clozapine and propanolol. In spite of the widespread variability of human hepatic liver microsomes, CYP1A2 activity

548

Hepatotoxicity

(7-ethoxyresorufin O-dealkylation) was also slightly higher in male donors (statistical significance, P = 0.030) (Parkinson et al., 2004). CYP1A2 in vivo activity, as determined by the 6 h urinary caffeine metabolic ratio in 378 twins after administrating a single dose of 200 mg caffeine, was higher in men than in women. CYP1A2 was also induced by smoking and was inhibited by oral contraceptives (Rasmussen et al., 2002). Although the data are sparse, the clearance of theophylline, caffeine, thiophene and thiothixene was also slower in women than men. Smoking induced CYP1A2 also increased the clearance of theophylline or thiothixene (Harris et al., 1995). This gender difference could increase the buildup of CYP1A2 drug metabolites in women and increase their risk of side effects. There is also a higher incidence of tobacco smoke-related lung cancer in women than men and the heterocyclic amines found in the smoke also induced CYP1A1 and CYP1A2 more readily in female rats than male rats (Iba et al., 1999). 22.3.1.2

CYP2A Enzymes

In humans, CYP2A6 represents up to 15 % of cytochrome 450 proteins in human liver microsomes and accounts for nearly all of human coumarin hydroxylase activity. The 7 hydroxy-coumarin formed is then detoxified by glucuronidation and could explain the lack of coumarin-induced hepatotoxicity in humans, in contrast to rats which are susceptible to hepatotoxicity and have much less CYP2A. Human CYP2A6 also detoxified nicotine and its product cotinine. Gender differences are difficult to assess as interindividual variability is reported to be very large, 10–100-fold (Pelkonen et al., 2000). It has been reported to be under the regulation of the constitutive androstane receptor (CAR) and is slightly induced by rifampicin and antiepileptic drugs, e.g. phenytoin and phenobarbital. Rifampicin primarily activates the pregnane X receptor (PXR), whereas phenobarbital activates the constitutive androastane receptor (CAR). Although CYP2A is gender-specific in minipigs, no gender effects have been reported yet in humans. 22.3.1.3

CYP2C Enzymes

In humans, CYP2C is responsible for metabolism of a variety of therapeutic drugs, such as warfarin, mephenytoin, omeprazole and antiinflammatory drugs (Ibeanu and Goldstein, 1995). Even though CYP2C is gender-specific in rats, it is unclear whether humans express gender differences in this enzyme, as only a few reports have been published on any gender differences involving CYP2C. A trend for higher CYP2C19 activity was seen in female human hepatic microsomes compared to male (Parkinson et al., 2004). A clinical investigation, based on (R)-mephobarbital metabolism showed that young women have lower CYP2C19 activity compared to males, which contradicted other clinical studies which showed no differences (Hooper and Qing, 1990). However, subsequent studies revealed that apparent gender differences were attributed to concomitant use of oral contraceptive steroids, which decreased CYP2C19 activity by two-thirds, and could explain why differences were only seen between young women (Meibohm et al., 2002). 22.3.1.4

CYP2D6 Enzymes

CYP2D6 accounts for an average of 2 % of hepatic cytochrome P450s, but plays a major role in the metabolism of 25 % of therapeutic drugs which include antidepressants, antipsychotics, beta blockers and cardioactive drugs. It is also the most susceptible P450 to genetic polymorphism and so there is a wide range of activities from nul to gene


549

duplication depending on the ethnic group. There was also a trend for higher CYP2D6 levels and increased dextromethorpan O-demethylation activity in male compared to female hepatic microsomes (Parkinson et al., 2004). The clearance data for clomipramine, desipramine, propanolol and ondansetron also suggest a slower clearance in women, although the clomipramine data was not normalized for body weight (Harris et al., 1995). Clearance studies with patients lacking CYP2D6 (i.e. poor metabolizers of debrisoquine) still need to be carried out to show that a gender-dependent CYP2D6 is the cause of these genderdependent clearance results. This gender difference could increase the buildup of CYP2D6 metabolized drugs in women and increase their risk of side effects. 22.3.1.5

CYP2E1 Enzymes

CYP2E1 only forms 7 % of total hepatic cytochrome P450s and only metabolizes a few drugs, e.g. halothane, isoniazid, chlorzoxazone and ethanol. However CYP2E1 is important in the metabolic activation of a number of environmental contaminants such as trichloroethylene, butadiene, and benzene which are significant teratogens or mutagens. Because of the large interindividual variability it has not yet been established that females have higher CYP2E1 activity than males. However CYP2E1 in inbred minipigs, an animal model for human tissue was four times higher in females than in males (Skaanild and Friis, 1999). Hepatotoxicity induced by the anesthetic halothane is believed to be caused by a CYP2E1-catalyzed metabolic oxidation of halothane to the reactive metabolite trifluoroacetyl chloride, which covalently binds to endoplasmic reticular proteins, including CYP2E1. Some of the bound proteins act as neoantigens which induce an immune response involving antibody formation. A survey of 105 pediatric and 53 general anesthesiologists found that female pediatric anesthesiologists had higher levels of CYP2E1 autoantibodies. Female general anesthesiologists had higher levels of ERp58 autoantibodies than male anesthesiologists. It was suggested that the higher antibody responses in females could be caused by estrogens or pituitary hormones (Whitacre et al., 1999). The majority of these individuals did not develop hepatitis suggesting that these autoantibodies do not directly cause hepatitis (Njoku et al., 2002). 22.3.1.6

CYP3A Enzymes

CYP3A accounts for nearly 60 % of human hepatic cytochrome P450s and contributes critically to the first-pass metabolism of about one half of all drugs prescribed. It is also largely independent of genetic polymorphism. By contrast to the above CYPs, some drugs that are mainly metabolized by CYP3A showed a higher clearance in women than men, e.g. cyclosporine, erythromycin, tiralazad, verapamil, nifedipine, diazepam, alfentanil, methylprednisolone and cortisol (Parkinson et al., 2004). Furthermore, even though there was almost a 50-fold variability between donors, the median CYP3A4 protein levels were about 2-fold higher in liver microsomes from 46 female donors than 48 male donors. A higher expression for CYP3A4 mRNA transcripts in female donors suggested a pre-translational mechanism. CYP3A4 mRNA levels also correlated with two thirds of pregnane X receptor (PXR) expression but no gender-dependent expression of mRNA was found. Microsomal verapamil N-dealkylation activity was also 50 % higher from female donors than male donors (Wolbold et al., 2003). Cryopreserved hepatocytes also showed a 2-fold higher CYP3A4 activity in hepatocytes isolated from female donors (Parkinson et al., 2004). One explanation proposed was that P-glycoprotein levels of hepatocytes from female donors

550

Hepatotoxicity

were 2-fold lower compared to male hepatocytes which could result in higher intracellular drug and drug metabolite levels in females (Schuetz et al., 1995). Drugs metabolized by CYP3A are also usually substrates or inhibitors of P-glycoprotein (Cummins et al., 2002). However, others have found no significant differences between women and men in their liver P-glycoprotein or MDR1 mRNA levels and also point out that nifedipine and alfentanil are not substrates for P-glycoprotein (Wolbold et al., 2003). This CYP3A gender difference could decrease the effectiveness in women of CYP3A-metabolized drugs. While CYP3A4 is the major CYP450 in humans, CYP3A2 is only a minor P450 in uninduced male rats with no activity in female rats. CYP2A2, 2C9 and 3A2 drug-metabolizing enzymes in rats are gender-specific that have been attributed to the growth hormone/STAT5 pathway. Gender-specific CYPs have not been found in humans and so the relevance of the rodent studies for human studies has been questioned. However, a similar sex-dependent pulsatile, Circadian pattern of growth hormone secretion occurs in humans (Winer et al., 1990). Furthermore, when a pulsatile administration of growth hormone was given to growth-hormone-deficient humans, then CYP3A4 activity was decreased (erythromycin breath test). However, an increase in CYP3A4 activity occurred with a continuous infusion of growth hormone (i.e. analogous to female rats) (Jaffe et al., 2002). Furthermore, it has been recently shown that the hepatocyte nuclear factor 4α, which mediates or modulates PXR/CAR-mediated transcription of the CYP3A4 gene, also belongs to the transcription growth hormone-regulated network in rat liver (Tirona et al., 2003). While the pharmacokinetics of many drugs is similar in men and women, some drugs do show differences, e.g. estradiol is oxidatively metabolized more rapidly in women than men. 22.3.2

Orphan Nuclear Receptor-Regulated CYP2A, CYP2B and CYP3A Enzyme Induction and Gender Effects

Induction of CYP2B6 or CYP2C9/CYP2C19 by incubating hepatocytes with phenobarbital or rifampin, respectively, were similar in hepatocyes from male versus female donors. Induction of CYP3A4 with rifampin or phenobarbital or the induction of CYP1A2 with β-naphthoflavone was slightly higher in hepatocytes from female donors but may be insignificant because of the small sample size (Parkinson et al., 2004). In human HepG2 cells transfected with NR1 and mCAR, steroid-dependent NR1 was activated by estradiol but was completely repressed by progesterone or androgens. Furthermore, this repression was fully restored by estradiol (Kawamoto et al., 2000). These effects could contribute to gender-based drug metabolism differences. 22.3.3 22.3.3.1

Gender-Based Phase II Xenobiotic Metabolism Glucuronosyl Transferases (UGTs)

There are five major human hepatic endoplasmic reticular glucuronosyl transferases that mostly detoxify xenobiotics to form water-soluble glucuronides. The UGTs with their probe substrate (catalyzing the glucuronidation of drugs) are UGT2B7 (morphine, carbamezapin), UGT1A1 (bilirubin), UGT1A4 (amitripyline), UGT1A6 (serotonin) and UGT1A9 (propofol). Of these, UGT2B7 catalyzes the glucuronidation of 40 % of all drugs that are glucuronidated, whereas the rest are equally responsible for a further 47 % of drugs that are glucuronidated by the liver (Burchell et al., 2005). In general, pharmacokinetics studies


551

show no gender differences for many drugs. However, drugs with gender differences are often less metabolized by phase II metabolism in women, resulting in higher plasma concentrations. Drugs with higher plasma concentrations in women include certain antibiotics, some tricyclic antidepressants, lithium, acetaminophen and aspirin. The higher plasma concentrations for acetaminophen in women were attributed to a 22 % lower clearance and glucuronidation (Abernethy et al., 1982). Later, this was attributed to a 50 % lower UGT1A6 content of liver microsomes obtained from female compared to male donors (Court et al., 2001). Higher plasma propanolol levels were also found in women than in men, which could also be attributed to poorer glucuronidation (Walle et al., 1989). Propanolol was later shown to be glucuronidated in the liver catalyzed by UGT2B7 (Coffman et al., 1998). Diflunisal clearance was also lower in women and was attributed to lower glucuronidation (Macdonald et al., 1990). The anxiolytic benzodiazepine drugs oxazepam and temazepam also had a lower clearance in women in vivo (Mugford and Kedderis, 1998). Furthermore, oxazepam was mostly metabolized by glucuronidation catalyzed by liver microsomal UGT2B15 that was 60 % less active in microsomes obtained from women than men donors (Court et al., 2004). However, with temazepam there was no significant difference between genders once plasma protein binding was corrected for. Other drugs cleared largely by glucuronidation were also not affected by gender, e.g. clofibric acid and ibuprofen. 22.3.3.2

Catechol-O-methyltransferase (COMT)

COMT catalyzes the methylation of catechols and detoxifies catecholamines such as levodopa and methyldopa. Woman have been reported to have a 20–30 % lower hepatic COMT activity than men (Boudikova et al., 1990). 22.3.4

Drug–Female Sex Hormone Interactions

Oral contraceptive steroids or hormone replacement therapy can affect the metabolism of other drugs. Hormone replacement therapy can inhibit P450 (CYP1A2) catalyzedoxidation of tacrine, thereby increasing plasma tacrine levels (Laine et al., 1999). Oral contraceptive steroids can also inhibit caffeine oxidation catalyzed by CYP1A2 (Patwardhan et al., 1980). Oral contraceptive steroids induced acetaminophen glucuronidation by 78 % and P450 catalyzed oxidation by 36 %. Acetaminophen clearance was also increased (Miners et al., 1983). The steroid pregnenolone 16α-carbonitrile induced hepatic UGT1A1, phenobarbital-inducible UGT and CYP3A1 in rats (Ejiri et al., 2005). 22.3.5 22.3.5.1

Gender-Based Hepatotoxic Drugs Acetaminophen

Acetaminophen is the most widely used over-the counter analgesic in the USA. Data from the US Acute Liver Failure Study Group, a registry of more than 700 patients with acute liver failure across the USA, found acetaminophen poisoning in nearly 51 % of all acute liver failures in 2003 up from 28 % reported in 1998. While 43 % of these cases were attributed to suicide attempts, 48 % of these cases were attributed to unintentional overdoses due to use of multiple acetaminophen-containing preparations or impulsive overdosing when pain relief was not forthcoming. As described above, the lower acetaminophen glucuronidation rate found in women could be a contributing factor as the minimal effective therapeutic

552

Hepatotoxicity

dose for women could be expected to be less than that for a man. It would then be easier for a woman to unintentially overdose. 22.3.5.2

Troglitazone

Troglitazone (Rezulin) was an oral hypoglycemic agent used for glucose control by nearly 2 million patients with type 2 diabetes. It was withdrawn from the market because of hepatotoxicity in February 2000. In a review of 83 patients reported in the USA with liver failure following troglitazone therapy from March 1997–February 2000, 63 % were female whereas 36 % were male (Faich and Moseley, 2001). The mechanism of this hepatotoxicity gender difference is not known. However, it is known that troglitazone is mostly detoxified by the phase II metabolising enzymes, SULT1A3 (Honma et al., 2002) and UGT 1A1 and UGT2B7 (Yoshigae et al., 2000). However, whether metabolism by the phase I enzyme CYP3A4 (He et al., 2004) involves activation to form a cytotoxic phenoxyl radical that causes oxidative stress toxicity (Tafazoli and O’Brien, 2004) or whether unmetabolized troglitazone contributes to hepatotoxicity (Honma et al., 2002) remains controversial. 22.3.5.3

Cancer Chemotherapy

Doxorubicin is twice as likely to cause cardiotoxicity in females, even in childhood, than males. Females also have a significantly lower clearance of doxorubicin and epirubicin (another anthracycline). 5-Fluorouracil also has a lower clearance in females and causes more bone marrow toxicity. However, dihydropyrimidine dehydrogenase activity was not lower in females (DeLeve and Kaplowitz, 2003). The incidence of methotrexate-induced hepatic fibrosis cirrhosis was not affected by gender (Zimmerman, 1999). 22.3.5.4

Other Drugs

The following drugs have also been found to cause an increased risk of developing an adverse hepatic reaction in females up to 2.3-fold more than that found in males : isoniazid, nitrofurantoin, minocycline (Maddrey, 2005), ecstasy, rifampicin, aminosalicylate, dantrolene and clarithomycin (Smith et al., 2005). The mortality rate for isoniazid hepatitis was higher for black women than men or whites of either sex (Zimmerman, 1999). Females are also more susceptible than males to liver disease induced by halothane, chlorpromazine, methyldopa, diclofenac, tetracycline and ticrynafen (Zimmerman, 1999).

22.3.6

Gender-Based Alcohol Toxicity

Alcohol abuse causes more hepatic damage in women than in men. In general, women have higher blood alcohol levels after drinking which has been attributed to lower gastric ADH4 activity (Baraona et al., 2001). About 50 % lower ADH4 activity (Jelski et al., 2002) was reported. While the molecular basis for this is not known, women aged 20–40 years have 50 % less gastric alcohol dehydrogenase (Lieber, 2000). However, male gastric ADH declined with age so that females aged 41–60 years were higher than men of this age group (Parlesak et al., 2002).


22.4 22.4.1

553

Gender-Based Hepatotoxicity and Regulatory Toxicology Risk Assessment

To consider the role of gender-based hepatotoxicity in regulatory toxicology, it is essential to delineate the framework of risk-assessment/management which, in its basic form, is represented by the following flowchart: Toxicity assessment Hazard identification

Risk assessment and characterization

Risk management

Exposure assessment

In general, the toxicity assessment procedure employed by regulatory agencies in North America and Europe (EPA, 1992; McColl, 1990; Renwick et al., 2003; WHO, 1999) for substances other than carcinogens consisted of the following steps, with more or less stringency: (1) Review available human and epidemiological data on the substance. Review animal studies conducted on the substance and determine the no-observed-adverse-effect level (NOAEL; EPA, 1995), or benchmark dose. (2) Determine the safety factors1 to be applied. A 100-fold default factor is routinely applied, consisting of a 10-fold uncertainty factor (UF1) for extrapolation from animal to human and a 10-fold factor for interindividual variation (UF2). On top of these two factors, other factors1 may also be applied where appropriate. These include up to 10-fold factors for less than chronic studies, the use of lowest-observed-effect level (LOAEL) rather than NOAEL, and for various data inadequacies. (3) Calculation of exposure limit (or intake) limit Exposure limit = NOAEL/UF1 × UF2 ×. . . . , etc. For a given substance or mixture, the calculated exposure limit is then considered together with the result of the exposure assessment work to arrive at an integrated risk assessment/ characterization conclusion. It is at this stage that additional toxicity information, such as nature and severity of toxicity and data on vulnerable population, is considered to arrive at a more complete risk assessment. Risk management is the final phase of the process and its advisory/regulatory action is based largely on the conclusion of the risk assessment deliberation. 22.4.2

Gender-Based Hepatotoxicity in Regulatory Toxicology

The liver plays a major role in the metabolism, detoxification and bioactivation of xenobiotics. It is often the first organ to show adaptive or more severe adverse response to toxic insults. The adverse changes may involve neoplastic or nonneoplastic changes to the liver 1 It should be noted that an additional 10-fold safety factor has been applied to the setting of pesticides tolerance levels for infants and children (EPA, 2002). This additional factor is distinguished from the traditional safety factors in that it is introduced at the risk characterization stage when data on toxicity and exposure are integrated, and it addresses concerns on pre- and postnatal toxicity and uncertainties in the exposure assessment.

554

Hepatotoxicity

Table 22.1 Examples of derivation of NOAEL using gender-based hepatic endpoints Chemical Methylene chloride

Sensitive gender Route/duration Female

Bromoform Male

Hepatic effects

Inhalation/chronic Fatty change, multinucleated hepatocytes, cytoplasmic vacuolation Oral/intermediate Hepatocellular vacuolation

NOAEL 50 ppm

Principle study/ ATSDR report Nitschke et al., 1988/ATSDR, 2000

25 mg/kg NTP 1989/ ATSDR, 2005

that seriously compromised hepatic structure and functions and may be life-threatening. For these reasons, hepatic toxicity has always been considered an important toxicity endpoint in risk assessment (Hartung and Durkin, 1986; Pohl and Chou, 2005). Within the framework of risk assessment there are several entry points at which genderbased heptatotoxicity could be addressed. First, national and international guidelines for toxicity testings invariably stipulated that animals of both sexes be included in the study. In this way, gender difference in toxic expression becomes an integral part of toxicity assessment because the most sensitive endpoints, be it in the male or female sex, are used for the determination of NOAEL or benchmark dose, which are in turn used for the calculation of exposure limits. As an example, Table 22.1 describes the approach taken by the US Agency for Toxic Substances and Disease Registry to establish minimal risk levels for two chemicals based on gender-specific hepatic endpoints. A second entry point at which gender-based hepatotoxicty may be considered is the 10fold uncertainty factors for interindividual difference used for the derivation of exposure limits. This factor is originally applied to accommodate the uncertainty of interindividual differences in general, but has now been subdivided into toxicokinetic (3.16 or 100.5 ) and toxicodynamic (3.16 or 100.5 ) components (WHO, 1994, 1999). The toxicokinetic component takes into consideration the variation in bioavailability, distribution, bioactivation, biotransformation and elimination. It is known that hepatic phase I and phase II xenobiotic metabolizing activities are critical determinants in bioactivation and biotransformation. It is also known that many of the enzymes involved in xenobiotic bioactivation and biotransformation are polymorphically distributed in the human population. Meta-analysis has been conducted to examine several human phase I and phase II enzyme pathways in terms of the relationship between inter-individual variability and the toxicokinetic uncertainty factor (Dorne et al., 2005). A physiologically based pharmacokinetic model has been applied to evaluate the impact of cytochrome P456 2E1 variability on the pharmacokinetic component of uncertainty factor (Lipscomb et al., 2003). These studies focused on human variability and on susceptible groups, such as children, the elderly, patients with liver or kidney diseases and persons with inherited enzyme deficiencies, but not on gender. Sections 22.2 and 22.3 have reviewed evidence of association between gender-related polymorphism in metabolizing enzymes and heptatotoxicity in animals and humans. It is not clear at this point to what extent this gender-based variable influences the derivation of the inter-human uncertainty factor.


555

Finally, gender-based hepatotoxicty may also be emphasized at the risk assessment and characterization stage where vulnerable and susceptible segments of the population are highlighted. In general, the deliberation may focus on children, the aged, gender, pregnant women, health and nutrition status, and population, with particular genetic makeup. The wider issues of gender-related social and occupational issues can also be considered at this stage (WHO, 2005). Such deliberation is important because it clearly identifies the populations at risk, communicates it to regulatory agencies, industries and the public at large, and hence directly influences the risk management of the substance or mixture of interest. 22.4.3

Mechanisms of Action

Mechanism of action is an important part of toxicity assessment. It helps in understanding the underlying causes of an adverse effect and the dosimetry. It also provides a mechanistic basis for the toxicity assessment of chemicals that share similar structures or similar toxic expression. The etiology of chemical-induced, male rat-specific and hyaline-droplet nephrotoxicity is a good example of the influence of mechanistic studies on risk assessment. The liver of adult male rat secretes large amounts of α2μ -globulin into the circulation. This protein is taken up by the epithelial cells in the kidney tubules where it is digested by lysosomal enzymes. A variety of chemicals, including branched aliphatic alkanes which bind to α2μ -globulin and the resultant complex, were highly resistant to protease digestion in the lysosmes. As the undigested α2μ -globulin accumulated in the lysosome it formed aggregates, or hyaline droplets, which damaged cells and interfered with kidney functions. The loss of epithelial cells leads to hyperplasia, and chronic cell proliferation is likely linked to neoplastic changes in the kidney. Thus, a chemical effect on a male-specific hepatic protein ultimately resulted in a toxic expression in the kidneys (Swenberg et al., 1989). The large volume of animal and mechanistic data forms the basis for a prevailing opinion that chemical-induced, male ratspecific α2μ -globulin nephrotoxicity may not be relevant to human risk assessment (EPA, 1991; IARC, 1998). In comparison to hyaline-droplet nephrotoxicity, the establishment of mechanisms of gender-based hepatotoxicity can be even more complex because the toxic expression is dependent on individual chemicals or drugs, influenced by hormonal effects, variation in metabolism and disposition, and enzyme polymorphism. Added to the complexity is the paucity of human data and the uncertainty of animal-to-human extrapolation. To enable the risk assessment process to fully consider the gender-based hepatotoxicity issue, more work along the line described in Section 22.2.6 correlating gender-based differences in enzymes and xenobiotic metabolism with hepatotoxic endpoints is needed. These studies should be both toxicological and pharmacokinetic in nature and should focus not only on animals, but also humans and non-human primates. In this regard, the use of animal and human hepatocytes could yield useful results.

22.5

Conclusions

Although rat hepatocytes differ from human hepatocytes in their P450 composition and relative proportions, they could provide a valuable model for human hepatocytes for studying the toxicity and metabolism of drugs catalyzed by CYP1A1/2 or CYP2E1 but not drugs

556

Hepatotoxicity

catalyzed by the gender-specific P450s such as CYP2A, CYP2C or CYP3A2. A more valuable model for human hepatocytes would be female minipig hepatocytes which have a similar P450 isozyme composition to human hepatocytes. The much larger interindividual differences and polymorphisms among humans than inbred research animals have made it difficult to compare these activities in female versus male humans, as much larger population numbers are needed. The molecular basis for the gender differences in laboratory animals is now much clearer. This gender difference is particularly marked in rats. However, as humans have similar regulatory and transcription pathways it is likely that the gender differences in humans is larger than has been shown so far but has been overshadowed by the larger interindividual differences. In particular, human drug metabolism would be expected to be most gender-specific during pregnancy. This is because of a high continuous level of growth hormone secretion by the placenta into the maternal compartment to such a degree that the circulatory growth hormone levels in rats and humans are now similar. Furthermore, during pregnancy the liver has to metabolize large amounts of circulating steroid hormones which would require increased steroid metabolizing enzyme expression (Mode and Gustafsson, 2006).

References D.R. Abernethy, M. Divoll, D.J. Greenblatt and B. Ameer (1982). Obesity, sex, and acetaminophen disposition. Clin. Pharmacol. Ther., 31, 783–790. L. Airoldi et al. (2005). 4-Aminobiphenyl–hemoglobin adducts and risk of smoking-related disease in ‘never smokers’ and ‘former smokers’ in the European Prospective Investigation into Cancer and Nutrition prospective study. Cancer Epidemiol. Biomarkers Prev., 14, 2118–2124. S. Anakk, W. Huang, J.L Staudinger, K. Tan, T.J. Cole, D.D Moore and H.W. Strobel. (2007). Gender dictates the nuclear receptormediated regulation of CYP3A4. Drug Metab Dispos., 35, 36–42. P. Anzenbacher et al. (1998). Presence and activity of cytochrome P450 isoforms in minipig liver microsomes. Comparison with human liver samples. Drug Metab Dispos., 26, 56–59. ATSDR (2000). Toxicological Profile for Methylene Chloride. U.S Agency for Toxic Substances and Disease Registry. September 2000. Atlanta, Georgia 30333. ATSDR (2005). Toxicological Profile for Bromoform and Dibromochloromethane. U.S Agency for Toxic Substances and Disease Registry. August, 2005. Atlanta, Georgia 30333. J. Axelrod (1956). The enyzmatic N-demethylation of narcotic drugs. J. Pharmacol. Exp. Ther., 117, 322–330. E. Baraona et al. (2001). Gender differences in pharmacokinetics of alcohol. Alcohol Clin. Exp. Res., 25, 502–507. A. Bhathena, C. Lee and D.S. Riddick (2002). Suppression of cytochrome P450 2C11 by aromatic hydrocarbons: mechanistic insights from studies of the 5 -flanking region of the CYP2C11 gene. Drug Metab Dispos., 30, 1385–1392. S.L. Born, A.S. Fix, D. Caudill and L.D. Lehman-McKeeman (1999). Development of tolerance to Clara cell necrosis with repeat administration of coumarin. Toxicol. Sci., 51, 300–309. R. Bort, X. Ponsoda, R. Jover, M.J. Gomez–Lechon and J.V. Castell (1999). Diclofenac toxicity to hepatocytes: a role for drug metabolism in cell toxicity. J. Pharmacol. Exp. Ther., 288, 65–72. B. Boudikova, C. Szumlanski, B. Maidak and R. Weinshilboum (1990). Human liver catechol-Omethyltransferase pharmacogenetics. Clin. Pharmacol. Ther., 48, 381–389. D.B. Buckely and C.D. Klaassen (2007). Tissue- and gender-specific mRNA expression of UDPglucuronosyltransferases (UGTs) in mice. Drug Metab Dispos, 35, 121–127.


557

B. Burchell, D.J. Lockley, A. Staines, Y. Uesa and M.W. Coughtria (2005). Substrate specificity of human hepatic udp-glucuronosyltransferases. Meth. Enzymol., 400, 46–57. V.A. Catania et al. (1995). Gender-related differences in the amount and functional state of rat liver UDP-glucuronosyltransferase, Biochem. Pharmacol., 50, 509–514. N. Chauret, A. Gauthier, J. Martin and D.A. Nicoll-Griffith (1997). In vitro comparison of cytochrome P450-mediated metabolic activities in human, dog, cat and horse. Drug Metab Dispos., 25, 1130– 1136. B.L. Coffman, C.D. King, G.R. Rios and T.R. Tephly (1998). The glucuronidation of opioids, other xenobiotics and androgens by human UGT2B7Y(268) and UGT2B7H(268). Drug Metab Dispos., 26, 73–77. M.H. Court et al. (2001). Interindividual variability in acetaminophen glucuronidation by human liver microsomes: identification of relevant acetaminophen UDP-glucuronosyltransferase isoforms. J. Pharmacol. Exp. Ther., 299, 998–1006. M.H. Court et al. (2004). UDP-glucuronosyltransferase (UGT) 2B15 pharmacogenetics: UGT2B15 D85Y genotype and gender are major determinants of oxazepam glucuronidation by human liver. J. Pharmacol. Exp. Ther., 310, 656–665. C.L. Cummins, C.Y. Wu and L.Z. Benet (2002). Sex-related differences in the clearance of cytochrome P450 3A4 substrates may be caused by P-glycoprotein. Clin. Pharmacol. Ther., 72, 474– 489. J.R. DeBaun, E.C. Miller and J.A. Miller (1970). N-hydroxy-2-acetylaminofluorene sulfotransferase: its probable role in carcinogenesis and in protein-(methion-S-yl) binding in rat liver. Cancer Res., 30, 577–595. L.D. DeLeve and N. Kaplowitz (2003). Drug-Induced Liver Disease, Marcel-Dekker, New York, NY, USA. J.L. Dorne, K. Walton and A.G. Renwick (2005). Human variability in xenobiotic metabolism and pathway-related uncertainty factors for chemical risk assessment: a review. Food Chem. Toxicol., 43, 203–216. V.A. Eagling, J.F. Tjia and D.J. Back (1998). Differential selectivity of cytochrome P450 inhibitors against probe substrates in human and rat liver microsomes. Br. J. Clin. Pharmacol., 45, 107– 114. N. Ejiri, K. Katayama, N. Kiyosawa, Y. Baba and K. Doi (2005). Microarray analysis on Phase II drug metabolizing enzymes expression in pregnant rats after treatment with pregnenolone-16alphacarbonitrile or phenobarbital. Exp. Mol. Pathol., 79, 272–277. EPA (1991). Alpha 2μ-globulin: Association with chemically induced renal toxicity and neoplasia in male rat. Risk Assessment Forum, Risk Assesment Forum, US Environmental Protection Agency, EPA/625/3-91/019F. EPA (1992). IRIS. Background document (4/1/91). Cincinnati, OH: Office of Health and Environmental Assessment, Environmental Criteria and Assessment Office. EPA (1995). The use of the benchmark dose approach in health risk assessment. Risk Assessment Forum, US Environmental Protection Agency, EPA/630/R-94/007. EPA (2002). Determination of the appropriate FQPA safety factor(s) in tolerance assessment, Office of Pesticide Programs, US Environmental Protection Agency. G.A. Faich and R.H. Moseley (2001). Troglitazone (Rezulin) and hepatic injury. Pharmacoepidemiol. Drug Saf, 10, 537–547. B.M. Forman et al. (1998). Androstane metabolites bind to and deactivate the nuclear receptor CARbeta. Nature, 395, 612–615. M. Gandhi, F. Aweeka, R.M. Greenblatt and T.F. Blaschke (2004). Sex differences in pharmacokinetics and pharmacodynamics. Annu. Rev. Pharmacol. Toxicol., 44, 499–523. M. Gillberg, M.T. Skaanild and C. Friis (2006). Regulation of gender-dependent CYP2A expression in pigs: involvement of androgens and CAR. Basic Clin. Pharmacol. Toxicol., 98, 480–487.

558

Hepatotoxicity

H.L. Gurtoo and N.B. Parker (1977). Sex-dependent regulation of benzo[a]pyrene and zoxazolamine metabolism in rat tissues. Drug Metab Dispos., 5, 474–481. R.Z. Harris, L.Z. Benet and J.B. Schwartz (1995). Gender effects in pharmacokinetics and pharmacodynamics. Drugs, 50, 222–239. R. Hartung and P.R. Durkin (1986). Ranking the severity of toxic effects: Potential application to risk assessment. Comm. Toxicol., 1, 49–63. K. He et al. (2004). Metabolic activation of troglitazone: identification of a reactive metabolite and mechanisms involved. Drug Metab Dispos., 32, 639–646. H. Holck, M. Kanan, L. Mills and E. Smith (1937). Studies upon sex differences in rats in tolerance to certain barbiturates and the nicotine. J. Pharmacol. Exp. Ther., 60, 323–346. W. Honma et al. (2002). Phenol sulfotransferase, ST1A3, as the main enzyme catalyzing sulfation of troglitazone in human liver. Drug Metab Dispos., 30, 944–949. W.D. Hooper and M.S. Qing (1990). The influence of age and gender on the stereoselective metabolism and pharmacokinetics of mephobarbital in humans. Clin. Pharmacol. Ther., 48, 633–640. A.L. Hubener and D. Amelung (1953). Enzymatische umwandlungen von steroiden. Hoppe Seylers. Z. Physiol Chem., 293, 137–141. IARC (1998). In Species differences in thyroid, kidney and urinary bladder carcinogenesis, 147. M.M. Iba, J. Fung, P.E. Thomas and Y. Park (1999). Constitutive and induced expression by pyridine and beta-naphthoflavone of rat CYP1A is sexually dimorphic. Arch. Toxicol., 73, 208–216. G.C. Ibeanu and J.A. Goldstein (1995). Transcriptional regulation of human CYP2C genes: functional comparison of CYP2C9 and CYP2C18 promoter regions. Biochemistry, 34, 8028–8036. T. Igarashi, T. Satoh, K. Ueno and H. Kitagawa (1983). Sex-related difference in the hepatic glutathione level and related enzyme activities in rat. J. Biochem. (Tokyo), 93, 33–36. C.A. Jaffe, D.K. Turgeon, K. Lown, R. Demott-Friberg and P.B. Watkins (2002). Growth hormone secretion pattern is an independent regulator of growth hormone actions in humans. Am. J. Physiol Endocrinol. Metab, 283, E1008–E1015. W. Jelski, L. Chrostek, M. Szmitkowski and W. Laszewicz (2002). Activity of class I, II, III and IV alcohol dehydrogenase isoenzymes in human gastric mucosa. Dig. Dis. Sci., 47, 1554–1557. T.F. Kalhorn, C.A. Lee, J.T. Slattery and S.D. Nelson (1990). Effect of methylxanthines on acetaminophen hepatotoxicity in various induction states. J. Pharmacol. Exp. Ther., 252, 112–116. R. Kato and T. Kamataki (1982). Cytochrome P-450 as a determinant of sex difference of drug metabolism in the rat. Xenobiotica, 12, 787–800. R. Kato and Y. Yamazoe (1992). Sex-specific cytochrome P450 as a cause of sex- and species-related differences in drug toxicity. Toxicol. Lett., 64-65 (Spec No), 661–667. T. Kawamoto, S. Kakizaki, K. Yoshinari and M. Negishi (2000). Estrogen activation of the nuclear orphan receptor CAR (constitutive active receptor) in induction of the mouse Cyp2b10 gene. Mol. Endocrinol., 14, 1897–1905. M.H. Kester et al. (2003). Characterization of rat iodothyronine sulfotransferases. Am. J. Physiol Endocrinol. Metab, 285, E592–E598. V. Kobliakov, N. Popova and L. Rossi (1991). Regulation of the expression of the sex-specific isoforms of cytochrome P-450 in rat liver. Eur. J. Biochem., 195, 585–591. K. Laine, S. Palovaara, P. Tapanainen and P. Manninen (1999). Plasma tacrine concentrations are significantly increased by concomitant hormone replacement therapy. Clin. Pharmacol. Ther., 66, 602–608. C.S. Lieber (2000). Ethnic and gender differences in ethanol metabolism. Alcohol Clin. Exp. Res., 24, 417–418. J.C. Lipscomb et al. (2003). The impact of cytochrome P450 2E1-dependent metabolic variance on a risk-relevant pharmacokinetic outcome in humans. Risk Anal., 23, 1221–1238. S. Lofgren, A.L. Hogbjork, S. Ekman, R. Fransson-Steen and Y. Terelius (2004). Metabolism of human cytochrome P450 marker substrates in mouse: a strain and gender comparison. Xenobiotica, 34, 811–834.


559

J.I. Macdonald, R.J. Herman and R.K. Verbeeck (1990). Sex-difference and the effects of smoking and oral contraceptive steroids on the kinetics of diflunisal. Eur. J. Clin. Pharmacol., 38, 175–179. W.C. Maddrey (2005). Drug-induced hepatotoxicity: 2005. J. Clin. Gastroenterol., 39, S83–S89. J.M. Maglich et al. (2003). Identification of a novel human constitutive androstane receptor (CAR) agonist and its use in the identification of CAR target genes. J. Biol. Chem., 278, 17277–17283. R.S. McColl (1990). Biological safety factors in toxicological risk assessment, Health and Welfare Canada. Environmental Directorate, Cat. H49-49/1990E. B. Meibohm, I. Beierle and H. Derendorf (2002). How important are gender differences in pharmacokinetics? Clin. Pharmacokinet., 41, 329–342. J.O. Miners, J. Attwood and D.J. Birkett (1983). Influence of sex and oral contraceptive steroids on paracetamol metabolism. Br. J. Clin. Pharmacol., 16, 503–509. C. Mitoma, T.J. Sorich and S.E. Neubauer (1968). The effect of caffeine on drug metabolism. Life Sci., 7, 145–151. M. Miyazawa, M. Shindo and T. Shimada (2002). Sex differences in the metabolism of (+)- and (−)limonene enantiomers to carveol and perillyl alcohol derivatives by cytochrome p450 enzymes in rat liver microsomes. Chem. Res. Toxicol., 15, 15–20. A. Mode and J.A. Gustafsson (2006). Sex and the liver – a journey through five decades. Drug Metab Rev., 38, 197–207. G. Morel, B. Cossec, A.M. Lambert and S. Binet (1999). Evaluation of rat hepatic 2E1 activity in function of age, sex and inducers: choice of an experimental model capable of testing the hepatotoxicity of low molecular weight compounds. Toxicol. Lett., 106, 171–180. C.A. Mugford and G.L. Kedderis (1998). Sex-dependent metabolism of xenobiotics, Drug Metab Rev., 30, 441–498. G.J. Mulder (1986). Sex differences in drug conjugation and their consequences for drug toxicity. Sulfation, glucuronidation and glutathione conjugation. Chem. Biol. Interact., 57, 1–15. V. Nedelcheva, I. Gut, P. Soucek and E. Frantik (1998). Cytochrome P450 catalyzed oxidation of monochlorobenzene, 1,2- and 1,4-dichlorobenzene in rat, mouse and human liver microsomes. Chem. Biol. Interact., 115, 53–70. I. Negishi and Y. Aizawa (1975). Sex difference in the development of fatty liver by orotic acid. Jpn. J. Pharmacol., 25, 289–294. J. Nicholas and D. Barron (1932). The use of sodium amytal in the production of anesthesia in the rat. J. Pharmacol. Exp. Ther., 46, 125–129. K.D. Nitschke, J.D. Burek, T.J. Bell, R.J. Kociba, L.W. Rampy and M.J. McKenna (1988). Methylene Chloride: A 2-year inhalation toxicity and oncogenidity study in rats. Fundam. Appl. Toxicol. 11, 48–59. D.B. Njoku et al. (2002). Autoantibodies associated with volatile anesthetic hepatitis found in the sera of a large cohort of pediatric anesthesiologists. Anesth. Analg., 94, 243–249. NTP (1989). Toxicology and carcinogenesis studies of tribromomethane (bromoform) in F344/N rats and B6C3F1 mice (gavage studies). U.S. Department if Health and Human Services, National Toxicology Program, Technical Report Series No. 350. S. Ohhira, M. Enomoto and H. Matsui (2006). Sex difference in the principal cytochrome P-450 for tributyltin metabolism in rats. Toxicol. Appl. Pharmacol., 210, 32–38. J.Y. Park, K.A. Kim and S.L. Kim (2003). Chloramphenicol is a potent inhibitor of cytochrome P450 isoforms CYP2C19 and CYP3A4 in human liver microsomes. Antimicrob. Agents Chemother., 47, 3464–3469. A. Parkinson, D.R. Mudra, C. Johnson, A. Dwyer and K.M. Carroll (2004). The effects of gender, age, ethnicity, and liver cirrhosis on cytochrome P450 enzyme activity in human liver microsomes and inducibility in cultured human hepatocytes. Toxicol. Appl. Pharmacol., 199, 193–209. A. Parlesak, M.H. Billinger, C. Bode and J.C. Bode (2002). Gastric alcohol dehydrogenase activity in man: influence of gender, age, alcohol consumption and smoking in a caucasian population. Alcohol Alcohol, 37, 388–393.

560

Hepatotoxicity

R.V. Patwardhan, P.V. Desmond, R.F. Johnson and S. Schenker (1980). Impaired elimination of caffeine by oral contraceptive steroids. J. Lab Clin. Med., 95, 603–608. O. Pelkonen, A. Rautio, H. Raunio and M. Pasanen (2000). CYP2A6: a human coumarin 7hydroxylase. Toxicology, 144, 139–147. T.S. Poet, C.A. McQueen and J.R. Halpert (1996). Participation of cytochromes P4502B and P4503A in cocaine toxicity in rat hepatocytes. Drug Metab Dispos., 24, 74–80. H.R. Pohl and C.H. Chou (2005). Health effects classification and its role in the derivation of minimal risk levels: hepatic effects. Regul. Toxicol. Pharmacol., 42, 161–171. L.O. Prince and T.C. Campbell (1982). Effects of sex difference and dietary protein level on the binding of aflatoxin B1 to rat liver chromatin proteins in vivo. Cancer Res., 42, 5053–5059. G.S. Rao, G. Haueter, M.L. Rao and H. Breuer (1977). Steroid glucuronyltransferases of rat liver. Properties of oestrone and testosterone glucuronyltransferases and the effect of ovariectomy, castration and administration of steroids on the enzymes. Biochem. J., 162, 545–556. B.B. Rasmussen, T.H. Brix, K.O. Kyvik and K. Brosen (2002). The interindividual differences in the 3-demthylation of caffeine alias CYP1A2 is determined by both genetic and environmental factors. Pharmacogenetics, 12, 473–478. A.G. Renwick et al. (2003). Risk characterisation of chemicals in food and diet. Food Chem. Toxicol., 41, 1211–1271. M.A. Runge-Morris (1994). Sulfotransferase gene expression in rat hepatic and extrahepatic tissues. Chem. Biol. Interact., 92, 67–76. M. Runge-Morris and J. Wilusz (1991). Age and gender-related gene expression of hydroxysteroid sulfotransferase-a in rat liver. Biochem. Biophys. Res. Commun., 175, 1051–1056. C. Sato, N. Izumi, T. Nouchi, Y. Hasumura and J. Takeuchi (1985). Increased hepatotoxicity of acetaminophen by concomitant administration of caffeine in the rat. Toxicology, 34, 95–101. E.G. Schuetz, K.N. Furuya and J.D. Schuetz (1995). Interindividual variation in expression of Pglycoprotein in normal human liver and secondary hepatic neoplasms. J. Pharmacol. Exp. Ther., 275, 1011–1018. S. Shen, S.J. Hargus, B.M. Martin and L.R. Pohl (1997). Cytochrome P4502C11 is a target of diclofenac covalent binding in rats. Chem. Res. Toxicol., 10, 420–423. A. Shimamoto, E. Tanaka, D. Mizuno and S. Misawa (1999). Age- and sex-related changes in toluene metabolism by rat hepatic microsomes in vitro. Res. Commun. Mol. Pathol. Pharmacol., 104, 265– 276. M.T. Skaanild and C. Friis (1999). Cytochrome P450 sex differences in minipigs and conventional pigs. Pharmacol. Toxicol., 85, 174–180. I.D. Smith, K.J. Simpson, O.J. Garden and S.J. Wigmore (2005). Non-paracetamol drug-induced fulminant hepatic failure among adults in Scotland. Eur. J. Gastroenterol. Hepatol., 17, 161– 167. P.K. Srivastava and D.J. Waxman (1993). Sex-dependent expression and growth hormone regulation of class alpha and class mu glutathione S-transferase mRNAs in adult rat liver. Biochem. J., 294, 159–165. S.I. Strasser, S.A. Smid, M.L. Mashford and P.V. Desmond (1997). Sex hormones differentially regulate isoforms of UDP-glucuronosyltransferase. Pharm Res. 14, 1115–1121. J.A. Swenberg, B. Short, S. Borghoff, J. Strasser and M. Charbonneau (1989). The comparative pathobiology of alpha 2μ-globulin nephropathy. Toxicol. Appl. Pharmacol., 97, 35–46. S. Tafazoli and P.J. O’Brien (2004). Prooxidant activity and cytotoxic effects of indole-3-acetic acid derivative radicals. Chem. Res. Toxicol., 17, 1350–1355. T. Takeuchi et al. (2004). Gender difference in serum bisphenol A levels may be caused by liver UDP-glucuronosyltransferase activity in rats. Biochem. Biophys. Res. Commun., 325, 549–554. R.G. Tirona et al. (2003). The orphan nuclear receptor HNF4alpha determines PXR- and CARmediated xenobiotic induction of CYP3A4. Nat. Med., 9, 220–224.


561

Y. Verma and S.V. Rana (2003). Gender differences in the metabolism of benzene, toluene and trichloroethylene in rat with special reference to certain biochemical parameters. J. Environ. Biol., 24, 135–140. Y. Verma and S.V. Rana (2004). Sex differences in oxidative stress induced by benzene in rats. Ind. J. Exp. Biol., 42, 117–120. T. Visalli, R. Turkall and M.S Abdel-Rahman (2005). Influence of gender on cocaine hepatotoxicity in CF-1 mice. Int J Toxicol. 24, 43–50. T.J. Visser, G.A. van Haasteren, E. Linkels, E. Kaptein, H. van Toor and W.J. de Greef (1996). Genderspecific changes in thyroid hormone-glucuronidating enzymes in rat liver during short-term fasting and long-term food restriction. Eur. J. Endocrinol. 135, 489–497. T. Walle, U.K. Walle, T.D. Coward and E.C. Conradi (1989). Pathway-selective sex differences in the metabolic clearance of propranolol in human subjects. Clin. Pharmacol. Ther., 46, 257–263. A.G. Wang, T. Hia, J. Yuan, R.A. Yu, K.D. Yang, X.M. Chen, W. Qu and M.P. Walkes (2004). Effects of phenobarbital on metabolism and toxicity of diclofenac sodium in rat hepatocytes in vitro. Food Chem. Toxicol., 42, 1647–1653. C.C. Whitacre, S.C. Reingold and P.A. O’Looney (1999). A gender gap in autoimmunity. Science, 283, 1277–1278. WHO (1994). Assessing human health risks of chemicals: derivation of guidance values for healthbased exposure limits, World Health Organization. Envrionmental Criteria, 170. WHO (1999). Principles for the assessment of risks to human health from exposure to chemicals, World Health Organization, Envrionmental Criteria, 210. WHO (2005). Gender in lung cancer and smoking research. World Health Organization, Gender and Health Research Series, 170. L.M. Winer, M.A. Shaw and G. Baumann (1990). Basal plasma growth hormone levels in man: new evidence for rhythmicity of growth hormone secretion. J. Clin. Endocrinol. Metab, 70, 1678–1686. R. Wolbold et al. (2003). Sex is a major determinant of CYP3A4 expression in human liver. Hepatology, 38, 978–988. Y. Yoshigae, K. Konno, W. Takasaki and T. Ikeda (2000). Characterization of UDPglucuronosyltransferases (UGTS) involved in the metabolism of troglitazone in rats and humans. J. Toxicol. Sci., 25, 433–441. M. Yoshizuka, N. Mori, Y. Doi, A. Kawahara and S. Fujimoto (1992). Studies on the hepatotoxicity induced by bis (tributyltin) oxide. Arch. Toxicol., 66, 182–187. B.T. Zhu, L.A. Suchar, M.T. Huang and A.H. Conney (1996). Similarities and differences in the glucuronidation of estradiol and estrone by UDP-glucuronosyltransferase in liver microsomes from male and female rats. Biochem. Pharmacol., 51, 1195–1202. H.J. Zimmerman (1999). Hepatotoxicity: the Adverse Effects of Drugs and Other Chemicals on the Liver, Lippincott Williams & Wilkins, Philadelphia, PA, USA.

