Food Factors for Health Promotion (Forum of Nutrition Vol 61)

Food Factors for Health Promotion

Forum of Nutrition Vol. 61

Series Editor

Ibrahim Elmadfa

Vienna

Food Factors for Health Promotion Volume Editor

Toshikazu Yoshikawa Kyoto Prefectural University of Medicine, Kyoto

53 figures, 1 in color and 12 tables, 2009

Basel · Freiburg · Paris · London · New York · Bangalore · Bangkok · Shanghai · Singapore · Tokyo · Sydney

Toshikazu Yoshikawa Molecular Gastroenterology and Hepatology Kyoto Prefectural University of Medicine Graduate School of Medical Science Kyoto, Japan

Library of Congress Cataloging-in-Publication Data International Conference on Food Factors for Health Promotion (2007 : Tokyo, Japan) Food factors for health promotion / volume editor, Toshikazu Yoshikawa. p. ; cm. -- (Forum of nutrition, ISSN 1660-0347 ; v. 61) Includes bibliographical references and index. ISBN 978-3-8055-9097-6 (hard cover : alk. paper) 1. Functional foods--Congresses. I. Yoshikawa, Toshikazu. II. Title. [DNLM: 1. Food--Congresses. 2. Nutritive Value--Congresses. 3. Metabolic Phenomena--Congresses. 4. Nutrition Therapy--Congresses. 5. Nutritional Physiological Phenomena--Congresses. W1 BI422 v.61 2009 / QU 145.5 I615f 2009] QP144.F85I58 2009 613.2--dc22 2009010218

Bibliographic Indices. This publication is listed in bibliographic services, including Current Contents® and PubMed/MEDLINE. Disclaimer. The statements, opinions and data contained in this publication are solely those of the individual authors and contributors and not of the publisher and the editor(s). The appearance of advertisements in the book is not a warranty, endorsement, or approval of the products or services advertised or of their effectiveness, quality or safety. The publisher and the editor(s) disclaim responsibility for any injury to persons or property resulting from any ideas, methods, instructions or products referred to in the content or advertisements. Drug Dosage. The authors and the publisher have exerted every effort to ensure that drug selection and dosage set forth in this text are in accord with current recommendations and practice at the time of publication. However, in view of ongoing research, changes in government regulations, and the constant flow of information relating to drug therapy and drug reactions, the reader is urged to check the package insert for each drug for any change in indications and dosage and for added warnings and precautions. This is particularly important when the recommended agent is a new and/or infrequently employed drug. All rights reserved. No part of this publication may be translated into other languages, reproduced or utilized in any form or by any means electronic or mechanical, including photocopying, recording, microcopying, or by any information storage and retrieval system, without permission in writing from the publisher. © Copyright 2009 by S. Karger AG, P.O. Box, CH–4009 Basel (Switzerland) www.karger.com Printed in Switzerland on acid-free and non-aging paper (ISO 9706) by Reinhardt Druck, Basel ISSN 1660–0347 ISBN 978–3–8055–9097–6 e-ISBN 978–3–8055–9098–3

Contents

VII XI

List of Contributors Preface Yoshikawa, T. (Kyoto) Genomics

1 10

25

Genomics for Food Functionality and Palatability Abe, K. (Tokyo) Lipid Metabolism and Nutrigenomics – Impact of Sesame Lignans on Gene Expression Profiles and Fatty Acid Oxidation in Rat Liver Ide, T.; Nakashima, Y.; Iida, H.; Yasumoto, S.; Katsuta, M. (Tsukuba) Genome Science of Lipid Metabolism and Obesity Takahashi, N.; Goto, T.; Hirai, S.; Uemura, T.; Kawada, T. (Kyoto) Proteomics

39

Oxidative Stress-Induced Posttranslational Modification of Proteins as a Target of Functional Food Naito, Y.; Yoshikawa, T. (Kyoto) Bioavailability and Safety

55 64

75

Absorption and Function of Dietary Carotenoids Nagao, A. (Tsukuba) Metabolism of Flavonoids Wang, Y.; Ho, C.-T. (Brunswick, N.J.) β-Carotene Degradation Products –Formation, Toxicity and Prevention of Toxicity Siems, W. (Bad Harzburg); Salerno, C.; Crifò, C. (Rome); Sommerburg, O. (Heidelberg); Wiswedel, I. (Magdeburg)

V

Antioxidants 87

Dietary Flavonoids as Antioxidants Terao, J. (Tokushima) Life-style Related Diseases

95

104 117 129

136 147

Inflammatory Components of Adipose Tissue as Target for Treatment of Metabolic Syndrome Yu, R.; Kim, C.-S.; Kang, J.-H.; (Ulsan) Soybean Isoflavones in Bone Health Ishimi, Y. (Tokyo) Probiotics in Primary Prevention of Atopic Dermatitis Ji, G.E. (Seoul) Astaxanthin Protects Neuronal Cells against Oxidative Damage and Is a Potent Candidate for Brain Food Liu, X.; Osawa, T. (Nagoya) Function of Marine Carotenoids Miyashita, K. (Hakodate) Exercise and Food Factors Aoi, W. (Kyoto) Chemoprevention and Cancer

156 170 182

193

204

217 218

VI

Molecular Basis for Cancer Chemoprevention by Green Tea Polyphenol EGCG Tachibana, H. (Fukuoka) Chemoprevention by Isothiocyanates: Molecular Basis of Apoptosis Induction Nakamura, Y. (Okayama) Ginger-Derived Phenolic Substances with Cancer Preventive and Therapeutic Potential Kundu, J.K.; Na, H.-K.; Surh, Y.-J. (Seoul) Chemoprevention with Phytochemicals Targeting Inducible Nitric Oxide Synthase Murakami, A. (Kyoto) Chemoprevention of Tocotrienols: The Mechanism of Antiproliferative Effects Wada, S. (Kyoto) Author Index Subject Index

Contents

List of Contributors

Professor Keiko Abe

Professor Chi-Tang Ho

Department of Applied Biological Chemistry Graduate School of Agricultural and Life Sciences The University of Tokyo 1-1-1 Yayoi Bunkyo-ku Tokyo, Japan

Department of Food Science Rutgers University 65 DudleyRoad New Brunswick, USA

Dr. Wataru Aoi Laboratory of Health Science Graduate School of Life and Environmental Sciences Kyoto Prefectural University 1-5 Hangi-cho Shimogamo Sakyo-ku Kyoto, Japan

Dr. Takashi Ide National Food Research Institute Kannondai 2-1-12 Tsukuba, Japan

Dr. Hiroshi Iida National Food Research Institute Kannondai 2-1-12 Tsukuba, Japan

Dr. Yoshiko Ishimi

Department of Biochemical Sciences University of Rome La Sapienza Piazzale Aldo Moro 5 Rome, Italy

Project for Bio-Index Nutritional Epidemiology Program, National Institute of Health and Nutrition 1-23-1 Toyama Shinjuku-ku Tokyo, Japan

Dr. Tsuyoshi Goto

Professor Geun Eog Ji

Laboratory of Molecular Functions of Foods Division of Food Science and Biotechnology Graduate School of Agriculture Kyoto University Uji Kyoto, Japan

College of Human Ecology Seoul National University Department of Food and Nutrition San 56-1 Shillim-Dong Kwanak-Ku Seoul, Korea

Professor Dr. Carlo Crifò

Dr. Shizuka Hirai Laboratory of Molecular Functions of Foods Division of Food Science and Biotechnology Graduate School of Agriculture Kyoto University Uji Kyoto, Japan

Dr. Ji-Hye Kang Department of Food Science and Nutrition University of Ulsan Mugeo-dong Ulsan, South Korea

VII

Dr. Masumi Katsuta

Dr. Akihiko Nagao

National Institute of Crop Science Kannondai 2-1-18 Tsukuba, Japan

National Food Research Institute National Agriculture and Food Research Organization 2-1-12 Kannondai Tsukuba Ibaraki, Japan

Professor Teruo Kawada Laboratory of Molecular Functions of Foods Division of Food Science and Biotechnology Graduate School of Agriculture Kyoto University Uj Kyoto, Japan

Dr. Chu-Sook Kim Department of Food Science and Nutrition University of Ulsan Mugeo-dong Ulsan, South Korea

Dr. Joydeb Kumar Kundu National Research Laboratory of Molecular Carcinogenesis and Chemoprevention College of Pharmacy Seoul National University Shillim-dong Kwanak-gu Seoul, South Korea

Dr. Xuebo Liu Laboratory of Food and Biodynamics Graduate School of Bioagricultural Science Nagoya University Furo-cho Nagoya, Japan

Dr. Kazuo Miyashita Graduate School of Fisheries Sciences Hokkaido University Hakodate, Japan

Professor Yuji Naito Molecular Gastroenterology and Hepatology Kyoto Prefectural University of Medicine 456 Kajii-cho Kamigyo-ku Kyoto, Japan

Dr. Yoshimasa Nakamura Faculty of Agriculture Okayama University 1-1-1 Tsushima-naka Okayama, Japan

Dr. Yasutaka Nakashima National Food Research Institute Kannondai 2-1-12 Tsukuba, Japan

Dr. Toshihiko Osawa Laboratory of Food and Biodynamics Graduate School of Bioagricultural Science Nagoya University Furo-cho Nagoya, Japan

Professor Costantino Salerno Laboratory of Clinical Biochemistry University of Rome La Sapienza Via dei Sardi 58 Rome, Italy

Dr. Werner Siems Dr. Akira Murakami Division of Food Science and Biotechnology Graduate School of Agriculture Kyoto University Kyoto, Japan

Research Institute of Physiotherapy & Gerontology, KortexMed Institute of Medical Education Hindenburgring 12 A Bad Harzburg, Germany

Dr. Hye-Kyung Na

Dr. Olaf Sommerburg

National Research Laboratory of Molecular Carcinogenesis and Chemoprevention College of Pharmacy Seoul National University Shillim-dong Kwanak-gu Seoul, South Korea

University Children´s Hospital Department III University of Heidelberg Im Neuenheimer Feld 153 Heidelberg, Germany

VIII

List of Contributors

Professor Young-Joon Surh

Dr. Sayori Wada

National Research Laboratory of Molecular Carcinogenesis and Chemoprevention College of Pharmacy Seoul National University Shillim-dong Kwanak-gu Seoul, South Korea

Laboratory of Health Science Kyoto Prefectural University 1-5 Hangi-cho Shimogamo Sakyo-ku Kyoto, Japan

Dr. Hirofumi Tachibana Division of Applied Biological Chemistry Department of Bioscience and Biotechnology Faculty of Agriculture Kyushu University Higashi-ku Hakozaki 6-10-1 Fukuoka, Japan

Dr. Nobuyuki Takahashi Laboratory of Molecular Functions of Foods Division of Food Science and Biotechnology Graduate School of Agriculture Kyoto University. Uji Kyoto, Japan

Dr. Yu Wang Massachusetts Institute of Technology 77 Massachusetts Avenue Room 56-731 Cambridge, USA

Dr. Ingrid Wiswedel Department of Pathological Biochemistry Institute of Clinical Chemistry and Pathological Biochemistry Otto-von-Guericke University Magdeburg Leipziger Str. 44 Magdeburg, Germany

Dr. Satoko Yasumoto National Agricultural Research Center Kannondai 3-1-1 Tsukuba, Japan

Professor Junji Terao

Professor Toshikazu Yoshikawa

Department of Food Science Graduate School of Nutrition and Bioscience The University of Tokushima 18-15 Kuramoto-cho 3 Tokushima, Japan

Molecular Gastroenterology and Hepatology Kyoto Prefectural University of Medicine 456 Kajii-cho Kamigyo-ku Kyoto, Japan

Dr. Taku Uemura

Dr. Rina Yu

Laboratory of Molecular Functions of Foods Division of Food Science and Biotechnology Graduate School of Agriculture Kyoto University Uji Kyoto, Japan

Department of Food Science and Nutrition University of Ulsan Mugeo-dong Ulsan, South Korea

List of Contributors

IX

Preface

Food factors are considered to be critical for human health promotion. It was a great pleasure for me to host the International Conference on Food Factors for Health Promotion (ICoFF 2007) that took place from November 27 to December 1, 2007, in Kyoto, Japan. The ICoFF2007 was organized mainly by the Japanese Society for Food Factors, which leads basic and clinical research in the field of functional food. The theme of the ICoFF2007 was ‘Food Factors for Health Promotion’. We thought that this was appropriate because food factors may function as the frontline in the prevention of lifestyle-related disProf. Toshikazu Yoshikawa eases as well as in health promotion. Prevention of lifestyle-related disease is listed as one of the priority research subjects not only in Japan but also in Western countries. Multiple factors are involved in the development of these diseases. A key future challenge is to clarify these factors, invent a method of detecting any change in the initial phase and establish a diagnostic approach that applies to prevention studies involving food factors. Although the function of food factors has been the main focus of this conference, other areas of interest in recent basic and clinical research also include bioavailability, metabolism, and safety of foods and their ingredients. This meeting provided an important opportunity to promote the development of food factor science. We invited several opinion leaders and young researchers from the USA, EU, Korea and other countries.

XI

This book includes an introductory overview of food factors and perspectives on bioavailability, chemistry and biomarkers that were presented during the ICoFF2007. We thank all the authors for their contributions and efforts in the preparation of this book. We also extend our sincere gratitude to the many scientists who reviewed the chapters found herein. Prof. Toshikazu Yoshikawa, Kyoto

XII

Yoshikawa

Genomics Yoshikawa T (ed): Food Factors for Health Promotion. Forum Nutr. Basel, Karger, 2009, vol 61, pp 1–9

Genomics for Food Functionality and Palatability Keiko Abe The University of Tokyo, Graduate School of Agricultural and Life Sciences, Tokyo, Japan

Abstract In the 1980s, Japan proposed the termiology of ‘functional food’ and its concept [1], and since then the importance of conducting basic and applied studies on food functionality has been emphasized globally. Functional foods in particular as well as common foods in general are constituted with a variety of components including functional factors, and it has been recognized as difficult to evaluate their functionalities by usual chemical, biochemical and physiological methodologies [2]. Against this backdrop, nutrigenomics came into being as a new method of evaluating functional foods, as well as nutrients, in a holistic manner. Meanwhile the endowed chair, Functional Food Genomics, was established at the University of Tokyo with the aegis of 32 food companies in Japan. This academia-industry collaboration has been working well to disclose why and how some particular functional foods elicit their effects in the body. These include soy protein isolate, cocoa polyphenol, sesamin as a lignan of sesame origin, and many others. On the other hand, food safety has been gaining public attention, and we applied genomics for assessment of the wholesomeness of newly developed hypoallergenic wheat flour compared with normal flour. The application of this way of holistic evaluation suggested that the new product was basically the same as the normal product in terms of all-gene expression profiles. The same method was applied to a new sweet protein, neoculin, which resembled toxic lectins in conformation. The result indicated that neoculin had lost its lectin activity, possessing no particular toxic effect. It is thus likely that genomics can be applied to a variety of foods in general for the Copyright © 2009 S. Karger AG, Basel purpose of simultaneously assessing their functionality.

In the 1980s, Prof. Soichi Arai, Department of Agricultural Chemistry at The University of Tokyo took the initiative of successfully launching, with the help of his colleagues, including myself, the Priority Area Research Project on Functional Food supported by the Grant-in-Aid from the Ministry of Education, Science, Sports and Culture. Food has traditionally been thought of as having two functions: the primary function of providing nutrition to the body and the secondary function of providing culinary enjoyment. The research team demonstrated that there exists a third function which contributes to reducing the risk of developing lifestyle-related diseases. The team declared to the world that it had named the type of food with this function, ‘functional food’ [1].

This basic and applied research opened the new field of food studies and attracted tremendous attention from inside and outside Japan, triggering a boom of research activities in this area. The advent of the aging society has accelerated this trend. The Department of Agricultural Chemistry had, as its traditional research principle, the extension of studies in life sciences from basic to applied research. It is a very natural consequence that the Department of Applied Biological Chemistry, a successor of this idea, has been leading functional food science to launch this endowed chair. What is more, given that cooperative research projects between industry and academia are today being promoted by the government, functional food science may even serve as a pioneering model. The other reason stems from the characteristics of food studies. Functional foods have some elements that help reduce the risk of disease, but unlike pharmaceuticals, each of them is a complex system consisting of multiple elements. Functional food factors may be affected by other ingredients contained in the same food, and their functions could be increased or decreased. In addition, we continuously consume a large amount of food every day, and the compounds resulting from metabolism in the body are therefore plentiful both in quantity and in quality. An excessive intake may exert adverse effects on physical health. It might cancel the positive effect of functional food factors. Even if any benefit is evident, the part that receives the benefit varies extensively: the digestive tract, the liver, kidneys, blood, muscles, the brain and others. Also, the duration of the effect is not even. What we need to do is comprehensively verify the overall effect of eating a single kind of functional food (table 1). In this sense, the traditional styles of physiology, biochemistry, molecular and cell biology, etc. are so much aimed at the deep investigation of individual matters that they are not appropriate for food research. This is where genomics comes in as a new science for analyzing the information of all genes. In 2002, nutrigenomics was established as applied genomics in the science of nutrition. Behind this event was a major global progress in functional food science as a new domain of the nutrition sciences [3]. The notable aspect of this progress is the attempt to measure the effects of ingested functional foods by means of simple indicators such as biomarkers prior to examining those by human intervention tests. The emergence of nutrigenomics came to represent these efforts. At the beginning, transcriptomics was the mainstay [4]. Lee et al. [5] found that old mice have a stronger expression of inflammatory and stress genes and a weaker expression of protein metabolism- and growth-related genes than young mice, and regarded the genes as indicators of aging. They also verified that the fluctuations in genes serving as markers of age can be controlled by limiting the daily calorie intake by 30%. In other words, the hypothesis that restricting calorie intake slows aging has been substantiated at the genetic level. This was the first ever nutrigenomics research report. The present author had introduced a DNA microarray system to start research before the term nutrigenomics was established in Europe. There was the belief that

2

Abe

Table 1. Characteristics of food compared with drug Item

Food

Drug

Chemistry Composition Source Structure Reactivity

multiple natural heterogeneous high

simple mostly designed homogeneous low

Intake Purpose Time Period Daily consumption

for nutrition usual lifelong large

for remedy when needed generally short trace

Physiology Taste Absorption Remaining time Target organ or tissue Efficiency/efficacy Secondary effect Metabolite Synergy Individual difference

important various various non-specific slow impossible multiple large large

unimportant easy short specific quick possible simple small large

this exhaustive analytical approach would be indispensable in the overall elucidation of the diverse range of effects that the consumption of food, which is a multicomponent complex system, has on living bodies (fig. 1). From the perspective of the industry, the approaches to food development are fundamentally different from the methods for developing pharmaceutical products. Simply put, the former is centrifugal while the latter is concentric (fig. 1). Nutrigenomics is significant to analysis in a wide range of centrifugal studies.

Functional Food Genomics

In the second half of 2003, about 30 member companies of a nonprofit organization called the International Life Sciences Institute of Japan (ILSI Japan) made a joint investment in the launch of an Endowed Chair of Functional Food Genomics run by a guest associate professor, Dr. Ichiro Matsumoto, Graduate School of Agricultural and Life Sciences, in which the author herself works to initiate studies on

Genomics for Food Functionality

3

Target tissue/cell

Functional food

Genes

Proteins

$$ %%

Physiological functions • Improved lipid metabolism • Resistance to obesity • Resistance to infection • Antioxidation and antiinflammation • Intestinal modulation and antiinfection • Immunotolerance and hypoallergenicity

&&

Analysis

Evaluation

Data

Fig. 1. Nutrigenomics-based evaluation of functional food.

nutrigenomics in cooperation with academia and industry. Major research achievements are as follows.

Soy Protein Functioning to Improve Lipid Metabolism In joint research with Fuji Oil Co., Ltd., the gene expression in rat liver was analyzed after a long-term intake of soy protein isolate (SPI) in order to determine what is behind the improvement in dislipidemia and hypercholesteremia as one of the multiple functions performed by soy protein (fig. 1) [6]. It confirmed that the concentrations of cholesterol and triglyceride in the blood were lower in the group with a 2-month intake of SPI than in the control group of casein intake. A DNA microarray analysis of the liver found that rats fed SPI had higher expression of genes for the antioxidant and sterol metabolism systems and lower expression of genes for cell proliferation, structural protein, amino acid metabolism and sterol metabolism systems than casein-fed rats. A majority of genes with at least 50% difference are concerned with cholesterol metabolism, fatty acid metabolism and antioxidation. The SPI intake was confirmed to have the most significant impact on fat metabolism and antioxidation systems. The genes involved included those relating to the fatty acid synthesis systems, which are considered to have reduced the concentration of triglyceride in the blood. On the other hand, the blood cholesterol concentration was significantly

4

Abe

reduced while the gene expression in the sterol synthesis system was significantly increased. Considering the significant rise in the excretion of acid steroid (bile acid) and neutral steroid (cholesterol), it is assumed that SPI exhibited a function of facilitating steroid excretion in a short time and that this effect lasted for a long time to eventually up-regulate sterol synthesis genes for the purpose of constantly controlling the level of cholesterol in the body.

Antiobesity Function of Cocoa Polyphenol Morinaga & Co., Ltd., has found that the group of rats fed with a high-fat diet (27% lard) and 12.5% cocoa polyphenol had a lower rate of weight increase and a lower weight of visceral fat with statistical significance than rats fed with a cocoa substitute (control) [7]. Also, the high cocoa group tended to have a lower level of serum triacylglycerol. To elucidate the reason for this by examining gene expression in the liver and fat cells, this study performed a DNA microarray analysis of the liver and mesenteric white fat cells. It has confirmed that (1) the antiobesity effect on the liver is caused by the decline in the blood triacylglycerol concentration after down-regulation of fatty acid biosynthesis, and that (2) the same effect on fat cells is produced by inhibition of the fatty acid transport system after down-regulation of PPARγ, inhibition of the fatty acid synthesis system after down-regulation of SREBP-1c, and the decline in cumulative fat after combustion of fatty acid following up-regulation of uncoupling proteins that act as exothermic factors.

Function of Sesamin to Regulate β-Oxidation and to Boost Alcohol Metabolism Sesamin is a principal lignan contained in sesame seeds. Under joint research with Suntory Limited, an analysis was conducted of gene expression changes in the liver of a rat fed with sesamin for 3 days to observe whether its intake up-regulated the expression of genes for various fat metabolism enzymes associated with β-oxidation and lipogenesis [8]. More interestingly, in the light of the up-regulation of genes for aldehyde dehydrogenase (ALDH) acting on alcohol catabolism, it was demonstrated that sesamin exhibits the function of regulating alcohol metabolism as activated ALDH accelerates the decomposition of acetaldehyde.

Royal Jelly Functioning to Facilitate Osteogenesis At the initiative of Nagaragawa Research Center of API Co., Ltd., an analysis was conducted of the osteogenetic effect of royal jelly (RJ) [9]. An increase in the mineral weight of the neck bone was confirmed in a mouse administered with RJ for 2

Genomics for Food Functionality

5

months. An analysis of gene expression in a femur observed a significant up-regulation of some 300 genes in the RJ-fed group in comparison with the non-RJ-fed group. Seventy percent of these genes were upregulated in the group receiving subcutaneous administration of 17β-estradiol (E2). This implies that RJ produces an estrogenic effect.

Safety Assessment of Low Allergen Wheat As a nutrigenomic case study on food safety assessment, an analysis of hypoallergenic wheat flour was carried out [10]. A team led by Dr. Hisanori Kato, Associate Professor, Graduate School of Agricultural and Life Sciences, The University of Tokyo, examined the gene expression in the livers and small intestines of rats fed a 12% protein diet containing hypoallergenic wheat flour or ordinary wheat flour. It was found that very few genes had changes in fold and that none of these genes were concerned with any function adverse to living bodies, such as toxicity. Today, increasing attention is focused on food safety. This study has confirmed that DNA microarray analysis can be used in the safety assessment. Just recently, nutrigenomics incorporated proteomics and metabolomics into transcriptomics. In other words, it is necessary to use these three in a manner where they are coordinated. Known as ‘coordinative genomics,’ it is now considered of indispensable significance to research on the overall effect of a complicated heterosystem, namely food, on living bodies. In the future, it is necessary to perform a study on personalized functional food in consideration of the variations among individuals. It is thought from a molecular perspective that the individual difference results from single nucleotide polymorphisms generated by the partial variation of individual genes. It is anticipated that we are sure to see tailor-made functional foods in the near future. Steps towards second-generation nutrigenomics are already being taken at a steady pace.

Food Safety Genomics with Special Reference to Neoculin as a New Sweet Protein and Plant Lectins – A Pilot Study

While the majority of sweet substances are of low molecular weight, there are six proteins, brazzein, thaumatin, mabinlin, monellin, neoculin (NCL) and pentadin, which elicit sweetness to humans. These substances as nonglycemic sweeteners will provide a potential use for people with obesity, diabetes and other metabolic syndromes. NCL occurring in the tropical fruits of Curculigo latifolia is currently the only protein that has both sweetness and a taste-modifying activity to convert sourness into sweetness [11]. The strong sweetness of NCL makes it possible to use this substance as a tool for basic taste signaling research as well. However, it has remained unclear how NCL induces this unique sensation.

