Liver Regeneration And Carcinogenesis: Molecular and Cellular Mechanisms

Liver Regeneration a n d C a r c l' n o g e n e s l's Molecular and Cellular Mechanisms This Page Intentionally Left...

Author: Randy L. Jirtle

33 downloads 620 Views 21MB 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

Liver Regeneration a n d C a r c l' n o g e n e s l's

Molecular and Cellular Mechanisms

This Page Intentionally Left Blank

L~ver Regeneration and Carcinogenesis Molecular and Cellular Mecbanisms

Edited by Randy L. Jirtle Departmentof Radiation Oncology Duke UniversityMedical Center Durham, North Carolina

AcademicPress San Diego New York Boston London Sydney Tokyo Toronto

Cover illustration: A Modification of Figure 1 from Chapter 4 of this volume.

This book is printed on acid-free paper.

Copyright 9 1995 by ACADEMIC PRESS, INC. All Rights Reserved. No part of this publication may be reproduced or transmitted in any form or by any means, electronic or mechanical, including photocopy, recording, or any information storage and retrieval system, without permission in writing from the publisher. Academic Press, Inc. A Division of Harcourt Brace & Company 525 B Street, Suite 1900, San Diego, California 92101-4495

United Kingdom Edition published by Academic Press Limited 24-28 Oval Road, London NW1 7DX Library of Congress Cataloging-in-Publication Data Liver regeneration and carcinogenesis : molecular and cellular mechanisms / edited by Randy L. Jirtle. p. cm. Includes index. ISBN 0-12-385355-9 (case : alk. paper) 1. Liver--Cancer. 2. Liver--Regeneration. 3. Pathology, Molecular. 4. Liver--Molecular aspects. I. Jirtle, Randy L. [DNLM: 1. Liver Regeneration. 2. Liver Neoplasms--genetics 702 L7848 1995] RC280.L5L587 1995 6516.3'62--dc20 DNLM/DLC for Library of Congress 95-14344 CIP PRINTED IN THE UNITED STATES OF AMERICA 95 96 97 98 99 00 QW 9 8 7 6

5

4

3

2

1

WI

This book is dedicated to my wife Nancy and my children, Bonnie and ]ames.


Contents Contributors xvii Preface xxi

1 Liver Regeneration Then and Now Nancy L. R. Bucher

I. II. III. IV. V.

Landmarks 1 Normal Adult Rat Liver Liver Regeneration 3 Hepatocyte Priming 5 Regeneration Signals 7 A. Hormones 7 B. Growth Factors 12 C. Cytokines 17 D. Interactions 18 VI. Conclusions 19 References 19

2 Hepatocyte Growth Factor (HGF) and Its Receptor (Met) in Liver Regeneration, Neoplasia, and Disease George K. Michalopoulos

I. Introduction 27 II. Structural and Functional Aspects of HGF and the HGF Receptor 28 III. HGF Localization 30 A. In Liver 30 B. In Extra Hepatic Tissues 31

vii

viii

Contents

32 IV. Liver and the Processing of HGF A. HGF and Liver Regeneration 33 V. HGF and the Early Proteolytic Events Following Partial Hepatectomy 37 40 VI. HGF Localization A. In Liver Embryogenesis 40 B. In Liver Disease 41 C. In Liver Carcinogenesis 42 VII. Summary 43 References 44

3 Structure and Functions of the HGF Receptor (c-Met) Paol0 M. Comogli0 Elisa Vigna

I. Hepatocyte Growth Factor and Scatter Factor 51 II. HGF Receptor 53 53 A. Encoding by the c - m e t Oncogene 54 B. Post-translational Modifications C. Positive and Negative Regulation 55 D. Signal Transduction 57 E. Tissue Distribution and Subcellular Localization 59 III. Regulation of c - m e t Expression 60 IV. Role of HGF in Tissue Regeneration and Embryogenesis 61 V. Role of c - m e t in Carcinogenesis 62 References 63

4 Expression and Function of Growth-Induced Genes during Liver Regeneration Rebecca Taub

I. Liver Regeneration: The Important Questions 71 II. Immediate-Early Gene Expression in Hepatic Cells 72 III. Modification of Preexisting Transcription Factors

ix

Contents

IV.

V.

VI.

VII. VIII.

IX.

X.

Immediately Following Partial Hepatectomy Turns on Immediate-Early Genes 76 Induction Patterns of 70 Genes Following Partial Hepatectomy Define the Temporal Course of Liver Regeneration 78 Transcription Factors Induced in the Regenerating Liver 80 A. The LRF-1/JunB Story 80 B. RNR-1, a Novel Nuclear Receptor That Acts through the NGFI-B Half-Site 82 Immediate-Early Genes Involved in Signal Transduction 84 A. PRL-1, a Member of a Novel Class of Protein-Tyrosine Phosphatases 84 Immediate-Early Genes That Are Secreted Proteins 86 Liver-Specific Immediate-Early Genes: Relationship to the Maintenance of Hepatocyte Differentiation and Metabolism 87 A. Identification of CL-6 88 Immediate-Early Genes in H35 Cells That Are Expressed as Delayed-Early Genes in Regenerating Liver 89 A. Delayed-Early Genes That Encode RNA Binding Proteins 89 Conclusions 91 References 93

5 Stem Cells and Hepatocarcinogenesis Snorri S. Th0rgeirss0n

I. Introduction 99 II. Cellular Biology of the Hepatic Stem Cell Compartment 101 A. Experimental Systems in Vivo 101 B. Experimental Systems in Vitro 102 III. Neoplastic Development in the Liver 104 A. Hepatic Stem Cells and Hepatocarcinogenesis 104 B. Transformation of Liver Derived Epithelial (Oval) Cells 106 IV. Conclusions 108 References 109

x

Contents

6 Contributions of Hepadnavirus Research to Our Understanding of Hepatocarcinogenesis Charles E. R0gler Leslie E. R0gler Deyun Yang Silvana Breiteneder-Geleef Shih Gong Haiping Wang

I. General Overview of Hepadnavirus Animal Models and Hepatocarcinogenesis 113 II. Hepatitis B Virus Envelope Protein (HBsAg) Transgenic Mice 117 A. HBsAg (Line 50-4) Transgenic Model of Hepatocarcinogenesis 117 B. Noncytopathic HBsAg Transgenic Mice: Role of Cytokines in Gene Regulation 118 III. Woodchuck Hepatitis Virus (WHV) Model of Hepatocarcinogenesis 122 A. Toxic Oxygen Radicals and WHV Persistent Infection 122 B. Insertional Mutagenesis 122 C. Integration and Human HCC 125 D. Analysis of Precancerous Lesions and HCCs: The Case for a Role of IGF2 in Tumor Promotion IV. Hepadnavirus X Gene Encodes an Oncogenic Transcriptional Transactivator 129 A. Background 129 B. X Gene Transactivation Mechanism 131 C. In Vitro Assays for Hepadnavirus Transforming Activity 132 D. HBx Transgenic Mice Develop HCC 132 V. Conclusions 134 References 134

7 Apoptosis and Hepatocarcinogenesis R01f Schulte-Hermann Bettina Grasl-Kraupp Wilfried Bursch

I. Apoptosis and Other Types of Active Cell Death 141

125

Contents

II. Active Cell Death in the Liver 145 A. Detection and Quantification 145 B. Models 146 C. Duration of Apoptosis in the Liver 151 III. Biochemical and Molecular Aspects of Apoptosis 152 A. Events Associated with Cell Killing 153 B. Events Associated with Preparation for Active Cell Death 154 C. Signal Factors 156 IV. Active Cell Death in the Stages of Hepatocarcinogenesis 158 A' Cancer Prestages in the Liver 158 B. Kinetic Aspects of Cell Proliferation and Death in Cancer Prestages 160 C. Active Cell Death and Initiation 161 D. Active Cell Death and Tumor Promotion 163 E. Active Cell Death and Tumors 165 V. Conclusions 166 References 167

8 Liver Tumor Promotion and the Suppression of p53-Dependent Cell C y c l e C h e c k p o i n t F u n c t i o n Yingchun Zhang Chia Chia0 Laura L. Byrd David G. Kaufman William K. Kaufmann I~ Introduction 179 II. Mechanisms of Cell Cycle Control 180 III. Cell Cycle Checkpoints, Lifespan Extension, and Genetic Instability 181 IV. Isolation of EL/EGV Hepatocytes and Promotion of Hepatocarcinogenesis in Vitro 183 V. Immortal Rat Hepatocytes Require PB for Clonal Expansion 186 VI. Mechanisms of Promotion of Hepatocarcinogenesis by Phenobarbital 191 VII. Conclusions 193 References 194

xi

xii

Contents

9 Mechanisms of Liver Tumor Promotion JeremyJ. Mills RandyL.Jirtle IvanJ. Boyer I. Introduction 199 II. Stages of Liver Carcinogenesis 201 A. Initiation 201 B. Promotion 202 C. Progression 203 III. Cell Cycle Regulation and Liver Carcinogenesis 204 A. Cyclins and Cyclin-Dependent Kinases B. Rb Gene 206 C. p53 Gene 207 IV. TGF[3 and Liver Carcinogenesis 208 A. TGFI3 and TGFI3 Receptors 210 B. M6P/IGF2 Receptor 212 C. Experimental Results 214 V. Apoptosis and Liver Carcinogenesis 216 VI. Summary 217 References 218

10 Hypomethylation of DNA: An Epigenetic Mechanism That Can Facilitate the Aberrant Oncogene Expression Involved in Liver Carcinogenesis JenniferL. Counts JayI. Goodman 227 I~ Introduction II. Epigenetics 230 230 III. DNA Methylation IV. Working Hypothesis and Experimental Model 235 V. Liver Tumor Promotion: A Role for Hypomethylation of DNA 236 VI. Methyl Deficient Diets 238 VII. DNA Damage and Altered DNA Methylation 240

205

~

Contents

VIII. DNA Methylation and Chemoprevention 241 IX. Differences in DNA Methylation between Rodents and Humans 243 A. Capacity to Maintain DNA Methylation 243 B. Methylation of the 5' Flanking Region in H a - r a s 244 X. Conclusions 245 A. DNA Methylation and Multistage Carcinogenesis 245 B. DNA Methylation and Risk Assessment 246 C. DNA Methylation and Chemoprevention 246 XI. Summary 246 References 247

11

Transgenic Models of Hepatic Growth Regulation and Hepatocarcinogenesis Eric P. Sandgren I. Transgene-Based Strategies for Studying Liver Growth, Development, and Cancer 257 II. Oncogenic Transgenes and Hepatic Neoplasia 259 A. Viral Oncogenes 259 B. Cellular Oncogenes 264 C. Oncogene-Transformed Immortalized Cell Lines 267 D. Conclusions 267 III. Growth Factor Transgenes and Hepatic Neoplasia 268 A. Transforming Growth Factor Alpha 268 B. Insulin-like Growth Factors 270 C. Hepatocyte Growth Factor 271 D. Conclusions 271 IV. Hepatotoxic Transgenes and Liver Neoplasia 271 A. Hepatitis B Virus Surface Antigen 271 B. Urokinase-Type Plasminogen Activator 273 C. Alpha- 1-Antitrypsin 274 D. Conclusions 275

xm

xiv

Contents

276 V. Transgenes and Multistage Carcinogenesis A. Cooperating Events in Transgene-Induced Hepatocarcinogenesis 276 276 B. Coexpression of Multiple Transgenes 278 C. Transgenes and Chemical Carcinogens D. Transgene-Based Mutagenesis Assay Systems 280 E. Conclusions 281 282 VI. Transgenes and Hepatic Growth Regulation A. Liver Gene Expression 282 286 B. Fetal and Neonatal Liver Development C. Liver Regeneration 287 D. Conclusions 288 289 VII. Assessment and Future Directions References 290

12 Genetic Susceptibility to Liver Cancer Norman R. Drinkwater Gang-H0ngLee

I. Introduction 301 II. Genetics of Human Liver Cancer 302 A. Genetic Epidemiology 302 B. Genetic Diseases Associated with an Increased Risk for Liver Cancer 303 III. Genetics of Experimental Liver Cancer 305 A. Variation among Inbred Strains 305 B. Specific Mutations Affecting Hepatocarcinogenesis 313 IV. Conclusion 314 References 315

13 Surgical Treatment of Hepatic Tumors and Its Molecular Basis Ravi S. Chari R. Daniel Beauchamp

I. Introduction 323 II. Diagnosis of Surgical Liver Tumors 324 A. Evaluation of Asymptomatic Liver Mass

324

xv

Contents

III. Indications for Surgical Treatment of Liver Tumors 326 A. Cystic Liver Disease 326 B. Neoplasms of the Liver 328 IV. Surgical Anatomy 336 A. Surgical Resection and Transplantation V. Hepatic Regeneration after Resection and Transplantation: Current Clinical Concepts References 345

339 342

14 Gene Therapy for the Treatment of Inherited and Acquired Diseases of the Liver Brian E. Huber

I. Human Gene Therapy--A Definition 351 II. Strategies for Liver-Directed Gene Therapy 352 A. Gene Replacement (Repair) or Excision Therapy 352 B. Gene Addition Therapy 353 C. Gene Addition Therapy and the Hepatocyte 355 D. Gene Addition Therapy--Ex Vivo versus in Vivo Liver-Directed Gene Therapy 356 III. Gene Transfer Techniques for Liver-Directed Gene Therapy 361 A. Chemical Transfer Methods 362 B. Electroporation 362 C. Microinjection into the Nucleus . . . . D. Scrape Loading 364 E. Macroinjection 364 E Ballistic Barrage 364 G. Receptor-Mediated Gene Delivery 364 H. Liposomal Gene Delivery 366 I. Retroviruses 366 J. Adenoviruses 368 IV. Clinical Applications of Gene Therapy Directed to the Hepatic Compartment 369 A. Metabolic and Plasma Protein Disorders 369

xvi

Contents

B. Primary and Metastatic Liver Cancer 374 C. Viral Diseases of the Liver 375 D. Hepatocellular Transplantation 376 V. Conclusions 377 References 378

Index

385

Contributors Numbers in parentheses indicate the pages on which the authors" contributions begin.

R. Daniel Beauchamp (323) Department of Surgery, Vanderbilt University, Nashville, Tennessee 77550 Ivan J. Boyer (199) MITRE Corporation, Inc., McLean, Virginia 22102 Silvana Breiteneder-Geleff (113) Institute of Clinical Pathology, University of Vienna Medical School, A-1090 Vienna, Austria Nancy L. R. Bucher (1) Department of Pathology, Boston University School of Medicine, Boston, Massachusetts 02118

Wilfried Bursch (141) Institute ffir Tumor Biologie-Cancer Research, A-1090 Wien, Austria Laura L. Byrd (179) Department of Pathology, and Lineberger Comprehensive Cancer Center, University of North Carolina at Chapel Hill, Chapel Hill, North Carolina 27599

Ravi S. Chari (323) Department of Surgery, Duke University Medical Center, Durham, North Carolina 27710 Chia Chiao (179) Department of Pathology, and Lineberger Comprehensive Cancer Center, University of North Carolina at Chapel Hill, Chapel Hill, North Carolina 27599 Paolo M. Comoglio (51) Institute for Cancer Research and Treatment (IRCC), University of Torino School of Medicine, 10126 Torino, Italy Jennifer L. Counts (227) Department of Pharmacology and Toxicology, Michigan State University, East Lansing, Michigan 48824 Norman R. Drinkwater (301) McArdle Laboratory for Cancer Research, University of Wisconsin Medical School, Madison, Wisconsin 53076 Shih Gong (113) Marion Bessin Liver Research Center, Albert Einstein College of Medicine, Bronx, New York 10461 Jay I. Goodman (227) Department of Pharmacology and Toxicology, Michigan State University, East Lansing, Michigan 48824 xvii

xviii

Contributors

Bettina Grasl-Kraupp (141) Institute ffir Tumor Biologie-Cancer Research, A- 1090 Wien, Austria Brian E. Huber (351) Wellcome Research Laboratories, Research Triangle Park, North Carolina 27709 Randy L. Jirtle (199) Department of Radiation Oncology, Duke University Medical Center, Durham, North Carolina 27710 David G. Kaufman (179) Department of Pathology, and Lineberger Comprehensive Cancer Center, University of North Carolina at Chapel Hill, Chapel Hill, North Carolina 27599

William K. Kaufmann (179) Department of Pathology, and Lineberger Comprehensive Cancer Center, University of North Carolina at Chapel Hill, Chapel Hill, North Carolina 27599 Gang-Hong Lee (301) Department of Pathology, Asahikawa Medical College, Asahikawa 078, Japan George K. Michalopoulos (27) Department of Pathology, University of Pittsburgh Medical Center, Pittsburgh, Pennsylvania 15261 Jeremy J. Mills (199) Department of Radiation Oncology, Duke University Medical Center, Durham, North Carolina 27710 Charles E. Rogler (113) Marion Bessin Liver Research Center, Albert Einstein College of Medicine, Bronx, New York 10461 Leslie E. Rogler (113) Marion Bessin Liver Research Center, Albert Einstein College of Medicine, Bronx, New York 10461

Eric P. Sandgren (257) Department of Experimental Pathology, School of Veterinary Medicine, University of Wisconsin-Madison, Madison, Wisconsin 53 706 Rolf Schulte-Hermann (141) Institute fiir Tumor Biologie-Cancer Research, A-1090 Wien, Austria Rebecca Taub (71) Department of Genetics and Medicine, Howard Hughes Medical Institute, University of Pennsylvania School of Medicine, Philadelphia, Pennsylvania 19104 Snorri S. Thorgeirsson (99) Laboratory of Experimental Carcinogenesis, Division of Cancer Etiology, National Cancer Institute, National Institutes of Health, Bethesda, Maryland 20892

Elisa Vigna (51) Institute for Cancer Research and Treatment (IRCC), University of Torino School of Medicine, 10126 Torino, Italy Haiping Wang (113) Marion Bessin Liver Research Center, Albert Einstein College of Medicine, Bronx, New York 10461

Contributors

xix

Deyun Yang (113) Marion Bessin Liver Research Center, Albert Einstein College of Medicine, Bronx, New York 10461 Yingchun Zhang (179) Department of Pathology, and Lineberger Comprehensive Cancer Center, University of North Carolina at Chapel Hill, Chapel Hill, North Carolina 27599


Preface The fact that the liver regenerates following injury has been known since ancient times. According to Greek mythology, Prometheus was chained to a rock for defying Zeus by stealing fire from Mount Olympus. He was subjected to an eagle eating his liver during the day, and it regenerated by night. Because of this marked capacity to regenerate and the ability of chemical carcinogens and viruses to transform hepatocytes, the liver is used extensively as a model for investigating the molecular mechanisms of normal cell proliferation and carcinogenesis. Recently, striking advances have occurred in our understanding of hepatocyte growth regulation and how chemical agents and viruses alter these normal growth regulatory pathways during the genesis of liver tumors. Both in vitro and in vivo models of neoplastic transformation have established that the pathway to neoplasia has multiple steps. Aberrant expression of proto-oncogenes or the expression of mutant forms of these genes (i.e., oncogenes) can lead to neoplastic transformation. The loss of tumor suppressor gene function is also involved in the development of both rodent and human liver cancer. This book demonstrates that the determination of the factors involved in controlling the regeneration of normal liver has led to a clearer understanding of the mechanisms involved in the formation of various liver diseases, including hepatocellular carcinomas. The chapters in this book are grouped into three main research areas. The first section of the book covers the subject of liver regeneration. Liver regeneration has been investigated for many years, particularly since 1931 when Higgins and Anderson published a classic paper describing a partial hepatectomy technique in rats that reproducibly stimulates the liver to regeneration. An overview of liver regeneration research both then and now is described in the first chapter by one of the premier scientists in this field of investigation, Dr. Nancy Bucher. The growth factor that plays a central role in stimulating hepatocyte proliferation following liver injury, hepatocyte growth factor, and its receptor (c-met) are discussed in Chapters 2 and 3, respectively. The immediate-early response genes that are expressed in the early stages of liver regeneration are described in Chapter 4. The second section of this book deals with the subject of liver carcinogenesis. The topics were chosen to provide an overview of this immense subject. xxi

xxii

Preface

Chapter 5 covers the role that hepatic stem cells play in both liver regeneration and carcinogenesis. It is well known that hepatitis virus infection is a major risk factor for the development of human hepatocellular carcinomas. The contribution of hepadnavirus research to our understanding of liver cancer is discussed in Chapter 6. Programmed cell death, apoptosis, has been increasingly shown to be mechanistically involved in liver tumor promotion and carcinogenesis as discussed in Chapter 7. Transforming growth factor 13and the suppression of p5 3-dependent cell cycle checkpoint function as they relate to liver tumor promotion are discussed in Chapters 8 and 9. In Chapter 10 it is postulated that hypomethylation of the DNA may represent an epigenetic mechanism for hepatocyte transformation. The use of transgenic animals to ascertain directly the role of growth factors, oncogenes, etc., in liver carcinogenesis is a powerful molecular technique thoroughly presented in Chapter 11. It is also important to realize that humans vary in their susceptibility to tumor formation. Chapter 12 describes the importance of genetics in the development of liver tumors in both humans and animals. The final section of this book deals with two vastly different techniques for treating liver cancer. Although hepatocellular carcinoma is one of the most common neoplasms in the world, primary cancer of the liver is relatively rare in the United States. Hepatocellular carcinomas are, however, still of significance in the United States because they are highly lethal; the 5-year survival rate is less than 10%. One method of treating primary and metastatic liver tumors is surgical resection. When the malignancy is not amenable to a simple resection, the whole liver can now be transplanted. These surgical procedures for the treatment of liver cancer are described in Chapter 13. As the genes involved in liver carcinogenesis are elucidated, it becomes more probable that this information will be used to develop successful molecular strategies for cancer treatment. The potential of using gene therapy approaches to treat liver tumors and genetic liver diseases is discussed in Chapter 14. In conclusion, the veritable explosion of scientific information that has occurred recently in liver biology encompasses many research disciplines. This book is an attempt to bring these diverse research results together in a coherent manner. Scientists who will benefit from this book include toxicologists, virologists, molecular biologists, cell biologists, cancer biologists, pharmacologists, pathologists, surgeons, and gastroenterologists who are interested in furthering their understanding of the molecular mechanisms controlling liver regeneration and hepatocellular carcinogenesis. I would like to thank Ms. Charlotte Brabants at Academic Press for her continued support and patience, the contributors of this book for writing their chapters when it is increasingly difficult to find the time for such

Preface

xxiii

scholarly endeavors, and Ms. Roxanne Scroggs for her secretarial assistance.

Randy L. Jirtle


1 Liver Regeneration Then and Now N a n c y L. R. B u c h e r Department of Pathology Boston UniversitySchool of Medicine, Boston, Massachusetts02118

I. Landmarks As is well known to devotees of liver regeneration, according to the Prometheus myth the ancient Greeks were aware not only that the liver rapidly grows to restore lost tissue, but also that it continues to do so after repeated insults. After that, as far as we know, it was not until late in the nineteenth century that Canalis carried out the first scientifically motivated partial hepatectomy. By 1894 the phenomenon of regeneration had been investigated in rats, mice, rabbits, and dogs (Bresnick, 1971). Now, 100 years later we still do not fully understand the mechanisms involved despite accelerating progress. An early breakthrough was the report of Higgins and Anderson (1931) that detailed a simple surgical procedure for performing a partial hepatectomy in the rat, thereby opening the way for reproducible quantitative studies. The rat is a creature well designed for the study of liver regeneration. The structure of its liver is such that excision of the two main lobes consistently removes 68% of the whole organ; this percentage is independent of rat strain and animal age (Bucher and Swaffield, 1964). Moreover, although the liver regenerates in all Species that have been examined so far, the regrowth is the most rapid and dramatic in the rat; even in the mouse it is slower by about a day. Although most studies have been performed in rats and mice; dogs, and rabbits have also been used. Investigation of the regeneration of human liver is limited to clinical and cell culture studies, which have received

Liver Regeneration and Carcinogenesis Copyright 9 1995 by Academic Press, Inc. All rights of reproduction in any form reserved.

2

Nancy L. R. Bucher

recent impetus from the availability of hepatocytes from livers to be used for transplants (see Chapter 13). In 1936 Brues and colleagues (Brues et al., 1936) carried out the first really basic scientific studies on the regenerating liver, showing that liver mass begins to increase within a few hours after partial hepatectomy, whereas restoration of cell number starts a day later. In addition, they showed that in the regenerating liver the rate of DNA synthesis is greatly increased, as determined by labeling with 32pi, then newly available and heralding the modern era of radioisotopes (Brues et al., 1944). The advent of 14C and 3H soon thereafter was indeed a boon to those investigators who had of necessity until then depended entirely on counting mitotic figures to quantify growth rates. A third landmark, occurring in the late 1960s, was the development by Berry and Friend (1969) of the collagenase perfusion for rat hepatocyte isolation and culture, opening whole new horizons for liver research. This enzymatic process was latter modified for isolating human hepatocytes from surgically removed and discarded pieces of human liver (Strom et al., 1982). Thus, hepatocytes from various species can now be studied in a controlled environment, since they are released from growth regulatory influences imposed by the body. There are numerous review articles on liver regeneration that provide a chronological overview of the progression of this research. The following list includes many, but not all of these reviews: Fishback, 1929; Mann, 1944; Harkness, 1957; Weinbren, 1959; Leduc, 1964; Bresnick, 1971; Bucher and Malt, 1971; Hays, 1974; Lewan et al., 1977; Starzl and Terblanche, 1979; Bucher, 1963, 1967a,b, 1982, 1987; Alison, 1986; Fleig, 1988; Leffert et al., 1988; Fausto and Mead, 1989; Fausto, 1990; Michalopoulos, 1990; Fausto, 1991; Bucher, 1991; Andrus et al., 1991; Bucher and Strain, 1992; DuBois et al., 1994. The present review focuses on regeneration of rat liver in vivo.

II. N o r m a l A d u l t R a t Liver In normal adult rat liver 60% of the cells are hepatocytes and the remainder are various types of nonparenchymal cells. The hepatocytes, however, are much larger and constitute about 90% of the liver volume. Hepatocytes are not all alike. There are many well-studied functional differences between mature, differentiated hepatocytes which occur in a gradation of zones from the periportal to the pericentral regions of the liver lobule (Gumucio and Chianale, 1988; Gumucio and Berkowitz, 1992). Hepatocytes are also normally long lived, and the liver continues to grow as the animal ages, albeit at a diminishing rate. A peculiarity of rodent livers is that although at birth the

1.

Liver Regeneration Then and N o w

3

hepatocytes are mononuclear and diploid, as the animals mature many cells become polyploid; in the rat about /30% become binucleate, 50 to 70% tetraploid, and I to 2% octaploid (Wilson and Leduc, 1948; Bucher, 1967a, b; Tamura et al., 1992). During liver regeneration the number of binucleate cells decreases, but the overall ploidy again increases, peaking at around 72 hr, and then gradually reverting to normal. Even in young animals all but a tiny fraction (0.01 to 0.05%) of the hepatocytes are noncycling G o phase cells, and, despite the polyploidy, they require a new round of DNA synthesis before they can divide. Although for most purposes these few proliferating cells can be ignored, in the early years careful counting of mitotic figures revealed a diurnal oscillation of activity, increasing 2- to 5-fold from a nadir at around 8 A M tO a peak at about 8 P M (Harkness, 1957; Bucher, 1963, 1967; DuBois et al., 1994). In contrast, in regenerating livers the mitotic activity increases 50- to 100-fold. So why is this minor oscillation worthy of note? The answer is that certain genes currently being investigated also exhibit diurnal periodicity, as illustrated by the wide swings in D binding protein (DBP) expression (Mueller et al., 1990). ICER, an isoform of the cyclic AMP-response element modulator (CREM) and the 3-hydroxy-3-methylglutaryl (HMG) CoA reductase genes behave similarly (Masquilier et al., 1993; Bach et al., 1969; Edwards et al., 1972). The diurnal periodicity program also carries over into the regenerating liver, so that although the interval from partial hepatectomy to the peak of DNA synthesis in young adult rats regularly occurs at 22 to 24 hr, its magnitude varies depending on the time of day that the liver samples are taken (Klinge and Mathyl, 1969). It is entrained by illumination and feeding patterns; rats are nocturnal and the feeding and activity programs can be reversed by reversing the light/dark cycle. Food intake is also a major determinant; if access to food is restricted to 2 to 4 hr at a constant time of day for 3 weeks followed by partial hepatectomy, the peak rate of DNA synthesis is increased approximately three-fold (Hopkins et al., 1973a,b; Bucher et al., 1978a).

III. Liver Regeneration It is generally thought that the liver regenerates in response to an excessive metabolic workload imposed by the body. The questions are what sort(s) of workload(s)? How is metabolic work translated into a growth stimulus? What starts the growth, and what stops it? What are the growth effectors? What are the molecular mechanisms? These questions are now being productively addressed in in vitro systems, but hepatocyte culture methodology was not developed until around 1970. Consequently, all of the earlier exper-

4

Nancy L. R. Bucher

iments were carried out in the whole animal. Whole animal studies were valuable for delineating the salient features of the growth process, are central to an understanding of hepatic growth regulation, and continue to deserve attention because they involve physiological aspects of the growth regulation that do not exist in cultures. The term "regeneration" is firmly embedded in the literature, but is inaccurate in that the excised lobes do not regrow. Instead, the remaining lobes enlarge, undergoing what is more accurately termed a compensatory hyperplasia. The histological architecture is preserved throughout this process. In the normal adult rat liver, the cells are highly differentiated and essentially all in a state of growth arrest, or G 0. They are induced to enter the cell cycle by cell loss or functional inadequacy, for example due to surgical resection, infectious, toxic or physical injury, or to metabolic imbalances caused by severe diabetes, to drastic changes in nutrients, or to pregnancy in rats and mice where multiple fetuses occur in a small animal. When the excess metabolic workload is removed, the liver shrinks back to normal size by apoptosis (See Chapter 7). Partial hepatectomy is widely used to induce liver regeneration, because it is fast and easy to perform, well tolerated, delivers a quantifiable stimulus and is free of the side effects and damage to surviving cells associated with carbon tetrachloride (CCI4) or other commonly used toxic agents. Following the standard 68% hepatectomy, the potentially interrelated growth and stress associated changes start almost immediately. The earliest documented events include monovalent cation fluxes, changes in intracellular pH, alterations in amino acid transport (Leffert et al., 1988), and activation of certain liver function specific and immediate-early genes (see Chapter 4) (Fausto, 1990; Mohn et al., 1991a,b). Early events also involve activation of second messengers and signaling pathways and further changes in gene expression and attendant alterations in the biochemistry of the cells as they undergo the transition from G Oto the G 1 phase of the cell cycle and continue through G 1 into the S phase. DNA synthesis, marking the S phase of the cell cycle, begins at about 14 to 16 hr following a partial hepatectomy in young adult rats, rises steeply to a sharp peak at 22 to 24 hr then falls and continues at a diminished level until the original liver complement is restored in a little more than a week. Mitosis follows DNA synthesis 6 to 8 hr later. DNA synthesis and mitosis are observed to occur first in hepatocytes in the periportal area, subsequently spreading centrally. During the first 72 hr, when the major regeneration has occurred, about 80% of the new hepatocytes are formed in the periportal region of the liver lobules. DNA synthesis in littoral and ductal cells follows that in hepatocytes and peaks 12 to 24 hr later (Harkness, 1957; Grisham, 1962). The regenerative process is most rapid in weanling animals and slowest in old animals (Bucher, 1967).

1.

Liver Regeneration Then and Now

5

In regenerating livers DNA synthesis first appears periportally, but progresses in time to involve the whole lobule with the possible exception of a very few cells close to the central vein (Grisham, 1962). The fact that all hepatocytes have proliferative potential has been shown in animals given [3H]thymidine repeatedly for several days, during which essentially all of the hepatocytes become labeled (Michalopoulos, 1990). When rats are subjected to partial hepatectomies at monthly intervals for a year, the livers regenerate repeatedly. Although at first the existing lobules enlarge, new lobules ultimately develop after additional hepatectomies (Ingle and Baker, 1957; Simpson and Finkh, 1963). The rate at which regrowth occurs is proportional to the extent of liver loss. If the deficit is small, the liver regrows more slowly, despite its high growth potential (Bucher, 1967; Bucher and Malt, 1971). If the liver resection is increased beyond the usual 68% up to 80 to 90%, however, the animals are severely stressed and DNA synthesis is not further increased (Caruana et al., 1986).

IV. Hepatocyte Priming Despite years of effort by numerous investigators, uncertainty still persists regarding the actual signals that initiate, maintain, regulate, and terminate liver regeneration. The answers are likely to remain equivocal until the growth process is clearly delineated at the molecular level; however, earlier work with animal models has brought this problem into clearer focus. Cross-circulation of blood between partially hepatectomized and normal rats, initially via capillaries (parabiosis) and later via arteriovenous shunting through polyethylene cannulas, demonstrates that blood from a hepatectomized partner will stimulate DNA synthesis in the intact liver of the normal partner. The stimulus is proportional to the amount of liver excised in the hepatectomized partner (Bucher et al., 1951; Moolten and Bucher, 1967; Fleig, 1988). Additional evidence of blood borne signaling comes from experiments in which autologous liver grafts and transplanted hepatocytes are stimulated to grow by partial resection of the liver of the host animal. Thus, although initially controversial, it is now well established that liver regeneration is tightly regulated by substances carried in the blood (Grisham et al., 1964; Jirtle and Michalopoulos, 1982; Bucher and Strain, 1992). In the cross-circulation experiments it was found necessary to maintain the blood exchange for at least 12 to 14 hr; shorter exchanges failed to stimulate DNA synthesis. However, if the exchange was interrupted for 2 to 4 hr during the early period of a total elapsed time of 14 hr, DNA synthesis still occurred in the hepatocytes of the intact partner (Bucher et al., 1969). It seemed that initiation of DNA synthesis must progress in a stepwise fashion,

6

Nancy L. R. Bucher

i.e., that the first step activates the process and a booster is needed to move the cells forward into the S phase. Further support for this notion came from the finding that although DNA synthesis following partial hepatectomy is constant at 22 to 24 hr in young adult rats, this interval could be shortened by about 8 hr if sham operations were performed from several hours to several days prior to the partial hepatectomy. Thus, regeneration seems to gain a head start by a surgical pretreatment, which by itself does not induce DNA synthesis. This finding suggests that response to stress may be an integral part of the liver regenerative process (Moolten et al., 1970; Fausto, 1990). This stepwise induction of liver growth within the animal has now been amply confirmed in cell culture, and found widely applicable to cell growth in general. The process of inducing hepatocytes to undergo transition from the G Oto the G 1 phase of the cell cycle is termed "priming." Further stimulation is required for progression through G 1 into the S phase. Activation of the heat shock protein gene, hsp 70, occurs during the prereplicative period of liver regeneration (Ohmori et al., 1990), and it may play a role in regulating the cell cycle (Schlesinger, 1990). It was shown in 1949 (Leduc) that when mice were changed from low to high protein diets, the liver exhibited a wave of mitotic activity, the peak occurring earlier when the protein enrichment was greater. Similarly, when rats were fed only 20% glucose for 3 days, followed by a balanced amino acid meal (casein hydrolysate by gavage), the ensuing peak of DNA synthesis appeared about 8 hr earlier than in normally fed hepatectomized controis, and was about 70% as high (Bucher et al., 1978a). The hepatocytes could thus be primed by means other than surgical stress. These earlier studies have been confirmed and extended by recent research. The protein-free diet induces the expression of mRNAs of several growth associated immediate-early genes, now recognized as evidence that priming has taken place. Prominent among immediate early genes commonly expressed following growth stimulation in most systems, including regenerating liver, are c-myc, c-jun and c-fos. In the protein deprivation model, however, c-fos mRNA is not expressed (Horikawa et al., 1986; Mead et al., 1990; Fausto, 1990). Are only particular ones of these genes required for liver growth? Do the various genes involved in priming and progression through G1 differ depending on the metabolic status of the cells or the nature of the stimulus? Do various pathways converge to control liver growth? These fundamental questions still need to be answered. Although pretreatments prior to partial (68 %) hepatectomy serve to head start the growth response, either extra stress or glucose administration at the time of the hepatectomy moderately depresses liver regeneration (Moolten et al., 1970; Carter et al., 1989; Holecek et al., 1991; Bucher and Strain, 1992). Following excision of 90% of the liver, survival is low and regen-

1.


7

eration does not occur without supportive treatment. Administration of glucose permits survival and regeneration, but it is no greater than the diminished regeneration occurring in glucose treated 68% hepatectomized controls (Gaub and Iversen, 1984; Caruana et al., 1986). From the foregoing discussion, it is clear that stress and the metabolic status of the cells are involved in the priming process, so that they are then responsive to subsequent growth signaling; however, following liver loss, only the liver grows. Whether this tissue growth specificity is determined by a unique hepatic response to these seemingly nonspecific stimuli, or to more specific stimulation of previously primed cells at a later stage remains an unanswered question (see Chapter 2).

V. Regeneration Signals The signals that initiate regeneration are clearly derived from extrahepatic sources, since they are transmitted by the blood. The signaling molecules are generally thought to be combinations of interacting hormones, growth factors, and cytokines, although direct action of nutrients or metabolites has not been ruled out. Table 1 lists a number of the effectors postulated to be mechanistically involved in regulating liver regeneration. The complexity of this problem is compounded by the fact that the same growth factor can act both positively and negatively; transforming growth factor 13(TGFI3), hepatocyte growth factor (HGF), and glucagon are well-known examples (see Chapter 2) (Michalopoulos, 1990; Sporn and Roberts, 1990). Infusion of insulin and epidermal growth factor (EGF) individually into normal rats weakly stimulates DNA synthesis, but when combined, the stimulus is substantial. Addition of glucagon to this combination is inhibitory, but when insulin is omitted glucagon becomes weakly stimulatory. Similar effects are demonstrable in hepatocyte cultures. Thus, it appears that hormones and growth factors interact in synergistic or antagonistic combinations, and the same substance may act positively or negatively, depending on the circumstances. A. Hormones 1. Insulin and Glucagon

Early work pointed to the importance of a portal blood supply, and emphasized insulin as a major hepatocyte growth modulator (Hays, 1974; Starzl and Terblanche, 1979). In splanchnic eviscerated rats (i.e., ablation of the gastrointestinal tract, pancreas, and spleen), partial hepatectomy results in a delayed and weakened DNA synthesis--about 20% of the post partial

8

Nancy L. R. Bucher

Table 1 Growth Effectors Proposed to Control Liver Regeneration I. HORMONES Insulin and glucagon Norepinephrine, vasopressin, angiotensin II, and neurotensin Growth hormone and insulin-likegrowth factors 1 and 2 (IGF1 and 2) Parathyroid hormone, calcitonin, and dihydroxycholecalciferol Glucocorticoids Thyroid hormones Prolactin Prostaglandins Estrogens and androgens II. GROWTH FACTORS A. Stimulators

Epidermal growth factor (EGF) Transforming growth factor oL(TGFa) Hepatocyte growth factor (HGF) Acidic fibroblast growth factor (aFGF) Hepatocyte stimulatory substance (HSS) B. Inhibitors

Transforming growth factor 13(TGF13) Activin Hepatocyte proliferation inhibitor (HPI) Platelet derived growth inhibitors a and 13(PDGI a and 13) HI. CYTOKINES Interleukins 1 and 6 (IL-1 and IL-6) Tumor necrosis factor el (TNFel)

hepatectomy response of control animals. Thus, in severely stressed animals, the liver still responds weakly to partial hepatectomy, even in the apparent absence of portal blood factors including insulin (Bucher and Swaffield, 1973). Treatment of these rats with insulin and glucagon separately has minimal effects, but in combination these two hormones dramatically restored the DNA synthesis to that in normal regenerating livers (Bucher and Swaffield, 1975). Administration of insulin and glucagon at the same doses to control rats, however, does not stimulate liver growth (Bucher and Strain, 1992). Taken together, these observations suggest that insulin and glucagon enhance the action of prior growth initiators. Other whole animal studies also point to modulating roles for these two hormones. For example, administration of neutralizing insulin antibodies diminishes but does not abolish liver regeneration. Whether under appropriate conditions insulin can serve as a prime mover in liver regeneration cannot be totally excluded. An example is severe diabetes, in which a brief administration of insulin potently stimulates hepatocyte proliferation (Bucher and Strain, 1992). Glucagon has also been reported to stimulate hepa-

1.


9

tocyte DNA synthesis (Hasegawa and Koga, 1977). In the experiments described, pharmacological doses of glucagon were administered, whereas insulin was used in the physiological range. Nevertheless, definitive evidence for insulin and glucagon being primary stimulators of liver growth is elusive, because ancillary actions of unknown growth effectors cannot be excluded in whole animal studies. 2. Norepinephrine, Vasopressin, Angiotensin II, and Neurotensin

Norepinephrine and vasopressin, which have similar effects on hepatocytes, also appear to be modulators of liver regeneration. Norepinephrine acts at the early stage of liver regeneration as evidenced by a reduction of DNA synthesis by prazosin, a highly specific c~-adrenergic antagonist. Moreover, adrenergic innervation is found to be substantially increased in regenerating liver, and the growth process is greatly reduced by surgical or chemical sympathetic denervation. Within 2 hr post partial hepatectomy plasma catecholamines increase. A shift from the largely ~1 type to the [3-type adrenergic receptors occurs after regeneration is underway, but 13-adrenoreceptor blocking agents fail to inhibit and [3-agonists fail to enhance regeneration. Similar observations have been made in EGF-stimulated hepatocyte cultures. In addition, the growth inhibitory action of TGF[31 is counteracted by the addition of norepinephrine (Michalopoulos, 1990; Bucher and Strain, 1992). Liver regeneration is also depressed in the hereditary vasopressin-deficient Brattleboro rat strain, and vasopressin administration largely restores the hepatic proliferative capability (Russell and Bucher, 1983). In the rat, vasopressin acts through the Ca 2+ mediated V-1 receptor, which is abundant in rat hepatocytes, whereas these high affinity receptors appear to be barely detectable in rabbit and human hepatocytes. Moreover, human hepatocyte cultures are unresponsive to this hormone (Bucher and Strain, 1992; DuBois et al., 1994). Possibly other hormonal substances carry out in human hepatocytes whatever interactive functions vasopressin fulfills in the rat. In EGF-stimulated hepatocyte cultures the effects of norepinephrine, vasopressin, angiotensin II, and neurotensin (Hasegawa et al., 1994) or other neurotransmitters resemble, to slightly differing extents, the effects of e~ladrenergic agents in vivo. These hormones also have similarities with regard to hepatic carbohydrate metabolism, suggesting that like insulin and glucagon they may function as regulators of nutrient availability and utilization in the regenerating livers. Epinephrine appears to be the most potent growth effector of this hormone group, at least in vitro.

10

Nancy L. R. Bucher

3. Growth Hormone and Insulin-Like Growth Factors 1 and 2

Growth hormone amplifies DNA synthesis following partial hepatectomy and accelerates it by acting as a priming agent (Moolten et al., 1970). Conversely, hypophysectomy delays and diminishes the regenerative process (Bucher and Strain, 1992; Ekberg et al., 1992; DuBois et al., 1994). Most, but not all of the effects of growth hormone are mediated through the insulin-like growth factors, IGF1 and IGF2, which are mainly produced by the liver. Although hepatocytes have been reported to contain few IGF1 receptors (IGFlr), one of the most highly expressed immediate-early genes in regenerating liver is insulin-like growth factor binding protein-1 (IGFBP-1) (see Chapter 4). The IGFBP-1 peptide has been implicated in modulating the mitogenic effect of IGFs on tissues. It is not expressed in mitogen-stimulated fibroblasts and may be specific to regenerating liver (Mohn et al., 1991b); the mRNA level is unaffected by sham operation (Ghahary et al., 1992). Although the IGFBP-1 mRNA is found in both parenchymal and nonparenchymal cells following partial hepatectomy, the protein appears to be present only in the hepatocytes (Lee et al., 1994). High levels of the mannose 6-phosphate/insulin-like growth factor 2 receptor (M6P/IGF2r) are also found in hepatocytes during liver regeneration (Scott et al., 1990; Jirtle et al., 1991). IGF2 interacts with these receptors which may, in conjunction, with IGFBP-1, mediate possible actions of the IGFs in regenerating liver (Lee et al., 1994). The latent complex of TGFI31 also binds to the M6P/IGF2r, thereby facilitating the activation of TGFI31 (Dennis and Rifkin, 1991). Therefore, it appears that M6P/IGF2r is involved in the liver growth process, but its function has not been totally defined (Jirtle et al., 1991). 4. Parathyroid Hormone, Calcitonin, Calcium, and Dihydroxycholecalciferol

This group of hormones is also reported to influence the course of liver regeneration, but not to initiate it. Liver regeneration is diminished in parathyroidectomized rats, and is restored by calcium injection. Secretion of both calcitonin and parathyroid hormone increase soon after partial hepatectomy, and the calmodulin level also rises. Both hormones, as well as the vitamin D metabolite 1-oL,25-dihydroxycholecalciferol [1,25-(OH)2-D3], which is actually a steroid hormone, probably all exert their regulatory function through calcium ions. The 1,25-(OH)2-D 3, has been suggested to participate in the regulation of several DNA replication-linked genes (c-myc, c-myb, and histone H4) that are essential for cell proliferation

1.


