nutrigenomics

Nutrigenomics – Opportunities in Asia

Forum of NutritionVol. 60

Series Editor

Ibrahim Elmadfa, Vienna

Nutrigenomics –Opportunities in Asia

Basel · Freiburg · Paris · London · New York ·

Bangalore · Bangkok · Singapore · Tokyo · Sydney

Volume Editors

E. Shyong Tai, Singapore

Peter J. Gillies, Newark, Del.

26 figures, 1 in color, and 10 tables, 2007

Dr. E. Shyong Tai Dr. Peter J. GilliesDepartment of Endocrinology DuPont Haskell Laboratory for Health

Singapore General Hospital and Environmental Sciences

Singapore Newark, Del. (USA)

Bibliographic Indices. This publication is listed in bibliographic services, including Current Contents® and

Index Medicus.

Disclaimer. The statements, options and data contained in this publication are solely those of the individ-

ual 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 2007 by S. Karger AG, P.O. Box, CH–4009 Basel (Switzerland) and ILSI Southeast Asia

Region, Singapore

www.karger.com

Printed on acid-free paper

ISSN 1660–0347

ISBN 978–3–8055–8216–2

Library of Congress Cataloging-in-Publication Data

ILSI International Conference on Nutrigenomics (1st: 2005: Singapore)

Nutrigenomics : opportunities in Asia / volume editors, E.S. Tai, P.J.

Gillies.

p. ; cm. – (Forum of nutrition, ISSN 1660–0347 ; v. 60)

Includes bibliographical references and indexes.

ISBN-13: 978–3–8055–8216–2 (hard cover : alk. paper)

1. Nutrition–Genetic aspects–Congresses. 2.

Nutrition–Asia–Congresses. 3. Genomics–Asia–Congresses. I. Tai, E.S.

(E. Shyong) II. Gillies, P. J. (Peter J.) III. Title. IV. Series.

[DNLM: 1. Genomics–Asia–Congresses. 2. Nutrition

Index Physiology–genetics–Asia–Congresses. 3. Nutritional

Sciences–Asia–Congresses. W1 BI422 v.60 2007 / QU 145 I29n 2007]

QP144.G45I47 2005

612.3–dc22

2007012335

This book is dedicated to the international community of scientistswho believe in the promise of nutrigenomics!

VII

Contents

XI PrefaceTai, E.S. (Singapore); Gillies, P.J. (Newark, Del.)

Concepts and Methods in Nutrigenomics

1 Nutrition in the ‘Omics’ EraMilner, J.A. (Rockville, Md.)

25 NutrigeneticsEl-Sohemy, A. (Toronto, Ont.)

31 Epigenomics and NutritionCobiac, L. (Adelaide)

42 Early Nutrition: Impact on EpigeneticsMathers, J.C. (Newcastle)

49 Nutrition and Genome HealthFenech, M. (Adelaide)

66 Nutrition: Ethics and Social ImplicationsSlamet-Loedin, I.H.; Jenie, U.A. (Jakarta)

80 ProteomicsThongboonkerd, V. (Bangkok)

91 Diet and Genomic StabilityYoung, G.P. (Adelaide)

97 High-Throughput GenotypingLee, J.-E. (Seoul)

Nutrigenomics and Health

102 Nutrient-Gene Interactions in Lipoprotein Metabolism – An OverviewOrdovas, J.M. (Boston, Mass.); Corella, D. (Boston, Mass./Valencia);

Kaput, J. (Chicago, Ill.)

110 The Genetics of Lipoprotein Metabolism and Heart DiseaseTai, E.S. (Singapore)

118 Gene-Environment Interactions and the Diabetes Epidemic in IndiaMohan, V.; Sudha, V.; Radhika, G.; Radha, V.; Rema, M.; Deepa, R. (Chennai)

127 Gene Expression in Low Glycemic Index Diet – Impact on Metabolic ControlTakeda, E.; Arai, H.; Muto, K.; Matsuo, K.; Sakuma, M.; Fukaya, M.;

Yamanaka-Okumura, H.; Yamamoto, H.; Taketani, Y. (Tokushima)

140 Genetic Polymorphisms in Folate-Metabolizing Enzymes and Risk of Gastroesophageal Cancers: A Potential Nutrient-GeneInteraction in Cancer DevelopmentLin, D.; Li, H.; Tan, W.; Miao, X.; Wang, L. (Beijing)

146 Dietary Quercetin Inhibits Proliferation of Lung Carcinoma CellsHung, H. (Singapore)

158 Osteoporosis: The Role of Genetics and the EnvironmentOngphiphadhanakul, B. (Bangkok)

168 Application of Nutrigenomics in Eye HealthDelcourt, C. (Bordeaux)

Nutrigenomics – Applications to the Food Industry

176 Nutrigenomics of Taste – Impact on Food Preferences and Food ProductionEl-Sohemy, A.; Stewart, L.; Khataan, N.; Fontaine-Bisson, B.; Kwong, P.;

Ozsungur, S.; Cornelis, M.C. (Toronto, Ont.)

183 Prospects for Improving the Nutritional Quality of Dairy and Meat ProductsCoffey, S.G. (St. Lucia)

Contents VIII

196 Functionality of Probiotics – Potential for Product DevelopmentDekker, J.; Collett, M.; Prasad, J.; Gopal, P. (Palmerston North)

Conclusion

209 Developing the Promise of Nutrigenomics through Complete Science and International CollaborationsKaput, J. (Chicago, Ill./Davis, Calif.)

Executive Summary

224 ILSI’s First International Conference on Nutrigenomics:Opportunities in AsiaFlorentino, R.F. (Metro Manila)

242 Author Index

243 Subject Index

Contents IX

Preface

Nutrition plays an important role in optimizing human health and manag-

ing disease. Unfortunately, the human response to diet is so incredibly variable

that nutritional counseling beyond that of general advice is a complex and chal-

lenging task. Nutrigenomics seeks to understand the variability of the individ-

ual’s response to food and the underlying mechanisms whereby foods exert

their health-promoting activities. The promise of nutrigenomics is that with a

deeper molecular understanding of nutrition we may some day be able to design

diets that truly maximize an individual’s potential for health and wellness.

Asia is home to two thirds of the world’s population. Many societies within

Asia are undergoing rapid socioeconomic development and are experiencing an

attendant transition in diet-related morbidity and mortality. Paradoxically, the

problem of under- and overnutrition coexists in Asia. This, combined with the

tremendous diversity in diet, dietary intake patterns, local culture, and nutri-

tional needs, makes the identification and provision of an optimal diet relevant

to all the people living in Asia an extraordinary challenge. This same diversity,

however, provides opportunities to ask and answer scientific questions which

cannot be investigated elsewhere in the world.

Recognizing the special nutrition science research opportunities afforded

in Asia, the International Life Sciences Institute (ILSI) hosted an exciting 3-day

meeting in Singapore on December 7–9, 2005. This conference enjoyed the

support and guidance of the Commonwealth Scientific and Industrial Research

Organization of Australia, the National Institutes of Health in the United States,

and the Genome Institute of Singapore. The first ILSI international conference

on nutrigenomics, with a focus on opportunities in Asia, was an international

XI

gathering of scientists from the academia, government and industry that

attracted speakers and attendees from around the world with everyone coming

to share their experience and knowledge in the area of nutrigenomics. This book

is a culmination of the efforts of all those who organized and participated in this

conference.

The book includes an elegant and articulate summary of the conference

that Rodolfo Florentino was kind enough to provide and closes with an invited

article by Jim Kaput that provides a road map for international collaboration in

nutrigenomics. The core of the book starts off with concepts and methods in

nutrigenomics designed to give those interested in this field a general overview;

this is followed by specific examples of the applications of these concepts and

methods to specific disease states. Unfortunately, it was not possible to include

all the presentations from the meeting. Respectful apologies are offered to those

speakers and presenters whose work could not be included, but without whose

participation the meeting could not have been such a success!

For those of you who were able to attend the meeting we hope this book

reinforces your memories of the exciting science and collegiality of the confer-

ence; for everyone else we hope the book encourages you to engage in nutri-

genomic research and to attend the next ILSI conference on nutrigenomics.

Dr. E. Shyong Tai, SingaporeDr. Peter J. Gillies, Newark, Del.

Preface XII

Tai ES, Gillies PJ (eds): Nutrigenomics – Opportunities in Asia.

Forum Nutr. Basel, Karger, 2007, vol 60, pp 1–24

Nutrition in the ‘Omics’ Era

J.A. Milner

Nutritional Science Research Group, Division of Cancer Prevention,

National Cancer Institute, National Institutes of Health, Department of

Health and Human Services, Rockville, Md., USA

AbstractConsumers throughout the world are increasingly questioning the quality and safety of

their diets and how the foods they eat are influencing their health. Much of this interest stems

from mounting evidence that bioactive food components cannot only influence one’s ability

to achieve one’s genetic potential, but can also have a significant influence on the quality of

life as measured by both physical and cognitive performance, and modify the risk and/or

severity of a variety of disease conditions. During the past century, a wealth of evidence has

pointed to dietary habits as a determinant of premature death, including that associated with

heart disease, stroke, diabetes, liver disease, atherosclerosis and cancer, although consider-

able variability in response is evident. Several factors may account for these discrepancies

including individual variability due to genetic and epigenetic regulation of cellular proteins

and associated small-molecular-weight compounds. This interrelationship between the food

components and the ‘omics’ (genomics, epigenomics, transcriptomics, proteomics and

metabolomics) will be briefly reviewed as a factor contributing to the variability among stud-

ies. Expanded knowledge about these omics interrelationships will not only define the mole-

cular target for food components but will also assist in identifying those individuals who are

likely to respond maximally.

Copyright © 2007 S. Karger AG, Basel

Belief in the medicinal powers of foods and their components is not a new

concept, but has been handed down for generations. Historically, Hippocrates is

often quoted as suggesting almost 2,500 years ago to ‘let food be thy medicine

and medicine be thy food’. Today, consumers throughout the world are bom-

barded by a host of health messages about the benefits of foods and/or food

components for promoting health and/or reducing the risk of a variety of

chronic diseases. Undeniably, evidence continues to mount that the use of foods

and/or dietary supplements can assist in achieving one’s ‘genetic potential’,

Concepts and Methods in Nutrigenomics

Milner 2

improve physical and cognitive performance, and reduce chronic diseases [1]. It

is logical that such attributes could have profound societal and economic conse-

quences by not only reducing premature deaths but also by reducing overall

health care cost. The societal implications are reflected in a 2005 report by the

World Health Organization which suggested that at least 80% of premature

heart disease, stroke and type 2 diabetes, and 40% of cancer in Southeast Asia

could be prevented through a healthy diet, regular physical exercise and avoid-

ing tobacco products [2]. In this same report, it was estimated that a 2% annual

reduction in deaths due to chronic disease might save over 8 million lives dur-

ing the next 10 years. Within India alone, it was estimated this reduction in mor-

tality could result in an economic gain of USD 15 billion during the upcoming

decade [2]. While the richness of the scientific literature makes it difficult to

discount an involvement of dietary habits in health promotion, it is less obvious

who might benefit most, which foods are most important, the circumstances

which dictate benefits or risk, and if anyone might be placed at risk by dietary

change. These along with other issues are of paramount importance in moving

the science of nutrition forward. Unquestionably, it will not be simple to define

the conditions and circumstance(s) for achieving the greatest benefit from

foods or their isolated components. Nevertheless, the accumulating science

makes it believable that such a personalized approach is feasible. Inherent to

this concept is that individuals will vary in their response to a food, food com-

ponent or dietary pattern. Thus, the overriding assumption of this concept is

that unique preemptive strategies exist for the use of diet to retard or reverse the

progression of a disease which are dependent on the genetics and lifestyle of the

consumer. This article is devoted to providing basic principles about how

increasing knowledge about genomics can assist with the unraveling of incon-

sistencies in the scientific literature about the importance of the diet in health

promotion and disease prevention, and for predicting who will benefit most or

be placed at risk by dietary change.

Controversies Involving Nutrition and Health

Several recent meta-analyses illustrate the mixed messages that can sur-

face about what role, if any, diet has in health promotion [3–6]. While these

findings frequently reflect disagreements in interpretation among scientists,

they also lead to confusion among consumers and can erode the trust that they

have for the scientific enterprise. Thus, greater attention needs to be given to

the totality of the information rather than to individual studies in defining the

health significance of the diet. It would be truly disappointing if a simple sum-

mation of evidence-based nutrition studies became the ‘gold standard’ since the

Nutrition and Genomics 3

majority of case-control and cohort studies are not simple repeats of previous

undertaking and thus vary enormously in experimental design, tested popula-

tions, and outcome measures; all of this can surely influence overall interpreta-

tions. Unfortunately, conclusions based on a meta-analysis may even depend on

the method the authors used to select trials for inclusion in the analysis. Thus,

summaries of evidence which do not consider the biological response, plausi-

bility and consequences are doomed to create more confusion than they resolve.

What is increasingly clear is that inadequate long-term intervention studies

exist for making definitive conclusions about who will benefit and who might

be placed at risk by dietary change. Indisputably, well-controlled long-term

intervention studies which incorporate the newest technologies hold the great-

est promise for unraveling the complex interplay between diet and health.

Future clinical studies must incorporate genomics in the study design, and not

just use it as an analytic approach to confounders to data interpretation.

Considerable preclinical evidence linking diet to health outcomes centers

on the response to a single bioactive component as a modulator of a key cellu-

lar process or series of critical processes [7, 8]. Both in vitro and in vivo studies

suggest that multiple targets are likely responsible for the phenotypic response

to foods and/or dietary supplements [8, 9]. These targets may be involved in cell

division, inflammation, apoptosis, compound bioactivation or a host of other

biological processes which influence the phenotype. Focusing on a process

which can be modified by one or more bioactive food components will help

with a systematic approach to understanding the role of diet in health promo-

tion. However, in some cases, research suggests that whole foods may offer

advantages over isolated components, possibly indicating that multiple food

components or multiple targets are needed to bring about a desired effect.

Nutrient-nutrient, as well as nutrient-drug interactions can be significant determi-

nants of the overall phenotypic response. For example, the ability of n–3 fatty

acids to increase the sensitivity to anthracyclines is dependent on vitamin E

intake [10] or the benefits of calcium are generally dependent on the intake of

vitamin D [11]. While not as frequently examined, negative interactions among

food components or nutrient-drug interactions are also possible and such lines

of investigation deserve added attention to assist with the identification of

potentially vulnerable individuals.

Since the quantity of exposure can markedly influence the outcome, it is

imperative that nutrition studies use physiologically relevant concentrations and

consider the totality of the diet as a factor influencing the overall response

[7, 8]. Sadly, multiple exposures in humans are hampered by the availability of

definitive biomarkers that reflect a long-term health outcome. Finally, it is

worth noting that studies which examine the impact of dietary change through

more than one phase of life are exceedingly rare, yet are desperately needed if

Milner 4

sense is to be made out of the diet-health conundrum and the opportune time for

intervention to bring about a desired change. A narrow and simplistic view of

dietary patterns is obviously problematic and may contribute to misconceptions

and thus the confusion that exist today about the importance of functional foods

and health promotion and disease prevention.

Clearly, predicative biomarkers which can evaluate long-term conse-

quences are fundamental to resolving dietary issues which cannot be addressed

for practical or ethical reasons. At present, few validated biomarkers are avail-

able for assessing the impact of diet in health promotion. Similar to environ-

mental toxicity research, it is likely that at least 3 different types of biomarkers

will be needed to assess the impact of diet (fig. 1) [12]. Foremost among these

is the need to accurately identify exposures to foods and their components.

Obviously, if the effective concentration does not reach the target tissue, there is

little hope that it will be effective in bringing about a desired effect. Likewise,

sensitive and reliable biomarkers for identifying the impact of bioactive food

components are in short supply. These ‘effect’ biomarkers should provide sensi-

tive and time-/dose-dependent information about how the food component(s)

modifies/modify one or more specific cellular processes [6–8]. Assuming this/

these molecular target(s) can be analyzed in the affected or surrogate tissue, it

can ideally be an effective biomarker for assessing the response to physiologi-

cally relevant exposures to foods or their components. Finally, it is clear that we

Absorbeddose

Inactive metabolite

Biologicallyeffective

dose

Dietaryexposure

Susceptibilityfactors

Earlybiologiceffect

Moleculartarget

Alteredstructure/function

Health consequences‘positive ornegative’

Fig. 1. Three types of biomarkers are needed to evaluate the response to foods and

dietary supplements.


need to assess susceptibility factors including genetic and epigenetic markers

which can reflect an individual’s responsiveness to the biological response to a

food, food patterns or dietary supplements. The use of such information may

assist with improving the usefulness of standard 24-hour recall and food fre-

quency questionnaires by developing predictive models that take into account

genetic factors influencing absorption, metabolism and excretion. These suscep-

tibility biomarkers (fig. 1) will again provide important clues about responders,

both positive and negative, in the molecular target to diet-induced phenotypic

change.

Health Implication of Individual Genomic Variability

The human genome is a complicated blueprint of information. While all

DNA has four relatively simple bases (adenine, guanine, cytosine, and thymi-

dine), their sequence can have a pronounced effect on what ultimately evolves.

The nearly 3 billion base pairs (3.2 Gb) in the human genome constitute what is

sometimes affectionately called the ‘genome encyclopedia’. If a gene is analo-

gous to a word, then a chromosome must be a chapter and the genome the whole

book. Similar to a word, a gene may have a single or multiple meanings, and can

be influenced by the context in which it is expressed. Like a chapter, a chromo-

some is a large collection of genes organized into a linear string of information.

The complete set of chapters is necessary to form the ‘book’ of information that

comprises the genetic blueprint of each and every organism.

The size of this blueprint is illustrated by assuming each DNA basis cre-

ates a series of words each of which contain 5 characters. Thus, about 600 mil-

lion words could be generated from the human genome. If these words were

compiled to an average of 12 words per line then an equivalent of about 50,000

text lines would be generated. Since an average page only has about 70 lines,

this would mean the human genome would contribute about 700,000 pages. If

these pages were assembled into an encyclopedia with 1,000 pages in each vol-

ume, there would be about 700 volumes for late-night reading and enjoyment!

Even this analogy is overly simplistic, since it does not take into consider-

ation genetics, epigenetics, proteomics and metabolomics variations that occur

within and among individuals. Human genetic predictions are exceedingly

complicated by the presence of comparatively long and variable intron sequences

[13]. These intron sequences (noncoding DNA regions) interrupt the sequences

containing instructions for making a protein (exons). The panoramic views of

the human genome have already begun to reveal a wealth of information and

some early surprises. While much remains to be deciphered in this vast informa-

tion source, several fundamental principles have emerged. It is safe to conclude

Milner 6

that the more we learn about the human genome, the more there will be to learn.

Interestingly, the coding region of a gene (exons) which is the portion of DNA

that is transcribed into mRNA and translated into proteins only constitutes

about 1.5% of the human genome [14]. Furthermore, transcription units, con-

sisting of exons, introns, and the regulatory region, constitute about 20% of the

entire human genome. One must wonder if the remainder is simply a filler or

has a yet to be defined role. Evidence already exists that multiple gene mes-

sages can be derived from a single stretch of DNA based on alternative uses of

promoters, exons, and termination sites. Adding to these overlapping transcrip-

tion units, somatic recombination events and the existence of highly similar

gene families and pseudogenes make it difficult to identify and categorize

genes. Regardless, the fundamental premise of genomics is that DNA reading

results in the formation of messenger RNA which then codes from proteins

which ultimately bring about changes in small-molecular-weight compounds

and in cellular processes (fig. 2).

It is already known that human genome variability can arise for several

reasons including single nucleotide changes (polymorphisms), deletions, inser-

tions, and translocations. Translocations and gross deletions are important

causes of both cancer and inherited disease. Such gene rearrangements are non-

randomly distributed in the human genome as a consequence of selection for

growth advantage and/or the inherent potential of some DNA sequences to be

frequently involved in breakage and recombination. Alu insertional elements,

the most abundant class of short interspersed nucleotide elements in humans,

are dimeric sequences approximately 300 bp in length derived from the 7SL

RNA gene. About 500,000 to 1 � 106 Alu units are dispersed throughout the

Essential andnonessential

foodcomponents

Transcriptionfactor

DNA target Gene

Nucleus

Biological response incell process(es)

Change mRNAProtein(Change in activity or

abundance)

Fig. 2. Relationship between dietary components and genomic regulation.


human haploid genome primarily in AT-rich neighborhoods located within

larger GC dense chromosomal regions. These sequences contain a bipartite

RNA polymerase III promoter, a central poly-A tract, a 3� poly-A tail, numer-

ous CpG islands and are bracketed by short direct repeats. Such insertions are

associated with a number of disease states [15].

Restriction fragment length polymorphisms, short tandem repeats, and

variable-number tandem repeats are also present in the genome. Intraspecies

variation in the length of DNA fragments generated by the action of restriction

enzymes or caused by mutations that alter the sites at which these enzymes act

can change the length, number, or production of fragments. Restriction frag-

ment length polymorphism is a term used in two related contexts: as a charac-

teristic of DNA molecules (arising from their differing nucleotide sequences)

by which they may be distinguished, and as the laboratory technique which uses

this characteristic to compare DNA molecules. The short tandem repeats are

tandemly repeated DNA sequences of a pattern of length from 2 to 10 bp [for

example (CA)n(TG)n in a genomics region] and the total size is lower than

100 bp. Repeated sequences represent a large part of eukaryotic genomes.

Single nucleotide polymorphisms (SNPs) are the most common DNA

sequence variation. They occur when a single nucleotide in the genome is

altered. A variation in the incidence must occur in at least 1% of the population

to be considered an SNP. Huntington’s disease, cystic fibrosis, and muscular

dystrophy are examples of diseases which are linked to a single gene polymor-

phism [16]. While we have known about the genetics of these diseases for a

number of years, reliable and effective therapies have remained largely elusive.

Cancer and possibly several other chronic diseases are likely a result of multiple

genetic shifts and thus present an even more daunting task for understanding

the disease but are important for developing strategies for prevention and/or

therapy.

Fortunately, the majority of SNPs do not appear to cause disease; however,

they may assist in determining the likelihood that a particular abnormality may

occur [17]. Nevertheless, some SNPs have been linked to an increased disease

risk. For example, a gene associated with Alzheimer’s disease is apolipoprotein

E. This gene can contain two SNPs which may result in three possible alleles:

E2, E3, and E4. Each allele differs by one DNA base, and the protein product of

each gene differs by one amino acid. Typically an individual inherits one mater-

nal and one paternal copy of a gene. Research has shown that an individual who

inherits at least one E4 allele has a greater risk of developing Alzheimer’s dis-

ease, presumably as a result of the one amino acid substation in the E4 protein

which influences its structure and function. Inheriting the E2 allele, on the other

hand, appears to protect against Alzheimer’s disease. Of course, SNPs are not

absolute since those inheriting two E4 alleles do not always develop Alzheimer’s

Milner 8

disease [18]. Thus, other factors or events, including the environment (diet),

may affect the disease risk. It is certainly possible that either internal or external

stressors set the stage for when bioactive food components are most effective.

Thus, expanded knowledge about genetic and environmental interactions is

fundamental to unraveling global variation in the incidence and/or severity of

several disease states. Evidence is already surfacing that genetic variation can

influence the propensity for the initiating event, the progression to a clinical

disease state, and the trajectory of several diseases. For example, the interleukin

1 family of cytokines has a critical role in mediating inflammation, which is

considered a factor in many chronic diseases, including coronary artery dis-

ease, rheumatoid arthritis and cancer. Recent research has identified several

sequence variations in the regulatory DNA of the genes coding for important

members of the interleukin 1 family, and these variations are associated with

differential effects on the inflammatory response [17]. While inconclusive, evi-

dence is beginning to surface that the physiological relevance of such genetic

variation can be modified by the foods that are consumed.


Nutrigenomics represents the dynamic interface between nutrition and

genomic-regulated processes [18, 19]. A set of fundamental principles exist

which underpins the nutrigenomics concept. The first is that the genotype can

influence the ability of a food component to influence cellular processes associ-

ated with health and/or disease. Second, numerous dietary components are

capable of influencing, singly or in combination, the gene expression patterns

involved in multiple cellular processes. Third, the observed cellular response is

dependent on the amount and duration of exposure to a specific or blend of

food components. Finally, the ability of a bioactive food component to influ-

ence cellular processes will depend, in some cases, on the stage of the life cycle.

Collectively, nutrigenomics embodies the interrelationships occurring among

variation in DNA base sequences, epigenetics events and transcriptomics. Such

interactions may influence not only the magnitude, but sometimes the direction,

of the response to specific bioactive food components [6, 19–22]. Inappropriate

dietary habits may tip the scale from a healthy condition to a state of disease

progression. Thus, appropriate dietary intake of food components is fundamen-

tal to regulating normal physiological processes, as well as the squelching of

potential pathologic conditions. The scientific literature already provides evi-

dence that the response to food components can vary from tissue to tissue, as

well as a function of the time and duration of intervention [23–25]. Undeniably,

the capturing of this genomic-diet information is critical to the identification of


those individuals who will benefit from intervention strategies and those who

might be placed at risk because of a dietary change. The incorporation of this

information will allow for nutritional preemption strategies which utilize foods

or their components to enhance normal processes and/or to retard or reverse

cellular events that lead to aberrant conditions including those associated with

chronic disease.

The 30,000 genes in the human genome are responsible for more than

100,000 functionally distinct proteins, and likely 3–5 times that number of

small-molecular-weight cellular constituents (such as metabolic intermediates,

hormones and other signaling molecules, and secondary metabolites) which

collectively can enhance or suppress a number of physiological processes.

Understanding how foods and their components influence each step in the cas-

cade of events leading to a phenotype (fig. 3) is a daunting task, but holds great

promise in helping improve the quality of life and reduce the risk of diseases.

Numerous Bioactive Components Are Involved

A host of dietary components, encompassing both essential and nonessen-

tial constituents, are reported to influence both genetic and epigenetic processes.

Phytochemicals arising from plants, along with zoochemicals occurring in ani-

mal products, fungochemicals from mushrooms, and bacterochemicals from

Metabolomics

Proteomics

Nutritionaltranscriptomics

Nutritionalepigenetics

Nutrigenetics

Bioactivefood

component(s)

DNA

RNA

Protein

Metabolite

Phenotype

Nutrigenomics

Fig. 3. The influence of nutrigenetics, epigenetics, transcriptomics, proteomics, and

metabolomics on the phenotypic response to food components.

Milner 10

bacteria may be physiologically relevant modifiers of health. Compounds encom-

passing such diverse categories as minerals, amino acids, carbohydrates, fatty

acids, carotenoids, dithiolthiones, flavonoids, glucosinolates, isothiocyanates,

and allyl sulfurs may influence multiple pathways associated with growth, devel-

opment and disease resistance [1, 6–8]. Some have estimated that typical diets

may contain more than 25,000 bioactive food components. If this is the case,

there will likely be many additional compounds which will be identified with

physiological relevance. Future research must focus on the effective dose

required to bring about a response and when during the life span the maximum

response is achievable. Likewise, it will be important to tease apart the effective

intake of multiple foods where active components are influencing a common

molecular target. Is it possible that consumption of one food can mask the pro-

tection provided by another and possibly account for the confusion in findings

which attempt to relate a food to a specific health condition?

DNA Polymorphisms Influence Response to Food Components

The metabolism of folate serves as an appropriate example of nutrigenetics

or how genetic polymorphisms may cause individual responsiveness to the diet.

Folate is recognized to be an important factor in DNA synthesis, stability, and

integrity. Its availability can also modulate DNA methylation, which is an

important epigenetic determinant of gene expression (an inverse relationship),

and the maintenance of DNA integrity and stability. Several SNPs in the folate

metabolic pathway have been identified and characterized [26]. The complexity

of understanding the importance of SNPs in folate homeostasis comes from

their involvement in folate absorption (glutamate carboxypeptidase II), intra-

cellular folate uptake (folate receptors and reduced folate carriers), intracellular

folate retention (folylpolyglutamate synthetase) and catabolism and efflux (glu-

tamyl hydrolase), methionine cycle (methionine synthase, methionine synthase

reductase), maintenance of the intracellular folate pool (dihydrofolate reduc-

tase, serine hydroxymethyltransferase), and nucleotide biosynthesis (thymidy-

late synthase) [27]. Nevertheless, one or more of these SNPs may influence the

requirements for this vitamin B and thereby modify its regulatory effects on

epigenetic processes. While much remains to be learned about the functional

ramifications of these SNPs, there is evidence that its occurrence in 5,10-meth-

ylenetetrahydrofolate reductase may increase disease risk, including cardiovas-

cular disease and certain kinds of birth defects [28]. This variant is relatively

common worldwide and is characterized by a replacement of the nucleotide

cytosine with thymine at position 677 in the 5,10-methylenetetrahydrofolate

reductase gene. This change leads the 5,10-methylenetetrahydrofolate reductase


gene to produce a form of methylenetetrahydrofolate reductase which has

reduced activity at higher temperatures (thermolabile). Individuals with the

thermolabile form of the enzyme have increased levels of blood homocysteine,

a biomarker which has been identified as a risk factor for heart disease. While

low serum or red cell folate does not appear to be a requirement for neural tube

defects in humans, it is possible that the response to folate fortification reflects

tissue variability in the uptake and/or utilization of this vitamin B [29]. An

expanded mechanistic understanding of the role of SNP-regulated folate status

in both DNA methylation and DNA stability warrants greater attention [27–30].

Another nutrigenetic example of a dietary link with SNPs arises from stud-

ies on the intake of calcium and vitamin D associated with cancer risk.

Decreased dietary exposures to calcium and vitamin D have been suggested as

risk factors for human colon cancer. Feeding a Western-style diet containing

reduced calcium, vitamin D and increased fat to an animal model of familial

adenomatous polyposis, a common genetic alteration in colon cancer [31],

decreased survival compared to an ideal semipurified diet. This effect was mag-

nified when the p21 gene, which is an important regulator of the cell cycle, was

inactivated [32]. More recently, this group has found that feeding this diet to

normal C57/Bl6 mice led to hyperproliferation, hyperplasia and whole-crypt

dysplasias in the colon [33]. Further modification of folic acid, methionine,

choline and vitamin B12 content of the diet was found to result in adenoma and

carcinoma development in normal mouse colon [34]. The results indicate that a

semipurified rodent diet designed to mimic the human Western diet can induce

colonic tumors in normal mice without carcinogen exposure. Likewise, these

studies clearly demonstrate the modifying role that genes [Apc, p21, WAF1/

cip1, and p27 (Kip1)] can have in influencing the magnitude of the tumor prob-

lem.

During the past few years, vitamin D has received increasing attention as a

possible deterrent to cancer. At least part of the response to vitamin D may

depend on the vitamin D receptor (VDR) and the downstream events that it

influences. The VDR, which is known to dimerize with the retinoid X receptor,

binds to 1�,25(OH)2D3 promoter sites to regulate the transcription of many

genes in more than 30 tissues. When bound to the hormonally active form of

vitamin D, 1,25(OH)2D3, VDR transactivates genes that inhibit proliferation, or

promote differentiation and apoptosis [35–37]. VDR has also been shown to

bind carcinogenic bile acids and thereby transactivates the CYP3A gene. The

CYP3A gene product mediates bile acid degradation, thus anticarcinogenic

effects of VDR might also be mediated through bile acid ligands [38]. Whatever

the ligand VDR, transactivation efficiency may be influenced by a start codon

polymorphism (FokI) which affects the length of the N-terminal VDR transac-

tivation domain. The FokI f allelle results in a VDR protein that is three amino

Milner 12

acids longer than the protein produced from the F allele [39], and the latter has

been shown to be more efficient in transactivating target genes at least in some

studies. In a case-control study nested within a large cohort of Singapore

Chinese it was noted that individuals carrying the FF FokI genotype had a 51%

increased risk of colorectal cancer compared with the Ff genotype and a 84%

increase compared to those with the ff genotype [40]. The risk appeared to be

modified by both dietary calcium and fat. Among those with a low calcium or

low fat intake (below the median values in controls), the risk for colorectal can-

cer increased in a gene-dose-dependent manner such that individuals possess-

ing the ff genotype displayed an approximately 2.5-fold increased risk. There

was little evidence of a VDR genotype-colorectal cancer association among

subjects with higher than median values of either dietary fat or calcium.

Polymorphisms in manganese superoxide dismutase (MnSOD) is another

example of how genetics can be modified by the foods consumed. This enzyme

is essential for life as evident by the neonatal lethality in knockout mice. Mice

expressing only 50% of the normal complement of MnSOD demonstrate

increased susceptibility to oxidative stress and severe mitochondrial dysfunc-

tion resulting from elevation of reactive oxygen species. Numerous studies have

shown that MnSOD can be induced to protect against pro-oxidant insults result-

ing from cytokine treatment, ultraviolet light, irradiation, certain tumors, amy-

otrophic lateral sclerosis, and ischemia/reperfusion. In addition, overexpression

of MnSOD has been shown to protect against proapoptotic stimuli as well as

ischemic damage. Conversely, several studies have reported declines in MnSOD

activity during diseases including cancer, aging, progeria, asthma, and trans-

plant rejection. The molecular mechanisms involved in this loss in activity are

not well understood. Certainly, MnSOD gene expression or other defects could

play a role in such inactivation. Based on recent studies of the susceptibility of

MnSOD to oxidative inactivation, it is also likely that post-translational modifi-

cation may be an important determinant of the response to dietary components

[41]. A polymorphism in MnSOD [valine (V) to alanine (A)] has been reported

to be associated with increased prostate cancer risk. This polymorphism

becomes a significantly modified risk when prediagnostic plasma antioxidants

are considered [42]. In men with the AA genotype, a high selenium concentra-

tion (4th versus 1st quartile) was found to be associated with a relative risk of

0.3 [95% confidence interval (CI) 0.2–0.7]. In contrast, in men with the VV/VA

genotype, the relative risks were 0.6 (95% CI 0.4–1.0) and 0.7 (95% CI

0.4–1.2) for total and clinically aggressive prostate cancer. These patterns were

similar for lycopene and �-tocopherol and were particularly strong when com-

bined with information about selenium status. Thus, nutrigenetic information

may provide important clues about those who should be particularly vigilant

in assuring adequate antioxidant intake. Collectively, evidence from this and


several other studies suggests that nutrigenomic information can be useful

for identifying those who may benefit most from adequate intake of specific

nutrients.

Genotypes and Haplotypes

Slight variations in our DNA sequences can have a major impact on

whether or not a disease develops and on the response to environmental factors

such as infectious microbes, toxins, and diet consumed. Literally millions of

SNPs are known to occur within the human genome making it unlikely that a

single base change will be found that is sufficient to account for a number of

chronic diseases. Thus, examining commonly linked genomic changes seems to

be a reasonable solution. Genetic variants are often inherited together in seg-

ments of DNA called haplotypes which are shared by a majority of the human

population. Thus, they may be useful in deciphering the genetic differences that

make some people more susceptible to disease than others and likewise how

diet will impact their susceptibility [43]. The International HapMap Project

(http://www.hapmap.org/) may be particularly useful in teasing out genetic dif-

ferences that determine the response to specific foods and their components.

This consortium of scientists from six countries is devoted to constructing a

map of the patterns of SNPs that occur across populations in Africa, Asia, and

the United States. It is hoped that dramatically decreasing the number of indi-

vidual SNPs by using haplotyping will provide a shortcut for discovery of the

DNA regions associated with common complex diseases such as cancer, heart

disease, diabetes, and some forms of mental illness. The new map may be par-

ticularly useful in providing clues about how genetic variation can be incorpo-

rated into understanding the response to environmental factors, including diet.

Dietary Modulation of Epigenomics

Epigenetics refers to the study of heritable changes in gene expression that

occur without a change in DNA sequence. Basically it is the study of heritable

changes in gene expression resulting from mitotic (the process in cell division

by which the nucleus divides through prophase, metaphase, anaphase, and

telophase, normally resulting in two new nuclei, each of which contains a com-

plete copy of the parental chromosomes) and meiotic processes (the process of

cell division in sexually reproducing organisms that reduces the number of

chromosomes in reproductive cells from diploid to haploid, leading to the pro-

duction of gametes in animals and spores in plants). Thus, epigenetic mechanisms

Milner 14

provide an ‘extra’ layer of control in gene expression regulation. These regula-

tion processes are critical components in the normal development and growth

of cells. Evidence continues to arise that epigenetic abnormalities are causative

factors in cancer, genetic disorders and pediatric syndromes as well as con-

tributing factors in autoimmune diseases and aging. Abnormal methylation pat-

terns are a nearly universal finding in cancer, as changes in DNA methylation

have been observed in many cancer tissues (e.g. colon, stomach, uterine cervix,

prostate, thyroid, and breast). Site-specific alterations in DNA methylation have

also been observed in cancer and are thought to have a significant role in gene

regulation and tumor behavior. Hypermethylation is often observed at some 5�gene or promoter regions of neoplastic cells that are largely unmethylated in

normal somatic tissues [44]. For many of these genes, this hypermethylation

has been linked to transcriptional silencing.

Three distinct mechanisms are intricately related to epigenetics: DNA

methylation, histone modification and RNA-associated silencing [45]. The

observation in human cells that silence RNA can function to suppress gene

expression at the level of transcription has created an interesting new paradigm

shift in thoughts about mammalian RNA regulation [46]. Because epigenetic

events can be changed, they offer another explanation for how environmental

factors, including diet, can influence biological processes and phenotypes.

Collectively, nutritional epigenomics refers to the ability of dietary components

to influence each of these distinct mechanisms of regulation. Both essential

and nonessential dietary components have been reported to influence DNA

methylation patterns [47]. The effects of these food components can occur at

four different sites. First, dietary factors are undeniably important in providing

and regulating the supply of methyl groups available for the formation of

S-adenosylmethionine, the universal methyl donor. Second, dietary factors may

modify the utilization of methyl groups by processes including shifts in DNA

methyltransferase activity. A third plausible mechanism relates to DNA deme-

thylation activity. Finally, the DNA methylation patterns themselves may influ-

ence the response by regulating genes which influence absorption, metabolism

or the site of action for the bioactive food component.

The importance of maternal methyl donor supply in the diet has been

examined in terms of DNA methylation and methylation-dependent epigenetic

phenotypes in the offspring. Evidence in the agouti mouse model that supple-

mentation of choline, betaine, folic acid, vitamin B12, methionine, and zinc to

the maternal diet increases the level of DNA methylation in the agouti gene and

induces a change in the color pattern of the hair coat is particularly striking

[48]. This phenotypic change, along with that caused by genistein, coincides

with a lower susceptibility to obesity, diabetes, and cancer [49]. These types of

studies suggest that in utero exposure to dietary components may not only


influence embryonic development but may also have profound and long-term

health consequences.

A process that regulates chromatin structure is through its attachment to

histones. It is increasingly apparent that a diverse array of enzymes modify his-

tones through acetylation, phosphorylation, methylation, and ubiquitylation.

Perturbations in histones can influence gene expression patterns. Indeed, aber-

rant histone acetylation is commonly observed in cancer. In general, histone

acetylation leads to chromatin remodeling and a derepression of transcription.

Naturally occurring histone deacetylase inhibitors such as butyrate can arise

from foods that are consumed or from bacterial fermentation of carbohydrate

within the gastrointestinal tract. Likewise, ingestion of bioactive components

such as sulforaphane from broccoli or genistein from soy [50, 51] can influence

histone deacetylase activity. While histones can be modified in a variety of ways,

most current information about the role of diet as a modifier comes from studies

involving acetylation. Clearly other types of modification deserve attention.

Diet and Transcriptomics

Genomic and epigenomic shifts do not totally account for the role that

dietary factors can have on a person’s phenotype since changes in the rate of

transcription of genes (nutritional transcriptomics) can also be exceedingly

important [52]. Several bioactive food components have been reported to be

important regulators of gene expression patterns both in vitro and in vivo.

Vitamins, minerals, and various phytochemicals have been reported to significantly

influence gene transcription and translation in a dose- and time-dependent

manner. These changes are likely key to the ability of food components to

influence one or more biological processes including cellular energetics, cell

growth, apoptosis, and differentiation, all of which are important in regulating

disease risk and consequences. The development of microarray technology con-

tinues to provide a powerful tool for examining potential sites of action of food

components. One merit of this approach is that it allows for a genome-wide

monitoring of expression for the simultaneous assessment of tens of thousands

of genes and their relative expression. While microarray technologies provide

an important tool to discover expression changes that are linked to cellular

processes, it must be remembered that any response is likely cellular dependent

and may vary from health to diseased conditions. Greater attention needs to be

given to why shifts in multiple genes are occurring simultaneously. Could it be

that these changes are a reflection of mRNA stability, pH, intracellular calcium

or some other intracellular signal? Regardless, transcriptomics holds promise to

assist with the discovery of sensitive biomarkers.

Milner 16

Animal models may also provide important clues about the physiological

importance of shifts in the expression of specific genes. Knockouts have

already been used to identify specific sites of action of bioactive food compo-

nents such as the nuclear factor E2 p45-related factor 2 (Nrf2) and the Kelch

domain-containing partner Keap1 which are modified by sulforaphane [53].

Gene expression profiles from wild-type and Nrf2-deficient mice fed sul-

foraphane have shown several novel downstream events and thus more clues

about the true biological response to this food component. The upregulation of

glutathione S-transferase, nicotinamide adenine dinucleotide phosphate:quinone

reductase, �-glutamylcysteine synthetase, and epoxide hydrolase, occurring

because of changes in Nrf2, likely explains the ability of sulforaphane to influ-

ence multiple processes including those involving xenobiotic-metabolizing

enzymes, antioxidants, and biosynthetic enzymes of the glutathione and glu-

curonidation conjugation pathways. Nrf2 also appears to provide protection

against oxidative stress and influences inflammatory processes, both of which

contribute to several disease conditions. Observations that Nrf2-deficient mice

are refractory to the protective actions of some food components and drugs

highlight the importance of the Keap1-Nrf2-ARE signaling pathway as a poten-

tially important molecular target.

Because microarray technologies provide only a single snapshot, overinter-

pretation of data is certainly possible. Mammals are known to be adaptive to

excess exposure to foods or components through shifts in absorption or metab-

olism. Thus, the quantity and duration of exposure must be considered when

evaluating the response in gene expression patterns following exposure to foods

or components. Molecular studies have already shown that specific events in

cell cycle progression that are modified by energy restriction can rather quickly

be reversed by refeeding [54]. Likewise, the ability of diallyl disulfide, a bioac-

tive component in garlic, to retard cell proliferation can be reversed by its

removal from incubation media [55]. Again, the ability of a cell or host to adapt

will dictate the frequency by which interventions will be needed to bring about

a desired effect. Another challenge with microarray analysis is how to analyze

the massive amounts of data that are generated. Because the number of genes

whose expression can be modified by dietary components may be enormous, a

hierarchical cluster analysis might be particularly useful. Although most studies

use a 50% change in gene expression patterns as a cutoff point for statistical

significance, a shift in mRNA expression in much lower amounts may have

physiological significance. As advances in bioinformatics occur, the impor-

tance of subtle changes in mRNA expression may help with predicting health

and disease risk and identifying responders from nonresponders to diet change.

Another new technology is RNA interference, which can be used to stop

the expression of a particular gene [56, 57]. This technology has recently been


used to investigate which genes are involved in explaining the actions of bioac-

tive food components and characteristics of diseases and conditions. The use of

RNA interference has allowed for the discovery of genes in the worm model

Caenorhabditis elegans which are associated with body fat and leanness [58].

Likewise, this technology has been used to identify sites of action of isothio-

cyanates from broccoli compounds [59]. These tools will surely help to delin-

eate the roles of various cellular factors in health and diseased conditions [56,

59]. They may also provide a new and exciting foundation for gene-nutrient

discovery.

Diet and Proteomics

The constellation of proteins in a cell is referred to as the proteome. Unlike

the relatively stable genome, the dynamic proteome changes from minute to

minute in response to tens of thousands of intra- and extracellular environmen-

tal signals, including ingested nutrients. A protein’s chemistry and behavior are

specified by the gene sequence and by the number and identities of other pro-

teins made in the same cell at the same time and with which it associates and

reacts. Studies to explore protein structure and activities, known as proteomics,

will likely be the focus of much research for decades to come and will help elu-

cidate the molecular basis of health and disease. Proteomics is an integral part

and key player in the family of ‘omics’ disciplines as there are genomics (gene

analysis), transcriptomics (gene expression analysis) and metabolomics (metabo-

lite profiling). Considering the complexity, dynamics and protein concentration

range of any given proteome, proteomics is an exceedingly challenging ‘omics’

discipline and generally requires the most sophisticated analytical approaches.

It should also be noted that proteomics does not always correlate with tran-

scriptomics. Several factors may account for this disconnect including alterna-

tive splicing producing multiple proteins from a single gene. Post-translational

modification (i.e. glycosylation, phosphorylation, oxidation, reduction) may

also generate multiple protein products originating from a single gene or a sin-

gle transcript. These modified proteins can vary tremendously in their biologi-

cal activities. Furthermore, protein expression inside a cell is not only regulated

by transcription of mRNA but also by translation efficiencies and degradation

rates.

While dietary inadequacy has long been recognized to influence protein

synthesis and degradation, recent evidence suggests that the nutritional pro-

teomics areas hold promise in explaining a number of subtle changes brought

about by slight shifts in eating behaviors. Preclinical studies have already

shown that dietary fish oil, conjugated linoleic acid, or elaidic acid can influence

Milner 18

lipoprotein metabolism and insulin levels as indicated by changes in proteomics

[60]. Overall, this approach identified 65 cytosolic and 8 membrane proteins

that were modified by dietary components, many of which were related to lipid

and glucose metabolism and to oxidative stress [60]. The importance of these

findings is in the merging of proteomics and physiologic measurements which

will in the future likely provide key insights into the mechanisms by which

dietary components regulate several metabolic processes and ultimately change

phenotypes [61].

Several other dietary components have also been shown to modify the pro-

teome through changes in post-translational events. For example, dietary sele-

nium has been shown to alter post-translationally proteins and thus influence a

number of metabolic pathways presumably by initially altering extracellular

signal-regulated kinase activity [62]. Western blot analysis revealed that the

response to selenium was not because of a change in protein content per se, but

resulted from an increase in extracellular signal-regulated kinase phosphoryla-

tion [63]. Likewise, adding allyl sulfur to cells in culture has been shown to

modify the phosphorylation of selected proteins and possibly accounts for the

ability of it to block the cell in the G2/M phase of the cell cycle [55]. Thus shifts

in phosphorylation may be the result of subtle changes in the activity of both

kinases and phosphatases which regulate a host of cellular events. The transi-

tory nature of these proteomic changes may indicate that some food compo-

nents will be needed at more frequent intervals in the diet to achieve a desired

outcome.

Diet and Metabolomics

Metabolomics is the study of the metabolome, which is the entire meta-

bolic content of a cell or organism at a given moment [64, 65]. Metabolomics

researchers have generally focused their attention on biofluids, including serum

and urine, and have paid far less attention to tissues and/or cells. The use of

exfoliated cells in the feces or urine as well as buccal cells may offer unique

insights into the specific role that diet has in changing small-molecular-weight

components and thereby cellular processes. Recent metabolomics studies have

utilized a number of sophisticated analytical tools including nuclear magnetic

resonance spectroscopy, mass spectrometry, chromatographic analysis, and

metabolic network analysis to estimate cellular metabolic fluxes.

Metabolomics has been studied in microorganisms and in plants, but the

literature lacks systematic studies with animals or humans [64, 66]. Quantitative

lipid metabolome data are beginning to reveal differential effects of dietary fats

on cardiac and liver phospholipid metabolism and hold promise for predicting


the response to other foods and their components [67, 68]. This approach

mapped changes in the concentration of lipid metabolites to their biochemical

pathways and the impact of drugs on their distribution and metabolism [68].

Because many effects of dietary macromolecules on tissue metabolism are

reflected in the plasma lipid metabolome, metabolomics has excellent potential

for evaluating subtle differences in the metabolic response to diet between

individuals.

Metabolomic approaches have been used for years to understand the

dynamic relationships between plasma amino acid inadequacy and cellular

processes. Noguchi et al. [69] suggested that correlation-based analyses could

be useful in the analysis of metabolomic data to determine which metabolites

may be responsible for the biological effects of adequate and excessive intakes

of amino acids. More recently, Shi et al. [70] indicated the merits of using high-

performance liquid chromatography separations coupled with coulometric

array detectors to detect low-molecular-weight, redox-active compounds that

differ between dietary-restricted and ad-libitum-fed states. Thus, metabolomic

approaches hold promise to detect subtle differences in the biological conse-

quences of consuming too little or too much of individual food constituents.