6

Abe

Neoculin

180°

Garlic lectin

Fig. 2. X-ray crystallography.

10,000

PHA-E

1,000 100 10 1 1

100

10,000

10,000

WGA

1,000 100 10 1 1

100

10,000

Neoculin

10,000 1,000 100 10 1 1

100 10,000 Control

Neoculin

WGA

Fig. 3. DNA microarray analysis.

Genomics for Food Functionality

7

Recently, we quantitatively evaluated the acid-induced sweetness of NCL by a cellbased assay [12]. In brief, human sweet taste receptor hT1R2-hT1R3 was functionally expressed, together with chimeric G␣, in cultured cells. The cells responded to NCL pH dependently under acidic conditions. The pH-response relationship reflected a sigmoidal, imidazole titration curve, suggesting the involvement of histidine residues in the acid-induced sweetness [13]. Actually, an NCL variant in which all the five histidine residues were replaced with alanine elicited strong sweetness at neutral as well as acidic pH. The His →Ala variant is apparently a novel sweet protein genetically engineered. Both the primary and overall tertiary structures of NCL resemble those of monocot mannose-binding lectins [14]. This study investigated differences in biochemical properties between NCL and the lectins. Structural comparison between the mannose-binding site of lectins and the corresponding regions of NCL showed that there is at least one amino acid substitution at each site in NCL, suggesting a reason for the lack of its mannose-binding ability (fig. 2). This was consistent with hemagglutination assay data demonstrating that NCL had no detectable agglutinin activity. DNA microarray analysis indicated that NCL had no significant influence on gene expression in the Caco-2 cell, whereas kidney bean lectin (Phaseolus vulgaris agglutinin) greatly influenced various gene expressions [15] (fig. 3). These data strongly suggest that NCL has no lectin-like properties, encouraging its practical use in the food industry.

References 1 Swinbanks D, O’Brien J: Japan explores the boundary between food and medicine. Nature 1993;364: 180. 2 Arai S, Yasuoka A, Abe K: Functional food science and food for specified health use policy in Japan: state of the art. Curr Opin Lipidol 2008;19: 69–73. 3 Roberfroid MB: Global view on functional foods: European perspectives. Brit J Nutr 2002;88:S133– 138. 4 Müller M, Kersten S: Nutrigenomics: goals and strategies. Nat Rev 2003;4:315–322. 5 Lee C-K, Klopp RG, Weindruch R, Prolla TA: Gene expression profile of aging and its retardation by caloric restriction. Science 1999;285:1390–1393. 6 Tachibana N, Matsumoto I, Fukui K, Arai S, Kato H, Abe K, Takamatsu KJ: Intake of soy protein isolate alters hepatic gene expression in rats. Agric Food Chem 2005;53, 4253–4257.

8

7 Matsui N, Ito R, Nishimura E, Yoshikawa M, Kato M, Kamei M, Shibata H, Matsumoto I, Abe K, Hashizume S: Ingested cocoa can prevent high-fat diet-induced obesity by regulating the expression of genes for fatty acid metabolism. Nutrition 2005;21: 594–601. 8 Tsuruoka N, Kidokoro A, Matsumoto I, Abe K, Kiso Y: Modulating effect of sesamin, a functional lignan in sesame seeds, on the transcription levels of lipidand alcohol-metabolizing enzymes in rat liver: a DNA microarray study. Biosci Biotechnol Biochem 2005;69:179–188. 9 Narita Y, Nomura J, Ohta S, Inoh Y, Suzuki KM, Araki Y, Okada S, Matsumoto I, Isohama Y, Abe K, Miyata T, Mishima S: Royal jelly stimulates bone formation: physiologic and nutrigenomic studies with mice and cell lines. Biosci Biotechnol Biochem 2006;70:2508–2514. 10 Narasaka S, Endo Y, Fu ZW, Moriyama M, Arai S, Abe K, Kato, H: Safety evaluation of hypoallergenic wheat flour by using a DNA microarray. Biosci Biotechnol Biochem 2006;70, 1464–1470.

Abe

11 Shirasuka Y, Nakajima K, Asakura T, Yamashita H, Yamamoto A, Hata S, Nagata S, Abo M, Sorimachi H, Abe K: Neoculin as a new taste-modifying protein occurring in the fruit of Curculigo latifolia. Biosci Biotechnol Biochem 2004;68:1403–1407. 12 Nakajima K, Asakura T, Oike H, Morita Y, ShimizuIbuka A, Misaka T, Sorimachi H, Arai S, Abe K: Neoculin, a taste-modifying protein, is recognized by human sweet taste receptor. Neuroreport 2006;17: 1241–1244. 13 Nakajima K, Morita Y, Koizumi A, Asakura T, Terada T, Ito K, Shimizu-Ibuka A, Maruyama J, Kitamoto K, Misaka T, Abe K: Acid-induced sweetness of neoculin is ascribed to its pH-dependent agonistic-antagonistic interaction with human sweet taste receptor. FASEB J 2008;22:2323–2330.

14 Shimizu-Ibuka A, Morita Y, Terada T, Asakura T, Nakajima K, Iwata S, Misaka T, Sorimachi H, Arai S, Abe K: Crystal structure of neoculin: insights into its sweetness and taste-modifying activity. J Mol Biol 2006;359:148–158. 15 Shimizu-Ibuka A, Nakai Y, Nakamori K, Morita Y, Nakajima, K, Kadota K, Watanabe H, Okubo S, Terada T, Asakura T, Misaka T, Abe K: Biochemical and genomic analysis of neoculin compared to monocot mannose-binding lectins. J Agric Food Chem 2008;56:5338–5344.

Professor Keiko Abe Department of Applied Biological Chemistry Graduate School of Agricultural and Life Sciences The University of Tokyo 1-1-1 Yayoi, Bunkyo-ku Tokyo 113-8657 (Japan) Tel. + 81 3 5841 5129, Fax +81 3 5841 8006, E-Mail [email protected]

Genomics for Food Functionality

9

Genomics Yoshikawa T (ed): Food Factors for Health Promotion. Forum Nutr. Basel, Karger, 2009, vol 61, pp 10–24

Lipid Metabolism and Nutrigenomics – Impact of Sesame Lignans on Gene Expression Profiles and Fatty Acid Oxidation in Rat Liver Takashi Idea ⭈ Yasutaka Nakashimaa ⭈ Hiroshi Iidaa ⭈ Satoko Yasumotob ⭈ Masumi Katsutac a National Food Research Institute, bNational Agricultural Research Center, and cNational Institute of Crop Science, Tsukuba, Japan

Abstract The impact of sesamin, episesamin and sesamolin (sesame lignans) on hepatic gene expression profiles was compared with a DNA microarray. Male Sprague-Dawley rats were fed experimental diets containing 0.2% sesamin, episesamin or sesamolin, and a control diet free of lignans for 15 days. Compared to a lignan-free diet, a diet containing sesamin, episesamin and sesamolin caused more than 1.5- and 2-fold changes in the expression of 128 and 40, 526 and 152, and 516 and 140 genes, respectively. The lignans modified the mRNA levels of not only many enzymes involved in hepatic fatty acid oxidation, but also proteins involved in the transportation of fatty acids into hepatocytes and their organelles, and in the regulation of hepatic concentrations of carnitine, CoA and malonylCoA. It is apparent that sesame lignans stimulate hepatic fatty acid oxidation by affecting the gene expression of various proteins regulating hepatic fatty acid metabolism. The changes in the gene expression were generally greater with episesamin and sesamolin than with sesamin. In terms of amounts accumulated in serum and the liver, the lignans ranked in the order sesamolin, episesamin and sesamin. The differences in bioavailability among these lignans appear to be important to their divergent physiological activities. We also confirmed that dietary sesame seed affected the expression of genes related to fatty acid oxidation in a manner similar to isolated lignan compounds. Copyright © 2009 S. Karger AG, Basel

Sesame seeds contain compounds known collectively as lignans. Sesamin and sesamolin are fat soluble, and sesame seeds and their oil contain these two lignans at a ratio of about 2:1 [1]. Another major lignan of sesame is sesaminol which exists as glucosides [2], and is not extractable in oil. In the refining of sesame oil, sesamin is epimerized to form episesamin, and most of the sesamolin is degraded [3]. The lignan preparation obtained as a by-product of the refining of edible sesame oil therefore consists of sesamin and episesamin at a ratio of 1:1. This preparation has

been tested for its physiological activity by many investigators. We previously demonstrated that this preparation strongly increased the activity and gene expression of enzymes involved in fatty acid oxidation in the rat liver [4]. This may account for its serum lipid-lowering effect [4–6]. We subsequently demonstrated that episesamin [7] and sesamolin [8], compared to sesamin, are much stronger at increasing the activity and gene expression of enzymes involved in fatty acid oxidation. These results indicate that large differences exist among various lignans in their effect on hepatic fatty acid oxidation. However, no study has simultaneously compared the physiological activities of these three lignans. The DNA microarray is a powerful tool for genome-wide analyses of gene expression patterns. Using DNA microarrays containing about 8,000 probes, Tsuruoka et al. [9] confirmed our previous finding [4] that a lignan preparation containing equivalent amounts of sesamin and episesamin strongly increased the mRNA expression of various enzymes involved in fatty acid oxidation in the rat liver. In addition, they showed that it increased the gene expression of aldehyde dehydrogenase 1 family members (Aldh1a1 and 1a7). They suggested that upregulation of these enzymes may be responsible for the preventive effect of the lignan preparation on the ethanol-induced liver damage observed previously [10]. As they used a lignan preparation containing both sesamin and episesamin in their experiment, information on different effects of various lignans affecting hepatic gene expression profiles is still lacking. The recent development of DNA microarray technology has enabled the examination of more widespread changes in gene expression profiles. Here, we compared the physiological activity of dietary sesamin, episesamin and sesamolin in affecting gene expression profiles in the rat liver using a microarray containing more than 30,000 oligonucleotide probes. In addition, we also examined the effect of sesame seeds on hepatic gene expression.

Methods Animals and Diets Male Sprague-Dawley rats obtained from Charles River Japan, Kanagawa, Japan, at 5 weeks of age were divided into 4 groups with equal mean body weights consisting of 7 animals each and fed either a diet free of lignan or diets containing 0.2% lignan (sesamin, episesamin or sesamolin) for 15 days (experiment 1). In a second experiment (experiment 2), three groups of rats consisting of 7 animals each were fed either a diet containing 10 or 20% sesame seed powder or a diet without sesame for 15 days. We used a sesame line rich in lignan developed by Sirato-Yasumoto et al. [1] in this experiment. Diets containing 10 and 20% sesame seed powder had 0.165 (0.116% as sesamin and 0.049% as sesamolin) and 0.33% (0.232% as sesamin and 0.098% as sesamolin) lignan, respectively. The basal composition of the experimental diet was the same as described previously [1, 8]. Upon termination of the experimental period, animals were anesthetized using diethyl ether and killed by bleeding from the abdominal aorta, after which livers were excised. This study was approved by the review board of animal ethics of our institute, and we followed the institute’s guidelines in the care and use of laboratory animals.

Sesame Lignans and Gene Expression

11

Affymetrics GeneChip and GeneSpring Analyses RNA extracted from the livers of 5 and 6 rats from each group was subjected to microarray analyses in experiments 1 and 2, respectively. Rats with the highest and lowest body weights in experiment 1, and the animals with the lowest body weight in experiment 2 in each group at the time of killing were eliminated from the microarray analyses. RNA was processed using kits supplied by Affymetrix (Santa Clara, Calif., USA) to prepare fragmented biotinylated cRNA for hybridization to the Rat Genome 230 2.0 Array. Analyses of the DNA microarray data were performed using GeneSpring GX v7.3 software (Agilent Technologies Inc., Santa Clara, Calif., USA).

Real-Time PCR Quantification of Hepatic mRNA The quantification of mRNA by real-time PCR was performed as detailed previously [11]. mRNA abundance was calculated as a ratio to the mRNA level of β-actin in each cDNA sample and expressed as fold change, assigning a value of 1 for rats fed a diet free of lignan (experiment 1) or sesame seed (experiment 2).

Analyses of Lignans and Carnitine Concentrations of lignans in the liver and serum were analyzed by HPLC as detailed previously [12]. The hepatic concentration of carnitine was analyzed by the method of Pearson et al. [13].

Results

Impact of Sesame Lignans (Sesamin, Episesamin, and Sesamolin) and Sesame Seed on Gene Expression Profile in Rat Liver In total, 679 genes were found to be significantly (p < 0.05) up- or downregulated more than 1.5-fold by either sesamin, episesamin or sesamolin (experiment 1). Compared with a lignan-free diet, a diet containing sesamin caused changes more than 1.5- and 2-fold in the expression of 128 and 40 genes, respectively. More of the genes were affected by episesamin and sesamolin. Episesamin and sesamolin caused changes more than 1.5-fold in the expression of 526 and 516 genes, respectively. The numbers of genes up- or downregulated by dietary lignans more than 2-fold are shown as Venn diagrams (fig. 1). The diagram in figure 1a shows that many genes were commonly upregulated by various lignans. Actually, 97% of genes upregulated more than 2-fold by sesamin were also upregulated by one or both of the other lignans. This was the same for episesamin and sesamolin. Therefore, lignans resembled each other with respect to affecting genes whose expression was upregulated. The proportions were much lower for the genes downregulated by lignans. However, detailed analyses indicated that the situation was similar to that for the upregulated genes. A subsequent Tukey’s test revealed that 93 and 90% of the genes downregulated more than 2-fold by episesamin and sesamolin, respectively, were also significantly down-regulated by

12

Ide · Nakashima · Iida · Yasumoto · Katsuta

Episesamin

Sesamin

30 55

2

33

3

0

21

3

a

Sesamin

1

0

8

Episesamin

Sesamolin

4 0

28

b

Sesamolin

Fig. 1. Venn diagrams of genes up- or downregulated in the livers of rats fed various sesame lignans. Number of genes upregulated (a) and downregulated (b) more than 2.0-fold. Each of the circles represents the genes affected by the respective lignan. The numbers in the spaces between overlapping circles represent the number of genes that were affected by the lignan species represented by the circles. The numbers in the outer portion of each circle represent the number of genes that were exclusively affected by the specific lignan represented by that particular circle.

one or both of the other lignans, although the extent of the changes was not necessarily greater than 2-fold. This evaluation was rather difficult for sesamin because only 9 genes were downregulated more than 2-fold by this lignan. The functions of genes significantly up- or downregulated more than 1.5-fold by treatment with various lignans were clarified using annotations and information supplied by GeneSpring and several databases (Rat Genome Database, Entrez Gene, and PubMed). Miscellaneous genes with diverse functions were up- or downregulated by lignans. Previous studies showed that dietary lignans profoundly affect hepatic fatty acid oxidation [4–9, 12]. Therefore, genes related to fatty acid oxidation whose expression was up- or downregulated more than 1.5-fold were selected and are listed in table 1. Dietary lignans increased the gene expression of many mitochondrial and peroxisomal enzymes related to fatty acid oxidation, including those involved in acylcarnitine biosynthesis (Crat, Crot and Cpt1b), the conversion of acylcarnitine to acylCoA (Cpt2), β-oxidation (Ehhadh, Ech1, Acaa1 and 2, Acox1, Hadha, Hadhb and Acadvl), the auxiliary pathway of β-oxidation (Dci, and Decr1 and 2), ω-oxidation (Cyp4a1; Cyp4a10 and Cyp4a3), and ketogenesis (Acat1 and Hmgcl). In addition, the genes for Pex11a, presumed to play a role in peroxisome membrane biogenesis [14], and for Fxc1, which mediates the import and insertion of hydrophobic membrane proteins into the mitochondrial inner membrane [15], were activated by lignans. The three lignans decreased the expression of Acaca involved in the biosynthesis of malonyl-CoA to a similar level. In contrast, they increased the expression of Mlycd, which catalyzes the breakdown of malonyl-CoA. The increases were greater with episesamin and sesamolin than with sesamin. Episesamin and sesamolin, but not sesamin, significantly increased the mRNA expression of Mgll, which may be involved in the hydrolysis of intracellular triacylglycerol [16], and Pank1, which catalyzes a rate-limiting step in CoA biosynthesis [17].

Sesame Lignans and Gene Expression

13

Table 1. Microarray analyses of genes for proteins involved in the regulation of fatty acid oxidation and those involved in the biogenesis of mitochondria and peroxisomes whose expression was affected (>1.5fold) by either sesamin, episesamin or sesamolin (experiment 1) or by a diet containing 10 or 20% sesame seed (experiment 2). Accession number

Gene symbol

Gene name

Enzymes involved in fatty acid oxidation AI411979

Crat

Carnitine acetyltransferase

NM_133606

Ehhadh

Enoyl-coenzyme A, hydratase/3- hydroxyacyl coenzyme A dehydrogenase

J02844

Crot

Carnitine O-octanoyl transferase

NM_017306

Dci

Dodecenoyl- coenzyme A delta isomerase

NM_031987

Crot

Carnitine O-octanoyl transferase

NM_022594

Ech1

Enoyl coenzyme A hydratase 1, peroxisomal

NM_013200

Cpt1b

Carnitine palmitoyl transferase 1β, muscle isoform

NM_016999

Cyp4a1; Cyp4a10

Cytochrome P450, family 4, subfamily a, polypeptide 1; 10

NM_ 012489

Acaa1

Acetyl-coenzyme A acyltransferase 1, peroxisomal

AA899304

Acat1

Acetyl-coenzyme A acyltransferase 1, mitochondrial

NM_012930

Cpt2

Carnitine palmitoyl transferase 2

NM_057197

Decr1

2,4-dienoyl CoA reductase 1, mitochondrial

NM_017340

Acox1

Acyl-coenzyme A oxidase 1, palmitoyl

AF044574

Decr2

2,4-dienoyl CoA reductase 2, peroxisomal

NM_133618

Hadhb

Hydroxyacyl-coenzyme A dehydrogenase/3-ketoacyl-coenzyme A thiolase/enoyl-coenzyme A hydratase (trifunctional protein), β-subunit

M33936

Cyp4a3

Cytochrome P450, family 4, subfamily a, polypeptide 3

AA893326

Cyp4a3

Cytochrome P450, family 4, subfamily a, polypeptide 3

AA800240

Hadha

Hydroxyacyl-coenzyme A dehydrogenase/3-ketoacyl-coenzyme A thiolase/enoyl-coenzyme A hydratase (trifunctional protein), α-subunit

NM_130433

Acaa2

Acetyl-coenzyme A acyltransferase 2 (mitochondrial 3-oxoacylcoenzyme A thiolase)

D13921

Acat1

Acetyl-coenzyme A acyltransferase 1, mitochondrial

NM_024386

Hmgcl

3-hydroxy-3-methylglutaryl-coenzyme A lyase

NM_012891

Acadvl

Acyl-coenzyme A dehydrogenase, very long chain

14

Ide · Nakashima · Iida · Yasumoto · Katsuta

Experiment 1

Experiment 2

fold change (dietary lignans)

fold change (dietary sesame)

lignan-free

sesamin

episesamin

sesamolin

0%

10%

20%

1.0 ± 0.3a

3.7 ± 0.7a

13 ± 2b

12 ± 3b

1.0 ± 0.1a

2.1 ± 0.2b

4.5 ± 0.6c

1.0 ± 0.1a

2.9 ± 0.3b

8.5 ± 0.7c

8.2 ± 0.9c

1.0 ± 0.1a

3.3 ± 0.4b

8.4 ± 0.4c

1.0 ± 0.1a

3.1 ± 0.4b

6.6 ± 1.0c

6.9 ± 0.7c

1.0 ± 0.1a

3.0 ± 0.3b

5.3 ± 0.5c

1.0 ± 0.1a

3.8 ± 0.3b

6.4 ± 0.4c

6.6 ± 0.4c

1.0 ± 0.1a

3.6 ± 0.2b

4.5 ± 0.2c

1.0 ± 0.1a

3.3 ± 0.2b

5.3 ± 0.4c

5.5 ± 0.3c

1.0 ± 0.0a

2.0 ± 0.1b

2.5 ± 0.1c

1.0 ± 0.1a

3.0 ± 0.2b

5.2 ± 0.2c

4.9 ± 0.3c

1.0 ± 0.1a

3.5 ± 0.1b

4.9 ± 0.1c

1.0 ± 0.1a

1.1 ± 0.1a

3.2 ± 0.6b

4.2 ± 0.6b

1.0 ± 0.0a

1.2 ± 0.1a

2.7 ± 0.5b

1.0 ± 0.0a

2.6 ± 0.1b

3.7 ± 0.1c

4.0 ± 0.2c

1.0 ± 0.1a

3.6 ± 0.2b

5.3 ± 0.2c

1.0 ± 0.1a

1.6 ± 01b

3.3 ± 0.2c

3.5 ± 0.2c

1.0 ± 0.0a

2.0 ± 0.2b

3.3 ± 0.1c

1.0 ± 0.1a

1.9 ± 0.2b

2.4 ± 0.3bc

2.7 ± 0.3c

1.0 ± 0.1a

1.8 ± 0.1b

2.5 ± 0.1c

1.0 ± 0.1a

1.8 ± 0.1b

2.9 ± 0.2c

2.8 ± 0.1c

1.0 ± 0.1a

2.0 ± 0.2b

2.7 ± 0.2c

1.0 ± 0.0a

1.8 ± 0.1b

2.6 ± 0.1c

2.6 ± 0.2c

1.0 ± 0.1a

1.6 ± 0.1b

2.0 ± 0.1c

1.0 ± 0.0a

1.5 ± 0.0b

2.5 ± 0.1c

2.4 ± 0.1c

1.0 ± 0.0a

1.5 ± 0.1b

1.9 ± 0.1c

1.0 ± 0.0a

1.8 ± 0.1b

2.3 ± 0.2c

2.3 ± 0.2c

1.0 ± 0.0a

1.9 ± 0.1b

2.5 ± 0.1c

1.0 ± 0.0a

1.5 ± 0.1b

2.2 ± 0.1c

2.1 ± 0.1c

1.0 ± 0.0a

1.6 ± 0.1b

2.0 ± 0.1c

1.0 ± 0.1a

1.7 ± 0.1b

2.0 ± 0.1bc

2.1 ± 0.2c

1.0 ± 0.1a

1.3 ± 0.1b

1.6 ± 0.1b

1.0 ± 0.1a

1.3 ± 0.1b

1.3 ± 0.1b

1.5 ± 0.1b

1.0 ± 0.1a

1.3 ± 0.1b

1.6 ± 0.1b

1.0 ± 0.0a

1.4 ± 0.0a

1.9 ± 0.1b

1.9 ± 0.1b

1.0 ± 0.1a

1.4 ± 0.1b

1.9 ± 0.0c

1.0 ± 0.1a

1.2 ± 0.1a

1.9 ± 0.1b

1.8 ± 0.1b

1.0 ± 0.0a

1.3 ± 0.0b

1.6 ± 0.1c

1.0 ± 0.1a

1.5 ± 0.1b

1.8 ± 0.1c

1.8 ± 0.1bc

1.0 ± 0.1a

1.5 ± 0.1b

1.9 ± 0.1c

1.0 ± 0.1a

1.2 ± 0.1a

1.6 ± 0.1b

1.5 ± 0.1b

1.0 ± 0.0a

1.4 ± 0.1b

1.8 ± 0.1c

1.0 ± 0.0a

1.3 ± 0.1b

1.6 ± 0.1c

1.5 ± 0.1c

1.0 ± 0.0a

1.3 ± 0.1b

1.5 ± 0.0c

Sesame Lignans and Gene Expression

15

Table 1. Continued Accession number

Gene symbol

Gene name

U88294

Cpt1a

Carnitine palmitoyltransferase 1a, liver

BI296347

Acad11

Acyl-coenzyme A dehydrogenase family, member 11

Proteins involved in biogenesis of mitochondria and peroxisomes NM_053487

Pex11a

Peroxisomal biogenesis factor 11A

BI273703

Pex11a

Peroxisomal biogenesis factor 11A

AW141617

Pex19

Peroxisomal biogenesis factor 19

AF061242

Fxc1

Fractured callus expressed transcript 1

Enzymes and transporters involved in the regulation of fatty acid oxidation NM_022193

Acaca

Acetyl-coenzyme A carboxylase-α

NM_053477

Mlycd

Malonyl-CoA decarboxylase

AY081195

Mgll

Monoglyceride lipase

AI713204

Mgll

Monoglyceride lipase

BG372713

Mgll

Monoglyceride lipase

AA850195

Pank1

Pantothenate kinase 1

NM_019269

Slc22a5

Solute carrier family 22 (organic cation transporter), member 5

NM_053965

Slc25a20

Solute carrier family 25 (carnitine/acylcarnitine translocase), member 20

NM_012804

Abcd3

ATP-binding cassette, subfamily D (ALD), member 3

AF072411

Cd36

Cd36 antigen

Values represent means ± SE for 5 and 6 rats for experiments 1 and 2, respectively. Values with different superscripts differ significantly at p < 0.05.