11

(Bucher and Strain, 1992). These hormones are of interest because of mounting evidence for the importance of calcium in early signaling events associated with activation of growth in various types of cells including hepatocytes. S. Other Classic Hormones

Through the years, on the basis of both in vivo and in vitro studies, most of the classic hormones have been put forward as having possible, though not clearly defined roles in regulating liver regeneration. However, with a few exceptions most of the evidence lacks firm underpinnings. Nevertheless, they cannot be excluded form participating in liver regeneration because of possible interactions with other effectors. a. Glucocorticoids In vivo studies on the possible role of glucocorticoids in liver regeneration are equivocal, and suggest that the glucocorticoids are relatively unimportant except under special conditions (Bucher and Strain, 1992; DuBois et al., 1994). In hepatocyte cultures, however, these hormones are widely used to promote cell survival and function; effects on growth vary with dosage and culture conditions, and these vary considerably among different laboratories. Consequently, the extent of their influence on the regeneration process remains unclear. b. Thyroid Hormones The effects of thyroid hormones on liver regeneration are likely to be primarily dependent on their effects on metabolism. c. Prolactin regeneration.

Prolactin lacks definitive support as a regulator of liver

d. Prostaglandins Prostaglandins (PG) may be involved in liver regeneration according to a few reports. PGE 2 reportedly undergoes a transient increase during liver regeneration that is inhibited by indomethacin, which inhibits DNA synthesis as well. Additional work with inhibitors of prostaglandin and thromboxane production in regenerating liver and with arachidonati, PGE2, and PGF2~ in cultures suggests that these hormones may act at a later stage in the cell cycle (Bucher, 1991; Bucher and Strain, 1992). e. Estrogens and Androgens On the basis of a number of somewhat discordant studies, estrogens and androgens appear not to play a major role in liver regeneration. Certain sexually dimorphic aspects of liver function are recognized, and estrogens rather then androgens seem to exert a modest influence (Bucher and Strain, 1992). In a recent report, estradiol has been

12

Nancy L. R. Bucher

found to enhance the stimulatory effects of EGF, transforming growth factor oL (TGFcx), HGF, and acidic fibroblast growth factor (aFGF) by about two-fold in hepatocyte cultures (Ni and Yager, 1994), but long-term exposure of rats to ethinyl estradiol reduces the proliferative capacity of hepatocytes (see Chapter 9) (Yager et al., 1994). In general, hormones seem to be growth modulators. The hormones of most interest and most studied so far are insulin, glucagon, and norepinephrine. Both in vivo and in vitro, they substantially influence the activity of other growth effectors toward hepatocytes. Therapeutic aspects of liver transplantation have revived interest in these hormones. B. Growth Factors

"Cytokine" is used interchangeably with "growth factor" by some investigators, but it is restricted to products of the immune system by others. These substances are polypeptides which act at short range in several modes, designated as paracrine, autocrine, and juxtacrine. They are thereby distinguished from the classical endocrine hormones, which act at long range. Because of limited availability, most studies of these growth effectors largely focused on their molecular biology, and have been carried out primarily in cell culture. While this work has contributed measurably to our understanding of the mechanisms through which these substances operate, the studies are conducted on cells in artificial environments that influence cellular behavior. In vitro studies can tell us what cells can do under specific conditions, but not necessarily what they actually do in vivo. Consequently, they are largely beyond the scope of this review, which is concerned with hepatocyte performance within the animal. Among the generally positive acting growth factors, HGF and TGFcx are the most potent. The somewhat less effective factors are EGF and aFGF, which is probably the least potent. These comparative potencies are of necessity based on in vitro studies. Therefore, they are not necessarily meaningful, because under physiological conditions various growth effectors, including hormones, growth factors, cytokines, nutrients, and metabolites, as well as cell-cell and cell-extracellular matrix interactions are integrated to control and fine tune the liver regenerative process. HGF and aFGF, as well as a recently described form of EGF termed heparin binding EGF (HB-EGF) bind to heparin, and also to heparan sulfate proteoglycans (HSPGs) in the extracellular matrix, where they may be sequestered, and interact with other HSPGs as well as with specific receptors on the cell surface. In a preliminary report, HB-EGF stimulates rat hepatocyte growth in vitro as potently as HGF, and like HGF, its mRNA is expressed in the nonparenchymal cells but not in hepatocytes (Higashiyama et

1.


13

al., 1991, 1993; Ito et al., 1994). Of the generally negative acting growth factors, the most extensively studied is TGFI31. 1. Epidermal Growth Factor

EGF is an effective mitogen for a wide variety of cells. It is originally isolated from the submaxillary salivary gland of the male mouse where it is present in high abundance, but it occurs in various other tissues as well. It has high homology and essentially identical biological activity to urogastrone, which is now considered to be the human form of EGE Other members of the EGF superfamily include HB-EGF, and TGFa which bind to the EGF receptor. The biological effects of EGF and TGFa are similar but not identical (Carpenter and Wahl, 1990). EGF was the first growth effector found to be mitogenic for hepatocytes, demonstrated both in vivo and in vitro, and greatly augmented by combination with insulin (Bucher and Strain, 1992). In addition to initiating liver regeneration, EGF also promotes fat accumulation, glycogen synthesis, gluconeogenesis, and hepatocyte motility. Furthermore, it elicits additional biological responses that are unrelated to mitogenesis (St. Hilaire and Jones, 1982; Carpenter and Wahl, 1990; Carpenter and Cohen, 1992). The EGF/TGFa receptor decreases during the prereplicative period, and continues to decline through the initial peak of DNA synthesis, reaching a nadir at 36 to 48 hr (DuBois et al., 1994). Several studies have dealt with the uptake of EGF from the blood, its binding to the receptor, internalization, and subsequent intracellular processing of the receptor complex (Michalopoulos, 1990). Liver regeneration in sialoadenectomized mice is delayed and DNA synthesis reduced; it is restored to normal by EGF treatment (Noguchi et al., 1991). In a recent preliminary report, involving a highly sensitive reverse transcriptase-polymerase chain reaction (RT-PCR) assay, EGF mRNA has been found to increase 30-fold within 15 min after partial hepatectomy, followed by a steep decline to subnormal levels in 4 to 8 hr. This occurs only in the hepatocyte population (Mullhaupt et al., 1993). Although EGF has been extensively studied in hepatocytes both in vivo and especially in vitro, there are many divergencies and thus its role in liver regeneration remains unclear (see Chapter 2) (Fausto, 1990, 1991; Michalopoulos, 1990; Selden and Hodgson, 1991; Bucher and Strain, 1992). 2. Transforming Growth Factor

In hepatocyte cultures, TGFa is more potent than EGF in stimulating DNA synthesis (Fausto, 1991; Webber et al., 1993). The precursors of EGF,

14

Nancy L. R. Bucher

TGFcx, and certain other growth factors are transmembrane glycoproteins that are cleaved to yield soluble growth factors, which act locally in paracrine or autocrine modes. The membrane bound forms, however, are also active and can interact with receptors on the surface of adjacent cells, thereby sustaining cell-cell adhesion as well as direct cell-cell (juxtacrine) stimulation (Massagu6, 1990). TGFa appears to be a more likely physiological regulator of liver regeneration than EGF (Fausto, 1990, 1991; Michalopoulos, 1990; Selden and Hodgson, 1991; Bucher and Strain, 1992). TGF(x mRNA increases within the hepatocytes at 4 to 8 hr after partial hepatectomy, reaching a maximum of 8- to 10-fold above sham-hepatectomized controis by 12 to 24 hr. The mature form of the TGFcx peptide increases by 4 hr after partial hepatectomy, is maximal at 18 hr, and remains significantly increased at 48 hr. These changes slightly precede the DNA synthesis peak. In a later study, although the TGF(x content increased prior to the DNA synthesis peak, the mature form was not detected until 48 hr post partial hepatectomy, and was still present at 96 hr. These observations suggest that the membrane anchored precursor and the mature forms of TGF(x may have different functions in liver regeneration (Russell et al., 1993). 3. H e p a t o c y t e G r o w t h F a c t o r / S c a t t e r Factor

HGF activity was first detected with the use of hepatocyte cultures, and was shown to be present in serum, plasma, and platelets from normal and partially hepatectomized rats (see Chapter 2) (Strain et al., 1982; Michalopoulos et al., 1982; Russell et al., 1984a,b; Nakamura et al., 1984), and serum from patients in fulminant hepatic failure (Ghoda et al., 1988). It was subsequently isolated and purified from rat blood platelets and serum (Nakamura et al., 1987; Zarnegar and Michalopoulos, 1989; Matsumoto and Nakamura, 1991, 1992). HGF was finally cloned and sequenced by Nakamura et al. (1989). HGF was later found to be identical to scatter factor (Naldini et al., 1991), a motility promoting substance secreted by fibroblasts of mouse or human origin, and identified and purified about the same time as HGF (Stoker et al., 1987). HGF is produced by various cells of mesenchymal origin in the liver, pancreas, brain, thyroid, salivary and Brunner's glands, kidney, and lung, but not by the epithelial cells in these organs. Within the liver, HGF is produced principally by the nonparenchymal cells of Ito (fat-storing cells). The high-affinity signaling receptor for HGF is the protein product of the c-met proto-oncogene (see Chapters 2 and 3), which is present in hepatocytes and a variety of other epithelial cells which HGF stimulates to proliferate. Like various other growth factors, HGF also acts negatively, inhibiting growth of several tumors, including HepG 2 cells and even normal

1.


15

hepatocytes when used at high concentrations (Michalopoulos, 1990; Matsumoto and Nakamura, 1991, 1992; DuBois et al., 1994). The serum concentration of HGF rises dramatically within an hour after partial hepatectomy, apparently from extrahepatic sources. About 12 hr later, HGF mRNA increases in the nonparenchymal cells, becoming maximal by 24 hr. The HGF activity also increases during this time, suggesting a paracrine mode of action specifically involving the liver. Although the HGF mRNA also increases in various nonhepatic tissues (e.g., lung) following partial hepatectomy, only the liver cells actively proliferate (DuBois et al., 1994). Similar increases in HGF are found in rat livers following other means of inducing regenerative activity, including liver deficiency induced by damage with CC14, galactosamine, ischemia, or physical injury. The timing of the HGF response is commensurate with the overall type of injury and extent of damage induced (Matsumoto and Nakamura, 1991, 1992). It appears likely from these observations, coupled with the remarkably high levels of HGF in patients with fulminant hepatic failure, as well as a large number of in vitro studies, that HGF has a significant physiological role in liver regeneration, but how it fits into the overall scheme still needs clarification. 4. Acidic Fibroblast Growth Factor

Both aFGF (also termed heparin binding growth factor-I) and HGF bind heparin, and consequently bind to heparin-like sites in the extracellular matrix, notably heparan sulfate proteoglycans, where they may serve as reservoirs from which they can be released as soluble growth factors. Alternatively, they may remain matrix bound and interact with heparan sulfate proteoglycans or specific receptors on the cell surface. In regenerating liver, expression of aFGF precedes the expression of TGFo~, and occurs in both the parenchymal and nonparenchymal cells, persisting for 7 days after the partial hepatectomy. It is secreted by hepatocytes and nonparenchymal cells, the maximal rate approximately coinciding with the peak rate of DNA synthesis. In hepatocyte cultures, the aFGF peptide binds to high affinity receptors and stimulates DNA synthesis with about one-third the potency of EGF (Kan et al., 1989; Michalopoulos, 1990; Matsumoto and Nakamura, 1992). S. Hepatocyte Stimulatory Substance

Hepatocyte stimulatory, substance (HSS) is another hepatocyte mitogen. It was originally observed in 1975 in extracts of weanling and regenerating rat liver, and subsequently in rabbit, mouse, dog, and pig livers (LaBrecque, 1994; LaBrecque et al., 1987). Purified preparations have so far not stimu-

16

Nancy L. R. Bucher

lated growth in any cell type other than hepatoma cells. However, because HSS has not yet become available in sufficient quantity and purity its cell type specificity has not been widely tested (Michalopoulos, 1990; Bucher and Strain, 1992; DuBois et al., 1994). 6. T r a n s f o r m i n g G r o w t h Factor f3

The TGFf3 family of growth factors comprises three isoforms, which are widespread and multifunctional. They can stimulate, or reversibly inhibit proliferation, tending to stimulate growth in mesenchymal cells and inhibiting it in epithelial cells. With regard to hepatocytes, TGFf31 is by far the most widely studied. TGFf31 is secreted in a latent form which is biologically inactive (DuBois et al., 1994), and the proteolytic activation of TGFf31 by plasmin is facilitated by binding of the latent complex of TGF[31 to the M6P/IGF2r (Dennis and Rifkin, 1991). During liver regeneration, expression of all three TGF[3 mRNA isoforms increases, especially TGF[31. The TGF[31 mRNA becomes detectable by 4 hr, increases further during the peak of DNA synthesis, rises steeply to a maximum at about 72 hr and then slowly declines to normal (Fausto and Mead, 1989). The mRNA is found only in the nonparenchymal cells. In contrast, the TGFIB types I, II, and III receptors are rapidly downregulated following a partial hepatectomy reaching a nadir at 24 hr post partial hepatectomy, the time of maximum DNA synthesis (Chari et al., 1995). Furthermore, the kinetics of type II receptor expression suggest it may be primarily responsible for ultimately limiting hepatocyte proliferation. Injection of TGFf31 at the time of partial hepatectomy does not affect the induction or course of DNA synthesis. In contrast, the injection of TGFf3 prior to the start of DNA synthesis substantially reduces the usual peak of DNA synthesis at 24 hr. The inhibition is transient, and continuation of treatment does not prevent the ultimate progression of the regeneration to completion. These results suggest that there is a TGF[31 sensitive restriction point during the late G 1 phase of the cell cycle (see Chapters 8 and 9) (Russell et al., 1988; DuBois et al., 1994). It has also been repeatedly shown that TGF[31 potently but reversibly inhibits DNA synthesis in growth factor stimulated hepatocytes in culture (Fausto, 1990; Michalopoulos, 1990; Bucher, 1991; Bucher and Strain, 1992; Matsumoto and Nakamura, 1992). Immunostaining of regenerating livers for TGF[31 protein is intensely positive in the periportal hepatocytes just before the start of DNA synthesis which also begins in these cells. The TGFf31 localization then progresses in a wave-like fashion toward the pericentral region of the liver lobule just before appearance of hepatocytes undergoing DNA synthesis (Jirtle et al., 1991). Thus, TGFIB appears to have a role in maintaining a ba!anced regenerative response, but a much fuller understanding of the underlying molecu-

1.


17

lar mechanisms is needed before the complexities of TGFf3 function can be interpreted into a coherent description of the regulation of liver regeneration. 7. Other Negative Growth Factors

These have not been extensively studied and are dealt with in recent reviews. They include activin, which is a peptide with homology to TGFI3, hepatocyte proliferation inhibitor (HPI) and platelet derived growth inhibitors cx and 13 (PDGIc~ and PDGI[3) (Fausto, 1990; Michalopoulos, 1990; Bucher, 1991; Bucher and Strain 1992; Matsumoto and Nakamura, 1992).

C. Cytokines 1. Interleukins

Recent studies implicate products of activated nonparenchymal cells in the regenerative process (i.e., cytokines). Some of these substances appear to act during the prereplicative period of hepatocyte regeneration. The relationship of cytokines to the regenerative response is provided by the indirect evidence that partial hepatectomy releases cytokines such as the interleukins (e.g., IL-1 and IL-6) and tumor necrosis factor c~ (TNFc~). Pretreatment of normal mice with lipopolysaccharide (an activator of cytokine production) 24 hr before partial hepatectomy augments DNA synthesis, whereas it is depressed in germ-free euthymic and athymic mice. These observations suggest that cytokines play an important role in stimulating liver regeneration (Cornell, 1990a,b). Moreover, serum IL-6 concentrations increase significantly after partial hepatectomy, and this increase is inhibited by antibodies to TNFa, a factor known to trigger IL-6 release (Akerman et al., 1992). Injection of IL-lc~ or IL-6 in combination with glucagon and ammonium chloride is reported to significantly increase the mitotic index in livers of normal adult rats. In contrast, control animals given only ammonium chloride and glucagon are unaffected (Koga and Ogasawara, 1991). IL-113, and to a lesser extent IL-6, inhibit the proliferation of hepatocytes in culture. Neither cytokine is as effective in this regard as TGF[31 (Michalopoulos, 1990; Bucher, 1991; Matsumoto and Nakamura, 1992). 2. Tumor Necrosis Factor c~

Administration of TNFe~ to normal rats has been reported to stimulate DNA synthesis. At the optimal dose of 25 I~g, DNA synthesis increases by about four-fold over the level in control animals; this is a very low DNA synthetic stimulus compared to partial hepatectomy where the increase is in

18

Nancy L. R. Bucher

the range of 50-fold (Feingold et al., 1988). It was subsequently reported, that the proliferative response is restricted to the nonparenchymal cells, especially the macrophages (Feingold et al., 1991). Meanwhile, intraperitoneal injection of polyclonal antibodies to TNFc~ 1 hr before partial hepatectomy significantly inhibits DNA synthesis in both the parenchymal and nonparenchymal cell populations. The conclusion is that TNFc~ positively modulates liver regeneration (Akerman et al., 1992). Thus, TNFc~ appears to be another of many possible effectors already proposed to function in the process of liver regeneration. D. Interactions

A finely balanced, integrated process such as liver regeneration must require a continuously shifting interplay of many growth effectors to meet the changing demands imposed by a varying array of bodily functions during the activation, progression, and subsidence of a major outburst of proliferative activity in a tissue that is normally in a state of growth arrest. Unraveling of the physiological and molecular biological complexities of this process started more than 100 years ago but still has a long way to go. Instances of hormonal, growth factor, and cytokine interactions have already been mentioned in relation to liver growth. These substances also interact in various ways to influence many metabolic functions of the liver, including amino acid, protein, lipid, carbohydrate, and probably other metabolic processes as well (Andrus et al., 1991). This is relevant, because the metabolic state of the cell can significantly affect its response to various growth effectors. Although hormones are generally available, nearly all of the growth factors and cytokines of interest, with a few very recent exceptions, have been obtainable only in small amounts. As a result, nearly all growth factor studies have been conducted in hepatocyte cultures. Hepatocytes, however, when freshly isolated for culture, are found to already express mRNAs of growth associated immediate-early genes (see Chapter 4). This indicates that cultured cells are primed before the growth factors are added, having been induced to undergo the G O to G 1 transition during the enzymatic preparative procedure (Etienne et al., 1988). Therefore, although in vitro studies may point to differences between effectors that potentiate mitogenesis and those that primarily modulate the process, in most instances their ability to initiate proliferative activity in unprimed (G O arrested) cells has been largely untested. In a recent report, however, EGF, TGF(x, and HGF have been infused directly into the portal vein of normal rats for 24 hr, resulting in remarkably small effects on DNA synthesis. If these growth factors are infused instead into rats previously subjected to a 33% hepatectomy, significant stimula-

1. Liver Regeneration Then and Now

19

tion results. Compared to 33% hepatectomized controls, EGF plus insulin causes DNA synthesis increases of up to four-fold, TGFc~ of up to eight-fold, and HGF of about five-fold. These results suggest that the 33% hepatectomy, which by itself causes little DNA synthesis, primes the normally quiescent hepatocytes, and potentiates the stimulation effects of the growth factors (Webber et al., 1994). This is at variance with an earlier study in which infusion of EGF plus insulin did stimulate hepatic DNA synthesis in normal rats. The discrepancy is probably due to priming induced by the stresses resulting from the less sophisticated infusion technique used in the earlier work (Bucher et al., 1978b). As regards combinations of growth factors, although HGF and TGFc~ augment each other in stimulating DNA synthesis in hepatocyte cultures, the combination is actually less stimulatory than either factor by itself in the in vivo 33 % hepatectomy model (Webber et al., 1994). No model is perfect, but despite the unquestionable value of hepatocyte cultures, it is occasionally worthwhile to emphasize their deficiencies (Bucher et al., 1990), and the enlightening perspectives to be gained from in vivo experiments.

VI. Conclusion After a long slow start, research on liver regeneration is gaining momentum at an accelerating rate that, to one who started long ago, is breathtaking. The framework, based on whole animal studies, is in place. Many of the parts that function within it are recognized and their potential roles are currently under intensive study at the molecular level. Although the complexities that are emerging may appear overwhelming, the technologies are at hand, and the pieces that fit into the puzzle are coming into sharper focus. The final step, assembly of the parts into the integrated whole that encompasses liver regeneration in vivo, is about to begin. Stay tuned!

References Akerman, R, Cote, R, Yang, S. Q., McClain, C., Nelson, S., Bagby, G. J., and Diehl, A. M. (1992). Antibodies to tumor necrosis factor inhibit liver regeneration after partial hepatectomy. Am. J. Physiol. 263, G579-585. Alison, M. R. (1986). Regulation of hepatic growth. Physiol. Rev. 66, 500-541. Andrus, T., Bauer, J., and Gerok, W. (1991). Effects of cytokines on the liver. Hepatology (Baltimore) 13, 364-375. Bach, R, Hamprecht, B., and Lynen, E (1969). Regulation of cholesterol biosynthesis in rat liver: Diurnal changes of activity and influence of bile acids. Arch. Biochem. Biophys. 133, 11-21. Berry, M. N., and Friend, D. S. (1969). High yield preparations of rat liver parenchymal cells. A biochemical and fine structural study. J. Cell Biol. 43, 506-520.

20

Nancy L. R. Bucher

Bresnick, E. (1971). Regenerating liver: An experimental model for the study of growth. Methods Cancer Res. 6, 347-397. Brues, A. M., Drury, D. R., and Brues, M. C. (1936). A quantitative study of cell growth in regenerating liver. Arch. Pathol. 22, 658-673. Brues, A. M., Tracy, M. M., and Cohn, W. E. (1944). Nucleic acids of rat liver and hepatoma: Their metabolism and turnover in relation to growth. J. Biol. Chem. 155, 619-633. Bucher, N. L. R. (1963). Regeneration of mammalian liver. Int. Rev. Cytol. 15, 245-300. Bucher, N. L. R. (1967a). Experimental aspects of hepatic regeneration. N. Engl. J. Med. 277, 686-696 and 738-746. Bucher, N. L. R. (1982). Thirty years of liver regeneration: A distillate. In "Cold Spring Harbor Conference on Cell Proliferation: Growth of Cells in Hormonally Defined Media" (G. H. Sato and R. Ross, eds.), Vol. 9, pp. 15-26. Cold Spring Harbor Laboratory, Cold Spring Harbor New York. Bucher, N. L. R. (1987). Regulation of liver growth: Historical perspectives and future directions. In "The Isolated Hepatocyte: Use in Toxicology and Xenobiotic Transformations" (E. J. Rauckman and G. M. Padilla, eds.), pp. 1-19. Academic Press, New York. Bucher, N. L. R. (1991). Liver regeneration: An overview. J. Gastroenterol. Hepatol. 6, 615624. Bucher, N. L. R., and Malt, R. A. (1971). "Regeneration of Liver and Kidney," pp. 1-176. Little Brown, Boston. Bucher, N. L. R., and Strain, A. J. (1992). Regulatory mechanisms in hepatic regeneration. In "Wright's Liver and Biliary Disease" (G. H. Millward-Sadler, R. Wright, and M. J. P. Arthur, eds.), 3rd Ed., pp. 258-274. Saunders, London. Bucher, N. L. R., and Swaffield, M. N. (1964). The rate of incorporation of labeled thymidine into the deoxyribonucleic acid of regenerating rat liver in relation to the amount of liver excised. Cancer Res. 24, 509-512. Bucher, N. L. R., and Swaffield, M. N. (1973). Regeneration of liver in rats in the absence of portal splanchnic organs and a portal blood supply. Cancer Res. 33, 3189-3194. Bucher, N. L. R., and Swaffield, M. N. (1975). Regulation of hepatic regeneration in rats by synergistic action of insulin and glucagon. Proc. Natl. Acad. Sci. U.S.A. 72, 11571160. Bucher, N. L. R., Scott, J. E, and Aub, J. C. (1951). Regeneration of the liver in parabiotic rats. Cancer Res. 11,457-465. Bucher, N. L. R., Schrock, T. R., ,and Moolten, E L. (1969). An experimental view of hepatic regeneration. Johns Hopkins Med. J. 125, 150-257. Bucher, N. L. R., McGowan, J. A., and Patel, U. (1978a). Hormonal regulation of liver growth. "ICN/UCLA Symposium: Molecular Cell Biology," Vol. 12, pp. 661-670. Bucher, N. L. R., Patel, U., and Cohen, S. (1978b). Hormonal factors concerned with liver regeneration. "Hepatotrophic Factors" Ciba Foundation Symposium No. 55, pp. 95-107. Elsevier, New York. Bucher, N. L. R., Robinson, G. S., and Farmer, S. R. (1990). Effects of extracellular matrix on hepatocyte growth and gene expression: Implications for hepatic regeneration and the repair of liver injury. Semin. Liver Dis. 10, 11-19. Carpenter, G., and Cohen, S. (1992). Epidermal growth factor. J. Biol. Chem. 265, 7709-7712. Carpenter, G., and Wahl, M. I. (1990). The epidermal growth factor family. In "Handbook of Experimental Pharmacology" (M. B. Sporn and A. B. Roberts, eds.), Vol. 1, pp. 69-171. Springer-Verlag, Berlin. Carter, E. A., Kirkham, S. E., Tompkins, R. G., and Burke, J. E (1989). Inhibition of in vivo DNA synthesis in regenerating rat liver following thermal injury. Biochem. Biophys. Res. Commun. 160, 196-201. Caruana, J. A., Whalen, D. A., Anthony, W. P., Sunby, C. R., and Ciechoski, M. P. (1986).

1.


21

Paradoxical effects of glucose feeding on liver regeneration and survival after partial hepatectomy. Endocr. Res. 12, 147-156. Chari, R. S., Price, D. T., Sue, S. R., Meyers, W. C., and Jirtle, R. L. (1995). Down-regulation of transforming growth factor 13type I, II, and III during liver regeneration. Am. J. Surg. 169, 126-132. Cornell, R. P. (1990a). Depressed liver regeneration after partial hepatectomy of germ-flee athymic and lipopolysaccharide-resistant mice. Hepatology (Baltimore) 11, 916-922. Cornell, R. P. (1990b). Acute phase responses after acute liver injury by partial hepatectomy in rats as indicators of cytokine release. Hepatology (Baltimore) 11, 923-931. Dennis, P. A., and Rifkin, D. B. (1991). Cellular activation of latent transforming growth factor 13 requires binding to the cation-independent mannose 6-phosphate/insulin-like growth factor type II receptor. Proc. Natl. Acad. Sci. U.S.A. 88, 580-584. DuBois, R. N., Hunter, E. B., and Russell, W. E. (1994). Molecular aspects of hepatic regeneration. In "Molecular Basis of Medicine" (C. V. Dang and A. M. Feldman, eds.), in press. Edwards, P. A., Muroya, N., and Gould, R. G. (1972). In vivo demonstration of the circadian rhythm of cholesterol biosynthesis in the liver and intestine of the rat. J. Lipid Res. 13,396401. Ekberg, S., Luther, M., Nakamura, T., and Jansson (1992). Growth hormone promotes early initiation of hepatocyte growth factor gene expression on the liver of hypophysectomized rats after partial hepatectomy. J. Endocrinol. 35, 59-67. Etienne, P. L., Baffet, G., Desvergne, B., Boisnard-Rissel, M., Glaise, D., and GuguenGuillouzo, C. (1988). Transient expression of c-los and constant expression of c-myc in freshly isolated and cultured normal adult rat hepatocytes. Oncog. Res. 3, 255-262. Fausto, N. (1990). Hepatic regeneration. In "Hepatology" (D. Zakim and T. D. Boyer, eds.), 2nd Ed., pp. 49-95. Saunders, London. Fausto, N. (1991). Growth factors in liver development, regeneration and carcinogenesis. Prog. Growth Factor Res. 3, 219-234. Fausto, N., and Mead, M. E. (1989). Regulation of liver growth: Protooncogenes and transforming growth factors. Lab. Invest. 60, 4-13. Feingold, K. R., Soued, M., and Grunfeld, C. (1988). Tumor necrosis factor stimulates DNA synthesis in the liver of intact rats. Biochem. Biophys. Res Commun. 153, 576-582. Feingold, K. R., Barker, M., Jones, A. L., and Grunfeld, C. (1991). Localization of tumor necrosis factor-stimulated DNA synthesis in the liver. Hepatology (Baltimore) 13,773-779. Fishback, E C. (1929). A morphologic study of regeneration of the liver after partial removal. AMA Arch. Pathol. 7, 955-977. Fleig, W. E. (1988). Liver-specific growth factors. Scand. J. Gastroenterol. Suppl. 23, 31-36. Gaub, J., and Iversen, J. (1984). Rat liver regeneration after 90% partial hepatectomy. Hepatology (Baltimore) 4, 902-904. Ghahary, A., Minuk, G. Y., Luo, J., Gauthier, T., and Murphy, L. M. (1992). Effects of partial hepatectomy on hepatic insulin-like growth factor binding protein-1 expression. Hepatology (Baltimore) 15, 1125-1131. Ghoda, E., Tsubouchi, H., Nakayama, H., Hirono, S., Sakiyama, O., Takahashi, K., Miyazaki, H., Hashimoto, S., and Daikuhara, Y. (1988). Purification and partial characterization of hepatocyte growth factor from plasma of a patient with fulminant hepatic failure. J. Clin. Invest. 81, 414-419. Grisham, J. W. (1962). A morphologic study of deoxyribonucleic acid synthesis and cell proliferation in regenerating rat liver; autoradiography with thymidine-3H. Cancer Res. 22,842-849. Grisham, J. W., Leong, G. E, and Hole, B. V. (1964). Heterotopic partial autotransplantation of rat liver. Technique and demonstration of structure and function of the graft. Cancer Res. 24, 1474-1482. Gumucio, J. J., and Berkowitz, C. M. (1992). Structural organization of the liver and function

22

Nancy L. R. Bucher

of the hepatic acinus. In "Liver and Biliary Diseases" (N. Kaplowitz, ed.), 3rd Ed., pp. 217. Williams & Wilkins, Baltimore, Maryland. Gumucio, J. J., and Chianale, J. (1988). Liver cell heterogeneity and liver function. In "The Liver: Biology, and Pathobiology" (I. M. Arias, W. B. Jakoby, H. Popper, D. Schachter, and D. A. Shafritz, eds.), 2nd Ed., pp. 931-947. Raven, New York. Harkness, R. D. (1957). Regeneration of liver. Br. Med. Bull. 13, 87-93. Hasegawa, K., and Koga, M. (1977). Induction of liver cell proliferation in intact rats by amines and glucagon. Life Sci. 21, 1723-1728. Hasegawa, K., Sidhartha, K., and Carr, B. I. (1994). Stimulation of hepatocyte DNA synthesis by neurotensin. J. Cell. Physiol. 159, 215-222. Hays, D. M. (1974). Surgical aspects of hepatic regeneration. Surg. Gynecol. Obstet. 139, 609-619. Higashiyama, S., Abraham, J. A., Miller, J., Fiddes, J. C., and Klagsbrun, M. (1991). A heparinbinding growth factor secreted by macrophage-like cells that is related to EGE Science 251, 936-939. Higashiyama, S., Abraham, J. A., and Klagsbrun, M. (1993). Heparin-binding EGF-like growth factor stimulation of smooth muscle cell migration: Dependence on interactions with cell surface heparan sulfate. J. Cell Biol. 122, 933-940. Higgins, G. M., and Anderson, R. M. (1931). Experimental pathology of the liver. I. Restoration of the liver of the white rat following partial surgical removal. Arch. Pathol. 12, 186202. Holecek, M., Simek, J., Palicka, V., and Zadak, Z. (1991). Effect of glucose and branched chain amino acid (BCAA) infusion on onset of liver regeneration and plasma amino acid pattern in partially hepatectomized rats. J. Hepatol. 13, 14-20. Hopkins, H. A., Bonney, R. J., Walker, P. R., Yager, J. D., and Potter, V. R. (1973a). Food and light as separate entrainment signals for rat liver enzymes. Adv. Enzyme Regul. 11, 169191. Hopkins, H. A., Campbell, H. A., Barbiroli, B., and Potter, V. R. (1973b). Thymidine kinase and deoxyribonucleic acid metabolism in growing and regenerating livers from rats on controlled feeding schedules. Biochem. J. 136, 955-966. Horikawa, S., Sakata, K., Hatanaka, M., and Tsukada, K. (1986). Expression of c-myc oncogene in rat liver by dietary manipulation. Biochem. Biophys. Res. Commun. 140, 574580. Ingle, D. J., and Baker, B. L. (1957). Histology and regenerative capacity of liver following multiple partial hepatectomies. Proc. Soc. Exp. Biol. Med. 95, 813-815. Ito, N., Kawata, S., Tamura, S., Kiso, S., Tsushima, H., Damm, D., Abraham, J. A., Higashiyama, S., Taniguchi, N., and Matsuzawa, Y. (1994). Heparin-binding EGF-like growth factor is a potent mitogen for rat hepatocytes. Biochem. Biophys. Res. Commun. 198, 25-31. Jirtle, R. L., and Michalopoulos, G. (1982). Effects of partial hepatectomy on transplanted hepatocytes. Cancer Res. 42, 3000-3004. Jirtle, R. L., Cart, B. I., and Scott, C. D. (1991). Modulation of insulin-like growth factor 2/mannose 6-phosphate receptors and transforming growth factor-J31 during liver regeneration. J. Biol. Chem. 266, 22444-22450. Kan, M., Huang, J., Mansson, P.-E., Yasumitsu, H., Carr, B., and McKeehan, W. L. (1989). Heparin-binding growth factor type I (acidic fibroblast growth factor): A potential biphasic autocrine and paracrine regulator of heplatocyte regeneration. Proc. Natl. Acad. Sci. U.S.A. 86, 7432-7436. Klinge, O., and Mathyl, J. (1969). Tagezeitliche mitose-rhythmen in der rattenleber nach modifizierter teilhepatektomierten rattenleber. Klin. Wochenschr. 46, 395-396. Koga, M., and Ogasawara, H. (1991). Induction of hepatocyte mitosis in intact adult rat by interleukin-le~ and interluekin-6. Life Sci. 49, 1263-1270.

1.


23

LaBrecque, D. (1994). Liver regeneration: A picture emerges from the puzzle. Am. J. Gastroenterol. 89(Suppl. 8), $86-$96. LaBrecque, D. R., Steele, G., Fogerty, S., Wilson, M., and Barton, J. (1987). Purification and physical-chemical characterization of hepatic stimulator substance. Hepatology (Baltimore) 7, 100-106. Leduc, E. H. (1949). Mitotic activity in the liver of the mouse during inanition followed by refeeding with different levels of protein. Am. J. Anat. 84, 397-430. Leduc, E. H. (1964). Regeneration of the liver. In "The Liver. Morphology, Biochemistry and Physiology" (C. Rouiller, ed.), Vol. 2, pp. 63-89. Academic Press, New York. Lee, J., Greenbaum, L., Haber, B. A., Nagle, D., Lee, V., Niles, V., Mohn, L. L., Bucan, M., and Taub, R. (1994). Structure and localization of the IGFBP-1 gene and its expression during liver regeneration. Hepatology (Baltimore) 19, 656-665. Leffert, H. L., Koch, L. S., Lad, P. J., Shapiro, I. P., Skelly, H., and deHemptinne, B. (1988). Hepatocyte regeneration, replication and differentiation. In "The Liver: Biology and Pathobiology" (I. M. Arias, W. B. Jakoby, H. Popper, D. Schachter, and D. A. Shafritz, eds.), 2nd Ed., pp. 833-850. Raven, New York. Lewan, L., Ynger, T., and Engelbrecht (1977). The biochemistry of the regenerating liver. Int. J. Biochem. 8, 477-487. Mann, E C. (1944). Restoration and pathologic reactions of the liver. J. Mr. Sinai Hosp. (N. Y.) 11, 65-74. Masquilier, D., Foulkes, N. S., Mattei, M.-G., and Sassone-Corsi, P. (1993). Human CREM gene: Evolutionary conservation, chromosomal localization, and inducibility of the transcript. Cell Growth Differ. 4, 931-937. Massagu6, J. (1990). Transforming growth factor cx. J. Biol. Chem. 265, 21393-21396. Matsumoto, K., and Nakamura, T. (1991). Hepatocyte growth factor: Molecular structure and implications for a central role in liver regeneration. J. Gastroenterol. Hepatol. 6, 509519. Matsumoto, K., and Nakamura, T. (1992). Hepatocyte growth factor: Molecular structure, roles in liver regeneration, and other biological functions. Crit. Rev. Oncogen. 3, 27-54. Mead, J. E., Braun, D. A., Martin, D. A., and Fausto, N. (1990). Induction of replicative competence ("priming") in normal liver. Cancer Res. 50, 7023-7030. Michalopoulos, G. K. (1990). Liver regeneration: Molecular mechanisms of growth control. FASEB J. 4, 176-187. Michalopoulos, G. K., Cianciulli, H. D., Novotny, A. R., Kligerman, A. D., Strom, S. C., and Jirtle, R. L. (1982). Liver regeneration studies with rat hepatocytes in primary cultures. Cancer Res. 42, 4673-4682. Mohn, K. L., Laz, T. M., Hsu, J.-C., Melby, A. E., Bravo, R., and Taub, R. (1991a). The immediate-early growth response in regenerating liver and insulin-stimulated H-35 cells: Comparison to serum-stimulated 3T3 cells and identification of 41 novel immediate-early genes. Mol. Cell. Biol. 11, 381-390. Mohn, K. L., Melby, A. E., Teware, D. E., La, T. M., and Taub, R. (1991b). The gene encoding rat insulin like growth factor binding protein-1 is rapidly and highly induced in regenerating liver. Mol. Cell. Biol. 11, 1393-1401. Moolten, E L., and Bucher, N. L. R. (1967). Regeneration of rat liver: Transfer of "humoral" agent by cross circulation. Science 158, 272-274. Moolten, E L., Oakman, N. J., and Bucher, N. L. R. (1970). Accelerated response of hepatic DNA synthesis to partial hepatectomy in rats pre-treated with growth hormone or surgical stress. Cancer Res. 30, 2353-2357. Mueller, C. R., Maire, P. N., and Schibler, U. (1990). DBP, a liver-enriched transcriptional activator, is expressed late in ontogeny and its tissue specificity is determined post transcriptionally. Cell (Cambridge, Mass.) 61,279-291. Mullhaupt, E., Fodor, A. Feren, A., and Jones, A. (1993). The steady-state level of hepatocyte

24

Nancy L. R. Bucher

epidermal growth factor RNA rapidly increases in the prereplicative phase of liver regeneration. Hepatology (Baltimore) 18, 148A. Nakamura, T., Nawa, K., and Ichihara, A. (1984). Partial purification and characterization of hepatocyte growth factor from serum of hepatectomized rats. Biochem. Biophys. Res. Commun. 122, 1450-1459. Nakamura, T., Nawa, K., Ichihara, A., Kaise, N., and Nishino, T. (1987). Purification and subunit structure of hepatocyte growth factor from rat platelets. FEBS Lett. 224, 311-316. Nakamura, T., Nishizawa, T., Hagiya, M., Seki, T., Shimonishi, M., Sugimura, A., Tashiro, K., and Shimizu, S. (1989). Molecular cloning and expression of human hepatocyte growth factor. Nature (London) 342, 440-443. Naldini, L., Weidner, K. M., Vigna, E., Gaudino, G., Bardelli, A., Ponzetto, C., Narsimhan, R. P., Hartmann, G., Zarnegar, R., Michalopoulos, G. K., Birchmeier, W., and Comoglio, P. M. (1991). Scatter factor and hepatocyte growth factor are indistinguishable ligands for the MET receptor. EMBO J. 10, 2867-2878. Ni, N., and Yager, J. D. (1994). Co-mitogenic effects of estrogens on DNA synthesis induced by various growth factors in cultured female rat hepatocytes. Hepatology (Baltimore) 19, 183-192. Noguchi, S., Ohba, Y., and Oka, T. (1991). Influence of epidermal growth factor on liver regeneration after partial hepatectomy in mice. J. Endocrinol. 128, 425-431. Ohmori, H., Murakami, A. E, Higashi, K., Hirano, H., Gotoh, S., Kuroiwa, A., Masui, A., Nakamura, T., and Amalric, E (1990). Simultaneous activation of heat shock protein (hsp 70) and nucleolin genes during in vivo and in vitro prereplicative stages of rat hepatocytes. Exp. Cell Res. 189, 227-232. Russell, W. E., and Bucher, N. L. R. (1983). Vasopressin modulates liver regeneration in the Brattleboro rat. Am. J. Physiol. 245, G321-G324. Russell, W. E., McGowan, J. A., and Bucher, N. L. R. (1984a). Partial characterization of a hepatocyte growth factor from rat platelets. J. Cell. Physiol. 119, 183-192. Russell, W. E., McGowan, J. A., and Bucher, N. L. R. (1984b). Biological properties of a hepatocyte growth factor from rat platelets. J. Cell. Physiol. 119, 193-197. Russell, W. E., Coffey, R. J., Ouellette,, A. J., and Moses, H. L. (1988). Type beta transforming growth factor reversibly inhibits the early proliferative response to partial hepatectomy in the rat. Proc. Natl. Acad. Sci. U.S.A. 85, 5126-5130. Russell, W. E., Dempsey, P. J., Sitaric, S., Peck, A. E, and Coffey, R. G. (1993). Transforming growth factor-e~ (TGF(x) concentrations increase in regenerating rat liver: Evidence for a delayed accumulation of mature TGFe~. Endocrinology (Baltimore) 133, 1731-1738. Schlesinger, M. J. (1990). Heat shock proteins. J. Biol. Chem. 265, 12111-12114. Scott, C. D., Ballesteros, M., and Baxter, R. C. (1990). Increased expression of insulin-like growth factor-II/mannose-6-phosphate receptor in regenerating rat liver. Endocrinology (Baltimore) 127, 2210-2216. Selden, A. C., and Hodgson, H. J. E (1991). Growth factors and the liver. Gut 32, 601-603. Simpson, G. E. C., and Finkh, E. S. (1963). Pattern of regeneration of rat liver after repeated partial hepatectomies. J. Pathol. Bacteriol. 86, 361-370. Sporn, M. B., and Roberts, A. B. (1990). TGFf3: Problems and prospects. Cell Regul. 1, 875882. St. Hilaire, R. J., and Jones, A. L. (1982). Epidermal growth factor: Its biological and metabolic effects with emphasis on the hepatocyte. Hepatology (Baltimore) 2, 601-613. Starzl, T. E., and Terblanche, J. (1979). Hepatotrophic substances. Prog. Liver Dis. 6, 135151. Stoker, M., Gherardi, E., Perryman, M., and Gray, J. (1987). Scatter factor is a fibroblastderived modulator of epithelial cell mobility. Nature (London) 327, 239-242. Strain, A. J., McGowan, J. A., and Bucher, N. L. R. (1982). Stimulation of DNA synthesis in

1.


