NUTRIGENOMICS AND PROTEOMICS IN HEALTH AND DISEASE€¦ · Section I Introduction 1. Nutrigenomics...

NUTRIGENOMICS ANDPROTEOMICS IN

HEALTH AND DISEASEFood Factors and Gene Interact ions

Yoshinori Mine, Kazuo Miyashita and Fereidoon Shahidi EDITORS

Functional Food Science and Technology Series

Fereidoon Shahidi SERIES EDITOR

BLBS028-Mine January 31, 2009 13:26

ii


Nutrigenomics and Proteomicsin Health and Disease

Food Factors and Gene Interactions

i


ii


Nutrigenomics and Proteomicsin Health and Disease

Food Factors and Gene Interactions

EditorsYoshinori Mine

Kazuo MiyashitaFereidoon Shahidi

Functional Food Science and Technology Series

Fereidoon Shahidi SERIES EDITOR

A John Wiley & Sons, Ltd., Publication

iii


Edition first published 2009C© 2009 Wiley-Blackwell

Blackwell Publishing was acquired by John Wiley & Sons in February 2007. Blackwell’s publishing program hasbeen merged with Wiley’s global Scientific, Technical, and Medical business to form Wiley-Blackwell.

Editorial Office2121 State Avenue, Ames, Iowa 50014-8300, USA

For details of our global editorial offices, for customer services, and for information about how to apply forpermission to reuse the copyright material in this book, please see our website atwww.wiley.com/wiley-blackwell.

Authorization to photocopy items for internal or personal use, or the internal or personal use of specific clients,is granted by Blackwell Publishing, provided that the base fee is paid directly to the Copyright Clearance Center,222 Rosewood Drive, Danvers, MA 01923. For those organizations that have been granted a photocopy licenseby CCC, a separate system of payments has been arranged. The fee codes for users of the TransactionalReporting Service are ISBN-13: 978-0-8138-0033-2/2009.

Designations used by companies to distinguish their products are often claimed as trademarks. All brand namesand product names used in this book are trade names, service marks, trademarks or registered trademarks of theirrespective owners. The publisher is not associated with any product or vendor mentioned in this book. Thispublication is designed to provide accurate and authoritative information in regard to the subject matter covered.It is sold on the understanding that the publisher is not engaged in rendering professional services. If professionaladvice or other expert assistance is required, the services of a competent professional should be sought.

Library of Congress Cataloguing-in-Publication Data

Nutrigenomics and proteomics in health and disease : food factors and gene interactions / edited by YoshinoriMine, Kazuo Miyashita, Fereidoon Shahidi.

p. ; cm.Includes bibliographical references and index.ISBN-13: 978-0-8138-0033-2 (alk. paper)ISBN-10: 0-8138-0033-1 (alk. paper)1. Nutrition–Genetic aspects. 2. Proteomics. I. Mine, Yoshinori. II. Miyashita, Kazuo. III. Shahidi,Fereidoon, 1951-[DNLM: 1. Nutrigenomics. 2. Nutrition Physiology–genetics. 3. Diet Therapy. 4. Obesity–genetics.5. Proteomics. 6. Receptors, Cytoplasmic and Nuclear–genetics. QU 145 N97059 2009]QP144.G45N885 2009613.2–dc22

2008041831

A catalog record for this book is available from the U.S. Library of Congress.

Set in 9.5/11 pt Times New Roman by Aptara R© Inc., New Delhi, IndiaPrinted and Bound in the United States.

1 2009

iv

http://www.wiley.com/wiley-blackwell


Contents

Preface viiContributors ix

Section I Introduction1. Nutrigenomics and Proteomics in Health and Disease: An Overview 3

Yoshinori Mine, Kazuo Miyashita, and Fereidoon Shahidi2. Omics in Nutrition and Health Research 11

Michael Affolter, Frederic Raymond, and Martin Kussmann

Section II Genomics and Proteomics in Health and Diseases3. Toward Personalized Nutrition and Medicine: Promises and Challenges 33

Baitang Ning and Jim Kaput4. Obesity and Nuclear Receptors: Effective Genomic Strategies in Functional Foods 47

Teruo Kawada, Tsuyoshi Goto, Shizuka Hirai, Rina Yu, and Nobuyuki Takahashi5. Inflammatory Genes Involved in Obesity-Induced Inflammatory Responses and Pathologies 59

Rina Yu6. Genomics and Proteomics in Allergy 67

Icy D’Silva and Yoshinori Mine

Section III Food Factors–Gene Interactions7. Beneficial Effects of Conjugated Linoleic Acid 85

Kazunori Koba and Teruyoshi Yanagita8. Regulation of Gene Transcription by Fatty Acids 97

Jean-Paul Pegorier9. Nonnutrient Functionality of Amino Acids 115

Yoshinori Mine and Connie J. Kim10. Functional Bioactive Proteins and Peptides in Nutrigenomics 129

Denise Young and Yoshinori Mine11. Antiobesity Effect of Allenic Carotenoid, Fucoxanthin 145

Kazuo Miyashita and Masashi Hosokawa12. Control of Systemic Inflammation and Chronic Diseases—The Use of Turmeric

and Curcuminoids 161Stig Bengmark

v


vi Contents

13. Alteration in Gene Expression and Proteomic Profiles by Soy Isoflavone 181Fazlul H. Sarkar and Yiwei Li

14. Green Tea Polyphenol-Modulated Genome Functions for Protective Health Benefits 201Molay K. Roy, Mahendra P. Kapoor, and Lekh R. Juneja

15. Oat Avenanthramides: A Novel Antioxidant 239Li Li Ji, Ryan Koenig, and Mitchell L. Wise

16. Cancer-Preventive Effects and Molecular Actions of Anthocyanins 251De-Xing Hou

17. Food Components Activating Capsaicin Receptor TRPV1 263Tatsuo Watanabe, Yusaku Iwasaki, Akihito Morita, and Kenji Kobata

18. New Therapeutic Effects of Anthocyanins: Antiobesity Effect, Antidiabetes Effect,and Vision Improvement 273Takanori Tsuda and Hitoshi Matsumoto

19. Licorice Flavonoids 291Shinichi Honda, Yuji Tominaga, and Shinichi Yokota

20. Isoprenols 301Tsuyoshi Goto, Nobuyuki Takahashi, Shizuka Hirai, and Teruo Kawada

21. Anti-inflammatory and Anticarcinogenesis Potentials of Citrus Coumarinsand Polymethylated Flavonoids 311Akira Murakami and Hajime Ohigashi

22. Probiotics: Food for Thought 325Kingsley C. Anukam and Gregor Reid

Section IV Advanced Analytical Techniques for Nutrigenomics and Proteomics23. Microarrays: A Powerful Tool for Studying the Functions of Food and Its Nutrients 339

Hiroshi Mano, Jun Shimizu, and Masahiro Wada24. Challenges and Current Solutions in Proteomic Sample Preparations 351

Feng Tao25. Computational Methods in Cancer Gene Networking 367

Edwin Wang26. Peptidomics 375

Yoshinori Mine

Index 387


Preface

Recent advances in the areas of functional foods, nu-traceuticals, and natural health products have beenculminated by those in modern molecular nutrition.Thus, the advent of nutrigenomic, proteomic, andmetabolomic has resulted in a leap toward individu-alized nutrition, hopefully in the near future. In thisconnection, the present book on nutrigenomic andproteomic is expected to provide links and informa-tion relevant to health promotion and disease risk re-duction. Many of the bioactive components presentin foods or produced upon ingestion or upon process-ing under conditions mincing digestion are found toimprove health status related to cardiovascular dis-eases, certain types of cancer, inflammatory disordersand immune response, diabetes, gastrointestinal tractconditions as well as various psychological problems,and the metabolic syndrome. The techniques used tostudy such benefits have improved over the recentyears and unique tools have now become available

that facilitate undertaking of challenges thought im-possible only a decade ago.

This book provides a state-of-the-art compilationof the most recent developments in the exciting fieldof nutrigenomics and proteomics. It is of special in-terest to nutritionists, food scientists, biochemists,pharmacologists and biologists, among others. Thisbook serves as a reference compendium for scientistsin academia, industry, and government laboratories.It may also be used as a text for senior undergradu-ate and graduate students in multidisciplinary areaslisted. We are indebted to world-renowned scientistsfor their excellent contributions that made the publi-cation of this book possible.