Lignans affected the gene expression of transporters involved in lipid metabolism. All the lignans significantly increased the mRNA expression of Slc22a5, which mediates high-affinity sodium-dependent carnitine transport [18]. The increases were stronger with episesamin and sesamolin than with sesamin, and comparable between the two former compounds. Episesamin and sesamolin, but not sesamin, significantly increased the mRNA levels of Slc25a20, which transports acylcarnitine into the mitochondrial matrix. Episesamin and sesamolin increased the mRNA expression of Cd36, a long-chain fatty acid transporter located in the plasma membrane. However, sesamin did not cause a significant increase in this parameter. Episesamin and sesamolin, but not sesamin, significantly increased the gene expression of Abcd3, which

16

Ide · Nakashima · Iida · Yasumoto · Katsuta

Experiment 1

Experiment 2

fold change (dietary lignans)

fold change (dietary sesame)

lignan-free

sesamin

episesamin

sesamolin

0%

10% a

20% b

1.3 ± 0.1

1.8 ± 0.2c

–

–

–

–

1.0 ± 0.1

–

–

–

–

1.0 ± 0.1a

1.4 ± 0.1b

1.9 ± 0.2c

1.0 ± 0.1a

4.0 ± 0.6b

15 ± 2c

13 ± 3c

1.0 ± 0.2a

6.0 ± 0.7b

11.5 ± 1.4c

1.0 ± 0.1a

2.9 ± 0.3a

7.0 ± 0.6b

6.2 ± 1.3b

1.0 ± 0.1a

3.5 ± 0.4b

5.2 ± 0.4c

–

–

–

–

1.0 ± 0.1a

1.4 ± 0.1b

1.9 ± 0.2c

1.0 ± 0.1a

1.1 ± 0.0a

1.6 ± 0.1b

1.7 ± 0.1b

1.0 ± 0.0a

1.2 ± 0.0b

1.7 ± 0.1c

1.0 ± 0.2b

0.61 ± 0.06a

0.61 ± 0.04a

0.54 ± 0.04a

1.0 ± 0.0b

0.56 ± 0.07a

0.49 ± 0.06a

1.0 ± 0.1a

1.3 ± 0.0b

1.7 ± 0.0c

1.6 ± 0.1c

–

–

–

a

1.0 ± 0.1

1.6 ± 0.1

2.4 ± 0.2

3.2 ± 0.5

1.0 ± 0.1

1.7 ± 0.2

2.5 ± 0.3c

1.0 ± 0.1a

1.4 ± 0.1a

2.3 ± 0.2b

3.1 ± 0.4c

1.0 ± 0.1a

1.5 ± 0.2a

2.1 ± 0.2b

1.0 ± 0.1a

1.4 ± 0.1a

2.3 ± 0.1b

2.9 ± 0.4b

1.0 ± 0.1a

1.6 ± 0.2b

2.1 ± 0.2b

1.0 ± 0.1a

1.2 ± 0.1a

1.6 ± 0.1b

1.7 ± 0.2b

1.0 ± 0.0a

1.4 ± 0.1b

1.6 ± 0.1c

1.0 ± 0.0a

2.2 ± 0.2b

4.3 ± 0.3c

3.8 ± 0.5c

1.0 ± 0.1a

2.1 ± 0.1b

3.2 ± 0.3c

1.0 ± 0.1a

1.3 ± 0.1a

2.0 ± 0.1b

2.1 ± 0.2b

1.0 ± 0.0a

1.6 ± 0.1b

2.1 ± 0.1c

1.0 ± 0.1a

1.1 ± 0.1a

1.6 ± 0.1b

1.6 ± 0.2b

–

–

–

a

a

b

b

1.0 ± 0.1

ab

1.4 ± 0.3

bc

3.6 ± 0.6

c

3.9 ± 1.0

a

a

1.0 ± 0.1

b

a

1.2 ± 0.1

4.1 ± 0.5b

is involved in the import of fatty acids and/or fatty acyl-CoAs into peroxisomes [19]. The sesamin-dependent change was not significant. In experiment 2, a diet containing 10% sesame seed, compared with a control diet free of sesame seed, caused changes more than 1.5- and 2-fold in the expression of 135 (91 genes were up- and 44 downregulated) and 55 genes (43 genes were upand 12 downregulated), respectively. The number of genes affected approximately doubled with a diet containing 20% of sesame seed. This diet caused changes more than 1.5- and 2-fold in the expression of 302 (212 genes were up- and 90 downregulated) and 123 genes (101 genes were up- and 22 downregulated), respectively. As expected, sesame seed dose-dependently increased the mRNA expression of many

Sesame Lignans and Gene Expression

17

hepatic enzymes involved in fatty acid oxidation (table 1). The magnitude of the increases observed with a diet containing 20% sesame seed (this diet had 0.232% sesamin and 0.098% sesamolin) was comparable to that observed with a diet containing 0.2% episesamin or sesamolin in experiment 1. We also observed that sesame seed increased mRNA levels of Pex11a and Fxc1 which are involved in the biogenesis of peroxisomes and mitochondria, respectively. In addition, sesame seed dosedependently increased the mRNA expression of Pex19 involved in the proliferation of peroxisomes [20]. As observed with dietary lignans, sesame seed increased mRNA levels of Mgll, Pank1, Slc22a5, Slc25a20, and Cd36, but decreased the mRNA level of Acaca. Sesame-dependent changes in the mRNA expression of Mlycd and Abcd3 were attenuated compared with those obtained with sesame lignans. The mRNA level of Mlycd was comparable between rats fed a control diet (1.0 ± 0.1-fold) and those fed a diet containing 10% sesame seed (1.1 ± 0.1-fold). The value was slightly but significantly higher in rats fed a diet containing 20% sesame seed (1.3 ± 0.1-fold) than in the animals fed a control diet. However, no significant differences were observed in the mRNA expression of Abcd3 among the groups. Effect of Sesame Lignans and Sesame Seed on the Serum and Liver Concentrations of Lignans, Liver Concentrations of Carnitine, and Liver mRNA Levels Sesamin, episesamin or sesamolin, but no other lignan, was detected in rats fed the respective lignan (experiment 1; table 2). Lignans were detected neither in serum nor in liver in rats fed a lignan-free diet. Serum episesamin and sesamolin concentrations in rats fed the corresponding lignan were 4.4- and 11.9-fold higher, respectively, than the sesamin concentration in rats given sesamin. Also, the serum sesamolin concentration was 2.7-fold the concentration of episesamin in rats fed the respective lignan. Sesamolin compared with both episesamin and sesamin, and episesamin compared with sesamin, accumulated more in the liver as well. Hepatic levels of episesamin and sesamolin were 3.5- and 6.6-fold greater, respectively; the levels of sesamin and sesamolin were 1.9-fold that of episesamin. Sesamin and sesamolin were detected in serum and liver in rats fed diets containing sesame seed (experiment 2). Although the concentration of sesamin in sesame seed was twice that of sesamolin, the sesamolin concentration greatly exceeded the sesamin concentration in both serum and liver. Sesame seed dose-dependently increased the serum and liver concentrations of these lignans. Sesame seed greatly increased the hepatic concentration of carnitine (experiment 2). Values were about 3 and 5 times higher in rats given diets containing 10 and 20% sesame seed, respectively, than in the animals fed a control diet. We also analyzed mRNA levels of Crot, Acox1, Ehhadh, Cpt2, Hadha and Hadhb involved in hepatic fatty acid oxidation by real-time PCR in both experiments. The results obtained using this methodology were consistent with those obtained with the DNA microarray (table 1). In addition, real-time PCR analyses in experiment 2 confirmed the findings of the microarray analysis that sesame seed increased the mRNA

18

Ide · Nakashima · Iida · Yasumoto · Katsuta

expression of Mgll, Slc22a5, Slc25a20 and Cd36. All these changes ranged within the expected values, as observed using the microarray.

Discussion

Using DNA microarray technology, we have confirmed previous findings [4, 7–9, 12] that various lignans increased the gene expression of hepatic enzymes involved in fatty acid oxidation. Moreover, we showed that the physiological activities enhancing the gene expression of hepatic enzymes involved in fatty acid oxidation were much stronger with episesamin and sesamolin than with sesamin, and comparable between the former lignans. As lignans increased mRNA levels of Cyp4a1, Cyp4a10 and Cyp4a3, they may also stimulate ω-oxidation of fatty acids [21, 22]. Since lignans, especially episesamin and sesamolin, upregulated the gene expression of Fxc1 and Pex11a, they may also stimulate the proliferation of mitochondria and peroxisomes. We observed that lignans, particularly episesamin and sesamolin, affected the expression of not only genes for the enzymes involved in fatty acid oxidation, but also those for various proteins related to fatty acid metabolism, as summarized in figure 2. The upregulation of the mRNA expression of Cd36 by lignans indicated that the compounds increase the hepatic uptake of fatty acid from the blood stream to supply the substrate for fatty acid oxidation. Dietary lignans also increased the gene expression of mitochondrial acylcarnitine transporter (Slc25a20) [23] and peroxisomal fatty acid/acyl-CoA transporter (Abcd3) [19]. These changes should facilitate the delivery of fatty acids into these organelles. It has been suggested that Mgll is involved in complementing the action of lipoprotein lipase and hepatic lipase in degrading triacylglycerol from lipoproteins in the liver [16]. The upregulation by lignans of Mgll expression may increase the supply of free fatty acid as a substrate for β-oxidation. All the lignans significantly increased the mRNA expression of Slc22a5, and the increases were greater with episesamin and sesamolin than with sesamin. Upregulation in the liver of the expression of this protein is expected to increase the availability of carnitine and hence stimulate mitochondrial transport of fatty acid to be oxidized in this organelle [18]. Episesamin and sesamolin, but not sesamin, caused a significant increase in the expression of Pank1. Therefore, it is expected that these lignans also increase the availability of CoA in the liver to activate fatty acids [17]. All these observations indicated that sesame lignans, especially episesamin and sesamolin, not only upregulate the gene expression of various enzymes involved in hepatic fatty acid oxidation, but also coordinately modify the expression of various proteins involved in regulating fatty acid transport into hepatocytes and their organelles, as well as the availability of substances required for fatty acid oxidation (carnitine and CoA) to stimulate hepatic fatty acid oxidation. A reduction in hepatic lipogenesis appears to be an alternative mechanism for the lipid-lowering effect of lignans. As various lignans decreased the mRNA expression

Sesame Lignans and Gene Expression

19

Table 2. Effect of sesame lignans (experiment 1) and sesame seed (experiment 2) on the serum and liver concentrations of lignans, liver concentrations of carnitine, and liver mRNA levels Experiment 1

Experiment 2

Dietary lignans

Dietary sesame

lignan-free

sesamin

episesamin

sesamolin

0%

10%

20%

Sesamin

ND

11.2 ± 3.9

ND

ND

ND

1.11 ± 0.18a 2.13 ± 0.40b

Episesamin

ND

ND

49.3 ± 1.6

ND

ND

ND

ND

Sesamolin

ND

ND

ND

132 ± 14

ND

19.5 ± 4.1a

35.9 ± 5.7b

Total

ND

11.2 ± 3.9a

49.3 ± 1.6b

132 ± 14c

ND

20.6 ± 4.2a

38.1 ± 6.0b

Sesamin

ND

2.26 ± 0.30

ND

ND

ND

0.463 ± 0.101a

1.13 ± 0.18b

Episesamin

ND

ND

6.17 ± 0.99

ND

ND

ND

Sesamolin

ND

ND

ND

11.8 ± 1.4

ND

1.16 ± 0.23a 1.96 ± 0.28b

Total

ND

2.26 ± 0.30a 6.17 ± 0.99b 11.8 ± 1.4c

ND

1.63 ± 0.33a 3.09 ± 0.38b

–

–

–

95.5 ± 4.2a

305 ± 15b

495 ± 14c

Lignans Serum, μg/dl

Liver, μg/g

Liver carnitine, nmol/g mRNA level in the liver (fold change) Crot

1.0 ± 0.0a

2.6 ± 0.2b

6.7 ± 0.4c

6.4 ± 0.6c

1.0 ± 0.1a

2.6 ± 0.1b

4.2 ± 0.4c

Acox1

1.0 ± 0.1a

1.6 ± 0.1b

3.7 ± 0.2c

3.3 ± 0.2c

1.0 ± 0.1a

1.6 ± 0.1b

2.5 ± 0.1c

Ehhadh

1.0 ± 0.1a

2.9 ± 0.4b

12 ± 1d

9.8 ± 0.8c

1.0 ± 0.1a

3.0 ± 0.4b

8.2 ± 1.1c

Cpt2

1.0 ± 0.0a

1.8 ± 0.1b

3.0 ± 0.1c

3.0 ± 0.2c

1.0 ± 0.1a

2.0 ± 0.2b

2.7 ± 0.2c

Hadha

1.0 ± 0.0a

1.4 ± 0.1b

2.0 ± 0.1c

2.0 ± 0.1c

1.0 ± 0.1a

1.5 ± 0.0b

1.7 ± 0.1c

Hadhb

1.0 ± 0.0a

1.6 ± 0.1b

3.0 ± 0.2c

2.6 ± 0.2c

1.0 ± 0.0a

1.7 ± 0.1b

1.9 ± 0.1c

Mgll

–

–

–

–

1.0 ± 0.1a

1.7 ± 0.2b

2.3 ± 0.3b

Slc22a5

–

–

–

–

1.0 ± 0.0a

1.8 ± 0.1b

2.4 ± 0.2c

Slc25a20

–

–

–

–

1.0 ± 0.0a

1.9 ± 0.1b

2.2 ± 0.3b

Cd36

–

–

–

–

1.0 ± 0.1a

1.6 ± 0.0b

3.7 ± 0.6c

Values represent means ± SE for 7 rats. Values with different superscripts differ significantly at p < 0.05. ND = Not detected.

20

Ide · Nakashima · Iida · Yasumoto · Katsuta

Plasma membrane

Citric acid

Mitochondria

Ketone bodies Acat1 Hmgcl

TCA cycle Acetyl-CoA Mlycd

Acetyl-CoA

Citric acid

Acaca

␤-Oxidation Malonyl-CoA

Slc22a5

Acyl-CoA Carnitine

Carnitine

Acyl-carnitine Cpt1a Cpt1b Slc25a20

Cpt2 Acyl-carnitine

Acadvl, Hadha, Hadhb, Acaa2, Decr1, Dci Downregulation

Acyl-CoA Pantothenic acid

Slc25a20

Pank1

Upregulation Acyl-carnitine

CoA Fatty acid

Crot

Abcd3

Cd36 Fatty acid Mgll Monoacylglycerol

Acyl-CoA

Acyl-CoA

Acyl-CoA ␤-Oxidation (medium chain)

Acox1, Ech1, Ehhadh, Acaa1, Decr2 Peroxisomes

Triacylglycerol

Fig. 2. Sesame lignan-dependent changes in the expression of genes involved in the regulation of hepatic fatty acid oxidation.

of Acaca to a similar level, it is thought that they decreased the production of malonyl-CoA as the substrate for lipogenesis to a similar extent. In addition, we observed that lignans increased the mRNA level of Mlycd, which catalyzes the breakdown of malonyl-CoA [24]. We found that the increases were stronger with episesamin and sesamolin than with sesamin. Therefore, it is likely that the availability of malonylCoA as a substrate for lipogenesis is lower in rats fed episesamin and sesamolin than in those fed sesamin. Apart from its role as a precursor for fatty acid biosynthesis, malonyl-CoA plays an important role in regulating hepatic fatty acid oxidation as a potent inhibitor of carnitine palmitoyltransferase I [24]. Therefore, the lignan-dependent changes in the gene expression of Acaca and Mlycd may stimulate mitochondrial transport of fatty acids as substrates for β-oxidation. Observations of hepatic gene expression profiles in the present study support the idea that various sesame lignans are the natural agonists for peroxisome proliferatoractivated receptor-α (PPARα). Studies have showed that various enzymes involved in fatty acid oxidation located in peroxisomes and mitochondria are upregulated by PPARα [25]. More recently, information suggests that PPARα agonists affect the

Sesame Lignans and Gene Expression

21

expression of diverse genes in addition to the genes related to fatty acid oxidation. Some of the changes may be the result of secondary metabolic events occurring after the activation of PPARα. In the present study, we observed that lignans not only upregulated the gene expression of enzymes involved in fatty acid oxidation, but also affected the expression of many genes involved in lipid and carbohydrate metabolism. The mRNA expression of appreciable numbers of these genes has also been observed to be affected in a similar fashion by various synthetic PPARα agonists in the liver of experimental animals or in cultured hepatocytes [14, 17, 18, 21, 24, 26, 27]. Among the three lignans tested in the present study, episesamin and sesamolin were generally stronger than sesamin in altering the expression of various genes. The binding affinity for, and hence the ability to activate PPARα may be greater for episesamin and sesamolin than for sesamin. However, the observation that liver and serum sesamin levels were much lower than episesamin and sesamolin levels in rats fed the respective lignan raises the possibility that a difference in bioavailability is a crucial factor responsible for the divergent effects of lignans on hepatic gene expression. However, even though twice as much sesamolin as episesamin accumulated in the liver, the impact on hepatic gene expression was comparable between these compounds. Therefore, it is expected that episesamin has a more profound effect on gene expression than sesamolin if these compounds are indistinguishable in terms of bioavailability. As expected, diets containing lignans in the form of sesame seeds also strongly increased the mRNA expression of many enzymes involved in hepatic fatty acid oxidation. In addition, sesame seed affected the gene expression of various proteins related to the regulation of hepatic fatty acid metabolism in a manner similar to that observed with sesame lignans in experiment 1. In addition, we observed that the sesame seed-dependent increase in the gene expression of a protein involved in carnitine transport across the plasma membrane (Slc22a5) in the liver was associated with a large increase in the concentration of carnitine in this tissue. A diet containing 20% sesame seed had about 0.2% sesamin and 0.1% sesamolin. However, concentrations of lignans in the liver and serum in rats fed sesame seed appeared considerably lower than those expected from the results obtained among the animals fed various lignans. In fact, serum and liver concentrations of sesamin in rats given a diet containing this compound were about 5 and 2 times higher, respectively, than the values in the animals fed a diet containing 20% sesame seed despite the fact that these diets contained comparable amounts of sesamin. Also, although sesamolin concentration was about 0.1% in the diet containing 20% sesame seed, serum and liver concentrations of this compound in the animals fed this diet were merely 27 and 17% of those observed with a diet containing 0.2% sesamolin. Therefore, it is suggested that lignans were less absorbable when supplied in the diet as sesame seed than when the purified compounds were added to diets. Despite the fact that the accumulation of lignans in serum and liver was attenuated in the rats fed sesame seed compared to the animals fed purified lignan compounds, the increases in the gene expression of various hepatic fatty acid oxidation enzymes in rats fed a diet containing 20% sesame

22

Ide · Nakashima · Iida · Yasumoto · Katsuta

seed were comparable to those obtained with the diet containing 0.2% episesamin or sesamolin. Therefore, there is a possibility that compound(s) other than sesamin and sesamolin are also involved in the sesame seed-dependent increase in the gene expression of enzymes related hepatic fatty acid oxidation. In conclusion, the DNA microarray analysis showed that dietary sesame lignans (sesamin, episesamin and sesamolin) profoundly affect gene expression profiles in the liver. The analysis not only confirmed the previous finding that lignans increased the gene expression of enzymes involved in hepatic fatty acid oxidation, but also showed that they modified the expression of proteins involved in the transportation of fatty acids into hepatocytes and their organelles, and regulated hepatic concentrations of carnitine, CoA and malonyl-CoA. All these changes should facilitate the oxidation of fatty acids in hepatocytes. The changes were generally greater with episesamin and sesamolin than with sesamin. The differences in bioavailability among these lignans appear to be important in terms of the divergent physiological activity of these compounds. The diets containing sesame seed also affected the gene expression of proteins related to fatty acid oxidation in a manner similar to that observed with diets containing purified sesame lignans. However, analyses of serum and liver concentrations of sesamin and sesamolin in rats fed sesame seed raised the possibility that compound(s) other than these lignans are also involved in the sesame seeddependent increase in the gene expression of hepatic fatty acid oxidation enzymes.

References 1 Sirato-Yasumoto S, Katsuta M, Okuyama Y, Takahashi Y, Ide T: Effect of sesame seeds rich in sesamin and sesamolin on fatty acid oxidation in rat liver. J Agric Food Chem 2001;49:2647–2651. 2 Katsuzaki H, Kawakishi S, Osawa T: Sesaminol glucosides in sesame seeds. Phytochemistry 1994;35: 773–776. 3 Fukuda Y, Nagata M, Osawa T, Namiki M: Contribution of lignan analogues to antioxidative activity of refined unroasted sesame seed oil. J Am Oil Chem Soc 1986;63:1027–1031. 4 Ashakumary L, Rouyer IA, Takahashi Y, Ide T, Fukuda N, Aoyama T, Hashimoto T, Mizugaki M, Sugano M: Sesamin, a sesame lignan, is a potent inducer of hepatic fatty acid oxidation in the rat. Metabolism 1999;48:1303–1313. 5 Hirose N, Inoue T, Nishihara K, Sugano M, Akimoto K, Shimizu S, Yamada H: Inhibition of cholesterol absorption and synthesis in rats by sesamin. J Lipid Res 1991;32:629–638. 6 Hirata F, Fujita K, Ishikura Y, Hosoda K, Ishikawa T, Nakamura H: Hypocholesterolemic effect of sesame lignan in humans. Atherosclerosis 1996;122:135– 136.

Sesame Lignans and Gene Expression

7 Kushiro M, Masaoka T, Hageshita S, Takahashi Y, Ide T, Sugano M: Comparative effect of sesamin and episesamin on the activity and gene expression of enzymes in fatty acid oxidation and synthesis in rat liver. J Nutr Biochem 2002;13:289–295. 8 Lim JS, Adachi Y, Takahashi Y, Ide T: Comparative analysis of sesame lignans (sesamin and sesamolin) in affecting hepatic fatty acid metabolism in rats. Br J Nutr 2007;97:85–95. 9 Tsuruoka N, Kidokoro A, Matsumoto I, Abe K, Kiso Y: Modulating effect of sesamin, a functional lignan in sesame seeds, on the transcription levels of lipidand alcohol-metabolizing enzymes in rat liver: a DNA microarray study. Biosci Biotechnol Biochem 2005;69:179–188. 10 Akimoto K, Kigawa Y, Akamatsu T, Hirose N, Sugano M, Shimizu S, Yamada H: Protective effect of sesamin against liver damage caused by alcohol or carbontetrachloride in rodents. Ann Nutr Metab 1993;37:218–224. 11 Ide T: Interaction of fish oil and conjugated linoleic acid in affecting hepatic activity of lipogenic enzymes and gene expression in liver and adipose tissue. Diabetes 2005;54:412–423.

23

12 Kushiro M, Takahashi Y, Ide T: Species differences in the physiological activity of dietary lignan (sesamin and episesamin) in affecting hepatic fatty acid metabolism. Br J Nutr 2004;91:377–386. 13 Pearson DJ, Chase JFA, Tubbs PK: The assay of (-)-carnitine and its O-acyl derivatives. Method Enzymol 1969;14:612–622. 14 Schrader M, Reuber BE, Morrell JC, JimenezSanchez G, Obie C, Stroh TA, Valle D, Schroer TA, Gould SJ: Expression of PEX11β mediates peroxisome proliferation in the absence of extracellular stimuli. J Biol Chem 1998;273:29607–29614. 15 Rothbauer U, Hofmann S, Mühlenbein N, Paschen SA, Gerbitz KD, Neupert W, Brunner M, Bauer MF: Role of the deafness dystonia peptide 1 (DDP1) in import of human Tim23 into the inner membrane of mitochondria. J Biol Chem 2001;276:37327– 37334. 16 Karlsson M, Contreras JA, Hellman U, Tornqvist H, Holm C: cDNA cloning, tissue distribution, and identification of the catalytic triad of monoglyceride lipase. Evolutionary relationship to esterases, lysophospholipases, and haloperoxidases. J Biol Chem 1997;272:27218–27223. 17 Ramaswamy G, Karim MA, Murti KG, Jackowski S: PPARα controls the intracellular coenzyme A concentration via regulation of PANK1α gene expression. J Lipid Res 2004;45:17–31. 18 Luci S, Geissler S, König B, Koch A, Stangl GI, Hirche F, Eder K: PPARα agonists up-regulate organic cation transporters in rat liver cells. Biochem Biophys Res Commun 2006;350:704–708. 19 Wanders RJ, Visser WF, van Roermund CW, Kemp S, Waterham HR: The peroxisomal ABC transporter family. Pflugers Arch 2007;453:719–734. 20 Sacksteder KA, Jones JM, South ST, Li X, Liu Y, Gould SJ: PEX19 binds multiple peroxisomal membrane proteins, is predominantly cytoplasmic, and is required for peroxisome membrane synthesis. J Cell Biol 2000;148:931–944.