One study that did use a metabolomic approach found that plasma profiles

of healthy premenopausal women before and after consumption of 60 g of soy

provided clues about shifts in energetics [71]. Despite the presence of substan-

tial intersubject variability, the metabolomic analysis revealed that soy interven-

tion changed the plasma lipoprotein, amino acid, and carbohydrate profiles

suggesting soy-induced alterations in fat, protein, and carbohydrate metabo-

lism. It is certainly possible to expand this approach to allow for the identifica-

tion of individuals, based on their metabolic abilities, who would benefit from

the ingestion of a variety of individual foods and/or specific food patterns.

The use of knockout and transgenetic models for defining the importance

of metabolomic events will surely provide critical clues about cell regulation as

influenced by dietary habits. For example, Griffiths and Stubbs [72] used a

mutant cell with a transcription factor for hypoxia-inducible factor 1, a crucial

mediator of tumor progression because it upregulates a number of genes

involved in the formation and regulation of blood vessels, iron metabolism, glu-

cose and energy metabolism, cellular proliferation, differentiation, apoptosis

and matrix metabolism, to detect an upregulation in anabolic synthesis of

purine rings required to make adenosine triphosphate. More recently, these sci-

entists [73] have demonstrated the merits of using knockout mice to evaluate

proteomic and metabolomic changes during atherogenesis as they relate to

immune-inflammatory responses, oxidative stress, and energy metabolism.

Thus, metabolomic approaches can help elucidate and establish conclusions

about gene-nutrient interactions and the resulting shift in small-molecular-weight

Milner 20

cellular constituents which bring about physiologically relevant events involved

in disease processes.

Summary and Conclusions

Genomics is increasingly recognized as a determinant of the biological

consequences of the foods and supplements that are consumed. Each of the

omics disciplines offers unique opportunities for understanding how bioactive

food components might be used to improve health and decrease the risk of dis-

ease. While the complexity of the interrelationships between food components

and the omics disciplines cannot be overemphasized, a greater understanding of

this dynamic interrelationship will help identify those who will benefit most

from dietary change. The future road map for nutrition research must incorpo-

rate omics signals for understanding the twists and turns needed for health and

the cautions and stops necessary for disease prevention.

References

1 Milner JA: Functional foods and health: a US perspective. Br J Nutr 2002;88(suppl 2):S151–S158.

2 World Health Organization Global Report 2005. Preventing Chronic Disease: A Vital Investment.

Geneva, World Health Organization, 2005.

3 Chuwa EW, Seow-Choen F: Dietary fibre. Br J Surg 2006;93:3–4.

4 Forman D, Altman D: Vitamins to prevent cancer: supplementary problems. Lancet 2004;364:

1193–1194.

5 Truswell AS: Some problems with Cochrane reviews of diet and chronic disease. Eur J Clin Nutr

2005;59(suppl 1):S150–S154.

6 Trujillo E, Davis C, Milner J: Nutrigenomics, proteomics, metabolomics, and the practice of

dietetics. J Am Diet Assoc 2006;106:403–413.

7 Milner JA: Molecular targets for bioactive food components. J Nutr 2004;134:2492S–2498S.

8 Davis CD, Hord NG: Nutritional ‘omics’ technologies for elucidating the role(s) of bioactive food

components in colon cancer prevention. J Nutr 2005;135:2694–2697.

9 Gohil K: Functional genomics identifies novel and diverse molecular targets of nutrients in vivo.

Biol Chem 2004;385:691–696.

10 Colas S, Germain E, Arab K, Maheo K, Goupille C, Bougnoux P: Alpha-tocopherol suppresses

mammary tumor sensitivity to anthracyclines in fish oil-fed rats. Nutr Cancer 2005;51:178–183.

11 Zeni SN, Ortela Soler CR, Lazzari A, Lopez L, Suarez M, Di Gregorio S, Somoza JI, de Portela

ML: Interrelationship between bone turnover markers and dietary calcium intake in pregnant

women: a longitudinal study. Bone 2003;33:606–613.

12 Milner JA: Functional foods: the US perspective. Am J Clin Nutr 2000;71(6 suppl):1654S–1659S.

13 Lander ES, Linton LM, Birren B, Nusbaum C, Zody MC, Baldwin J, Devon K, Dewar K, Doyle

M, FitzHugh W, Funke R, Gage D, Harris K, Heaford A, Howland J, Kann L, Lehoczky J, LeVine

R, McEwan P, McKernan K, Meldrim J, Mesirov JP, Miranda C, Morris W, Naylor J, Raymond C,

Rosetti M, Santos R, Sheridan A, Sougnez C, Stange Thomann N, Stojanovic N, Subramanian A,

Wyman D, Rogers J, Sulston J, Ainscough R, Beck S, Bentley D, Burton J, Clee C, Carter N,

Coulson A, Deadman R, Deloukas P, Dunham A, Dunham I, Durbin R, French L, Grafham D,

Gregory S, Hubbard T, Humphray S, Hunt A, Jones M, Lloyd C, McMurray A, Matthews L,


Mercer S, Milne S, Mullikin JC, Mungall A, Plumb R, Ross M, Shownkeen R, Sims S, Waterston

RH, Wilson RK, Hillier LW, McPherson JD, Marra MA, Mardis ER, Fulton LA, Chinwalla AT,

Pepin KH, Gish WR, Chissoe SL, Wendl MC, Delehaunty KD, Miner TL, Delehaunty A, Kramer

JB, Cook LL, Fulton RS, Johnson DL, Minx PJ, Clifton SW, Hawkins T, Branscomb E, Predki P,

Richardson P, Wenning S, Slezak T, Doggett N, Cheng JF, Olsen A, Lucas S, Elkin C, Uberbacher

E, Frazier M, Gibbs RA, Muzny DM, Scherer SE, Bouck JB, Sodergren EJ, Worley KC, Rives

CM, Gorrell JH, Metzker ML, Naylor SL, Kucherlapati RS, Nelson DL, Weinstock GM, Sakaki Y,

Fujiyama A, Hattori M, Yada T, Toyoda A, Itoh T, Kawagoe C, Watanabe H, Totoki Y, Taylor T,

Weissenbach J, Heilig R, Saurin W, Artiguenave F, Brottier P, Bruls T, Pelletier E, Robert C,

Wincker P, Smith DR, Doucette-Stamm L, Rubenfield M, Weinstock K, Lee HM, Dubois J,

Rosenthal A, Platzer M, Nyakatura G, Taudien S, Rump A, Yang H, Yu J, Wang J, Huang G, Gu J,

Hood L, Rowen L, Madan A, Qin S, Davis RW, Federspiel NA, Abola AP, Proctor MJ, Myers RM,

Schmutz J, Dickson M, Grimwood J, Cox DR, Olson MV, Kaul R, Raymond C, Shimizu N,

Kawasaki K, Minoshima S, Evans GA, Athanasiou M, Schultz R, Roe BA, Chen F, Pan H, Ramser

J, Lehrach H, Reinhardt R, McCombie WR, de la Bastide M, Dedhia N, Blocker H, Hornischer K,

Nordsiek G, Agarwala R, Aravind L, Bailey JA, Bateman A, Batzoglou S, Birney E, Bork P,

Brown DG, Burge CB, Cerutti L, Chen HC, Church D, Clamp M, Copley RR, Doerks T, Eddy SR,

Eichler EE, Furey TS, Galagan J, Gilbert JG, Harmon C, Hayashizaki Y, Haussler D, Hermjakob

H, Hokamp K, Jang W, Johnson LS, Jones TA, Kasif S, Kaspryzk A, Kennedy S, Kent WJ, Kitts P,

Koonin EV, Korf I, Kulp D, Lancet D, Lowe TM, McLysaght A, Mikkelsen T, Moran JV, Mulder N,

Pollara VJ, Ponting CP, Schuler G, Schultz J, Slater G, Smit AF, Stupka E, Szustakowski J,

Thierry-Mieg D, Thierry-Mieg J, Wagner L, Wallis J, Wheeler R, Williams A, Wolf YI, Wolfe KH,

Yang SP, Yeh RF, Collins F, Guyer MS, Peterson J, Felsenfeld A, Wetterstrand KA, Patrinos A,

Morgan MJ, de Jong P, Catanese JJ, Osoegawa K, Shizuya H, Choi S, Chen YJ; International

Human Genome Sequencing Consortium: Initial sequencing and analysis of the human genome.

Nature 2001;409:860–921.

14 Leo CP, Hsu SY, Hsueh AJ: Hormonal genomics. Endocr Rev 2002;23:369–381.

15 Kulski JK, Dunn DS: Polymorphic Alu insertions within the Major Histocompatibility Complex

class I genomic region: a brief review. Cytogenet Genome Res 2005;110:193–202.

16 Church DM, Stotler CJ, Rutter JL, Murrell JR, Trofatter JA, Buckler AJ: Isolation of genes from

complex sources of mammalian genomic DNA using exon amplification. Nat Genet 1994;6:98–105.

17 Duff GW: Evidence for genetic variation as a factor in maintaining health. Am J Clin Nutr

2006;83:431S–435S.

18 Kaput J, Rodriguez RL: Nutritional genomics: the next frontier in the postgenomic era. Physiol

Genomics 2004;16:166–177.

19 Davis CD, Milner J: Frontiers in nutrigenomics, proteomics, metabolomics and cancer prevention.

Mutat Res 2004;551:51–64.

20 Nicodemus KK, Stenger JE, Schmechel DE, Welsh-Bohmer KA, Saunders AM, Roses AD,

Gilbert JR, Vance JM, Haines JL, Pericak-Vance MA, Martin ER: Comprehensive association

analysis of APOE regulatory region polymorphisms in Alzheimer disease. Neurogenetics 2004;5:

201–208.

21 Ordovas JM, Corella D: Nutritional genomics. Annu Rev Genomics Hum Genet 2004;5:71–118.

22 Ordovas JM: The quest for cardiovascular health in the genomic era: nutrigenetics and plasma

lipoproteins. Proc Nutr Soc 2004;63:145–152.

23 Fong LY, Zhang L, Jiang Y, Farber JL: Dietary zinc modulation of COX-2 expression and lingual

and esophageal carcinogenesis in rats. J Natl Cancer Inst 2005;97:40–50.

24 Crott JW, Choi SW, Branda RF, Mason JB: Accumulation of mitochondrial DNA deletions is age,

tissue and folate-dependent in rats. Mutat Res 2005;570:63–70.

25 Hilakivi-Clarke L, Cho E, Cabanes A, DeAssis S, Olivo S, Helferich W, Lippman ME, Clarke R:

Dietary modulation of pregnancy estrogen levels and breast cancer risk among female rat off-

spring. Clin Cancer Res 2002;8:3601–3610.

26 Smulders YM, Stehouwer CD: Folate metabolism and cardiovascular disease. Semin Vasc Med

2005;5:87–97.

27 Kim YI: Folate, colorectal carcinogenesis, and DNA methylation: lessons from animal studies.

Environ Mol Mutagen 2004;44:10–25.

Milner 22

28 Curtin K, Bigler J, Slattery ML, Caan B, Potter JD, Ulrich CM: MTHFR C677T and A1298C polymor-

phisms: diet, estrogen, and risk of colon cancer. Cancer Epidemiol Biomarkers Prev 2004;13:285–292.

29 Wen SW, Walker M: An exploration of health effects of folic acid in pregnancy beyond reducing

neural tube defects. J Obstet Gynaecol Can 2005;27:13–19.

30 Kim YI: 5,10-Methylenetetrahydrofolate reductase polymorphisms and pharmacogenetics: a new

role of single nucleotide polymorphisms in the folate metabolic pathway in human health and dis-

ease. Nutr Rev 2005;63:398–407.

31 Yang K, Edelmann W, Fan K, Lau K, Leung D, Newmark H, Kucherlapati R, Lipkin M: Dietary

modulation of carcinoma development in a mouse model for human familial adenomatous poly-

posis. Cancer Res 1998;58:5713–5717.

32 Yang WC, Mathew J, Velcich A, Edelmann W, Kucherlapati R, Lipkin M, Yang K, Augenlicht LH:

Targeted inactivation of the p21WAF1/cip1 gene enhances Apc-initiated tumor formation and the

tumor-promoting activity of a Western-style high-risk diet by altering cell maturation in the

intestinal mucosa. Cancer Res 2001;61:565–569.

33 Newmark HL, Yang K, Lipkin M, Kopelovich L, Liu Y, Fan K, Shinozaki H: A Western-style diet

induces benign and malignant neoplasms in the colon of normal C57Bl/6 mice. Carcinogenesis

2001;22:1871–1875.

34 Yang K, Yang W, Mariadason J, Velcich A, Lipkin M, Augenlicht L: Dietary components modify

gene expression: implications for carcinogenesis. J Nutr 2005;135:2710–2714.

35 Kallay E, Pietschmann P, Toyokuni S, Bajna E, Hahn P, Mazzucco K, Bieglmayer C, Kato S, Cross

HS: Characterization of a vitamin D receptor knockout mouse as a model of colorectal hyperpro-

liferation and DNA damage. Carcinogenesis 2001;22:1429–1435.

36 Liu M, Lee MH, Cohen M, Bommakanti M, Freedman LP: Transcriptional activation of the Cdk

inhibitor p21 by vitamin D3 leads to the induced differentiation of the myelomonocytic cell line

U937. Genes Dev 1996;10:142–153.

37 Mathiasen IS, Lademann U, Jaattela M: Apoptosis induced by vitamin D compounds in breast

cancer cells is inhibited by Bcl-2 but does not involve known caspases or p53. Cancer Res

1999;59:4848–4856.

38 Makishima M, Lu TT, Xie W, Whitfield GK, Domoto H, Evans RM, Haussler MR, Mangelsdorf

DJ: Vitamin D receptor as an intestinal bile acid sensor. Science 2002;296:1313–1316.

39 Miyamoto K, Kesterson RA, Yamamoto H, Taketani Y, Nishiwaki E, Tatsumi S, Inoue Y, Morita K,

Takeda E, Pike JW: Structural organization of the human vitamin D receptor chromosomal gene

and its promoter. Mol Endocrinol 1997;11:1165–1179.

40 Wong HL, Seow A, Arakawa K, Lee HP, Yu MC, Ingles SA: Vitamin D receptor start codon poly-

morphism and colorectal cancer risk: effect modification by dietary calcium and fat in Singapore

Chinese. Carcinogenesis 2003;24:1091–1095.

41 Macmillan-Crow LA, Cruthirds DL: Invited review: manganese superoxide dismutase in disease.

Free Radic Res 2001;34:325–336.

42 Li H, Kantoff PW, Giovannucci E, Leitzmann MF, Gaziano JM, Stampfer MJ, Ma J: Manganese

superoxide dismutase polymorphism, prediagnostic antioxidant status, and risk of clinical signif-

icant prostate cancer. Cancer Res 2005;65:2498–2504.

43 Velasquez JL, Lipkin SM: What are SNPs and haplotypes and how will they help us manage the

prevention of adult cancer? Curr Oncol Rep 2005;7:475–479.

44 Jones PA, Baylin SB: The fundamental role of epigenetic events in cancer. Nat Rev Genet

2002;3:415–428.

45 Jones PA: Overview of cancer epigenetics. Semin Hematol 2005;42(suppl 2):S3–S8.

46 Grunweller A, Hartmann RK: RNA interference as a gene-specific approach for molecular medi-

cine. Curr Med Chem 2005;12:3143–3161.

47 Ross SA: Diet and DNA methylation interactions in cancer prevention. Ann NY Acad Sci

2003;983:197–207.

48 Cooney CA, Dave AA, Wolff GL: Maternal methyl supplements in mice affect epigenetic varia-

tion and DNA methylation of offspring. J Nutr 2002;132(8 suppl):2393S–2400S.

49 Dolinoy DC, Weidman JR, Waterland RA, Jirtle RL: Maternal genistein alters coat color and pro-

tects Avy mouse offspring from obesity by modifying the fetal epigenome. Environ Health Perspect

2006;114:567–572.


50 Myzak MC, Karplus PA, Chung FL, Dashwood RH: A novel mechanism of chemoprotection by

sulforaphane: inhibition of histone deacetylase. Cancer Res 2004;64:5767–5774.

51 Ye R, Goodarzi AA, Kurz EU, Saito S, Higashimoto Y, Lavin MF, Appella E, Anderson CW, Lees-

Miller SP: The isoflavonoids genistein and quercetin activate different stress signaling pathways

as shown by analysis of site-specific phosphorylation of ATM, p53 and histone H2AX. DNA

Repair (Amst) 2004;3:235–244.



53 Yu X, Kensler T: Nrf2 as a target for cancer chemoprevention. Mutat Res 2005;591:93–102.

54 Hursting SD, Lavigne JA, Berrigan D, Donehower LA, Davis BJ, Phang JM, Barrett JC, Perkins

SN: Diet-gene interactions in p53-deficient mice: insulin-like growth factor-1 as a mechanistic

target. J Nutr 2004;134:2482S–2486S.

55 Knowles LM, Milner JA: Diallyl disulfide induces ERK phosphorylation and alters gene expres-

sion profiles in human colon tumor cells. J Nutr 2003;133:2901–2906.

56 Behlke MA: Progress towards in vivo use of siRNAs. Mol Ther 2006;13:644–670.

57 Gartel AL, Kandel ES: RNA interference in cancer. Biomol Eng 2006;23:17–34.

58 Ashrafi K, Chang FY, Watts JL, Fraser AG, Kamath RS, Ahringer J, Ruvkun G: Genome-wide

RNAi analysis of Caenorhabditis elegans fat regulatory genes. Nature 2003;421:268–272.

59 Singh SV, Herman-Antosiewicz A, Singh AV, Lew KL, Srivastava SK, Kamath R, Brown KD,

Zhang L, Baskaran R: Sulforaphane-induced G2/M phase cell cycle arrest involves checkpoint

kinase 2-mediated phosphorylation of cell division cycle 25C. J Biol Chem 2004;279:

25813–25822.

60 de Roos B, Duivenvoorden I, Rucklidge G, Reid M, Ross K, Lamers RJ, Voshol PJ, Havekes LM,

Teusink B: Response of apolipoprotein E*3-Leiden transgenic mice to dietary fatty acids: com-

bining liver proteomics with physiological data. FASEB J 2005;19:813–815.

61 de Roos B, Rucklidge G, Reid M, Ross K, Duncan G, Navarro MA, Arbones-Mainar JM,

Guzman-Garcia MA, Osada J, Browne J, Loscher CE, Roche HM: Divergent mechanisms of cis9,

trans11 and trans10, cis12-conjugated linoleic acid affecting insulin resistance and inflammation

in apolipoprotein E knockout mice: a proteomics approach. FASEB J 2005;19: 1746–1748.

62 Hu H, Jiang C, Li G, Lu J: PKB/AKT and ERK regulation of caspase-mediated apoptosis by

methylseleninic acid in LNCaP prostate cancer cells. Carcinogenesis 2005;26:1374–1381.

63 Adams A: Metabolomics: small molecule ‘omics’. Scientist 2003;18:38–40.

64 Watkins SM, German JB: Toward the implementation of metabolomic assessments of human

health and nutrition. Curr Opin Biotechnol 2002;13:512–516.

65 German JB, Bauman DE, Burrin DG, Failla ML, Freake HC, King JC, Klein S, Milner JA, Pelto

GH, Rasmussen KM, Zeisel SH: Metabolomics in the opening decade of the 21st century: build-

ing the roads to individualized health. J Nutr 2004;134:2729–2732.

66 Fiehn O: Combining genomics, metabolome analysis, and biochemical modelling to understand

metabolic networks. Comp Funct Genom 2001;2:155–168.

67 Watkins SM, Lin TY, Davis RM, Ching JR, DePeters EJ, Halpern GM, Walzem RL, German JB:

Unique phospholipid metabolism in mouse heart in response to dietary docosahexaenoic or

�-linolenic acids. Lipids 2001;36:247–254.

68 Fitzgerald DA: Drug discovery: lipid profiling for studying the metabolome. Genet Eng News

2001;21:32–36.

69 Noguchi Y, Sakai R, Kimura T: Metabolomics and its potential for assessment of adequacy and

safety of amino acid intake. J Nutr 2003;133(suppl 1):2097S–2100S.

70 Shi H, Vigneau-Callahan KE, Shestopalov AI, Milbury PE, Matson WR, Kristal BS: Characterization

of diet-dependent metabolic serotypes: primary validation of male and female serotypes in inde-

pendent cohorts of rats. J. Nutr 2002;132:1039–1046.

71 Solanky KS, Bailey NJ, Beckwith-Hall BM, Davis A, Bingham S, Holmes E, Nicholson JK,

Cassidy A: Application of biofluid 1H nuclear magnetic resonance-based metabonomic techniques

for the analysis of the biochemical effects of dietary isoflavones on human plasma profile. Anal

Biochem 2003;323:197–204.

72 Griffiths JR, Stubbs M: Opportunities for studying cancer by metabolomics: preliminary observa-

tions on tumors deficient in hypoxia-inducible factor 1. Adv Enzyme Regul 2003;43:67–76.

Milner 24

73 Mayr M, Chung YL, Mayr U, Yin X, Ly L, Troy H, Fredericks S, Hu Y, Griffiths JR, Xu Q:

Proteomic and metabolomic analyses of atherosclerotic vessels from apolipoprotein E-deficient

mice reveal alterations in inflammation, oxidative stress, and energy metabolism. Arterioscler

Thromb Vasc Biol 2005;25:2135–2142.

Dr. John A. Milner

Nutritional Science Research Group, Division of Cancer Prevention

National Cancer Institute, 6130 Executive Boulevard, Suite 3164

Rockville, MD 20892 (USA)

Tel. �1 301 496 0118, Fax �1 301 480 3925, E-Mail [email protected]



Nutrigenetics

Ahmed El-Sohemy

Department of Nutritional Sciences, Faculty of Medicine,

University of Toronto, Toronto, Ont., Canada

AbstractNutrients interact with the human genome to modulate molecular pathways that may

become disrupted, resulting in an increased risk of developing various chronic diseases.

Genetic polymorphisms affect the metabolism of dietary factors, which in turn affects the

expression of genes involved in a number of important metabolic processes. Genetic poly-

morphisms affecting nutrient metabolism may explain some of the inconsistencies among

epidemiological studies relating diet to chronic diseases such as cancer, diabetes, rheumatoid

arthritis, osteoporosis and cardiovascular disease. Understanding how genetic variations

influence nutrient digestion, absorption, transport, biotransformation, uptake and elimina-

tion will provide a more accurate measure of exposure to the bioactive food ingredients

ingested. Furthermore, genetic polymorphisms in the targets of nutrient action such as recep-

tors, enzymes or transporters could alter molecular pathways that influence the physiological

response to dietary interventions. Among the candidate genes with functional variants that

affect nutrient metabolism are those that code for xenobiotic-metabolizing enzymes (also

called drug-metabolizing enzymes). These enzymes are involved in the phase I and II

biotransformation reactions that produce metabolites with either increased or decreased

biological activity compared to the parent compound. A number of dietary factors are known

to alter the expression of these genes that, in turn, metabolize a vast array of foreign chemi-

cals including dietary factors such as antioxidants, vitamins, phytochemicals, caffeine,

sterols, fatty acids and alcohol. Knowledge of the genetic basis for the variability in response

to these dietary factors should result in a more accurate measure of exposure of target tissues

of interest to these compounds and their metabolites. Examples of how ‘slow’ and ‘fast’

metabolizers respond differently to the same dietary exposures will be discussed. Identifying

relevant diet-gene interactions will benefit individuals seeking personalized dietary advice as

well as improve public health recommendations by providing sound scientific evidence link-

ing diet and health.


El-Sohemy 26

Background

Nutrition plays an important role in the development of chronic diseases

such as osteoporosis, diabetes, rheumatoid arthritis, cancer and cardiovascular

disease. Nutrigenomics is an emerging branch of nutritional science that uses

genomic information along with techniques in molecular biology and genetics

to address issues important to nutrition and health [1, 2]. One approach used to

explore how genetic and dietary factors interact to influence various health out-

comes is to examine how diet alters the expression of genes that regulate impor-

tant cellular processes. Another is to examine how sequence variations in genes

that regulate metabolic pathways affect responsiveness to specific dietary factors.

Depending on the study design, a ‘response’ could be risk of disease, biomark-

ers of disease, or even biochemical indicators of nutrient intake. Nutritional

genomics is another term that has been used to describe the complex interac-

tions between nutrition and the genome [2]. Knowledge of the genetic basis for

the variability in response to certain dietary factors should result in a more

accurate measure of exposure of target tissues of interest to these compounds

and their metabolites.

Single nucleotide polymorphisms are the most common form of genetic

variation and occur at about 1 in every 500–2,000 bases throughout the human

genome [3, 4]. They are normally found in at least 1% of the population,

although common polymorphisms that occur in 10–50% of the population may

be more relevant from a public health perspective. Other types of genetic alter-

ations such as gene deletions or insertions can also occur and have significant

phenotypic effects. The complex chronic diseases that occur commonly in

developed countries usually take many years (decades) to develop, and their

multifactorial etiologies make it difficult to unravel the role of specific dietary

factors. Thus, the combined contribution of in vitro, animal, clinical and epi-

demiologic studies is necessary to understand the role of specific bioactive food

ingredients in maintaining optimal health.

Epidemiologic studies are of particular interest because they examine the

effects of a dietary factor or genetic variant in a human population. The etiol-

ogy of complex chronic diseases clearly involves both environmental and

genetic factors, with environmental influences such as diet likely having a

greater influence on individuals with a genetic predisposition. Genetic associ-

ation studies that link genotype frequencies to health outcomes have been lim-

ited by the failure to reproduce many results in subsequent studies that are

conducted in different populations [5]. The apparent inconsistencies between

studies, however, highlight the critical role that the environment contributes

towards the expression of a genetic variant. Despite the limitations of genetic

association studies, unexpected findings can sometimes point to environmental

Nutrigenetics 27

exposures that might previously have been overlooked [6]. Furthermore, if a

polymorphism that decreases the expression levels of a metabolic enzyme is

associated with a decreased risk of a disease, then identifying dietary factors

that inhibit the enzyme or decrease the expression levels of its gene in target

tissues might be a worthwhile strategy for optimizing health and preventing

disease (fig. 1).

Functionally significant polymorphisms that affect the absorption, metab-

olism or disposition of nutrients or other substances found in the diet will ulti-

mately affect the level of exposure of target tissues to the dietary compound and

its metabolites. The polymorphic xenobiotic-metabolizing enzymes include the

phase I and II biotransformation reactions that metabolize a vast array of chem-

icals into products that have either increased or decreased biological activity

compared to the parent compound. Some dietary factors are known to alter the

expression of genes that code for these enzymes that, in turn, transform them

into metabolites with altered biological activity. In addition to dietary sub-

stances that might be harmful, a number of nutrients and bioactive food ingre-

dients with purported health benefits are also metabolized by specific isoforms.

Genetic polymorphisms have been associated with altered rates of enzymatic

activities that affect circulating concentrations and ultimately the effectiveness

of dietary chemicals and their metabolites.

Cytochrome P450s (CYP) are a diverse group of enzymes that play an

essential role in the oxidative biotransformation of steroids, prostaglandins,

nutrients, drugs, chemicals and carcinogens. Despite their known function in

the metabolism of dietary factors and the wide variability in activity associated

with specific genetic variants, the use of CYP genotypes in nutritional epidemi-

ologic studies has been limited. Several dietary factors affect the expression of

CYP isoforms [7], which display wide individual variability in inducibility and

Dietary factor Digestion

Transport

Metabolism

Uptake

Biotransformation

Genome

Health

outcome

Fig. 1. The effects of diet on health

outcomes could occur directly or indirectly

through interactions with the genome (e.g.

DNA methylation, gene expression). Genetic

variations affecting nutrient digestion, absorp-

tion, metabolism, uptake or biotransforma-

tion can modify the effects of dietary factors

on various health outcomes.

El-Sohemy 28

activity because of their polymorphic sequences. CYP1A2 plays an important

role in the metabolism of a wide range of drugs as well as chemical substances

found in the diet. CYP1A2 is known to activate dietary carcinogens such as aro-

matic amines, but also detoxifies compounds such as caffeine [8]. A recent

study linking the low-activity CYP1A2 genotype to an increased risk of

myocardial infarction [9] suggests that a substance that is detoxified, rather than

activated, by this enzyme may be an important risk factor in that population.

Indeed, a subsequent study revealed that individuals with a low-activity

CYP1A2 genotype are at a greater risk of coffee-associated heart disease [10].

Since caffeine is the only major substance in coffee that is known to be detoxi-

fied by CYP1A2, this observation suggests that caffeine may be an important

risk factor for heart disease in certain populations.

Glutathione S-transferases (GSTs) are a superfamily of enzymes that play

a central role in the detoxification of several dietary compounds [11]. GSTs

are divided into several distinct classes with partially overlapping substrate

specificities. GSTM1, GSTT1, and GSTP1 are isoforms of the mu, theta, and

pi class, respectively. The GSTM1 and GSTT1 null genotypes have been asso-

ciated with both an increased as well as a decreased risk of certain forms of

cancer [12]. Inconsistencies among genetic association studies may be related

to the dual role of these enzymes in eliminating both harmful mutagens as well

as potentially beneficial compounds such as dietary isothiocyanates that are

found in cruciferous vegetables [13]. Indeed, a protective effect of the GSTM1

null genotype on colon and lung cancer has been related to lower urinary

excretion of glutathione-conjugated phytochemicals, suggesting that they are

not rapidly excreted [6, 14]. GSTT1 plays a similar role to GSTM1 in elimi-

nating potentially beneficial phytochemicals found in cruciferous vegetables

[15, 16]. Furthermore, vegetables rich in certain phytochemicals such as isoth-

iocyanates increase the expression of GSTs [17], which conjugate them to

more water-soluble forms that are more readily excreted. Although genetic

variants in xenobiotic-metabolizing enzymes are unlikely to have clinical sig-

nificance on their own, they may shed light on the role of potential substrates

in disease development. Genetic polymorphisms have also been identified in

catechol-O-methyltransferase, sulfotransferase and UDP-glucuronosyltrans-

ferase that result in marked differences in enzyme activity [18–20]. These

enzymes metabolize a number of dietary compounds which may be more

strongly associated with various health outcomes among individuals who

‘detoxify’ them less efficiently. For example, the intake of green tea was asso-

ciated with a lower risk of breast cancer only in women with the low-activity

allele for catechol-O-methyltransferase [21]. This enzyme catalyzes the

methylation of catechins found in green tea, which makes them more rapidly

eliminated.

Nutrigenetics 29

Conclusion

Genetic polymorphisms can affect the rate of digestion, absorption, metabo-

lism, and uptake of dietary factors, which could explain some of the inconsistent

findings relating diet to various health outcomes. In addition to providing a more

rational basis for giving personalized dietary advice, the knowledge gained by

applying genomic information to nutrition research will also improve the quality

of evidence used for making population-based dietary recommendations.

Discoveries made using genomic information should translate into more effective

dietary strategies to improve overall health by identifying unique targets for pre-

vention. Genetic polymorphisms of xenobiotic-metabolizing enzymes can affect

the biological activity of numerous dietary substances, and single nucleotide

polymorphisms in genes that code for these enzymes should be included in nutri-

tional epidemiologic studies to provide a more accurate measure of exposure of

target cells to the dietary compounds or their metabolites. Incorporating genetic

markers in the design of nutritional epidemiologic studies will help clarify the

role of both genetic and lifestyle factors in the development of chronic diseases.

Acknowledgements

This research was supported by a grant from the Canadian Institutes of Health Research

(No. MOP-77741) and the Advanced Foods and Materials Network (M&E-B-4). A. El-Sohemy

holds a Canada Research Chair in Nutrigenomics.

References

1 Muller M, Kersten S: Nutrigenomics: goals and strategies. Nat Rev 2003;4:315–322.


3 Sachidanandam R, Weissman D, Schmidt SC, et al: A map of human genome sequence variation

containing 1.42 million single nucleotide polymorphisms. Nature 2001;409:928–933.

4 Thorisson GA, Stein LD: The SNP Consortium website: past, present and future. Nucleic Acids

Res 2003;31:124–127.

5 Ioannidis JP, Ntzani EE, Trikalinos TA, Contopoulos-Ioannidis DG: Replication validity of genetic

association studies. Nat Genet 2001;29:306–309.

6 Lin HJ, Probst-Hensch NM, Louie AD, et al: Glutathione transferase null genotype, broccoli, and

lower prevalence of colorectal adenoma. Cancer Epidemiol Biomarkers Prev 1998;7:647–652.

7 Riddick DS, Lee C, Bhathena A, et al: Transcriptional suppression of cytochrome P450 genes by

endogenous and exogenous chemicals. Drug Metab Dispos 2004;32:367–375.

8 Gu L, Gonzalez FJ, Kalow W, Tang BK: Biotransformation of caffeine, paraxanthine, theobromine

and theophylline by cDNA-expressed human CYP1A2 and CYP2E1. Pharmacogenetics 1992;2:

73–77.

9 Cornelis MC, El-Sohemy A, Campos H: Genetic polymorphism of CYP1A2 increases risk of

myocardial infarction. J Med Genet 2004;41:758–762.

10 Cornelis MC, El-Sohemy A, Kabagambe EK, Campos H: Coffee, CYP1A2 genotype and risk of

myocardial infarction. JAMA 2006;295:1135–1141.

El-Sohemy 30

11 Armstrong RN: Structure, catalytic mechanism and evolution of the glutathione S-transferases.

Chem Res Toxicol 1997;10:2–18.

12 Cotton SC, Sharp L, Little J, Brockton N: Glutathione S-transferase polymorphisms and colorec-

tal cancer: a HuGe review. Am J Epidemiol 2000;151:7–32.

13 Kolm RH, Danielson UH, Zhang Y, Talalay P, Mannervik B: Isothiocyanates as substrates for

human glutathione transferases: structure-activity studies. Biochem J 1995;311:453–459.

14 London SJ, Yuan J-M, Chung FL, et al: Isothiocyanates, glutathione S-transferase M1 and T1 poly-

morphisms, and lung-cancer risk: a prospective study of men in Shanghai, China. Lancet 2000;356:

724–729.

15 Seow A, Shi CY, Chung FL, et al: Urinary total isothiocyanate (ITC) in a population-based sample

of middle-aged and older Chinese in Singapore: relationship with dietary total ITC and glu-

tathione S-transferase M1/T1/PI genotypes. Cancer Epidemiol Biomarkers Prev 1998;7:775–781.

16 Lin HJ, Zhou H, Dai A, et al: Glutathione transferase GSTT1, broccoli, and prevalence of colorec-

tal adenomas. Pharmacogenetics 2002;12:175–179.

17 Lampe JW, Chen C, Li S, et al: Modulation of human glutathione S-transferase by botanically

defined vegetable diets. Cancer Epidemiol Biomarkers Prev 2000;9:787–793.

18 Coughtrie MW, Johnston LE: Interactions between dietary chemicals and human sulfotransferases-

molecular mechanisms and clinical significance. Drug Metab Dispos 2001;29:522–528.

19 Miners JO, McKinnon RA, Mackenzie PI: Genetic polymorphisms of UDP-glucuronosyltransferases

and their functional significance. Toxicology 2002;181–182:453–456.

20 Mannisto PT, Kaakkola S: Catechol-O-methyltransferase (COMT): biochemistry, molecular biology,

pharmacology, and clinical efficacy of the new selective COMT inhibitors. Pharmacol Rev 1999;51:

593–628.

21 Wu A, Tseng C-C, Van Den Berg D, Yu MC: Tea intake, COMT genotype and breast cancer in

Asian-American women. Cancer Res 2003;63:7526–7529.

Dr. A. El-Sohemy

Department of Nutritional Sciences, Faculty of Medicine

University of Toronto, 150 College St.

Toronto, Ont. M5S 3E2 (Canada)




Epigenomics and Nutrition

Lynne Cobiac

Preventative Health National Research Flagship, CSIRO, Adelaide, Australia

AbstractEpigenomics or epigenetics refers to the modification of DNA that can influence the

phenotype through changing gene expression without altering the nucleotide sequence of the

DNA. Two examples are methylation of DNA and acetylation of the histone DNA-binding

proteins. Dietary components – both nutrients and nonnutrients – can influence these epige-

netic events, altering genetic expression and potentially modifying disease risk. Some of these

epigenetic changes appear to be heritable. Understanding the role that diet and nutrition play

in modifying genetic expression is complex given the range of food choices, the diversity of

nutrient intakes, the individual differences in genetic backgrounds and intestinal physiological

environments where food is metabolized, as well as the impact on and acceptance of new tech-

nologies by consumers.


Nutrigenomics is an all-encompassing term that covers the interaction of

diet with DNA, chromatin or RNA expression. Our diet is a complex mixture of

nutrients with known functions and nonnutritive bioactives or food com-

ponents, not all of which are known, with varying levels of bioactivity and

bioavailability, and which can be either protective or nonprotective against the

risk of developing disease or ill-health. The study of nutrigenomics thus extends

beyond the traditional perspective of nutritional science.

Just as there is diversity in the diet, there is considerable diversity in the

genome (polymorphisms) and in its expression. Genomic variation can occur in

single nucleotides [either as single nucleotide polymorphisms or as point muta-

tions], through insertion or deletion of a few up to several hundred bases or

through gross chromosomal rearrangements like those occurring in Down’s

syndrome.

‘Nutrigenetics’ is used to refer to the nutrient response at the level of the

single gene and is concerned with the effects of specific gene variants on

Cobiac 32

an individual’s response to diet, and ‘nutrigenomics’ refers to the response

to nutrients at the level of the whole genome or how diet affects genetic

expression across the whole genome. Nutrigenetics takes into consideration

an individual’s genetic background with recognition that there are allelic

variations in genes that respond differently to bioactives from foods. Genetic

polymorphisms may influence the nutrient or bioactive response in several

ways.

• Genetically based differences in absorption, disposition, metabolism and

excretion (e.g. the ability to remove the dietary potentially carcinogenic

heterocyclic amines is influenced by the presence of polymorphisms in the

CYP1A2 gene and/or the N-acetyltransferase gene) [1].

• Nutrient-gene interactions – an individual’s genetic background can deter-

mine which gene products are expressed that modulate the physiological

response to specific nutrients (e.g. polymorphisms in the angiotensinogen

or angiotensin-converting enzyme gene may influence the effectiveness of

changing dietary intake to modify blood pressure) [2].

• Nutrients may affect gene expression and thus the tissue or cellular

level of a protein that is part of the causal pathway of a disease and, as

such, could modify an individual’s susceptibility to that disease (e.g.

dietary fatty acids appear to play a role in the regulation of COX-2 expr-

ession and thus influence the inflammatory response which may eventu-

ally lead to diseases such as arthritis, cardiovascular disease and cancer)

[3, 4].

Epigenomics

Traditionally it was thought that phenotypic traits were determined solely

by genetic mutations and recombinations and that the nucleotide sequence was

the sole driver of heredity. However, the study of epigenetics/genomics has dra-

matically changed this point of view. Epigenomics/genetics refers to modifica-

tion of the DNA and DNA-binding proteins that can influence the phenotype

without altering the nucleotide sequence of the DNA. Methylation of cytosine

residues in DNA and modification of the histone proteins are two common

examples. The other major epigenetic mechanism is RNA-associated interfer-

ence with gene expression (e.g. small interfering RNAs [5]). The key issues are

that some of these epigenetic changes appear to be heritable, associated with

different disease states and modifiable by dietary factors. This short paper will

provide an overview of DNA methylation, histone modification and genomic

instability, linking to dietary factors and some broader food- and diet-related

issues.

Epigenomics and Nutrition 33

DNA MethylationDNA methylation generally occurs with the transfer of a methyl group

from S-adenosylmethionine to the carbon-5 position of cytosine in a CpG dinu-

cleotide sequence catalyzed by DNA methyltransferase. The overall frequency

of CpG dinucleotides in the genome is lower than expected from the base com-

position and they tend to be more common in the promoter regions of genes and

in repeat sequences. In general, they tend to be heavily methylated except for

those in the promoter regions of transcriptionally active genes where they tend

to be less methylated. Methyl-CpG-binding proteins bind specifically to methy-

lated DNA and cause chromatin remodelling. Thus the level of DNA methyla-

tion can affect the level of genetic expression with hypermethylation generally

having a net silencing effect.

Cancer has been associated with disregulation of epigenetic control [6]

with resultant changes to DNA methylation patterns such as global hypomethy-

lation [7], hypermethylation of tumor suppressor genes (thus silencing the sup-

pressive effect) [8] or hypomethylation in the promoter region of an oncogene

(allowing the oncogene to express) which may be an early event for cancer if

this occurs or hypomethylation of repeat sequences that are rich in CpG islands.

Changes in methylation (such as global hypomethylation) can also result in

chromosome instability [9, 10].

With aging, we also see an epigenetic drift with aberrant methylation of

CpG islands in the promoter regions of DNA repair and tumor suppressor

genes, which is likely to increase tissue vulnerability to neoplastic transforma-

tion and may in some way account for the increased incidence of several can-

cers, such as colorectal cancer, with advancing age.

Furthermore, Alzheimer’s disease has been associated with different pat-

terns of DNA methylation. DNA hypomethylation has been reported in the

genetic regions associated with the expression of amyloid precursor protein,

presenilin 1 and �-secretase [11–13]. Such hypomethylation could theoretically

lead to increased transcription with the net result being elevated expression

of these proteins that appear to be central to the development of Alzheimer’s

disease.

Diet and DNA MethylationDietary folate, methionine and choline are important methyl donors feed-

ing into the methionine cycle, but pyridoxine, vitamin B12, riboflavin and zinc

are also important cofactors for the cycle to operate efficiently. Looking at indi-

vidual nutrients alone may be too simplistic an approach – dietary deficiency

of combinations of nutrients may be important [10]. Other nonnutrient die-

tary components such as tea polyphenols may directly affect the activity of

DNA methyltransferase or its gene [14]. Selenium can also affect methylation

Cobiac 34

status [15]. Thus differences in dietary intakes of these compounds may affect

the methylation status of our DNA, influence genetic expression and hence our

predisposition towards disease. Different polymorphisms in the 5,10-methyl-

enetetrahydrofolate reductase affect cancer risk, providing additional evidence

of the importance of methylation [16] and the relationship between diet and the

methionine cycle [17].

However, there is now also evidence that epigenetic events such as DNA

methylation can be determined prior to birth, so-called fetal programming or

genetic imprinting, and can even be transmitted over generations [18, 19].

Programming of DNA methylation can occur in utero through maternal diet, the

methylation state is then transmitted through mitosis and meiosis with further

interaction possible with postnatal environmental factors causing additional

epigenetic events. The net effect is to ‘reset’ the basal expression of genes and

thus affect disease risk. Fetal malnutrition is associated with an increased risk

of diabetes, metabolic syndrome and heart disease – the Barker hypothesis –

and epigenetic programming during critical times such as germ cell and early

embryo development [20–22] may play a key part in this. The dietary environ-

ment of our ancestors may thus be influencing our genetic expression today.

Histone ModificationThe second common epigenomic mechanism relates to modification of

histone proteins that mediate the folding of DNA into chromatin which sup-

ports and influences genomic transcription. Modifications such as acetylation,

methylation, phosphorylation, ubiquitination, polyADP-ribosylation, sumoyla-

tion and biotinylation appear to determine the transcription of genes and are

referred to as the histone or chromatin code.

DNA forms nucleosome structures where 146 bp of DNA are wrapped

around an octamer of small basic proteins called histones consisting of two

copies of the histone proteins H2A, H2B, H3 and H4. The remaining bases – the

linker DNA – link the nucleosomes, and can vary in length from 8 to 114 bp.

This variation is species specific, but variation in linker DNA length has also

been associated with the developmental stage of the organism or specific

regions of the genome.

Histone proteins consist of a globular C-terminal domain and a flexible

N-terminal tail. The amino terminus protrudes from the nucleosomal surface –

lysines, arginines, serines and glutamates in the N-terminus are targets for the

modifications mentioned above. Multiple residues can be modified on one his-

tone and one residue can be modified in different ways. For example, lysine can

be either acetylated or methylated. Although considerable research has been

undertaken on modifications to the histone tails, there is also evidence that the

globular domains can be modified [23]. Within a histone, different modifications


can prevent or enhance further modifications and modifications on different

histones can interact.

Chromosomes consist of sections that are transcriptionally active or non-

active. If the chromatin is highly condensed (heterochromatin), these areas are

generally transcriptionally inactive whereas if the chromatin is more open, less

condensed (euchromatin), it tends to be transcriptionally active. Modification

of the histones can alter the chromatin structure and influence transcription

activity either positively or negatively. Acetylation of histones is associated

with transcription and methylation with transcription repression. The histones

act as binding sites for effector proteins (such as heterochromatin protein 1) and

modification will thus alter the ability of the effector proteins to facilitate DNA

transcription – the locus of the histones on the chromatin and the specific pat-

terns of histone modifications will influence what transcription complexes are

recruited which will in turn determine the expression profile of specific genes.

The modifications of the histones appear to be reversible and enzymically dri-

ven with many enzymes only recently discovered and probably many more still to

be identified. Some examples include acetyltransferase and deacetyltransferase,

kinases and phosphatases, ubiquitin ligases, poly(ADP-ribose) polymerases, bio-

tinidase and holocarboxylase, methyltransferase and demethylase, and SUMO

(small ubiquitin-related modifier)-specific ubiquitin-like ligase. Histones (and

their modifications) are also important for the regulation of chromosome segrega-

tion during cell division (both meiosis and mitosis) and DNA repair [24].

The two most common epigenetic events, DNA methylation and histone

modification, can interact and appear to be interdependent in some instances

such that methylation of DNA can trigger local histone modifications [6, 25].

Evidence is accumulating on the role of the histone or chromatin code in

the etiology of several chronic diseases. Cancer is associated with global

changes in histone modifications, for example loss of monoacetylation and

trimethylation of H4 [26] and clinical trials with histone deacetylase (HDAC)

inhibitors as anticancer agents show promise [27]. Furthermore, there is evi-

dence that histone modifications may play a role in the development of

Alzheimer’s disease and some mental disorders [28, 29] and diabetes [30].

Diet and Histone ModificationExamples of dietary influences on the histone code include HDAC

inhibitors such as the short-chain fatty acid butyrate from resistant starch or

fiber fermentation, diallyl sulfide from garlic and sulforaphane [31]. These

dietary compounds may thus result in hyperacetylation of histones, altering

chromatin structure by opening it up and allowing transcription which may lead

to re-expression of silenced genes such as tumor suppressor genes and activa-

tion of proapoptotic genes so that the cells can reduce damaged cells exposed to

Cobiac 36

environmental insults. Although the exact mechanisms are still to be delineated,

some HDAC inhibitors, via hyperacetylation of histones, can activate the tran-

scription factors such as NF�B which can derepress such genes as p21 and

BAX and thus potentiate apoptosis selectively in cancer cells which appear to

be more sensitive to HDAC inhibitors than noncancerous cells [31].

Other nutrients that may influence histone modification include folate,

choline, methionine, vitamins B6 and B12, riboflavin and zinc which are all nec-

essary for the production of methyl donors for histone methylation; biotin which

is a cofactor in fatty acid synthesis by carboxylase enzymes and is required for

histone biotinylation, and tryptophan and niacin which supply NAD� for

polyADP-ribosylation of histones (NAD� is also required for class III HDACs).

Genomic InstabilityFrom conception, our DNA is potentially a target for damage which may

lead to genomic instability and to genome damage which can increase the risk

of some diseases, for example colorectal cancer [9]. Dietary factors may play a

role in the prevention of DNA damage and its repair. For example, when methyl

group donors are low or unavailable, uracil is incorporated into DNA instead of

thymine which can lead to a mutagenic lesion by predisposing that region to the

formation of double-strand breaks in the DNA [32]. In other words, there is

damage to the DNA.

There are many forms of genome damage mainly caused by faulty DNA

metabolism and repair which can be linked to nutrient deficiency, oxidative

stress and excess calories. Strand breaks in DNA, DNA misrepair and mitotic

malfunction result from chromosomal breakage, loss, translocation, amplifica-

tion, apoptosis and necrosis [32]. Genomic damage markers such as micro-

nuclei have been associated with increased cancer incidence as well as with

other degenerative diseases such as Alzheimer’s disease, Parkinson’s disease,

diabetes and vascular disease [33–35].

Micronutrients such as folate, nicotinic acid and calcium appear to protect

against genome damage whereas others (e.g. biotin) may not be protective and

may even elevate genomic damage [32]. More research is needed to fully under-

stand the associations between the different measures of genomic damage, the

actual risk of developing specific diseases and the identification of food com-

ponents that are potentially protective.