25

primary cultures of adult rat hepatocytes by rat platelet-associated substance(s). In Vitro 18, 106-116. Strom, S. C., Jirtle, R. L., Jones, R. S., Rosenberg, M. R., and Michalopoulos, G. (1982). Isolation, culture and transplantation of human hepatocytes. J. Natl. Cancer Inst. 68, 771778. Tamura, J., Tanaka, J., Fujita, K.-I., Yoshida, M., Kasamatsu, T., Arii, S., and Tobe, T. (1992). Cell kinetics of regenerating liver after 70% hepatectomy in rats. 2-color flow cytometric analysis. HPB Surg. 5, 103-114. Webber, E. M., FiztGerald, M. J., Brown, P. I., Bartlett, M. H., and Fausto, N. (1993). Transforming growth factoroc~ expression during liver regeneration after partial hepatectomy and toxic injury, and potential interactions between transforming growth factor-e~ and hepatocyte growth factor. Hepatology (Baltimore) 18, 1422-1431. Webber, E. M., Godowski, P. J., and Fausto, N. (1994). In vivo response of hepatocytes to growth factors requires an initial priming stimulus. Hepatology (Baltimore) 19, 489-497. Weinbren, K. (1959). Regeneration of the liver. Gastroenterology 37, 657-668. Wilson, J. W., and Leduc, E. H. (1948). The occurrence and formation of binucleate and multinucleate cells and polyploid nuclei in the mouse liver. Am. J. Anat. 82, 353-392. Yager, J. D., Zurlo, J., Sewall, C. H., Lucier, G. W., and He, H. (1994). Growth stimulation followed by growth inhibition in livers of female rats treated with ethinyl estradiol. Carcinogenesis 15, 2117-2124. Zarnegar, R., and Michalopoulos, G. K. (1989). Purification and biological characterization of human hepatopoietin A, a polypeptide growth factor from serum of hepatectomized rats. Biochem. Biophys Res. Commun. 122, 1450-1459.


2 Hepatocyte Growth Factor (HGF) and Its Receptor (Met) in9 Liver Rege neratlon, " Neoplasia, and Disease George K. Michalopoulos Department of Pathology University of Pittsburgh Medical Center Pittsburgh, Pennsylvania15261

I. Introduction Hepatocyte growth factor (HGF) is a multifunctional cytokine that has pleiotropic effects on several cells and tissues. It was originally identified and isolated based on its capacity to stimulate mitogenesis in cultured hepatocytes (Michalopoulos et al., 1982, 1984; Nakamura et al., 1984). Isolation of HGF was followed shortly by a description of its structure. This was accomplished by deriving the amino acid sequence of the molecule directly (Zarnegar et al., 1989) as well as by cloning and sequencing the HGF gene (Nakamura et al., 1989; Miyazawa et al., 1989). It was subsequently shown that the protein encoded by the proto-oncogene c - m e t was the receptor for HGF (Naldini et al., 1991a; Bottaro et al., 1991). These fundamental discoveries opened the door for several studies that have now thoroughly documented that HGF and its receptor are important in the function of most organs and tissues. Given this realization, it is relevant to consider the question of whether there is any special relationship between HGF and the liver. Are there any hepatic functions that are regulated by HGF and is the mode of action of HGF as a regulator of these functions unique and specific to the liver? This chapter emphasizes the aspects of the relationship Liver Regeneration and Carcinogenesis Copyright 9 1995 by Academic Press, Inc. All rights of reproduction in any form reserved.

27

28

George K. Michalopoulos

between HGF and the liver that play a regulatory role in liver growth and function. There have been several comprehensive reviews that cover the current knowledge of the structural and functional characteristics of HGF and its receptor (see Chapter 3) (Michalopoulos and Zarnegar, 1992; Gherrardi et al., 1993; Matsumoto and Nakamura, 1992; Ponzetto et al., 1994; Weidner et al., 1993). Thus, these topics will only be covered to the extent that they are essential to our understanding of the biological effects of HGE

II. Structural and Functional Aspects of HGF and the HGF Receptor HGF is a protein of 100-kDa molecular weight (Figure 1). It is synthesized as a precursor molecule composed of a single polypeptide chain and activated by proteolytic cleavage (Nakamura et al., 1989). The activation is carried out by urokinase (uPA) (Mars et al., 1993) as well as by HGF activator, a recently identified molecule that has substantial sequence homology to coagulation Factor XII (Miyazawa et al., 1993). Activated HGF is a heterodimer consisting of two chains. The heavy (alpha) chain is composed of a hairpin loop domain followed by four kringle domains. The light (beta) chain has several amino acid sequences that are identical to the consensus sequences for serine proteases (Nakamura et al., 1989). Although the consensus sequences are present, the key amino acids which define the catalytic site for serum proteases are mutated. The final result is a pseudo protease structure that has no documented functional protease activity. The structure of the heterodimeric molecule has been characterized by sitedirected mutagenesis (Lokker et al., 1992), and by studying the properties and function of variant HGF molecules that are naturally produced by alternate splicing of the HGF mRNA (Chan et al., 1991). These studies have have shown that the hairpin loop domain and the second kringle are important for binding HGF to its receptor. HGF is a potent mitogen for many epithelial cells. The list currently includes hepatocytes, keratinocytes, urinary bladder epithelial cells, mammary epithelial cells, bronchial epithelial cells, bile duct cells, and endothelial cells. Given the experience with other growth factors, it is likely that HGF will prove to be a widespread mitogen for all epithelial cells of ectodermal and endodermal origin. In addition to its mitogenic properties, HGF is also a cell motogen. It was independently identified as a scatter factor that increases the motility of several normal and neoplastic cells (Stoker et al., 1987), and the identity between scatter factor iSF) and HGF has now been unequivocally confirmed (Naldini et al., 1991b). Although the effects of HGF on cell motility have been investigated extensively, little is still known

2.

Hepatocyte Growth Factor (HGF) and Its Receptor (Met)

29

Figure 1 Diagrammatic structure of hepatocyte growth factor (HGF). The two chains are shown held together by disulfide bonds (S-S). The amino terminal of the alpha chain contains a hairpin loop structure and the four kringle structures (K1, K2, K3, and K4). The hairpin loop and kringle 2 are the most important determinants for the binding of HGF to its receptor. The beta chain has the structure of a pseudo protease. The numbered amino acids in the beta chain are surrounded by consensus sequences characteristic of serine proteases. These amino acids have replaced the three amino acids normally seen within the sequences in true proteases. The Arg494-Va1495 site contains the peptide bond cleaved by HGF activating molecules (urokinase or factor XII homologous HGF activator), thus converting HGF from an inactive (single chain) form to the two chain active heterodimeric form composed of an alpha and beta chain.

a b o u t the signal t r a n s d u c t i o n p a t h w a y s that mediate this aspects of H G F function. The c o m b i n a t i o n of mitogenic and m o t o g e n i c effects p r o b a b l y underlies the recently d e m o n s t r a t e d m o r p h o g e n i c effects of H G F on different cell types. Studies of endothelial cells have s h o w n that H G F functions in angiogenesis (Bussolino et al., 1992; Rosen et al., 1993b). H G F also stimulates kidney t u b u l a r epithelial cells ( M D C K cells) to f o r m tubules in a collagen substrate (Pepper et al., 1992), and stimulates hepatocytes to f o r m plates in type I collagen gels ( M i c h a l o p o u l o s et al., 1993). In m o r e recent studies, H G F has been s h o w n to induce the f o r m a t i o n of ductular structures in

30


cultures of hepatocytes maintained in a medium that allows for long-term hepatocyte proliferation (Michalopoulos, 1992). The HGF receptor (HGFr) is the protein encoded for by the protooncogene c - m e t (see Chapter 3) (Naldini et al., 1991b; Bottaro et al., 1991). The Met protein is synthesized in a precursor form that becomes activated by cleavage almost immediately after its synthesis (Zhen et al., 1994). The cleavage results in a heterodimeric molecule composed of both a heavy chain and a light chain attached to one another by disulfide bonds. The heavy (beta) chain contains an extracellular domain, a transmembrane domain, and an intracellular domain, and the light (alpha) chain is attached to the extracellular domain. The mode of action of the HGFr has been thoroughly described in recent papers and reviews by Comoglio and colleagues (see Chapter 3) (Ponzetto et al., 1994). Upon binding to its ligand, HGF, and HGFr dimerizes and cross phosphorylates its tyrosine kinase sites. The activated tyrosine kinase sites then become the docking site for signal transduction proteins that are further phosphorylated at tyrosine residues and migrate to other regions of the cell to transmit the stimulatory signals of the activated HGFr (Ponzetto et al., 1994). Although the HGFr belongs to the overall family of tyrosine kinase receptors, it has unique peculiarities in the structure of its activated region. Phosphorylation of two tyrosine residues results in the formation of a single docking site on which many proteins with src homology (i.e., SH2 and SH3) recognition sites bind (see Chapter 3) (Ponzetto et al., 1994). This contrasts with other members of the tyrosine kinase family which have two docking sites in their active configuration. The receptors encoded by the genes c-sea and c - r o n have substantial sequence homology with c - m e t and a similar structure of the activate tyrosine kinase sites (Gaudino et al., 1994).

III. H G F L o c a l i z a t i o n A. In Liver Several studies have shown that both HGF mRNA and HGF protein are produced by the nonparenchymal cells in the liver. The identity of the HGF producing cells was initially disputed; however, it has now been clearly demonstrated that the liver cells responsible for the production of HGF are the cells of Ito (Noji et al., 1990; Schirmacher et al., 1992). These cells contain lipid droplets that are rich in vitamin A. They also produce other growth factors including acidic fibroblast growth factor (aFGF), transforming growth factor 131 (TGFI31), etc. In addition, these cells are responsible for producing much of the scant connective tissue present around hepatocytes. Ito cells change into myofibroblasts in culture as well as in cirrhosis,

2.


31

and when they become myofibroblasts they lose the capacity to store vitamin A (Schirmacher et al., 1992). Since Ito cells are responsible for the production of HGF as well as matrix production in the liver, it is not surprising that HGF is heavily deposited in the hepatic biomatrix. When isolated liver is perfused with high concentrations of NaC1, large amounts of HGF are extracted (Masumoto and Yamamoto, 1991). The localization of growth factors in the biomatrix is a well-documented phenomenon and is partly related to the heparin affinity binding sites that most growth factors possess. Typically, these binding sites are characterized by high capacity and low affinity. In addition to the binding of HGF to high capacity low affinity sites [e.g., heparin sulfated proteoglycans, (HSPG)], HGF also binds to specific matrix sites that have relatively high affinity (Zarnegar et al., 1990). When HGF is injected into the liver there is a discrepancy between the lobular distribution of HGF and the distribution of the high affinity binding receptor, HGFr. The HGFr is uniformly distributed throughout the hepatic lobule (Liu et al., 1994a) whereas, HGF bound to high affinity sites is predominantly localized in the periportal region of the liver lobules; eidermal growth factor (EGF) binding has also been shown to have the same lobular pattern of localization (St. Hilaire et al., 1983). This suggests that the bulk of HGF attached to high affinity sites is not bound to the HGFr. Previous studies have shown that there are two high affinity binding sites for HGF in the liver (Zarnegar et al., 1990). The highest affinity binding site is the HGFr (Met protein). The other lesser, but still relatively high affinity binding site, has not been clearly characterized although recent studies have shown that HGF binds with relatively high affinity to sulfoglycolipids compounds such as galactosylceramide sulfate (SM4), lactosylceramide sulfate (SM3), and gangliotriaosylceramide bis-sulfate (Kobayashi et al., 1994). The precise role that these relatively high affinity nonsignaling receptor binding sites play in the function of HGF is not yet clear. However, the localization of the bulk of the HGF in the periportal region of the liver lobule along with an identical localization for EGF correlates with hepatocyte proliferation starting in the periportal region of the liver lobule following a partial hepatectomy (Rabes et al., 1976). The presence of the two most powerful hepatic mitogens at high concentrations at the lobular location where liver regeneration begins suggests a cause and effect relationship between the presence of these growth factors and hepatocyte recruitment into the cell cycle during liver regeneration. B. In Extra Hepatic Tissues

As previously stated, HGF is produced primarily by mesenchymal cell types. These cells include the Ito cells in the liver and the myofibroblasts in other


32

organs. These are cells that often surround blood vessels, have contractile properties, and produce the connective tissue that provides the structural framework of different organs. Immunohistochemical studies, however, show that HGF is also present in most types of epithelial cells (Wolf et al., 1991). The epithelial distribution of HGF, however, closely corresponds to that for the HGFr which is also present in most epithelial cells (Giordano et al., 1989) suggesting that the presence of HGF in epithelial cells reflects receptor uptake rather than cellular production. Both HGF and HGFr are also present in neurons of the brain cortex and hippocampus (Jung et al., 1994; Schirmacher, 1994). In contrast, endothelial cells express the HGFr but contain no detectable HGF (Rosen et al., 1993a). IV. L i v e r

and the Processing of HGF

When radiolabeled HGF is injected into the systemic circulation, it can be traced to several organs (Appasamy et al., 1993; Zioncheck et al., 1994). Approximately 40 to 50% of the injected radioactivity becomes bound in high capacity low affinity sites in the skin, muscle, and bone; most of the remaining HGF is found in the liver. Approximately 30 to 40% of the injected HGF is localized in the liver within 10 min after injection; studies have shown that similar temporal and spatial relationships exist for other growth factors such as EGF and TGF~I. It is not clear at this point, however, whether the uptake of HGF by the liver proceeds through a lysosomal dependent pathway. An unexpected finding was that a portion of the injected HGF becomes secreted intact in the bile (Liu et al., 1994a). Furthermore, approximately 20% of the HGF present in the bile is intact. The percentage of intact HGF increases after treatment with leupeptin, but it is not affected by the administration of chloroquine (Liu and Michalopoulos, 1993). Chloroquine inhibits lysosomal function by raising the intra lysosomal pH. Therefore, the combined evidence suggests that most of the processing of HGF within the hepatocyte proceeds through nonlysosomal pathways. The mechanisms by which intact HGF is secreted into the bile are not clear at this time. In general, liver has a high capacity for the uptake of HGE When increasing amounts of HGF are injected through the portal circulation, a large amount (0.157 + 0.012 ~g/g liver) is taken up by the liver before its capacity to bind HGF becomes saturated (Liu iet al., 1994a). Similar studies have shown that following partial hepatectomy, the capacity of regenerating liver to take up HGF increases more than two-fold. Regenerating liver can still secrete intact HGF in the bile; however, shortly after partial hepatectomy the secreted forms of HGF appear to be processed by alternate pathways (Liu et al., 1994a).

2. Hepatocyte Growth Factor (HGF) and Its Receptor (Met)

33

A. HGF and Liver Regeneration The relationship between HGF and liver regeneration has been at the center of the investigation of the functions of HGF from the beginning of its discovery and it originally drove the process that led to its isolation. Several past studies, repeated by multiple laboratories and investigators, have demonstrated that during liver regeneration there appears in the blood factor(s) that transmit a mitogenic signal to hepatocytes present anywhere in the body (see Chapter 1). Fragments of hepatic tissue engrafted into extra hepatic sites responded with DNA synthesis and growth when the in situ liver is subjected to a partial hepatectomy (Leong et al., 1964). When rat pairs are joined together in parabiosis, a partial hepatectomy of the liver in one member leads to liver regeneration in situ as well as in the intact liver of the unoperated member of the parabiotic pair (Moolten and Bucher, 1967). The maximal regenerative stimulation of the unoperated liver is seen when the liver of other member of the pair is removed entirely (Fisher et al., 1971). Partial hepatectomy also leads to DNA synthesis in isolated hepatocytes engrafted into the adipose tissue (Jirtle and Michalopoulos, 1982). These and other studies provided abundant evidence for the existence of circulating hepatotrophic factors, and led to the studies that eventually concluded with the isolation of HGF and the identification of its structure and function. In the last decade, liver regeneration has been shown to proceed very rapidly, with specific changes in hepatocytes demonstrable immediately after partial hepatectomy. Hyperpolarization of the hepatic plasma membrane occurs within 10 min after partial hepatectomy (de Hemptinne et al., 1985), and an enhanced expression of multiple new mRNAs occurs within the first 30 min after partial hepatectomy (see Chapter 4) (Mohn et al., 1990). The cluster of specific genes expressed immediately after partial hepatectomy is referred to as "immediate early genes" and their nature and control mechanisms are currently being investigated. Intense glycogenolysis also occurs and depletes the liver of glycogen within an hour after partial hepatectomy. The clonogenicity of transplated hepatocytes is dependent on the time between when the recipient animals are partially hepatectomized and when the hepatocytes are injected, with maximum clonogenicity occurring approximately 1 hr after partial hepatectomy (Jirtle and Michalopoulos, 1982). All of these phenomena suggest that the response to a partial hepatectomy is rapid and that it defines a course of events that commits the hepatocytes to proliferate. In view of the above findings, any changes related to HGF following partial hepatectomy should be viewed in terms of whether they are sufficient and capable of explaining the above biological phenomena. As previously mentioned, HGF was originally isolated from both human and rat plasma.

34


Thus, it definitely exists as an identifiable growth factor in the peripheral blood. Lindroos et al. (1991) demonstrated that between 1 and 2 hr after partial hepatectomy, the plasma HGF level rapidly rises approximately 15to 20-fold above that found in control animals; a 10-fold increase in the plasma HGF level is already seen within 30 min after the operation. These original studies have now been corroborated by several other groups (Kinoshita et al., 1991; Tomiya et al., 1992b). Similar changes are also seen when the liver is damaged by carbon tetrachloride (CC14). The rise in HGF is immediate and equally as sustained as after partial hepatectomy, but it is elevated for a longer period. Thus, in animals following either a partial hepatectomy or CCI 4 exposure, the rise in plasma HGF occurs rapidly and precedes the peak in hepatocyte DNA synthesis by approximately 20 to 24 hr (Lindross et al., 1991). Furthermore, in terms of a temporal correlation, the rise in HGF occurs during the time frame that is compatible with the induction in immediate early gene expression (see Chapter 4). The fact that the plasma HGF level significantly increases in the peripheral blood also makes HGF compatible with the early findings documenting the rapid emergence of blood borne hepatotrophic factors following a partial hepatectomy (Moolten and Bucher, 1967; Jirtle and Michalopoulos, 1982). The role of HGF as a potential initiator of the hepatic regenerative process was further strengthened by recent studies that showed HGF stimulates the expression of some of the immediate early genes in primary cultures of hepatocytes (Tewari et al., 1992). A characteristic marker of this process is the transcription factor known as liver regeneration factor (LRF), a transcription factor that functions similarly to I kappa B (IKB). Although L R F gene expression is stimulated in cultured hepatocytes by HGF (Weir et al., 1994), other factors such as EGF also enhance the expression of LRF. This suggests that this effect of HGF is not specific, but rather relates to the overall initiation of hepatocyte growth. The number of growth factors that are capable of initiating DNA synthesis in hepatocytes in chemically defined serum-free medium is rather limited. These factors include HGF, EGF, transforming growth factor cx (TGFcx), and aFGF (Michalopoulos, 1990). Of these growth factors, HGF is the only one whose plasma level rises significantly after partial hepatectomy. There is no strong evidence for changes in the plasma EGF concentration after partial hepatectomy, and the TGFcx level rises only during the late stages of regeneration (Tomiya and Fujiwara, 1993); there is no evidence for any change in the plasma level of aFGE It should be pointed out, however, that although the plasma EGF level does not rise following a partial hepatectomy, EGF is continually available to the liver through the portal circulation. EGF is produced by multiple sites in the gastrointestinal tract, including the Brunner's glands of the duodenum (Skov Olsen et al., 1985). Furthermore, there is indirect evidence which suggests that in addition to HGF, EGF may play a

2.

Hepatocyte Growth Factor (HGF) and Its RecePtor (Met)

35

role in the early stages of the regenerative process. A characteristic finding subsequent to the interaction of most growth factors with their receptors is their downregulation. This has been shown to be the case with the EGF receptor after partial hepatectomy (Earp and O'Keefe, 1981), suggesting that EGF may also participate in triggering the phenomena seen during the earliest stages of the regenerative process. In addition to the rise of HGF in the plasma, norepinephrine, a comitogen for hepatocytes also increases substantially during liver regeneration (Cruise et al., 1987). Circulating norepinephrine is degraded primarily in liver. Thus, the rapid elevation of norepinephrine may reflect the release of catecholamines following the stress of partial hepatectomy as well as a decreased rate of degradation of circulating catecholamines by the reduced mass of hepatic tissue. Norepinephrine has a synergistic interaction with HGF in stimulating hepatocyte DNA synthesis (Lindroos et al., 1991). Therefore, the elevation of HGF plus the synergistic effect of norepinephrine may play an interactive role in stimulating hepatocytes to proliferate. Other growth factors also increase after partial hepatectomy, including TGF]31 (Jirtle and Michalopoulos, 1994). Despite the well-documented rise of HGF after partial hepatectomy, the reasons for this phenomenon are not clearly understood. An obvious hypothesis derives from the fact that the liver is the major site for degradation and processing of HGE Accordingly, a sharp decrease in liver mass would result in a decreased clearance of HGF, leading to a rise in the plasma HGF level. However, when plasma HGF clearance was compared between normal animals and those that had received a partial hepatectomy, it was found that partial hepatectomy only led to a 50% decrease in HGF clearance (Appasamy et al., 1993). This does not account for the observed 15- to 20fold increase in HGF plasma concentration during liver regeneration. It also does not explain the significant rise in the plasma level of TGF~I. Both findings are more compatible with a hypothesis that the rise in HGF occurs because of its release from preexisting storage sites. This hypothesis will be discussed subsequently in this chapter as it relates to the early proteolytic events following partial hepatectomy. Although HGF protein rises sharply in the plasma, HGF mRNA in the liver does not change until approximately 3 to 4 hr after partial hepatectomy (Zarnegar et al., 1991). At that time, the HGF mRNA level in the liver increases and remains elevated for approximately 36 hr after partial hepatectomy. Although the site of production of the HGF mRNA in regenerating liver has not been fully investigated, it is assumed to be the same cellular site as in the normal liver, namely the cells of Ito. The signals that lead nonparenchymal cells to synthesize HGF as well as TGFI31 after partial hepatectomy are not clear. Additionally, HGF mRNA expression is increased in the lungs and kidneys following a partial hepatectorny (Yanagita et al., 1992).

36


This led to the hypothesis that a specific substance, termed "injurin" was produced shortly after partial hepatectomy and was responsible for the increase in the level of HGF mRNA in different tissues (Okazaki et al., 1994). Despite extensive studies, however, the existence of this molecule has still not been documented. Interestingly, recent studies have shown that IL-1 (Matsumoto et al., 1992) as well as IL-6 (Moghul et al., 1994) can stimulate production of HGF and its receptor in responding cell lines. IL-6 also increases after partial hepatectomy or during chronic liver failure (Matsunami et al., 1992). Therefore, it is possible that some of the observed systemic changes in the HGF mRNA expression after partial hepatectomy are mediated by IL-1 and/or IL-6. Of interest is that stimulation of HGF production by IL-6 is matched by the existence of several IL-6 responding elements in the regulatory domain of the H G F gene as well as in the regulatory domain of the c - m e t gene coding for the HGFr (Liu et al., 1994c). In fact, IL-6 has been shown to increase the expression of both H G F and c - m e t in cells in which either of these genes are normally expressed. Thus, IL-6 may be a major regulator for reparative growth responses directed by HGF subsequent to acute inflammation in tissues and organs. IL-6 and other interleukins are produced during inflammation by macrophages and cells in liver and other peripheral tissues. IL-6 induction of H G F and c - m e t expression may, in a coordinate fashion, control tissue repair and angiogenesis following tissue damage. The interaction between IL-6 and HGF makes HGF a growth factor of major importance in local tissue repair and raises further questions about the role of IL-6 as one of the early stimulators of the regenerative process. The importance of the increase in HGF mRNA expression, and the presumed production of new HGF following partial hepatectomy, in the sustenance or further enhancement of growth of hepatocytes is not clear. Hepatocytes are already in the G1 phase of the cell cycle by the time HGF mRNA expression reaches its peak; the same is true for the mRNA expression for TGF~ and aFGE It has been hypothesized that these growth factors, produced by hepatocytes or by Ito cells, may further promote the proliferation of hepatocytes. Supportive of this postulate is the finding that sustained growth of hepatocytes, and eventual appearance of neoplasia, is seen in the liver of TGFc~ transgenic mice (Lee et al., 1992). This has not been the case, however, in HGF transgenic mice (Shiota et al., 1994). It is possible that in addition to having effects on hepatocytes themselves, these growth factors may also direct the proliferation of the nonparenchymal cells which occurs 24 hr after the first cycle of hepatocyte DNA synthesis (Grisham, 1962). In addition, HGF has morphogenic effects on hepatocytes (see sections below on HGF and embryogenesis and liver disease). Sustained HGF production by elements in the liver may be important for further directing the late stage morphogenic effects that occur during liver regeneration (i.e.,

2.


37

the formation of acinar structures and mature hepatic plates) (Stamatoglou and Hughes, 1994). The transformation of hepatocytes into acinar structures and hepatic plates is an effect induced by HGF in hepatocyte cultures (Michalopoulos, 1992). Thus, the role of HGF in liver regeneration should be seen as an extended one to encompass aspects of hepatic morphogenesis that occur late in the regenerative process and mimic in many regards the changes seen during embryogenesis.

V. HGF and the Early Proteolytic Events Following Partial Hepatectomy If the rise in the plasma concentration of HGF and norepinephrine provides the main stimuli leading to liver regeneration, a derivative prediction is that the injection of HGF in sufficient amounts would initiate hepatocyte proliferation in the livers of normal rats. However, when HGF is administered in large concentrations to normal rats, either through the portal vein or systemically, stimulation of DNA synthesis is seen only in the hepatocytes surrounding the portal triads. In contrast, if HGF is injected into rats following special nutritional manipulations or surgical procedures (30% hepatectomy), it induces a relatively large number of hepatocytes to undergo DNA synthesis (Webber et al., 1994). The precise nature of these experimental "priming" events is not clear, since these manipulations by themselves do not stimulate DNA synthesis. These findings suggest, however, that most hepatocytes, especially those in midzonal and centralobular regions of the liver, are incapable of responding to HGF unless subjected to a prior "priming" modification. In another study (Liu et al., 1994b), it was shown that if minute amounts of collagenase (approximately 1/100 of the amount required for liver dissociation) are injected into the liver prior to the injection of HGF, the proliferative effect of HGF is dramatically enhanced and a large percentage of the quiescent hepatocytes undergo DNA synthesis. Furthermore, the cells recruited into DNA synthesis come from all areas of the lobule, including the hepatocytes present in the midzonal and central lobular regions. The enhanced recruitment of hepatocytes into DNA synthesis by HGF following collagenase treatment is of interest since other studies where hepatocytes isolated from the liver by collagenase treatment show enhanced expression of immediate early genes and behave as though they were already in the G 1 phase of the cell cycle (Kost and Michalopoulos, 1990). Taken together, these findings suggest that proteolytic events, presumably mimicked by collagenase, occur very shortly after partial hepatectomy and render hepatocytes responsive to the influx of HGF. If there is acute proteolysis following partial hepatectomy, it may also be

38


responsible for the sharp rise in HGF, as well as other cellular matrix bound factors such as TGFJ31 and hyaluronic acid (HA) (Figure 2). As mentioned above, large amounts of HGF are bound in the liver matrix. These HGF binding sites range from the very low affinity (HSPG) to relatively higher affinity glycosulfopholipids. Proteolytic degradation of these sites may release a large mount of HGF into the circulation that when returned to the liver via the vasculature exerts a mitogenic effect. Currently, the evidence for proteolytic events early after partial hepatectomy is rather small. Enhanced expression of the urokinase receptor (uPAr) in the plasma membrane fraction

Figure 2 Outline of the proposed scheme of the early events triggering liver regeneration. The emergence of the urokinase receptor (uPAr) in the plasma membrane (discussed in the text) leads to activation of urokinase (uPA) which can further activate type IV collagenase and generate plasmin. Abundant literature has shown that plasmin is involved in the activation of matrix metalloproteinases. It is hypothesized that the activation of the matrix metalloproteinases leads to the degradation of the connective tissue matrix around hepatocytes resulting in the observed increases in the plasma levels of the biomatrix bound growth factors HGF and TGFIB. Hyaluronic acid (HA), a well-known matrix component, has also been found to be increased in the plasma soon after partial hepatectomy. This composite of proven facts and hypothesis is most compatible with the finding that collagenase treatment prior to HGF injection into the portal circulation leads to the stimulation of hepatocyte DNA synthesis comparable to that seen in liver regeneration.

2.


39

of liver occurs within 5 min after partial hepatectomy (Mars and Michalopoulos, 1994). The enzyme involved in HGF activation, uPA, also converts plasminogen into plasmin. The latter protease is known to activate a variety of tissue bound metalloproteinases (e.g., type IV collagenase, stromelysin, interstitial collagenase, etc.) (Rifkin, 1992; Duffy, 1992; Blasi, 1993; KeskiOja et al., 1993). uPA also directly activates type IV collagenase (Duffy, 1992; Kleiner and Stetler-Stevenson, 1993; Blasi, 1993; Keski-Oja et al., 1992). uPA has a catalytic effect in simple solutions, but its activity is dramatically enhanced after binding to the uPAr. Therefore, the enhanced presence of the uPAr within minutes after partial hepatectomy should increase uPA activity. Given recent evidence about the rapid cascade of events in plasma membrane surface proteolysis (Rifkin, 1992; Duffy, 1992; Blasi, 1993), the elevated presence of the uPAr in hepatic plasma membranes has the potential of leading to a quick cascade of proteolytic events mediated by matrix metalloproteinases activated by uPA. That could further lead to matrix degradation around the hepatocytes, causing the release of matrix bound growth factors such as TGF~31 and HGF (Figure 2). Subsequent to the release of TGF~31 from the matrix, it would be inactivated by cx2-macroglobulin (O'Connor-McCourt and Wakefield, 1987) thereby blocking its mitoinhibitory effects on hepatocytes. In contrast, matrix bound, inactive, single chain HGF in the presence of enhanced uPA activity would be converted to a heterodimeric activated molecule that could stimulate hepatocyte proliferation. This hypothesis would also explain the paradox of the unique effect of plasma HGF on hepatocyte proliferation. Hepatocytes are unique, among all the potential epithelial targets of HGF since they are stimulated into mitogenesis following a partial hepatectomy. HGF in the plasma should theoretically be capable of being mitogenic to all the potential cellular targets. The restricted effect of HGF on hepatocytes could be explained if the biomatrix surrounding the epithelial cells needs to first be degraded so that HGF and other growth factors have a mitogenic effect in tissues. A competition between matrix binding and signaling receptor (i.e., HGFr) binding may also be an issue that relates peculiarly to HGE In contrast to other growth factors, in addition to the very low affinity and high capacity binding sites, there are also relatively high affinity, nonsignaling receptor binding sites for HGF in the biomatrix which may offer a competition between binding of HGF to HGFr and making HGF not available to function in a mitogenic manner. This competition would also be eliminated by early proteolytic events leading to the destruction of such relatively high affinity, nonsignaling HGF matrix binding sites. Although this hypothesis is attractive and may explain some of the peculiar events related to the rise of HGF in the plasma, and the unique targeting

40


of the HGF mitogenic effect to the liver, additional investigations are still required. These would include defining the proteolytic targets as well as the enzymes involved in carrying out the actual proteolysis.

VI. H G F Localization A. In Liver Embryogenesis Given the strong mitogenic, motogenic, and morphogenic effects of HGF in several tissues as well as the high concentration of HGF in the placenta, it is likely that HGF plays a major role in embryogenesis. In recent studies involving HGF implanted into the embryo, abnormal release of HGF led to abnormalities in embryonic polarity with the formation of multiple notochords (Stern and Ireland, 1993). Specifically in relationship to the liver, HGF as well as its receptor are intensely expressed during hepatic embryogenesis. Of the multiple tissues in which HGF and the HGFr are expressed, liver is by far the one with the most expression (Sonnenberg et al., 1993; Katyal and Michalopoulos, 1995). Several authors have pointed out that embryogenesis of the liver proceeds through characteristic stages (Stamatoglou and Hughes, 1994). During most of embryogenesis, hepatocytes do not possess the adult canalicular sinusoidal surface orientation. Toward the end of embryogenesis (1 to 2 days prior to birth) and during the first 2 to 3 days of the postnatal period, the apolar hepatocytes become organized into acinar structures. These acinar/ductal structures then eventually become oriented into typical adult looking hepatic plates that are composed of polar hepatocytes with sinusoidal and canalicular surfaces. The high concentrations of HGF and its receptor in the embryonic liver and during the early stages of the postnatal period suggest that HGF may be involved in these morphogenic processes. This postulate is further strengthened by the studies showing that HGF can affect the morphological structures formed by hepatocytes in primary culture. When hepatocytes are cultured in type I collagen gels and exposed to HGF in normal media, hepatic plates form after an intense period of motogenesis and mitogenesis. If the HGF concentration is increased in media that allows for long-term hepatocyte proliferation, acinar structures composed of ductular hepatocytes are formed (Michalopoulos et al., 1993; Bloc and Michalopoulos, 1995). In standard monolayer cultures with the same medium, HGF induces hepatocytes to transform into hepatoblasts. These studies suggest that HGF is involved not only in hepatocyte proliferation in embryonic liver, but also in the overall transformation of an embryonic liver composed of apoiar hepatocytes into an organ of hepatic plates. However, to further investigate the

2.


41

role of HGF in liver embryogenesis studies need to be performed with HGF transgenic mice and HGF "knockout" mice. B. In Liver Disease

Acute liver disease (viral hepatitis) or chronic liver disease (chronic active hepatitis, cirrhosis) leads to a rise in the plasma level of HGF (Tsubouchi et al., 1989). This correlates with the severity of the disease, and usually the level of HGF is inversely correlated with the prognosis of the disease; the patients who fare better tend to have lower concentrations of HGF (Tomiya et al., 1992a). Even more striking is the elevation of HGF in patients with fulminant hepatic failure. This syndrome is precipitated by a variety of offending insults (Wiesner, 1991; Riegler and Lake, 1993; Sherlock, 1993). It can be seen in patients infected with hepatitis A, B, or C viruses, patients exposed to toxins (acetaminophen, amanitin poisoning, other chemical poisons, etc.), patients with autoimmune liver disease, etc. Regardless of the etiology, the injury to the liver proceeds with a rapid destruction of a very large number of hepatocytes. If the process is unfettered, it will lead to eventual destruction of all hepatocytes in the liver, resulting in a condition described in standard pathology textbooks as "acute yellow atrophy" of the liver. The liver volume shrinks, liver weight decreases, and hepatic function indicators plummet. This condition occurs in approximately I to 2% of patients with a variety of acute hepatic inflammatory conditions. A percentage of the patients affected with fulminant hepatitis (20 to 30%) spontaneously recover. Histological examination of the liver in patients with this condition is now often performed because many of these patients are treated with orthotopic liver transplantation (see Chapter 13). The resected livers therefore have allowed for histological studies of the processes leading to fulminant hepatitis. Initially, cell proliferation is seen both in hepatocytes and bile ductal epithelial (Wolf and Michalopoulos, 1992). At late stages of the disease, hepatocytes cease to proliferate. Concomitantly, there is proliferation of hepatocytes arranged in acinar structures as well as exuberant proliferation of frank bile duct epithelial cells. These acinar structures composed of hepatocytes are the basis for the term "ductular hepatocytes" which is reserved for hepatocytes participating in the formation of these acinar structures and those having markers of mixed differentiation between bile ducts and hepatocytes (Sirica et al., 1994). In the context of the above description of the role of HGF in embryology, it is perhaps reasonable to speculate that the observed acinar transformation of hepatocytes is recapitulating the embryonic events occurring in the liver. Given the propensity of HGF to induce the same changes in hepatocytes in primary culture, it is reasonable to

42

George K. MichaIopoulos

speculate that the very high levels of HGF present in fulminant hepatitis are responsible for the observed transformation of hepatocytes into acinar structures. The effect of HGF in fulminant hepatitis, however, may be even more insidious. In one of its early incarnations, HGF was isolated as "sarcoma cytotoxic factor" (Higasho and Shima, 1993). Sarcoma cytotoxic factor was subsequently found to be identical to HGF, and it has now been shown that high concentrations of HGF have a mitoinhibitory effect on a variety of normal and neoplastic tissues (Shiota et al., 1992). At high concentrations, HGF also loses its mitogenic effect on hepatocytes in primary culture (Michalopoulos, 1992). The high concentrations of HGF present during fulminant hepatitis could be responsible for the eventual cessation in hepatocyte proliferation, the transformation of hepatocytes into acinar structures, and the eventual decline in hepatocyte-supported liver function. This is an avenue worth exploring because it may allow a therapeutic approach for fulminant hepatitis aimed at restricting the very high concentrations of HGF found in this disease state. The very high levels of HGF in the plasma of patients with fulminant hepatic failure appear to render meaningless the pursuit of the notion that this disease can be treated by injecting high concentrations of HGF to stimulate liver regeneration. Hepatocyte proliferation throughout most of the early and middle stages of this disease is even higher than that which occurs following a partial hepatectomy. Therefore, it is doubtful that further stimulation of the process can be achieved by injecting a substance that is already present in high concentrations in the plasma and available to the hepatocytes. It should be pointed out, however, that there have been no detailed studies of the nature of the circulating HGE If the circulating HGF is in the inactive (single chain) form, then activation of HGF may be a problem. This needs to be further studied. C. In Liver Carcinogenesis The role of HGF in liver regeneration, embryogenesis, and liver disease is well documented; however, its role in hepatic carcinogenesis is not as clear. The HGFr is expressed in many hepatocellular carcinomas, in some instances at increased levels and in other instances at decreased levels. Hepatoma cell lines that express the HGFr respond to HGF slightly, intensely or not at all, for reasons which at this time remain obscure. In contrast to TGFcx, which is expressed by the majority of hepatocellular carcinomas, no hepatocellular carcinoma has yet been described that expresses HGE Furthermore, in transgenic mice where HGF-expression was targeted to the liver under the direction of the albumin promoter, there was no observed neoplasms (Shiota et al., 1994). In fact when HGF cDNA is transfected into

2.


43

the human hepatoma cell line, HepG2, the cells expressing the HGF cDNA are less tumorigenic than the normal HepG2 cells (Shiota et al., 1992). Growth inhibitory effects of HGF have also been shown in several other tumor cell lines from a wide range of cell origins (Tajima et al., 1991). It should be pointed out, however, that there is no example of a specific neoplasia in which all of the tumors are either entirely negatively or positively affected by H G E Typically, as for most types of neoplasms, some neoplasms are inhibited and others are stimulated. Nevertheless, there is no doubt that HGF and its receptor have a prominent role in neoplasia in general. The motogenic effects of HGF make it likely that it functions in liver metastases. In fact, the very high concentrations of HGF in the liver matrix would make liver an appealing site for neoplasms whose growth is stimulated by HGE Thus, the HGF bound to the liver biomatrix may enhance the metastatic properties of neoplasms which respond positively to HGE These neoplasms may include squamous cell carcinomas, melanomas, breast cancer, uroepithelial neoplasms, and prostatic cancers. In recent studies, we found that when HGF is infused into livers containing a large number of small primary tumors induced by diethylnitrosamine, the proliferation of most of the tumors was reduced by HGF; however, a small number of tumors were found that were totally resistant to this mitoinhibitory effect (Liu and Michalopoulos, 1993). This suggests that HGF may be useful as a component of therapeutic regimens for hepatocellular carcinomas. In view of competitive molecular analogues of HGF naturally produced by mRNA splicing or artificially produced by site directed mutagenesis (Lokker et al., 1992), it may also be possible to affect the growth of HGF-dependent neoplasms by using HGF competitive molecular analogues.

VII. Summary Although HGF is now a recognized multifunctional cytokine with target effects in different organs and tissues, it has some unique aspects in its relationship toward the liver. The paracrine effects of HGF probably regulate local tissue responses in a variety of organs, but the endocrine effects of HGF appear to be limited primarily to the liver. No documented changes in HGF plasma levels are seen in any other conditions comparable to those seen during liver regeneration, fulminant hepatic failure, or chronic liver disease. The very high expression of HGF and its receptor in embryonic liver, totally out of proportion to that seen in the rest of the embryonic organs, however, may obscure the important role that HGF also plays in the development of the central nervous system, another site where HGF and its receptor are highly expressed (Jung et al., 1994). The endocrine effect of

44


HGF and its involvement in events related to liver disease make it likely that in contrast to other growth factors, HGF and its molecular competitive analogues may be useful in therapy. Clearly, HGF is involved in the earliest stages of liver regeneration. Undoubtedly, the regenerative process is a very complex one, but the evidence so far indicates that HGF is the best candidate for the initial trigger of hepatocyte proliferation following partial hepatectomy; it is capable of explaining both the phenomena of the blood borne hepatotrophic factors as well as immediate early gene expression. As we learn more about the other complex parameters that are undoubtedly involved in liver growth regulation, the role of HGF is likely to remain an eminent one and will provide substantial justification for the adoption of a hepatic derived name for this otherwise promiscuous cytokine. References Appasamy, R., Tanabe, M., Murase, N., Zarnegar, R., Venkataramanan, R., Van Thiel, D. H., and Michalopoulos, G. K. (1993). Hepatocyte growth factor, blood clearance, organ uptake, and biliary excretion in normal and partially hepatectomized rats. Lab. Invest. 68, 270-276. Blasi, E (1993). Urokinase and urokinase receptor: A paracrine/autocrine system regulating cell migration and invasivenesss. BioEssays 15, 105-111. Bloc, G., and Michalopoulos, G. K. (1995). Unpublished results. Bottaro, D. P., Rubin, J. S., Faletto, D. L., Chan, A. M. L., Kmiecick, T. E., Vande Woude, G. E, and Aaronson, S. A. (1991). Identification of the hepatocyte growth factor receptor as the c-met proto-oncogene. Science 251,802-804. Bussolino, E, Di Renzo, M. E, Ziche, M., Bocchietto, E., Olivero, M., Naldini, L., Gaudino, G., Tamagnone, L., Coffer, A., and Comoglio, P. M. (1992). Hepatocyte growth factor is a potent angiogenic factor which stimulates endothelial cell motility and growth. J. Cell Biol. 119, 629-641. Chan, A. M., Rubin, J. S., Bottaro, D. P., Hirschfield, D. W., Chedid, M., and Aaronson S. A. (1991). Identification of a competitive HGF antagonist encoded by an alternative transcript. Science 254, 1382-1385. Cruise, J. L., Knechtle, S. J., Bollinger, R. R., Kuhn, C., and Michalopoulos, G. K. (1987). Adrenergic effects and liver regeneration. Hepatology (Baltimore) 7, 1189-1194. de Hemptinne, B., Lorge, E, Kestens, P. J., and Lambotte, L. (1985). Hepatocellular hyperpolarizing factors and regeneration after partial hepatectomy in the rat. Acta Gastroenterol. Belg. 48, 424-431. Duffy, M. J. (1992). The role of proteolytic enzymes in cancer invasion and metastasis. Clin. Exp. Metastasis 10, 145-155. Earp, H. S., and O'Keefe, E. J. (1981). Epidermal growth factor receptor number decreases during rat liver regeneration. J. Clin. Invest. 67, 1580-1583. Fisher, B., Szuch, P., Levine, M., and Fisher, E. R. (1971). A portal blood factor as the humor agent in liver regeneration. Science 171,575-577. Gaudino, G., Follenzi, A., Naldini, L., Collesi, C., Santoro, M., Gallo, K. A., Godowski, P. J., and Comoglio, P. M. (1994). RON is a heterodimeric tyrosine kinase receptor activated by the HGF homologue MSP. EMBO J. 13, 3524-3532.

2.