Yoshinori MineKazuo Miyashita

Fereidoon Shahidi

vii


viii


Contributors

Michael AffolterFunctional Genomics GroupDepartment of BioAnalytical SciencesNestle Research CenterVers-chez-les-BlancCH-1000 Lausanne 26Switzerland

Kingsley C. AnukamCanadian Research & Development Centre

for ProbioticsLawson Health Research Institute, London, Ontario,

Canada

Stig BengmarkMember Academia EuropeaeEmeritus professor Lund University, Lund, SwedenHonorary Visiting Professor to Department of

Hepatology,University College London (UCL),London University,

69-75 Chenies Mews, London, WC1E 6HX,United Kingdom

Icy D’SilvaDepartment of Food ScienceUniversity of GuelphGuelph Ontario N1G2W1, Canada

Tsuyoshi GotoLaboratory of Molecular Function of FoodDivision of Food Science and BiotechnologyGraduate School of Agriculture, Kyoto

UniversityUji Kyoto 611-0011, Japan

Shizuka HiraiLaboratory of Molecular Function of FoodDivision of Food Science and BiotechnologyGraduate School of AgricultureKyoto UniversityUji Kyoto 611-0011, Japan

Shinichi HondaFrontier Biochemical and Medical Research

Laboratories, Kaneka Corporation,1-8 Miyamae-machi, Takasago-cho,Takasago-shi, Hyogo 676-8688, Japan

Masashi HosokawaFaculty of Fisheries SciencesHokkaido UniversityHakodate 041-8611, Japan

De-Xing HouDepartment of Biochemical Science

and TechnologyFaculty of AgricultureKagoshima University, Korimoto 1-21-24Kagoshima City, 890-0065 Japan

Yusaku IwasakiGraduate School of Nutritional and Environmental

Sciences and Global COE ProgramUniversity of ShizuokaShizuoka 422-8526, Japan

ix


x Contributors

Li Li JiThe Biodynamics Laboratory and the

InterdisciplinaryGraduate Program of Nutritional ScienceUniversity of Wisconsin-MadisonWI 53706, USA

Lekh R. JunejaTaiyo Kagaku Co. LtdResearch & Development,1-3 Takaramachi, YokkaichiMie 510-0844, Japan

Mahendra P. KapoorTaiyo Kagaku Co. LtdResearch & Development,1-3 Takaramachi, YokkaichiMie 510-0844, Japan

Jim KaputDivision of Personalized Nutrition and MedicineFDA/National Center for Toxicological Research3900 NCTR RoadJefferson, AR 72079

Teruo KawadaLaboratory of Molecular Function of FoodDivision of Food Science and BiotechnologyGraduate School of AgricultureKyoto UniversityUji Kyoto 611-0011, Japan

Connie J. KimDepartment of Food ScienceUniversity of Guelph50 Stone Road EastGuelph, OntarioN1G 2W1, Canada

Kazunori KobaDepartment of Nutritional ScienceUniversity of Nagasaki, SieboldNagasaki, 851-2195, Japan

Kenji KobataGraduate School of Nutritional and Environmental


Ryan KoenigThe Biodynamics Laboratory andUniversity of WisconsinMadison, WI 53706, USA

Martin KussmannGroup Leader Functional GenomicsDepartment of BioAnalytical SciencesNestle Research CenterVers-chez-les-BlancCH-1000 Lausanne 26Switzerland

Yiwei LiDepartment of PathologyKarmanos Cancer InstituteWayne State University School of MedicineDetroit, MI, USA

Hiroshi ManoFaculty of Pharmaceutical ScienceDepartment of Clinical Dietetics and Human

NutritionJosai University1-1 Keyakidai, SakadoSaitama 350-0295, Japan

Hitoshi MatsumotoFood & Health R&D LaboratoriesMeiji Seika Kaisha, Ltd.5-3-1 Chiyoda SakadoSaitama 350-0289, Japan

Yoshinori MineDepartment of Food ScienceUniversity of Guelph50 Stone Road EastGuelph, ON N1G 2W1, Canada

Kazuo MiyashitaFaculty of Fisheries SciencesHokkaido UniversityHakodate 041-8611, Japan

Akihito MoritaGraduate School of Nutritional and Environmental


Akira MurakamiDivision of Food Science and BiotechnologyGraduate School of AgricultureKyoto UniversityKyoto 606-8502, Japan


Contributors xi

Baitang NingDivision of Personalized Nutrition and MedicineFDA/National Center for Toxicological Research3900 NCTR RoadJefferson, AR 72079

Hajime OhigashiFaculty of Biotechnology, FukuiPrefectural University4-1-1, Konejojima, MatsuokaEiheiji-Town, Fukui, 910-1195, Japan

Jean-Paul PegorierDepartement d’Endocrinologie, Metabolisme

et CancerInstitut Cochin, IFR Alfred JOSTINSERM U567, CNRS UMR8104, Universite

Rene Descartes24 rue du Faubourg Saint Jacques75014 Paris, France

Frederic RaymondFunctional Genomics GroupDepartment of BioAnalytical SciencesNestle Research CenterVers-chez-les-BlancCH-1000 Lausanne 26Switzerland

Gregor ReidCanadian Research & Development Centre for

ProbioticsLawson Health Research Institute268 Grosvenor Street, London, Ontario,Canada, N6A 4V2

Molay K. RoyTaiyo Kagaku Co. LtdResearch & Development,1-3 Takaramachi, YokkaichiMie 510-0844, Japan

Fazlul H. SarkarDepartment of PathologyKarmanos Cancer InstituteWayne State University School of MedicineDetroit, MI, USA

Fereidoon ShahidiDepartment of BiochemistryMemorial University of NewfoundlandSt. John’s, NL A1B 3X9, Canada

Jun ShimizuFaculty of Pharmaceutical ScienceDepartment of Clinical Dietetics and Human


Nobuyuki TakahashiLaboratory of Molecular Function of FoodDivision of Food Science and BiotechnologyGraduate School of Agriculture, Kyoto UniversityUji Kyoto 611-0011, Japan

Feng TaoPressure BioSciences, Inc.14 Norfolk AvenueSouth Easton, MA 02375

Yuji TominagaFrontier Biochemical and Medical Research

LaboratoriesKaneka Corporation1-8 Miyamae-machi, Takasago-choTakasago-shi, Hyogo 676-8688, Japan

Takanori TsudaCollege of Bioscience and BiotechnologyChubu University, 1200 Matsumoto-cho, KasugaiAichi 487-8501, Japan

Masahiro WadaFaculty of Pharmaceutical ScienceDepartment of Clinical Dietetics and Human


Edwin WangBiotechnology Research InstituteNational Research Council of Canada6100 Royalmount AvenueMontreal, Quebec, CanadaandCenter for BioinformaticsMcGill UniversityMontreal, Quebec, Canada

Tatsuo WatanabeGraduate School of Nutritional and Environmental

Sciences and Global COE ProgramUniversity of Shizuoka3Shizuoka 422-8526, Japan


xii Contributors

Mitchell L. WiseUnited States Department of Agriculture Cereal CropResearch LaboratoryMadison, WI 53706, USA

Teruyoshi YanagitaDepartment of Applied Biochemistry and Food

ScienceSaga UniversitySaga, 840-8502, Japan

Shinichi YokotaFrontier Biochemical and Medical Research

LaboratoriesKaneka Corporation1-8 Miyamae-machi, Takasago-choTakasago-shi, Hyogo 676-8688, Japan

Denise YoungDepartment of Food ScienceUniversity of GuelphGuelph, Ontario N1G2W1, Canada

Rina YuLaboratory of Clinical Nutrition & ImmunologyDepartment of Food Science and NutritionUniversity of UlsanUlsan 680-749, South Korea


Section I

Introduction

1


2


1Nutrigenomics and Proteomics inHealth and Disease: An Overview

Yoshinori Mine, Kazuo Miyashita, and Fereidoon Shahidi

Association between diet and chronic diseases haslong been recognized through epidemiological stud-ies. Modern molecular nutrition focuses on healthpromotion, disease risk reduction, and performanceimprovement through diet and lifestyle considera-tions (Kussmann and Blum 2007; Ronteltap et al.2008). New genomic, proteomic, and metabolomictechniques are now enabling us to find out moreabout the basis of these associations through exami-nation of the functional interactions of food with thegenome at the molecular, cellular, and systemic levels(Corthesy-Theulaz et al. 2005; Kato 2008; Mariman2006). The human genome is estimated to encodeover 30,000 genes and to be responsible for generat-ing more than 100,000 functionally distinct proteins.While traditional nutrition research has dealt withproviding nutrients to nourish populations, nowa-days it focuses on improving health of individualsthrough diet. Modern nutritional research is aimingat health promotion and disease risk reduction andon performance improvement (Trujillo et al. 2006).Nutrigenetics questions as to how individual geneticdisposition, manifesting as single nucleotide poly-morphisms, copy-number polymorphisms, and epi-genetic phenomena, affects susceptibility to diet. Nu-trigenomics addresses the inverse relationship; thatis, how diet influences gene transcription, proteinexpression, and metabolism. Metabolomics is a di-agnostic tool for metabolic classification of individu-als. A major methodological challenge and first pre-requisite of nutrigenomics is integrating genomics(gene analysis), transcriptomics (gene expressionanalysis), proteomics (protein expression analysis),and metabolomics (metabolite profiling) to define a“healthy” phenotype (Kussmann et al. 2006; Milner2004). The long-term deliverable of nutrigenomicsis personalized nutrition for maintenance of indi-

vidual health and prevention of disease (Fay et al.2008; Kaput 2008; Ronteltap et al. 2008). “Nutrige-nomics” may offer a new approach for understand-ing the beneficial effects of dietary compounds onthe development of severe polygenic diseases, suchas cardiovascular disease, diabetes, and hypertension(Keusch 2006).

This book aims to compile current science-basednutrigenomics and proteomics in food and health.The book comprises four sections: (I) Introduction,(II) Genomics and Proteomics in Health and Dis-eases, (III) Food Factors–Gene Interactions, and (IV)Advanced Analytical Techniques for Nutrigenomicsand Proteomics.