Section 7 Hepatotoxicity and Hepatocarcinogenicity

23 Hepatotoxicity in Oncology Drug Development† Wei Chen, Kenneth Hastings and John K. Leighton

23.1

Introduction

Cancer is a leading cause of death in the USA. There remains a demand for drugs to be available for patients to treat this heterogeneous disease and prevent recurrence. Diversity of cellular targets and biological molecules, along with increased pace of drug discovery and review for approval provide the promise for rapid development of novel therapeutics but may produce an incomplete evaluation of drug safety prior to full approval. Thus, development of drugs with expected efficacy and acceptable side effects is a central issue at all stages of drug discovery through to marketing. Liver as the main organ that metabolizes and transforms drugs attracts much attention. It is estimated that drug induced liver injury (DILI) constitutes about 10 % of all diseases detected by elevated liver enzymes and 50 % of acute hepatic failure (Lee, 2003). Antibiotics, non-steroidal anti-inflammatory drugs (NSAID) and neuropsychotropic medications are most cited drugs causing unacceptable hepatotoxicity in postmarketing evaluations. However, the mechanism of action and clinical use of most antineoplastic drugs developed over the past two decades requires a special focus on hepatotoxicity during patient monitoring. These drugs typically have a narrow therapeutic window, with dosing decisions often made based on dose-limiting toxicity. Although data from some non human species over predict hepatic toxicity, cytotoxic drugs by and large cause qualitatively similar toxicity in

† Disclaimer: The views expressed in this article are those of the authors and do not necessarily reflect the views of the U.S. Food and Drug Administration.


566

Hepatotoxicity

animals and in human (Owens, 1962a,b; Schein et al., 1970; Schein, 1977; Rozencweig et al., 1981; Olson et al., 2000). Oncology drugs have often demonstrated more predictable dose-dependent hepatotoxicity than other therapeutic areas and escalating doses can induce signs of liver toxicity in humans (King and Perry, 2001). It remains to be seen whether this pattern holds for the targeted therapeutics now entering clinical evaluation, where dosing may be to a biological endpoint rather than the maximum tolerated dose. The intended population of chemotherapeutic agents is an added concern in oncology drug development. Cancer patients are often frail, receiving on average of 4–6 drugs, and susceptible to renal and hepatic injury (Blower et al., 2005). Reduced metabolism and elimination of drugs in elderly patients may lead to a 3- to 10-fold increase in the incidence of adverse drug reactions compared to younger patients (Braverman, 1982). Even in cancer patients with normal liver function, altered hepatic clearance may result in increased risk of nonhepatic toxicity. It is not the intent of this article to exhaustively review the current state of hepatotoxicity evaluation, as many reviews have been written on this topic (Navarro and Senior, 2006). Rather, criteria currently used by the Division of Drug Oncology Products for assessing nonclinical drug-induced hepatotoxicity during IND review, the implications for this evaluation, and the potential application of new technology for hepatotoxicity evaluation will be discussed. Several examples of drug-related liver toxicity will be presented.

23.2

Mechanism of Hepatic Adverse Drug Reactions

Adverse drug reactions (ADR) are usually described as either dose-dependent or idiosyncratic. In fact, some authorities have proposed classification schemes based on this apparent dichotomy. For example, Edwards and Aronson (2000) proposed a system of ADR types that has been adopted by the World Health Organisation (WHO). Under most circumstances, however, the Rawlins and Thompson (1991) scheme is more useful since it is simple and seems to fit most ADR: Type A reactions are predictable and dose-related, Type B reactions are neither, and in most cases are assumed to have a metabolic or immunological basis. Dose-dependent (Type A) liver toxicities are often uncovered in early preclinical or clinical testing of drugs. In contrast, idiosyncratic responses (Type B) are likely due to chemically reactive parent drug and/or metabolites which may render the drug toxic to a specific organ or covalently bind to endogenous proteins to form immunogenic neoantigens which can induce drug-assocated immunopathology (Boelsterli, 2003). These reactions are typically described as not dose-dependent. Type B reactions do appear to occur more often if given in high doses chronically, which implies that there is at least a necessary threshold exposure (Park et al., 2000, 2005a,b). These reactions are of low or sporadic incidence, with uncertain temporal relationship to drug exposure, and in general are currently unpredictable from in vitro or animal studies (Park et al., 2005a). Underlying hepatic disease may also contribute to idiosyncratic injury, although this would not be obvious from nonclinical evaluation as toxicology of a molecule is typically investigated in healthy animals. Based on histology criteria, DILI involves either hepatocyte or cholestatic injuries, and mixed disorders (Grunhage et al., 2003). Alterations of hepatic morphology produced by chemotherapeutic agents are diverse and can include hepatocyte necrosis, fatty liver,

Hepatotoxicity in Oncology Drug Development

567

cholestasis and granuloma formation. The molecular mechanisms of Types A and B ADRs of anticancer agents-are likely to be similar to those of non-oncology drugs. Cytotoxicity of most chemotherapeutic drugs is relatively non-specific to tumor cells, and escalating dosages almost inevitably cause dose-dependent liver cell injury unless toxicity is evident in other vital organs first; therefore cytotoxic chemotherapeutic agents are considered to possess a dose-dependent intrinsic toxicity to liver cells. Transient elevations of liver enzymes in cancer patients are common manifestations during chemotherapy, and adjustment or modification of dosage is a rule to avoid DILI and other organ damage (Eklund et al., 2005; Floyd et al., 2006). The requirement of liver metabolism for the anti-tumor activity for many oncology drugs (e.g. cyclophosphamide and many antimetabolites) may also increase the likelihood of Type B ADRs.

23.3 23.3.1

Mechanisms: Case Studies Green Tea

In recent years, a variety of green tea extracts have been marketed in the United States as dietary supplements and green tea components, most notable epigallocatechin gallate (EGCG), are being investigated as anticancer drugs and for other indications as well (see clinicaltrials.gov). Despite the long history of safe consumption, individual compounds from green tea can potentially have toxic effects. Hepatotoxicity following use of several green tea extracts has been reported (Bonkovsky, 2006; Gloro et al., 2005). Sales of one green tea product were suspended by the French and Spanish Advisory Boards in 2003 and later withdrawn from the market completely by the manufacturer after 13 cases of hepatotoxicity were associated with its use (www.who.int/medicines/library/pnewslet/3news2003.pdf). Hepatic necrosis has been reported as a toxic response to EGCG and polyphenon E in mice (Chang et al., 2003; Goodin and Rosengren, 2003) and rats (Galati et al., 2006; Schmidt et al., 2005). No adverse effects were noted when up to 500 mg/kg/day EGCG was administered to pre-fed dogs for 13 weeks, but a similar or lower dose of EGCG caused morbidity when administered to fasted dogs (Isbrucker et al., 2006). All of these dogs had marked body weight loss and were generally anorexic over the three weeks preceding death. In addition to the other treatment-related toxicities such as proximal tubular necrosis in the kidneys, liver necrosis with elevation of serum AST, ALT and bilirubin were noted in male and female dogs treated with EGCG in the fasted state. A biodistribution study revealed that when the concentration of EGCG in tissues was determined 1 h after an intravenous administration (25 mg/kg) after 27 days of oral treatment with EGCG (250 mg/kg/day) to mimic chronic consumption of tea, EGCG was distributed to a variety of epithelial tissues with the highest concentrations in the liver and gastrointestinal tract (Swezey et al., 2003). Maximum plasma catechin concentration following oral administration of a single dose of polyphenon E was significantly different between healthy volunteers in a fed condition vs. fasting state (4–5 fold) (Chow et al., 2005). Although EGCG is thought to be a major contributor to the cytotoxic effect of green tea extracts in cultured hepatocytes and the levels required to elicit such effects were extremely high, the mechanism of heptotoxicity has not been established in vitro (Schmidt et al., 2005). As a

568

Hepatotoxicity

result of these nonclinical and clinical findings, additional monitoring of hepatic function is warrented and an updated informed consent provided to patients. 23.3.2

Oligonucleotides

Antisense oligodeoxynucleotides (AS ODN) offer the promise of therapeutic effect in cancer treatment by virtue of their high selectivity (Patil et al., 2005). The safety of these compounds has usually been assessed in rodents and non-human primates. Studies have indicated that several toxicities have been attributable to the chemical structure of some AS ODN, and not to the particular targeted mRNA sequence, and have included doselimiting toxicities related to hepatic toxicities (Jason et al., 2004; Winquist et al., 2006; Marshall et al., 2004). Nonclinical studies conducted to support the IND submissions have provided evidence of AS ODN-induced drug-related hepatic toxicities in chronic or subchronic non-clinical studies in both rodents and non-rodents (Farman and Kornbrust, 2003; Monteith et al., 1998). Dose-limiting toxicities of AS ODN have been related to hepatocellular degeneration leading to increased serum transaminases and decreased levels of alkaline phosphatase, albumin, total protein and cholesterol (Levin, 1999; Agrawal and Zhao, 1997). The underlying mechanism is not entirely clear, but specific liver uptake of AS ODN may contribute to this toxicity (Tavitian et al., 2002) and may involve selective concentration in Kupffer cells reslting in pro-inflammatory effects (Farman and Kornbrust, 2003). Due to the nature of hepatic and nonhepatic toxicities observed relative to therapeutic efficacy, investigations should explore next-generation antisense products with different structural variations in an attempt to overcome the limitations observed with these products. Appropriate monitoring of hepatic function should be included in nonclinical and clinical evaluations to assess possible class-effects of these products. 23.3.3

Liposome Formulation

Liposomes and drug/lipid complexes have been approved for the delivery of the anticancer drugs doxorubicin and daunorubicin and are subject to active investigation for many other cytoxic drugs. Pharmacokinetic studies indicate that a second dose of liposomes is rapidly cleared from the blood and accumulates in the liver when injected twice in the same rat or mouse at several-day intervals, a process termed accelerated blood clearance (ABC) phenomenon (Dams et al., 2000; Wang et al., 2005; Ishida et al., 2003a,b, 2004). The concern with ABC is the possible accumulation of toxic drugs in the liver, with subsequent hepatotoxicity. The accelerated clearance of subsequently injected liposomes is not solely induced by PEGylated liposomes (which have been structurally modified with polyethylene glycol), but can also be observed when conventional liposomes are used (Ishida et al., 2005; Wang et al., 2005). The mechanisms underlying the ABC phenomenon are not clear yet, but it has been reported that hepatosplenic macrophages play an essential role. In rats, the increased accumulation of liposomes in macrophages of the liver decreased their phagocytic capacity. Two weeks were necessary to restore phagocytic capacity. During this period, blood clearance of Klebsiella pneumoniae was significantly decreased (Daemen et al., 1995; Laverman et al., 2001). These findings suggested that accelerated blood clearance of liposomes not only compromise the therapeutic efficacy of liposomal drug formulations, but it could also cause undesirable pathological side-effects (such as immunosuppression). Although the alterations in pharmacokinetic behavior of subsequently injected liposomes


569

have not been confirmed in humans, ABC in animals, if hepatotoxicity is confirmed, may have a significant impact on the clinical evaluation of liposomal formulations that are administered repeatedly.

23.4

Current in Vitro Methodologies Used For Safety Assessment of Hepatotoxicity

Generation of reactive metabolites through cytochrome P450s can be rapidly evaluated through use of cell lines and hepatocyte preparations combined with appropriate analytical methods, and early characterization of these products is a major goal of the industry in order to reduce attrition (Ward, 2005). Approximately 90 % of drugs are metabolized through CYP450 (Watkins, 1990) so this evaluation is critical to understanding a drug’s potential for therapeutic success. These assays may also assist with evaluation of potential drug–drug interactions (Dambach et al., 2005). These studies, while often conducted early in development, are not currently required as part of the minimal recommendations to initiate an IND in oncology patients with end-stage disease. Formation of a reactive metabolite appears to be the first step necessary for inducing idiosyncratic drug reactions (Ju and Uetrecht, 2002). Besides cell lines, simple or complex cell culture system, liver slices, covalent binding assays and liver perfusion bioreactors are current in vitro methods of assessing hepatotoxicity (Ulrich et al., 1995; Vickers and Fisher, 2004; Farkas and Tannenbaum, 2005) although none of these methods have yet reached their full potential. The use of cytotoxicity assays to predict clinical hepatotoxicity is useful for direct or metabolism-mediated hepatic necrosis, and not immune-mediated or cholestatic injury. The rationale is based on data suggesting that about 50 % of all drug-related hepatic injuries involved direct hepatocyte necrosis (Zimmerman, 1999). Modeling cholestatic injury is more complex and requires a minimum of co-culture systems that contain several different cells (Zeilinger et al., 2002). In vitro modeling of immune-mediated hepatotoxicity is not practical at this time. Compounds with potential to activate cell death pathway or immune mediated hepatotoxicity can be screened by covalent binding assays, which apply radiochemical and immunological techniques to detect and quantify binding for a drug or its metabolite(s) to liver proteins (Coen et al., 2004; Boelsterli, 1993). Relevant to oncology drug development, alkylation of nucleic acids and proteins is required for cytotoxicity of alkylating chemotherapeutic agents. An understanding of metabolic pathways and potential interactions is important in nonclinical safety assessment (Park et al., 2000). Despite the new technologies and screening strategies for DILI, the study of hepatotoxicity using in vitro model systems is hindered by the difficulty of maintaining hepatocytes in culture because of the lack of detailed current knowledge of microenvironment surrounding these cells (Dambach et al., 2005; Farkas and Tannenbaum, 2005). The interpretation of data obtained is dependent to a great deal by the medium and matrix composition of the model system. Therefore, in vitro toxicity studies are suggestive and mainly useful when investigated drugs bear similar chemical structures to those known to have hepatotoxicity. In recent years, high-throughput molecular or cellular screening approaches such as toxicogenomics/proteomics assays have been applied at an early stage of drug development to screen for hepatotoxicity, and the challenge of these new technologies will be discussed later.

570

23.5

Hepatotoxicity

In vivo Studies and Regulatory Considerations

For small molecules, general toxicology studies are typically conducted in two laboratory animals (rodent and non-rodent). The dose-range used in toxicology studies should be chosen to allow a thorough qualitative and quantitative characterization of a drug’s toxicity. Most oncology drugs are administered to test species to dose levels that identify doselimiting toxicities. The dosing schedule in animal toxicity studies should reflect the clinical schedule. For oncology drugs, toxicology studies longer than 28 days are rarely needed to support the initial or continuing clinical trials. Reversibility of toxicity is evaluated by including animals in high dose and control groups which are left untreated for 2 to 4 additional weeks. The parameters used for detection of hepatic changes are clinical chemistry, morphologic pathology (histopathology, gross pathology and liver weight), and in-life findings including changes of body weight and food consumption. These parameters are evaluated at phases of the study that allow identification of acute, chronic, persistent, transient and/or reversible hepatic change. Increased serum levels of both aminotransferases and bilirubin in a clinical pathology study are considered significant for predicting meaningful hepatic damage or death, based on ‘Hy’s law’, a prognostic rule of drug-induced hepatotoxicity. Usually, a dose-dependent increase in serum transaminases of at least 100 % over control values is a trigger of concern. The hepatotoxicity must be assessed in context of onset, severity, duration and reversibility. If hepatotoxicity occurs in the general toxicology studies, specialized studies such as ultrastructural pathology, morphometrics, histologic staining, or methods for antibody detection may be necessary to characterize the details of the mechanism of toxicity, as well as the extent of liver injury. It must be emphasized that in the oncology setting, for patients with end-stage disease, investigating the potential mechanism of hepatoxicity is generally not considered a prerequisite to continued development of a drug in the context of the IND or NDA. However, such studies might be useful to avoid co-administration of combinations of drugs with similar mechanisms of toxicity. The potential DILI identified by nonclinical studies should be used to define the inclusion and exclusion criteria for patient recruitment, dose escalation scheme and liver function monitoring strategies in the trial design. For compounds with strikingly dissimilar metabolic profiles in humans and nonclinical animal species, consideration should be given to performing a more careful investigation of the role of metabolites in the toxicity profile of the drug candidate. Although liver is a common site of human organ toxicity caused by pharmaceuticals in development and on the market, hepatobiliary toxicity is among the most difficult to predict (Olson et al., 2000). Since interspecies differences in exposure, metabolic differences between animals and man, and pharmacological unresponsiveness in animal species could result in a false negative prediction, none of which were considered singly to account for the finding, lack of abnormality in animal toxicity studies does not necessarily translate into lack of future human toxicity. An additional explanation for the lack of nonclinical correlation to human DILI may be the lack of sensitive and reliable biochemical or histological markers in current in vivo studies using conventional animal species. The commonly used liver aminotransferases (AST and ALT) were considered relatively insensitive markers in animals. However, the correlation to human liver toxicity improved by incorporating histopathology data into the analysis.


571

In some cases, even with well designed clinical evaluation, DILI may be difficult to fully evaluate due to complicating factors. Many chemotherapeutic regimens require combination treatment which makes it more difficult to predict which drug in the regimen may be responsible for DILI. For example, vincristine itself is not hepatotoxic, but produces hepatotoxicity used in combination with radiation therapy (el Saghir and Hawkins, 1984). In light of the fact that most oncology drugs are used at close to the maximum tolerated dose, changes in patient status, such as the addition of a second drug or therapy, or presence of metastatic disease in the liver, may have profound effects on patient susceptibility to DILI and other toxicities. Special studies in these patients have been required. However, toxicology studies of drug or radiation combinations, or studies in animal models with hepatic impairment, while useful, have generally not been required as part of normal nonclinical drug development.

23.6

Promises and Limitations of New Technologies for Liver Toxicity Assessment

Assessment of drug safety in nonclinical studies, if these are to be useful in risk management, should identify a sensitive biomarker or biomarkers for tissue injury. For DILI, ideal biomarkers may be molecules from damaged cells, or combinatorial enzyme–drug adducts which could reflect or correlate with the mechanism of cell injuries or the histological locations of injury. Traditional biomarker development for toxicity assessment, however, has been hindered because of complex injury mechanisms of most DILI. Toxic metabolite formation, mitochondria enzyme inhibition with subsequent depletion of ATP molecules, liver cell death due to apoptosis or necrosis, and uncommonly autoimmune reaction may all participate in the liver injury process. Comparisons of toxicogenomic profiles among drugs tested on a ‘liver chip’ are useful in providing information about potential underlying mechanisms for liver cell toxicity (Hamadeh et al., 2002; Eubanks, 2005). It also opens the door for future predictive toxicology and mechanism-based risk assessment (Minami et al., 2005; Oberemm et al., 2005). It was estimated that 25 to 90 % of the variability in drug responses has a genetic basis. The application of pharmacogenomics may help identify key genes that exhibit polymorphic variability to drug responses. As stated in the FDA Guidance on pharmacogenomic data submission, only known valid biomarkers generated from pharmacogenomic studies would be used for regulatory decision making. These biomarkers may be used in the drug labeling in an informative manner, to choose a dose and dose schedule, to identify patients at risk and/or to identify patients likely to respond to therapy. These exploratory systems need to be validated and until such time, are primarily used in compound selection and not in regulatory decision making. Metabonomic/metabolomic analysis of metabolite profiles and pathways differentially regulated in liver with oxidative stress, enlarged liver weight, and hyperplasia have attracted attention in early drug discovery (Azmi et al., 2005; Baillie et al., 2002; Griffin and Bollard, 2004; Mortishire-Smith et al., 2004). The effects of polymorphisms of CYP450 system, immune system and individual genetics can be integrated and analyzed to distinguish viable from nonviable drug candidates. A sensitive, multiparameter biomarker for liver toxicity may well emerge from these new approaches of ‘Omics’ technologies that use systems biology combining both in vitro and in vivo data (Ekins et al., 2005).

572

Hepatotoxicity

At its experimental stage, high-throughput technology requires careful multilevel data mining, data reduction and interpretation through complex statistical multivariate models. Without solid hypotheses in drug development and toxicity assessment during the nonclinical stage, data interpretation generated through ‘Omics’ is not currently useful for regulatory purposes regarding drug safety (Leighton, 2005). Over the past several years, the FDA’s Critical Path Initiative (www.fda.gov/oc/initiatives/criticalpath/whitepaper.html), in line with rapid biomedical research achievement, has emphasized early detection of potential safety problems by ‘modernizing’ Critical Path tools through identifying predictive and confirmative safety biomarkers, including imaging biomarkers for organ-specific toxicity. The Hepatotoxicity Clinical Research Network established by the National Institute of Diabetes and Digestive and Kidney Diseases (NIDDK) will help to develop a prospective database on drug-related hepatotoxicity. The Liver Toxicology Biomarker Study (LTBS), as part of the FDA’s Critical Path Initiative, aims to discover signs of human liver toxicity in a standard test used in the initial stages of drug development. The formation of the Predictive Safety Testing Consortium is another path on the way to understand the safety of potential new drugs earlier. All efforts seek to modernize drug development by making the process more predictable and successful, and less costly.

23.7

Conclusions

Prediction of human dose-dependent hepatotoxicity during drug discovery and development requires an overall evaluation of information about the chemical structure, mechanism of action, cellular target and nonclinical and clinical experience gained in various phases of drug development. Potential for DILI is evaluated in each stage of the drug discovery and development process. The potential for hepatotoxicity is initially assessed in the discovery phase of drug development using in vitro methods and further tested in short-term animal toxicity screening studies. Despite such efforts, drug-induced hepatotoxicity is occasionally observed in IND-enabling studies but the rare, serious idiosyncratic ADRs are most often seen later in the course of clinical trials or after marketing approval. The fate of a chemical compound when liver toxicities are discovered in the preclinical stage depends on the context of such toxicity.

References Agrawal S and Zhao Q (1997). Toxicologic effects of an oligodeoxynucleotide phosphorothioate and its analogs following intravenous administration in rats. Antisense Nucleic Acid Drug Dev 7(6):575–584. Azmi J, Griffin JL, Shore RF, Holmes E and Nicholson JK (2005). Chemometric analysis of biofluids following toxicant induced hepatotoxicity: a metabonomic approach to distinguish the effects of 1-naphthylisothiocyanate from its products. Xenobiotica 35:839–852. Baillie TA, Cayen MN, Fouda H, Gerson RJ, Green JD, Grossman SJ, Klunk LJ, LeBlanc B, Perkins DG and Shipley LA (2002). Drug metabolites in safety testing. Toxicol Appl Pharmacol 182:188– 196. Blower P, de WR, Goodin S and Aapro M (2005). Drug–drug interactions in oncology: why are they important and can they be minimized? Crit Rev Oncol Hematol 55:117–142. Boelsterli UA (1993). Specific targets of covalent drug–protein interactions in hepatocytes and their toxicological significance in drug-induced liver injury. Drug Metab Rev 25:395–451.


573

Boelsterli UA (2003). Diclofenac-induced liver injury: a paradigm of idiosyncratic drug toxicity. Toxicol Appl Pharmacol 192:307–322. Bonkovsky HL (2006). Hepatotoxicity associated with supplements containing Chinese green tea (Camellia sinensis). Ann Intern Med 144:68–71. Braverman AM (1982). Therapeutic considerations in prescribing for elderly patients. Eur J Rheumatol Inflamm 5:289–293. Chang PY, Mirsalis J, Riccio ES, Bakke JP, Lee PS, Shimon J, Phillips S, Fairchild D, Hara Y and Crowell JA (2003). Genotoxicity and toxicity of the potential cancer-preventive agent polyphenon E. Environ Mol Mutagen 41:43–54. Chow HH, Hakim IA, Vining DR, Crowell JA, Ranger-Moore J, Chew WM, Celaya CA, Rodney SR, Hara Y and Alberts DS (2005). Effects of dosing condition on the oral bioavailability of green tea catechins after single-dose administration of Polyphenon E in healthy individuals. Clin Cancer Res 11:4627–4633. Coen M, Wilson ID, Nicholson JK, Tang H and Lindon JC (2004). Probing molecular dynamics in chromatographic systems using high-resolution 1H magic-angle-spinning NMR spectroscopy: interaction between p-xylene and C18-bonded silica. Anal Chem 76:3023–3028. Daemen T, Hofstede G, Ten Kate MT, Bakker-Woudenberg IA and Scherphof GL (1995). Liposomal doxorubicin-induced toxicity: depletion and impairment of phagocytic activity of liver macrophages. Int J Cancer 61:716–721. Dambach DM, Andrews BA and Moulin F (2005). New technologies and screening strategies for hepatotoxicity: use of in vitro models. Toxicol Pathol 33:17–26. Dams ET, Laverman P, Oyen WJ, Storm G, Scherphof GL, van der Meer JW, Corstens FH and Boerman OC (2000). Accelerated blood clearance and altered biodistribution of repeated injections of sterically stabilized liposomes. J Pharmacol Exp Ther 292:1071–1079. Edwards IR and Aronson JK (2000). Adverse drug reactions: definitions, diagnosis, and management. Lancet 365(9237):1255–1259. Ekins S, Nikolsky Y and Nikolskaya T (2005). Techniques: application of systems biology to absorption, distribution, metabolism, excretion and toxicity. Trends Pharmacol Sci 26:202–209. Eklund JW, Trifilio S and Mulcahy MF (2005). Chemotherapy dosing in the setting of liver dysfunction. Oncology (Williston Park) 19:1057–1063. el Saghir NS and Hawkins KA (1984). Hepatotoxicity following vincristine therapy. Cancer 54:2006– 2008. Eubanks M (2005). MIT toxicogenomics research program. Environ Health Perspect 113:A234– A235. Farkas D and Tannenbaum SR (2005). In vitro methods to study chemically induced hepatotoxicity: a literature review. Curr Drug Metab 6:111–125. Farman C and Kornbrust D (2003). Oligodeoxynucleotide studies in primates: antisense and immune stimulatory indications. Toxicol Pathol 31:(Suppl): 119–122. Floyd J, Mirza I, Sachs B and Perry MC (2006). Hepatotoxicity of chemotherapy. Semin Oncol 33:50–67. Galati G, Lin A, Sultan AM and O’Brien PJ (2006). Cellular and in vivo hepatotoxicity caused by green tea phenolic acids and catechins. Free Radic Biol Med 40:570–580. Gloro R, Hourmand-Ollivier I, Mosquet B, Mosquet L, Rousselot P, Salame E, Piquet MA and Dao T (2005). Fulminant hepatitis during self-medication with hydroalcoholic extract of green tea. Eur J Gastroenterol Hepatol 17:1135–1137. Goodin MG and Rosengren RJ (2003). Epigallocatechin gallate modulates CYP450 isoforms in the female Swiss–Webster mouse. Toxicol Sci 76:262–270. Griffin JL and Bollard ME (2004). Metabonomics: its potential as a tool in toxicology for safety assessment and data integration. Curr Drug Metab 5:389–398. Grunhage F, Fischer HP, Sauerbruch T and Reichel C (2003). Drug- and toxin-induced hepatotoxicity. Z Gastroenterol 41:565–578.

574

Hepatotoxicity

Hamadeh HK, Bushel PR, Jayadev S, Martin K, DiSorbo O, Sieber S, Bennett L, Tennant R, Stoll R, Barrett JC, Blanchard K, Paules RS and Afshari CA (2002). Gene expression analysis reveals chemical-specific profiles. Toxicol Sci 67:219–231. Isbrucker RA, Bausch J, Edwards JA and Wolz E (2006). Safety studies on epigallocatechin gallate (EGCG) preparations. Part 1: genotoxicity. Food Chem Toxicol 44:626–635. Ishida T, Maeda R, Ichihara M, Irimura K and Kiwada H (2003a). Accelerated clearance of PEGylated liposomes in rats after repeated injections. J Control Release 88:35–42. Ishida T, Masuda K, Ichikawa T, Ichihara M, Irimura K and Kiwada H (2003b). Accelerated clearance of a second injection of PEGylated liposomes in mice. Int J Pharm 255:167–174. Ishida T, Ichikawa T, Ichihara M, Sadzuka Y and Kiwada H (2004). Effect of the physicochemical properties of initially injected liposomes on the clearance of subsequently injected PEGylated liposomes in mice. J Control Release 95:403–412. Ishida T, Harada M, Wang XY, Ichihara M, Irimura K and Kiwada H (2005). Accelerated blood clearance of PEGylated liposomes following preceding liposome injection: effects of lipid dose and PEG surface-density and chain length of the first-dose liposomes. J Control Release 105:305– 317. Jason TL, Koropatnick J and Berg RW (2004). Toxicology of antisense therapeutics. Toxicol Appl Pharmacol 201:66–83. Ju C and Uetrecht JP (2002). Mechanism of idiosyncratic drug reactions: reactive metabolite formation, protein binding and the regulation of the immune system. Curr Drug Metab 3:367–377. King PD and Perry MC (2001). Hepatotoxicity of chemotherapy. Oncologist 6:162–176. Laverman P, Carstens MG, Boerman OC, Dams ET, Oyen WJ, van RN, Corstens FH and Storm G (2001). Factors affecting the accelerated blood clearance of polyethylene glycol–liposomes upon repeated injection. J Pharmacol Exp Ther 298:607–612. Lee WM. (2003). Drug-induced hepatotoxicity. N Engl J Med 349:474–485. Leighton JK. (2005). Application of emerging technologies in toxicology and safety assessment: regulatory perspectives. Int J Toxicol 24:153–155. Levin AA. (1999). A review of the issues in the pharmacokinetics and toxicology of phosphorothioate antisense oligonucleotides. Biochim Biophys Acta 1489:69–84. Marshall J, Chen H, Yang D, Figueira M, Bouker KB, Ling Y, Lippman M, Frankel SR and Hayes DF (2004). A phase I trial of a Bcl-2 antisense (G3139) and weekly docetaxel in patients with advanced breast cancer and other solid tumors. Ann Oncol 15:1274–1283. Minami K, Saito T, Narahara M, Tomita H, Kato H, Sugiyama H, Katoh M, Nakajima M and Yokoi T (2005). Relationship between hepatic gene expression profiles and hepatotoxicity in five typical hepatotoxicant-administered rats. Toxicol Sci 87:296–305. Monteith DK, Geary RS, Leeds JM, Johnston J, Monia BP and Levin AA (1998). Nonclinical evaluation of the effects of a novel antisense compound targeting C-raf kinase in mice and monkeys. Toxicol Sci 46:365–375. Mortishire-Smith RJ, Skiles GL, Lawrence JW, Spence S, Nicholls AW, Johnson BA and Nicholson JK (2004). Use of metabonomics to identify impaired fatty acid metabolism as the mechanism of a drug-induced toxicity. Chem. Res Toxicol 17:165–173. Navarro V and Senior J (2006). Drug-related hepatotoxicity. N Engl J Med 354:731–739. Oberemm A, Onyon L and Gundert-Remy U (2005). How can toxicogenomics inform risk assessment? Toxicol Appl Pharmacol 207:592–598. Olson H, Betton G, Robinson D, Thomas K, Monro A, Kolaja G, Lilly P, Sanders J, Sipes G, Bracken W, Dorato M, Van DK, Smith P, Berger B and Heller A (2000). Concordance of the toxicity of pharmaceuticals in humans and in animals. Regul Toxicol Pharmacol 32:56–67. Owens AH, Jr (1962a). Predicting anticancer drug effects in man from laboratory animal studies. J Chronic Dis 15:223–228.


575

Owens AH, Jr (1962b). Predicting effects of anticancer drugs in man, from laboratory animal studies. Cancer Chemother Rep 16:57–60. Park BK, Kitteringham NR, Powell H and Pirmohamed M (2000). Advances in molecular toxicology – towards understanding idiosyncratic drug toxicity. Toxicology 153:39–60. Park BK, Kitteringham NR, Maggs JL, Pirmohamed M and Williams DP (2005a). The role of metabolic activation in drug-induced hepatotoxicity. Annu Rev Pharmacol Toxicol 45:177–202. Park K, Williams DP, Naisbitt DJ, Kitteringham NR and Pirmohamed M (2005b). Investigation of toxic metabolites during drug development. Toxicol Appl Pharmacol 207:425–434. Patil SD, Rhodes DG and Burgess DJ (2005). DNA-based therapeutics and DNA delivery systems: a comprehensive review. AAPS J 7:E61–E77. Rawlins M and Thompson W (1991). Mechanisms of adverse drug reactions. In: Davies D, (ed). Textbook of Adverse Drug Reactions. Oxford University Press, New York, 18–45. Rozencweig M, Von Hoff DD, Staquet MJ, Schein PS, Penta JS, Goldin A, Muggia FM, Freireich EJ and DeVita VT, Jr (1981). Animal toxicology for early clinical trials with anticancer agents. Cancer Clin Trials 4:21–28. Schein PS (1977). Nonclinical toxicology of anticancer agents. Cancer Res 37:1934–1937. Schein PS, Davis RD, Carter S, Newman J, Schein DR and Rall DP (1970). The evaluation of anticancer drugs in dogs and monkeys for the prediction of qualitative toxicities in man. Clin Pharmacol Ther 11:3–40. Schmidt M, Schmitz HJ, Baumgart A, Guedon D, Netsch MI, Kreuter MH, Schmidlin CB and Schrenk D (2005). Toxicity of green tea extracts and their constituents in rat hepatocytes in primary culture. Food Chem Toxicol 43:307–314. Swezey RR, Aldridge DE, LeValley SE, Crowell JA, Hara Y and Green CE (2003). Absorption, tissue distribution and elimination of 4-[(3)h]-epigallocatechin gallate in beagle dogs. Int J Toxicol 22:187–193. Tavitian B, Marzabal S, Boutet V, Kuhnast B, Terrazzino S, Moynier M, Dolle F, Deverre JR and Thierry AR (2002). Characterization of a synthetic anionic vector for oligonucleotide delivery using in vivo whole body dynamic imaging. Pharm Res 19:367–376. Ulrich RG, Bacon JA, Cramer CT, Peng GW, Petrella DK, Stryd RP and Sun EL (1995). Cultured hepatocytes as investigational models for hepatic toxicity: practical applications in drug discovery and development. Toxicol Lett 82-83:107–115. Vickers AE and Fisher RL (2004). Organ slices for the evaluation of human drug toxicity. Chem Biol Interact 150:87–96. Wang XY, Ishida T, Ichihara M and Kiwada H (2005). Influence of the physicochemical properties of liposomes on the accelerated blood clearance phenomenon in rats. J Control Release 104:91–102. Ward KW (2005). Recent advances in pharmacokinetic extrapolation from preclinical data to humans. Expert Opin Drug Metab Toxicol 1:583–594. Watkins PB (1990). Role of cytochromes P450 in drug metabolism and hepatotoxicity. Semin Liver Dis 10:235–250. Winquist E, Knox J, Ayoub JP, Wood L, Wainman N, Reid GK, Pearce L, Shah A and Eisenhauer E (2006). Phase II trial of DNA methyltransferase 1 inhibition with the antisense oligonucleotide MG 98 in patients with metastatic renal carcinoma: a National Cancer Institute of Canada Clinical Trials Group investigational new drug study. Invest New Drugs 24:159–167. Zeilinger K, Sauer IM, Pless G, Strobel C, Rudzitis J, Wang A, Nussler AK, Grebe A, Mao L, Auth SH, Unger J, Neuhaus P and Gerlach JC (2002). Three-dimensional co-culture of primary human liver cells in bioreactors for in vitro drug studies: effects of the initial cell quality on the long-term maintenance of hepatocyte-specific functions. Altern Lab Anim 30:525–538. Zimmerman H (1999). Hepatotoxicity Lippincott Williams & Wilkins, 2nd Edition, Philadelphia, PA, USA.

24 The Potent Rat Hepatocarcinogen Methapyrilene: An Hypothesis Regarding its Hepatotoxicity Daniel A. Casciano

24.1

Introduction

Methapyrilene hydrochloride (MP), Figure 24.1, is a histamine H1 -receptor antagonist that was the active ingredient in many over-the-counter allergy, cold medications and sleeping aids. In the late 1970s, MP was identified as a potent liver carcinogen in Fischer 344 rats (Lijinsky et al., 1980) and consequently was removed from the market by the Food and Drug Administration. At relatively high concentrations (1000 ppm) in food, the chemical induced almost 100 % incidence of hepatocellular carcinomas and cholangiocarcinomas. This finding was a surprise to the authors because the data were derived from one of their control populations. Lijinsky was interested in the formation of nitrosamines in the gut upon administration of nitrite and amine drugs such as MP and so one of his controls was animals exposed to the antihistamine alone. Follow-up in vivo ultrastructural studies by this laboratory suggested that MP induced a significant increase in mitochondria of periportal hepatocytes, as well as a pronounced conformation change of these organelles, while the lipid, glycogen and smooth endoplasmic reticulum was greatly reduced (Reznik-Schuller and Lijinsky, 1981; Reznik-Schuller and Reuber, 1986), an observation that seemed unique to this hepatocarcinogen. They concluded from their findings that MP interacts with cell constituents other than DNA and postulated that this chemical-induced cancer occurs via an indirect, epigenetic mechanism. At the time, this interpretation intrigued the scientific community and many felt that understanding the mechanism(s) by which MP induced


578

Hepatotoxicity

Figure 24.1

Structure of methapyrilene, a tertiary amine H1 -receptor antagonist

hepatic tumors would lead to a universal understanding of the epigenetic mechanism(s) of cancer.

24.2

Follow-Up Carcinogenicity Studies

Couri et al. (1982) measured the effect of MP on initiation and promotion of λ-glutamyltranspeptidase-positive foci and demonstrated its potential to enhance enzyme-altered foci formation following initiation by nitrosodiethylamine. They also showed that MP did not act as an initiator in this system. Furuya et al. (1983) showed that MP alone could not induce neoplasms but acted as a promoter when the liver was initiated by a genotoxic carcinogen. However, in contrast to Couri et al. (1982), they found that MP alone produced a significant incidence of enzyme-altered foci. In a later study, they proposed that MP acted synergistically with 2-acetylaminoflorene (2-AAF) when MP was given before the 2-AAF (Furuya and Williams, 1984). Glauert and Pitot (1989) studied the ability of MP to initiate hepatocarcinogenesis in female F344 rats and their data indicate that it may act as a weak initiator. Hernandez et al. (1989) reported a novel result indicating that MP, using doses that induced tumors, produced an increase in DNA methylation. This observation, although novel, conflicts with other findings that indicate hypomethylation and not hypermethylation constitutes an essential step in carcinogenesis. Graichen et al. (1984) evaluated the short-term treatment of F344 rats with MP and found that P-450 dependent mixed-function oxidase activities were decreased while epoxide hydrolase, DT-diaphorase and liver serum enzymes, indicative of a hepatotoxic response, were elevated. Perera et al. (1985) found that short-term feeding of MP resulted in lipid peroxidation in both mitochondrial and microsomal membranes and suggested that membrane lipid peroxidation may play a role in promoting/initiating the carcinogenic effect of MP. Steinmetz et al. (1988) investigated the potential of MP to induce DNA damage in vivo using the in vivo/in vitro UDS methodology. They found that MP was inactive; however, they observed a dose-dependent increase in hepatic DNA S-phase synthesis in rats and mice. Cunningham et al. (1995) demonstrated that MP induced sustained hepatocellular replication and mitochondrial protein alterations in F344 rats in a 13-week feeding study, suggesting that inappropriate hepatocellular proliferation may contribute to the mechanism of carcinogenesis of MP. Additionally, two studies demonstrated that MP did not induce DNA adducts in F344 rat livers as measured by 32 Ppostlabeling techniques (Casciano et al., 1988; Lijinsky and Yamashita, 1988), suggesting a nongenotoxic mechanism of carcinogenesis.

The Potent Rat Hepatocarcinogen Methapyrilene

24.3

579

Genotoxicity Studies

Methapyrilene has been evaluated in a number of short-term in vitro genotoxicity assays that detect primary DNA damage, single gene mutations, chromosomal mutations and also cellular transformation resulting in conflicting reports. Probst and Neal (1980), measuring primary DNA damage as unscheduled DNA synthesis (UDS) in cultured primary rat hepatocytes using autoradiographic analysis, demonstrated that MP was inactive in this system. Budroe et al. (1984), using similar analysis confirmed their results. Althaus et al. (1982), however, showed that MP-induced DNA damage in rat hepatocytes could be detected when UDS was quantified using scintillometric analysis. These results are considered questionable because of the method of analysis. Iype et al. (1982) evaluated primary DNA damage by assessing the ability of MP to induce sister chromatid exchange (SCE) in several established cell lines derived from various organs. Methapyrilene, even at the maximum tolerated dose, did not induce SCE in Chinese hamster ovary (CHO) cells or hamster lung fibroblasts in the presence or absence of exogenous metabolic activation (S9) systems. However, other data suggest that under certain conditions, MP may induce SCE in CHO (National Toxicology Program, unpublished data). Andrews et al. (1980) and Probst et al. (1981) observed no mutagenic activity due to MP in Salmonella typhymurium. Kammerer et al. (1986) tested MP and some of its metabolites in the same system and reported that only the nitrosated derivatives were mutagenic. We also tested MP and some of it metabolites in several of the Ames tester strains and determined that these compounds were inactive under the conditions of assay (unpublished data). However, Ashby et al. (1988) found that MP may be a very weak mutagen in strain 1535, although the interpretation is open to question. Reports of the ability of MP to induce mutations in somatic cells are also conflicting. Casciano and Schol (1984) reported that MP failed to induce mutations at the hypoxanthineguanine phosphoribosyl transferase (HPRT) X-linked locus in CHO co-cultured with intact rat hepatocytes or in the presence or absence of S9. Oberly et al. (1984) reported a negative response in the L5178Y mouse lymphoma mutagenesis assay that measures mutations in the thymidine kinase (TK) autosomal gene. These negative results were observed in the presence and absence of S9. Blazak et al. (1986) demonstrated a significant increase in mutation frequency in L5178Y cells following treatment with MP in the presence of S9 and found that both large and small colonies were recovered, the larger fraction of which were small colonies suggesting that chromosomal rather than single gene mutants were predominantly recovered. In addition, Turner et al. (1987) observed reproducible positive responses in L5178Y cells and suggested that MP’s activity and potency was a function of the freshness of the S9 preparations. They found that lower induced mutant frequencies were observed when fresh S9 was used versus when aged S9 was used. They also reported that the majority of the recovered mutant colonies were small, again suggesting that the derived mutants were of chromosomal origin rather than single gene or point mutations. This observation was the basis of a study by Casciano et al. (1991) initiated to determine if DNA adducts were responsible for the MP-induced small colonies. Although MP-induced mutations were again derived, no adducts were detected in the DMSO controls or in the MPexposed L5178Y experimental samples. Our data clearly indicated that the MP-induced mutations at the TK locus resulted in the absence of direct DNA damage and affirming the large body of literature suggesting that MP is a complete carcinogen that does not act through a direct genotoxic mechanism, i.e. electrophilic reaction with the DNA.

580

Hepatotoxicity

Although this potent hepatocarcinogen has not been classified as a genotoxin when evaluated in conventional short term assays (Mirsalis, 1987), it has been considered by some as acting via an indirect secondary genotoxic mechanism. This interpretation is mainly driven by the evidence presented above which indicates that, under certain conditions of assay, MP apparently induces mutations in some well-validated in vitro systems (Blazak et al., 1986; Turner et al., 1987; Ashby et al., 1988; Casciano et al., 1991). Indirect mechanisms, such as intercalation or oxidative damage that can induce homologous recombination or gene conversion or alterations of enzymes or structural chromosomal proteins involved in maintaining the dynamic structural and functional integrity of DNA (e.g. topisomerase I and II), have been considered. Many have thought that understanding the mechanism of action of methapyrilene and its mechanism of action of these events would provide valuable insights into those classes of carcinogens operating through an indirect secondary genotoxic mechanism.

24.4 24.4.1

Autophagy and Endocytosis Autophagy

Autophagy is the mechanism by which cellular cytoplasm is degraded and turned over, serving to restrict cell size and cell growth, and to supply protein-derived amino acids when needed by the organism (Seglen, 1997). This activity is regulated by growth conditions, growth factors, hormones and metabolites, particularly amino acids that exert a negative feedback control (Seglen and Bohley, 1992). Under certain conditions, autophagy accounts for a major part of intracellular protein degradation, as well as a mechanism to turn over cellular organelles, i.e. endoplasmic reticulum, mitochondria, etc. Autophagy is initiated by sequestering organelles known as a phagophore (Seglen, 1987) which in liver hepatocytes typically consist of multiple flattened cisternae that may be derived from endoplasmic reticulum. The phagophore engulfs a part of the cytoplasm to form a closed vacuole which is called an ‘autophagosome’. The autophagosome is generally recognized as the first well-defined step in the autophagic/lysosomal pathway of intracellular protein degradation. These autophagosomes can fuse with small lysosomes to form autolysosomes (Seglen and Solheim, 1985). Within these vesicles, the cytoplasm, containing protein and organelles, is degraded to monomers available for synthesis of new proteins and membranous vesicles. Kovacs et al. (1982) initiated a study to determine the effect of a number of inhibitors on the accumulation of autophagosomes in cultured primary rat hepatocytes. They showed that endogenous protein degradation was strongly inhibited by an amino acid mixture as well as by the inhibitors propylamine (a lysomotropic amine), leupeptin (a protease inhibitor) and vinblastine (a microtubule poison). Amino acids, leupeptin and vinblastine all produced an approximate 65 % inhibition, whereas propylamine inhibited 75 %. They concluded that these treatments represented an essentially complete inhibition of the autophagic/lysosomal pathway of protein degradation. Each of the inhibitors of degradation induced characteristic ultrastructural changes in isolated hepatocytes. Treatment resulted in an accumulation of autophagosome-like vesicles that resembled electron-dense lysosomes containing partially digested cytoplasmic


581

constituents, lysosomal vacuolization containing electron-dense lysosomes with an accumulation of secretory vesicles containing lipoprotein granules and a marked dilation of the smooth endoplasmic reticulum. It was suggested that these degradation inhibitors produce lysosomes with reduced fusion capacity, perhaps by dissimilar mechanisms. Vinblastine impairs microtubular function, leupeptin inhibits lysosomal proteases, propylamine, because of its pK , causes an increase in lysosomal pH by ‘tying-up’ lysosomal protons. This latter mechanism will be discussed further in the section on methapyrilene and accumulation of autophagosome-like vesicles. 24.4.2

Endocytosis

Many macromolecules gain entry into the cell via a process of receptor-mediated endocytosis (Sorkin and van Zastrow, 2002). Ligand–receptor complexes formed on the cell surface can be selectively recruited into clathrin-coated pits (small indentations of the plasma membrane that can invaginate inward and pinch off vesicles into the cytoplasm). This action resembles what was known as ‘pinocytosis’. Endocytic vesicles that are formed by invagination and the pinching of clathrin-coated pits become uncoated in the cytoplasm and fuse with specialized membrane organelles, called ‘endosomes’. These endosomes contain the ligand complexes and serve as the vessel by which the various receptor-mediated complexes are ferried to their various intracellular locations. Iron, an essential nutrient for the hepatocyte, is introduced into the cell via the transferin receptor located on the surface of the cell. Iron complexes with its receptor and the membrane/ligand complex pinches off through an endocytic/pinocytotic mechanism forming an enclosed vesicle within the cytoplasm. Like many such complexes, the iron is released from the receptor by means of a decrease in the pH of the endosome and thus becomes available to accomplish its function, i.e. nutrition, enzyme co-factor, etc. The endosome containing a functional receptor then traverses to the membrane and becomes available once again to interact with its ligand through a process resembling reverse pinocytosis. Several inhibitors can alter the receptor-mediated endocytosis process, including cellular ion perturbation, manipulations that interfere with the assembly of a functional coated pit and chlorpromazine, as well as chemicals that alter the intracellular pH of the functional endosome.