Diet and Nutrition Considerations/Impacts

In order to identify foods, nutrients or bioactives that can prevent or delay

the onset of disease in population groups, it will be critical to understand the


effect of food components on gene expression and genome stability as well as

the impact of genetic diversity within that population. If we wish to personalize

a dietary approach for an individual, we will need to determine the levels of

genetic damage, the capability and extent of genetic repair, as well as obtain a

measure of that person’s genetic background and the extent of epigenetic

changes [36].

However, nutrition or diet is an exceedingly complex environmental factor

even before we consider the complex interactions with the genome. Not only is

there a vast range of foods and beverages consumed both across and within

populations, but there are differences in food preparation and processing, vary-

ing amounts of protective or potentially damaging bioactives present with vary-

ing bioavailabilities in foods and nutrient-nutrient interactions within the

human body. Nutrients or bioactives can affect carcinogen metabolism (e.g.

fiber), DNA damage (e.g. heterocyclic amines) and repair (e.g. folate), cell

cycle and apoptosis (e.g. butyrate, n–3 fatty acids), proliferation (e.g. resvera-

trol) and signal transduction (e.g. phytoestrogens) [32, 37–40].

Add to this the activity of 1–2 kg of bacteria per person of around 500

species of colonic microflora, many of which have not yet been identified, with

their nutrient-gene and potential gene-gene interactions and it becomes a sig-

nificant science challenge to understand the genetics of the human host and the

profile of the gut microflora, interspecies interaction, cross-feeding, quorum

sensing, and their interaction with dietary substrates, the products released and

their relationships with the host’s mucosal layer and immune system to develop

dietary approaches to improve health and lower disease risk. One example of

the link between microflora and epigenetic effects is that there are butyrate-

producing bacteria in the colon. Butyrate has been shown to affect histone

acetylation and promote apoptosis thus influencing gene expression and elimi-

nation of cells with DNA damage, respectively. There are different pathways to

producing butyrate and it is feasible that not all butyrate-producing bacteria

have yet been identified.

As we think strategically about developing foods that may be protective

against disease, we may need to become more innovative with delivery mecha-

nisms, in order to deliver bioactives to targeted sites of the intestine where their

release will maximize the physiological response. Two examples are (1)

microencapsulation [41] that can protect the bioactive until a trigger releases its

payload at the desired site and (2) butyrate starch ester that resists some upper

intestinal digestion and reaches the colon where the butyrate is released where

it may help to promote apoptosis and so potentially minimize the risk of devel-

oping colorectal cancer [42].

We really cannot consider diet or food without considering cultural issues

and these are also linked to health outcomes. We need to understand dietary

Cobiac 38

behaviors, lifestyle factors and food choices. There may be genetic differences

in sensory sensitivities that will influence what is consumed. We need to under-

stand attitudes and beliefs that might impinge on the acceptance of these new

biomic technologies and foods developed from our greater understanding of

nutrition at the molecular level. Ethical issues are likely to be raised as well as

whether or not there will be public health benefits from this approach in addi-

tion to privacy issues around acquiring genetic information on individuals and

who has access to this information [43]. Education and communication to the

consumer will be important.

Most countries are establishing evidence-based dietary guidelines and rec-

ommended nutrient reference values. By their very nature, they are targeted

towards general recommendations and populations. How do we integrate the

finding that there are specific subgroups in the population that may be more or

less sensitive to nutrients or bioactives in the diet? Do we need to revisit some

of the nutrient reference values where in some instances it appears that there are

much higher levels of for example folate needed to maintain genome stability

[44]?

Nutrients may have a protective effect on one pathway and a nonprotective

effect on another and these complex interplays need to be understood. Further-

more, nutrients or bioactives may be protective at one level and nonprotective

at a higher level and this effect may depend on the genetic background of the

individual.

What is key for all of this is that the appropriate human clinical trials must

be undertaken before we can make clear recommendations. With clinical trials

we need the right biomarkers that are meaningful, that are valid, measurable and

changed with dietary intervention. They need to be suitably specific for differ-

ent disease states, validated to represent an increased risk of disease or ill-

health, and sensitive enough to be perturbed by relatively modest changes in

dietary intakes.

If we are to begin to tailor diets for specific genetic backgrounds of indi-

viduals, we need to be able to have the appropriate cost-effective diagnostics

and the accompanying capabilities and technologies to predict disease suscepti-

bility and identify individuals who need more or less of particular nutrients or

bioactives.

The inherent complexity of this approach is significant and potentially

problematic. One enormous challenge is to understand holistically the role of

nutrition and epigenetics in polygenic diseases such as obesity, cancer, diabetes –

this increases the complexity significantly and is one of the challenges of the

future.

There is a range of advanced technologies and the emergence of new

knowledge that will increase our understanding. Advanced imaging – including


molecular imaging – will provide new insights [45], as will understanding the

role of miRNAs or siRNAs in controlling gene expression in obesity and col-

orectal cancer for example and the links to nutrition [46]. We need to ensure

there is an appropriate use of bioinformatics on the array data being generated

which can be highly variable, produce thousands of variables, the results of

which are highly dependent on experimental technique.

Conclusion

Dietary intake is a complex social and bioactive mixture where more of a

potentially protective nutrient is not always better. Different genetic polymor-

phisms affect disease risk and dietary recommendations may need to be modified

accordingly. Individual measures of genetic variation, epigenetic changes and

genome stability, nutrient metabolism and gut flora populations may mean that

we should have tailored preventative diets, especially if our diet is likely to affect

gene expression and hence health outcomes of our successors. We may need spe-

cific nutrigenomic approaches for the prevention and even treatment of different

disease states. One thing that is emerging is that it is increasingly unlikely that one

set of dietary recommendations is going to suit all of us in the future.

Acknowledgements

Thanks must go to Trevor Lockett and Shelly Hope for their help with the manuscript.

References

1 Dietz AC, Zheng W, Leff MA, Gross WQ, Wen MA, Doll GH, Xiao GH, Folsom AR, Hein DW:

N-acetyltransferase-2 genetic polymorphism, well-done meat intake and breast cancer risk among

postmenopausal women. Cancer Epidemiol Biomarkers Prev 2000;9:905–910.

2 Svetkey LP, Moore TJ, Simons-Morton DG, Appel LJ, Bray GA, Sacks FM, Ard JD, Mortensen RM,

Mitchell SR, Conlin PR, Kesari M: Angiotensinogen genotype and blood pressure response in the

Dietary Approaches to Stop Hypertension (DASH) study. J Hypertens 2001;19:1949–1956.

3 Simopoulos AP: Omega-6/Omega-3 essential fatty acid ratio and chronic diseases. Food Rev Int

2004;20:77–90.

4 Yu W, Murray NR, Weems C, Chen L, Guo H, Ethridge R, Ceci JD, Evers BM, Thompson EA,

Fields AP: Role of cyclooxygenase 2 in protein kinase C beta II-mediated colon carcinogenesis.

J Biol Chem 2003;278:11167–11174.

5 Morris KV: siRNA-mediated transcriptional gene silencing: the potential mechanism and a possi-

ble role in the histone code. Cell Mol Life Sci 2005;62:3057–3066.

6 Kondo Y, Issa JP: Epigenetic changes in colorectal cancer. Cancer Metastasis Rev 2004;23:29–39.

7 Johanning GL, Heimbergur DC, Piyathilake CJ: DNA methylation and diet in cancer. J Nutr 2002;

132:3814S–3818S.

Cobiac 40

8 Fang MZ, Jin Z, Wang Y, Liao J, Yang GY, Wang LD, Yang CS: Promoter hypermethylation and

inactivation of O(6)-methylguanine-DNA methyltransferase in esophageal squamous cell carcino-

mas and its reactivation in cell lines. Int J Oncol 2005;26:615–622.

9 Raptis S, Bapat B: Genetic instability in human tumors. EXS 2006;96:303–320.

10 Davis CD, Ulthus EO: DNA methylation, cancer susceptibility, and nutrient interactions. Exp Biol

Med (Maywood) 2004;229:988–995.

11 Abdolmalecky HM, Smith CL, Faraone SV, Shafa R, Stone W, Glatt SJ, Tsuang MT: Methylomics

in psychiatry: modulation of gene-environment interactions may be through DNA methylation.

Am J Med Genet B Neuropsychiatr Genet 2004;127:51–59.

12 Fuso A, Seminara L, Cavallaro RA, D’Anselmi F, Scarpa S: S-adenosylmethionine/homocysteine

cycle alterations modify DNA methylation status with consequent deregulation of PS1 and BACE

and beta-amyloid production. Mol Cell Neurosci 2005;28:195–204.

13 West RL, Lee JM, Maroun LE: Hypomethylation of the amyloid precursor protein gene in the

brain of an Alzheimer’s disease patient. J Mol Neurosci 1995;6:141–146.

14 Fang MZ, Wang Y, Ai N, Hou Z, Sun Y, Lu H, Welsh W, Yang CS: Tea polyphenol (–)epigallocatechin-

3-gallate inhibits DNA methyltransferase and reactivates methylation-silenced genes in cancer cell

lines. Cancer Res 2003;63:7563–7570.

15 Ferguson LR, Karunasinghe N, Philpott M: Epigenetic events and protection from colon cancer in

New Zealand. Environ Mol Mutagen 2004;44:36–43.

16 Curtin K, Bigler J, Slattery ML, Caan B, Potter JD, Ulrich CM: MTHR C667T and A1298C poly-

morphisms: diet, estrogen and risk of colon cancer. Cancer Epidemiol Biomarkers Prev 2004;13:

285–292.

17 Bailey LB: Folate methyl-related nutrients, alcohol and the MTHR 677C-T polymorphism affect

cancer risk: intake recommendations. J Nutr 2003;133:3748S–3753S.

18 Pembrey ME, Bygren LO, Kaati G, et al: Sex-specific, male-line transgenerational responses in

humans. Eur J Hum Genet 2006;14:159–166.

19 Rakyan VK, Chong S, Champ ME, et al: Transgenerational inheritance of epigenetic states in the

murine AxinFU allele occurs following maternal and paternal transmission. Proc Natl Acad Sci

USA 2003;100:2538–2543.

20 McMillen IC, Robinson JS: Developmental origins of the metabolic syndrome: prediction, plas-

ticity and programming. Physiol Rev 2005;85:571–633.

21 Kelly TLJ, Trasler JM: Reproductive epigenetics. Clin Genet 2004;65:247–260.

22 Gallou-Kabani C, Junien C: Nutritional epigenomics of metabolic syndrome: new perspective

against the epidemic. Diabetes 2005;54:1899–1906.

23 Cosgrove MS, Wolberger C: How does the histone code work? Biochem Cell Biol 2005;83:

468–476.

24 Fernandez-Capetillo O, Lee A, Nussenzweig M, Nussenzweig A: H2AX: the histone guardian of

the genome. DNA Repair (Amst) 2004;3:959–967.

25 Fahrner JA, Eguchi S, Herman JG, Baylin SB: Dependence of histone modifications and gene

expression on DNA hypermethylation in cancer. Cancer Res 2002;62:7213–7218.

26 Fraga MF, Ballestar E, Villar-Garea A, Boix-Chornet M, Espada J, Schotta T, et al: Loss of acety-

lation at Lys16 and trimethylation at Lys20 of histone H4 is a common hallmark of human cancer.

Nat Genet 2005;37:391–400.

27 Acharya MR, Sparreboom A, Venitz J, Figg WD: Rational development of histone deacetylase

inhibitors as anticancer agents: a review. Mol Pharmacol 2005;68:917–932.

28 Mattson MP: Methylation and acetylation in nervous system development and neurodegenerative

disorders. Ageing Res Rev 2003;2:329–342.

29 Simoni MV, Camargo LM, Dong E, Maloku E, Veldic M, Costa E, Guidotti A: The benzamide

MS-275 is a potent, long-lasting brain region-selective inhibitor of histone deacetylases. Proc Natl

Acad Sci USA 2006;103:1587–1592.

30 Gray ST, DeMeyts P: Role of histone and transcription factor acetylation in diabetes pathogenesis.

Diabetes Metab Res Rev 2005;21:416–433.

31 Dashwood RH, Myzak MC, Ho E: Dietary HDAC inhibitors: time to rethink weak ligands in can-

cer chemoprevention? Carcinogenesis 2006;27:344–349.


32 Fenech M: The Genome Health Clinic and genome health nutrigenomics concepts: diagnosis and

nutritional treatment of genome and epigenome damage on an individual basis. Mutagenesis

2005;20:255–269.

33 Bonassi S, Hagmar L, Stromberg U, et al: Chromosomal aberrations in lymphocytes predict

human cancer independently of exposure to carcinogens. Cancer Res 2000;60:1619–1625.

34 Migliore L, Botto N, Scarpato R, Petrozzi L, Cipriani G, Bonuccelli U: Preferential occurrence of

chromosome 21 malsegregation in peripheral blood lymphocytes of Alzheimer’s disease patients.

Cytogenet Cell Genet 1999;87:41–46.

35 Migliore L, Scarpato R, Coppede F, Petrozzi L, Bonucelli U, Rodilla V: Chromosome and oxida-

tive damage biomarkers in lymphocytes of Parkinson’s disease patients. Int J Hyg Environ Health

2001;204:61–66.

36 Mathers JC: The biological evolution – towards a mechanistic understanding of the impact of diet

on cancer risk. Mutat Res 2004;551:43–49.

37 Reddy L, Odhav B, Bhoola KD: Natural products for cancer prevention: a global perspective.

Pharmacol Ther 2003;99:1–13.

38 Schneider Y, Vincent F, Duranton B, Badolo L, Gosse F, Bergmann C, Seiler N, Raul F: Anti-

proliferative effect of resveratrol, a natural component of grapes and wine, on human colonic

cancer cells. Cancer Lett 2000;158:85–91.

39 Augenlicht LH, Mariadason JM, Wilson A, Arango D, Yang W, Heerdt BG, Velcich A: Short chain

fatty acids and colon cancer. J Nutr 2002;132:3804S–3808S.

40 Ng Y, Barhoumi R, Tjalkens RB, Fan YY, Kolar S, Wang NY, Lupton JT, Chapkin RS: The role of

docosahexaenoic acid in mediating mitochondrial membrane lipid oxidation and apoptosis in

colonocytes. Carcinogenesis 2005;26:1914–1921.

41 Augustin MA, Sanguansri L, Bode O: Maillard reaction products as encapsulants for fish oil pow-

ders. J Food Sci 2006;71:E25–E32.

42 Annison G, Illman RJ, Topping DL: Acetylated, proprionylated and butyrylated starches raise

large bowel short-chain fatty acids preferentially when fed to rats. J Nutr 2003;133:3523–3528.

43 Chadwick R: Nutrigenomics, individualism and public health. Proc Nutr Soc 2004;63:161–166.

44 Fenech M: Recommended dietary allowances (RDAs) for genomic stability. Mutat Res 2001;480:

51–54.

45 Moskaug JO, Carlsen H, Myhrstad M, Blomhoff R: Molecular imaging of the biological effects of

quercetin and quercetin-rich foods. Mech Ageing Dev 2004;125:315–324.

46 Wynter CV: The dialectics of cancer: a theory of the initiation and development of cancer through

errors in RNAi. Med Hypotheses 2006;66:612–635.

Dr. Lynne Cobiac

Dept. of Nutrition and Dietetics

Flinders University

GPO Box 2100, Adelaide 5001 (Australia)




Early Nutrition: Impact on Epigenetics

John C. Mathers

Human Nutrition Research Centre, School of Clinical Medical Sciences,

University of Newcastle, Newcastle, UK

AbstractBackground/Aims: (1) To outline the findings that alterations in nutrition in utero and

in early postnatal life influence health in later life. (2) To review the evidence that alterations in

epigenetic markings may be a means by which the genome records environmental (including

nutritional) exposure resulting in changes in gene expression and cell function which underlie

susceptibility to disease. Methods: Literature review. Results: There is strong evidence that

low birth weight, especially when followed by accelerated growth in childhood and greater cen-

tral adiposity in adulthood, is a risk factor for a range of common diseases including cardiovas-

cular disease and type 2 diabetes. Such observations provide the basis for the ‘programming’

hypothesis and present a challenge to discover the mechanisms by which nutritional insults in

early life are received, recorded, remembered and then revealed in later life. Emerging evidence

suggests that alterations in epigenetic marking of the genome may be a key mechanism by

which nutritional exposure in utero can influence gene expression, and therefore, phenotype.

Conclusion: Early life nutrition has the potential to change chromatin structure, to alter gene

expression and to modulate health throughout the life course. Whether later interventions can

reverse adverse epigenetic markings remains to be discovered.


The Developmental Origins of Adult Disease Hypothesis

McCance [1] was the pioneer in this field with his observations that under-

nutrition in the early life of experimental animals (rats and pigs) had lifetime

effects. However, these observations had little impact on human nutrition until

Barker [2] and others found that low birth weight (LBW) is associated with

increased risk of coronary heart disease, stroke, hypertension and type 2 dia-

betes (T2D). Whilst LBW may also increase the risk of other diseases, e.g.

osteoporosis [3], and appears to reduce longevity [4], there is support for the

idea that higher birth weight may increase the risk of some cancers [5].

Nutrition and Epigenetics 43

However, postnatal events may modify the influence of poor prenatal develop-

ment. For example, evidence is accumulating that the increased risk of coronary

heart disease, metabolic syndrome and associated diseases in those born small

is greater in those who show childhood catch-up growth [6] or who become

obese in later life [7]. In addition, Lucas et al. [8] have shown that both the

quantity and quality of infant nutrition may have long-term effects on cognitive

function. From studies in rural Gambia, Moore et al. [9] reported that the sea-

son of birth (a surrogate for the extent of maternal undernutrition) predicted

mortality several years later and suggested that undernutrition in utero may

have profound long-term effects on immune defences.

Such research has reinvigorated interest in the impact of early life nutrition

on health throughout the life course and has initiated a paradigm shift in under-

standing of the role of nutrition in determining health. Importantly, aspects of

these epidemiological observations have been reproduced in animal models

exposed to various forms of undernutrition in utero including low-protein [10]

and low-iron [11] maternal diets. These and other observations underpin the

concept of ‘programming’, a term which describes the impact of a stimulus or

insult during a critical or sensitive time window resulting in long-term struc-

tural and/or functional changes in the organism [12]. Programming is an exam-

ple of developmental plasticity, i.e. that a given genotype can give rise to

different phenotypes depending on environmental conditions [13].

If the impact on long-term health of poor nutrition during development

resulting in LBW is exacerbated by overnutrition in childhood, then the popula-

tions at greatest risk will be those with a high prevalence of LBW which

are undergoing a nutritional transition to overnutrition. Several countries of

Southeast Asia have a high LBW prevalence (defined as birth weight less than

2.5 kg) but also an increasing incidence of T2D associated with increasing obe-

sity rates [14]. The risk of T2D appears to be amplified in those born small who

experience accelerated childhood growth and who have greater abdominal adi-

posity in adulthood [14].

Overview of Epigenetics

Epigenetics describes changes to the genome which are inherited from one

cell generation to the next which alter gene expression but which do not involve

changes in the primary DNA sequence. The main features of epigenetic mark-

ing of the genome are (1) DNA methylation and (2) histone ‘decoration’.

Changes in DNA methylation are an essential part of normal development. In

the early embryo, there is a wave of DNA demethylation after fertilization fol-

lowed by a wave of de novo methylation upon embryo implantation [15–17].

Mathers 44

DNA methylation is responsible for Y chromosome inactivation, for determin-

ing which of the parental alleles is expressed in the case of imprinted genes [18]

and for the regulation of expression of particular genes in particular cell types

[19]. A proportion of the cytosine residues is modified after translation by

attachment of a methyl group to position 5 on the cytosine ring. Such methy-

lated cytosines are usually found where the cytosine is next to a guanine

residue, i.e. in a CpG dinucleotide. In about half the genes in the human

genome, unmethylated CpGs are found clustered at the 5� ends of genes in

domains known as CpG islands. When the CpGs in such islands are unmethy-

lated, gene transcription proceeds normally but when some or all of the CpGs

become methylated, the genes are switched off.

There is extensive covalent modification (by methylation, acetylation,

phosphorylation and ubiquitination) of the amino-terminal tails of histones that

protrude from the globular nucleosome core. Emerging evidence suggests that

this histone ‘decoration’ is the basis for a histone code which extends considerably

the information potential of the genetic (DNA) code [20]. Current hypotheses

suggest that the pattern of histone modifications in any region of the genome

together with the recruitment of several proteins alter the chromatin structure

and DNA methylation and thus control the access to the associated DNA by the

proteins necessary for transcription [16]. A range of nutrients may regulate

gene expression by altering histone post-translational modifications and/or

DNA methylation [21].

Malleability of Epigenetic Markings

Although monozygotic twins are genetically identical, in many cases there

are considerable phenotypic differences between members of such monozygotic

pairs including differences in disease experience [22]. In a recent elegant study

of epigenetic markings in monozygotic twins, Fraga et al. [23] observed that in

early life the twins are epigenetically indistinguishable as assessed by global

and locus-specific differences in DNA methylation and in histone acetylation.

However, with age there was increasing discordance in epigenetic marking and,

equally important, associated divergence in gene expression patterns [23].

The nature of the environmental signals which are capable of inducing

changes in epigenetic markings is very poorly understood, but it seems likely

that a wide range of exposures can have such effects. For example, they may

include a range of maternal behaviours. Liu et al. [24] observed that the off-

spring of rat dams which were more attentive to their pups through licking,

grooming and arched-back nursing show more modest hypothalamic-pituitary-

adrenal responses to stress. More recently, Weaver et al. [25] demonstrated that


these differences in stress response were due to stable alterations in DNA

methylation (altered cytosine methylation within the promoter of the glucocor-

ticoid receptor) and chromatin structure in the hippocampus of the offspring

and that the epigenomic markings were reversible by treatment with the histone

deacetylase inhibitor trichostatin A. These findings provide the first evidence

that maternal behaviour in the early postnatal period may have profound effects

on stress responses throughout life in the offspring and that these phenotypic

changes are mediated by changes in gene expression due to stable alterations in

DNA methylation and chromatin structure [25].

A number of dietary components may influence DNA methylation and

gene expression. For example, molecular modelling studies have shown that the

tea polyphenol (�)-epigallocatechin-3-gallate (EGCG) fits into the catalytic

pocket of DNA methyltransferase 1 (DNMT1) [26] and in doing so acts as a

competitive inhibitor. DNMT1 is the ‘house-keeping’ enzyme which ensures

that patterns of DNA methylation are copied to the daughter strands each time

DNA is replicated. Fang et al. [26] have shown that EGCG suppresses DNMT1

activity with a Ki of 6.89 �M and that treatment of tumour cell lines with EGCG

results in a time- and dose-dependent demethylation of CpG islands in the pro-

moters of several cancer-related genes. Importantly this reversal of aberrant

methylation resulted in re-expression of the silenced genes [26].

Impact of Dietary Factors in Early Life on Epigenetic Markings

Intra-uterine growth retardation (IUGR) can be induced experimentally in

rats by bilateral uterine artery ligation at day 19 of gestation and results in ani-

mals born at term which are significantly smaller than normal but without any

change in litter size [27]. The IUGR rats have altered tissue structure and func-

tion including decreases in kidney glomeruli number [28] and reduced small

intestinal growth [29]. Although IUGR is a transient prenatal insult, the effects

persist into adult life and have been shown to be associated with altered one-

carbon metabolism (as illustrated by significantly increased concentrations of

homocysteine and S-adenosylhomocysteine) which may be responsible for

changes in DNA methylation and histone acetylation [27].

More direct evidence of the effects of alterations in one-carbon metabolism

during pregnancy was provided by the study by Waterland and Jirtle [30] in which

they gave dietary supplements of methyl group donors and cofactors (folic acid,

vitamin B12, choline and betaine) to viable yellow agouti (Avy) mice for 2 weeks

before mating and throughout pregnancy and lactation. Dietary supplementa-

tion of the dams resulted in phenotypic alterations in the offspring with a higher

proportion having a darker coat colour (mottled and pseudoagouti) [30]. This

Mathers 46

coat colour change was due to silencing of the expression of the agouti gene as

a result of increased CpG methylation at the Avy locus [30]. The authors specu-

lated that this shift in epigenotype may have occurred during early embryonic

development and demonstrated that the altered epigenetic marking observed at

weaning (21 days) was stable for several months [30].

Rats fed a low-protein (90 g casein/kg) diet during pregnancy produce

smaller offspring which develop increased blood pressure in adulthood [10].

Such maternal undernutrition significantly increased the expression of the glu-

cocorticoid receptor and of the peroxisome proliferator-activated receptor-� in

the liver of offspring, which was associated with significantly reduced methyla-

tion of CpGs in the promoter regions of the same genes [31]. Interestingly, these

alterations in epigenetic marking and gene expression were prevented if the

low-protein diet was supplemented with folic acid [31]. The limited evidence to

date suggests that these effects of altered maternal nutrition on DNA methyla-

tion and gene expression in the offspring appeared to be gene specific since the

nutritional manipulations had no effect on peroxisome proliferator-activated

receptor-� [31]. These observations suggest that altered one-carbon metabolism

may play a central role in the molecular mechanisms through which IUGR is

sensed by the cell. In addition, the emerging evidence suggests that cellular

‘memory’ of such developmental insults may be encoded in altered epigenetic

marking of the genome and may be manifested as altered phenotype later in life

through changes in the pattern of gene expression.

Conclusions and Suggestions for Further Research

It is now clear that LBW, especially when followed by accelerated growth

in childhood and greater central adiposity in adulthood, is a risk factor for a

range of common diseases. Whilst maternal undernutrition remains a serious

issue throughout the world, an increasing proportion of mothers are overweight

or obese and research is needed to ascertain whether the offspring of such

mothers carry a long-term health penalty. In addition to the public health impli-

cations, the demonstration of ‘programming’ by nutrition in early life raises

many fundamental biological questions. These include the mechanisms by

which the early life experiences are received, recorded, ‘remembered’ and then

revealed in later life. If, as seems likely, changes in epigenetic marking of the

genome play a role in this process then it will be important to discover (1) the

critical ‘windows’ when epigenetic markings are most susceptible to perturba-

tion, (2) which nutrients or other food constituents can influence epigenetic

markings, (3) what DNA domains are susceptible to altered epigenetic marking

by nutritional factors, (4) whether alterations in epigenetic markings occurring


in early life are reversible by nutritional or other interventions in later life and

(5) the molecular mechanisms by which nutrition can influence epigenetic

markings. The latter will require a much better understanding of the linkages

between nutrient-sensitive signalling pathways and chromatin structure and

function.

Acknowledgements

Research in my laboratory on early nutrition and its impact on epigenetics is funded by

the World Cancer Research Fund (2001/37) and by the Biotechnology and Biological

Sciences Research Council through the Centre for Integrated Systems Biology of Ageing and

Nutrition (CISBAN) (BB/C008200/1).

References

1 McCance RA: Food, growth and time. Lancet 1961;ii:671–676.

2 Barker DJP: The developmental origins of well-being. Philos Trans R Soc Lond B Biol Sci 2004;359:

1359–1366.

3 Yarborough DE, Barrett-Connor E, Morton DJ: Birth weight as a predictor of adult bone mass in

postmenopausal women: the Rancho Bernardo Study. Osteoporos Int 2000;11:626–630.

4 Andersen AM, Osler M: Birth dimensions, parental mortality, and mortality in early adult age. Int

J Epidemiol 2004;33:92–99.

5 Ahlgren M, Melbye M, Wohlfahrt J, Sorensen TIA: Growth patterns and risk of breast cancer in

women. N Engl J Med 2004;351:1619–1626.

6 Eriksson JG, Forsen T, Tuomilehto J, Winter PD, Osmond C, Barker DJP: Catch-up growth in

childhood and death from coronary heart disease: longitudinal study. BMJ 1999;318:427–431.

7 Frankel S, Elwood P, Sweetnam P, Yarnell J, Davey Smith G: Birthweight, body-mass index in

middle age and death from coronary heart disease. Lancet 1996;348:1478–1480.

8 Lucas A, Morley R, Cole TJ: Randomised trial of early diet in preterm babies and later intelligence

quotient. BMJ 1998;317:1481–1487.

9 Moore SE, Cole TJ, Poskitt EM, Sonko BJ, Whitehead RG, McGregor IA, Prentice AM: Season of

birth predicts mortality in rural Gambia. Nature 1997;388:434.

10 Langley SC, Jackson AA: Increased systolic blood pressure in adult rats induced by fetal exposure

to maternal low protein diets. Clin Sci (Lond) 1994;86:217–222.

11 Gambling L, Dunford S, Wallace DI, Zuur G, Solanky N, Srai SK, McArdle HJ: Iron deficiency

during pregnancy affects postnatal blood pressure in the rat. J Physiol 2003;552:603–610.

12 Lucas A: Programming by early nutrition in man. Ciba Found Symp 1991;156:38–55.

13 Bateson P, Barker D, Clutton-Brock T, Deb D, D’Udine B, Foley RA, Gluckman P, Godfrey K,

Kirkwood T, Lahr MM, McNamara J, Metcalfe NB, Monaghan P, Spencer HG, Sultan SE:

Developmental plasticity and human health. Nature 2004;430:419–421.

14 Yajnik CS: Early life origins of insulin resistance and type 2 diabetes in India and other Asian

countries. J Nutr 2004;134:205–210.

15 Reik W, Dean W, Walter J: Epigenetic reprogramming in mammalian development. Science 2001;

293:1089–1093.

16 Robertson KD: DNA methylation and chromatin – Unravelling the tangled web. Oncogene

2002;21:5361–5379.

17 Morgan HD, Santos F, Green K, Dean W, Reik W: Epigenetic reprogramming in mammals. Hum

Mol Genet 2005;14(suppl 1):R47–R58.

Mathers 48

18 Ferguson-Smith AC, Surani MA: Imprinting and the epigenetic asymmetry between parental

genomes. Science 2001;293:1086–1089.

19 Ehrlich M: Expression of various genes is controlled by DNA methylation during mammalian

development. J Cell Biochem 2003;88:899–910.

20 Jenuwein T, Allis CD: Translating the histone code. Science 2001;293:1074–1080.

21 Oommen AM, Griffin JB, Sarath G, Zempleni J: Roles for nutrients in epigenetic events. J Nutr

Biochem 2005;16:74–77.

22 Cardno AG, Rijsdijk FV, Sham PC, Murray RM, McGuffin P: A twin study of genetic relation-

ships between psychotic symptoms. Am J Psychiatry 2002;159:539–545.

23 Fraga MF, Ballestar E, Paz MF, Ropero S, Setien F, Ballestar ML, Heine-Suñer D, Cigudosa JC,

Urioste M, Benitez J, Boix-Chornet M, Sanchez-Aguilera A, Ling C, Carlsson E, Poulsen P, Vaag A,

Stephan Z, Spector TD, Wu Y-Z, Plass C, Esteller M: Epigenetic differences arise during the life-

time of monozygotic twins. Proc Natl Acad Sci USA 2005;102:10604–10609.

24 Liu D, Dioro J, Tannenbaum B, Caldji C, Francis D, Freedman A, Sharma S, Pearson D, Plotsky PM,

Meaney MJ: Maternal care, hippocampal glucocorticoid receptors, and hypothalamic-pituitary-

adrenal responses to stress. Science 1997;277:1659–1662.

25 Weaver ICG, Cervoni N, Champagne FA, D’Alessio AC, Sharma S, Seck JR, Dymov S, Szyf M,

Meaney MJ: Epigenetic programming by maternal behaviour. Nat Neurosci 2004;7:847–854.

26 Fang MZ, Wang Y, Ai N, Hou Z, Sun Y, Lu H, Welsh W, Yang CS: Tea polyphenol (�)-epigallo-

catechin-3-gallate inhibits DNA methyltransferase and reactivates methylation-silenced genes in

cancer cell lines. Cancer Res 2003;63:7563–7570.

27 MacLennan NK, James SJ, Melnyk S, Piroozi A, Jernigan S, Hsu JL, Janke SM, Pham TD, Lane RH:

Uroplacental insufficiency alters DNA methylation, one-carbon metabolism, and histone acetyla-

tion in IUGR rats. Physiol Genom 2004;18:43–50.

28 Pham TD, MacLennan NK, Chiu CT, Laksana GS, Hsu JL, Lane RH: Uteroplacental insufficiency

increases apoptosis and alters p53 gene expression in the full-term IUGR rat kidney. Am J Physiol

Regul Integr Comp Physiol 2003;285:R962–R970.

29 Baserga M, Bertlotto C, MacLennan NK, Hsu JL, Pham T, Laksana GS, Lane RH: Uteroplacental

insufficiency decreases small intestine growth and alters apoptotic homeostasis in term intrauter-

ine growth retarded rats. Early Hum Dev 2004;79:93–105.

30 Waterland RA, Jirtle RL: Transposable elements: targets for early nutritional effects on epigenetic

gene regulation. Mol Cell Biol 2003;23:5293–5300.

31 Lillycrop KA, Phillips ES, Jackson AA, Hanson MA, Burdge GC: Dietary protein restriction of

pregnant rats induces and folic acid supplementation prevents epigenetic modification of hepatic

gene expression in the offspring. J Nutr 2005;135:1382–1386.

Dr. John C. Mathers

Human Nutrition Research Centre, School of Clinical Medical Sciences

William Leech Building, University of Newcastle

Newcastle, NE2 4HH (UK)

Tel. �44 1912226912, Fax �44 1912228943, E-Mail [email protected]



Nutrition and Genome Health

Michael Fenech

CSIRO Human Nutrition, Adelaide, Australia

AbstractThe link between genome damage and adverse health outcomes is compelling. There is

increasing evidence indicating that genome instability, in the absence of overt exposure to

genotoxins, is itself a sensitive marker of nutritional deficiency. We have shown that above-

average intake of certain micronutrients (i.e. calcium, vitamin E, retinol, folate, vitamin B12

and nicotinic acid) is associated with a reduced genome damage rate measured using the

micronucleus assay. Genome health nutrigenomics is an emerging and important new field

of nutritional science because it is increasingly evident that optimal concentration of

micronutrients for the prevention of genome damage is dependent on genetic polymorphisms

that alter the function of genes involved directly or indirectly in DNA repair and metabolism.

Essentially this also means that the dietary ‘nutriome’ (i.e. nutrient profile and composition)

recommendations should be matched to an individual’s functional genome to optimise

genome health maintenance. Development of functional foods and dietary patterns that are

specifically designed to improve genome health maintenance in humans with specific

genetic backgrounds are expected to provide an important contribution to a new health strat-

egy based on the diagnosis and individualised nutritional treatment of genome instability (i.e.

Genome Health Clinics).


The central role of the genetic code in determining health outcomes such

as developmental defects and degenerative diseases such as cancer is well

established. In addition, it is evident that DNA metabolism and repair is depen-

dent on a wide variety of dietary factors that act as cofactors or substrates in

these fundamental metabolic pathways [Ames, 2001; Ames and Wakimoto,

2002; Fenech and Ferguson, 2001]. DNA is continuously under threat of major

mutations from conception onwards by a variety of mechanisms which include

point mutation, base modification due to reactive molecules such as the

hydroxyl radical, chromosome breakage and rearrangement, chromosome loss

or gain, gene silencing due to inappropriate methylation of CpG at promoter

Fenech 50

sequences, activation of parasitic DNA expression due to reduced methylation

of CpG as well as accelerated telomere shortening [Egger et al., 2004; Fenech,

2002, 2005; Rajagopalan and Lengauer, 2004]. It is true to say that all of the

above mechanisms of genome damage occur spontaneously due to the effects

of endogenously generated mutagens and/or due to deficiency in cofactors

required for DNA metabolism and repair and/or exposure to environmental

genotoxins. However, it is also true that genetic defects in DNA metabolism and

repair, the latter involving more than 100 genes in humans [Lindahl and Wood,

1991; Thompson and Schild, 2002], are also a key factor. While much has been

learnt of the genes involved in DNA metabolism and repair and their role in a

variety of pathologies, such as defects in BRCA1 and BRCA2 genes that cause

increased risk for breast cancer [Nathanson et al., 2001; Thompson and Schild,

2002], much less is known of the impact of cofactor and/or micronutrient defi-

ciency on DNA repair. Put simply, a deficiency in a micronutrient required as a

cofactor or as an integral part of the structure of a DNA repair gene (e.g. Zn as

a component of the DNA repair glycosylase OGG1 involved in removal of oxi-

dised guanine or Mg as a cofactor for several DNA polymerases) could mimic

the effect of a genetic polymorphism that reduces the activity of that enzyme

[Ames, 2001, 2003]. Therefore, nutrition has a critical role in DNA metabolism

and repair and this awareness is leading to the development of the new field of

genome health nutrigenomics [Fenech, 2004, 2005].

Evidence Linking Genome Damage with Adverse Health Outcomes

Genome damage impacts on all stages of life. There is good evidence to

show that infertile couples exhibit a higher rate of genome damage than fertile

couples [Trkova et al., 2000] when their chromosomal stability is measured in

lymphocytes using the micronucleus (MN) assay [Fenech, 2000] (fig. 1).

Infertility may be due to a reduced production of germ cells because genome

damage effectively causes programmed cell death or apoptosis which is one of

the mechanisms by which grossly mutated cells are normally eliminated [Hsia

et al., 2003; Narula et al., 2002; Ng et al., 2002]. When the latter mechanism

fails, reproductive cells with genomic abnormalities may survive leading to

serious developmental defects [Liu et al., 2002; Vinson and Hales, 2002].

That an elevated rate of chromosomal damage is a cause of cancer has

been demonstrated by ongoing prospective cohort studies in Italy and the

Scandinavian countries which showed a 2- to 3-fold increased risk of cancer in

those whose chromosomal damage rate in lymphocytes was in the highest ter-

tile when measured 10–20 years before cancer incidence was measured

[Bonassi et al., 2000].

Nutrition and Genome Health 51

Chromosomal damage is also associated with accelerated ageing and neu-

rodegenerative diseases. Several studies have shown that chromosomal abnor-

malities, including MN frequency (fig. 2), increase progressively with age in

somatic cells [Bonassi et al., 2001; Fenech, 1998]. Accelerated ageing and

cancer-prone syndromes, such as progeria, Bloom’s syndrome, Fanconi’s anaemia

and Werner’s syndrome, exhibit increased chromosomal instability and/or

accelerated telomere shortening due to defects in a variety of genes essential for

DNA repair and telomere maintenance such as ATM, PARP, BRCA1, BRCA2and DNA helicases [Joenje and Patel, 2001; Lansdorp, 2000; Shen and Loeb,

2001; Thompson and Schild, 2002]. Equally interesting is the observation that

neurodegenerative diseases such as Alzheimer’s disease and Parkinson’s disease

MN formation – chromosome breakage or loss

Nucleoplasmic bridge – chromosome translocation

Cytochalasin B

block

cytokinesis block

Fig. 1. Expression of MNs and nucleoplasmic bridges during nuclear division. MNs

originate from either (a) lagging whole chromosomes (top panel) that are unable to engage with

the mitotic spindle due to a defect in the spindle, or a defect in the centromere/kinetochore

complex required to engage with the spindle or (b) an acentric chromosome fragment originat-

ing from a chromosome break (top and bottom panel) which lags behind at anaphase because it

lacks a centromere/kinetochore complex. Misrepair of two chromosome breaks may lead to an

asymmetrical chromosome rearrangement producing a dicentric (i.e. two centromeres) chro-

mosome and an acentric fragment (bottom panel) – frequently the centromeres of the dicentric

chromosome are pulled to opposite poles of the cell at anaphase resulting in the formation of a

nucleoplasmic bridge between the daughter nuclei. Nucleoplasmic bridges are frequently

accompanied by an MN originating from the associated acentric chromosome fragment.

Because MNs and nucleoplasmic bridges are only expressed in cells that have completed

nuclear division, it is necessary to score these genome instability biomarkers specifically in

once-divided cells. This is readily accomplished by blocking cytokinesis using cytochalasin B

(for more detailed explanation refer to Thomas et al. [2003] and Fenech [2002]).

Fenech 52

exhibit much higher rates of MN frequency in human peripheral blood lympho-

cytes [Migliore et al., 1999, 2001]. Those individuals with accelerated ageing

syndromes or suboptimal DNA repair may be particularly susceptible to the

genome damaging effects of moderate micronutrient deficiency.

The Concept of Genome Damage as a Marker of Nutritional Deficiency

Spontaneous chromosomal rates (more than one cell in a thousand exhibit-

ing a major chromosomal mutation) are high in humans even in the absence of

overt exposure to known carcinogens and there is a wide variation in rates of

mutation even among individuals of the same age (fig. 2). One therefore has to

consider whether genetic factors and diet may be the main determinants of vari-

ation in background mutation rate. There is overwhelming evidence that several

micronutrients (vitamins and minerals) are required as cofactors for enzymes or

as part of the structure of proteins (metalloenzymes) involved in DNA synthesis

and repair, prevention of oxidative damage to DNA as well as maintenance

methylation of DNA. The role of micronutrients in maintenance of genome sta-

bility has recently been extensively reviewed [Ames and Wakimoto, 2002;

Fenech, 2003, 2005; Fenech and Ferguson, 2001]. Examples of micronutrients

involved in various genome stability processes are given in table 1. The main

point is that genome damage caused by moderate micronutrient deficiency is of

the same order of magnitude as the genome damage levels caused by exposure

18–2

5

26–3

5

36–4

5

46–5

5

56–6

5

66–7

5

76–9

0

0

10

20

30

40

50

Age (years)

MN

per

1,0

00 B

N c

ells

18–2

5

26–3

5

36–4

5

46–5

5

56–6

5

66–7

5

76–9

0

0

10

20

30

40

50

60

70

80

90

100

Age (years)

MN

per

1,0

00 B

N c

ells

a b

Fig. 2. Variation in chromosome DNA damage rates of healthy non-smoking males

(n � 495) (a) and females (n � 511) (b) within and between age groups measured using the

cytokinesis block MN assay. BN � Binucleated; MN � micronuclei.


Table 1. Examples of the role and the effect of deficiency of specific micronutrients on genomic stability

Micronutrient/s Role in genomic stability Consequence of deficiency

Vitamins C and E Prevention of oxidation of DNA and lipid Increased baseline level of

oxidation [Claycombe and Meydani, 2001; DNA strand breaks,

Halliwell, 2001] chromosome breaks and

oxidative DNA lesions and lipid

peroxide adducts on DNA

[Claycombe and Meydani,

2001; Halliwell, 2001]

Folate and vitamins Maintenance methylation of DNA; Uracil misincorporation in

B2, B6 and B12 synthesis of dTMP from dUMP and DNA, increased chromosome

efficient recycling of folate [Fenech, breaks and DNA

2001] hypomethylation [Fenech,

2001]

Niacin, nicotinic acid Required as substrate for poly(ADP- Increased level of unrepaired

ribose) polymerase which is involved in nicks in DNA, increased

cleavage and rejoining of DNA and chromosome breaks and

telomere length maintenance [Boyonoski rearrangements, and

et al., 1999; Hageman and Stierum, sensitivity to mutagens

2001] [Boyonoski et al., 1999;

Hageman and Stierum, 2001]

Zinc, manganese Zn, required as a cofactor for Cu/Zn Increased DNA oxidation,

and selenium superoxide dismutase, endonuclease IV, DNA breaks and elevated

function of p53, Fapy glycosylase and in chromosome damage rate

Zn finger proteins such as poly(ADP- [Ambrosone et al., 1999;

ribose) polymerase [Dreosti, 2001; Dreosti, 2001; El-Bayoumy,

Ho and Ames, 2003]; Mn, required as 2001; Ho and Ames, 2003;

a component of mitochondrial Mn Keen and Zidenberg-Cherr, 1996]

superoxide dismutase [Ambrosone et al.,

1999; Keen and Zidenberg-Cherr, 1996];

Se required as a component of

peroxidases, e.g. glutathione peroxidase

[El-Bayoumy, 2001]

Iron Required as a component of Reduced DNA repair capacity

ribonucleotide reductase and and increased propensity for

mitochondrial cytochromes oxidative damage to

[Walter et al., 2002] mitochondrial DNA

[Walter et al., 2002]

Calcium and Mg, required as cofactor for a variety of Reduced fidelity of DNA

magnesium DNA polymerases, in nucleotide excision replication; reduced DNA

repair, base excision repair and mismatch repair capacity; chromosome

repair; essential for microtubule segregation errors; survival of

polymerisation and chromosome genomically aberrant cells

Fenech 54

to significant doses of environmental genotoxins such as chemical carcinogens,

ultraviolet radiation and ionising radiation. An example from our laboratory is

the observation that the chromosomal damage in cultured human lymphocytes

caused by reducing folate concentration from 120 to 12 nmol/l is equivalent to

that induced by an acute exposure to 0.2 Gy of low linear energy transfer ionis-

ing radiation (e.g. X-rays), a dose of radiation which is approximately ten times

greater than the annual allowed safety limit of exposure for the general popula-

tion [International Atomic Energy Agency, 1986] (fig. 3). The normal folate

concentration in plasma is only 10–30 nmol/l which may be adequate to prevent

anaemia but insufficient to minimise chromosomal damage.

Table 1. (continued)

Micronutrient/s Role in genomic stability Consequence of deficiency

segregation [Hartwig, 2001]; Ca, [Hartwig, 2001; Honda et al.,

plays an important role in chromosome 2004; Xu et al., 2003]

segregation and is required for apoptosis

[Honda et al., 2004; Xu et al., 2003]

dTMP � Deoxythymidylic acid; dUMP � deoxyuridylic acid. For information on other micronutrients (e.g.

carotenoids, vitamin D, polyphenols, and copper) refer to other papers in Fenech and Ferguson [2001].

0

10

20

30

MN

ed B

N c

ells

per

1,0

00 B

Nce

lls

X-raysFolic acid

120 60 24 12 Folic acid (nM)

0 5 10 20 X-rays (rad)

Fig. 3. A comparison of the dose-response effect on MN induction in cytokinesis-

blocked cultured lymphocytes caused by (a) acute exposure to X-rays up to a maximum dose of

20 rad, equivalent to 10 times the annual exposure safety limit for the general public [IAEA,

2001] and (b) folic acid deficiency within the ‘normal’ physiological range of 12–120 nM

concentration. Data are from Fenech and Morley [1986] and Crott et al. [2001a, b].

MNed � Micronucleated; BN � binucleated. Results represent the mean �1 SEM; n � 6 for

X-rays and n � 20 for folic acid experiments. 1 rad � 0.01 Gy.


Results from a Recent Epidemiological Study Suggest that There Are atLeast Nine Micronutrients That Affect Genome Health Maintenance

We have recently reported the results of an epidemiological study on 190

healthy individuals (mean age 47.8 years, 46% males) designed to determine

the association between dietary intake, measured using a food frequency ques-

tionnaire, and genome damage in lymphocytes, measured using the MN assay

(fig. 1). Multivariate analysis of baseline data showed that (a) the highest tertile

of intake of vitamin E, retinol, folic acid, nicotinic acid (preformed) and cal-

cium was associated with significant reductions in MN frequency, i.e., �28,

�31, �33, �46, and �49%, respectively (all p values �0.005), relative to the

lowest tertile of intake and (b) the highest tertile of intake of riboflavin, pan-

tothenic acid and biotin was associated with significant increases in MN fre-

quency, i.e., �36% (p � 0.054), �51% (p � 0.021), and �65% (p � 0.001),

respectively, relative to the lowest tertile of intake (fig. 4). Mid-tertile

�-carotene intake was associated with an 18% reduction in MN frequency

(p � 0.038); however, the highest tertile of intake (�6,400 �g/day) resulted in

an 18% increment in MN frequency. We were interested in investigating the

combined effects of calcium or riboflavin with folate consumption because epi-

demiological evidence suggests that these dietary factors tend to interact in

modifying the risk of cancer [Giovannucci, 2003; Lamprecht and Lipkin, 2003;

Willet, 2001; Xu et al., 2003] and they are also associated with reduced risk of

osteoporosis and hip fracture [Cagnacci et al., 2003; MacDonald et al., 2004;

Sato et al., 2005]. Interactive additive effects were observed such as the protec-

tive effect of increased calcium intake (�46%) and the exacerbating effect of

riboflavin (�42%) on increased genome damage caused by low folate intake.

The results from this study illustrate the strong impact of a wide variety of

micronutrients and their interactions on genome health depending on the level

of intake.