Chapter 1 summarizes aims and scope as well asoverall highlights of this book. Chapter 2 consists ofintroductory omics in nutrition and health research.Nutrigenomics contains the three omics disciplinesgene, protein, and metabolite profiling (transcrip-tomics, proteomics, and metabolomics) as appliedto the field of nutrition and health. Furthermore, nu-trigenomics forms the scientific basis for develop-ing nutrition adapted to the specific needs of (ratherlarge) consumer groups, be they healthy, at risk, ordiseased. The three omics platforms are introducedin this chapter that also describes their application innutritional research. Microarray-based gene expres-sion analysis is the most mature genome-wide profil-ing platform. Consequently, transcriptomics in nutri-tional studies is widely applied when it comes to basicand preclinical research in either cell culture systemsor animal models. Proteomics has evolved as an ana-log to genomics, from identifying all proteins presentin a given sample at a given time to a global molecu-lar analysis platform addressing functional aspects ofbiological systems. Comparing such variations in theproteome enables the discovery of key proteins and

3


4 Section I: Introduction

the identification of modulated pathways involved,for example, in specific nutrition-related processes.Over the last two decades, proteomics has devel-oped into an established technology for biomarkerdiscovery, clinical applications, disease profiling anddiagnostics, and the study of protein interactions andof the dynamics of signaling pathways. Metabolitesrepresent the endpoints of metabolism and can pro-vide information on the molecular events associatedwith the adaptations of the body to increased ordecreased fluxes of nutrients through metabolic path-ways. Metabolomics in nutrition addresses the chal-lenge of characterizing food-related metabolic mod-ulations. Moreover, individual metabolites such ascholesterol, glucose, and homocysteine are consid-ered as markers for health or disease status. Nutrige-nomics and nutrigenetics are key science platformsto promote health and prevent disease through nu-trition that better meets the requirements and con-straints of consumer groups with specific health con-ditions and particular lifestyles. Section II comprisesthree chapters on the impact of nutrigenomic and pro-teomic interventions on health and diseases. Chapter3 deals with personalized nutrition and medicine. Theconcept of personalizing nutrition and medicine—and therefore healthcare—emerged from the humangenome and haplotype projects. The results of theselarge-scale, international initiatives offered the hopethat nutrition and medicine could be tailored to theindividual. The significant advances in understand-ing complex biological processes relied on reduc-tionistic approaches: hold all variables but one con-stant. While this strategy was successful for certainmonogenic phenotypes, understanding complex sys-tems requires analytical approaches that incorporaterather than avoid complexity. The key challenge forpersonalizing healthcare then is not the complex-ity of the data sets, but acquiring those data sets ina manner to reduce noise and increase true signals.This might best be accomplished by preselecting phe-notypes based on quantitative data, or alternatively,preselecting genotypes that maximize differences inallele frequencies of candidate genes involved in nu-trient metabolism or other physiological traits. Theintegrative whole system analyses of the data sets andnew visualization methods such as shown with net-work analysis tools provide a path not only to performthese complex experiments, but also to develop bio-logical insight into the outcomes. The developmentof nutrigenomics and genetics and the applicationof this knowledge will provide strategies for main-taining health and improving medical treatment ofchronic diseases.

Chapters 4 and 5 discuss obesity and nuclear re-ceptors and inflammatory genes involved in obesity-

induced inflammatory responses and pathologies.Obesity is the state of excessive formation of adi-pose tissues. Recent research has clarified the dif-ferentiation of adipocytes, the level of subsequentfat accumulation, and the secretion of the biologi-cally active adipocytokines by adipocytes). In par-ticular, it has been clarified that adipocytokines se-creted by adipocytes play a significant role in thepathogenesis of diseases such as diabetes and car-diovascular diseases and are closely associated withthe pathogenesis and exacerbation of ailments arisingfrom obesity. This chapter discusses obesity and themetabolic syndrome and then describes the nuclearreceptors that are most important in adipocyte differ-entiation and the mechanism underlying the expres-sion of function of adipocytes affecting obesity fromthe viewpoint of nutrigenomics. Obesity is also alow-grade systemic chronic inflammatory condition,characterized by abnormal cytokine production, in-creased acute phase proteins, and other inflammatorymediators. Obesity-induced inflammation consists ofa set of inflammatory immune components and in-flammatory signaling pathways similar to those in-volved in classical inflammation, such as inflamma-tory cells like macrophages, inflammatory mediatorslike cytokines and chemokines, as well as inflamma-tory signaling molecules. Obesity-induced inflam-mation is considered to serve as the potential mech-anism linking obesity to obesity-related pathologiessuch as insulin resistance, type 2 diabetes, fatty liverdisease, atherosclerosis, some immune disorders,and several types of cancer. Chapter 5 specificallyfocuses on obesity-induced inflammatory compo-nents, linking to obesity-related pathologies. Adiposetissue-derived inflammatory genes/proteins such asadipocytokines and signaling molecules and the in-flammatory cross-talk within adipose tissue cellsthrough adipocytokine. Allergies affect almost 20%of the population in the developed world and allergiescan be life-threatening. Individuals may be allergicto a variety of natural or synthetic molecules, suchas foods, drugs, chemicals, dust, pollen, and metals.Genomic and proteomic methods are powerful tech-niques for the identification, characterization, andin vitro diagnosis of allergies. Chapter 6 describesmolecular mechanisms of allergy and gene interac-tions and susceptibility to allergic responses. It alsoreports on recent therapeutic approaches for allergiesusing recombinant DNA techniques.

Section II includes various food factors–gene in-teractions and their impact in health and diseases.This section consists of 16 chapters that cover lipids,proteins/peptides/amino acids, cartenoids, phyto-chemicals, and probiotics. Chapters 7 and 8 dealwith the beneficial effects of conjugated linoleic


Chapter 1: Nutrigenomics and Proteomics in Health and Disease: An Overview 5

acid (CLA) and regulation of gene transcription byfatty acids. CLA has been shown to exert variousphysiological functions, other than antimutagenicity,such as anticarcinogenic and antiobesity (reductionof body fat mass) activities, prevention of atheroscle-rosis, enhancement of immune function, andsuppression of blood pressure, despite the fact thatphysiological properties of CLA are still limited. Thephysiological effects of CLA are also described alongwith potential health benefits of conjugated linolenicacid. Dietary fat is an important macronutrient re-quired for the growth and development of all organ-isms. Excessive levels of dietary fat or imbalancein its composition (saturated versus unsaturated fat)have been related to the onset or development ofseveral chronic diseases such as coronary artery dis-ease, obesity, and type 2 diabetes as well as certaintypes of cancer. The biological functions of lipidsare mainly carried out by fatty acids and/or derivedsignaling molecules such as ceramides, diacylglyc-erols, eicosanoids, and coenzyme A thioesters (acyl-CoA). The last two decades have provided evidencethat major (glucose, fatty acids, amino acids) or mi-nor (iron, vitamin, etc.) dietary constituents regu-late gene expression in a hormone-independent man-ner. The molecular mechanisms by which fatty acidsand/or their metabolites control the transcription ofgenes involved in their own metabolism or in carbo-hydrate metabolism are also described. These effectsare mediated either by direct binding on transcrip-tion factors such as PPARs, LXR, HNF-4, RXR, etc.(each belonging to the nuclear receptor superfam-ily) or alternatively through modifications in nuclearabundance and/or activity of numerous transcriptionfactors such as SREBP-1c, ChREBP, and NF-κB.

Chapter 9 focuses on amino acid biological func-tions as nonnutrient. Although amino acids arewidely known as the building blocks of proteins, theirfunctions in living organisms are vast as they can in-teract with the endocrine, neuronal, and immune sys-tems to influence the balance between health and dis-ease. These systems, particularly in diseased states,affect the amino acid availability and may inducepathways to alter protein synthesis. The underlyingmechanism of the regulation of the biological func-tions is partially due to amino acid control of geneexpression. This chapter reviews the importance ofamino acid balance and the consequences of aminoacid imbalance at the genetic level. Health and dis-ease implications through amino acid deficiency andsupplementation was explored. Many amino acidstudies have reported health benefits during diseasedstates, such as cancer, inflammatory disorders, dia-betes, gastrointestinal disorders, and muscular wast-ing diseases. Understanding the mechanism of amino

acid control of genes, both singly and in unison,may provide its involvement in disease progressionand prevention. Many researchers have reported thatfood proteins and their peptides express a variety offunctions in the body, including a reduction of bloodpressure, antimicrobial activity, antioxidative, anti-inflammatory, antisatiety, anticancer, antiobesity, an-tiallergy, modulation of immune cell functions, andregulation of nerve functions. Bioactive peptides arepeptide sequences present in the intact protein thatunder normal circumstances do not have biologicalproperties, but when they are released as peptidesin vitro or in vivo, they exert biological activities.There is increasing commercial interest in the pro-duction of bioactive peptides from various sourcessuch as egg, milk, cereal, and fish proteins. Chap-ter 10 summarizes recent advances of food-derivedbioactive peptides–gene interactions and their mech-anisms of actions. Although their properties andphysiological effects have not been completely ex-plored, bioactive peptides can broadly be divided intotwo categories: (1) peptides that exert their effectsby direct physical interaction with another molecule,and (2) peptides that interfere with gene expression.Bioactive peptides that alter gene expression can doso by (1) epigenetic modification of the proteins thatattach to the DNA, (2) alteration of the cell’s primarysignaling ligand to indirectly influence transcriptionfactor activity, and (3) interference with cell signal-ing and gene expression via direct binding of peptideligand to receptor. Understanding the behavior of di-etary proteins and peptides in the intestine is alsoimportant for designing functional foods with phys-iological functions.