24.5

Methapyrilene and Autophagy

During the late 1980s and early 1990s, MP became a favorite chemical to study because it induced liver cancer in 100 % of Fischer 344 rats and it was ‘an over-the-counter’ medication to which many humans had been exposed. Many studies, including ours, suggested that this antihistamine apparently was inducing cancer through a nongenotoxic or epigenetic mechanism. Investigators, including our laboratory, were convinced that this chemical would serve as a model by which nongenotoxic mechanisms may be better understood. Even more intriguing was the availability of close structural analogs that were either mildly hepatotoxic or non-toxic, providing a model whereby the biological activity could be related to the structure of the molecule. Additionally, MP appeared to be hepatocarcinogenic only in the Fischer 344 rat and was less potent in other rat strains and non-hepatotoxic in the

582

Hepatotoxicity

mouse. These findings allow determination of species differences and perhaps identification of pathways associated with MP susceptibility. In our studies to detect MP-induced DNA damage in primary hepatocytes, we noticed that this chemical induced a peculiar morphological alteration within living cells observed using phase contrast microscopy (Figure 24.2). In the unexposed cultures, the cells appeared very healthy and did not contain any abnormal morphology (Figure 24.2(a)). However, when rat hepatocytes were exposed to 10 μM MP for approximately 24 hours, the living cells displayed an unusual, never before observed, vacuolization (Figures 24.2(b) and 24.2(c)). Although the cells retained this unusual morphology for at least another 24 h in culture, the cells were still viable. Initial interpretation, from the phase-contrast microscopy, was that MP induces an accumulation of vesicles filled with lipid or lipid-like substances, perhaps indicating that we discovered an in vitro model for studying toxicity as a function of fatty livers. However, when these cell cultures were examined ultrastructurally using transmissionelectron microscopy, a completely different interpretation emerged. It was apparent that one result of cellular exposure to MP is the induction of autophagy, leading to the accumulation of autophagosomes. Figure 24.3 shows the results of ultrastructural examination of control populations and MP exposed hepatocytes in primary culture. Figures 24.3(a) and 24.3(b) depict transmission-electron micrographs of control cells. Figure 24.3(a) shows a rat hepatocyte 24 h post-perfusion of the intact liver. The view is a slice from the top of the cell that didn’t include the nucleus. Notice the intact plasma membrane and healthy mitochondria and prominent rough and smooth endoplasmic reticulum. Also obvious in the micrograph are intact ribosomes, indicating a cell very active in protein synthesis. In addition, several lipid-like inclusions are evident. Figure 24.3(b) is an electron micrograph of another independent control cell. This particular cell was ultramicrotomed to an inner layer of the cell, indicating a view of the intact viable cell with a healthy and prominent nucleus and mitochondria. Abundant endoplasmic reticulum and ribosomes indicate that the cell is metabolically active. Evident also in this micrograph are bile canaliculi and desmosomes, suggesting active intercellular communication through ‘tight’ junctions. Figure 24.3(c) shows low magnification of a single intact viable cell that is filled with autophagosomes or non-functional lysosomes. One can see that these vesicles contain mitochondria, endoplasmic reticulum and other membranous organelles and probably soluble macromolecules that are indigenous to the cytoplasm (cf. figures in Kovacs et al., 1982). Figures 24.3(d) and 24.3(e) display increasingly higher magnifications of a viable MPexposed cell and its vesicular contents. The membranous content of the vesicles are easily identified as intracellular organelles that are in the process of recycling and turning over, a normal physiological mechanism the cell uses to renew macromolecules and functional organelles. It appears that MP alters the completion of this normal process by inhibiting lysis of macromolecules to monomers and interrupting the traverse of the lytic vesicles to the plasma membrane. Therefore, the cell cannot turn over its contents by the reverse pinocytotic pathway, resulting in the accumulation of these vesicles. How does MP interfere with this essential function of the hepatocyte? We suggest that MP or one of its metabolites alters lysosomal function by disrupting intralysosomal pH. Methapyrilene is a tertiary amine and has the ability to interfere with the lysosomal acidic microenvironment by sequestering lysosomal protons, resulting in an increased pH. This

The Potent Rat Hepatocarcinogen Methapyrilene (a)

583

(b)

(c)

Figure 24.2 (a) Phase-contrast micrograph of a control population of primary cultured rat hepatocytes. The living cells were observed microsopically 24 h post-isolation. (b) Phasecontrast micrograph of primary cultured rat hepatocytes exposed to 10 μM methapyrilene for 24 h. Cells are located in a ‘less-densely’ populated part of the Petri plate (notice the prominent holes in these living cells). (c) Phase-contrast micrograph of primary cultured rat hepatocytes exposed to 10 μM methapyrilene for 24 h. Cells are located in a ‘more-densely’ populated part of the Petri plate (notice the many prominent holes in most of the living healthy cells)

584

Hepatotoxicity

(a)

(b)

(d)

(e)

(c)

Figure 24.3 (a) Transmission-electron micrograph (4000×) of a control cultured primary rat hepatocyte 24 h post-isolation. This view does not include the nucleus (notice the healthy, intact mitochondria and prominent endoplasmic reticulum). (b) Transmission-electron micrograph (4000×) of a control cultured primary rat hepatocyte 24 h post-isolation. This view includes healthy and prominent nucleus and mitochondria and tight junctions with neighboring cells. (c) Transmission-electron micrograph (4000×) of a cultured primary rat binucleated hepatocyte exposed to 10 μM methapyrilene for 24 h. In addition to healthy and prominent nuclei, mitochondria and other organelles, notice the large number of autophagosomes containing recycled mitochondria and endoplasmic reticulum. (d) Transmission-electron micrograph (8000×) of a cultured primary rat hepatocyte exposed to 10 μM methapyrilene for 24 h. In addition to the healthy mitochondria and other organelles, notice the autophagosomes containing recycled mitochondria and endoplasmic reticulum. (e) Transmission-electron micrograph (18 000×) of a cultured primary rat hepatocyte exposed to 10 μM methapyrilene for 24 h. This enlarged view demonstrates a single autophagosome containing recycled mitochondria and other organelles

pH alteration inhibits lysis or degradation of intralysosomal contents, in this case causing an accumulation of autophagosomes or lytic vesicles as seen in Figures 24.2 and 24.3. How does interruption of lysosomal function relate to the toxic response of hepatocytes to MP? We hypothesize that MP inhibits the essential cellular functions dependent upon a normally functional endocytic pathway. As mentioned earlier in this review, most receptormediated processes, especially those related to entry of essential nutrients like iron, utilize the endocytic pathway. As hypothesized, MP or its metabolite(s) could increase the pH of endosomes containing transferin iron complexes (or any other essential nutrient or hormonal


585

complex), thereby inhibiting ligand release resulting in starvation of the cell for iron or some other essential nutrient. The consequence of the interruption of these essential physiological processes is cell death. It is hypothesized that this mechanism of cell toxicity results in inappropriate cellular proliferation that then leads to the initiation of the carcinogenesis process.

24.6

Summary

The antihistamine methapyrilene has been of interest to investigators because it can serve as a model for the understanding of epigenetic mechanisms associated with the induction of cancer. In addition, because millions of humans had been exposed to MP, it was necessary to determine if it only induced hepatotoxicity and cancer in rodents. Many investigators suggested that MP did not initiate cancer via a genotoxic mechanism. Early ultrastructural studies suggested that MP altered mitochondrial function and apparently induced mitochondrial proliferation and somehow this response was related to cancer induction. This observation launched several studies using genomic and proteomic technologies in an attempt to understand this process at the biochemical and molecular level. However, it is our contention that the in vivo ultrastructural data were misinterpreted and the response observed was not mitochondrial proliferation but MP-induced accumulation of autophagosomes or nonfunctional lysosomal vesicles unable to complete the reverse pinocytotic process required for cellular protein and organelle turn over. The inhibition of this process was a consequence of MP’s pK a resulting in an alteration of the pH of these vesicles, rendering them nonfunctional. It is hypothesized that MP alters the pH of other vesicles, such as endosomes, resulting in starvation of the cell for essential nutrients leading to cell death. This induced toxicity-caused inappropriate hepatocyte proliferation. This cellular regeneration may lead to spontaneous mutations in oncogenes or suppressor genes, leading to the onset of cancer.

References Althaus FR, Lawrence SD, Sattler DG and Longfellow DG, Pitot HC (1982). Chemical quantification of unscheduled DNA synthesis in cultured hepatocytes as an assay for the rapid screening of potential carcinogens. Cancer Res., 42: 3010–3015. Andrews AW, Fornwald AJ and Lijinsky W (1980). Nitrosation and mutagenicity of some amine drugs. Toxicol. Appl. Pharmacol., 52: 237–244. Ashby J, Callander RD, Paton D, Zeiger E and Ratpan F (1988). Weak and unexpected mutagenicity to Salmonella of the rat hepatocarcinogen methapyrilene. Environ. Mol. Mutagen., 12: 243–252. Blazak WF, Stewart BE, Galperin I, Allen KL, Rudd CJ, Mitchell AD and Caspary WJ (1986). Chromosome analysis of trifluorothymidine-resistant L5178Y mouse lymphoma cell colonies. Environ. Mut., 8: 229–240. Budroe JD, Shaddock JG and Casciano DA (1984). A study of the potential genotoxicity of methapyrilene and related antihistamines using the hepatocyte/DNA repair assay. Mut. Res., 135: 131–137. Casciano Da and Schol MH (1984). Methapyrilene is inactive in the hepatocyte-mediated chinese hamster ovary/hypoxanthine-guanine phosphoribosyl transferase mutational assay. Cancer Lett 21: 337–341.

586

Hepatotoxicity

Casciano DA, Talaska G and Clive D (1991). The potent hepatocarcinogenn methapyrilene induces mutations in L5178Y mouse lymphoma cells in the apparent absence of DNA adduct formation. Mut. Res., 263: 127–132. Casciano DA, Shaddock JG and Talaska G (1988). The potent hepatocarcinogen methapyrilene does not induce DNA adducts in livers of Fischer 344 rats. Mut. Res., 208: 129–135. Couri D, Wilt SR and Milks MM (1982). Methapyrilene effects on initiation and promotion of λglutamyl-transpeptidase positive foci in rat liver. Res. Commun. Chem. Pathol. Pharmacol., 35: 51–61. Cunningham L, Pippin LL, Anderson NL and Wenk ML (1995). The hepatocarcinogen methapyrilene but not the analog pyrilamine induces sustained hepatocellular replication and protein alterations in F344 rats in a 13-week feed study. Toxicol. Appl. Pharmacol., 131: 216–223. Furuya K and Williams GM (1984). Neoplastic conversion in rat liver by the antihistamine methapyrilene demonstrated by a sequential syncarcinogenic effect with N -2-flurenylacetamide. Toxicol. Appl. Pharmacol., 74: 63–69. Furuya K, Mori H and Williams GM (1983). An enhancing effect of the antihistaminic drug methapyrilene on rat liver carcinogenesis by previously administered N -2-flurenylacetamide. Toxicol. Appl. Pharmacol., 70: 49–56. Glauert HP and Pitot HC (1989). Effect of the antihistamine methapyrilene as a initiator of hepatocarcinogenesis in female rats. Cancer Lett., 46: 189–194. Graichen ME, Neptun DA, Dent JD, Popp JA and Leonar TB (1984). Effects of methapyrilene on rat hepatic xenobiotic metabolizing enzymes and liver morphology. Fund. Appl. Toxicol., 5: 165–174. Hernandez L, Allen PT, Poirier LA and Lijinsky W (1989). S-adenosylmethionine, Sadenosylhomocysteine and DNA methylation levels in the liver of rats fed methapyrilene and analogs. Carcinogenesis, 10: 557–562. Iype PT, Ray-Chaudhuri R, Lijinsky W and Kelley SP (1982). Inability of methapyrilene to induce sister chromatid exchanges in vitro and in vivo. Cancer Res., 42: 4614–4618. Kammerer RC, Froines JR and Price T (1986). Mutagenicity studies of selected antihistamines, their metabolites and products of nitrosation. Food Chem. Toxicol., 24: 981–985. Kovacs AL, Reith A and Seglen PO (1982). Accumulation of autophagosomes after inhibition of hepatocytic protein degradation by vinblastine, leupeptin or a lysomotropic amine. Exp. Cell Res., 137: 191–201. Lijinsky W and Yamashita K (1988). Lack of binding of methapyrilene and similar antihistamines to rat liver DNA examined by 32 P-postlabeling. Cancer Res., 48: 6475–6477 . Lijinsky W, Reuber MD and Blackwell BN (1980). Liver tumors induced in rats by oral administration of the antihistaminic methapyrilene hydrochloride. Science, 209: 817–819. Mirsalis JC (1987). Genotoxicity, toxicity and carcinogenicity of the antihistamine methapyrilene. Mut. Res., 185: 309–317. Oberly JJ, Bewsey BJ and Probst GS (1984). An evaluation of the L51784TR +/− mouse lymphoma forword mutation assay using 42 chemicals. Mut. Res., 125: 291–306. Perera MIR, Katyal SL and Shinozuka H (1985). Methyprilene-induced membrane lipid peroxidation of rat liver cells. Carcinogenesis, 6: 925–927. Probst GS and Neal SB (1980). The induction of unscheduled DNA synthesis by antihistamines in primary hepatocyte cultures. Cancer Lett., 10: 67–73. Probst GS, McMahon RE, Hill LE, Thompson CZ, Epp JK and Neal SB (1981). Chemically induced unscheduled DNA synthesis in primary rat hepatocyte cultures: a comparison with bacterial mutagenicity using 218 compounds. Environ. Mut., 3: 11–32. Reznik-Schuller HM and Lijinsky W (1981). Morphology of early changes in liver carcinogenesis induced by methapyrilene. Arch. Toxicol., 49: 79–83.


587

Reznik-Schuller HM and Reuber MD (1986). Ultrastructure of liver tumors induced in F344 rats by methapyrilene. J. Environ. Pathol. Toxicol. Oncol., 7: 181–195. Seglen PO (1987). Regulation of autophagic protein degradation in isolated liver cells, in Lysosomes: Their Role in Protein Breakdown, Glaumann H and Ballard FJ (Eds), Academic Press, London, UK, pp. 369–414. Seglen PO (1997). DNA ploidy and autophagic protein degradation as determinants of hepatocellular growth and survival. Cell Biol. Toxicol., 13: 301–315. Seglen PO and Bohley P (1992). Autophagy and other vacuolar protein degradation systems. Experientia, 48: 158–172. Seglen PO and Solheim AE (1985). Conversion of dense lysosomes into light lysosomes during hepatic autophagy. Exp. Cell Res., 157: 550–555. Sorkin A and von Zastrow M (2002), Signal transduction and endocytosis:close encounter of many kinds. Nat. Rev. Mol. Cell Biol., 3: 800–814. Steinmetz KL, Tyson CK, Meierhenry EF, Spalding JW and Mirsalis JC (1988). Examination of genotoxicity, toxicity and morphologic alterations in hepatocytes following in vivo or in vitro exposure to methapyrilene. Carcinogenesis, 9, 959–963. Turner NT, Wooley JL, Hozier JC, Sawyer JR and Clive D (1987). Methapyrilene is a genotoxic carcinogen: Studies on methapyrilene and pyrilamine in the L5178Y/TK+/− mouse lymphoma assay. Mut. Res., 189: 285–297.

Section 8 Hepatotoxicity and Botanical Supplements

25 Botanical Supplements and Hepatotoxicity Shabana Khan, Ikhlas Khan and Larry Walker

25.1

Introduction

In the United States the use of dietary supplements, including botanical products, has risen dramatically over the last two decades. In 1993, surveys estimated that 3 % of adults in the USA used herbal remedies (Eisenberg et al., 1993). This had grown to 12 % by 1997 (Eisenberg et al., 1998) and a recent compilation by the National Center for Complementary and Alternative Medicine (Barnes et al., 2004) suggests that almost 20 % of surveyed Americans had used alternative treatments in the form of ‘natural products’ within the past year. Estimates are published that 20 % to 30 % of patients attending hepatology clinics used herbal remedies, often explicitly for the treatment of liver disease (Berk et al., 1999; Schuppan et al., 1999; Seeff et al., 2001; Strader et al., 2002). The popularity of such alternative treatments has been attributed to several factors, including dissatisfaction with costs or effectiveness of conventional medical treatment, the desire of patients to be more involved in medical decisions and treatment, the general public perception of ‘natural, therefore safe’, and creative (and sometimes misleading) marketing. Botanical supplements are especially aggressively marketed for promotion of weight-loss, enhanced sexual performance, or as ‘anti-aging’ therapies. Use is also exceptionally high among individuals with life-threatening illnesses such as cancer (Sparber and Wootton, 2001) or HIV (Wootton and Sparber, 2001). Such patients are typically using botanical supplements as adjuncts to conventional therapies, often with the rationale that ‘it can’t hurt, but it may help’. These products are often taken without medical consultation or disclosure (Eisenberg et al., 2002). There is a popular perception that ‘because they are natural, they are safe’


592

Hepatotoxicity

or ‘they have been used for centuries/millennia without harmful effects’. Such perceptions of herbal remedies are often reinforced by individual patient experience, since most of these agents do in fact prove to be relatively innocuous, and some may have real beneficial effects. However, the collective experience reflected in the literature certainly admits of some notable exceptions, and several such ‘remedies’ have resulted in severe, and even fatal, complications. For a variety of reasons, some of which will be discussed in this chapter, ‘history of use’ for a botanical is not a guarantee of safety, particularly with longterm use, at high doses, or with co-medications and/or co-morbidities.

25.2

General Safety Issues with Botanicals

With this brief background, it can be appreciated that careful attention is warranted to potential safety concerns over botanical dietary supplements. With widespread use, some serious adverse events have been encountered, including myocardial infarction, stroke, renal and hepatic failure. Among the most serious safety concerns for botanical dietary supplements is the potential for liver injury. A growing number of botanical/herbal products such as kava, germander, chaparral, greater celandine, and comfrey have been associated in case reports with rare but severe cases of liver injury, as summarized below. Several of these cases have occasioned considerable controversy with regard to causation, but clearly there are grounds for concern that use of some botanicals is associated with hepatotoxicity. There is a critical need to develop better understanding of the determinants of sensitivity to botanical supplement-associated liver injury, the mechanisms involved, and how to better assess/predict the toxic potential of botanical products. Several serious botanical–drug interactions have also been reported, the best studied of which is the induction of cytochrome P450 3A mediated drug metabolism with coadministration of St. John’s wort. This combination has resulted in very important clinical sequelae, including transplant rejection in patients on immunosuppressant therapy and interference with antiretroviral therapy.

25.3

Complexities in Chemical and Biological Characterization of Botanicals

A major complicating issue for understanding the safety and efficacy of botanical supplements is their chemical complexity. While a prescription pharmaceutical preparation involves a single, pure, well-characterized and well-studied chemical entity (along with known and innocuous excipients and carriers used in their formulation), for the botanical products, even within a single herb, there may be dozens or hundreds of constituents, depending on the source or preparation. In addition, many products contain not just a single plant, but combinations of several herbs; this is particularly common in Ayurvedic or Traditional Chinese Medicine remedies. This complicates our understanding of many aspects of supplement use, from the basic knowledge of what exactly is being ingested, to interpretation of clinical efficacy, to monitoring of adverse events. Furthermore, in research on biological activities of these supplements, many factors contribute to variable results, namely the uncertain identities of raw materials used (misidentified species, adulterant

Hepatotoxicity of Botanicals

593

or contaminant plants), the variability in constituents within a species, depending on plant part, developmental stage, and growth conditions, and the variables introduced with harvest, processing and product formulation.

25.4

Regulatory Aspects

Botanical supplements in the United States are regulated according to the Dietary Supplement Health and Education Act (DSHEA) (http://www.fda.gov/opacom/laws/dshea. html#sec3), which became law in 1994. A dietary supplement under this law is basically a product (other than tobacco) that is intended to supplement the diet, contains one or more dietary ingredients (including vitamins, minerals, herbs or other botanicals, amino acids and other substances) or their constituents, is intended to be taken by mouth as a pill, capsule, tablet or liquid, and is labeled as being a dietary supplement. Botanical supplements are marketed as fresh or dried crude plant material or in the form of various types of extracts, including purified or partially purified components. The definition excludes any drugs and any drug claims (to diagnose, treat, mitigate, cure or prevent disease). These supplements are thus regulated essentially as foods under federal law. Safety assessment requirements for dietary ingredients are quite different from that required for pharmaceuticals. With the exception of new dietary ingredients (those introduced to the market since 1994), dietary supplements are not subject to FDA approval before marketing, nor are products and manufacturers required to register with the FDA. Under DSHEA, botanical dietary supplements cannot be removed from the market by the FDA unless there is evidence of harmful effects or mislabeling. This regulatory approach is in contrast to the drug approval process in which both proof of efficacy and safety must be established in humans.

25.5

Understanding Xenobiotic-induced Hepatotoxic Injury

The metabolic function of the liver is primarily responsible for the detoxification of structurally diverse xenobiotics. Phase I enzymes in the liver are responsible for ‘functionalization’ of xenobiotics, in the introduction or exposure of functional groups, e.g. hydroxylation. Phase II enzymes typically are responsible for coupling reactions with endogenous polar conjugates, e.g. glucuronic acid. Both phase I and II enzymes can be induced or inhibited by other conditions, including drugs, dietary factors, inflammation and infection. These metabolic routes typically reduce the activity or toxicity of the xenobiotic and facilitate its excretion via bile or urine. There are also, however, numerous examples of xenobiotics for which these enzymes, particularly phase I enzymes, lead to metabolic activation of protoxicants. These factors can predispose the liver to injury with certain xenobiotics. Although this is widely appreciated for drug-induced hepatotoxicity, the same principles are relevant for botanical products. In fact, the picture can be much more complicated, since the botanical preparations contain many constituents that have the potential for interacting with liver enzyme systems, depending on dose and duration of exposure. Many hepatotoxic reactions are idiosyncratic – that is, having an unpredictable, relatively low incidence, not necessarily dose-related (Kaplowitz and DeLeve, 2003). Idiosyncratic reactions can be allergic or non-allergic, depending on the presence of clinical features such

594

Hepatotoxicity

as fever, rash, eosinophilia and other symptoms related to the adaptive immune system. Often idiosyncratic drug reactions are set against a background mild liver injury; this has led to an hypothesis that transient, mild liver injury might progress to severe injury depending on certain genetic or environmental factors, and may involve associated reactions of inflammation and cell death (Kaplowitz, 2005). Drug-induced liver injury may be manifested as acute hepatitis (from hepatocellular injury and inflammation) or cholestasis (canalicular or ductular injury and inflammation) and very often presents as a mixed pattern. The innate immune system’s adaptive responses after injury may mitigate initial mild insult. However, it is believed that in some circumstances, the injury can sensitize hepatic cells to the toxic effects of cytokines or other inflammatory mediators, leading to progressive injury and cellular death. Furthermore, some offending metabolites may cause ‘haptenization’, thus eliciting aberrant immune responses to hepatic cells. Predictable liver toxicities are often discovered in preclinical and clinical testing for drugs. However, no such testing is currently required for botanicals marketed as dietary supplements. Furthermore, because the chemistry of these plant-derived products can be quite variable and the use sometimes sporadic, even the occurrence of ‘predictable’ hepatotoxicities may be rare. Unpredictable hepatotoxicity associated with either drugs or botanicals is also extremely difficult to recognize or assess because estimates of prevalence of adverse reactions ranges between 1 in 10 000 and 1 in 100 000 against a large background (Hampson and Harvey 2002). Recognition of causality in idiosyncratic hepatotoxicity is further confounded by environmental (underlying disease, co-medication, inflammation or tissue injury) or genetic factors (drug metabolism or transport differences), induction or inhibition of metabolizing enzymes and the molecular nature of the toxicant (Boelsterli, 2003). A field of growing interest and importance with regard to idiosyncratic drug-induced hepatotoxic reactions is the mechanistic role of metabolic, inflammatory and other processes in the liver injury. (Ganey et al., 2004; Tafazoli et al., 2005). Although some hepatotoxicants are certainly directly toxic to hepatic parenchymal cells in vitro, there is strong evidence that for many of them, more complex (and variable) mechanisms are involved in the liver injury in vivo. There has in the past two decades been an increasing awareness of the potential role that pharmacogenomic susceptibility may play, especially with regard to population variations in the CYP isoforms that mediate phase I transformations (Watkins, 2003) or in the transferases that carry out conjugation reactions. More recently, however, additional factors have been highlighted as pivotal in some circumstances. Robert Roth and colleagues have reported a series of compelling studies indicating that other cell types and several inflammatory mediators play key roles in the insult, including Kupffer cells, neutrophils, endothelial cells, coagulation factors, tumor necrosis factor-α, cyclooxygenase-2, macrophage inflammatory protein-2 and others (Copple et al., 2002, 2003; Yee et al., 2000, 2002, 2003a,b; Ganey et al., 2004; Luyendyk et al., 2003, 2005, 2006). Focusing on the best developed example from this work, ranitidine is an antagonist of histamine-2 receptors, first marketed as a prescription drug for treatment of symptoms of gastric hyperacidity. Ranitidine is associated with idiosyncratic liver toxicity, occurring in a very small fraction of users. Roth’s group established that although ranitidine is not normally hepatotoxic in rats, co-administration of low doses of bacterial lipopolysaccharide (LPS) predisposed dramatically to liver injury by ranitidine (Luyendyk et al., 2003). This reaction was mitigated by heparin or by neutrophil depletion, and appeared to be dependent


595

on coagulation-mediated hypoxia (Luyendyk et al., 2005). Transcript profiling (Luyendyk et al., 2006) indicated the enhanced expression in LPS/ranitidine-treated rats of several gene products capable of regulating neutrophils, hypoxia responses and hemostasis. In addition, the profiles suggested the involvement of the p38/MAP kinase-activated protein kinase-2 pathway in the augmentation of LPS-induced gene expression by ranitidine. Importantly, no liver injury and no such gene expression changes were observed in rats treated with LPS and famotidine, a histamine-2 receptor antagonist that does not elicit idiosyncratic hepatotoxicity in man. Similar mechanisms may be involved for plant-derived hepatotoxins. Monocrotaline (MCT), a pyrrolizidine alkaloid (see below), causes pronounced necrosis of hepatic parenchymal cells by a mechanism involving endothelial cells (Copple et al., 2002), activation of the coagulation system, and centrilobular hypoxia (Copple et al., 2002, 2003). Although larger doses of MCT do not require inflammation to cause liver injury, when rats were treated with doses of lipopolysaccharide (LPS) sufficient to induce a modest inflammatory response, subsequent administration of sub-threshold doses of MCT elicited pronounced hepatotoxicity (Yee et al., 2000). The enhanced sensitivity appears to require Kupffer cells, neutrophils and tumor necrosis factor-α, as well as an activated coagulation system (Yee et al., 2002, 2003a,b). This raises the strong possibility that ‘idiopathic’ liver injury due to botanicals may be due to the convergence of factors in addition to the herbal product; consumers who experience an infection or inflammatory episode may be at greater risk for developing liver injury. Predictable liver toxicities are often discovered in preclinical and clinical testing for drugs. However, no such testing is currently required for botanicals marketed as dietary supplements. Furthermore, because the chemistry of these plant-derived products can be quite variable and the use sometimes sporadic, even the occurrence of ‘predictable’ hepatotoxicities may be rare. Unpredictable hepatotoxicity associated with either drugs or botanicals is also extremely difficult to recognize or assess because estimates of prevalence of adverse reactions ranges between 1 in 10 000 and 1 in 100 000 against a large background (Hampson and Harvey, 2002). Recognition of causality in idiosyncratic hepatotoxicity is further confounded by environmental (underlying disease, co-medication, inflammation or tissue injury) or genetic factors (drug metabolism or transport differences), induction or inhibition of metabolizing enzymes and the molecular nature of the toxicant (Boelsterli, 2003).

25.6

Hepatotoxicity of Botanical Products

Herbal products that have been associated with hepatotoxicity have been the subject of several reviews (Stedman, 2002; Pageaux and Larrey, 2003; Zhou et al., 2004; Stickel et al., 2005; Dasgupta and Bernard 2006) – a listing of some of the known or suspected agents, with identified toxic principles (where known) is provided in Table 25.1. 25.6.1

Comfrey and Other Pyrrolizine Alkaloid (PA)-Containing Plants

Botanicals containing pyrrolizidine alkaloids (PAs) have been recognized for their hepatotoxicity for over 70 years (Stedman, 2002). These alkaloids cause a characteristic venoocclusion, resulting in hepatic congestion and centrilobular necrosis. PAs can be found in

Gum Thistle Pennyroyal

Jin Bu Huan (Chinese herbal) Senna

Impila Ephedra, ma huang

Indian pennywort (Gotukola) Noni juice Green tea

Atractylis gummifera Hedeoma, Mentha

Lycopodium serratum ´ a and Cassia acutifolia C.angustifoliaa

Callilepsis laureolaa Ephedra sinicaa

Centella asiaticaa Morinda citrifoliaa Camellia sinensisa

Suggestive evidence only, based on a limited number of case reports.

Rattleweed, sunhemp White or garden heliotrope Comfrey Germander Kava kava Greater Celandine Chaparral

Crotalaria spp. Heliotropium spp. Symphytum spp. Teucrium chamaedrys Piper methysticum Chelidonium majus Larrea tridentata

a

Common names

Botanical

? Anthraquinones? ?

Atractylosides ?

Alkaloid? Anthraquinone glycosides (sennosides)

Loeper et al., 1994 Teschke et al., 2003; Schulze et al., 2003 Benninger et al., 1999; Stickel et al., 2005 Sheikh et al., 1997; Lambert et al., 2002

Neoclerodane diterpenes Kavapyrones? Unidentified alkaloid(s)? Nordihydroguaiaretic acid (NDGA) quinone? Atractylosides, gummiferin Pulegone

Stickel et al., 2005 Nadir et al., 1996; Bajaj et al., 2003; Neff et al., 2004 Jorge and Jorge, 2005 Millonig et al., 2005; Stadlbauer et al., 2005 Bonkovsky, 2006

Stickel et al., 2005 Sztajnkrycer, 2003; Anderson et al., 1996; Bakerink et al., 1996 Woolf et al., 1994; McRae et al., 2002 Vanderperren et al., 2005

Stedman, 2002 Wang et al., 2005

References

Pyrrolizidine alkaloids

Suspected toxic principles

Table 25.1 Botanicals with established or putative hepatotoxic potential, along with their toxic principles (see also Stedman, 2002; Pageaux and Larrey, 2003; Stickel et al., 2005)


597

more than 350 plant species, but Heliotropium, Senecio, Symphytum and Crotalaria species have been most clearly associated with liver injury. Their content may vary in particular plants and clinical courses show striking differences due to varying amounts of ingested toxins. Human exposure typically occurs from teas, contaminated foods or herbal preparations. Although PAs found in Heliotropium, Senecio and Crotalaria were considered toxic, extracts of Symphytum (comfrey) leaves and roots which were widely sold in USA were also found to cause liver injury. Because of the presence of at least nine hepatotoxic PAs found in comfrey, companies have been advised by the FDA to remove comfrey products for internal consumption from the market (www.cfsan.fda.gov/∼dms/dspltr06.html). The mechanism of hepatotoxicity exerted by PAs is not fully understood but it is known to be related to biotransformation by cytochrome P450s. Metabolic transformation of PAs is enhanced in presence of dexamethasone and inhibited by troleandomycin (Wang et al., 2005). A genotoxic mechanism has been related to the tumorigenicity of retrorsine and monocrotaline (retronecine-type PAs) through formation of a common metabolite DHP (dihydropyrrolizidine derivative) and DHP-derived DNA adducts in liver of treated rats. However it is not known if this mechanism accounts for veno-occlusive injury or is generalizable to PAs of other types.

25.6.2

Germander (Teucrium chamaedrys L)

Germander, a member of Labiatae family, is a traditionally used remedy that was approved in France in 1986 as a drug for the treatment of obesity and mild diarrhoea. As a result of high intake, in the early 1990s, several reports appeared about germander-associated hepatoxicity, with a range of classification and severity including acute, chronic and fulminant hepatitis. Hepatitis mostly developed after two months of permanent medication. The chemical constituents of germander include saponins, glycosides, flavonoids and a number of furan-containing neoclerodane diterpenoids. Furans are well-known to be carcinogens in animal models. Several furano compounds are well known hepatotoxins after being metabolized to reactive epoxides and unsaturated aldehydes by cytochrome P450s (CYP3A). Hepatotoxicity of germander mediated by its furano neoclerodane diterpenoid is proven in animal (mice) studies. Formation of hepatotoxic metabolites by CYP3A was observed which was enhanced by inducers of CYP3A and glutathione depletion. Toxicity was prevented by pretreatment of a single dose of troleandomycin and enhanced by pretreatment with dexamethasone or clotrimazole (Loeper et al., 1994). Variability in human CYP3A includes polymorphisms in CYP3A5 and CYP3A7 that occur at different frequencies in various populations. These polymorphisms can be major determinants of the percentage of the total CYP3A expressed in human liver (Kuehl et al., 2001; Burk et al., 2002). If CYP3A5 and 3A7 play a role in the germander CYP3A metabolism, certain individuals with polymorphisms in these isoforms could be predisposed to Germander-induced hepatotoxicity. Germander was withdrawn from the market in 1992. Another plant of the same genus, Teucrium polium, has been used for anti-inflammatory and antimicrobial properties, as well as for treatment of scars. It was also reported to cause cholestatic hepatitis (Mazakopakis et al., 2004).

598

25.6.3

Hepatotoxicity

Kava Kava (Piper methysticum)

Kava is a shrub native to the South Pacific. Its roots and rhizomes have been used for generations to prepare a psychoactive beverage. Use of dietary supplements containing kava extracts for treatment of anxiety, stress and insomnia has gained popularity in the western world in the past few decades. Kava lactones are the major components which are pharmacologically active. Products containing kava pyrones were withdrawn in Germany in June 2002, after suggestions of hepatotoxicity. Controversy surrounding these cases continued as patients’ case histories were re-analyzed and the regulatory decision was questioned (Teschke et al., 2003; Schulze et al., 2003). As a result, kava-containing herbal supplements were confirmed to be associated with acute hepatotoxicity. Seven case reports suggested the development of hepatitis after kava intake. Later on, analysis of 29 more cases of hepatitis between 1990 and 2002 established adverse hepatic reactions of hepatic necrosis and cholestatic hepatitis. Nine patients developed fulminant liver failure (Stickel et al., 2003a). Kava products are banned in several countries around the world. Kava lactones, as well as extracts, were found to inhibit several CYP450 enzymes when incubated with liver microsomes (Mathews et al., 2002), suggesting a high potential for drug interactions (Anke and Ramzan, 2004). It is also suspected that some alkaloids might contribute to its hepatotoxicity (Nerurkar et al., 2004). In HepG2 cells, pipermethystine (alkaloid from kava) caused cell death by disrupting mitochondrial function and inducing apoptosis, indicating that it may contribute to rare but severe hepatotoxic reactions. Interactions with co-ingested medication is one of the many hypotheses postulated for kava related hepatotoxicity (Singh, 2005). Genetic polymorphism of cytochromes P450 may be an important factor for the discrepancy in hepatotoxic response to kava from Caucasians kava users and Pacific Islanders. The mechanism of toxicity is still not clear but in vitro data suggest that kava lactones do not appear to be activated to toxic metabolites by drug-metabolizing enzymes (Zou et al., 2004). 25.6.4

Chaparral (Larrea tridentata)

Chaparral is a desert plant commonly referred to as ‘creosote bush’ or ‘greasewood’ and has been used by native Americans to treat a number of conditions including tuberculosis, arthritis and cancer (Tyler, 1993). It is also used for its anti-inflammatory and blood purifying potential. Eighteen case reports of adverse events associated with the ingestion of chaparral were reported to FDA between 1992 and 1994 and evidence of hepatotoxicity was seen in thirteen cases. These were associated with marked increase in serum liver enzymes, 3–52 weeks after ingestion of chaparral. The predominant pattern of liver injury was characterized as toxic or drug-induced cholestatic hepatitis progressing to cirrhosis (in four individuals) and liver failure (in two individuals) (Sheikh et al., 1997). Nordihydriguaiaretic acid (NDGA), a major lignan found in the leaves and twigs of L. tridentata (10–15 % by dry weight), seems to contribute to its hepatotoxicity (Lambert et al., 2002). Acute exposure of mice to NDGA resulted in a dose-dependent lethality. The toxicity was also marked by a time- and dose-dependent increase in serum alanine aminotransferase (ALT) levels. A number of naturally occurring methylated and acetylated derivatives of NDGA are also found in L. tridentata (Gnabre et al., 1995). These compounds may undergo O-demethylation or de-esterification in vivo and further increase the amount of polyphenolic moieties available for oxidation. Glucuronidation was identified as a potential detoxification mechanism for


599

NDGA. Within minutes of iv administration of NDGA to mice, mono and diglucuronides r was apappeared in plasma (Lambert et al., 2002). A topical form of NDGA (Actinex) proved by the FDA for the treatment of keratoses but was withdrawn from the market due to skin hypersensitivity (Kulp-Shorten et al., 1993). 25.6.5

Atractylis and Other Plants Containing Atractylosides

Atractylis gummifera is used as antipyretic, emetic and diuretic. A whitish fluid secreted from the plant is also chewed by children as gum. Over two dozen plant species grow around the Meditarranean and the African continent. Symptoms such as nausea, abdominal pain, diarrhea, headache and convulsions, followed by acute hepatitis and fatal liver failure have been reported after a few hours of ingestion. In the spring, toxins are concentrated in the roots, which are sometimes confused with wild artichoke, consumption of which causes serious problems. Atractylosides and gummiferin are responsible for toxicity. Tissues of high metabolic activity are the main target organ. These compounds interact with a mitochondrial protein, the adenine nucleotide translocator that is involved in mitochondrial membrane permeabilization. They inhibit mitochondrial function and exert oxidative stress as a result of glutathione depletion and increased lipid peroxidation. In vitro experiments showed that verapamil or DTT could protect against the toxic effects of atractyloside, if administered before atractyloside exposure (Daniele et al., 2005; Hamouda et al., 2004). Callilepsis laureola is another plant that contains atractylosides which has been associated with severe cases of fulminant hepatitis among Zulus in South Africa (Stickel et al., 2000). 25.6.6

Jin Bu Huan (Lycopodium serratum)

The Chinese herbal product Jin Bu Huan, widely used as remedies to various ailments, can also cause liver injury. Cases of acute hepatitis associated with this product were reported between 7 and 52 weeks of Jin Bu Huan ingestion, with symptoms of fever, fatigue, nausea, pruritis abdominal pain and signs of jaundice and hepatomegaly. No conclusive evidence was provided to confirm what caused hepatitis as these products may contain adulterants. Although the hepatotoxic mechanisms are not defined for this product, they may include hypersensitive or idiosyncratic reactions or direct toxicity to active metabolites. One active ingredient in Jin Bu Huan tablets is levo-tetrahydropalmatine, which is present in the plant genera Stephania and Corydalis. The hepatotoxic mechanism of levo-tetrahydropalmatine is unknown (Woolf et al., 1994; McRae et al., 2002). 25.6.7

Pennyroyal (Mentha pulegium and Hedeoma pulegoides)

Pennyroyal, also referred to as squawmint oil or ‘poejo’, is used in folk medicine, as an abortifacient and as pesticides against fleas, for centuries. Pulegone, a monoterpene ketone, is the principal component of the essential oil of pennyroyal. Pennyroyal oil ingestion, at high doses, has been associated with severe hepatotoxicity and death (Anderson et al., 1996; Bakerink et al., 1996). Its primary constituent, pulegone, is considered to be responsible for hepatotoxicity as a result of metabolic conversion to toxic intermediates (e.g. menthofuran and p-cresol) via hepatic cytochrome P450s (Sztajnkrycer, et al., 2003) and by producing oxidative stress. The in vivo metabolism of pulegone is complicated and about 14 phase

600

Hepatotoxicity

I and 10 phase II metabolites have been identified and characterized (Chen et al., 2001). Pennyroyal is still widely available and its use will continue to be a public health concern. 25.6.8

Greater Celandine (Chelidonium majus)

Herbal preparations containing Greater celandine are frequently used in Europe for their cholerectic and antispasmodic properties to treat gastric and biliary disorders. Greater celandine is a mixture of 20 different alkaloids. Potential hepatotoxicity of this herb has been reported; however, the exact mechanism is still unclear and rapid recovery was observed in all patients after the withdrawal (Benninger et al., 1999; Stickel et al., 2003b). Despite the evidence of toxicity, Greater celandine preparations are still in the market (Stickel et al., 2005). 25.6.9

Preparations Containing Usnic Acid

Usnic acid, a lichen alkaloid, is a component found in crude medicines and dietary suppler used for weight loss. Durazo et al. (2004) recently reported ments, including LipoKinetix on a case of hepatoxicity requiring liver transplant in a patient taking pure usnic acid. The r (www.cfsan.fda.gov/∼dms/dsFDA has recently warned against the use of LipoKinetix r contains sodium usniate, norephedrine, lipo.html) (Novak and Lewis, 2003). LipoKinetix 3,5-diiodothyronine, yohimbine and caffeine. Patients taking this supplement developed heptatotoxic symptoms to varying degrees, including elevated ALT levels, jaundice, acute hepatitis, massive necrosis and fulminant hepatic failure (Favreau et al., 2002; Estes et al., 2003; Neff et al., 2004). In vitro, usnic acid was found to exert potent direct cytotoxic effects on cultured hepatocytes and caused necrosis (Han et al., 2004). 25.6.10

Other Botanicals with Potential Hepatotoxicity Concerns

Case reports have appeared recently describing liver injury which the authors attributed to various herbal products: Cassia acutifolia and C. angustifolia (dry Senna fruits), Centella asiatica (Gotu kola), Ephedra sinica (ma huang), Camellia sinensis (green tea) and Morinda citrifolia (Noni juice) (Neff et al., 2004; Vanderperren et al., 2005; Jorge and Jorge, 2005; Stadlbauer et al., 2005, Millonig et al., 2005). The associations in most of these cases are more circumstantial and it remains to be seen whether a true causative relationship can be established. Ephedra has been used for thousands of years for medicinal purposes in ancient Chinese medicines, where it is known as ‘ma huang’. Ephedra came to be widely used in the USA for its weight-reducing and energizing effects. The main active principle, the alkaloid ephedrine, is a sympathomimetic amine with agonist activity at adrenergic receptors, an effect deemed responsible for cardiovascular (stroke, myocardial infarction) and neurological (seizure)-adverse events. Since April 2004, sales of Ephedra-containing dietary supplements are banned by the FDA due to concerns over these adverse effects (Rados, 2004; Keisler and Hosey, 2005; Clark and Schofield, 2005). A few cases of hepatotoxicity associated with ma huang use have been reported (Nadir et al., 1996; Bajaj et al., 2003; Neff et al., 2004). However, many of these reports have confounding conditions and it is presently unclear if there is an association of liver injury with Ephedra, and if so, whether they are attributable to ephedrine.


601

Senna seeds in the form of herbal tea are widely used as laxatives. Severe hepatotoxicity related to its abuse is unusual but could be explained by the exposure of liver to unusual amounts of toxic metabolites of anthraquinone glycosides (sennosides). High concentrations of cadmium were also found in the urine of patients suggesting a contamination of herbal tea by metals (Vanderperren et al., 2005). Centella asiatica has been used in Ayurvedic and other traditional systems for wound healing, treatment of leprosy and to promote general health and longevity. It has also been used for dementias, cognitive disorders, diabetic microangiopathy and atherosclerotic plaques, as well as psoriasis. Pentacyclic triterpenic saponosides are considered the active principles. Three clinical cases of hepatotoxicity related to Centella asiatica ingestion were reported (Jorge and Jorge, 2005). The liver biopsies showed marked eosinophilic degeneration and cellular necrosis. Noni juice, prepared from the fruit of Morinda citrifolia, is gaining increasing popularity as a ‘wellness drink’, claimed to be beneficial for many illnesses. Not much was known about its toxicity until recently three cases of hepatotoxicity were reported in Austria (Stadlbauer et al., 2005; Millonig et al., 2005). Anthroquinones are likely to be hepatotoxic components. Although preliminary, these cases raise the awareness about the potential hepatotoxicity of Noni juice. Recent reports have appeared attributing hepatitis to the use of green tea preparations (Camellia sinensis) (Bonkovsky, 2006). In this letter to the editor, the author tabulates a listing of nine case reports of subjects who developed hepatitis after taking green tea extract preparations, although as pointed out by the author, several of the supplement products listed contained a number of ingredients, sometimes unspecified, and the times after exposure varied greatly, making it difficult to establish a causal relationship. The author suggested that the effect may be due to the high exposure levels to tea constituents due to extraction and concentration in the dosage form. These will, no doubt, elicit some further assessment regarding green tea’s potential contributions. It should be mentioned that there is a large body of literature demonstrating protective effects of green tea in animal models of chemical- or drug-induced liver injury (see, for example, Chen et al., 2004; Oz et al., 2004; Sai et al., 1998).

25.7

Conclusions

Hepatic injury due to herbal products, while still relatively uncommon, is a serious problem in that the use of botanical supplements is extremely widespread (and often unreported) and based on the common perception of safety, supplements may be taken at high doses and over a long period of time. These characteristics are especially likely for weight-loss products, where users will often aggressively ‘self-medicate’ to accelerate the desired outcome. The complexity and variability of botanical products, insufficient rigorous quality-control standards in manufacturing and the limited statutory regulation of dietary supplements under the DSHEA make the problem a more difficult one to manage than is the case with prescription drugs. The proposed FDA rules for ‘Good Manufacturing Practices’ http://www.cfsan.fda.gov/∼lrd/tr07625a.html recently once implemented, will be an important step toward minimizing the incidence of hepatotoxic and other adverse events and facilitate tracking offending agents in cases that do occur.

602

Hepatotoxicity

More awareness and broader utilization of avenues for adverse event reporting (e.g. FDA’s MedWatch program for health care professionals (see http://www.fda.gov/medwatch/ index.html) will facilitate the identification of problem plants and their toxic constituents. Such reporting is critical and it should, for the sake of balance, be pointed out that the growing awareness in the medical community has resulted in a proliferation of case reports in the clinical literature. These often draw unwarranted conclusions where adverse event causation is difficult to establish. So, while these reports should serve to raise our awareness and promote additional exploration of these herbal products, it can in reality be counterproductive to use a few isolated case reports to classify a botanical as a hepatotoxin (particularly those with limited information about the botanical or its constituents, or those where there are multiple potentially confounding causes). However, with more comprehensive and diligent use of national reporting systems, offending agents can be more quickly pinpointed. Advances in our understanding of drug-associated idiosyncrasies hold great promise for application to botanical products. Thus, clarifying the bases for variations in individual susceptibility to hepatic injury – whether due to genetic, environmental, co-medication or pathophysiological factors – will certainly help us to better understand the mechanisms involved and ultimately lead to better prevention and treatment.