The results concerning folate are consistent with (a) studies showing that

folate deficiency leads to hypomethylation of DNA and excessive incorporation

of uracil into DNA which are two of the known underlying molecular events

that cause chromosomal instability and MN formation [Choi and Mason, 2002;

Crott et al., 2001a and 2001b; Fenech, 2003; Mashiyama et al., 2004] and (b) the

observation from controlled intervention studies that folate intakes greater than

200 �g/day are required for chromosomal stability [Fenech et al., 1998; Fenech,

2001]. To our knowledge, there are no previous data showing an association

between dietary calcium and chromosomal instability. However, calcium plays

an important role in chromosome segregation [Honda et al., 2004; Xu et al.,

2003]; it restrains cell proliferation, and induces apoptosis and cell differentia-

tion, which may explain, in part, why reduced calcium intake is associated with

Fenech 56

increased colorectal cancer risk [Cho et al., 2004; Giovannucci, 2003;

Lamprecht and Lipkin, 2003].

The Concept of a Genome Health ‘Nutriome’

As shown in table 2, the concentration of genome-protective micronutrients

varies greatly between foods and, therefore, one has to start considering foods and

diets in terms of their ‘nutriome’, i.e. their content of genome-protective nutrients

because this will determine which foods and combinations of foods are likely to

be most beneficial for genome health maintenance. For example, certain cereal,

vegetable and dairy foods are particularly rich in those micronutrients that

have been shown in our epidemiological study to be protective against genome

damage. A preliminary food group analysis of the data from the same study sug-

gests that above-average intake of dairy foods, cereals and vegetables are inde-

pendently associated with a 10–20% reduction in genome damage (data not

Vita

min

E

Cal

cium

Fola

te

Ret

inol

Nic

otin

ic a

cid

�-C

arot

ene

Rib

ofla

vin

Pan

toth

enic

aci

d

Bio

tin

�50

�25

0

25

50

75

*

*

** *

*

*

*

*

*

**p�0.006

Per

cent

var

iatio

n in

gen

ome

dam

age

Mid-tertileHighest tertile

Fig. 4. Percent variation in genome damage rate for the mid-tertile and highest tertile

of intake of vitamin E, calcium, folate, retinol, nicotinic acid, �-carotene, riboflavin, pan-

tothenic acid and biotin relative to the lowest tertile of intake. Genome damage rate was mea-

sured in peripheral blood lymphocytes using the cytokinesis block MN assay. For more

information, refer to Fenech et al. [2005].


shown, paper in preparation). However, further detailed analyses and placebo-

controlled trials are required to identify those specific foods (and levels of intake)

with greatest potential for optimising genome health. It is interesting to note that

increased intake of fruits, which are relatively poor in nutrients required for

genome health maintenance (table 2), is associated with increased MN frequency

Table 2. Examples of commonly consumed foods and their content of micronutrients that were found

to be associated with improved genome stability [Fenech et al., 2005]

Calcium Folate Niacin Vitamin E �-Carotene Retinol

mg/100 g �g/100 g mg/100 g mg/100 g �g/100 g �g/100 g

Cereals

Wheat bran 110 260 29 1.6 0 0

Soya flour 210 345 2 1.5 0 0

Wholemeal wheat flour 40 57 6 1.4 0 0

Dairy

Cheddar cheese 800 20 0.1 0.8 205 310

Fresh whole milk 120 5 0.1 0.1 20 35

Parmesan cheese 1,220 20 0.3 0.9 195 325

Nuts

Almonds 250 96 2 20 0 0

Peanuts 40 76 11 5.6 0 0

Walnuts 60 66 1 0.8 0 0

Fish

Sardines (canned) 500 7 7 1.1 Tr Tr

Cod (baked) 11 12 1.5 0.6 Tr Tr

Tuna (canned) 7 15 13 6.3 Tr Tr

Meat

Beef (mince cooked) 20 16 5 0.3 Tr Tr

Chicken (boiled) 10 8 6.5 0.1 Tr Tr

Lamb liver (cooked) 10 240 15 0.3 Tr 26,780

Fruit

Banana 7 22 0.6 0.2 200 0

Orange 30 28 0.2 0.2 40 0

Strawberry 20 20 0.4 0.2 30 0

Vegetables

Broccoli (boiled) 80 110 0.6 1.1 2,500 0

Spinach (boiled) 600 140 0.4 2 6,000 0

Peas (frozen) 30 78 1.5 Tr 300 0

Tr � Trace amounts only. Micronutrient content data from Paul and Southgate [1978].

Fenech 58

in the same study, suggesting that a strong reliance on fruit within a dietary pat-

tern may actually deplete the body of genome-protective nutrients.

Genome Health – A New Paradigm for Recommended Dietary Allowances

Current recommended dietary allowances (RDAs) for vitamins and min-

erals are based largely on the prevention of diseases of deficiency such as

scurvy in the case of vitamin C, anaemia in the case of folic acid and pellagra

in the case of niacin. However, these diseases of deficiency are rare in the

developed world but degenerative disease and developmental disease are very

important. The dietary allowance for folic acid for the prevention of neural

tube defects has been revised to more than double the original RDA [Centers

for Disease Control, 1992]. There is a strong international awareness that it is

also necessary to redefine RDAs for the prevention of degenerative disease

(such as cancer, cardiovascular disease and Alzheimer’s disease) and compres-

sion of the morbidity phase during old age. Because diseases of development,

degenerative disease and ageing itself are partly caused by damage to DNA

[Ames, 1998; Holliday, 1995], it seems logical that we should rather focus our

attention on defining optimal requirements of key minerals and vitamins for

preventing damage to both nuclear and mitochondrial DNA. To date, our

knowledge on optimal micronutrient levels for genomic stability is scanty and

disorganised. Table 1 lists some of the most important minerals and vitamins

required for DNA maintenance and prevention of DNA damage and the

DNA lesions that could be induced by inadequate intake of these antimuta-

genic vitamins.

Supplementation of diet with appropriate minerals and vitamins could, in

some cases, help overcome inherited metabolic blocks in key DNA mainte-

nance pathways. A good example are the recent studies on optimisation of

folate and vitamin B12 status for genome health maintenance. Both in vitro and

in vivo studies with human cells clearly show that folate deficiency, vitamin

B12 deficiency and elevated plasma homocysteine are associated with expres-

sion of chromosomal fragile sites, chromosome breaks, excessive uracil in

DNA, MN formation and DNA hypomethylation (table 2) [Blount and Ames,

1995; Blount et al., 1997; Cravo et al., 1994; Crott et al., 2001a, b; Duthie and

Hawdon, 1998; Fenech et al., 1998; Jacky et al., 1983; Jacob et al., 1998;

Titenko-Holland et al., 1998]. It is notable that four of eight known human gly-

cosylases are dedicated to the removal of uracil from DNA, the mutation

caused by folate deficiency [Lindahl and Wood, 1999]. In vitro experiments

indicate that DNA breaks in human cells are minimised when folic acid


concentration in culture medium is greater than 180 nmol/l [Duthie and

Hawdon, 1998; Jacky et al., 1983]. Recently, we have shown that uracil in

DNA, chromosome breakage, chromosome rearrangement and gene amplifi-

cation in human lymphocytes cultured for 9 days are minimised at a folic acid

concentration of 120 nmol/l [Crott et al., 2001a, b]. Intervention studies in

humans taking folate and/or vitamin B12 supplements show that DNA

hypomethylation, chromosome breaks, uracil misincorporation and MN for-

mation are minimised when plasma concentration of vitamin B12 is greater

than 300 pmol/l, plasma folate concentration is greater than 34 nmol/l, red cell

folate concentration is greater than 700 nmol/l folate and plasma homocysteine

is less than 7.5 �mol/l [Blount and Ames, 1995; Blount et al., 1997; Cravo

et al., 1994; Fenech et al., 1998; Jacob et al., 1998; Titenko-Holland et al.,

1998]. These concentrations are only achievable at intake levels in excess of

current RDAs, i.e. more than 400 �g folic acid per day and more than 2 �g

vitamin B12 per day. Dietary intakes above the current RDA may be particu-

larly important in those with defects in the absorption and metabolism of these

vitamins, for which aging is a contributing factor. For example, it has recently

been shown that vitamin B12 (active corrinoid) bioavailability is significantly

reduced in Alzheimer’s disease patients suggesting a higher requirement for

vitamin B12 in these individuals [McCaddon et al., 2001]. The defect in utilis-

ing vitamin B12 could explain the significantly elevated DNA damage rate

(MN frequency) observed in both sporadic and familial Alzheimer’s disease

patients [Trippi et al., 2001] because MN frequency is significantly related to

vitamin B12 status [Fenech et al., 1998; Fenech, 2001].

It is important to note the conflict between the traditional RDA and the

new genome health nutrigenomics concepts. The former is designed as a rec-

ommendation for 95% of the population to prevent deficiency of a particular

nutrient while the latter is intended to provide recommendations on an individ-

ual basis by matching the nutriome to the genome to optimise genome health

with the expected outcome of improved well-being throughout life and com-

pressing morbidity in the ‘twilight’ years of life.

Genome Health Nutrigenomics

One of the important emerging areas of nutrition science is the field of

nutrigenomics, i.e. the effect of diet on gene expression and chromosomal

structure and the extent to which genetic differences between individuals influ-

ence response to a specific dietary pattern, functional food or supplement in

terms of a specific health outcome. The specific field of genome health nutrige-

nomics [Fenech, 2005] has been proposed on the premise that a more useful

Fenech 60

approach to prevention of diseases caused by genome damage is to take into

consideration the genotype of individuals with a focus on common genetic

polymorphisms that alter the bioavailability of micronutrients and/or the affin-

ity of key enzymes involved in DNA metabolism for their micronutrient cofac-

tor. Supplementation of diet with appropriate minerals and vitamins could, in

some cases, help overcome inherited metabolic blocks in key DNA mainte-

nance pathways [Ames, 2003, 2004]. Increasing concentration of a cofactor by

supplementation is expected to be particularly effective when a mutation (poly-

morphism) in a gene decreases the binding affinity for its cofactor resulting in

a lower reaction rate. The interaction between genotype and diet in modulating

risk is emerging as an exciting area of research as regards micronutrient effects

on DNA. This is illustrated by recent research on the common mutations in the

methylenetetrahydrofolate reductase (MTHFR) gene and other genes in the

folate/methionine cycle with regard to developmental defects and cancer risk

[Brody et al., 2002; Skibola et al., 1999]. The product of the MTHFR gene

determines the availability of folate for the synthesis of thymidylic acid from

deoxyuridylic acid. Polymorphisms in the MTHFR gene, such as the C677T

mutation, that reduce activity of the MTHFR enzyme are predicted to minimise

uracil misincorporation into DNA whilst making less methylfolate available for

the synthesis of S-adenosyl methionine, the common methyl donor [Ames,

1999; Fenech, 2001]. Epidemiological studies have suggested that individuals

homozygous for the C677T polymorphism (i.e. TT genotype, which causes

reduced MTHFR activity) may be protected against colorectal cancer and acute

lymphocytic leukaemia relative to those with the wild-type CC genotype [Chen

et al., 1999; Skibola et al., 1999]. Recent results from our laboratory have

shown that there are important significant interactions between the MTHFRC677T polymorphism, its cofactor riboflavin and folic acid with respect to

chromosomal instability [Kimura et al., 2004]. This is illustrated by (a) the

reduction in nuclear bud frequency (a biomarker of gene amplification) in TT

homozygotes relative to CC homozygotes for the MTHFR C677T mutation and

(b) the observation that high riboflavin concentration increases nuclear bud fre-

quency under low folic acid conditions (12 nM folic acid) probably by increas-

ing MTHFR activity which diverts folate away from dTTP synthesis, increasing

the odds for uracil incorporation into DNA synthesis, the generation of breakage-

fusion-bridge cycles and subsequent gene amplification and nuclear bud

formation. Other common polymorphisms, such as the manganese superoxide

dismutase alanine-to-valine change in the �9 position, which disables transport

of this enzyme to the mitochondrion where it is normally located [Ambrosone

et al., 1999], increase susceptibility to oxidative stress and breast cancer risk.

Individuals with this manganese superoxide dismutase mutation appear to

benefit more than controls from a higher intake of fruits and vegetables and/or


vitamin C-rich foods in terms of protection against breast cancer [Ambrosone

et al., 1999]. In the past, considerable attention has been given to gene-environment

interaction as it relates to mutagen/carcinogen exposure and genotoxicity or

cancer risk. However, it is probable that gene-diet interaction as it relates to

efficacy of DNA repair/DNA metabolism and antioxidant response may be

equally important in determining genomic stability and its consequent impact

on fertility, development, cancer risk and the rate of ageing.

The Genome Health Clinic Concept – A Paradigm Shift in DiseasePrevention Based on the Diagnosis and Nutritional Treatment ofGenome and Epigenome Damage

The advances in our knowledge described above have opened up a new

opportunity in disease prevention based on the concepts that (a) excessive

genome damage is the most fundamental cause of developmental and degener-

ative disease, (b) genome damage caused by micronutrient deficiency is pre-

ventable, (c) accurate diagnosis of genome instability using DNA damage

biomarkers that are sensitive to micronutrient deficiency is technically feasible

and (d) it is possible to optimise nutritional status and verify efficacy by diag-

nosis of a reduction in genome and epigenome damage rate after intervention.

Given the emerging evidence that dietary requirements of individuals may

depend on their inherited genes, we can anticipate (a) important scientific

developments in the understanding of the relationships between dietary require-

ment and genetic background to optimise genome stability and (b) that the

accumulated knowledge on dietary requirements for specific genetic subgroups

will be used to guide decisions by the practitioners of this novel preventive

medicine in what might be called ‘Genome Health Clinics’. In other words, one

can envisage that instead of diagnosing and treating diseases caused by genome

and/or epigenome damage, health/medical practitioners will be trained, in the

near future, to diagnose and nutritionally prevent a most fundamental initiating

cause of developmental and degenerative disease, i.e. genome and epigenome

damage. This novel approach also opens up the possibility for the massive num-

bers of health-conscious consumers to be able to assess directly the effect of

their dietary and nutritional supplement choices on their genome and that of

their children. In addition, there will be scope to develop new functional foods

and supplements for genome health that can be mixed and matched so that the

dietary intake nutriome is appropriately tailored to an individual’s genotype and

genome status. The conceptual framework of the diagnostics and databases

required to implement this complementary preventive medicine are described

in more detail in Fenech [2003, 2005].

Fenech 62

References

Ambrosone CB, Freudenheim JL, Thompson PA, Bowman E, Vena JE, Marshall JR, Graham S,

Laughlin R, Nemoto T, Shields PG: Manganese superoxide dismutase (MnSOD) genetic polymor-

phisms, dietary antioxidants and risk of breast cancer. Cancer Res 1999;59:602–606.

Ames BN: Micronutrients prevent cancer and delay ageing. Toxicol Lett 1998;102–103:5–18.

Ames BN: Cancer prevention and diet: help from single nucleotide polymorphisms. Proc Natl Acad Sci

USA 1999;96:12216–12218.

Ames BN: DNA damage from micronutrient deficiencies is likely to be a major cause of cancer. Mutat

Res 2001;475:7–20.

Ames BN: The metabolic tune-up: metabolic harmony and disease prevention. J Nutr 2003;133(suppl 1):

1544S–1548S.

Ames BN: A role for supplements in optimizing health: the metabolic tune-up. Arch Biochem Biophys

2004;423:227–234.

Ames BN, Wakimoto P: Are vitamin and mineral deficiencies a major cancer risk? Nat Rev Cancer

2002;2:694–704.

Bailey LB: Folate, methyl-related nutrients, alcohol, and the MTHFR 677C → T polymorphism affect

cancer risk: intake recommendations. J Nutr 2003;133(suppl 1):3748S–3753S.

Blount BC, Ames BN: DNA damage in folate deficiency. Bailleres Clin Haematol 1995;8:461–478.

Blount BC, Mack MM, Wehr CM, MacGregor JT, Hiatt RA, Wang G, Wickramasinghe SN, Everson RB,

Ames BN: Folate deficiency causes uracil misincorporation into human DNA and chromosome

breakage: implications for cancer and neuronal damage. Proc Natl Acad Sci USA 1997;94:

3290–3295.

Bonassi S, Fenech M, Lando C, Lin YP, Ceppi M, Chang WP, Holland N, Kirsch-Volders M, Zeiger E,

Ban SY, Barale R, Bigatti MP, Bolognesi C, Jia C, DiGiorgio M, Ferguson LR, Fucic A, Lima OG,

Hrelia P, Krishnaja AP, Lee TK, Migliore L, Mikhalevich L, Mirkova E, Mosesso P, Muller WU,

Odagiri Y, Scarfi MR, Szabova E, Vorobtsova I, Vral A, Zijno A: Human Micronucleus Project:

international database comparison for results with the cytokinesis-block micronucleus assay in

human lymphocytes. 1. Effect of laboratory protocol, scoring criteria and host factors on the fre-

quency of micronuclei. Environ Mol Mutagen 2001;37:31–45.

Bonassi S, Hagmar L, Stromberg U, Montagud AH, Tinnerberg H, Forni A, Heikkila P, Wanders S,

Wilhardt P, Hansteen IL, Knudsen LE, Norrpa H: Chromosomal aberrations in lymphocytes

predict human cancer independently of exposure to carcinogens. Cancer Res 2000;60:

1619–1625.

Boyonoski AC, Gallacher LM, Apsimon MM, Jacobs RM, Shah GM, Poirier GG, Kirkland JB: Niacin

deficiency increases sensitivity of rats to the short and long term effects of ethylnitrosourea treat-

ment. Mol Cell Biochem 1999;193:83–87.

Brody LC, Conley M, Cox C, Kirke PN, McKeever MP, Mills JL, Molloy AM, O’Leary VB, Parle-

McDermott A, Scott JM, Swanson DA: A polymorphism, R653Q, in the trifunctional enzyme

methylenetetrahydrofolate dehydrogenase/methenyltetrahydrofolate cyclohydrolase/formylte-

trahydrofolate synthetase is a maternal genetic risk factor for neural tube defects: Report of the

Birth Defects Research Group. Am J Hum Genet 2002;71:1207–1215.

Cagnacci A, Baldassari F, Rivolta G, Arangino S, Volpe A: Relation of homocysteine, folate, and vitamin B12

to bone mineral density of postmenopausal women. Bone 2003;33:956–959.

Centers for Disease Control: Recommendations for the use of folic acid to reduce the number of cases of

spina bifida and other neural tube defects. MMWR Recomm Rep 1992;41:1–7.

Chen J, Giovannucci EL, Hunter DJ: MTHFR polymorphisms, methyl-replete diets and risk of colorec-

tal carcinoma and adenoma among US men and women: an example of gene-environment inter-

actions in colorectal tumorigenesis. J Nutr 1999;129:560S–564S.

Cho E, Smith-Warner SA, Spiegelman D, Beeson WL, van den Brandt PA, Colditz GA, Folsom AR,

Fraser GE, Freudenheim JL, Giovannucci E, Goldbohm RA, Graham S, Miller AB, Pietinen P,

Potter JD, Rohan TE, Terry P, Toniolo P, Virtanen MJ, Willett WC, Wolk A, Wu K, Yaun SS,

Zeleniuch-Jacquotte A, Hunter DJ: Dairy foods, calcium, and colorectal cancer: a pooled analysis

of 10 cohort studies. J Natl Cancer Inst 2004;96:1015–1022.


Choi SW, Mason JB: Folate status: effects on pathways of colorectal carcinogenesis. J Nutr 2002;132

(suppl):2413S–2418S.

Claycombe KJ, Meydani SN: Vitamin E and genomic stability. Mutat Res 2001;475:37–44.

Cravo M, Fidalgo P, Pereira AD, Gouveia-Oliviera A, Chaves P, Selhub J, Mason JB, Mira PC, Leitao CN:

DNA methylation as an intermediate biomarker in colorectal cancer: modulation by folic acid

supplementation. Eur J Cancer Prev 1994;3:473–479.

Crott JW, Mashiyama ST, Ames BN, Fenech M: MTHFR C677T polymorphism does not alter folic acid

deficiency-induced uracil incorporation into primary lymphocyte DNA in vitro. Carcinogenesis

2001a;22:1019–1025.

Crott JW, Mashiyama ST, Ames BN, Fenech MF: Folic acid deficiency increases chromosome breakage

and rearrangement, gene amplification and DNA-uracil content in human lymphocytes in vitro:

effect of MTHFR C677T polymorphism. Cancer Epidemiol Biomarkers Prev 2001b;10:1089–1096.

Dreosti IE: Zinc and the gene. Mutat Res 2001;475:161–168.

Duthie SJ, Hawdon A: DNA instability (strand breakage, uracil misincorporation, and defective repair)

is increased by folic acid depletion in human lymphocytes in vitro. FASEB J 1998;12:1491–1497.

Egger G, Liang G, Aparicio A, Jones PA: Epigenetics in human disease and prospects for epigenetic

therapy. Nature 2004;429:457–463.

El-Bayoumy K: The protective role of selenium on genetic damage and on cancer. Mutat Res

2001;475:123–139.

Fenech M: Chromosomal damage rate, ageing and diet. Ann NY Acad Sci 1998;854:23–36.

Fenech M: The in vitro micronucleus technique. Mutat Res 2000;455:81–95.

Fenech M: The role of folic acid and vitamin B12 in genomic stability of human cells. Mutat Res

2001;475:56–67.

Fenech M: Chromosomal biomarkers of genomic instability relevant to cancer. Drug Discov Today

2002;7:1128–1137.

Fenech M: Nutritional treatment of genome instability: a paradigm shift in disease prevention and in the

setting of recommended dietary allowances. Nutr Res Rev 2003;16:109–122.

Fenech M: Genome health nutrigenomics: nutrition and the science of optimal genome maintenance.

Asia Pac J Clin Nutr 2004;13(suppl):S15.

Fenech M: The Genome Health Clinic and Genome Health Nutrigenomics concepts: diagnosis and nutri-

tional treatment of genome and epigenome damage on an individual basis. Mutagenesis 2005;20:

255–269.

Fenech M, Aitken C, Rinaldi J: Folate, vitamin B12, homocysteine status and DNA damage in young

Australian adults. Carcinogenesis 1998;19:1163–1171.

Fenech M, Baghurst P, Luderer W, Turner J, Record S, Ceppi M, Bonassi S: Low intake of calcium,

folate, nicotinic acid, vitamin E, retinol, �-carotene and high intake of pantothenic acid, biotin and

riboflavin are significantly associated with increased genome instability – results from a dietary

intake and micronucleus index survey in South Australia. Carcinogenesis 2005;26:991–999.

Fenech M, Ferguson LR: Micronutrients and genomic stability. Mutat Res 2001;475:45–56.

Fenech M, Morley AA: Cytokinesis-block micronucleus method in human lymphocytes: effects of in

vivo ageing and low dose X-irradiation. Mutat Res 1986;161:193–198.

Giovannucci E: Diet, body weight and colorectal cancer: a summary of epidemiological evidence.

J Womens Health 2003;12:173–182.

Hageman GJ, Stierum RH: Niacin, poly(ADP-ribose) polymerase-1 and genomic stability. Mutat Res

2001;475:45–56.

Halliwell B: Vitamin C and genomic stability. Mutat Res 2001;475:29–35.

Hartwig A: Role of magnesium in genomic stability. Mutat Res 2001;475:113–121.

Ho E, Ames BN: Low intracellular zinc induces oxidative DNA damage, disrupts p53, NFkappaB and

AP1 binding and affects DNA repair in a rat glioma cell line. Proc Natl Acad Sci USA 2002;99:

16770–16775.

Holliday R: Understanding Ageing. Cambridge, Cambridge University Press, 1995.

Honda S, Marumoto T, Hirota T, Nitta M, Arima Y, Ogawa M, Saya H: Activation of m-calpain is

required for chromosome alignment on the metaphase plate during mitosis. J Biol Chem 2004;279:

10615–10623.

Fenech 64

Hsia KT, Millar MR, King S, Selfridge J, Redhead NJ, Melton DW, Saunders PT: DNA repair gene Ercc1

is essential for normal spermatogenesis and oogenesis and for functional integrity of germ cell

DNA in the mouse. Development 2003;130:369–378.

International Atomic Energy Agency: Biological Dosimetry: Chromosomal Aberration Analysis for

Dose Assessment. Technical Reports Series No 260. Vienna, International Atomic Energy Agency,

1986, pp 13–27.

Jacky PB, Beek B, Sutherland GR: Fragile sites in chromosomes: possible model for the study of spon-

taneous chromosome breakage. Science 1983;220:69–70.

Jacob RA, Gretz DM, Taylor PC, James SJ, Pogribny IP, Miller BJ, Henning SM, Swendseid ME:

Moderate folate depletion increases plasma homocysteine and decreases lymphocyte DNA methy-

lation in postmenopausal women. J Nutr 1998;128:1204–1212.

Joenje H, Patel JK: The emerging genetic and molecular basis of Fanconi anaemia. Nat Rev Genet.

2001;2:446–457.

Keen CL, Zidenberg-Cherr S: Manganese; in Zeigler EE, Filer LJ (eds): Present Knowledge in

Nutrition, ed 7. Washington, ILSI Press, 1996, pp 334–343.

Kimura M, Umegaki K, Higuchi M, Thomas P, Fenech M: MTHFR C677T polymorphism, folic acid and

riboflavin are important determinants of genome stability in cultured human lymphocytes. J Nutr

2004;134:48–56.

Lamprecht SA, Lipkin M: Chemoprevention of colon cancer by calcium, vitamin D and folate: molecu-

lar mechanisms. Nat Rev Cancer 2003;3:601–614.

Lansdorp PM: Repair of telomeric DNA prior to replicative senescence. Mech Ageing Dev 2000;118:

23–34.

Lindahl T, Wood RD: Quality control by DNA repair. Science 1999;286:1897–1905.

Liu L, Blasco M, Trimarchi J, Keefe D: An essential role for functional telomeres in mouse germ cells

during fertilization and early development. Dev Biol 2002;249:74–84.

McCaddon A, Hudson P, Abrahamsson L, Olofsson H, Regland B: Analogues, ageing and aberrant

assimilation of vitamin B12 in Alzheimer’s disease. Dementia Geriatr Cogn Disord 2001;12:

133–137.

MacDonald HM, McGuigan FE, Fraser WD, New SA, Ralston SH, Reid DM: Methylenetetrahydrofolate

reductase polymorphism interacts with riboflavin intake to influence bone mineral density. Bone

2004;35:957–964.

Mashiyama ST, Coutemanche C, Elson-Schwab I, Crott J, Lee BL, Ong CN, Fenech M, Ames BN:

Uracil in DNA, determined by an improved assay, is increased when deoxynucleotides are added

to folate-deficient cultured human lymphocytes. Anal Biochem 2004;330:58–69.

Migliore L, Botto N, Scarpato R, Petrozzi L, Cipriani G, Bonucelli U: Preferential occurrence of chro-

mosome 21 segregation in peripheral blood lymphocytes of Alzheimer disease patients. Cytogenet

Cell Genet 1999;87:41–46.

Migliore L, Scarpato R, Coppede F, Petrozzi L, Bonucelli U, Rodilla V: Chromosome and oxidative

damage biomarkers in lymphocytes of Parkinson’s disease patients. Int J Hyg Environ Health

2001;204:61–66.

Narula A, Kilen S, Ma E, Kroeger J, Goldberg E, Woodruff TK: Smad4 overexpression causes germ cell

ablation and leydig cell hyperplasia in transgenic mice. Am J Pathol 2002;161:1723–1734.

Nathanson KL, Wooster R, Weber BL, Nathanson KN: Breast cancer genetics: what we know and what

we need. Nat Med 2001;7:552–556.

Ng JM, Vrieling H, Sugasawa K, Ooms MP, Grootegoed JA, Vreeburg JT, Visser P, Beems RB, Gorgels

TG, Hanaoka F, Hoeijmakers JH, van der Horst GT: Developmental defects and male sterility

in mice lacking the ubiquitin-like DNA repair gene mHR23B. Mol Cell Biol 2002;22:

1233–1245.

Paul AA, Southgate DAT: McCance and Widdowson’s ‘The Composition of Foods’. London, HMSO,

1978.

Rajagopalan H, Lengauer C: Aneuploidy and cancer. Nature 2004;432:338–341.

Sato Y, Honda Y, Iwamoto J, Kanoko T, Satoh K: Effect of folate and mecobalamin on hip fractures in

patients with stroke: a randomized controlled trial. JAMA 2005;293:1082–1088.

Shen J, Loeb LA: Unwinding the molecular basis of Werner syndrome. Mech Ageing Dev 2001;122:

921–944.


Skibola CF, Smith MY, Kane E, Roman E, Rollinson S, Cartwright RA, Morgan G: Polymorphisms in

the methylenetetrahydrofolate reductase gene are associated with susceptibility to acute leukaemia

in adults. Proc Natl Acad Sci USA 1999;96:12810–12815.

Thomas P, Umegaki K, Fenech M: Nucleoplasmic bridges are a sensitive measure of chromosome

rearrangement in the cytokinesis-block micronucleus assay. Mutagenesis 2003;18:187–194.

Thompson LH, Schild D: Recombinational DNA repair and human disease. Mutat Res 2002;509:49–78.

Titenko-Holland N, Jacob RA, Shang N, Balaraman A, Smith MT: Micronuclei in lymphocytes and

exfoliated buccal cells of postmenopausal women with dietary changes in folate. Mutat Res

1998;417:101–114.

Trippi F, Botto N, Scarpato R, Petrozzi L, Bonuccelli U, Latorraca S, Sorbi S, Migliore L: Spontaneous

and induced chromosome damage in somatic cells of sporadic and familial Alzheimer disease

patients. Mutagenesis 2001;16:323–327.

Trkova M, Kapras J, Bobkova K, Stankova J, Mejsnarova B: Increased micronuclei frequencies in cou-

ples with reproductive failure. Reprod Toxicol 2000;14:331–335.

Vinson RK, Hales BF: DNA repair during organogenesis. Mutat Res 2002;509:79–91.

Walter PB, Knutson MD, Paler-Martinez A, Lee S, Xu Y, Viteri FE, Ames BN: Iron deficiency and iron

excess damage mitochondria and mitochondrial DNA in rats. Proc Natl Acad Sci USA

2002;99:2264–2269.

Willett WC: Diet and cancer: one view at the start of the millennium. Cancer Epidemiol Biomarkers

Prev 2001;10:3–8.

Xu N, Luo KQ, Chang DC: Ca2� signal blockers can inhibit M/A transition in mammalian cells by inter-

fering with the spindle checkpoint. Biochem Biophys Res Commun 2003;306:737–745.

Dr. Michael Fenech

CSIRO Human Nutrition

PO Box 10041

Adelaide BC, 5000 (Australia)




Nutrition: Ethics and Social Implications

Inez H. Slamet-Loedin, Umar A. Jenie

Indonesian Institute of Sciences, National Bioethics Commission, Jakarta, Indonesia

AbstractIn October 2003, the general conference of UNESCO adopted the International

Declaration on Human Genetic Data, followed by the adoption of the Universal Declaration

on Bioethics and Human Rights in October 2005 to ensure the respect of human dignity and

the protection of human rights and fundamental freedoms in the collection, processing, use

and storage of human genetic data with the requirement of equality, justice and solidarity.

Nutrigenomics studies the relationship between specific nutrients or diet and polymorphisms

and gene expression; therefore, eventually diet can be tailored for each individual. The

dietary intervention is based on collected human genetic data that eventually build knowl-

edge of nutritional requirements, and the nutritional status of different human genotypes.

This knowledge can be used to prevent, mitigate or cure chronic diseases. As in another

branch of posthuman genome science, it is a global concern that the collected data should not

be misused or create inequity. Some ethical issues raised and discussed in this paper are: (1)

consent and confidentiality issues in the collection and storage of data, (2) genetic screening

and how to prevent inequity, (3) regulatory oversight and in a wider context the need to

improve public confidence in biotechnology-related science, (4) other social issues. The eth-

ical issues in nutrigenomics need clear and concise guidelines developed in accordance with

the universally adopted declarations and ethical concern needs to be integrated in the scien-

tific design. Efforts to improve the public awareness, public participation and consultation

need to be made at the early stage of the development of nutrigenomics.


Nutrigenomics has emerged as a new ‘omics’ technology developed as a

more complex functional analysis compared to the basic sequence information

provided by the Human Genome Projects. This branch of science in the area of

genomics allows us to understand the relationship between specific nutrients or

diet and gene expression which can eventually facilitate the development of a bet-

ter strategy for the prevention or treatment of diseases and a personalized individ-

ual diet. The science of nutrigenomics allows a better molecular understanding of

Nutrition: Ethics and Social Implications 67

how common dietary chemicals affect health by altering the expression of an

individual’s genetic makeup [Mathers, 2005].

Like in other areas developed to seek further functional studies of genomic

research, such as pharmacogenomics, the use of samples from identified popula-

tions resulted in the inextricable linkage of social, ethical and scientific issues.

Ethical concerns have become a major global issue in the development of postge-

nomic science in humans. As a later branch of postgenomic science, the scien-

tific community should learn from other experiences that ethical issues need to

be addressed at an early stage of the development of this technology. The experi-

ence of pharmacogenetics will therefore give an indication of whether people are

prepared to allow the use of their genetic information for the purposes of

research and, ultimately, treatment [Burton and Steward, 2005].

Although the integration of ethical concerns into a scientific approach can

be a serious challenge due to the complexity of the issues and the fact that the

approach may be different in various societies, nowadays it is inevitable that the

ethical concerns should be identified and addressed even in the experimental

design process. The International HapMap Project is one example of this approach

[International HapMap Consortium, 2004]. This HapMap Consortium Project

raised many ethical issues because it allowed researchers to compare patterns of

variation among both individuals and populations. They decided to have a sep-

arate team of ethics advisers that worked collaboratively throughout the project

with geneticists to address the ethical issues. In a smaller project, it is still

advisable to address ethical issues from early on in conjunction with the scien-

tific decision to avoid a legal implication afterwards.

Regarding nutrigenomics, the general public would most probably support

the view that obtaining knowledge about the observation that, under certain cir-

cumstances and in some individuals, diet can be a serious risk factor for the onset

and incidence of a number of diseases is a useful effort for the benefit of society.

The paradigm of a preventive approach to health has been promoted even from

the Hippocrates era on (‘leave your drugs in the chemist’s pot if you can heal the

patient with food’). However, the fears that the collected genetic data will be mis-

used and that the results of the genomic research based on the collected data may

generate social discrimination have raised a major concern. The way the data are

collected for a specific purpose and the possible misuse and misinterpretation of

the collected data can lead to the violation of basic human rights such as privacy,

equality and justice. The challenge is how to conduct genetic variation research

that uses identified populations in an ethical way, including how to involve mem-

bers of a population in evaluating the risks and benefits for everyone who shares

that identity [International HapMap Consortium, 2004].

The Universal Declaration on Bioethics and Human Rights adopted in

2005 gave a general principle, i.e. ethical issues raised by the rapid science and

Slamet-Loedin/Jenie 68

technological applications should be examined with universal respect for the

inherent dignity of the human person and with universal respect for, and obser-

vance of, human rights and fundamental freedoms [UNESCO, 2005].

In a global context, ethical issues in human genetics have been discussed

by the general conferences of UNESCO and resulted in the adoption of the

Universal Declaration on the Human Genome and Human Rights on November

11, 1997, followed by the International Declaration of Human Genetic Data on

October 16, 2003, and recently the adoption of the Universal Declaration on

Bioethics and Human Rights in October, 2005.

Chadwick [2004] defined nutrigenomics as the relationship between spe-

cific nutrients or diet and gene expression and it is envisaged that it will facili-

tate the prevention of diet-related common diseases, while nutrigenetics is

concerned with the effects of individual genetic variation (single nucleotide

polymorphisms) in response to diet, which in the longer term may lead to per-

sonalized dietary guidance. The Nuffield report on nutrigenomics [2004] stated

that the distinction between nutrigenomics and nutrigenetics provided a useful

basis for discussion about what the new technologies could do and future prob-

lems. In this paper, the discussion of ethical issues of nutrigenomics will not be

separated from nutrigenetics. The major issues discussed here are (1) the data

collection process and storage, (2) how to prevent inequity and (3) regulatory

oversight and in a wider context the need to improve public confidence in

biotechnology-related science.

Data Collection: Consent, Privacy and Confidentiality

To a certain extent, there are similarities in ethical issues between nutrige-

nomics and pharmacogenomics, mainly with regard to privacy and confiden-

tiality. In order to reveal an association between either diseases or nutrient

response and a genetic polymorphism, we need population genetic data. The

major difference between nutrigenomics and pharmacogenomics is the fact

that pharmaceuticals are well-defined compounds aimed at specific targets;

foods, however, are complex substances that have multiple effects on different

pathways in the body [Muller and Kersten, 2003]; therefore, the size of popu-

lation data for nutrigenomics studies may need to be even greater [Chadwick,

2004].

Consent, privacy and confidentiality are considered to be the principal

issues in relation to the genetic database. The most commonly expressed fear is

that genetic information will be used in ways that could deny people access to

health insurance, employment, education, and even loans. This concern is partly

increased by the growing recognition that health information is not entirely


private [Clayton, 2003]. One of the issues to be resolved would be who ought to

have access to nutrigenomics information and under what conditions.

The fear is also related to the misconception of genetic determinism that

the genetic ‘blueprint will tell you all about your future’, while in reality human

characteristics are the product of complex interactions over time between genes

and the environment. As is true for so many conditions in medicine, clinicians

have a variable but usually limited ability to predict when, how severely, and

even whether a person with a genetic predisposition to a certain illness is going

to become ill [Guttmacher, 2001]. The reality shows that individuals carrying

certain genes may be prone to the onset of the disease but far from guaranteed

to develop the illness.

Collection and management of nutrigenomics data involves data collection

starting from obtaining consent, processing, use and storage of a massive amount

of data of individuals and populations. The more individualized the promises of

genetics, the more collective action is required in the form of population-based

research to enable discernment of the differences at the genetic level between

individuals that will affect susceptibility [Chadwick, 2004].

ConsentIn the summary of the Nuffield report on pharmacogenetics ethical issues

[Nuffield Council on Bioethics, 2003], it was stated that there are numerous

codes of practice and guidance regarding the conduct of clinical research which

includes consent. It is a common practice to require consent for the collection

and banking of tissue and DNA samples of participants in research, especially if

it is intended to combine genetic information with other information from the

patient’s medical record.

As mentioned above, the nutrigenomics paradigm, as the paradigm of other

genomic-based sciences, is proactive, rather than reactive, and will provide advice

that is predictive, preventive, and personalized. Standardization of the method

that will allow genetic screening in the future will require a large number of pop-

ulations and nations, including eastern nations, as well as individual consent.

Individual consent is obligatory in most of the available guidelines and

regulations. There have been arguments that in the case of public interest, indi-

vidual consent could be overruled. Another argument is that different societies

have different concepts of individual freedom and autonomy; the values of indi-

viduality are interwoven with the cultural consciousness. In some countries, it is

acceptable that local domestic authorities or local leaders give consent on

behalf of their people (community consent) in accordance with prevailing local

cultural norms, but still in any case domestic law should be upheld.

Consent has to be voluntary in nature. The Universal Declaration on Bioethics

and Human Rights [UNESCO, 2005] requires informed consent (article 10) and


the International Declaration on Human Genetic Data [UNESCO, 2003]

strongly states that prior, free, informed and expressed consent, without induce-

ment by financial or other personal gains, should be obtained for the collection

of human genetic data, human proteomic data or biological samples, whether

through invasive or noninvasive procedures, and for their subsequent process-

ing, use and storage, whether carried out by a public or private institution (arti-

cle 8). Limitations on this principle of consent should only be prescribed for

compelling reasons by domestic law consistent with the international law of

human rights.

The last sentence in the stipulation leaves room to a certain degree of flex-

ibility in domestic regulation, but the next section strengthens that authorization

should be obtained from the legal representative in accordance with domestic

law; this legal representative should act to the best interest of the person con-

cerned. It was also stated in the International Declaration on Human Genetic

Data [UNESCO, 2003] that consent can be withdrawn without any penalty.

Individual consent increasingly becomes an important human right issue.

Consent has to be free without inducement by financial or other personal gains.

The recent legal case of stem cell research in Korea initially started with an

issue of consent, due to the fact that the donor was one of the junior researchers

in the project, in which power may have played a role, before it later on also

developed in other directions.

In the case of pharmacogenetics, the Nuffield report [Nuffield Council on

Bioethics, 2003] stated that there is a serious question regarding whether volun-

tary consent to pharmacogenetic testing can truly be obtained in the context of

clinical trials or in clinical practice. If researchers require genotyping as a con-

dition of enrolment in a study, patients might not feel able to refuse, especially

if they think it is possible that they may get some personal benefit. In some

cases, taking part in a clinical trial may be the only way for a patient to have a

chance of obtaining a particular medicine, while actually, as stated by Corrigan

[2005], the pharmacogenetics study is of no direct benefit to the patient and

thus compromises in the consent process cannot be offset against the potential

therapeutic benefit. For nutrigenomics, the issue of data collection might be

slightly different. First the nutritional information to prevent the onset of a dis-

ease can be directly beneficial for the patients or at least the possible benefit is

more direct, and second the public tends to think that health problems related to

food are less daunting compared to other types of diseases.

Regarding informed consent and avoiding the issue of financial gain, the

International HapMap Project [International HapMap Consortium, 2004], in

which samples were taken from the Yoruba people (from Nigeria), Japanese and

Chinese individuals, and from residents of Utah in the United States, could

serve as an example. Blood donors had to be adults (as legally defined in each


country) and competent to provide informed consent. Although donors in

Nigeria were each given an equivalent of approximately USD 8.00 and multivi-

tamins worth approximately USD 4.00 to compensate them for their time and

travel – a standard amount for the participation in research involving blood

draws in that part of Africa – the prospective donors were not told that they

would be compensated until after they had arrived to donate, to guard against

the possibility that they would be induced to participate by the prospect of

material benefit. Although this may not be entirely correct, it would also be

unjust not to give any compensation to the people for their time and travel.

Another important point is that the consent issue cannot be separated from

the public awareness of what the consent is for and more generally the local

public opinion on the research aims. Castro [2005] studied the InternationalEthical Guidelines for Biomedical Research Involving Human Subjects pub-

lished by the Council for International Organizations of Medical Sciences and

found 26 items related to consent in these guidelines that needed to be under-

stood by the participating individuals. This long list suggests the amount of

detail that has to be understood before an individual gives his/her consent. It is

also very important to include local communities in the public consultation

process.

Whether consent has to be classified as ‘broad’ or ‘limited’ is another pub-

lic issue. Limited or narrow consent refers to instances where a sample is only

to be used for a restricted range of purposes, as for example a single research

project, while broad consent entails that patients agree that their sample may be

used for a variety of future studies which it may not be possible to specify in

any detail at the time of consent. Allowing broad consent may be of significant

benefit to researchers and to society’s interest in the acquisition of knowledge

about health and disease, but it may not be beneficial for the individual. The

Nuffield pharmacogenomics ethical team recommended/considered that it is

permissible to request broad consent to the use of samples which are anony-

mous or anonymized; however, where samples collected for a particular study

are coded or identified, broad consent to future research may also be permissi-

ble, but should be sought separately from consent to the initial study. This sepa-

rate consent may be obtained when the samples are originally taken, or at a later

date [Nuffield Council on Bioethics, 2004]. In any case of requesting a broad

consent to future research, the prospective donors should be given a compre-

hensive explanation regarding the risks and benefits to allow them to have a

clear understanding of the possible future implications.

Privacy and ConfidentialityAlthough confidentiality is often used interchangeably with privacy,

according to Anderlik and Rothstein [2001], privacy subsumes at least four


categories: (1) access to persons and to personal spaces; (2) access to informa-

tion by third parties; (3) third-party interference with personal choices, espe-

cially in intimate spheres such as procreation, and (4) ownership of materials

and information derived from persons. Privacy is also a term with deep emo-

tional resonance, while confidentiality describes the duties that accompany the

disclosure of nonpublic information to a third party within a professional, fidu-

ciary, or contractual relationship. By law, social norm, or contract, and usually

by some combination of these, the third party entrusted with the information is

prohibited from redisclosing it or discussing it outside the confines of the rela-

tionship except under very restricted circumstances. Security refers to the mea-

sures taken to prevent unauthorized access to persons, places, or information.

The measures used to achieve the goal of security vary according to the context

and the state of technology.

As regards privacy, the value of privacy in eastern society may differ from

that in western society; however, generally the public favors protecting privacy.

Anderlik and Rothstein [2001] explained that privacy has both an intrinsic and

instrumental value. The intrinsic value is linked with the ethical principle of

autonomy of individual self-governance. The genetic information is connected

to personal and group identity, and protecting the privacy of genetic informa-

tion is an important individual and social priority. The instrumental value of pri-

vacy is usually termed as utilitarianism. In the approach to ethics most closely

associated with the philosopher Immanuel Kant, each person has the duty to

respect the autonomy of others. This means that it is wrong to treat a fellow

human being solely as a means to an end, even if that end is something noble

like the advancement of science or the cure of disease and the prevention of suf-

fering. In addition, the rules that govern society should be consistent with the

principle of respect for autonomy.

How to ensure confidentiality is a challenge. Most participants in a genetic

data collection worry that employers and insurance companies might get access

to the data. Article 14 of the International Declaration on Human Genetic Data

[UNESCO, 2003] on privacy and confidentiality stipulates that states should

endeavor to protect the privacy of individuals; genetic and proteomics data

should not be disclosed to third parties, in particular employers, insurance com-

panies, and educational institutions. Article 11 of the Universal Declaration on

Bioethics and Human Rights [2005] regarding privacy and confidentiality also

states that any decision/practice should respect privacy.

It is in fact important to know and acknowledge set differences in order to

be able to enhance assets and reduce liabilities, not only in the field of health

care, but also in several other fields. Human interaction is not only marked by

rationality, cooperation and complimentarily, but also by fierce competition,

rivalry and misuse of other people’s weaknesses. It is therefore absolutely


necessary when collecting extensive genetic data to safeguard the rights to pri-

vacy and confidentiality both of individuals and groups, and to respect the right

of individuals to withhold and withdraw consent.

The Nuffield report on pharmacogenetic ethics [Nuffield Council on

Bioethics, 2003] mentioned that the implications for patients of DNA samples

being used in research differ depending on how easily their samples can be

traced back to them, and whether the research is likely to give rise to informa-

tion that may be of personal clinical relevance. The report’s view is that it is

generally possible to obtain genetic and clinical information about a patient

during a clinical trial and then to anonymize the samples so that the code link-

ing the sample with the patient is destroyed. It was also stated that researchers

should explain to prospective participants the implications of the manner in

which samples will be stored for those participants.

In the International HapMap Projects [International HapMap Consortium,

2004], for example, no personal identifiers or medical information about sam-

ple donors were included, but each sample was identified by the population

from which it came. The scientific rationale for identifying the populations is

that differences in haplotype frequencies and lengths among populations will

be important for how data are used in the research project. They decided to

name the population since from an ethical point of view, removing population

identifiers could create a false sense of protection from collective risks, because

it would be easy to guess the populations from which donors were recruited. It

would not be difficult either to discern from previously collected data sets the

identity of these populations. Rather than allowing donors to assume that their

population identities were protected or allowing other researchers to infer those

identities, they concluded that naming the populations was thought to be more

ethically appropriate.

As mentioned above, security refers to the measures taken to prevent

unauthorized access to persons, places, or information. The challenge is the fact

that this technology deals with massive amounts of data; therefore, storage of

data that can guarantee confidentiality needs to be carefully designed. Another

challenge is who legally is allowed to have access to the individual data: key

researchers, private doctors, parents of underage children? All these issues need

to be made clear in the ethical guidelines.

Large amounts of the data were collected by private sectors and in many

examples these companies were then sold to a bigger company or merged. It

should be made sure that changes in the personnel do not jeopardize the data

security. To keep the public confidence, care should be taken to ensure the secu-

rity of the data.

Another concern is the potential for accidental release through system

malfunction or access by hackers as well as a lapse of vigilance on the part of


a range of persons, including data processors and researchers as well as health

care professionals.

Genetic Screening, How to Prevent Inequity and Some Other Social Issues

Genetic Screening and How to Prevent InequityGenetic screening is typically defined as the determination of the preva-

lence of a gene in an asymptomatic population or population group [Chadwick,

2004]. Genetic screening entails identifying individuals affected by a disease, at

risk for developing a disease, or at risk for having a child affected by a disease.

It involves a population-based approach, and populations are selected with spe-

cific goals and strategies in mind [McCabe and McCabe, 2004].