21 Hardwick JP, Song BJ, Huberman E, Gonzalez FJ: Isolation, complementary DNA sequence, and regulation of rat hepatic lauric acid ω-hydroxylase (cytochrome P-450LAω). Identification of a new cytochrome P-450 gene family. J Biol Chem 1987;262: 801–810. 22 Kimura S, Hardwick JP, Kozak CA, Gonzalez FJ: The rat clofibrate-inducible CYP4A subfamily. II. cDNA sequence of IVA3, mapping of the Cyp4a locus to mouse chromosome 4, and coordinate and tissue-specific regulation of the CYP4A genes. DNA 1989;8:517–525. 23 Indiveri C, Iacobazzi V, Giangregorio N, Palmieri F: The mitochondrial carnitine carrier protein: cDNA cloning, primary structure and comparison with other mitochondrial transport proteins. Biochem J 1997;321:713–719. 24 Lee GY, Kim NH, Zhao ZS, Cha BS, Kim YS: Peroxisomal-proliferator-activated receptor α activates transcription of the rat hepatic malonyl-CoA decarboxylase gene: a key regulation of malonylCoA level. Biochem J 2004;378:983–990. 25 Schoonjans K, Staels B, Auwerx J: Role of the peroxisome proliferator-activated receptor (PPAR) in mediating the effects of fibrates and fatty acids on gene expression. J Lipid Res 1996;37:907–925. 26 Motojima K, Passilly P, Peters JM, Gonzalez FJ, Latruffe N: Expression of putative fatty acid transporter genes are regulated by peroxisome proliferator-activated receptor α and γ activators in a tissue- and inducer-specific manner. J Biol Chem 1998;273: 16710–16714. 27 Cherkaoui-Malki M, Meyer K, Cao WQ, Latruffe N, Yeldandi AV, Rao MS, Bradfield CA, Reddy JK: Identification of novel peroxisome proliferator-activated receptor α (PPARα) target genes in mouse liver using cDNA microarray analysis. Gene Expr 2001;9:291–304.

Dr. Takashi Ide Laboratory of Nutritional Function, Division of Food Functionality, National Food Research Institute Kannondai 2-1-12 Tsukuba 305-8642 (Japan) Tel. +81 29 838 8083, Fax +81 29 838 7996, E-Mail [email protected]

24

Ide · Nakashima · Iida · Yasumoto · Katsuta

Genomics Yoshikawa T (ed): Food Factors for Health Promotion. Forum Nutr. Basel, Karger, 2009, vol 61, pp 25–38

Genome Science of Lipid Metabolism and Obesity Nobuyuki Takahashi ⭈ Tsuyoshi Goto ⭈ Shizuka Hirai ⭈ Taku Uemura ⭈ Teruo Kawada Laboratory for Molecular Function of Food, Graduate School of Agriculture, Kyoto University, Kyoto, Japan

Abstract Abnormalities in lipid metabolism cause obesity leading to metabolic syndrome. Thus, improvement of the abnormalities is significant for the treatment of metabolic syndrome. Nuclear receptors activated by specific ligands regulate lipid metabolism at the genomic level. The expression of lipid metabolism-related enzymes is increased or decreased by the activity of various nuclear receptors. The regulation of enzyme expression is mediated by specific response elements to each nuclear receptor in promoters of target genes. Many food factors acting as agonists or antagonists control the activities of nuclear receptors. Here, we provide several examples of food factors acting as agonists or antagonists, which are useful for the management of obesity accompanied by lipid metaboCopyright © 2009 S. Karger AG, Basel lism abnormalities.

Recently, metabolic syndrome has become a severe health and social problem. The syndrome is accompanied by obesity with lipid metabolism abnormalities which are often due to daily excess energy intake. Therefore, improvement of these abnormalities is significant for the treatment of obesity and the metabolic syndrome. Lipid metabolism is regulated by various factors such as glucose availability, energy expenditure rate, and stored lipid amounts. Short-term regulation of lipid metabolism (in terms of seconds and minutes) is mediated by changes in enzymatic activities. On the other hand, in the middle- and long-term ranges (in terms of hours and days), the increase or decrease in the mRNA expression of genes encoding lipid metabolism-related enzymes is regulated. Many transcriptional factors are involved in the regulation of mRNA expression. The most significant factor is the nuclear receptor superfamily (table 1). The nuclear receptors have common structures and functions. The transcriptional factors are activated by small hydrophobic molecules acting as ligands, which the plasma membrane is permeable to, and then positively or negatively regulate

Table 1. Nuclear receptor superfamily Steroid hormone receptors Estrogen receptor Androgen receptor Progesterone receptor Glucocorticoid receptor Mineralocorticoid receptor Homodimer orphan receptors Retinoid X receptor EAR Hepatocyte nuclear factor Monomer orphan receptors Receptor tyrosine kinase-like orphan receptor Estrogen receptor-related receptor Rev-erb Retinoid X receptor heterodimer receptors PPAR LXR FXR PXR Retinoic acid receptor Thyroid hormone receptor Vitamin D receptor Constitutive androstane receptor

mRNA expression of target genes. Members of the superfamily have two distinct and conserved functional domains, an N-terminal DNA-binding domain (DBD) and a C-terminal ligand-binding domain (LBD). The structure of the N-terminal DBD (approximately 80 amino acids) is similar among the members (fig. 1). This domain is composed of two typical Zn finger motifs recognizing specific sequences with six nucleotides. Most nuclear receptors generally form a hetero- or homodimer, so that a nuclear receptor complex binds to two copies of the sequence in a promoter region. LBD is a C-terminal region with approximately 250 amino acids containing 12 α-helixes, which contributes to not only the binding of ligands but also the formation of a dimer and the interaction of other regulatory proteins. The nuclear receptors require other accessory proteins for the regulation of the target gene expression. After a ligand binds to a nuclear receptor, various cofactors are recruited to the ligand-nuclear receptor complex. For the induction of target gene expression, coactivators should interact with the nuclear receptor (fig. 2). The association of coactivators is dependent on ligand binding to a nuclear receptor (therefore, the

26

Takahashi · Goto · Hirai · Uemura · Kawada

DBD About 80 amino acids

LBD About 250 amino acids

Binding to coactivators

Binding to coactivators

Binding to a response element Dimerization Nuclear translocation

Binding to a ligand Dimerization Binding to cofactors

Fig. 1. Common structure of nuclear receptors. Nuclear receptors have two distinct domains: DBD and LBD. Other regions that are significant for nuclear receptor activation are shown.

Coactivators

Ligands

Coactivators RNA polymerase complex

Target gene expression

Response element

Promoter (i.e. TATA box)

Fig. 2. Nuclear receptor complex. Ligand-bound nuclear receptors bind to a response element on the target gene promoter with various cofactors such as coactivators. Some cofactors mediate the recruitment of the RNA polymerase complex. Then, the expression of target genes is induced.

association is used to examine whether a compound serves as a ligand for nuclear receptors: ‘coactivator pull-down assay’). General coactivators bind to many nuclear receptors, which contribute to the chromatin remodeling due to histone acetylation and the association to other transcriptional complexes such as RNA polymerase. The general coactivators include CREB-binding protein (CBP)/p300, steroid receptor coactivator (SRC), and transcriptional intermediary factor-1 (TIF1). Other coactivators associate only to restricted nuclear receptors. For example, peroxisome proliferator-activated receptor-γ (PPARγ) coactivator-1 (PGC1), which was originally identified as a PPARγspecific coactivator, interacts with only a few nuclear receptors such as PPARα and EER. These coactivators have distinct in vivo expression patterns. Thus, nuclear receptors interact with different coactivators in different types of cell. This might be the reason why nuclear receptors induce the expression of different genes in different cells. The ligands of nuclear receptors are numerous, including sterol-, fatty acid-, and carotenoid-related compounds. For example, PPARγ is activated by fatty acids, a

Genome Science of Lipid Metabolism and Obesity

27

prostaglandin J2 derivative, isoprenols, and several terpenoids. Moreover, fatty acids with different carbon chains and/or saturations show different activities as the PPARγ ligand (generally, long-chain and/or unsaturated fatty acids are more highly active than short-chain and/or saturated ones). Thus, nuclear receptors sense various indicators (ligands) and respond at the cellular level to changes in the indicators. Moreover, the ligands are not only endogenous but also exogenous compounds including food factors, which have structures similar to those of endogenous ligands. Therefore, such food factors can regulate nuclear receptors as agonists or antagonists. In this review, the genomic regulation of lipid metabolism by various food factors in vitro and in vivo is described.

Lipid Metabolism Regulation by Nuclear Receptor Activity

Many nuclear receptors are involved in the regulation of lipid metabolism. Nuclear receptors are involved in the changes in cellular gene expression. Particularly, the expression of rate-limiting enzymes is often controlled by nuclear receptors. These expression controls are important targets of food factors to improve the lipid metabolism abnormalities. Here, we focus on lipid metabolism-related enzymes and explain how the enzyme expression is regulated by nuclear receptors in vitro and in vivo.

Fatty Acid and Triglyceride Synthesis (SREBP1, FAS, ACC and SCD) Sterol response element-binding protein-1 (SREBP1) is a key transcriptional factor in de novo fatty acid synthesis. Many target genes of SREBP are involved in the de novo fatty acid synthesis. The SREBP1 protein is firstly expressed as a membrane-integrated precursor. When activated, specific proteases cleave the precursor to release a cytoplasmic domain serving as an active transcriptional factor to induce expression of target genes. SREBP1 stimulates the expression of fatty acid synthase (FAS) and acyl-CoA carboxylase (ACC), which are important enzymes for de novo fatty acid synthesis. Functional FAS complex is a homodimer of two identical subunits with multiple catalytic properties encoded by a single-copy gene. A substrate of the FAS-mediated fatty acid synthesis is a malonyl-CoA derived from acetyl-CoA. This reaction to produce malonyl-CoA is catalyzed by ACC. Thus, both FAS and ACC are indispensable enzymes for the de novo fatty acid synthesis. Palmitic acid (C16:0), an end product of the synthesis, is then elongated by a specific enzyme to produce stearic acid (C18:0). Finally, these saturated fatty acids are desaturated to produce oleic acid (C18:1). This is mediated by stearoyl-CoA desaturase-1 (SCD1). This enzymatic activity is significant for the maintenance of fatty acid composition.

28

Takahashi · Goto · Hirai · Uemura · Kawada

Liver X receptor (LXR) and pregnane X receptor (PXR) stimulate de novo fatty acid synthesis in the liver. The LXR/PXR-dependent stimulation is mediated by SREBP through LXR/PXR-dependent induction of hepatic SREBP1 expression. The SREBP1 promoter has both an LXR response element (LXRE) and a PXR response element [1]. Therefore, the activation of both nuclear receptors induces SREBP1 expression. The effects of LXR and PXR on the accumulation of hepatic triglyceride (TG) are mainly mediated by SREBP1 activity. The induced SREBP1 increases the mRNA levels of FAS via SREBP response elements in promoter regions of the FAS gene [2]. The FAS promoter also has an LXRE, indicating that LXR directly stimulates FAS expression [3]. In addition, FAS expression is also induced by farnesoid X receptor (FXR), which is activated by bile acid in the liver [4]. This is mediated via LXRE because FXR can bind to LXRE in many gene promoters. Although the physiological significance of this is yet unclear, the FXRdependent regulation of FAS expression is very interesting in the context of crosstalk between lipogenesis and cholesterol homeostasis in the liver. Finally, PPARγ also induces FAS expression in adipocytes, although the FAS promoter has no typical PPAR response element (PPRE) [5]. The mechanism of the PPARγ-dependent regulation is unknown. However, FAS activity is required for TG accumulation in adipocytes and actually increases during adipocyte differentiation. ACC is also an SREBP1 target gene in the liver. The ACC promoter has a sterol response element (SRE) [6], indicating that ACC expression is induced by SREBP1. This suggests the possibility that LXR and PXR could stimulate ACC expression via SREBP1 induction. Actually, it has been reported that hepatic ACC expression is stimulated by LXR activation via SREBP1 induction, although PXR-dependent stimulation of ACC expression has not yet been reported. Moreover, ACC is directly regulated by LXR through LXRE in the promoter region, as well as FAS [7]. However, ACC activity is controlled by phosphorylation at the enzyme activity level rather than at the mRNA transcription level. Therefore, the nuclear receptor-mediated regulation of ACC might be an accessory-regulating system in total ACC activity. Expression of SCD1 is induced by SREBP1 through SRE of the SCD1 promoter. Therefore, LXR and FXR regulate SCD1 expression via SREBP1 induction as well as FAS and ACC expressions [8]. On the other hand, it has been well known that polyunsaturated fatty acids (PUFAs) suppress SCD1 expression in the liver. This suggests that PPARα, which is activated by PUFAs in the liver, might be involved in the hepatic regulation of SCD1 expression. Actually, treatment with WY-14643 (a synthetic PPARα agonist) decreased the mRNA level of the SCD1 gene in the liver [9]. However, in vitro analysis of the SCD1 promoter demonstrated that a PPRE of the SCD1 promoter is distinct from a PUFA response element which is responsible for the PUFA-dependent suppression of SCD1 expression [10]. This suggests that the PUFA-dependent suppression of SCD1 expression is independent of PPARα activity in hepatocytes. At least it is clear that PPARα activation suppresses SCD1 expression in the liver, although the details remain unclear.

Genome Science of Lipid Metabolism and Obesity

29

Lipolysis (HSL and ATGL) Adipocytes release free fatty acid (FFA) derived from accumulated TG to supply energy to other tissues such as cardiac and skeletal muscles. The liver absorbs the adipocyte-derived FFA to metabolize it and releases ketone bodies for neurons, which can use ketone bodies under hypoglycemic conditions. Therefore, lipase activity to hydrolyze TG into FFA and glycerols is indispensable for the maintenance of systemic energy balance. Adipocyte lipolysis is mediated by two distinct lipases: hormone-sensitive lipase (HSL) and adipocyte-specific triglyceride lipase (ATGL). HSL is activated through activation of adrenergic receptors to induce hormone-induced lipolysis. ATGL is concerned with adipocyte basal lipolysis to regulate the size of lipid droplets. Both lipases are controlled at mRNA levels together with enzymatic activity levels. HSL is activated by hormones like noradrenalin to increase the FFA blood level under fasting conditions. Adipocyte differentiation induces this lipase expression, suggesting that HSL expression is regulated by PPARγ in adipocytes. Actually, thiazolidinedione (TZD) treatment increases the mRNA expression level of HSL in adipocytes [11]. However, this induction is not direct. Transcription factor specificity protein-1 (Sp1) through the GC box of the HSL promoter mediates the induction of HSL in adipocytes. Deletion of the GC box or inhibition of Sp1 DNA-binding activity markedly reduces the PPARγ-dependent induction of HSL in adipocytes. Therefore, although PPARγ requires the involvement of Sp1, HSL is a PPARγ target gene in adipocytes. Another lipase in adipocytes, ATGL, is also a PPARγ target gene, which is a TG-specific lipase that is induced during adipogenesis and remains highly expressed in mature adipocytes. TZD induces ATGL mRNA [12], although it is still unknown whether the promoter of the ATGL gene has PPRE. Because ATGL is an important enzyme that interacts with adipocyte lipid droplets and controls the volume of the droplets, the promoter of the ATGL gene should be analyzed as soon as possible to elucidate the mechanism of ATGL transcriptional regulation in adipocytes.

Fatty Acid Oxidation (CPT1 and ACO) FFA as a fuel molecule is metabolized via the hepatic β-oxidation pathway to generate NADH used for oxidative phosphorylation in mitochondria. Absorbed FFA is activated to acyl-CoA by acyl-CoA synthase (ACS) as described later. Then, acyl-CoA is transported into mitochondria where the metabolic pathway occurs. Because the mitochondrial membrane is impermeable to acyl-CoA, the fatty acyl-CoA should be transformed to a mitochondrial membrane-permeable form to enter mitochondria directly for oxidation. This transformation is mediated by carnitine palmitoyl transferase-1 (CPT1), an outer mitochondrial membrane-integrated enzyme that catalyzes the transfer of long-chain acyl groups from CoA to carnitine. Newly produced acyl-

30

Takahashi · Goto · Hirai · Uemura · Kawada

carnitine can enter the mitochondrial matrix, where acyl-CoA is reproduced through the opposing reaction mediated by CPT2, another isoform of CPT. Mitochondrial fatty acid oxidation is limited by the CPT1-mediated step. On the other hand, mammalian cells have another organelle for fatty acid oxidation: peroxisomes. In hepatocytes, peroxisomes are considered to contribute up to 50% of the total fatty acid oxidative activity of the liver under average conditions. The fatty acid oxidation in peroxisomes is slightly different from that in mitochondria. In the peroxisomal pathway, the rate-limiting enzyme is acyl-CoA oxidase (ACO) that catalyzes the first step of the pathway. Therefore, CPT1 and ACO expressions are important for the stimulation of mitochondrial and peroxisomal fatty acid oxidations. The expressions of both genes are known to be regulated by nuclear receptors. Expressions of CPT1 and ACO are stimulated by PPAR activation. In hepatocytes, fibrates, PPARα agonists, induce the mRNA expressions of both genes in a manner of PPARα expression [13, 14]. Although the induction is mediated by a functional PPRE in the promoter of each gene, PPARγ cannot induce their expressions in adipocytes. This might be partly because PPRE has isoform specificity and because PPAR activity shows cell specificity dependent on cell-specific cofactors. The difference in target genes between PPAR isoforms results in the difference in fatty acid utilization between hepatocytes and adipocytes, in which fatty acid is degraded into acyl-CoA and utilized for TG synthesis, respectively. In addition, CPT1 expression is suppressed by PXR activation in the liver [15]. Because PXR activation causes hepatic TG accumulation, it is reasonable to suppress hepatic CPT1 expression by PXR activity.

Others (LPL, Fatty Acid Transporters and ACS) Circulating TG in lipoproteins should be hydrolyzed by a specific lipase before cellular uptake via fatty acid transporters. The indispensable step is mediated by lipoprotein lipase (LPL). The expression of this lipase is upregulated during adipocyte differentiation, suggesting that PPARγ is involved in the regulation in adipocytes. Actually, TZD and other PPARγ agonists induce LPL expression in adipocytes. This regulation is mediated through PPRE in the promoter region [16]. In addition, expression of LPL is enhanced by LXR activation [17]. LXR agonist-fed mice exhibit a significant increase in LPL gene expression in the liver and macrophages, but not in adipose tissues and muscle. The LXR-dependent regulation is also mediated by LXRE in the LPL promoter region. The mechanism of the PPAR- and LXR-dependent induction of LPL is necessary for the accumulation of lipid in adipocytes and hepatocytes, respectively. Hydrolyzed FFAs are absorbed into cells by both passive and facilitated diffusions. The facilitated diffusion is mediated by fatty acid transporter proteins. Thus, two distinct proteins are identified as fatty acid transporter proteins: CD36/FAT (fatty acid translocase) and FATPs (fatty acid transport proteins). CD36/FAT is a multifunctional protein that has been reported as a scavenger receptor (an oxidized-LDL receptor)

Genome Science of Lipid Metabolism and Obesity

31

and a thrombospondin receptor. This gene expression is regulated by PPAR activity in various tissues including liver, skeletal muscle, and adipose tissues [18], whereas several reports have shown PPAR-independent expression of the CD36/FAT gene in vitro and in vivo. Therefore, regulation of this gene is very complicated. Actually, the CD36/FAT gene has 3 distinct promoters in the 5⬘-noncoding region and the intron between exons 1 and 2 [19]. However, the promoter of CD36/FAT has a PPRE in the mouse [20]. In our experiments, TZD treatment induces CD36/FAT gene expression by adipocytes in vitro and in vivo. Thus, the differences in the control of this gene expression might be due to the specificity of cells and/or experimental conditions. Recently, it has been reported that LXR and PXR also control the expression of CD36/ FAT in hepatocytes [21]. The PPAR activation induces the expression of another fatty acid transporter, FATP, in adipocytes, liver, and intestine [22]. This transporter is significant in fatty acid uptake in intestinal epithelial cells. Thus, it is appropriate that PPAR, which is activated by dietary fatty acids, regulates FATP in the intestine. Thus far, there is no report that the expression of the FATP gene is regulated by LXR and/ or PXR. Whether LXR/PXR also upregulates FATP gene expression as well as CD36/ FAT is a very interesting question. In addition, ACS is a key enzyme in the metabolism of absorbed fatty acid. The absorbed fatty acid should be transformed into acyl-CoA to be metabolized for both TG synthesis and β-oxidation. This first step in the cellular utilization of the absorbed fatty acid is mediated by ACS. Regulation of ACS expression is regulated by PPAR in hepatocytes and adipocytes. The promoter region of this gene has typical PPRE [22]. It is likely that the PPAR-dependent induction of ACS contributes to β-oxidation in hepatocytes (PPARα dependent) and to TG synthesis in adipocytes (PPARγ dependent). Interestingly, fatty acyl-CoA such as stearoyl-CoA and palmitoyl-CoA serves as a PPAR antagonist [23]. Therefore, the antagonistic effect of acyl-CoA on PPAR activity might be feedback regulation to prevent excess cellular utilization of fatty acid.

Food Factors That Regulate Nuclear Receptor Activity

Agonistic Effects of Food Factors Thus far, our and other groups have identified various agonistic food factors that regulate the activity of nuclear receptors to control dysfunctions of lipid metabolism [24–28]. Particularly, many food factors with PPAR agonistic effects have been found. PPARα is involved in fatty acid oxidation in the liver. Therefore, application of the food factors with PPARα agonist activity enhances the uptake and oxidation of fatty acid in hepatocytes. As a result, hepatic and circulating TG amounts decrease. Therefore, the food factors with PPARα agonist activity are valuable for the treatment of obesity-related hyperlipidemia. On the other hand, PPARγ agonists such as TZD

32

Takahashi · Goto · Hirai · Uemura · Kawada

improve insulin resistance to decrease blood glucose level. This is because they promote adipocyte differentiation to increase the relatively high number of insulin-sensitive small adipocytes. Recently, the combinational application of PPARγ and PPARα agonists has been useful for the treatment of both hyperglycemia and hyperlipidemia. Although PPARγ agonists often cause body weight gain, PPARα activity suppresses the gain through the increase in hepatic lipid consumption. Thus, dual agonists to activate both PPARs are valuable. Example 1: Isoprenols for PPARs Farnesol (a C15 branched-chain fatty alcohol), an isoprenol, serves as a dual agonist for PPARα and PPARγ in ‘in vitro’ experiments [24]. In luciferase ligand assay for PPARs, farnesol activated both PPARα and PPARγ, although the activity of PPARα was more potent than that of PPARγ. Farnesol-related isoprenols such as geranylgeraniol (C20) also activated PPARs, whereas geraniol, another isoprenol (C10), activated PPARγ but not PPARα. Next, to examine the effect of farnesol and geranylgeraniol on the mRNA expression of PPAR target genes, the compounds were added to HepG2 hepatocytes or 3T3-L1 adipocytes. In PPARα-expressing HepG2 cells, both farnesol and geranylgeraniol induced the mRNA expression of PPARα target genes such as ACO and CPT1A. Moreover, the expression of PPARγ target genes in differentiated 3T3-L1 adipocytes (aP2 and LPL) was induced by the addition of farnesol and geranylgeraniol. These results indicate that isoprenols such as farnesol and geranylgeraniol serve as PPAR dual agonists in vitro. Then, to examine the in vivo effects of the isoprenols, we performed animal experiments using diabetic obese KK-Ay mice. In the experiments, we used farnesol, the most potent effect to activate PPARs. Farnesol was fed for 4 weeks with HFD. Body weight and organ weights showed no difference between the farnesol-fed and control HFD-fed mice. Although it had been expected that farnesol is effective on both adipose tissues and liver, only hepatic PPARα target genes were induced in the farnesol-fed mice [Goto and Kawada, unpubl. data]. In the liver, farnesol increased the expression levels of ACO and CPT1A mRNAs and decreased the hepatic TG content, whereas farnesol did not increase the expression levels of aP2 and LPL mRNAs in adipose tissues. However, the blood glucose level during fasting was decreased by the farnesol feeding. The improvement of hyperglycemia might be caused by the increase in the fatty acid oxidation level in the liver to reduce insulin resistance in other tissues. The reason why farnesol was effective only in the liver, but not in adipose tissues, might be due to the degradation of farnesol in the liver so that farnesol itself could not be delivered to adipose tissues. Branched fatty acids and alcohols including farnesol and geranylgeraniol are metabolized via the microsomal α-oxidation pathway. Hepatocytes show very high activity of this pathway to degrade substrates very fast. Of course, we cannot deny the possibility that farnesol has no effect on PPARγ activation in vivo. Further investigations are necessary to answer this question.

Genome Science of Lipid Metabolism and Obesity

33

Example 2: Phytol and Phytanic Acid In the previous section on farnesol, it has been mentioned that the effect of hepatic metabolism might be important for the functions of food factors in vivo. In this context, phytol and phytanic acid are very interesting food factors. We have reported that phytol (a component of chlorophyll) activates PPARα in the luciferase ligand assay and induces mRNA expression of PPARα target genes in hepatocytes [25]. Although phytol showed PPARγ activation activity, the activity was very moderate in comparison with that of PPARα activation. Phytol is metabolized to phytanic acid in the liver. The metabolite can be detected in the blood. Therefore, phytanic acid is a circulating form of phytol after the hepatic metabolism. Interestingly, it has been reported that phytanic acid activates PPARγ and retinoid X receptor-α [29]. In those reports, PPARα activation by phytanic acid is very weak in comparison with PPARγ activation. On the basis of these facts, we can propose an attractive model of functions of phytol as a food factor. First, phytol activates PPARα to induce the expression of its target genes and enhance fatty acid oxidation in the liver to decrease the hepatic and circulating lipid levels. Next, phytol itself is metabolized to phytanic acid to be delivered to adipose tissues, in which phytanic acid induces the expression of PPARγ target genes leading to the improvement of insulin resistance. Interestingly, it has been shown that the gene expression of enzymes involved in the transformation of phytol to phytanic acid is upregulated by PPARα activity in the liver [30]. Therefore, phytol enhances its own transformation to phytanic acid. This model is very attractive because the consideration of metabolites expands the possibility of food factors as PPAR regulators, but we should perform some additional experiments to prove the model.