References Anderson IB, Mullen WH, Meeker JE, Khojasteh-Bakht SC, Oishi S, Nelson SD and Blanc PD (1996). Pennyroyal toxicity: measurement of toxic metabolites in two cases and review of the literature. Ann Intern Med 124:725–734. Anke J and Ramzan I (2004). Pharmacokinetic and pharmacodynamic drug interactions with Kava (Piper methysticum Forst. f.). J Ethnopharmacol. 93:153–60. Bajaj J, Knox JF, Komorowski R and Saeian K (2003). The irony of herbal hepatitis: Ma-Huanginduced hepatotoxicity associated with compound heterozygosity for hereditary hemochromatosis. Dig Dis Sci 48:1925–1928. Bakerink JA, Gosper SM, Dimand RJ and Eldridge MW (1996). Multiple organ failure after ingestion of pennyroyal oil from herbal tea in two infants. Pediatrics 98:944–947. Barnes PM, Powell-Griner E, McFann K and Nahin RL (2004). Complementary and alternative medicine use among adults: United States, 2002, Advance Data From Vital and Health Statistics, No 343, National Center for Health Statistics, Hyattsville, MD, USA. Benninger J, Schneider HT, Schuppan D, Kirchner T and Hahn EG (1999). Acute hepatitis induced by greater celandine (Chelidonium majus). Gastroenterology 117:1234–1237. Berk BS, Chaya C, Benner KC and Flora KD (1999). Comparison of herbal therapy for liver disease: 1996 versus 1999. Hepatology 30:478A. Boelsterli, UA (2003). Idiosyncratic drug hepatotoxicity revisited: New insights from mechanistic toxicology. Toxicol Mech Meth 13:3–20. Bonkovsky HL (2006). Hepatotoxicity associated with supplements containing Chinese green tea (Camellia sinensis). Ann Intern Med. 144:68–71. [Erratum in: Ann Intern Med. 2006 Mar 7;144(5):380]. Burk O, Teguda H, Koch I, Hustert E, Wolbold R, Glaeser H, Klein K, Fromm MF, Nuessler AK, Neuhaus P, Zanger UM, Eichelbaum M and Wojnowski L (2002) Molecular mechanisms of polymorphic CYP3A7 expression in adult human liver and intestine. J Biol Chem 277:24280–24288.


603

Chen LJ, Lebetkin EH and Burka LT (2001). Metabolism of (R)-(+)-pulegone in F344 rats. Drug Metab Dispos. 29:1567–1577. Chen JH, Tipoe GL, Liong EC, So HS, Leung KM, Tom WM, Fung PC and Nanji AA (2004). Green tea polyphenols prevent toxin-induced hepatotoxicity in mice by down-regulating inducible nitric oxide-derived prooxidants. Am J Clin Nutr 80:742–751. Clark BM and Schofield RS (2005). Dilated cardiomyopathy and acute liver injury associated with combined use of ephedra, gamma-hydroxybutyrate, and anabolic steroids. Pharmacotherapy 25:756–761. Copple BL, Banes A, Ganey PE and Roth RA (2002). Endothelial cell injury and fibrin deposition in rat liver after monocrotaline exposure. Toxicol Sci 65:309–318. Copple BL, Ganey PE and Roth RA (2003). Liver inflammation during monocrotaline hepatotoxicity. Toxicology 190:155–169. Daniele C, Dahamna S, Firuzi O, Sefaki N, Saso L and Mazzanti G (2005). Atractylis gummifera L. poisoning: an ethnopharmacological review. J Ethnopharmacol 97:175–181. Dasgupta A and Bernard DW (2006). Herbal remedies: effects on clinical laboratory tests. Arch Pathol Lab Med 130:521–528. Durazo FA, Lassman C, Han SB, Saab S, Lee NP, Kawano M, Saggi B, Gordon S, Farmer DG, Yersiz H, Goldstein RL, Ghobrial M and Busuttil RW (2004). Fulminant liver failure due to usnic acid for weight loss. Am J Gastroenterol 99:950–952. Eisenberg DM, Kessler R, Foster C, Norlock FE, Calkins DR and Delbanco TL (1993). Unconventional medicine in the United States. Prevalence, costs and patterns of use. N Engl J Med 328:246–252. Eisenberg DM, Davis RB, Ettner SL, Appel S, Wilkey S, Van Rompay M and Kessler RC (1998). Trends in alternative medicine use in the United States, 1990–1997: results of a follow-up national survey. JAMA 280:1569–1575. Eisenberg DM, Kessler RC, Van Rompay MI, Kaptchuk TJ, Wilkey SA, Appel S and Davis RB (2001). Perceptions about complementary therapies relative to conventional therapies among adults who use both: results from a national survey. Ann Intern Med 135:344–351. Estes JD, Stolpman D, Olyaei A, Corless CL, Ham JM, Schwartz JM and Orloff SL (2003). High prevalence of potentially hepatotoxic herbal supplement use in patients with fulminant hepatic failure. Arch Surg 138:852–858. Favreau JT, Ryu ML, Braunstein G, Orshansky G, Park SS, Coody GL, Love LA and Fong, T-L (2002). Severe hepatotoxicity associated with the dietary supplement LipoKinetix. Annal Intern Med 136:590–595. Ganey PE, Luyendyk JP, Maddox JF and Roth RA (2004). Adverse hepatic drug reactions: inflammatory episodes as consequence and contributor. Chem Biol Interact 150:35–51. Gnabre JN, Brady JN, Clanton DJ, Ito Y, Dittmer J, Bates RB and Huang RC (1995). Inhibition of human immunodeficiency virus type 1 transcription and replication by DNA sequence-selective plant lignans. Proc Natl Acad Sci USA 92:11239–11343. Hampson JP and Harvey JN (2002). Letters: Postmarketing surveillance and black box warnings. J Am Med Assoc 288: 956. Hamouda C, Hedhili A, Ben Salah N, Zhioua M and Amamou M (2004). A review of acute poisoning from Atractylis gummifera L. Vet Hum Toxicol 46:144–146. Han D, Matsumara K, Rettori D and Kaplowitz N (2004). Usnic acid-induced necrosis of cultured mouse hepatocytes: inhibition of mitochondrial function and oxidative stress. Biochem Pharmacol 67:439–451. Jorge OA and Jorge AD (2005). Hepatotoxicity associated with the ingestion of Centella asiatica. Rev Esp Enferm Dig 97:115–124. Kaplowitz N (2005). Idiosyncratic drug hepatotoxicity. Nat Rev Drug Discov 4:489–499.

604

Hepatotoxicity

Kaplowitz N, and DeLeve LD (2003). Drug-Induced Liver Disease, Marcel Dekker, New York, NY, USA. Keisler BD and Hosey RG (2005). Ergogenic aids: an update on ephedra. Curr Sports Med Rep 4:231–235. Kuehl P, Zhang J, Lin Y, Lamba J, Assem M, Schuetz J, Watkins PB, Daly A, Wrighton SA, Hall SD, Maurel P, Relling M, Brimer C, Yasuda K, Venkataramanan R, Strom S, Thummel K, Boguski MS and Schuetz E (2001). Sequence diversity in CYP3A promoters and characterization of the genetic basis of polymorphic CYP3A5 expression. Nat Genet 27:383–391. Kulp-Shorten CL, Konnokov N and Callen JP (1993). Comparative evaluation of the efficacy and safety of masoprocol and 5-fluorouracil cream for the treatment of multiple actinic keratoses of the head and neck. J Geriatric Dermatol 1:161–168. Lambert JD, Zhao D, Meyers RO, Kuester RK, Timmermann BN and Dorr RT (2002). Nordihydroguaiaretic acid: hepatotoxicity and detoxification in the mouse. Toxicology 40:1701–1708. Loeper J, Descatoire V, Letteron P, Moulis C, Degott, C, Dansette P, Fau D and Pessayre D (1994). Hepatotoxicity of germander in mice. Gastroenterolgy 106:464–472. Luyendyk JP, Maddox JF, Cosma GN, Ganey PE, Cockrell GL and Roth RA (2003). Ranitidine treatment during modest inflammatory response precipitates idiosyncrasy-like liver injury in rats. J Pharmacol Exp Ther 314:1023–1031. Luyendyk JP, Shaw PJ, Green CD, Maddox JF, Ganey PE and Roth RA (2005). Coagulation-mediated hypoxia and neutrophil-dependent hepatic injury in rats given lipopolysaccharide and ranitidine. J Pharmacol Exp Ther 307:9–16. Luyendyk JP, Lehman-McKeeman LD, Nelson DM, Bhaskaran VM, Reilly TP, Car BD, Cantor GH, Maddox JF, Ganey PE and Roth RA (2006). Unique gene expression and hepatocellular injury in the lipopolysaccharide-ranitidine drug idiosyncrasy rat model: comparison with famotidine. Toxicol Sci 90:569–585. Mathews JM, Etheridge AS and Black SR (2002). Inhibition of human cytochrome P450 activities by kava extract and kavalactones. Drug Metab Dispos 30:1153–1157. Mazokopakis E, Lazaridou S, Tzardi M, Mixaki J, Diamantis I and Ganotakis E (2004). Acutecholestatic hepatitis caused by Teucrium polium L. Phytomedicine 11:83–84. McRae CA, Agarwal K, Mutimer D and Bassendine MF (2002). Hepatitis associated with Chinese herbs. Eur J Gastroenterol Hepatol 14:559–562. Millonig G , Stadlmann S and Vogel W (2005). Herbal hepatotoxicity : acute hepatitis caused by a Noni preparation (Morinda citrifolia). Eur J Gastroenterol Hepatol 17:445–447. Nadir A, Agrawal S, King PD and Marshall JB (1996). Acute hepatitis associated with the use of a Chinese herbal product, Ma-hung. Am J Gastroenterol 91:2647–2648. Neff GW, Reddy KR, Durazo FA, Meyer D, Marrero R, and Kaplowitz N (2004). Severe hepatotoxicity associated with the use of weight loss diet supplements containing ma huang or usnic acid. J Hepatol 41:1062–1064. Nerukar PV, Klaus D and Tang C-S (2004). In vitro toxicity of kava alkaloid, pipermethystine, in HepG2 cells compared to kavalactones. Toxicol Sci 79:106–111. Novak D and Lewis JH (2003). Drug-induced liver disease. Curr Opin Gastroenterol 19:203–215. Oz HS, McClain CJ, Nagasawa HT, Ray MB, de Villiers WJ and Chen TS (2004). Diverse antioxidants protect against acetaminophen hepatotoxicity. J Biochem Mol Toxicol 18:361–368. Pageaux G.-P and Larrey D (2003). Alternative medicine, vitamins and natural hepatotoxins, in DrugInduced Liver Disease, N Kaplowitz and LD DeLeve (Eds), Marcel Dekker, New York, NY, USA, pp. 709–724. Rados C (2004). Ephedra ban: No shortage of reasons. FDA Consum 38:6–7. Sai K, Kai S, Umemura T, Tanimura A, Hasegawa R, Inoue T and Kurokawa Y (1998). Protective effects of green tea on hepatotoxicity, oxidative DNA damage and cell proliferation in the rat liver induced by repeated oral administration of 2-nitropropane. Food Chem Toxicol 36:1043–1051.


605

Seeff LB, Lindsay KL, Bacon BR, Kresina TF and Hoofnagle JH (2001). Complementary and alternative medicine in chronic liver disease. Hepatology 34:595–604. Schulze J, Raasch W and Siegers CP (2003). Toxicity of kava pyrones, drug safety and precautions – a case study. Phytomedicine 4:68–73. Schuppan D, Jia J-D, Brinkhaus B and Hahn EG (1999). Herbal products for liver diseases: a therapeutic challenge for the new millennium. Hepatology 30:1099–1104. Sheikh NM, Philen RM and Love LA (1997). Chaparral-associated hepatotoxicity. Arch Intern Med 157:913–919. Singh YN (2005). Potential for interaction of kava and St. John’s wort with drugs. J Ethnopharmacol. 100:108–113. Sparber A and Wootton JC (2001). Surveys of complementary and alternative medicine: Part II. Use of alternative and complementary cancer therapies. J Altern Complement Med 7:281–287. Stadlbauer V, Fickert P, Lackner C, Schmerlaib J, Krisper P, Trauner M and Stauber RE (2005). Hepatotoxicity of NONI juice: Report of two cases. World J Gastroenterol 11:4758–4760. Stedman C (2002). Herbal Hepatotoxicity. Sem Liver Dis 22:195–206. Stickel F, Egerer G and Seitz HK (2000). Hepatotoxicity of botanicals. Public Health Nutr 3:113–24. Stickel F, Baumuller HM, Seitz K, Vasilakis D, Seitz G, Seitz HK and Schuppan D (2003a). Hepatitis induced by Kava (Piper methysticum rhizome). J Hepatol 39:62–7 Stickel F, Poschl G, Seitz HK, Waldherr R, Hahn EG and Schuppan D (2003b). Acute hepatitis induced by greater celandine (Chelidonium majus). Scand J Gastroenterol 38:565–568. Stickel F, Patsenkerg E and Schuppan D (2005). Herbal hepatotoxicity. J Hepatol 43:901–910. Strader DB, Bacon BR, Lindsay KL, La Brecque DR, Morgan T, Wright EC, Allen J, Khokar MF, Hoofnagle JH and Seeff LB (2002). Use of complementary and alternative medicine in patients with liver disease. Am J Gastroenterol 97:2391–2397. Sztajnkrycer MD, Otten EJ, Bond GR, Lindsell CJ and Goetz RJ (2003). Mitigation of pennyroyal oil in hepatotoxicity in the mouse. Acad Emerg Med 10:1024–1028. Tafazoli S, Spehar DD and O’Brien PJ (2005). Oxidative stress mediated idiosyncratic drug toxicity. Drug Metab Rev 37:311–325. Teschke R, Gaus W and Loew D (2003). Kava extracts: Safety and risks including rare hepatotoxicity. Phytomedicine 10:440–446. Tyler VE (1993). Chaparral, in The Honest Herbal: A Sensible Guide to the Use of Herbs and Related Remedies, Third Edition, Pharmaceutical Products Press, NY, pp. 87–88. Vanderperren B, Rizzo M, Angenot L, Haulfroid V, Jadoul M and Hanston P (2005). Acute liver failure with renal impairment related to the abuse of senna anthraquinone glycosides. Ann Pharmacother 39:1353–1357. Wang YP, Fu PP and Chou MW (2005). Metabolic activation of the tumorigenic pyrrolizidine alkaloid, retrorsine, leading to DNA adduct formation in vivo. Int J Environ Res Public Health 2:74–79. Watkins PB (2003). The role of cytochrome P450s in drug-induced liver disease, in Drug-Induced Liver Disease, N Kaplowitz and LD DeLeve (Eds), Marcel Dekker, New York, NY, USA, pp. 15–33. Willett KL, Roth RA and Walker LA (2004). Workshop overview: Hepatotoxicity assessment for botanical dietary supplements. Toxicol Sci 79:4–9. Woolf GM, Petriovic LM, Rojter SE, Wainwright S, Villamil FG, Katkov WN, Michieletti P, Wanless IR, Stermitz FR, Beck JJ and Vierling JM (1994). Acute hepatitis associated with the Chinese herbal product jin bu huan. Ann Intern Med 121:729–735. Wootton JC and Sparber A (2001). Surveys of complementary and alternative medicine: Part III. Use of alternative and complementary therapies for HIV/AIDS. J Altern Complement Med 7:371–377. Yee SB, Kinser S, Hill DA, Barton CC, Hotchkiss JA, Harkema, JR, Ganey PE and Roth RA (2000). Synergistic hepatotoxicity from coexposure to bacterial endotoxin and the pyrrolizidine alkaloid monocrotaline. Toxicol Appl Pharmacol 166:173–185.

606

Hepatotoxicity

Yee SB, Ganey PE and Roth RA (2002). The role of Kupffer cells and TNF-alpha in monocrotaline and bacterial lipopolysaccharide-induced liver injury. Toxicol Sci 71:124–132. Yee SB, Hanumegowda UM, Hotchkiss JA, Ganey PE and Roth RA (2003a). Role of neutrophils in the synergistic liver injury from monocrotaline and bacterial lipopolysaccharide exposure. Toxicol Sci 72:43–56. Yee SB, Harkema JR, Ganey PE and Roth, RA (2003b). The coagulation system contributes to synergistic liver injury from exposure to monocrotaline and bacterial lipopolysaccharide. Toxicol Sci 74:457–469. Zhou S, Koh H-L, Gao Y, Gong Z-y and Lee EJD (2004). Herbal bioactivation: The good, the bad and the ugly. Life Sci 74:935–968. Zou L, Karkey MR, Henderson GL and Dike LE (2004). Kava does not display metabolic toxicity in a homogeneous cellular assay. Planta Med 70:289–292.

Section 9 Risk Analysis of Hepatotoxins

26 Physiologically Based Pharmacokinetic Modeling and Risk Assessment of Hepatotoxicants Kannan Krishnan

26.1

Introduction

Most in vivo, in vitro and genomic studies contribute to enhance the scientific basis of risk and safety assessments for chemicals and drugs. There are increasing examples of the use of such studies in improving our ability to screen for hepatotoxicants and to assess the health risks associated with human exposures (e.g. Jarabek et al., 1994; Hamadeh et al., 2002; Heinloth et al., 2004; Luyendyk et al., 2006). The health risk assessment process for hepatotoxicants, as for other xenobiotics, is conducted in four steps: (i) hazard identification, (ii) exposure assessment, (iii) dose–response assessment and (iv) risk characterization (NAS, 1981; Barnes and Dourson, 1988). For characterizing the dose–response relationship of hepatotoxicants in test animals, high doses of chemicals are frequently administered by routes often different from anticipated human exposures. In such cases, the challenge becomes one of extrapolation i.e. high dose to low dose, route-to-route and animals to humans (Clewell et al., 2002; Slikker et al., 2004). Given that hepatotoxicants are systemically acting chemicals, the various extrapolations may be conducted on the basis of equivalent internal doses (Andersen et al., 1987a). In other terms, for risk assessment purposes, the response level (e.g. benchmark dose) or the no-response level (e.g. no observed adverse effect level (NOAEL)) may be extrapolated between scenarios and species on the basis of equivalent measures of internal dose (often referred to as ‘dose metrics’). Table 26.1 lists some of the dose metrics that have been used in conducting such extrapolations for


610

Hepatotoxicity

Table 26.1 Examples of dose metrics of hepatotoxicants evaluated with PBPK models for use in risk assessment processes Chemical

Dose metric

Reference

Chloroform

Amount of metabolites covalently bound to biological macromolecules L liver per day; % cell kill/day AUC of parent chemical in liver Time-weighted average concentration in liver over lifetime Liver AUC Area under the liver, concentration vs. time curve Rate of glutathione conjugates produced/L liver/time Time-weighted receptor occupancy Up/down regulation of receptor occupancy Fraction of cells induced

Reitz et al. (1990b)

Chloropentafluorobenzene 1,4-Dioxane

Methyl chloroform Methylene chloride Tetrachlorodibenzodioxin

Trichloroethylene

Vinyl chloride

Amount metabolized/kg/day; AUC for trichloroacetic acid or dichloroacetic acid/L plasma; production of thioacetylating intermediate from dichlorovinylcysteine in kidney mg metabolized/L liver; mg metabolite produced/L liver/day

Clewell and Jarnot (1994) Leung and Paustenbach (1990) Reitz et al. (1990a) Reitz et al. (1988) Andersen et al. (1987a) Andersen et al. (1993) Portier et al. (1993) Conolly and Andersen (1997) Clewell et al. (2000); Fisher and Allen (1993)

Clewell et al. (2001); Reitz et al. (1996a)

hepatotoxicants. The identification of appropriate dose metrics for risk assessment applications is facilitated with the use of physiologically based pharmacokinetic (PBPK) models (Clewell et al., 2002). The PBPK models are mathematical constructs of the absorption, distribution, metabolism and elimination of chemicals in biota (Teorell, 1937; Bischoff et al., 1971; Bischoff, 1987; Clewell and Andersen, 1987; Andersen, 2003; Krishnan and Andersen, 2001). The mechanistic basis of these models allows them to be used confidently in the conduct of various extrapolations essential for risk assessment purposes (Clewell and Andersen, 1987; Krishnan and Andersen, 2001). This chapter provides an overview of how the PBPK models are developed and applied in the risk assessment of hepatotoxicants.

Modeling and Risk Assessment of Hepatotoxicants

611

Lungs Oral dose

GI

Fat Slowly Perfused Tissues Richly Perfused Tissues

Arterial Blood

Venous Blood

Liver

SKIN

iv dose

Chemical in air contacting skin

Figure 26.1 Conceptual representation of a physiologically based pharmacokinetic model (GI, gastrointestinal tract)

26.2

Development of PBPK Models

The PBPK models are developed on the basis of hypotheses or knowledge of mechanisms relating to absorption, distribution, metabolism and excretion. Typically, PBPK models represent the body as a set of tissue compartments (Figure 26.1). These tissues reflect the target organ, portals of entry, as well as the sites of elimination (i.e. metabolism and excretion) and accumulation (e.g. adipose tissues). The tissues are either described as individual compartments or lumped together on the basis of similarity of perfusion and tissue-partitioning characteristics (Gerlowski and Jain, 1983; Krishnan and Andersen, 2001; Reddy et al., 2005). Each compartment in the model is then described either as a heterogeneous or homogenous (well-stirred) reactor. For high-molecular-weight compounds, diffusion is often the rate-limiting process such that the computation of their tissue uptake requires that the cellular matrix and blood sub-compartments be considered and described separately. For all other compounds, the well stirred model is applied, at least to start with. In this approach, the tissue compartment is considered as being homogenous and described with a mass-balance differential equation of the following type: Rate of change = Input rate − Output rate − Rate metabolized

(26.1)

Dimensionally, the unit for each of these terms is amount per time (or concentration per unit time). Each term, in turn, is based on organ-specific clearance and concentration. For example, the input rate equals the influx clearance and the arterial blood concentration, with the influx clearance being the tissue perfusion rate. Similarly, the output rate is equal to the efflux clearance (= tissue blood flow rate) times the venous concentration (= amount

612

Hepatotoxicity

in tissue divided by the volume of distribution specific to the tissue). The volume of distribution for each tissue compartment equals the product of the tissue volume and the tissue:blood partition coefficient. As such, in order to solve the mass balance equations for non-metabolizing tissues, information on tissue perfusion rates and volume of distribution (i.e. tissue volumes and tissue:blood partition coefficients) are required. For metabolizing tissues, the information on clearance is additionally required in order to compute the rate of metabolism in liver (or another metabolizing organ). The volumetric flow of blood from which a chemical is removed per unit time can be described using one of the various hepatic clearance models (e.g. venous equilibration model, parallel tube model, distributed model and transit-time model (Johansen and Keiding, 1981; Robinson, 1992; Sirianni and Pang, 1997; Liu and Pang, 2006)). Each of these models is based on certain assumptions on physiological processes in the liver, particularly the determinants of clearance, namely the blood flow and chemical concentration across the sinusoids. The venous equilibration description is commonly used in PBPK models. Even though the different clearance descriptions may give different results for certain scenarios, if the average tissue concentration or some other less-sensitive dose surrogate (e.g. venous blood concentration of parent chemical) is the entity of interest, then the choice of the clearance models may not make very much of a difference. However, when the objective of a modeling study is to evaluate the dose to specific regions of the liver (Andersen et al., 1997a,b), then it becomes critical that the appropriate clearance model be used with proper information on the distribution of the metabolizing enzymes (Figure 26.2). Figure 26.3 presents the mass balance differential equations and algebraic expressions required for developing a two-compartmental PBPK model to simulate the pharmacokinetics of an inhaled volatile organic chemical. Here, the inhaled hepatotoxicant equilibrates with alveolar air and then enters the systemic circulation to be metabolized by the enzymes in the liver. The free chemical exiting the tissues then forms the mixed venous concentration. For obtaining simulations of dose surrogates of relevance to risk assessment, the equations shown in Figure 26.3 need to be solved. For this purpose, knowledge of various input parameters – namely physiological parameters (alveolar ventilation rate (Q p ), cardiac output (Q c ), tissue blood flow rates (Q t ) and tissue volumes (Vt )), physico-chemical parameters (blood:air (Pba ), tissue:air (Pta ) and tissue:blood (Pt ) partition coefficients) as well as metabolic clearance parameters (maximal velocity (Vmax ) and Michaelis constant (K m )) are required.

26.3

Estimating the Parameters of PBPK Models

The range of numerical values of the physiological parameters (Q p , Q c , Q t , Vt ) for various species is available in the biomedical literature (e.g. Krishnan and Andersen, 2001; Brown et al., 1997; Haddad et al., 2001a; Price et al., 2003a; Gentry et al., 2004). However, for each new hepatoxicant to be modeled, the values of its physico-chemical (Pba , Pta , Pt ) and biochemical (Vmax , K m ) parameters need to be known additionally. Metabolism parameters required for PBPK relate to the hepatic extraction ratio (E) or the Michaelis–Menten parameters (Vmax (maximal velocity) and K m (Michaelis constant)) (Johanson and Naslund, 1988; Yamaguchi et al., 1996; Krishnan and Andersen, 2001). Even


613

Blood Qb

Fat Qf

Slowly Perfused Qs

Richly Perfused Qr

13% 26%

34% Liver

20% 7%

Ql Periportal

Centrilobular

Figure 26.2 Conceptual representation of a physiologically based pharmacokinetic model: for 2,3,7,8-tetrachlorodibenzo-p-dioxin. Reprinted from Toxicol. Appl. Pharmacol., 144, M. E. Andersen, C. R. Eklund, J. J. Mills, H. A. Barton and L. S. Birnbaum, ‘A multicompartment geometric model of the liver in relation to regional induction of cytochrome P450s’, 135–144, Copyright (1997), with permission from Elsevier

though approaches using E or Vmax and K m give identical outputs (Poulin and Krishnan, 1998, 1999) for first-order conditions, the latter form describes the dose-dependent (saturable) nature of metabolism. A number of in vitro and in vivo models are available for estimating Vmax and K m for hepatic and extrahepatic metabolism of xenobiotics (Krishnan and Andersen, 2001; Brandon et al., 2003). Issues and examples of in vitro to in vivo extrapolation of metabolic constants using the physiological modeling approach may be found in Iwatsubo et al. (1996, 1997) and van Eljkeren (2002) . Care should be exercised while interpreting in vitro data on metabolism rates because metabolism under in vivo conditions may be limited by liver blood flow (Wilkinson and Shand, 1975) and the free concentration of chemical in vivo might be different compared to the in vitro system. Initial estimates of blood concentration profiles, particularly for inhaled hepatotoxicants, may be obtained by setting E values to 0 and 1 in PBPK models (Poulin and Krishnan, 1998, 1999). The predictions obtained with this animal-alternative approach should, in principle, encompass

614

Hepatotoxicity

Ca =

QpCi + QcCv Qc +

dAt dt

= Ql(Ca−Cvl) −

Qp Pba

VmaxCvl

dAt

Km+Cvl

dt

t

t

Al = ∫dAl

Ab = ∫dAb

o

Cl =

Cvl =

= Qb(Ca−Cvb)

o

Al

Cb =

Vl Cl Pl

Cvb =

Ab Vb Cb Pb

QbCvb+QlCvl Cv = Qc

Figure 26.3 Mathematical representation of a simple physiologically based pharmacokinetic model: C, Q, P and V refer to concentrations, flows, partition coefficients and volumes; a, b, b:a, c, i, l, p, v, vb and vl correspond to arterial, rest of body, blood:air, cardiac, inhaled, liver, pulmonary, mixed venous, venous blood leaving rest of body and venous blood leaving liver; dA/dt, K m and Vmax refer to the rate of change in the amount, Michaelis constant and maximal velocity for metabolism

all experimental data. The E value in conjunction with the hepatic blood flow gives an indication of the hepatic clearance, the numerical value of which should be between 0 and hepatic blood flow. Figure 26.4 presents the hepatic clearance for a number of volatile organic chemicals and compares it with the pulmonary clearance. The hepatic extraction ratio, E, for several chemicals is presented in Figure 26.5 and it can be seen that the value of E is essentially between 0 and 1. The third set of parameters, namely partition coefficients, used in PBPK models reflect the ratio of chemical concentration between two phases at equilibrium (e.g. liver and blood, liver:air, blood and air). Table 26.2 lists the liver:air partition coefficients of some volatile organic chemicals. The greater the partition coefficient, the higher the affinity of the chemical for the liver tissue. This might be one reason for a chemical to accumulate preferentially in the liver where it causes toxicity. The accumulation in liver should also be more appropriately assessed as a function of the liver:blood partition coefficients. A number of in vivo, in vitro and animal-alternative methods are available for estimating these parameters (reviewed in Krishnan and Andersen 2006). For chemicals that do not bind significantly to tissue or blood proteins, the concentration in a biological matrix is a function of the chemical concentration in its key components, namely neutral lipids and water (Falk et al., 1990). Accordingly, the following algorithm has been used to compute tissue:blood partition coefficients (Pt ) based on tissue and blood composition data (Poulin

Modeling and Risk Assessment of Hepatotoxicants 180 160

615

Respiratory Metabolic

Clearance (L /h)

140 120 100 80 60 40 20

2T 1, etra 1, 1 , 1, 1- ch 2 , T lo 2- r ic ro T h et 1, etra loro han 1, c 2- hlo eth e T r an 1, r ic h oet e 1- lo ha D r n 1, ich oet e 2- lo ha D ro n ich e e lo tha ro n Ca et e rb h on B an e te nz e tra e c ne D C hlor ic hl id hl or e or of om or et m ha Te -X ne tra y ch S lene lo tyr r o en et hy e Tr le ic ne T hl or olu oe en th e yl en e

0

1,

1,

1,

m

Substance

Figure 26.4 Hepatic and pulmonary clearance of several substances. Data from Sato and Nakajima (1987) 1.0 Hepatic extraction ratio

0.9 0.8 0.7 0.6 0.5 0.4 0.3 0.2 0.1 D ic hl or Te om tra et ch ha lo ne ro et hy le ne D io xa ne To lu en e -X yl en C e ar St bo yr n en te e tra c Et hlo rid hy e lb en ze ne C hl Tr or ic of hl or or m oe th Vi y ny len e lc hl or id e

0.0

m

Substance

Figure 26.5 Hepatic extraction ratios of some volatile organic chemicals. Data from Ramsey and Andersen (1984), Paustenbach et al. (1988), Andersen et al. (1991), Fisher et al. (1991), Allen and Fisher (1993), Tardif et al. (1993, 1995, 1997), Reitz et al. (1996a,b) and Ward et al. (1998)

616

Hepatotoxicity

Table 26.2 Experimental liver:air partition coefficients (Pt : a ) and n-octanol:water partition coefficients (Po : w ) for several chemicals (Fiserova-Bergerova, 1983, 1984; FiserovaBergerova and Diaz, 1986; Perbellini et al., 1985; Filser et al., 1993; Gearhart et al., 1993; Lapare et al., 1993; Gargas et al., 1986, 1989; Andersen et al., 1987a; Medinsky et al., 1994; Peterson and Mackay 1986). Po : w values were computed using the software KOWWIN of Syracuse Research Corporation, Inc. Experimental Pt : a

Chemical

Po : w from KOWWIN

Human

Rat

Diethyl ether 1,3-Butadiene Dichloromethane 2-Chloro-1, 1, 1-trifluoroethane Isoflurane Benzene Chloroform Methyl chloroform Toluene Trichloroethylene m-Xylene n-Heptane Tetrachloroethylene n-Hexane Cyclohexane

4.88 8.40 21.98 57.14 140.89 169.09 118.93 393.33 603.43 666.27 1690.10 2250.00 2701.27 5961.54 29 300.00

11.0 0.7 ± 0.1 7.2 ± 0.1 2.3 ± 0.4 4.1 ± 0.4 22.6 ± 4.5 17.0 16.5 ± 2.8 48.0 29.4 ± 5.1 79.7 10.8 ± 1.0 61.1 5.2 ± 0.7 10.8 ± 0.9

6.8 ± 0.54 1.19 ± 0.1 14.2 ± 1.2 1.84 ± 0.14 4.07 ± 0.2 17.0 ± 1.3 21.1 ± 1.5 8.6 ± 0.9 83.6 ± 5.8 27.2 ± 3.4 90.9 ± 4.4 15.0 ± 4.7 70.0 ± 9 5.2 ± 0.8 7.9 ± 0.59

and Krishnan, 1995a,b, 1996a,b): Pt =

Po:w (Fnlt + 0.3Fpt ) + (Fwt + 0.7Fpt ) Po:w (Fnlb + 0.3Fpb ) + (Fwb + 0.7Fpb )

(26.2)

where Po:w = n-octanol:water partition coefficient, Fnlt = volume fraction of neutral lipids in tissue, Fnlb = volume fraction of neutral lipids in blood, Fwt = volume fraction of water in tissue, Fwb = volume fraction of water in blood, Fpt = volume fraction of phospholipids in tissue and Fpb = volume fraction of phospholipids in blood. PBPK models, following parametrization, can be used to simulate the tissue dose associated with the exposure concentrations (or administered dose) in test species. Using the simulated tissue dose (or internal dose) as the basis, various extrapolations essential for risk assessment purposes have been conducted (Clewell and Andersen, 1987; Krishnan and Andersen, 2001).

26.4 26.4.1

PBPK Model-Based Extrapolations in Health Risk Assessment High-Dose to Low-Dose Extrapolation

PBPK models have been used for the conduct of high dose–low dose extrapolation of the dose–response curve on the basis of tissue dose measures. High-dose to low-dose extrapolations, particularly for carcinogens, are often conducted on the basis of linear models. These models may not be appropriate, either because the tissue dose is not proportional to


617

the administered dose across the dose range of interest, or because toxicodynamic processes are dose-dependent. Non-linearity due to saturable metabolism, enzyme induction, enzyme inactivation, depletion of glutathione reserves, etc. have been successfully described using PBPK models (e.g. Clewell and Andersen, 1987, D’Souza et al., 1988, Krishnan et al., 1992; Slikker et al., 2004). Since these models incorporate mechanism-based descriptions of the non-linear processes, predictions of pharmacokinetic behavior of chemicals from one scenario to another, as well as from high dose to low dose, are facilitated by PBPK models (Lapare et al., 1993, 1995). However, the choice of appropriate dose surrogate (e.g. area under the concentration vs. time curve in target tissue) for this process is critical, as for all other extrapolations based on PBPK models (U.S. EPA 2006). 26.4.2

Route-to-Route Extrapolation

The route-to-route extrapolation of the point of departure (e.g. NOAEL, benchmark dose (BMD)) may be essential in the case of risk assessment of certain hepatotoxicants, particularly if there is a lack of POD for the exposure route of interest. Route-to-route extrapolation of toxicologically equivalent doses may be conducted on the basis of administered dose, absorbed dose or target tissue dose (Pepelko and Withey, 1985; Pepelko, 1987). The extrapolation of the kinetic behavior and POD from one exposure route to another (e.g. oral to inhalation) can be performed by adding appropriate equations to represent each exposure pathway in the PBPK model (Figure 26.6). Once the pathways and equations describing the route-specific entry of chemicals into systemic circulation are included in the model, it is possible to conduct route-to-route extrapolation of POD on the basis of equivalent internal dose (e.g. Clewell and Andersen, 1987; Krishnan and Andersen, 2001; Sarangapani et al., Inhaled air

Expired air Lung Media

Fat Richly perfused tissues Poorly perfused tissue Liver

oral Inh derm

Arterial blood

Venous blood

Skin Computer Program

Oral ingestion

Inhaled concentration Dermal dose Oral dose

C

T

Gut Metabolism Structure

Input Parameters

Simulations

Figure 26.6 Schematic representation of the use of PBPK models in simulating the internal dose associated with specific exposure routes

618

Hepatotoxicity Validated animal PBPK model

Evaluation of animal dose metrics

Formulation of a human PBPK model Model refinement

DR model

Risk assessment

Adequate simulation

Inadequate simulation

Prediction of human dose metrics

Figure 26.7 Schematic representation of the process of using PBPK models in the conduct of interspecies extrapolation for risk assessment purposes DR, dose-response.

2003). The US EPA (2000) risk assessment for vinyl chloride, nicely illustrates the use of PBPK models in deriving the POD from one exposure to another route, on the basis of an appropriate internal dose (i.e. amount of metabolite per liver volume). 26.4.3

Interspecies Extrapolation

In the risk assessment of hepatotoxicants, as for other chemicals, the interspecies extrapolation of administered dose is performed either on the basis of body surface area (carcinogens) or using an uncertainty factor of 10 (non-carcinogens) (Dourson et al., 1996). These conventional approaches are adequate if all relevant pharmacokinetic and pharmacodynamic determinants vary as a function of body surface between test animals and humans (Kalberlah et al., 2003). This is often not the case and therefore tools such as PBPK models that take into account the qualitative and quantitative differences in the various determinants (e.g. physiology, partition coefficients, metabolism rates) are relevant (Clewell and Andersen, 1987; Krishnan and Andersen, 2001; Pelekis and Krishnan, 2004). For conducting interspecies extrapolation of toxicologically equivalent doses, PBPK models for both species are constructed on the basis of species-specific model parameter values (i.e. partition coefficients, physiological parameters and metabolic rate constants) (Figure 26.7). Some of these parameters may be species-invariant whereas others might scale from one species to another (Krishnan and Andersen, 2001). Interspecies scaling of metabolism constants in PBPK models is challenging, since they do not vary in any coherent manner between one species to another. When the human data for a particular chemical are not available for validation/calibration purposes, a corollary or parallelogram approach permitting the use of in vitro observations along with human data on similar chemicals (e.g. Barton et al., 2000). 26.4.4

Intraspecies Extrapolation

Toxicologically equivalent doses may not vary in a predictable fashion between individuals, particularly if the pharmacokinetic and pharmacodynamic determinants vary in different


Arterial blood concentration (μg/L)

2

619

6 years old

1,8

10 years old

1,6

14 years old

1,4 Adult

1,2 1 0,8 0,6 0,4 0,2 0

0

10

20 Time (h)

30

40

Figure 26.8 PBPK model simulations of the arterial blood concentrations of furan associated with inhalation exposure to 1 μg/L of this compound (during 30 h) in children and adults. Reproduced by permission of Taylor & Francis from K. Price, S. Haddad and K. Krishnan (2003), Physiological modeling of age-specific changes in the Pharmacokinetics of organic chemicals in children, J. Toxicol. Environ. Health A, 66, 417–433

directions (Calabrese, 1985; Bruckner, 2000). An intrapsecies uncertainty factor of 10 has been used to derive the safe dose of toxicants for the entire population, based on acceptable dose to the ‘typical’ human adult (Dourson et al., 1996; Barnes and Dourson, 1998; Burin and Saunders, 1999; Kalberlah et al., 2003). Various modeling approaches (Bayesian modeling, Monte Carlo simulation or subject-specific physiological models) have been proposed for the quantitation of intraspecies uncertainty factors (Portier and Kaplan, 1989; Gelman et al., 1996; El-Masri et al., 1999; Johanson et al., 1999; Pelekis et al., 2001; Lipscomb and Kedderis, 2002; Jonsson and Johanson, 2001; Nong et al., 2006). Figure 26.8 shows the blood concentration profiles of a hepatotoxicant (furan), following inhalation exposures in adults and children of various age groups, obtained using the subject-specific PBPK modeling approach (Price et al., 2003b). Such approaches allow the quantitation of the magnitude of adult–child differences in the tissue dose associated with hepatoxicants and other substances.

26.5

Examples of the Application of the PBPK Model in the Risk Assessment of Hepatotoxicants

The use of PBPK models in the risk assessment of hepatotoxicants relates to high dose to low dose, interspecies and intraspecies extrapolation of pharmacokinetics and tissue dose. Additionally, these models have been used in the conduct of route-to-route extrapolation of

620

Hepatotoxicity

PODs, in certain cases. Considering the non-cancer risk assessments for hepatotoxicants, the reference dose (or reference concentration) is established by dividing the POD (e.g. NOAEL, LOAEL, BMD) by a series of factors of 10 each to represent uncertainty associated with interspecies, intraspecies, database, susceptibility of infants, study quality and study duration (Dourson et al., 1986). Of these uncertainty factors, the PBPK models have specifically been used to quantify the magnitude of the interspecies and intraspecies uncertainty factors. When a PBPK model is used to conduct the above extrapolations, residual uncertainty (i.e. pharmacodynamic component) associated with the interspecies and intraspecies extrapolations is accounted for by using a smaller factor (3.16 each) (US EPA, 2006). In the case of risk assessments for hepatocarcinogens based on animal studies, high dose to low dose extrapolation of the dose–response relationship precedes interspecies dose adjustment for establishing risk-specific dose for humans (e.g. corresponding to a risk level of 1 in a million). The potential usefulness of PBPK models in enhancing the scientific basis of the high dose to low dose and interspecies extrapolations based on tissue dose simulations for liver tumorigens was initially demonstrated with dichloromethane (Andersen et al., 1987a, 1991; US EPA, 2006). This chemical (dichloromethane, DCM) was reported to cause liver and lung tumors in mice exposed for lifetime to 2000 or 4000 ppm for 6 h/day. The carcinogenicity of DCM is correlated with the flux through the glutathione (GSH)-pathway rather than to the flux through the oxidative (P-450) pathway (Figure 26.9). Therefore, the high dose to low dose and interspecies extrapolations of DCM liver cancer risk were conducted by Andersen et al. (1987a) using the model simulations of the tissue dose, reflective of the flux through the GSH-pathway. This PBPK-based cancer risk assessment for DCM predicted human low dose risk about 100–200 times less than that predicted by initial assessments of the US EPA (1987). Subsequently, however, both the US EPA and OSHA have altered their risk assessments for DCM on the basis of PBPK model simulations of tissue dose. Similar approaches have been demonstrated for the assessment of hepatic cancer risk associated with human exposures to chloroform and 1,4-dioxane. 1,4-Dioxane has been shown to cause hepatic tumors in rats and mice exposed to high concentrations of this chemical in drinking water (reviewed by Stickney et al., 2003). Reitz et al. (1990a) developed a PBPK model for the conduct of dose–response assessment of 1,4-dioxane. The dose surrogates evaluated were (i) the daily average area under the liver dioxane concentration vs. time curve and (ii) the daily average area under the metabolite concentration vs. time. The tumor incidence observed at high doses were correlated with dioxane iteslf, the behaviour being explained by the disproportionate increase in parent chemical concentration due to saturation of hepatic metabolism. The PBPK-based threshold approach suggested that 120 000 ppb in drinking water and 3700 ppb in air are unlikely to cause adverse effects in humans. On the other hand, the PBPK model combined with linearized multistage (LMS) model established an upper bound risk estimate of 1 × 10−5 for human exposures to dioxane at 740 ppb in air or 20 000 ppb in drinking water. Somewhat comparable results based on modeling of dose–response relationships for dioxane-induced liver tumors were reported by Leung and Paustenbach (1990). Chloroform has been shown to increase the number of hepatic and renal tumors in rats and mice receiving this chemical by corn oil gavage. The hepatotoxicity has been thought to be associated with the metabolites rather than the parent chemical itself. Reitz et al. (1990b) developed a PBPK model to estimate the risk of liver cancer in humans. Two dose surrogates were evaluated by these authors: (1) integrated amount of chloroform metabolites covalently


621

Tissue dose (GSH pathway)

Tumor incidence (%)

90 80 70 60 50 40 30 20 10 0 0

2000

4000

0

2000

4000

0

2000

4000

2000 1800 1600 1400 1200 1000 800 600 400 200 0

Tissue dose (P450 pathway)

4000 3500 3000 2500 2000 1500 1000 500 0 Exposure concentration (ppm)

Figure 26.9 Relationship between exposure concentration of dichloromethane and the tumor response as well as tissue dose in B6C3F1 mice. Based on data from Andersen et al. (1987a)

bound to macromolecules and (ii) the rate of hepatocyte kill. The PBPK model output of the rate of hepatocyte kill, rather than the administered dose of chloroform, successfully describe the dose–response relationship. Using the cell kill as the dose surrogate, the PBPK model-based approach estimated acceptable exposure concentration for humans (2840 ppm in air; 13 900 ppb in water). These numbers were four to five orders of magnitude higher than those calculated by the US EPA. A recent Health Canada assessment is based on amount metabolized, calculated with PBPK models (Meek et al., 2002).

622

Hepatotoxicity

There are other examples of evaluations of appropriate dose surrogates and conduct of risk assessments for hepatotoxicity and hepatocarcinogenicity induced by other chemicals (e.g. trichloroethylene (Clewell and Andersen, 2004), dioxins (Maruyama and Aoki, 2006), vinyl chloride (US EPA, 1995, 2000; Clewell et al., 1995). Parameters and summaries of past PBPK modeling efforts are also available for a number of other hepatotoxicants (e.g. vinyl fluoride (Cantoreggi and Keller, 1997), perchloroethylene (Clewell et al., 2005), PCBs (Emond et al., 2005), bromodichloromethane (Lilly et al., 1998)). Even though the PBPK models permit the estimation of target tissue dose of chemicals to improve the scientific basis of risk assessments, the ultimate interest is to develop integrated kinetic–dynamic models for hepatotoxicity and hepatocarcinogenicity. These kinds of integrated biologically based dose–response models might increasingly replace the use of default approaches in risk assessment. These modeling tools integrate quantitative descriptions of pharmacokinetics and pharmacodynamics, on the basis of mode of action of chemicals (Krishnan et al., 1992). In the case of chloroform, for example, hepatotoxicity was correlated with the rate of metabolism of this chemical and not to the total amount of metabolites bound to tissue macromolecules (Corley et al., 1990), suggesting the possibility that the transient and maximal concentrations of metabolites are likely to be the determinants of chloroform-induced hepatotoxicity (Reitz et al., 1990b). Accordingly, the likelihood of liver cell death following chloroform exposure has been linked to the rate of metabolism via a statistical distribution, as follows (Reitz et al., 1990b): dN /dt = −SENS × N × k(death)

(26.3)

where SENS = statistical distribution of cell sensitivity to cytotoxicity as a function of the rate of metabolites produced per unit liver volume (i.e. proportion of cells at risk any any time), N = total number of viable hepatocytes and k(death) = rate at which cells at risk die. Mode-of-action based toxicodynamic descriptions have been produced for both carbon tetrachloride-induced hepatotoxicity, and chloroform-induced hepatocarcinogenesis (Conolly et al., 1988; Paustenbach et al., 1988). The cancer model is essentially a 2-stage model (Moolgavkar–Knudson–Venzon (MVK) model) that specifies two mutational events along with the birth and death rate of cell populations (Moolgavkar and Knudson, 1981; Moolgavkar and Venzon, 1979). Such models have been applied to study the hepatic cancer incidence associated with exposure to dioxins and diethylnitrosmaine. (Paustenbach et al., 1991; Travis et al., 1991). Thse kinds of integrated kinetic–dynamic models allow unique sensitivity analyses to identify exposure scenarios and dose levels that might affect the events that are critical to cancer induction and further evaluate whether they are of relvance to human exposure situations (Andersen et al., 1992). In this regard, the availability of improved quantitative descriptions of physiological processes influenced by hepatotoxicants would be valuable (e.g. Glicklis et al., 2004).

26.6

Physiological Modeling and Risk Assessment of Hepatotoxicant Mixtures

Mixed exposures to chemicals can result in an increase, decrease or no change in the toxicity compared to single exposures (Krishnan and Brodeur, 1991, 1994). A number of studies have shown that the hepatotoxicity of chemicals may be modified during combined


623

exposures, due to pharmacokinetic or pharmacodynamic interactions. Metabolic interactions have often been reported to be the mechanism of interaction, leading to synergistic, potentiating or antagonistic hepatotoxic interactions. Initial observations of potentiation of hepatotoxicity were based on combined exposure studies involving alcohols, ketones and carbon tetrachloride (Plaa, 1981; Plaa and Hewitt, 1982; Plaa et al., 1982). Several other studies of hepatotoxic interactions among environmental agents have also appeared (e.g. Reynolds et al., 1975; Krishnan and Brodeur, 1991; Charbonneau et al., 1997; Da Silva et al., 2000; St Pierre et al., 2003). A major challenge regarding these mixed exposure studies relates to the ability to address the question of threshold of interactions. Due to limited resources, not all combinations of chemicals are evaluated in the mixture studies. Therefore, the PBPK model has been used as a tool to study binary interactions and determine the threshold of interactions. Such modeling studies of some hepatotoxic mixtures have appeared (e.g. carbon tetrachloride + kepone (El-Masri et al., 1996), bromodichloromethane and mirex, Phenobarbital or chlordecone (Thakore et al., 1991), ethanol and trichloroethylene (Sato et al., 1991)). With respect to modeling of binary chemical interactions, the approach involves the construction of PBPK models for two single chemicals, and their linkage via interaction mechanisms described at the level of a tissue or blood compartment. This is accomplished by modifying the numerical value(s) of mechanistic determinant(s) in the mathematical expressions of absorption, distribution, metabolism or excretion (Andersen et al., 1987a; Krishnan et al., 1994). Metabolic interactions occur when one chemical competes directly with another chemical for an enzymatic binding site (competitive inhibition), when a chemical binds directly to the enzyme–substrate complex but not to the free enzyme (uncompetitive), or when it does both of these (non-competitive). The inhibitory effect of one chemical on another is modeled by including a term that describes the quantitative manner in which the Michaelis affinity constant for metabolism (K m ) and/or the maximal velocity for metabolism (Vmax ) are modified (Table 26.3). High dose to low dose extrapolations of pharmacokinetic interactions is feasible with PBPK modeling because the mathematical descriptions employed in these models account for the non-linear kinetic behavior of the interacting chemicals (Andersen et al., 1987b; Tardif et al., 1995; Pelekis and Krishnan, 1997). The ability to conduct high dose to low dose extrapolation of pharmacokinetic interactions using PBPK models may be examined with metabolic induction/inhibition as the mechanism. In such cases, the binary pharmacokinetic interaction model accounts for the non-linearity arising from two phenomena: Table 26.3 Mathematical representation of the modifications of maximal velocity for metabolism (Vmax ) and Michaelis affinity constant (K m ) during metabolic inhibitiona Mechanism of inhibition Competitive Uncompetitive Noncompetitive a

Modification of K m

Modification of Vmax

Increase Decrease None

None Decrease Decrease

The magnitude of change (decrease/increase) is related to (1 + Cvli /K i ), where Cvli represents the free concentration of inhibitor in hepatic venous blood and K i represents the inhibition constant.