Newborn screening serves as a model for all genetic screening. Newborn

screening identifies affected individuals and carriers [McCabe and McCabe,

2004]. There is also an example of the possible population screening beyond the

newborn period. Prevention is one goal of screening for adult-onset disorders

such as type 2 diabetes mellitus. One form of type 2 diabetes is maturity-onset

diabetes of the young (MODY) (27). Diagnosing MODY relies on three genera-

tions with autosomal dominant diabetes and 2 patients with onset at �25 years

of age. Thirteen percent of screened family members of Scandinavian patients

had a mutation in one of the four MODY genes [Lehto et al., 1999]. If screening

could be performed before adolescence, it might be possible to prevent obesity

associated with MODY through diet and exercise. Many of the interventions for

disorders such as diabetes mellitus, obesity, heart disease, and even some forms

of cancer involve lifestyle changes including exercise and diet.

Genetic screening in some instances may easily lead to prejudice and dis-

crimination against a whole group or population. The same things can happen

in the case of nutrigenetics. The example of public reactions to a publication by

Rushton [1995] that suggests Asians in general have a higher IQ compared to

Caucasians shows how sensitive this kind of issue is, although this example is

not a result of genetic screening. The same kind of reaction may occur when a

certain conclusion based on extensive genetic survey reveals significant differ-

ences not only between individuals but also between groups and populations.

McCabe and McCabe [2004] stated that informed population screening

applies the technology for the benefit of populations and the individuals com-

posing those populations, and avoids health care disparities and there must be

adequate information for each ethnic or cultural group regarding mutation fre-

quency and penetrance. However, the dilemma is that this could lead to the

stigmatization of certain ethnic groups concerning certain diseases. One of the


differences within food and medicine is that individuals make food choices for

a variety of reasons which are nonmedical and are more related to social issues

such as expressing their sense of identity.

The issue of prevention of inequity is also very much interwoven with pri-

vacy. Anderlik and Rothstein [2001] explained that rules protecting the privacy

of genetic information are intended to prevent, lessen, or eliminate negative

consequences of the new genetics. For example, recent policy statements from

the American Society of Human Genetics and other organizations have addressed

the appropriate use of predictive genetic testing of children. These statements

discourage testing of children for adult-onset disorders and disorders that can-

not be ameliorated or cured.

It was also written in the same paper that adopting rules that restrict access

to genetic information by third parties greatly reduces the potential for genetic

discrimination in insurance and employment and a range of negative outcomes,

including the destabilization of the insurance market, the sidelining of produc-

tive employees, and an erosion of social solidarity, culminating in the creation

of a ‘genetic underclass’ of uninsurables and unemployables. Prohibitions

against genetic discrimination are virtually meaningless without limitations on

the access to genetic information. Laws protecting the privacy of health infor-

mation and prohibiting genetic discrimination are included in most jurisdic-

tions, but there are gaps in these laws and in the social safety net. McCabe and

McCabe [2004] suggested avoiding health care inequalities; furthermore,

screening should be offered at a reasonable price or as part of a free public

health program.

Other Social IssuesData interpretation, drawing conclusions and giving a balanced recom-

mendation are also a major issue. As mentioned earlier, nutrigenomics in cer-

tain respects is more complex compared to pharmacogenomics; therefore, the

amount of data has to be greater, and consequently interpretation and drawing

conclusions of statistic data has to be done very carefully. Genetic screening

can help individuals to take certain preventions including choosing a certain

type of diet, but this knowledge may also lead individuals to take drastic pre-

ventions that may not be necessary such as the case of BRCA1 testing in fami-

lies with BRCA1-linked hereditary breastovarian cancer. Genetic screening

clearly requires a proper consultation and balanced recommendation. Certain

types of diet may increase the chances of onset of a particular disease in a cer-

tain genetic makeup, but the chances that this happens is actually very small;

however, the same diet may be important for other health reasons.

In relation to health risks, Suzuki and Knudtson [1989] gave examples in

the case of occupational genetic screening; individual differences in DNA often


reveal statistical risks for occupational diseases that are inconsequential com-

pared to the routine health risks faced by workers every day at home or on the

job. A genotype that makes a worker 10 times more likely to suffer from an

occupational disorder that only occurs in 1 out of 10,000 people may be intrigu-

ing, but in fact far less risky than commuting in busy traffic to a factory.

Regarding public health issues, there is a concern that the interest in this

new field of research could well overshadow other important necessities in the

field of nutrition worldwide [Chadwick, 2004]. The question of whether it is

ethical to allocate resources for a certain nutrigenomics research should be

answered at the beginning of the project. Resources of developing countries

should first of all be allocated to the tackling of basic public health problems

addressed in the millennium goals such as malnutrition, diet-related diseases as

well as contaminated food; this, however, should not exclude research in the

field of nutrigenomics in developing countries. It only means that allocation of

resources should be balanced.

A way of assessing when applications of nutrigenomics in public health

would be worthwhile is needed [Chadwick, 2004]. From a public health perspec-

tive, one might even go one step further and ask whether people will actually use

the test results to alter their behavior in ways that improve health [Clayton, 2003].

Another concern is how the commercialization of this technology affects

the type of applications that will be developed, and who will have access to

them. It has to be shown that the results of this technology will not only benefit

the developing countries or small groups of people. An example of nutrige-

nomics research such as green tea consumption that can reduce the chances of

developing breast cancer shows the type of information that is useful for the

general public. This confirms that this technology does not have to be applica-

ble to the privileged few only.

Regulatory Oversight and Public Awareness

International ethics committees in medical research such as the Human

Genome Organization Ethics Committee have already released a set of ethical

guidelines for conducting research in these areas. As mentioned above, the

general conferences of UNESCO have adopted three declarations, i.e. the

Universal Declaration on the Human Genome and Human Rights [UNESCO,

1997], the International Declaration of Human Genetic Data [UNESCO, 2003]

and the Universal Declaration on Bioethics and Human Rights [UNESCO,

2005], while at present in many countries there are no specific ethical guide-

lines available yet for the implementation of ethical principles in nutrigenomics

research. As far as possible, domestic regulations and guidelines should be


made available in accordance with the UNESCO declarations. However, certain

adaptations to the local culture and legal framework cannot always be avoided.

Efforts to improve the public awareness, public consultation and participa-

tion need to be made at the early level of development of nutrigenomics, by giving

balanced information and allowing the public to be involved in the decision-

making process. The example of the current public reactions in some countries

to genetically modified food shows how important food issues are for many

individuals; these issues are not only centering on the food safety but also

include the freedom of choice. In general, there is an urgent need to improve

public confidence in biotechnology-related science.

Public communication of the emerging science with the public should be

pursued. Without public participation, there is a very real risk that the public

will turn against this emerging genetic research if projects come to light that

violate public expectations of the protection of privacy and autonomy [Anderlik

and Rothstein, 2001]. When conducting public consultation at different stages

of the research and policy setup, researchers and policy makers need to recog-

nize the importance of local cultures and social systems, values and beliefs.

The International HapMap Consortium [2004] shared their experience that

asking people respectfully about participating in projects of this type, providing

complete, balanced and accurate information, giving them a chance to express

their views, and, where possible, incorporating their input need not unduly

impede research. Indeed, it can create a climate in which research proceeds in

an atmosphere of openness and trust.

Information has to be relayed to the patients, donors, health care workers

and the general public. The Internet may be an important means apart from oth-

ers since it can reach various lay audiences [Guttmacher, 2001]. Information

about the ‘central dogma’ of genetics to those seeking to inform concerned cit-

izens about ethical, legal, and social aspects has to be easily accessible. The

Web has been seen as an egalitarian way to relay information, but on the other

hand it has to be realized that the Internet should not be the only means to

inform the public since there is still limited Internet access in some localities in

the developing and less developed countries.

Concluding Remarks

Ethical issues raised by rapid advances in science and technology applica-

tions should be examined with respect for the inherent dignity of the human

person and with universal respect for, and observance of, human rights and fun-

damental freedoms, as mentioned in the Universal Declaration on Bioethics and

Human Rights [UNESCO, 2005].


Consent, privacy and confidentiality as well as equitable access to the

technology are the major ethical issues that need to be considered when coun-

tries develop their guidelines, regulation and laws related to nutrigenomics.

Public consultations at different stages of research and the development of a

nutrigenomic project involving a large population need to be carried out to ful-

fill the right to know and to create an atmosphere of openness and trust.

A concise and clear set of guidelines to undertake research in nutrige-

nomics needs to be developed consistent with the Universal Declaration on

Bioethics and Human Rights [UNESCO, 2005] and the International

Declaration on Human Genetic Data [UNESCO, 2003].

A way of assessing when applications of nutrigenomics in public health

would be worthwhile is needed. Efforts to improve the public awareness and

public participation in nutrigenomics need to be made in this new branch of sci-

ence right from the start. Nutrigenomics research that is beneficial for the gen-

eral public needs to be undertaken to prove that this technology does not have to

be applicable to privileged people only.

References

Anderlik MR, Rothstein MA: Privacy and confidentiality of genetic information: what rules for the new

science? Annu Rev Genomics Hum Genet 2001;2:401–433.

Burton H, Steward A: Nutrigenomics: Report of a Workshop Hosted by The Nuffield Trust and

Organized by The Public Health Genetics Unit in February 2004. London, The Nuffield Trust, 2005.

Castro LD: Informed consent: what information? Whose consent? Proc Int Joint Bioethics Congr on

Intercult Bioethics Asia and the West, Sanliurfa, November 2005.

Chadwick R: Nutrigenomics, individualism and public health. Proc Nutr Soc 2004;63:161–166.

Clayton EW: Ethical, legal, and social implications of genomic medicine. N Engl J Med 2003;349:

562–569.

Corrigan OP: Pharmacogenetics, ethical issues: review of the Nuffield Council on Bioethics Report.

J Med Ethics 2005;31:144–148.

Guttmacher AE: Human genetics on the web. Annu Rev Genomics Hum Genet 2001;2:213–233.

International HapMap Consortium: Integrating ethics and science in the International HapMap Project.

Nat Rev Genet 2004;5:467–475.

Kaput J: Decoding the pyramid: a systems-biological approach to nutrigenomics. Ann NY Acad Sci

2005;1055:64–79.

Lehto M, Wipemo C, Ivarsson SA, Lindgren C, Lipsaren-Nyman M, et al: High frequency of mutations

in MODY and mitochondrial genes in Scandinavian patients with familial early-onset diabetes.

Diabetologia 1999;42:1131–1137.

Mathers J: The science of nutrigenomics; in Burton H, Steward A (eds): Nutrigenomics: Report of a

Workshop Hosted by The Nuffield Trust and Organized by The Public Health Genetics Unit on 5

February 2004. London, The Nuffield Trust, 2005.

McCabe LL, McCabe ERB: Genetic screening: carriers and affected individuals. Annu Rev Genomics

Hum Genet 2004;5:57–69.

Muller M, Kersten S: Nutrigenomics: goals and strategies. Nat Rev Genet 2003;4:315–322.

Nuffield Council on Bioethics: Pharmacogenetics: ethical issues. Summary and recommendations.

2003. www.nuffieldbioethics.org/file/library/pdf/pharm_summary_chapter.pdf.


Rushton P: Race, Evolution and Behaviour. A Life History Perspectives. New Brunswick, Transaction

Publishers, 1995.

Suzuki D, Knudtson P: Genethics. The Ethics of Engineering Life. Cambridge, Harvard University

Press, 1989.

UNESCO: Universal Declaration on the Human Genome and Human Rights. 1997.

UNESCO: International Declaration on Human Genetic Data. 2003.

UNESCO: Universal Declaration on Bioethics and Human Rights. 2005.

Dr. Inez H. Slamet-Loedin

Indonesian Institute of Sciences, National Bioethics Commission

Sasana Widya Sarwono, Jl, Gatot Subroto 10

Jakarta 12710 (Indonesia)

Tel. �62 875 4627/5873, Fax �62 875 4588, E-Mail [email protected]



Proteomics

Visith Thongboonkerd

Siriraj Proteomics Facility, Medical Molecular Biology Unit, Office for Research and

Development, Faculty of Medicine Siriraj Hospital, Mahidol University, Bangkok,

Thailand

AbstractProteomics has been widely applied to several biomedical fields in recent years. The

high-throughput capability of proteomics allows simultaneous examination of numerous pro-

teins and offers the possibility of a global analysis of proteins in cells, tissues or biofluids.

The rapid progress in the field of proteomics is based primarily on the success of protein

separation sciences (either gel-based or gel-free techniques) and recent advances of mass

spectrometry. Unlike the genome, the proteome is dynamic and varies according to cell type

and functional state of the cell. In addition, gene expression does not always correlate with

protein expression as one gene can be modified to be several products or proteins that

directly govern cellular function. Thus, proteome analysis is expected to provide a wealth of

useful information in nutrition research on the effects of nutrients or food components on

metabolic pathways. Such research allows experts to explore the regulatory mechanisms for

maintaining normal homeostasis during nutritional imbalance, to better understand the path-

ogenic mechanisms and pathophysiology of nutritional disorders, to define molecular targets

of bioactive food components and to identify biomarkers that can be used as diagnostic, pre-

dictive or prognostic factors. This paper will provide a brief overview of proteomics, a sum-

mary of current proteomic technologies and an example of proteomic application to nutrition

research. Finally, the concept of systems biology, which involves integrative ‘omics’ (i.e.,

combining genomics, transcriptomics, proteomics, lipomics and metabolomics) as well as

bioinformatics and modeling, will be discussed. Due to the extent of information that can be

obtained from systems biology, this ideal approach holds great promise for future nutrition

research.


It is well known that nutrition is closely related to health status and illness.

Dietary habits and bioactive food components can modulate some medical dis-

eases. Current knowledge on how these food components can modulate health

status or diseases remains unclear. Because enzymes in metabolic pathways and

Proteomics 81

most molecular targets for nutrients are proteins, it is most likely that extensive

analysis on these proteins will lead to a better understanding of molecular

effects of nutrients or bioactive food components on health status and diseases.

Traditional study of proteins involved in nutritional sciences has relied on con-

ventional biochemical methods. This approach is based primarily on hypothesis-

driven research, focusing on a specific pathway of nutritional metabolism.

Although successful, it is time-consuming and the molecular targets to be stud-

ied require a priori assumption. Thus, the number of molecular targets currently

known is underestimated and additional methods are required to screen for a

large number of novel targets simultaneously. In the postgenomic era, when the

Human Genome Project has been completed, several biotechnologies have

been developed to utilize the genomic information to examine other cellular

compositions (e.g. proteins, transcripts, metabolites, and lipids) on the genomic

scale. Since then, respective ‘omics’ fields (e.g. proteomics, transcriptomics,

metabolomics, and lipomics) have been defined and widely applied to biomed-

ical research. The successfulness of these omics fields is mainly due to

advances in separation sciences (either gel-based or gel-free methods) and mass

spectrometry (MS). The great contribution of MS technology to life science has

been confirmed as the 2002 Nobel prize in chemistry went to John B. Fenn

(who developed electrospray ionization; ESI) and Koichi Tanaka (who devel-

oped matrix-assisted laser desorption/ionization; MALDI) [1].

Brief Overview of Proteomics

Since the term ‘proteome’ was coined for the first time in the public by

Marc Wilkins in 1994, proteomics, which is a subject or an area of sciences to

study the proteome, has been widely applied to several biomedical fields. For

human studies, proteomics has been used with five main objectives: (1) to bet-

ter understand human physiology; (2) to explore and better understand the com-

plexity of pathogenic mechanisms and the pathophysiology of diseases; (3) to

define new therapeutic targets for better therapeutic outcome; (4) to identify

novel biomarkers for earlier diagnosis, and (5) to identify molecular targets for

vaccine development for disease prevention. From 1995 through the end of

November 2005, approximately 11,000 proteomics-related articles can be found

in PubMed (using the key words ‘proteomics’, ‘proteome’ or ‘proteomic’).

Interestingly, half of them appear in the last 2 years. These numbers in a wide

variety of international journals imply the rapid progress of the proteomics

field and its utility, as well as applicability.

The concept of proteomics can be differentiated from that of conventional

protein chemistry. Protein chemistry aims to extensively examine the protein

Thongboonkerd 82

structure and physicochemical properties of each protein in detail, whereas pro-

teomics aims to simultaneously examine a large number of proteins or the entire

proteome in a complex mixture of biological samples (more details for protein

chemistry, whereas broader contents for proteomics). Both of them are aimed to

better understand the cellular biology and physiology. There are some overlaps

between these two fields, e.g. the area of structural proteomics. Therefore, these

two fields are different but closely related and complementary. Although they

can be distinguished based on their concepts, these two fields cannot be com-

pletely separated in practice or in a study on a large number of proteins.

When should proteomics be applied? First, the investigators need to have

clear specific aims and future plans for the proteomics project. Otherwise, the

technology can be misused. The investigators need to design whether they

really need the high-throughput analysis. Second, the nature of the study project

can determine the type of technologies to be applied. Proteomics is suitable for

fishing expedition or screening for candidate proteins, whereas the conven-

tional methods are suitable for hypothesis-driven research in which some leads

or prior results are required before proceeding. Third, funding support should

be sufficient for a proteomic study as it is costly. Finally, specific instrumenta-

tions such as two-dimensional polyacrylamide gel electrophoresis (2D-PAGE),

liquid chromatography (LC), capillary electrophoresis (CE), protein chip instru-

mentation, as well as a mass spectrometric system are needed.

‘Proteome analysis’ (or ‘proteomic analysis’, which will be used inter-

changeably in this article) can be divided into two categories based upon analy-

tical strategies. The ‘classical approach’ involves extensively and systematically

examining proteins for their expression and function, whereas the ‘alternative

approach’ bypasses the complicated analytical procedures in the classical approach

and involves proteome profiling just to differentiate the types of biological sam-

ples (e.g. normal versus disease; a specific disease versus other diseases). The

former is suitable for unraveling the pathophysiology of diseases, whereas the lat-

ter is suitable for clinical diagnostics and biomarker screening. Techniques that

are used in the alternative approach are microarrays, surface-enhanced laser des-

orption/ionization time-of-flight (SELDI-TOF), and CE coupled to MS.

Current Proteomic Technologies

Gel-Based Methods

Currently, 2D-PAGE is the most commonly used method in proteomic

studies. The principles of protein separation by 2D-PAGE are based on differen-

tial pH or isoelectric point (pI) for the first dimension and differential molecular

Proteomics 83

size (Mr) for the second dimensional separation [2]. Resolved proteins in 2D

gel can be visualized by various stains (e.g. Coomassie brilliant blue, silver, flu-

orescence). Visualized protein spots can then be excised, in-gel digested with

proteolytic enzymes (e.g. trypsin, chymotrypsin, Arg-C, Asp-N, Lys-C, pepsin A,

V8-E, V8-DE), and identified by MALDI-MS followed by peptide mass finger-

printing. The most common type of mass analyzer employed in MALDI analy-

sis is time-of-flight (TOF). MALDI-TOF-MS provides a high-throughput

manner of protein identification; hundreds of proteins can be identified within

a day [3–5]. Consequently, MALDI-TOF-MS has become an integral part of

today’s modus operandi in proteome analysis. Even with lots of advantages, the

gel-based approach has some limitations. 2D-PAGE procedures are time-

consuming, and low-abundant, transmembrane, and highly hydrophobic pro-

teins may not be detectable in a 2D gel.

Gel-Free Methods

Liquid Chromatography Coupled to Tandem Mass SpectrometryCoupling of high-performance LC to ESI-tandem MS (MS/MS) has

gained a wide acceptance for gel-free proteomic analysis and become a method

of choice for the analysis of membrane and low-abundant proteins [6–8]. ESI is

the process of ionization from the electrospray source, whereas MS/MS refers

to the strategy of multistep mass analyses. When compared to the 2D-PAGE

approach, LC-based methods are more effective for the analysis of small pro-

teins and peptides, as well as membrane and highly hydrophobic proteins.

Recently, a high-throughput LC approach has been developed, namely ‘multi-

dimensional protein identification technology’ or 2D-LC-MS/MS [9]. This

approach involves proteolytic digestion of the total protein mixture to obtain a

set of protein-derived peptides that are then separated by strong cation

exchange chromatography (‘bottom-up’ approach). Peptides present in frac-

tions from this strong cation exchange step are separated further by reversed-

phase LC and then sequenced by MS/MS. Several thousand of peptides can be

sequenced in this way in a relatively short time. In another methodology (‘top-

down’ approach), which is in contrast to the bottom-up approach, the undi-

gested proteins in the complex mixture are separated by high-performance LC

prior to digestion with proteolytic enzymes and MS/MS sequencing [10, 11].

Surface-Enhanced Laser Desorption/Ionization or Protein Chip TechnologyThis method is suitable for proteome profiling. SELDI-TOF-MS combines

MALDI-TOF-MS with surface retentate chromatography. A protein sample is

Thongboonkerd 84

applied onto a chip surface carrying a functional group (e.g. normal phase,

hydrophobic, cation or anion exchange). After incubation, proteins that do not

bind to the surface are removed by a simple washing step and bound peptides/

proteins are analyzed by a TOF mass spectrometer. The detection of a protein by

SELDI-TOF-MS is critically determined by its concentration in the sample, its

binding to the chromatographic surface and its ionization process within the

mass spectrometer. This approach reduces the complexity of the sample being

analyzed by selecting only a subset of proteins. Only 5–10 �l of sample are

needed for a single analysis and this method can be readily automated, making

it particularly useful for high-throughput studies.

Capillary Electrophoresis Coupled to Mass SpectrometryCE-MS is another method suitable for proteome profiling. It offers some

advantages as it is fairly robust, uses inexpensive capillaries and is compatible

with essentially all buffers and analytes [12–15]. In contrast to LC, CE gener-

ally has no flow rate but requires a closed electric circuit. Various MS coupling

techniques can be applied to CE [16, 17]. The predominant ionization method

for CE-MS is ESI, while MALDI has also been used extensively [18, 19]. The

main advantages of MALDI appear to be the enhanced stability as well as eas-

ier handling compared to ESI. Additionally, once the analytes are deposited on

the target, they can be reanalyzed several times without the need of a new CE

run. Moreover, the deposited analytes can be subsequently manipulated. The

disadvantages of MALDI are certainly the decreased dynamic range in compar-

ison to ESI and the higher sensitivity towards signal suppression. For the detec-

tion of the narrow CE-separated analyte zones, a fast and sensitive mass

spectrometer is required. Both ion trap and TOF systems appear adequate.

While ion trap MS acquires data over a suitable mass range with the rate of sev-

eral spectra per second, the resolution is generally too low to resolve the single

isotope peaks of �3-fold charged molecules. Consequently, assignment of

charge to these spectra is hampered. Modern ESI-TOF mass analyzers record

up to 20 spectra per second and provide the resolution of more than 10,000 and

a mass accuracy of better than 5 ppm. Therefore, the most suitable mass spec-

trometer for this type of analysis, to date, is ESI-TOF-MS.

Mass Spectrometric Immunoassay Mass spectrometric immunoassay combines immunoassays with MALDI-

TOF-MS [20–33]. Proteins are first captured by microscale affinity techniques

and are subsequently examined qualitatively and quantitatively using MALDI-

TOF-MS [34]. This approach has the potential for greatly extending the range,

utility and speed of biological research and clinical assays. In the initial phase

of development, agarose beads (derivatized with an affinity ligand) were used

Proteomics 85

to create a microliter-volume column inside a micropipettor tip (thus, creating

an affinity pipette) [20, 22]. More recently, tailored affinity micropipettes with

a high flow rate and a high binding capacity have been manufactured and used

in combination with robotic platforms for the preparation of up to 96 samples in

parallel [24, 28, 32]. Using this approach, the proteins of interest are selectively

retained and concentrated by repeat flowing through the affinity pipette. After

washing to remove unspecified compounds, the retained proteins are eluted,

mixed with a MALDI matrix (i.e. �-cyano-4-hydroxycinnamic acid), and tar-

geted onto the MALDI plate. The eluted proteins are then analyzed with a mass

spectrometer (TOF-MS).

Applications of Proteomics to Nutrition Research

Proteomics applied to nutrition research or ‘nutriproteomics’ is now at its

infancy phase as compared to proteomics applied to other biomedical fields.

Even with several review articles on nutriproteomics during the past few years

[35–45], the number of original research on nutriproteomics is small [46, 47].

Herein, I provide an example of proteomic application to nutrition research

using an animal model of chronic potassium (K�) depletion induced by inade-

quate K� intake [fed with a K�-depleted (KD) diet] [46].

K� is one of the most important electrolytes that are crucial for maintain-

ing the normal homeostasis of living cells, tissues and organs. The kidney is the

major organ responsible for regulating normal K� homeostasis. Abnormal K�

balance may occur when K� intake and its output are imbalanced. Inadequate

dietary K� intake, renal K� loss (excessive urinary K� excretion) and/or extr-

arenal K� loss (e.g. diarrhea and vomiting) can lead to hypokalemia, which is

widely defined as a serum K� level of less than 3.5 mmol/l. Hypokalemia may

be asymptomatic if the deficit is temporary and the degree of the deficit is mod-

est to mild (3.0–3.5 mmol/l), but can be a cause of death when the degree of the

deficit is severe (�3.0 mmol/l) and the dysregulation is left untreated.

Prolonged K� deficiency can affect several organ systems, especially hemody-

namic, cardiovascular, muscular, gastrointestinal and renal systems [48–50].

Renal involvement of prolonged K� depletion has been defined as ‘hypoka-

lemic nephropathy’, a disease known for half a century [51–53], and is associ-

ated with metabolic alkalosis, growth retardation, hypertension, polydipsia,

polyuria, enlarged kidney, progressive tubulointerstitial injury and ultimately

renal failure or end-stage renal disease [54–59]. Even with a long history of the

disease, the pathophysiology of hypokalemic nephropathy remains unclear.

Although K� repletion can reverse renal ultrastructural changes that occur in

an acute K� deficiency state [60], these changes and renal dysfunction remain

Thongboonkerd 86

in some cases of chronic K� deficiency, even with the repletion therapy

[54, 61].

A recent study by our group has applied proteomic technology to discover

previously unknown changes in renal protein expression that are associated with

hypokalemic nephropathy [46]. Hypokalemia was induced by giving ad libitum

KD diet to BALB/c mice for 8 weeks, whereas the control mice received normal

K� chow. The KD mice displayed many characteristics of human hypokalemic

nephropathy, including severe hypokalemia, growth retardation, polydipsia,

polyuria, markedly enlarged kidneys, severe tubular dilatation, intratubular

deposition of amorphous and laminated hyaline materials, as well as tubular

atrophy. Gel-based, differential proteome analysis of the kidney (using 2D-

PAGE and quantitative intensity analysis) revealed altered expression of 33 renal

proteins in KD mice. Using MALDI-TOF-MS and quadrupole-TOF-MS/MS,

30 of the altered proteins were identified, including metabolic enzymes (e.g. car-

bonic anhydrase II, aldose reductase, glutathione S-transferase GT41A), signal-

ing proteins (14-3-3�, 14-3-3� and cofilin 1) and cytoskeletal proteins (�-actin

and tropomyosin) [46]. Some of these altered proteins, particularly metabolic

enzymes and signaling proteins, have been demonstrated to be involved in meta-

bolic alkalosis, polyuria and renal tubular injury. Our findings may lead to a new

road map for research on hypokalemic nephropathy and to a better understand-

ing of the pathophysiology of this medical disease.

In addition to dietary-induced K� depletion, nutriproteomics is expected to

provide a wealth of useful information in nutrition research to study effects of

nutrients or food components on metabolic pathways, to explore the regulatory

mechanisms for maintaining normal homeostasis during nutritional imbalance,

to better understand the pathogenic mechanisms and pathophysiology of nutri-

tional disorders, to define molecular targets of bioactive food components and

to identify biomarkers that can be used as diagnostic, predictive or prognostic

factors.

Integrative Omics and Systems Biology

It is unlikely that the complexity of nutrition science will be completely

understood only by proteomics or by any other single omics approach.

Integrating all of them is required for future nutrition research. Recently, the

concept of ‘systems biology’ has been emerging for the global evaluation of

biological systems and has included ‘integrative omics’ as one of analytical

processes [62, 63]. Systems biology has been defined by Weston and Hood [64]

as ‘the analysis of the relationships among the elements in a system in response

to genetic or environmental perturbations, with the goal of understanding the

Proteomics 87

system or the emergent properties of the system’. A system may be a few pro-

tein molecules carrying out a particular task such as galactose metabolism

(termed a biomodule), a complex set of proteins and other molecules working

together as a molecular machine such as the ribosome, a network of proteins

operating together to carry out an important cellular function such as giving the

cell shape (protein network), or a cell or group of cells carrying out particular

phenotypic functions. Thus, a biological system may encompass molecules,

cells, organs, individuals, or even ecosystems [64]. Advances in the high-

throughput platforms of biotechnologies have allowed the simultaneous study

of a large complement of genes, transcripts, proteins, lipids or other elements.

Systems biology or integrative omics is, thus, the ideal approach for future

nutrition research.

Conclusions

The current knowledge in nutrition science can be enhanced by recent

advances in the postgenomic biotechnologies, particularly proteomics. Using

gel-based and/or gel-free proteomic methodologies, a large number of proteins

can be examined simultaneously in a single experiment. The high-throughput

capability of proteomics, thus, holds a potential promise in nutrition research.

However, it is unlikely that the complete dynamic image of nutrition science

will be obtained by a single omics approach. With the concept of systems biol-

ogy, integrating proteomics into the other omics sciences is, therefore, the ideal

approach for future nutrition research.

References

1 Cho A, Normile D: Nobel Prize in Chemistry. Mastering macromolecules. Science 2002;298:

527–528.

2 Klein E, Klein JB, Thongboonkerd V: Two-dimensional gel electrophoresis: a fundamental tool for

expression proteomics studies; in Thongboonkerd V, Klein JB (eds): Proteomics in Nephrology.

Contrib Nephrol. Basel, Karger, 2004, vol 141, pp 25–39.

3 Henzel WJ, Watanabe C, Stults JT: Protein identification: the origins of peptide mass fingerprint-

ing. J Am Soc Mass Spectrom 2003;14:931–942.

4 Pierce WM, Cai J: Applications of mass spectrometry in proteomics; in Thongboonkerd V, Klein JB

(eds): Proteomics in Nephrology. Contrib Nephrol. Basel, Karger, 2004, vol 141, pp 40–58.

5 Karas M, Gluckmann M, Schafer J: Ionization in matrix-assisted laser desorption/ionization:

singly charged molecular ions are the lucky survivors. J Mass Spectrom 2000;35:1–12.

6 McCormack AL, Schieltz DM, Goode B, Yang S, Barnes G, Drubin D, Yates JR 3rd: Direct analy-

sis and identification of proteins in mixtures by LC/MS/MS and database searching at the low-

femtomole level. Anal Chem 1997;69:767–776.

7 Wu CC, MacCoss MJ, Howell KE, Yates JR 3rd: A method for the comprehensive proteomic

analysis of membrane proteins. Nat Biotechnol 2003;21:532–538.

Thongboonkerd 88

8 Gygi SP, Rist B, Griffin TJ, Eng J, Aebersold R: Proteome analysis of low-abundance proteins

using multidimensional chromatography and isotope-coded affinity tags. J Proteome Res 2002;1:

47–54.

9 Washburn MP, Wolters D, Yates JR 3rd: Large-scale analysis of the yeast proteome by multidi-

mensional protein identification technology. Nat Biotechnol 2001;19:242–247.

10 Mabuchi H, Nakahashi H: Profiling of urinary medium-sized peptides in normal and uremic urine

by high-performance liquid chromatography. J Chromatogr 1982;233:107–113.

11 Heine G, Raida M, Forssmann WG: Mapping of peptides and protein fragments in human urine

using liquid chromatography-mass spectrometry. J Chromatogr A 1997;776:117–124.

12 Jensen PK, Pasa-Tolic L, Peden KK, Martinovic S, Lipton MS, Anderson GA, Tolic N, Wong KK,

Smith RD: Mass spectrometric detection for capillary isoelectric focusing separations of complex

protein mixtures. Electrophoresis 2000;21:1372–1380.

13 Gelpi E: Interfaces for coupled liquid-phase separation/mass spectrometry techniques. An update

on recent developments. J Mass Spectrom 2002;37:241–253.

14 Schmitt-Kopplin P, Frommberger M: Capillary electrophoresis-mass spectrometry: 15 years of

developments and applications. Electrophoresis 2003;24:3837–3867.

15 Hernandez-Borges J, Neususs C, Cifuentes A, Pelzing M: On-line capillary electrophoresis-mass

spectrometry for the analysis of biomolecules. Electrophoresis 2004;25:2257–2281.

16 Kolch W, Neususs C, Pelzing M, Mischak H: Capillary electrophoresis-mass spectrometry as a

powerful tool in clinical diagnosis and biomarker discovery. Mass Spectrom Rev 2005;24:

959–977.

17 Simpson DC, Smith RD: Combining capillary electrophoresis with mass spectrometry for appli-

cations in proteomics. Electrophoresis 2005;26:1291–1305.

18 Musyimi HK, Narcisse DA, Zhang X, Stryjewski W, Soper SA, Murray KK: Online CE-MALDI-

TOF MS using a rotating ball interface. Anal Chem 2004;76:5968–5973.

19 Zuberovic A, Ullsten S, Hellman U, Markides KE, Bergquist J: Capillary electrophoresis off-line

matrix-assisted laser desorption/ionisation mass spectrometry of intact and digested proteins

using cationic-coated capillaries. Rapid Commun Mass Spectrom 2004;18:2946–2952.

20 Nelson RW, Krone JR, Bieber AL, Williams P: Mass spectrometric immunoassay. Anal Chem 1995;67:

1153–1158.

21 Nelson RW, Krone JR, Dogruel D, Tubbs KA, Granzow R, Jansson O: Interfacing biomolecular

interaction analysis with mass spectrometry and the use of bioreactive mass spectrometer probe

tips in protein characterization; in Marshak DR (ed): Techniques in Protein Chemistry. San Diego,

Academic Press, 1997, vol 8, pp 493–504.

22 Krone JR, Nelson RW, Williams P: Mass spectrometric immunoassay. Proc Soc Photo Opt Instrum

Eng 1996;2680:415–421.

23 Tubbs KA, Nedelkov D, Nelson RW: Detection and quantification of beta-2-microglobulin using

mass spectrometric immunoassay. Anal Biochem 2001;289:26–35.

24 Niederkofler EE, Tubbs KA, Gruber K, Nedelkov D, Kiernan UA, Williams P, Nelson RW:

Determination of beta-2 microglobulin levels in plasma using a high-throughput mass spectromet-

ric immunoassay system. Anal Chem 2001;73:3294–3299.

25 Niederkofler EE, Tubbs KA, Kiernan UA, Nedelkov D, Nelson RW: Novel mass spectrometric

immunoassays for the rapid structural characterization of plasma apolipoproteins. J Lipid Res

2003;44:630–639.

26 Kiernan UA, Nedelkov D, Tubbs KA, Niederkofler EE, Nelson RW: Selected expression profiling

of full-length proteins and their variants in human plasma. Clin Proteomics 2004;1:7–16.

27 Kiernan UA, Nedelkov D, Tubbs KA, Niederkofler EE, Nelson RW: Proteomic characterization of

novel serum amyloid P component variants from human plasma and urine. Proteomics 2004;4:

1825–1829.

28 Kiernan UA, Tubbs KA, Gruber K, Nedelkov D, Niederkofler EE, Williams P, Nelson RW: High-

throughput protein characterization using mass spectrometric immunoassay. Anal Biochem 2002;

301:49–56.

29 Kiernan UA, Tubbs KA, Nedelkov D, Niederkofler EE, McConnell E, Nelson RW: Comparative

urine protein phenotyping using mass spectrometric immunoassay. J Proteome Res 2003;2:

191–197.

Proteomics 89

30 Kiernan UA, Tubbs KA, Nedelkov D, Niederkofler EE, Nelson RW: Comparative phenotypic

analyses of human plasma and urinary retinol binding protein using mass spectrometric immuno-

assay. Biochem Biophys Res Commun 2002;297:401–405.

31 Kiernan UA, Tubbs KA, Nedelkov D, Niederkofler EE, Nelson RW: Detection of novel truncated

forms of human serum amyloid A protein in human plasma. FEBS Lett 2003;537:166–170.

32 Nedelkov D, Tubbs KA, Niederkofler EE, Kiernan UA, Nelson RW: High-throughput comprehen-

sive analysis of human plasma proteins: a step toward population proteomics. Anal Chem 2004;76:

1733–1737.

33 Nelson RW, Nedelkov D, Tubbs KA, Kiernan UA: Quantitative mass spectrometric immunoassay

of insulin like growth factor 1. J Proteome Res 2004;3:851–855.

34 Nelson RW, McLean MA, Hutchens TW: Quantitative determination of proteins by matrix-assisted

laser desorption/ionization time-of-flight mass spectrometry. Anal Chem 1994;66:1408–1415.

35 Daniel H: Genomics and proteomics: importance for the future of nutrition research. Br J Nutr

2002;87(suppl 2):S305–S311.

36 Go VL, Butrum RR, Wong DA: Diet, nutrition, and cancer prevention: the postgenomic era. J Nutr

2003;133:3830S–3836S.

37 Milner JA: Molecular targets for bioactive food components. J Nutr 2004;134:2492S–2498S.

38 Kim H, Page GP, Barnes S: Proteomics and mass spectrometry in nutrition research. Nutrition

2004;20:155–165.

39 Arab L: Individualized nutritional recommendations: do we have the measurements needed to

assess risk and make dietary recommendations? Proc Nutr Soc 2004;63:167–172.

40 Ross SA, Srinivas PR, Clifford AJ, Lee SC, Philbert MA, Hettich RL: New technologies for nutri-

tion research. J Nutr 2004;134:681–685.

41 Barnes S, Kim H: Nutriproteomics: identifying the molecular targets of nutritive and non-nutritive

components of the diet. J Biochem Mol Biol 2004;37:59–74.

42 Davis CD, Milner J: Frontiers in nutrigenomics, proteomics, metabolomics and cancer prevention.

Mutat Res 2004;551:51–64.



44 Kim H: New nutrition, proteomics, and how both can enhance studies in cancer prevention and

therapy. J Nutr 2005;135:2715–2718.

45 Fuchs D, Winkelmann I, Johnson IT, Mariman E, Wenzel U, Daniel H: Proteomics in nutrition

research: principles, technologies and applications. Br J Nutr 2005;94:302–314.

46 Thongboonkerd V, Chutipongtanate S, Kanlaya R, Songtawee N, Sinchaikul S, Parichatikanond P,

Chen ST, Malasit P: Proteomic identification of alterations in metabolic enzymes and signaling

proteins in hypokalemic nephropathy. Proteomics 2006;6:2273–2285.

47 Mitchell BL, Yasui Y, Lampe JW, Gafken PR, Lampe PD: Evaluation of matrix-assisted laser

desorption/ionization-time of flight mass spectrometry proteomic profiling: identification of

alpha 2-HS glycoprotein B-chain as a biomarker of diet. Proteomics 2005;5:2238–2246.

48 Gennari FJ: Hypokalemia. N Engl J Med 1998;339:451–458.

49 Gennari FJ: Disorders of potassium homeostasis. Hypokalemia and hyperkalemia. Crit Care Clin

2002;18:273–288, vi.

50 Antes LM, Kujubu DA, Fernandez PC: Hypokalemia and the pathology of ion transport mole-

cules. Semin Nephrol 1998;18:31–45.

51 Relman AS, Schwartz WB: The nephropathy of potassium depletion: a clinical and pathological

entity. N Engl J Med 1956;255:195–203.

52 Hollander W Jr: The effect of potassium depletion on the kidneys. N C Med J 1957;18:505–509.

53 Weissmann G, Ludwig AM: Potassium depletion nephropathy. J Mt Sinai Hosp NY 1958;25:

454–458.

54 Tolins JP, Hostetter MK, Hostetter TH: Hypokalemic nephropathy in the rat. Role of ammonia in

chronic tubular injury. J Clin Invest 1987;79:1447–1458.

55 Berl T, Linas SL, Aisenbrey GA, Anderson RJ: On the mechanism of polyuria in potassium deple-

tion. The role of polydipsia. J Clin Invest 1977;60:620–625.

56 Ray PE, Suga S, Liu XH, Huang X, Johnson RJ: Chronic potassium depletion induces renal injury,

salt sensitivity, and hypertension in young rats. Kidney Int 2001;59:1850–1858.

Thongboonkerd 90

57 Kaufman AM, Kahn T: Potassium-depletion alkalosis in the rat. Am J Physiol 1988;255:

F763–F770.

58 Peterson LN, Carpenter B, Guttierrez GA, Fajardo C, Levine DZ: Potassium depletion enhances

renal compensatory hypertrophy in the nephrectomized rat. Miner Electrolyte Metab 1987;13:

57–62.

59 Stetson DL, Wade JB, Giebisch G: Morphologic alterations in the rat medullary collecting duct

following potassium depletion. Kidney Int 1980;17:45–56.

60 Ordonez NG, Toback FG, Aithal HN, Spargo BJ: Zonal changes in renal structure and phospho-

lipid metabolism during reversal of potassium depletion nephropathy. Lab Invest 1977;36:33–47.

61 Fourman P, McCance RA, Parker RA: Chronic renal disease in rats following a temporary defi-

ciency of potassium. Br J Exp Pathol 1956;37:40–43.

62 Morel NM, Holland JM, van der Greef J, Marple EW, Clish C, Loscalzo J, Naylor S: Primer on

medical genomics. 14. Introduction to systems biology – a new approach to understanding disease

and treatment. Mayo Clin Proc 2004;79:651–658.

63 Ge H, Walhout AJ, Vidal M: Integrating ‘omic’ information: a bridge between genomics and

systems biology. Trends Genet 2003;19:551–560.

64 Weston AD, Hood L: Systems biology, proteomics, and the future of health care: toward predic-

tive, preventative, and personalized medicine. J Proteome Res 2004;3:179–196.

Dr. Visith Thongboonkerd

Siriraj Proteomics Facility, Medical Molecular Biology Unit

Office for Research and Development

Faculty of Medicine Siriraj Hospital, Mahidol University

Bangkok (Thailand)

Tel./Fax �66 2 4184793, E-Mail [email protected]



Diet and Genomic Stability

Graeme P. Young

Department of Medicine, Flinders University, Adelaide, Australia

AbstractCancer results from a disordered and unstable genome – the degree of abnormality pro-

gresses as the process of oncogenesis proceeds. Such genomic instability appears to be subject

to control by environmental factors as evidenced by the number of cancers that are either

caused by specific environmental agents (lung, skin, cervix) or else regulated by a broader

range of agents such as effect of diet on gastric and colorectal cancers. Dietary factors might

interact in several ways with the genome to protect against cancer. An agent might interact

directly with the genome and regulate expression (as a genetic or epigenetic regulator) or indi-

rectly by influencing DNA ‘repair’ responses and so improve genomic stability. Research now

shows that diet-genomic interactions in cancer go beyond interactions with the normal genome

and involve enhancement of normal cellular responses to DNA damage such that genome sta-

bility is more effectively maintained. Activation of apoptosis may be a key to protection.


Cancer results from a disordered and unstable genome – the degree of

abnormality progresses as the process of oncogenesis proceeds [1, 2]. Such

genomic instability appears to be subject to control by environmental factors as

evidenced by the number of cancers that are either caused by specific environ-

mental agents (lung, skin, cervix) or else regulated by a broader range of agents

such as effect of diet on gastric and colorectal cancers. How might such envi-

ronmental regulators interact with the genomic events that give rise to cancer?

Furthermore, what is the nature of such interactions in the context of the abnor-

mal genome that arises during oncogenesis?

The model for oncogenesis termed ‘multistep carcinogenesis’ appears to

better explain the genomic instability inherent in cancer [1, 2]. The process is

largely driven by a broad range of genetic alterations, randomly accumulating

in no given sequence, at multiple sites on DNA. It could be thought of as multi-

ple, superimposed, initiation-promotion models, but such would not adequately

Young 92

allow for the biological complexity or widespread genomic instability charac-

teristic of oncogenesis in the colon [1]. Changes in DNA thus drive the process

of oncogenesis and these might arise in several ways [3]. They might arise as a

result of chance events or mistakes especially at the time of DNA replication.

Alternatively, they might be caused by an environmental carcinogen where an

adduct forms from interaction of the carcinogen with the DNA base or by

irradiation.

The consequence of such events will then depend on several factors but

most important is whether or not the cell can act to ‘repair’ the damage [4]. If

damage occurs to DNA, and provided that cellular recognition and surveillance

systems detect this, the cell responds in two main ways. One is cell cycle arrest

to allow DNA repair [5] through enzymes such as the alkylguanine alkyl trans-

ferases (e.g. MGMT). The other involves activation of apoptosis if the mutation

cannot be repaired [6]. If both fail, further checkpoint repair systems may come

into play when the cell attempts to proliferate (S-phase). This is shown dia-

grammatically in figure 1. If repair is effective, it would abort any downstream

consequences while an unrepaired and thus mutated cell resulting from failed

repair might develop into a pro-oncogenic clone.

Dietary factors might therefore interact in several ways with the genome to

protect against cancer. An agent might interact directly with the genome (as a

genetic or epigenetic regulator) or indirectly by influencing ‘repair’ responses.

Furthermore, dietary factors might interact with normal DNA to keep it stable

or to create a biological setting where progression to oncogenesis is less likely.

But it is also possible that stabilizing an already unstable genome might be a

mechanism of protection. The purpose of this review is to explore this latter

possibility.

Potential Mechanisms for the Role of Diet in Maintaining Genome Stability

A cell reacts to a chance mistake or an induced DNA adduct in an effort to

‘repair’ the damage. Some proteins, such as p53, are critical in the surveillance

mechanisms that detect such abnormalities. Others, such as MLH1 and Bub1,

form an essential function of coupling DNA repair to surveillance for muta-

tions. As a result of these processes, two main events occur. To effect repair of a

DNA base, cell cycle arrest is triggered, through proteins such as p21, and

repair enzymes, such as DNA alkyl transferases, restore a gene to normal. The

alternative is destruction of the DNA-damaged cell through activation of

programmed cell death (apoptosis) pathways involving the caspases [7–9]. If

DNA repair is error prone or apoptosis fails, a viable cell may remain that

Diet and Genomic Stability 93

carries a mutation – such might create a biotype that is more prone to progres-

sion to cancer. Cell cycle arrest/DNA repair and apoptosis have been thought to

be sequential with the latter being activated in mitosis if error-prone repair is

detected. Recent evidence points to apoptosis also being a primary response to

DNA damage [4].

Is it possible then that dietary factors might act to enhance these inherent

homeostatic responses to damage? To answer this question, it would be neces-

sary to study the impact of protective agents on the various components of the

cellular response to DNA damage. Despite the strong evidence for dietary fac-

tors protecting against cancers such as colorectal cancer, there has been little

work in this area.

Normal DNA

DNA adduct

Surveillance for damage and/or cell cycle

checkpoints

Error-prone or failed repair

Genomic instability

Cancer

Error-free repair

Apoptotic death

Fig. 1. Diagrammatic representation of a model that could account for the control of

mutations contributing to colorectal oncogenesis. The three shaded boxes represent key

events in the process that act to control the consequences of DNA adduct formation. The

three heavy arrows indicate the major outcomes of inherent surveillance mechanisms for

controlling DNA fidelity in response to adduct formation. Failed repair results in adduct ‘fix-

ation’ as a mutation that is passed on to cell progeny. Genomic instability can itself compro-

mise all control mechanisms. Epigenetic regulation can apply at all stages.

Young 94

A range of dietary factors have been shown to be proapoptotic in vitro.

Some of these are protective in vivo. Butyrate is strongly proapoptotic in vitro

[10]. It achieves this by epigenetic regulation (inhibition of histone deacetylase)

of a gene that leads to activation of the caspase cascade. When dietary fiber is

fed to rodents, active colonic fermentation generates high levels of butyrate

which in turn are associated with an augmentation of the colonic apoptotic

response to a methylating carcinogen (azoxymethane) and protection against

cancer [11–15]. Feeding fermentable wheat bran to rats increases the acute

apoptotic response to genotoxic carcinogen (AARGC) while feeding nonfer-

mentable methylcellulose does not (fig. 2) [7]. Feeding type 2 resistant starch

(such as high-amylose maize starch) also enhances AARGC; furthermore,

butyrate levels in the feces have been shown to correlate significantly with

AARGC in distal colonic crypts suggesting that butyrate is the mediator of this

effect [15].

A model of defective apoptotic response to a methylating carcinogen has

also been developed based on the p53�/� mouse [16]. It has been shown that a

nonsteroidal anti-inflammatory agent (sulindac) restores this defective response

to damage and reduces the risk of cancer caused by the methylating carcinogen.

This same principle is now under test with various dietary agents. These find-

ings raise the possibility that a proapoptotic effect characterizes one class of

protective agents found in the diet.