Antagonistic Effects of Food Factors Agonistic activity of food factors for PPARα is very significant for the management of hyperlipidemia and hyperglycemia as described above. On the other hand, it has been suggested that hyperactivation of PPARγ causes hypertrophy of adipose tissues to induce obesity and obesity-related insulin resistance [31]. Therefore, PPARγ agonists such as TZD should be used together with dietary control to restrict energy uptake for the suppression of obesity. This indicates that the PPARγ activation is not beneficial when there is a high concentration of agonists, whereas the activation is very useful when PPARγ agonists are restricted. This concept is confirmed by the fact that PPARγ hetero-knockout mice (PPARγ+/–) show resistance to HFD-induced hyperglycemia [31]. In the PPARγ hetero-knockout mice, a decrease in PPARγ gene expression causes a decrease in PPARγ activity. The decrease in PPARγ activity suppresses the HFD-induced adipocyte hypertrophy to improve insulin resistance. Therefore, it is likely that PPARγ antagonists are also useful for long-term treatment in the presence of endogenous PPARγ agonists.

34

Takahashi · Goto · Hirai · Uemura · Kawada

LXR also has significant effects on lipid and glucose metabolism, particularly hepatic metabolism. It has been reported that hyperglycemia is improved by LXR activation. Although this mechanism is unknown, this might be partly because LXR activation induces adipocyte differentiation to increase insulin-sensitive small adipocytes, as TZD shows. On the other hand, the hepatic LXR activation induces lipogenesis to cause hepatic TG accumulation (hepatic steatosis). This is due to the LXR-dependent induction of SREBP1 regulating the expression of lipogenesis-related genes in the liver. The hepatic induction of SREBP1 by LXR leads to an increase in circulating lipid content. Therefore, when we focus on the improvement of hyperlipidemia, it is beneficial to suppress the LXR activity in the liver. Example 1: β-Cryptoxanthine for PPARs Dietary foods contain various carotenoids such as carotenes and xanthophylls. β-Cryptoxanthine is a major carotenoid in citrus fruits. We have recently found that this carotenoid serves as a PPARγ antagonist in vivo and in vitro [Ohyama and Kawada, unpubl. data]. Diabetic obese KK-Ay mice were fed HFD containing β-cryptoxanthine for 4 weeks. The β-cryptoxanthine-fed mice showed no significant difference in body and organ weights. However, their fasting blood glucose level decreased. In addition, oral glucose tolerance test demonstrated improvement in insulin resistance after 4 weeks of feeding with β-cryptoxanthine. The number of small adipocytes increased and that of large ones decreased in the adipose tissues of the β-cryptoxanthine-fed mice in comparison with those of control HFD-fed mice. Interestingly, the gene expression of enzymes involved in fatty acid oxidation in the liver was upregulated in the β-cryptoxanthine-fed mice, such that the hepatic TG content of the β-cryptoxanthine-fed mice was lower than that of the control mice. These data of the β-cryptoxanthine-fed mice were very similar to those of the PPARγ hetero-knockout mice. The reason for the hepatic TG decrease may be the increase in the level of circulating leptin secreted by adipocytes. Although the details of the mechanism are yet unknown, the β-cryptoxanthine feeding caused increases in both mRNA and protein levels of leptin in adipose tissues. Actually, the TG content of skeletal muscle also decreased in the β-cryptoxanthine-fed mice accompanied by the increase in the expression of fatty acid oxidation-related enzymes. These results indicate that β-cryptoxanthine inhibits adipocyte hypertrophy to maintain insulin sensitivity and affects the release of adipocytokines such as leptin to improve the insulin resistance of other tissues. To investigate the mechanism of the β-cryptoxanthine effects on the improvement in lipid metabolism, we performed in vitro experiments using culture cells. Luciferase ligand assay showed β-cryptoxanthine-dependent inhibition of TZD-induced PPARγ activation in a dose-dependent manner. Moreover, the β-cryptoxanthine treatment suppressed lipid accumulation and GPDH enzyme activity (a differentiation marker enzyme involved in TG synthesis) 10 days after differentiation induction in 3T3-L1 adipocytes. The expression of PPAR target genes was also inhibited

Genome Science of Lipid Metabolism and Obesity

35

in 3T3-L1 adipocytes by the β-cryptoxanthine treatment. These data suggest that β-cryptoxanthine serves as an antagonist of PPARγ. Although more details should be elucidated for both in vivo and in vitro effects of β-cryptoxanthine, it appears that PPARγ antagonists such as β-cryptoxanthine are valuable for the management of obesity-related insulin resistance. Example 2: Diosgenin for LXRs Diosgenin is the main aglycon of saponin in fenugreek, which is a spice used in the Eastern Mediterranean region to Central Asia and Ethiopia and is widely cultivated in India, Pakistan, and China. Thus far, diosgenin and fenugreek have been reported as food factors that improve hyperglycemia and hyperlipidemia. However, their detailed mechanisms remain unknown. Thus, we examined the effects of fenugreek or diosgenin on lipid metabolism in diabetic obese mice and HepG2. Diabetic obese KK-Ay mice fed fenugreek for 4 weeks inhibited HFDinduced hepatic steatosis in comparison with control HFD-fed mice. This is due to decreased mRNA levels of lipogenesis-related genes such as SREBP1, FAS, and SCD1 in hepatocytes. These results indicate that fenugreek suppressed SREBP1 expression in the liver to inhibit hepatic lipogenesis leading to the improvement of the HFD-induced hepatic steatosis. Interestingly, the fenugreek-fed mice showed significant improvement in insulin resistance in OGTT. Therefore, it is likely that the hepatic suppression of steatosis by fenugreek is sufficient to affect other tissues in terms of insulin sensitivity. SREBP1 is regulated by LXR in the liver, suggesting the possibility that fenugreek containing diosgenin affects LXR activation in hepatocytes. Thus, LXR ligand assay on HepG2 hepatocytes was performed using diosgenin. As expected, diosgenin inhibited LXR activation in a dose-dependent manner in HepG2 cells. This inhibitory effect of diosgenin was sufficient to suppress LXR agonistinduced SREBP1 expression in HepG2 cells. The diosgenin treatment also inhibited SREBP1-dependent expression of the lipogenesis-related genes such as FAS and SCD1 in HepG2 cells. LXR activation increases cellular TG content in HepG2 cells. However, the diosgenin treatment decreased the cellular TG content in the presence of an LXR agonist. These data indicate that diosgenin inhibits LXR activation to suppress the lipogenesis-related gene expression in hepatocytes and the LXR-dependent increase in the hepatic TG content. It is suggested that fenugreek containing diosgenin is useful for the treatment of hyperlipidemia and hepatic steatosis.

36

Takahashi · Goto · Hirai · Uemura · Kawada

Acknowledgements The authors thank Kana Ohyama, Chikako Ando, Noriko Mizoguchi, Michihiro Takada, Aki Teraminami, Tomoya Sakamoto, and Rino Kimura for their support in the preparation of the manuscript, and to Sayoko Shinotoh for her secretarial support. The works described in this review were supported by the Research and Development Program for New Bio-industry Initiation.

References 1 Repa JJ, Liang G, Ou J, Bashmakov Y, Lobaccaro JM, Shimomura I, Shan B, Brown MS, Goldstein JL, Mangelsdorf DJ: Regulation of mouse SREBP-1c by oxysterol receptors, LXRalpha and LXRbeta. Genes Dev 2000;14:2819–2830. 2 Latasa MJ, Moon YS, Kim KH, Sul HS: Nutritional regulation of the fatty acid synthase promoter in vivo: sterol regulatory element binding protein functions through an upstream region containing a sterol regulatory element. Proc Natl Acad Sci USA 2000;97:10619–10624. 3 Joseph SB, Laffitte BA, Patel PH, Watson MA, Matsukuma KE, Walczak R, Collins JL, Osborne TF, Tontonoz P: Direct and indirect mechanisms for regulation of fatty acid synthase gene expression by liver X receptors. J Biol Chem 2002;277:11019–25. 4 Zhou J, Zhai Y, Mu Y, Gong H, Uppal H, Toma D, Ren S, Evans RM, Xie W: A novel pregnane X receptor-mediated and sterol regulatory element-binding protein-independent lipogenic pathway. J Biol Chem 2006;281:15013–15020. 5 Schadinger SE, Bucher NL, Schreiber BM, Farmer SR: PPARgamma2 regulates lipogenesis and lipid accumulation in steatotic hepatocytes. Am J Physiol Endocrinol Metab 2005;288:E1195–E1205. 6 Lopez JM, Bennett MK, Sanchez HB, Rosenfeld JM, Osborne TF: Sterol regulation of acetyl coenzyme A carboxylase: a mechanism for coordinate control of cellular lipid. Proc Natl Acad Sci USA 1993;93:1049– 1053. 7 Talukdar S, Hillgartner FB: The mechanism mediating the activation of acetyl-coenzyme A carboxylase-alpha gene transcription by the liver X receptor agonist T0-901317. J Lipid Res 2006;47:2451–2461. 8 Tabor DE, Kim JB, Spiegelman BM, Edwards PA: Identification of conserved cis-elements and transcription factors required for sterol-regulated transcription of stearoyl-CoA desaturase 1 and 2. J Biol Chem 1999;274:20603–20610. 9 Miller CW, Ntambi JM: Peroxisome proliferators induce mouse liver stearoyl-CoA desaturase 1 gene expression. Proc Natl Acad Sci USA 1996;93:9443– 9448.

Genome Science of Lipid Metabolism and Obesity

10 Ntambi JM: Regulation of stearoyl-CoA desaturase by polyunsaturated fatty acids and cholesterol. J Lipid Res 1999;40:1549–1558. 11 Deng T, Shan S, Li PP, Shen ZF, Lu XP, Cheng J, Ning ZQ: Peroxisome proliferator-activated receptor-gamma transcriptionally up-regulates hormonesensitive lipase via the involvement of specificity protein-1. Endocrinology 2006;147:875–884. 12 Kershaw EE, Schupp M, Guan HP, Gardner NP, Lazar MA, Flier JS: PPARgamma regulates adipose triglyceride lipase in adipocytes in vitro and in vivo. Am J Physiol Endocrinol Metab 2007;293:E1736– E1745. 13 Napal L, Marrero PF, Haro D: An intronic peroxisome proliferator-activated receptor-binding sequence mediates fatty acid induction of the human carnitine palmitoyltransferase 1A. J Mol Biol 2005; 354:751–759. 14 Tugwood JD, Issemann I, Anderson RG, Bundell KR, McPheat WL, Green S: The mouse peroxisome proliferator activated receptor recognizes a response element in the 5⬘ flanking sequence of the rat acyl CoA oxidase gene. EMBO J 1992;11:433–439. 15 Nakamura K, Moore R, Negishi M, Sueyoshi T: Nuclear pregnane X receptor cross-talk with FoxA2 to mediate drug-induced regulation of lipid metabolism in fasting mouse liver. J Biol Chem 2007;282: 9768–9776. 16 Auwerx J, Schoonjans K, Fruchart JC, Staels B: Regulation of triglyceride metabolism by PPARs: fibrates and thiazolidinediones have distinct effects. J Atheroscler Thromb 1996;3:81–89. 17 Zhang Y, Repa JJ, Gauthier K, Mangelsdorf DJ: Regulation of lipoprotein lipase by the oxysterol receptors, LXRalpha and LXRbeta. J Biol Chem 2001; 276:43018–43024. 18 Sato O, Kuriki C, Fukui Y, Motojima K: Dual promoter structure of mouse and human fatty acid translocase/CD36 genes and unique transcriptional activation by peroxisome proliferator-activated receptor alpha and gamma ligands. J Biol Chem 2002;277:15703–15711.

37

19 Sato O, Takanashi N, Motojima K: Third promoter and differential regulation of mouse and human fatty acid translocase/CD36 genes. Mol Cell Biochem 2007;299:37–43. 20 Teboul L, Febbraio M, Gaillard D, Amri EZ, Silverstein R, Grimaldi PA: Structural and functional characterization of the mouse fatty acid translocase promoter: activation during adipose differentiation. Biochem J 2001;360:305–312. 21 Zhou J, Febbraio M, Wada T, Zhai Y, Kuruba R, He J, Lee JH, Khadem S, Ren S, Li S, Silverstein RL, Xie W: Hepatic fatty acid transporter Cd36 is a common target of LXR, PXR, and PPARgamma in promoting steatosis. Gastroenterology 2008;134:556–567. 22 Cha BS, Ciaraldi TP, Carter L, Nikoulina SE, Mudaliar S, Mukherjee R, Paterniti JR Jr, Henry RR: PPARgamma and RXR agonists have complementary effects on glucose and lipid metabolism in human skeletal muscle. Diabetologia 2001;44:444– 452. 23 Murakami K, Ide T, Nakazawa T, Okazaki T, Mochizuki T, Kadowaki T: Fatty-acyl-CoA thioesters inhibit recruitment of steroid receptor co-activator 1 to alpha and gamma isoforms of peroxisome-proliferator-activated receptors by competing with agonists. Biochem J 2001;353:231–238. 24 Takahashi N, Kawada T, Goto T, Yamamoto T, Taimatsu A, Kimura K, Saitoh M, Hosokawa T, Miyashita K, Fushiki T: Dual action of isoprenols to activate both PPARγ and PPARα in 3T3-L1 adipocytes and HepG2 hepatocytes. FEBS Lett 2002;514: 315–322. 25 Goto T, Takahashi N, Kato S, Egawa K, Ebisu S, Moriyama T, Fushiki T, Kawada T: Phytol, a phytochemical compound, activates PPARα and regulates lipid metabolism in PPARα-expressing cells. Biochem Biophys Res Commun 2005;337:440–445.

26 Kuroyanagi K, Kang MS, Goto T, Kusudo T, Hirai S, Yu R, Yano M, Sasaki T, Takahashi N, Kawada T: Citrus auraptene acts as an agonist for PPARs and enhances adiponectin production and MCP-1 reduction in 3T3-L1 adipocytes. Biochem Biophys Res Commun 2008;366:219–225. 27 Kang MS, Hirai S, Goto T, Kuroyanagi K, Ezaki Y, Takahashi N, Kawada T: Dehydroabietic acid, herbal terpenoid, acts as ligands for PPARs in macrophages to regulate inflammation. Biochem Biophys Res Commun 2008;369:333–338. 28 Takahashi N, Kawada T, Goto T, Kim CS, Taimatsu A, Egawa K, Yamamoto T, Jisaka M, Nishimura K, Yokota K, Yu R, Fushiki T: Abietic acid activates PPARγ in RAW264.7 macrophages and 3T3-L1 adipocytes to regulate gene expression involved in inflammation and lipid metabolism. FEBS Lett 2003;550:190–194. 29 Zomer AW, van Der Burg B, Jansen GA, Wanders RJ, Poll-The BT, van Der Saag PT: Pristanic acid and phytanic acid: naturally occurring ligands for the nuclear receptor peroxisome proliferator-activated receptor alpha. J Lipid Res 2000;41:1801–1807. 30 Gloerich J, van den Brink DM, Ruiter JP, van Vlies N, Vaz FM, Wanders RJ, Ferdinandusse S: Metabolism of phytol to phytanic acid in the mouse, and the role of PPARalpha in its regulation. J Lipid Res 2007;48:77–85. 31 Kubota N, Terauchi Y, Miki H, Tamemoto H, Yamauchi T, et al: PPARgamma mediates high-fat diet-induced adipocyte hypertrophy and insulin resistance. Mol Cell 1999;4:597–609.

Professor Teruo Kawada Laboratory of Molecular Functions of Food Division of Food Science and Biotechnology Graduate School of Agriculture Kyoto University, Uj Kyoto 606-8502 (Japan) Tel. +81 75 753 6262, Fax +81 75 753 6264, E-Mail [email protected]

38

Takahashi · Goto · Hirai · Uemura · Kawada

Proteomics Yoshikawa T (ed): Food Factors for Health Promotion. Forum Nutr. Basel, Karger, 2009, vol 61, pp 39–54

Oxidative Stress-Induced Posttranslational Modification of Proteins as a Target of Functional Food Yuji Naito ⭈ Toshikazu Yoshikawa Molecular Gastroenterology and Hepatology, Kyoto Prefectural University of Medicine, Kyoto, Japan

Abstract In lifestyle-related diseases including metabolic syndrome, atherosclerosis, and cancer, oxidative stress is indicated by several markers, among which are lipid peroxides, aldehydes, and nitrotyrosine. We hypothesized that identification of proteins that are posttranslationally modified due to oxidative stress would lead to a greater understanding of some of the potential molecular mechanisms involved in degeneration and inflammation in these disorders. Proteomics is an emerging method for identification of proteins and their modification residues, and its application to food factor science is just beginning. Especially, we can obtain several monoclonal antibodies to detect specifically oxidized proteins, which can be applied to analysis by immunostaining or immunoblot. In this review, we present the use of these monoclonal antibodies in several diseases, from which new insights have emerged into mechanisms of metabolism and inflammation in these disorders that are Copyright © 2009 S. Karger AG, Basel associated with oxidative stress.

It is well known that oxidative stress is involved in the pathogenesis of lifestyle-related diseases, including atherosclerosis, hypertension, diabetes mellitus, ischemic diseases, and malignancies. Oxidative stress, which refers to a state of elevated levels of reactive oxygen species (ROS), occurs form a variety conditions that stimulate either ROS production or a decline in antioxidant defenses. During oxidative stress, the oxidation of cellular components results in the modification of DNA, proteins, lipids, and carbohydrates [1]. In the case of proteins, numerous posttranslational modifications have been characterized as resulting either from direct oxidation of amino residues or through the formation of reactive intermediates by the oxidation of other cellular components. The oxidative stress-induced posttranslational modification (OPTM) of 20 kinds of basic amino acids takes an important role in the manifestation of the function of many proteins. OPTM may be subdivided into two general forms: reversible OPTM and irreversible OPTM (fig. 1). The oxidation of cysteine to sulfenic, sulfinic,

Oxidative stress

Inflammatory cells Drugs Ischemia Hyperglycemia Metals

Detection of OPTM proteins

Functional analysis of OPTM proteins

Association with diseases

Oxidative stress-specific modification HNE, HEL addition Nitration Chlorination Bromonation Necrosis Apoptosis Innate immune reaction Irreversible OPTM Decrease in defense system • Degeneration • Chronic inflammation • Cytokine storm • Carcinogenesis

Oxidative stress NO stress AGEs Target cells Reversible OPTM

Redox regulation Signal modification Mitochondrial dysfunction Abnormal cell proliferation

Oxidation of cystein residues S-thiolation Glutationation Nitrosylation

Fig. 1. Induction of oxidative stress, detection and functional analysis of OPTM proteins, and their association with lifestyle-related diseases. AGEs = Advanced glycation end products.

and sulfonic acids has been shown to occur frequently and sulfenic and sulfinic acids often can be reduced enzymatically. Some of the lipid peroxidation products exhibit a facile reactivity with proteins, generating a variety of intra- and intermolecular covalent adducts. Furthermore, especially in the situation of inflammation, nitration by reactive nitrogen species (RNS), chlorination by hypochlorous acid (HOCl), and bromonation by hypobromous acid of the targeted protein are frequently detected. In this review, we focused on the detection of OPTM proteins in lifestyle-related diseases and their animal models, and introduced several investigations for functional foods by using OPTM proteins as their target.

Oxidative Stress-Induced Posttranslational Modification

As the relationship between oxidative stress and various diseases becomes clear, OPTM proteins are now attracting attention. The transduction of an oxidant signal into a biological response can be mediated in several ways, but one principal mechanism involves the oxidation of protein cysteine residues. The thiol (-SH) moiety on the side chain of the amino acid cysteine is particularly sensitive to redox reactions

40

Naito · Yoshikawa

and is an established redox sensor. As for cysteine residue-specific OPTM of the protein (S-thiolation), it is found to act as a switch regulating biological function like phosphorylation-dephosphorylation [2]. For example, oxidative modification of free reactive thiols (S-thiolation) on the small G protein Ras increases Ras activity and thereby promotes ROS-dependent hypertrophic signaling in cardiac myocytes [3]. There are an increasing number of proteins that are known to be regulated by S-thiolation. Proteins regulated by cysteine oxidation include ion translocators, structural proteins, metabolic enzymes, DNA isomerases, and signaling proteins. Signal transduction proteins that are directly regulated by oxidative modification of cysteine residues include protein phosphatases, protein kinases, G-proteins, and membrane receptors [4]. Recent studies used biotinylated cysteine (biotin-cysteine) as a probe for proteins that are S-thiolated during oxidative stress. Biotin-cysteine rapidly crosses the plasma membrane and can be used to detect, quantify, purify, and identify proteins susceptible to oxidation in cells. Using this method, Ishii et al. [5] have identified 26 proteins including glycolytic enzymes, cytoskeletal proteins, redox enzymes, and stress proteins as substrates for reversible cysteine-targeted oxidation when human neuroblastoma SH-SY5Y cells were exposed to 15-deoxydelta12,15-prostaglandin J2. OPTM of amino acid residues in a peptide may lead to structural changes, ranging from a slight conformational change to a severe denaturation accompanied by fragmentation. It may lead to either activation or inhibition of the protein activities. Mild oxidative stress can induce modification of cysteine such as reversible S-thiolation, disulfide formation, glutathionylation, and S-nitrosylation. Although OPTM of cysteine residue inactivates protein targets, such S-thiolation can also result in a gain of function for certain stress-responsive signaling systems. For example, transcriptional activation of antioxidant-responsive element (ARE)-containing genes is upregulated by S-thiolation. Several genes containing ARE are activated by the activation of Nrf-2-Keap 1 system by oxidative stress. The S-thiolation of Keap 1, the cytoplasmic inhibitor of Nrf2, by ROS or electrophiles results in the dissociation of the Keap 1-Nrf2 complex. Once freed from inhibition by Keap 1, Nrf2 translocates to the nucleus and activates the expression of ARE-containing genes. In addition to these reversible OPTM, it becomes clear that irreversible OPTM for histidine, lysin, and cysteine residues means generation of abnormal protein associated with the etiology of lifestyle related diseases. Moderate oxidation of amino acid side chains can give rise to a number of adducts. Many reactions result in the formation of direct oxidation of arginine and proline residues (semialdehyde formation), sulfoxide and nitrosylation formation, oxo-, hydroxy-, nitro-, and chloroderivatives of amino acids, and finally covalent cross-links within and between protein molecules [6]. Furthermore, lipid peroxidation products can diffuse across cellular membranes, allowing the reactive aldehyde-containing lipids to covalently modify proteins localized throughout the cell and relatively far away from the initial site of ROS formation. The most reactive aldehydes generated from polyunsaturated

Oxidative Stress-Induced Posttranslational Modification of Proteins

41

fatty acid oxidation are 4-hydroxy-2-nonenal (HNE), 4-oxo-2-nonenal, and acrolein. OPTM proteins have been shown to have a wide variety of effects on cells in vitro depending upon the concentration utilized, and as such, interpretation of experimental results must be considered cautiously. Because the side chains of cysteine, histidine, and lysine are often used in catalysis, the most common effect of OPTM is enzyme inactivation. Lipid peroxidation products can inactivate or modify the function of several proteins (Na+-K+-ATPase [7], glucose transporter [8], adipocyte fatty acid-binding protein [9, 10], NADP+-dependent isocitrate dehydrogenase [11], thioredoxin and thioredoxin reductase [12–14], glutathione peroxidase [15], and heat shock protein 90 [16]). A large body of evidence supports OPTM via endogenous and exogenous ROS compounds as an important factor in the induction of autoimmunity [17–19]. OPTM epitope causes the failure of acquired tolerance for autoantigen by what is shown as a new antigen. Recent studies suggest that OPTM protein is a cause of autoimmune disease and inflammatory disorder [19–21]. OPTM proteins could be the targets of B cell-mediated immune responses and induce T cell responses and add the potential of certain aldehydes to induce autoimmunity by breaking the B cell tolerance to nonmodified proteins. It has been shown that the OPTM of self-proteins by lipid peroxidation products indeed results in a break of a tolerance to self-proteins. The presence of autoantibodies against OPTM protein (glutamic acid decarboxylase) in patients with type I diabetes as well as in patients with Stiffman syndrome [22] has also been demonstrated. Recently, Toyoda et al. [19] have reported that the HNEspecific epitopes can be a triggering antigen of anti-DNA response, and the monoclonal antibodies against HNE-specific epitopes can bind to two structurally distinct antigens (i.e. native DNA and 4-oxo-2-nonenal-modified proteins). Their findings provide the evidence to suspect an etiologic role for the lipid peroxidation in autoimmune disease. One of the most important processes for maintaining homeostasis is the ability of proteolytic systems to eliminate OPTM proteins. Cells degrade OPTM proteins via the proteasome pathway and the lysosomal pathway. The ubiquitin-proteasome system is considered the major pathway responsible for the degradation and elimination of mildly oxidized proteins, and is involved in regulating proteins involved in several cellular activities like cell cycles. Interestingly, a proteomic analysis of OPTM proteins shows that the major intracellular target of protein carbonylation is one of the regulatory subunits in 26S proteasome, S6 ATPase [23]. Removal of OPTM proteins is crucial for cell survival. Indeed, if OPTM proteins are not eliminated either through proteasome or lysosomal pathways, they are able to begin to accumulate and potentially aggregate. These aggregated proteins can alter cell functions and lead to necrosis or apoptosis. Furthermore, a recent study clearly showed the possibility that the decrease in autophagy function may cause the autoimmune inflammation in the intestine, because autophagy is a primary mechanism to resolve these OPTM proteins [24].