624

Hepatotoxicity

(i) the saturable nature of the metabolism of individual chemicals, and (ii) the relative importance of the metabolic interaction mechanism as a function of substrate concentration. Since both the saturable nature of the metabolism of mixture components, and the quantitative mechanism of the interactive effects are incorporated within the PBPK models, these models are useful for conducting high dose to low dose extrapolations of the consequence of pharmacokinetic interactions. Application of PBPK models along these lines would facilitate a priori characterization of the threshold of interactions in humans, and identification of those binary interactions that may be of concern to humans exposed to low concentrations, by various routes and scenarios (Kanamitsu et al., 2000). However, chemicals do not just co-exist in two but in multiple numbers, and as such the toxicity of two interacting chemicals might further be altered by other components of the mixture. The toxicity of complex mixtures then is determined by the outcome of interactions not only at the binary level, but also at other higher levels (e.g. ternary, quaternary). In this regard, PBPK models have been shown to be uniquely useful in conducting extrapolations of the change in tissue dose during mixed exposures (Tardif et al., 1997). The approach involves identifying and linking all individual chemical PBPK models via interaction terms. PBPK models for mixtures of any level of complexity can then be created as long as the quantitative information on the mechanism for each interacting chemical pair is available or can be hypothesized. According to this methodology then, for modeling the kinetics of the components of complex mixtures, plausible binary interactions need only be characterized. In a mixture of three chemicals, for example, there are three two-way interactions. The first step here is to write the models for each component of the mixture. Then, the single chemical models should be interconnected at the binary level by modifying the appropriate equations. If we consider competitive metabolic inhibition as the mechanism of interaction, then the equation for calculating the rate of the amount metabolized of each component should be modified appropriately (Haddad et al., 1999, 2000). Logically, this PBPK modeling approach should be applicable to predict higher-order interactions in mixtures of any complexity. It is important to note that all linkages involving mixture components are of binary nature only (Figure 26.10). If we consider interactions at the binary level alone, then how is it possible to simulate the consequence of a higher-order interaction (e.g. involving three chemicals)? This is where the unique usefulness of PBPK modeling becomes evident. Let’s assume that the binary chemical interaction between A and B has been modeled. Then, following the addition of another chemical, C, the PBPK model not only simulates the binary interactions involving C (i.e. C–A, C–B), but also the modulatory effect of C on the interaction between A and B (Figure 26.10). Once we describe the inhibitory effect of C on B, this would result in a reduction in the rate of B metabolized and consequently an increase in its concentration in venous blood leaving the liver (CB ). CB is the numerator of the term representing the inhibitory effect of B on A (i.e. 1 + CB /K mB ) (Table 26.3). Since the exposure to chemical C increases C B , this then translates into a modification of the magnitude of the interactive effect of B on A. Similarly, C may also affect the concentration of A, which would then result in a change in the magnitude of the interactive effect of A on B. The PBPK model framework can also simulate similar phenomena affecting the concentration of C, since all components of the mixture are linked. Based on this analogy, it will be possible to predict the influence of the addition of another chemical D to the ternary mixture, and so forth. When a fourth chemical, D, is added to an existing ternary mixture PBPK model of chemicals A, B and C,


625

T X B

D E Figure 26.10 Illustration of the linkage of individual chemical models through binary level connections, using a physiologically based pharmacokinetic model framework for mixtures (B, benzene; T, toluene; E, ethyl benzene; X, m-xylene; D, dichloromethane)

we only need to consider three binary interactions additionally (i.e. D–A, D–B, D–C). By doing this, the modulating effect of D on the C–A and B–A interactions will be automatically simulated since all components are linked with each other within the PBPK framework. The effect of D on the kinetics of A will in turn affect the kinetics of B, C and D. Any modulation of a binary interaction will affect the kinetics of other chemicals that are part of the network of binary interactions present in the mixture. The same considerations are applicable while another chemical, E, is added to the quaternary mixture. After adding the four new binary interactions (E–A, E–B, E–C, E–D), chemical E will become an integral part of the network of the components of the mixture and any modulation of a binary interaction involving E will have repercussions on all the others (Figure 26.10). The novel thing about this approach is that it only requires data on binary interaction mechanisms for predicting the magnitude and consequence of multiple interactions within complex mixtures (Tardif et al., 1997). The magnitude of the modulation of interactions invoked upon the addition of another chemical to an existing ‘network’ of binary interactions depends on its inhibition potency, and also on its free concentration (Cvl ). With increasing complexity of mixtures, the K i for binary interactions is not modified; rather the Cvl is increased according to the potency and number of the inhibitors. The increasing effective concentration of chemicals in a mixture is due to a cascade of inhibitory events at the binary level. The use of these kinds of mechanism-based physiological models should be useful for evaluating the impact of increasing number of inhibitors, on the metabolism and risk associated with hepatotoxicants at specific exposure concentrations (Haddad et al., 2001a)

26.7

Concluding Remarks

Physiologically based modeling involves the development of mathematical descriptions of the inter-relationships among critical parameters and processes determining the uptake, disposition and toxicodynamics of chemicals. Such models have been developed for several hepatotoxicants and applied to enhance the scientific basis of the risk assessments for such

626

Hepatotoxicity

chemicals. The uncertainties associated with the high dose to low dose, species to species and route-to-route extrapolations can be reduced with the use of PBPK models. These models are also particularly useful for integration of various in vitro and in vivo observations, identification of critical data gaps, and estimation of risk numbers, along with attendant appreciation of areas of significant biological uncertainty, to facilitate focused use of limited resources.

References Allen, B. C. and Fisher, J. W. (1993). Pharmacokinetic modeling of trichloroethylene and trichloroacetic acid in humans. Risk Anal. 13, 71–86. Andersen, M. E. (2003). Toxicokinetic modeling and its applications in chemical risk assessment. Toxicol. Lett. 138, 9–27. Andersen, M. E., Clewell, H. J. III, Gargas, M. L., Smith, F. A. and Reitz, R. H. (1987a). Physiologically based pharmacokinetics and the risk assessment process for methylene chloride. Toxicol. Appl. Pharmacol. 87, 185–205. Andersen, M. E., Gargas, M. L., Clewell, H. J. III and Severyn, K. M. (1987b). Quantitative evaluation of the metabolic interactions between trichloroethylene and 1,1-dichloroethylene in vivo using gas uptake methods. Toxicol. Appl. Pharmacol. 89, 149–157. Andersen, M. E., Clewell, H. J. III, Gargas, M. L., MacNaughton, M. G., Reitz, R. H., Nolan, R. J. and McKenna, M. J. (1991). Physiologically based pharmacokinetic modeling with dichloromethane, its metabolite, carbon monoxide, and blood carboxyhemoglobin in rats and humans. Toxicol. Appl. Pharmacol. 108, 14–27. Andersen, M. E., Krishnan, K., Conolly, R. B. and McClellan, R. O. (1992). Biologically based modeling in toxicology research. Arch. Toxicol. 15(Suppl.), 217–227. Andersen, M. E., Mills, J. J., Gargas, M. L., Kedderis, L., Birnbaum, L. S., Neubert, D. and Greenlee, W. F. (1993). Modeling receptor-mediated processes with dioxin: implications for pharmacokinetics and risk assessment. Risk Anal. 13, 25–36. Andersen, M. E., Eklund, C. R., Mills, J. J., Barton, H. A. and Birnbaum, L. S. (1997a). A multicompartment geometric model of the liver in relation to regional induction of cytochrome P450s. Toxicol. Appl. Pharmacol. 144, 135–144. Andersen, M. E., Birnbaum, L. S., Barton, H. A. and Eklund, C. R. (1997b). Regional hepatic CYP1A1 and CYP1A2 induction with 2,3,7,8-tetrachlorodibenzo- p-dioxin evaluated with a multicompartment geometric model of hepatic zonation. Toxicol. Appl. Pharmacol. 144, 145–155. Barnes, D. G. and Dourson, M. (1988). Reference dose (RfD): description and use in health risk assessments. Regul. Toxicol. Pharmacol. 8, 471–486. Barton, H. A., Deisinger, P. J., English, J. C., Gearhart, J. N., Faber, W. D., Tyler, T. R., Banton, M. I., Teeguarden, J. and Andersen, M. E. (2000). Family approach for estimating reference concentrations/doses for series of related organic chemicals. Toxicol. Sci. 54, 251–261. Bischoff, K. B. (1987). Physiogically based pharmacokinetic modeling. Drink. Water Health 8, 36–64. Bischoff, K. B., Dedrick, R. L., Zakharo, D. S. and Longstreth, J. A. (1971). Methotrexate pharmacokinetics. J. Pharm. Sci. 60, 1128–1133. Brandon, E. F., Raap, C. D., Meijerman, I., Beijnen, J. H. and Schellens, J. H. (2003). An update on in vitro test methods in human hepatic drug biotransformation research: pros and cons. Toxicol. Appl. Pharmacol. 189, 233–246. Brown, R. P., Delp, M. D., Lindstedt, S. L., Rhomberg, L. R. and Beliles, R. P. (1997). Physiological parameter values for physiologically based pharmacokinetic models. Toxicol. Ind. Health 13, 407– 484.


627

Bruckner, J. V. (2000). Differences in sensitivity of children and adults to chemical toxicity: the NAS panel report. Regul. Toxicol. Pharmacol. 31, 280–285. Burin, G. J. and Saunders, D. R. (1999). Addressing human variability in risk assessment – the robustness of the intraspecies uncertainty factor. Regul. Toxicol. Pharmacol. 30, 209–216. Calabrese, E. J. (1985). Uncertainty factors and interindividual variation. Regul. Toxicol. Pharmacol. 5, 190–196. Cantoreggi, S. and Keller, D. A. (1997). Pharmacokinetics and metabolism of vinyl fluoride in vivo and in vitro. Toxicol. Appl. Pharmacol. 143, 130–139. Charbonneau, M., Plaa, G. L. and Krishnan, K. (1997). Patterns of liver injury induced by mixtures of halogenated hydrocarbons: a predictable event? Drug Metab. Rev. 29, 137–151. Clewell, H. J. III and Andersen, M. E. (1987). Dose Species and Route Extrapolations Using Physiologically Based Pharmacokinetic Models, National Academy Press, Washington, DC, USA. Clewell, H. J. and Andersen, M. E. (2004). Applying mode-of-action and pharmacokinetic considerations in contemporary cancer risk assessments: an example with trichloroethylene. Crit. Rev. Toxicol. 34, 385–445. Clewell, H. J. III and Jarnot, B. M. (1994). Incorporation of pharmacokinetics in noncancer risk assessment: example with chloropentafluorobenzene. Risk Anal. 14, 265–276. Clewell, H. J., Gentry, P. R., Gearhart, J. M., Allen, B. C. and Andersen, M. E. (1995). Considering pharmacokinetic and mechanistic information in cancer risk assessments for environmental contaminants: examples with vinyl chloride and trichloroethylene. Chemosphere 31, 2561– 2578. Clewell, H. J. III, Gentry, P. R., Covington, T. R. and Gearhart, J. M. (2000). Development of a physiologically based pharmacokinetic model of trichloroethylene and its metabolites for use in risk assessment. Environ. Health Perspect. 108(Suppl. 2), 283–305. Clewell, H. J., Gentry, P. R., Gearhart, J. M., Allen, B. C. and Andersen, M. E. (2001). Comparison of cancer risk estimates for vinyl chloride using animal and human data with a PBPK model. Sci. Total Environ. 274, 37–66. Clewell, H. J. III, Andersen, M. E. and Barton, H. A. (2002). A consistent approach for the application of pharmacokinetic modeling in cancer and noncancer risk assessment. Environ. Health Perspect. 110, 85–93. Clewell, H. J., Gentry, P. R., Kester, J. E. and Andersen, M. E. (2005). Evaluation of physiologically based pharmacokinetic models in risk assessment: an example with perchloroethylene. Crit. Rev. Toxicol. 35, 413–433. Conolly, R. B. and Andersen, M. E. (1997). Hepatic foci in rats after diethylnitrosamine initiation and 2,3,7,8-tetrachlorodibenzo-p-dioxin promotion: evaluation of a quantitative two-cell model and of CYP 1A1/1A2 as a dosimeter. Toxicol. Appl. Pharmacol. 146, 281–293. Conolly, R. B., Reitz, R. H., Clewell, H. J. III and Andersen, M. E. (1988). Pharmacokinetics, biochemical mechanism and mutation accumulation: a comprehensive model of chemical carcinogenesis. Toxicol. Lett. 43, 189–200. Corley, R. A., Mendrala, A. L., Smith, F. A., Staats, D. A., Gargas, M. L., Conolly, R. B., Andersen, M. E., and Reitz, R. H. (1990). Development of a physiologically based pharmacokinetic model for chloroform. Toxicol. Appl. Pharmacol. 103, 512–527. Da Silva, M. L., Charest-Tardif, G., Krishnan, K. and Tardif, R. (2000). Evaluation of the pharmacokinetic interactions between orally administered trihalomethanes in the rat. J. Toxicol. Environ. Health A 60, 343–353. Dourson, M. L., Hertzberg, R. C. and Stara, J. F. (1986). A new method for determining allowable daily intakes. Fund. Appl. Toxicol. 6, 182–184. Dourson, M. L., Felter, S. P. and Robinson, D. (1996). Evolution of science-based uncertainty factors in noncancer risk assessment. Regul. Toxicol.Pharmacol. 24, 108–120.

628

Hepatotoxicity

D’Souza, R. W., Francis, W. R. and Andersen, M. E. (1988). Physiological model for tissue glutathione depletion and increased resynthesis after ethylene dichloride exposure. J. Pharmacol. Exp. Ther. 245, 563–568. El-Masri, H. A., Thomas, R. S., Sabados, G. R., Phillips, J. K., Constan, A. A., Benjamin, S. A., Andersen, M. E., Mehendale, H. M. and Yang, R. S. (1996). Physiologically based pharmacokinetic/pharmacodynamic modeling of the toxicologic interaction between carbon tetrachloride and Kepone. Arch. Toxicol. 70, 704–713. El-Masri, H. A., Bell, D. A. and Portier, C. J. (1999). Effects of glutathione transferase theta polymorphism on the risk estimates of dichloromethane to humans. Toxicol. Appl. Pharmacol. 158, 221–230. Emond, C., Charbonneau, M. and Krishnan, K. (2005). Physiologically based modeling of the accumulation in plasma and tissue lipids of a mixture of PCB congeners in female Sprague–Dawley rats. J. Toxicol. Environ. Health A 68, 1393–1412. Falk, A., Gullstrand, E., Löf, A. and Wigaeus-Hjelm, E. (1990). Liquid/air partition coefficients of four terpenes. Br. J. Ind. Med. 47, 62–64. Filser, J. G., Johanson, G., Kessler, W., Kreuzer, P. E., Stei, P., Baur, C. and Csanady, A. (1993). A pharmacokinetic model to describe toxicokinetic interactions between 1,3-butadiene and styrene in rats: predictions for human exposure. IARC Sci. Publ. 127, 65–78. Fiserova-Bergerova, V. (1983). Modeling of Inhalation Exposure to Vapors. Uptake, Distribution and Elimination, CRC Press, Boca Raton, FL, USA. Fiserova-Bergerova, V. (1984). Inhibitory effect of isoflurane upon oxidative metabolism of halothane. Anesth. Analg. 63, 399–404. Fiserova-Bergerova, V. and Diaz, M. L. (1986). Determination and prediction of tissue–gas partition coefficients. Int. Arch. Occup. Environ. Health 58, 75–87. Fisher, J. W. and Allen, B. C. (1993). Evaluating the risk of liver cancer in humans exposed to trichloroethylene using physiological models. Risk Anal. 13, 87–95. Fisher, J. W., Gargas, M. L., Allen, B. C. and Andersen, M. E. (1991). Physiologically based pharmacokinetic modeling with trichloroethylene and its metabolite trichloroacetic acid in the rat and mouse. Toxicol. Appl. Pharmacol. 109, 183–195. Gargas, M. L., Clewell, H. J. and Andersen, M. E. (1986). Metabolism of inhaled dihalomethanes in vivo: Differentiation of kinetic constants for two independent pathways. Toxicol. Appl. Pharmacol. 82, 211–223. Gargas, M. L., Burgess, R. J., Voisard, D. E., Cason, G. H. and Andersen, M. E. (1989). Partition coefficients of low molecular weight volatile chemicals in various liquids and tissues. Toxicol. Appl. Pharmacol. 98, 87–99. Gearhart, J. M., Seckel, C., and Vinegar, A. (1993). In vivo metabolism of chloroform in B6C3F1 mice determined by the method of gas uptake: the effects of body temperature on tissue partition coefficients and metabolism. Toxicol. Appl. Pharmacol. 119, 258–266. Gelman, A., Bois, F. and Jiang, J. (1996). Physiological pharmacokinetic analysis using population modeling and informative prior distributions. J. Am. Stat. Assoc. 91, 1400–1412. Gentry, P. R., Haber, L, T., McDonald, T. B., Zhao, Q., Covington, T., Nance, P., Clewell, H.J. III. and Lipscomb, J. C. (2004). Data for physiologically based pharmacokinetic modeling in neonatal animals: Physiological parameters in mice and Sprague–Dawley rats. J. Child Health 2, 363–411. Gerlowski, L. E. and Jain, R. K. (1983). Physiologically based pharmacokinetic modeling principles and applications. J. Pharm. Sci. 72, 1103–1127. Glicklis, R., Merchuk, J. C., and Cohen, S. (2004). Modeling mass transfer in hepatocyte spheroids via cell viability, spheroid size and hepatocellular functions. Biotechnol. Bioeng. 86, 672–680. Haddad, S., Tardif, R., Charest-Tardif, R. and Krishnan, K. (1999). Physiological modeling of pharmacokinetic interactions in a quaternary mixture of hydrocarbons. Toxicol. Appl. Pharmacol. 161, 249–257.


629

Haddad, S., Charest-Tardif, G., Tardif, R. and Krishnan, K. (2000). Validation of a physiological modeling framework for simulating toxicokinetic interactions in chemical mixtures. Toxicol. Appl. Pharmacol. 167, 199–209. Haddad, S., Restieri, C. and Krishnan, K. (2001b). Characterization of age-related changes in body weight and organ weights from birth to adolescence in humans. J. Toxicol. Environ. Health A 64, 453–464. Hamadeh, H. K., Knight, B. L., Haugen, A. C., Sieber, S., Amin, R. P., Bushel, P. R., Stoll, R., Blanchard, K., Jayadev, S., Tennant, R. W., Cunningham, M. L., Afshari, C. A. and Paules, R. S. (2002). Methapyrilene toxicity: anchorage of pathologic observations to gene expression alterations. Toxicol. Pathol. 30, 470–482. Heinloth, A. N., Irwin, R. D., Boorman, G. A., Nettesheim, P., Fannin, R. D., Sieber, S. O., Snell, M. L., Tucker, C. J., Li, L., Travlos, G. S., Vansant, G., Blackshear, P. E., Tennant, R. W., Cunningham, M. L. and Paules, R. S. (2004). Gene expression profiling of rat livers reveals indicators of potential adverse effects. Toxicol. Sci. 80, 193–202. Iwatsubo, T., Hirota, N., Ooie, T., Suzuki, H. and Sugiyama, Y. (1996). Prediction of in vivo drug disposition from in vitro data based on physiological pharmacokinetics. Biopharm. Drug Dispos. 17, 273–310. Iwatsubo, T., Hirota, N., Ooie, T., Suzuki, H., Shimada, N., Chiba, K., Ishizaki, T., Green, C. E., Tyson, C. A. and Sugiyama, Y. (1997). Prediction of in vivo drug metabolism in the human liver from in vitro metabolism data. Pharmacol. Ther. 73, 147–171. Jarabek, A. M., Fisher, J. W., Rubenstein, R., Lipscomb, J. C., Williams, R. J., Vinegar, A. and McDougal, J. N. (1994). Mechanistic insights aid the search for CFC substitutes: risk assessment of HCFC-123 as an example. Risk Anal. 14, 231–250. Johansen, S. and Keiding, S. (1981). A family of models for the elimination of substrate in the liver. J. Theor. Biol. 89, 549–556. Johanson, G. and Naslund, P. H. (1988). Spreadsheet programming: A new approach in physiologically based modeling of solvent toxicokinetics. Toxicol. Lett. 41, 115–127. Johanson, G., Jonsson, F. and Bois, F. (1999). Development of new technique for risk assessment using physiologically based pharmacokinetic models. Am. J. Ind. Med. 36(Suppl. 1), 101– 103. Jonsson, F. and Johanson, G. (2001). A Bayesian analysis of the influence of GSTT1 polymorphism on the cancer risk estimate for dichloromethane. Toxicol. Appl. Pharmacol. 174, 99–112. Kalberlah, F., Schneider, K. and Schuhmacher-Wolz, U. (2003). Uncertainty in toxicological risk assessment for non-carcinogenic health effects. Regul. Toxicol. Pharmacol. 37, 92–104. Kanamitsu, S., Ito, K. and Sugiyama, Y. (2000). Quantitative prediction of in vivo drug-drug interactions from in vitro data based on physiological pharmacokinetics: use of maximum unbound concentration of inhibitor at the inlet to the liver. Pharm. Res. 17, 336–343. Krishnan, K. and Andersen, M. E. (2001). Physiologically based phamacokinetic modeling in toxicology, in Principles and Methods in Toxicology, W. Hayes (Ed.), Taylor & Francis, New York, NY, USA, pp. 193–241. Krishnan, K. and Andersen, M. E. (2007). Physiologically based pharmacokinetic and toxicokinetic modeling, In Principles and Methods of Toxicology, A Wallace Hayes (ed), CRC Press, Philadelphia, PA, USA, in press. Krishnan, K. and Brodeur, J. (1991). Toxicological consequences of combined exposure to environmental pollutants. Arch. Complex Environ. Stud. 3, 1–106. Krishnan, K. and Brodeur, J. (1994). Toxic interactions among environmental pollutants: corroborating laboratory observations with human experience. Environ. Health Perspect. 102(Suppl 9), 11–17. Krishnan, K., Gargas, M. L., Fennell, T. R. and Andersen, M. E. (1992): A physiologically based description of ethylene oxide dosimetry in the rat. Toxicol. Ind. Health 8, 121–140.

630

Hepatotoxicity

Krishnan, K., Clewell, H. J. III, Andersen, M. E. and Yang, R. S. H. (1994). Physiologically based pharmacokinetic modeling of chemical mixtures’, in: Toxicology of Complex Mixtures, R. S. H. Yang (Ed.), Academic Press, New York, NY, USA, pp. 399–437. Lapare, S., Tardif, R. and Brodeur, J. (1993). Effect of various exposure scenarios on the biological monitoring of organic solvents in alveolar air. I. Toluene and m-xylene. Int. Arch. Occup. Environ. Health 64, 569–580. Lapare, S., Tardif, R. and Brodeur, J. (1995). Effect of various exposure scenarios on the biological monitoring of organic solvents in alveolar air. II. 1,1,1-Trichloroethane and trichloroethylene. Int. Arch. Occup. Environ. Health 67, 375–394. Leung, H. W. and Paustenbach, D. J. (1990). Cancer risk assessment for dioxane based upon a physiologically based pharmacokinetic approach. Toxicol. Lett. 51, 147–162. Lilly, P. D., Andersen, M. E., Ross, T. M. and Pegram, R. A. (1998). A physiologically based pharmacokinetic description of the oral uptake tissue dosimetry and rates of metabolism of bromodichloromethane in the male rat. Toxicol. Appl. Pharmacol. 150, 205–217. Lipscomb, J. C. and Kedderis, G. L. (2002). Incorporating human interindividual biotransformation variance in health risk assessment. Sci. Total Environ. 288, 13–21. Liu, L. and Pang, K. S. (2006). An integrated approach to model hepatic drug clearance. Eur. J. Pharm. Sci. 29, 215–230. Luyendyk, J. P., Lehman-McKeeman, L. D., Nelson, D. M., Bhaskaran, V. M., Reilly, T. P., Car, B. D., Cantor, G. H., Maddox, J. F., Ganey, P. E. and Roth, R. A. (2006). Unique gene expression and hepatocellular injury in the lipopolysaccharide–ranitidine drug idiosyncrasy rat model: comparison with famotidine. Toxicol. Sci. 90, 569–585. Maruyama, W. and Aoki, Y. (2006). Estimated cancer risk of dioxins to humans using a bioassay and physiologically based pharmacokinetic model. Toxicol. Appl. Pharmacol. 214, 188–198. Medinsky, M. A., Leavens, T. L., Csanady, G. A., Gargas, M. L. and Bond, J. A. (1994). In vivo metabolism of butadiene by mice and rats: a comparison of physiological model predictions and experimental data. Carcinogenesis 15, 1329–1340. Meek, M. E., Beauchamp, R., Long, G., Moir, D., Turner, L. and Walker, M. (2002). Chloroform: exposure estimation, hazard characterization, and exposure–response analysis. J. Toxicol. Environ. Health B Crit. Rev. 5, 283–334. Moolgavkar, S. H. and Knudson, A. G. Jr (1981). Mutation and cancer: a model for human carcinogenesis. J. Natl. Cancer Inst. 66, 1037–1052. Moolgavkar, S. H. and Venzon, D. J. (1979). Two-event models for carcinogenesis: incidence curves for childhood and adult tumors. Math. Biosci. 47, 55–77. National Academy of Sciences (NAS). 1981. Risk Assessment in the Federal Government: Managing the Process. National Research Council. National Academy Press, Washington, D.C. Nong, A., McCarver, D. G., Hines, R. N. and Krishnan, K. (2006). Modeling interchild differences in pharmacokinetics on the basis of subject-specific data on physiology and hepatic CYP2E1 levels: a case study with toluene. Toxicol. Appl. Pharmacol. 214, 78–87. Paterson, S. and Mackay, D. (1986). A pharmacokinetic model of styrene inhalation with the fugacity approach. Toxicol. Appl. Pharmacol. 82, 444–453. Paustenbach, D. J., Clewell, H. J. III, Gargas, M. L. and Andersen, M. E. (1988). A physiologically based pharmacokinetic model for inhaled carbon tetrachloride. Toxicol. Appl. Pharmacol. 96, 191–211. Paustenbach, D. J., Layard, M. W., Wenning, R. J. and Keenan, R. E. (1991). Risk assessment of 2,3,7,8-TCDD using a biologically based cancer model: a reevaluation of the Kociba et al., bioassay using 1978 and 1990 histopathology criteria. J. Toxicol. Environ. Health 34, 11–26. Pelekis, M. and Krishnan, K. (1997). Assessing the relevance of rodent data on chemical interactions for health risk assessment purposes: a case study with dichloromethane–toluene mixtures. Regul. Toxicol. Pharmacol. 25, 79–86.


631

Pelekis, M. and Krishnan, K. (2004). Magnitude and mechanistic determinants of the interspecies toxicokinetic uncertainty factor for organic chemicals. Regul. Toxicol. Pharmacol. 40, 264–271. Pelekis, M., Gephart, L. A. and Lerman, S. E. (2001). Physiological-model-based derivation of the adult and child pharmacokinetic intraspecies uncertainty factors for volatile organic compounds. Regul. Toxicol. Pharmacol. 33, 12–20. Pepelko, W. E. (1987). Feasibility of route extrapolation in risk assessment. Br. J. Ind. Med. 44, 649–651. Pepelko, W. E. and Withey, J. R. (1985). Methods for route-to-route extrapolation of dose. Toxicol. Ind. Health 1, 153–170. Perbellini, L., Brugnone, F., Caretta, D. and Maranelli, G. (1985). Partition coefficients of some industrial aliphatic hydrocarbons (C5–C7) in blood and humans tissues. Br. J. Ind. Med. 42, 162– 167. Plaa, G. L. (1981). Potentiation of haloalkane-induced hepatotoxicity and nephrotoxicity: role of biotransformation, in Industrial and Environmental Xenobiotics, I. Gut, M. Cikrt and G. L. Plaa (Eds), Springer-Verlag, New York, NY, USA, pp. 96–110. Plaa, G. L. and Hewitt, W. R. (1982). Potentiation of liver and kidney injury by ketones and ketogenic substances, in Advances in Pharmacology and Therapeutics II, H. Yoshida, Y. Hagihara and S. Ebashi (Eds), Pergamon Press, New York, NY, USA, pp. 65–75. Plaa, G. L., Hewitt, W. R. du, Souich, P., Caille, G. and Lock, S. (1982). Isopropanol and acetone potentiation of carbon tetrachloride-induced hepatotoxicity: single versus repetitive pretreatments in rats. J. Toxicol. Environ. Health 9, 235–250. Portier, C. J. and Kaplan, N. L. (1989). Variability of safe estimated when using complicated models of carcinogenic processes. A dose study: methylene chloride. Fundam. Appl. Toxicol. 13, 533– 544. Portier, C., Tritscher, A., Kohn, M., Sewall, C., Clark, G., Edler, L., Hoel, D. and Lucier, G. (1993). Ligand/receptor binding for 2,3,7,8-TCDD: implications for risk assessment. Fundam. Appl. Toxicol. 20, 48–56. Poulin, P. and Krishnan, K. (1995a). An algorithm for predicting tissue:blood partition coefficients from n-octanol:water partition coefficient data. J. Toxicol. Environ. Health 46, 101–113. Poulin, P. and Krishnan, K. (1995b) A biologically-based algorithm for predicting human tissue:blood partition coefficients of organic chemicals. Human Exp. Toxicol. 14, 273–290. Poulin, P. and Krishnan, K. (1996a). A mechanistic algorithm for predicting blood:air partition coefficients of organic chemicals with the consideration of reversible binding in hemoglobin. Toxicol. Appl. Pharmacol. 136: 131–137. Poulin, P. and Krishnan, K. (1996b). A tissue composition based algorithm for predicting tissue:air partition coefficients of organic chemicals. Toxicol. Appl. Pharmacol. 136: 126–130. Poulin, P. and Krishnan, K. (1998). A quantitative structure toxicokinetic relationship model for highly metabolized chemicals ATLA 26, 45–90. Poulin and Krishnan (1999). Molecular structure based prediction of the toxicokinetics of inhaled vapors in humans. Int. J. Toxicol. 18, 7–18. Poulin, P., Beliveau, M. and Krishnan, K. (1999). Mechanistic animal-replacement approaches for predicting pharmacokinetics of organic chemicals, in: Animal Alternatives in Toxicology: Recent Advances, H. Salem and S. A. Katz (Eds), Humana Press, Totowa, NY, USA, pp. 115–139. Price, P. S., Conolly, R. B., Chaisson, C. F., Gross, E. A., Young, J. S., Mathis, E. T. and Tedder, D. R. (2003a). Modeling interindividual variation in physiological factors used in PBPK models of humans. Crit. Rev. Toxicol. 33, 469–503. Hadded, S., Beliveau, M., Tardif, R. and Krishnan, K. (2001a). A PBPK model-based approach for the risk assessment of chemical mixtures. Toxicol. Sci. 63, 125–135. Price, K., Haddad, S. and Krishnan, K. (2003b). Physiological modeling of age-specific changes in the pharmacokinetics of organic chemicals in children. J. Toxicol. Environ. Health A 66, 417–433.

632

Hepatotoxicity

Ramsey, J. C. and Andersen, M. E. (1984). A physiologically based description of the inhalation pharmacokinetics of styrene in rats and humans. Toxicol. Appl. Pharmacol. 73, 159–175. Reddy, M. B., Yang, R. S. H., Clewell, H. J. III and Andersen, M. E. (2005). Physiologically Based Pharmacokinetic Modeling: Science and Application, John Wiley & Sons, Ltd, Chichester, UK. Reitz, R. H., McDougal, J. N., Himmelstein, M. W., Nolan, R. J. and Schumann, A. M. (1988). Physiologically based pharmacokinetic modeling with methylchloroform: implications for interspecies, high dose/low dose and dose route extrapolations. Toxicol. Appl. Pharmacol. 95, 185–199. Reitz, R. H., McCroskey, P. S., Park, C. N., Andersen, M. E. and Gargas, M. L. (1990a). Development of a physiologically based pharmacokinetic model for risk assessment with 1,4-dioxane. Toxicol. Appl. Pharmacol. 105, 37–54. Reitz, R. H., Mendrala, A. L., Corley, R. A., Quast, J. F., Gargas, M. L., Andersen, M. E., Staats, D. A. and Conolly, R. B. (1990b). Estimating the risk of liver cancer associated with human exposures to chloroform using physiologically based pharmacokinetic modeling. Toxicol. Appl. Pharmacol. 105, 443–459. Reitz, R. H., Gargas, M. L., Andersen, M. E., Provan, W. M. and Green, T. L. (1996a). Predicting cancer risk from vinyl chloride exposure with a physiologically based pharmacokinetic model. Toxicol. Appl. Pharmacol. 137, 253–267. Reitz, R. H., Gargas, M. L., Mendrala, A. L. and Schuman, A. M. (1996b). In vivo and in vitro studies of perchloroethylene metabolism for physiologically based pharmacokinetic modeling in rats, mice and humans. Toxicol. Appl. Pharmacol. 136, 289–306. Reynolds, E. S., Moslen, M. T., Szabo, S., Jaeger, R. J. and Murphy, S. D. (1975). Hepatotoxicity of vinyl chloride and 1,1-dichloroethylene. Role of mixed function oxidase system. Am. J. Pathol. 81, 219–236. Robinson, P. J. (1992). Physiologically based liver modeling and risk assessment. Risk Anal. 12, 139–148. Sarangapani, R., Teeguarden, J., Andersen, M. E., Reitz, R. H. and Plotzke, K. P. (2003). Routespecific differences in distribution characteristics of octamethylcyclotetrasiloxane in rats: analysis using PBPK models. Toxicol. Sci. 71, 41–52. Sato, A. and Nakajima, T. (1987). Pharmacokinetics of organic solvent vapors in relation to their toxicity. Scand. J. Work Environ. Health 13, 81–93. Sato, A., Endoh, K., Kaneko, T. and Johanson, G. (1991). Effects of consumption of ethanol on the biological monitoring of exposure to organic solvent vapours: a simulation study with trichloroethylene. Br. J. Ind. Med. 48, 548–556. Sirianni, G. L. and Pang, K. S. (1997). Organ clearance concepts: new perspectives on old principles. J. Pharmacokinet. Biopharm. 25, 449–470. Slikker, Jr W., Andersen, M. E., Bogdanffy, M. S., Bus, J. S., Cohen, S. D., Conolly, R. B., David, R. M., Doerrer, N. G., Dorman, D. C., Gaylor, D. W., Hattis, D., Rogers, J. M., Woodrow, Setzer, R., Swenberg, J. A. and Wallace, K. (2004): Dose-dependent transitions in mechanisms of toxicity. Toxicol. Appl. Pharmacol. 201, 203–225. Stickney, J. A., Sager, S. L., Clarkson, J. R., Smith, L. A., Locey, B. J., Bock, M. J., Hartung, R. and Olp, S. F. (2003). An updated evaluation of the carcinogenic potential of 1,4-dioxane. Regul. Toxicol. Pharmacol. 38, 183–195. St-Pierre, A., Krishnan, K. and Tardif, R. (2003). Evaluation of the influence of chloroacetic acids on the pharmacokinetics of trihalomethanes in the rat. J. Toxicol. Environ. Health A 66, 2267–2280. Tardif, R., Lapare, S., Krishnan, K. and Brodeur J (1993). Physiologically based modeling of the toxicokinetic interactions between toluene and m-xylene in the rat. Toxicol. Appl. Pharmacol. 120, 266–273. Tardif, R., Lapare, S., Charest-Tardif, G., Brodeur, J. and Krishnan, K. (1995). Physiologically based pharmacokinetic modeling of a mixture of toluene and xylene in humans. Risk Anal. 15, 335–342.


633

Tardif, R., Charest-Tardif, G., Brodeur, J. and Krishnan, K. (1997). Physiologically based pharmacokinetic modeling of a ternary mixture of alkyl benzenes in rats and humans. Toxicol. Appl. Pharmacol. 144, 120–134. Teorell, T. (1937). Kinetics of distribution of substances administered to the body. I. The extravascular modes of administration. Arch. Int. Pharmacodyn. 57, 205–225. Thakore, K. N., Gargas, M. L., Andersen, M. E., and Mehendale, H. M. (1991). PBPK derived metabolic constants, hepatotoxicity, and lethality of BrCCL3 in rats pretreated with chlordecone, phenobarbital, or mirex. Toxicol. Appl. Pharmacol. 109, 514–528. Travis, C. C., McClain, T. W. and Birkner, P. D. (1991). Diethylnitrosamine-induced hepatocarcinogenesis in rats: a theoretical study. Toxicol. Appl. Pharmacol. 109, 289–304. USEPA (1987). Update to the health assessment document and addendum for dichloromethane (methylene chloride): pharmacokinetics, mechanism of action and epidemiology, EPA/600/887/030A, Office of Health and Environmental Assessment, Washington, DC, USA. USEPA (1995). The development and validation of a physiologically based pharmacokinetic model for vinyl chloride and its application in a carcinogenic risk assessment for vinyl chloride (ICF Kaiser). Office of Health and Environmental Assessment and the Occupational Safety and Health Administration, Directorate of Health and Standards Programs, US Environmental Protection Agency, Washington, DC, USA. USEPA (2000). Toxicological review of vinyl chloride: in support of summary information on the integrated risk information system (IRIS), EPA/635/R-00/004, 1-197, National Center for Environmental Assessment, Washington, DC, USA. USEPA (2006). Approaches for the application of physiologically based pharmacokinetic (PBPK) models and supporting data in risk assessment, EPA 600/R-05/043A, Office of Health and Environmental Assesment, Washington, DC, USA. van Eijkeren, J. C. (2002). Estimation of metabolic rate constants in PBPK-models from liver slice experiments: what are the experimental needs? Risk Anal. 22, 159–173. Ward, R. C., Travis, C. C., Hetrick, D. M., Andersen, M. E. and Gargas, M. L. (1988). Toxicokinetics of tetrachloroethylene. Toxicol. Appl. Pharmacol 93, 108–117. Wilkinson, G. R. and Shand, D. G. (1975). A physiological approach to hepatic drug clearance. Clin. Pharmacol. Ther. 18, 377–390. Yamaguchi, T., Yabuki, T. M., Saito, S., Watanabe, T., Nishimura, H., Isobe, N., Shono, F. and Matsuo, M. (1996). Research to develop a predicting system of mammalian subacute toxicity (3) construction of a predictive toxicokinetic model. Chemosphere 33, 2441–2468.

Index

A1 396 A chip 513, 518, 520 ABCB 196, 218 ABCC 196, 218 ABCG 194, 198, 212, 218 ABI Prism T100 sequence detection 9 abrogated metal-induced cytokine 247 abscesses 128 accelerated blood clearance (ABC) 566–7 acetaldehyde 204–5, 245, 377 acetaminophen (APAP) 124–5, 211–12, 317–20, 346–50, 551 cellular stress 234–5, 237 chemokines and cytokines 247 drug discovery 58–9, 63, 64–6 drug metabolism and interaction 76–7, 80–1, 84 drug safety 95 experimental models 120, 124–5 gender differences 540, 543, 550–1 genomic profiling 468, 469, 473, 474, 476–7 in vitro applications 38, 40 in vitro studies 3–5, 25, 29, 34 Kupffer cells 317–20, 321–2 lipids 241–2 mechanism of TLI 192, 219–21, 230

mitochondria 231 overdose 4, 124–5, 219–20, 226–7, 234–5, 346, 524, 551 oxidative stress 108 sinusoidal cells 341–3, 346–50, 354 survival and repair 249 TGP database 510, 515–16, 524–5 TNF–á 249 toxicogenomics 457, 460 transgenic model 223–8 acetone 79, 224 acetonitrile 79 acetylaminofluorene (AAF) 209, 542, 576 acetyltransferase 72 acidosis 233, 545 acinar structures 7, 9, 11, 203, 316, 491 acrylamide (AM) 224, 225 acryl-Co-oxidase 494 acrylonitrile (AN) 224, 227 actin 193, 206, 220, 231, 241, 501–3 Actinex 597 actinomycin D 40 activator protein-1 (AP-1) 205, 210, 215–16, 236, 240 APAP 348 bile acids and apoptosis 402, 407 cellular stress 235, 236


636

Index

acyl-coenzyme A oxidase (ACOX) 158, 164–5, 326 acyl-coenzyme A oxidase (CoA) dehydrogenase 64 acyl glucuonide 434 adenocarcinoma 92, 252–3 adenosine triphosphatase 440 adenoviruses 374 adenylate cyclase 122 adenylate dehydrogenase 63 adipokines 378 adiponectin 378–9 adipocytes 378 adipose differentiation-related protein 25 ADMET 97 ADPGK (ATP-dependent glucokinase) 164–5 adrenal glands 211 adriamycin 292 Affymetrix 491 Affymetrix GeneChips 450, 452, 509–10, 512–13, 520, 526 Affymetrix GCOS 450 aflatoxicosis 178–9 aflatoxin–albumin (AF–alb) 180 aflatoxin B1 (AFB1 ) 27, 108, 178–80 biomarkers of mycotoxins 178–80, 181, 184 gender differences 540, 542, 545 toxicogenomics 454–5, 460 aflatoxins 178–80, 233 age 4, 8, 70, 120, 515–16 botanical supplements 589 cellular stress 234, 238 enzyme and gene expression 23 gender differences 546, 547, 552, 554 genomic profiling 467, 470 intrahepatic cholestasis 38 mitochondria 232 oncology drugs 564 PBPK models 617 TGP database 515–16, 519–20 Agilent Technologies 491 Ahr (aromatic hydrocarbon receptor) 225, 523, 543 AHR (aryl hydrocarbon receptor) 25, 76, 207–8, 210, 218, 518 AHR-mediated aromatic hydrocarbon response element (AHRE) 207 AHR nuclear translocator (ARNT) 207, 519, 543 AhRR (aryl hydrocarbon repressor) 207

AIDS (acquired immune deficiency syndrome) 531 Akt 221, 374, 400-2, 406 alanine aminotransferase see ALT albumin 154, 214–15, 248, 372, 518 DNA arrays 500, 503 in vitro applications 30 in vitro prediction 148, 152 in vitro studies 20–2, 24, 28, 29 oncology drugs 566 sinusoidal cells 344, 349 alcohol 24–5, 129–30, 141, 376–7, 552 apoptosis 245 bile acids and apoptosis 393 cirrhosis 126, 381 ConA 129 cytokines 374, 375, 376–7, 381 DNA arrays 494–5, 499–500 experimental models 120, 124, 126, 129–30 gender differences 110, 539, 552 isoniazid 535 Kupffer cells 320–1, 329 mitochondria 233 PBPK models 621 quinone 289 steatosis 199 transgenic models 228 alcohol dehydrogenase (ADH) 25, 499–500, 503, 552 alcoholic hepatitis 321, 376, 394 alcoholic liver disease (ALD) 130, 376–7, 379, 381 aldehyde 38, 122, 239, 240, 249, 344, 595 aldehyde dehydrogenase (ALDH) 25 aldolase B 304 alfentanil 549 alginate 147, 152 aliphatic alkanes 554 alkaline phosphatase (AP) 158, 193, 437, 566 alkanes 122, 554 alkylation 241, 288–9, 290–1, 293, 567 alkyl hydroperoxide 289 alleles 531, 533 allergies 511, 575, 591 allopurinol 130, 252, 327 allyl alcohol 5, 110, 247, 455, 473 alpha 1-acid glycoprotein (AGP) 247–8, 540 alpha-1-antitrypsin 20, 22 alpha-actinin 241 alpha fetoprotein (AFP) 21, 158, 475

Index alpha ketoisocyanate derivate 302 alpha-macroglobin 343, 49 alpha methyldopa 95 alpha naphthoflavone 541 alpha tocopherol 290 ALT (alanine transaminase) 30, 196, 216, 299 biomarkers of mycotoxins 182 botanical supplements 596, 598 DNA arrays 494 intrahepatic cholestasis 436 isoniazid 531 oncology drugs 565, 568 troglitazone 303 toxicogenomics 458 Amanita phalloides 231 amantadine 192 amebiasis 120, 128 Ames-assay 153, 577 amidase 531 amineptine 5, 95, 457 amines 540, 543, 547, 598 methapyrilene 575–6, 578, 580 amino acids 78, 217, 238, 439, 503 bile acids 392, 410 botanical supplements 591 carbon tetrachloride 350 experimental models 121, 122 genomic profiling 467, 474 methapyrilene 578 aminobiphenyl 223, 543 aminocamptothecin (AC) 90 aminoguanidine 347, 352 aminophospholipid transferase 196 aminopyrine 541 aminosalicylate 552 aminotransferase 223, 226, 534, 568 amiodarone (AMD or AD) 5, 38, 95, 231, 242, 457 drug discovery 58–9, 63–5 amitripyline 550 ammonia 21, 503 amnion cells 19, 22 amobarbital 540 Amodiaquine 74, 80 amoxixillin 193 amphetamine 410 amphiphilic drugs 230 amphotericin 148 amyloid-â (Aâ) 406, 499 amytal 540

637

anaemia 81, 521 analgesics 124, 219, 317, 346–50, 551 androgen receptor 40 androgens 540–1, 544, 545, 550 androstanol 544 androstenol 544 androsterol 209 androstone receptor 216 anesthetics 201, 545 angiogenesis 349, 353 angiotensin 378 aniline 4-hydroxylation 74 animals 3–4, 142, 159, 182, 223–8, 248 APAP 347–8 apoptosis 244, 246 bile acids 392, 410 biomarkers of mycotoxins 178–80, 181, 182 botanical supplements 595, 599 cholestasis 194–5, 197 cytokines 376, 378, 379 DNA arrays 489–94 drug metabolism and interaction 70–1 drug safety 90, 91, 101 experimental models 120–1, 123–30 food-related hepatotoxins 105–8, 110 gender differences 539–55 genomic profiling 467, 470 intrahepatic cholestasis 421, 436, 441 in vitro prediction 142–4, 152, 166 in vivo prediction 142, 143 lipids 241 oncology drugs 563–4, 567–70 PBPK models 607, 611–12, 616, 618 quinone 287, 291 sinusoidal cells 342, 345, 347–8 TGP database 507–9, 511–12, 515, 518, 520, 527 toxicogenomics 449, 452, 456–7 troglitazone 299–300, 303, 307 see also mice; rats anion exchanger 2 immunoreactivity 125 ANIT 195–7, 211 annexins 501, 503 ANOVA F-test 452, 517 anoxia 392 anthracycline 551 anthraquinone glycosides 599 antiapoptotics 243–6, 247, 250, 402–9 bile acids 396–7, 400–1, 402–9, 410, 411 antibacterials 349