45

Gherardi, E., Sharpe, M., Lane, K., Sirulnik, A., and Stoker, M. (1993). Hepatocyte growth factor/scatter factor (HGF/SF), the c-met receptor and the behaviour of epithelial cells. Symp. Soc. Exp. Biol. 47, 163-181. Giordano, S., Ponzetto, C., Di Renzo, M. E, Cooper, C. S., and Comoglio, P. M. (1989). Tyrosine kinase receptor indistinguishable from the c-met protein. Nature (London) 339, 155-156. Grisham, J. W. (1962). A morphologic study of deoxyribonucleic acid synthesis and cell proliferation in regenerating liver; autoradiography with thymidine-H3. Cancer Res. 22, 842849. Higashio, K., and Shima, N. (1993). Tumor cytotoxic activity of HGF-SE EXS 65, 351-368. Jirtle, R. L., and Michalopoulos, G. K. (1994). unpublished observations. Jirtle, R. L., and Michalopoulos, G. (1982). Effects of partial hepatectomy on transplanted hepatocytes. Cancer Res. 42, 3000-3004. Jung, W., Castren, E., Odenthal, M., Vande Woude, G. E, Ishii, T., Dienes, H. P., and Lindholm, D., and Schirmacher, P. (1994). Expression and functional interaction of hepatocyte growth factor-scatter factor and its receptor c-met in mammalian brain. J. Cell Biol. 126, 485-494. Katyal, S., and Michalopoulos, G. K. (1995). unpublished results. Keski-Oja, J., Lohi, J., Tuuttila, A., Tryggvason, K., and Vartio, T. (1992). Proteolytic processing of the 72,000-Da type IV collagenase by urokinase plasminogen activator. Exp. Cell Res. 202, 471-476. Kinoshita, T., Hirao, S., Matsumoto, K., and Nakamura, T. (1991). Possible endocrine control by hepatocyte growth factor of liver regeneration after partial hepatectomy. Biochem. Biophys. Res. Commun. 177, 330-335. Kleiner, D. E., Jr., and Stetler-Stevenson, W. G. (1993). Structural biochemistry and activation of matrix metalloproteases. Curr. Opin. Cell Biol. 5, 891-897. Kobayashi, T., Honke, K., Miyazaki, T., Matsumoto, K., Nakamura, T., Ishizuka, I., and Makita, A. (1994). Hepatocyte growth factor specifically binds to sulfoglycolipids. J. Biol. Chem. 269, 9817-9821. Kost, D. P., and Michalopoulos, G. K. (1990).Effect of epidermal growth factor on the expression of protooncogenes c-myc and c-Ha-ras in short-term primary hepatocyte culture. J. Cell. Physiol. 144, 122-127. Lee, G. H., Merlino, G., and Fausto, N. (1992). Development of liver tumors in transforming growth factor alpha transgenic mice. Cancer Res. 52, 5162-5170. Leong, G. E, Grisham, J. W., Hole, B. V., and Albright, M. L. (1964). Effect of partial hepatectomy on DNA synthesis and mitosis in heterotopic partial autografts of rat liver. Cancer Res. 24, 1496-1501. Lindroos, P. M., Zarnegar, R., and Michalopoulos, G. K. (1991). Hepatocyte growth factor (hepatopoietin A) rapidly increases in plasma before DNA synthesis and liver regeneration stimulated by partial hepatectomy and carbon tetrachloride administration. Hepatology (Baltimore 13, 743-750. Liu, M. L., and Michalopoulos, G. K. (1993). unpublished results. Liu, M. L., Mars, W. M., Zarnegar, R., and Michalopoulos, G. K. (1994a). Uptake and distribution of hepatocyte growth factor in normal and regenerating adult rat liver. Am. J. Pathol. 144, 129-140. Liu, M. L., Mars, W. M., Zarnegar, R., and Michalopoulos, G. K. (1994b). Collagenase pretreatment and the mitogenic effects of hepatocyte growth factor and transforming growth factor-alpha in adult rat liver. Hepatology (Baltimore) 19, 1521-1527. Liu, Y., Michalopoulos, G. K., and Zarnegar, R. (1994c). Structural and functional characterization of the mouse hepatocyte growth factor gene promoter. J. Biol. Chem. 269, 41524160.

46


Lokker, N. A., Mark, M. R., Luis, E. A., Bennett, G. L., Robbins, K. A., Baker, J. B., and Godowski, P. J. (1992). Structure-function analysis of hepatocyte growth factor: Identification of variants that lack mitogenic activity yet retain high affinity receptor binding. EMBO J. 11, 2503-2510. Mars, W. M., and Michalopoulos, G. K. (1994). unpublished results. Mars, W. M., Zarnegar, R., and Michalopoulos, G. K. (1993). Activation of hepatocyte growth factor by the plasminogen activators uPA and tPA. Am. J. Pathol. 143, 949-958. Masumoto, A., and Yamamoto, N. (1991). Sequestration of a hepatocyte growth factor in extracellular matrix in normal adult rat liver. Biochem. Biophys. Res. Commun. 174, 90-95. Matsumoto, K., and Nakamura, T. (1992). Hepatocyte growth factor: Molecular structure, roles in liver regeneration, and other biological functions. Crit. Rev. Oncog. 3, 27-54. Matsumoto, K., Okazaki, H., and Nakamura, T. (1992). Up-regulation of hepatocyte growth factor gene expression by interleukin-1 in human skin fibroblasts. Biochem. Biophys. Res. Commun. 188, 235-243. Matsunami, H., Kawasaki, S., Ishizone, S., Hashikura, Y., Ikegami, T., Makuuchi, M., Kawarasaki, H., Iwanaka, T., Nose, A., and Takemura, M. (1992). Serial changes of h-HGF and IL-6 in living-related donor liver transplantation with special reference to their relationship to intraoperative portal blood flow. Transplant. Proc. 24, 1971-1972. Michalopoulos, G. (1992). unpublished results. Michalopoulos, G. K. (1990). Liver regeneration: Molecular mechanisms of growth control. FASEBJ 4, 176-187. Michalopoulos, G. K., and Zarnegar, R. (1992). Hepatocyte growth factor (editorial). Hepatology (Baltimore) 15, 149-155. Michalopoulos, G., Cianciulli, H. D., Novotny, A. R., Kligerman, A. D., Strom, S. C., and Jirtle, R. L. (1982). Liver regeneration studies with rat hepatocytes in primary culture. Cancer Res. 42, 4673-4682. Michalopoulos, G., Houck, K. A., Dolan, M. L., and Luetteke, N. C. (1984). Control of hepatocyte replication by two serum factors. Cancer Res. 44, 4414-4419. Michalopoulos, G. K., Bowen, W., Nussler, A. K., Becich, M. J., and Howard, T. A. (1993). Comparative analysis of mitogenic and morphogenic effects of HGF and EGF on rat and human hepatocytes maintained in collagen gels. J. Cell. Physiol. 156, 443-452. Miyazawa, K., Tsubouchi, H., Naka, D., Takahashi, K., Okigaki, M., Arakaki, N., Nakayama, H., Hirono, S., Sakiyama, O., Takahashi, K., Gohoda, E., Daikuhara, Y., and Kitamura, N. (1989). Molecular cloning and sequence analysis of cDNA for human hepatocyte growth factor. Biochem. Biophys. Res. Commun. 163, 967-973. Miyazawa, K., Shimomura, T., Kitamura, A., Kondo, J., Morimoto, Y., and Kitamura, N. (1993). Molecular cloning and sequence analysis of the cDNA for a human serine protease responsible for activation of hepatocyte growth factor. Structural similarity of the protease precursor to blood coagulation factor XII. J. Biol. Chem. 268, 10024-10028. Moghul, A., Lin, L., Beedle, A., Kanbour-Shakir, A., DeFrances, M. C., Liu, Y., and Zarnegar, R. (1994). Modulation of c-met proto-oncogene (HGF receptor) mRNA abundance by cytokines and hormones: Evidence for rapid decay of the 8 kb c-met transcript. Oncogene 9, 2045-2052. Mohn, K. L., Laz, T. M., Melby, A. E., and Taub, R. (1990). Immediate-early gene expression differs between regenerating liver, insulin-stimulated H-35 cells, and mitogen-stimulated Balb/c 3T3 cells. Liver-specific induction patterns of gene 33, phosphoenolpyruvate carboxykinase, and the jun, fos, and egr families. J. Biol. Chem. 265, 2914-2921. Moolten, E L., and Bucher, N. L. R. (1967). Regeneration of rat liver: Transfer of humoral agents by cross circulation. Science 158, 272-274. Nakamura, T., Nawa, K., and Ichihara, A. (1984). Partial purification and characterization of hepatocyte growth factor from serum of hepatectomized rats. Biochem. Biophys. Res. Commun. 122, 1450-1459.

2.


47

Nakamura, T., Nishizawa, T., Hagiya, M., Seki, T., Shimonishi, M., Sugimura, A., Tashiro, K., and Shimizu, S. (1989). Molecular cloning and expression of human hepatocyte growth factor. Nature (London) 342, 440-443. Naldini, L., Vigna, E., Narsimhan, R. P., Gaudino, G., Zarnegar, R., Michalopoulos, G. K., and Comoglio, P. M. (1991a). Hepatocyte growth factor (HGF) stimulates the tyrosine kinase activity of the receptor encoded by the proto-oncogene c-met Oncogene 6, 501-504. Naldini, L., Weidner, K. M., Vigna, E., Gaudino, G., Bardelli, A., Ponzetto, C., Narsimhan, R. P., Hartmann, G., Zarnegar, R., Michalopoulos, G. K., Birchmeier, W., and Comoglio, P. M. (1991b). Scatter factor and hepatocyte growth factor are indistinguishable ligands for the MET receptor. EMBO J. 10, 2867-2878. Noji, S., Tashiro, K., Koyama, E., Nohno, T., Ohyama, K., Taniguchi, S., and Nakamura, T. (1990). Expression of hepatocyte growth factor gene in endothelial and Kupffer cells of damaged rat livers, as revealed by in situ hybridization. Biochem. Biophys. Res. Commun. 173, 42-47. O'Connor-McCourt, M. D., and Wakefield, L. M. (1987). Latent transforming growth factor-J3 in serum. A specific complex with ot2-macroglobulin. J. Biol. Chem. 262, 14090-14099. Okazaki, H., Matsumoto, K., and Nakamura, T. (1994). Partial purification and characterization of "injurin-like" factor which stimulates production of hepatocyte growth factor. Biochim. Biophys. Acta 1220, 291-298. Pepper, M. S., Matsumoto, K., Nakamura, T., Orci, L., and Montesano, R. (1992). Hepatocyte growth factor increases urokinae-type plasminogen activator (u-PA) and u-PA receptor expression in Madin-Darby canine kidney epithelial cells. J. Biol. Chem. 267, 2049320496. Ponzetto, C., Bardelli, A., Zhen, Z., Maina, E, dalla Zonca, P., Giordano, S., Graziani, A., Panayotou, G., and Comoglio, P. M. (1994). A multifunctional docking site mediates signaling and transformation by the hepatocyte growth factor/scatter factor receptor family. Cell (Cambridge, Mass.) 77, 261-271. Rabes, H. M., Wirsching, R., Tuczek, H. V., and Iseler, G. (1976). Analysis of cell cycle compartments of hepatocytes after partial hepatecomy. Cell Tissue Kinet. 6, 517-532. Riegler, J. L., and Lake, J. R. (1993). Fulminant hepatic failure. Med. Clin. North Am. 77, 1057-1083. Rifkin, D. B. (1992). Plasminogen activator expression and matrix degradation. Matrix Suppl. 1, 20-22. Rosen, E. M., Grant, D. S., Kleinman, H. K., Goldberg, I. D., Bhargava, M. M., Nickoloff, B. J., Kinsella, J. L., and Polverini, P. (1993a). Scatter factor (Hepatocyte growth factor) is a potent angiogenesis factor in vivo. Symp. Soc. Exp. Biol. 47, 227-234. Rosen, E. M., Zitnik, R. J., Bhargava, M. M., Wines, J., and Goldberg, I. D. (1993b). The interaction of HGF-SF with other cytokines in tumor invasion and angiogenesis. EXS 65, 301-310. St. Hilaire, R. J., Hradek, G. T., and Jones, A. L. (1983). Hepatic sequestration and biliary secretion of epidermal growth factor: Evidence for a high-capacity updtake system. Proc. Natl. Acad. Sci. U.S.A. 80, 3797-3801. Schirmacher, P. (1994). Expression and functional interaction of hepatocyte growth factorscatter and its receptor c-met in mammalian brain. J. Cell Biol. 126, 485-494. Schirmacher, P., Geerts, A., Pietrangelo, A., Dienes, H. P., and Rogler, C. E. (1992). Hepatocyte growth factor/hepatopoietin A is expressed in fat-storing cells from rat liver but not myofibroblast-like cells derived from fat-storing cells. Hepatology (Baltimore) 15, 5-11. Sherlock, S. (1993). Fulminant hepatic failure. Adv. Intern. Med. 38, 245-267. Shiota, G., Rhoads, D. B., Wang, T. C., Nakamura, T., and Schmidt, E. V. (1992). Hepatocyte growth factor inhibits growth of hepatocetlular carcinoma cells. Proc. Natl. Acad. Sci. U.S.A. 89, 373-377. Shiota, G., Wang, T. C., Nakamura, T., and Schmidt, E. (1994). Hepatocyte growth factor in

48


transgenic mice: Effects on hepatocyte growth, liver regeneration and gene expression. Hepatology (Baltimore) 19, 962-972. Sirica, A. E., Gainey, T. W., and Mumaw, V. R. (1994). Ductular hepatocytes: Evidence for a bile ductular cell origin in furan-treated rats. Am. J. Pathol. 145, 375-383. Skov Olsen, P., Poulsen, S. S., and Kirkegaard, P. (1985). Adrenergic effects of secretion of epidermal growth factor from Brunner's glands. Gut 26, 920-927. Sonnenberg, E., Meyer, D., Weidner, K. M., and Birchmeier, C. (1993). Scatter factor/hepatocyte growth factor and its receptor, the c-met tyrosine kinase, can mediate a signal exchange between mesenchyme and epithelial during mouse development. J. Cell Biol. 123,223-235. Stamatoglou, S. C., and Hughes, R. C. (1994). Cell adhesion molecules in liver function and pattern formation. FASEBJ. 8, 420-427. Stern, C. D., and Ireland, G. W. (1993). HGF-SF: A neural inducing molecule in vertebrate embryos? EXS 65,369-380. Stoker, M., Gherardi, E., Perryman, M., and Gray, J. (1987). Scatter factor is a fibroblastderived modulator of epithelial cell mobility. Nature (London) 327, 239-242. Tajima, H., Matsumoto, K., and Nakamura, T. (1991). Hepatocyte growth factor has potent anti-proliferative activity in various tumor cell lines. FEBS Lett. 291, 229-232. Tewari, M., Dobrzanski, P., Mohn, K. L., Cressman, D. E., Hsu, J. C., Bravo, R., and Taub, R. (1992). Rapid induction in regenerating liver of RL/IF-1 (an I kappa B that inhibits NFkappa B, RelB-p50, and c-Rel-p50) and PHF, a novel kappa B site-binding complex. Mol. Cell. Biol. 12, 2898-2908. Tomiya, T., and Fujiwara, K. (1993). Serum levels of transforming growth factor-alpha in patients after partial hepatectomy as determined with an enzyme-linked immunosorbent assay. Hepatology (Baltimore) 18, 304-308. Tomiya, T., Nagoshi, S., and Fujiwara, K. (1992a). Significance of serum human hepatocyte growth factor levels in patients with hepatic failure. Hepatology (Baltimore) 15, 1-4. Tomiya, T., Tani, M., Yamada, S., Hayashi, S., Umeda, N., and Fujiwara, K. (1992b). Serum hepatocyte growth factor levels in hepatectomized and nonhepatectomized surgical patients. Gastroenterology 103, 1621-1624. Tsubouchi, H., Horono, S., Gohda, E., Nakayama, H., Takahashi, K., Sakiyama, O., Miyazaki, H., Sugihara, J., Eiichi, T., Muto, Y., Dakuhara, Y., and Hashimoto, S. (1989). Clinical significance of human hepatocyte growth factor in blood from patients with fulminant hepatic failure. Hepatology (Baltimore) 9, 875-881. Webber, E. M., Godowski, P. J., and Fausto, N. (1994). In vivo response of hepatocytes to growth factors requires an initial priming stimulus. Hepatology (Baltimore) 19, 489-497. Weidner, K. M., Hartmann, G., Sachs, M., and Birchmeier, W. (1993). Properties and functions of scatter factor/hepatocyte growth factor and its receptor c-Met. Am. J. Respir. Cell Mol. Biol. 8, 229-237. Weir, E., Chen, Q., DeFrances, M. C., Bell, A., Taub, R., and Zarnegar, R. (1994). Rapid induction of nRNAs for liver regeneration factor and insulin-like growth factor binding protein-1 in primary cultures of rat hepatocytes by hepatocyte growth factor and epidermal growth factor. Hepatology (Baltimore) 20, 955-960. Wiesner, R. H. (1991). Acute fulminant hepatic failure. Transplant. Proc. 23, 1892-1894. Wolf, H. K., and Michalopoulos, G. K. (1992). Hepatocyte regeneration in acute fulminant and non-fulminant hepatitis: A study of proliferating cell nuclear antigen expression. Hepatology (Baltimore) 15, 707-713. Wolf, H. K., Zarnegar, R., and Michalopoulos, G. K. (1991). Localizatoin of hepatocyte growth factor in human and rat tissues: An immunohistochemical study. Hepatology (Baltimore) 14, 488-494. Yanagita, K., Nagaike, M., Ishibashi, H., Niho, Y., Matsumoto, K., and Nakamura, T. (1992). Lung may have an endocrine function producing hepatocyte growth factor in response to injury of distal organs. Biochem. Biophys. Res. Commun. 182, 802-809.

2.


49

Zarnegar, R., Muga, S., Enghild, J., and Michalopoulos, G. (1989). NH2-terminal amino acid sequence of rabbit hepatopoietin A, a heparin-binding polypeptide growth factor for hepatocytes. Biochem. Biophys. Res. Commun. 163, 1370-1376. Zarnegar, R., DeFrances, M. C., Oliver, L., and Michalopoulos, G. (1990). Identification and partial characterization of receptor binding sites for HGF on rat hepatocytes. Biochem. Biophys. Res. Commun. 173, 1179-1185. Zarnegar, R., DeFrances, M. C., Kost, D. P., Lindroos, P., Mich~lopoulos, G. K. (1991). Expression of hepatocyte growth factor mRNA in regenerating rat ~rer after partial hepatectomy. Biochem. Biophys. Res. Commun. 177, 559-565. Zhen, Z., Giordano, S., Longati, P., Medico, E., Campiglio, M., and Comoglio, P. M. (1994). Structural and functional domains critical for constitutive activation of the HGF receptor (Met). Oncogene 9, 1691-1697. Zioncheck, T. E, Richardson, L., DeGuzman, G. G., Modi, N. B., Hansen, S. E., and Godowski, P. J. (1994). The pharmacokinetics, tissue localization, and metabolic processing of recombinant human hepatocyte growth factor after intravenous administration in rats. Endocrinology (Baltimore) 134, 1879-1887.


3 Structure and Functions of the HGF Receptor (c-Met) Paolo M. Comoglio Elisa Vigna Institute for Cancer Research and Treatment, University of Torino School of Medicine, 10126 Torino, Italy

I. Hepatocyte Growth Factor and Scatter Factor Hepatocyte growth factor (HGF) and scatter factor (SF) are molecules secreted by mesenchymal cells acting in a paracrine way on epithelial cells. The factors were isolated independently because they show different, and apparently unrelated, biological effects. HGF was purified from several sources, such as rat platelets and human and rabbit serum, using a biological assay based on hepatocyte proliferation in vitro (Nakamura et al., 1986; Godha et al., 1988; Zarnegar and Michalopoulos, 1989). SF was isolated from cell culture supernatants, given its ability to induce dissociation of the epithelial monolayers--the so-called "scatter" effect (Stoker et al., 1987; Gherardi et al., 1989; Weidner et al., 1990). Subsequently, biochemical purification and cDNA cloning demonstrated that the two molecules are indistinguishable, and have the same biological activities, the same biochemical and immunological properties, and ultimately the same sequence (Weidner et al., 1991; Naldini et al., 1991c). H G F is a disulfideqinked heterodimer made of a heavy subunit (~) of 60 kDa and a light subunit (13) of 32 to 34 kDa. The two chains are glycosylated and differences in glycosylation generate different forms of the chain. HGF is homologous to serine proteases involved in the blood clotting cascade (e.g., 38% homology with plasminogen). The e~ chain contains a "hairpin" loop and four "kringle" modules. The ~ chain has the same Liver Regeneration and Carcinogenesis Copyright 9 1995 by Academic Press, Inc. All rights of reproduction in any form reserved.

51

Paolo M. Comoglio and Elisa Vigna

52

overall structure of the serine proteases, but critical amino acids in the catalytic domain are mutated (Figure 1) (Miyazawa et al., 1989; Nakamura et al., 1989; Weidner, et al., 1990, 1991). No evidence for proteolytic activity of HGF has ever been reported. Interestingly, two independent reports have shown the existence of another protein related to HGF, called "HGFlike" (Han et al., 1991) or macrophage stimulating protein (MSP) (Skeel et al. 1991; Yoshimura et al., 1993), that shares with HGF the overall fourkringle/protease-like structure. HGF is synthesized and secreted as a single chain precursor devoid of biological activity (Naldini et al., 1992; Hartmann et al., 1992; Lokker et al., 1992; Naka et al., 1992; Mizuno et al., 1992). A serine protease in the extracellular environment cleaves the precursor between Arg 494 and Va149s to generate the two-chain mature form. Urokinase-type plasminogen activator (uPA) has been shown to correctly process HGF in vitro (Naldini et al., 1992). A serum-derived serine protease, homologous to coagulation factor XII, can also cleave the HGF precursor (Miyazawa et al., 1993). This enzyme must first, however, be proteolytically activated by thrombin (Shimomura et al., 1993). It is thus possible to postulate two different paths for the activation of HGF, one mediated by uPa, normally present in plasma and tissues, the other activated only during the coagulation process. Recently, a rather broad spectrum of target cells and biological activities have been assigned to HGF. It induces mitogenesis in different epithelial cell

HPL

~

K1

K2

K3

K4 I R

V

I

Q

(46%)

(60%)

K1

K2

(66%)

(49%)

K3

K4

I

HGF

(45%)

(50%)

I

D

-S

~ R

V

I

Q

! I

Q

I

MSP

Y

Figure 1 Schematic structure of hepatocyte growth factor (HGF) and macrophage stimulating protein (MSP). HPL designates the hairpin loop and K1 to K4 are the four kringle domains of HGF and MSP. The mature oLand 13chains originate from proteolytic cleavage of a single-chain precursor, at a specific site between Arg 494 (R) and Val ags (V). The estimated position of the disulfide bond (S---S) is indicated. The HGF 13 chain is devoid of proteolytic activity due to substitution of critical residues (Q, D, Y) in the catalytic domain; similarly, Q, Q, Y substitutions have occurred in MSP. Numbers in parentheses indicate the overall and the partial homology of the various domains.

3.

Structure and Functions of the HGF Receptor (c-Met)

53

lines (Kan et al., 1991; Rubin et al., 1991) and in primary culture of hepatocytes (Zarnegar and Michalopoulos, 1989), melanocytes (Matsumoto et al., 1991a; Halaban et al., 1992), and keratinocytes (Matsumoto et al., 1991b). HGF also induces growth and wound healing in endothelial cells and behaves as a potent angiogenic factor in vivo (Bussolino et al., 1992; Grant et al., 1993). HGF stimulates invasion of the extracellular matrix by epithelialderived cancer cells (Weidner et al., 1990), and may have a paradoxical growth-inhibitory effect (Higashio et al., 1990; Tajima et al., 1991; Shiota et al., 1992a). A morphogenic effect has also been demonstrated. HGF induces tubule formation by kidney epithelial cells in vitro (Montesano et al., 1991; Santos and Nigam, 1993), and is involved in development of the spinal cord during embryogenesis (Stern et al., 1990). Recently, we showed that HGF also activates and promotes the phenotypic transition between monocytes and macrophages (Galimi et al., 1993), and that it stimulates the growth and differentiation of erythroid hematopoietic precursors (Galimi et al., 1994).

II. HGF Receptor A. Encoding by the c-met Oncogene The expression of the c-met proto-oncogene in the liver, in addition to several other epithelial organs, prompted us to study the possible identity between the Met protein (p190 Met) and the HGF receptor (HGFr). A gastric carcinoma cell line, GTL-16, in which the receptor is overexpressed (Giordano et al., 1989b; Ponzetto et al., 1991), was used as source of the Met protein. Cross-linking experiments showed a specific interaction between the HGF c~ chain and one of the two subunits (IB) of p190 Met. Binding experiments showed a high affinity interaction between the two molecules, with a calculated dissociation constant (KA) of approximately 0.3 nM (NaP dini et al., 1991c; Higuchi and Nakamura, 1991; Zarnegar et al., 1990; Lokker et al., 1992). Another binding site was identified, with low affinityhigh capacity binding properties (K a) = 3 nM) (Naldini et al., 1991c), which corresponds to membrane sulfoglycolipids and/or heparin-sulfate proteoglycans associated to the extracellular matrix (Naldini et al., 1991c; Kobayashi et al., 1994). Formal proof of the identity between the transmembrane tyrosine kinase encoded by the m e t oncogene and the high affinity HGFr was obtained by transfer of the binding activity to insect cells (Spodoptera frugiperda) infected with a recombinant baculovirus carrying human m e t cDNA (Naldini et al., 1991c). Moreover, HGF is capable of stimulating autophosphorylation in tyrosine of p190 Met (Rubin et al., 1991; Bottaro et al., 1991; Naldini et al., 1991b, 1991c).


54

1. Structure a n d Biosynthesis

The c-met proto-oncogene encodes a protein with tyrosine kinase activity (Park et al., 1987). This molecule is a dimer of two subunits, an extracellular a chain (50 kDa) and a transmembrane 13chain (145 kDa) (Giordano et al., 1989a). The 13chain contains an extracellular, a transmembrane, and an intracellular domain endowed with the catalytic function. The a and 13 chains are encoded by the same gene and linked by disulfide bonds. This was proved by transfecting the cloned human M e t cDNA into simian cells; the entire exogenous p]90 Met was expressed and the two-chain structure correctly processed (Giordano et al., 1993). The HGFr (pl90 Met) is synthesized as a single chain precursor of 170 kDa, cleaved and N-glycosylated to generate the mature 190-kDa oLI3heterodimer (Giordano et al., 1989a). The cleavage site (K303RKKR-S308) amino acid numbers are derived from the sequence of the major m e t transcript, (EMBL/GenBank Accession number: X54559, Ponzetto et al., 1991) is a canonical consensus for the endoplasmic reticulum protease furin (Barr, 1991). Experiments of site-directed mutagenesis confirmed that the critical residues for cleavage are the two arginines in position 304 and 307 (Mark et al., 1992). The HGFr is the prototype of a distinct subfamily of heterodimeric tyrosine kinases, including the putative receptors c-Ron and c-Sea (Figure 2) (Ronsin et al., 1993; Huff et al., 1993). These receptors share significant sequence similarities including the extracellular cleavage site between oLand [3 chains and the location of the cysteine residues. In the cytoplasmic domain, the three family members are highly homologous in the kinase region and share an eleven-amino acid motif containing two tyrosines that behaves as a multifunctional docking site for SH2-transducers (Ponzetto et al., 1994). B. Post-Translational Modifications

Two different forms of HGFr devoid of kinase activity were identified using monoclonal antibodies against extracellular epitopes of the Met protein (Prat et al., 1991a). The molecular weight of these truncated forms are 140 and 130 kDa. The first, p140 Met, is composed of the canonical c~ chain linked to an 85 kDa 13 chain lacking most of the cyotplasmic domain. The second, p130 Met, is a soluble protein released from the cell; it differs from the p140 Met by lacking the transmembrane domain. NIH3T3 fibroblasts transfected with the full-length human M e t cDNA express the full-size p190 Met and both p140 Met and P130 Met. This indicates that post-translational events, and not alternative mRNA splicing, are responsible for generation of the truncated receptor forms (Giordano et al., 1993). The generation of truncated receptors was investigated in detail

3.

Structure and Functions o f the HGF Receptor (c-Met)

55

Figure 2 The HGF receptor family. Schematic representation of the heterodimeric (or-J3) structure of the HGF receptor (c-Met), compared with the structure of the putative receptors encoded by the proto-oncogenes c-ron and c-sea. The cross-hatched boxes represent cysteinerich domains and the solid boxes the tyrosine kinase (TK) domains. The major tyrosine phosphorylation sites (yn) are indicated.

(Crepaldi et al., 1994a). In the endoplasmic reticulum, a fraction of the single-chain p170 Met precursor is cleaved at the cytosolic side, generating a second precursor of 120 kDa. A further proteolytic event, occurring in the trans-Golgi network, converts both precursors into the mature heterodimers, p190 Met and p140 Met, respectively. The soluble p130 Met is generated by proteolytic cleavage at the cell surface (Prat et al., 1991a). Under physiological conditions the p170 Met precursor is inactive. Conditions that cause its accumulation in the endoplasmic reticulum (e.g., overexpression) lead to activation of the tyrosine kinase. Consequently, an accumulation of p120 Met is observed. Overproduction of p120 Met does not occur in cells overexpressing a transfected kinase-dead receptor mutant. Proteolytic cleavage of the cytoplasmic domain of p170 Met may thus represent a safety mechanism aimed at preventing the effect of ligand-independent intracellular activation of the receptor kinase (Crepaldi et al., 1994a).

C. Positive and Negative Regulation The HGFr binds ATP to Lys 111~ which is located in the catalytic pocket of the tyrosine kinase domain. Mutation of this residue leads to a kinase-

56


defective receptor (Crepaldi et al., 1994a). Autophosphorylation upregulates the tyrosine kinase activity of the p190 Met receptor (Naldini et al., 1991a). An increase in substrate phosphorylation rate is observed due to a higher Vmax of the phosphotransfer reaction catalyzed by the enzyme. The K m for ATP is not affected while a slight increase in the Km for the peptide substrate is observed. The major autophosphorylation sites of the HGFr have been mapped to y1234 and y1235 (Figure 3) (Ferracini et al., 1991; Longati et al., 1993). This tyrosine doublet is present at homologous locations in the proteins encoded by ron and sea, two putative receptors belonging to the M e t gene family (Ronsin et al., 1993; Huff et al., 1993). The doublet is also conserved and phosphorylated in other tyrosine kinases, including the insulin receptor (Tornqvist et al., 1987, 1988). Mutant receptors have been constructed by substitution of y1234 and/or y1235 with phenylalanine (Longati et al., 1993). These studies showed that single mutations of either y1234 or y1235 generate a receptor with reduced but not abolished kinase activity; the residual activity was sufficient for activation of the kinase by autophosphorylation. However, when both tyrosines were mutated together the re-

Figure 3 Positive and negative regulation of the HGF receptor tyrosine kinase (TK) activity. Inhibition of catalytic activity is observed after Ser 98s phosphorylation, induced either by protein kinase C (PKC) or by calcium-calmodulin kinase III (CAM-K-III). The latter responds to variations of the intracellular Ca + + concentration. Enhancement of catalytic activity follows autophosphorylation of Tyr 1234 and Tyr 1235. The cross-hatched box represents a cysteine rich domain and the solid box a tyrosine kinase (TK) domain.

3.


57

ceptor lost its kinase activity. These two sites are also the first tyrosines phosphorylated in response to ligand stimulation. Stable transfectants expressing the receptor with single substitutions (either y1234 or y1235) show an impaired response to HGE Phosphorylation on serine and/or threonine residues negatively modulates the tyrosine kinase activity of receptors, such as those for insulin, insulin-like growth factor I (IGF1), and epidermal growth factor (EGF) (Jacobs et al., 1983; Takayama et al., 1988; Davis and Czech, 1984). Stimulation of protein kinase C (PKC) by 12-O-tetradecanoyl-phorbal-13-acetate (TPA) is followed by serine phosphorylation and concomitant decrease in tyrosine kinase activity of the HGFr (Gandino et al., 1990). An independent pathway has been shown to have a similar effect. This negative regulatory pathway is triggered by increased intracellular Ca 2+ concentration, and is mediated by a serine kinase with the biochemical properties of Ca 2+calmodulin kinase III (CAM-K-III)(Gandino et al., 1991). Interestingly, the target residue phosphorylated in both instances is Ser 98s (Figure 3). This residue is located within the juxtamembrane domain of the HGFr, and fits into a canonical consensus sequence for phosphorylation by both PKC and CAM-K-III (Kennelly and Krebs, 1991). The inhibitory effect of either TPA or Ca +2 ionophores on the HGFr kinase activity was abolished if Ser 98s was substituted with an alanine residue by site-directed mutagenesis (Gandino et al., 1994).

D. Signal Transduction Upon ligand binding, tyrosine kinase receptors dimerize and each monomer is transphosphorylated at multiple sites (Ullrich and Schlessinger, 1991). In the HGFr, sites located within the tyrosine kinase domain (i.e., y1234 and y1235) are phosphorylated at high stoichiometry. These are responsible for regulatory functions (see above). Other sites, located outside the kinase domain, are phosphorylated at low stoichiometry and create docking sites for binding to cytoplasmic effectors. These effectors recognize phosphotyrosines via a conserved structural module, known as the Src homology-2 (SH2) domain (Cantley et al., 1991; Koch et al., 1991). Different SH2 domains bind different sequences containing a phosphotyrosine residue. In these sequences, amino acids in position + 1, +2, and + 3 (with respect to phosphotyrosine) are critical in determining the selectivity of interaction with the different SH2 domains (Fantl et al., 1992; Cantley et al., 1991; Rotin et al., 1992; Songyang et al., 1993; Waksman et al., 1992, 1993; Eck et al., 1993). The HGFr docking site for SH2-signal transducers is endowed with unconventional binding properties. Using synthetic phosphopeptides and BIAcore biosensor analysis to measure intermolecular bindings, we showed that

58


a phosphotyrosine pair in the sequence y*1349VHVNATY*1356VNV (Y* = phosphorylated tyrosine) located in the receptor C-terminal tail mediates specific association with multiple SH2-containing cytoplasmic effectors (Ponzetto et al., 1994). These include phosphatidylinositol 3-kinase (PI3K), phospholipase C-~/1 (PLC~/), pp60 c-src, the SH2 adapter, and the growth factor receptor binding protein 2/son of sevenless (GRB2/SOS) complex. All the above transducers interact directly with both phosphotyrosine y*1349VHV and y*1356VNV. GRB2, which has a strong requirement for asparagine in the +2 position (Songyang et al., 1993), binds selectively to the motif following y*13s6. Comparison of the HGFr sequence Y V H / N V with a series of binding motifs, optimal for each transducer (Songyang et al. 1993), indicates the Y V H / N V represents a degenerate site permissive for multiple SH2 domains ("SH2-supersite"). All bindings are characterized by fast association and even faster dissociation rates, indicating that the "SH2-supersite" can efficiently and rapidly exchange the bound effectors. The multispecific supersite represents a variation from the common theme of docking sites found in other tyrosine kinase receptors, where a phosphotyrosine embedded within a specific sequence interacts with one effector molecule at a time (Figure 4). In other receptors, the phosphorylation of different specific sites is likely to determine which intracellular transducer will be activated (Pawson and Schlessinger, 1993). For the HGFr, one must postulate a different model; phosphorylation of the single multifunctional site triggers a pleiotropic response involving multiple signal transducers. Modulation of this response may, however, take place by variations in the level of receptor phosphorylation affecting the binding rates of transducers with different affinities. Formal proof for the crucial role of the HGFr multifunctional docking site is the loss of function following mutation of y1349 and y1356. Substitution of both tyrosines with phenylalanine in TPR-MET, the oncogenic counterpart of the receptor (Cooper et al., 1984), abolishes its transforming ability. Single substitution of either tyrosine was less effective for the inhibition, indicating that maximal signaling efficiency of HGFr results from the cooperation between the two tyrosines (Ponzetto et al., 1994). The supersite motif Y-hydrophobic-X-hydrophobic-(X)3-Y-hydrophobic-N-hydrophobic is also conserved in the otherwise divergent C-terminal regions of the Metrelated putative receptors, Sea and Ron (Huff et al., 1993; Ronsin et al., 1993). Thus, the multifunctional docking site (in three subtle variations) represents the main transductional switch for all members of this receptor family. In response to ligand binding in vivo, the tyrosine-phosphorylated HGF receptor activates PI3K (Graziani et al., 1991; Ponzetto et al., 1993), PLC~/, SRC, (Ponzetto et al., 1994), as well as Ras, through stimulation of a guanine nucleotide exchange factor (Graziani et al., 1993). Following HGF

3.


59

Figure 4 HGF receptor docking "supersite". Signaling by tyrosine kinase receptors is mediated by selective interactions between individual SH2 domains of cytoplasmic effectors and specific phosphotyrosine residues in the activated receptor. In the HGF receptor (HGFr), phosphorylation of a multifunctional docking site made of the tandemly arranged degenerate sequence YVH/NV (residue y1349 and y1356) mediates high-affinity interactions with multiple (src homology 2 (SH2)-containing signal transducers, including phosphotyrosine phosphatase 2 (PTP2), GTPase activating protein (GAP), Nck (Nishimura et al., 1993), phosphatidylinositol-3-OH kinase (PI3K), phospholipase C-~/1 (PLC~/), Src, and the growth factor receptor binding protein 2/son of sevenless (GRB2/SOS) complex. These transducers bind to individual docking sites on the epidermal growth factor receptor (EGFr) tail or are scattered along the entire cytoplasmic domain of the platelet derived growth factor receptor (PDGFr). The crosshatched boxes represent cysteine rich domains and the solid boxes the tyrosine kinase (TK) domains.

stimulation by association to the receptor, a tyrosine phosphatase is also activated (Villa-Moruzzi et al., 1993). E. Tissue Distribution and Subcellular Localization The H G F r is mainly expressed in tissues of epithelial origin. In h u m a n s , high levels of expression are found in the liver and in epithelial cells of the gastrointestinal tract. Significant levels of expression are also found in the kidney, ovary, and endometrium. Low a m o u n t s of H G F r are detectable in lung epithelium, in the prostate, seminal vesicles, m a m m a r y gland, and in keratinocytes and melanocytes (Di Renzo et al., 1991; Prat et al., 1991b; Natali et al., 1993; M a t s u m o t o et al., 1991a,b; H a l a b a n et al., 1992). In


60

these cells mRNA and protein levels are correlated. In the thyroid follicular epithelium, which expresses low levels of mRNA, no protein is detectable. This suggests regulation of HGFr expression at both the translational and post-translational level (Di Renzo et al., 1991, 1992b). The HGFr is also expressed by oval cells, considered to be the hepatic stem cell compartment (see Chapter 5), and the receptor level increases during liver regeneration after partial hepatectomy (Hu et al., 1993). The HGFr expression in the brain is restricted to a specific cell subset, the microglia, a macrophagederived cell population (Di Renzo et al., 1992a). The HGFr is expressed in some nonepithelial cells, such as endothelial cells (Bussolino et al., 1992), monocytes (Galimi et al., 1993), and hemopoietic precursors (Galimi et al., 1994). The subcellular distribution of HGFrs in polarized epithelial cells has been studied using a variety of techniques. The reported apical distribution of the receptor, observed by immunohistochemistry (Tsarfay et al., 1992), was unexpected as the receptor should have access to HGF that is present in the blood (Nakamura et al., 1986; Zarnegar and Michalopoulos, 1989) and is stored in the pericellular matrix (Matsumoto and Yamamoto, 1991; Naldini et al., 1992). In contrast, studies using domain selective surface biotinylation and immunoprecipitation on cultured polarized kidney epithelial cells showed that HGFrs are selectively exposed at the basolateral surface; the receptors were colocalized with the basolateral marker molecule E-cadherin around the cell-cell contact zone (Crepaldi et al., 1994b). An extensive analysis of human organ frozen sections, using monoclonal antibodies against different epitopes, also showed a consistent localization of the receptors at the basolateral surface and not at the apical surface of epithelia lining the lumen of organs (Crepaldi et al., 1994b).

III. Regulation of

c-met

Expression

The oncogene encoding the HGFr, c-met, is an inducible gene and its expression is controlled at the transcriptional level. The c-met regulatory sequences have been identified and cloned (Gambarotta et al., 1994). The minimal promoter region is included within the first 300 bp upstream to the start site, it lacks TATAA consensus and CCAAT-boxes, and is characterized by GC rich boxes. The promoter contains sequences for binding the transcriptional factors, AP2 and PEA3; these motifs function as PKC responsive elements (Imagawa et al., 1987; Xin et al., 1992). Accordingly, the c-met promoter region responds to the phorbol ester, TPA, in vitro (Gambarotta et al., 1994), and TPA treatment in vivo results in a strong upregulation of the receptor expression, at both the mRNA and protein levels (Boccaccio et al., 1994).

3.


61

HGFr levels are regulated in the different phases of growth. In tissue culture, confluent growth-arrested epithelial cells express low amounts of p190 Met. Recruitment into the cell cycle by stimulation with fresh serum, however, results in a significant increase of both mRNA and protein. Interestingly, HGF can also transiently induce expression of its own receptor, in a time- and dose-dependent manner. The appearance of specific c-met mRNA follows c-fos and c-jun induction. It is reduced by the protein synthesis inhibitor, puromycin, but is increased by cycloheximide alone (superinduction) (Boccaccio et al., 1994). Thus, in the response to serum or growth factors, c-met behaves as a delayed-early response gene (see Chapter 4). Using this autoamplification mechanism the ligand seems able to enhance its intracellular signal.

IV. Role of HGF in Tissue Regeneration and Embryogenesis Long before molecular cloning, HGF was known to stimulate hepatocyte growth in physiological as well as pathological conditions. The HGF mRNA markedly increases in nonparenchymal liver cells of rodents treated with carbon tetrachloride or other hepatotoxic compounds within the first 10 hr (Kinoshita et al., 1989). Consequently, the HGF biological activity increases in the plasma, with a peak at 30 hr (Lindroos et al., 1991). Other agents causing liver damage, such as ischemia, mechanical crush (Hamanoue et al., 1992), and fulminant hepatitis (Tsubouchi et al., 1989), cause HGF mRNA increases with similar fast kinetics. Circulating HGF also increases in the plasma of patients with liver cirrhosis (Shimizu et al., 1991). HGF is primarily synthesized in the nonparenchymal liver cells, like Kupffer cells, sinusoidal endothelia, and Ito cells (Kinoshita et al., 1989; Noji et al., 1990; Ramadori et al., 1992); thus, a paracrine mechanism seems to be responsible for hepatocyte growth following liver injuries. If liver regeneration is induced by partial hepatectomy, and not by direct damage of hepatocytes, the peak of liver HGF mRNA expression is observed at 24 hr. In contrast, the plasma level of HGF maximizes within 2 hr. This indicates that hepatocytes respond to exogenous HGF produced and released into the circulation by other distal noninjured organs (see Chapter 2) (Kinoshita et al., 1991). There are thus two possible mechanisms of HGF action: one paracrine and one endocrine. In this respect, one group reported the identification of a protein, named injurin, as the possible inducer of HGF mRNA synthesis in distant organs after partial hepatectomy (Matsumoto et al., 1992). Evidence that HGF acts as a potent hepatotropic factor in vivo was obtained in animals injected intravenously with the factor (Ishiki et al.,

62


1992). In these animals, HGF induced hepatocyte DNA synthesis and protected the liver from subsequent injuries. HGF also strongly stimulates DNA synthesis of rabbit kidney tubular epithelial cells in tissue culture (Igawa et al., 1991). Unilateral nephrectomy in experimental animals, or treatment with CC14, markedly increases HGF mRNA in the contralateral kidney. In situ hybridization shows the appearance of HGF mRNA in endothelial cells of the organ (Nagaike et al., 1991). These data suggest that HGF also functions in a paracrine manner as a renotrophic factor after kidney injury. In proliferating hepatocytes or renal tubular cells the specific binding of HGF to its receptor is greatly inhibited. Scatchard analysis showed a decrease in receptor number without changes in affinity, suggesting that HGFr downmodulation occurs during liver and kidney regeneration (Higuchi and Nakamura, 1991; Nagaike et al., 1991). The mitogenic and motogenic activity of HGF, associated with its ability to stimulate the invasion of extracellular matrices and the organization of tubular structures, strongly suggest that the factor may behave as a potent morphogen during embryonal development. Experimental administration of exogenous HGF in chicken embryos affects the development of the spinal cord (Stern et al., 1990). In situ hybridization studies in mouse embryos also demonstrate the nonrandom distribution of HGF and its receptor in developing organs. Epithelial cells express the HGFr, while mesenchymal cells in close vicinity express HGF, suggesting a paracrine relationship (Sonnemberg et al., 1993). Signals of mesenchymal origin are known to govern differentiation and morphogenesis of many epithelial organs. HGF may be one of the long-sought factors mediating these paracrine interactions.

V. R o l e of c - m e t in

Carcinogenesis

The HGFr gene, c-met, was originally identified as an oncogene activated in an osteosarcoma cell line treated with a chemical carcinogen (Cooper et al., 1984). In this cell line, a genomic rearrangement generated a fusion protein containing the aminoterminal sequence of a gene called tpr and the truncated carboxyterminal sequence of c-met, encoding the tyrosine kinase domain (Park et al., 1986). The tpromet rearrangement was also found in human xeroderma pigmentosum (XP) cells after treatment with a chemical carcinogen (Michelin et al., 1993), and at low frequency in naturally occurring human tumors (Soman et al., 1990, 1991). The tpr-met encoded tyrosine kinase is constitutively active and is transforming. Deletion of the tpr moiety yields a protein with reduced transforming potential, indicating a direct role for this region in the mechanism of activation. Two leucine zipper motifs within the tpr region were demonstrated to mediate dimerization of the Tpr-Met receptor; constitutive dimerization allows transphosphoryla-

3.