42

Naito · Yoshikawa

4-Hydroxy-2-Nonenal

A growing body of evidence suggests that many of the effects of cellular dysfunction under oxidative stress are mediated by products of the peroxidative degradation of polyunsaturated fatty acids. When unsaturated fatty acid is exposed to oxidative stress, a lipid peroxidation response progresses like a chain reaction, and various kinds of degradation products by lipid peroxidation responses are generated. Many past studies showed that aldehyde molecules, an end product of lipid peroxidation, are implicated as causative agents in cytotoxic processes initiated by the exposure of biological systems to oxidizing agents. In contrast to free radicals, aldehydes are relatively stable and can diffuse within or even escape from the cell and attack targets far from the site of the original event [25]. HNE (fig. 2), among the reactive aldehydes, is believed to be largely responsible for the cytopathological effects observed during oxidative stress. The most common and reliable approach for detection of HNE adducts is the use of antibodies which recognize HNE bound to amino acid sidechains of proteins. In 1995, Toyokuni et al. [26] raised a monoclonal antibody directed at the HNEmodified protein. This monoclonal antibody, which is now commercially available, has been attested to be specific for the HNE-histidine Michael adduct. The development of specific antibodies against protein-bound HNE has made it possible for us to obtain highly probable evidence for the occurrence of oxidative stress in vivo. Toyokuni et al. [27] demonstrated the presence of HNE-modified proteins in vivo in the iron-nitrolotriacetate (Fe3+-NTA)-induced renal carcinogenesis, which is one of the best characterized in vivo models of oxidative stress. The immunoreactive HNE adducts were markedly generated in the renal proximal tubules with degeneration several hours after administration of Fe3+-NTA. The HNE-modified proteins have also been detected in other animal and human studies, including hyperglycemia injury of pancreatic β-cells [28], ischemia-reperfusion injury [29], carbon tetrachloride-induced liver injury [30], ischemic heart or renal failure [31, 32], inflammatory bowel disease [33], chronic hepatitis type C [34], nonalcoholic steatohepatitis [35], atherosclerotic lesion [36], Alzheimer’s disease [37], Parkinson’s disease [38], amyotrophic lateral sclerosis [39], exercise-induced muscular injury [40]. Collectively, the results of these studies document the association of HNE-derived epitopes with chemical-induced oxidative stress or disease states linked directly or indirectly with chronic inflammation. However, the mechanistic role of HNE-modified proteins in the cellular damage accompanying oxidative stress is not well delineated. To examine the relationship between oxidative stress, HNE-protein modification, and cellular signaling, we treated cultured human mesangial cells with high (25 mM)concentration glucose and identified proteins modified by HNE using immunoblotting [41]. Interestingly, specific targets with estimated molecular masses of 60, 80, 85, and 105 kDa were strongly stained by anti-HNE antibody in mitochondrial fraction of high glucose-treated mesangial cells (fig. 3). In addition, fluorescent intensity of

Oxidative Stress-Induced Posttranslational Modification of Proteins

43

␻-6 PUFAs

R

R’ Peroxidation

Fig. 2. Formation/detection of HNE-modified proteins. HNE is generated during lipid peroxidation of ω–6 polyunsaturated fatty acids (PUFAs). The Michael type addition of HNE to protein is formed in vivo. X represents the side chain of nucleophilic amino acids, such as cysteine, histidine, and lysine. Protein-bound HNE can be detected by immunochemical methods and mass spectrometry using an anti-HNE antibody.

OH O

HNE

Protein

O HNE-specific modification of proteins

OH Anti-HNE Ab

X Protein

RedoxSensor CC-1 was increased in high glucose-exposed mesangial cells and the merged images with MItoTracker Green FM clearly indicated that mitochondria are the major source of ROS production in high glucose-exposed mesangial cells. It is conceivable that these specific targets identified in vitro are involved in the mechanism of HNE-induced cell signaling and would be of interest in identifying the proteins. Moreover, these specific targets may be candidates for biomarkers to evaluate the antioxidative properties of food factors. We have found that astaxanthin, a carotenoid, reduced the increases of these HNE-modified proteins in mitochondrial fraction as well as inhibited the ROS production from the mitochondria in high glucose-exposed mesangial cells. These molecular-based data strongly support an in vivo evidence that the treatment with astaxanthin inhibits diabetic nephropathy via reducing oxidative stress in a mouse model [42].

Nε-(Hexanonyl)lysine

Nε-(hexanonyl)lysine (HEL) has been found in the reaction between linoleic hydroperoxide and lysine moiety. It has been shown that the formation of HEL is a good marker for oxidative modification by oxidized ω–6 fatty acids such as linoleic acid and arachidonic acid [43]. Presence of HEL is reported by the immunostaining by using monoclonal and polyclonal antibodies against HEL. Kato et al. [44] evaluated muscular oxidation injury by excessive exercise using an anti-HEL antibody and reported that the functional food factor flavonoid is useful in reduction of this oxidation injury by showing that this compound clearly reduced the HEL-positivity in

44

Naito · Yoshikawa

WB: MnSOD (25 kDa)

25 kDa

WB: HNE (4-hydroxy-2-nonenal)

105 kDa

75 kDa

D-Glucose ASX (10–6 M)

5 –

25 5 25 – – +

Mitochondria

5 25 5 25 + – + + Cytosol

Fig. 3. Astaxanthin inhibited high glucose-induced production of HNE-modified proteins in mitochondria of normal human mesangial cells. Representative Western blotting image out of three independent experiments is shown. Anti-HNE antibody was used to detect protein adducts in mitochondrial and cytosolic fractions of high glucose (25 mM)-treated normal human mesangial cells. Anticytochrome C oxidase complex IV subunit antibody was used as a mitochondrial protein marker. Arrows indicate bands stained with anti-HNE antibody. M = Marker. Data are from Manabe et al. [41].

muscular tissues. Fukuchi et al. [45] have reported that HEL is detected in foam cellrich areas in atherosclerotic specimens obtained from New Zealand White rabbits, where they are primarily colocalized with C-reactive protein (CRP)-positive cells, suggesting that the generation of oxidative stress marker may be mediated by CRP in atherosclerotic lesions. Osakabe et al. [46] have demonstrated that rosmarinic acid, a major polyphenolic component of Perilla frutescens, reduces lipopolysaccharideinduced liver injury in d-galactosamine-sensitized mice, and also demonstrates that its cytoprotective action may be derived from antioxidative action of the ingredient by showing the decrease in HEL production in liver as a biomarker of lipid peroxidation. The antioxidative action of cacao liquor proanthocyanidin polyphenols has been also shown by the in vivo investigation using several biomarkers including nitrotyrosine, HEL, and HNE adducts in mice lung [47]. We have recently demonstrated that astaxanthin (1) increased the utilization of lipids as an energy substrate during exercise, and (2) improved muscle lipid metabolism in exercise via inhibitory effect of oxidative carnitine parmitoyltransferase I (CPT I) modification by HEL [48] (fig. 4). A rate-limiting step of lipid metabolism in myocytes is the entry of long-chain fatty acids into mitochondria. CPT I located on the mitochondrial membrane plays an important role in the entry of fatty acid. Recent studies have shown that FAT/CD36 is associated with CPT I on the mitochondrial membrane and elevates its function. Our study has shown that an increase in the interaction between CPT I and FAT/CD36 in muscle during exercise is facilitated by

Oxidative Stress-Induced Posttranslational Modification of Proteins

45

IB:HEL

IB:FAT/CD36 *

2.5

*

*

2.0

Normalized ratio

Normalized ratio

*

2.5

*

1.5 1.0 0.5

*

2.0 1.5 1.0 0.5 0

0

CONT

a

AST

Sedentary

CONT

CONT

AST

Running

*

b

AST

Sedentary

CONT

AST

Runing

Fig. 4. Amount of fatty acid translocase (FAT/CD36) that coimmunoprecipitated with carnitine palmitoyltransferase I (CPT I; a) and HEL-modified CPT I (b) in skeletal muscle. A single session of exercise was performed at 30 m/min for 30 min on the final day of the experiment. Lysate protein from the muscle collected immediately after running was immunoprecipitated with CPT I antibody. Immunoprecipitates were separated by SDS-PAGE and membranes probed for FAT/CD36 (a) or HEL (b). Values are mean ± SE obtained from 6 mice. * p < 0.05. Data are from Aoi et al. [40].

astaxanthin, which would be one of mechanisms involved in the promotion of lipid metabolism. Exercise-induced oxidative stress is mainly derived from mitochondrial production of ROS associated with ATP generation and thus CPT I located on the mitochondria membrane is easily exposed to oxidative stress. We found that astaxanthin limits the modification of CPT I by HEL during exercise. Modification of CPT I by HEL may alter the colocalization of CPT I with FAT/CD36 by changing the CPT I molecule. These data indicate that an increase in fatty acyl-CoA uptake into the mitochondria via CPT I during exercise may be involved in the promotion of lipid metabolism by antioxidant activity of astaxanthin. Moreover, our data show the possibility that the HEL modification of CPT I may be a good biomarker for the evaluation of antioxidative properties of food factors in vivo.

Neutrophil-Dependent Oxidative Stress

In activated neutrophils, NADPH oxidase in cell membranes becomes activated, and an electron transfer takes place from NADPH in cells to oxygen inside and outside cells, and the oxygen that received electrons becomes superoxide radicals (O2–˙), which is rapidly converted to hydrogen peroxides (H O ) by spontaneous dismutation or enzymatic SOD, and hydroxyl radicals (˙OH), which are formed nonenzymatically in the presence of Fe2+ as a secondary reaction [1, 49]. On the other hand, the toxicity of H O is enhanced by the activity of myeloperoxidase (MPO). MPO is abundant in primary azurophilic granules of leukocytes including neutrophils and monocytes/ 2

2

46

2

2

Naito · Yoshikawa

macrophages, and secreted into the phagolysosomal compartment following phagocyte activation. In combination with H2O2, MPO can oxidize the halides and the pseudohalide thiocyanate (SCN–) to their corresponding hypohalous acids. H2O2 + HX → HOX + H2O (X = Cl–, Br–, I–, or SCN–)

Owing to its high concentration in biological fluids, Cl– is the major substrate for MPO; consequently, HOCl is a major oxidation product. It has several known targets. One of these targets is the molecule ‘tyrosine’ in both its free and bound forms. Upon reaction with HOCl, tyrosine is converted into the 3-chlorotyrosine molecule (fig. 5). This molecule could also cause tissue damage or dysfunction in many inflammatory conditions. In addition, 3-chlorotyrosine is unique because it is heat stable and is not readily formed by an artificial mechanism, which makes it an excellent marker, a ‘molecular fingerprint’, for MPO-induced oxidation [50]. Recent in vivo studies indicate that assessment of 3-chlorotyrosine protein adduct formation by immunohistochemistry could be a useful marker of neutrophil-induced cell injury in druginduced liver injury [51], inflammatory neuronal degeneration [52], atherosclerosis [53, 54], and cystic fibrosis [54]. If we want to evaluate the anti-inflammatory effect, especially in neutrophil-associated inflammation, these chlorotyrosine molecules may be a good marker for the function of food factors. Recently, Kunitomo et al. [55, 56] have demonstrated that the administration of coenzyme Q10 or corosolic acid, a constituent of banaba leaves, significantly attenuates the increase in 3-chlorotyrosine in serum, reduces the elevated serum insulin levels and elevated blood pressure, and finally improves endothelial dysfunction in the mesenteric arteries in a rat model of metabolic syndrome.

Nitrosative Protein Tyrosine Modification

Nitration and nitrosylation of tyrosine residues are mediated by RNS produced during inflammation, aging, and oxidative stress. Increases in RNS production result from excess or deregulating nitric oxide (NO) reacting with reactive oxygen species (ROS). NO, as a signal molecule, is generated during inflammation by neutrophils and phagocytes, and it reacts with superoxide to generate RNS, including peroxynitrite (ONOO–) and NO itself, which in turn reacts with tyrosyl radicals to add –NO2 or –NO to tyrosine residues, forming 3-nitrotyrosine or 3-nitrosotyrosine residues, respectively. As one of the most important mediators of MPO, nitrotyrosine plays a key role in the process of oxidation seen early in inflammatory conditions including ulcerative colitis and rheumatoid arthritis. One important aspect about the formation of nitrotyrosine that must be noted is that it is not solely generated by peroxynitrite (ONOO), a product of the interaction between NO and O2˙–. MPO can also independently

Oxidative Stress-Induced Posttranslational Modification of Proteins

47

HOOC

OH

NH2

OH Cl

CH2

Cl

Cl

OH CH2 HO

H2N

CH2

COOH

H2N

COOH

CH2 H2N

COOH

HOCl

OH

ROS MPO NO

CH2 H2N ONOO

COOH

HOBr

–

OH

OH

OH

NO2

Br

CH2

CH2 H2N

Br

Br

COOH

H2N

COOH

CH2 H2N

COOH

Fig. 5. Neutrophil-dependent modification of tyrosine residues of proteins.

oxidize nitrite (NO2–), a stable end product of NO metabolism, to form nitrogen dioxide (NO2). NO is also a RNS that in turn can nitrogenate tyrosine (fig. 6). A third pathway utilizes MPO-generated HOCl, which also oxidizes NO2– to nitryl chloride (NO2Cl), which is also a RNS. Several recent investigations validate the importance of nitrotyrosine (table 1). For example, protein immunoprecipitation and Western blotting revealed MnSOD as a target of 3-nitrotyrosine formation in traumatic brain injury [57], adriamycininduced neurotoxicity [58], and Alzheimer’s disease [59]. During inflammatory response, ONOO– reacts with the metal center of MnSOD to generate reactive free radicals. The production of such nitrating agents, which is close to Tyr34 in the active site of the enzyme, facilitates site-specific nitration [60, 61]. In addition to the transitional metal centers, protein tyrosine nitration is affected by several conditions: the proximity to the site of nitrating agent generation, the acidity, the near presence of amino acids (tryptophan, cysteine, methionine) that compete for nitrating agents, and the location of tyrosine residue [62]. Heme oxygenase 1 (HO-1) protein may be one of target of ONOO–. The HO system is the rate-limiting enzymatic step that catalyzes the breakdown of heme into equimolar amounts of biliverdin, an antioxidant rapidly converted to bilirubin, and carbon monoxide, an antiapoptotic vasodilator, with the release of its iron moiety. Kruger et al. [63] have demonstrated that ONOO– generates 2

48

Naito · Yoshikawa

Classical pathway NO˙ ONOO– NO2˙

O2˙–

OH NO2

MPO-dependent pathway H2O2

MPO

NO2– NO2Cl

HOCl

H2N

CH2 COOH 3-Nitrotyrosine

MPO NO˙

NO2–

NO2˙

Fig. 6. Formation of nitrotyrosine by the classical and MPO-dependent pathway.

3-nitrotyrosine within an HO-1 and HO-1 immunoprecipitate and decreases HO activity, supporting their hypothesis of HO-1 inactivation by ONOO–. Serum levels of nitrotyrosine may be a good stable marker for NO stress in vivo. In a clinical field, Musso et al. [64] have compared the serum nitrotyrosine levels and many clinical parameters in 64 nonobese nondiabetic patients with nonalcoholic fatty liver disease (NAFLD) and 74 control subjects. Persons with NAFLD had greater systemic nitrosative stress than did control subjects, but the two groups did not differ significantly in any other features. Nitrotyrosine and adiponectin concentrations and vitamin A intakes independently predicted alanine aminotransferase concentrations in NAFLD patients and liver histology in a subgroup of 29 subjects with biopsy-proven nonalcoholic steatohepatitis. Devaraj et al. [65] have demonstrated that γ-tocopherol supplementation alone and in combination with α-tocopherol alters biomarkers of oxidative stress, especially nitrotyrosine in serum, and inflammation in subjects with metabolic syndrome. These data may indicate the need to test γ-tocopherol in prospective clinical trials to elucidate its utility in cardiovascular disease prevention, although prospective clinical trials with α-tocopherol have not yielded positive results.

Conclusion

Identification of OPTM proteins in lifestyle diseases allows one to determine which proteins are more affected by oxidation in these diseases and, consequently, more prone to inactivation, and thus represents a significant step in linking well-established inflammation, degeneration, or carcinogenesis with oxidative events at a protein level.

Oxidative Stress-Induced Posttranslational Modification of Proteins

49

Table 1. Target proteins of 3-nitrotyrosine formation Cell/tissue

Target proteins

Tyrosine number

Function

First author

Cancer cells/ tissue

tissue inhibitor of MMP-4

Y114, Y195, Y188, Y190

enzyme inactivation

Donnini, 2008

Muscle

carbonic anhydrase III

aging

Chen, 2008

Retina

TrkA receptor

impairment of NGF signaling

Ali, 2008

Brain

MnSOD

decrease in mitochondrial respiration

Tangpong, 2008; Bayir, 2007; Anantharaman, 2006; Pittman, 2002

Brain

ferric-human neuroglobin

scavenge/ neuroprotection

Nicolis, 2007

Muscle

desmin

neurodegeneration

Janue, 2007

Mesothelioma

HIF-1α, p53

enzyme inactivation Thomas, 2006

Heart/tissue

human myoglobin

Y103, Y146

scavenge/little effect on enzyme

Macrophage

Indoleamine 2,3-dioxygenase

Y15 (Y345,Y353)

enzyme inactivation Fujigaki, 2006

Aorta

heme oxygenase-1

PC12 cell

α-tubulin

Y161, Y357

neuronal differentiation

Recombinant protein

cytochrome P450 2B1

Y190

enzyme inactivation Lin, 2005, 2003

Muscle

creatine kinase

Y82, (Y14, Y20)

physiological?

Kanski, 2005

Muscle

α-enolase

Y43

interference of phosphorylation

Casoni, 2005

Cardiac myocyte

α-actinin

contractile dysfunction

Borbely, 2005

Atheroma

apolipoprotein A-I

proatherogenic HDL Zheng, 2004

Lung

MnSOD

enzyme inactivation Gray, 2004

Macrophage

iron regulatory protein-1

self-protecting

Gonzalez, 2004

Heart

carnitine palmitoyl transferase I

endotoxin toxicity

Fukumoto, 2004

50

Nicolis, 2006

enzyme inactivation Kruger, 2006 Tedeschi, 2005

Naito · Yoshikawa

Table 1. Continued Cell/tissue

Target proteins

Hippocampus

Tyrosine number

Function

First author

synaptophysin

impairment of ACh release

Tran, 2003

Blood

hemoglobin

major target of peroxynitrite in blood

Pietraforte, 2003

Brain

Na,K-ATPase

undefined marker

Golden, 2003

PC12 cell

α-tubulin

NGF-induced differentiation

Cappelletti, 2003

Isolated protein

glutathione S-transferases

enzyme inactivation Wong, 2001

Heart

myofibrillar creatine kinase

heart failure

Mihm, 2001

Glioma cell

p53

loss of p53 DNA binding ability

Cobbs, 2001

Endothelial cell

extracellular matrix

endothelial transcytosis of MPO

Baldus, 2001

Blood

oxyhemoglobin

physiological scavenger

Minetti, 2000

Pancreatic cancer

c-Src kinase

enhanced tyrosine kinase signaling

MacMillan-Crow, 2000

Mitochondria

cytochrome c

changes in redox properties

Cassina, 2000

Yeast

GAPDH

enhancement of chaperone expression and ubiquitination

Buchczyk, 2000

Neuroblastoma

p130cas

toxicity target

Saeki, 1999

Y67

NGF = Nerve growth factor; GAPDH = glyceraldehyde-3-phosphate dehydrogenase; HDL = high-density lipoprotein; MnSOD = manganese superoxide dismutase; MMP = matrix metalloproteinase; HIF-1α = hypoxiainducible factor 1α.

Oxidative Stress-Induced Posttranslational Modification of Proteins

51

References 1 Yoshikawa T, Naito Y: What is oxidative stress? JMAJ 2002;45:271–276. 2 Rhee SG: Cell signaling. H2O2, a necessary evil for cell signaling. Science 2006;312:1882–1883. 3 Kuster GM, Siwik DA, Pimentel DR, et al: Role of reversible, thioredoxin-sensitive oxidative protein modifications in cardiac myocytes. Antioxid Redox Signal 2006;8:2153–2159. 4 Eaton P, Byers HL, Leeds N, et al: Detection, quantitation, purification, and identification of cardiac proteins S-thiolated during ischemia and reperfusion. J Biol Chem 2002;277:9806–9811. 5 Ishii T, Uchida K: Induction of reversible cysteinetargeted protein oxidation by an endogenous electrophile 15-deoxy-delta12,14-prostaglandin J2. Chem Res Toxicol 2004;17:1313–1322. 6 Cloos PA, Christgau S: Post-translational modifications of proteins: implications for aging, antigen recognition, and autoimmunity. Biogerontology 2004;5:139–158. 7 Miyake H, Kadoya A, Ohyashiki T: Increase in molecular rigidity of the protein conformation of brain Na+-K+-ATPase by modification with 4hydroxy-2-nonenal. Biol Pharm Bull 2003;26: 1652–1656. 8 Mark RJ, Pang Z, Geddes JW, et al: Amyloid betapeptide impairs glucose transport in hippocampal and cortical neurons: involvement of membrane lipid peroxidation. J Neurosci 1997;17:1046–1054. 9 Bennaars-Eiden A, Higgins L, Hertzel AV, et al: Covalent modification of epithelial fatty acid-binding protein by 4-hydroxynonenal in vitro and in vivo. Evidence for a role in antioxidant biology. J Biol Chem 2002;277:50693–50702. 10 Grimsrud PA, Picklo MJ Sr, Griffin TJ, et al: Carbonylation of adipose proteins in obesity and insulin resistance: identification of adipocyte fatty acid-binding protein as a cellular target of 4hydroxynonenal. Mol Cell Proteomics 2007;6:624– 637. 11 Yang JH, Yang ES, Park JW: Inactivation of NADP+dependent isocitrate dehydrogenase by lipid peroxidation products. Free Radic Res 2004;38:241–249. 12 Cassidy PB, Edes K, Nelson CC, et al: Thioredoxin reductase is required for the inactivation of tumor suppressor p53 and for apoptosis induced by endogenous electrophiles. Carcinogenesis 2006;27:2538– 2549. 13 Fang J, Holmgren A: Inhibition of thioredoxin and thioredoxin reductase by 4-hydroxy-2-nonenal in vitro and in vivo. J Am Chem Soc 2006;128:1879– 1885.

52

14 Park YS, Misonou Y, Fujiwara N, et al: Induction of thioredoxin reductase as an adaptive response to acrolein in human umbilical vein endothelial cells. Biochem Biophys Res Commun 2005;327:1058– 1065. 15 Park YS, Koh YH, Takahashi M, et al: Identification of the binding site of methylglyoxal on glutathione peroxidase: methylglyoxal inhibits glutathione peroxidase activity via binding to glutathione binding sites Arg 184 and 185. Free Radic Res 2003;37:205– 211. 16 Carbone DL, Doorn JA, Kiebler Z, et al: Modification of heat shock protein 90 by 4-hydroxynonenal in a rat model of chronic alcoholic liver disease. J Pharmacol Exp Ther 2005;315:8–15. 17 Ahsan H, Ali A, Ali R: Oxygen free radicals and systemic autoimmunity. Clin Exp Immunol 2003;131: 398–404. 18 Kovacic P, Jacintho JD: Systemic lupus erythematosus and other autoimmune diseases from endogenous and exogenous agents: unifying theme of oxidative stress. Mini Rev Med Chem 2003;3:568– 575. 19 Toyoda K, Nagae R, Akagawa M, et al: Proteinbound 4-hydroxy-2-nonenal: an endogenous triggering antigen of antI-DNA response. J Biol Chem 2007;282:25769–25778. 20 Chang MK, Binder CJ, Miller YI, et al: Apoptotic cells with oxidation-specific epitopes are immunogenic and proinflammatory. J Exp Med 2004;200: 1359–1370. 21 Chou MY, Hartvigsen K, Hansen LF, et al: Oxidation-specific epitopes are important targets of innate immunity. J Intern Med 2008;263:479–488. 22 Trigwell SM, Radford PM, Page SR et al: Islet glutamic acid decarboxylase modified by reactive oxygen species is recognized by antibodies from patients with type 1 diabetes mellitus. Clin Exp Immunol 2001;126:242–249. 23 Ishii T, Sakurai T, Usami H, et al: Oxidative modification of proteasome: identification of an oxidationsensitive subunit in 26 S proteasome. Biochemistry 2005;44:13893–13901. 24 Saitoh T, Fujita N, Jang MH et al: Loss of the autophagy protein Atg16L1 enhances endotoxininduced IL-1beta production. Nature 2008;456:264– 268. 25 Uchida K: Protein-bound 4-hydroxy-2-nonenal as a marker of oxidative stres. J Clin Biochem Nutr 2005;36:1–10. 26 Toyokuni S, Miyake N, Hiai H, et al: The monoclonal antibody specific for the 4-hydroxy-2-nonenal histidine adduct. FEBS Lett 1995;359:189–191.