638

Index

antibiotics 218, 229, 543, 550 cholestasis 125, 195 cytokines 377 hepatitis 201 intrahepatic cholestasis 438, 439 in vitro prediction 148 oncology drugs 563 TGP database 511 anticancer agents 90, 209, 500, 563–70 dry safety 90 quinone 292, 293–4 TGP database 510–11 anti-CD4 antibodies 129 antidepressants 218, 548, 550 antidiabetics 82, 209, 458 cholestasis 195, 425, 427–8, 439 troglitazone 299–307 antidiuretics 201 anti-epileptics 81, 548 anti-estrogens 221 antifibrotics 124, 126 antigens 93, 342, 344, 429, 503 Kupffer cells 315, 316, 324, 342 antihelminthic 321 antihistamine 575–83 antihypertensives 202 anti-liver(kidney microsome antibodies type 2 (LKM2) 202 anti-lysozyme 314 antimetabolites 565 antimetastatics 316, 569 antimicrobials 595 antimycotic agents 148 antineoplastics 563 antioxidant depletors 455 antioxidant response element (ARE) 235, 236, 455, 460 antioxidants 78, 123, 232, 252 APAP 347 apoptosis 245, 400 carbon tetrachloride 352 chemokines and cytokines 247 fibrosis 204 lipids 241–2 oxidative stress 108, 234–6 PPARs 212 quinone 289–90, 294 survival and repair 249 troglitazone 301 antipsychotics 548

antipyretics 211, 219, 231, 317, 346, 597 antiretrovirals 230, 437–8, 531, 590 antisense oligodeoxynucleotides (AS ODN) 565 antistaphylococcal activity 229 antitiberculostatic drugs 221 antivirals 38, 349, 375–6 anxiolytic 550 Apaf-1 (apoptotic protease-activating factor 1) 244, 395, 397, 399 aplastic anaemia 81 apoA-1 211 apocrine glands 544 apolipoproteins 158, 497, 499 apoproteins 15, 248 apoptosis 5, 154, 242–6, 391–411 bile acids 391–411 botanical supplements 596 carbon tetrachloride 350, 353 cellular stress 235, 237–8, 242 chemokines and cytokines 247 cholestasis 194 cytokines 375–7, 380 drug safety 97, 99 experimental models 120, 121, 123, 124–5, 129 fibrosis 204, 206 food-related hepatotoxins 108, 109 genomic profiling 474, 476 high content screening 31, 34–5 in vitro prediction 141, 152 Kupffer cells 315–17, 322, 324, 327–8, 342 lipids 240–2 mechanism of TLI 191, 207, 214–17, 219, 222, 242–6, 253 mitochondria 230–1, 233 oncology drugs 569 regeneration 250–1 sinusoidal cells 342, 345, 350, 353 steatosis 199–200 TGP database 523 transgenic models 225, 226, 227 troglitazone 304–7 apoptosis-inducing factor (AIF) 124, 243 apoptosis signal-regulating kinase 1 (ASK1) 238 apoptosomes 244, 395, 397 aquaporin 25 arachidonic acid 208

Index Aroclor1254 90, 153 arrhythmia 217 arsenic 227, 228 arsenite 196 arthritis 18, 251, 596 aryl alcohol 511 arylation 288–9, 290–1, 292, 293 aryl hydrocarbon 544 aryl hydrocarbon receptor see AHR aryl hydrocarbon repressor (AhRR) 207 aryl sulfotransferase 545 aryltransferases 545 asbestos 317 ascites 123, 126 ascorbic acid 290 asparagine synthase 504 aspartate aminotransferase 182, 196 aspartic proteases 503 Aspergillus flavus 178–9 Aspergillus parasiticus 178–9 aspirate transaminase see AST aspirin 124, 212, 231, 473, 550 AST (aspartate transaminase) 30, 216, 303, 436, 458, 525 oncology drugs 565, 568 astemizole 217 ATF 236 ATP 26, 73, 78, 124, 154, 161–2, 252 APAP 346 bile acid and apoptosis 396, 400 binding cassette transporters 212 carbon tetrachloride 351 cholestasis 126, 193, 194, 422 drug safety 95 genomic profiling 474, 476 intrahepatic cholestasis 422, 425, 431, 434–41 in vitro applications 30 in vitro prediction 146, 153 Kupffer cells 317 mitochondria 231 necrosis 242 oncology drugs 569 troglitazone 303, 305 ATPases 122, 123, 351 ATP-dependent BSEP 422 atractylis and atractylosides 597 autolysosomes 578 autophagosomes 578–80, 582–3 autophagy 191, 250, 393, 578–83

639

autoradiographic analysis 577 Ayurvedic medicine 590, 599 azathioprine 251-2 B cells 201, 225, 316 Bacillus Calmette-Guerin (BCG) 322 Bacillus subtilis 510 bacteria and bacterial infections 108, 247–8, 344, 376, 392 genomic profiling 465, 476 Kupffer cells 315, 320–2 quinone 287, 293 tuberculosis 535 baculosomes 145 Bad 250, 305, 475 apoptosis 243, 396, 399, 401, 406, 411 Bak 243, 250, 397 BaP 224, 225 barbital 540 basic fibroblast growth factors (bFGF) 20 Bax 124, 230, 250 apoptosis 395, 396–7, 399–400, 402, 404–6, 408–9 steatosis 200 troglitazone 305 Bay 243 Bcl-2 230, 243–4, 250, 493 apoptosis 395–7, 399, 401, 404, 406, 408–9, 411 Bcl-W 396 Bcl-X(L) 206, 230, 502 apoptosis 243, 395–6, 399, 401, 408–9 benign recurrent intrahepatic cholestasis (BRIC) 441 Benoxaprofen 80 benzafibrate 214, 455 benzbromarone 455 benzene 224, 460, 473, 545 PBPK models 613–14, 623 benzoate X receptors (BXRs) 208 benzo[a]pyrene 37, 224, 225, 542 benzodiazepine 550 benzoic acid 545 benzoquinone (BQ) 289, 291, 292, 293 benzphetamine 541 beta-actin 63 beta-blockers 206, 548 beta D-glucan 347 beta-glucuronidase 302 beta-hydroxysteroid dehydrogenase 211

640

Index

beta-naphthoflavone (BNF) 76, 78, 153, 161, 163, 550 beta-oxidation 38, 200–1, 231, 233 BH domains 396 bicarbonate 438 Bid 124, 305 apoptosis 243–4, 246, 394–7, 399 Bik 396 bile 25, 77, 119, 154, 217–18 APAP 346 cholestasis 38, 125–7, 141, 193–8, 422 cytokines 372, 379–80 excretion of drugs 69, 81, 372, 591 food-related hepatotoxins 108 genomic profiling 467 in vitro prediction 143, 144, 146, 152 troglitazone 301 bile acids 24, 26, 28, 155, 197–8, 399–411 apoptosis 246, 391–411 cellular stress 236 cholestasis 125, 126–7, 193–8 DNA arrays 502, 503 drug safety 95 genomic profiling 467, 474 intrahepatic cholestasis 422–3, 425–6, 430–2, 436–42 in vitro applications 38 mechanisms of TLI 191, 208, 210, 211, 216–17 mitochondria 231 sulfotransferases 545–6 toxicogenomics 458 bile canaliculi 11, 580 cholestasis 125, 127, 193, 195, 196, 198 in vitro prediction 143–5, 147–51 bile duct ligation (BDL) 125–6, 130, 379–80, 406 cholestasis 197, 206 experimental models 120, 125–6 bile ducts 193, 287, 324, 424, 457 cholestasis 125–6, 193, 196, 197–8 experimental models 119, 123 fibrosis 202, 203, 206 bile salt export pump (BSEP) 24, 26, 38–9, 303, 421–42 cholestasis 126, 194–8, 421–42 drug discovery 441–2 regulation of expression 440 species difference 439–40 troglitazone 303, 304

bile salts 5, 246 cholestasis 127, 193, 195, 196, 198 biliary epithelial cell (BEC) 197, 469 bilirubin 38, 208, 210, 392, 543, 550 cholestasis 127, 194, 195 oncology drugs 565, 568 bilirubin glucuronide 422 biliverdin 249 Bim 243, 396 bioactivation 79–80, 84, 201–2, 221, 230 gender differences 540, 554 intrahepatic cholestasis 431, 433 Kupffer cells 319, 320, 326 bioinformatics 61, 66, 159, 450, 524 genomic profiling 466, 472 biological exposure indices (BEIs) 178 biomarkers 61–2, 63–5, 177–84, 459–60, 525–6 drug metabolism and interaction 73, 83 drug safety 96 gene expression 159 genomic profiling 466–7, 473 oncology drugs 569–70 TGP database 511, 525–6 toxicogenomics 449, 459–61 biotransformation 8, 13–14, 16, 18, 143, 217–19, 221–2 acute phase reaction 248 botanical supplements 595 cellular stress 234 drug metabolism and interaction 72, 84 experimental models 119, 121, 130 gender differences 540, 554 genomic profiling 467, 474 mechanisms of TLI 192, 207–10, 215, 217–30 toxicogenomics 460 transgenic models 222–8 troglitazone 302 bisphenol-A 546 bisphoglycerate synthetase 220 bladder 539 blebbing 154, 243 Blk 396 Bmf 396 Bnip 396 Bod 396 Bok 396 bone marrow 19, 20–1, 37, 225, 313 gender differences 545, 551 Boo/DIVA 396

Index borborygmi 181 bosentan 422, 424, 426, 429 botanical supplements 589–600 bowel disease 251–2 see also colon and colorectal cancer brain and neurotoxicity 93, 291, 467, 542 see also neurodegenerative diseases breast cancer 221 breast cancer resistance protein (BCRP) 77, 218 bridging biomarkers 61–2 BRIRB 222 bromfenac 80 bromobenzene (BB) 247, 457, 459–60, 473, 476, 525 bromodichloromethane 620–1 bromoform 553 BSA 77 BSO 524 BSP 233 bufuralol 1 -hydroxylation 74 Bupropion-hydroxylation 74 1,3-butadiene 540, 614 2-butoxyethanol 326, 328 Byler disease 194, 196, 440 bZip 214–15, 236 C. elegans 410 C-reactive protein (CRP) 222, 247-8 cadmium 318–19, 322–3, 599 cadmium chloride (CdCl2 ) 95, 247, 323 caffeine 74, 542–3, 547, 551, 598 calcineurin 235 calcium 122–3, 220, 241, 244, 251, 305, 476 bile acids and apoptosis 396–7, 401, 404–6 carbon tetrachloride 350–1 experimental models 121, 122–3, 124 high content screening 30–1, 33, 34–5, 36 in vitro prediction 141, 146 Kupffer cells 321 mitochondria 231 nuclear receptors 208–9 troglitazone 305, 306 Callilepsis laureola 594, 597 calves 110 calpain 124, 405, 503 calreticulin 304 CAM lactone 90 canaliclular BSEP 303 cancer and carcinoma 24–6, 107, 218, 221, 250, 325–9, 496

641

apoptosis 246 bile acids and apoptosis 393, 394, 400, 404, 406–7 biomarkers of mycotoxins 179–80, 181 botanical supplements 596 cellular stress 239 DNA arrays 500, 503 drug development 563–70 drug-related liver injury 252–3 experimental models 121 food-related hepatotoxins 109 gender differences 542, 545, 547, 551 genomic profiling 465, 467, 475 Kupffer cells 325–9 methapyrilene 575–83 PBPK models 618, 620 quinone 287 transgenic models 225 see also carcinogens and carcinogenesis; hepatoma; HepG2 carbamazepine 5, 95, 201, 550 carbamolyphosphate synthetase (CPS1) 21 carbocation 221 carbohydrates 211, 342, 344, 372 genomic profiling 467, 474, 476 carbon monoxide 249 carbon tetrachloride (CCl4 ) 5, 121–4, 214, 321–2, 350–4 acute phase reaction 247–8 cellular stress 237, 239 cholestasis 126, 194 cirrhosis 130 cytokines 374, 380–1 drug discovery 58–9, 63–5 experimental models 120, 121–4 food-related hepatotoxins 110 gender differences 544 genomic profiling 473, 474, 476 Kupffer cells 318–19, 321–2 lipids 239 mechanisms of TLI 192, 215–16, 219 PBPK models 613, 620–1 sinusoidal cells 341–3, 350–4 survival and repair 249 TGP database 510–11, 515 toxicogenomics 455, 457 transgenic model 224, 226, 227 carbonic anhydrase 63 carbonyls 122, 226 carboxylesterase 64, 209, 304

642

Index

carboxyl-terminal propeptides 205 carcinogens and carcinogenesis 207–8, 214, 216, 325–9 apoptosis 246 bile acids and apoptosis 406, 408–9 biomarkers of mycotoxins 178–9, 181 botanical supplements 595 cellular stress 234, 239 endpoint of culture analysis 159 experimental models 121 fibrosis 203 food-related 108, 109, 111 gender differences 542–3, 545–6, 552 genomic profiling 468, 473, 474–6 Kupffer cells 313–29 methapyrilene 575–83 mycotoxins 178, 181 PBPK models 614, 616, 618, 620 survival and repair 250 TGP database 520 toxicogenomics 454, 457, 460 transgenic models 225, 228 troglitazone 301 cardioactive drugs 548, 598 cardiotoxicity 3, 18, 93, 222, 251–2 botanical supplements 598 DNA arrays 497 drug metabolism and interaction 76 drug safety 97 gender differences 551 high content screening 37 genomic profiling 467 TGP database 511 cardiomyocytes 93 carnitine palmitoyl transferase (CPT) 164–5, 211 carvedilol 347 caspases (cysteine-aspartase-specific proteases) 124, 154, 220 apoptosis 243–4, 246 bile acids and apoptosis 393–5, 398–9, 401, 404–6, 409, 411 cellular stress 237 chemokines and cytokines 247 fibrosis 204, 206 high content screening 33 lipids 240 mitochondria 231 steatosis 200 troglitazone 305

castration 544, 546 catabolic enzymes 123, 351 catabolites 392 catalase 64, 226, 234, 247, 502 quinone 289 sinusoidal cells 343, 352 catechin 565 catecholamines 550 catechol-O-methyltransferase (COMT) 550 catechols 550 cathepsin 200, 244, 398, 475 cationic amphiphilic drugs 31, 230 cations 290 cats 120 caudate 119 CCAAT/enhancer-binding proteins (C(EBP)157, 214–16, 250 CCMF 81–2 CD14 320, 351,377 cDNA 9, 84, 146, 215 DNA arrays 490–1, 493, 496, 497, 499, 502–3 microarrays 96, 451 toxicogenomics 450–1, 457 cell adhesion 20, 73, 147, 342, 346, 371 see also intercellular adhesion molecules (ICAM) cell aggregation 106 cell-based assay technologies 62, 65–7, 142–53 cell–cell interaction 66, 143, 145–7, 150, 152 cell cycle 372, 406, 474, 476, 498 arrest 306 cell death 108, 141, 154, 242–6, 305, 392–7 bile acids and apoptosis 392–7, 398, 391, 400 botanical supplements 592, 596–7 cellular stress 236, 238 cholestasis 198 cytokines 380 DNA arrays 498 drug metabolism and interaction 70 experimental models 123, 124, 129 food-related hepatotoxins 108, 109 Kupffer cells 317, 324 mechanisms of TLI 212, 219, 222 methapyrilene 583 mitochondria 231–2 oncology drugs 569 PBPK models 619–20 quinone 289, 290, 293 regeneration 250

Index sinusoidal cells 351 steatosis 199–200 TGP database 523 TNF-á 248 troglitazone 304, 305 see also necrosis cell lines 24–7, 41, 83, 107, 145, 153, 157 DNA arrays 489–91, 496, 498–500, 504 drug discovery 58, 60 food-related hepatotoxins 106, 107 in vitro applications 37 in vitro prediction 142, 145, 151, 153 methapyrilene 577 oncology drugs 567 troglitazone 304, 306 cell–matrix interaction 66, 143, 145–7, 236 cell numbers 34–5, 36, 154 cellomics 96 cell transfection array technologies 62 cellular FLICE inhibitory protein (cFLIP) 398 cellular stress 231–2, 233–42, 253, 514 see also oxidative stress Centella asiatica 594, 599 central nervous system (CNS) 121, 222, 511, 545 ceramide synthase 181 cerivastatin 31, 33 c-fos 215, 348, 402, 475 chaparral 109, 590, 594, 596–7 chaperone genes 305-6, 407 chaperone proteins 493 Chee’s medium 77 chemicals 191, 214, 218, 554–5 biomarkers of mycotoxins 177–8 botanical supplements 590–1, 599 cellular stress 236 cytokines 374 DNA arrays 489, 494 fibrosis 203 food-related hepatotoxins 106, 107, 109, 111 genomic profiling 471–3 in vitro predictions 141, 143 Kupffer cells 329 methapyrilene 575, 579–80 PBPK models 607, 610–16, 618, 620–4 sinusoidal cells 354 TGP database 507–11, 518–23, 525–6 toxicogenomics 456, 457–9 chemokines 198, 246–7, 322, 371, 394, 503 sinusoidal cells 341, 344, 345, 348–50

643

chemotaxis 237, 240, 249, 343, 345, 346 chemotherapy 252–3, 551, 563, 565, 569 chenodeoxycholate 422–3 chenodeoxycholic acid 193, 392, 407 children 179, 554, 617–18 Chinese hamster ovary (CHO) cells 577 chloral hydrate 540 chloramphenicol 34, 542 chlordecone 621 2-chloro-1,1,1-trifluoroethane 614 chloroform 5, 110, 121, 219, 473 PBPK models 608, 613–14, 618–20 chloromethylfluorescein diacetate (CMFDA) 39 chloropentafluorobenzene 608 chloroxazone 6-hydroxylation 74, 90 chlorpromazine 125, 457, 543, 552, 579 chlorpropamide 426–7 chlorzoxazone 548 cholangiocarcinoma 495, 575 cholangiocytes 125, 194, 195, 198 apoptosis 245–6, 393, 402, 406 cholangitis 197 cholate 22–3 choleresis 438–9 cholestasis 125–7, 141, 193–8, 208, 397–8 apoptosis 246 bile acids and apoptosis 392–3, 397–8, 400–2, 406–7, 411 botanical supplements 592, 595–6 drug induced 5–6, 24, 26 drug safety 95 experimental models 120, 125–7 fibrosis 379 food-related hepatotoxins 108 genomic profiling 469 in vitro applications 37, 38 nuclear receptors 210, 211 oncology drugs 564–5, 567 toxicogenomics 457–8 troglitazone 303, 304 see also extrahepatic cholestasis; intrahepatic cholestasis cholesterol 200, 211, 217, 460, 566 bile acids 391, 392, 406–7 cholestasis 125, 127, 193, 195, 196 genomic profiling 467, 474, 476 cholesterol 7 alpha hydroxylase 196, 215, 392 cholesterol hydroperoxide 289 cholic acid 193, 392, 407 choline 224

644

Index

cholyltaurine 438 chondroitin sulphate 351 CHOP-10 216 chromatid 229 chromatin 124, 243, 393, 409 chromosomes 532, 577–8 chronic effects 161, 191 bile acids and apoptosis 393, 396, 397, 401, 410 botanical supplements 595 chemokines and cytokines 246 cytokines 376, 379, 381 drug safety 93–4, 96 in vitro prediction 142, 143, 144, 148 oncology drugs 566, 568 TNF-á 249 toxicogenomics 449 troglitazone 304 chrysin 209 cIAP1 399, 401 cIAP2 401 cimetidine 542 ciprofibrate 76, 212 ciprofloxacin 229 circadian rhythmns 467, 470, 518–19, 547, 549 cirrhosis 123–4, 239, 246, 397, 402 alcohol 129, 376 botanical supplements 596 carbon tetrachloride 350–1, 353 cholestasis 125–6 cytokines 246, 372, 376, 379–81 experimental models 120–2, 123–4 fibrosis 203 gender differences 551 genomic profiling 468–9, 471–2 Kupffer cells 321 lipids 241 steatosis 199 troglitazone 304 cis-acting enhancer 236 cisapride 217 CisH 493 cis–regulatory elements 214 cisplatin 227 cis-platinum 95, 227 9-cis retinoic acid 210 citalopram 495 CITCO 544 c-Jun 205, 215, 238, 348

c-Jun N-terminal kinase (JNK) 235, 237–8, 305 c-kit 20 clarithromycin 218, 552 clathrin 579 clavulanic acid 193 CLO 13–15 clodronate 316–18, 320, 324–5 clofibrate 8, 76, 212, 218, 473, 476 toxicogenomics 456 transgenic models 228 clofibric acid 473, 550 clomipramine 548 clones 60, 153, 327, 329, 422, 434, 498 Clontech 491 clotrimazole 76, 209, 543, 595 cloxacillin 439 clozapine 547 cLPMV (canalicular liver plasma membrane vesicle) 425, 432, 435, 439, 442 cluster analysis 456, 494, 501–2 c-myc 211, 402, 475 CoA 38, 63–4, 211, 212, 476 see also fatty acyl CoA oxide (FACO) coagulation 592–3 cocaine 95, 110 co-culture 145–7, 151–2, 567 co-factors 6, 372 collagens 6–7, 106, 154, 157–60, 351–2, 500–3 cellular stress 239 cytokines 372 fibrosis 202, 204–6 DNA arrays 492, 497, 498–9, 500–3, 504 drug metabolism and interaction 72, 77 in vitro applications 161–3 in vitro prediction 147, 148–52 Kupffer cells 319, 322 sinusoidal cells 343, 344–5, 351–3 collagenase 106, 341, 343, 353, 411, 512 colloid 323 colon and colorectal cancer 76, 252, 495 bile acids and apoptosis 400, 404, 406–7 colony-stimulating factor 315, 345 coma 545 combinatorial chemistry 57 Comet assay 329 comfrey 590, 593–5 common housekeeping gene 9, 12, 16, 452, 494 Comparative Toxicogenomics Database (CTB) 159 complementary medicine 589

Index compound X 161–2 conazole fungicides 473, 476 concanavalin A (ConA) 120, 129, 376, 455 connective tissue growth factor (CTGF) 228, 347 constitutive androstane receptor (CAR) 18, 40, 164–5, 210–11, 440 cholestasis 127, 196–7 drug metabolism and interaction 76–7 gender differences 541, 543–4, 548–9 mechanisms of TLI 207–8, 210–12, 218–20 transgenic models 227 copper 199, 232, 239–40, 379 apoptosis 244, 245, 393 copper zinc (CuZn) superoxide dismutase 247, 289 corn oil 512, 520 corticosteroids 252, 347 corticosterone 209 cortisol 549 Corynebacterium parvum 320, 322 Cos-7 cells 404 costs of drug development 89, 507, 589 cotinine 540, 548 coumarin 455, 457–8, 473, 525 gender differences 540–1, 547 coumarin-7-hydroxylation 74, 90, 541, 544, 547 councilman bodies 243, 398 covalent binding assays 567 CRIg receptors 342 cRNA 450–1 Crotalaria 594–5 cryopreservation 19, 71, 72, 75, 77, 82, 512 drug safety 92, 94 food-related hepatotoxins 106–7, 111 gender differences 549 in vitro prediction 145, 147, 153 troglitazone 302 CT method 158, 163, 164 cumene hydroperoxide 434 CuraGen 459 C-X-C chemokines 247 CXCR2 350 cyanide 227 cycle threshold (Ct) 9, 12 cyclic adenosine monophosphate (cAMP) 401 cyclin-dependent kinase (cdk) 250, 306 cyclin-dependent kinase inhibitor 1A (cdkn 1a) 184 cyclins 157, 184, 228, 306, 406, 475, 502

645

cyclohexamide 95, 473 cyclohexane 614 cyclohydrolase 504 cyclooxegenases 217, 249, 326, 328, 343, 348, 592 cyclophilin 9 cyclophosphamide 27, 38, 40, 95, 565 cyclophylin (CYC) 9–12, 425 cyclosporin A (CsA) 5, 26, 95, 400, 457, 549 cholestasis 125, 195 intrahepatic cholestasis 422, 424, 425–6, 439, 442 cyprofibrate 13 Cys 238 cysteine 81, 236–7, 288–9, 302, 404 cytochrome c 124, 220, 231 apoptosis 243–4 bile acids and apoptosis 394–7, 399-400, 402–6, 408 lipids 240 mitochondria 109, 233, 394 quinone 293 steatosis 200 troglitazone 305 cytochrome c oxidase 232 cytochrome isozymes 121, 153, 230 cytochrome monooxygenases 248, 251 cytochrome oxidase (COX) 233 cytochrome P450 (CYP450) 25–9, 72–3, 109, 161–2, 184, 287–95 acute phase reaction 248 bile acids 392 biomarkers of mycotoxins 179 botanical supplements 590, 595–7 DNA arrays 497, 499, 504 experimental models 121, 122, 125, 127 gender differences 110–11, 540–5, 547–9, 551, 555 genomic profiling 469, 475–6 in vitro DILI 6–11, 13–18, 21–3 in vitro prediction 143 intrahepatic cholestasis 429, 433, 437 isoniazid 531 Kupffer cells 315, 316 mechanisms of TLI 207, 210, 217, 219–21 methapyrilene 576 oncology drugs 567, 569 PBPK models 618–19 RNAi 40 sinusoidal cells 341–2, 346, 349, 350

646

Index

cytochrome (Cont.) transgenic models 224 troglitazone 301 cytochrome P450 monooxygenase 72, 75, 79, 422 cytochrome P450 oxygenase 121 cytochromes (CYPs) 7–11, 13–18, 20–3, 26–7, 40, 75–9, 195–8 acute phase reaction 248 bile acids 407 biomarkers of mycotoxins 179–80 botanical supplements 592 BSEP 440 DNA arrays 493, 497, 499, 501, 503–4 drug metabolism and interaction 71–2, 73–9, 83–4 drug safety 92 endpoints of culture analysis 154, 155, 157–60, 161–5 engineered cells 24–7, 28, 29 fibrosis 204 gender differences 540–5, 547–8, 554 genetic polymorphisms 251 genomic profiling 469, 475–6 hepatitis 202 in vitro applications 161–3 in vitro prediction 145–6, 148, 152, 153 Kupffer cells 321 mechanisms of TLI 207–12, 215, 217–19, 220–8 oncology drugs 569 oxidative stress 242 TGP database 515, 523, 525 toxicogenomics 460 transgenic models 224 troglitazone 300–3, 304 cytofluorometry 403 cytokeratins 21, 22, 501 cytokine-induced neutrophil chemoattractant (CINC) 249, 345 cytokines 246–7, 371–82 acute phase reaction 247–8 bile acids and apoptosis 393–4, 401, 407 botanical supplements 592 cellular stress 236, 237 cholestasis 194, 198 drug safety 94, 100 experimental models 121, 129 fibrosis 203–6 in vitro prediction 151, 152

Kupffer cells 315–18, 320–5, 327, 329, 342, 344 lipids 241 mechanisms of TLI 192, 208, 210–11, 215–16, 220–2, 229, 253 regeneration 250 sinusoidal cells 341–2, 344–5, 347–9, 350, 352–4 TNF-á 248–9 transgenic models 227 troglitazone 305 cytolytic hepatitis 199 cytolytic reactions 191, 201 cytometric measurements 30 cytomics 159 cytoplasm 6, 96, 342, 346, 578–80 bile acids and apoptosis 398, 401 cellular stress 235, 236, 238 gender differences 543, 553 in vitro prediction 148, 150, 152 Kupffer cells 314, 322–3, 324 quinone 289 troglitazone 305 cytoskeleton 123, 147, 193, 351, 397 DNA arrays 501, 503 genomic profiling 474 lipids 241 necrosis 242 TGP database 524 cytosol 73, 146, 341, 351, 543, 546 apoptosis 243, 396–7, 400, 402, 406, 408–9 cholestasis 195 in vitro prediction 143, 145, 146 intrahepatic cholestasis 431 Kupffer cells 315 lipids 241 sulfotransferases 545–6 transgenic models 228 cytotoxicity 5, 108, 237, 249, 376 apoptosis 245 bile acids 391, 398, 401, 406, 411, 423 botanical supplements 598 drug discovery 59 drug safety 93–5, 100–1 food-related hepatotoxins 106–9 gender differences 541, 551 high content screening 30–1, 33–7 intrahepatic cholestasis 434, 436 in vitro prediction 141–2, 143, 152, 154, 161–2, 166

Index Kupffer cells 315–16, 318 lipids 241 oncology drugs 563, 565–7 PBPK models 620 quinone 287, 291, 293 regeneration 250 sinusoidal cells 341, 343, 345, 354 TGP database 510 toxicogenomics 454 transgenic models 224 troglitazone 301 danazol 34 dantrolene 552 dATP 397 daunorubicin 566 D-binding protein 214 DDI 424 D-dopachrome tautomerase 64 death 69, 82, 179, 201, 425, 449, 568 botanical supplements 589 experimental models 121, 123, 126–8, 130 generic polymorphisms 252 troglitazone 299–300 death effector domain (DED) 394 death-inducing signalling complex (DISC) 243, 394, 398 death receptors (DR) 194, 200, 206, 238, 394–5 apoptosis 243–6, 394–5, 398–9, 400, 404 debrisoquine 548 debrisoquine 4-hydrolation 74 dehydrocholate 422–3 dendrogram tree 453 2 -deoxyadenosine triphosphate 232 deoxycholate 422–3 deoxycholic acid (DCA) 246, 392, 398, 400, 402–4, 407 2 -deoxycytidine 5 -triphosphate 232 2-deoxyguanosine 239 deoxyribonuclease 227 depression 539 dermatan 351 dermatology 510 desipramine 548 desmin 204 desmosomes 580 dexamethasone (DEX) 6, 21, 29, 148, 151, 159 bile acids 407 botanical supplements 595 drug metabolism and interaction 76, 77–8

647

gender differences 543–4 PXR 209 dextran sulfate 347 dextromethorphan N-demethylation 74 dextromethorphan O-demethylation 74, 548 D-galactosamine 376, 455, 473, 477 diabetes 199, 213–14, 299–307, 599 intrahepatic cholestasis 427–9 troglitazone 299–307, 551 dialdehyde phenolate 180 diarrhea 181, 184, 595, 597 diazepam 495, 549 dicer 39 dichloroacetate 455 dichloroacetic acid 608 dichlorobenzene 457, 545 2,6-dichlorobenzonitrile 540 dichlorodihydrofluorescein 39 dichloroethane 613 dichlorofluorescein 109 dichlorohydroquinone 545 dichloromethane (DCM) 613–14, 618–19, 623 dichloromethylene diphosphonate 316, 347 dichloropropanol 27 dichlorovinylcysteine 608 diclofenac 5, 27, 34, 38, 74, 476 gender differences 541, 552 dicoumarol 290 dieldrin 473 diet and food 4, 105–11, 178–9, 379, 440 apoptosis 246, 392 biomarkers of mycotoxins 177–9, 181–4 botanical supplements 589–600 cellular stress 233–4 experimental models 120, 129–30 feeding states 467, 470 gender differences 543, 546, 547 genomic profiling 467–8, 470 Kupffer cells 315, 321–2, 329 oncology drugs 568 quinone 287 TGP database 512 transgenic models 224 troglitazone 301 see also alcohol diethylaminoethoxyhexestrol (DEAH) 38, 95 diethyl ether 614 diethylhexylphthalate (DEHP) 214, 473 diethylnitrosamine (DEN) 181, 468, 620 diethylstilbestrol 539

648

Index

diflunisal 455, 473, 550 difluorinated side-chain 229 digestion see gastrointestinal system digitonin-collagenase perfusion 147 dihydralazine 202 dihydroethidium 32, 36 dihydro-8-(NT-guanyl)-9-hydroxy-AFB1 (AFB1 -N7-Gua) 179–80 dihydropyrimidine dehydrogenase (DPD) 252–3, 551 dihydropyrrolizidine (DHP) 595 diisonylphthalate 473 dimers 207, 236 dimethoxy-1,4-naphthoquinone (DMNQ) 290–3 dimethylbenz[a]anthracene (DMBA) 225 dimethylnitrosamine (DMN) 37, 454–5, 460, 473 dimethyl sulfoxide (DMSO) 8, 64, 79, 106, 157–8, 494 methapyrilene 577 TGP database 512 dinitrophenyl 437 dione 209 dioxane 608, 613, 618 dioxin 207, 543, 620 dioxin response element (DRE) 207, 543 diphenyleneiodonium sulfate 320 diquat 226 discriminant analysis 526 disease and health states 4, 70, 440, 494–5, 592–3 gender differences 539, 547, 554 genomic profiling 465–9, 471–3, 477 disulfide 234, 291 disulfide isomerise 201, 304 disulfiram 38 dithiothreitol 291 DKO 225 DM-160 medium 148 DMF 494 DNA 27, 99, 142, 159, 489–504, 533–4 APAP 346 apoptosis 242–4, 393, 406, 407, 409 biomarkers of mycotoxins 179–80 botanical supplements 595 cellular stress 235, 236–9 cytokines 372–4 experimental models 121, 122, 124 food-related hepatotoxins 108

gender differences 545 genomic profiling 465, 474–6 high content screening 31–3 Kupffer cells 319, 325, 326–9 lipids 240, 242 mechanisms of TLI 207, 211, 213–6, 221, 229 methapyrilene 575–8, 580 mitochondria 231–3 quinone 290 TGP database 509–10, 512 toxicogenomics 450, 454, 456, 460 transgenic models 223–6 dogs 4, 90, 110, 120, 458, 541, 565 donors 8, 13–15, 17, 19, 72, 82, 163 DNA arrays 489–92, 495–6, 498 drug safety 95, 96 food-related hepatotoxins 107 gender differences 547, 549 in vitro prediction 143, 146, 147 troglitazone 302 dopamine 410 dosage 3–4, 71, 143, 192, 229, 435–6, 614–15 APAP 124, 317, 346–7 biomarkers 178, 180, 182, 183 botanical supplements 590, 591–3, 595–7, 599 carbon tetrachloride 350 cellular stress 239 cholestasis 127, 197 DNA arrays 494 drug discovery 59, 61 drug safety 93–4 experimental models 121, 124, 127, 129 gender differences 540, 552–3, 554 genetic polymorphisms 252 genomic profiling 470–1, 473–4, 476–7 high content screening 33, 36 intrahepatic cholestasis 421, 423, 426–7, 432, 435–41 isoniazid 535 Kupffer cells 316–17, 324 lipids 241 methapyrilene 576 mitochondria 231 multiple 516–17 nefazodone 436 oncology drugs 563–6, 568–70 PBPK models 607–8, 610–11, 614–22, 624

Index sulindac 438–9 TGP database 508–10, 512–22, 524–5, 527 toxicogenomics 450, 457–60 transgenic models 224, 225 troglitazone 300–1, 303 doxorubicin 219, 292, 294, 473, 551, 566 doxycycline 5 DP5 396 DR5 398 drosha 39 drug development 3, 7, 19, 41, 57, 93, 563–70 gender differences 539 in vitro applications 161, 162 in vitro prediction 141, 166 metabolism and interaction 71–3, 75, 81, 84 oncology 563–70 safety 89, 93, 101–2 TGP database 507, 511, 526–8 toxicogenomics 449–50, 453 drug discovery 7, 19, 28, 41, 57–67, 89, 441–2 intrahepatic cholestasis 421, 441–2 oncology drugs 563, 570 Systems Biology 161 toxicogenomics 450, 452–3 drug–drug interactions and co-medications 4, 18, 69–84, 157, 161, 435–6 botanical supplements 590, 592–3, 596, 600 genetic polymorphisms 252–3 intrahepatic cholestasis 421–3, 425–6, 429, 435–6, 438 in vitro prediction 141, 144, 148 mechanisms of TLI 212, 217, 253 oncology 567–9 safety 91 tuberculosis 531, 535 DT-diaphorase 576 DTT 597 Dubin-Johnson syndrome 194 ducks 182 Dulbecco’s modified Eagle medium (DMEM) 148 Dunnett’s T-test 9 duration of exposure to hepatotoxins 142, 161–3, 166, 121, 253 botanical supplements 591, 595, 599 carbon tetrachloride 350 DNA arrays 490, 496, 500 drug discovery 60–1 drug metabolism and interaction 73, 78–9

649

drug safety 93–4 gender differences 553 gene expression 156–60 genomic profiling 470–1, 474, 476 high count screening 31 intrahepatic cholestasis 421, 423 Kupffer cells 322 methapyrilene 580–1 mitochondria 231 oncology drugs 564 PBPK models 607, 609, 614, 618, 620–1 quinone 291, 292, 293 toxicogenomics 450 troglitazone 304 dysentery 128 dysglycaemia 229 dyslipidaemia 213–14 E2F1 408, 409 E. histolytica 128 early growth response factor-1 (egr-1) 194, 211, 251, 348, 503 Ebrotidine 80 ecstasy 552 ED1 314, 317, 324, 351 ED2 314, 316, 324, 351 edema 126, 179 efavirenz 423, 437–8 efavirenz hydroxylase 74 efficacy 3, 70, 89, 93, 142, 177, 511 APAP 346 botanical supplements 590–1 high count screening 33, 37 quinone 293 eicanosoids 219, 324, 342, 352 eicosanoids 211, 315, 318, 324 sinusoidal cells 341–5, 347, 352, 354 Eigenvector value 526–7 elastin 352 electron micrographs 9, 11 electron microscopy 96, 128 electrophiles 221, 234, 476, 545, 577 drug metabolism and interaction 70, 81 metabolites 39, 81–2 quinone 289 electrophilic response element (EpRE) 455, 460 ELISA (enzyme linked immunosorbent assay) 154–5 elongation factor 1-ã 64 EM 64

650

Index

embryos 21, 23, 107, 207, 210, 222, 225 apoptosis 242 calf serum 149 fibrosis 205–6 food-related hepatotoxins 107 reactive metabolites 81 stem cells (ES) 19, 21 troglitazone 306 EMD 162–4 endobiotics 422 endocrine system 542 endocytosis 315, 344, 349, 578, 579, 582 endoglin 351 endonucleases 39, 123–4, 243, 244, 351 endoplasmic reticulum (ER) 11, 243, 322, 524, 548, 550 bile acids and apoptosis 396, 397, 404–5 experimental models 121, 123 methapyrilene 575, 578–80, 582 quinone 287, 289, 291 sinusoidal cells 344, 351 troglitazone 306 endosomes 579, 582–3 endothelial cells 29, 344, 348–9, 393, 474, 495 botanical supplements 592–3 chemokines and cytokines 247, 371–2, 374 in vitro prediction 147, 151 Kupffer cells 313–14, 316, 319, 322, 325–6, 328–9, 342 sinusoidal 242–3, 344, 348–9, 351, 354 endothelin (ET) 343–5, 351, 353 endotoxemia 377 endotoxin 195, 226, 473–4, 476 Kupffer cells 315, 319, 320, 321–2, 329 Engelbreth–Holm–Swarm (EHS) mouse sarcoma 7, 148 entactin 7 enterotoxin receptor 497 environment 70, 96, 177–8, 207, 454, 621 botanical supplements 592–3, 600 cadmium 322 Environment, Drugs and Gene Expression (EDGE) database 473 enzymes 8, 30, 39, 69, 141, 162, 219, 243 bile acids and apoptosis 392–3, 397, 401, 404, 407 botanical supplements 592–3, 596 BSEP 440 cellular stress 234, 235, 236 cholestasis 197

cytokines 372, 378 DNA arrays 493, 503, 504 drug-induced liver injury 4, 8–9, 13–18, 21–2, 25–7, 29 drug metabolism and interaction 69–73, 75–80 drug safety 92, 94–5 endpoints of culture analysis 154, 157 experimental models 121, 122, 123 fibrosis 203, 205 food-related hepatotoxins 106, 108, 109, 110 gender differences 539–46, 547–50, 554–5 genetic polymorphisms 251–3 genomic profiling 469–70 hepatitis 202 high content screening 34–5 intrahepatic cholestasis 421–2, 424, 429, 439–40 in vitro prediction 144–5, 146, 147, 148, 153 Kupffer cells 314–15, 322, 328, 342 lipids 240 mechanisms of TLI 191–3, 207, 211–12, 217–21, 230, 252–3 methapyrilene 576, 578–9 mitochondria 230, 231–2 NAT2 531–5 oncology drugs 563, 565, 569 PBPK models 610, 615, 621 quinone 287–8, 289–91 reactive metabolite 79–81 RNAi 40 sinusoidal cells 341–2, 344, 349, 353 steatosis 199–200 survival and repair 25 TGP database 525 toxicogenomics 456, 460 transgenic animals 223 troglitazone 303–4 see also Phase I enzymes; Phase II enzymes eosinophilia 193, 201, 592 eosinophils 129, 223, 520, 599 Ephedra 109, 594, 598 ephedrine 598 epidermal growth factor (EGF) 148, 374 epidermal growth factor receptor (EGFR) 394, 398, 402, 404 epifluorescence 30, 39 epigallocatechin gallate (EGCG) 565 epigenetics 181, 575–6, 583 epirubicin 551

Index epithelial cells 6, 20, 26, 108, 554, 565 epoxide hydrolase (mEH) 227, 302, 455, 545, 576 epoxides 224, 301–2, 541–2, 595 Era 195 ercamptothecin (CAM) 90 EROD metabolism 15 erythrocytes 289, 314, 315 erythromycin 74, 95, 125, 218, 549 Escherichia coli 248 E-selectin 344 esophageal cancer 181 esterification 200 estradiol 540, 541, 544, 545–6, 549–50 estradiol-17â-D-glucuronide 5, 127, 437–8 estriol 545 estrogen 126–7, 208, 209, 221, 437 cholestasis 125, 126–7, 195 gender differences 539–42, 545–6, 549 estrogen 17á-ethinylestradiol (EE2) 195, 437 estrogen metabolites 437 estrogen receptor 207 estrogen receptor a´ (ERá) 127, 195 estrone 544, 545–6 estrone-3-sulfate 303 ethane 240 ethanol 25, 79, 129–30, 211, 320–1 apoptosis 244–5 bile acids and apoptosis 404 cellular stress 237 cytokines 376, 378–9 drug safety 95 fibrosis 204–5 food-related hepatotoxins 110 gender differences 539, 548 Kupffer cells 318–19, 320–1 lipids 239, 241 mechanisms of TLI 218, 229, 230 PBPK models 621 steatosis 199–200 transgenic models 224, 228 ethenobases 239 etheno-deoxyadenosine 239 etheno (epsilon)- modified DNA 239 ethics 91, 142, 490, 507 ethidium 32, 403 ethinyl estradiol 5 ethionine 248 ethnicity 49, 532, 534, 548, 552, 596–7 ethoxycoumarin-O-deethylase 20

651

7-ethoxyresorufin O-dealkylation 547 7-ethoxyresorufin-O-deethylation 74 ethylbenzene 613, 623 ethylhexanol 212 ethylmorphine 541, 542 17á-ethynylestradiol (EE) 126–7 etoposide 455 Euclidian distance 453 eukaryotic imitation factor 2á (eIF2á) 306 excretion or clearance of hepatotoxins 191, 217–18, 232, 234, 372, 431 APAP 346 bile acids and apoptosis 73–5, 402, 439 botanical supplements 591 drug metabolism and interaction 69–70, 73–5, 77, 81 gender differences 548–51, 554 intrahepatic cholestasis 430–1, 434–6, 437–9, 441 Kupffer cells 317 nefazodone 436 oncology drugs 564, 566–7 PBPK models 608–10, 612–13, 621 sulindac 439 troglitazone 301, 303, 430–1 troglitazone sulphate 434–5 exon 1 410 expressed sequence tags (ESTs) 450–1 extracellular matrix (ECM) 6–7, 141, 159, 379–81 apoptosis 243 cellular stress 239 cytokines 372, 379–81 DNA arrays 501 fibrosis 141, 202–6 genomic profiling 470 in vitro prediction 144, 146–51, 152 Kupffer cells 321 lipids 241 sinusoidal cells 341, 344–5, 347, 351 extracellular signal-regulated kinase (ERK) 238, 373–4, 502 bile acids and apoptosis 399–402, 404, 406 fibrosis 204, 205 troglitazone 305 extrahepatic biliary atresia (EBA) 125 extrahepatic cholestasis 125, 196, 206, 246 FACS analysis 144 F-actin 241

652

Index

FADH2 234 falciform ligament 119 famotidine 593 farnesoid X receptor (FXR) 207, 211, 216, 407–8, 440 cholestasis 127, 196–7 Fas 40, 200, 206, 208, 247 apoptosis 243, 245–6 bile acids and apoptosis 394–5, 398, 400, 402, 404–5, 411 sinusoidal cells 348, 353 Fas-associated protein with death domain (FADD) 243, 394–5, 398–9 Fas ligand (FasL) 208, 247, 348, 353 apoptosis 243, 245, 246, 394–5, 398, 405 fat 8, 224, 240, 321, 378, 391 cytokines 372, 377–8 sinusoidal cells 344, 350 fatty acids 38, 121, 219, 240, 377–8, 503 apoptosis 246, 391 beta oxidation 455 binding protein (FABP) 158 carbon tetrachloride 350 genomic profiling 470 mitochondria 230 omega-hydroxylase 211 oxidation 146, 153, 199, 470 oxidation inhibitors 230 PPARs 211, 213 steatosis 199–201 synthase (FAS) 158 toxicogenomics 458, 460 transgenic models 224 fatty acyl CoA oxide (FACO) 8–11, 13–14, 16, 215 fatty liver 5, 109, 199–200, 378, 393, 580 alcohol 129 gender differences 545 oncology drugs 564 see also steatosis Fc receptor 342, 344 felbamate 34, 80–2 fenbufen 455, 473 fenestrae 313, 344, 348 fenofibrate 33, 36, 163–4, 213 Fenton reaction 239, 328 ferryl species 328, 540 fetal cells see embryos fever 128, 193, 201, 592, 597 fialuridine 38

fibrates 212 fibril 351–2 fibroblastic growth factors (FGFs) 203, 343–4, 353 fibroblasts 20, 24, 29, 150–1, 202–6 bile acids and apoptosis 404 cholestasis 198 cytokines 247, 372, 380 DNA arrays 503 Kupffer cells 316 lipids 241 methapyrilene 577 sinusoidal cells 344, 352 fibrogenesis 239, 345, 352, 354, 379–80 fibrolysis 379 fibromyalgia 539 fibronectin 202, 205, 239, 343, 345, 351–2 fibrosis 8, 141, 202–6, 208, 379–81 alcohol 129–30 apoptosis 246, 393 cellular stress 237, 239, 242 chemokines and cytokines 246, 372, 379–82 cholestasis 126, 194, 197–8 experimental models 121, 123–4, 126, 129–30 gender differences 551 genomic profiling 469 Kupffer cells 316, 319, 322 lipids 241–2 mitochondria 230 sinusoidal cells 345, 350, 351–3 steatosis 199 fish 474, 540 Fishrer’s exact test 535 five-lipoxygenase activating protein (FLAP) 322 flavin-containing monooxygenase (FMO) 72, 184, 497 flavin-dependent monooxygenase 92 flavoenzymes 287, 290 flavone 542–3 flavonoid myricetin 108 favonoids 209, 595 flavoprotein P450 reductase 219, 292 FLICE-inhibitory protein (FLIP) 237 flucloxacillin 5 flufenamic acid 476 flunitrazepam 110 fluo4 33, 36 fluorescence 63–4, 154–5, 450