63

tion of the kinase domains and contributes to oncogenic activation (Rodriguez and Park, 1993). Although tpr enhances the transforming activity of the m e t kinase domain, the simple truncation of the extracellular and the transmembrane portion is sufficient to generate an active oncogenic protein. Recent findings show that a short sequence in the juxtamembrane domain is critical for constitutive activation of the truncated met. This sequence, 36 amino acids long, could contain dimerization motifs, as in tpr, or previously unrecognized sequences critical for transformation (Zhen et al., 1994). Transfection of full-length human m e t c-DNA in mouse or rat fibroblasts does not lead to malignant transformation (Giordano et al., 1993; Zhen et al., 1994). In contrast, transfection of human m e t together with human HGF cDNA is transforming; an autocrine loop has been suggested (Rong et al., 1992). In a gastric human carcinoma, the m e t proto-oncogene is amplified and overexpressed and the tyrosine kinase is constitutively activated (Giordano et al., 1989a,b). Because no modification in the sequence has been found in the derived cell line (Ponzetto et al., 1991), constitutive activation of the Met kinase may be initiated by simple overexpression. Overexpression has indeed been found in spontaneously arising human cancers. In a significant fraction of gastrointestinal tract carcinomas, the amount of m e t m R N A and protein are increased (Di Renzo et al., 1991). In thyroid tumors, overexpression of m e t is observed in carcinomas derived from the follicular epithelium and correlates with an aggressive phenotype (Di Renzo et al., 1992b).

Acknowledgments The experimental work reviewed in this paper is the result of the collaborative effort of our coworkers E Di Renzo, G. Gaudino, C. Ponzetto, M. Prat, T. Crepaldi, S. Giordano, A. Graziani, L. Naldini, C. Boccaccio, G. Gambarotta, and P. Longati, who are gratefully acknowledged. The manuscript was written with the excellent assistance of A. Cignetto and E. Wright. Our work is supported by grants from the Associazione Italiana Ricerche Cancro (AIRC) and from the Italian National Research Council (CNR), progetto finalizzato ACRO.

References Barr, P. J. (1991). Mammalian subtilisins: The long-sought dibasic processing endoproteases. Cell (Cambridge, Mass.) 66, 1-3. Boccaccio, C., Gaudino, G., Gambarotta, G., Galimi, E, and Comoglio, P. M. (1994). Hepatocyte growth factor receptor expression is inducible and is part of the delayed-early response to HGE J. Biol. Chem. 269, 12846-12851. Bottaro, D. P., Rubin, J. S., Faletto, D. L., Chan, A. M. L., Kmiecick, T. E., Vande Woude, G. E,

64


and Aaronson, S. A. (1991). Identification of the hepatocyte growth factor receptor as the c-met proto-oncogene. Science 251, 802-804. Bussolino, E, Di Renzo, M. E, Ziche, M., Bocchietto, E., Olivero, M., Naldini, L., Gaudino, G., Tamagnone, L., Coffer, A., Marchisio, P. C., and Comoglio, P. M. (1992). Hepatocyte growth factor is a potent angiogenic factor which stimulates endothelial cell motility and growth. J. Cell Biol. 119, 629-641. Cantley, L. C., Auger, K., Carpenter, C., Duckworth, B., Graziani, A., Kapeller, R., and Soltoff, S. (1991). Oncogenes and signal transduction. Cell (Cambridge, Mass.) 64, 281-302. Cooper, C. S., Park, M., Blair, D., Tainsky, M. A., Huebner, K., Croce, C. M., and Vande Woude, G. E (1984). Molecular cloning of a new transforming gene from a chemically transformed human cell line. Nature (London) 311, 29-33. Crepaldi, T., Prat, M., Giordano, S., and Comoglio, P. M. (1994a). Generation of a truncated hepatocyte growth factor receptor in the endoplasmic reticulum. J. Biol. Chem. 269, 17501755. Crepaldi, T., Pollack, A. L., Prat, M., Zborek, A., Mostov, K., and Comoglio, P. M. (1994b). Targeting of the SF/HGF receptor to the basolateral domain of polarized epithelial cells. J. Cell Biol. 125, 312-320. Davis, R. J., and Czech, M. P. (1984). Tumor-promoting phorbol diesters mediate phosphorylation of the epidermal growth factor receptor. J. Biol. Chem. 259, 8545-8549. Di Renzo, M. E, Narsimhan, R. P., Olivero, M., Bretti, S., Giordano, S., Medico, E., Gaglia, P., Zara, P., and Comoglio, P. M. (1991). Expression of the met/HGF receptor in normal and neoplastic human tissues. Oncogene 6, 1997-2003. Di Renzo, M. E, Bertolotto, A., Olivero, M., Putzolu, P., Crepaldi, T., Schiffer, D., Pagni, C. A., and Comoglio, P. M. (1992a). Selective expression of the met/HGF receptor in human central nervous system microglia. Oncogene 8, 219-222. Di Renzo, M. E, Olivero, M., Ferro, S., Prat, M., Bongarzone, I., Pilotti, S., Belfiore, A., Costantino, A., Vignari, R., Pierotti, M. A., and Comoglio, P. M. (1992b). Overexpression of the c-met/HGF receptor gene in human thyroid carcinomas. Oncogene 7, 2549-2553. Eck, M. J., Shoelson, S. E., and Harrison, S. C. (1993). Recognition of a high affinity phosphotyrosyl peptide by the scr homology 2 domain of p56 lck. Nature (London) 362, 87-91. Fantl, W. J., Escobedo, J. A., Martin, G. A., Turck, C. W., Rosario, M., McCormick, E, and Williams, L. T. (1992). Distinct phosphotyrosines on a growth factor receptor bind to specific molecules that mediate different signaling pathways. Cell (Cambridge, Mass.) 69, 413-423. Ferracini, R., Longati, P., Naldini, L., Vigna, E., and Comoglio, P. M. (1991). Identification of the major autophosphorylation site of the met/hepatocyte growth factor receptor tyrosine kinase. J. Biol. Chem. 266, 19558-19564. Galimi, E, Brizzi, M. E, and Comoglio, P. M. (1993). The hepatocyte growth factor and its receptor. Stem Cells 11, 22-30. Galimi, E, Bagnara, G. P., Bonsi, L., Cottone, E., Follenzi, A., Simeone, A., and Comoglio, P. M. (1994). Hepatocyte growth factor induces proliferation and differentiation of multipotent and erythroid hemopoietic progenitors. J. Cell Biol. 127, 1743-1754. Gambarotta, G., Pistoi, S., Giordano, S., Comoglio, P. M., and Santoro, C. (1994). Structure and inducible regulation of the human met promoter. J. Biol. Chem. 269, 12852-12857. Gandino, L., Di Renzo, M. E, Giordano, S., Bussolino, E, and Comoglio, P. M. (1990). Protein kinase-C activation inhibits tyrosine phosphorylation of the c-met protein. Oncogene, 5, 721-725. Gandino, L., Munaron, L., Naldini, L., Magni, M., and Comoglio, P. M. (1991). Intracellular calcium regulates the tyrosine kinase receptor encoded by the met oncogene. J. Biol. Chem. 266, 16098-16104. Gandino, L., Longati, P., Medico, E., Prat, M., and Comoglio, P. M. (1994). Phosphorylation

3.


65

of serine 985 negatively regulates the hepatocyte growth factor receptor kinase. J. Biol. Chem. 269, 1815-1820. Gherardi, G., Gray, J., Stoker, M., Perryman, M., and Furlong, R. (1989). Purification of scatter factor, a fibroblast-derived basic protein that modulates epithelial interaction and movement. Proc. Natl. Acad. Sci. U.S.A. 86, 5844-5848. Giordano, S., Di Renzo, M. E, Narsimhan, R., Cooper, C. S., Rosa, C., and Comoglio, P. M. (1989a). Biosynthesis of the protein encoded by the c-met proto-oncogene. Oncogene 4, 1383-1388. Giordano, S., Ponzetto, C., Di Renzo, M. E, Cooper, C. S., and Comoglio, P. M. (1989b). Tyrosine kinase receptor indistinguishable from the c-MET protein. Nature (London) 339, 155-156.

Giordano, S., Zhen, Z., Medico, E., Gaudino, G., Galimi, E, and Comoglio, P. M. (1993). Transfer of motogenic and invasive response to scatter factor/hepatocyte growth factor by transfection of human met proto-oncogene. Proc. Natl. Acad. Sci. U.S.A. 90, 649-653. Gohda, E., Tsubouchi, H., Nakayama, H., Hirono, S., Sakiyama, O., Takahashi, K., Miyazaki, H., Hashimoto, S., and Daikuhara, Y. (1988). Purification and partial characterization of hepatocyte growth factor from plasma of a patient with hepatic failure. J. Clin. Invest. 81, 414-419. Grant, D. S., Kleinman, H. K., Goldberg, I. D., Bhargava, M., Nickoloff, B. J., Kinsella, J. L., Polverini, P. J., and Rosen, E. M. (1993). Scatter factor induces blood vessel formation in vivo. Proc. Natl. Acad. Sci. U.S.A. 90, 1937-1941. Graziani, A., Gramaglia, D., Cantley, L. C., and Comoglio, P. M. (1991). The tyrosine phosphorylated hepatocyte growth factor/scatter factor receptor associates with phosphatidylinositol 3-kinase. J. Biol. Chem. 266, 22087-22090. Graziani, A. Gramaglia, D., Dalla Zonca, P., and Comoglio, P. M. (1993). Hepatocyte growth factor/scatter factor stimulates the ras-guanine nucleotide exchanger. J. Biol. Chem. 268, 9165-9168.

Halaban, R., Rubin, J., Funasaka, Y., Cobb, C., Boulton, T., Faletto, D., Rosen, E., Chan, A., Yoko, K., White, W., Cook, C., and Moellmann, G. (1992). Met and hepatocyte growth factor/scatter factor signal transduction in normal melanocytes and melanoma cells. Oncogene 7, 2195-2206.

Hamanoue, M., Kawaida, K., Takao, S., Shimazu, H., Noji, S., Matsumoto, K., and Nakamura, T. (1992). Rapid and marked induction of hepatocyte growth factor during liver regeneration after ischemic or crush injury. Hepatology (Baltimore) 16, 1485-1492. Han, S., Stuart, L. A., and Friezner Degen, S. J. (1991). Characterization of the DNF15S2 locus on human chromosome 3: Identification of a gene coding for four kringle domains with homology to hepatocyte growth factor. Biochemistry 30, 9768-9780. Hartmann, G., Naldini, L., Weidner, M., Sachs, M., Vigna, E., Comoglio, P., and Birchmeier, W. (1992). A functional domain in the heavy chain of scatter factor/hepatocyte growth factor binds the c-met receptor, induces cell dissociation but not mitogenesis. Proc. Natl. Acad. Sci. U.S.A. 89, 11574-11578. Higashio, K., Shima, N., Goto, M., Itagaki, Y., Nagao, M., Yasuda, H., and Morinaga, T. (1990). Identity of a tumor cytotoxic factor from human fibroblasts and hepatocyte growth factor. Biochem. Biophys. Res. Commun. 170, 397-404. Higuchi, O., and Nakamura, T. (1991). Identification and change in the receptor for hepatocyte growth factor in rat liver after partial hepatectomy or induced hepatitis. Biochem. Biophys. Res. Commun. 176, 599-607. Hu, Z., Evatrs, R. P., Fujio, K., Marsden, E. R., and Thorgeirsson, S. (1993). Expression of hepatocyte growth factor and c-met gene during hepatic differentiation and liver development in the rat. Am. J. Pathol. 142, 1823-1830. Huff, J. I., Jelinek, A. M., Borgman, C. A., Lansing, T. J., and Parsons, J. T. (1993). The proto-

66


oncogene c-sea encodes a transmembrane protein-tyrosine kinase related to the met/hepatocyte growth factor/scatter factor receptor. Proc. Natl. Acad. Sci. U.S.A. 90, 6140-6144. Igawa, T., Kanda, S., Kanetake, H., Saitoh, Y., Ichihara, A., Tomita, Y., and Nakamura, T. (1991). Hepatocyte growth factor is a potent mitogen for cultured rabbit renal tubular epithelial cells. Biochem. Biophys. Res. Commun. 174, 831-838. Imagawa, M., Chiu, R., and Karin, M. (1987). Transcription factor AP-2 mediates induction by two different signal transduction pathways: Protein kinase C and cAMP. Cell (Cambridge, Mass.) 51, 251-260. Ishiki, Y., Ohnishi, H., Muto, Y., Matsumoto, K., and Nakamura, T. (1992). Direct evidence that hepatocyte growth factor is a hepatotrophic factor for liver regeneration and for potent anti-hepatitis action in vitro. Hepatology (Baltimore) 16, 1227-1235. Jacobs, S., Shayoun, N. E., Saltiel, A. R., and Cuatrecasas, P. (1983). Phorbol esters stimulate the phosphorylation of receptors for insulin and somatomedin C. Proc. Natl. Acad. Sci. U.S.A. 80, 6211-6213. Kan, M., Zhang, G. H., Zarnegar, R., Michalopoulos, G., Myoken, Y., Mckeehan, W. L., and Stevens, J. L. (1991). Hepatocyte growth factor/hepatopoietin A stimulates the growth of rat kidney proximal tubule epithelial cells (RPTE), rat nonparenchymal liver cells, human melanoma cells, mouse keratinocytes and stimulates anchorage-independent growth of SV40-transformed RPTE. Biochem. Biophys. Res. Commun. 174, 331-337. Kennelly, P. J., and Krebs, E. G. (1991). Consensus sequences as substrate specificity determinants for protein kinases and protein phosphatases. J. Biol. Chem. 266, 15555-15558. Kinoshita, T., Tashiro, K., and Nakamura, T. (1989). Marked increase of HGF and mRNA in nonparenchymal liver cells of rat treated with hepatotoxin. Biochem. Biophys. Res. Commun. 165, 1229-1234. Kinoshita, T., Hirao, S., Matsumoto, K., and Nakamura, T. (1991 ). Possible endocrine control by hepatocyte growth factor of liver regeneration after partial hepatectomy. Biochem. Biophys. Res. Commun. 177, 330-335. Kobayashi, T., Honke, K., Miyazaki, T., Matsumoto, K., Nakamura, T., Ishizuka, I., and Makita, A. (1994). Hepatocyte growth factor specifically binds to sulfoglycolipids. J. Biol. Chem. 269, 9817-9821. Koch, C. A., Anderson, D., Moran, M. E, Ellis, C., and Pawson, T. (1991). SH2 and SH3 domains: Elements that control interactions of cytoplasmic signaling proteins. Science 252, 668-674. Lindroos, P. M., Zarnegar, R., and Michalopoulos, G. K. (1991). Hepatocyte growth factor (hepatopoietin A) rapidly increases in plasma prior to DNA synthesis and liver regeneration stimulated by partial hepatectomy and C C I 4 administration. Hepatology (Baltimore) 13, 743-750. Lokker, N. A., Mark, M. R., Luis, E. A., Bannet, G. L., Robbins, K. A., Baker, J. B., and Godowski, P. J. (1992). Structure-function analysis of hepatocyte growth factor: Identification of variants that lack mitogenic activity yet retain high affinity receptor binding. EMBO J. 11, 2503-2510. Longati, P., Bardelli, A., Ponzetto, C., Naldini, L., and Comoglio, P. M. (1993). Tyr o s i n e s 1234-1235 a r e critical for activation of the tyrosine kinase encoded by the met protooncogene (HGF receptor). Oncogene 9, 49-57. Mark, M. R., Lokker, N. A., Zioncheck, T. E, Luis, E. A., and Godowski, P. J. (1992). Expression and characterization of hepatocyte growth factor receptor-IgG fusion proteins. J. Biol. Chem. 36, 26166-26171. Matsumoto, A., and Yamamoto, N. (1991). Sequestration of a hepatocyte growth factor in extracellular matrix in normal adult liver. Biochem. Biophys. Res. Commun. 174, 90-95. Matsumoto, K., Tajima, H., and Nakamura, T. (1991a). Hepatocyte growth factor is a potent stimulator of human melanocytes DNA synthesis and growth. Biochem. Biophys. Res. Commun. 176, 45-51.

3.


67

Matsumoto, K., Hashimoto, K., Yoshikawa, K., and Nakamura, T. (1991b). Marked stimulation of growth and motility of human keratinocytes by hepatocyte growth factor. Exp. Cell Res. 196, 114-120. Matsumoto, K., Tajima, H., Hamanoue, M., Kohno, S., Kinoshita, T., and Nakamura, T. (1992). Identification and characterization of "injurin", an inducer of the gene expression of hepatocyte growth factor. Proc. Natl. Acad. Sci. U.S.A. 89, 3800-3804. Michelin, S., Daja-Grosjean, L., Sureau, F., Said, S., Sarasin, A., and Suarez, H. G. (1993). Characterization of a c-met proto-oncogene activated in human xeroderma pigmentosum cells after treatment with N-methyl N-nitro N-nitrosoguanidina (MNNG). Oncogene, 8, 1983-1991. Miyazawa, K., Tsubouchi, H., Naka, D., Takahashi, K., Okigaki, M., Arakaki, N., Nakayama, H., Hirono, S., Sakiyama, O., Takahashi, K., Gohda, E., Daikuhara, Y., and Kitamura, N. (1989). Molecular cloning and sequence analysis of cDNA for human hepatocyte growth factor. Biochem. Biophys. Res. Commun. 163, 967-973. Miyazawa, K., Shimomura, T., Kitamura, A., Kondo, J., Morimoto, Y., and Kitamura, N. (1993). Molecular cloning and sequence analysis of the c-DNA for a human serine protease responsible for activation of hepatocyte growth factor. J. Biol. Chem. 268, 1002410028. Mizuno, K., Takehara, T., and Nakamura, T. (1992). Proteolytic activation of a single-chain precursor of hepatocyte growth factor by extracellular serine-protease. Biochem. Biophys. Res. Commun. 189, 1631-1638. Montesano, R., Matsumoto, K., Nakamura, T., and Orci, L. (1991). Identification of a fibroblast-derived epithelial morphogen as hepatocyte growth factor. Cell (Cambridge, Mass.) 7, 901-908. Nagaike, M., Hirao, S., Tajima, H., Noji, S., Taniguchi, S., Matsumoto, K., and Nakamura, T. (1991). Renotropic functions of hepatocyte growth factor in renal regeneration after unilateral nephrectomy. J. Biol. Chem. 266, 22781-22784. Naka, D., Ishii, T., Yoshiyama, Y., Miyazawa, K., Hara, H., Hishida, T., and Kitamura, N. (1992). Activation of hepatocyte growth factor by proteolytic conversion of a single chain form to a heterodimer. J. Biol. Chem. 267, 20114-20119. Nakamura, T., Teramoto, H., and Ichihara, A. (1986). Purification and characterization of a growth factor from rat platelets for mature parenchymal hepatocytes in primary cultures. Proc. Natl. Acad. Sci. U.S.A. 86, 6489-6493. Nakamura, T., Nishizawa, T., Hagiya, M., Seki, T., Shimonishi, M., Sugimura, A., Tashiro, K., and Shimizu, S. (1989). Molecular cloning and expression of human hepatocyte growth factor. Nature (London) 342, 440-443. Naldini, L., Vigna, E., Ferracini, R., Longati, P., Gandino, L., Prat, M., and Comoglio, P. M. (1991a). The tyrosine kinase encoded by the met proto-oncogene is activated by autophosphorylation. Mol. Cell. Biol. 11, 1793-1803. Naldini, L., Vigna, E., Narshiman, R. P., Gaudino, G., Zarnegar, R., Michalopoulos, G., and Comoglio, P. M. (1991b). Hepatocyte growth factor (HGF) stimulates the tyrosine kinase activity of the receptor encoded bythe proto-oncogene c-met. Oncogene, 6, 501-504. Naldini, L., Weidner, M., Vigna, E., Gaudino, G., Bardelli, A., Ponzetto, C., Narsimhan, R., Hartmann, G., Zarnegar, R., Michalopoulos, G., Birchmeier, W., and Comoglio, P. M. (1991c). Scatter factor and hepatocyte growth factor are indistinguishable ligands for the met receptor. EMBO J. 10, 2867-2878. Naldini, L., Tamagnone, L., Vigna, E., Sachs, M., Hartmann, G., Birchmeier, W., Daikuhara, Y., Tsubouchi, H., Blasi, E, and Comoglio, P. M. (1992). Extracellular proteolytic cleavage by urokinase is required for activation of hepatocyte growth factor/scatter factor EMBO J. 11, 4825-4833. Natali, ~P. G., Nicotra, M. R., Di Renzo, M. E, Prat, M., Bigotti, A., Cavaliere, R., and Comoglio, P. M. (1993). Expression of the c-met/HGF receptor in human melanocytic

68


neoplasms: Demonstration of the relationship to malignant melanoma tumour. Br. J. Cancer 68, 746-750. Nishimura, R., Li, W., Kashishian, A., Mondino, A., Zhou, M., Cooper, J., and Schlessinger, J. (1993). Two signaling molecules share a phosphotyrosine-containing binding site in the platelet-derived growth factor receptor. Mol. Cell. Biol. 13, 6889-6896. Noji, S., Tashiro, K., Koyama, E., Nohno, T., Osyama, K., Taniguchi, S., and Nakamura, T. (1990). Expression of the hepatocyte growth factor gene in endothelial and Kupffer cells of damaged rat liver, as revealed by in situ hybridization. Biochem. Biophys. Res. Commun. 173, 42-47. Park, M., Dean, M., Cooper, C. S., Schmidt, M., O'Brian, S. J., Blair, D. G., and Vande Woude, G. E (1986). Mechanism of met oncogene activation. Cell. (Cambridge, Mass.) 45, 895904. Park, M., Dean, M., Kaul, K., Braun, M. J., Gonda, M. A., and Vande Woude, G. (1987). Sequence of met proto-oncogene cDNA has features characteristic of the tyrosine kinase family of growth-factor receptors. Proc. Natl. Acad. Sci. U.S.A. 84, 6379-6383. Pawson, T., and Schlessinger, J. (1993). SH2 and SH3 domains. Curr. Biol. 3, 434-442. Ponzetto, C., Giordano, S., Peverali, F., Della Valle, G., Abate, M., Vaula, G., and Comoglio, P. M. (1991). c-met is amplified but not mutated in a cell line with an activated MET tyrosine kinase. Oncogene 6, 553-559. Ponzetto, C., Bardelli, A., Maina, F., Longati, P., Panayotou, G., Dhand, R., Waterfield, M. D., and Comoglio, P. M. (1993). A novel recognition motif for phosphatiylinositol 3-kinase binding mediates its association with the hepatocyte growth factor/scatter factor receptor. Mol. Cell. Biol. 13, 4600-4608. Ponzetto, C., Bardelli, A., Zhen, Z., Maina, F., Dalla Zonca, P., Giordano, S., Graziani, A., Panayotou, G., and Comoglio, P. M. (1994). A multifunctional docking site mediates signalling and transformation by hepatocyte growth factor/scatter factor (HGF/SF). receptor family. Cell (Cambridge, Mass.) 77, 261-271. Prat, M., Crepaldi, T., Gandino, L., Giordano, S., Longati, P., and Comoglio, P. M. (1991a). C-terminal truncated forms of met, the hepatocyte growth factor receptor Mol. Cell. Biol. 11, 5954-5962. Prat, M. P., Narsimhan, R. P., Crepaldi, T., Nicotra, M. R., Natali, P. G., and Comoglio, P. M. (1991b). The receptor encoded by the human c-met oncogene is expressed in hepatocytes, epithelial cells and solid tumors. Int. J. Cancer 49, 323-328. Ramadori, G., Neubauer, K., Odenthal, M., Nakamura, T., Knittel, T., Schwogler, K. H., and zum Buschenfelde, M. (1992). The gene of hepatocyte growth factor is expressed in fatstoring cells of rat liver and is downregulated during cell growth and by transforming growth factor-beta. Biochem. Biophys. Res. Commun. 183, 739-742. Rodriguez, G. A., and Park, M. (1993). Dimerization mediated through a leucine zipper activates the oncogenic potential of the MET receptor tyrosine kinase. Mol. Cell. Biol. 13, 6711-6722. Rong, S., Bodescot, M., Blair, D., Dunn, J., Nakamura, T., Mizuno, K., Park, M., Chan, A., Aaronson, S., and Vande Woude, G. F. (1992). Tumorigenicity of the met proto-oncogene and the gene for hepatocyte growth factor. Mol. Cell. Biol. 12, 5152-5158. Ronsin, C., Muscatelli, F., Mattei, M. G., and Breathnach, R. (1993). A novel putative receptor tyrosine kinase of the met family. Oncogene 8, 1195-1202. Rotin, D., Margolis, B., Mohammadi, M., Daly, R. J., Daum, G., Li, N., Fischer, E. H., Burgess, W. H., Ullrich, A., and Schlessinger, J. (1992). SH2 domain prevent tyrosine dephosphorylation of the EGF receptor: Identification of tyr 992 as the high affinity binding site for SH2 domains of phospholipase C gamma. EMBO J. 11,559-567. Rubin, J. S., Chan, A. M. L., Bottaro, D. P., Burgess, W. H., Taylor, W. G. Cech, A. C., Hirschfield, D. W., Wong, J., Miki, T., Finch, P. W., and Aaronson, S. A. (1991). A broad-

3.


69

spectrum human lung fibroblast-derived mitogen is a variant of hepatocyte growth factor. Proc. Natl. Acad. Sci. U.S.A. 88, 415-419. Santos, O. E P., and Nigam, S. K. (1993). HGF-induced tubulogenesis and branching of epithelal cells is modulated by extracellular matrix and TGF-beta. Dev. Biol. 160, 293302. Shimizu, I., Ichihara, A., and Nakamura, T. (1991). Hepatocyte growth factor in ascites from patients with cirrhosis. J. Biochem. (Tokyo) 109, 14-18. Shimomura, T., Kondo, J., Ochiai, M., Naka, D., Miyazawa, K., Morimoto, Y., and Kitamura, N. (1993). Activation of the zymogen of hepatocyte growth factor activator by thrombin. J. Biol. Chem. 268, 22927-22932. Shiota, G., Rhoads, D. B., Wang, T. C., Nakamura, T., and Schmidt, E. V. (1992a). Hepatocyte growth factor inhibits growth of hepatocellular carcinoma cells. Proc. Natl. Acad. Sci. U.S.A. 89, 373-377. Skeel, A., Yoshimura, T., Showalter, S. D., Tanaka, A., Appella, E., and Leonard, E. J. (1991). Macrophage stimulating protein: Purification, partial aminoacid sequence, and cellular activity. J. Exp. Med. 173, 1227-1234. Soman, N. R., Wogan, G. N., and Rhim, J. S. (1990). Tpromet oncogenic rearrangement: Detection by polymerase chain reaction amplification of the transcript and expression in human tumor cell lines. Proc. Natl. Acad. Sci. U.S.A. 87, 738-742. Soman, N. R., Correa, P., Ruiz, B. A., and Wogan, G. N. (1991). The tpromet oncogenic rearrangement is present and expressed in human gastric carcinoma and precursor lesions. Proc. Natl. Acad. Sci. U.S.A. 88, 4829-4896. Songyang, Z., Shoelson, S. E., Chaudhuri, M., Gish, G., Pawson, T., Haser, W. G., King, E, Roberts, T., Ratnofsky, S., Lechleider, R. J., Neel, B. G., Birge, R. B., Fajardo, J. E., Chou, M. M., Hanafusa, H., Schaffhausen, B., and Cantley, L. C. (1993). SH2 domains recognize specific phosphopeptide sequences. Cell (Cambridge, Mass.) 72, 1-20. Sonnemberg, E., Meyer, D., Weidner, M. K., and Birchmeier, C. (1993). Scatter factor/hepatocyte growth factor and its receptor the c-met tyrosine kinase can mediate a signal exchange between mesenchyme and epithelia during mouse embryogenesis. J. Cell Biol. 123, 223235. Stern, C. D., Ireland, G. W., Herrick, S. E., Gherardi, E. Gray, J., Perryman, M., and Stoker, M. (1990). Epithelial scatter factor and development of the chick embryonic axis. Development (Cambridge, UK) 110, 1271-1284. Stoker, M., Gherardi, E., Perryman, M., and Gray, J. (1987). Scatter factor is a fibroblastderived modulator of epithelial cell mobility. Nature (London) 327, 239-242. Tajima, H., Matsumoto, K , and Nakamura, T. (1991). Hepatocyte growth factor has a potent anti-proliferative activity for various tumor cell lines. FEBS Lett. 291,229-232. Takayama, S., White, M. F., and Kahn, C. R. (1988). Phorbol ester-induced serine phosphorylation of the insulin receptor decreases its tyrosine kinase activity. J. Biol. Chem. 263, 3440-3447. Tornqvist, H. E., Pierce, M. W., Frackleton, A. R., Nemenoff, R. A., and Avruch, J. (1987). Identification of insulin receptor tyrosine residues autophosphorylated in vitro. J. Biol. Chem. 262, 10212-10219. Tornqvist, H. E., Gunsalus, J. R., Nemenoff, R. A., Frackleton, A. R., Pierce, M. W., and Avruch, J. (1988). Identification of the insulin receptor tyrosine residues undergoing insulin-stimulated phosphorylation in intact rat hepatoma cells. J. Biol. Chem. 263, 350359. Tsarfay, I., Resau, J. H., Rulong, S., Keydar, I., Faletto, D. L., and Vande Woude, G. E (1992). The met proto-oncogene receptor and lumen formation. Science 257, 1258-1261. Tsubouchi, H., Hirono, S., Godha, E., Nakayama, H., Takahashi, K., Sakiyama, O., Miyazaki, H., Sugihara, J., Tomita, E., Muto, Y., Daikuhara, Y., and Hashimoto, S. (1989). Clinical

70


significance of human hepatocyte growth factor in blood from patients with fulminant hepatic failure. Hepatology (Baltimore) 9, 875-881, Ullrich, A., and Schlessinger, J. (1991). Signal transduction by receptors with tyrosine kinase activity. Cell (Cambridge, Mass.) 61,203-212. Villa-Moruzzi, E., Lapi, S., Prat, M., Gaudino, G., and Comoglio, R M. (1993). A protein tyrosine phosphatase activity associated with the hepatocyte growth factor/scatter factor receptor. J. Biol. Chem. 268, 18176-18180. Waksman, G., Kominos, D., Robertson, S. R., Pant, N., Baltimore, D., Birge, R. B., Cowburn, D., Hanafusa, H., Mayer, B. J., Overduin, M., Resh, M. D., Rios, C. B., Silverman, L., and Kuriyan, J. (1992). Crystal structure of the phosphotyrosine recognition domain SH2 of v-src complexed with tyrosine-phosphorylated peptides. Nature (London), 358, 646-653. Waksman, G., Shoelson, S., Pant, N., Cowburn, D., and Kuriyan, J. (1993). Binding of a high affinity phosphotyrosil peptide in the src SH2 domain: Crystal structures of the complexed and peptide-free forms. Cell (Cambridge, Mass.) 72, 779-790. Weidner, K. M., Behrens, J., Vandekerckove, J., and Birchmeier, W. (1990). Scatter factor: Molecular characteristics and effect on the invasiveness of epithelial cells. J. Cell Biol. 111, 2097-2108. Weidner, K. M., Arakaki, N., Vandekerchove, J., Weingart, S., Hartmann, G., Rieder, H., Fonatsch, C., Tsubouchi, H., Hishida, T., Daikuhara, Y., and Birchmeier, W. (1991). Evidence for the identity of human scatter factor and hepatocyte growth factor. Proc. Natl. Acad. Sci. U.S.A. 88, 7001-7005. Xin, J. H., Cowie, A., Lachance, R, and Hassel, J. A. (1992). Molecular cloning and characterization of PEA3, a new member of the ETS oncogene family that is differentially expressed in mouse embryonic cells. Gene Dev. 6, 481-496. Yoshimura, T., Yuhki, N., Wang, M.-H., Skeel, A., and Leonard, E. J. (1993). Cloning, sequencing, and expression of human macrophage stimulating protein (MSP, MST1) confirms MSP as a member of the family of the kringle proteins and locates the MSP gene on chromosome 3. J. Biol. Chem. 268, 15461-15468. Zarnegar, R., and Michalopoulos, G. (1989). Purification and biological characterization of human hepatopoietin A, a polypeptide growth factor for hepatocytes. Cancer Res. 49, 3314-3320. Zarnegar, R., De Frances, M. C., Oliver, L., and Michalopoulos (1990). Identification and partial characterization of receptor binding sites of HGF on rat hepatocytes. Biochem. Bioph. Res. Commun. 173, 1179-1185. Zhen, Z. Giordano, S., Longati, R, Medico, E., Campiglio, M., and Comoglio, R M. (1994). Structural and functional domains critical for constitutive activation of the HGF receptor (MET) Oncogene 9, 1691-1697.

4 Expression and Function of Growth-Induced Genes during Liver Regeneration Rebecca Taub Department of Genetics and Medicine Howard Hughes Medical Institute University of Pennsylvania, School of Medicine Philadelphia, Pennsylvania 19104

I. Liver Regeneration: The Important Questions Although scientists have been studying liver regeneration for many years, until recently there has been little research directed toward understanding its molecular basis. Ultimately, scientists would like to understand the mechanisms that (1) trigger regeneration, (2) allow the liver to concurrently grow and maintain differentiated function, and (3) terminate cell proliferation once the liver has reached the appropriate mass. These questions are fundamentally important to help us understand the mechanisms of cell growth and differentiation. Furthermore, they are directly important in understanding some aspects of clinical liver disease since the liver is required to regenerate following liver transplantation and hepatic damage resulting from hepatitis virus infection and exposure to chemical toxins. In addressing these questions, it is important to first realize some basic facts about cell cycle kinetics in the regenerating liver (see Chapters I and 2) (Fausto and Mead, 1989; Michalopoulos, 1990; Fausto, 1994). Within minutes after a partial hepatectomy, the majority of the cells in the remnant liver undergo a transition from the quiescent or Go state into the G1 phase of the cell cycle (Figure 1). The signals that mediate this transition are not clearly understood, but a number of different growth factors and other signaling Liver Regeneration and Carcinogenesis Copyright 9 1995 by Academic Press, Inc. All rights of reproduction in any form reserved.

71

72

Rebecca Taub

FigureI Modelof the cell cycle during liver regeneration followingpartial hepatectomyshowing the timing of induction of immediate-early and delayed-early genes.

events have been implicated in this process. Following rapid intracellular signal transduction in hepatic cells, preexisting transcription factors are modified resulting in their activation. These activated transcription factors are then responsible for initiating the transcription of primary or immediateearly response genes within minutes after the partial hepatectomy in a protein synthesis independent manner. As discussed below, immediate-early genes represent diverse functional classes and include transcription factors, growth factors, signal transduction regulators, and other types of proteins (A1mendral et al., 1988; Herschman, 1991; Lau and Nathans, 1985, 1987; Mohn et al., 1991a; Zipfel et al., 1989). Immediate-early genes encode proteins that regulate later phases in G 1 including the induction of the delayed-early response genes. Delayed-early response genes are induced within a few hours of the partial hepatectomy, but their transcription also requires protein synthesis.

II. Immediate-Early Gene Expression in Hepatic Cells To begin to understand the complex interplay between the onset of cell growth and the maintenance of differentiated function during liver regeneration, one must understand the molecular bases of the events that occur during the hepatic cell cycle. A subset of the primary or immediate-early response genes and delayed-early response genes induced in the regenerating liver are likely to be central players in regulating this process. Early studies in regenerating liver showed that immediate-early genes like c-fos and c - m y c that are induced in mitogen-treated cells, are also rapidly induced in regen-

4.

Expression and Function of Growth-Induced Genes during Liver Regeneration

73

erating liver (Goyette et al., 1984; Kruijer et al., 1986; Thompson et al., 1986). Initially, our laboratory specifically examined immediate-early gene expression in H35 cells treated with insulin (Taub et al., 1987). H35 cells represent a minimal deviation cell line that becomes quiescent under serumdeprived conditions and rapidly reenters the cell cycle in response to physiologic concentrations of insulin. We demonstrated that c-fos, c-myc, and ~3-actin are induced as immediate-early genes following insulin treatment of serum-deprived H35 cells (Mohn et al., 1990; Taub, et al., 1987). Thus, H35 cells represent an in vitro liver system with some properties of the regenerating liver in which the growth environment could be precisely defined. To further examine the spectrum of immediate-early gene induction in insulin-treated H35 cells and regenerating liver, we performed differential cDNA cloning studies in regenerating liver and insulin-treated H35 cells. We examined more than 1000 differentially expressed clones and found 52 nonoverlapping sequences that were induced in hepatic cells. Of these, 41 were novel genes or genes not previously known to be induced as immediateearly genes (Mohn et al., 1991a). We then extended these studies further by examining a large number of other known and novel immediate-early genes that were isolated by differential cDNA cloning of mitogen-treated fibroblasts. In total we identified more than 70 genes induced as immediateearly genes in regenerating liver and insulin-treated H35 cells. Of these, a small subset was induced as delayed-early genes in regenerating liver and as immediate-early genes in H35 cells. Table 1 presents an updated list of the immediate- and delayed-early genes expressed in the regenerating liver that have been examined in our laboratory. In the past few years, the sequence of a number of these novel genes has been determined in our laboratory or coincidentally in other laboratories. In addition, a number of immediateearly genes induced in regenerating liver has been identified in other labs including plasminogen activator inhibitor (PAI) (Schneiderman et al., 1993; Thornton et al., 1994), metallothionein (Tohyama et al., 1993), and adherens junction proteins (Gliick et al., 1992). Following the initial identification, members of this group of novel growthincluded genes that fell into interesting functional categories were studied in detail. These include novel transcription factors, proteins involved in signal transduction, potential growth factors, liver-specific immediate-early genes, and delayed-early genes (Figure 2). The fact that immediate-early genes represent diverse functional classes is not surprising, because cell growth requires simultaneous activation of multiple different intracellular pathways. The detailed study of any one particular gene may ultimately explain how that protein functions during liver regeneration and other types of cellular growth. In this way, proteins that have central roles in controlling liver growth may be identified.

Table 1 Gene Expression during Liver Regeneration

Pattern Growth regulated

Induction

Gene

Tissue; Hep, NP

Peak exp. (h)

Pattern

Induction

Peak 2 (h)

Gene

Tissue; Hep, NP

ATP synthase B2 microglobulin IGFl ubiquitin

Mult Mult Mult; Hep Mult

0-216 0-216 0-216 0-216

eck ClEBP alpha CL-34 CL-58

Mult Mult; Hep Liver Liver

48-216 60-216 2-216 2-216

c-fos IGFBP-1 Ikpa pip92( CL-14) MKP-1(RL-30) LRF-1 IP-10 CL-6 CL-73 CL-142 G6Pase(RL- 1) PC3(RL-98)

Mult Liver Mult Mult Mult Mult Mult; Hep Mult; Hep Mult; Hep Liver Liver Mult

1 2 1 2 1 2 0.5 3 2 2 0.5 2

Constitutive

IE

beta actin Gene 33 fibronectin junB c-jun JE RL-9 (viral env.protein) albumin PRL-1 (SL-314) RNR-1 (SL-332) egr-1 PEPCK KC c-myc c-ets C/EBPbeta CL-8 p68 RNA helicase(CL-36) CL-97 CL-180 dCBP(CL-183) CL-211 RL-27 RL-53 SL-339 SL-371

Mult Mult Mult Mult Mult Mult Liver; Hep Liver Mult; NP Liver Mult Liver; Hep Mult; Hep Mult Mult; Hep Mult Mult Mult Mult; Hep Liver Mult Mult Mult Mult Mult Mult

L

2 L

1 2 2 6 3 1

Max expression after growth phase

L

L

2 2 I

2 2 2 2 2 nd 1 nd 2 1 nd

Cell-cycle regulated

IE

48 36 48 48 48 48 42 48 48 36 60 48

SnkKinase(B28) Thrombospondinl(B10) Gly96 (M45) elF-2B (S48) HLH462 (S68) MyD 118 (Tr5) Tis7/PC4 (TT40) FK506BP(CL-87) nur77 DE

HRS (CL-4) MHC class 1 tropomyosin alpha FNR beta FNR pre-rnRNABP(CL-20) CL-22 HIBBA18(CL-31) CL-38 CL-39 CL-61 CL-120 FBRNP(CL- 141) CL-182 RL-104

Mult Mult Mult Mult Mult Mult

Mult

nd nd nd nd nd nd nd

nd Mult

nd

Mult Mult; Hep Mult Mult Mult Mult Mult Mult Mult; NP Liver Mult Mult Mult Mult Liver

6 8 24 24 24 24 24 24 24 8 8 24 24 24 24

C

DE S phase

SL-353

Mult

1

48

gGCS(CL-131)

Liver

1

24

Histone H 3

Mult

16-24

Undetectable

EGF EGF receptor IGFl receptor egr-2 alpha-fetoprotein

Pattern of expression, type of induction (IE, immediate-early; DE, delayed-early), tissue expression (Hep, hepatocytes; NP, nonparenchymal; Mult, multiple tissues), and peak expression in hours (Peak exp.) is given for each gene. References for most genes are provided in references of this chapter, and all papers are from Dr. Taub's laboratory except for the following: p68 RNA helicase (Lemaire and Heinlein, 1993); d cBP (Ito et al., 1993); SnkKinase (Simmons et al., 1992);M45, S48, S68, TT5, TT40, original clone numbers from Almendral et al. (1988);HLH462 (Christy et al., 1991);Tis7 (Varnum et al., 1989); FK506BP (Nelson et al., 1991); pre-mRNABP (Adams et al., 1993); HIBBAl8 (Adams et al., 1993); FBRNP (Takiguchi et a/., 1993); MKP-1 (Sun et al., 1993); G6Pase (Shelly et al., 1993); PC3 (Bradbury eta!., 1991);gGCS (Huang et d., 1993); CL-6 is also identical to a recently identified gene RESP-18 (Bloomquist et al., 1994).

76

Rebecca Taub

Figure 2 Categoriesof function of immediate-early and delayed-earlygenes.

III. Modification of Preexisting Transcription Factors Immediately Following Partial Hepatectomy Turns on Immediate-Early Genes As immediate-early genes are induced in a protein synthesis independent fashion, their transcription must be activated by transcription factors that are preexisting in hepatic cells. Little is presently known about what preexisting transcription factors are activated by signal transduction pathways in mitogen-stimulated cells. However, in analyzing the c-fos promoter, it was noted that a serum-response element was important in mediating c-los gene

4.


77

transcription following mitogen stimulation of 3T3 fibroblasts. Similar elements have been noted in the promoter regions of several other immediateearly genes, and it is now known that serum response factor (SRF) and a cofactor ets-like protein (ELK-l) which bind to the serum response element are activated by phosphorylation following mitogen stimulation (Marais et al., 1993). These transcription factors act in a coordinated fashion to activate c-fos gene transcription. It is likely that members of the SRF family are also activated during liver regeneration, but they have never been studied. Our laboratory also found that members of the Rel transcription factor family that preexist in hepatic cells are activated within minutes post partial hepatectomy. Clues to this activation were provided by the identification of one of the immediate-early genes expressed in the regenerating liver. This gene encodes inhibitor of kappa Bc~ IKB0t, a specific inhibitor of p65/RelA and other Rel family members (Grilli et al., 1993; Tewari et al., 1992). In examining the expression of Rel family members during liver regeneration, we discovered a post hepatectomy factor (PHF) that binds to a nuclear factor-kappa B (NF-KB) site and is induced within minutes after partial hepatectomy in the remnant liver in a protein synthesis independent manner; PHF is not induced by sham surgery. This is one of the most rapid responses that has been measured in the regenerating liver. Since its isolation, increased PHF activity has been used as the first indicator of signal transmission following growth factor addition in isolated hepatocytes and in liver perfusion studies. Further characterization of PHF indicates that it contains NF-KB subunits, but has some differences from standard p65-p50 NF-KB (Haber et al., 1995). Moreover, unlike the NF-KB DNA binding activity that is induced for extended times in many cells following cytokine or mitogen treatment, high molecular weight PHF complexes rapidly disappear 1 hr post partial hepatectomy and only lower molecular weight complexes persist. This occurs through physiologic nuclear proteolysis involving proteasomes. The target genes of PHF during liver regeneration are not known. However, genes such as IKB~ and K C which are immediate-early genes in regenerating liver and are known to be regulated by NF-KB subunits are likely to be induced by PHE Interestingly, NF-KB has been shown to interact with the CAAT enhancer binding protein (C/EBP) and other transcription factors to modulate the activity of certain promoters (Stein and Baldwin, 1993). As both active C/EBP and NF-KB subunits are found in the regenerating liver, a growth-induced factor (i.e., PHF) and a constitutive factor (i.e., C/EBP) could act in a coordinate fashion to regulate target genes in the regenerating liver. Another preexisting transcription factor that is activated by growth factor or interferon treatment of cells is signal transducer and activator of transcription I (SIF/Statl), which binds to the serum inducible element first identified in the c-fos promoter (Ruff-Jamison et al., 1993). This transcrip-

78

Rebecca Taub

tion factor family is activated by phosphorylation on a specific tyrosine residue that results in its nuclear translocation and ability to bind to DNA. Interestingly, epidermal growth factor (EGF) treatment of animals induces Stat activity in the liver. In unpublished studies, we have found that partial hepatectomy also induces Stat3 DNA binding within 30 min of partial hepatectomy (Cressman et al., 1995). Thus, it represents another early transcription factor that is activated immediately after the initial growth signal. PHF and Stat3 are activated by the earliest signaling pathways that trigger liver regeneration. Thus, they may be useful as markers to trace back to the triggering mechanism. They also are interesting as potential regulators of immediate-early growth response genes in regenerating liver.