Naito · Yoshikawa

27 Toyokuni S, Uchida K, Okamoto K, et al: Formation of 4-hydroxy-2-nonenal-modified proteins in the renal proximal tubules of rats treated with a renal carcinogen, ferric nitrilotriacetate. Proc Natl Acad Sci USA 1994;91:2616–2620. 28 Ihara Y, Toyokuni S, Uchida K, et al: Hyperglycemia causes oxidative stress in pancreatic beta-cells of GK rats, a model of type 2 diabetes. Diabetes 1999;48:927–932. 29 Yamagami K, Yamamoto Y, Kume M, et al: Formation of 8-hydroxy-2⬘-deoxyguanosine and 4-hydroxy-2-nonenal-modified proteins in rat liver after ischemia-reperfusion: distinct localization of the two oxidatively modified products. Antioxid Redox Signal 2000;2:127–136. 30 Hartley DP, Kroll DJ, Petersen DR: Prooxidantinitiated lipid peroxidation in isolated rat hepatocytes: detection of 4-hydroxynonenal- and malondialdehyde-protein adducts. Chem Res Toxicol 1997;10:895–905. 31 Blasig IE, Grune T, Schonheit K, et al: 4-Hydroxynonenal, a novel indicator of lipid peroxidation for reperfusion injury of the myocardium. Am J Physiol 1995;269:H14–H22. 32 Eschwege P, Paradis V, Conti M, et al: In situ detection of lipid peroxidation by-products as markers of renal ischemia injuries in rat kidneys. J Urol 1999; 162:553–557. 33 Nair J, Gansauge F, Beger H, et al: Increased ethenoDNA adducts in affected tissues of patients suffering from Crohn’s disease, ulcerative colitis, and chronic pancreatitis. Antioxid Redox Signal 2006;8: 1003–1010. 34 Kageyama F, Kobayashi Y, Kawasaki T, et al: Successful interferon therapy reverses enhanced hepatic iron accumulation and lipid peroxidation in chronic hepatitis C. Am J Gastroenterol 2000;95: 1041–1050. 35 Serviddio G, Bellanti F, Tamborra R, et al: Uncoupling protein-2 (UCP2) induces mitochondrial proton leak and increases susceptibility of non-alcoholic steatohepatitis (NASH) liver to ischaemia-reperfusion injury. Gut 2008;57:957– 965. 36 Uchida K, Toyokuni S, Nishikawa K, et al: Michael addition-type 4-hydroxy-2-nonenal adducts in modified low-density lipoproteins: markers for atherosclerosis. Biochemistry 1994;33:12487–12494. 37 Reed T, Perluigi M, Sultana R, et al: Redox proteomic identification of 4-hydroxy-2-nonenal-modified brain proteins in amnestic mild cognitive impairment: insight into the role of lipid peroxidation in the progression and pathogenesis of Alzheimer’s disease. Neurobiol Dis 2008;30:107–120.

38 Selley ML: (E)-4-hydroxy-2-nonenal may be involved in the pathogenesis of Parkinson’s disease. Free Radic Biol Med 1998;25:169–174. 39 Perluigi M, Fai Poon H, Hensley K, et al: Proteomic analysis of 4-hydroxy-2-nonenal-modified proteins in G93A-SOD1 transgenic mice–a model of familial amyotrophic lateral sclerosis. Free Radic Biol Med 2005;38:960–968. 40 Aoi W, Naito Y, Sakuma K, et al: Astaxanthin limits exercise-induced skeletal and cardiac muscle damage in mice. Antioxid Redox Signal 2003;5:139– 144. 41 Manabe E, Handa O, Naito Y, et al: Astaxanthin protects mesangial cells from hyperglycemiainduced oxidative signaling. J Cell Biochem 2008; 103:1925–1937. 42 Naito Y, Uchiyama K, Aoi W, et al: Prevention of diabetic nephropathy by treatment with astaxanthin in diabetic db/db mice. Biofactors (Oxford) 2004; 20:49–59. 43 Kato Y, Mori Y, Makino Y, et al: Formation of Nepsilon-(hexanonyl)lysine in protein exposed to lipid hydroperoxide. A plausible marker for lipid hydroperoxide-derived protein modification. J Biol Chem 1999;274:20406–20414. 44 Kato Y, Miyake Y, Yamamoto K, et al: Preparation of a monoclonal antibody to N(epsilon)-(Hexanonyl) lysine: application to the evaluation of protective effects of flavonoid supplementation against exercise-induced oxidative stress in rat skeletal muscle. Biochem Biophys Res Commun 2000;274: 389–393. 45 Fukuchi Y, Miura Y, Nabeno Y, et al: Immunohistochemical detection of oxidative stress biomarkers, dityrosine and N(epsilon)-(hexanoyl)lysine, and C-reactive protein in rabbit atherosclerotic lesions. J Atheroscler Thromb 2008;15:185–192. 46 Osakabe N, Yasuda A, Natsume M, et al: Rosmarinic acid, a major polyphenolic component of Perilla frutescens, reduces lipopolysaccharide (LPS)induced liver injury in d-galactosamine (d-GalN)sensitized mice. Free Radic Biol Med 2002;33: 798–806. 47 Yasuda A, Takano H, Osakabe N, et al: Cacao liquor proanthocyanidins inhibit lung injury induced by diesel exhaust particles. Int J Immunopathol Pharmacol 2008;21:279–288. 48 Aoi W, Naito Y, Takanami Y, et al: Astaxanthin improves muscle lipid metabolism in exercise via inhibitory effect of oxidative CPT I modification. Biochem Biophys Res Commun 2008;366:892–897. 49 Naito Y, Takano H, Yoshikawa T: Oxidative stressrelated molecules as a therapeutic target for inflammatory and allergic diseases. Curr Drug Targets 2005;4:511–515.

Oxidative Stress-Induced Posttranslational Modification of Proteins

53

50 Winterbourn CC, Kettle AJ: Biomarkers of myeloperoxidase-derived hypochlorous acid. Free Radic Biol Med 2000;29:403–409. 51 Gujral JS, Hinson JA, Jaeschke H: Chlorotyrosine protein adducts are reliable biomarkers of neutrophil-induced cytotoxicity in vivo. Comp Hepatol 2004;3(suppl 1):S48. 52 Ryu JK, Tran KC, McLarnon JG: Depletion of neutrophils reduces neuronal degeneration and inflammatory responses induced by quinolinic acid in vivo. Glia 2007;55:439–451. 53 Heinecke JW: The HDL proteome: a marker – and perhaps mediator – of coronary artery disease. J Lipid Res 2008; Epub ahead of print. 54 Shao B, Oda MN, Bergt C, et al: Myeloperoxidase impairs ABCA1-dependent cholesterol efflux through methionine oxidation and site-specific tyrosine chlorination of apolipoprotein A-I. J Biol Chem 2006;281:9001–9004. 55 Kunitomo M, Yamaguchi Y, Kagota S, et al: Beneficial effect of coenzyme Q10 on increased oxidative and nitrative stress and inflammation and individual metabolic components developing in a rat model of metabolic syndrome. J Pharmacol Sci 2008;107:128–137. 56 Yamaguchi Y, Yamada K, Yoshikawa N, et al: Corosolic acid prevents oxidative stress, inflammation and hypertension in SHR/NDmcr-cp rats, a model of metabolic syndrome. Life Sci 2006;79: 2474–2479. 57 Bayir H, Kagan VE, Clark RS, et al: Neuronal NOSmediated nitration and inactivation of manganese superoxide dismutase in brain after experimental and human brain injury. J Neurochem 2007;101:168– 181. 58 Tangpong J, Cole MP, Sultana R, et al: Adriamycinmediated nitration of manganese superoxide dismutase in the central nervous system: insight into the mechanism of chemobrain. J Neurochem 2007; 100:191–201.

59 Anantharaman M, Tangpong J, Keller JN, et al: Beta-amyloid mediated nitration of manganese superoxide dismutase: implication for oxidative stress in a APPNLH/NLH X PS-1P264L/P264L double knock-in mouse model of Alzheimer’s disease. Am J Pathol 2006;168:1608–1618. 60 MacMillan-Crow LA, Crow JP, Thompson JA: Peroxynitrite-mediated inactivation of manganese superoxide dismutase involves nitration and oxidation of critical tyrosine residues. Biochemistry 1998; 37:1613–1622. 61 Yamakura F, Taka H, Fujimura T, et al: Inactivation of human manganese-superoxide dismutase by peroxynitrite is caused by exclusive nitration of tyrosine 34 to 3-nitrotyrosine. J Biol Chem 1998;273:14085– 14089. 62 Yeo WS, Lee SJ, Lee JR, et al: Nitrosative protein tyrosine modifications: biochemistry and functional significance. BMB Rep 2008;41:194–203. 63 Kruger AL, Peterson SJ, Schwartzman ML, et al: Up-regulation of heme oxygenase provides vascular protection in an animal model of diabetes through its antioxidant and antiapoptotic effects. J Pharmacol Exp Ther 2006;319:1144–1152. 64 Musso G, Gambino R, De Michieli F, et al: Nitrosative stress predicts the presence and severity of nonalcoholic fatty liver at different stages of the development of insulin resistance and metabolic syndrome: possible role of vitamin A intake. Am J Clin Nutr 2007;86:661–671. 65 Devaraj S, Leonard S, Traber MG, et al: Gammatocopherol supplementation alone and in combination with alpha-tocopherol alters biomarkers of oxidative stress and inflammation in subjects with metabolic syndrome. Free Radic Biol Med 2008; 44: 1203–1208.

Professor Yuji Naito Molecular Gastroenterology and Hepatology, Kyoto Prefectural University of Medicine Kamigyo-ku Kyoto 602-8566 (Japan) Tel. +81 75 251 5519, Fax +81 75 251 0710, E-Mail [email protected]

54

Naito · Yoshikawa

Bioavailability and Safety Yoshikawa T (ed): Food Factors for Health Promotion. Forum Nutr. Basel, Karger, 2009, vol 61, pp 55–63

Absorption and Function of Dietary Carotenoids Akihiko Nagao National Food Research Institute, National Agriculture and Food Research Organization, Tsukuba, Japan

Abstract Carotenoids are highly hydrophobic pigments with yellow to red color and their major dietary sources are fruits and vegetables. They have an essential physiological function as a vitamin A precursor and also have antioxidant, anticancer, immune enhancement and antiobesity activities related to prevention of degenerative diseases. The release of carotenoids from food matrix, their dispersion within the digestive tract, and their solubilization in mixed micelles are important steps for carotenoid bioaccessibility. Solubilized carotenoids are taken up by epithelial cells of the small intestine by simple diffusion and/or transporter-mediated processes and then secreted to lymph as chylomicron. Carotenoids accumulated in tissues are thought to be metabolized to small molecules by enzymatic cleavage and/or chemical oxidation with active oxygen species at conjugated double bonds. The hydroxyl group of xanthophylls can be oxidatively metabolized to carbonyl group. Carotenoids with long chain of conjugated double bonds physically quench singlet oxygen and scavenge oxygen radicals, particularly under low oxygen pressure, and thereby they have been thought to work as lipophilic antioxidants for human health. In addition to antioxidant activities, each carotenoid has characteristic functions such as cell cycle inhibition, induction of cell differentiation and apoptosis, and enhancement of gap-junctional communication. However, the detailed mechanisms of these biological actions have not been fully revealed yet and deserve future studies. Copyright © 2009 S. Karger AG, Basel

Carotenoid usually consists of eight units of isoprene and has a symmetrical skeleton of 40 carbon atoms with a long chain of conjugated carbon double bonds (fig. 1). The conjugated double bonds endow carotenoid molecules with a characteristic yellow to red color and antioxidant activities. They are synthesized by plants and microorganisms, and can be found at high concentration in photosynthetic organelle. Animals accumulate carotenoids in their tissues through dietary intake of carotenoids. Carotenoids play essential physiological roles in harvesting light energy and preventing oxidative stress in photosynthetic tissues. Metabolites of carotenoids such as abscisic acid and trisporic acid work as hormones in plants and fungi. In vertebrates,

␤-Carotene

␣-Carotene

Lycopene

HO ␤-Cryptoxanthin OH

HO Lutein

Fig. 1. Major carotenoids present in human plasma.

some carotenoids are the precursors of vitamin A, which is involved in cell differentiation, growth and vision. In addition to these essential physiological roles, carotenoids have been thought to have beneficial effects on human health due to their biological actions [1]. Many epidemiological studies have shown that intake of carotenoid-rich fruits and vegetables or carotenoid level in serum is negatively correlated with incidence of degenerative diseases. Animal studies as well as in vitro studies indicate that carotenoids have antioxidant, anticancer, immune-enhancing activities. In this chapter, recent studies on intestinal absorption of dietary carotenoids, their metabolism and functions are described.

Bioaccessibility and Intestinal Absorption

Bioavailability of dietary carotenoids is much lower than that of triacylglycerols. Solubilization in gastrointestinal tract and absorption by epithelial cells of small

56

Nagao

Fruits and vegetables Release from food matrix

Emulsion Fats and oils

Epithelial cells Lipolytic enzymes

Phosphatidylcholine Bile acids

Mixed micelles

Chylomicron

Solubilization Intestinal absorption

Fig. 2. Absorption of dietary carotenoids.

intestine largely affect bioavailability of carotenoids. Dietary carotenoids are solubilized as follows: release from food matrix, dispersion in gastrointesinal tract, and subsequent solubilization in mixed micelles. Solubilized carotenoids are taken up by the epithelial cells and secreted to lymph after incorporation into the chylomicron (fig. 2) [2]. Mechanical destruction of food matrix and heating during cooking and processing enhance the release of carotenoids from foods. In particular, carotenoids in vegetables are hard to release because they have rigid cell walls. For example, ingestion of tomato paste showed 2.5-fold higher lycopene concentration in human plasma than that of fresh tomato [3]. After carotenoids are released from food matrices, they interact with digestive fluids in gastrointestinal tract. As carotenoids are insoluble in aqueous medium due to their high hydrophobicity, their dispersion in digestive tract is attained by emulsification with dietary lipids and bile containing phospholipids and bile acids. After hydrolysis of dietary and biliary lipids, carotenoids are solubilized in mixed micelles, which consist of bile acid, fatty acid, monoacylglycerol, cholesterol and phospholipid. The mixed micelle is a disc-like particle, with a diameter of 40–200 nm. Carotenoids solubilized in the micelles are thought to be accessible to epithelial cells of the small intestine. Thus, the ratio of the amount of carotenoids in the mixed micelles to that present in food ingested represents the bioaccessibility.

Absorption of Carotenoids

57

Dietary fats and oils are involved in the carotenoid solubilization processes as described above. Firstly, they facilitate dispersion of carotenoids during cooking and gastrointestinal digestion, although solubility of carotenoids in fats and oils is limited: 0.112–0.141% in case of β-carotene [4]. Solubilization of spinach β-carotene (ca. 4 mg/100 g fresh weight) needs at least 2.8–3.6 g oil/100 g fresh weight. Secondarily, hydrolysis products of fats and oils serve as components of the mixed micelles. Moreover, intake of fats and oils induces the secretion of bile and pancreatic lipases, which are essential for the formation of the mixed micelles. Thus, dietary fats and oils enhance the bioaccessibility of dietary carotenoids. After solubilization in the mixed micelles, carotenoids become accessible to the epithelial cells of the jejunum. Carotenoid transfer from the mixed micelles to the epithelial cells has been thought to occur by simple diffusion across the phospholipid bilayers of the cytoplasmic membrane. Uptake of 14C-labelled β-carotene by epithelial cells was linearly proportional to its extracellular concentration and independent of temperature. The incubation with excess unlabelled β-carotene did not inhibit the uptake [5]. Uptake of various carotenoids from the mixed micelles by Caco-2 human intestinal cells was correlated with their lipophilicity [6]. Lysophosphatidylcholine in the mixed micelles enhanced the carotenoid uptake by Caco-2 cells, suggesting that the physical perturbation of membrane integrity by lysophosphatidylcholine favors the transfer of carotenoids across the plasma membrane [7]. These results support the simple diffusion mechanism for carotenoid uptake by the epithelial cells of the intestine. However, recent studies have suggested that transporters are involved in intestinal absorption of carotenoids. Plasma lycopene level in mice overexpressing scavenger receptor class B type I (SR-B1) increased 10-fold in comparison with wildtype mice [8]. Treatment of Caco-2 cells with anti-SR-B1 antibody and an SR-B1 inhibitor partially inhibited the cellular uptake of lycopene and lutein. SR-B1 knockout mouse showed lower intestinal absorption of β-carotene than wild-type mouse [9]. However, impairment of SR-B1 did not completely block carotenoid absorption. Therefore, it is most likely that one part of carotenoids is absorbed by simple diffusion and the other part by transporter-mediated mechanism through SR-B1 as well as other unknown transporters. After uptake of carotenoids by the epithelial cells, intracelluar trafficking of carotenoids and incorporation into the chylomicron occur before secretion to lymph. The chylomicron is assembled with apolipoproteins, phospholipids and triacylglycerols resynthesized from digests of fats and oils. Therefore, dietary fats and oils would affect the secretion of carotenoids to lymph as well as the solubilization process described above. However, details of secretion process after carotenoid uptake by epithelial cells has not been clarified yet. It has been well known that accumulation of carotenoids in tissues is largely varied among mammal species. Humans accumulate both carotenes and xanthophylls, whereas cows accumulate only carotenes and rodents accumulate little carotenoids. It has not been revealed yet whether these variations are due to the differences in intestinal absorption or metabolism in the body.

58

Nagao

Metabolic Transformation

Provitamin A carotenoids are oxidized at the central double bond to two molecules of retinal by 15,15⬘-dioxygenase in the intestinal epithelium [10]. Retinal is reduced to retinol and then converted to retinyl esters, which are transported to the liver after incorporation into the chylomicron. The dioxygenase is present in several tissues other than the intestine and would work to produce vitamin A directly from carotenoids accumulated in the peripheral tissues. Recently, a gene encoding another dioxygenase which can oxidize 9⬘,10⬘-double bond of carotenoids was discovered [11]. The enzyme expressed in Escherichia coli converted lycopene as well as provitamin A carotenoids to the corresponding carbonyl compounds. However, in vivo formation of metabolites derived by this asymmetric cleavage enzyme reaction has not been confirmed yet. In addition to these enzymatic cleavage reactions, chemical oxidation of carotenoids in vitro are known to produce carbonyl compounds, which are produced by cleavage at any position of conjugated double bonds [12]. Thus, cleavage reaction might occur in vivo whether it is mediated by enzyme or reactive oxygen species. Oxidation products of astaxanthin, which might be formed by cleavage at 9, 10-double bond, were found in the human plasma after intake of astaxanthin [13]. Moreover, 3-hydroxy-β-ionone and 3-hydroxy-14⬘-apocarotenal were detected as cleavage products of zeaxanthin in the human macula. These cleavage products found in human tissues indicate that the cleavage reaction would occur in vivo. Various carotenoid metabolites have been found in human plasma and breast milk. Khachik et al. [14] found 34 carotenoids including geometrical isomers, among which 9 were thought to be metabolites because they were not present in foods. 2,6-cyclolycopene-1,5-diol was identified as a metabolite of lycopene and suggested to be formed via 2,6-cyclolycopene-1,5-epoxide from lycopene-1,2-epoxide, an oxidation product of lycopene, by enzymatic or nonenzymatic hydrolysis [14]. One of the major metabolites of lutein, anhydrolutein, was found in human plasma and thought to be formed by dehydration under acidic condition in the stomach. Other metabolites of lutein in human plasma were 3-hydroxy-β,ε-carotene-3⬘-one, 3⬘-hydroxy-ε,ε-carotene-3-one, ε,ε-carotene-3,3⬘-dione, and 3⬘-epilutein [15]. These metabolites as well as zeaxanthin were thought to be formed from lutein by repeated oxidation, reduction and isomerization. Moreover, canthaxanthin and capsanthone were found in human plasma after intake of 4, 4’-dimethoxy-β-carotene [16] and capsanthin [17], respectively. Amarouciaxanthin A was found in plasma and liver of mouse fed fucoxanthin [18], a major xanthophyll in brown algae. Fucoxanthin was hydrolyzed to fucoxanthinol, and then oxidatively converted to amarouciaxanthin A by NAD-dependent dehydrogenase present in liver microsomes. These results suggest that mammals have a metabolic activity to oxidize the secondary hydroxyl group to carbonyl group in xanthophylls. Crocetin, a C20 apocarotenoid, is present in saffron stigmas and gardenia fruits and has been used as yellow food colorings. It has a characteristic structure with carboxyl

Absorption of Carotenoids

59

groups at both ends of the carbon chain, which is shorter than usual carotenoids. In mice, crocetin was rapidly absorbed into the blood circulation and metabolized to glucuronide conjugates, which would be excreted to urine [19]. This metabolism suggests that carotenoids are finally eliminated from the body as glucuronide conjugates after cleavage to apocarotenoids. Thus, in mammals, carotenoids are transformed to various derivatives with C40 carbon chain as well as smaller molecules. However, biological actions of these metabolites remain largely unknown except for vitamin A.

Functions

Provitamin A carotenoid present in foods is an important source of vitamin A, which plays essential roles in cell differentiation, growth and vision. This physiological function is not described here, because it is beyond the scope of this chapter. Most carotenoids have a long chain of conjugated double bonds, which endow these molecules with antioxidant activities. Carotenoids that have more than 9 conjugated double bonds can quench singlet oxygen by physically receiving its energy, and then, the excited carotenoids turn to ground state by releasing energy as heat. Thus, one molecule of the carotenoids can quench singlet oxygen repeatedly [20]. The reactivity of the carotenoids with singlet oxygen is much higher than those of vitamin E and C, which quench singlet oxygen by chemical reaction. Therefore, carotenoids are excellent singlet oxygen quenchers among dietary antioxidants and have been thought to protect tissues from singlet oxygen, which is generated in skin exposed to UV irradiation through the endogenous photosensitizers and in macrophage phagocytosis through myeloperoxidase reaction. Carotenoids work as antioxidants by scavenging oxygen radicals as well as quenching singlet oxygen. Lipid peroxy radicals make a stable adduct to conjugated double bonds of carotenoids. Under the low oxygen pressure, the carbon radical adduct is stabilized by the conjugation system. However, reactivity of carotenoids with lipid peroxy radicals is lower than that of vitamin E, which is present in tissues at higher concentration than carotenoids. Therefore, carotenoid is not a major antioxidant to scavenge oxygen radicals in lipid peroxidation in vivo. On the other hand, carotenoid was found to scavenge peroxynitrite in isolated LDL, suggesting the protection of blood lipoproteins from peroxynitrite and prevention of atherosclerosis initiation by carotenoids [21]. These antioxidant properties of carotenoids as well as vitamin E and C are thought to prevent oxidative lesions which would lead to cancer and cardiovascular diseases. In addition to antioxidant and provitamin A activities, individual carotenoids have been reported to have various biological activities. In animal models, β-carotene, α-carotene, lycopene, β-cryptoxanthin and lutein have been reported to suppress carcinogenesis of the liver, colon, skin, and lung initiated with chemicals. Feeding rats with β-carotene reduced the number of aberrant crypt foci and incidence of colon cancer induced by azoxymethane.