Carotenoids represent a large group of isoprenoidstructures with many different structural characteris-tics and biological activities. To date, a wide range ofcarotenoids have been isolated, identified, and quan-tified from the extracts of fruits and vegetables com-monly consumed in the world. The best known bi-ological function of carotenoids is their establishedrole as pro-vitamin A. Chapter 11 describes the nu-trigenomic study on the anti-obesity effect of alleniccarotenoids from seaweeds and vegetables, with spe-cial reference to their regulations on relative geneand protein expressions. Fucoxanthin and neoxan-thin are the major carotenoids present in chloroplastsof brown seaweeds and higher plants, respectively.Fucoxanthin is the most abundant of all carotenoids,accounting for >10% of the estimated total naturalproduction of carotenoids. The key for success of fu-coxanthin will be induction of uncoupling protein 1in white adipose tissue (WAT) and downregulationof adipokines such as TNFα. The regulatory effectof fucoxanthin on PPARγ and γ3-AR in WAT is



correlated with its antiobesity and antidiabeticeffects. Furthermore, the relationship betweencarotenoid structure and suppressive effect on thedifferentiation of 3T3-L1 adipose cells shows thatcarotenoids containing an allene bond and an addi-tional hydroxyl substituent on the side group mayshow the characteristic antiobesity activity.

Chapter 12 deals with the control of systemicinflammation and chronic diseases by the useof turmeric and curcumenoids. Numerous plant-derived, but also microbially derived, substances,often referred to as chemopreventive agents, havedocumented anti-inflammatory effects and are be-lieved to reduce the rate of aging and prevent degen-erative malfunctions of organs and also developmentof acute and chronic diseases. Among these are vari-ous curcumenoids, the active ingredients in turmericand curry-containing foods, and thousands more ofhitherto little or totally unexplored substances. Thischapter focuses on documented experimental andclinical effects of supplementation of turmeric, vari-ous curcumenoids, and pure curcumin. The Food andDrug Administration (FDA) has approved a healthclaim for soy-based food products for health benefitsprimarily based on epidemiological data indicatingthat high soy consumption is associated with a lowerrisk of cardiovascular diseases. Soy isoflavones alsoshow a beneficial role in obesity, diabetes, coronaryartery disease, and osteoporosis in postmenopausalwomen. Soy isoflavones have been shown to inhibitcarcinogenesis and cancer cell growth in vivo andin vitro. It has also been found that soy isoflavoneslower total cholesterol and low-density lipoproteincholesterol, suggesting the effect of isoflavones oncardiovascular disease risk reduction. Chapter 13presents gene expression and proteomic profiling bysoy isoflavones. It has been found that soy isoflavonesregulate the expression of genes that are related to es-trogen regulation, organ differentiation, and fat andbone metabolism in normal cells. Soy isoflavonesalso inhibit the growth of cancer cells through themodulation of genes, which control cell proliferation,cell cycle, apoptosis, oncogenesis, transcription reg-ulation, and cell signal transduction system. In thischapter, current evidence on the molecular effects ofsoy isoflavones as documented by nutrigenomic andnutriproteomic research is provided.

Over the last two decades more than five thou-sand peer-reviewed articles and tens of thousands ofnews articles have provided evidence for enhancedhealth benefits of tea consumption. At present, mul-tiple evidences have proven the involvement of teabeverages in health promotion that are directly linkedto its polyphenol content. Green tea has firmly es-

tablished its powerful strength in reducing oxidativestress, suppressing cancer-related risks, cardiovascu-lar disease, neuronal damage, and hepatic disorders,among others. Epidemiological and clinical studieshave also proven that individuals consuming tea ormany form of tea polyphenols benefit from a lower in-cidence of cancers and other lifestyle-related diseasessuch as diabetes, obesity, and cardiovascular disease,among others. However, the question that is duly con-tinued to be answered is how green tea polyphenolsexert their health beneficial effect? Chapter 14 ex-plores how green tea polyphenols modulate genomefunctions for protective health benefits. Is it a simplesite-specific activity or alteration of a pathway thatultimately lead to altered activity of one or more sec-ondary molecules required to maintain normal cellfunction or enhancing the meaningful roles of themolecules to maintain cell machinery systems? Thischapter reviews how green tea polyphenols modu-late genome function, gene repair, protecting genes,and exerting the roles considered auspicious, thateven remained unknown until a decade ago. Thischapter also lists the latest evidences in accordancewith the enhanced philological functions. Reactiveoxygen species are generated ubiquitously in aerobicorganisms. When these cytotoxic agents overwhelmendogenous antioxidant defense systems, serious ox-idative stress and damage occur as reflected by theoxidative modification of macromolecules such aslipid, protein, and DNA. Thus, it is critical that cellsmaintain optimal antioxidant defenses in order to re-duce oxidative damage. Dietary supplementation andtherapeutic use of antioxidants are emerging mea-sures to prevent and treat oxidative stress-induceddiseases. Chapter 15 describes oat avenanthramidesas novel antioxidants. Oat (Avena sativa), althoughconsumed in considerably lower quantities world-wide than wheat and rice, has a highly edible qualityand contains high nutritional value compared to otherminor grains. Over the past decade, interest of restor-ing oat as a natural antioxidant additive in food hasbeen on the rise. Other than tocopherols, tocotrienols,and flavonoids, oat contains a unique group of ap-proximately 40 different types of polyphenolic com-pounds called avenanthramides (AVA) that consist ofan anthranilic acid derivative and a hydroxycinnamicacid derivative linked by an amide bond similar tothose found in peptides. There is strong evidencethat AVA are potent inhibitors of cell proliferationand inflammatory processes, especially in the en-dothelial cells and smooth muscle cells of blood ves-sels. These effects have been shown to be mediatedby its inhibition of proinflammatory cytokine pro-duction and signaling. AVA have also been reported



to modulate endogenous antioxidant defense such asincreasing plasma glutathione level and upregulatingtissue superoxide dismutase activity, the mechanismsof which remain to be elucidated. Chapter 16 reviewscancer-preventive effects and molecular actions ofanthocyanins. Anthocyanins are naturally occurringpolyphenolic compounds that confer an intense colorto many fruits and vegetables. A few population-based investigations have highlighted the potency ofanthocyanins or anthocyanin-containing mixtures oncancer prevention or cancer risk reduction. Studies onanimal models have revealed that high intake of an-thocyanins or anthocyanin-containing mixtures pro-tects against tumorigenesis of colon, skin, and mam-mary glands. Extensive studies in cancer cell lineshave shown the inhibitory effects of anthocyaninsor anthocyanin-containing mixtures on the growthof cancer cells derived from malignant human tis-sues including vulva, stomach, colon, lung, breast,leukemia, uterus, mouth, and prostate. Recent molec-ular data have demonstrated that anthocyanins couldmodulate oncogenic cellular signaling transductionpathways (MAPK and EGFR), transcriptional fac-tor activations (AP-1, NF-κB, p53), and downstreamgene expressions (COX-2, iNOS, Bax). These molec-ular actions are involved in the processes of cell trans-formation, inflammation, and apoptosis, which pro-vide molecular basis for the cancer-preventive effectsof anthocyanins.

Chapter 17 deals with how food components ac-tivate capsaicin receptor, transient receptor poten-tial vanilloid subtype 1 (TRPV1). Capsaicin is apungent principle of hot pepper. Capsaicin exertsseveral biological activities such as causing burn-ing sensation, stimulating primary afferent neuronsconducting chemical pain or hotness, enhancing en-ergy metabolism, showing protection against stom-ach mucosa, inducing apoptosis in some cancer cells,and so on. Many of them are exerted through cap-saicin receptor activation. Because obesity is oneof the serious factors on lifestyle-related diseasessuch as hypertension, stroke, diabetes, and hyper-lipemia, this chapter focuses on the thermogenicaction or body fat lowering effect of capsaicin. Ther-mogenic action of capsaicin is thought to be exhib-ited through activation of TRPV1. From the dis-covery of TRPV1 gene in 1997, food componentsactivating TRPV1 have been vigorously investigated.There are lists of capsaicinoids of hot pepper, piper-ine of black pepper, eugenol of clove, ginsenosidesof Asian ginseng, and evodiamine of Evodia rutae-carpa, among others. Capsiate inhibits accumula-tion of body fat in humans. Anthocyanins are thelargest group of water-soluble pigments in the plant