IV. Induction Patterns of 70 Genes Following Partial Hepatectomy Define the Temporal Course of Liver Regeneration The large collection of growth-induced genes provides a useful tool for defining the temporal course of liver regeneration at the level of gene expression. Using dot blot technology, we denatured and immobilized more than 100 different cDNAs corresponding to immediate-early, delayed-early, and liver-specific genes on a nylon membrane (Haber et al., 1993). We then hybridized these filters with labeled remnant liver mRNA corresponding to different times post partial hepatectomy. The signal obtained for each immobilized cDNA correlates with the level of expression for that gene at the given time point. We then were able to use Northern blots to confirm the temporal course of expression of a subset of these genes. The kinetics of gene induction during liver regeneration is shown in Figure 3. Approximately 12 hr after partial hepatectomy, S phase is entered by some hepatocytes and peak DNA synthesis occurs at 24 hr. DNA synthesis peaks significantly later in nonparenchymal cells. The major portion of the mass of the liver is reconstituted within 72 hr and the process stops completely after 7 to 10 days (Grisham, 1962). We found that many of these "early" genes are expressed for extended periods during the hepatic growth response (Haber et al., 1993). Several patterns of expression of immediateearly, delayed-early, and liver-specific genes were defined during the 9-day period post partial hepatectomy. One pattern of induction parallels the major growth period of the liver that ends at 60 to 72 hr post partial hepatectomy. A second pattern has two peaks coincident with the first and second G 1 phases of the two hepatic cell cycles. A third group which in-

4.


79

Figure 3 Representation of patterns of regulated gene expression during liver regeneration. The pattern of gene expression is indicated for growth-regulated genes (e.g., f3-actin), cell cycle regulated genes (e.g., IGFBP-1), and genes with maximal expression after the major growth phase (e.g., C/EBPe~). Hours after partial hepatectomy and patterns of DNA synthesis in hepatocytes (H) and nonparenchymal cells (NP) and the regrowth of liver mass are indicated. cludes liver-specific genes such as C/EBPcx shows maximal expression after the growth period. Although the peak in DNA synthesis in nonparenchymal cells occurs 24 hr later than in hepatocytes, most of the genes studies demonstrate similar induction in both cell types. This finding suggests that the G0/G 1 transition occurs simultaneously in all cells in the liver, but that the G1 phase of nonparenchymal cells may be relatively prolonged. These studies define the temporal boundary between proliferation and return to quiescence in the post partial hepatectomized liver. For regeneration to be precisely carried out, multiple hepatic cell types must proliferate in a coordinated fashion, and it is likely that cell-cell interactions and paracrine effects of growth factors are involved. Therefore, it is interesting to compare the immediate-early gene response in the two cell types. We found that the majority of immediate-early genes induced in the regenerating liver are expressed in a cell type independent fashion, but

80

Rebecca Taub

approximately one-third show cell type restricted expression and are induced primarily in hepatocytes.

V. Transcription Factors Induced in the Regenerating Liver Members of many different families of transcription factors are induced as immediate-early genes in several different cell types. Transcription factors induced as immediate-early genes in the regenerating liver include members of the large Jun-Fos leucine-zipper family, the Rel family, the nuclear receptor family, the helix-loop-helix family, and the zinc finger family. Immediateearly genes in the regenerating liver and insulin-treated H35 cells include many members of the transcription factor families that are induced differently in mitogen-treated fibroblasts (Mohn et al., 1990). For instance, in mitogen-treated fibroblasts the zinc finger genes, egr-1 and egr-2, are induced, but in H35 cells and regenerating liver, egr-1 is induced but not egr-2. Likewise, in the large Jun-Fos family of transcription factors, c-fos and all of the jun family members are induced in the two liver systems. However, many of the fos family members (e.g., fra-1, fosB, fra-2) that are immediateearly genes in mitogen-treated fibroblasts are not induced in the two liver systems. Instead, liver regeneration factor-1 (LRF-1), another member of this family that was identified in our laboratory, is highly induced in the two experimental liver systems we have used (discussed below) (Hsu et al., 1991, 1992). These findings imply that similar transcription factors responsible for transactivating a specific subset of immediate-early genes may be present in the two liver systems, and that a similar intracellular milieu is responsible for immediate-early gene induction in both liver systems. A. The LRF-1/JunB Story Members of the Jun and Fos families of transcription factors are thought to have a role in activating the transcription of delayed-early genes expressed subsequently in the growth response. In regenerating liver and insulintreated H35 cells, LRF-1, junB, c-jun and c-los among jun/fos/LRF-1 family members are induced post partial hepatectomy. We first identified LRF-1 through differential screening of a regenerating-liver cDNA library (HSU et al., 1991). We found that LRF-1 is a rapidly and highly induced novel gene encoding a 21-kDa leucine-zipper protein. LRF-1 is also highly induced following insulin-treatment of H35 cells, and is induced at a lower level in mitogen-treated fibroblasts. As such, it may be a more important regulator of hepatic than nonhepatic growth.

4. Expression and Function of Growth-Induced Genes during Liver Regeneration

81

LRF-1 has no homology with other leucine-zipper proteins outside the basic leucine-zipper domains. LRF-1 alone can bind DNA, but it preferentially forms heteromeric complexes with c-Jun and JunB and does not interact with c-Fos. In solution, it binds with highest affinity to cAMP response elements but also has affinity for related sites like activator protein-1 (AP-1) elements (i.e., phorbol ester response elements). Early cotransfection studies showed that LRF-1 in combination with c-Jun strongly activates an AP-1 element-containing promoter (Hsu et al., 1991). The induction of the LRF-1 gene in regenerating liver and insulin-treated H35 cells greatly increases the potential variety of heterodimeric combinations of leucine-zipper transcription factors. While LRF-1 mRNA is rapidly induced in the absence of protein synthesis, its peak induction is later than that of c-fos mRNA, suggesting that LRF-1 may regulate responsive genes at a later point in the cell cycle. To further explore the coordinate temporal expression of LRF-1, JunB, c-Jun and c-Fos, we examined the levels of these proteins in hepatic cells following partial hepatectomy or insulin treatment (Hsu et al., 1992). We found that the peak levels of c-Fos and c-Jun occur early in G1, within 1 hr of the stimulus and that the peak levels of JunB and LRF-1 occur later in G 1. Using immunoprecipitations with specific antisera, we found that in insulintreated H35 cells, high levels of c-Fos/c-Jun, c-Fos/JunB, LRF-1/c-Jun, and LRF-1/JunB complexes are present for several hours after the G0/G 1 transition. The relative level of LRF-1/JunB complex increases during G 1. We found dramatic differences in promoter-specific activation by LRF-1 and c-Fos containing complexes. LRF-1 in combination with either Jun protein strongly activates a cyclic AMP response element-containing promoter which c-Fos/Jun does not activate. LRF-1/c-Jun, c-Fos/c-Jun, and c-Fos/JunB activate specific AP-1 and activator of transcription factor (ATF) site-containing promoters. In contrast, LRF-1/JunB potently represses c-Fos and c-Junmediated activation of these promoters (Figure 4). Repression is dependent on a region in LRF-1 that includes amino acids just proximal to the basic domain and the DNA binding domain. We extended these studies to examine the activity of JunB which interacts with LRF-1 to mediate inhibition of AP-1 sites (Hsu et al., 1993). We identified separate regions of JunB required for cAMP element mediated transactivation and AP-1/ATF site-mediated repression. Deletion analysis showed that the region involved in transactivation function is highly conserved among all Jun family members, and corresponds to activator domain (A1) of c-Jun. In contrast, repression is maximal in the presence of both the DNA binding domain and a region proximal to the basic region that is highly divergent among Jun proteins. The regions required for repression in LRF-1 and JunB are positioned similarly just upstream of the DNA binding domain. It is

82

Rebecca Taub

Figure 4 Model of the actions of LRF-1/JunB heterodimers in regulation of target genes controlled by AP-1 and cAMP (CRE) promoter elements.

likely that the configuration of the LRF-1/JunB heterodimer varies depending on whether it is interacting with cAMP or AP-1 response elements, thereby determining if the heterodimer will be an activator or repressor. Functional distinctions between Jun proteins during the growth response may be accounted for by promoter-specific activation and repression mediated by regional differences in Jun family proteins. Thus, through complex interactions among LRF-1, JunB, c-Jun, and c-Fos, control of delayed-early gene expression may be established for extended time during the G 1 phase of hepatic growth (Figure 5). As the relative level of LRF-1/JunB complexes increases post partial hepatectomy and following insulin treatment of H35 cells, the c-Fos/Jun-mediated ATF and AP-1 site activation is likely to decrease with simultaneous increased transcriptional activation of the many liver-specific genes whose promoters contain cyclic AMP response element sites. In this way the liver-specific phenotype of the liver or H35 cells can be maintained during the growth response. It is likely that Jun/Fos/LRF-1 interact with other transcription factors such as members of the steroid receptor family and constitutively expressed hepatic factors in regulating target genes during the hepatic growth response. B. RNR-1, a Novel Nuclear Receptor That Acts through the NGFI-B Half-Site

During screening of a subtracted cDNA library of immediate-early genes induced during liver regeneration, we identified a novel member of the

4.


83

Figure 5

Model of the regulation of delayed-early genes containing AP-1 and CRE enhancers by LRF-1, JunB, c-Jun, and c-Fos heterodimers during the G 1 phase of liver regeneration and insulin-induced mitogenesis in H35 cells. Various heterodimers are found at the indicated times post partial hepatectomy. Relative activation ("ON," "OFF, .... DECREASED") of delayedearly gene promoters is indicated (adapted from Hsu et al., 1993).

thyroid/steroid receptor superfamily, regenerating liver nuclear receptor-1 (RNR-1) (Scearce et al., 1993). This gene is not expressed in quiescent liver but is rapidly induced following partial hepatectomy and is specific to hepatic growth as it is not induced in other mitogen-treated cells; RNR-1 is also expressed in brain. A full-length cDNA clone of RNR-1 encodes a 66kDa 597-amino acid protein that is highly homologous to nerve growth factor inducible clone B/nuclear receptor clone 77 (r-NGFI-B/m-Nur77) particularly in the DNA binding (94%) and putative ligand binding (59%) domains. Interestingly, m-nur77 is also induced as an immediate-early gene in regenerating liver. RNR-1 specifically binds to the NGFI-B DNA half-site and forms a complex very similar in size to the m-Nur77 complex, suggesting that RNR-1 also may bind as a monomer. Consistent with this finding, the A box region important in mediating half-site binding is 100% conserved between RNR-1 and r-NGFI-B/m-Nur77. Both RNR-1 and r-NGFIB/Nur77 strongly transactivate a reporter driven by a consensus r-NGFIB/m-Nur77 binding site, and their effect together is additive. As both the RNR-1 and r-NGFI-B/m-nur77 genes are induced during liver regeneration, it is possible that RNR-1 acts concomitantly with r-NGFI-B/m-Nur77 in regulating the expression of delayed-early genes during liver regeneration. These factors are also intriguing because they can cooperate with other transcription factor families to modulate transactivation of specific promoters, and they are regulated by specific ligands, many of which have not yet been defined. Such ligands could be, however, important unidentified circulating factors that help regulate regeneration.

84

Rebecca Taub

VI. Immediate-Early Genes Involved in Signal Transduction Signal transduction pathways are rapidly activated following mitogen or cytokine addition to cells. The activity of signal transduction molecules is frequently controlled by phosphorylation. For instance, when a growth factor binds to a tyrosine kinase receptor, the tyrosine kinase receptor becomes activated and autophosphorylates itself and may phosphorylate other downstream molecules. Active tyrosine phosphorylated proteins may interact with other downstream targets via Src homology 2 (SH2) domains that interact with the tyrosine phosphate residue. Activation of downstream modulators such as Ras, Raf, mitogen-activated protein (Map) kinase, and phosphoinositol kinase then ensues. Recently, a subset of these pathways has been dissected and is briefly summarized (Figure 6) (Blenis, 1993). Tyrosine kinases and tyrosine phosphorylated substrates can also be regulated by specific tyrosine phosphatases (Brautigan, 1992). In the regenerating liver, we have identified two immediate-early genes, map kinase phosphatase (MKP-1) (Sun et al., 1993), and phosphatase of regenerating liver (PRL-1) (Diamond et al., 1994) that encode distinct tyrosine protein phosphatases.

A. PRL-1, a Member of a Novel Class of Protein-Tyrosine Phosphatases P R L - 1 is a particularly interesting immediate-early gene because it is in-

duced in mitogen-stimulated cells and regenerating liver, but is constitutively expressed in insulin-treated rat H35 hepatoma cells which otherwise show normal regulation of immediate-early genes (Diamond et al., 1994). Sequence analysis revealed that P R L - 1 encodes a novel 20-kDa protein that contains the eight amino acid consensus protein-tyrosine phosphatase (PTPase) active site. PRL-1 is able to dephosphorylate phosphotyrosine substrates, and mutation of the cysteine residue in the active site abolishes this activity. Because P R L - 1 has no homology to other PTPases outside the active site, it is a member of a new class of PTPases. PRL-1 migrates as a 21kDa protein, and is located primarily in the triton-insoluble fraction of the cell nucleus. Stably transfected cells which overexpress PRL-1 demonstrate altered cellular growth and morphology, and a transformed phenotype. Thus far, we have not identified specific intracellular targets for the PRL-1 phosphatase. Tyrosine phosphatases, particularly those of the unique classes such as PRL-1, MKP-1, and the cyclin, cdc25, may have very specific intracellular targets (Figure 6). Several different sites have been identified in

4. Expression and Function of Growth-Induced Genes during Liver Regeneration

85

Figure6 Model of signal transduction pathway and potential sites of interaction of phosphatases (PTPases).

the signal transduction pathway where tyrosine phosphorylation is important. For instance, cdc25 is a regulator of cdc2 that dephosphorylates cdc2 on a specific tyrosine and threonine residue resulting in its activation (Dunphy and Kumagai, 1991). The cyclin, cdc2, and related molecules bind to specific cyclins, and in the active state are important in cell cycle transition between G1/S and G2/M (see Chapter 8). Thus, cdc25 is important in mediating transition through G1/S and G2/M. Likewise, MKP-1 appears to specifically dephosphorylate Map kinase and is important in turning off the initial signal transduction cascade. Because MKP-1 is highly induced in regenerating liver with a peak in expression at 30 min post partial hepatectomy, it is likely that this pathway is important in the regenerating liver. PRL-1 acts at a distinct site in this pathway, or in as yet unidentified pathways. Like MKP-1, it appears that PRL-1 is important in normal cellular growth control during mitogenesis. The specific role of PRL-1 in hepatic growth is implied by the very high level of its expression in liver regeneration and development, and in hepatoma cells. Thus, it is likely that PRL-1 plays a central role in regulating growth during liver regeneration and the development of some hepatic malignancies.

86

Rebecca Taub

VII. Immediate-Early Genes That Are Secreted Proteins Several identified immediate-early genes encode secreted proteins. Some of these are known cytokines or growth factors and others are secreted proteins of unknown function. The importance of such proteins in modulating the growth response is that they may be acting as paracrine or autocrine factors in directly stimulating cell growth. Immediate-early genes identified in regenerating liver encode the intercrines, JE, KC, (clone names-PDGF inducible), and interferon inducible protein (IP-10). Their function in liver regeneration is not known but interestingly, expression of both KC and IP-10 is restricted to hepatocytes (Haber et al., 1993). Potentially, they could be important in recruitment of other cells to the regenerating liver or in growth stimulation of neighboring cells. Of the secreted proteins induced in the regenerating liver, insulin-like growth factor binding protein-1 (IGFBP-1) is the most highly induced. IGFBPs are important modulators of the insulin-like growth factors (IGF) that may have both positive and negative effects on the ability of IGFs to stimulate cell growth. One of the most abundant liver-specific immediate-early genes is IGFBP-1 (Mohn et al., 1991b), which is rapidly induced more than 100-fold with peak expression at 1 hr post partial hepatectomy. In a different study, a more modest elevation of IGFBP-1 gene expression was seen after partial hepatectomy in fasted rats (Ghahary et al., 1992). IGFBP-1 is similar to other IGFBPs which have been shown to have important roles in modulating the actions of IGFs (Andress and Birnbaum, 1992; Brinkman et al., 1988; Conover, 1992; Elgin et al., 1987; Hsu and Olefsky, 1992). In particular, IGFBP-1 has been shown to either enhance or inhibit the mitogenic effect of IGFs in certain tissues. We have shown that the 1GFBP-1 gene is transcriptionally activated during liver regeneration, but its transcription is rapidly depressed by insulin treatment of hepatic cells (Mohn et al., 1991b). IGFBP-1 is highly expressed in the liver during fetal development, and is normally tissue restricted with high levels of expression only in the liver and kidney (Mohn et al., 1991b). During liver regeneration, there is a second peak of 1GFBP-1 expression at 36 to 60 hr that corresponds to the second round of cell division, and there appears to be expression in both parenchymal and nonparenchymal cell types (Haber et al., 1993). During liver regeneration, the level of hepatic IGF1 mRNA remains constant, and insulin-like growth factor 1 receptor (IGFlr) gene expression is virtually undetectable (Mohn et al., 1991b). However, because IGFBP-1 increases, the amount of bioactive IGF1 available to tissues may be altered. The mechanism of IGFBP-1 action is postulated to be through modulating the delivery of IGF1 to the IGFlr. In the liver, these receptors are present primarily

4.


87

on nonparenchymal cells which therefore may be the targets for IGFBP-1/ IGF1 actions during regeneration. As previously stated, IGFBP-1 may play an important role in both liver growth and metabolism. To begin to examine the regulation of this gene, we cloned and sequenced the entire mouse IGFBP-1 gene (Lee et al., 1994). Its structure is highly similar to the human gene, and in addition to the exonic regions, the two genes are highly conserved within specific regions in the promoter and first intron. Analysis of this conservation allows us to predict important regulatory sites that define the tissue specific and insulinmediated regulation of the gene, as well as to identify potential sites that might be important for the transcriptional induction during liver regeneration. The mouse gene is located on mouse chromosome 11, and is found at the boundary between regions in the mouse genome syntenic to human chromosomes 22 and 7. We have found IGFBP-1 mRNA in both parenchymal and nonparenchymal RNA following partial hepatectomy. By in situ hybridization to IGFBP-1 mRNA in regenerating rat liver tissue, IGFBP-1 transcripts were found to be present in multiple cell types. We find that IGFBP-1 gene induction following partial hepatectomy is paralleled by protein expression, and that the IGFBP-1 protein is found only in hepatocytes post partial hepatectomy. Unlike IGFBP-1 mRNA, serum levels of IGFBP-1 are elevated for a relatively short time with a peak at 2 to 3 hr post partial hepatectomy. Increased levels of IGFBP-1 could be important in modulating IGF1 effects on metabolism and growth during liver regeneration. Studies that modulate the level of IGFBP-1 in the liver during development and regeneration will be an important means of determining its function in the hepatic growth response.

VIII. Liver-Specific Immediate-Early Genes" Relationship to the Maintenance of Hepatocyte Differentiation and Metabolism During liver regeneration the liver must maintain normal glucose homeostasis despite the abrupt loss of two-thirds of its mass. As a result, immediately following partial hepatectomy, insulin levels rapidly fall and glucagon levels rise (Bucher et al., 1975; Shelly et al., 1993). Thus, it is not surprising that some liver-restricted immediate-early genes are important in maintaining metabolic homeostasis. For instance, two immediate-early genes restricted to liver and kidney encode phosphoenol pyruvate carboxykinase (PEPCK) and the recently identified glucose-6-phosphatase ( G6Pase) gene which are important in gluconeogenesis (Haber et al., 1995; Shelly et al., 1993). Likewise, IGFBP-1 has been proposed to have some role in

88

Rebecca Taub

metabolic homeostasis. Expression of all three of these genes is elevated in diabetics that lack insulin. A sign for the potential involvement of a particular gene in glucose homeostasis is indicated by distinct differences in its regulation in insulin-treated H35 cells and regenerating liver (Diamond et al., 1993; Mohn et al., 1990, 1991a,b). These differences include suppression of expression of two genes, IGFBP-1 and PEPCK by insulin and their induction in regenerating liver. When insulin levels return to normal a few hours post partial hepatectomy, IGFBP-1, G6Pase, and PEPCK expression also return to normal. Thus, immediate-early genes that are specific to the hepatic growth response may also have some role in the maintenance of hepatic function during liver regeneration. A. Identification of CL-6 Some genes were more highly induced in insulin-treated H35 cells than in the regenerating liver (Diamond et al., 1993). Some of these genes, including hepatic Arg-Ser protein (HRS), are induced as delayed-early genes in the regenerating liver (discussed below). In the differential screening analyses we performed in insulin/cycloheximide treated H35 cells, 12% of over 200 differentially expressed cDNA clones represented HRS clones and 25% represented CL-6 clones (clone name-insulin inducible). These clones were never isolated from differential or subtraction screening of regenerating liver libraries. Both mRNAs showed low level immediate-early induction in serum-treated 3T3 fibroblasts, but were not isolated from 3T3 cell cDNA libraries (Almendral et al., 1988) despite extensive differential screening analyses. Interestingly, CL-6 mRNA shows immediate-early expression in H35 cells, regenerating liver and 3T3 fibroblasts, while HRS shows immediateearly induction in H35 cells and delayed-early (protein synthesis dependent) induction in regenerating liver. CL-6 is much more highly induced in hepatic cells including the regenerating liver than it is in mitogen-treated fibroblasts. Sequence analysis of CL-6, the most abundant insulin-induced gene, resulted in the identification of a highly hydrophobic hepatic protein. CL-6 demonstrates hepatocyte and epithelial organ-specific expression. It is induced in the regenerating liver and fetal liver in the perinatal period (Haber et al., in press). CL-6 migrates as a 43-kDa protein following in vitro translation of synthetic CL-6 mRNA, and on immunoblots of H35 and regenerating liver extracts. Interestingly, although CL-6 mRNA shows strong induction in H35 cells and regenerating liver, the protein level is constitutive in H35 cells, but increases during liver regeneration. Although it is highly hydrophobic, it is not yet clear if CL-6 is a membrane-associated protein. However, because of its high level expression in hepatic tissues, CL-6 is likely to have a role in the tissuespecific aspect of cellular growth, perhaps being involved in the mainte-

4. Expressionand Function of Growth-Induced Genes during Liver Regeneration

89

nance of normal liver architecture or metabolism during regeneration and fetal development.

IX. Immediate-Early Genes in H35 Cells That Are Expressed as Delayed-Early Genes in Regenerating Liver Another group of genes shows immediate-early expression in insulin-treated H35 cells, but delayed-early expression in regenerating liver (Table 1) (Diamond et al., 1993). In general, the peak expression in insulin-treated H35 cells occurs a few hours before peak expression in regenerating liver, but later than the induction of many other immediate-early genes (Diamond et al., 1993; Mohn et al., 1990, 1991a). In regenerating liver, the earliest elevation of delayed-early expression occurs at 3 hr post partial hepatectomy, while peak expression occurs between 8 and 24 hr, consistent with late G 1 to mid S phase. Little is known about the delayed-early growth response genes, the kinetics of their induction, or their role as part of the regulatory cascade that results in cell growth (Lemaire and Heinlein, 1993). However, the finding that some cyclins are members of this class establishes the importance of delayed-early genes in cell cycle regulation (Lanahan et al., 1992; Lu et al., 1992). Cyclins are important in cell cycle events (Sherr, 1993). The timing of the induction of the delayed-early genes we have identified indicates that they are first expressed approximately 3 hr post partial hepatectomy in a protein synthesis dependent fashion; however, peak expression frequently occurs several hours later. Therefore, whatever stimuli are responsible for elevating the level of delayed-early gene mRNAs, they are present for extended times during G1 and perhaps continue into the S phase. A. Delayed-Early Genes That Encode RNA Binding Proteins Given the dramatic increase in RNA production during late G1, proteins that control RNA processing are likely to have important roles in cell cycle regulation. Thus far, three delayed-early genes that we have identified fall into this family including HRS, fetal brain ribonucleic protein (FBRNP), and pre messenger R N A binding protein (m-RNABP) (Adams et al., 1993; Diamond et al., 1993; Takiguchi et al., 1993). H R S is of particular interest because it is highly induced by insulin, shows delayed-early induction in regenerating liver, and has two different transcripts [i.e., HRS-Short Form(SF) (1.7 kb) and HRS-Long Form(LF) (3.2 kb)] that demonstrate different temporal patterns of induction. The complete sequence of HRS-SF was determined and an alignment between

90

Rebecca Taub

HRS-SF and the protein data bank demonstrated that HRS is a member of the family of regulators of alternative pre-mRNA splicing. It is most similar to Drosophila protein splicing factor, serine arginine p55 (B52/SRp55), and human splicing factor 2/alternative splicing factor (SF2/ASF), and appears to be identical to splicing factor, serine arginine p40 (SRp40) (Ge et al., 1991; Hedley and Maniatis, 1991; Krainer et al., 1991; Li and Bingham, 1991; Mattox et al., 1992; Mayeda et al., 1992; Mayeda and Krainer, 1992; Zahler et al., 1992, 1993a,b). These proteins are structurally similar with two amino RNA binding domains and a carboxyl Arg/Ser domain. These proteins have been shown to function in in vitro splicing assays, and are members of the larger class of Arg/Ser-rich proteins that contain RNA binding regions including the small splicing RNAs U1 and U2, small nuclear ribonucleoproteins (snRNPS) polyadenylation factors, and other splicing factors such as sex lethal (Sxl) and Drosophila developmental splicing factor (Tra-2) which are important in developmental regulation. These Arg/Ser domain proteins target to the speckled subnuclear compartment where mRNA processing occurs. We found a low level of HRS mRNA in most tissues except spleen and thymus in which HRS mRNA is abundant and is composed of both mRNA forms. Interestingly, unlike other tissues, spleen and thymus contain actively proliferating cells. Another interesting feature of HRS expression is the delayed appearance of a higher molecular weight RNA species (HRS-LF). Several splicing regulators have been found to regulate the production of their own spliced forms. For example, in the Drosophila su(w)a system this results in the accumulation of larger differentially processed su(w)a mRNA molecules in which the open reading frame is disrupted by the alternatively spliced exon (Hedley and Maniatis, 1991). Likewise, it seems possible that HRS-SF autoregulates HRS pre-mRNA processing resulting in the production of HRS-LF mRNA which appears later than HRS-SF mRNA following insulin treatment or partial hepatectomy. We identified several longer FIRS cDNA clones derived from HRS-LF mRNA. These cDNAs have an approximately 500 bp "insert" with stop codons in all three reading frames that disrupt the open reading frame of HRS-SF mRNA (Figure 7) (Diamond et al., 1993). This "insert" (E'2) and a smaller cDNA insert (E'I) found in the 5' untranslated region of HRS-SF cDNA may represent alternative exons. In fact, genomic cloning and sequence analyses indicate that E'2 appears to be an alternative exon because it is flanked in the genome by consensus splice acceptor and donor sequences (Figure 7). HRS has all of the features of a RNA binding protein that regulates alternative pre-mRNA splicing. In the absence of such regulators, default splicing pathways are chosen based on recognition of consensus splice site sequences by the general splicing machinery. Regulators like Tra-2, Sxl, SF2/ASF, and B52/SRp55 may function partly by binding to specific sequences in the pre-mRNA to prevent or to enhance splicing at specific

4.


91

Figure 7 Modelof hypothetical interaction of hepatic Arg-Serprotein HRS with its own premRNA. The structure of the HRS gene is shown with exons surrounding the E'2 exon alternative exon that is present in HRS-LF mRNA. Splice donor (SD) and acceptor (SA) sites are indicated. Following interaction between HRS and its pre-mRNA, alternative splicing with retention of the E'2 exon could occur resulting in HRS-LF mRNA.

positions. For instance, HRS could autoregulate the production of H R S - L F mRNA by interacting with the pre-mRNA and regulating splice site selection. The interesting question is what are the targets of HRS during hepatic growth. I would propose that one way to regulate cellular events is by activating alternative splicing of existent hepatic mRNAs. For instance, fibronectin mRNA is induced during liver regeneration, and apparently, alternatively spliced forms of fibronectin mRNA are produced (Enrich et al., 1988; Huh and Hynes, 1993). Different forms of fibronectin may be important in defining the hepatic architecture during regeneration. Potentially, fibronectin pre-mRNA could be a target for the HRS splicing factor, resulting in the production of alternatively spliced forms of fibronectin during liver regeneration. The identification of delayed-early genes involved in RNA production is consistent with previous studies that have demonstrated that rRNA, mRNA, and protein levels increase dramatically during the late phase of G1. This is compatible with one of the major functions of G 1 which is to double cell mass. The identification of several RNA processing proteins induced in the regenerating liver that are also insulin-regulated is particularly intriguing because of the involvement of insulin in stimulating mRNA, rRNA, and protein synthesis--major components of insulin-regulated growth (Hutson et al., 1987; Jefferson, 1980; Peavy et al., 1985). Because insulin has been proposed as a major regulator of growth during liver regeneration, and because insulin levels renormalize by the time RNA processing protein genes are expressed, their expression could be tied to insulin regulation of growth during liver regeneration.

X. Conclusions In this chapter, I have tried to demonstrate how the study of immediateearly and delayed-early growth response genes induced in the regenerating

92

Rebecca Taub

Figure 8 Summaryof proposed sites of cell cycle regulation by various immediate-early and delayed-early gene products during liver regeneration.

liver can lead to new insights into the central questions regarding liver regeneration. Specifically, analysis of the temporal expression of a large group of these genes defines the time course of liver regeneration and identifies the liver-specific aspects of the growth response. The study of preexisting transcription factors that are activated by partial hepatectomy will provide insights into the triggering mechanisms (Figure 8). Examination of the interplay of growth-induced and constitute transcription factors can provide insight into how the liver maintains its hepatic phenotype during regeneration. Analysis of signal transduction molecules will lead to a greater understanding of the specific regulation of growth processes and the induction of abnormal hepatic growth that occurs in malignancies. Study of liver-specific and secreted proteins induced in the primary growth response provides insight into how the liver maintains its metabolic function and architecture during regeneration. Finally, analysis of certain delayed-early genes allows insight into regulation of later phases of G 1 which may be accomplished not just by transcriptional control but by post-transcriptional regulation of the form of mRNA that are produced. Liver regeneration is a complex process that requires the interplay of many different cellular events. It is only through the analysis of each of these events on a individual basis that we may begin to dissect this complex process, and to form the larger picture of how the liver regenerates.

Acknowledgments I would like to thank the members of my laboratory past and present who were responsible for the data reviewed in this chapter including Kenneth L. Mohn, Thomas M. Laz,Jui-Chou Hsu, Dinesh S. Tewari, Manorama Tewari, Barbara A. Haber, Robert H. Diamond, L. Marie Scearce, Simon E. Chin, Linda E. Greenbaum, Drew E. Cressman, Vashti Miles, Jehyuk Lee, and Leyla Naji. Work in this chapter was in part supported by grants from the Juvenile Diabetes Foundation and NIH (DK44237).

4.


93

References Adams, M. D., Soares, M. B., Kerlavage, A. R., Fields, C., and Venter, J. C. (1993). Rapid cDNA sequencing (expressed sequence tags) from a directionally cloned human infant brain cDNA library. Nat. Genet. 4, 373-380. Almendral, J. M., Sommer, D., MacDonald-Bravo, H., Burckhardt, J., Perera, J., and Bravo, R. (1988). Complexity of the early genetic response to growth factors in mouse fibroblasts. Mol. Cell. Biol. 8, 2140-2148. Andress, D. L., and Birnbaum, R. S. (1992). Human osteoblast-derived insulin-like growth factor (IGF) binding protein-5 stimulates osteoblast mitogenesis and potentiates IGF action. J. Biol. Chem. 267 22467-22472. Blenis, J. (1993). Signal transduction via the Map kinases: Proceed at your own RSK. Proc. Natl. Acad. Sci. U.S.A. 90, 5889-5892. Bloomquist, B. T., Darlington, D. N., Mains, R. E., and Eipper, B. A. (1994). RESP18, a novel endocrine secretory protein transcript, and four other transcripts are regulated in parallel with pro-opiomelanocortin in melanotropes. J. Biol. Chem. 269, 9113-9122. Bradbury, A., Possenti, R., Shooter, E. M., and Tirone, E (1991). Molecular cloning of PC3, a putatively secreted protein whose mRNA is induced by nerve growth factor and depolarization. Proc. Natl. Acad. Sci. U.S.A. 88, 3353-3357. Brautigan, D. L. (1992). Great expectations: protein tyrosine phosphatases in cell regulation. Biochim. Biophys. Acta 1114, 63-77. Brinkman, A., Groffen, C., Kortleve, D. J., Geurts van Kessel, A., and Drop, S. L. S. (1988). Isolation and characterization of cDNA encoding the low molecular weight insulin-like growth factor binding protein (IGFBP-1). EMBO J. 7, 2417-2423. Bucher, N. L. R., and Swaffield, M. N. (1975). Regulation of hepatic regeneration in rats by synergistic action of insulin and glucagon. Proc. Natl. Acad. Sci. U.S.A. 72, 11571160. Christy, B. A., Sanders, L. K., Lau, L. E, Copeland, N. G., Jenkins, N. A., and Nathans, D. (1991). An Id-related helix-loop-helix protein encoded by a growth factor-inducible gene. Proc. Natl. Acad. Sci. U.S.A. 88, 1815-1819. Conover, C. A. (1992). Potentiation of insulin-like growth factor (IGF) action by IGF-binding protein-3: Studies of underlying mechanism. Endocrinology (Baltimore) 130, 31913199. Cressman, D. E., Diamond, R. H., and Taub, R. (1995). Rapid activation of the Stat3 transcription complex in liver regeneration. Hepatology (Baltimore) 21, 1443-1449. Diamond, R. H., Du, K., Lee, V. M., Mohn, K. L., Haber, B. A., Tewari, D. S., and Taub, R. (1993). Novel delayed-early and highly insulin-induced growth response genes. J. Biol. Chem. 268, 15185-15192. Diamond, R. H., Cressman, D. E., Laz, T. M., Abrams, C. S., and Taub, R. (1994). PRL-1 a unique nuclear protein tyrosine phosphatase, affects cell growth. Mol. Cell. Biol. 14, 37523762. Dunphy, W. G., and Kumagai, A. (1991). The cdc25 protein contains an intrinsic phosphatase activity. Cell (Cambridge, Mass.) 67, 189-196. Elgin, R. G., Busby, W. H., and Clemmons, D. R. (1987). An insulin-like growth factor (IGF) binding protein enhances the biologic response to IGF1. Proc. Natl. Acad. Sci. U.S.A. 84, 3254-3258. Enrich, C., Evans, W. H., and Gahmberg, C. G. (1988). Fibronectin isoforms in plasma membrane domains of normal and regenerating rat liver. FEBS Lett. 228, 135-138. Fausto, N. (1994). Liver Regeneration, In "The Liver: Biology and Pathobiology" (I. M. Arias, W. B. Jakoby, H. Popper, D. Schachter, and D. A. Shafritz, eds.), pp. 1059-1084. Raven, New York.

94

Rebecca Taub

Fausto, N., and Mead, M. E. (1989). Regulation of liver growth: Proto-oncogenes and transforming growth factors Lab. Invest. 60, 4-13. Ge, H., Zuo, P., and Manley, J. L. (1991). Primary structure of the human splicing factor ASF reveals similarities with Drosophila regulators. Cell (Cambridge, Mass.) 66, 373-382. Ghahary, A., Minuk, G. Y., Luo, J., Gauthier, T., and Murphy, L. J. (1992). Effects of partial hepatectomy on hepatic insulin-like growth factor binding protein-1 expression. Hepatology (Baltimore) 15, 1125-1131. Gliick, U., Fernandez, J. L. R., Pankov, R., and Ben-Ze'ev, A. (1992). Regulation of adherens junction protein expression in growth-activated 3T3 cells and in regenerating liver. Exp. Cell Res. 202, 477-486. Goyette, M., Petropoulos, C., Shanko, P., and Fausto, N. (1984). Regulated transcription of c-Ki-ras and c-myc during compensatory growth of rat liver Mol. Cell. Biol. 4, 1493-1498. Grilli, M., Chiu, J. J.-S., and Lenardo, M. J. (1993). NF-KB and Rel: Participants in a multiform transcriptional regulatory system. Int. Rev. Cytol. 143, 1-62. Grisham, J. W. (1962). A morphologic study of deoxyrobonucleic acid synthesis and cell proliferation in regenerating rat liver: Autoradiography with thymidine-3H. Cancer Res. 22, 842-849. Haber, B. A., Mohn, K. L., Diamond, R. H., and Taub, R. (1993). Induction patterns of 70 genes during nine days after hepatectomy define the temporal course of liver regeneration. J. Clin. Invest. 91, 1319-1326. Haber, B. A., Chin, S., Chuang, E., Buikhuisen, W., Naji, A., and Taub, R. (1995). High levels of glucose-6-phosphtase gene and protein expression reflect an adaptive response in proliferating liver and diabetes. J. Clin. Invest. in press. Haber, B. A., Naji, L., Cressman, D. E., and Taub, R. (1995). Coexpression of liver-specific and growth-induced genes in perinatal and regenerating liver: Attainment and maintenance of the differentiated state during cell growth. Hepatology (Baltimore), in press. Hedley, M. L., and Maniatis, T. (1991). Sex-specific splicing and polyadenylation of dsx premRNA requires a sequence that binds specifically to tra-2 protein in vitro. Cell (Cambridge, Mass.) 68, 365-375. Herschman, H. R. (1991). Primary response genes induced by growth factors and tumor promoters. Annu. Rev. Biochem. 60, 281-319. Hsu, D., and Olefsky, J. M. (1992). Characterization of insulin-like growth factor (IGF) binding proteins and their role in modulating IGF-I action in BHK cells. J. Biol. Chem. 267, 25576-25582. Hsu, J.-C., Laz, T. M., Mohn, K. L., and Taub, R. (1991). LRF-1, a novel leucine-zipper protein rapidly and highly induced in regenerating liver. Proc. Natl. Acad. Sci. U.S.A. 88, 35113515. Hsu, J.-C., Bravo, R., and Taub, R. (1992). Interactions between LRF-1, JunB, c-Jun and c-Fos define a regulatory program in the G 1 phase of liver regeneration. Mol. Cell. Biol. 12, 4654-4665. Hsu, J.-C., Cressman, D. E., and Taub, R. (1993). Promoter-specific trans-activation and inhibition mediated by JunB. Cancer Res. 53, 3789-3794. Huang, C.-S., Anderson, M. E., and Meister, A. (1993). Amino acid sequence and function of the light subunit of rat kidney ~/-glutamylcysteine synthetase. J. Biol. Chem. 268, 2057820583. Huh, G. S., and Hynes, R. O. (1993). Elements regulating an alternatively spliced exon of the rat fibronectin gene. Mol. Cell. Biol. 13, 5301-5314. Hutson, S. M., Stinson-Fisher, C., Shiman, R., and Jefferson, L. S. (1987). Regulation of albumin synthesis by hormones and amino acids in primary cultures of rat hepatocytes. Am. J. Physiol. 252, E291-E298.

4.


95

Jefferson, L. S. (1980). Role of insulin in the regulation of protein synthesis. Diabetes 29, 487496. Krainer, A. R., Mayeda, A., Kozak, D., and Binns, G. (1991). Functional expression of cloned human splicing factor SF2: Homology to RNA-binding proteins, U1 70K, and Drosophila splicing regulators. Cell (Cambridge, Mass.) 66, 383-394. Kruijer, W., Skelly, H., Botteri, E, van der Putten, H., Barber, J., Verma, I., and Leffert, H. (1986). Proto-oncogene expression in regenerating liver is simulated in cultures of primary adult rat hepatocytes. J. Biol. Chem. 261, 7929-7933. Lanahan, A., Williams, J. B., Sanders, L. K., and Nathans, D. (1992). Growth factor-induced delayed early response genes. Mol. Cell. Biol. 12, 3919-3929. Lau, L. E, and Nathans, D. (1985). Identification of a set of genes expressed during the G0/G1 transition of cultured mouse cells. EMBO J. 4, 3145-3151. Lau, L. E, and Nathans, D. (1987). Expression of a set of growth-related immediate early genes in BALB/c 3T3 cells: Coordinate regulation with c-fos or c-myc. Proc. Natl. Acad. Sci. U.S.A. 84, 1182-1186. Lee, J., Greenbaum, L., Haber, B. A., Nagle, D., Lee, V., Miles, V., Mohn, K. L., Bucan, M., and Taub, R. (1994). Structure and localization of the IGFBP-1 gene and its expression during liver regeneration. Hepatology (Baltimore) 19, 656-665. Lemaire, L., and Heinlein, U. A. O. (1993). High-level expression in male germ cells of murine P68 RNA helicase mRNA. Life Sci. 52, 917-926. Li, H., and Bingham, P. M. (1991). Arginine/serine-rich domains of the su(wa) and tra RNA processing regulators target proteins to a subnuclear compartment implicated in splicing. Cell (Cambridge, Mass.) 67 335-342. Lu, X. P., Koch, K. S., Lew, D. J., Dulic, V., Pines, J., Reed, S. I., Hunter, T., and Leffert, H. L. (1992). Induction of cyclin mRNA and cyclin-associated histone H1 kinase during liver regeneration. J. Biol. Chem. 267, 2841-2844. Marais, R., Wynne, J., and Treisman, R. (1993). The SRF accessory protein Elk-1 contains a growth factor-regulated transcriptional activation domain. Cell (Cambridge, Mass.) 73, 381-393. Mattox, W., Ryner, L., and Baker, B. S. (1992). Autoregulation and multifunctionality among trans-acting factors that regulate alternative pre-mRNA processing. J. Biol. Chem. 267, 19023-19026. Mayeda, A., and Krainer, A. R. (1992). Regulation of alternative pre-mRNA splicing by hnRNP A1 and splicing factor SF2. Cell (Cambridge, Mass.) 68, 365-375. Mayeda, A., Zahler, A. M., Krainer, A. R., and Roth, M. B. (1992). Two members of a conserved family of nuclear phosphoproteins are involved in pre-mRNA splicing. Proc. Natl. Acad. Sci. U.S.A. 89, 1301-1304. Michalopoulos, G. K. (1990). Liver regeneration: Molecular mechanisms of growth control FASEB J. 4, 176-187. Mohn, K. L., Laz, T. M., Melby, A. E., and Taub, R. (1990). Immediate-early gene expression differs between regeneration liver, insulin-stimulated H-35 cells, and mitogen~ 3T3 cells: Liver specific induction patterns of gene 33, PEPCK, and the jun, fos and egr families. J. Biol. Chem. 265, 21914-21921. Mohn, K. L., Laz, T. M., Hsu, J.-C., Melby, A. E., Bravo, R., and Taub, R. (1991a). The immediate-early growth response in regenerating liver and insulin-stimulated H-35 cells: Comparison to serum-stimulated 3T3 cells and identification of 41 novel immediate-early genes. Mol. Cell. Biol. 11, 381-390. Mohn, K. L., Melby, A. E., Tewari, D. S., Laz, T. M., and Taub, R. (1991b). The gene encoding rat insulin like growth factor-binding protein i is rapidly and highly induced in regenerating liver. Mol. Cell. Biol. 11, 1393-1401.