60

Nagao

In the cell culture system, carotenoids have been reported to suppress propagation of cancer cells by cell cycle inhibition, induction of differentiation and apoptosis, and enhancement of gap-junctional communication, since Murakosi et al. [22] first reported that α-carotene suppressed the propagation of human neuroblastoma cells in 1989. Lycopene inhibited cell growth of androgen-independent human prostate cancer cells at the G0/G1 phase and of normal epithelial cells of prostate by suppressing expression of cyclin D [23]. These biological activities of lycopene would support the epidemiological observation that lycopene intake is associated with lowered incidence of prostate cancer. β-Carotene inhibited propagation of several colon cancer cells and the inhibition might be mediated by suppression of cyclooxygenase 2 expression and formation of prostaglandin E2. Crocetin, crocin, lutein, β-carotene, and lycopene inhibited propagation of human promyelocytic leukemia cells by inducing cell differentiation. Certain carotenoids cause apoptosis induction, which is an important function in prevention and treatment of cancer. In 1995, Muto et al. [24] first reported that β-carotene inhibited cell growth of human cervical dysplastic cells by inducing apoptosis, which is mediated by suppression of EGF receptor protein level. β-Carotene as well as retinoic acid induced apoptosis in DU145 human prostate cancer cells, and lycopene at high concentration induced apoptosis in LNCaP human prostate cancer cells by enhancing permeability of the mitochondrial membrane. Indeed, intake of tomato products containing lycopene by subjects of prostate cancer and hyperplasia increase the ratio of apoptotic cells in the prostate [25]. Polar xanthophylls, neoxanthin and fucoxanthin, which have a characteristic allenic bond and are present in green leafy vegetables and edible brown algae, respectively, also induced apoptosis of human prostate cancer cells as well as colon cancer cells [26]. Gap-junctional communication among cells is involved in maintaining tissue homeostasis. In cancer tissues, cell propagation is not properly regulated due to the reduced communication caused by a decreased gap junction. β-Carotene, canthaxanthin, lutein, lycopene and retinoic acid suppressed transformation of mouse fibroblast C3H/10T1/2 treated with methylcholanthrene by increasing gap-junctional communication [27], which was mediated by increased expression of connexin 43 encoding a gap-junction protein. The increased gap junction was also found in the liver of rats fed α-carotene, β-carotene, and lycopene. In humans, mRNA level of connexin 43 increased in the colon mucosa of subjects who ingested β-carotene. Immune enhancement by carotenoid intake has been thought to be an important biological activity in cancer prevention, in particular, for aged people, who have less immune ability and high risk of cancer. Intake of β-carotene by aged people for a long period enhanced activity of natural killer cells engaged in cell-mediated immune response [28]. After administration of β-carotene to nonsmoking healthy men, monocytes expressing MHC class II (HLA-DR) and adhesion molecules (ICAM-1 and LFA-3) increased, and secretion of TNF-α by the isolated monocytes was enhanced, indicating enhancement in antigen presentation [29]. As lycopene,

Absorption of Carotenoids

61

lutein, canthaxanthin, and astaxanthin as well as β-carotene have been reported to show immune enhancement, it is not attributed to vitamin A activity, but to intrinsic properties of each carotenoid. The detailed mechanism of immune enhancement by carotenoids has not been revealed yet. The carotenoids as antioxidants are thought to protect immune cells, which are vulnerable to oxidative stress, and to modulate formation of arachidonate metabolites, which suppress immune functions. In addition to immune enhancement, carotenoids may have suppressive effect on allergic reaction. Simultaneous intake of vitamin E and β-carotene was reported to suppress IgE production in mice [30]. Carotenoids have been reported to show several biological activities other than those mentioned above. Inhibition of platelet aggregation, growth inhibition of human aortic smooth muscle cells, and suppression of adhesion molecule expression in human aortic endothelial cells were reported. These effects as well as antioxidant activity of carotenoids are thought to be involved in prevention of cardiovascular diseases. In a recent study, fucoxanthin was found to reduce fat accumulation in mice. Thus, carotenoids showed various biological activities and are expected to have beneficial effects on human health. However, the exact mechanisms have not been fully revealed and deserve future studies.

References 1 Stahl W, Sies H: Bioactivity and protective effects of natural carotenoids. Biochim Biophys Acta 2005; 1740:101–107. 2 Yonekura L, Nagao A: Intestinal absorption of dietary carotenoids. Mol Nutr Food Res 2007;51: 107–115. 3 Gartner C, Stahl W, Sies H: Lycopene is more bioavailable from tomato paste than from fresh tomatoes. Am J Clin Nutr 1997;66:116–122. 4 Borel P, Grolier P, Armand M, Partier A, Lafont H, Lairon D, Azais-Braesco V: Carotenoids in biological emulsions: solubility, surface-to-core distribution, and release from lipid droplets. J Lipid Res 1996;37:250–261. 5 Scita G, Aponte GS, Wolf G: Uptake and cleavage of beta carotene by cultures of rat small intestinal cells and human lung fibroblasts. J Nutr Biochem 1992;3: 118–123. 6 Sugawara T, Kushiro M, Zhang H, Nara E, Ono H, Nagao A: Lysophosphatidylcholine enhances carotenoid uptake from mixed micelles by Caco-2 human intestinal cells. J Nutr 2001;131:2921–2927. 7 Yonekura L, Tsuzuki W, Nagao A: Acyl moieties modulate the effects of phospholipids on beta-carotene uptake by Caco-2 cells. Lipids 2006;41:629– 636.

62

8 Moussa M, Landrier JF, Reboul E, Ghiringhelli O, Comera C, Collet X, Frohlich K, Bohm V, Borel P: Lycopene absorption in human intestinal cells and in mice involves scavenger receptor class B type I but not Niemann-Pick C1-like 1. J Nutr 2008;138: 1432–1436. 9 van Bennekum A, Werder M, Thuahnai ST, Han CH, Duong P, Williams DL, Wettstein P, Schulthess G, Phillips MC, Hauser H: Class B scavenger receptor-mediated intestinal absorption of dietary carotene and cholesterol. Biochemistry 2005;44: 4517–4525. 10 Nagao A, During A, Hoshino C, Terao J, Olson JA: Stoichiometric conversion of all trans-beta-carotene to retinal by pig intestinal extract. Arch Biochem Biophys 1996;328:57–63. 11 Kiefer C, Hessel S, Lampert JM, Vogt K, Lederer MO, Breithaupt DE, von Lintig J: Identification and characterization of a mammalian enzyme catalyzing the asymmetric oxidative cleavage of provitamin A. J Biol Chem 2001;276:14110–14116. 12 Kim SJ, Nara E, Kobayashi H, Terao J, Nagao A: Formation of cleavage products by autoxidation of lycopene. Lipids 2001;36:191–199.

Nagao

13 Kistler A, Liechti H, Pichard L, Wolz E, Oesterhelt G, Hayes A, Maurel P: Metabolism and CYP-inducer properties of astaxanthin in man and primary human hepatocytes. Arch Toxicol 2002;75:665–675. 14 Khachik F, Spangler CJ, Smith JC Jr, Canfield LM, Steck A, Pfander H: Identification, quantification, and relative concentrations of carotenoids and their metabolites in human milk and serum. Anal Chem 1997;69:1873–1881. 15 Khachik F, de Moura FF, Zhao DY, Aebischer CP, Bernstein PS: Transformations of selected carotenoids in plasma, liver, and ocular tissues of humans and in nonprimate animal models. Invest Ophthalmol Vis Sci 2002;43:3383–3392. 16 Zeng S, Furr HC, Olson JA: Metabolism of carotenoid analogs in humans. Am J Clin Nutr 1992;56: 433–439. 17 Etoh H, Utsunomiya Y, Komori A, Murakami Y, Oshima S, Inakuma T: Carotenoids in human blood plasma after ingesting paprika juice. Biosci Biotechnol Biochem 2000;64:1096–1098. 18 Asai A, Sugawara T, Ono H, Nagao A: Biotransformation of fucoxanthinol into amarouciaxanthin A in mice and HepG2 cells: formation and cytotoxicity of fucoxanthin metabolites. Drug Metab Dispos 2004;32:205–211. 19 Asai A, Nakano T, Takahashi M, Nagao A: Orally administered crocetin and crocins are absorbed into blood plasma as crocetin and its glucuronide conjugates in mice. J Agric Food Chem 2005;53:7302– 7306. 20 Di Mascio P, Kaiser S, Sies H: Lycopene as the most efficient biological carotenoid singlet oxygen quencher. Arch Biochem Biophys 1989;274:532– 538. 21 Panasenko OM, Sharov VS, Briviba K, Sies H: Interaction of peroxynitrite with carotenoids in human low density lipoproteins. Arch Biochem Biophys 2000;373:302–305. 22 Murakoshi M, Takayasu J, Kimura O, Kohmura E, Nishino H, Iwashima A, Okuzumi J, Sakai T, Sugimoto T, Imanishi J, Iwasaki R: Inhibitory effects of alpha-carotene on proliferation of the human neuroblastoma cell line GOTO. J Natl Cancer Inst 1989;81:1649–1652.

23 Obermuller-Jevic UC, Olano-Martin E, Corbacho AM, Eiserich JP, van der Vliet A, Valacchi G, Cross CE, Packer L: Lycopene inhibits the growth of normal human prostate epithelial cells in vitro. J Nutr 2003;133:3356–3360. 24 Muto Y, Fujii J, Shidoji Y, Moriwaki H, Kawaguchi T, Noda T: Growth retardation in human cervical dysplasia-derived cell lines by beta-carotene through down-regulation of epidermal growth factor receptor. Am J Clin Nutr 1995;62:1535S-1540S. 25 Kim HS, Bowen P, Chen LW, Duncan C, Ghosh L, Sharifi R, Christov K: Effects of tomato sauce consumption on apoptotic cell death in prostate benign hyperplasia and carcinoma. Nutr Cancer 2003;47: 40–47. 26 Kotake-Nara E, Kushiro M, Zhang H, Sugawara T, Miyashita K, Nagao A: Carotenoids affect proliferation of human prostate cancer cells. J Nutr 2001; 131:3303–3306. 27 Zhang LX, Cooney RV, Bertram JS: Carotenoids enhance gap junctional communication and inhibit lipid peroxidation in C3H/10T1/2 cells: relationship to their cancer chemopreventive action. Carcinogenesis 1991;12:2109–2114. 28 Santos MS, Meydani SN, Leka L, Wu D, Fotouhi N, Meydani M, Hennekens CH, Gaziano JM: Natural killer cell activity in elderly men is enhanced by beta-carotene supplementation. Am J Clin Nutr 1996;64:772–777. 29 Hughes DA, Wright AJ, Finglas PM, Peerless AC, Bailey AL, Astley SB, Pinder AC, Southon S: The effect of beta-carotene supplementation on the immune function of blood monocytes from healthy male nonsmokers. J Lab Clin Med 1997;129:309– 317. 30 Bando N, Yamanishi R, Terao J: Inhibition of immunoglobulin E production in allergic model mice by supplementation with vitamin E and beta-carotene. Biosci Biotechnol Biochem 2003;67:2176–2182.

Dr. Akihiko Nagao National Food Research institute, National Agriculture and Food Research Organization Kannondai 2–1-12 Tsukuba, Ibaraki, 305-8642 (Japan) Tel. +81 298 838 8039, Fax +81 298 838 7996, E-Mail [email protected]

Absorption of Carotenoids

63

Bioavailability and Safety Yoshikawa T (ed): Food Factors for Health Promotion. Forum Nutr. Basel, Karger, 2009, vol 61, pp 64–74

Metabolism of Flavonoids Yu Wang ⭈ Chi-Tang Ho Department of Food Science, Rutgers University, New Brunswick, N.J., USA

Abstract Flavonoids, the most abundant polyphenolic compounds in foods, can be classified into flavanols, flavones, flavonols, flavanones, isoflavones and anthocyanidins. They have been demonstrated to possess strong antioxidant and disease-preventing properties especially for various degenerative diseases such as cancers and cardiovascular diseases in in vitro and in vivo models. However, flavonoids undergo metabolic transformation such as methylation, sulfation and glucuronidation, and consequently changes of their structures and biological activities. In order to reveal a relationship between parent compounds and their metabolites, the changes of structure and bioactivity of variCopyright © 2009 S. Karger AG, Basel ous flavonoids are reviewed in the vitro and vivo models.

Flavonoids are a class of plant secondary metabolites widely distributed in the leaves, seeds, bark and flowers of plants. They have attracted great deal of attention in the past few years, because they have been demonstrated to occur ubiquitously and play potentially protective roles in human health. Flavonoids are benzo-γ-pyrone derivatives consisting of phenolic and pyrane rings, and are classified into flavanols, flavones, flavonols, flavanones, isoflavones and anthocyanidins. Flavonols are the most abundant flavonoids in foods. Quercetin and kaempferol, which are the major representatives of flavonols, are very rich in onions (up to 1.2 g/kg fresh weight), curly kale, leeks, broccoli and blueberries [1]. Those compounds generally occur in the glycosylated forms, which are preferable to glucose and rhamnose. Compared to flavonols, flavones are found much less in fruits and vegetables, and the most common edible source of flavones is parsley and celery which mainly contain glycosides of luteolin and apigenin. However, polymethoxyflavones (PMF), a general term for flavones, bear four or more methoxy groups. PMFs exist almost exclusively in citrus plants [2]. Isoflavones are one class of phytoestrogens. Due to their structural similarly to estradiol, they can bind to estrogen receptors. The major sources of isoflavones in the human diet are soya and its products such as soy flour, miso, tofu and soy milk which contain three major compounds, genistein, daidzein, and glycitein at a concentration ratio of 1:1:0.2 [3]. Flavanones mostly occur as the glycosylation

with the disaccharides, which may impact bitter taste. However, the aglycones of flavanones could be found such as naringenin in grapefruit, hesperetin in oranges, and eriodictyol in lemons [4]. Flavanols are discussed in their monomers (catechins) and polymers (proanthocyanidins). Apricots contain 250 mg catechins/kg fresh weight and are the fruits containing the highest level of catechins; they are followed by cherries, grapes and apples. However, green tea and chocolate, which contain 800 and 610 mg catechins/kg fresh weight, respectively, are by far its richest sources, particularly gallocatechin (GC), epigallocatechin (EGC), epicatechin gallate (ECG), and epigallocatechin gallate (EGCG) in tea [1, 5, 6]. Compared with other classes of flavonoids, flavanols are not glycosylated in foods. Proanthocyanidins are dimers, oligomers and polymers of catechins which are condensed together at positions 4 and 8 (or 6). Anthocyanins exist in different chemical structures which show various colors according to pH. Aglycones of anthocyanins are sensitive to light, pH or oxygen, but they can be stabilized by glycosation generally with glucose at position 3, or esterification with organic acids and phenolic acids. Anthocyanins are abundant in foods with intensive color such as blackcurrants or blackberries (4,000 mg/kg fresh weight) [7]. Flavonoids have been demonstrated to be protective against degenerative diseases such as cancers, cardiovascular diseases and neurodegenerative disorders in in vitro and in vivo models. Studies show that their beneficial effects are due to their ability to scavenge free radicals, enhance or inhibit some specific enzymes, induce apoptosis in tumor cells, and regulate the immune system. Quercetin has been shown to be a potent antioxidant and anti-inflammatory agent that protects blood vessels, cells and their structures from the harmful effects produced by free radicals. Quercetin also modulates phase 1 and phase 2 enzymes and inhibits azoxymethane-induced colorectal carcinogenesis. Kaempferol causes G2/M arrest and induces apoptosis in human esophageal adenocarcinoma cells, and inhibits STAT1 (signal transducer and activator of transcript 1) and NK-κB (nuclear factor κB) activation in activated macrophages. Green tea catechins have been the most studied health-promoting flavonoids in recent years. Tea is one of the most widely consumed beverages in the world. Green tea and its catechin constituents have been extensively studied both in vitro and in animal models of carcinogenesis. Cyanidin, an anthocyanidin present in cherry and strawberry, exhibited significant decrease in CCl4-induced lipid and protein peroxidation and induced G2/M arrest and apoptosis in U937 cells. Delphinidin, an anthocyanidin present in dark fruit, is believed to contribute to the inhibition of cyclooxygenase-2 expression by blocking mitogen-activated protein kinase signaling and nuclear factor-κB, activator protein-1 and C/EBPδ nuclear translocation. Health benefits of soybean and its products have been recognized in recent years. Genistein, which is an isoflavone, is considered to be the main nutraceutical in soybeans. As a phytoestrogen, genistein lacks estrogenic activity and exhibits antiestrogenic activity. Genistein induced apoptosis by activation of calpain-caspase and ASK-1 signaling pathway in human breast cancer MCF-7 cells [8].

Metabolism of Flavonoids

65

Considering health benefits of flavonoids, it becomes of importance to understand their metabolism and bioavailability which may provide sufficient information on how these compounds are utilized. Bioavailability is an overall effect of absorption, biotransformation (metabolism), distribution, and excretion. One of the most important factors of bioavailability is metabolism, which is a detoxification process and can be grouped into phase I and phase II enzyme systems. Xenobiotic substances are transformed into more hydrophilic molecules by those enzymes so that they can be easily excreted from the body. Phase I enzymes involve oxidation and reduction reactions as well as hydrolysis of esters, amides and ether linkages, and in turn introduce stronger hydrophilic groups such as –OH, –NH2, –SH, and –CO2H. Phase II enzymes, also called transferases or conjugation enzymes, link the hydrophilic groups described above to even stronger hydrophilic groups, such as glucuronic acid or sulphate. Consequently, the polarity and hydrophilicity of thus formed molecules are significantly increased and the tendency of these molecules to be easily absorbed and excreted is also greatly increased [9]. It has been indicated that the biological activities of flavonoids in vivo depend on bioavailability and their metabolites generated during metabolism. Although native flavonoids have been described as powerful beneficial agents, these compounds undergo extensive metabolism in liver, small intestine and colon, leading to biotransformation such as methylation, glucuronidation, sulfation as well as degradation into phenolic acids or other compounds. Biotransformation can dramatically alter the biological properties. Therefore, it is still hard to attribute these health effects to native forms of flavonoids, and characterization of metabolism and verification of metabolites should be accomplished systematically. This review summarizes current knowledge on the metabolic fate of flavonoids in in vitro and in vivo models, and provides insight into biotransformation of parent compounds and biological activities of metabolites.

Metabolism of Flavonoids

After oral administration, flavonoids are extensively metabolized during absorption in the small intestine, and subsequently go though liver metabolism, which includes conjugates of O-methylation, sulfation and glucuronidation. The nonabsorbed fraction, particularly flavonoid glycosides, reach the colon, degraded by colonic microflora, and through the circulation extensively metabolized in the liver. However, during the metabolism of flavonoids, there are some exceptional cases. EGCG can be hydrolyzed to EGC in human saliva by the esterase [10]. Luteolin-7-glucoside, kaempferol-3-glucoside and quercetin-3-glucoside are hydrolyzed and absorbed in the small intestine, but rhamnoglucoside and diglucosides cannot be cleaved, indicating that the activity of β-glucosidase in the small intestine is dependent on the location and structure of sugar moiety [11]. In addition, anthocyanin glycosides can be absorbed intact [12]. All those metabolites are eliminated through biliary or urinary routes.

66

Wang · Ho

Large conjugates are more likely to be excreted through the biliary way, whereas small conjugates preferentially follow the urinary route. Based on this evidence, metabolism of flavonoids is grouped into three categories: (1) phase І biotransformation of flavonoids including hydroxylation and demethylation, (2) flavonoid conjugates, and (3) colonic metabolism. All these metabolites are different from their parent compounds in the structure and bioactivity.

Phase І Biotransformation of Flavonoids

The metabolism of flavonoids, particularly hydroxylation and demethylation is considered to be mediated by cytochrome 450. Human liver microsomes transformed the soy isoflavone daidzein (7,4⬘-dihydroxyisoflavone) to three monohydroxylated and three dihydroxylated metabolites namely, 6,7,4⬘-trihydroxyisoflavone, 7,3⬘,4⬘trihydroxyisoflavone, 7,8,4⬘-trihydroxyisoflavone as well as 7,8,3⬘,4⬘-tetrahydroxyisoflavone, 6,7,8,4⬘-tetrahydroxyisoflavone, and 6,7,3⬘,4⬘-tetrahydroxyisoflavone. Genistein (5,7,4⬘-trihydroxyisoflavone) was metabolized by human liver microsomes into six hydroxylation products. The main metabolites were the three aromatic monohydroxylated products 5,6,7,4⬘-tetrahydroxyisoflavone, 5,7,8,4⬘-tetrahydroxyisoflavone and 5,7,3⬘,4⬘-tetrahydroxyisoflavone [13]. The human liver microsomal study of flavonoid metabolism showed that galangin (3,5,7-trihydroxyflavone) and kaempferide (3,5,7-trihydroxy-4⬘-methoxyflavone) could be hydroxylated and demethylated, respectively, to kaempferol (3,5,7,4⬘-tetrahydroxyflavone) [14]. Demethylation of 3⬘-O-methylquercetin has been observed to generate quercetin in the fibroblasts [15]. In fact, P450-related metabolism of flavonoids especially demethylation is mainly focused on the subclass of PMFs. The interaction between PMFs and liver cytochrome P-450 isozymes has been observed. For example, early studies showed that the activity of some cytochrome P450 enzymes, such as 7-ethoxyresorufin-O-deethylase (classified as CYP 1A) and nifidifine oxidase (CYP 3A4) in human liver microsomes, was inhibited by tangeretin in a noncompetitive manner. Cytochrome P450 (CYP) is the key enzyme system involved in the metabolism of PMFs and is capable of catalyzing hydroxylation and demethylation reactions. The metabolic pathway of PMFs is considered to be identical across the species. The 3⬘ and 4⬘ positions on the B ring of PMFs are the primary site of biotransformation. The number and position of the hydroxyl and methoxy groups on the B ring of PMFs have a great influence on the metabolism of PMFs [2]. In an in vitro experiment, when tangeretin was incubated with Aroclor-induced rat liver microsomes, the major metabolites were identified as 4⬘-demethyltangeretin, 3⬘,4⬘-dihydroxy-5,6,7,8-tetramethoxyflavone and 5,4⬘-didemethyltangeretin or 5,4⬘- dihydroxy-6,7,8-trimethoxyflavone; whereas in the in vivo biotransformation study of tangeretin where rats were submitted to repeated gavage, the major metabolites found in rat urine and feces were characterized as 4⬘-demethyltangeretin and

Metabolism of Flavonoids

67

3⬘,4⬘-dihydroxy-5,6,7,8-tetramethoxyflavone, though other metabolites with intact 4⬘-methoxy group and demethylation at various positions on the A ring were also observed. Hence, it can be concluded that the 4⬘-methoxy group of tangeretin is the primary site for demethylation and the 3⬘ position is the most vulnerable site for hydroxylation or oxidation by phase I enzymes (fig. 1). Also, urine analysis determined that 38% of the tangeretin metabolites were excreted as conjugates of glucuronates and sulfates [2]. In the in vitro biotransformation study of nobiletin, nobiletin was treated with rat liver S-9 mixture for 24 h, and 3⬘-demethylnobiletin was identified as the major metabolite. It has also been found that the demethylation rate of nobiletin is slow with a half-life of >24 h, suggesting that the half-life of nobiletin in an in vivo system might be considerably long. The in vivo metabolic study of nobiletin in male SpragueDawley rats identified the dominant metabolite as 3⬘-demethylnobiletin with two other mono-demethylated nobiletins and two di-demethylnobiletin products. Also in this in vivo experiment, 3⬘-demethylnobiletin was the only metabolite that was detected in serum. In a different nobiletin biotransformation study on male SpragueDawley rats, 4⬘-demethylnobiletin was isolated and characterized with two other minor metabolites. A recent metabolic study of nobiletin in CD-1 mice identified three metabolites in urine and plasma: 3⬘-demethylnobiletin, 4⬘-demethylnobiletin and 3⬘,4⬘-didemethylnobiletin (or 3⬘,4⬘-dihydroxy-5,6,7,8-tetramethoxyflavone) [2].

Flavonoid Conjugates

The P450-mediated metabolism of flavonoids has been shown in liver microsomes, but the metabolism of conjugates is thought to be the competitor to the phase І metabolism. Methylation and conjugation with glucuronic acid or sulfate are the major forms of conjugated metabolites of flavonoids. Because of the catechol and pyrogallol structures, flavonoids are metabolized primarily to the 3⬘-O-methyl and 4⬘-O-methyl derivatives, respectively. Quercetin is metabolized to 3⬘-O-methylquercetin as the primary metabolite and 4⬘-O-methylquercetin as the secondary metabolite [16]. Likewise, anthocyanins show the same trend. The metabolite of delphinidin 3-O-βd-glucopyranoside is 4⬘-O-methyldelphinidin 3-O-β-d-glucopyranoside, indicating that methylation of the 4⬘-OH on the flavonoid B-ring is a common metabolic way for flavonoids carrying the pyrogallol [17]. EGC and EGCG can be methylated to 4⬘-O-methyl-EGC and 4⬘⬘-O-methyl-EGCG or 4⬘,4⬘⬘-O-dimethyl-EGCG, respectively, by the catechol-O-methyltransferase (COMT) which shows the higher activity in rat liver cytosol than in human and mouse liver cytosol [10]. Moreover, the dimethylated metabolite is preferable at lower concentrations of EGCG, whereas at higher concentrations, monomethylation is dominant. Besides methylation, glucuronidation and sulfation are also common conjugated metabolites. Glucuronidation of flavanols, flavones, flavonols, flavanones has been

68

Wang · Ho

OMe

OMe MeO

O

MeO

–Me

OH

OMe MeO

O

OH

OMe [O]

MeO

O

OH

MeO

MeO OMe O

OMe O

Tangeretin

4'-demethyltangeretin

OMe O 3', 4'-dihydroxy-5, 6, 7, 8tetramethoxyflavone

Fig. 1. Biotransformation pathway of tangeretin in rat.

observed. The number of the available hydroxyl groups may influence the extent of glucuronidation. Flavonoids containing catechol or pyrogallol B ring are susceptible to glucuronidation, whereas those containing monohydroxylated B ring are less susceptible to glucuronidation [18]. EGCG-4⬘⬘-O-glucuronide and EGC-3⬘-Oglucuronide are the major metabolites formed in human, rat and mouse microsomes [10]. However, EC undergoes sulfation in the human liver and intestine. In addition, conjugated EGCG (methylated, glucuronated, or sulfated) can be further conjugated with methylation, glucuronidation and/or sulfation to form mixed EGCG metabolites [10]. Glucuronide and/or sulfate conjugates of quercetin have been reported in rat and human models after administration of quercetin [19, 20]. The affinity of each position (4⬘-, 3⬘-, 7-, 3- and 5-) to the UDP-glucuronosyltransferases has been demonstrated following the order 4⬘-