kingdom. In the human diet, they are derived pri-marily from a wide variety of plant sources includ-ing crops, beans, fruits, vegetables, and red wine,and their effects are also diverse and important tohealth promotion. Chapter 18 focuses on blackcur-rant (Ribes nigrum L.) anthocyanins because black-currant is rich in it and blackcurrant is consumedin many countries. This chapter provides a reviewof the newly discovered effects of anthocyanins in-cluding their antiobesity effect, antidiabetes effect,and vision improvement. Chapter 19 describes var-ious biological activities of licorice. Licorice, theroot of the leguminous Glycyrrhiza plant species,is one of the most useful and popular plants in bothAsia and Europe, and the history of its consump-tion as a traditional medicine and food goes back toover 4,000 years to the era of ancient Mesopotamiaand Egypt. Licorice contains triterpenes and pheno-lic constituents such as glycyrrhizin, a well-knowntypical active constituent of licorice, and the species-specific constituents glabridin, glycycoumarin, andlicochalcone A in G. glabra, G. uralensis, and G.inflate, respectively. In G. glabra, the species spe-cific compound is glabridin. Various studies haveshown the biological effects of glabridin, licorice,or its extracts. These include antioxidative, estrogen-like, anti-inflammatory and anti-Helicobacter pyloriactivities. Hydrophobic flavonoids from G. glabraare extracted and concentrated, and the resulting ex-tract is referred to as licorice flavonoid oil (LFO).DNA microarray analysis suggests that the antiobe-sity effects of LFO are attributable to suppressed fattyacid synthesis and activated fatty acid catabolismin the liver. LFO has also received FDA approvalas a new dietary ingredient in the United States in2006. Therefore, further studies that elucidate themechanism of LFO containing licorice hydrophobicflavonoids would contribute to the efficient applica-tion of LFO in the treatment of metabolic syndrome.Isopentyl diphosphate and its isomer dimethylallyldiphosphate are the universal five-carbon precursorsof isoprenoids. Isoprenoids are contained in manyherbal plants, and several isoprenoids have beenshown to be available for pharmaceuticals, for ex-ample, artemisinin and taxol as malaria and can-cer medicines, respectively. Various isoprenoids arecontained in many plants not only for herbal usebut also for dietary consumption. Chapter 20 reportson several bioactive isoprenoids, contained in herbalor dietary plants, which have possibilities to ame-liorate metabolic disorders ia activation of ligand-dependent transcription factors, that is, nuclear re-ceptors. Chapter 21 reviews anti-inflammatory andanticarcinogenic potential of citrus coumarins and



polymethylated flavonoids. Citrus fruits are wellknown to contain an array of secondary metabolitesin terms of their chemical structures and biologi-cal activities, which biosynthesize monoterpenes (d-limonene, etc.), triterpenes (limonoids), flavonoids(nobiletin, hesperidin, etc.), coumarins (auraptene,bergamottin, etc.), and carotenoids (β-carotene, β-cryptoxanthin, etc.). Ample evidence obtained fromin vitro and in vivo experiments as well as epidemi-ological surveys indicates that frequent intake of cit-rus fruits is beneficial to human health. These citruscompounds are hydrophobic and thus tend to local-ize in gastrointestinal mucosa in rodents as com-pared to general polyphenols present. Thus, abun-dant data have revealed both auraptene and nobiletinto be highly promising citrus components with anti-inflammatory and anticancer activities, with notableaction mechanisms and effects on metabolism. Oneof the distinct characteristics of citrus fruits, as com-pared with other foods, is the variety of active con-stituents in terms of chemical characteristics andbioactivities. Thus, combination studies using differ-ent types of citrus components for enhancing eachefficacy are warranted, such as combining nobiletin(targeting COX-2 transcription) and auraptene (tar-geting COX-2 translation) to determine their additiveor even synergistic effects.

Food and Agricultural Organization of the UnitedNations and the World Health Organization defineprobiotics as “live microorganisms which when ad-ministered in adequate amounts confer a health bene-fit on the host.” The majority of probiotics are strainsof lactobacilli or bifidobacteria and they are admin-istered in food products such as yogurt, milk drinks,and cheese, as well as capsules and tablets. The effectthat beneficial microbes have on health maintenanceis becoming more and more recognized, given the re-alization that so many organisms reside in the humanbody. The reintroduction of beneficial organisms(probiotics) to the host has mostly been via food anddietary supplement products, and thus relevant to nu-trigenomics. Chapter 22 discusses some examples ofhow probiotic microbes and their proteinaceous andother by-products contribute to health. As more hu-man and microbial genomic information emerges, itwill become clearer under what conditions probioticorganisms interface with the host in an optimal way.

Section IV highlights recent advances in analyticaltechniques for nutrigenomic and proteomic researchin food and health. Chapter 23 describes microarrayas a powerful tool for studying the functions of foodand its nutrients. Microarray is a high-throughputgenomic tool. It can be used for profiling and mon-itoring the expression levels of tens and thousands

of genes (entire genomes). It can also be used to de-termine the influence of food nutrients and/or bioac-tive compounds (food factors) on metabolic path-ways and to understand how food nutrients andfactors maintain homeostatic control of gene ex-pression levels. Microarray technology is a “nutrige-nomics” tool and can be used to investigate the lev-els of transcripts in particular. Typically, food isa complex and variable mixture of nutrients andother components. Most food factors are weak di-etary signals and must be considered in the con-text of chronic exposure. Microarray analysis clearlyindicates the effects exerted by food factors andnutrients on metabolic pathways via transcriptomemodifications. Moreover, the results of microarrayanalysis suggest that food factors and nutrients in-fluence the metabolome because alterations in thetranscriptome cause changes in the metabolome.Therefore, microarray analysis is one of the mostconvenient tools for inferring the proteome andmetabolome. This technology will enhance under-standing of the manner in which food and nutri-tion influence metabolic pathways and how thesefactors maintain homeostasis under normal condi-tions or diet-related or non-diet-related disease con-ditions. Chapter 24 highlights challenges and currentsolutions in proteomic sample preparation. Pro-teomics is a discipline of relatively short history, but itholds great promise in elucidating biochemical infor-mation via quantitative determinations of the wholecollection or representative proteins. One of the com-mon objectives in proteomic studies is the discov-ery of biomarkers. Although biological systems areextremely complex, and the technology challengesare still many, hundreds and thousands of biomarkercandidates are being discovered with advancementsmade in proteomic technologies. One of the majorhurdles in proteomics is the identification of truebiomarkers via analytical and clinical validation stud-ies. This chapter reviews some critical aspects ofbiomarker determination using proteomic methodsand some examples of new developments in the pro-teomic sample preparation techniques, particularlythe “pressure cycling technology.” In the past fewyears, many high-throughput techniques have beendeveloped and applied in biological studies. Thesetechniques such as “next generation” genome se-quencing, chip-on-chip, and microarray, among oth-ers, can be used to measure gene expression andgene regulatory elements in a genome-wide scale.Moreover, as these technologies become more af-fordable and accessible, they have become a drivingforce in modern biology. Traditionally, biologists de-scribed these relationships between a limited number



of genes or proteins using a descriptive language.With the huge amount of data produced by high-throughput techniques, biologists have to deal withthousands of biological relations in a single exper-iment. In this situation, the traditionally descriptiveways for biological relations are not sufficient to dealwith the huge number of relations under study. Theonly way to deal with a large amount of relationsis through mathematical representations and compu-tations by researchers in biological sciences. Chap-ter 25 first introduces basic computational conceptsand then illustrates the procedures and computationaltechniques for high-throughput data analysis, usingexamples from cancer research. Proteomics is cen-tral to nutrigenomics and has the potential to ex-plain many of the physiological changes associatedwith nutritional stimuli. In proteomics, all proteinsexpressed in a cell or tissue are analyzed to iden-tify the presence or absence of some key proteinsthat provide information about the early stages ofdisease or different conditions. However, a compre-hensive analysis of peptides and small proteins of abiological system corresponding to the respective ge-nomic information was missing in proteomics. Chap-ter 26 introduces the concept of peptidomics. Theterm peptidomics was first introduced as a subset ofproteomics for the description of peptides as geneproducts in February 2000 at the ABRF conference“From Singular to Global Analysis of Biological Sys-tems.” This was coined as a short version of “pep-tide proteomics” and was defined as the technologyfor comprehensive qualitative and quantitative de-scription of peptides in a biological sample. Studiesof peptidomics cover peptides with low-molecular-weight and small proteins (0.5–15 kDa), since pep-tides among the families of hormones, cytokines, andgrowth factors play a central role in many physiologi-cal processes. In addition, application of peptidomicsknowledge to the nutrient effect may yield potentialinformation about the diet-induced peptide changesand may act as good biomarkers. However, the fieldof peptidomics is relatively new and has potential toprogress in future with the advent of high-throughputmass spectrometry-based technologies coupled withbioinformatics and genomic databases.

Completion of human genome project coupledwith the advancement in “omic” technologies en-abling researchers to analyze the complex interplayof metabolism, gene expression, and function, andmore broadly, genetic diversity within and betweenhuman populations. Nutrition science has broadenedto the new discipline of nutrigenomics, which allowsan in-depth understanding of metabolism, health, andpathophysiology of disease that ultimately could be

used to prevent or treat diseases. The major goal ofthis book is to comprehensively understand the re-sponse of the body’s genes to diets and food fac-tors through various omics technologies such astranscriptomics, proteomics, and metabolomics. Thiswill contribute to the development of new preven-tive and therapeutic strategies for both pharmacolog-ical and nutritional interventions (Bauer et al. 2004;Mariman 2006; Milner 2007).

The editors have succeeded in bringing togethermany renowned international experts in nutrige-nomics and proteomics in health and diseases. Weare grateful to all the authors for their state-of-the-artcompilation of recent rapid development in this field.We believe that this book certainly deserve a broadreadership in the disciplines of nutrition, pharma-cology, nutraceutical/functional foods, food science,biology, biochemistry, biotechnology, and lifescience. This book could also be used as a referencebook by senior undergraduate and graduate studentsas well as nutraceutical and pharmaceutical industry.

REFERENCES

Bauer M, Hamm A, and Pankratz MJ. 2004. Linkingnutrition to genomics. Biol Chem. 385(7):593–596.