The possibility that dietary agents might act to enhance other aspects of

the normal cellular response to DNA damage is unclear. It is not clear if alkyl

transferases can be regulated. Some protective agents serve to slow prolifera-

tion, e.g. calcium in colorectal cancer [17]. This is thought to be effective

00.51.01.52.02.53.03.54.04.55.0

8 12Time (h)

Ap

opto

tic c

ells

/cry

pt WB

MC

Fig. 2. Effects of fermentable fiber wheat bran (WB) and nonfermentable fiber

methylcellulose (MC) on genotoxin-induced apoptosis in response to azoxymethane, in the

distal colon, at 8 and 12 h after administration of azoxymethane [7]. The difference in the

effects of fiber on the distal colon was significant (p � 0.01).

Diet and Genomic Stability 95

because DNA is less subject to exogenous damage in a more slowly proliferat-

ing cell. On the other hand, DNA repair mechanisms might also become more

effective.

Conclusions

Diet-genomic interactions in cancer seem likely to go beyond interactions

with the normal genome and involve enhancement of normal cellular responses

to DNA damage such that genome stability is more effectively maintained.

Activation of apoptosis may be a key to protection.

References

1 Boland CR: Malignant tumors of the colon; in Yamada T, Alpers DH, Kaplowitz N, Laine L,

Owyang C, Powell DW (eds): Textbook of Gastroenterology, ed 4. Philadelphia, Lippincott,

Williams and Wilkins, 2003, pp 1940–1989.

2 Carethers JM, Boland CR: Neoplasia of the gastrointestinal tract; in Yamada T, Alpers DH,

Kaplowitz N, Laine L, Owyang C, Powell DW (eds): Textbook of Gastroenterology, ed 4.

Philadelphia, Lippincott, Williams and Wilkins, 2003, pp 557–583.

3 Grady WM: Genomic instability and colon cancer. Cancer Metastasis Rev 2004;23:11–27.

4 Young GP, Ying Hu, Le Leu R, Nyskohus L: Dietary fibre and colorectal cancer: a model for envi-

ronment-gene interactions. Mol Nutr Food Res 2005;49:571–584.

5 Lane DP: Cancer. p53, guardian of the genome. Nature 1992;358:15–16.

6 Hong MY, Chapkin RS, Wild CP, Morris JS, et al: Relationship between DNA adduct levels, repair

enzyme, and apoptosis as a function of DNA methylation by azoxymethane. Cell Growth Differ

1999;10:749–758.

7 Hu Y, Martin J, Le Leu R, Young GP: The colonic response to genotoxic carcinogens in the rat:

regulation by dietary fibre. Carcinogenesis 2002;23:1131–1137.

8 Potten CS, Grant HK: The relationship between ionizing radiation-induced apoptosis and stem

cells in the small and large intestine. Br J Cancer 1998;78:993–1003.

9 Renehan AG, Bach SP, Potten CS: The relevance of apoptosis for cellular homeostasis and tumori-

genesis in the intestine. Can J Gastroenterol 2001;15:166–176.

10 Medina V, Edmonds B, Young GP, James R, et al: Induction of caspase-3 protease activity and

apoptosis by butyrate and trichostatin A (inhibitors of histone deacetylase): dependence on protein

synthesis and synergy with a mitochondrial/cytochrome c-dependent pathway. Cancer Res

1997;57:3697–3707.

11 Cassidy A, Bingham SA, Cummings JH: Starch intake and colorectal cancer risk: an international

comparison. Br J Cancer 1994;69:937–942.

12 Whitehead RH, Young GP, Bhathal PS: Effects of short chain fatty acids on a new human colon

carcinoma cell line (LIM1215). Gut 1987;27:1457–1463.

13 Stephen AM, Cumming JH: Mechanism of action of dietary fibre in the human colon. Nature

1980;284:283–284.

14 Boffa LC, Lupton JR, Mariani MR: Modulation of colonic epithelial cell proliferation, histone

acetylation, and luminal short chain fatty acids by variation of dietary fiber in rats. Cancer Res

1992;52:5906–5912.

15 Le Leu RK, Hu Y, Young GP: Effects of resistant starch and nonstarch polysaccharides on colonic

lumenal environment and genotoxin-induced apoptosis in the rat. Carcinogenesis 2002;23:

713–719.

Young 96

16 Hu Y, Le Leu RK, Young GP: Absence of acute apoptotic response to genotoxic carcinogens in

p53-deficient mice is associated with increased susceptibility to azoxymethane-induced colon

tumours. Int J Cancer 2005;115:561–567.

17 Bonithon-Kopp C, Kronborg O, Giacosa A, Rath U, Faivre J: Calcium and fibre supplementation

in prevention of colorectal adenoma recurrence: a randomised intervention trial. Lancet 2000;356:

1300–1306.

Dr. Graeme P. Young

Department of Medicine, Flinders University

Bedford Park

Adelaide 5042 (Australia)




High-Throughput Genotyping

Jong-Eun Lee

DNA Link Inc., Seoul, Korea

AbstractThere are many genetic variations in the human genome. The most abundant form of

genetic variation is the single nucleotide polymorphism (SNP). SNPs are thought to be

responsible for observable differences in biological processes among individuals of a popu-

lation. Genetic association studies utilizing SNP markers are expected to allow identification

of genetic factors responsible for complex phenotypes like chronic diseases and responses to

various nutritional elements. Success of such studies relies on detecting genetic markers

either directly responsible for the phenotype or the markers with a close relationship with

causative markers. There are over 10 million SNPs reported and each SNP contains limited

genetic information due to the limited number of alleles. To cover these limitations,

researchers have to genotype many SNP markers to find appropriate associations. As a result,

the need for efficient high-throughput SNP genotyping technologies is high and many effi-

cient high-throughput SNP genotyping technologies have been developed. Highly efficient

systems that can handle as many as 500,000 SNPs at a time have been developed and tech-

nological advances have transformed genome-wide association studies into reality.


Background

Genetic epidemiology is a new and rapidly expanding field of epidemiol-

ogy. Recent genetic epidemiologic research has increasingly focused on com-

plex, multifactorial disorders. Due to the development of the human genome

map and advances in molecular technology, the importance of genetic epidemi-

ologic applications has been enlarged. Large-scale population-based studies

requiring close integrative cooperation of genetic and epidemiologic research

will play a key role in near-future research.

Single nucleotide polymorphisms (SNPs) are single-base differences in the

DNA sequences on the genome that can be observed among individuals of a

population and are the most abundant form of DNA variation in the human

Lee 98

genome. SNPs are thought to be associated with many phenotypes like disease

susceptibility, drug responses and differential responses against environmental

factors such as food intake. Each individual has a large number of SNPs on

his/her genome. On average, every 1 kb has an SNP which can be used as a

marker on the genome. There are over 10 million SNPs deposited in the SNP

database (www.ncbi.nlm.nih.gov/SNP/) which provides valuable information

about human diversity [1]. Detection methods for SNPs are amenable to automa-

tion. Recent developments of high-throughput SNP genotyping technologies

have enabled researchers to build high-density SNP maps of the human genome.

Through the use of these high-density maps, researchers will be able to find sub-

tle genetic factors which contribute to the phenotype of interest.

Approaches in Association Studies Using Single Nucleotide Polymorphism Markers

There are two general approaches for association studies using SNP mark-

ers: the candidate gene approach and whole-genome scan. The candidate gene

approach is the most widely used method for exploring an association between a

certain gene and a phenotype. The candidate gene approach is a hypothesis-

based approach aimed to investigate the role of a particular susceptibility gene in

disease etiology, where gene polymorphisms are considered as a risk factor.

Under this method, studies are usually based on assumptions about the role of

specific known polymorphisms in the candidate gene and their effects on the

pathophysiology of a disease. This approach is considered the method of choice

where there is a lack of information about SNPs or affordable genotyping meth-

ods are lacking. Many reliable SNP genotyping technologies were developed for

this type of study design. Technologies like SNaPShot, TaqMan, SNP-IT, Mass

Array, and Invader assays were developed [2–6]. Most of these technologies rely

on polymerase chain reactions to amplify signals except the Invader assay. These

technologies are designed to genotype a single SNP on a single sample at a time

except the Mass Array assay which can genotype up to 10 SNPs on a single sam-

ple simultaneously. Due to the abundant nature of the SNPs and as more SNPs

are discovered and deposited in the SNP database even on a specific gene region,

the need for multiplex genotyping technologies which can handle multiple SNPs

at a single reaction grew. Also the low information content on a single SNP com-

pared to the other types of markers like short tandem repeats forced researchers

to interrogate more SNPs to find genetic signals associated with the phenotype

of interest. To meet these needs, many clever multiplex genotyping systems were

developed. Degrees of multiplexing range from a few SNPs (e.g. 10 for Mass

Array, 12 for SNPstream) to thousands of SNPs (1,536 for Illumina’s BeadArray

High-Throughput Genotyping 99

and 25,000 for Molecular Inversion Probes from ParAllele) [7–10]. Low-end

multiplexing systems like SNPstream and Mass Array are based on single-base

extension assays while the others use allele-specific extension and ligation as the

main biochemical reactions. All of the technologies depend on polymerase chain

reactions to amplify signals for proper detection. Most of these multiplex geno-

typing assays except the Mass Array employ a form of tag arrays to sort out indi-

vidual SNP genotypes from a pool of multiplexed assay results. These tags are:

microarrayed tags on glass plates (SNPstream), tags on microbeads (BeadArray),

and tag array chips (ParAllele).

However, a major disadvantage of this candidate gene approach is that

prior knowledge of the pathogenesis of the disease is required. When knowl-

edge of the function and location of genes involved in a certain disease is lim-

ited, this approach cannot provide a complete identification of genetic variants.

An alternative approach is the whole-genome scan which scans for many SNP

markers distributed across different genes in the human genome. The whole-

genome scan approach makes no prior assumptions about the identity of the

genes to be found. Recent developments of new genotyping technologies and

international efforts like the HapMap Project enabled researchers to perform

genome-wide association studies by genotyping hundreds of thousands of SNP

markers at the same time at reasonable costs [11]. There are two major com-

mercial products on the market currently. One platform is the GeneChip system

developed by Affymetrix. There are several chips which can genotype from

10,000 up to 500,000 SNPs at a time [12]. The SNPs on the chip are randomly

selected SNPs which cover the whole genome at a density as high as 1 SNP

every 5 kb of the genome. This assay is based on hybridization of amplified

genomic DNA onto the SNP chips. Discrimination of single-base differences is

based on the hybridization strength between the oligonucleotides on the chip

and the genomic DNA. The other platform is the Infinium assay developed by

Illumina [13]. There are two whole-genome products, the 100,000-SNP chip

and the 300,000-SNP chip. The major difference between the Illumina assay

and the GeneChip system is that it does the allele-specific extension and signal

amplification on the chip instead of hybridization. Both assays deliver high-

quality genotype data at a reasonable cost per SNP.

Potential Problems in Genetic Association Studies

Although the ability to genotype multiple SNPs at a reasonable cost offers

obvious advantages, it can also present serious problems during data analysis.

To find which genes are major candidate genes among genes genotyped, multi-

ple testing is usually performed. When performing multiple testing and given

Lee 100

the large number of genes, adjustment of the significance level is very impor-

tant. This is due to the large number of false-positive findings that will be

obtained by conducting thousands of tests. The debate over adjustment for mul-

tiple testing is ongoing. Also, computational power required to handle the large

volume of genotype information is very high. Mining pieces of information

from many SNPs throughout the genome to get the whole picture of the genetic

makeup of a complex phenotype is very challenging. New genetic analysis

algorithms which can handle these kinds of data sets and figure out interactions

among networks of genes are in great need to identify genetic factors involved

in complex diseases.

Conclusions

Most of the SNPs in the human genome have been identified and their

relationship to each other on a genomic level have been elucidated. A linkage

disequilibrium map of the human genome and the tagging SNPs identified from

the HapMap Project have led researchers to try genome-wide association stud-

ies. Currently available marker panels on whole-genome genotyping systems

cover as much as 70% of genetic variation information of the human genome.

Even though it is still quite expensive to perform genome-wide association

studies, it is likely that genotyping cost will go down with time and we would

expect to see more results generated from such genome-wide association stud-

ies. Recent developments in genotyping technologies have lifted the bottleneck

on the generation of genotype data and shifted the pressure onto developing

new analysis algorithms to elucidate the network of subtle genetic factors which

are responsible for the phenotypes of interest.

References

1 Kruglyak L, Nickerson DA: Variation is the spice of life. Nat Genet 2001;27:234–236.

2 Kwok PY: Methods for genotyping single nucleotide polymorphisms. Annu Rev Genomics Hum

Genet 2001;2:235–258.

3 Livak KJ: Allelic discrimination using fluorogenic probes and the 5? nuclease assay. Genet Anal

1999;14:143–149.

4 Miller RD, Phillips MS, Jo I, et al; The SNP Consortium Allele Frequency Project: High-density

single-nucleotide polymorphism maps of the human genome. Genomics 2005;86:117–126.

5 Buetow KH, Edmonson M, MacDonald R, et al: High-throughput development and characteriza-

tion of a genome-wide collection of gene-based single nucleotide polymorphism markers by

matrix-assisted laser desorption/ionization time-of-flight mass spectrometry. Proc Natl Acad Sci

USA 2001;98:581–584.

6 Kwiatkowski RW, Lyamichev V, de Arruda M, Neri B: Clinical, genetic, and pharmacogenetic

applications of the Invader assay. Mol Diagn 1999;4:353–364.

High-Throughput Genotyping 101

7 Syvanen AC: Toward genome-wide SNP genotyping. Nat Genet 2005;37(suppl):S5–S10.

8 Bell PA, Chaturvedi S, Gelfand CA, et al: SNPstream UHT: ultra-high throughput SNP genotyp-

ing for pharmacogenomics and drug discovery. Biotechniques 2002;(suppl):70–72, 74, 76–77.

9 Fan JB, Oliphant A, Shen R, et al: Highly parallel SNP genotyping. Cold Spring Harb Symp Quant

Biol 2003;68:69–78.

10 Hardenbol P, Yu F, Belmont J, et al: Highly multiplexed molecular inversion probe genotyping:

over 10,000 targeted SNPs genotyped in a single tube assay. Genome Res 2005;15:269–275.

11 The International HapMap Consortium: The International HapMap Project. Nature 2003,426:

789–796.

12 Matsuzaki H, Dong S, Loi H, et al: Genotyping over 100,000 SNPs on a pair of oligonucleotide

arrays. Nat Methods 2004;1:109–111.

13 Gunderson KL, Sreemers FJ, Lee G, et al: A genome-wide scalable SNP genotyping assay using

microarray technology. Nat Genet 2005;37:549–554.

Dr. Jong-Eun Lee

Seodaemun-Gu Yonhee-Dong 15–1

Yonsei Milk Bldg. No. 106

Seoul 120–110 (Korea)




Nutrient-Gene Interactions in LipoproteinMetabolism – An Overview

Jose M. Ordovasa, Dolores Corellaa,c, James Kaputb

aNutrition and Genomics Laboratory, USDA Human Nutrition Research Center

on Aging at Tufts University, Boston, Mass., and bLaboratory of Nutrigenomic

Medicine, University of Illinois Chicago, Chicago, Ill., USA; cGenetic and

Molecular Epidemiology Unit, School of Medicine, University of Valencia,

Valencia, Spain

The effect of dietary changes on phenotypes (i.e., plasma lipid measures,

body weight and blood pressure) differs significantly between individuals

[1–3]. Some individuals appear to be relatively insensitive (hyporesponders) to

dietary intervention, whereas others (hyperresponders) have enhanced sensitiv-

ity [2]. This phenomenon has been more extensively researched in relation to

changes in dietary fat and plasma lipid concentrations for the prevention of

cardiovascular disease (CVD) compared to other pathological conditions.

Although common knowledge associates low-fat diets with reductions in total

and plasma low-density lipoprotein cholesterol (LDL-C), the clinical evidence

shows dramatic interindividual differences in response which may be one of

the underlying causes of the limited success of dietary recommendations in the

prevention of CVDs observed by randomized clinical trials [4].

A growing body of data supports the hypothesis that the interindividual

variability in response to dietary modification is determined by genetic factors,

especially for lipid and lipoprotein phenotypes [5]. Indirect evidence comes

from the general observation that the phenotypic response to diet is determined

partly by the baseline value of the phenotype that is itself affected by genetic

factors [2]. The main challenges are (1) how to uncover and elucidate the many

potential gene-diet interactions and (2) how potential epistatic interactions

(gene-gene) caused by differing ancestral backgrounds affect these gene-diet

interactions.

Several studies have found specific genes to be associated with the vari-

ability in response of LDL-C levels responding to changes in dietary fat but


Genes, Diet and Nutrition 103

so far, the findings have been highly inconsistent. These conflicting outcomes

probably reflect the complexity of the mechanisms involved in dietary

responses as well as the limitations of the experimental designs used to

address this problem. In addition to their effects on plasma LDL-C levels,

low-fat diets can result in reduced plasma high-density lipoprotein and/or

increased triacylglycerol (TAG) concentrations [3] that may be particularly

harmful for some persons. For example, it has been shown that individuals

with a predominance of small, dense LDL particles (subclass pattern B), a

phenotype that is associated with an increased risk of coronary heart disease,

benefit more from a low-fat diet [6] than do those with the subclass pattern A

(larger LDL particles). A significant proportion of the latter group unexpect-

edly exhibited a more atherogenic pattern B subclass after consuming a low-

fat diet. Intervention studies are increasingly focusing on the interindividual

differences in response to diet rather than on the mean effect analyzed for a

population. Moreover, new evidence indicates that the variability in response

is an intrinsic characteristic of the individual, rather than being the result of

different dietary compliance with the experimental protocols. Jacobs et al. [7]

found that individual TAG responses to a high-fat or to a low-fat diet were

vastly different, suggesting that many patients with hypertriglyceridemia are

not treated optimally if general advice for either a low-fat or a high-fat diet is

given. Studying the reasons for this variation will allow us to better identify

individuals who can benefit from a particular dietary intervention. Obviously,

this is not an easy task and some authors have already proposed different sta-

tistical algorithms in attempts to better predict the response of individuals to

different diets [8].

How Nutrients Communicate with Genes

Before presenting some of the current nutrigenetic evidences in the area of

lipid metabolism and CVD, it is helpful to gain an understanding of how nutri-

ents and other chemicals in the diet may influence gene expression and drive

gene-diet interactions. This, in fact, is the subject of nutrigenomics, which

seeks to understand gene-diet interactions in the context of the total genetic

makeup of each individual. Technological limitations in the past restricted the

investigator to a piecemeal approach: one gene, one gene product and one nutri-

ent at a time. Conceptual and technological advances are changing the playing

field. For the first time, researchers can cast a wide net in the form of micro-

arrays that can potentially capture the information about each one of the genes

expressed in a specific cell or tissue of interest. Despite these advances, the

challenges are not trivial given the chemical complexity of food, our incomplete

Ordovas/Corella/Kaput 104

knowledge about the various bioactive components present in food grown in

different climates at different times of the year, and our inability to assay gene

expression in the most appropriate target tissues in humans.

Regulation of the expression of genes involved in fatty acid metabolism

occurs when a dietary fat or metabolite binds to and activates specific fatty acid

transcription factors. These dietary chemically regulated transcription factors

are members of the nuclear receptor superfamily. This gene family consists of

48 mammalian transcription factors that regulate nearly all aspects of develop-

ment, inflammation, and metabolism. Two subclasses, the peroxisome proliferator-

activated receptors (PPARs) and liver X receptors, are lipid-sensing receptors

that have critical roles in lipid and glucose metabolism [9–11]. PPARs are

among the best-studied fatty acid-regulated nuclear receptors [12]. After uptake

into target cells, a subset of them are transported to the nucleus in association

with fatty acid-binding proteins, which facilitates their interaction with PPARs.

Several PPAR subtypes have been described. PPAR-� plays a key role in lipid

oxidation and inflammation, whereas PPAR-� is involved in cell (adipocyte)

differentiation, glucose lipid storage and inflammation. PPAR-� (also known as

PPAR-�), may play an important role in development, lipid metabolism and

inflammation. In addition to fatty acids, pharmacological agonists have been

developed for each receptor: PPAR-� binds fibrates, PPAR-� binds lipophilic

carboxylic acids, and PPAR-� binds glitazones. The fibrates are used to treat

hyperlipidemia. The glitazones are used to manage plasma glucose levels in

patients with insulin resistance [13].

Many of the previously published nutrigenetic [i.e., single gene/single

nucleotide polymorphism (SNP)] studies focused on genes that are regulated

by PPARs and other nuclear receptors [14]. Polymorphisms in promoter

regions of these genes may disrupt or at least alter the communication with

these transcription factors which would have significant consequences in a

person’s response to dietary factors that are ligands (i.e., polyunsaturated

fatty acids) of the transcription factors. It is also obvious that polymorphisms

within the transcription factors themselves will have a significant impact on

the way that each one of us responds to dietary factors. The evidence for

gene-diet interactions between common SNPs in candidate genes and

dietary factors related to lipid metabolism is increasing. However, caution is

needed before applying these results to clinical practice for three primary

reasons: (1) the meaning of ‘statistically significant results’ is subject to dif-

fering interpretations and often depends upon the study design, (2) many

initial gene-nutrient-phenotype associations are not replicated in subsequent

studies, and (3) gene variations may influence phenotypes differently in

individuals from different ancestral backgrounds due to gene-gene (epistatic)

interactions.


Results from Interventional StudiesInterventional studies in which subjects receive a controlled dietary intake

provide the best approach for conducting gene-nutrient-phenotype association

studies. However, these well-controlled feeding studies have several important

logistical limitations, most importantly the small number of participants and the

brief duration of the interventions. Scores of interventional studies examining

gene-diet interactions on different parameters of lipid metabolism have been

published. However, the level of replication among studies analyzing the same

genetic variation tends to be low. The lack of replication is most likely due to

the different characteristics (ethnicity, physical condition, age, lifestyle differ-

ences) of study subjects, length of intervention, sample size, and heterogeneity

in the design. In a systematic review (from 1966 to 2002), Masson et al. [15]

identified 74 relevant articles including dietary intervention studies that had

measured the lipid and lipoprotein response to diet in different genotype groups

and 17 reviews on gene-diet interactions. After a comparative analysis of the

individual findings, they concluded that there is evidence to suggest that (1)

variations in the APOA1, APOA4, APOB, and APOE genes contribute to the

heterogeneity in the lipid response to dietary intervention and (2) all of these

genes are regulated directly or indirectly by PPAR-� or other nuclear receptors.

However, the evidence suggested by Masson et al. [15] in relation to the above

genes comes from meta-analyses of the published data and describes the aver-

age effect. It should be noted that there is not total consistency of results among

individual studies.

More recently, one of our groups [16–18] reviewed this topic extensively

and included additional studies reported after 2002. The median for the sample

sizes in these more recent studies was in the range of 60 subjects. These small

sample sizes highlight one of the traditional problems for the lack of repro-

ducibility, specifically, the statistical power is low. In addition, the composition of

the dietary intervention in these studies varied considerably. We propose that the

design of future intervention studies should be standardized for key dietary intake

variables and phenotype measurements. A minimum set of variables would

include patients’ physical and genetic characteristics, medications, composition

and length of the dietary treatment, and sample size. Such standardization would

allow better comparison among studies and the possibility of conducting meta-

analyses, which is not possible under current experimental conditions.

Results from Observational StudiesObservational studies have the advantage of large numbers of subjects and

the ability to estimate long-term dietary habits. However, the level of evidence of

the results obtained from these studies has traditionally been considered to be


lower than that of experimental studies. Nevertheless, the level of confidence in

such studies can be increased by taking into consideration the principle of

Mendelian randomization [19]. This concept reflects the random assortment of

alleles at the time of gamete formation. Such randomization results in population

distributions of genetic variants that are generally independent of behavioral and

environmental factors that confound epidemiological associations between

potential risk factors and disease. This topic has been extensively reviewed

[16–18]. The median population size for recent observational studies is approxi-

mately 850. This sample size may be informative for traditional genotype-

phenotype association studies but, considering the higher measurement error

of dietary intake in comparison with experimental studies, it may not have

enough statistical power to address properly the complexity of gene-environment

interactions. As pointed out for intervention studies, replication of results is still

very low. In addition, these findings need the synergy of those studies examining

the effects of nutrients on gene expression (nutrigenomics) to provide the mech-

anistic knowledge that will support the reported statistical associations.

Genotype-nutrient-phenotype analyses may be improved by determining

ancestral backgrounds of each study participant. These additional data are nec-

essary since SNPs may be expressed differently among individuals of differing

ethnicities because of varying gene-gene and gene-nutrient interactions.

Determining the genetic architecture (that is, geographical origin of chromoso-

mal regions) in each study participant may reduce statistical noise caused by

mismatching cases and controls [20].

Gene-Diet Interactions in the Postprandial StateHuman beings living in industrialized societies spend most of the waking

hours in a nonfasting state because of meal consumption patterns and the amounts

of food ingested. Postprandial lipemia, characterized by a rise in TAG after eating,

is a dynamic, nonsteady-state condition [21]. Over 25 years ago, Zilversmit [22]

proposed that atherogenesis was a postprandial phenomenon since high concentra-

tions of lipoproteins and their remnants following food ingestion could deposit

into the arterial wall and accumulate in atheromatous plaques. Several studies have

investigated the potential interaction between some polymorphisms in candidate

genes and diet on postprandial lipids (for a review, see Ordovas and Corella [17]).

In postprandial studies, subjects usually receive a fat-loading test meal that has dif-

fering compositions depending on the nutrient(s) to be tested. After the test meal,

blood samples are taken to measure postprandial lipids which are then compared

with preprandial levels [21]. Consistency among studies is still very low and repli-

cation of findings is a major necessity. Postprandial studies often have low num-

bers of subjects (usually �50) with added complexity inducing even more bias

than for other experimental approaches.


The Road Map to Solidifying the Nutrigenomics FieldDespite the excitement brought up by an increasing number of findings

related to nutritional genomics, the progress of the field is hampered by the inad-

equacy of the current experimental approaches to efficiently deal with the bio-

logical complexity of the phenotype(s), the complexity of dietary intakes,

differing genetic backgrounds among participants, and the limitations of low sta-

tistical power of the studies. We and others have proposed that only a compre-

hensive, international nutritional genomics approach [23, 24] will yield short-

and long-term benefits to human health by: (1) revealing novel nutrient-gene

interactions, (2) developing new diagnostic tests for adverse responses to diets,

(3) identifying specific populations with special nutrient needs, (4) improving

the consistency of current definitions and methodology related to dietary assess-

ment, and (5) providing the information for developing more nutritious plant and

animal foods and food formulations that promote health and prevent, mitigate, or

cure disease. Achieving these goals will require an extensive dialogue between

scientists and the public about the nutritional needs of the individual versus

groups, local food availability and customs, analysis and understanding of

genetic differences between individuals and populations, and serious commit-

ment of funds from the public and private sectors. Nutritional genomics

researchers are seeking collaborations of scientists, scholars, and policy makers

to maximize the collective impact on global poverty and health by advancing our

knowledge of how genetics and nutrition can promote health or cause disease.

Conclusions

Although the current evidence from both experimental and observational

nutrigenetics studies is not enough to start making specific personalized nutri-

tional recommendations based on genetic information, there are a large number

of examples of common SNPs modulating the individual response to diet as

proof of concept of how gene-diet interactions can influence lipid metabolism.

It is critical that these preliminary studies go through further replication and

that subsequent studies be properly designed with sufficient statistical power

and careful attention to phenotype and genotype. The many challenges that lay

ahead are evident. This review has examined the vast world of nutrigenetics and

nutrigenomics only through the small keyhole of PPAR-� and dietary fat.

Analogous to the use of the X-ray diffraction patterns 50 years ago to determine

the structure of DNA, which led to today’s progress in sequencing the entire

human genome, these initial steps in understanding nutrigenomics will likely

lead to fundamental breakthroughs that will both clarify today’s mysteries and

pave the way for clinical applications. Hopefully, bringing nutrigenetics to the


state of becoming a practical and useful tool will not take 50 years. However, to

arrive at the point where it is possible to assess the modulation by specific SNPs

of the effects of dietary interventions on lipid metabolism, well-designed, ade-

quately powered, and adequately interpreted randomized controlled studies (or

their equivalent) of greater duration than current studies are needed, with care-

ful consideration given to which patients to include in such studies. Moreover,

research must also investigate the potential mechanisms involved in the gene-

diet interactions reported by nutrigenetic studies [23]. These imperative needs

can be achieved only through the collaboration of experts in the different fields

involved, which must include nutrition professionals [24].

One of the first situations where personalized nutrition is likely to be ben-

eficial is with dyslipidemic patients that require special intervention with

dietary treatment. It is known that these individuals will display dramatic het-

erogeneity in response to the currently recommended therapeutic diets and that

the recommendations will need to be adjusted individually. This process could

be more efficient and efficacious if the recommendations were carried out

based on genetic and molecular knowledge. Moreover, adherence to dietary

advice may increase when it is supported with information based on nutritional

genomics, and the patient feels that the advice is personalized. However, a num-

ber of important changes in the provision of health care are needed in order to

achieve the potential benefits associated with this concept, including a team-

work approach, with greater integration among physicians and nutrition profes-

sionals. Once more experience is gained from patients and/or individuals at

high risk, these approaches could be applied towards primary prevention.

Acknowledgments

This study was supported by NIH/NHLBI grant No. HL54776 (J.M.O.), NIH/NHLBI

contract No. 1-38038 (J.M.O.), and contracts 53-K06-5-10 and 58-1950-9-001 from the US

Department of Agriculture Research Service (J.M.O.) and National Center for Minority

Health and Health Disparities Center of Excellence in Nutritional Genomics (MD00222).

References

1 Jacobs DR Jr, Anderson JT, Hannan P, Keys A, Blackburn H: Variability in individual serum cho-

lesterol response to change in diet. Arteriosclerosis 1983;3:349–356.

2 Katan MB, Beynen AC, De Vries JH, Nobels A: Existence of consistent hypo- and hyperrespon-

ders to dietary cholesterol in man. Am J Epidemiol 1986;123:221–234.

3 Katan MB, Grundy SM, Willett WC: Should a low-fat, high-carbohydrate diet be recommended

for everyone? Beyond low-fat diets. N Engl J Med 1997;337:563–566.

4 Prentice RL, Caan B, Chlebowski RT, Patterson R, Kuller LH, Ockene JK, Margolis KL,

Limacher MC, Manson JE, Parker LM, Paskett E, Phillips L, Robbins J, Rossouw JE, Sarto GE,


Shikany JM, Stefanick ML, Thomson CA, Van Horn L, Vitolins MZ, Wactawski-Wende J, Wallace RB,

Wassertheil-Smoller S, Whitlock E, Yano K, Adams-Campbell L, Anderson GL, Assaf AR,

Beresford SA, Black HR, Brunner RL, Brzyski RG, Ford L, Gass M, Hays J, Heber D, Heiss G,

Hendrix SL, Hsia J, Hubbell FA, Jackson RD, Johnson KC, Kotchen JM, LaCroix AZ, Lane DS,

Langer RD, Lasser NL, Henderson MM: Low-fat dietary pattern and risk of invasive breast

cancer: the Women’s Health Initiative Randomized Controlled Dietary Modification Trial. JAMA

2006;295:629–642.

5 Loktionov A: Common gene polymorphisms and nutrition: emerging links with pathogenesis of

multifactorial chronic diseases. J Nutr Biochem 2003;14:426–451.

6 Krauss RM: Dietary and genetic effects on low-density lipoprotein heterogeneity. Annu Rev Nutr

2001;21:283–295.

7 Jacobs B, De Angelis-Schierbaum G, Egert S, Assmann G, Kratz M: Individual serum triglyceride

responses to high-fat and low-fat diets differ in men with modest and severe hypertriglyceridemia.

J Nutr 2004;134:1400–1405.

8 Parks EJ, Rutledge JC, Davis PA, Hyson DA, Schneeman BO, Kappagoda CT: Predictors of

plasma triglyceride elevation in patients participating in a coronary atherosclerosis treatment

program. J Cardiopulm Rehabil 2001;21:73–79.

9 Li AC, Glass CK: PPAR- and LXR-dependent pathways controlling lipid metabolism and the

development of atherosclerosis. J Lipid Res 2004;45:2161–2173.

10 Pegorier JP, Le May C, Girard J: Control of gene expression by fatty acids. J Nutr 2004;134:

2444S–2449S.

11 Jump DB: Fatty acid regulation of gene transcription. Crit Rev Clin Lab Sci 2004;41:41–78.

12 Clarke SD: The multi-dimensional regulation of gene expression by fatty acids: polyunsaturated

fats as nutrient sensors. Curr Opin Lipidol 2004;15:13–18.

13 Berger JP, Akiyama TE, Meinke PT: PPARs: therapeutic targets for metabolic disease. Trends

Pharmacol Sci 2005;26:244–251.

14 Mandard S, Müller M, Kersten S: Peroxisome proliferator-activated receptor alpha target genes.

Cell Mol Life Sci 2004;61:393–416.

15 Masson LF, McNeill G, Avenell A: Genetic variation and the lipid response to dietary intervention:

a systematic review. Am J Clin Nutr 2003;77:1098–1111.

16 Corella D, Ordovas JM: Single nucleotide polymorphisms that influence lipid metabolism: inter-

action with dietary factors. Annu Rev Nutr 2005;11;25:341–390.


18 Ordovas JM, Corella D: Genes, diet and plasma lipids: the evidence from observational studies.

World Rev Nutr Diet 2004;93:41–76.

19 Tobin MD, Minelli C, Burton PR, Thompson JR: Development of Mendelian randomization: from

hypothesis test to ‘Mendelian deconfounding’. Int J Epidemiol 2004;33:26–29.

20 Campbell CD, Ogburn EL, Lunetta KL, Lyon HN, Freedman ML, Groop LC, Altshuler D, Ardlie KG,

Hirschhorn JN: Demonstrating stratification in a European American population. Nat Genet 2005;37:

868–872.

21 Ordovas JM: Genetics, postprandial lipemia and obesity. Nutr Metab Cardiovasc Dis 2001;11:

118–133.

22 Zilversmit DB: Atherogenesis: a postprandial phenomenon. Circulation 1979;60:473–485.

23 van Ommen B, Stierum R: Nutrigenomics: exploiting systems biology in the nutrition and health

arena. Curr Opin Biotechnol 2002;13:517–521.

24 Kaput J, Ordovas JM, Ferguson L, Ommen BV, Rodriguez RL: The case for strategic international

alliances to harness nutritional genomics for public and personal health. Br J Nutr 2005;94:

623–632.

Dr. Jose M. Ordovas

Nutrition and Genomics Laboratory

USDA Human Nutrition Research Center on Aging at Tufts University

711 Washington St., Boston, MA 02111 (USA)




The Genetics of Lipoprotein Metabolismand Heart Disease

E. Shyong Tai

Singapore General Hospital, Singapore

AbstractBlood lipids are major determinants of risk for cardiovascular disease. Lipid-lowering

therapies have been demonstrated to reduce the risk of cardiovascular disease in humans.

Genetic variants in many candidate genes are associated with blood lipids. In some

instances, such as the association between APOE variants and low-density lipoprotein cho-

lesterol, the associations are similar from population to population. However, for others,

the associations may differ between populations. In some instances, these differences

related to interactions between the genetic variants and environmental factors. The exami-

nation of such associations/interactions tells us something about the biology of human

lipoprotein metabolism. However, the utility of genetic variants for predicting cardiovascu-

lar disease is currently limited. To date, none of these genetic variants have been shown to

improve the ability of predictive functions to discriminate between those at high and low

risk of heart disease. To do this, the genetic variants should connote some aspect of risk that

is not included in existing predictive functions. Alternatively, they should modify the risk

associated with the risk factors in existing functions. Research to determine the impact of

these genetic markers as predictors of disease is an important area that is currently under-

explored.


Blood lipid levels are among the most important, modifiable risk fac-

tors for coronary artery disease (CAD). Of those commonly measured, high

levels of low-density lipoprotein cholesterol (LDL-C) are associated with

increased risk of CAD. In contrast, high concentrations of high-density

lipoprotein cholesterol (HDL-C) are protective. Pharmacological treatment to

lower LDL-C [1] and raise HDL-C [2–4] have been shown to reduce the risk of

CAD.

The Genetics of Lipoprotein Metabolism and Heart Disease 111

Genetic Variation and Ethnic Differences in Blood Lipid Levels

In Singapore, we have observed significant ethnic differences in both the

levels of lipids in the blood and the risk of CAD [5–7]. Indians living in

Singapore have approximately a threefold higher risk of CAD than Chinese

people [8, 9]. They also exhibit lower serum levels of HDL-C. Malays, on the

other hand, have elevated serum levels of LDL-C and their risk of CAD is inter-

mediate between that of the Chinese and Indians. It is tempting, when compar-

ing ethnic groups, to think that genetic differences could explain differences

between ethnic groups. The most common genetic variants in the human

genome comprise single nucleotide polymorphisms (SNPs) [10]. These repre-

sent a single-base change in the DNA sequence. Such changes can result in

changes in the level or function of the protein encoded by the genetic locus.

Alternatively, associations between the presence of particular SNPs and a dis-

ease could result from linkage disequilibrium between the SNP examined and a

causative mutation in another part of the locus.

SNPs at the APOE locus have been studied extensively in relation to blood

lipids and CAD. Two common polymorphisms occur at sites in codons 112 and

158 of the APOE gene locus. The �2 allele contains a cysteine at position 158,

and the �4 contains an arginine at position 112. When we examined the associ-

ation between polymorphisms at the APOE locus in the three ethnic groups liv-

ing in Singapore, the frequency of the APOE4 allele was highest in Malays

compared to Chinese and Indians [11]. The presence of the APOE4 allele was

also associated with a higher LDL-C concentration, irrespective of the ethnic

group. As such, the APOE4 allele contributes to the higher serum concentra-

tions of LDL-C in Malays in Singapore.

However, this explanation for the ethnic differences seems overly simplis-

tic for several reasons. Firstly, the effect of the APOE4 allele is relatively small

and the differences in LDL-C concentration between the ethnic groups remained

even after accounting for the differences in the APOE4 allele frequency.

Secondly, when it comes to other genetic loci, the frequency of the ‘at risk’

allele is not always higher in those ethnic groups with greater risk. In an attempt

to understand the reasons for low HDL-C concentrations in Asian Indians, we

also genotyped Chinese, Malays and Indians for the presence of the TaqIB

polymorphism at the CETP locus [12] (encoding cholesteryl ester transfer pro-

tein). The B2 allele has been associated with higher concentrations of HDL-C

in other populations. In our population, the same association with HDL-C was

observed in the various ethnic groups [12]. However, contrary to our expecta-

tions, the frequency of the B2 allele was highest in Indians, the ethnic group

with the lowest HDL-C concentration.

Tai 112

Genes, Diet and Ethnic Differences in Blood Lipids

It is important to remember that ethnicity is a construct that encompasses

both genetic and cultural differences. This means that, over and above any

genetic differences, the different ethnic groups are subject to different environ-

mental exposures. Diet is one of the environmental exposures that differs

between ethnic groups. However, dietary factors do not appear to act in isola-

tion either. Differences in the dietary intake of various macronutrients, includ-

ing dietary fat, failed to explain the ethnic differences in blood lipids in

Singapore [13]. Furthermore, family and twin studies suggest that plasma lipid

concentrations show significant hereditability assuming that, although genetic

factors are unlikely to explain the ethnic differences observed, they may still

play a role [14]. In fact, it appears that the effect of diet on blood lipid levels

appears to be partially determined by an individual’s genetic makeup. The evi-

dence for this is as follows. Although relationships between dietary changes and

serum lipid changes are well founded and predictable for groups, a striking

variability in the response of serum cholesterol to diet between subjects has

been reported [15–17]. In some individuals, plasma cholesterol levels dramati-

cally decrease following consumption of a low-fat diet, while they remain

unchanged in others [16–19]. It has been shown in elegant studies in nonhuman

primates that the serum lipoprotein response to dietary manipulation has a sig-

nificant genetic component [20–22]. In this regard, several variants at various

genetic loci seem to modulate the association between dietary fat and serum

lipid concentrations [23].

In line with this hypothesis, we have shown that the association with the

presence of the B2 allele at the CETP locus (described above) is not a straight-

forward one. We have shown that this polymorphism interacts with dietary cho-

lesterol intake and that these jointly modulate HDL-C concentration [12]. Other

studies have shown that this polymorphism interacts with other environmental

factors including obesity/insulin resistance [24], smoking [25] and alcohol

intake [26] enhancing the effect of these environmental factors on HDL-C con-

centration. It has been suggested that, over and above its direct effect on HDL-

C, the B2 allele connotes an increased susceptibility to the effect of environmental

exposure on HDL-C concentration.

As another example, we have examined the �514C → T polymorphism at

the hepatic lipase (LIPC) locus. The presence of the T allele is associated with

elevated serum HDL-C concentration in many studies [27–30]. This was also

the case in Singapore [31]. More recently, investigators from the Framingham

Offspring Study have shown that this polymorphism modulated the association

between dietary fat intake and serum HDL-C concentration [32]. In those with

the TT genotype, high dietary fat intake was associated with low HDL-C


concentration whereas in those with the CC genotype, the opposite was true.

The investigators suggested that the TT genotype may identify a group of indi-

viduals who are maladapted to a high-fat diet in relation to cardiovascular dis-

ease risk. The findings in Singapore were similar. We found that a high-fat diet

had an adverse effect on the serum lipid profile in the form of hypertriglyc-

eridemia in all three ethnic groups and that this occurred primarily in those with

the TT genotype [31]. In addition, our findings in Asian Indians replicated

those in the Framingham Offspring Study in relation to dietary fat intake and

low HDL-C concentration.

The Relevance of These Findings to the Risk of Coronary Artery Disease

Gene nutrient interactions of this nature also alter the risk of CAD. For

example, ADH3 encodes a protein which metabolizes ethanol. A polymor-

phism at this locus reduces the catabolism of alcohol resulting in higher HDL-C

levels in persons who drink alcohol regularly and carry the polymorphism

[33]. These individuals were also shown to have a reduced risk of CAD [33,

34]. Unfortunately, few studies have demonstrated that the gene-diet interac-

tions in relation to blood lipids apply to the risk of CAD as might be expected

by the association with blood lipids. The findings related to blood lipids may

not translate directly into risk of CAD. For example, the T allele at position

�514 at the LIPC locus is associated with increased blood concentrations of

HDL-C, and would be expected to be associated with a reduced risk of

CAD. However, in a study that examined the association of the �514C → T

polymorphism at the LIPC locus with the degree of coronary artery stenosis,

the T allele was associated with an increase in coronary stenosis whereas those

carrying the C allele experienced less stenosis [35]. This suggests that the

T allele connotes an increased risk of CAD. Therefore, based on current

knowledge, the relevance of many of the described gene-diet interactions, as

they pertain to CAD, is unclear. Many more studies incorporating the assess-

ment of dietary intake, genetic analysis and an assessment of CAD events are

required.

It has also been suggested that genetic markers may be useful predictors of

chronic disease. There are several reasons why this is not likely, at least in the

near future. Firstly, as described in this study, the association between a genetic

marker and disease is not a straightforward one. Instead the data suggest signif-

icant modifications of the effect of genetic variants by environmental expo-

sures. Therefore, to fully utilize a genetic marker to predict disease, we would

also have to take into account the environmental exposure. As discussed above,

Tai 114

there are insufficient data to allow this at this time. Secondly, as predictors of

chronic disease, genetic markers should not be considered in isolation. Rather,

they need to be considered in the context of the other known risk factors for the

disease [36]. For CAD, several risk factors such as blood glucose, lipids, blood

pressure, age and gender are well established. In fact, these risk factors, when

combined into a predictive function, have been shown to operate well in several

different populations. To be clinically useful, these variables should improve the

ability of existing predictive functions, such as the Framingham predictive

function, to discriminate between individuals with differing risks of CAD. This

was carried out for the APOE variants. Although it is fairly well established that

the APOE4 allele is associated with increased risk of CAD, when information

on the APOE genotype was added to an existing predictive function for CAD, it

provided little, if any, additional discriminatory value.

What then is the relevance of the study of gene-diet interactions? I believe

that this lies in two areas. Firstly, knowledge of gene-diet interactions may facil-

itate the development of dietary recommendations which are individualized to

optimize the benefit obtained. While this is unlikely to be useful for large seg-

ments of the populations, in those who do not respond to conventional recom-

mendations for reducing CAD risk, it may be a useful adjunct to current

treatment and may reduce the need to resort to drug treatment. Secondly, the

elucidation of gene-diet interactions can help us understand the biological path-

ways through which diet alters the risk of chronic disease. For example, in rela-

tion to HDL-C concentration, several dietary factors have been shown to alter

HDL metabolism. However, the pathways through which they act are not clear.

Whilst it is possible to carry out experiments in cells and animals to try to elu-

cidate these pathways, these findings may not always be relevant to human

physiology. The examination of gene-environment interactions offers us an

opportunity to understand the pathways involved in humans. Genetic factors

which interact with dietary factors to determine a particular phenotypic trait are

likely to lie along the metabolic pathway by which the dietary factor acts on the

phenotypic trait. As an example, it is known that the substitution of saturated fat

in the diet with unsaturated fat results in a decrease in serum HDL-C concen-

tration [37]. However, the pathways involved are unclear. We have found that

polymorphisms at the peroxisome proliferator-activated receptor-� (PPAR-�)

locus interact with dietary polyunsaturated fatty acid intake to determine

HDL-C concentrations in Chinese women [38]. This suggests that polyunsatu-

rated fatty acid may act on HDL-C through PPAR-� and genes that are regu-

lated by this PPAR-�. There is some hope that the elucidation of these pathways

in humans may identify molecular targets through which we could mimic the

beneficial effects of nutrients in those who require additional pharmacological

treatment.


Conclusion

Dyslipidemia represents an example of a chronic disorder which appears

to relate to a combination of genetic and environmental factors. These genetic

and environmental factors interact to produce the ultimate phenotype. However,

the relevance to CAD risk is not clear at this time. More studies incorporating

sufficient numbers of clinical events are required to establish this link. To be

useful for the prediction of CAD risk, we also need to evaluate the gain in dis-

crimination resulting from the addition of these new genetic markers to existing

predictive functions.

Nevertheless, the elucidation of gene-diet interactions may facilitate (1)

the development of individualized dietary recommendations for individuals

who fail to response to conventional dietary recommendations, and (2) help us

identify the biological pathways involved in the beneficial effects of certain

nutrients on human physiology. However, these studies need to be large, and

include significant numbers of individuals who experience clinical events. For

these reasons, there is an urgent need to develop collaborative studies with stan-

dardized methodology to ensure adequate sample sizes and statistical power.

References

1 Briel M, Nordmann AJ, Bucher HC: Statin therapy for prevention and treatment of acute and

chronic cardiovascular disease: update on recent trials and metaanalyses. Curr Opin Lipidol

2005;16:601–605.

2 Brown BG, Zhao XQ, Chait A, et al: Simvastatin and niacin, antioxidant vitamins, or the combi-

nation for the prevention of coronary disease. N Engl J Med 2001;345:1583–1592.

3 Canner PL, Berge KG, Wenger NK, et al: Fifteen year mortality in Coronary Drug Project

patients: long-term benefit with niacin. J Am Coll Cardiol 1986;8:1245–1255.

4 Rubins HB, Robins SJ, Collins D, et al: Gemfibrozil for the secondary prevention of coronary

heart disease in men with low levels of high-density lipoprotein cholesterol. Veterans Affairs

High-Density Lipoprotein Cholesterol Intervention Trial Study Group. N Engl J Med 1999;341:

410–418.

5 Hughes K, Yeo PP, Lun KC, et al: Cardiovascular diseases in Chinese, Malays, and Indians in

Singapore. 2. Differences in risk factor levels. J Epidemiol Community Health 1990;44:29–35.

6 Tai ES, Emmanuel SC, Chew SK, Tan BY, Tan CE: Isolated low HDL cholesterol: an insulin-resis-

tant state only in the presence of fasting hypertriglyceridemia. Diabetes 1999;48:1088–1092.

7 Tan CE, Emmanuel SC, Tan BY, Jacob E: Prevalence of diabetes and ethnic differences in cardio-

vascular risk factors. The 1992 Singapore National Health Survey. Diabetes Care 1999;22:

241–247.

8 Lee J, Heng D, Chia KS, Chew SK, Tan BY, Hughes K: Risk factors and incident coronary heart

disease in Chinese, Malay and Asian Indian males: the Singapore Cardiovascular Cohort Study.

Int J Epidemiol 2001;30:983–988.

9 Mak KH, Chia KS, Kark JD, et al: Ethnic differences in acute myocardial infarction in Singapore.

Eur Heart J 2003;24:151–160.