Index fluorogenic cDNA probes (TaqMan) 9, 11–12, 14, 17–18, 533 low density arrays 158, 163, 164 fluorofelbamate metabolite (F-CCMF) 81–2 fluorometric analysis 15 5-fluorouracil (5-FU) 252–3 fluoropyrimidine capecitabine 252 fluoroquinolones 229–30, 539 5-fluorouracil 551 fluoxetine O-dealkylation 74 flurbiprofen 74, 476 flutamide 34, 223 forkhead family 401 box subclass A (FOXA) 214–15 Fos 211, 236 free radicals 39, 121, 124, 409–10, 476 carbon tetrachloride 350–1 cellular stress 234, 239 Kupffer cells 319, 320, 327, 329 fumonisin B1 (FB1 ) 110, 178, 181–4, 248 fumonisins 178–9, 181, 183 functional gradients 469–70 fungi and fungicides 178, 218, 473, 476, 545 fungizone 148 furan 595, 617 fuructose-bisphosphate adolase 63 Fusarium verticillioides 178–9, 181, 182, 184 fusidate 422, 424, 438 GADD (growth arrest and DNA-damage) 157–8, 216, 306, 475, 493 gadolinium 455 gadolinium chloride 195, 347–8, 377, 379–80 Kupffer cells 316–18, 320–5 galactosamine 226, 248, 473, 477, 524 gall bladders 119, 392 gamma glutamyl cysteine synthetase 220, 235, 455 gamma glutamyl transaminase (GGT) 181, 182, 440, 458 gamma glutamyl transpeptidase 546 gastrointestinal (GI) system 105, 119, 125, 130, 476 bacterial endotoxin 315, 319–21, 476 bile acids 391, 392 botanical supplements 592, 598 cytokines 377 drug metabolism and interaction 76 experimental models 119 FXR 211

653

gender differences 541, 552 micro-organisms 4, 69 oncology drugs 565 PBPK models 609 quinone 291 GCRMA 452 gelatinase 345 gemfibrozil 212, 456, 473 gender 4, 70, 107, 110–11, 120, 183, 539–55 alcohol 129, 552 DNA arrays 494 genomic profiling 467, 470 intrahepatic cholestasis 430–2 toxicogenomics 452 Genedata Expressionist Analyst 155–6, 160 gene expression 13, 60–1, 76, 155–60, 162–6, 475, 492–7 acute phase reaction 248 bile acids 391, 401, 407, 410 biomarkers of mycotoxins 177, 180, 184 botanical supplements 593 cellular stress 235 DNA arrays 489–90, 492–7, 498–504 drug discovery 60–1, 62, 66–7 drug safety 96–8 fibrosis 204–6 food-related hepatotoxins 107, 109–10 genomic profiling 465–77 in vitro prediction 142, 144, 147, 152, 153 Kupffer cells 315, 323, 326–7 mechanisms of TLI 210, 219, 229, 253 regeneration 250–1 sinusoidal cells 345, 348 TGP database 508–10, 512–27 toxicogenomics 450–4, 455–61 GeneLogic ToxExpress 459 Gene Ontology (GO) Slim 493 genes and genetics 4, 6–7, 9, 13–17, 26, 302–3, 440–1 acute phase reaction 247 alcohol 376 bile acids and apoptosis 398, 401, 406–10 biomarkers of mycotoxins 177, 179, 180, 184 botanical supplements 592–3, 600 cellular stress 235, 236–7, 239 chemokines and cytokines 247, 374–6, 378, 381 cholestasis 195, 196, 198 DNA arrays 490, 492–6, 498–504 drug discovery 57, 60–1

654

Index

genes and genetics (Cont.) drug safety 96–9 endpoints of culture analysis 154 gender differences 539–55 genomic profiling 465–8, 470–7 intrahepatic cholestasis 423, 437–8, 440–1 Kupffer cells 327, 329, 352 lipids 240 mechanisms of TLI 207–9, 210–15, 220–8 methapyrilene 577, 583 oncology drugs 569 polymorphisms 4, 13, 70, 251–3, 303, 531–5, 549 regeneration 250–1 RNAi 39–40 silencing 40, 62, 327 sinusoidal cells 342, 346, 348–9, 352–3 stem cell-derived hepatocytes 19, 21–3 survival and repair 249 TGP database 507, 509–11, 513–27 toxicogenomics 450–4, 455–6, 458–61 transgenic models 223–8 troglitazone 299, 302–3, 306, 307 GeneSpring 450 genetic algorithm(K-nearest neighbor (GA(KNN) 456 genetically engineered cells 23, 24–7, 41 genomics and genomes 4, 177, 180–1, 184, 449–61 bile acids and apoptosis 397 drug discovery 57 endpoints of culture analysis 159, 161 Kupffer cells 327 methapyrilene 583 PBPK models 607 profiling of liver injury 465–77 RNAi 39 TGP database 507–28 tuberculosis 535 genotoxicity 177–9, 181, 221, 224, 577–8 botanical supplements 595 methapyrilene 576, 577–8, 583 geranyltransferase 503 gerbils 120, 128 germander 109, 590, 594, 595 Gilbert’s syndrome 251, 253 glial cells 93 glibenclamide 26, 95, 125, 195, 425–6, 428, 457

intrahepatic cholestasis 422, 424, 426, 428–9, 436, 439–40, 442 gliclazide 427 glipizide 426, 428 glitazars 214 glitazones 83, 213, 439–40 global normalization 510 glucocorticod receptor (GR) 210–11, 218, 350, 543 antagonist 209, 210, 407 bile acids 407, 409 glucocorticoid sulfotransferase 545–6 glucocorticoids 148, 151, 77, 214, 407 glucokinase (GCK) 164–5 glucose 28–9, 213, 299–300, 470, 474, 551 glucose-6 phosphate 21, 164–5, 214 glucuronic acid 218, 287, 290, 591 glucuronic sulfate 290 glucuronidation 217, 221, 251, 253, 501 botanical supplements 596–7 gender differences 541, 546, 550–1 troglitazone 300–2 glucuronide 38, 221, 346, 429–30, 434, 550 glucuronosyl transferase 234, 422, 546, 550 glue sniffing 545 glutamate 406 glutamate-cysteine ligase 235, 455 glutamate dehydrogenase 63, 64 glutamine 8 glutamine synthetase (GS) 21, 476 glutathione (GSH) 29, 211–13, 220–1, 546 APAP 346, 349 apoptosis 245, 400 botanical supplements 595, 597 cellular stress 234–7 chemokines and cytokines 247 cholestasis 195, 196 cytokines 377 depletion 6, 29, 33, 39, 154, 235, 237, 240–2 DNA arrays 494, 501 drug safety 95 endpoints of culture analysis 154 experimental models 124–5 food-related hepatotoxins 110 gender differences 546 genomic profiling 470, 474, 476 high content screening 33 intrahepatic cholestasis 433, 437 Kupffer cells 322 lipids 240–2

Index mitochondria 231–2 PBPK models 608, 615, 618–19 quinone 287–90 reactive metabolites 80–2 steatosis 200 TGP database 515, 521, 524–5 toxicogenomics 455, 460 transgenic models 223–8 troglitazone 302, 433 glutathione disulfide (GSSG) 125, 235, 289, 290, 546 glutathione disulfide reductase 235, 546 glutathione peroxidise (GPX) 63–4, 148, 226, 289, 502 cellular stress 234 gender differences 546 lipids 242 glutathione reductase 234, 494, 521–2 glutathione-S-transferase (GST) 8–11, 13–14, 16, 25, 184, 207, 211 acute phase reaction 248 alpha 158 cellular stress 234 DNA arrays 501, 502 drug metabolism and interaction 72 gender differences 545, 546 genomic profiling 475 in vitro prediction 146, 153 PXR 209 toxicogenomics 455 transgenic models 226, 228 troglitazone 302–3 glutathione synthetase 72, 81, 92, 235 glyburide 426 glyceraldehyde-3-phosphate dehydrogenase (GAPDH) 220 glycidamide 225 glycine 193, 316, 321–2, 326, 392, 404 glycochenodeoxycholate 398, 402 glycochenodeoxycholic acid 397, 401, 404 glycocholate 439 glycogen 11, 474, 575 glycogen phosphorylase 64 glycolysis 474 glycooxidation 238 glycoproteins 344, 351, 373 glycosides 595 glycosis 220 glycoursodeoxycholic acid (GUDCA) 402 Golgi complex 245, 398

655

Gotu kola 598 gout 539 G-protein 371, 400, 503 granulocytes 322 granuloma 565 grapefruit juice 218 GRE 409 greater celandine 590, 594, 598 green tea 347, 565–6, 594–5, 598–9 growth factors 20, 106, 148, 373–4, 400, 578 growth hormone (GH) 13, 215 gender differences 541, 544, 546, 547, 549, 555 guanine 179, 239 guanosine 239 guanylate cyclase 497 guinea pigs 120 gummiferin 597 gum thistle 594 haloalkanes 121, 122, 350 halothane 5, 95, 192, 219, 201–2, 221 gender differences 548, 552 Ham’s medium 77 hamsters 120, 128, 327, 577 hapten hypothesis 93 haptenization 429, 592 heart see cardiotoxicity heat maps 156, 521–2, 524–5 heat shock 157, 206 proteins 63, 158, 207, 493 heavy metals 78, 107, 236, 322, 468 HeLa cells 62, 404 heliotrope (Heiliotropium) 594–5 hemangiosarcoma 328–9 hematopoietic cells 20 heme complexes 121, 195, 236, 249–50, 291 heme oxygenase (HO) 235, 236, 247, 249–50 DNA arrays 497, 502 sinusoidal cells 343, 352 TGP database 514–16, 525 hemochromatosis 239, 379 hemoglobin beta chain complex 519 hemolysis 328–9 hemoprotein peroxidases 234 hemosiderin 314, 326, 328 heparan 351 heparan sulfate proteoglycans 7 HepaRG cell line 27, 153, 157 heparin 353, 592

656

Index

hepatectomy 123, 193, 229, 323–5, 373–4, 477 Kupffer cells 320, 323–5, 328, 329 regeneration 4, 250, 373 hepatic arteries 119, 130, 203, 314, 468 hepatic nuclear factors (HNFs) 157, 158, 164–5, 210, 214–15 cholestasis 195 fibrosis 205 hepatic stellate cells (HSCs; Ito cells) 22, 204–6, 215, 217, 344–5, 352–3 alcohol 130 apoptosis 245–6, 393 carbon tetrachloride 352–3 cellular stress 239 cholestasis 198 cytokines 372, 374, 379–80, 381–2 DNA arrays 495 fibrosis 202, 203–6 genomic profiling 469–70 in vitro prediction 147, 151 Kupffer cells 313–14, 319, 322, 325, 342, 352 lipids 241 sinusoidal cells 342–3, 344–5, 351, 352–4 hepatitis 201–2, 246, 375–6 alcohol 321, 376, 394 bile acids and apoptosis 393–4, 404 botanical supplements 592, 595–6, 597, 598–9 cellular stress 239 chemokines and cytokines 246 cytokines 374–6, 379, 381 gender differences 541, 549, 552 genetic polymorphism 251 genomic profiling 472 in vitro prediction 141 isoniazid 531, 534–5, 552 Kupffer cells 321 mechanisms of TLI 221, 229 transgenic models 223 troglitazone 304, 436 see also immune hepatitis; viral hepatitis hepatitis B 179–80 hepatitis C 108, 110, 307, 375–6 hepatoblastoma 108 hepatocyte growth factor (HGF) 20, 250, 373–4 Kupffer cells 316, 320, 322, 325, 351 sinusoidal cells 342–3, 349, 351, 353 hepatocytes 4, 141–66, 468–9 acute phase reaction 248

APAP 347–9 apoptosis 242, 244–6 application of cultures 161–6 bile acids and apoptosis 391, 393–409, 411 botanical supplements 598 carbon tetrachloride 351–3 chemokines and cytokines 247 cholestasis 193–8 cytokines 372–8, 380–1 DNA arrays 492, 494, 495–6, 498–502 drug metabolism and interaction 69–84 drug safety 91–8, 100 endpoints for analysis of cultures 153–61 experimental models 122–3, 125–7, 129 fibrosis 202–6 food-related hepatotoxins 105–11 gender differences 541–2, 544, 547, 549–50, 553, 555 genetic polymorphism 252 genomic profiling 468–70, 474 high content screening 30–7 intrahepatic cholestasis 422, 429, 431, 436–8, 440 in vitro prediction 142–53 Kupffer cells 313–28, 342 lipids 240–1 mechanisms of TLI 210–16, 220–1, 229 methapyrilene 575, 577–80, 582–3 mitochondria 230, 231, 232 necrosis 242 oncology drugs 564, 567 PBPK models 619–20 quinone 287–8, 291–3 regeneration 250 sinusoidal cells 341–2, 344–5, 354 steatosis 200 stem cell-derived 18–24 TGP database 509–12, 515, 524 three dimensional bioreactors 27–30 TNF-á 249 toxicogenomics 457–9 transgenic models 224 troglitazone 301–4 see also parenchymal cells; primary hepatocytes hepatoma 24–5, 153, 157, 215, 375 bile acids and apoptosis 402, 409 cell line 27, 153, 157 DNA arrays 491, 493, 496, 499–501, 504 intrahepatic cholestasis 434

Index troglitazone 304, 306, 434 see also HepG2 cells hepatomegaly 208, 210, 223, 476, 597 hepatosis in horses 181 hepatosteatosis see steatosis HepG2 cells 153 24, 25–6, 37, 83–4, 496–7 bile acids and apoptosis 404 biomarkers of mycotoxins 180 botanical supplements 596 DNA arrays 491, 493, 499–500, 504 drug discovery 61 drug safety 92 gender differences 550 hepatocyte cultures 107, 108, 109 high screening content 31–3, 36 toxicogenomics 454, 460 troglitazone 306 see also hepatoma herbal products 109, 111, 421, 589–91, 595, 599–600 herbicides 231 HE staining 520 heterodimers 207, 210, 214, 216, 375, 543 nuclear receptors 210, 211 hexachlorobenzene 223, 511 hexobarbital 540 hierarchical algorithms 453 high content analysis (HCA) 30, 38 high content assays 96–7, 102 high content imaging (HCI) 154, 166 high content screening (HSC) 25–6, 30-7, 41, 142, 154, 166 high mobility group box 1 (HMGB1) 350 high performance liquid chromatography (HPLC) 15, 73, 155, 179 high resolution analysis 30 high throughput screening 144–6, 466, 471, 152, 567, 570 high throughput toxicology (HTP-Tox) 57, 67 highly active antiretroviral therapy (HAART) 233 histamines 575, 592–3 histone deacetylase 352, 503 HIV 209, 233, 535 HIV protease 437–8 HMG-CoA reductase inhibitors 212, 476 HMG-CoA synthase 64, 211 Hoechst stain 32, 36, 403 hollow fibre reactors 28–9, 152 homodimers 214

657

homogenates 153, 510 hormesis 33 hormone replacement therapy 126, 195, 551 hormones 78, 106, 148, 215, 452, 511 bile acids 392, 407–8 cholestasis 196 cytokines 378, 381 DNA arrays 492, 493–4 gender differences 540, 543, 545–6, 549, 551, 555 methapyrilene 578 see also growth hormones; steroid hormones horses 120, 181, 541 HSP 227, 408–9, 455 5-HT6 receptor antagonists 458 HUH-7 cell line 27, 404, 503–4 Human-600 array 458 human embryonic stem cells (HESCs) 19, 22 humans 3–5, 71–3, 77–9, 142, 183, 494–6, 547–52 alcohol 129–30 apoptosis 244, 245 bile acids 392, 397, 402, 406–7 biomarkers of mycotoxins 177–81, 183, 184 botanical supplements 591, 593, 595 cholestasis 125–6, 196 cytokines 247, 374, 376, 377, 379–382 DNA arrays 489–95, 496, 498–500, 503 drug discovery 61–2, 64–5 drug-induced liver injury 3–30 drug metabolism and interaction 70–3, 75, 77–9, 81–4 drug safety 89–102 endpoints of culture analysis 154, 156 experimental models 119–21, 123–30 food-related hepatotoxins 106–11 gender differences 539–55 genomic profiling 465–6, 468, 475 hepatitis 201 high content screening 30–7 intrahepatic cholestasis 426, 429, 431, 433–41 in vitro applications 30–40, 161–4, 166 in vitro prediction 142, 143–8, 152, 153 Kupffer cells 321, 329, 342 mechanism of TLI 207–8, 210, 211, 215–16 methapyrilene 579 mitochondria 233 oncology drugs 563–4, 567–8, 570 PBPK models 607, 614, 616–20, 622

658

Index

humans (Cont.) sinusoidal cells 342 TGP database 507–9, 511–12, 528 toxicogenomics 454, 458–9 troglitazone 300–4, 307 huntingtin gene 410 Huntington’s disease (HD) 409–10 hyaline 124, 377, 554–5 hyaluronic acid 344, 351 hyaluronin 503 hydralazine 34 hydrazine 457 hydrocarbons 540 hydrocortisone (HC) 7–8, 16–17 hydrogen peroxide (H2 O2 ) 200, 204, 216, 220, 232, 247 cellular stress 234, 238, 239 Kupffer cells 315, 328, 342 quinone 289–91 sinusoidal cells 342, 345, 354 hydrolases 217, 243 hydrolytic enzymes 353 hydrophilics 217, 539 bile acids and apoptosis 246, 391–2, 397, 399, 402, 407–8, 411 hydrophobic bile acids 422 hydroquinone 290, 292, 545 6-hydroxychromane moiety 434 8-hydroxydeoxyguanosine (8-oxoG) 108, 232–3 6-hydroxydopamine 410 hydroxyethyl radical 204 hydroxyl groups 392 hydroxyl radicals 234, 239, 289–90, 315, 328 hydroxylases 392 hydroxymethylglutary-CoA synthase 64 hydroxynonenal (HNE) 38, 122, 204, 232, 240, 249 hydroxysteroid sulfotransferase (ST) 10–11, 545–6 hydroxytamoxifen 221 hyperbilirubinaemia 251, 424, 438 hyperforin 209 hyperlipidaemia 213 hypermethylation 576 hyperplasia 123, 210, 245, 457, 554, 569 Kupffer cells 323, 326, 329 hypertension 123, 352, 495 hypertriglycerinaemia 199 hypertrophy 210, 521, 524

hypoglycaemic effect 301, 551 hypolipidaemic effect 301 hypomethylation 122–4, 576 hypophysectomy 540, 541, 546 hypoxanthine guanine phosphoribosyl transferase (HGPT) 494, 577 hypoxia 127, 130, 321, 378, 593 hypoxia-inducible factor (HIF) 207 Hy’s law 568 ibufenac 80 ibuprofen 34, 455, 473, 550 idiosyncratic hepatotoxicity 217, 221, 229, 251 botanical supplements 591–3, 597, 600 drug safety 4, 93, 96 high content screening 31 intrahepatic cholestasis 429, 436 Kupffer cells 317 oncology drugs 564, 567 toxicogenomics 449 troglitazone 299–300, 303–4, 307 IFNAR 375 ifosfamide 40 IãB – kinase (IKK) 237, 401 image analysis 30 imidazol 251 imipramine 32, 34, 95, 547 immune hepatitis 201–2, 219, 221, 246 apoptosis 246, 404 chemokines and cytokines 246 drug-related liver injury 251 immune system 4–5, 125, 129, 141, 201–2, 208 apoptosis 244, 246, 393 botanical supplements 592 cytokines 371–2, 374–5, 378 DNA arrays 493, 499 drug metabolism and interaction 70 drug-related liver injury 251 drug safety 93 fibrosis 202–3 gender differences 548 intrahepatic cholestasis 421, 429 Kupffer cells 315, 342 mechanism of TLI 191–2, 193 oncology drugs 564, 567, 569 sinusoidal cells 342, 344, 345 TNF-á 248 transgenic models 223 troglitazone 301, 304

Index immune thrombocytopenic purpura 539 immunoaffinity chromatography 179 immunochemical studies 128 immunogenicity 238 immunoglobulin heavy-chain binding protein (BiP) 306 immunohistochemisty 22, 324, 469 immunoprecipitation 79 immunosuppressants 195, 251–2, 347, 422, 566, 590 impila 594 in silico prediction system 63, 67 in vitro studies 3–41, 142–53, 229, 492–500, 567 apoptosis 245, 246, 400, 410 applications of cultures 161–6 applications of models 5, 30–40 biomarkers of mycotoxins 180 botanical supplements 592, 596–7, 598 cellular stress 239 culture conditions 500–4 cytokines 373, 375, 380 DNA arrays 490–504 drug discovery 57–67 drug-induced liver injury 3–41 drug metabolism and interaction 71–84 drug safety 89–102 endpoints of culture analysis 153, 154–61 experimental models 120 fibrosis 205 food-related hepatotoxins 105–11 gender differences 543 genetic polymorphisms 252 genomic profiling 469 intrahepatic cholestasis 422–3, 425–7, 429, 431–2, 434, 436–9, 441–2 Kupffer cells 326–9 lipids 240 media changes 504 methapyrilene 576–8, 580 oncology drugs 564–5, 567, 570 PBPK models 607, 611–12, 616, 624 TGP database 507, 510–12, 528 toxicogenomics 457–9 troglitazone 300–1, 307 in vitro–in vivo strategy (IVIVS) 91 Invitrogen 491 in vivo studies 24–9, 58–60, 229, 497–500, 568–9 APAP 348 applications 30, 37, 39–41, 161, 162–6

659

bile acids and apoptosis 400, 402, 406 biomarkers of mycotoxins 180 botanical supplements 592, 596–7 cellular stress 239 cytokines 373–4, 380 DNA arrays 490, 496, 497–500, 504 drug discovery 58–60, 61, 63–5 drug-induced liver injury 6–9, 11, 16, 20, 24–9 drug metabolism and interaction 71–3, 75, 83 drug safety 91–3, 97, 101–2 endpoints of culture analysis 157, 159, 161 experimental models 120, 128 food-related hepatotoxins 105–11 gender differences 542, 547, 550 genomic profiling 469, 474 intrahepatic cholestasis 423, 425–7, 429, 431–2, 438, 441–2 Kupffer cells 316, 322, 326–8 lipids 240 methapyrilene 575–6, 583 nuclear receptors 208, 210 oncology drugs 568–70 PBPK models 607, 611–12, 624 prediction 142–4, 146–7, 153 regeneration 250 TGP database 509–12, 520, 528 toxicogenomics 459–61 troglitazone 300, 303, 304, 307 Incyte Pharmaceutical 491 Indian ink 324 indomethacin 34 inducible NO synthetase (iNOS) 247, 316, 323 infections 4, 8, 178, 372, 374, 524, 591–3 tuberculosis 531, 535 see also bacteria and bacterial infections inflammation 4, 38, 124, 152, 201–2, 342–4, 354 acute phase reaction 248 alcohol 129 amebiasis 128 APAP 346–51 apoptosis 243, 245 bile acids and apoptosis 393, 395–6, 401, 407 botanical supplements 591–3, 596 carbon tetrachloride 350, 353 cellular stress 236, 237, 239, 242 chemokines and cytokines 246–7, 371–2, 374–7, 379, 381 cholestasis 126, 194, 196

660

Index

inflammation (Cont.) DNA arrays 503 drug discovery 59, 61 drug-related liver injury 251–2 drug safety 94–7, 99 fibrosis 202–6 gender differences 547, 548 genomic profiling 474 hepatitis Kupffer cells 314–17, 321, 323, 342–4 lipids 240–2 mechanisms of TLI 192–3, 208, 210–17, 220, 222, 229, 253 mitochondria 230, 232 oncology drugs 566 regeneration 250 sinusoidal cells 341–5, 354 steatosis 199 stem cell-derived hepatocytes 18 survival and repair 249 TGP database 511, 524 TNF-á 248–9 toxicogenomics 454–5, 457 transgenic models 226 troglitazone305 infliximab 381 inhibitors of apoptosis proteins (IAPs) 237, 243–4, 395, 397 inosine nucleotide 252 inosine triphosphate pyrophosphastase (ITPase) 252 insulin 6–8, 77, 148, 345, 378–9 bile acids and apoptosis 400, 402 intrahepatic cholestasis 429 liver-enriched transcription factors 214 PPARs 213–14 troglitazone 299 insulin growth factor-I (IGF-I) 380, 475 insulin-like growth factor 497, 502 insulin resistance 199–200, 213–14, 299, 377–9 Integrated Discrete Multiple Organ Co-culture (IdMOC) 97, 100–1, 102 intercellular adhesion molecules (ICAMs) 194, 249, 344 Kupffer cells 319, 321 sinusoidal cells 344–5, 348, 350–1 interferon (IFN) 342–4 interferon-á (IFN-á) 375, 381 interferon-â (IFN-â) 372 interferon-ã (IFN-ã) 129, 215, 222, 375–6

cytokines 372, 375–6, 380–1 Kupffer cells 316 sinusoidal cells 345, 349 TNF-á 249 interferon-ã inducible protein 10 (IP-10) 343, 349 interindividual differences 4, 106–7, 192, 616–17, 618 biomarkers of mycotoxins 182 DNA arrays 490, 494–6, 498–9 gender 548–9, 553–5 PBPK models 616–17 troglitazone 303 interlaminar fibrosis 124 interleukin-1 (IL-1) 215, 220, 222 APAP 346 cytokines 371–2 fibrosis 203, 205 Kupffer cells 315–16, 342 sinusoidal cells 342–6 TNF-á 248 interleukin-1á (IL-1á) 315, 319 interleukin-1â (IL-1â) 210, 222, 343, 375, 401 cholestasis 194–5 Kupffer cells 315–16, 319, 320, 323, 325, 351 interleukin-2 (IL-2) 222 interleukin-4 (IL-4) 129, 222, 248 interleukin-5 (IL-5) 129, 222 interleukin-6 (IL-6) 129, 210, 215, 229 cholestasis 194–5 cytokines 373, 376, 378, 381 fibrosis 203, 205 Kupffer cells 315–16, 20, 324–5, 344 sinusoidal cells 342–5, 348, 349, 351 interleukin-8 (IL-8) 222, 247, 316, 342, 376–7 interleukin-10 (IL-10) 129, 228 cytokines 372, 375, 376, 380–1 Kupffer cells 320, 325, 342, 351 sinusoidal cells 342–3, 345, 347, 351, 353 interleukin-12 (IL-12) 129 interleukin-13 (IL-13) 222, 343, 351, 372 interleukin-15 (IL-15) 129 interleukin-18 (IL-18) 129, 342, 376–7 interleukin-22 (IL-22) 129 intermediate acetylators (IAs) 532, 534–5 International Agency for Research on Cancer (IARC) 178, 181

Index intestinal matrix 202 intrahepatic cholestasis 38, 125–6, 196, 210, 303, 422 BSEP 421–42 drug induced 422-39, 441–2 genetic origin 440–1 pregnancy (ICP) 437 ionophores 316 Iproniazid 80 irinotecan 252–3 iron 192, 239, 249, 379, 579, 582–3 apoptosis 393 Kupffer cells 315, 328–9 lipids 239–40 mitochondria 231, 232 steatosis 199 transgenic models 223 irritable bowel syndrome 539 Isaxonine 80, 95 ischaemia and reperfusion 120, 127–8, 229, 374 acute phase reaction 248 apoptosis 242, 393, 402, 409 cholestasis 195 necrosis 242 steatosis 200 TNF-á 248 isocyanates 236 isoflurane 614 isolated perfused liver 142, 143, 144, 146–7 isolated perfused rat liver (IPRL) 430–1, 438 isoniazid (INH)34, 193, 218, 221, 241 gender differences 531–5, 548, 552 TGP database 510, 515 isotropes 433 isozymes and isoenzymes 26–7, 291, 431 itraconazole 218 Ito cells see hepatic stellate cells (HSCs) Janus kinase (JAK) 373, 375–6 Japan Pharmaceutical Manufacturers Association 507–8 jaundice 38, 251, 424, 436, 597, 598 jin bu huan 594, 597 jun 211, 236, 402 jun N-terminal kinase (JNK) 200, 222, 240, 253, 402 bile acids and apoptosis 397, 401–2 cellular stress 235, 237–8 troglitazone 305

661

kappa beta kinase 378 kava 109, 590, 594, 596 Keap-1 236 kepones 621 keratinocyte-derived chemokine (KC) 343, 349 keratoses 597 ketoconazole 5, 218, 292–4, 541, 542 ketones 621 kidneys 7, 37, 76, 229, 497 botanical supplements 590 drug safety 92, 93, 100 FXR 211 gender differences 539, 542, 545, 554–5 genomic profiling 467 Kupffer cells 321, 324 lipids 241 mitochondria 231 oncology drugs 563, 565 PBPK models 608 quinone 291 TGP database 509–12 transgenic models 223, 224, 227 troglitazone 304 Ki-M2R 314 kinases 205, 235, 236–7, 373, 375–6, 378 Klebsiella pneumoniae 566 Krk 396 Kupffer cell receptor (KCR) 316, 324 Kupffer cells (KCs) 7, 313–29, 342, 344–5, 346-8, 351–2 alcohol 130 apoptosis 245, 393, 396 botanical supplements 592–3 cholestasis 195 ConA 129 cytokines 372–3, 376–7, 379–80, 381 DNA arrays 495, 498 fibrosis 203–4, 206 genomic profiling 469, 474, 476 hepatitis 201 in vitro prediction 147, 151, 152 mechanisms of TLI 214, 217, 220, 229 oncology drugs 566 regeneration 250 TNF-á 249 lactacidaemia 233 lactate dehydrogenase (LDH) 30, 59, 78, 154, 196, 293, 512 lactic acidosis 233

662

Index

lactoferrin 344 lamda-glutamyl-transpeptidase 576 lamellar bodies 230 laminin 7, 148, 202, 351 lamotrigine 5 lanosterol 503 lansoprazole 78 laser capture microdissection 469 lauric acid 11-hydroxylation 74 laxatives 599 LBD 407, 409 LDL cholesterol 212 learning vector quantization 453 leflunomide 34 leptin 378–9, 381 leucin zipper (bZIP) 214–15, 236 leukaemia 545 leukocytes 126, 319, 341, 344, 345, 354 leukocytopenia 224 leukoencephalomalacia 181 leukotrienes 315, 319, 438, 434–5, 497 leupeptin 578–9 levodopa 550 levo-tetrahydropalmatine 597 lidocaine 5 ligand activated transcription factors 207, 219 ligands 299, 315, 440, 543–4, 579, 583 apoptosis 243, 394–5, 398–9, 404, 407–9 cytokines 374, 375 mechanisms of TLI 206–8, 210, 211, 213, 218–19 limonene 542 lindane 326, 328 linear discriminant analysis (LDA) 456 linearized multistage (LMS) 618 linoleic acid 77 lipid aldehyde 205 lipid hydroperoxides 289 lipidomics 159 lipidosis 96 lipid peroxidation 38, 121–2, 193, 235, 238, 239–42 bile acids and apoptosis 400 botanical supplements 597 cellular stress 239–42 experimental models 121–2 fibrosis 204–5, 379 food-related hepatotoxins 108 gender differences 545

genomic profiling 476 intrahepatic cholestasis 434 Kupffer cells 319, 321–2, 328 methapyrilene 576 mitochondria 232 quinone 290 sinusoidal cells 345, 350–1 steatosis 201 TNF-á 249 transgenic models 227 lipid peroxyl radical 234 lipids 5, 6, 38–9, 239–42, 354, 503 bile acids and apoptosis 392, 397, 404 carbon tetrachloride 350–1 cellular stress 238, 239–42 cholestasis 193 drug safety 95, 96 endpoints of culture analysis 154 experimental models 121, 122 fibrosis 203–5 food-related hepatotoxins 108, 109 genomic profiling 467, 470, 474, 476 in vitro prediction 142 Kupffer cells 317 mechanisms of TLI 191–2, 211–13 methapyrilene 575, 580 mitochondria 231, 232–3 oncology drugs 566 PBPK models 612, 614 quinone 290 steatosis 199–201 TGP database 520 transgenic models 224 troglitazone 299, 307 lipocytes 344 LipoKinetix 109, 598 lipolysis 474 lipophility 229, 422 lipopolysaccharide (LPS) 222, 229, 473–4, 476–7 acute phase reaction 247–8 botanical supplements 592–3 cytokines 376–7 food-related hepatotoxins 110 Kupffer cells 315, 316, 323, 326, 329 TGP database 524 TNF-á 248–9 lipoproteins 109, 122, 344, 579 liposomes 40, 347, 566–7 Kupffer cells 316–18, 320, 324–5

Index liquid chromatography combined with mass spectrometry (LC-MS) 72, 81–2, 95 lisinopril 435 Listeria monocytogenes 107, 108 lithium 550 lithocholic acid (LCA) 197, 210, 216, 392, 407, 457 liver-enriched transcription factors (LETFs) 214–16, 499–500 liver failure 3–4, 141, 213, 221, 229, 253 APAP 346 apoptosis 245, 393 botanical supplements 596–7, 598 drug metabolism and interaction 69 drug safety 93 experimental models 121, 124 gender differences 110, 551 genetic polymorphisms 251 Kupffer cells 317 oncology drugs 563 reactive metabolites 81 TNF-á 248 toxicogenomics 449 liver receptor homolog (LRH) 127 liver slices 71, 73, 142, 143, 144, 146 DNA arrays 489–92, 497–8 drug discovery 58, 60–1 endpoints of culture analysis 155–6 oncology drugs 567 liver X receptor (LXR) 196, 211, 216, 219 LLC-PK1 439, 441 long-chain fatty acids 38, 248 lovastatin 422 lowest observable adverse effect level (LOAEL) 34–5, 36, 553, 618 LPO 228, 241 lungs 76, 100, 231, 467, 497, 577 gender differences 539, 543, 545, 547 Kupffer cells 324 PBPK models 609, 618 quinone 291 lupus 539 lymph node hyperplasia 123 lymphoblastoma 225 lymphocyte function-associated antigen-3 (LFA-3) 344 lymphocytes 76, 245, 291, 316, 345, 371–2 lymphoma 225. 577 lysine 180, 201 lysis 242, 393, 582

663

lysosomal cysteine protease 200, 398 lysosomal enzymes 128, 230, 554 lysosomal hydrolases 230 lysosomes 6, 230, 253, 344 apoptosis 244, 398 high content screening 30–2 Kupffer cells 314, 315 lipids 240 methapyrilene 578–80, 582–3 necrosis 242 steatosis 199–201 Lysotracker Green (stain) 32 macrolide antibiotics 439, 543 macromolecules 95, 467, 474, 579–80, 608, 619 macrophage activator (MA) 455, 475–6 macrophage chemoattractant protein-1 (MCP-1) 343, 345, 346, 349 macrophage colony stimulating growth factor (M-CSF) 203 macrophage inhibitory protein-1 (MIP-1) 343, 346 macrophage inhibitory protein-1á 349 macrophage inhibitory protein-2 (MIP-2) 194, 247, 321, 342, 349 macrophages 221, 342-4, 346–8 APAP 346–50 bile acids and apoptosis 393–4 botanical supplements 592 carbon tetrachloride 351–2 chemokines and cytokines 247, 371–2, 379 DNA arrays 498–9 fibrosis 204 genomic profiling 474, 476 in vitro prediction 151 Kupffer cells 313–18, 321–2, 324, 326, 329, 346–8 lipids 240 oncology drugs 566 sinusoidal cells 342–4, 345, 354 TNF-á 249 Maf proteins 235, 236 ma huang 594, 598 malaria 287, 293 malonaldehyde (MDA) 38, 204, 240 mammals 7, 39–40, 62, 289, 396, 467 gender differences 540, 542 troglitazone 304, 306 manganese superoxide dismutase (MnSOD) 234, 289, 502

664

Index

MAP kinase (MAPK) 216, 222, 229, 235, 305, 373, 399 bile acids and apoptosis 399, 401, 402, 404, 406, 411 botanical supplements 593 cellular stress 235, 236, 237 fibrosis 204, 205 troglitazone 305 mass spectrometry 63, 81 mast cells 247 matrigel 6–9, 11, 13–15, 18, 77 DNA arrays 492, 496, 499, 500 in vitro prediction 147, 148, 151 matrix metalloproteinase (MMP) 229, 348 sinusoidal cells 341, 343, 345, 347–8 maximal velocity 610–12, 621 mCAR 550 Mcl-1 243, 396 Mdm-2 408–9 mechanisms of action 554–5, 570 mechanistic toxicology 454–5 Med Watch 600 megalocytes 120 megamitochondria 199, 219 MEK (MAPK and ERK kinase) 373 membrane n-6 polyunsaturated fatty acids 240 menadione 32, 225, 290, 455 mephenytoin 548 6-mercaptopurine (6MP) 251–2 mesenchymal stem cells(MSCs) 20–1 mesenteric fat lobulation 123 Met 374 metabolites 217–18, 219–21, 230, 251–2 bile acids 409 biomarkers of mycotoxins 177, 179 botanical supplements 592, 595, 597–9 cholestasis 127, 193–4, 422 DNA arrays 504 drug metabolism and interaction 69–73, 75 drug safety 90, 93, 94, 96, 97, 100 endpoints of culture analysis 155 gender differences 540 hepatitis 201 intrahepatic cholestasis 429, 431, 433, 437, 441 in vitro prediction 145, 148 Kupffer cells 328 methapyrilene 577, 578, 582 mitochondria 233

nuclear receptors 209–10, 213 oncology drugs 564, 568, 569 PBPK models 608, 616, 618, 620 quinone 291 sinusoidal cells 341 TGP database 508, 515, 524–5 transgenic models 223–4 troglitazone 300–4, 429–30 see also reactive metabolites metabolomics 459, 461, 569 metabonomics 96, 159, 184, 472, 569 drug discovery 58, 63, 64 metalloproteases 503 metallothionein 227, 228, 323 metals see heavy metals metformin 161–2 methacrylonitrile (MAN) 227 methanol 79 methapyrilene (MP) 457, 473, 520–1, 575–83 methide 433–4 methionine 224 methotrexate 551 methylcellulose 494, 512, 520 3-methylcholanthrene (3MC) 8, 13–16, 20, 148, 161, 163 drug metabolism and interaction 78 gender differences 543–4 methyl chloroform 608, 614 methyldopa 34, 550, 552 methylene chloride 553, 608 methylene dianiline 457 methyl methanesulfonate 216 4-(methylnitrosamine)-1-3-pyridyl)-1-butanone (NNK) 540 4-(methylnitrosamino)-1-(3-pyridyl)-1butanone 454, 460 methyl palmitate 316, 326 methylprednisolone 549 methyltransferase 72 metoprolol 495 metyrapone 542 MHC 201, 344 mibefradil 80 mice19, 21, 196–7, 404, 410 APAP 346–50 biomarkers of mycotoxins 181, 184 botanical supplements 595–7 carbon tetrachloride 352–3 cytokines 247, 373, 376, 378–80 DNA arrays 490–2, 493–4

Index drug discovery 65–6 drug safety 90 experimental models 120, 123, 128, 129 fibrosis 206 food-related hepatotoxins 107, 108, 110 gender differences 540–1, 544 genomic profiling 468, 474–5 intrahepatic cholestasis 439 Kupffer cells 314, 317–18, 320, 323–5, 327–8, 352 mechanisms of TLI 208, 211–13, 215–16, 221–2 methapyrilene 576, 580 oncology drugs 565–6 PBPK models 618–19 quinone 292 regeneration 250 RNAi 40 steatosis 200 TGP database 509 toxicogenomics 458, 460 transgenic models 223–8 troglitazone 306 mice (knockout) 200, 221–2, 223–8, 245 APAP 347, 349 carbon tetrachloride 350, 352, 353 cholestasis 194–5, 196, 197, 439 cytokines 247, 373–4, 376–7, 380 Kupffer cells 318, 320–1, 324, 327 mechanisms of TLI 208, 211–12, 215, 220–1 mice (severe combined immunodeficient, SCID) 128 mice (wildtype) 184, 200, 206, 380, 410 acute phase reaction 247 cholestasis 194, 195, 197 Kupffer cells 324 mechanisms of TLI 210–12, 215, 221 quinone 292 sinusoidal cells 349, 350 transgenic models 223–8 micelle 397 Michaelis constant 610–12, 621 microarray analysis 96, 109, 375, 494, 533–4 genomic profiling 465–6, 470–2, 475, 477 TGP database 508, 513, 525 toxicogenomics 450–3, 456–9 microbial pathogens and microbes 105–6, 108–11 micro-electromechanical system (MEMS) 29–30

665

microfabrication 28, 66 micro RNA (miRNA) 39–40, 307 microscopy analysis 8, 9, 11 microsomes 15, 90–1, 145, 201, 341, 455 botanical supplements 596 DNA arrays 496 drug metabolism and interaction 71–3, 75, 78–9, 81, 84 gender differences 541–5, 547–50 in vitro prediction 142, 145, 153 intrahepatic cholestasis 431, 433 Kupffer cells 322 lipids 240 methapyrilene 576 quinone 291, 293 troglitazone 304 microvesicular steatosis 193, 199 midazolam 1-hydroxylation 74 midecamycin 439 Mig 343, 349, 475 migraine 539 mineralocorticoid receptor (MR) 407, 409 minipigs 541–2, 544, 548, 555 minocycline 35, 552 mirex 621 MIT Liverclip 28 mitochondria 25, 38, 109, 230, 231–3, 305, 395–7 bile acids and apoptosis 243–6, 394–406, 408–11 botanical supplements 596–7 cellular stress 234–5 drug discovery 59–60 drug-induced liver injury 5–6, 11, 25, 28 drug safety 95, 98 endpoints of culture analysis 154 experimental models 121, 123, 124 gender differences 542 genetic polymorphism 252 genomic profiling 476–7 high content screening 30–6 in vitro prediction 141, 146, 153 intrahepatic cholestasis 422–3, 429 Kupffer cells 319, 322 lipids 240–1 mechanisms of TLI 191, 193, 219–20, 230, 231–3, 253 methapyrilene 575–6, 580, 582–3 necrosis 242 oncology drugs 569

666

Index

mitochondria (Cont.) quinone 287, 289 reactive metabolites 81 respiration 59, 200–1, 231–3, 234, 287, 289, 409 sinusoidal cells 346, 351 steatosis 199–201 troglitazone 305 mitochondrial DNA (mtDNA) 38, 199–200, 231, 233 high content screening 30, 31–2, 36 mitochondrial DNA polymerase gamma 31 mitochondrial permeability transition (MPT) 233, 241, 242 apoptosis 244, 245, 396–7, 399-400 mitochondrial permeability transition pores (MPT-pores)231, 244 mitogen activated kinase see MAP kinase (MAPK) mitogen-platelet-derived growth factor 204, 205, 401 mitotracker Deep Red (stain) 36 mitotracker Far Red (stain) 32 MMH-GH 153 modified Chee’s medium (MCM) 148 molybdenum-containing oxidase (Mo-CO) 72 monkeys 4, 182, 300 monoamine oxidase (MAO) 72, 92, 503 monocarboxylate 503 monochlorobimane 6, 39 monocrotaline (MCT) 593, 595 monocyte chemotactic protein-1 (MCP-1) 240, 249 monolayer culture 60, 77–8, 161–2, 441 DNA arrays 492, 497, 500, 503 endpoints of culture analysis 155–60 in vitro prediction 147–50, 151–2, 166 monomeric receptors 207 monooxygenases 157, 217–19, 221, 291 Moolgavkar–Knudson–Venzon (MVK) model 620 morphine 540, 550 mRNA 7–9, 12–14, 16–17, 23, 28, 40 applications of culture 161, 163, 164 cytokines 376 DNA arrays 496, 499–501, 503–4 drug metabolism and interaction 75, 79 drug safety 96 fibrosis 204 gender differences 543, 546, 549

gene expression 155–60, 470 genomic profiling 469, 470 Kupffer cells 316, 320, 324–6 oncology drugs 566 sinusoidal cells 345, 353 TGP database 508, 510, 525 TNF-á 249 toxicogenomics 459 Mtd 396 MTS analysis 73, 78 MTT assays 9, 73, 78 mucositis 252 multi-drug resistance proteins (MRPs) 76–7, 148, 158, 197, 209, 217–18 cholestasis 194, 195–6 DNA arrays 500 intrahepatic cholestasis 424, 434–40 PXR 209 multi-drug resistance transporter (MDR) 76, 158, 164–5, 194 cholestasis 196 DNA arrays 494, 497, 503 gender differences 549 intrahepatic cholestasis 439–40 multi-hit hypothesis 5 multiple organ interactions 97, 100–1 multi-spectral cytometric analysis 6 muraglitazar 163, 164 muscles 37, 76, 222 see also smooth muscle mutagens 109, 111, 216 m-xylene 613–14, 623 myclobutanil 473, 476 Mycobacterium tuberculosis 531, 535 mycotoxins 109, 111, 177–84, 248 myeloid progenitors 90 myelomonocytic cells 20 myelosuppression 252 myocardiac infarction (MI) 495, 590, 598 myofibroblasts 198, 204, 206, 241, 345, 380 myosin 241 N-acetylaminotransferase 92 N-acetylation 251 N-acetylcysteine (NAC) 81, 124, 237, 247, 347 N-acetyl-p-benzoquinone imine (NAPQI) 29, 124, 219–20 APAP 346 cellular stress 234 drug metabolism and interaction 77, 80

Index Kupffer cells 317 transgenic models 226 N-acetyltransferase (NAT) 146, 503 N-acetyltransferase-2 (NAT2) 531–5 NAD 200, 220 NADH 200, 207, 232, 234 NADP-dependent malic enzyme 63 NADPH 72, 218, 231, 288, 290 cellular stress 235 gender differences 540 quinone 288, 290, 292, 293–4 reactive metabolites 81 toxicogenomics 455 NADPH chinonoxidoreductase 234 NADPH cytochrome b5 reductase 290, 292, 294 NADPH cytochrome P450 reductase 290, 292–4, 540 NADPH menadione oxidoreductase(diaphorase 455 NADPH oxidase 228, 245, 320, 345, 400 NADPH quinone oxidoreductase 290, 292 NADPH quinone reductase 234, 288 nafenopin 326–7 nanotechnologies 66 naphthoquinone 291 narcoleptics 231 National Center for Toxicogenomics 471 National Disease Research Interchange (NDRI) 8 natural killer cells 245, 316, 345, 349, 380 sinusoidal cells 343, 345, 349, 354 natural killer T (NKT) cells 129, 201, 343, 345, 349 NBD-PC (stain) 38 NBD-PE (stain) 38 NDRG1-related protein 63 nearest-neighbor analysis 453, 456 necrapoptosis 243 necrosis 5, 141, 242–3 alcohol 129–30 APAP 346, 348, 350 bile acids and apoptosis 393, 396, 398, 400 biomarkers of mycotoxins 179 botanical supplements 593, 596, 598–9 carbon tetrachloride 350–1 cellular stress 234–5 cholestasis 196, 197 cytokines 374, 381 experimental models 120-1, 123-4, 125–6, 129–30

667

food-related hepatotoxins 108 genetic polymorphisms 251 genomic profiling 468, 476–7 intrahepatic cholestasis 422 in vitro prediction 147, 152 Kupffer cells 316, 321, 322, 325 lipids 240–1 mitochondria 231 oncology drugs 564–5, 567, 569 quinone 289, 291 sinusoidal cells 342, 350, 351, 353 steatosis 199 toxicogenomics 455, 457, 459–60 transgenic models 226, 227 nefazodone 26, 80, 218, 422, 424, 436–7 neoantigens 95, 201, 548, 564 neoclerodane diterpenoids 595 neophobia 410 neoplasms 110, 328–9, 553–4, 576 neopterin 377 nephrotoxocity see kidneys nerve growth factor (NGF) 343, 353, 394 neural networks 453 neural tube defects (NTDs) 183 neurodegenerative disorders 238, 391, 409–11 neuroleptics 201 neurones 93, 215, 222, 252 neuropsychotropic drugs 563 Neutral Red 73, 154 neutropenia 252 neutrophils 129, 249, 350, 353–4, 592–3 apoptosis 245, 393 cholestasis 126, 194 Kupffer cells 316, 319, 321 sinusoidal cells 343, 349, 350, 353–4 nevirapine 201, 437–8, 531 new chemical entities (NCEs) 7, 16, 18, 23, 41 new molecular entities 441–2 n-heptane 614 n-hexane 614 N-hydroxy-2-acetylaminofluorene 545–6 nicotine 540–1, 543, 547 nicotine c-oxidation 74 NIDDM 495 nifedipine 74, 549 Nile Red (stain) 95–6 nimesulide 5, 38 nimodipine 321 Nip 396