Corthesy-Theulaz I, den Dunnen JT, Ferre P, GeurtsJM, Muller M, van Belzen N, and van Ommen B.2005. Nutrigenomics: The impact of biomicstechnology on nutrition research. Ann Nutr Metab.49(6):355–365.

Fay LB and German JB. 2008. Personalizing foods: Isgenotype necessary? Curr Opin Biotechnol. 19(2):121–128.

Kaput J. 2008. Nutrigenomics research forpersonalized nutrition and medicine. Curr OpinBiotechnol. 19(2):110–120.

Kato H. 2008. Nutrigenomics: The cutting edge andAsian perspectives. Asia Pac J Clin Nutr. 17(Suppl1):12–15.

Keusch GT. 2006. What do -omics mean for thescience and policy of the nutritional sciences? Am JClin Nutr. 83(2):520S–522S.

Kussmann M and Blum S. 2007. OMICS-derivedtargets for inflammatory gut disorders:Opportunities for the development of nutritionrelated biomarkers. Endocr Metab Immune DisordDrug Targets. 7(4):271–287.

Kussmann M, Raymond F, and Affolter M. 2006.OMICS-driven biomarker discovery in nutrition andhealth. J Biotechnol. 124(4):758–787.

Mariman EC. 2006. Nutrigenomics and nutrigenetics:The ‘omics’ revolution in nutritional science.Biotechnol Appl Biochem. 44(3):119–128.



Milner JA. 2004. Molecular targets for bioactive foodcomponents. J Nutr. 134(9):2492S–2498S.

Milner JA. 2007. Nutrition in the ‘omics’ era. ForumNutr. 60:1–24.

Ronteltap A, van Trijp JC, and Renes RJ. 2008.Consumer acceptance of nutrigenomics-based

personalised nutrition. Br J Nutr. 15:1–13.

Trujillo E, Davis C, and Milner J. 2006.Nutrigenomics, proteomics, metabolomics, and thepractice of dietetics. J Am Diet Assoc.106(3):403–413.


2Omics in Nutrition and Health Research

Michael Affolter, Frederic Raymond, and Martin Kussmann

INTRODUCTION

Nutrients and genomes interact. Nutrition is the mostimportant lifelong environmental impact on humanhealth status. While nutrigenetics addresses how anindividual’s genetic makeup predisposes for suscep-tibility for dietary intake, nutrigenomics rather askshow nutrition influences the expression of a givengenome.

Nutrigenomics contains the three omics disci-plines—gene, protein, and metabolite profiling(transcriptomics, proteomics, and metabolomics)—as applied to the field of nutrition and health.Together, they are a prerequisite for nutritionalsystems biology; that is, the understanding of thedynamic interaction between food components andthe entire diet with cells, organs, and the whole body.Nutrigenomics furthermore forms the scientific basisfor developing nutrition adapted to the specific needsof (rather large) consumer groups, be they healthy,at risk, or diseased. This chapter introduces the threeomics platforms and describes their applicationin nutritional research. We also discuss currentlimitations, recommend future developments, andhighlight the opportunities for omics integration andcorrelation with genetics in a nutritional context.

TRANSCRIPTOMICS INNUTRITION AND HEALTHRESEARCH

Microarray-based gene expression analysis is themost mature genome-wide profiling platform. Con-sequently, transcriptomics in nutritional studies iswidely applied when it comes to basic and preclini-cal research in either cell culture systems or animalmodels. The mRNA profiling bears the potential to

identify specific transcript changes as a response tothe administration of a nutrient or non-nutrient com-pound, or to a treatment or dietary intervention ina well-defined experimental setting. The observedchanges in mRNA level are not necessarily causalmarkers; they might rather represent a pattern ofexpressed transcripts that changes in a characteris-tic and reproducible way. Gene expression profilinghas the character of a screening process coveringthousands of potential indicators of the (changed)metabolic status and, therefore, it often also revealsunexpected findings.

MICROARRAY-BASED GENE EXPRESSION

PROFILING

Although microarrays are not the only available tech-nology for genome-wide gene expression profiling,it has established itself by far as the most widelydeployed in research. This is mainly due to a rangeof commercially available platforms and the mean-while high level of standardization. Today’s microar-ray platforms are based on either single long or mul-tiple short oligonucleotides as probes. They havedifferent manufacturing procedures and use differ-ent labeling methods. The arrays display probes thathybridize with high sensitivity and specificity withtheir counterparts from the sample. Commercial plat-forms such as those provided by Affymetrix (Baroneet al. 2001) and Agilent (Wolber et al. 2006) rely onin situ synthesis of the probes. Affymetrix oligonu-cleotide arrays consist of 25-mer probes, whereasthose of Agilent use longer 60-mer probes. Multi-ple short oligonucleotides per gene can better dis-criminate between related sequences, but the longerprobes provide better sensitivity. Illumina introducedin 2004 a new microarray technology for quantitative

11



gene expression profiling on the basis of randomlyassembled arrays of beads with each bead carryinga gene-specific probe sequence but multiple copiesof each sequence-specific bead in an array (Kuhnet al. 2004). This new platform seems to providean increased sensitivity compared to the other ma-jor commercial microarray platforms, with a largernumber of detected genes, and it only needs 25 ng ofRNA as starting material for analysis.

Depending on not only the platforms and tech-nologies but also on the RNA source, numerous prob-lems can arise in transcript profiling approaches. Onemore general challenge is the signal-to-noise ratio,which can be improved with multiple short probesper gene or the same long probe per gene presentin multiple copies on the same array. In order to im-prove interlaboratory comparability, standards for re-porting microarray data have been established underMIAME (minimum information about a microar-ray experiment) (Brazma et al. 2001). This stan-dard contains information required to consistentlydescribe microarray data so that the results de-rived from its analysis can be independentlyverified.

Compromised reproducibility of expression pro-files is a severe problem that has been meanwhileaddressed by a number of studies. However, dif-ferent conclusions were drawn ranging from goodconcordance of results across analysis platforms topoor comparability between platforms and laborato-ries. Recent comparisons of different array platforms(Barnes et al. 2005; Bosotti et al. 2007) revealedthat the signal concordance significantly improvedwith increasing amount of expressed transcript. Theconcordance was excellent when probes on differentplatforms could be identified as likely to target thesame set of transcripts of a given gene. It appearsnow that the main factors contributing to result vari-ability are the natural differences between biologicalsamples, rather than the techniques per se. However,variations in sample preparation have been observedwhen experiments are conducted by different opera-tors. Therefore, the use of automated systems shouldbe envisaged to reduce the “human factor” in techni-cal variability (Raymond et al. 2006).

The challenge of performing microarray studieshas today moved from data generation to analysisand interpretation. Analyses restricted to lists of sig-nificantly expressed genes with p values and foldchanges are insufficient to fully understand the un-derlying biology of metabolic adaptations. A singlehighly regulated gene does not necessarily have animportant biological meaning by itself. Therefore,understanding the biological meaning of the many

observed gene changes requires their assembly tomotifs of regulation. This can be achieved eithervia cluster analysis as a data-driven approach or byknowledge-based annotation analysis.

Cluster analysis uses statistical algorithms to or-ganize genes according to their similarity in patternof expression. The result is displayed graphically inan intuitive form for biologists and the highlightedpatterns can be interpreted as indications of the cellstatus. Moreover, coexpression of unknown geneswith well-characterized genes can bring suggestionson the functions of genes that are not well describedyet.

Annotation-based analysis refers to a set of struc-tured and precisely defined vocabularies, called on-tologies, used to characterize genes and gene prod-ucts. This kind of analysis aims at finding outhow genes are involved in different molecular func-tions, biological processes, and cellular componentsbased on the gene annotation. Thus, ontologiesare structured in a form that represents a networkof linked terms. Cluster analysis may miss subtlechanges, whereas annotation-based analysis appearsfrequently too restrictive. This is why the combina-tion of both analyses is often used to better interpreta gene expression data set.

A large number of tools are available to the sci-ence community for microarray data analysis. Clas-sical software for gene expression experiments usu-ally provides cluster and annotation-based analysis.A new generation of tools also offers metabolicand regulatory pathway analysis based on literature-derived information, enabling to interpret data in alarger context (Joyce and Palsson 2006; Khatri andDraghici 2005; Weniger et al. 2007; Yi et al. 2006).Despite the fact that the basic standards for report-ing microarray analyses are set under MIAME andthat a large variety of commercial and public do-main software tools for data interpretation are avail-able, there are still limitations in microarray dataanalysis.

A more specific problem in nutritional applica-tions is the often low signal-to-noise ratio: in con-trast to pharmaceutical or toxicological studies inwhich usually a limited number of target genesshow rather robust changes, in nutrition studies tran-script levels typically change more subtly, but thenumber of affected genes is often surprisingly highand this renders interpretation challenging. There-fore, a sound interpretation of the data needs in-dependent confirmation by assessing protein levelsby classical techniques or proteomics (see next partof this chapter) in combination with physiologicalreadouts.


Chapter 2: Omics in Nutrition and Health Research 13

MICROARRAY-BASED TRANSCRIPTOMICS IN

STUDIES ON HUMAN NUTRITION AND HEALTH

Applications of the different omics technologies ap-pear unlimited when utilizing cells in culture ormodel organisms, but they are constrained when itcomes to studies in humans. Expression profiling atthe mRNA level is restricted by the limited avail-ability of vital cells or tissues for analysis. Althoughtissue samples may be obtained via biopsies, espe-cially in nutrition research, these invasive techniquesare restricted in use and require ethical approval inevery study—in other words, it is very difficult toobtain a biopsy from a control sample.