10 Syvanen AC: Accessing genetic variation: genotyping single nucleotide polymorphisms. Nat Rev

Genet 2001;2:930–942.

Tai 116

11 Tan CE, Tai ES, Tan CS, et al: APOE polymorphism and lipid profile in three ethnic groups in the

Singapore population. Atherosclerosis 2003;170:253–260.

12 Tai ES, Ordovas JM, Corella D, et al: The TaqIB and �629C → A polymorphisms at the choles-

teryl ester transfer protein locus: associations with lipid levels in a multiethnic population. The

1998 Singapore National Health Survey. Clin Genet 2003;63:19–30.

13 Deurenberg-Yap M, Li T, Tan WL, van Staveren WA, Chew SK, Deurenberg P: Can dietary factors

explain differences in serum cholesterol profiles among different ethnic groups (Chinese, Malays

and Indians) in Singapore? Asia Pac J Clin Nutr 2001;10:39–45.

14 Heller DA, de Faire U, Pedersen NL, Dahlen G, McClearn GE: Genetic and environmental influ-

ences on serum lipid levels in twins. N Engl J Med 1993;328:1150–1156.

15 Cobb MM, Teitlebaum H: Determinants of plasma cholesterol responsiveness to diet. Br J Nutr

1994;71:271–282.

16 Jacobs DR, Anderson JT, Hannan P, Keys A, Blackburn H: Variability in individual serum choles-

terol response to change in diet. Arteriosclerosis 1983;3:349–356.

17 Katan MB, Beynen AC, de Vries JH, Nobels A: Existence of consistent hypo- and hyperrespon-

ders to dietary cholesterol in man. Am J Epidemiol 1986;123:221–234.

18 Cobb MM, Risch N: Low-density lipoprotein cholesterol responsiveness to diet in normolipi-

demic subjects. Metabolism 1993;42:7–13.

19 MA OH, Rosner B, Bishop LM, Sacks FM: Effects of inherent responsiveness to diet and day-to-

day diet variation on plasma lipoprotein concentrations. Am J Clin Nutr 1996;64:53–59.

20 Mahaney MC, Blangero J, Rainwater DL, et al: Pleiotropy and genotype by diet interaction in a

baboon model for atherosclerosis: a multivariate quantitative genetic analysis of HDL subfractions

in two dietary environments. Arterioscler Thromb Vasc Biol 1999;19:1134–1141.

21 Rainwater DL, Kammerer CM, Hixson JE, et al: Two major loci control variation in beta-lipopro-

tein cholesterol and response to dietary fat and cholesterol in baboons. Arterioscler Thromb Vasc

Biol 1998;18:1061–1068.

22 Rainwater DL, Kammerer CM, VandeBerg JL: Evidence that multiple genes influence baseline

concentrations and diet response of Lp(a) in baboons. Arterioscler Thromb Vasc Biol

1999;19:2696–2700.

23 Ordovas JM: Gene-diet interaction and plasma lipid responses to dietary intervention. Biochem

Soc Trans 2002;30:68–73.

24 Vohl MC, Lamarche B, Pascot A, et al: Contribution of the cholesteryl ester transfer protein gene

TaqIB polymorphism to the reduced plasma HDL-cholesterol levels found in abdominal obese

men with the features of the insulin resistance syndrome. Int J Obes Relat Metab Disord

1999;23:918–925.

25 Freeman DJ, Griffin BA, Holmes AP, et al: Regulation of plasma HDL cholesterol and subfraction

distribution by genetic and environmental factors. Associations between the TaqI B RFLP in the

CETP gene and smoking and obesity. Arterioscler Thromb 1994;14:336–344.

26 Fumeron F, Betoulle D, Luc G, et al: Alcohol intake modulates the effect of a polymorphism of the

cholesteryl ester transfer protein gene on plasma high density lipoprotein and the risk of myocar-

dial infarction. J Clin Invest 1995;96:1664–1671.

27 Deeb SS, Peng R: The C-514T polymorphism in the human hepatic lipase gene promoter dimin-

ishes its activity. J Lipid Res 2000;41:155–158.

28 Guerra R, Wang J, Grundy SM, Cohen JC: A hepatic lipase (LIPC) allele associated with high

plasma concentrations of high density lipoprotein cholesterol. Proc Natl Acad Sci USA 1997;94:

4532–4537.

29 Jansen H, Verhoeven AJ, Weeks L, et al: Common C-to-T substitution at position –480 of the

hepatic lipase promoter associated with a lowered lipase activity in coronary artery disease

patients. Arterioscler Thromb Vasc Biol 1997;17:2837–2842.

30 Murtomaki S, Tahvanainen E, Antikainen M, et al: Hepatic lipase gene polymorphisms influence

plasma HDL levels. Results from Finnish EARS participants. European Atherosclerosis Research

Study. Arterioscler Thromb Vasc Biol 1997;17:1879–1884.

31 Tai ES, Corella D, Deurenberg-Yap M, et al: Dietary fat interacts with the �514C → T polymor-

phism in the hepatic lipase gene promoter on plasma lipid profiles in a multiethnic Asian popula-

tion: the 1998 Singapore National Health Survey. J Nutr 2003;133:3399–3408.


32 Ordovas JM, Corella D, Demissie S, et al: Dietary fat intake determines the effect of a common

polymorphism in the hepatic lipase gene promoter on high-density lipoprotein metabolism: evidence

of a strong dose effect in this gene-nutrient interaction in the Framingham Study. Circulation

2002;106:2315–2321.

33 Hines LM, Stampfer MJ, Ma J, et al: Genetic variation in alcohol dehydrogenase and the benefi-

cial effect of moderate alcohol consumption on myocardial infarction. N Engl J Med 2001;344:

549–555.

34 Younis J, Cooper JA, Miller GJ, Humphries SE, Talmud PJ: Genetic variation in alcohol dehydro-

genase 1C and the beneficial effect of alcohol intake on coronary heart disease risk in the Second

Northwick Park Heart Study. Atherosclerosis 2005;180:225–232.

35 Zambon A, Deeb SS, Brown BG, Hokanson JE, Brunzell JD: Common hepatic lipase gene pro-

moter variant determines clinical response to intensive lipid-lowering treatment. Circulation 2001;103:

792–798.

36 Humphries SE, Ridker PM, Talmud PJ: Genetic testing for cardiovascular disease susceptibility: a

useful clinical management tool or possible misinformation? Arterioscler Thromb Vasc Biol

2004;24:628–636.

37 Lichtenstein AH, Ausman LM, Jalbert SM, Schaefer EJ: Effects of different forms of dietary

hydrogenated fats on serum lipoprotein cholesterol levels. N Engl J Med 1999;340:1933–1940.

38 Chan E, Tan CS, Deurenberg-Yap M, Chia KS, Chew SK, Tai ES: The V227A polymorphism at

the PPARA locus is associated with serum lipid concentrations and modulates the association

between dietary polyunsaturated fatty acid intake and serum high density lipoprotein concentra-

tions in Chinese women. Atherosclerosis 2006;187:309–315.

Dr. E. Shyong Tai

Singapore General Hospital

1, Hospital Drive

Singapore 169608 (Singapore)

Tel. �65 321 3654, Fax �65 227 3576, E-Mail [email protected]



Gene-Environment Interactions and theDiabetes Epidemic in India

V. Mohan, V. Sudha, G. Radhika, V. Radha, M. Rema, R. Deepa

Madras Diabetes Research Foundation and Dr. Mohan’s Diabetes Specialities Centre,

Gopalapuram, Chennai, India

AbstractThe prevalence of diabetes is rising rapidly in all developing countries and India already

has the largest number of people with diabetes. Evidence for the rising prevalence of diabetes

in India comes from recent population-based studies such as the Chennai Urban Population

Study (n � 1,262) and the Chennai Urban Rural Epidemiology Study (n � 26,001). These two

studies revealed that the current age-standardized prevalence of diabetes in Chennai in adults

�20 years of age is 14.3%, which is 70% higher than that seen in the year 1989 (8.3%). In the

Chennai Urban Population Study, we observed that the higher-income group who consumed

excess fat and calorie-rich food had an increased prevalence of diabetes compared to the lower-

income group. There was also a linear increase in the prevalence of diabetes with an increase in

visible fat consumption. In addition, we observed that visible fat consumption and physical

inactivity showed a cumulative effect on increasing the prevalence of diabetes. We carried

out gene-diet interaction studies, which revealed that the adiponectin gene polymorphism

(�10211T → G) contributed to insulin resistance and diabetes and this was exaggerated in

those consuming diets with higher glycemic loads. These subjects also had an increased risk for

hypoadiponectinemia. Similarly, the Ala54Thr polymorphism of the fatty acid-binding protein

2 gene showed a synergistic effect with a high glycemic load increasing the risk for hyper-

triglyceridemia. These studies indicate that gene-diet interactions could play a major role in

increasing the risk for diabetes. However, given the imprecision in measuring dietary intake,

very large sample sizes would be needed for meaningful conclusions to be drawn.


The problem of diabetes is growing in epidemic proportions and is taking a

toll of millions of lives worldwide. Recent statistics from the World Health

Organization show that in the year 2000, 171 million people had diabetes globally

and these numbers are expected to increase to 366 million by the year 2030 [1].

The top three countries in terms of numbers of people with diabetes are India,

Gene-Environment Interaction in Diabetes 119

China, and the USA and it is predicted that these countries would continue to top

the list even by the year 2030 [1]. India currently has about 40 million people with

diabetes and by the year 2030, this would increase to nearly 80 million. In this

paper, we first show evidence for the epidemic in India and then demonstrate evi-

dence for both genetic variations and environment in its causation.

Many of the studies conducted in India documenting the prevalence of dia-

betes mellitus have been conducted in different geographical locations, which

limits our ability to look at trends in disease prevalence over time. We con-

ducted two population-based studies in the Chennai city which has a history of

well-conducted epidemiological studies that have documented the prevalence of

diabetes. The Chennai Urban Population Study (CUPS) focused on the intraur-

ban difference in the prevalence of diabetes, while the Chennai Urban Rural

Epidemiology Study (CURES) was aimed at determining the prevalence of dia-

betes and its complications.

The Chennai Urban Population Study

The CUPS was conducted in two residential areas representing the lower-

and middle-income group in Chennai (formerly Madras) in South India. All

individuals aged over 20 years living in these two colonies were requested to

participate in the study. Of the total of 1,399 eligible subjects (age �20 years),

1,262 (90.2%) participated in the study. The study subjects underwent a glucose

tolerance test and were categorized as having normal glucose tolerance,

impaired glucose tolerance (IGT) or diabetes. Twelve-lead ECG was also per-

formed and coronary artery disease was diagnosed based on a previous medical

history of coronary artery disease and/or Minnesota coding of ECGs [2].

The overall prevalence of diabetes in the CUPS was 12%, which included

91 subjects (7.2%) with known diabetes and 61 (4.8%) with undiagnosed dia-

betes [3]. An additional 74 (5.9%) subjects were detected to have IGT. The

prevalence of diabetes among the high-income group was twofold higher than

that observed in the middle-income group (12.4 vs. 6.5%). Similarly, all the

metabolic abnormalities were higher in the middle-income group compared to

the low-income group [3]. The study thus suggested that the risk for diabetes

increased with increase in income.

The Chennai Urban Rural Epidemiology Study

The CURES is an ongoing epidemiological study conducted on a repre-

sentative population (aged �20 years) of Chennai (formerly Madras), the

Mohan/Sudha/Radhika/Radha/Rema/Deepa 120

fourth-largest city in India [4]. The sampling for the CURES was based on the

model of systematic random sampling, wherein, of the 155 wards, 46 wards

were selected to represent all the 10 corporation zones. The total sample size of

26,001 individuals was selected from these 46 wards. The sample distribution

in each ward within these zones was based on the proportion of the population

in that particular zone. Further, within each ward, every third lane or road, fol-

lowing the right-hand rule, was surveyed. Such a sampling approach was cho-

sen as it enabled an equitable distribution of the entire Chennai population

while ensuring that the sampling error was kept to a minimum. All men and

women �20 years of age were considered eligible for the study.

Fasting capillary blood glucose was determined using a One Touch Basic

glucose meter in all subjects. Blood tests were done in all but 184 subjects (i.e.

in 99.3% of the study population). Subjects were categorized as having normal

fasting glucose, impaired fasting glucose or diabetes based on the American

Diabetes Association capillary blood glucose criteria [5]. Subjects were also

classified as ‘known diabetic subjects’ if they stated that they had diabetes and

were on treatment. In phase 2 of the CURES, all the known diabetic subjects

(n � 1,529) were invited to the center for detailed studies on vascular compli-

cations. In phase 3, every tenth study subject in phase 1 was invited to the cen-

ter for more detailed studies.

Evidence for Rising Prevalence of Diabetes in the Indian Subcontinent

The CURES gave us a unique opportunity to determine the prevalence of

diabetes in Chennai, and compare the secular trends in the same city over the

last two decades. The crude prevalence rate obtained in the study was age

standardized to the 1991 census of India and compared with other studies

which provided prevalence standardized to the same census. In 1989, the age-

standardized prevalence of diabetes was 8.3%; this rose to 11.6% in 1995 and to

13.5% in 2000, while in the present study (2003–2004), it is 14.3%. Thus,

within a span of 14 years, the prevalence of diabetes had increased by over 70%

(p � 0.001). From 1989 to 1995, it increased by 39.8%, between 1995 to 2000

by 16.3%, and between 2000 to 2004 by 6.0% [6].

Why the Diabetes Epidemic in India?

Although there are several reports that highlight the high prevalence of

diabetes in Indians, the exact reasons for the epidemic of diabetes in this ethnic


group are still not clear. Population-based studies conducted by us [3, 6] point

to the role of both genetic and environmental factors in contributing to the dia-

betes epidemic in India.

Heritability and Diabetes

Earlier studies on migrant Indians and Europeans conducted in the UK

showed that 10% of Asian Indians with diabetes reported that both parents were

diabetic compared to 1% of Europeans, suggesting that the inheritance is

stronger among Indians [7].

In the CUPS, the prevalence of diabetes was higher among subjects who

had a positive family history of diabetes (18.2%) compared to subjects without

a family history of diabetes (10.6%, p � 0.0015). When subjects with diabetes

and IGT were grouped together as glucose intolerance, the prevalence of glu-

cose intolerance among subjects whose parents were both diabetic (55%) was

significantly higher than the prevalence among those with one diabetic parent

(22.1%, p � 0.005), which in turn was higher than the prevalence among those

with no parental history of diabetes (15.6%, p � 0.0001) [8].

Studies have also shown Asian Indians to be more insulin resistant compared

to Europeans [9]. Indeed, hyperinsulinemia has been demonstrated even among

Indian neonates in contrast to white Caucasians [10]. Over 10 studies from differ-

ent parts of the world have confirmed that Indians have a higher degree of insulin

resistance [11]. Furthermore, Asian Indians also tend to have increased abdomi-

nal obesity compared to other ethnic populations. These studies suggest that there

could be a genetic predisposition to insulin resistance and diabetes.

Genetic Polymorphism and Diabetes

In order to answer the question whether any variants at specific genetic loci

contribute to increased susceptibility to insulin resistance and diabetes in India,

we undertook a systematic search of genes along the insulin action pathway.

In the CURES population, we studied several candidate genes for diabetes

and insulin resistance. Some showed an association with diabetes similar to that

seen in other ethnic groups while a few failed to show an association indicating

ethnic differences in genetic susceptibility to diabetes.

Three candidate genes implicated in insulin resistance and type 2 diabetes,

namely plasma cell glycoprotein 1 (PC-1), peroxisome proliferator-activated

receptor-� coactivator 1� (PGC-1�) and peroxisome proliferator-activated

receptor-� (PPAR-�) were examined in detail in the CURES population.


The Lys121Gln (K121Q) polymorphism of the PC-1 gene was studied in

South Asians living in India (n � 679) and this was compared with South

Asians living in Dallas (n � 1,083) and Caucasians (n � 858). The study result

suggests that the Lys121Gln (K121Q) polymorphism of the PC-1 gene predicts

genetic susceptibility to type 2 diabetes in both South Asians and Caucasians

[12]. The prevalence of the Lys121Gln (K121Q) polymorphism was 25% in the

nondiabetic group and 34% in the diabetic group of South Asians living in

Chennai (p � 0.01). The prevalence in the nondiabetic and diabetic groups was

33 and 45% (p � 0.01) for the South Asians living in Dallas and 26 and 39%

(p � 0.003) for the Caucasians [12]. We also observed that the ‘A’ allele of the

Thr394Thr (G → A) silent polymorphism of the PGC-1� gene was associated

with type 2 diabetes in Asian Indians and that the ‘XA’ genotype confers 1.6

(95% CI 1.264–2.241, p � 0.0004) times higher risk for type 2 diabetes com-

pared to the ‘GG’ genotype in this population [13]. The Pro12Ala polymor-

phism of the PPAR-� gene had a protective effect against diabetes in

Caucasians (20% in nondiabetic subjects vs. 9% in diabetic subjects,

p � 0.006), whereas there were no significant differences between diabetic and

nondiabetic subjects among the South Asians living in Dallas (20 vs. 23%) and

in India (19 vs. 19.3%) [14], showing that this polymorphism does not protect

South Asians against diabetes (table 1).

Table 1. Genetic polymorphism studied in the CURES [12–14, 16, 17]

Gene Polymorphism Type of Reference

association with

diabetes

PC-1 Lys121Gln susceptible [12]

(K121Q)

PGC-1� Thr394Thr (G → A) susceptible [13]

Gly482Ser (G → A) no association

�A2962G no association

PPAR-� Pro12Ala no association [14]

Insulin receptor substrate 2 Gly1057Asp gene-obesity [16]

interaction in

diabetes

Hepatocyte nuclear factor 2 Ala/Val association with [17]

lower age at

onset of type 2

diabetes


Though these studies indicate that some genetic variants at key candidate

genetic loci are associated with diabetes mellitus, genetic factors alone cannot

fully explain such a rapid rise in the prevalence of diabetes in India, as it is

highly unlikely that the frequency of genetic variation has changed significantly

in this population during the last 30 years. This points to the role of environ-

mental factors in the causation of diabetes.

Epidemiological Transition

India is undergoing rapid epidemiological transition with increasing

urbanization. Presently 35% of India is urbanized in contrast to 15% in the

1950s. Urbanization has led to rapid changes in lifestyle, with more white-col-

lar jobs leading to decreased physical activity and affluence associated with

consumption of fast foods rich in fat, sugar and calories. This epidemiological

transition has lead to a paradigm shift in the health patterns in the country, from

communicable to noncommunicable diseases. Of the latter, diabetes is one of

the most frequent.

The CUPS showed that only 5% of the Chennai residents exercised regu-

larly. Further, studies on physical activity showed that among subjects who per-

formed light-grade activity, the prevalence of diabetes was 17.0%, which was

significantly higher than that observed in subjects who performed heavy-grade

activity (5.6%) [15]. This association persisted even after adjusting for age.

As it is well known that economic transition results in nutritional changes

with increased consumption of fatty foods, we examined the relation between

visible fat (visible fats are fats and oils derived from animal and vegetable fats

that are added during cooking/processing – e.g. butter, ghee, vegetable cooking

oil and hydrogenated fat) in diet and the prevalence of diabetes. With an

increase in quartiles of visible fat consumption, the prevalence of diabetes and

IGT increased indicating a strong relation between diabetes and visible fat

intake (fig. 1). These findings implicate reduced physical activity and increased

fat intake as some of the environmental factors that may be involved in the

pathogenesis of the diabetes epidemic in India.

Gene and Environment Interaction

It is likely that both genes and environment act together and have a cumu-

lative effect on the prevalence of diabetes. To look at this aspect, we studied the

combined effect of dietary (�physical inactivity) factors and genetic factors in

increasing the risk of diabetes.


Figure 2 shows that the prevalence of glucose intolerance was 13.0% among

subjects who consumed excess visible fat but had no family history of diabetes

compared to 6.6% among those with no family history of diabetes and who

consumed less visible fat (�50% median). However, the highest prevalence of

diabetes occurred among subjects who consumed increased visible fat and had

a positive family history of diabetes. Logistic regression analysis revealed that

this effect was additive. Since family history involves both heritability and envi-

ronmental effects, to examine the interaction of heritability with diet, we further

studied specific gene-diet interactions.

0

1

2

3

4

5

6

7

8

9

Pre

vale

nce

(%)

1st quartile 2nd quartile 3rd quartile 4th quartile

Visible fat consumption

Diabetes*

IGT*

Fig. 1. Visible fat consumption and prevalence of newly diagnosed diabetes and IGT –

the CUPS. *p � 0.05 for trend.

6.6

9.4

13

20.8

0

5

10

15

20

25

Pre

vale

nce

of g

luco

sein

tole

ranc

e (%

)

Family history negative

� visible fat (�median)

Family history positive


Family history negative


Family history positive


Fig. 2. Synergistic effect of heritability and visible fat on the prevalence of glucose

intolerance – the CURES.


Gene-Diet Interactions

Gene-diet interaction was examined by conducting studies on adiponectin

gene polymorphism and fatty acid-binding protein 2 polymorphism. We found

that the adiponectin gene polymorphism (�10211T → G) contributes to insulin

resistance and diabetes and in addition also showed a cumulative effect with

dietary glycemic load. Subjects in the highest tertile of glycemic load who also

had the �10211T → G polymorphism of the adiponectin gene had an increased

risk for hypoadiponectinemia compared to those without (n � 26; OR: 51.7,

95% CI: 6.00–445.9, p � 0.0001). Similarly, the Ala54Thr polymorphism in

fatty acid-binding protein 2 showed a synergistic effect with high glycemic

load and increased the risk for hypertriglyceridemia. When obesity (BMI �25)

was introduced into the model, the risk for hypertriglyceridemia was further

increased (OR: 23.26, 95% CI: 2.38–226.43, p � 0.007). These studies indicate

that gene-diet interactions could play a major role in increasing the risk for dia-

betes in Asian Indians. Further work is required to explore whether similar

gene-environment interactions occur in individuals of different ethnic origin.

Moreover, as the number of studies is small, the confidence intervals are very

wide. Much larger studies of nutrigenomics are needed to draw meaningful

conclusions given the imprecision in measuring dietary intakes.

In summary, the present study shows that a combination of genetic suscep-

tibility (family history of diabetes) along with lifestyle changes with consump-

tion of higher amounts of visible fats combined with physical inactivity is

responsible for the diabetes epidemic in India. Preventive strategies aimed at

increasing physical activity and decreasing calorie and fat intake could be the

key for prevention of diabetes in India and other developing countries.

References

1 Wild S, Roglic G, Green A, Sicree R, King H: Global prevalence of diabetes: estimates for the year

2000 and projections for 2030. Diabetes Care 2004;27:1047–1053.

2 Mohan V, Deepa R, Shanthi Rani S, Premalatha G: Prevalence of coronary artery disease and its

relationship to lipids in a selected population in south India. The Chennai Urban Population Study

(CUPS No 5). J Am Coll Cardiol 2001;38:682–687.

3 Mohan V, Shanthirani S, Deepa R, et al: Intra-urban differences in the prevalence of the metabolic

syndrome in southern India – The Chennai Urban Population Study (CUPS No 4). Diabet Med

2001;18:280–287.

4 Deepa M, Pradeepa R, Rema M, Mohan A, Deepa R, Shanthirani S, Mohan V: The Chennai Urban

Rural Epidemiology Study (CURES) – study design and methodology (urban component)

(CURES-I). J Assoc Physicians India 2003;51:863–870.

5 Expert Committee on the Diagnosis and Classification of Diabetes Mellitus: Report of the

expert committee on the diagnosis and classification of diabetes mellitus. Diabetes Care 1997;20:

1183–1197.


6 Mohan V, Deepa M, Deepa R, Shanthirani CS, Farooq S, Ganesan A, Datta M: Secular trends in

the prevalence of diabetes and impaired glucose tolerance in urban South India – the Chennai

Urban Rural Epidemiology Study (CURES-17). Diabetologia 2006;49:1175–1178.

7 Mohan V, Sharp PS, Aber VR, Mather HM, Kohner EM: Family histories of Asian Indian and

Europeans non-insulin-dependent diabetic patients. Pract Diabetes 1986;3:254–256.

8 Mohan V, Shanthirani CS, Deepa R: Glucose intolerance (diabetes and IGT) in a selected South

Indian population with special reference to family history, obesity and life style factors – the

Chennai Urban Population Study (CUPS 14). J Assoc Physicians India 2003;51:771–777.

9 Sharp PS, Mohan V, Levy JC, Mather HM, Kohner EM: Insulin resistance in patients of Asian

Indian and European origin with non-insulin dependent diabetes. Horm Metab Res 1987;19:

84–85.

10 Mohan V, Sharp PS, Cloke HR, Burrin JM, Schumer B, Kohner EM: Serum immunoreactive

insulin responses to a glucose load in Asian Indian and European type 2 (non-insulin-dependent)

diabetic patients and control subjects. Diabetologia 1986;29:235–237.

11 Misra A, Vikram NK: Insulin resistance syndrome (metabolic syndrome) and Asian Indians. Curr

Sci 2002;83:1483–1496.

12 Abate N, Chandalia M, Satija P, Adams-Huet B, Grundy SM, Sandeep S, Radha V, Deepa R,

Mohan V: ENPP1/PC-1 K121Q polymorphism and genetic susceptibility to type 2 diabetes.

Diabetes 2005;54:1207–1213.

13 Vimaleswaran KS, Radha V, Ghosh S, Majumder PP, Deepa R, Babu HN, Rao MR, Mohan V:

Peroxisome proliferator-activated receptor-gamma co-activator-1alpha (PGC-1alpha) gene poly-

morphisms and their relationship to type 2 diabetes in Asian Indians. Diabet Med 2005;22:

1516–1521.

14 Radha V, Vimaleswaran KS, Babu HN, Abate N, Chandalia M, Satija P, Grundy SM, Ghosh S,

Majumder PP, Deepa R, Rao SM, Mohan V: Role of genetic polymorphism peroxisome proliferator-

activated receptor-gamma2 Pro12Ala on ethnic susceptibility to diabetes in South-Asian and

Caucasian subjects: evidence for heterogeneity. Diabetes Care 2006;29:1046–1051.

15 Mohan V, Gokulakrishnan K, Deepa R, Shanthirani CS, Manjala D: Association of physical inac-

tivity with components of metabolic syndrome and coronary artery disease – the Chennai Urban

Population Study (CUPS No 15). Diabet Med 2005;22:1206–1211.

16 Bodhini D, Radha V, Deepa R, Ghosh S, Majumder PP, Rao MR, Mohan V: The G1057D poly-

morphism of IRS-2 gene and its relationship with obesity in conferring susceptibility to type 2

diabetes in Asian Indians. Int J Obes (Lond) 2007;31:97–102.

17 Anuradha S, Radha V, Deepa R, Hansen T, Carstensen B, Pedersen O, Mohan V: A prevalent amino

acid polymorphism at codon 98 (Ala98Val) of the hepatocyte nuclear factor-1alpha is associated

with maturity-onset diabetes of the young and younger age at onset of type 2 diabetes in Asian

Indians. Diabetes Care 2005;28:2430–2435.

Dr. V. Mohan

Madras Diabetes Research Foundation and

Dr. Mohan’s Diabetes Specialities Centre

4, Conran Smith Road, Gopalapuram

Chennai 600 086 (India)




Gene Expression in Low Glycemic Index Diet – Impact on Metabolic Control

Eiji Takeda, Hidekazu Arai, Kazusa Muto, Kaoru Matsuo, Masae Sakuma,Makiko Fukaya, Hisami Yamanaka-Okumura, Hironori Yamamoto, Yutaka Taketani

Department of Clinical Nutrition, Institute of Health Biosciences, University of

Tokushima Graduate School, Tokushima, Japan

AbstractBackground: Correcting postprandial hyperglycemia forms an important part of the

prevention and management of type 2 diabetes. Methods: A low-glycemic-index liquid for-

mula designated as Inslow was prepared by replacing dextrin in the standard balanced for-

mula (SBF) with 55.7% palatinose. Long-term administration of Inslow prevented fatty liver

and improved insulin resistance in rats. Expressions of mRNA of factors involved in glucose

and lipid metabolism were determined to clarify its mechanism. Results: Analysis of mRNA

expressions revealed that Inslow increased the expression of enzymes involved in �-oxidation

and peroxisome proliferator-activated receptor-� (PPAR-�) in the liver, and increased PPAR-�,

adiponectin and uncoupling protein 2 as well as decreased tumor necrosis factor � in adipose

tissue in comparison with those of SBF. Conclusions: Inslow may induce improvement of

insulin resistance by accelerated �-oxidation through increased expression of the hepatic

PPAR-� gene and adipocyte PPAR-� gene. Therefore, Inslow is a functional food which pre-

vents and treats type 2 diabetes.


Postprandial Hyperglycemia and Diabetic Complications

Excessive energy intake with concomitant obesity and physical inactivity

are the main risk factors for type 2 diabetes mellitus. In addition, diets high in

fat and saturated fatty acids, but low in dietary fiber, and diets with a high

glycemic index (GI) increase the risk for type 2 diabetes. Type 2 diabetes

develops after years of insulin resistance and eventual pancreatic �-cell failure

and loss.

Takeda/Arai/Muto/Matsuo/Sakuma/Fukaya/Yamanaka-Okumura/Yamamoto/Taketani 128

The GI was introduced by Jenkins et al. [1, 2] as a quantitative assessment

of foods based on postprandial blood glucose response, expressed as a percent-

age of the response to an equivalent carbohydrate portion of a reference food,

such as white bread or glucose [3]. A high-GI food with an equivalent carbo-

hydrate content as a low-GI food induces a larger area under the glucose curve

over the postprandial period. As a consequence of the induced insulin response,

intake of a high-GI food may result in lower blood glucose concentrations over

the late (2–3 h) postprandial period than intake of a low-GI food [4].

The profile of postprandial hyperglycemia is determined by many factors

including the timing, quantity and composition of the meal, carbohydrate con-

tent, insulin and glucagon secretion. The effects of carbohydrate on health may

be better described on the basis of the ability to raise blood glucose levels,

which depend on the type of the constituent sugars, the physical form of the carbo-

hydrate, the nature of the starch and other food components [5].

There is a considerable body of evidence from epidemiological studies

supporting the concept that postprandial glucose excursions are strongly asso-

ciated with the development of macrovascular disease, the chief cause of

morbidity and mortality in patients with type 2 diabetes. A meta-analysis of

20 studies in over 80,000 subjects found a progressive relationship between

fasting plasma glucose (PG) and 2-hour glucose and cardiovascular disease

mortality [6].

A meta-analysis of randomized controlled trials investigating the effect of

low-GI versus high-GI diets on markers for carbohydrate and lipid metabolism

with a crossover or parallel design published in English between 1981 and 2003

has recently been reported [7]. Literature searches identified 16 studies that met

the strict inclusion criteria. Low-GI diets significantly reduced fructosamine

by �0.1 mmol/l (95% CI �0.20 to 0.00; p � 0.05), HbA1c by 0.27% (95%

CI �0.5 to �0.03; p � 0.03), total cholesterol by �0.33 mmol/l (95% CI �0.47

to �0.18; p � 0.0001) and tended to reduce low-density lipoprotein cholesterol

in type 2 diabetic subjects by �0.15 mmol/l (95% CI �0.31 to �0.00;

p � 0.06) compared with high-GI diets. No changes were observed in high-

density lipoprotein cholesterol and triacylglycerol concentrations. This analysis

supports the use of the GI as a scientifically based tool to enable selection of

carbohydrate-containing foods to reduce total cholesterol and improve overall

metabolic control of diabetes [2].

The GI of foods is now considered to be an important feature in the develop-

ment of insulin resistance as determined by the homeostasis model assessment

of insulin resistance. After adjustment for potential confounding variables, total

but also fruit and cereal dietary fiber intakes were inversely associated with

the homeostasis model assessment of insulin resistance in the Framingham

Offspring Study [8].

Gene Expression in Low Glycemic Index Diet 129

Mechanisms Linking Postprandial Hyperglycemia and the Risk ofDiabetes Mellitus and Its Complications

The precise mechanisms linking hyperglycemia and its associated compli-

cations are unclear at this time. However, in what follows, two potential mecha-

nisms are described.

Oxidative StressInflammation is known to be involved not only in acute illnesses, but also

in chronic conditions such as obesity, diabetes, or atherosclerotic disorders

[25]. Repeated mental stress may lead to chronic alterations in cortisol and cat-

echolamine concentrations, and to insulin resistance. Furthermore, chronically

elevated cortisol concentrations may favor the development of abdominal obe-

sity and metabolic syndrome [26, 27]. The link between mental stress and meta-

bolic disorders may be an inflammatory response that may contribute, over the

long term, to the development of central obesity and insulin resistance.

The stress response is characterized by a stimulation of the sympathetic ner-

vous system and increased secretion of both epinephrine from the adrenal medulla

and glucocorticoids from the adrenal cortex. All these stimuli can be expected to

reduce insulin sensitivity. In addition, increased glucocorticoid levels may be

linked to central obesity, an essential feature of the metabolic syndrome [26, 34].

Hyperglycemia can increase oxidative stress through several pathways

(fig. 1). A major mechanism appears to be the hyperglycemia-induced intracel-

lular reactive oxygen species (ROS), produced by the proton electromechanical

gradient generated by the mitochondrial electron transport chain and resulting

in an increased production of superoxide [28]. These results suggest that adi-

pose tissue is the major source of the elevated plasma ROS levels. Oxidative

stress is known to impair both insulin secretion by pancreatic �-cells and glu-

cose transport in muscle and adipose tissue [29–31]. Increased oxidative stress

in vascular walls is involved in the pathogenesis of hypertension and athero-

sclerosis [32]. Oxidative stress also underlies the pathophysiology of hepatic

steatosis [33]. Thus, oxidative stress locally produced in each of the above tis-

sues seems to be involved in the pathogenesis of these diseases.

Rapid increases in glucose and lipid levels after ingestion of a meal are

also likely to trigger carbonyl stress which, either independently or by potentia-

tion of oxidative stress, contributes to the development of both microvascular

and macrovascular complications [35].

Changes in Serotonin MetabolismUnder conditions of acute stress, increases in brain serotonin may improve

stress adaptation and thus may contribute to the initiation as well as termination


of a cortisol response by way of different serotonergic pathways in the brain

[36]. If an increased serotonin level constitutes a biological condition to

improve stress adaptation in stress-prone subjects, serotonin activity might be

continuously increased after chronic stress experiences. Ultimately, this may

lead to a functional shortage in serotonin, causing a subsequent deficiency of

brain serotonin activity [37].

Dietary carbohydrate enhances the uptake of circulating tryptophan into

the brain mediated by modifying the plasma amino acid pattern. Insulin has

little or no effect on plasma tryptophan levels, but it markedly lowers the

plasma levels of the large neutral amino acids (LNAAs), which compete with

tryptophan for passage across the blood-brain barrier. This decrease allows

more tryptophan to enter the brain and resolves the paradox of why dietary car-

bohydrates, which lack tryptophan, should increase brain levels of this amino

acid while protein-rich foods fail to do so. Dietary proteins raise plasma trypto-

phan levels. However, since tryptophan tends to be the least abundant of the 22

amino acids in proteins, this rise is small relative to the increases in other, more

abundant LNAAs, such as leucine, isoleucine, and valine. The carbohydrate-

rich, protein-poor diet caused a significant 42% increase in plasma trypto-

phan/�LNAAs compared with the protein-rich, carbohydrate-poor diet [37].

Young adult mice with a targeted mutation of the serotonin 5-HT2c recep-

tor gene consume more food despite normal responses to exogenous leptin

�20 00

20

40

60

80

10060 min15 min0 min

Controlglucose 5.6 mM

(100 mg/dl)

High glucose25 mM

(450 mg/dl)

Highmannitol25 mM

20 40

Time (min)

Highglucose

Fluo

resc

ence

inte

nsity

Control

Highmannitol

60 80 100

Fig. 1. ROS production by high glucose exposure in endothelial cells.


administration [38]. Chronic hyperphagia leads to a ‘middle-aged’-onset obe-

sity associated with a partial leptin resistance of late onset [38]. In addition,

older mice develop insulin resistance and impaired glucose tolerance. Levels of

food intake do not change during obesity development, indicating that 5-HT2c

receptor mutant mice undergo an age-dependent reduction in their ability to

compensate for chronic moderate hyperphagia.

Whichever of these, or other, mechanisms operates, correcting postpran-

dial hyperglycemia forms an important part of the strategy for the prevention

and management of complications in patients with type 2 diabetes.

Suppression of Postprandial Hyperglycemia by the Food

Postprandial hyperglycemia can be reduced by altering the carbohydrate

content of the diet. In the remainder of this paper, I will discuss the role of

palatinose as a source of carbohydrates in the diet.

Preparation of Palatinose-Based FoodPalatinose is a naturally occurring disaccharide composed of �-1,6-linked

glucose and fructose. Commercial palatinose is produced from sucrose by

enzymatic rearrangement and has been used as a sugar in Japan since 1985. In

vivo studies with rats and pigs indicate that palatinose is completely hydrolyzed

and absorbed in the small intestine. This is supported by in vitro studies show-

ing that intestinal disaccharidases from various species (including man) can

hydrolyze palatinose. The rate of hydrolysis, however, is very slow compared

with sucrose and maltose. Blood glucose and insulin levels in humans after oral

administration rise more slowly and reach lower maxima than after sucrose

administration. Thus, palatinose is completely cleaved and absorbed, and the

hydrolysis of palatinose by a homogenate of human intestinal mucosa is one

fourth that of sucrose [9, 10]. A previous study demonstrated that the increase

in PG and immunoreactive insulin (IRI) after palatinose ingestion was signifi-

cantly smaller than that after sucrose [11].

The novel enteral liquid formula designated as Inslow was prepared by

replacing dextrin in the standard balanced formula (SBF) with 55.7% palati-

nose (table 1). Inslow contains palatinose, branched dextrin, xylitol and other

dietary fiber carbohydrates, and mixed carbohydrates from raw material as the

principal carbohydrates, and the percentages of protein, fat, and carbohydrate in

the formula are 20, 29.7 and 50.3%, respectively. The commercially available

SBF that was used for comparison contains dextrin and sucrose as the principal

carbohydrates, and the percentages of protein, fat, and carbohydrate are 16, 25

and 59%, respectively.


Effect of Palatinose-Based Food on Glucose and Lipid Metabolism in MenIn the human study, peak PG and IRI levels at 30 min after Inslow loading

were lower than after SBF loading. Postprandial fat oxidation rates in the

Inslow group were higher than those in the SBF group. The free fatty acid con-

centration in the Inslow group immediately before lunch was significantly

lower than that in the SBF group. PG and IRI levels in the Inslow group after

standard lunch were lower than those in the SBF group, though the peak levels

in these groups were not different [12].

Effect of Palatinose-Based Food on Glucose and Lipid Metabolism in RatsThe effect of Inslow on carbohydrate and lipid metabolism in Sprague-

Dawley rats was compared with that of SBF [13]. After a bolus intragastric

injection of each formula equivalent to 0.9 g/kg carbohydrate, the peak levels of

PG and IRI in the femoral vein of the Inslow group were significantly smaller

than those of the SBF group (fig. 2). The values of the total incremental area

(area under the curve) of PG and IRI from the basal level 120 min after Inslow

ingestion were significantly smaller than after SBF ingestion.

From 20 to 27 weeks of age, daily food intake and body weight did not dif-

fer significantly among the Inslow and SBF groups. After ingestion of Inslow or

Table 1. Composition of Inslow and standard balanced

formula (SBF)

Inslow SBF

Energy, kcal/ml 1 1Protein, % 20.0 16.0Fat, % 29.7 25.0

SFA 9.5 9.0

MUFA 68.5 45.0

PUFA 16.8 40.0

Carbohydrate, % 50.3 59.0Maltodextrin 22.8 Sucrose 2.8

Xylitol 8.9 Dextrin 97.2

Palatinose 68.3

SFA � Saturated fatty acid; MUFA � monounsaturated

fatty acid; PUFA � polyunsaturated fatty acid.


SBF for 2 months, fasting PG levels were not different among the two groups,

but the IRI level in the Inslow group was significantly lower than that in the SBF

group. Serum triglyceride (TG) level markedly decreased by 34% in the Inslow

group and increased by 23% in the SBF group. The TG level of the Inslow group

was significantly lower than that of the SBF group. The concentrations of serum

free fatty acid and total cholesterol did not differ among the two groups.

The weights of epididymal, mesenteric, and retroperitoneal adipose tissues

were significantly lower in the Inslow group than in the SBF group (table 2). The

concentration of TG in the liver in the Inslow group was significantly lower than

that in the SBF group. Insulin sensitivity in the Inslow and SBF groups was eval-

uated by the hyperinsulinemic euglycemic clamp test with oral glucose load. The

glucose infusion rate, which reflected the insulin sensitivity in peripheral tissues,

of the Inslow group was significantly higher than that of the SBF group. The rate

of hepatic glucose uptake, which might reflect insulin sensitivity in the liver, was

significantly higher in the Inslow group than in the SBF group.

Expressions of mRNA of Genes Involved in Glucose and Lipid MetabolismAfter the acclimatization period, rats were divided into two groups: spray-

dried Inslow powder-fed rats and spray-dried SBF powder-fed rats as reported

previously [13]. Both groups were administrated each diet (80 kcal/day) with a

pair-feeding condition and water ad libitum for 8 weeks. All rats were sacrificed

after 8 weeks on the experimental diets to extract total RNA. To assess diet-

induced changes in gene expressions, mRNA levels of genes involved in

0

2

4

6

8

10

12

0 15 30 60 90 120

Pla

sma

gluc

ose

(mm

ol/l)

**

0

40

80

120

160

0 15 30 60 90 120

Pla

sma

insu

lin (p

mol

/l)

Time (min)

**

Time (min)a b

Fig. 2. Changes in plasma glucose (a) and plasma insulin (b) levels in the femoral vein

after oral administration of Inslow (white circles) and SBF (black squares). Values are

means � SE for n � 10. *p � 0.001 (vs. Inslow).


glucose and lipid homeostasis were determined by reverse transcription of total

RNA followed by PCR analysis [14].

Analysis of mRNA expressions in the liver revealed that Inslow did not

change the level of expression of sterol response element-binding protein 1c but

increased the expression of peroxisome proliferator-activated receptor-�(PPAR-�) in fasting conditions (table 3). The mRNA levels of 3,2-trans-enoyl-

CoA isomerase, carnitine palmitoyl transferase 1, acyl-CoA oxidase, and

uncoupling protein 2 in the liver of the Inslow group were significantly higher

than those of the SBF group. In contrast, there were no differences in the

mRNA levels of hepatic glucokinase, pyruvate kinase, glucose 6-phosphatase,

Table 2. Effects of long term administration of Inslow on amounts of body fat and

hepatic TG in rats

Fat, g/kg bodyweight Hepatic TG

epididymal mesenteric retroperitoneal

mmol/g tissue

SBF 23.8 � 2.4 22.0 � 1.3 29.7 � 1.9 136.7 � 20.3

Inslow 14.5 � 1.2** 11.6 � 0.6** 19.6 � 1.2** 77.9 � 10.1*

*p � 0.05 (vs. SBF); **p 0.01 (vs. SBF).

Table 3. Effect of long-term administration of Inslow on gene expression in rat liver

and adipose tissue

LiverFat oxidation: increased

PPAR-�, carnitine palmitoyl transferase 1,

acyl-CoA oxidase, 3,2-trans-enoyl-CoA isomerase

Fatty acid synthesis: no change

Sterol response element-binding protein 1c, fatty acid synthase, glucokinase, pyruvate

kinase, glucose 6-phosphatase, phosphoenolpyruvate carboxykinase

Energy expenditure, antioxidant activity: increased

Uncoupling protein 2

Adipose tissueEnergy expenditure, antioxidant activity: increased

PPAR-�, adiponectin, carnitine palmitoyl transferase 1,

3,2-trans-enoyl-CoA isomerase, uncoupling protein 2

Inflammation: decreased

TNF-�


phosphoenolpyruvate carboxykinase and fatty acid synthetase (FAS) in both

groups. Administration of Inslow increased the expression of PPAR-�, 3,2-

trans-enoyl-CoA isomerase, carnitine palmitoyl transferase 1, adiponectin and

uncoupling protein 2 mRNAs and decreased tumor necrosis factor � (TNF-�)

mRNA in adipose tissue in comparison with those of the SBF group.

In a previous report, the chronic effects of high-GI diet and low-GI diet on

the lipogenic enzymes, FAS and lipoprotein lipase have been evaluated in nor-

mal and diabetic (streptozotocin-injected on day 2 of life) male Sprague-

Dawley rats [15]. After 3 weeks, neither body weights nor relative epididymal

fat pad weights differed between the normal and diabetic rats, and high-GI diet

induced high basal plasma insulin levels. Plasma TGs were not significantly

affected by diet in either normal or diabetic rats. Adipose tissue and liver

lipoprotein lipase activities were not modified by the types of GI diet. In normal

rats, FAS activity and gene expression in epididymal adipose tissue but not in

the liver were greater in rats consuming high-GI diet than in those consuming

low-GI diet. High-GI diet compared with low-GI diet resulted in lower hepatic

phosphoenolpyruvate carboxykinase mRNA in both normal and diabetic rats.

Therefore, high-GI diet is implicated in stimulating FAS activity and lipogene-

sis and might have undesirable long-term metabolic effects.

Functions of Palatinose-Based FoodPPAR-� is an important lipid sensor and regulator of cellular energy

metabolism. It was shown to be a critical player in the regulation of cellular

uptake and �-oxidation of fatty acid. PPAR-� triggers the expression of two

proteins that transport fatty acids across the cell membrane: the fatty acid

transport protein and fatty acid translocase [16], suggesting a role in cellular

uptake and lipid homeostasis. Activation of PPAR-� also directly upregulates

the transcription of the long-chain fatty acid acyl-CoA synthetase and of vari-

ous enzymes of the peroxisomal �-oxidation pathways, such as acyl-CoA

oxidase, acyl-CoA hydratase and dehydrogenase, and keto-acyl-CoA thiolase

[17, 18].

The adipose tissue-derived cytokine TNF-� could be involved in the devel-

opment of diabetes through several mechanisms, and elevated levels of this

cytokine have been shown to be linked to the risk of diabetes [19]. Uncoupling

proteins have also been linked to the development of both obesity and type 2

diabetes [20, 21]. Adiponectin is believed to improve insulin resistance, since it

inhibits the expression of TNF-� [22] and decreases the content of TG in tissues

by enhancing oxidation of fatty acids in skeletal muscles [23]. Furthermore,

plasma adiponectin levels are inversely associated with several risk factors for

the metabolic syndrome, including adiposity, insulin resistance, diastolic blood

pressure, TG concentrations and TNF-� receptor concentrations [24].


These findings suggest that Inslow is a functional food that is effective for

the prevention and treatment of obesity, diabetes and metabolic syndrome

because it regulates gene expression and consequently glucose and lipid

homeostasis.

Conclusions

A diet based on low-GI foods may contribute to the prevention of diabetes

mellitus and its complications. Reducing the rate of carbohydrate absorption by

lowering the GI of the diet may have several health benefits, such as a reduced

insulin demand, improved blood glucose control and reduced blood lipid con-

centrations [5]. Our data suggest that these may be mediated through pathways

involved in fatty acid metabolism and regulated by PPAR-�. The metabolic

changes and the subsequent physiological processes evoked by Inslow suggest

that the substitution of carbohydrates in foods by palatinose may play a role in

the prevention and the management of cardiovascular disease, stroke and type 2

diabetes (fig. 3).

Noteworthy of the emerging evidence is that, in most studies, it is not only

the consumption of fruit and vegetables that is associated with a reduced risk of

type 2 diabetes, but also the consumption of whole-grain foods. The latter, how-

ever, does not appear to play a major role in the regulation of postprandial glu-

cose metabolism, which would suggest that the protective effect of whole grain

High energy intake, high fat diet, high-GI diet, secretion of 1st phase insulin defect

Prevention and management of cardiovascular disease, stroke and type 2 diabetes

Postprandial hyperglycemia

ROS production

Low-GI diet

Fig. 3. Effects of low-GI diet on the prevention and the management of cardiovascular

disease, stroke and type 2 diabetes.


against type 2 diabetes is mediated by as yet complex and incompletely eluci-

dated mechanisms [39, 40].

Acknowledgement

This study was supported by a Grant-in-Aid for Scientific Research from the Ministry

of Education, Science, and Culture, Japan, and the 21st Century COE Program, Human

Nutritional Science on Stress Control, Tokushima, Japan.