96

Rebecca Taul)

Nelson, P. A., Lippke, J. A., Murcko, M. A., Rosborough, S. L., and Peattie, D. A. (1991). cDNA encoding murine FK506-binding protein (FKBP): Nucleotide and deduced amino acid sequences. Gene 109, 255-258. Peavy, D. E., Taylor, J. M., and Jefferson, L. S. (1985). Time course of changes in albumin synthesis and mRNA in diabetic and insulin-treated diabetic rats. Endocrinol. Metal). Clin. North Am. 248,E656-E663. Ruff-Jamison, S., Chen, K., and Cohen, S. (1993). Induction by EGF and interferon-~ of tyrosine phosphorylated DNA binding proteins in mouse liver nuclei. Science 261, 17331736. Scearce, L. M., Laz, R. M., Hazel, T. G., Lau, L. E, and Taub, R. (1993). RNR-1, a nuclear receptor in the NGFI-B/Nur-77 family that is rapidly induced in regenerating liver. J. Biol. Chem. 268, 8855-8861. Schneiderman, J., Sawdey, M., Craig, H., Thinnes, T., Bordin, G., and Loskutoff, D. J. (1993). Type I plasminogen activator inhibitor gene expression following partial hepatectomy. Am. J. Pathol. 143, 753-762. Shelly, L. L., Lei, K.oJ., Pan, C.-J., Sakata, S. E, Ruppert, S., Shutz, G., and Chou, J. Y. (1993). Isolation of the gene for murine glucose-6ophosphatase, the enzyme deficient in glycogen storage disease type 1A. J. Biol. Chem. 268, 21482-21485. Sherr, C. J. (1993). Mammalian G1 cyclins Cell (Cambridge, Mass.) 75, 839-841. Simmons, D. L., Neel, B. G., Stevens, R., Evett, G., and Erikson, R. L. (1992). Identification of an early-growth-response gene encoding a novel putative protein kinase. Mol. Cell. Biol. 12, 4164-4169. Stein, B., and Baldwin, Jr., A. S. (1993). Distinct mechanisms for regulation of the interleukin-8 gene involve synergism and cooperativity between C/EBP and NF-KB. Mol. Cell Biol. 13, 7191-7198. Sun, H., Charles, C. H., Lau, L. E, and Tonks, N. K. (1993). MKP-1 (3CH134), an immediate early gene product, is a dual specificity phosphatase that dephosphorylates Map kinase in vivo. Cell (Cambridge, Mass.) 75, 487-493. Takiguchi, S., Tokino, T., Imai, T., Tanigami, A., Koyama, K., and Nakamura, Y. (1993). Identification and characterization of a cDNA, which is highly homologous to the ribonucleoprotein gene, from a locus (D10S102) closely linked the MEN2 (multiple endocrine neoplasia type 2). Cytogenet. Cell Genet. 64, 128-130. Taub, R., Roy, A., Dieter, R., and Koontz, J. (1987). Insulin as a growth factor in rat hepatoma cells: Stimulation of protooncogene expression. J. Biol. Chem. 262, 10893-10897. Tewari, M., Dobrzanski, P., Mohn, K. L., Cressman, D. E., Hsu, J.oC., Bravo, R., and Taub, R. (1992). Rapid induction in regenerating liver of RL/IF-1, an IKB that inhibits NH-KB, Relo p50; and PHF, a novel KB site binding complex. Mol. Cell. Biol. 12, 2898-2908. Thompson, N. L., Mead, J. E., Braum, L., Goyette, M., Shank, P. R., and Fausto, N. (1986). Sequential protooncogene expression during rat liver regeneration. Cancer Res. 46, 31113117. Thornton, A. J., Bruzdzinski, C. J., Raper, S. E., and Gelehrter, T. D. (1994). Plasminogen activator inhibitor-1 is an immediate early response gene in regenerating rat liver. Cancer Res. 54, 1337-1343. Tohyama, C., Suzuki, J. S., Hemelraad, J., Nishimura, N., and Nishimura, H. (1993). Induction of metallothionein and its localization in the nucleus of rat hepatocytes after partial hepatectomy. Hepatology (Baltimore) 18, 1193-1201. Varnum, B. C., Lim, R. W., and Herschman, H. R. (1989). Characterization of TIS7, a gene induced in Swiss 3T3 cells by the tumor promoter tetradecanoyl phorbol acetate. Oncogene 4, 1263-1265. Zahler, A. M., Lane, W. S., Stolk, J. A., and Roth, M. B. (1992). SR proteins: A conserved family of pre-mRNA splicing factors. Genes Dev. 6, 837-847.

4.


97

Zahler, A. M., Neugebauer, K. M., Stolk, J. A., and Roth, M. B. (1993a). Human SR proteins and isolation of a cDNA encoding SRp75. Mo/. Cell. Biol. 13, 4023-4028. Zahler, A. M., Neugebauer, K. M., Stolk, J. A., and Roth, M. B. (1993b). Alternative premessenger RNA splicing by SR proteins. Science 260, 219-222. Zipfel, P. F., Irving, S. G., Kelly, K., and Siebenlist, U. (1989). Complexity of the primary genetic response to mitogenic activation of human T cells. Mo/. Cell. Biol. 9, 1041-1048.


5 Stem Cells and Hepat0carc nogenes s Snorri S. Thorgeirsson Laboratory of Experimental Carcinogenesis Division of Cancer Etiology National Cancer Institute National Institutes of Health Bethesda, Maryland 20892

I. Introduction The existence of hepatic stem cells has been, and no doubt will continue to be, a matter of considerable controversy. This controversy is partly fueled by the fact that cell turnover in the liver is very slow and that the two major types of hepatic epithelial cells, hepatocytes and biliary epithelia, are capable of proliferation and can, at least in a healthy liver, meet replacement demands of cellular loss from these two differentiated populations. The best example of the capacity of adult hepatocytes and bile epithelial cells to proliferate is seen following partial hepatectomy in rats and mice in which the compensatory hyperplasia of these cells in the remaining lobes restores the liver mass (see Chapters 1 and 2). The increased use and success of liver transplantation in clinical medicine has shown that these animal models also correctly reflect the capacity of the human liver to regenerate (see Chapter 13) (Van Thiel et al., 1989). The major part of the hepatic stem cell controversy may, however, be due to the failure of recognizing that the adult organism contains many kinds of stem cells. These cells may exist at different stages of differentiation and have very different capacities for generating multilineage progeny.

Liver Regenerationand Carcinogenesis

99

100

Snorri S. Thorgeirsson

The capacity for self maintenance is a fundamental and common trait of all stem cells. A cell population that has an extensive self-maintaining capacity is the only definition that applies to all stem cells (Lajtha, 1979). In this context, the adult liver, which has an extensive capacity for maintaining parenchymal cell number throughout the life span of the organism, can be viewed as a single lineage stem cell system in which the hepatocyte is the stem cell. Recent data from hepatic cell transplantation experiments in a transgenic mouse system (Rhim et al., 1994) have demonstrated the enormous growth potential of adult hepatocytes (12 to 16 doublings per donor cell), further supporting the notion of the liver parenchyma as a single lineage or unipotential stem cell system (see Chapter 11). The endodermal origin of the liver, and the fact that the early fetal hepatocytes or hepatoblasts are progenitor for both adult hepatocytes and bile epithelial cells, suggests that the hepatoblasts are at least bipotential precursors. The question then arises whether either or both of the cell lineages derived from the hepatoblast retain the "bipotential capacity" of the precursor cells. At present there is no substantial evidence to indicate that adult hepatocytes are more than a unipotential stem cell system. The possible exception to this generalization is the concept of ductular metaplasia or transformation of hepatocytes into ductules. This topic has been extensively discussed by Desmet and colleagues (for review see Van Eyken and Desmet, 1992). Studies of chronic cholestatic diseases in humans employing both enzyme histochemistry and cytokeratin immunohistochemistry have provided evidence for gradual transformation of hepatocytes into "bile duct-type" cells (Van Eyken et al., 1989). Evidence showing that these "bile duct-type" cells also exhibit functional characteristics of normal bile epithelium is still lacking. It is therefore questionable but still possible that ductal metaplasia of hepatocytes seen in cholestatic diseases may reflect multipotential or at least bipotential capacity of the hepatocytes. In contrast to the hepatocyte system, there is strong evidence indicating that the bile epithelium harbors a compartment of cells that are capable of differentiating into several lineages including bile epithelia, hepatocytes, intestinal epithelia, and possibly exocrine pancreas (Thorgeirsson, 1993; Fausto, 1990; Sell, 1990; Sigal et al., 1992). Also, as pointed out by Sell (1990), there may exist a periductal system of stem cells capable of differentiating into all the hepatic lineages. These ductular/periductal cells are frequently referred to as the hepatic stem cell system. In light of these observations, the liver system should be viewed as being composed of two stem cell systems: the unipotential (possibly bipotential) hepatocytic and the multipotential nonparenchymal epithelial (ductular) systems (Figure 1). This chapter will review the general cellular biology of the hepatic stem cell compartment, and the possible involvement of these cells in hepatocarcinogenesis.

5. Stem Cells and Hepatocarcinogenesis

101

Figure I Possiblelineage pathways in the liver.

n. Cellular Biology of the Hepatic Stem Cell Compartment What then is the evidence that a multipotential stem cell compartment exists in the liver? The existence of hepatic stem cells was first postulated by Wilson and Leduc in 1958 based on experiments involving liver regeneration in the mouse after chronic injury induced with a methionine-rich basal diet mixed with an equal amount of bentonite (Wilson and Leduc, 1958). The authors concluded "that prolonged and severe injury to the liver may make direct restoration by division of preexisting parenchymal cells impossible, and that, when this occurs, the new parenchyma is derived from the indifferent cholangiole cells." The major support for the existence of hepatic stem cells has, perhaps not surprisingly in light of the earlier work by Wilson and Leduc, come from extensive studies of hepatic carcinogenesis (Fausto, 1990; Sell, 1990; Sigal et al., 1992; Marceau, 1990). A. Experimental Systems in Vivo The rat has been most extensively used to generate experimental systems in which to study the cell biology of the hepatic stem cell compartment. The three most used model systems are (1) feeding of choline deficient diet with or without 0.1% ethionine (CDE) (Shinozuka et al., 1978); (2) combination of 2-acetylaminofluorene (AAF) treatment and partial hepatectomy (Tatematsu et al., 1985); and (3) administration of hepatotoxic doses of d-galac-

102


tosamine (Lemire et al., 1991). Common to all these experimental systems is the extensive destruction of liver parenchyma and/or prevention of regeneration following loss of liver mass as well as compromised function of the remaining hepatocytes. The observation that the regenerative capacity of the hepatocytes has to be compromised before the hepatic stem cell compartment becomes activated and starts to contribute to the regeneration of the liver parenchyma led Grisham (1980) to propose that this be referred to as a facultative stem cell compartment. This is a particularly important concept since, as pointed out above, healthy hepatocytes have extensive capacity for regeneration and should be looked on as an unipotential stem cell system. A common cellular response in rats subjected to the experimental systems listed above is the proliferation of small periportal cells with scant cytoplasm and ovoid nuclei that have been termed oval cells (Farber, 1984, 1990; Evarts, et al., 1987; Marceau, 1990). The existence of similar cells has been reported in human liver (Gerber et al., 1983; Hsia et al., 1992). These oval cells are thought to represent a progeny of the hepatic stem cell compartment, and in some instances to be a precursor for hepatic tumors (Sell, 1990; Sigal et al., 1992; Hixson et al., 1990). The precise anatomical location of the hepatic stem cell compartment in normal liver is still unclear, but present data suggest that the terminal ductule cells connecting the canals of herring with the bile canaliculi and/or a distinct population of periductal cells constitute the hepatic stem cell compartment (Factor et al., 1994; Fausto, 1990; Sell, 1990; Sigal et al., 1992; Wilson and Leduc, 1958). Recent studies in the rat liver have shown that oval cells are capable of differentiating into, in addition to bile epithelium, at least two lineages in vivo including hepatocytes (Evarts et al., 1987; Lemire et al., 1991) and intestinal type epithelium (Tatematsu et al., 1985). Furthermore, isolated oval cells in culture can be induced to differentiate into both hepatocyte-like and biliary types of cells (Hayner et al., 1984; Germain et al., 1985, 1988). These data and other results not reviewed here support the notion that oval cells have lineage options similar to hepatoblasts in the early stages of liver development. As such, the oval cells can be regarded as "bipotential precursors" for the two hepatic parenchymal cell lineages. In addition, they may exhibit a capacity to differentiate into lineages particularly when the hepatic microenvironment is drastically disrupted (Thorgeirsson and Evarts, 1992). B.

ExperimentalSystems in

Vitro

In addition to the in vivo data mentioned above, significant support for the existence of the hepatic stem cell has come from results obtained in studies on hepatic cell cultures. Several research groups have been able to isolate and establish long-term cultures of small, morphologically and functionally

5.

Stem Cells and Hepatocarcinogenesis

103

simple epithelial cells by enzymatic perfusion of fetal and adult rat liver and utilizing culture conditions that exclude differentiated hepatocytes (Grisham et al., 1974; Furukaura et al., 1987; Marceau et al., 1980; Schaeffer, 1980). These rat liver-derived epithelial (RLE) cells share some phenotypic properties with both bile duct epithelial cells and hepatocytes, but are phenotypically much closer to some of the oval cell lines (Hayner et al., 1984; Tsao et al., 1984; Marceau et al., 1986). The notion that oval cells are descendants from hepatic stem cells suggests that the RLE cells are also progeny from such a stem cell compartment because of their phenotypic similarities to oval cells. This suggestion is supported by data obtained from extensive use of RLE cells for in vitro transformation studies with both chemical carcinogens and oncogenes. These studies have demonstrated various transformation properties of RLE cells (Tsao and Grisham, 1987; Garfield et al., 1988; Schaeffer, 1980). Most importantly, when the transformed RLE cells are transplanted into either syngeneic rats or nude mice a spectrum of tumors are encountered that includes highly differentiated hepatocellular carcinomas, hepatoblastomas, cholangiocarcinoma as well as mixed epithelial-mesenchymal tumors (Tsao and Grisham, 1987). This suggests a blastic nature of the RLE cells and demonstrates the potential of the cells to differentiate via both the hepatocytic and biliary lineages. However, no in vitro culture system exists that conclusively demonstrates differentiation of the RLE cells into either hepatocytes or bile epithelial cells. A recent insight into the nature of the RLE cells and the potential role of hepatic stem cells in liver biology was provided by Coleman et al. (1993). They have successfully demonstrated, following intrahepatic transplantation of RLE cells carrying the Escherichia coli 13-galactosidase reporter gene, that the RLE cells integrate into hepatic plates and acquire the size and nuclear structure of mature hepatocytes. In addition, they showed that intrahepatic transplantation of an aneuploid, neoplastically transformed derivative of the RLE cells, which produces aggressively growing tumor when transplanted subcutaneously, does not produce tumors in the liver but rather integrates into the hepatic plates and morphologically differentiates. The results from this study raise several important issues for basic hepatic biology as well as clinical hepatology. Differentiated hepatocytes isolated from enzymatically prepared suspensions of murine liver cells have been shown already to integrate into the hepatic plate following transplantation into liver of congenic animals (Ponder et al., 1991; Gupta et al., 1991). This is, however, the first demonstration that primitive or "stem-like" RLE cells on intrahepatic transplantation also integrate into the hepatic plates and differentiate morphologically into hepatocytes. These results taken together with the large body of data on the blastic nature of the RLE cells strongly support both the existence of a hepatic stem cell compartment and the

104


notion that RLE cells are derived from this compartment. Furthermore, the present study has also reinforced the notion that both oval cells and RLE cells represent a progeny from the hepatic stem cell compartment.

III. Neoplastic Development in the Liver Genesis of liver tumors most probably occurs via multiple molecular mechanisms which depend on both the nature of the carcinogen and the lesions induced by it. As previously stated, the liver system should be viewed as being composed of two stem cell systems: the unipotential (possibly bipotential) hepatocytic and the multipotential nonparenchymal epithelial (ductular) systems (Figure 1). Therefore, it seems reasonable to expect that both cell systems could provide progenitor cells for the neoplastic process. There is no doubt that the hepatocyte frequently is the progenitor cell for liver tumors (Farber, 1992). The involvement of the nonparenchymal (ductular) system in the genesis of liver tumors, particularly hepatocellular carcinomas, is still hotly debated. On the one hand, it has been proposed that "the cell of origin of liver cancer is the putative liver stem cell or its progeny, the transitional duct cell" (Sell and Pierce, 1994). Alternatively, Farber (1992) has stated that "rare original mature hepatocytes in zone 1, 2, or 3 of the adult liver appearing after initiation with genotoxic carcinogens have been shown to be the cell of origin for foci or islands of altered hepatocytes and of nodules derived from these foci." The central issue in better understanding the involvement of the nonparenchymal epithelial (ductular) cells in the carcinogenic process is the characterization of the mechanisms that regulate both the proliferation of these cells after carcinogenic as well as noncarcinogenic insults, and the factors governing the lineage commitment processes in this compartment. A. Hepatic Stem Cells and Hepatocarcinogenesis A landmark contribution to the involvement of liver nonparenchymal epithelial cells in hepatocarcinogenesis was provided by Farber (1956)who documented a detailed description of the early histological changes during hepatocarcinogenesis caused by three chemical carcinogens. The carcinogens used by Farber, ethionine, 2-acetylaminofluorene (AAF), and 3'-methyl4-dimethylaminoazobenzene (Me-DAB), in spite of being structurally very different, caused similar histological alterations. The common features included (1) oval cell proliferation which progressively involved most of the liver lobule, beginning in the portal areas, (2) degenerative and hypertrophic changes in hepatocytes adjacent to proliferating oval cells, and (3) nodular regenerative hyperplasia of liver cells. There were, however, important dif-

s.


105

ferences in the time course of appearance and fate of the oval cells induced by these three hepatocarcinogens. Whereas, oval cells appeared early following ethionine and AAF administration (7 and 14 days, respectively), their appearance occurred significantly late after Me-DAB treatment (first seen at Day 21). More importantly, the fate of the oval cells in the Me-DAB treated animals were different from those induced by ethionine and AAE In the early stages the oval cells induced by Me-DAB were morphologically indistinguishable from those generated by ethionine and AAE However, at later stages areas of apparent transition between oval cells and hepatocytes were numerous in the Me-DAB treated animals but absent in those receiving ethionine and AAE These observations raise several important issues. First, and most importantly, it is now well established that many different chemical compounds capable of producing liver tumors in rats and mice, induce a similar sequence of histological changes in which oval cell hyperplasia is prominent (Dunsford et al., 1985). Secondly, if the transition from oval cells to hepatocytes can be morphologically observed after Me-DAB treatment, then it is in principle established that oval cells (or at least a subpopulation of oval cells) have the capacity to differentiate into hepatocytes. The fact that administration of ethionine or AAF did not provide the same clear morphological sequence as seen with Me-DAB in which the oval cells merge imperceptibly and were in continuity with the regenerating nodules, suggests that the compounds capable of inducing oval cell proliferation may greatly affect both the rate and extent of oval cell differentiation into hepatocytes. The fact that a large population of oval cells is cycling during the early stages of chemical hepatocarcinogenesis and that these cells can differentiate into hepatocytes strongly suggests that at least a percentage of the hepatocellular carcinomas are derived from oval cell progenitors. Recently there has been accumulating experimental evidence in support of this notion. Hixson and his colleagues (Hixson et al., 1990; Faris et al., 1991) have used a battery of monoclonal antibodies specific for antigens associated with bile duct cells, oval cells and fetal, adult, and neoplastic hepatocytes to analyze the phenotypic relationship between oval cells, foci, nodules, and hepatocellular carcinomas during chemical hepatocarcinogenesis. These investigators found, using the resistant hepatocyte model of Solt and Farber (1976), that oval cells, GGT-positive hepatocellular foci, persistent hepatocyte nodules, and primary hepatocellular carcinomas express both oval cell and hepatocyte antigens. This finding indicates a precursorproduct relationship between oval cells and carcinomas. Similar results were obtained by Dunsford et al. (1989) using different monoclonal antibodies raised against oval cells. These lineage relationships between oval cells and hepatocellular carcinomas also exist in other models of liver carcinogenesis. For example, animals maintained on a CDE diet display markers

106


for oval cells and hepatocytes in a significant percentage of nodules and hepatocellular carcinomas (Hixson et al., 1990; Faris et al., 1991). Also, metastatic foci in the lung from the animals harboring these liver tumors show essentially the same phenotype (Hixson et al., 1990). The evidence for oval or ductular cells as progenitors for hepatocellular carcinomas is not restricted to experimental models of chemical hepatocarcinogenesis in rodents. Results from Van Eyken et al. (1988) on the cytokeratin expression in 34 "classical" human hepatocellular carcinomas (HCC) using monospecific anticytokeratin antibodies show that all HCCs were positive for cytokeratins 8 and 18. However, in 17 cases a variable number of the tumor cells were positive for cytokeratin 7 (two cases), cytokeratin 19 (7 cases), or both 7 and 19 (8 cases). The authors also reported that only 3 of 11 well-differentiated tumors display an "unexpected" pattern of immunoreactivity as opposed to 7 of 7 poorly differentiated tumors. This is particularly important in light of the earlier observation by Denk et al. (1982) that cytokeratins continue to be expressed when hepatocytes become neoplastic. These observations are also highly relevant in light of the recent findings of Hsia et al. (1992) and Vandersteenhoven et al. (1990) who demonstrated immunohistochemically the presence of ductular "oval" type cells with characteristics of both bile ducts and hepatocytes in the liver of patients with end stage cirrhosis and/or tumors from hepatitis B infection. It is important at this stage to reemphasize that the relative percentage of primary hepatocellular carcinomas derived from oval cell progenitors may vary over a wide range depending on the carcinogenesis protocol and/or the chemical carcinogen used as well as the extent of oval cell involvement in the early stages of the process. B. Transformation of Liver Derived Epithelial (Oval) Cells The most direct evidence that oval cells and/or RLE cells can progress to hepatocellular carcinomas comes from in vitro transformation of these cells. Spontaneous transformation of RLE and oval cells as well as transformation with chemical carcinogens and dominant oncogenes results in the tumors displaying a wide range of phenotypes including well-differentiated hepatocellular carcinomas, cholangiomas, hepatoblastomas, and poorly differentiated or anaplastic tumors (Tsao and Grisham, 1987; Garfield et al., 1988; Fausto, 1990; Marceau, 1990). In one of the most comprehensive studies on the chemical transformation of the RLE cells by Tsao and Grisham (1987), a wide range of tumors described including carcinomas, sarcomas, mixed epithelia-mesenchymal tumors, and undifferentiated tumors. In addition, several tumors were morphologically indistinguishable from hepatocellular carcinomas. Also, the "mixed epithelial-mesenchymal" tumors reproduced most of the various histologic features of human hepatoblastomas (Table 1).

107

S. Stem Cells and Hepatocarcinogenesis

Table 1 Light Microscopic Classification of Tumors Produced by ChemicallyTransformed Rat Liver Epithelial Cells Tumor type Carcinomas (epithelial) Epidermoid Adenocarcinoma Hepatocellular Poorly differentiated/anaplastic Sarcomas (mesenchymal) Mixed epithelial-mesenchymaltumors Unclassified

Number of tumors 15 13 4 22 19 30 22

From Tsao and Grisham (1987).

We have recently demonstrated that cytokeratin 14 is expressed in several RLE cell lines (Bisgaard et al., 1994a). Although the partner for cytokeratins 8 and 14 has traditionally been found to be cytokeratins 18 and 5, respectively, it is now well documented that cytokeratins 8 and 14 can be expressed in the complete absence of their traditional partner (Bisgaard and Thorgeirsson, 1991; Bisgaard et al., 1993; Wirth et al., 1992). We have shown that in some RLE cell lines cytokeratins 8 and 14 form heterotypic filaments (Bisgaard et al., 1993). Also, we found that these cell lines express vimentin along with the cytokeratins (Wirth et al., 1992). However, the spontaneous transformation and differentiation of one of our RLE cell lines to a hepatoblast-like phenotype, forming a well-differentiated trabecular hepatocellular carcinoma, results in an abrogation of vimentin protein expression and a change in cytokeratin expression from which cytokeratin 14 was substituted by 18 (Bisgaard et al., 1994b). We have used this RLE transformation system to study the relationship between the expression of cytokeratins 14, 8, as well as 18 and oL-fetoprotein (AFP) during the process of proliferation and differentiation of the RLE cell line to a hepatoblast-like progeny. The steady-state levels of mRNA transcripts for cytokeratin 14 and AFP, as well as for cytokeratins 8 and 18 and vimentin show a significant change in the expression pattern during the process of transformation (Figure 2). Before the cells display morphological signs of transformation, a high steady-state level of cytokeratin 14 transcripts in addition to transcripts for cytokeratin 8 and vimentin is detected (Figure 2). During the process of transformation that occurred within 33 to 35 passages, the steady-state levels of cytokeratin 14 and vimentin abruptly declined, and could not be detected in later passages nor in the clonal transformed B5T cell line. The disappearance of the cytokeratin 14 and vimentin mRNA transcripts closely corresponds with the appearance of a 2.1-kDa transcript for AFP and a 1.4-

108


Figure 2 Differential expression of cytokeratins 8, 14, 18, vimentin, and oL-fetoprotein mRNA during spontaneous transformation of RLESF13 cells by prolonged passage in vitro. (From Bisgaard et al., 1994b. With permission).

kDa transcript for cytokeratin 18 (Figure 2). The mRNA transcripts for cytokeratins 8 and 18 as well as those for AFP are present in the transformed clonal B5T cell line (Figure 2). In contrast to the spontaneous transformation, these same RLE cells when transformed by dominant oncogenes yield primitive and anaplastic tumors (Garfield et al., 1988). These data indicate that the tumor phenotypes derived from RLE and/or oval cells may depend on both the mechanism of transformation and the stage of differentiation of the cells when the transformation occurs.

IV. C o n c l u s i o n s The adult organism contains many kinds of stem cells that exist at different stages of differentiation and have very different capacities for generating multilineage progeny. The capacity for self maintenance is a fundamental and common trait of all stem cells. A cell population that has an extensive self-maintaining capacity is the only definition that applies to all stem cells. It is proposed that the liver system be viewed as composed of two stem cell systems: the unipotential hepatocytic and the multipotential nonparenchy-

S.


109

mal epithelial (ductular) systems. Although the participation of the nonparenchymal epithelial (ductular) system in the development of liver tumors is still not fully defined, strong evidence now exists indicating that both these systems can and do provide progenitor cells for the neoplastic process in the liver. The central issue in better understanding the involvement of the nonparenchymal epithelial (ductular) cells in the carcinogenic process is the characterization of the mechanisms that regulate both the proliferation of these cells after carcinogenic as well as noncarcinogenic insults and the factors that govern the lineage commitment processes in this stem cell system.

References Bisgaard, H. C., and Thorgeirsson, S. S. (1991). Evidence for a common cell of origin for primitive epithelial cells isolated from rat liver and pancreas. J. Cell. Physiol. 147, 333343. Bisgaard, H. C., Parmelee, D. C., Dunsford, H. A., Sechi, S., and Thorgeirsson, S. S. (1993). Keratiin 14 protein in cultured nonparenchymal rat hepatic epithelial cells: Characterization of keratin 14 and keratin 19 as antigens for the commonly used mouse monoclonal antibody OV-6. Mol. Carcinog. 7, 60-66. Bisgaard, H. C., Nagy, P., Ton, P. T., Hu, Z., and Thorgeirsson, S. S. (1994a). Modulation of keratin 14 and ~-fetoprotein expression during hepatic oval cell proliferation and liver regeneration. J. Cell. Physiol. 159, 475-484. Bisgaard, H. C., Ton, P. T., Nagy, P., and Thorgeirsson, S. S. (1994b). Phenotypic modulation of keratins, vimentin, and cx-fetoprotein in cultured rat liver epithelial cells after chemical, oncogene, and spontaneous transformation. J. Cell. Physiol. 159, 485-494. Coleman, W. B., Wennerberg, A. E., Smith, G. T., and Grisham, J. W. (1993). Regulation of the differentiation of diploid and some aneuploid rat liver epithelial (stem like) cells by the hepatic microenvironment. Am. J. Pathol. 142, 1373-1382. Denk, H., Krepler, R., Lackinger, E., Artlieb, U., and Franke, W. W. (1982). Biochemical and immunocytochemical analysis of the intermediate filament cytoskeleton in human hepatocellular carcinomas and in hepatic neoplastic nodules of mice. Lab. Invest. 46, 584-596. Dunsford, H. A., Maset, R., Salman, J., and Sell, S. (1985). Connection of duct-like structures induced by a chemical hepatocarcinogen to portal bile ducts in the rat liver detected by injection of bile ducts and pigmented barium gelatin medium. Am. J. Pathol. 118,218-224. Dunsford, H. A., Karnasuta, C., Hunt, J. M., and Sell, S. (1989). Different lineages of chemically induced hepatocellular carcinoma in rats defined by monoclonal antibodies. Cancer Res. 49, 4894-4900. Evarts, R. P., Nagy, P., Marsden, E., and Thorgeirsson, S. S. (1987). A precursor relationship exists between oval cells and hepatocytes in rat liver. Carcinogenesis 8, 1737-1740. Factor, V. M., Radaeva, S. A., and Thorgeirsson, S. S. (1994). Origin and fate of oval cells in Dipin-induced hepatocarcinogenesis in the mouse. Am. J. Pathol. 145,409-422. Farber, E. (1956). Similarities in the sequence of early histological changes induced in the liver of the rat by ethionine, 2-acetylaminofluorene, and 3'-methyl-4-dimethylaminoazobenzene. Cancer Res. 16, 142-148. Farber, E. (1984). Cellular biochemistry of the stepwise development of cancer with chemicals. Cancer Res. 44, 5463-5474.

110


Farber, E. (1990). Clonal adaptation during carcinogenesis. Biochem. Pbarmacol. 39, 18371846. Farber, E. (1992). On cells of origin of liver cell cancer. In "The Role of Cell Types in Hepatocarcinogenesis" (A. E. Sirica, ed.), pp. 1-28. CRC Press, Boca Raton, Florida. Faris, R. A., Monfils, B. A., Dunsford, H. A., and Hixson, D. (1991). Antigenic relationship between oval cells and a subpopulation of hepatic foci, nodules, and carcinomas induced by the "resistant hepatocyte" model system. Cancer Res. 51, 1308-1317. Fausto, N. (1990). Oval cells and liver carcinogenesis: An analysis of cell lineages in hepatic tumors using oncogene transfection techniques. Prog. Clin. Biol. Res. 331,325-334. Furukaura, K., Schimada, T., England, P., Mochizuki, Y., and Williams, G. M. (1987). Enrichment and characterization of clonogenic epithelial cells from adult rat liver and initiation of epithelial cell strains. In Vitro Cell. Dev. Biol. 23, 339-348. Garfield, S., Huber, B. E., Nagy, P., Cordingley, M. G., and Thorgeirsson, S. S. (1988). Neoplastic transformation and lineage switching of rat liver epithelial cells by retrovirus-associated oncogenes. Mol. Carcinog. 1, 189-195. Gerber, M. A., Thung, S. N., Shen, S., Stromeyer, E W., and Ishak, K. G. (1983). Phenotypic characterization of hepatic proliferation: Antigenic expression of proliferating epithelial cells in fetal liver, massive hepatic necrosis, and nodular transformation of the liver. Am. J. Pathol. 110, 70-74. Germain, L., Goyette, R., and Marceau, N. (1985). Differential cytokeratin and ot-fetoprotein expression in morphologically distinct epithelial cells emerging at the early stages of rat hepatocarcinogenesis. Cancer Res. 45,673-681. Germain, L., Noel, H., Gourdeau, H., and Marceau, N. (1988). Promotion of growth and differentiation of rat ductular oval cells in primary culture. Cancer Res. 48, 368-378. Grisham, J. W. (1980). Cell types in long-term propagable cultures of rat liver. Ann. N.Y. Acad. Sci. 349, 128-137. Grisham, J. W., Thai, S. B., and Nagel, A. (1974). Cellular derivation of continuously cultured epithelial cells from normal rat liver. In "Gene Expression and Carcinogenesis in Cultured Liver" (L. E. Gerschenson and E. B. Thompson, eds.), pp. 1-23. Academic Press, New York. Gupta, S., Aragona, E., Vemuru, R. P., Bhargava, K. K., Burk, R. D., and Chowdhury, J. R. (1991). Permanent engraftment and function of hepatocytes delivered to the liver: Implications for gene therapy and liver repopulation. Hepatology (Baltimore) 14, 144-149. Hayner, N. T., Braun, L., Yaswen, R, Brooks, M., and Fausto, N. (1984). Isozyme profiles of oval cells, parenchymal cells, and biliary cells isolated by centrifugal elutriation from normal and preneoplastic livers. Cancer Res. 44, 332-338. Hixson, D. C., Faris, R. A., and Thompson, N. L. (1990). An antigenic portrait of the liver during carcinogenesis. Pathobiology 58, 65-77. Hsia, C. C., Evarts, R. P., Nakatsukasa, N., Marsden, E. R., and Thorgeirsson, S. S. (1992). Occurrence of oval-type cells in hepatitis B virus-associated human hepatocarcinogenesis. Hepatology (Baltimore) 16, 1327-1333. Lajtha, L. G. (1979). Stem cell concepts. Nouv. Rev. Fr. Hematol. 21, 59-65. Lemire, J. M., Shiojiri, N., and Fausto, N. (1991). Oval cell proliferation and the origin of small hepatocytes in liver injury induced by D-galactosamine. Am. J. Pathol. 139, 535-552. Marceau, N. (1990). Biology of disease. Cell lineage and differentiation programs in epidermal, urothelial and hepatic tissues and their neoplasms. Lab. Invest. 63, 4-20. Marceau, N., Goyette, R., Deschenes, J., and Valet, J.-P. (1980). Morphological differences between epithelial and fibroblast cells in rat liver cultures and the roles of cell surface fibronectin and cytoskeletal element organization in cell shape. Ann. N.Y. Acad. Sci. 349, 138-152. Marceau, N., Germain, L., Goyette, R., Noel, M., and Gourdeau, H. (1986). Cell of origin of

S.


111

distinct cultured rat liver epithelial cells, as typed by cytokeratin and surface component selective expression. Biochem. Cell Biol. 64, 788-802. Ponder, K. P., Gupta, S., Leland, E, Darlington, G., Finegold, M., DeMayo, J., Ledley, E D., Chowdhury, J. R., and Woo, S. L. C. (1991). Mouse hepatocytes migrate to liver parenchyma and function indefinitely after intrasplenic transplantation. Proc. Natl. Acad. Sci. U.S.A. 88, 1217-1221. Rhim, J. A., Sandgren, E. P., Degen, J. L., Palmiter, R. D., and Brinster, R. L. (1994). Replacement of diseased mouse liver by hepatic cell transplantation. Science 263, 1149-1152. Schaeffer, W. I. (1980). The long term culture of a diploid rat hepatocyte cell strain. Ann. N.Y. Acad. Sci. 349, 165-182. Sell, S. (1990). Is there a liver stem cell? Cancer Res. 50, 3811-3815. Sell, S., and Pierce, G. B. (1994). Biology of disease. Maturation arrest of stem cell differentiation is a common pathway for the cellular origin of teratocarcinomas and epithelial cancers. Lab. Invest. 70, 6-22. Shinozuka, H., Lombardi, B., Sell, S., and Iammarino, R. M. (1978). Early histological and functional alterations of ethionine liver carcinogenesis in rats fed a choline-deficient diet. Cancer Res. 38, 1092-1098. Sigal, S. H., Brill, S., and Reid, L. M. (1992). The liver as a stem cell and lineage system. Am. J. Physiol. 263, G139-G148. Solt, D. B., and Farber, E. (1976). New principle for the analysis of chemical carcinogenesis. Nature (London) 263, 701-703. Tatematsu, M., Kaku, T., Medline, A., and Farber, E. (1985). Intestinal metaplasia as a common option of oval cells in relation to cholangiofibrosis in the livers of rats exposed to 2-acetylamino-fluorene. Lab. Invest. 52, 354-362. Thorgeirsson, S. S. (1993). Commentary. Hepatic stem cells. Am. J. Pathol. 142, 1331-1333. Thorgeirsson, S. S., and Evarts, R. P. (1992). Growth and differentiation of stem cells in adult liver. In "The Role of Cell Types of Hepatocarcinogenesis" (A. E. Sirica, ed.), pp. 100-120. CRC Press, Boca Raton, Florida. Tsao, M.-S., and Grisham, J. W. (1987). Hepatocarcinomas, cholangiocarcinomas, and hepatoblastomas produced by chemically transformed cultured rat liver epithelial cells. A lightand electron-microscopic analysis. Am. J. Pathol. 127, 168-181. Tsao, M.-S., Smith, J. D., Nelson, K. G., and Grisham, J. W. (1984). A diploid epithelial cell line from normal adult rat liver with phenotypic properties of oval cells. Exp. Cell Res. 154, 38-52. Van Eyken, P., and Desmet, V. J. (1992). Development of intrahepatic bile ducts, ductular metaplasia of hepatocytes, and cytokeratin patterns in various types of human hepatic neoplasms. In "The Role of Cell Types in Hepatocarcinogenesis" (A. E. Sirica, ed.), pp. 227-263. CRC Press, Boca Raton, Florida. Van Eyken, P., Sciot, R., Paterson, A., Callea, E, Kew, M. C., and Desmet, V. J. (1988). Cytokeratin expression in hepatocellular carcinoma: An immunohistochemical study. Hum. Pathol. 19, 562-568. Van Eyken, P., Sciot, R., Callea, E, and Desmet, V. J. (1989). A cytokeratin-immunohistochemical study of focal nodular hyperplasia of the liver: Further evidence that ductular metaplasia of hepatocytes contributes to ductular "proliferation." Liver 9, 372-377. Van Thiel, D. H., Gavaler, J. S., Kam, I., Francavilla, A., Polimeno, L., Schade, P. R., Smith, J., Diven, W., Penkrot, R. J., and Starzl, T. E. (1989). Rapid growth of an intact human liver transplanted into a recipient larger than the donor. Gastroenterology 93, 1414-1419. Vandersteenhoven, A. M., Burchette, J., and Michalopoulos, G. (1990). Characterization of ductular hepatocytes in end-stage cirrhosis. Arch. PathoI. Lab. Med. 114, 403-406. Wilson, J. W., and Leduc, E. H. (1958). Role of cholangioles in restoration of the liver of the mouse after dietary injury. J. Pathol. Bacteriol. 76, 441-449.

112


Wirth, P. J., Luo, L.-D., Fujimoto, Y., and Bisgaard, H. C. (1992). Two-dimensional electrophoretic analysis of transformation-sensitive polypeptides during chemically, spontaneously, and oncogene-induced transformation of rat liver epithelial cells. Electrophoresis (Weinheim, Fed. Repub. Ger.) 13, 305-320.

6 Contributions of Hepadnavirus Research to Our Understanding of H ep ato carcin og en esis Charles E. Rogler Leslie E. Rogler Deyun Yang Silvana Breiteneder-Geleef Shih Gong Haiping Wang Marion Bessin Liver Research Center Albert Einstein College of Medicine Bronx, New York 10461

I. General Overview of Hepadnavirus Animal Models and Hepatocarcinogenesis The field of hepadnaviruses and their association with hepatocellular carcinomas has recently been the subject of several extensive reviews (Schirmacher et al., 1993; Rogler and Chisari, 1992; Tennant and Gerin, 1994). Rather than attempt to recapitulate these summaries, we will focus on those aspects of hepadnavirus research which have provided unique approaches to studying hepatocarcinogenesis and some of the insights they have provided into the process. Our discussion will focus on three animal models of hepadnavirus associated hepatocarcinogenesis. These include: (1) overexpression of the hepatitis B virus (HBV)envelope protein (HBsAg) in transgenic mice, (2) persistent infection of woodchucks with woodchuck hepatiLiver Regeneration and Carcinogenesis Copyright 9 1995 by Academic Press, Inc. All rights of reproduction in any form reserved.

113

114

Charles E. Rogler et al.

tis virus (WHV), and (3) overexpression of hepatitis B virus X gene (HBx) in transgenic mice. We will also review selected studies on the role of viral DNA integrations in hepatocellular carcinoma (HCC). Hepadnavirus animal models have upheld and reinforced many of the well-established principles of multistage hepatocarcinogenesis which were developed using chemical carcinogenesis animal models (Scherer, 1983; Farber and Sarma, 1987; Sell et al., 1987). The hepadnavirus animal models, as well as previously established models, are now poised to add new insights into the molecular genetics of hepatocarcinogenesis using newly developed tools made available through the murine and human genome projects. Table 1 is a brief list of some of the major cellular and molecular genetic processes which occur during hepatocarcinogenesis in the hepadnavirus associated models. While there are many common features of the overall pathophysiology between hepadnavirus models and chemical carcinogenesis models, the overall picture regarding molecular genetic changes is still unclear. The last section of Table 1 is an attempt to obtain a general consensus of possible changes in proto-oncogene, growth factor, and tumor suppressor gene expression in the hepadnavirus various models. Much of the data for this table was obtained from recent reviews (Tabor, 1994; Strom and Faust, 1990; Pasquinelli et al., 1992). At present, the fields of viral and chemical hepatocarcinogenesis would benefit from a comprehensive overview of all the literature to determine whether any clear general conclusions can be made; however, this is beyond the scope of this review. In spite of the confusion, the existing data suggest that malignant transformation of hepatocytes can be obtained by mutations or deregulation of genes linked to signal transduction pathways which are common to other cancers, and to alterations in tumor suppressor genes which perform critical functions in controlling cell cycle progression. Some of the signal transduction pathways implicated in hepadnavirus associated hepatocarcinogenesis are schematically represented in Figure 1. The interactions between members of these pathways continue to be elucidated, and new members of each interacting pathway are being discovered almost daily. One recent report which illustrates this concept is the discovery that the main function of Ras proto-oncogenes is to recruit the Raf proto-oncogene to the plasma membrane where it functions as a mitogenactivated protein (Map) kinase to activate an entire signal transduction pathway (Leevers et al., 1994). It will be productive for future studies to focus on common pathways which are perturbed in malignant hepatocytes and how these pathways interact with each other to lead to the malignant phenotype. We should not expect one specific link in the pathways to be altered in all cases. When viewed in this light, it is clear that the Ras signal transduction pathways, which are growth factor regulated, are altered at some point in nearly all hepatomas (Figure 1). The tumor suppressor gene

Table 1 Hepadnavirus Associated Hepatocarcinogenesis Virus: HBV Host: Man

1. Long-term persistent infection associated with H C C 2. Limited immune response to persistent infection 3. Persistent cycles of cell death and regeneration 4. Proliferation of oval cells in precancerous stages 5. Development of distinct precancerous lesions a. Altered hepatic foci b. Neoplastic nodules 6. Development of hepatocellular adenoma 7. Development of H C C types: a. Poorly differentiated b. Well differentiated 8. Viral DNA integrations a. Occur in precancerous liver b. Are clonal in HCCs c. Are associated with a commonly activated protooncogene d. Are associated with function genes e. Often contain viral gene 9. Viral X gene is a. Present in genome b. A viral transcription factor c. A cellular gene transcription factor d. Oncogenic

WHV Woodchuck

GSHV Ground squirrel

DHBV Duck

HBsAg Mouse Yes Yes Yes

HBxgene Mouse -

Yes Yes Yes

? Yes Yes

Yes Yes Yes Yes

Yes Yes Yes Yes

Yes Yes Yes ?

Yes Yes > >

Yes Yes Yes

Yes Yes Yes

Yes Yes Yes

? ? ?

Yes Yes

Yes Yes

?

?

?

?

Yes

-

Yes

Yes

Yes Yes No

Yes Yes Yes

?

?

Yes

Yes

?

?

NA NA NA

NA NA NA

Yes Yes

Yes Yes

? ?

? ?

? ?

? ?

Yes Yes Yes Yes

Yes Yes Yes >

Yes Yes Yes

No

Yes > >

Yes

?

Yes

?

> > ?

> >

? ?