Different types of blood cells or even whole bloodis therefore generally an interesting source of biolog-ical material in human transcriptomic studies. Bloodcells respond to dietary intervention, and more in-terestingly, they have different lifetimes, exhibit dif-ferent gene expression profiles, and can reach andoccupy different body compartments. In particular,peripheral blood mononuclear cells (PBMCs) aresampled for microarray-based identification of can-didate mRNA markers in human studies in responseto nutritional factors. However, attention should bepaid to the result interpretation because PBMCs com-prise different kinds of cells (B and T lymphocytesand monocytes), each showing a cell-type-specificgene expression signature. From a technical pointof view, care needs to be taken for sample storageand preparation, particularly when using peripheralblood cells for transcriptome analysis. It has beendemonstrated that sample handling and prolongedtransportation significantly alters gene expressionprofiles (Debey et al. 2004) and these procedureshave to be highly standardized for across-site com-parisons. More recently, also whole-blood RNA sam-ples are used for profiling purposes. These requirethe depletion of globin mRNA in order to detectlow-abundance transcripts. Various protocols haverecently been written to enhance sensitivity and qual-ity of mRNA detection from whole-blood samples(Field et al. 2007; Ovstebo et al. 2007) but, so far, thishas not yet been applied to human nutritional studies.

MICROARRAY-BASED TRANSCRIPTOMICS IN

HUMAN NUTRITIONAL INTERVENTION STUDIES

Whole-genome gene expression analysis is increas-ingly being deployed to assess the efficacy and safetyof food ingredients and to evaluate the molecular out-comes of dietary interventions.

Nutrients and genomes interact. Human geneticvariation influences nutrient bioavailability and bio-

efficacy. An individual’s genome predisposes the or-ganism with regard to the use of nutrients, and—viceversa—the nutrients can significantly alter the ex-pression of the genome. The next-generation tran-scriptomics technologies, that is, the sequencing-based gene expression analyses are promising meansto better determine the molecular mechanisms under-lying these interactions and their modification by ge-netic variation: these techniques enable both the anal-ysis of transcript abundance and its variation amongindividuals. Nutrigenetics and nutrigenomics are thedisciplines addressing these interactions and formthe scientific basis for the development of specificdiets that could prevent or delay disease and promotehealth and well-being. This is especially envisagedin chronic diseases because of the lifelong impact ofnutrition.

Consequently, a substantial part of these chronicdisease-related nutrigenomic/nutrigenetic studies fo-cused on cardiovascular disease, type 2 diabetes,or gastric disorders. While many of the early stud-ies had assumed that single nucleotide polymor-phisms (SNPs) were the main source of humangenetic variability, an increasing body of evidencesuggests the importance of additional layers ofvariability, including copy number polymorphisms(CNPs) and epigenetic regulation such as DNAmethylation. Many complex diseases like irritablebowel disease (Crohn’s disease and ulcerative coli-tis) have been shown to be related to SNPs on par-ticular chromosomal regions, but are also associatedwith copy number variation of certain other genes(McCarroll and Altshuler 2007; Shelling and Fergu-son 2007). Such discoveries suggest that a detaileddescription of the genetic background of complexdiseases is a challenging but necessary objective inorder to better prevent pathological development by,for example, adapted diets.

DNA methylation appears to provide a format forlong-term dietary imprinting of the genome (Water-land and Jirtle 2003). Some evidence shows thatchronic diseases present in adults are due to persis-tent adaptations to early-life nutrition. DNA methyla-tion would therefore be directly influenced by dietarymethyl supplementation, suggesting that nutritionalsupplementation may have unexpected adverse con-sequences on the gene regulation in human, and thatwell-adapted diets applied already at pre- and post-natal stage may exert a fundamental and long-lastingimpact.

Even cognitive development seems to be amenableto genetically counseled nutritional intervention.Studies have shown that nutrients, such as n-3 long-chain polyunsaturated fatty acids (n-3 LC-PUFAs),



can affect brain development and, therefore, the cog-nitive function. A recent publication showed that theassociation between breast-feeding and IQ is mod-erated by a genetic variant in a gene involved in thecontrol of fatty acid pathways (Caspi et al. 2007).

Some gene expression profiling approaches arealso used to understand the bioactivity of specificfood-derived components and to complement pre-vious epidemiological studies that suggested a po-tential health benefit. Thus, transcript profiling hasbeen used in nutritional interventions to assess theeffect of nutrients. Such studies dealt, for exam-ple, with antioxidants with the aim of mimickingthe benefits of caloric restriction (Lane et al. 2007);with plant-derived flavonoids like green tea catechins(McLoughlin et al. 2004; Vittal et al. 2004); and withsoy isoflavones and flavones (Fuchs et al. 2005a, b;Herzog et al. 2004), which have been shown toprovide remarkable biological effects as importantas cancer-preventive activity. Other health-beneficialnutrients like polyunsaturated fatty acids (Kitajkaet al. 2004; Lapillonne et al. 2004) or micronutri-ents like zinc (Kindermann et al. 2005; tom Diecket al. 2005) and vitamin E (Johnson and Manor 2004)have also been studied by transcriptomics to describetheir effect on the metabolism. The aim here is toidentify an affected set of genes whose regulationillustrates a metabolic adaptation. This type of fun-damental discoveries can then be used as a basis forthe development of adapted diets focusing on partic-ular (pathological) states of the organism.

Given the degree of complexity of genetic re-search, it appears evident that a combination of ge-netics and gene expression experiments applied to thesame subjects in the same studies can confer addedvalue, assuming that the analysis tools are ready tointegrate the related results. Therefore, the deploy-ment of microarray-based gene expression analysesshould and will be more and more complemented bySNP, CNP, or epigenetic studies. Nevertheless, thenumber of human studies in which only transcriptprofiling is applied to assess the biological effectsof nutritional intervention or to identify markers ofhealth continues to grow.

PROTEOMICS IN NUTRITIONAND HEALTH RESEARCH

Proteomics has evolved as an analogue to genomics,from identifying all proteins present in a given sam-ple at a given time to a global molecular analysisplatform addressing functional aspects of biologi-cal systems (Wilkins et al. 1996). In contrast to thegenome, the proteome is highly dynamic and con-

stantly changing in response to the environment of acell or an organism. Comparing such variations in theproteome enables the discovery of key proteins andthe identification of modulated pathways involved,for example, in specific nutrition-related processes.Over the last two decades, proteomics has developedinto an established technology for biomarker discov-ery (Lescuyer et al. 2007; Schrattenholz and Groebe2007), clinical applications (Mischak et al. 2007),disease profiling and diagnostics (Marko-Varga et al.2005; Vitzthum et al. 2005), the study of protein in-teractions (Gingras et al. 2007), and of the dynamicsof signaling pathways (Scholten et al. 2006). Nutri-tional proteomics is an emerging field in which thesetechnologies are applied to nutritional research. Itholds great promise to (a) profile and characterizedietary and body proteins, digestion, and absorp-tion, (a) identify biomarkers of nutritional status andhealth/disease condition, and (c) understand func-tions of nutrients and other dietary factors in growth,reproduction, and health (Wang et al. 2006a).

In the following, we briefly summarize the maintechnologies deployed for protein separation, iden-tification, and quantification. Then, we review pro-teomic studies with a specific focus on nutritionalinterventions.

PROTEOMICS TECHNOLOGIES

Numerous reviews of proteomic technologies and ap-plications have been published. Most recently, Na-ture Methods dedicated a special section on massspectrometry (MS) in proteomics that gives an ex-cellent overview on topics such as large-scale datageneration, analysis, and validation (Nesvizhskiiet al. 2007), elucidation of cellular networks of pro-tein interactions (Kocher and Superti-Furga 2007),mass spectrometric imaging (Cornett et al. 2007),“top-down” analysis of intact proteins (Siuti andKelleher 2007), and clinical research perspectives(Beretta 2007).

The proteomics workflow essentially con-sists of sample preparation and protein/peptide(pre)separation, identification, and quantification.The latter two encompass the complex interface be-tween data generation and processing/validation. De-spite tremendous progress at all levels of this work-flow, the term “proteome” remains—in contrast tothe genome—a theoretical entity, because proteomicstudies have to date never revealed an entire pro-teome. Recent efforts, for example, in Drosophilamelanogaster research, catalogued up to 63% of thepredicted proteome (Brunner et al. 2007). Cover-age in higher organisms, however, rarely reaches


Chapter 2: Omics in Nutrition and Health Research 15

more than 10%, with the numbers for quanti-fied proteins being even smaller (Bantscheff et al.2007). Nevertheless, open databases (e.g., PRIDE(http://www.ebi.ac.uk/pride) (Jones et al. 2008) orPeptideAtlas (http://www.peptideatlas.org) (Desiereet al. 2006)) have been established to convertdata and results from proteomic experiments intopublicly accessible information. The most recentaddition of this collective effort to standardizeand share protein data is the Human Proteinpedia(http://www.humanproteinpedia.org) (Mathivananet al. 2008). This portal provides an integrated viewof the human proteome, allowing users to con-tribute and edit proteomic data similar to the on-line encyclopedia Wikipedia (Giles 2005). HumanProteinpedia can accommodate data from diverseplatforms, including yeast two-hybrid screens, MS,peptide/protein arrays, immunohistochemistry, west-ern blots, co-immunoprecipitation, and fluorescencemicroscopy-type experiments.