668

Index

nitric oxide (NO) 152, 235, 247, 380, 404 APAP 347, 349 carbon tetrachloride 352–3 Kupffer cells 315–16, 318–19, 323, 342, 351–2 oxidative stress 108, 234, 238 sinusoidal cells 341–3, 345, 354 nitric oxide synthetase (NOS) 108, 228, 347–9, 350, 351–2 nitrites 575 nitrofluorene 454, 460 nitrofurantoin 35, 552 nitroguanine 239 nitrosamines 575 nitrosative stress 238 nitrosodiethylamine 576 nitrosthiols 234 NMR spectroscopy 65 N-nitro-L-arginine methyl ester (L-NAME) 349 N-nitrosodimethylamine 248, 540 3-nitropropionic acid (3-NP) 410 Nix 396 nomifensin 80 non-alcoholic fatty liver disease 393, 471 non-alcoholic steatohepatitis (NASH) 38, 199–200, 230, 377–9 apoptosis 245–6, 394 cytokines 377–9, 381 genomic profiling 472 non-esterified fatty acids (NEFAs) 199 noni juice 594, 598–9 nonneoplasm 553 nonopsonized particles 342 non-parenchymal cells 27, 29, 217, 342–5, 354 APAP 347 cytokines 372, 374 fibrosis 203 food-related hepatotoxins 107 genomic profiling 469–70 in vitro prediction 144, 147, 151, 152 Kupffer cells 313–29 TNF-á 249 non-steroidal anti-inflammatory drugs (NSAIDs) 231, 455, 563 gender differences 541 genomic profiling 473, 476 intrahepatic cholestasis 422, 438 nonylphenol 209 no observed adverse effect level (NOAEL) 552–3, 607, 615, 618

noradrenalin 378 nordihydrigaiaretic acid (NDGA) 596–7 Northern blotting 177 Noxa 243, 396 N-oxidation 542 NPSH 228 NQO 225, 460 Nrf 220, 228, 235, 236, 455, 460, 475 NRLclone 60, 498 NSR/hsp 90 dissociation 408 NTCP 24, 26, 437–41 N-terminal kinase 378 nuclear area 33-5, 36 nuclear DNA 30, 32 nuclear factor 4á 549 nuclear factor-IL-6 (nf-IL-6) 237 nuclear factor kappa B (NF-êB) 236–8, 249, 475 alcohol 130 apoptosis 244–5, 395, 399, 401–2, 404, 407, 411 biotransformation 220 cellular stress 235, 236–8, 240 cytokines 247, 373, 377 genetic polymorphisms 253 Kupffer cells 320, 326, 327, 352 mechanisms of TLI 208, 210, 213–15 regeneration 250–1 sinusoidal cells 344, 348, 352, 353–4 steatosis 200 toxicogenomics 455 nuclear hormone receptors (NHR) 18, 392 nuclear orphan receptors 543–4, 550 nuclear receptors (NRs) 40, 196–7, 206–17 bile acids 391, 392 BSEP 440 cholestasis 127, 196–7 drug metabolism and interaction 76–7 gender differences 541, 543–4, 550 mechanisms of TLI 191, 206–17, 218–20, 230 nuclear steroid receptors (NSRs) 406–9, 411 nuclear transcription factors 25, 195, 196, 198, 216 nucleic acid 39, 121, 191, 232, 567 sinusoidal cells 341, 350 nucleophiles 179, 288 nucleoside reverse transcriptase inhibitors (NRTIs) 31–2, 230, 233 nucleotides 350, 531, 545

Index obesity 199, 213, 377–9, 465, 595 obstructive cholestasis 125 occupation 178, 554 oestrogen see estrogen ofloxacin 229 o-hydroxyphenyl acetaldehyde 541 Oil Red O (stain) 6, 38 okadaic acid 404 Oligo 491 oligomers 243, 394–5, 397–9, 533 oligonucleotides 10, 450, 491, 509, 566 omeprazole (OMZ) 8, 16–17, 78, 513–15, 520–5, 548 omeprazole 5-hydroxylation 74 ‘omics’ 41, 58, 154, 159-61, 166, 570 Omi/HtrA2 397 ‘on-chip’ single-cell analysis systems 66 oncogenes 25, 327, 583 oncotis necrosis 124 oncotic pressures 214 ondansetron 548 ONO-1714 347 ophthalmology 510 opsonised particles 342 oral contraceptives 126, 195, 547–8, 551 organelles 200, 219, 230, 231, 240, 253 bile acids and apoptosis 394, 395–7, 403 experimental models 123 Kupffer cells 329 methapyrilene 575, 578–80, 582–3 quinone 289 reactive metabolites 81 organic anion transporter polypeptides (OATPs)24, 158, 164–5, 217–18 cholestasis 194, 198 drug metabolism and interaction 76–7 intrahepatic cholestasis 437–8, 440 nuclear receptors 208, 209 troglitazone 303 organic anions 434 organic cation transporter (OCT) 77, 217 organic solvents 543 organochlorine 209 organophosphate 209 orotic acid 545 orphan nuclear receptors 543–4, 550 osteopontin 321–2, 351, 475 sinusoidal cells 343, 349, 351, 353 osteoporosis 539 ovaries 543–4, 546

669

overdose 374, 508, 516, 524 acetaminophen 4, 124–5, 219–20, 226–7, 234–5, 346, 524, 551 oxacillin-resistant staphylococcal endocarditis 229 oxazepam 550 oxidative phosphorylation 199–200, 230–3, 542 oxidative stress 5–6, 25, 108–9, 141, 192–3, 233–42 apoptosis 245, 396, 397, 400 botanical supplements 597 chemokines ad cytokines 247, 377, 379, 380 drug discovery 59 drug metabolism and interaction 70 drug safety 98 endpoint of culture analysis 154 fibrosis 203–6 gender differences 545, 551 genetic polymorphism 252 genomic profiling 474, 475–6 high content screening 30–5 intrahepatic cholestasis 434 in vitro applications 38–9 Kupffer cells 315, 323, 327, 328–9 mechanisms of TLI 220–1, 233–42 oncology drugs 569 quinone 288–9, 290–1 sinusoidal cells 345 survival and repair 249 TGP database 521–3 TNF-á 249 toxicogenomics 455 transgenic models 226, 227, 228 troglitazone 302 oxidative stressors/reactive metabolites (Oss/RMs) 455, 475–6 oxidised glutathione (GSSG) 231–2 oxidised low-density lipoproteins (oxLDL) 239 oxidoreductase 217 oxoglutarate dehydrogenase complex 201 oxoisovalerate dehydrogenase 63 oxolithocholic acid 392 oxypenicillins 125 oxyphenisatin 90 P13 374 P14 408 P18 306 P21 306, 497, 503 P27 306

670

Index

P38 216, 222, 229, 253, 593 bile acid and apoptosis 399, 401–2, 404 cellular stress 235, 238 troglitazone 305 P47 phox 320 P50 352–3 P52 353 P53 status 37, 220, 240, 306, 475 bile acids and apoptosis 398, 402, 404, 408 P90 250 P450 isoforms 24, 61, 90, 92, 97, 542 P450 reductase 540 paclitaxel 76 PAM (prediction analysis of microarray) 513 pancreas 21, 37, 119, 215, 539 panomics 60–2, 67 parabens 231 paracetamol see acetaminophen paracrine 204, 315, 374, 376 paraptosis 393 paraquat 219, 231–2 parasites 130 parenchymal cells 21, 29, 129, 253 botanical supplements 592–3 cytokines 372, 381 drug metabolism and interaction 72 drug safety 92, 94 fibrosis 203 genomic profiling 469 in vitro prediction 147, 152 Kupffer cells 313–16, 329 regeneration 250 sinusoidal cells 344, 345, 347 Parkinson’s disease (PD) 410 PARP (poly (ADP-ribose) polymerase) 228, 403–4, 407 partition coefficients 610, 612, 614, 616 pattern recognition receptors 342 PCB153 223 PCNA 324 p-dimethylaminoazobenzene 250 pefloxacin 229 pemoline 80 pennyroyal 594, 597–8 pennywort 594 pentachlorodibenzofuran (PeCDF) 223 pentacyclic triterpenic saponosides 599 pentobarbital 540 pentoxifylline 248, 381 peptides 316

percellome 509 perchloroethylene 110, 620 percoll 8 perferryl species 328 perfluoroheptanoioc acid 473 perfluorooctanoate 212 perfluorooctanoic acid 473 perhexiline 5, 80, 95, 251 perhexiline maleate 38 pericellular fibrosis 124 perisinusoidal cells 344 PERK 306 perlecan 351 peroxides 234, 289–90 peroxidise 148, 289, 314 peroxiredoxin (Prx) 63–4, 242, 289 peroxisomal aˆ -oxidation 199 peroxisome proliferator activated receptors (PPARs) 18, 25, 211–14, 216, 218–19 applications of cultures 162–3 drug metabolism and interaction 76–7 genomic profiling 476 Kupffer cells 325–7 mechanisms of TLI 208, 211–14, 216, 218–19 sinusoidal cells 351 target genes 166 TGP database 524 TNF-á 248 toxicogenomics 455 transgenic models 228 troglitazone 299, 304, 306, 307 peroxisome proliferator response elements (PPREs) 211, 253 peroxisome proliferators (PPs) 455–7, 473–6 peroxisomes 212, 230, 289, 316, 326–7, 494 peroxynitrate 234–5 peroxynitrite 124, 226, 234, 239, 341–2 Kupffer cells 319, 342 pesticides 209, 543, 552, 597 PGD2 326, 328 PGE2 321, 326, 328 p-glucoprotein (P-gp) 217 P-glycoprotein 549 pH 233, 579–80, 582–3 phagocytes 313–14, 316, 341–2, 566 phagocytosis 343, 345, 346, 372 Kupffer cells 315, 316, 317, 324 phagophores 578 phalloidine 545

Index pharmacodynamics 539, 616, 618, 620–1 pharmacogenomics 57, 527–8, 569, 592 pharmacokinetics 71, 91, 110, 152, 535, 566 gender differences 539, 547, 549–50, 554–5 models 607–24 pharmacophores 93 Phase I enzymes 6–8, 13–14, 16, 18, 25, 27–8 biomarkers of mycotoxins 180, 184 botanical supplements 591–2, 598 DNA arrays 496, 498–9, 501, 504 drug discovery 61 drug metabolism and interaction 77 drug safety 92, 99 endpoints of culture analysis 157, 159 gender differences 551, 554 in vitro prediction 143, 145, 152 mechanisms of TLI 207, 209–10, 217–18 quinone 287 sinusoidal cells 341 Phase II enzymes 6–8, 13–14, 16, 18, 21, 25, 27 acute phase reaction 248 biomarkers of mycotoxins 180, 184 botanical supplements 591, 598 cellular stress 234 DNA arrays 496, 498, 501, 504 drug discovery 61 drug metabolism and interaction 77 drug safety 92, 97, 98 experimental models 125 gender differences 545, 550, 551, 554 in vitro prediction 143, 145–6, 152–3 mechanisms of TLI 209–10, 217–18, 220, 221 quinone 287 sinusoidal cells 341 toxicogenomics 455 Phase III enzymes 217–18 phase contrast 9, 11, 580–1 phenacetin 80, 223 phenacetin-O-dethylation 74 phenindione 5 phenobarbital (PB) 8, 13–16, 209–11, 216, 218 drug metabolism and interaction 76–8 gender differences 540, 541, 543–4, 548, 550 genomic profiling 473, 474, 476 in vitro prediction 148, 153 Kupffer cells 316, 326, 328 PBPK models 621 TGP database 510

671

toxicogenomics 456 transgenic models 22 phenol sulfotransferase 300–1, 545 phenotypic anchoring 470–1, 472, 524 phenoxy radical 302, 551 phenylalanine-4-hydroxylase 64 2-phenypropenal (ATPAL) 81–2 phenytoin 76, 78, 193, 473, 548 phenytoin-4-hydroxylation 74 phenytoin-O-deethylation 90 phorone 524 phosgene 219 phosphatase 122, 235, 289 phosphatidylinositol 3 -kinase (PI3K) 205, 373 phosphoenolpyruvate carboxykinase (PEPCK) 164–5 phosphoinositide-specific phospholipase C 240 phosphoinositol-3-kinase (PI3K) 221, 235–6 bile acids and apoptosis 399–402, 404, 406, 411 phosphokinase C (PKC) 236, 240, 406 phospholipases 123, 230, 240, 317, 351 phospholipid flippase 196 phospholipid hydroperoxide 289 phospholipidosis 6, 25, 38, 63–4, 230–1 drug safety 95, 96 high content screening 31–2 phospholipids 6, 38, 219, 230–1 cholestasis 125, 127 drug discovery 63 drug safety 96 Kupffer cells 317, 319 PBPK models 614 phosphor-Rb 306 phosphorylates 375, 401 photolithography 29 photomicrographs 30, 32 photosynthesis 287 p-hydroxybenzoic acid 231 physiologically based pharmacokinetic (PBPK) models 607–24 picogreen 32, 36 pigs 110, 539–40, 542 see also minipigs pimozide 217 pinocytosis 579–80, 583 pioglitazone (PGZ) 82–3, 163–4, 300–1, 458, 473, 476 drug safety 96–9 intrahepatic cholestasis 433, 439–40

672

Index

pipermethystine 596 piroxicam 35 pirprofen 5, 80 pit cells 313 pituitary glands 540, 541, 545, 547, 549 PKR 306 plasma membrane integrity (blebs) 154, 243 plasma retinol-binding protein 64 plasmids 40 plasminogen 374 plasminogen-activator inhibitor (PAI) 374 plastic Petri plate matrix 501–2 platelet activating factor (PAF) 125, 315–16, 342–5, 351, 353 platelet-derived growth factors (PDGFs) 203–4, 319, 380 sinusoidal cells 343, 345, 353 pleiotropin 343, 353 pluripotent cells 21–2 PM4 (5á-pregnant-3á-ol-20-one) 437 PMN 229 P-nitrophenol 3-hydroxylation 74 p-nitrophenol sulfotransferase 545–6 point of departure (POD) 615–16, 618 poly (ADP-ribose) polymerase see PARP polyamines 122 polychlorinated biphenyls 543 polycyclic aromatic hydrocarbons (PAHs) 207, 218, 229 polyethylene glycol 566 polyhalogenated biphenyls 207 polymerase chain reaction (PCR) 9–10 real time (RT-PCR) 13, 16–18, 22, 154, 158, 163–4 polymerase chain reaction-restriction fragment length polymorphism (PCR-RFLP) 532–3 polymerase gamma 233 polymorphonuclear cells 371 polymorphonuclear neutrophils (PMN) 129, 321 polymyxin B 322 polypeptides 238 polypharmacy 4 polyphenon E 565 polyunsaturated fats (PUFAs) 121 polyunsaturated lipids 238 ponies 182 portal vein 105, 119, 130, 392, 468 amebiasis 128 cholestasis 194

fibrosis 202, 203 Kupffer cells 314 portocabal anastomosis 123–4 post-mitochondrial fractions 71, 73, 75, 81 potassium 122, 206, 208, 243 pravastatin 434 predictive toxicology 454, 457–9 prednisolone 347, 381 pregnancy 125, 126, 183, 195, 437, 554–5 intrahepatic cholestasis (ICP) 437 pregnane-activated receptor (PAR) 208 pregnance-X-receptor (PXR) 18, 38, 164–5, 208–10, 440 bile acids and apoptosis 407 cholestasis 127, 196–7 drug metabolism and interaction 76–7 gender differences 543–4, 548–9 mechanisms of TLI 207, 208–10, 211, 216, 218 pregnenolone-16á-carbonitrile (PCN) 8, 16–17, 209–10, 347 gender differences 543–4, 551 primary bile acids 392 primary biliary cirrhosis (PBC) 125 primary hepatitis 201 primary hepatocytes 5–18, 23–4, 37–9, 144–5, 215, 495–6 apoptosis 246 applications of cultures 161–3, 166 bile acids and apoptosis 398, 400, 403–4 biomarkers of mycotoxins 180 bioreactors 27 cellular stress 235 cytokines 376 DNA arrays 489–92, 494, 495–9, 500–2 drug discovery 58–61, 63–5 drug safety 92, 94, 98, 102 endpoints of culture analysis 155–60 food-related hepatotoxins 106, 110 gender differences 543–4 genetically engineered cells 24–5 intrahepatic cholestasis 434, 440 in vitro prediction 142, 144–5, 148–9, 150–1 lipids 242 methapyrilene 577–8, 580–2 quinone 289–93 TGP database 509, 511 toxicogenomics 458–9 primary liver cancer (PLC) 180, 184 primary sclerosing cholestasis (PSC) 125

Index primary steatosis 199 primates 120–1, 128, 163, 300, 555, 566 principal components analysis (PCA) 155, 156, 453, 526–7 proadifen 541 procainamide 35 procaspases 216, 243–4, 394–5, 397, 405, 407 procollagens 204–5 procysteine 377 progenitor cells 22 progesterone 209, 437, 544, 550 progressive familial intrahepatic cholestasis (PFIC) 194, 196, 440–1 propanolol 231, 547–8, 550 propiconazole 473, 476 propofol 74, 550 propylamine 578–9 prostaglandins 23, 315, 321, 328, 376 prostanoids 315, 328 proteases 146–7, 374, 578–9 apoptosis 243, 411 DNA arrays 498, 500, 503 experimental models 123, 24 intrahepatic cholestasis 437 nuclear receptors 209 quinone 289 sinusoidal cells 351 steatosis 200–1 proteasomes 377 protein kinase C (PKC) 122, 238, 240, 327 351 apoptosis 245, 398, 401 proteinase 201, 218 proteins 5, 8, 13, 23–5, 28, 306 acute phase reaction 248 bile acids and apoptosis 243–4, 392, 394–8, 401–4, 407, 409 biomarkers of mycotoxins 177, 180 botanical supplements 592–3 cellular stress 234–9 cholestasis 194–5, 422 cytokines 371–2, 375–7, 380, 381 DNA arrays 493, 497, 498–500, 503 drug discovery 58–67 drug metabolism and interaction 70, 75, 78–81 drug safety 95–7 endpoints of culture analysis 154 experimental models 121, 122, 124, 125 fibrosis 203, 205–6 food-related hepatotoxins 108

673

gender differences 543–4, 546, 548, 550, 554 genomic profiling 467, 474, 476 hepatitis 201 high content screening 33 intrahepatic cholestasis 429, 431, 433–4, 437, 439, 441 in vitro prediction 142, 147, 148 Kupffer cells 315, 328, 352 lipids 240–2 mechanisms of TLI 191, 207, 214–17, 219–20, 247 methapyrilene 578, 580, 583 mitochondria 232–3 oncology drugs 566–7 oxidation 290 PBPK models 612 phosphorylation 210 quinone 288–91 regeneration 250 sinusoidal cells 341, 344–5, 349, 350–4 steatosis 199–200 survival and repair 250 TGP database 518, 520–1, 524 toxicogenomics 450, 454, 459 transgenic models 226, 228 troglitazone 304–6 protein-serine/threonine kinase 204 proteoglycans 351 proteolysis 238, 341–2, 344, 407 proteomics 96, 177, 184, 161, 567, 583 drug discovery 58–9, 63 gene expression 159 genomic profiling 467, 472, 474 toxicogenomics 459, 461 protocol enrichment 509, 512–20 proton pore 231 pruritus 210 P selectin 344 pseudo-septa 202 PTPase 400 pulegone 597 Puma 243, 396 pyrazole 78 pyrethroid pesticides 209 pyridine 543 pyrimidine nucleotide 545 pyrrolizidine alkaloids (Pas) 109, 593–5 pyruvate carboxylase 201 pyruvate kinase isoenzymes 64

674

Index

quadrate 119 quantitativity 509–10, 525–6 quaternary nitrogen bipyridinium 231 quinolones 229 quinone 219, 241, 287–95, 302, 433–4, 545 troglitazone 301–2, 433–4 quinone epoxide 302 quinoneimines 541, 543 quinone methide 433–4 quinone redox cycle 290 quinone-semiquinone interconversion 288 rabbits 110, 120, 439 racemic 90 radiation and radioactivity 234, 243, 380, 425, 569 radicals 234 see also free radicals Raf 373, 404 ranitidine 592–3 RANTES 343, 349 rapid acetylators (RAs) 532, 535 Rappaport fields 203 Ras 184, 373 rash 193, 201, 592 rats 7–8, 16–18, 24, 28, 493–4 acute phase reaction 248 alcohol 129–30, 376, 377 applications of cultures 162–4, 166 bile acids and apoptosis 400, 402–4, 406, 409–11 biomarkers of mycotoxins 180–2 botanical supplements 592–3, 595 cholestasis 195, 196, 197 cytokines 372, 376–7, 380–1 DNA arrays 490–2, 493–4, 496, 497–502 drug discovery 58–60, 63–5 drug safety 90 endpoints of culture analysis 155–60 experimental models 120, 122, 123, 125–7, 129–30 food-related hepatotoxins 106–11 gender differences 540–6, 547–9, 551, 554–5 genetic polymorphisms 252 genomic profiling 468, 473–4, 476–7 intrahepatic cholestasis 422, 424, 425–6, 429–35, 4 in vitro prediction 148, 149, 150–1 Kupffer cells 314, 317–18, 320–8 lipids 240–2

methapyrilene 575–83 oncology drugs 565–6 PBPK models 614, 618 quinone 290–3 mechanisms of TLI 215, 221, 229 sinusoidal cells 350, 353 TGP database 507–13, 515–16, 520, 522, 525–6 toxicogenomics 449, 454–60 troglitazone 301, 303 rattleweed 594 reactive metabolites 6, 26, 38–9, 70, 79–84, 141 bioactivation 79–80 drug safety 94–5, 96, 97 experimental models 124 gender differences 558 genomic profiling 475 identification 80–4 intrahepatic cholestasis 429, 431, 433–4, 437 Kupffer cells 317 mechanisms of TLI 191, 202, 221 oncology drugs 567 toxicogenomics 455 troglitazone 304 reactive nitrogen species (RNS) 216, 317–18 cellular stress 234, 238, 239 sinusoidal cells 341, 343–4, 346–7 reactive oxygen species (ROS) 38, 238–9, 289–90 apoptosis 243–5 bile acids and apoptosis 393, 396, 399–400, 402–3, 405 cellular stress 234–5, 237–9 cytokines 378 experimental models 124 fibrosis 204–5, 206 food-related hepatotoxins 108 high content screening 31, 36 Kupffer cells 315, 317–18, 320–1, 327, 329 lipids 240–2 mechanisms of TLI 191, 208, 216, 219–20, 230, 238–9, 253 mitochondria 231–3 quinone 289–91 sinusoidal cells 341, 343–4, 346–7, 353 steatosis 199, 201 TNF-á 249 transgenic models 223

Index redox 6, 192, 220, 249, 290–1 cellular stress 235, 236, 238 drug metabolism and interaction 70 experimental models 125 fibrosis 205 high content screening 31 lipids 240–1 quinone 288, 290–4 mitochondria 232 quinone 288, 290–4 steatosis 200 toxicogenomics 455 reductases 288, 290, 294 regeneration 4, 250–1, 372–4 apoptosis 242, 392–3, 400, 408 cytokines 372–4, 376, 381 experimental models 120, 121, 123 fibrosis 202–4, 206 genomic profiling 474 Kupffer cells 315, 320, 324–5, 329 lipids 241 mechanisms of TLI 191, 193, 207, 229, 253 nuclear receptors 211, 213, 214, 216 sinusoidal cells 341–2, 344, 348, 349 transgenic models 227 Rel A 401 Rel B 353 reproducibility 59–10 reproduction systems 542 reptiles 540 Research Genetics 491 resident cells 371 resistin 378 resveratrol 347 retinoic acid 23 retinoid acid receptor (RAR) 194, 208 retinoid X receptor (RXR) 194, 207, 227, 543 mechanisms of TLI 207, 210–11, 216, 219 retinyl esters 344 retrorsine 595 reverse transcriptase-polymerase chain reaction (RT-PCR) 79 Rezulin 299, 551 rhabdomyolysis 212, 217 rheumatoid arthritis 251 ribonuclease 155 ribosomal protein 306, 500, 503 ribosomal RNA methylation 122 ribosomal S kinase (RSK) 250 ribosomes 122, 305, 306, 322, 580

675

rifampicin (RIF or RFP)76–8,148, 218, 439 cholestasis 125, 195 drug-induced liver injury 8, 13–17, 26, 38 drug safety 95 gender differences 543, 548, 550, 552 high content screening 35 intrahepatic cholestasis 422, 424, 439 lipids 241 nuclear receptors 209–10 tuberculosis 531, 534 rifamycin 125, 422 rifamycin SV 439 rifapentine 78 risk assessment 552–3, 607–24 ritonavir 26, 76, 218, 423, 437–8 RMA 452 (R)-mephobarbital 548 RNA 9, 142, 240, 495, 509–10 toxicogenomics 450–2, 456 RNA-inducing silencing complex (RISC) 40 RNA interference (RNAi) 39–40, 41, 62 rodents 3–4, 328–9, 493–4, 495–8 APAP 346, 350 BSEP 26 cadmium 322 cellular stress 235 cytokines 247, 374, 378, 379 DNA arrays 489–90, 493–4, 495–6, 497–8 gender differences 539, 542, 549 genomic profiling 470, 476 methapyrilene 583 oncology drugs 566, 568 TGP database 511 toxicogenomics 449, 455 see also mice; rats rosiglitazone (RGZ) 82–3, 300–1, 304, 473, 476 drug safety 96–9 intrahepatic cholestasis 433, 439–40 toxicogenomics 458 rosiglitazone para-hydroxylation 74 route-to-route extrapolation 615–16, 617–18, 622, 624 St John’s wort 209, 590 Saccharomyces cerevisiae 180 S-adenosylmethionine (SAM) 122, 347, 381 S-oxidation 433 safety 19, 37, 89–102, 142, 567, 590 APAP 346 botanical supplements 589–91, 599

676

Index

safety (Cont.) drug discovery 57–8, 60–2, 67 drug metabolism and interaction 70–1 food-related hepatotoxins 105–6, 107, 108 gender differences 552–3 genetic polymorphism 251 intrahepatic cholestasis 421 oncology drugs 563, 565–6, 567, 569–70 PBPK models 607–24 TGP database 508, 526, 528 safingol 111 salicyclic acid (aspirin) 231 Salmonella typhymurium 577 sandwich culture 6–7, 23, 38, 106, 154–60, 161–3 DNA arrays 492, 497 drug discovery 60 drug metabolism and interaction 72, 77 intrahepatic cholestasis 436, 437–8 in vitro prediction 147–52, 166 Saos-2 404 SAPK(JNK 205 saponins 595 saquinavir 26, 437–8 scavenger receptors 315, 342, 344, 498–9 Schiff bases 180 scintillometric analysis 577 screening 18, 61, 120, 143 drug discovery 57, 58–9, 61, 65–6 drug metabolism and interaction 70, 72, 81, 83–4 drug safety 93, 95, 97 food-related hepatotoxins 105–11 high content (HSC) 25–6, 30-7, 41, 142, 154, 166 intrahepatic cholestasis 441 secondary bile acids 392 secondary hepatitis 201 secondary steatosis 199 selenium 77, 148 semiquinone 288, 290, 292, 294 Senecio 594–5 senescence marker protein-30 64 senna 594, 598–9 sensitivity and drug discovery 62 Sentrix Beadchip Arrays 155–6, 160 sepsis 247 septic shock 315 serine 237, 306, 371, 401 serine proteases 503

serotonin 550 serotonin receptor antagonist 218 serotransferrin 63–4 serzone 436 set correlation analysis 456 sexual performance enhancers 589 sheep 110, 120 short-chain 3-hydroxyacyl-CoA dehydrogenase 63–4 short hairpin RNA (shRNA) 40 short heterodimer partner (SHP) 127 signal cointegrator-2 (ASC-2) 40 signal transducer and activator of transcription (STAT) 215, 375–6, 381 cytokines 373 gender differences 549 genomic profiling 475 sinusoidal cells 344, 348 toxicogenomics 455 silicates 317 silicon 29, 152 simvastatin 435 single nucleotide polymorphisms (SNPs) 528, 532–3 sinusoidal endothelial cells (SECs) 203, 217, 245, 250, 498 genomic profiling 469, 474 Kupffer cells 313, 318, 320–1 sinusoidal free acidophile bodies 120 sinusoids 7, 27, 151, 341–54 cytokines 372 fibrosis 202 Kupffer cells 313–14 PBPK models 610 Sister chromatid exchange (SCE) 577 sisulfide 236 SKF 292–4 skin cancer 542 slow acetylators (SAs) 532, 534–5 Smac 124 Smac(DIABLO 243–4, 395, 397 small double-stranded RNA 40 small-interfering RNA (siRNA) 39–40, 62, 306 S-mephenytoin 4 -hydroxylation 74, 90 S-mephenytoin-N-demethylation 74 smoking and tobacco 494–5, 543, 547, 591 smooth muscle 240, 249, 291 alpha acton (SMA) 204, 352, 380, 398 S-nitrosylation 404

Index sodium 122, 502 sodium-dependent phosphorylation (NAD) 200, 220 sodium diethyldithiocarbamate 5 sodium salt 90 sodium-taurocholate co-transporting polypeptide (NTCP) 24 cholestasis 126–7, 193–5, 196–8 somatic cells 577 space of Disse 194, 202–3, 313–14, 344, 348 species differences 18, 23, 90, 91–3, 110, 142, 439–40, 616–17 BSEP 441 drug metabolism and interaction 70–1 drug safety 90, 91–3, 101 food-related hepatotoxins 110, 111 genomic profiling 466, 473 in vitro prediction 142, 166 Kupffer cells 328 methapyrilene 580 oncology drugs 568 PBPK models 616-17, 618, 624 TGP database 508, 509, 511–12, 528 spectral principal component analysis 65 spectroscope 404 spheroid cultures 144, 152, 503 sphinganine (Sa) 181–2, 183 sphingolipid metabolism 181–3 sphingomyelin phosphodiesterase 503 sphingomyelinase 245 sphingosine (So) 181–2, 183 spironolactone 209 spleen 119, 123, 223, 325 Sprague-Dawley (SD) rats 107, 434–5, 456, 497, 543 TGP database 509, 512 src-kinase 245 statins 31, 209, 212, 217 statistical analysis 9 stavudine 5 steatohepatitis 5, 38, 199, 240, 376, 377–9 steatosis 38, 109, 141, 199–201, 213, 230 apoptosis 246 carbon tetrachloride 350, 351 cytokines 378 drug-induced liver injury 6, 25 drug safety 95, 96 experimental models 120 lipids 240 mitochondria 233

677

toxicogenomics 457–8 see also fatty liver stem cells 18–24, 41, 306 steroid delta-isomerase 519 steroid hormones 195, 207, 209, 437 bile acids 407 gender differences 546, 555 TGP database 520 steroid hydroxylases 210, 541 steroid reductase 497 steroid and xenobiotics receptor (SXR) 76, 308, 407 steroids 23, 470, 474, 476, 497 gender differences 540–1, 543–6, 550 intrahepatic cholestasis 422, 438 sterol 196, 210, 392 stress proteins 154 strogens 120 stroke 409, 495, 539, 590, 598 stroma cells 20, 371 strychnine 540 Student’s t-test 293, 513–14 styrene 613 succinate dehydrogenase 410 sugars 78, 238 suicide 551 sulfamethoxazole 5, 35, 201 sulfasalazine 532 sulfate ester 221 sulfated proteoglycan 315 sulfates and sulfonation 217, 221, 300, 346, 437 sulfathiazole 80 sulfenic acid 291 sulfenic acid-á-keto-isocyanate 433 sulfhydryl 122, 241, 322–3 sulfobromophtalein 127 sulfonylurea 425–6, 427–8 sulfotransferase (ST or SULT) 8–9, 13–14, 16, 29, 146, 158, 545–6 DNA arrays 497, 502 drug metabolism and interaction 72, 76, 77 drug safety 92 gender differences 544, 545–6, 551 intrahepatic cholestasis 431–2 PXR 209 related protein 184 troglitazone 304 sulfoxide 302 sulfuric acid 287 sulfur nucleophiles 302

678

Index

sulindac 35, 422, 438–9 sunhemp 594 supernatants (S9) 142, 145, 153 superoxide anion 204, 232, 234, 247 Kupffer cells 314, 315, 318, 321, 342, 351 quinone 288–90, 292–3 sinusoidal cells 342, 345, 351, 354 superoxide dismutase (SOD) 158, 226, 347, 497 cellular stress 234, 242 quinone 289–90 superoxides 33, 219–20, 245 cellular stress 234–5, 238 Kupffer cells 316, 319, 326–7, 328 supersomes 143, 145 support vector machines (SVMs) 453, 457–8, 513, 526 suprofen 80 survival and repair 249–50 survival signalling pathways 400–2, 405–6, 411 susceptibility 120, 178, 251, 317, 434 methapyrilene 580 oncology drugs 569 troglitazone 303, 307 SV40 T antigen virus 26 S-warfarin 7-hydroxylation 74 Symphytum 594–5 systems biology 66–7, 154, 159–61, 166, 472, 570 tacrine 35, 74, 458, 551 tamoxifen (TAM) 5, 8, 16–17, 35, 101, 221 tandem mass spectrometry 39 TaqMan see fluorogenic cDNA probes (TaqMan) target cell initiation theory (TACIT) 94–7 target cells 90, 94, 96, 7 target genes 9–12, 16, 163, 166, 373, 401, 450 taurine 193, 392, 404, 410–11 taurochenodeoxycholate 401, 439 taurochenodeoxycholic acid 397, 406 taurocholate 6, 26, 38, 126, 194, 195, 303 intrahepatic cholestasis 422, 424–5, 427, 429–31, 434–5, 437, 439-41 taurocholic acid 425, 430 taurolithocholate 5 tauroursodeoxycholate 439 tauroursodeoxycholic acid (TUDCA) 402, 404–6, 410–11 taxol 78 taxol 6-hydroxylation 74

tBid 395 T cells 129, 202, 222, 345, 376 see also natural killer T cells TCPOBOP 76 temafloxacin 80, 229 temazepam 550 temperature 233 teratogenics 208, 225 terbinafine 35 terfenadine 217 terfenadine C-hydroxylation 74 test compound selection 509, 510–11 testes 289, 497, 544 testosterone 15, 28, 74, 90 gender differences 540, 541, 544, 546 testosterone 16â-hydroxylase 544 tetrachlorodibenzo-p-dioxin (TCDD)25, 76, 207–8 gender differences 543 PBPK models 608, 611 transgenic models 223, 224, 225 tetra chloroethane 613 tetrachloroethylene 613–14 tetracycline (TC) 5, 38, 95, 242, 552 drug discovery 58–9, 63–5 tetrazolium salts (MTT) 73, 154, 423 thalidomide 381, 508, 539 thansthyretin 21 theophylline 539, 542–3, 547 theophylline-N-demethylation 74 thiabendazole 110 thiazolidinediones (TZD) 82–3, 379, 439 ring scission 433 troglitazone 299–307, 429, 433–4 thiazolidinedione-sulfoxide 433 thioacetamide 5, 227, 455, 457, 473, 476, 520–1 thioacetylate 608 thiobarbituric acid 240 thioesters 215 thio-ITP 252 thiol 223, 288–9, 291, 431 cellular stress 234, 236, 237 thiolate anion 291 thiophene 547 thiopurine methyltransferase (TP/MT) 251–3 thiopurines 252 thioredoxin (TRX) 226, 235, 236–8, 289 thioredoxin 2 (Txn2) 158 thiothixene 547 THLE-CYP cell line 26

Index three-dimensional bioreactors 23–4, 27–30, 41, 144, 146, 152 three-dimensional configuration 7–8, 9, 11, 18, 143, 144, 152 three-step perfusion technique 8 threonine kinase 27, 306, 371, 401 thromboxane B2 (TxB2) 319 Thy-1 20 thymidine kinase (TK) 577 thymocyte 222 thymus 76, 224 thyroid cancer 539 tianeptine 5 ticrynafen 80, 202, 552 tienilic acid 5, 95, 201–2 tiralazad 549 tissue inhibitors of metallo proteinase (TIMP) 343, 345, 352–3 titer 304 T lymphocytes 129, 245, 316, 380, 394 TMRM 6, 36 tolbutamide 74, 90, 426–7 tolcapone 80 toll-like receptors (TLRs) 342, 344, 377 tolrestat 80 toluene 545, 613–14, 623 toluidine blue 64 topisomerases 578 topotecan (TPT) 90 torsades de pointes 229 toto-3 33, 36 TOX-chip 98 toxic hepatitis 120, 123 toxicodynamics 554, 615, 620, 623 toxicogenomics 96–8, 109–10, 111, 449–61, 466–77 applications 454–60, 473–7 DNA arrays 494 drug discovery 58, 60, 62 gene expression 159 genomic profiling 466–77 oncology drugs 567, 569 prediction using database 507–28 Toxicogenomics Project (TGP) database 507–28 toxicokinetics 4, 142, 554 toxicologically relevant genes (tox genes) 458 toxicopanomics 58, 60–2 toxicophore 93 toxicoproteomics 58, 62

679

TPP 423 Traditional Chinese Medicine 590, 594, 597, 598 TRADD (TNFR1-associated death domain protein) 237, 243–4, 394 TRAF 237, 244, 399, 401 TRAIL (TNF-related apoptosis-inducing ligand receptor) 243–4, 246, 394–5, 398 trans-4-hydroxy-2-nonenal 239 trans-acting liver enriched transcription factors 214 transaminases 129, 213, 221, 458, 566, 568 transcription factors 214–16, 219, 253 bile acids 392, 401, 407–8 cellular stress 235–6 cytokines 373 DNA arrays 503–4 endpoints of culture analysis 154, 157 gender differences 543 genomic profiling 471, 473–6 Kupffer cells 326–7 lipids 240 regeneration 250–1 sinusoidal cells 342, 344, 348 survival and repair 249 toxicogenomics 455 troglitazone 306 transcriptomes 466–7, 468–70, 476, 499, 516 transcriptomics 159, 465 transferase mediated dUTP nick-end labelling (TUNEL) 409, 411 transferases 72, 81, 592 transferin 77, 579, 582 transforming growth factor-á (TGF-á) 315, 373–4, 475 Kupffer cells 316 regeneration 250 sinusoidal cells 342, 345, 347 transforming growth factor-â (TGF-â) 208, 222, 240, 249–50, 345, 353 bile acids and apoptosis 404, 408 cytokines 372, 380, 381 fibrosis 203–5, 380–1 Kupffer cells 315–16, 319–20, 323–5, 342, 351 lipids 240 regeneration 250 sinusoidal cells 342–3, 345, 351–3 transgenic cells 142, 153, 347 trans-inhibition of BSEP 437

680

Index

transmission electron microscopy (TEM) 64, 580, 582 transplantation 28, 127, 141, 381 APAP 346 bile acids and apoptosis 393, 410 botanical supplements 590, 598 drug metabolism and interaction 69, 82 drug-related liver injury 252 transporters 24, 141, 209–10, 230, 253, 303 bile acids 95 cholestasis 127 DNA arrays 493, 502–3 drug safety 92, 95 endpoints of culture analysis 154–5 intrahepatic cholestasis 421–3, 434–5, 437–41 in vitro prediction 144–5, 147 proteins 37, 38 troglitazone 303 trapping agent 81 tree shrews 179, 180 triadimefon 473, 476 triazolam 4-hydroxylation 74 tributylin 542 trichloroacetic acid 608 trichloroethane 613 trichloroethylene 5, 110, 608, 613–14, 620–1 trichloromethyl radicals 121–2, 321, 350–1, 476 trichostatin A (TSA) 503 tricyclics 550 trifluoroacetyl chloride 201, 219, 548 trifluoroacetyl (TFA) protein 304 triglyceride 95, 200, 212, 214, 377, 476 triosephosphate 63 tripepidylpeptidase 456 troglitazone (TGZ) 80, 82–3, 161–2, 213, 299–307, 429–36, 551 drug metabolism and interaction 78 drug safety 96–9 gender differences 551 genomic profiling 473, 476 intrahepatic cholestasis 422, 424, 425–6, 429–36, 439–40 toxicogenomics 458 troglitazone glucuronide 300, 301–2, 429–30 troglitazone quinone 300–2, 304, 434 troglitazone sulfate (TGZ-S) 300–1, 303–4, 434–5 intrahepatic cholestasis 424, 425, 429–32, 434–6, 440

troleandomycin 595 trolox 252 trophozoites 128 tropoelastin 352 tropomyosin 503 trout 179, 180 trovafloxacin 80, 229 Trypan blue 154, 498 tryptophan 543 tuberculosis 531, 534–5, 596 tumor necrosis factor 235, 237, 320, 371–2 tumor necrosis factor-á 213, 228, 248–9, 253, 378 APAP 346–8 apoptosis 243–4, 245, 394, 401–2 botanical supplements 592–3 carbon tetrachloride 351–4 cellular stress 238 cholestasis 194–5 ConA 129 cytokines 247, 372–3, 375–81 fibrosis 203, 206 in vitro prediction 152 Kupffer cells 315–16, 319–21, 323–7, 329, 342, 344, 351–2 lipids 240 mechanism of TLI 215, 220–2, 229 regeneration 250–1 sinusoidal cells 342–5 steatosis 200 TGP database 524 transgenic models 228 tumor necrosis factor a´ receptor (TNFR1) 200, 204, 215, 250, 352 apoptosis 243–6, 394–5 cellular stress 237, 238 cytokines 247, 373 tumors 141, 153, 565, 595, 618–19 biomarkers of mycotoxins 177, 179, 180 gender differences 542 Kupffer cells 315, 325–9, 342 methapyrilene 575 toxicogenomics 449, 454 see also cancer and carcinoma turpentine oil 477 two-dimensional electrophoresis 63 Tyk2 375 Tyr 373 tyrosine 215, 234, 238, 319, 394, 503 tyrosine aminotransferase 20, 407

Index tyrosine hydroxylase 410 tyrosine kinase 238, 371, 374, 375, 400 tyrosine phosphatase 373 ultra violet (UV) irradiation 216 umbilical cord blood 19, 20 uncoupling protein 232 unscheduled DNA synthesis (UDS) 576, 577 untransformed transgenic hepatocyte cell lines 153 urea production 20–1, 28–9, 30, 40, 152, 154 uridine diphosphate-glucose dehydrogenase (UGDH) 217–18 uridine diphosphate-glucuronosyl transferase (UDPGT or UGT) 8–11, 13–14, 16, 25, 29, 546 BSEP 440 DNA arrays 497 drug metabolism and interaction 72, 75–6, 77 gender differences 543–4, 546, 550–1 in vitro prediction 145–6 mechanisms of TLI 207, 210, 217, 251, 253 PXR 209 toxicogenomics 460 transgenic models 228 troglitazone 301–2 uridine diphosphate dependent glucuronyl transferase 92 urine 69, 81, 182, 183, 301 APAP 346 biomarkers of mycotoxins 178–84 botanical supplements 591, 599 cellular stress 239 cholestasis 197 drug discovery 65–6 gender differences 544, 547 TGP database 525 transgenic models 224 troglitazone 302 urokinase-type plasminogen activator (uPA) 374 uroporphyrin 223, 224 ursodeoxycholic acid (UDCA) 210, 391–2, 402–6, 407–11 usnic acid 598 V79 cells 27 vacuoles 154, 314, 323, 344, 457, 553, 579–80 vagotomy 406 valences 234 valproate 35, 212

681

valproic acid 38, 193, 231, 455, 510 vanadium pentoxide 247 vascular adhesion protein (VAP-1) 344 vascular cell adhesion molecule (VCAM-1) 344–5 vascular endothelial cells 93, 100 vascular endothelial growth factor (VEGF) 349, 374 vena cava 119, 130 venous equilibration 610 venous outflow obstruction 379 verapamil 549, 597 very late antigen-5 (VLA-5) 344 ViaSpan 8 vimentin 502 vinblastine 578–9 vincristine 569 vinyl chloride 608, 613, 616, 620 vinyl fluoride 620 viral hepatitis 107, 229, 246, 375–6, 379, 381, 393 bile acids and apoptosis 393 cytokines 374, 375–6 troglitazone 304, 307 virtual cells 146, 161 viruses 129, 130, 316, 394 food-related hepatotoxins 108, 109-11 vitamin A 203, 322, 344–5, 372, 470 vitamin C 234 vitamin D 216 vitamin D receptor (VDR) 207, 208, 212, 216–17, 219 vitamin E 204, 234, 241, 252, 301, 434 vitamins 372, 392, 467, 474, 591 VX-745 222 warfarin 548 weight-loss remedies 589, 598–9 ‘wells in a chamber’ concept 100–1 Western immunoblotting 15, 23, 79, 154, 177 WIF-B9 cells 24–5 William’s medium E (WME) 8, 77, 148 Williams test 517 Wilson’s disease 199, 239, 245 Wistar rats 123, 149, 151 wortmannin 406 wound healing 202, 241, 379, 381, 599 sinusoidal cells 342, 345, 348 Wyeth 14,643 (WY14643) 13, 16, 456–7, 473 Kupffer cells 316, 326–7

682

Index

xanthine oxidase 252, 327 xenobiotic response element (XRE) 207 xenobiotics 6–7, 16–18, 23, 26, 40, 91 botanical supplements 591–3 cholestasis 197, 422 cytokines 372, 374 DNA arrays 501 drug metabolism and interaction 75, 77 endpoints of culture analysis 155 experimental models 119, 130 food-related hepatotoxins 105–6, 108–10 gender differences 539–55 genomic profiling 465, 466, 467–72, 474 intrahepatic cholestasis 434 in vitro prediction 141, 143, 153 Kupffer cells 315

mechanisms of TLI 206–7, 211–14, 216–19, 253 PBPK models 607, 611 quinone 287–9, 292 toxicogenomics 449, 456–7, 460 troglitazone 303 xenosensors 207, 235, 236 yeast 467 yolk sac 21 zalcitabine 32 zidovudine 5 zileutin 80 zimelidine 80 zomepirac 80 zonation-based bioreactor 29 zymosan 322

Hepatotoxicity: From Genomics to In Vitro and In Vivo Models

Hepatotoxicity: From Genomics to In Vitro and In Vivo Models

Nanotoxicity: From In Vivo and In Vitro Models to Health Risks

Protein Folding: In Vivo and In Vitro (Acs Symposium Series)

In Vivo Models of Inflammation

Pharmaceutical Product Development: In Vitro-In Vivo Correlation

In Vitro Transcription and Translation

In Vitro Methods in Toxicology

In Vitro Mutagenesis Protocols

In Vitro Mutagenesis Protocols

In Vitro Fertilization

In Vitro Toxicology

In Vitro Mutagenesis Protocols

In Vitro Fertilization

Angiogenesis: In Vitro Systems

In Vivo Neuromethods

Drug Absorption Studies: In Situ, In Vitro and In Silico Models

In Vitro Toxicity Testing Protocols

In-Vitro Fertilization, Third Edition

In Vivo Models of Inflammation, 2e, Vol. II

Introduction to Marine Genomics (Advances in Marine Genomics)

In Vitro Toxicity Testing Protocols

Xenobiotic Metabolism: In Vitro Methods

Introduction to Marine Genomics (Advances in Marine Genomics)

In Vivo NMR Spectroscopy: Principles and Techniques

In Vitro Methods in Pharmaceutical Research

Textbook of in vivo Imaging in Vertebrates

In Vitro Transcription and Translation Protocols

Metastasis Research Protocols Volume 2 Analysis of Cell Behavior In Vitro and in Vivo (Methods in Molecular Medicine)

Hepatotoxicity: From Genomics to In Vitro and In Vivo Models