(continues )

Table 1-Continued Virus: Host:

10. Frequency of alterations in cellular proto-oncogenes and growth factor expression in HCCs a. Ras point mutations b. Ras overexpression c. Myc overexpression via amplification d. Myc overexpression via insertional mutagenesis of viral DNA e. IGF2 overexpression f. TGFa overexpression 11. Tumor suppressor genes a. Frequent LOH in 17p b. Frequent point mutations in p53 c. Frequent loss of Rb expression

HBV Man

WHV Woodchuck

GSHV Ground squirrel

DHBV Duck

HBsAg Mouse

H B x gene Mouse

Low Med Med Low

?

?

?

Low High

High Low

? ? ?

Low Low Low Low

?

?

?

High High

High ?

?

>

?

?

High Low

? ?

Highllow Highllow Highllow

?

? ? ?

? ? ?

Low Low Low

? ?

?

?

?

? ?

?

LOW:Not believed to be major mechanism. Most studies do not reveal any change by this mechanism; Med: Some studies positive, others negative. Basically, there is some support for mechanism, but its importance is not firmly established. High: Several studies clearly linking this mechanism to hepatocarcinogenesis. HighILow: Results of studies can reveal high or low incidence depending on regions of the world.

6.

Contributions of Hepadnavirus Research

117

Figure I Schematic diagram of selected signal transduction pathways which lead to transcriptional activation of nuclear genes involved in growth control. In these pathways the following have been identified as oncogenes in various systems: (1) ligands IGF2 and TGFci, (2) receptor tyrosine kinases, (3) SH2-SH3 domain proteins (i.e., SOS), (4) membrane localized protooncogenes, c-Ras and c-Raf, and (5) the nuclear proto-oncogene c-Myc. This illustrates plasticity and redundancy in oncogenic pathways.

p53 which controls the cell cycle, is also clearly implicated in hepatocarcinogenesis (Harris, 1993).

II. Hepatitis B Virus Envelope Protein (HBsAg) Transgenic Mice A. HBsAg (Line 50-4) Transgenic Model of Hepatocarcinogenesis HBsAg expression in transgenic mice is normally non cytopathic; however, when the large envelope protein of HBV is overexpressed in relation to the small envelope protein, the HBsAg particles remain in the endoplasmic reticulum and cause a cytopathic effect (Chisari et al., 1989). In this murine model, chronic hepatocellular injury and inflammation lead to regenerative hyperplasia and eventually to the development of chromosomal abnormalities and HCC (Chisari et al., 1989). This transgenic mouse model thereby reiterates many of the pathophysiological events that occur prior to the

118


development of HCC in chronic HBV infection in humans (Schirmacher et al., 1993).

An extensive survey of the structure and expression of a large panel of proto-oncogenes and tumor suppressor genes in the precancerous liver and HCCs of these HBsAg animals did not reveal any commonly activated genes (Pasquinelli et al., 1992). The genes studied included multiple members of the m y c and ras gene families, raf, and others. These findings suggest that mechanisms other than transcriptional activation or mutation of the studied proto-oncogenes must be responsible for malignant transformation in this model (Pasquinelli et al., 1992). Sequence analysis of the frequently mutated regions of the p53 tumor suppressor gene did not reveal any point mutations in p53 in a large set of tumors from HBsAg mice (Pasquinelli et al., 1992). This is consistent with p53 mutations primarily being a late event in hepatocarcinogenesis (Pasquinelli et al., 1994). The authors suggest that the tumors in HBsAg, line 50-4 mice represent an early stage of hepatocellular tumorigenesis. Interestingly, the only growth factor which is commonly found to be overexpressed is insulin-like growth factor 2 (IGF2) (Schirmacher et al., 1992; Pasquinelli et al., 1994). No other changes are observed in the steady-state expression of hepatocyte growth controlling factors including insulin-like growth factor 1 (IGF1), epidermal growth factor (EGF), EGF receptor (EGFr), hepatocyte growth factor (HGF), c-met, transforming growth factor [3 (TGF[3) or transforming growth factor cx (TGFci) in the HCCs for line 50-4 transgenic mice (Pasquinelli et al., 1994). IGF2 is also expressed in HCCs from three other transgenic models suggesting that it may be a common growth factor involved in hepatocarcinogenesis (Schirmacher et al., 1992). This conclusion has been strongly supported by subsequent research from our group and others (see further discussion below).

B. Noncytopathic HBsAg Transgenic Mice: Role of Cytokines in Gene Regulation The cellular immune response to virus encoded antigens is thought to play an important role in viral clearance and the pathogenesis during acute and chronic infection (Moriyama et al., 1990; Ando et al., 1992; Mondelli et al., 1982). Transgenic mice which produce the small HBsAg in the liver do not develop a cytopathic lesion but rather secrete HBsAg into the blood (Chisari et al., 1985; Babinet et al., 1985; Burk et al., 1988). Chisari and colleagues have reconstructed a cellular immune response by adaptive transfer of HBsAg specific cytotoxic T lymphocyte (CTL) clones into the HBsAg transgenic mice (Moriyama et al., 1990, Ando et al., in press). Their studies show that an important component of the disease process upon adaptive transfer is mediated by inflammatory cytokines secreted by the CTLs, especially

6.


119

~/-interferon (IFN~/) (Guidotti et al., in press). One of the most interesting aspects of these studies, and several which have followed (Guilhot et al., 1993; Gilles et al., 1992; Guidotti et al., 1994), is that HBV gene expression is downregulated by the cytokines, IFN~/, interleukin-2 (IL-2) and tumor necrosis factor cx (TNFcx) which are secreted from HBV specific T cells and the inflammatory cells they recruit. The regulatory effect of IL-2 and TNFcx on HBV gene expression occurs at the post-transcriptional level and is specific for virus transcripts, suggesting that IL-2 and TNFcx activate genes that selectively destabilize HBV mRNA (Guidotti et al., 1994). Most likely, intrahepatic macrophages in the HBsAg mice release TNFcx in response to activation by IL-2, which is produced by activated class II restricted CD4positive T cells. Therefore, the intrahepatic cellular immune response to HBV may play a previously unsuspected role in the biology of HBV infection by modulating HBV gene expression without destroying infected cells (Guidotti et al., 1994). What relevance do these results have to hepatocarcinogenesis? Their relevance is perhaps strongest in helping us understand possible mechanisms by which gene expression is controlled in precancerous lesions in hepadnavirus carriers. This has relevance to any hepatocarcinogenesis regimen in which an inflammatory response occurs. The early stages of chemical hepatocarcinogenesis regimens include a short period in which there is an inflammatory response to liver damage (Scherer 1983; Farber and Sarma 1987; Sell et al., 1987). In WHV-infected liver, precancerous lesions described as altered hepatic foci (AHF) have the characteristics of polyclonal lesions (Table 2). AHF invariably contain focal accumulations of inflammatory cells within them and the steady-state mRNA level of several cellular genes are coordinately upregulated while others are downregulated in the woodchuck precancerous lesions (Yang and Rogler 1991; Yang et al., 1993) (Figures 2 and 3). Is it possible that cytokines secreted from inflammatory

Table 2 Characteristics of Altered Hepatic Foci (AHF) in Woodchucks 1. Histological analysis reveals: a. Hepatocytes strongly basophilic with clear cell lesions occurring only rarely b. Well-defined cord structure c. Altered nuclear phenotype d. Slight or no compression of surrounding parenchyma e. AHF extend from portal tracts to central veins f. Localized inflammatory cell accumulation in nearly all AHF g. Structural characteristics of polyclonal lesions 2. Nonpermissive or semipermissive for WHV transcription and replication 3. Heterogeneousexpression of gammaglutamyltranspeptidase (GGT)with expression strongest near portal tracts

Figure 2 (A) N-myc, IGF2, and WHV expression in persistently infected woodchuck livers. Autoradiograms of serial sections of two tissue blocks hybridized with WHV, N-myc, or IGF2 antisense riboprobes. Areas of positive hybridization appear as dark regions of the sections. (B) Microscopic analysis of gene expression in serial sections of a single altered hepatic focus (AHF). The AHF analyzed is the one marked by arrows in Figure 2A. In all the serial sections the long solid arrows denote the outer boundaries of the AHF, open arrows denote portal tracts, solid arrowheads denote central veins, and curved solid arrows denote specialized structures. The pattern of portal tracts and central veins provide landmarks to facilitate direct comparison of the serial sections. (1) Structural organization of the AHF and surrounding liver

Figure 3 Expression of genes involved in regulating the biological activity of IGF2 and N-myc in woodchuck tissues. Schematic illustration of representative results from in situ hybridization of antisense riboprobes to Normal, normal uninfected woodchuck liver; N § AHF, persistently infected woodchuck liver containing an altered hepatic loci (circular area within the square); T/N, T, hepatocellular carcinoma, N, peritumor liver. The intensity of hybridization is directly proportional to the darkness of the section. Dark circles within AHF note the presence of "foci within loci" which express higher or lower levels of the gene. (Note results for IGF2, N-myc and IGFBP-1.) IGFBP3 is expressed in Ito cells and therefore its pattern is strippled. When expression was uniform for an entire region, the area is uniformly darkened.

viewed by bright field microscopy (hematoxylin and eosin stained' X40). (2) Variable gamma glutamyl transpetidase (GGT) gene expression within the AHF (bright field microscopy, hematoxylin counter stained, X40). Panels 3 to 6 are dark field micrographs in which silver grains appear as bright spots (X40). (3) N-myc expression. (4) IGF2 expression. Curved arrow denotes the highly elevated expression of IGF2 in the subregion of the AHE (5) WHV expression detected with a WHV antisense riboprobe. Note the low and variable expression, which is strongest in GGT positive regions (compare to Panel 2). (6) Histone H3-2 expression. Note higher frequency of histone III positive cells in the foci compared to the surrounding liver tissue.

122


cell lesions in the AHF are responsible for the altered expression of liver specific genes and WHV mRNA described above for HBsAg mice (Guilhot et al., 1993; Gilles et al., 1992; Guidotti et al., 1994). Whether downregulation of a few critical genes required for the maintenance of the hepatocyte differentiated functions would release them from cell cycle controls and start them on a deregulated path toward immortalization and eventually HCCs is an area that needs further investigation. Furthermore, the cytokine effects, once established, may not require a continued cytokine presence. The results from the hepadnavirus models clearly suggest that the roles of various cytokines in the early stages of hepatocarcinogenesis should be more thoroughly investigated.

ni. Woodchuck Hepatitis Virus (WHV) Model of Hepatocarcinogenesis The WHV animal model has recently been the subject of a review (Tennant and Gerin, 1994). Therefore, we will focus on those aspects of the model which directly relate to viral activity in carcinogenesis, and the insights gained into the genetic events associated with precancerous lesions. A. Toxic Oxygen Radicals and WHV Persistent Infection

Point mutations in genes which are responsible for mismatch repair have recently been linked to hereditary nonpolyposis colorectal cancer (Leach et al., 1993). Similarly, point mutations in the MSH1 and MSH2 genes, which are responsible for mismatch repair, lead to microsatellite instability (MIN) within the host genome (Leach et al., 1993). Thus, a direct link has recently been drawn between specific genes involved in a basic genetic mechanism (i.e., mismatch repair), and the genetic instability (i.e., MIN) that predisposes cells to carcinogenic transformation. In this regard, the work of Tennant and his associates is particularly relevant (Liu et al., 1991, 1992, 1993). They have demonstrated that the production of toxic oxygen radicals, specifically nitric oxide (NO), increases in livers persistently infected with WHV. Clearly, the presence of toxic oxygen radicals in hepatocytes will present a risk factor for point mutations and possible development of MIN, especially if the normal detoxification mechanisms in the liver are also compromised. The frequency and the time course of development of MIN in liver of hepadnavirus carriers needs to be determined. B. Insertional Mutagenesis

In addition to predisposing the hepatocyte genome to point mutations, WHV has another mechanism for generating host genomic instability. This

6.


123

involves direct integration of its viral DNA genome into the host genome (Table 3). Fortunately for the host, viral integration is an infrequent event, since integration is not required for productive viral infection. However, viral DNA integration does occur and integrations can be cloned from precancerous liver tissue (Rogler and Summers, 1984). During persistent viral infection, open circular viral DNA molecules, containing a single stranded region resembling a stationary replication fork, are recycled into the nucleus at a low rate (Tuttleman et al., 1986a; Summers et al., 1990). These molecules are suicide substrates for the ubiquitous nuclear enzyme, Topoisomerase I (Topo I). This enzyme can cleave and linearize WHV DNA molecules in vitro and can also mediate their integration into the cellular genome (Wang and Rogler, 1991). The integration reaction has been duplicated in vitro using purified Topo I, purified WHV virion DNA, and a cellular target DNA (Wang and Rogler 1991). The in vitro generated WHV integrations were virtually identical to a subset of those which have been cloned from HCCs (Dejean et al., 1984; Yaginuma et al., 1987; Hino et al., 1989; Shih et al., 1987; Nagaya et al., 1987). The Topo I integration mechanism predicts that integrations cloned from HCCs should occur at preferred Topo I cleavage sites and a detailed analysis of cellular and viral sequences at integration sites has revealed that this is true (Schirmacher et al., 1994). The survey showed that 93% of the illegitimate recombination junctions of hepadnavirus DNA and host DNA occur within, or immediately adjacent to, preferred Topo I cleavage sites (Schirmacher et al., 1994). The idea that integrations generate genetic instability comes from the finding that integrations are found at the site of large and small deletions of

Table 3 Information About Hepadnavirus Integrations a 1. Although integration is not required for viral replication, clonally propagated integrations are present in most HCCs from hepadnavirus carriers. 2. Both linear and arranged viral genomes are present in integrations. 3. Integration is not specific for any viral DNA sequence yet highly preferred integration sites exist, especially in the immediate vicinity of the replication origin (DR1). 4. Topoisomerase I, an abundant nuclear enzyme, can mediate illegitimate recombination of hepadnavirus DNA with cellular DNA in vitro. The integrations produced in vitro closely resemble those which occur in vivo. 5. Preferred Topoisomerase I cleavage sites are present at over 90% of the crossover sites in hepadnavirus integrations cloned from HCCs. 6. Integration increases carcinogenic risk by causing micro and macro deletions, inverted and direct duplications, and translocations of host genomic DNA. 7. Activation of c-myc genes by insertion of hepadnavirus DNA is a commnon event in the genesis of HCC in woodchucks. 8. Integrations often contain viral X genes, and/or a truncated PreS2/S, both of which function as transcriptional transactivators; the X gene is also an oncogene. aSee text for reference citations and abbreviations.

124


cellular DNA (Rogler et al., 1985; Nakamua et al., 1988), at direct and inverted duplications of cellular DNA (Yaginuma et al., 1985; Mizusawa et al., 1985), and as linker molecules at sites of chromosome translocations (Hino et al., 1986). These findings do not allow us to distinguish whether HBV directly causes genome instability by integrating into the genome or becomes linked at sites in cellular DNA which have already been cleaved and not repaired correctly. Recent studies in our laboratory, using a cell line which replicates hepadnaviruses, have shown that viral DNA integrations can also excise from the cellular genome at a relatively high frequency (Gong and Rogler, unpublished data). Further measurements of the frequency of integration and excision of integrations in stable and unstable genetic backgrounds will help resolve the questions of cause and effect in regard to integrations and genome stability. One previous study has presented evidence for increased in vitro recombination of DNA adjacent to integrated HBV DNA (Hino et al., 1991). The presence of clonal viral DNA integrations, sometimes as the sole form of viral DNA in an HCC, fueled the speculation that they could activate expression of proto-oncogenes involved in hepatocarcinogenesis. Initial cloning experiments of WHV integrations from woodchuck HCCs did not reveal a common integration site (Ogston et al., 1982). However, after many years of diligent searching, the research group headed by Tiollais and Buendia discovered a proto-oncogene in the woodchuck genome which is commonly activated by WHV DNA integration in a large majority of woodchuck HCCs (Fourel et al., 1990; Wei et al., 1992). The chromosome locus identified turned out to be a functional retroposon of the cellular N-myc gene, which they designated N-myc-2 (Fourel et al., 1990). They demonstrated that WHV integration activated transcription of the normally silent N-myc-2 retroposon by an enhancer insertion mechanism. The WHV enhancer has many binding sites for liver specific transcription factors and is a strong enhancer, which when inserted into the 3' untranslated region of N-myc-2, activates transcription from a cryptic promoter that lies within the first exon of N-myc-2, upstream from the translation start site (Wei et al., 1992). Further work by their group has shown that additional integrations, very distant from N-myc-2, but at a common site on the same chromosome, may also activate N-myc expression in those cases where a WHV DNA integration is not found immediately 5' to the N-myc-2 gene, or within the 3' noncoding region of N-myc-2 (Fourel et al., 1993). In those tumors that do not express N-myc, c-myc is overexpressed due to either WHV DNA integration or another mechanism (Hansen et al., 1993; Hsu et al., 1990). Therefore, WHV DNA integration has revealed a cellular proto-oncogene, which may fit the criteria of a "gatekeeper gene" for HCC in woodchucks. The gatekeeper concept states that the gatekeeper gene is a gene whose expression must be altered in order for a tumor to

6. Contributions of Hepadnavirus Research

125

develop in a particular organ or cell type. Another example of a gatekeeper gene would be the APC gene in certain hereditary forms of colon carcinoma (Fearon and Vogelstein, 1990).

C. Integration and Human HCC As clear as the function of viral DNA integration appears to be in woodchucks, it is unclear in all the other animal models and in humans (Table 3). After analyzing the flanking sequences of HBV integrations from many tumors from many sources around the world, a common or even frequent integration site has not yet been identified (Nagaya et al., 1987). Chromosomes 17 and 11, both of which contain tumor suppressor genes, and protooncogenes, have been shown to have the highest frequency of HBV integrations. One of the HBV integrations in chromosome 11 was in the vicinity of the Wilms tumor suppressor gene on chromosome 11p13; the HBV integration caused a large deletion in that segment of the chromosome (Rogler et al., 1985). Another HBV integration was identified at a chromosome translocation in which chromosome 17q21-22 was linked to chromosome 18q 11 via HBV DNA (Hino et al., 1986). The breast tumor gene, B r a c l , was recently localized to chromosome 17q21 and ongoing studies are being conducted to determine whether the HBV associated chromosome breakpoints are within the small region that has been recently identified to contain the Bracl gene (Bowcock et al., 1993). Additional HBV integrations have been identified in association with a truncated cyclin A gene (Wang et al., 1990), and the retinoic acid receptor beta gene (deThe et al., 1987), as well as other cellular genes whose functions are not as well established (Etiemble et al., 1989; Ochiya et al., 1986). These data demonstrate that HBV integrations can be utilized to identify new cellular genes which are potentially involved in tumorigenesis. The cloning of a single HBV integration in a cyclin A gene provides a telltale sign that additional alterations in the numerous cell cycle control genes may be important in hepatocarcinogenesis. The HBV integration into the retinoic acid receptor beta gene suggests that altering the expression of genes that control hepatocyte differentiation may also be a step in hepatocarcinogenesis in HBV carriers. D. Analysis of Precancerous Lesions and HCCs: The Case for a Role of IGF2 in Tumor Promotion In order to study the early events of hepatocarcinogenesis in WHV carrier woodchuck liver, we have utilized in situ hybridization and immunocytochemistry to study the expression patterns of genes in AHE Some of the histological characteristics of AHF in woodchuck liver are summarized

126


in Table 2. AHF are clearly evident in woodchuck liver as foci of hepatocytes which have a strongly basophilic cytoplasm. The histological characteristics of the AHF strongly suggest that they are polyclonal in nature, although subregions (loci within foci) have characteristics of clonally propagated cells. A schematic summary of the in situ hybridization results to be discussed in this section is presented in Figure 3 and typical data used to support the schematic summary are presented in Figure 2. 1. R a t i o n a l e a n d Genetic Predisposition to H C C

Our original rationale for studying the expression of IGF2 in AHF and HCCs of woodchucks was based on the discovery of HBV integrations in chromosome 11 (Nagaya et al., 1987; Rogler et al., 1985), and the finding of a high frequency of loss of heterozygosity (LOH) on chromosome 11p in human HCCs (Wang and Rogler, 1988). In that study, we discovered that LOH was limited to the 11p15 region in some HCCs but in others it was limited to chromosome 11p13, which is the site of the Wilms tumor suppressor gene (WT1). The WT1 gene at 11p13 is not expressed in adult liver, leading to the hypothesis that a second tumor suppressor might be located at the distal end of chromosome 11 at 11p15 (Wang and Rogler, 1988). Genetic evidence also suggested that genes on chromosome 11p15 were involved in hepatocarcinogenesis because duplications of the distal end of chromosome 11, encompassing 11p15 are characteristic of Beckwith Wiedeman syndrome (Best and Hoeksha, 1981; Waziri et al., 1983). This genetic syndrome is associated with overgrowth and childhood tumors, and particularly with hepatomegaly and heptoblastoma (Koufos et al., 1985). Since hepatoblastoma arises in children with a fairly short latency, this strongly suggests that genes important in hepatocarcinogenesis are located in that region of chromosome 11p. Rapid developments in the recent past have strongly supported this hypothesis and pointed to specific genes in chromosome 11p15 as important candidates for roles as promoters and suppressors of tumorigenesis. Chromosome 11p15 contains a whole series of genes that have been linked to growth and malignancy. The H-ras proto-oncogene at 11p15.5 has been widely implicated in tumorigenesis, and overexpression of ras protooncogenes is a common event in HCCs (Tabor 1994). The other gene cluster of great interest at 11p15 includes the insulin, IGF2, and H 1 9 genes. Insulin is not expressed in normal liver whereas IGF2 is expressed at a low level in normal human liver and woodchuck liver but not in adult mouse liver. Since IGF2 has mitogenic activity in cell cultures (Cohick and Clemmons, 1993) and is expressed in a wide variety of tumors (Fu et al., 1988; Cullen et al., 1992), we focused on the expression of IGF2 in the woodchuck liver and HCCs. The H 1 9 gene produces an abundant developmentally regulated transcript of unknown function in normal embryos (Willison, 1991).


127

2. IGF2 Expression in A H F a n d H C C s

Studies of IGF2 expression in woodchuck liver and HCCs strongly suggest that it plays a role in hepatocarcinogenesis. IGF2 is overexpressed in over 90% of AHF in precancerous woodchuck liver (Yang and Rogler, 1991). Interestingly, two levels of IGF2 expression are present in AHF, a moderate level throughout the foci and, in about 5% of loci, we observe localized regions (foci within loci) of highly elevated expression which is equivalent to the high levels most often observed in HCCs (Figure 2). We have suggested that this may be clonal selection of high expressing cells within the AHF because the small regions of very high IGF2 expression in AHF do have higher mitotic indices (Yang and Rogler, 1993). Interestingly, similar findings of localized high level IGF2 expression in precancerous lesions and a higher degree of malignancy in IGF2 producing tumors has recently been reported for SV40 TAg associated pancreatic carcinoma (Chrislofori et al., 1994). 3. IGF2 Functions as a T u m o r P r o m o t e r in Transgenic M i c e

The above studies establish that the pattern of IGF2 expression in precancerous lesions and HCCs is consistent with a causal role in carcinogenesis. In order to determine whether IGF2 was causally involved in tumorigenesis, two new animal models were utilized. We developed IGF2 transgenic mice which overexpress IGF2 primarily in the liver (Rogler et al., 1994). These animals develop HCC plus tumors from organs that do not express the transgene suggesting that the IGF2 in our mice functions by both autocrine and endocrine mechanisms (Rogler et al., 1994). The tumors arise at a slightly higher incidence than in control animals (22% versus 5% in controis), and the tumors arise with a very long latency period (18 to 24 months). These data suggest that although IGF2 is a weak tumor initiator, it can promote the growth of tumors which arise in mice. In order to test this possibility we have taken two approaches. The first approach was to mate the IGF2 transgenic mice with TGFc~ transgenic mice. The TGFc~ mice develop liver tumors with a latency of approximately 13 to 16 months (Jhappan et al., 1990; Takagi et al., 1992) and initial results of such matings have shown that the presence of the IGF2 transgene decreases the latency time for development of liver tumors in the TGF0~ mice by approximately 4 to 6 months (C. E. Rogler, L. E. Rogler, and G. Merlino, unpublished data). Furthermore, at short latency times, a much higher percentage of the tumors have a clearly malignant phenotype. These data suggest that IGF2 expression plays a direct role in tumor progression in the liver and that it can function in cooperation with another liver growth factor. The second approach is to determine whether knocking out the IGF2 gene can prevent HCC. In this approach we crossed TGF0~ transgenic mice

128


with mice in which the IGF2 gene had been knocked out (mice kindly provided by Efstradiadis, Robertson and DeChiara, Columbia University). This experiment is nearing completion. However, a parallel experiment using the IGF2 knockout animals has been completed by Hanahan and colleagues at the University of California, San Francisco. Knocking out the IGF2 gene in mice that express the SV40 T A g in pancreatic islets significantly reduces both tumor growth in islets and the degree of tumor malignancy (Chrislofori et al., 1994). Tumors which developed in the SV40 TAg mice are clearly malignant whereas SV40 T Ag plus IGF2 knockout mice develop small benign tumors. The absence of IGF2 is associated with a fivefold increase in apoptosis in the benign tumors suggesting that the presence of IGF2 protects cells in tumors from apoptosis. However, knocking out IGF2 did not block tumor formation per se. Thus, tumor initiation in the SV40 TAg model depends on the action of SV40 TAg plus at least one additional unknown oncogene (Chrislofori et al., 1994). It was proposed that without IGF2 the action of this oncogene may cause cells to undergo apoptosis as opposed to mitogenesis. This hypothesis is based on work from other systems that shows the overexpression of c-myc can induce apoptosis in some immortalized cell lines (Shi et al., 1992). Cells can also be rescued from apoptosis by overexpression of growth factors or other cytokines which block the apoptotic pathway (Rodriguez-Tarduchy et al., 1992; Williams and Smith, 1993). The coordinate expression of IGF2 along with N-myc in AHF of woodchucks (Figure 2) may prevent hepatocytes from entering the apoptosis pathway. In fact, recent studies have shown that IGF2 can block apoptosis in several in vitro systems (Evans, 1994). Bcl-2 which also can prevent apoptosis is not normally expressed in liver and it is not induced in AHF of woodchucks (L. E. Rogler, unpublished data). 4. IGF2 Expression in the Context of Multiple Receptors, Regulatory Proteins, and Genetic Imprinting The biological activity of IGF2 is mediated by its receptors and insulin-like growth factor binding proteins (IGFBPs) (Table 4). In order to understand IGF2 expression in the context of its receptor and binding proteins in woodchuck liver, serial sections of the liver specimens that were used to determine IGF2 and N-myc expression were used to study expression of IGF receptor and IGFBP genes. The general conclusions of in situ hybridization studies are schematically summarized in Figure 3. The most striking results of this study are the coordinate upregulation of IGF2, N-myc, and IGFBP-4 and downregulation of IGFBPs 1 and 2 in AHF and HCCs and the localization of IGFBP-3 expression to Ito cells (represented by a speckled pattern in Figure 3). IGFBP-2 expression is generally downregulated in AHF; however, subregions of AHF sometimes expressed IGFBP-2 at normal levels. These

6.


129

Table 4 Major Proteins Which Regulate IGF Biological Activity Ligands

Binding proteins

Receptors

Insulin IGF1 IGF2

IGFBP-1 IGFBP-2 IGFBP-3 IGFBP-4 IGFBP-5 IGFBP-6 IGFBP-?

Insulin receptor IGF1 receptor M6P/IGF2 receptor

regions are immediately adjacent to portal tracts and are represented in Figure 3 as a dark loci within the AHE While some IGFBPs are known to interfere with IGF2 activity, others can enhance its activity in certain cell types. Further work is necessary to determine whether IGFBPs 1 and 2 interfere with IGF2 action in liver in which case downregulation in tissues would enhance the effectiveness of IGF2 or whether they enhance IGF2 activity and their loss reduces the biological activity of IGF2. In the liver, IGF2 signal transduction must be mediated through either the insulin receptor or the mannose 6-phosphate/insulin-like growth factor 2 receptor (M6P/IGF2r) since IGF1 receptors are not present (Cohick and Clemmons, 1993). Northern blot analysis of woodchuck liver and HCCs has shown that both insulin and M 6 P / I G F 2 r genes are expressed, as expected (Yang and Rogler, manuscript in preparation) (Figure 3). IGF2 polypeptides in woodchuck HCCs accumulate in the perinuclear region of cells and colocalize with M6P/IGF2r (Yang and Rogler, 1991). Some evidence exists that the M6P/IGF2r may signal through G proteins when bound by IGF2 (Nishimoto et al., 1989; Okamoto et al., 1990); however, they are generally believed to target IGF2 for degradation in lysosomes. Therefore, IGF2 media..ed signal transduction is most likely mediated through its cross-reaction with insulin receptors in the liver (Cohick and Clemmons, 1993).

IV. Hepadnavirus X Gene Encodes an Oncogenic Transcriptional Transactivator A. Background We will present a brief historical overview of major aspects of HBx gene research and then focus more in-depth on the recent transgenic and cell culture models that have clearly demonstrated that the X gene can function

130


as an oncogene (Conclusions, Table 5). After the HBV genome was cloned, sequence analysis revealed the presence of an open reading frame whose protein sequence did not match any of the viruses structural proteins (Galibert et al., 1979). Since this gene's function was u n k n o w n , it was given the name, X protein, as opposed to the name of the other viral genes, S (surface antigen), C (nucleocapsid core protein), and P (polymerase). Initial studies suggested that the X gene does not have a highly active promoter that produced a unique X gene message; however, later studies have clearly shown that it does have an active p r o m o t e r (Treinin and Laub, 1987). X gene mutagenesis studies and the lack of an X gene in duck hepatitis B virus (DHBV) initially led to the conclusion that the X gene is not required for viral replication. Recent in v i v o infection studies with m u t a n t W H V viruses lacking the X gene, however, demonstrate that the W H x gene is required for establishment of viral infection in v i v o (Chen et al., 1993). The position of the X gene in the m a m m a l i a n hepadnaviruses ( d o w n s t r e a m from the core and envelope genes and in the same transcriptional orientation), led to speculation that it might have originated as a cellular gene which was picked up by the m a m m a l i a n hepadnaviruses (Miller and Robinson, 1986). H o w ever, to this date a cellular homologue has not been identified. The presence of the complete X open reading frame in hepadnavirus D N A integrations cloned from H C C s while the core (C) and surface antigen (S) genes in the same integrations are rearranged (Nagaya et al., 1987; Ogston et al., 1982),

Table 5 Information about Hepadnavirus X Genea 1. X-ORF encodes a 17-kDa promiscuous transcriptional activator (transactivator). a. Examples of a few enhancers and promoters transactivated include: 1. HBV enhancer + core promoter 2. SV40 enhancer + early promoter 3. LTRs for HSV-TK, HIV~ and RSV 4. c-Myc 5. Many others 2. No cellular homologue yet identified. 3. X proteins of HBV, WHV, and GSHV are active in transactivation assays. 4. HBx acts through multiple cell type-specific transcription factors (AP-1, AP-2, CREB, ATF~ NFKB). 5. Transactivation of HBV and heterologous genes occurs when the X gene is expressed in its native state during productive infection. 6. X gene is required for WHV productive infection. 7. X gene is expressed during acute and persistent infection. 8. X gene is retained in many integrations in HCCs and transactivation activity is maintained. 9. X gene has transforming activity in SV40T Ag immortalized murine hepatocytes. 10. HBx transgenic mice that express high levels of X protein develop HCC after one year. 11. HBx may function through both the PKC and Raf signal transduction pathways. aSee text for reference citations and abbreviations.

6.


131

led to additional speculation that the X gene might function in hepatocarcinogenesis. The discovery that the X protein was a transcriptional transactivator added a great deal of focus to the field and provided a potential mechanism by which the X gene might function in viral replication and hepatocarcinogenesis (Twu and Schloemer, 1987; Spandau and Lee, 1988). Antibodies to X protein were detected after acute infection demonstrating that the X protein was expressed in productive infections (Feitelson et al., 1990; Pershing et al., 1986). In vitro replication systems were used to demonstrate that X proteins of the mammalian viruses [HBV, ground squirrel hepatitis virus (GSHV) and WHV] transactivate viral promoters during productive infections (Colgrove et al., 1989). Cloned integrations from HCCs that contain X genes were utilized to demonstrate that integrated X genes are transcribed and have transactivation activity (Yang et al., 1993; Wollersheim et al., 1988; Takada and Koike, 1990b). Many studies of the transactivation mechanism of the X protein demonstrate that it is a promiscuous transactivator that acts through multiple cell type specific transcription factors some of which include AP-1, AP-2, CREB, ATF-2, NFKB, cEBP, and others (Seto et al., 1990; Maguire et al., 1991). B. X Gene Transactivation Mechanism

Researchers in the field of X transactivation agree the X protein is not a DNA binding protein and that its activity is manifest through proteinprotein interactions. The search for proteins that interact with the X protein in vivo is ongoing in many laboratories. This line of research is important because it will give us insights into the molecular mechanism of transactivation and may provide valuable insights into which gene pathways are important to start hepatocytes on the path to malignant transformation. We can say this because the X protein has now been shown to have oncogenic activity in vivo in transgenic mice (Kim et al., 1991; Koike et al., 1994). X protein is not an acute transforming protein but rather appears to be a slow acting oncogene which induces metabolic alterations (e.g., glycogen accumulation) in hepatocytes in a zonal pattern (near central veins) when expressed in the liver under its own native enhancer-promoter (Koike et al., 1994). The mechanism of X protein transactivation is, and has been, the focus of a great deal of research. Work in several laboratories has pointed to two main pathways in which the X protein may function indirectly to regulate transcription of a variety of promoters. These include activation of the protein kinase C (PKC) pathway (Kekule et al., 1992) through mobilization of PKC to the plasma membrane, and activation of a protein kinase cascade pathway through activation of Raf (Cross et al., 1993), which serves a

132


serine threonine protein kinase kinase kinase (see signal transduction pathway outlined in Figure 1). Evidence also suggests that the HBx protein might function as a serine threonine protease inhibitor (Takada and Koike, 1990b). Such a function of X might fulfill the broad requirements for its promiscuous action on enhancers with unrelated sequence motifs. Interestingly, the early effects of X gene expression in transgenic mice include alterations in metabolism leading to clear cell foci with glycogen accumulation. This could reflect early actions on gene pathways involved in maintenance of hepatocyte differentiated functions. C. In Vitro Assays for Hepadnavirus Transforming Activity

Rather than extensively discuss the literature on X protein transformation activity, we will focus on a single experimental system dealing with SV40 T antigen immortalized hepatocytes (Paul et al., 1992). SV40 TAg is a powerful oncogene because it binds and functionally inactivates at least two important cellular proteins, Rb and p53 (Levine et al., 1994; Harris and Hollstein, 1993). The tumor suppressor functions of these proteins are due, at least in part, to their ability to control the cell cycle and the fact that their removal or inactivation deregulates the cell cycle. Cells which have experienced DNA damage progress through the cell cycle, when p53 is knocked out, rather than arresting in G 1 (Levine et al., 1994; Harris and Hollstein, 1993). Thus, mutations become fixed in the genome increasing the risk of malignant transformation of that cell. Expression of SV40 TAg in hepatocytes deregulates the cell cycle and the hepatocytes behave as immortalized cells. The immortalized phenotype is stable in culture and the immortalized cells do not exhibit a malignant phenotype (Paul et al., 1992). HBV expression in these cells causes malignant transformation as judged by growth in soft agar and production of HCCs in nude mice (H6hne et al., 1990). Furthermore, exclusive expression of HBx protein in the cells also causes malignant transformation (Seifer et al., 1991). Work on DHBV replication in primary hepatocytes shows that viral replication per se does not lead to malignant transformation (Tuttleman et al., 1986b). Therefore, once cell cycle controls, provided by tumor suppressors, are removed HBV and particularly the X protein have a greater oncogenic potential. The recent report that HBx protein can directly interact with the p53 tumor suppressor protein provides a potential mechanism by which HBx might directly deregulate cell cycle control in hepatocytes (Wang et al., 1994).

D. HBx Transgenic Mice Develop HCC There are now two reports that HBx expression in transgenic mice can lead to a multistep progression toward HCC (Kim et al., 1991; Koike et al.,


133

1994). Since the second report provides a more detailed description of the pathophysiology, we will focus our analysis on it. When the HBx gene was expressed under the influence of the natural HBV enhancer and X promoter, transgenic mice with a high level of HBx expression clearly developed HCC beginning around 13 months old. Mice expressing low levels of HBx develop HCC at a low frequency and with a longer latency. These results may explain why HCC has not been observed in some other lines of HBx transgenic mice which express only low levels of the HBx transgenes (Lee et al., 1990). In young HBx transgenic mice, a high level of HBx expression occurs in the centrilobular region (zone 3), and these regions subsequently develop increased vacuolation at which time increased HBx expression in the centrilobular region is evident (Kim et al., 1991; Koike et al., 1994). By the age of 4 months, aneuploid peaks of DNA are present in nuclear preparations of the HBx liver and this corresponds to the development of dysplastic hepatocytes with large nuclei which appear within the altered foci around central veins. At 7 months of age, the regions around the central veins that selectively express higher HBx protein also have an increased DNA synthesis as measured by incorporation of BrdU into newly replicated DNA. The presence of aneuploidy suggests that X protein expression leads to genetic instability in the liver, and that this could be due to alterations in the functional levels of p53 in hepatocytes (Koike et al., 1994). HBx protein expression did not induce an inflammatory response, or elevate glutamic-pyruvic transaminase (SGPT) in blood which is an indicator of ongoing liver damage. Thus cell death, which must occur, most likely occurs via apoptosis which does not elevate blood SGPT levels. Experiments to determine whether p53 is bound to the HBx protein in this transgenic model or whether p53 undergoes genetic alteration (i.e., point mutations or deletions) are of great interest. A change in frequency of apoptotic bodies was not detected in preneoplastic foci of HBx transgenic mice compared to control mice; however, extensive quantitative data on the apoptotic index of foci were not presented in this study. One might predict that if HBx protein really does bind and inactivate p53 in vivo, there would be no need to further mutate p53, and wild type p53 genes would be found in HCCs from these mice. As the HBx mice age and the liver lesions progress to HCC, the expression of HBx protein increases suggesting that a selection for high expressing cells occurs. The mice develop multifocal HCC beginning at 13 months of age. The absence of a distinct period of hepatocyte death and an inflammatory response in the liver, however, separates this model from the HBsAg model of hepatocarcinogenesis. The progressive series of precancerous lesions, adenoma, and HCC which occur in HBx mice are common features in all the other transgenic models. While HBx is capable of initiating metabolic and/or genetic alterations in hepatocytes which lead to HCC, it must func-

134


tion in conjunction with other oncogenes/growth factors and/or tumor suppressor genes for malignant transformation to be complete. The activity of HBx as a transactivator and oncogene provides a valuable new tool to begin to dissect the early events in hepatocarcinogenesis in the HBx transgenic mice.

V. C o n c l u s i o n s This article has not attempted to review all the literature on hepadnaviruses and HCC. Instead, we have chosen to focus on a few experimental models that have provided specific insights into hepatocarcinogenesis in hepadnavirus carriers which may be of general significance. The new transgenic models which utilize the HBsAg and the HBx genes should now be exploited to further understand the molecular genetics of HCC. These models are directly relevant to H C C in humans since both the HBsAg and HBx genes are expressed in precancerous liver and HCCs from humans. In addition, these models have at their disposal the vast resources of the human and murine genome projects. In addition, the W H V model provides many opportunities to further analyze molecular genetic changes that occur in a natural host-virus system which is also very relevant to the human disease. So far, work on these models has pointed to the m y c family of protooncogenes and the I G F 2 gene as important genes in the initiation of transformation and tumor growth and malignant progression, respectively. The effects of HBx on the PKC and Raf 1 signal transduction pathways may also lead to the identification of new genes of importance in hepatocarcinogenesis. When the individual genetic components (e.g., Ras, Raf, etc.) of the interactive signal transduction pathways are more fully understood, common mechanisms between HCC and other cancers will undoubtedly become more evident.

References Ando, K., Moriyama, T., Guidotti, L. G., Wirth, S., Schreiber, R. D., Schlicht, H. J., Huang, S., and Chisari, E V. (1992). Mechanisms of class I restricted immunopathology:A transgenic model for fulminant hepatitis. J. Exp. Med. 178, 1541-1554. Ando, K., Guidotti, L. G., Wirth, S., Ishikawa, T., Missale, G., Moriyama, T., Schreiber, R. D., Schlicht, H. J., Huang, S. N., and Chisari, E V. (1994). Class I-restricted cytotoxicT lymphocytes are directly cytopathic for their target cells in vivo. J. Immunol. 152, 3245-3253. Babinet, C., Farza, H., Morello, D., Hadchovel, M., and Pourcel, C. (1985). Specificexpression of hepatitis B surface antigen (HBsAg) in transgenic mice. Science 230, 1160-1163. Best, L. G., and Hoeksha, R. E. (1981). Wiedemann-Beckwith Syndrome:Autosomal dominant inheritance in a family. Am. J. Med. Genet. 9, 291-299. Bowcock, A. M., Anderson, L. A., Friedman, L. S., Black, D. M., Osborne-Lawrence, S.,

6.


135

Rowell, S. E., Hall, J. M., Solomon, E., and King, M. C. (1993). THRA1 and D17S183 flank an interval of

Liver Regeneration And Carcinogenesis: Molecular and Cellular Mechanisms

Atherosclerosis: Molecular and Cellular Mechanisms

Cellular Respiration and Carcinogenesis

Hypoxic pulmonary vasoconstriction: cellular and molecular mechanisms

Cancer metastatis, molecular and cellular mechanisms and clinical intervention

Molecular and Cellular Signaling

Molecular and Cellular Biophysics

Molecular and Cellular Signaling

Molecular Chaperones and Folding Catalysts: Regulation, Cellular Functions and Mechanisms

Cellular and Molecular Neurobiology,

Molecular and Cellular Biophysics

Cellular and Molecular Neurobiology

Molecular and Cellular Signaling

Ethanol and the Liver: Mechanisms and Management

Liver Regeneration (De Gruyter Textbook)

Chronic Obstructive Pulmonary Disease: Cellular and Molecular Mechanisms

Vascular Liver Disease: Mechanisms and Management

Brucella: Molecular and Cellular Biology

Molecular and Cellular Iron Transport

Molecular and Cellular MR Imaging

Cellulose. Molecular and Cellular Biology

Yersinia: Molecular and Cellular Biology

Molecular and Cellular Iron Transport

Integrated molecular and cellular biophysics

Molecular Carcinogenesis and the Molecular Biology of Human Cancer

Insulin Resistance and Cancer: Epidemiology, Cellular and Molecular Mechanisms and Clinical Applications

Molecular and Cellular Foundations of Biomaterials

Cellular and Molecular Immunology, 5th Edition

Cellular and Molecular Immunology, Updated 6th Edition

Cellular and Molecular Biology of Metals

Liver Regeneration And Carcinogenesis: Molecular and Cellular Mechanisms

Atherosclerosis: Molecular and Cellular Mechanisms

Cellular Respiration and Carcinogenesis

Hypoxic pulmonary vasoconstriction: cellular and molecular mechanisms

Cancer metastatis, molecular and cellular mechanisms and clinical intervention

Molecular and Cellular Signaling

Molecular and Cellular Biophysics

Molecular and Cellular Signaling

Molecular Chaperones and Folding Catalysts: Regulation, Cellular Functions and Mechanisms

Cellular and Molecular Neurobiology,

Molecular and Cellular Biophysics

Cellular and Molecular Neurobiology

Molecular and Cellular Signaling

Ethanol and the Liver: Mechanisms and Management

Liver Regeneration (De Gruyter Textbook)

Chronic Obstructive Pulmonary Disease: Cellular and Molecular Mechanisms

Vascular Liver Disease: Mechanisms and Management

Brucella: Molecular and Cellular Biology

Molecular and Cellular Iron Transport

Molecular and Cellular MR Imaging

Cellulose. Molecular and Cellular Biology

Yersinia: Molecular and Cellular Biology

Molecular and Cellular Iron Transport

Integrated molecular and cellular biophysics

Molecular Carcinogenesis and the Molecular Biology of Human Cancer

Insulin Resistance and Cancer: Epidemiology, Cellular and Molecular Mechanisms and Clinical Applications

Molecular and Cellular Foundations of Biomaterials

Cellular and Molecular Immunology, 5th Edition

Cellular and Molecular Immunology, Updated 6th Edition

Cellular and Molecular Biology of Metals

Recommend Documents