For nutritional studies, in vitro samples like cellsas well as ex vivo samples such as tissues and bodyfluids may be suitable. Cultivable primary cells, thatis, nontransformed cell lines, should be chosen overcancer cell lines as those have a number of dereg-ulated pathways as compared to normal cells as re-cently demonstrated by a systems biology-orientedapproach integrating transcriptomic and proteomicanalysis of buccal epithelial tumor cells (Staab et al.2007). Cell cultures, however, offer the advantageof virtually unlimited protein supply but the in vitromodels may be far from the in vivo situation. There-fore, intestinal epithelial cell lines have been com-pared at proteomic level to ex vivo recovered gutcells in different cellular stages (Lenaerts et al.2007).

Ex vivo proteomic samples encompass tissue sec-tions from gut (Lopes et al. 2008; Marvin-Guy et al.2005), liver (Edvardsson et al. 2003), biliary tract(Kristiansen et al. 2004), and muscle (Gelfi et al.2006) obtained by resection of biopsies. An impor-tant constraint of proteomic sampling for any nu-tritional study is the demand of being minimallyinvasive or noninvasive. Therefore, less invasivelysampled body fluids like blood plasma (Andersonet al. 2004) and urine (Adachi et al. 2006) are attrac-tive. The plasma proteome is characterized by thehighest complexity and the widest dynamic range,but the proteins in blood are by nature relatively sol-uble. The urinary proteome has revealed an astonish-ingly high number of intact proteins and is thereforean information-rich proteome source; however, trun-cation and degradation of urinary proteins add to thecomplexity.

Extensive sample preseparation, depletion, and en-richment strategies are required due to the complex-ity of a proteome and the technical limitations ofmodern MS.

Two-dimensional gel electrophoresis enabled forthe first time to separate, visualize, and quantify manyproteins simultaneously in one image and paved theway for proteomics (Gorg et al. 2004). The cur-rent state-of-the-art methodology is 2D-DIGE, whichstands for differential imaging gel electrophoresis(Tonge et al. 2001): the control and case “proteomes”are labeled each with a specific fluorescent dye, thenmixed and co-separated, and subsequently analyzedfor fluorescent color ratios in a similar way as DNAmicroarrays. Despite the remarkable improvementsof gel-based proteomics thanks to DIGE (Sellerset al. 2007), the gel approach still suffers from (a)difficult automation and thus limited throughput, (b)relatively narrow dynamic range, and (c) a discrim-ination of proteins with extreme physicochemicalproperties (size, pI, hydrophobicity).

Therefore, chromatography-based techniqueshave been developed for protein and peptide(pre)separation. The established term in this contextis multidimensional protein identification technol-ogy (MudPIT) (Motoyama et al. 2006), also called“shotgun proteomics” approach (reviewed by Wuand MacCoss 2002). Typically, two-dimensionalchromatography (ion exchange followed by reversedphase) is online coupled to electrospray ionization(ESI) MS (see below). The approach is based on“proteome” digestion upstream in the workflow andpeptide-level separation.

Often, protein and peptide preseparation alone isinsufficient for dealing with proteome-scale samplecomplexity. Therefore, the most abundant proteinsmay have to be depleted from the sample, that is,specifically be removed by affinity chromatographywithout affecting the remaining protein composi-tion, especially when analyzing human plasma (Gonget al. 2006) or serum (Bjorhall et al. 2005).The “Equalizer Technology” was described recently(Righetti et al. 2006): a combinatorial library of lig-ands bound to beads was shown to reduce the con-centration differences in human plasma and urine,essentially by binding less of the abundant and moreof the rare proteins.

A complementary strategy of gaining access tolow-abundant proteins is enrichment of the latter, forexample, when targeting subproteomes such as thephospho- or glycoproteome. Various chemical scav-engers have been developed to capture phosphopep-tides and proteins, such as IMAC (immobilized metalaffinity capture), titanium dioxide resins, or alumina



particles. Recently, Reinders et al. thoroughly com-pared the performance of these techniques (Rein-ders and Sickmann 2005). Glycosylated peptides andproteins can be enriched by lectin affinity (Vosselleret al. 2006) or different trapping reactions such as hy-drazide chemistry (Sun et al. 2007). Nandi et al. havedeveloped a so-called tagging-via-substrate (TAS)approach for global identification of O-GlcNAc-modified proteins, enabling O-glycosyl enrichment(Nandi et al. 2006).

Key characteristics of a modern mass spectrome-ter are sensitivity (today femto- to attomolar), massaccuracy (high to low ppm), mass resolution (10,000to millions), and speed of MS and MS/MS ac-quisition. MALDI (matrix-assisted laser desorption/ionization) (Tanaka 2003) and ESI (Fenn 2003) arethe most popular and powerful methods to producegas phase ions of proteins and peptides. Differentmass analyzers are used in proteomics experiments,such as triple quadrupole (QqQ) instruments, espe-cially for targeted multireaction monitoring (MRM)experiments to, for example, simultaneously quan-tify dozens of plasma proteins (Anderson and Hunter2006). Combined with time-of-flight (ToF) or ion-trap (IT) analyzers, they form hybrid systems such asQ–ToF and Q–IT instruments. Fourier transform ioncyclotron resonance mass spectrometers (FT-ICRs)(Nielsen et al. 2005) and the more recently introducedorbitrap system (Makarov et al. 2006) represent thehigh-end proteomics MS space. FT-ICR instrumentsoffer ultimate resolution (>100,000) and low-ppmmass accuracy that enables “top-down” analysis ofintact proteins as opposed to the more frequentlyemployed “bottom-up” approach (Siuti and Kelleher2007).

Protein quantification can be achieved throughstaining with protein dyes, and as discussed above,currently the most advanced technology at proteinlevel is 2D-DIGE (Sellers et al. 2007). A furtheroption is to incorporate stable isotopes into proteinsand/or peptides in a differential manner (heavy versuslight isotope) and to quantify the proteins/peptides bymass spectrometric comparison of the signals derivedfrom the light- and heavy-isotope-labeled sample.These methods have been summarized by the Reg-nier group (Julka and Regnier 2004) and assessedin real-life scenarios by Wu et al. (2006) and theHeck team (Kolkman et al. 2005). Introduction ofthe mass labels can also be achieved by metaboliclabeling (Beynon and Pratt 2005; de Godoy et al.2006). This approach is advantageous because ofits minimal interference with the biological system.Metabolic labeling is routinely performed with cul-tured cells ranging from bacteria and yeast to mam-

malian cells. This has been demonstrated in multicel-lular organisms such as Caenorhabditis elegans andD. melanogaster (Krijgsveld et al. 2003) and veryrecently even in rats (McClatchy et al. 2007). Chem-ical or enzymatic methods must be applied to labelex vivo recovered tissues or fluids. The chemical tag-ging concept has been introduced by Aebersold et al.under the name ICAT (isotope-labeled affinity tag)(Gygi et al. 1999). The iTRAQ (isotope tags for rela-tive and absolute quantification) method (Ross et al.2004) offers quadruplex (and soon eight-plex) anal-ysis, that is, four conditions can be compared in oneexperiment. In view of multiple chemical taggingmethods typically performed post-digestion and tar-geting one amino acid side chain at a time, our grouphas come up with a new concept termed AniBAL(aniline-benzoic acid labeling): the same tag is intro-duced into all amino and carboxyl functions alreadyat the protein level (Panchaud et al. 2008). This ap-proach minimizes sample bias, optimizes proteomecoverage, and is based on a simple and symmetricchemistry. Bowman et al. have extended the stable-isotope concept to quantitative glycomics by devel-oping a quadruplex derivatization scheme amenableto mass spectral readout (Bowman and Zaia 2007).Label-free approaches have been developed more re-cently, based on highly reproducible LC-MS condi-tions, which allow comparative peptide analysis ofcomplex samples (Old et al. 2005; Ono et al. 2006).

All quantification approaches discussed so fardeliver relative quantitative information. Absolutequantification (AQUA) of proteins/peptides was de-scribed by Gerber et al. (2003) employing the classi-cal isotope-labeled internal standard approach. Theconcept of proteotypic peptides takes this strategyto the proteome level by selection of the “best fly-ing” peptide for each protein as a unique identifier(Mallick et al. 2007). Recently, targeted quantifica-tion of more than 50 plasma proteins was demon-strated with this approach (Anderson and Hunter2006). However, design and production of suchisotope-labeled reference peptides still needs furtherimprovement for optimal exploitation (Mirzaei et al.2008).

MS data analysis requires sophisticated softwareto acquire, store, retrieve, process, validate, and in-terpret these data and to eventually transform theminto useful biological information. While peptide andprotein identification and database search programslike Mascot (Perkins et al. 1999), Sequest (Yates et al.1995), or Phenyx (Colinge et al. 2004) have a longand successful standing (reviewed by Nesvizhskiiet al. 2007), entirely new software infrastructures fordata processing and validation have been built, such

NUTRIGENOMICS AND PROTEOMICS IN HEALTH AND DISEASE€¦ · Section I Introduction 1. Nutrigenomics...

Documents

Transcript of NUTRIGENOMICS AND PROTEOMICS IN HEALTH AND DISEASE€¦ · Section I Introduction 1. Nutrigenomics...