References

1 Jenkins DJ, Wolever TM, Taylor RH, Barker H, Fielden H, Baldwin JM, Bowling AC, Newman HC,

Jenkins AL, Goff DV: Glycemic index of foods: a physiological basis for carbohydrate exchange.

Am J Clin Nutr 1981;34:362–366.

2 Jenkins DJA, Wolever TMS, Jenkins AL, Josse RG, Wong GS: The glycaemic response to carbo-

hydrate foods. Lancet 1984;ii:388–391.

3 Wolever TMS, Jenkins DJA, Jenkins AL, Josse RG: The glycemic index: methodology and clini-

cal implications. Am J Clin Nutr 1991;54:846–864.

4 Brand-Miller J, Nantel G, Slama G, Lang V: Glycemic Index and Health: the Quality of the

Evidence. Paris, John Libbey Eurotext, 2001.

5 Augustin LS, Franceschi S, Jenkins DJA, Kendall CWC, La Vecchia C: Glycemic index and

chronic disease: a review. Eur J Clin Nutr 2002;56:1049–1071.

6 Coutinho M, Gerstein HC, Wang Y, Yusuf S: The relationship between glucose and incident

cardiovascular events. A metaregression analysis of published data from 20 studies of 95,783

individuals followed for 12.4 years. Diabetes Care 1999;22:233–240.

7 Opperman AM, Venter CS, Oosthuizen W, Thompson RL, Vorster HH: Meta-analysis of the health

effects of using the glycaemic index in meal-planning. Br J Nutr 2004;92:367–381.

8 McKeown NM, Meigs JB, Liu S, Saltzman E, Wilson PW, Jacques PF: Carbohydrate nutrition,

insulin resistance, and the prevalence of the metabolic syndrome in the Framingham Offspring

Cohort. Diabetes Care 2004;27:538–546.

9 Dahlqvist A, Auricchio S, Semenza G, Prader A: Human intestinal disaccharidases and hereditary

disaccharide intolerance: the hydrolysis of sucrose, isomaltose, palatinose (isomaltulose), and 1,6-

�-oligosaccharide (isomalto-oligosaccharide) preparation. J Clin Invest 1963;42:556–562.

10 Lina BA, Jonker D, Kozianowski G: Isomaltulose (palatinose): a review of biological and toxico-

logical studies. Food Chem Toxicol 2002;40:1375–1381.

11 Kawai K, Okuda Y, Yamashita K: Changes in blood glucose and insulin after an oral palatinose

administration in normal subjects. Endocrinol Jpn 1985;32:933–936.

12 Arai H, Mizuno A, Sakuma M, Fukaya M, Matsuo K, Muto K, Sasaki H, Matsuura M, Okumura H,

Yamamoto H, Taketani Y, Doi T, Takeda E: Effects of a palatinose-based liquid diet (Inslow) on

glycemic control and the second-meal effect in healthy men. Metabolism 2007;56:115–121.

13 Arai H, Mizuno A, Matsuo K, Fukaya M, Sasaki H, Arima H, Matsuura M, Taketani Y, Doi T, Takeda E:

Effect of a novel palatinose-based liquid balanced formula (MHN-01) on glucose and lipid metabolism

in male Sprague-Dawley rats after short- and long-term ingestion. Metabolism 2004;53:977–983.

14 Matsuo K, Arai H, Muto K, Fukaya M, Sato T, Mizuno A, Sakuma M, Yamanaka-Okumura H,

Sasaki H, Yamamoto H, Taketani Y, Doi T, Takeda E: Antiobesity effect of long-term palatinose-

based formula (Inslow) administration mediated by hepatic PPAR-� and adipocyte PPAR-� gene

expression. J Clin Biochem Nutr 2007, in press.

15 Kabir M, Rizkalla SW, Quignard-Boulange A, Guerre-Miflo M, Boillot J, Ardouin B, Luo J, Slama G:

A high glycemic index starch diet affects lipid storage-related enzymes in normal and to a lesser

extent in diabetic rats. J Nutr 1998;128:1878–1883.


16 Motojima K, Passilly P, Peters J M, Gonzalez F J, Latruffe N: Expression of putative fatty acid

transporter genes are regulated by peroxisome proliferator-activated receptor alpha and gamma

activators in a tissue- and inducer-specific manner. J Biol Chem 1998;273:16710–16714.

17 Schoonjans K, Watanabe M, Suzuki H, Mahfoudi A, Krey G, Wahli W, Grimaldi P, Staels B,

Yamamoto T, Auwerx J: Induction of the acyl-coenzyme A synthetase gene by fibrates and fatty

acids is mediated by a peroxisome proliferator response element in the C promoter. J Biol Chem

1995;270:19269–19276.

18 Tugwood JD, Issemann I, Anderson RG, Bundell KR, McPheat WL, Green S: The mouse peroxi-

some 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.

19 Fasshauer M, Paschke R: Regulation of adipocytokines and insulin resistance. Diabetologia

2003;46:1594–1603.

20 Ricquier D, Bouillaud F: The uncoupling protein homologues: UCP1, UCP2, UCP3, StUCP and

AtUCP. Biochem J 2000;345:161–179.

21 Dalgaard LT, Pedersen O: Uncoupling proteins: functional characteristics and role in the patho-

genesis of obesity and type II diabetes. Diabetologia 2001;44:946–965.

22 Maeda N, Shimomura I, Kishida K, Nishizawa H, Matsuda M, Nagaretani H, Furuyama N, Kondo H,

Takahashi M, Arita Y, Komuro R, Ouchi N, Kihara S, Tochino Y, Okutomi K, Horie M, Takeda S,

Aoyama T, Funahashi T, Matsuzawa Y: Diet-induced insulin resistance in mice lacking

adiponectin/ACRP30. Nat Med 2002;8:731–737.

23 Yamauchi T, Kamon J, Waki H, Terauchi Y, Kubota N, Hara K, Mori Y, Ide T, Murakami K,

Tsuboyama- Kasaoka N, Ezaki O, Akanuma Y, Gavrilova O, Vinson C, Reitman ML, Kagechika H,

Shudo K, Yoda M, Nakano Y, Tobe K, Nagai R, Kimura S, Tomita M, Froguel P, Kadowaki T: The

fat-derived hormone adiponectin reverses insulin resistance associated with both lipoatrophy and

obesity. Nat Med 2001;7:941–946.

24 Fernandez-Real JM, Lopez-Bermejo A, Casamitjana R, Ricart W: Novel interactions of

adiponectin with the endocrine system and inflammatory parameters. J Clin Endocrinol Metab 2003;88:

2714–2718.

25 Grimble RF: Inflammatory status and insulin resistance. Curr Opin Clin Nutr Metab Care 2002;5:

551–559.

26 Bjorntorp P: The regulation of adipose tissue distribution in humans. Int J Obes Relat Metab

Disord 1996;20:291–302.

27 Masuzaki H, Paterson J, Shinyama H, Morton NM, Mullins JJ, Seckl JR, Flier JS: A transgenic

model of visceral obesity and the metabolic syndrome. Science 2001;294:419–422.

28 Nishikawa T, Edelstein D, Du XL, Yamagishi S, Matsumura T, Kaneda Y, Yorek MA, Beebe D,

Oates PJ, Hammes HP, Giardino I, Brownlee M: Normalizing mitochondrial superoxide produc-

tion blocks three pathways of hyperglycaemic damage. Nature 2000;404:787–790.

29 Matsuoka T, Kajimoto Y, Watada H, Kaneto H, Kishimoto M, Umayahara Y, Fujitani Y, Kamada T,

Kawamori R, Yamasaki Y: Glycation-dependent, reactive oxygen species-mediated suppression of

the insulin gene promoter activity in HIT cells. J Clin Invest 1997;99:144–150.

30 Maddux BA, See W, Lawrence JC Jr, Goldfine AL, Goldfine ID, Evans JL: Protection against

oxidative stress-induced insulin resistance in rat L6 muscle cells by micromolar concentrations of

�-lipoic acid. Diabetes 2001;50:404–410.

31 Rudich A, Tirosh A, Potashnik R, Hemi R, Kanety H, Bashan N: Prolonged oxidative stress

impairs insulin-induced GLUT4 translocation in 3T3-L1 adipocytes. Diabetes 1998;47:

1562–1569.

32 Ohara Y, Peterson TE, Harrison DG: Hypercholesterolemia increases endothelial superoxide anion

production. J Clin Invest 1993;91:2546–2551.

33 Roskams T, Yang SQ, Koteish A, Durnez A, DeVos R, Huang X, Achten R, Verslype C, Diehl AM:

Oxidative stress and oval cell accumulation in mice and humans with alcoholic and nonalcoholic

fatty liver disease. Am J Pathol 2003;163:1301–1311.

34 Bjorntorp P: Growth hormone, insulin-like growth factor-I and lipid metabolism: interactions with

sex steroids. Horm Res 1996;46:188–191.

35 Baynes JW, Thorpe SR: Role of oxidative stress in diabetic complications: a new perspective on an

old paradigm. Diabetes 1999;48:1–9.


36 Anisman H, Zacharko RM: Depression as a consequence of inadequate neurochemical adaptation

in response to stressors. Br J Psychiatry 1992;160(suppl 15):36–43.

37 Markus CR, Panhuysen C. Tuiten A, Koppeschaar H, Fekkes D, Peters M: Does carbohydrate-rich,

protein-poor food prevent a deterioration of mood and cognitive performance of stress-prone sub-

jects when subjected to a stressful task? Appetite 1998;31:49–65.

38 Nonogaki K, Strack A, Dallman M, Tecott L: Leptin-independent hyperphagia and type 2 diabetes

in mice with a mutated serotonin 5-HT2c receptor gene. Nat Med 1998;4:1152–1156.

39 Parillo M, Riccardi G: Diet composition and the risk of type 2 diabetes: epidemiological and clin-

ical evidences. Br J Nutr 2004;92:7–19.

40 Steyn NP, Mann J, Bennett PH, Temple N, Zimmet P, Tuomilehto J, Lindstrom J, Louheranta A:

Diet, nutrition and the prevention of type 2 diabetes. Public Health Nutr 2004;7:147–165.

Dr. Eiji Takeda

Department of Clinical Nutrition, Institute of Health Biosciences

University of Tokushima Graduate School, Kuramoto-cho 3–18–15

Tokushima 770–8503 (Japan)




Genetic Polymorphisms in Folate-Metabolizing Enzymes and Risk ofGastroesophageal Cancers: A PotentialNutrient-Gene Interaction in CancerDevelopment

Dongxin Lina, Hui Lib, Wen Tana, Xiaoping Miaoa, Li Wangb

aDepartment of Etiology and Carcinogenesis, Cancer Institute/Hospital andbDepartment of Epidemiology, Institute of Basic Medical Sciences, Chinese Academy

of Medical Sciences and Peking Union Medical College, Beijing, China

AbstractFolate deficiency has been associated with certain types of human cancer. We therefore

investigated the effects of genetic polymorphisms in folate-metabolizing enzymes on the risk

of developing gastroesophageal cancers in a Chinese population where folate deficiency is

common. We found that functional polymorphisms in methylenetetrahydrofolate reductase

(MTHFR) and thymidylate synthase (TS), two key enzymes involved in folate and methyl

group metabolism, were significantly associated with increased risk of esophageal squamous

cell carcinoma, gastric cardia carcinoma, and pancreatic carcinoma. The polymorphisms

modulate risk of these cancers associated with low folate status. Our results suggest that

MTHFR and TS genotypes may be determinant of gastroesophageal cancers in this at-risk

Chinese population.


Folate deficiency resulting from low consumption of vegetables and

fruits is associated with an increased risk of several cancers, including gas-

troesophageal cancer and pancreatic cancer [1–3]. As an essential cofactor for

the de novo biosynthesis of purines and thymidylate, folate plays a crucial

role in DNA synthesis, repair, and integrity [4]. Folate is also an essential

nutrient to provide methyl groups for intracellular DNA methylation reactions

[5]. Although the exact mechanism and extent of the relationship between

Genetic Polymorphisms and Risk of Gastroesophageal Cancers 141

folate deficiency and risk of these cancers have not been established, aberrant

DNA synthesis, repair, and methylation, which result in abnormal gene

expression, genome instability, and mutagenesis, may be involved in carcino-

genesis [4, 5].

To serve as a methylation mediator, folate requires metabolism catalyzed

by several enzymes. It has been suggested that genetic variants resulting in

changes in the expression or function of these enzymes may contribute to can-

cer risk, over and above that associated with folate deficiency.

Methylenetetrahydrofolate reductase (MTHFR) and thymidylate synthase

(TS) are two key enzymes involved in folate and methyl group metabolism.

MTHFR catalyzes the irreversible reduction of 5,10-methylenetetrahydrofo-

late to 5-methyltetrahydrofolate, the carbon donor for the final formation of

the cellular universal methyl donor, S-adenosyl methionine. TS catalyzes the

reductive methylation of dUMP by 5,10-methylenetetrahydrofolate to form

dTMP, a rate-limiting step in DNA synthesis. Accordingly, variation in

MTHFR and TS functions may contribute to the susceptibility to carcinogene-

sis through aberrant DNA methylation and/or diminished thymidylate synthe-

sis [4, 5]. Two single nucleotide polymorphisms in MTHFR, 677C → T and

1298A → C, have been associated with a phenotype that presents significant

reduction in enzyme activity [6, 7]. It was reported that individuals carrying

the variant MTHFR genotypes, especially in the context of inadequate folate

intake, have significantly reduced levels of global DNA methylation, com-

pared with those carrying the wild-type genotype [8–10]. The 5�-UTR of TS

contains a variable number of 28-bp tandem repeats, mainly 2 repeats (2R) and

3 repeats (3R), and has been associated with the efficiency of the gene expres-

sion [11]. Individuals with the 3R/3R genotype have higher TS RNA levels

compared with those with the 2R/2R genotype [12]. A G → C polymorphism

within the second repeat of the 3R allele has been found and this mutation dis-

rupts the USF-1-binding consensus element and consequently downregulates

TS expression [13, 14].

Because of the key roles of MTHFR and TS in folate biotransformation

linked to normal DNA methylation and genome integrality, we hypothesized

that genetic polymorphisms resulting in impaired expression or activity of

these two enzymes might confer individual susceptibility to cancer. To examine

this hypothesis, we have analyzed the associations between risks of develop-

ing esophageal squamous cell carcinoma (ESCC), gastric cardia adeno-

carcinoma, and pancreatic cancer and the polymorphisms in MTHFR and TS

in case-control studies in a Chinese population, where folate deficiency has

been shown to be common and 60% of the men are plasma folate deficient in

spring [15].

Lin/Li/Tan/Miao/Wang 142

Esophageal Squamous Cell Carcinoma

Hospital-based case-control studies were conducted to examine the associ-

ations between two functional genetic polymorphisms in MTHFR and the risk

of developing ESCC and gastric cardia adenocarcinoma. For the ESCC study,

240 patients and 360 sex- and age-frequency-matched controls were recruited.

Genotypes of the MTHFR 677C → T and 1298A → C polymorphic sites were

analyzed by using PCR-based restriction fragment length polymorphism meth-

ods. We observed that the allele frequency of MTHFR 677T was significantly

higher in patients than in controls (63% vs. 41%, p � 0.001). Subjects with the

677TT genotype had a more than 6-fold increased risk for developing ESCC

[odds ratio (OR) � 6.18, 95% confidence interval (CI) � 3.32–11.51] com-

pared with those with the 677CC genotype. Moreover, the increased ESCC risk

associated with the polymorphism was in an allele-dose relationship (trend test,

p � 0.0001), with ORs of 1.00, 3.14 (95% CI � 1.94–5.08), and 6.18 (95%

CI � 3.32–11.51) for the CC, CT, and TT genotype, respectively, after adjust-

ment for age, sex, and smoking status. The 1298CC genotype was extremely

rare in both controls (1.4%) and cases (2.9%) and was also associated with an

increased risk of ESCC (OR � 4.43, 95% CI � 1.23–16.02) compared with the

1298AA genotype [16]. The results were similar for gastric cardia adenocarci-

noma, a common cancer and more prevalent in areas of high risk of esophageal

cancer in China, in a case-control study consisting of 217 patients and 468

frequency-matched controls (matched for age and sex). We found that subjects

with the MTHFR 677TT variant genotype had a 2-fold increased risk for cancer

(95% CI � 1.28–3.26). Furthermore, a significantly elevated risk was also seen

among the 677CT heterozygotes (OR � 1.56, 95% CI � 1.03–2.36). However,

the 1298 polymorphism had no effect on the risk [17]. These findings suggest

that the MTHFR genotype may be a genetic determinant of gastroesophageal

cancers among this at-risk population where folate intake is low.

Gastroesophageal Cancer

We also performed an independent case-control study to investigate the

association between risk of gastroesophageal cancers and TS polymorphisms

along, and in interaction with serum folate status. A total of 555 patients (324

with ESCC and 231 with gastric cardia adenocarcinoma) and 492 controls were

analyzed for their TS genotypes of the 28-bp tandem repeats and G → C single

nucleotide polymorphism in the 5�-UTR and serum folate concentration.

We found that compared with the normal-expression TS genotype (3Rg/3Rg �3Rg/3Rc � 3Rg/2R), the low-expression TS genotype (3Rc/3Rc � 3Rc/


2R � 2R/2R) alone was significantly associated with an increased risk of

ESCC (OR � 1.47, 95% CI � 1.03–2.10) but not gastric cardia adenocarci-

noma (OR � 0.98, 95% CI � 0.68–1.40). More importantly, a significant

interaction between the TS polymorphisms and serum folate status in relation to

the risk of esophageal cancer and gastric cardia cancer was observed

(p � 0.002 and 0.029, respectively) [18]. Compared to individuals with normal

folate (�3 mg/dl) and the high-expression genotype, those with normal folate

and the low-expression genotype had an increased risk of ESCC (OR � 1.35,

95% CI � 0.85–2.14). Low folate was associated with an increased risk of

ESCC in the presence of the high-expression genotype (OR � 9.97, 95%

CI � 5.67–17.53). However, the combination of low folate with the low-

expression phenotype was associated with the highest risk of ESCC

(OR � 22.64, 95% CI � 10.44–49.05), greater than that expected if both geno-

type and folate status were simply independent risk factors. A similar interac-

tion was noted for gastric cardia cancer. The combination of low folate and the

low-expression genotype was associated with a higher risk of gastric cardia

cancer (OR � 4.08, 95% CI � 1.94–8.59) than either genotype (OR � 0.84,

95% CI � 0.56–1.27) or folate status (OR � 1.88, 95% CI � 1.18–3.24) alone.

Pancreatic Cancer

Although the results from epidemiologic studies are not all consistent,

prospective studies showed that low folate intake or low serum folate concen-

tration is associated with risk of pancreatic cancer. We therefore examined the

contribution of functional polymorphisms in MTHFR and TS to the risk of this

cancer in a case-control study consisting of 163 patients with pancreatic ductal

adenocarcinoma and 337 controls frequency-matched to the patients by sex and

age (�5 years). We observed a significantly increased risk of pancreatic cancer

associated with the MTHFR 677CT (OR � 2.60, 95% CI � 1.61–4.29, p � 0.005)

or 677TT (OR � 5.12, 95% CI � 2.94–9.10, p � 0.001) genotype compared

with the 677CC genotype. An increased risk of pancreatic cancer was also asso-

ciated with the TS 3Rc/3Rc genotype (OR � 2.19, 95% CI � 1.13–4.31,

p � 0.022) compared with the TS 3Rg/3Rg genotype. Furthermore, we found a

significant interaction between the MTHFR C677T polymorphism and smok-

ing (which depletes systemic and intracellular folate) or drinking (a well-known

folate antagonist) intensifying the risk of pancreatic cancer. The ORs for smok-

ing, the polymorphism and both factors combined were 0.70 (95%

CI � 0.30–1.63), 2.17 (95% CI � 1.17–4.21) and 3.10 (95% CI � 1.54–6.51),

respectively. This effect was much stronger in heavy smokers (OR � 6.69, 95%

CI � 3.39–13.63, p � 0.0001). The ORs for drinking, the polymorphism and

Lin/Li/Tan/Miao/Wang 144

both factors combined were 0.98 (95% CI � 0.40–2.30), 2.81 (95%

CI � 1.65–4.98) and 4.39 (95% CI � 2.25–8.78), respectively [19].

Conclusions

In summary, these findings demonstrate a significant association between

genetic polymorphisms in the folate-metabolizing genes MTHFR and TS and

risks of gastroesophageal and pancreatic cancers in the Chinese population.

These genetic variants modulate the risk of these cancers associated with

low folate status. Increased folate intake may overcome the effects of geneti-

cally determined reduction of MTHFR or TS activity, and suggests a potential

role of enhanced folate intake in the prevention of gastroesophageal and pan-

creatic cancers in an at-risk population of individuals carrying the variant

MTHFR and TS alleles.

References

1 Chang-Claude JC, Wahrendorf J, Liang QS, Rei YG, Munoz N, Crespi M, Raedsch R, Thurnham DI,

Correa P: An epidemiological study of precursor lesions of esophageal cancer among young per-

sons in a high-risk population in Huixian, China. Cancer Res 1990;50:2268–2274.

2 Mayne ST, Risch HA, Dubrow R, Chow WH, Gammon MD, Vaughan TL, Farrow DC, Schoenberg JB,

Stanford JL, Ahsan H, West AB, Rotterdam H, Blot WJ, Fraumeni JF Jr: Nutrient intake and risk of sub-

types of esophageal and gastric cancer. Cancer Epidemiol Biomarkers Prev 2002;10:1055–1062.

3 Howe GR, Burch JD: Nutrition and pancreatic cancer. Cancer Causes Control 1996;7:69–82.

4 Duthie SJ: Folic acid deficiency and cancer: mechanisms of DNA instability. Br Med Bull 1999;55:

578–592.

5 Choi SW, Mason JB: Folate and carcinogenesis: an integrated scheme. J Nutr 2000;130:129–132.

6 Frosst P, Blom HJ, Milos R, Goyette P, Sheppard CA, Matthews RG, Boers GJ, den Heijer M,

Kluijtmans LA, van den Heuvel LP, Roers GJ: A candidate genetic risk factor for vascular disease:

a common mutation in methylenetetrahydrofolate reductase. Nat Genet 1995;10:111–113.

7 Weisberg I, Tran P, Christensen B, Sibani S, Rozen R: A second genetic polymorphism in methyl-

enetetrahydrofolate reductase (MTHFR) associated with decreased enzyme activity. Mol Genet

Metab 1998;64:169–172.

8 Stren LL, Mason JB, Selhub J, Choi SW: Genomic DNA hypomethylation, a characteristic of most

cancers, is present in peripheral leukocytes of individuals who are homozygous for the C677T

polymorphism in the methylenetetrahydrofolate reductase gene. Cancer Epidemiol Biomarkers

Prev 2000;9:849–853.

9 Castro R, Rivera I, Ravasco P, Camilo ME, Jakobs C, Blom HJ, De Almeida IT: 5,10-methyl-

enetetrahydrofolate reductase (MTHFR) 677C → T and 1298A → C mutations are associated

with DNA hypomethylation. J Med Genet 2004;41:454–458.

10 Friso S, Choi S-W, Girelli D, Mason JB, Dolnikowski GG, Bagley PJ, Olivieri O, Jacques PF,

Rosenberg IH, Corrocher R, Selhub J: A common mutation in the 5,10-methylenetetrahydrofolate

reductase gene affects genomic DNA methylation through an interaction with folate status. Proc

Natl Acad Sci USA 2002;99:5605–5611.

11 Horie N, Aiba H, Oguro K, Hojo H, Takeishi K: Functional analysis and DNA polymorphism of

the tandemly repeated sequences in the 5�-terminal regulatory region of the human gene for

thymidylate synthase. Cell Struct Funct 1995;20:191–197.


12 Pullarkat ST, Stoehlmacher J, Ghaderi V, Xiong Y-P, Ingles SA, Sherrod A, Warren R, Tsao-Wei D,

Groshen S, Lenz H-J: Thymidylate synthase gene polymorphism determines response and toxicity

of 5-FU chemotherapy. Pharmacogenomics J 2001;1:65–70.

13 Mandola MV, Stoehlmacher J, Muller-Weeks S, Cesarone G, Yu MC, Lenz HJ, Ladner RD:

A novel single nucleotide polymorphism within the 5� tandem repeat polymorphism of the thymidy-

late synthase gene abolishes USF-1 binding and alters transcriptional activity. Cancer Res

2003;63: 2898–2904.

14 Kawakami K, Watanabe G: Identification and functional analysis of single nucleotide poly-

morphism in the tandem repeat sequence of thymidylate synthase gene. Cancer Res 2003;63:

6004–6007.

15 Hao L, Ma J, Stampfer MJ, Ren A, Tian Y, Tang Y, Willett WC, Li Z: Geographical, seasonal and

gender differences in folate status among Chinese adults. J Nutr 2003;133:3630–3635.

16 Song C, Xing D, Tan W, Wei Q, Lin D: Methylenetetrahydrofolate reductase polymorphisms

increase risk of esophageal squamous cell carcinoma in a Chinese population. Cancer Res 2001;

61:3272–3275.

17 Miao X, Xing D, Tan W, Qi J, Lu W, Lin D: Susceptibility to gastric cardia adenocarcinoma and

genetic polymorphisms in methylenetetrahydrofolate reductase in an at-risk Chinese population.

Cancer Epidemiol Biomarkers Prev 2002;11:1454–1458.

18 Tan W, Miao X, Wang L, Yu C, Xiong P, Liang G, Sun T, Zhou Y, Zhang X, Li H, Lin D:

Significant increase in risk of gastroesophageal cancer is associated with interaction between pro-

moter polymorphisms in thymidylate synthase and serum folate status. Carcinogenesis 2005;26:

1430–1435.

19 Wang L, Miao X, Tan W, Lu X, Zhao P, Zhao X, Shan Y, Li H, Lin D: Genetic polymorphisms in

methylenetetrahydrofolate reductase and thymidylate synthase and risk of pancreatic cancer. Clin

Gastroenterol Hepatol 2005;3:743–751.

Dr. Dongxin Lin

Department of Etiology and Carcinogenesis, Chinese Academy of Medical Sciences

Cancer Hospital/Institute, 17 Panjiayuan Nanli, Caoyang District

Beijing 100021 (China)




Dietary Quercetin Inhibits Proliferation of Lung Carcinoma Cells

Huynh Hung

Laboratory of Molecular Endocrinology, Division of Cellular and Molecular

Research, National Cancer Centre, Singapore, Singapore

AbstractRegular consumption of fruits and vegetables is strongly associated with reduced risk of

developing chronic diseases. It is estimated that one third of all cancer deaths in the USA could

be avoided through appropriate dietary modification. Several studies have indicated that fruits,

vegetables and whole grains contain significant amounts of bioactive phytochemicals that have

antiproliferative and antineoplastic properties. The bioactive phytochemicals may help protect

cellular systems from oxidative damage as well as reduce the risk of chronic diseases. Quercetin

and other related flavonoids have been shown to inhibit carcinogen-induced tumors in rodents.

In humans, the total average intake of quercetin and kaempferol is estimated at 20 mg/day and

consumption of quercetin from onions and apples was inversely correlated with lung cancer

risk. In this study, we report that quercetin-inhibited A549 lung carcinoma cell proliferation

was associated with activation of the extracellular signal-regulated kinase (ERK). Inhibition of

MEK1/2 but not PI3 kinase, p38 kinase or JNK abolished quercetin-induced apoptosis sug-

gesting MEK-ERK activation was required to trigger apoptosis.


The relationship between diet and cancers has been implicated in several

epidemiological studies [Block et al., 1992]. The cancer incidence is signifi-

cantly lower in people whose diets consist largely of fruits and vegetables to

those whose diets consist mainly of animal products [Block et al., 1992]. The

results from several studies indicate that vegetables and fruits contain large

amounts of bioactive phytochemicals that may help protect cellular systems

from oxidative damage as well as lower the risk of chronic diseases [Leighton et al.,

1992; Messina et al., 1994].

The current study of nutrient-modulated carcinogenesis involves exploring

the effects of flavonoids on target receptors and signal transduction pathways;

Quercetin and Lung Cancer 147

cell cycle control and checkpoint; apoptosis and antiangiogenic processes.

Flavonoids have been found to arrest cell cycle progression either at G1/S or at

G2/M boundaries (reviewed in Casagrande and Darbon [2001]). However, the

precise mechanism responsible for the cell cycle effect of these compounds is

not clearly understood yet. It is proposed that the 3�-OH of the phenyl ring

might be important for the level at which the cell cycle arrests. The presence of

the 3�-OH in quercetin and luteolin correlates with a G1 block while its absence

in kaempferol and apigenin correlates with a G2 block [Casagrande and

Darbon, 2001]. Flavonoids that upregulate both p21CIP1 and p27KIP1 (quercetin,

luteolin and daidzein) lead to G1 arrest. By contrast, the flavonoids which

poorly upregulate p27KIP1 and not p21CIP1 (kaempferol and apigenin) or which

induce only p21CIP1 (genistein) are unable to arrest cells in G1 [Casagrande and

Darbon, 2001].

The most common flavonoids found in the diet are quercetin, kaempferol,

rutin and robinin [Anton, 1988]. The total average intake of quercetin and

kaempferol is estimated at 16 mg and 4 mg/day, respectively. In the gastroin-

testinal tract, robinin is hydrolyzed to kaempferol by the �-glucosidase activity

of microorganisms [Bokkenheuser and Winter, 1988]. Among the dietary

flavonoids, quercetin has been studied extensively [Aligiannis et al., 2001;

Constantinou et al., 1995; Lee et al., 1998]. In addition, consumption of

quercetin from onions and apples is inversely correlated with lung cancer risk

[Le Marchand et al., 2000].

In the present study, we report that quercetin inhibits human A549 lung

cancer cell proliferation and induces apoptosis. In addition to the inhibition of

Akt activation and upregulation of Bax and Bad, extracellular signal-regulated

kinase (ERK) activation plays an important role in mediating quercetin-induced

apoptosis in A549 cells and ERK functions upstream of the caspase activation

to initiate the apoptotic signal.

Materials and Methods

ReagentsU0126, LY294002 and antibodies against phospho-MEK1/2 (Ser217/221), cleaved cas-

pase 7 (20 kDa), caspase 3, phospho-Akt (Ser473), phospho-p44/42 ERK (Thr202/Tyr204),

anti-Akt, anti-ERK1 and cleaved poly(ADP-ribose) polymerase (PARP) were from New

England Biolabs, Beverly, Mass., USA. Antibodies against Bax, phospho-c-Jun (Ser63),

phospho-JNK (Thr183/Tyr185), �-tubulin, Bcl-2, Bcl-xL, and Bad were obtained from Santa

Cruz Inc., Santa Cruz, Calif., USA. Horseradish peroxidase-conjugated secondary antibod-

ies were purchased from Pierce, Rockford, Ill., USA. Chemiluminescent detection system

was supplied by Amersham, Pharmacia Biotech, Arlington Heights, Ill., USA. Disposal tis-

sue culture plates and dishes were purchased from Nunc Inc., Naperville, Ill., USA. Quercetin

and 3-(4,5-dimethylthiazol-2-yl)-2,5-diphenyltetrazolium bromide (MTT) was purchased

Hung 148

from Sigma, Saint Louis, Mo., USA. Cell proliferation ELISA and in situ cell death detec-

tion kits were supplied by Roche Diagnostics Corporation, Indianapolis, Ind., USA. RPMI-

1640 medium, fetal bovine serum, TRIzol and penicillin-streptomycin were from Gibco-BRL,

Grand Island, N.Y., USA.

Cell Proliferation and Thymidine Incorporation AssaysHuman A549 lung epithelial cells were obtained from American Type Culture Collection

(Rockville, Md., USA). Cells were treated with indicated concentrations of quercetin in min-

imum essential medium. Hemocytometric counts and thymidine incorporation of triplicate

cultures were performed as previously described [Huynh et al., 1996]. Cell proliferation and

cell viability were determined at 24 and 48 h after treatment using the cell proliferation

ELISA kit and the MTT assay, respectively, as described in Huynh et al. [1996]. Experiments

were repeated at least 3 times, and the data were expressed as means � SD.

aa

bb

c c

dd e

e

00

0.2

0.4

Rat

e of

Brd

U

inco

rpor

atio

n (4

50nm

)

0.6

0.8

1.0

1.2

29

Quercetin (�M/ml)

14.5 43.5 58

24 h48 h

a

a

ab

b c

c d c d

00

0.3

Ab

sorb

ance

(570

nm)

0.6

0.9

1.2

29

Quercetin (�M/ml)

14.5 43.5 58

a

b

Fig. 1. Effects of quercetin on the viability and proliferation of A549 cells. Cells were

grown and treated with serum-free RPMI-1640 medium containing either 0.1% DMSO or indi-

cated doses of quercetin for 24 and 48 h as described in Materials and Methods. Cell prolifera-

tion (a) and cell viability (b) were determined by BrdU incorporation and MTT assay,

respectively. Bars with different letters are significantly different from one another at p � 0.01.


Western Blot AnalysisTo determine the changes in the expression of indicated proteins, cells were grown and

treated with indicated concentrations of quercetin. An equal amount of proteins (100 �g per

sample) was subjected to Western blot analysis as described in Huynh et al. [2002]. Blots

were visualized with a chemiluminescent detection system (Amersham) as described by the

manufacturer.

Control Quercetin (14.5�M)

Quercetin (29�M)

60

50

40

30

20

10

0

Rat

e of

ap

opto

sis

(%)

Quercetin (58�M)

a

0 14.5 29

Quercetin (�M)

43.5 58

b

c

d

d

b

a

Fig. 2. Induction of apoptosis by quercetin in A549 cells. Cells were grown and treated

with escalating doses of quercetin in SRF medium for 24 h. Apoptotic cells were determined

by the TUNEL assay as described in Materials and Methods. a Apoptotic cells were visual-

ized under a fluorescent microscope. b The rate of apoptosis expressed as a percentage of

total cells counted is shown. Bars with different letters are significantly different from one

another at p � 0.01 as determined by the Kruskal-Wallis test. Experiments were repeated

3 times with similar results.

Hung 150

Detection of ApoptosisA549 cells were plated onto 8-chamber slides at a density of 5 � 103 cells per well and

treated with indicated concentrations of quercetin for 48 h. Apoptosis was detected by the ter-

minal deoxynucleotidyl transferase-mediated dUTP nick-end labelling (TUNEL) assay

using the in situ cell death detection kit (Roche) as described by the manufacturer. Slides

were visualized with a laser confocal microscope (Zeiss) equipped with epifluorescence

optics and appropriate filters for FITC. Labelling indices were obtained by counting the

number of labelled cells among at least 500 cells per region expressed as a percentage value.

Results

For the time course and dose-response experiments, human A549 lung

cancer cells were treated with different concentrations of quercetin for 24 and

48 h. Figure 1 shows that quercetin caused a dose-dependent reduction in DNA

synthesis (fig. 1a) and cell viability (fig. 1b) (p � 0.01). A 50% reduction in

cell viability was seen at a dose of 29.0 �M after 48 h of incubation (fig. 1b).

Fold changed (cleaved caspase 3)

�-Tubulin

19kDa17kDa

1 5 13 13.5 8

1 1.7 30.4 38.4 37.8

1 4.27 8.2 13.9 10.7

0 14.5 29 43.5 58

19kDa

89kDa

20kDa

Caspase 3

Cleaved caspase 7

Fold changed (cleaved caspase 7)

Fold changed (cleaved PARP)

Quercetin (�M/ml)

Cleaved PARP

a

b

c

d

e

Fig. 3. Effects of quercetin on the levels of Bcl-2, Bax, Bad and Bcl-xL in A549 cells.

Cells were cultured as described in Materials and Methods. Cells were treated with vehicle or

indicated concentrations of quercetin for 24 h. Cells were harvested and lysed for Western

blot analysis as described in Materials and Methods. Blots were incubated with indicated

antibodies. Changes in the levels of Bax, Bad, Bcl-2 and Bcl-xL proteins are shown below

each blot. Experiments were repeated 3 times with similar results.


DNA fragmentation (fig. 2a) and a dose-dependent increase in apoptotic

cells (fig. 2b) were observed in quercetin-treated cells. As shown in figure 3,

quercetin induced a significant elevation in the expression of proapoptotic Bax

and Bad (p � 0.01). While Bcl-2 expression was slightly decreased, the anti-

apoptotic Bcl-xL expression was significantly increased by quercetin (p � 0.01)

(fig. 3d). An 89-kDa cleaved PARP fragment was detected in quercetin-treated

samples (fig. 4e). Figure 4c and d shows that the cleaved forms of caspase 3 and

7 fragments were readily detectable at a dose as low as 14.5 �M of quercetin and

reached high levels at a dose of 29.0 �M. Figure 5 shows that quercetin

decreased in total antiapoptotic Akt protein (fig. 5d) and its basal phosphoryla-

tion (fig. 5c). Treatment of A549 cells with quercetin also led to a dose-dependent

activation of MEK1/2 (fig. 6b) and ERK1/2 (fig. 6d).

To determine whether quercetin-induced apoptosis was mediated by the

activation of ERK, A549 cells were treated with MEK inhibitor to suppress

quercetin-induced ERK activation and its downstream effects. The TUNEL assay

showed that quercetin and combined quercetin-LY294002 caused apoptosis in

�-Tubulin

1 1.05 2.2 2.5 3.54

1 1.05 2.05 2.15 2.3

1 2 2.3 2.4 2.55

1

0

0.86 0.78 0.58 0.72

14.5 29 43.5 58

Bax

Bad

Fold changed (Bax)

Fold changed (Bad)

Bcl-xL

Fold changed (Bcl-xL)

Bcl-2

Fold changed (Bcl-2)

Quercetin (�M/ml)

a

b

c

d

e

Fig. 4. Effects of quercetin on the cleavage of caspase 3, caspase 7 and PARP in A549

cells. Cells were cultured and treated with quercetin as described in Materials and Methods.

Cells were harvested and lysed for Western blot analysis as described in Materials and

Methods. Blots were incubated with indicated antibodies. Changes in the levels of cleaved

caspase 3, cleaved caspase 7 and cleaved PARP are shown below each blot. Experiments

were repeated 3 times with similar results.

Hung 152

A549 cells (fig. 7d, f). Co-treatment of A549 cells with quercetin and U0126

completely blocked quercetin-induced apoptosis (fig. 7e). Co-treatment of cells

with quercetin and U0126 prevented quercetin-induced phosphorylation of

ERK, phosphorylation of c-Jun, and cleavage of caspase 3, caspase 7 and PARP

(fig. 8). Blocking PI3 kinase activity neither enhanced nor prevented quercetin-

induced apoptosis, and cleavage of caspase 3, caspase 7 and PARP (fig. 8). The

results suggest that activation of MEK-ERK plays an important role in

quercetin-induced apoptosis and ERK acts upstream of caspase 3 and caspase 7

to exert its apoptotic influence in the quercetin-treated A549 cells.

Discussion

In the present study, we have shown that quercetin inhibits human A549

lung cancer cell proliferation and induces apoptosis. In addition to the inhibi-

tion of Akt-1 phosphorylation, sustained activation of ERK is required for

quercetin-induced apoptosis in A549 cells. Quercetin treatment results in dose-

and time-dependent activation of ERK. That elevated ERK activity contributes

�-Tubulin

PI3 kinase

pAkt

Akt

Fold changed (pAkt) 1 0.57 0.16 0.13 0.07

Fold changed (Akt) 1 0.7 0.56 0.52 0.42

Quercetin (�M/ml) 0 14.5 29 43.5 58

a

b

c

d

Fig. 5. Effects of quercetin on the basal levels of p85 subunit of PI3 kinase, Akt-1 and

phosphorylated Akt (Ser473) in A549 cells. Cells were cultured and treated with quercetin as

described in Materials and Methods. Cells were harvested and lysed for Western blot analy-

sis as described in Materials and Methods. Blots were incubated with indicated antibodies.

Changes in the levels of Akt-1 and phospho-Akt-1 are shown below each blot. Experiments

were repeated 3 times with similar results.


to apoptosis by quercetin is supported by the observations that activation of

ERK by expression of activated MEK1 induces apoptosis while inhibition of

ERK by MEK inhibitors blocks quercetin-induced cell death (data not shown).

Quercetin-induced apoptosis is associated with cleavage of caspase 3, caspase 7

and PARP all of which can be reduced by treatment of A549 cells with the

MEK1/2 inhibitor. Our findings suggest that in addition to the inhibition of Akt

activation, ERK activation plays an important role in mediating quercetin-

induced apoptosis in A549 cells and ERK functions upstream of the caspase

activation to initiate the apoptotic signal.

In the present report, we observe that quercetin inhibits Akt expression and

Akt phosphorylation. Because Akt is a downstream target of PI3 kinase, the

�-Tubulin

pMEK1

MEK1

pMEK2

MAPK

Fold changed (pMEK1)

Fold changed (pMAPK)

Fold changed (p-c-Jun)

pJNK 50kDa

pJNK 40kDa

1 2.17 5.1 5.27 11.8

Fold changed (pMEK2) 1 0.5 0.4 0.09 0.05

Fold changed (pJNK 50kDa) 1 1.35 1.6 2.4 3.1

Fold changed (pJNK 40kDa) 1 1.17 1.3 1.4 1.3

Quercetin (�M/ml) 0 14.5 29 43.5 58

1 1.1 3.9 4.17 5.7

1 2.8 3.4 3.3 3.5

a

b

c

d

e

f

g

pMAPK

p-c-Jun

Fig. 6. Effects of quercetin on the levels of MEK1, ERK, and phosphorylated MEK1/2

(Ser217/221), phosphorylated ERK (Thr202/Tyr204), phosphorylated JNK (Thr183/

Tyr185), and phosphorylated c-Jun (Ser63) in A549 cells. Cells were cultured and treated

with various concentrations of quercetin as described in Materials and Methods. Cells were

harvested and lysed for Western blot analysis as described in Materials and Methods. Blots

were incubated with indicated antibodies. Changes in the levels of indicated proteins are

shown below each blot. Experiments were repeated 3 times with similar results.

Hung 154

observed inhibition of Akt phosphorylation suggests that quercetin also inhibits

PI3 kinase. This argument is supported by previous studies showing that

quercetin is an inhibitor of PI3 kinase and serine/threonine protein kinases

[Agullo et al., 1997; Gamet-Payrastre et al., 1999; Hagiwara et al., 1988; Yoshizumi

et al., 2001]. By suppressing the activation of Akt-1, quercetin can promote

apoptosis via several pathways. Inactivation of Akt would prevent Akt-1 from

phosphorylating Bad on serine 136. As a result, Bad binds to Bcl-2, and its

proapoptotic activity is effectively increased in the death regulation equation.

a

b

c

d

e

f

Fig. 7. Effects of MEK1/2 inhibitor U0126 and PI3 kinase inhibitor LY294002 on

quercetin-induced apoptosis in A549 cells. Cells were grown and treated with a vehicle (a),

10.0 �M of U0126 (b), 10.0 �M of LY294002 (c), 58.0 �M of quercetin (d), 58.0 �M of

quercetin plus 10.0 �M of U0126 (e), and 58.0 �M of quercetin plus 10.0 �M of LY294002 (f)for 24 h. Cells were subjected to the TUNEL assay as described in Materials and Methods.

Original magnification, �200.


In this study, we have provided evidence that activation of MEK-ERK

plays a dominant role in quercetin-induced activation of caspase 3 and 7 which

is necessary for the cleavage of PARP and apoptosis in A549 lung cancer cells.

Quercetin treatment results in high and sustained activation of ERK in A549

cells. One important difference between the quercetin- and IGF-1-induced ERK

activation is the time and duration of activity (data not shown). In the case of

IGF-1, ERK activation is rapid, occurring within minutes of treatment, and

transient. With quercetin, significant activation occurs at 3 h, but the activity

remains highly elevated throughout the experiment (up to 24 h). Utilizing

U0126 to modulate ERK activity, we find that inhibition of MEK-ERK activa-

tion abolishes quercetin-induced apoptosis. Our results are similar to other

studies which demonstrate that abrogation of the ERK pathway by Taxol delays

or fails to prevent Taxol-induced apoptosis [Kalechman et al., 2000; Lieu et al.,

1998]. However, it is not a universal feature of mammalian cells as activation of

the MEK-ERK pathway has been shown to contribute to cell proliferation and

survival [Ballif and Blenis, 2001], migration [Krueger et al., 2001] and transfor-

mation [Montesano et al., 1999]. Furthermore, inhibition of stress-induced signalling

via the MEK-ERK pathway increases the toxic effects of chemotherapeutic drugs

and irradiation [Yano et al., 1992]. Therefore, the ability of the MEK-ERK

�-Tubulin

Quercetin (58�M)

LY294002 (10�M)U0126 (10�M)

� � �

� � � �

� � � �

a

b

c

d

e

f

g

Phospho-MAPK

MAPK

Phospho c-Jun

Cleaved caspase 7

Cleaved PARP

Caspase 3

20kDa19kDa

89kDa

Fig. 8. Effects of MEK1/2 inhibitor U0126 and PI3 kinase inhibitor LY294002 on

quercetin-induced phosphorylation of ERK, c-Jun and cleavage of caspase 3, caspase 7 and

PARP in lung carcinoma cells. A549 cells were grown and treated with SRF medium con-

taining 0.1% DMSO and indicated concentrations of drugs for 24 h. Cells were harvested and

lysed for Western blot analysis as described in Materials and Methods. Blots were incubated

with indicated antibodies. Experiments were repeated 3 times with similar results.

Hung 156

pathway to regulate proliferation versus survival appears to depend on cell

types and the amplitude and duration of ERK activation. A short activation of

the MEK-ERK cascade by growth factors such as IGF-1 is associated with pro-

liferation while prolonged activation of ERK activity inhibits DNA synthesis.

Our data suggest that consumption of diets containing high levels of

quercetin may help to reduce the risk of and/or prevent lung cancer.

Acknowledgements

This work was supported by grants from the National Medical Research Council of

Singapore (NMRC/0541/2001), A*STAR-BMRC (LS/00/017) and A*STAR-BMRC (LS/00/

019) to Huynh Hung.

References

Agullo G, Gamet-Payrastre L, Manenti S, Viala C, Remesy C, Chap H, Payrastre B: Relationship

between flavonoid structure and inhibition of phosphatidylinositol 3-kinase: a comparison with

tyrosine kinase and protein kinase C inhibition. Biochem Pharmacol 1997;53:1649–1657.

Aligiannis N, Mitaku S, Mitrocotsa D, Leclerc S: Flavonoids as cycline-dependent kinase inhibitors:

inhibition of cdc 25 phosphatase activity by flavonoids belonging to the quercetin and kaempferol

series. Planta Med 2001;67:468–470.

Anton R: Flavonoids and Traditional Medicine. New York, Liss, 1988.

Ballif BA, Blenis J: Molecular mechanisms mediating mammalian MEK-MAP kinase cell survival sig-

nals. Cell Growth Differ 2001;12:397–408.

Block G, Patterson B, Subar A: Fruit, vegetables, and cancer prevention: a review of the epidemiological

evidence. Nutr Cancer 1992;18:1–29.

Bokkenheuser VD, Winter J: Hydrolysis of flavonoids by human intestinal bacteria; in Cody V,

Middleton E, Harbourne JB, Beretz A (eds): Plant Flavonoids in Biology and Medicine II. New

York, Alan R Liss, Inc, 1988.

Casagrande F, Darbon JM: Effects of structurally related flavonoids on cell cycle progression of human

melanoma cells: regulation of cyclin-dependent kinases CDK2 and CDK1. Biochem Pharmacol

2001;61:1205–1215.

Constantinou A, Mehta R, Runyan C, Rao K, Vaughan A, Moon R: Flavonoids as DNA topoisomerase

antagonists and poisons: structure-activity relationships. J Nat Prod 1995;58:217–225.

Gamet-Payrastre L, Manenti S, Gratacap MP, Tulliez J, Chap H, Payrastre B: Flavonoids and the inhibi-

tion of PKC and PI 3-kinase. Gen Pharmacol 1999;32:279–286.

Hagiwara M, Inoue S, Tanaka T, Nunoki K, Ito M, Hidaka H: Differential effects of flavonoids as

inhibitors of tyrosine protein kinases and serine/threonine protein kinases. Biochem Pharmacol

1988;37:2987–2992.

Huynh H, Chow PK, Ooi LL, Soo KC: A possible role for insulin-like growth factor-binding protein-3

autocrine/paracrine loops in controlling hepatocellular carcinoma cell proliferation. Cell Growth

Differ 2002;13:115–122.

Huynh H, Yang X, Pollak M: Estradiol and antiestrogens regulate a growth inhibitory insulin-like growth

factor binding protein 3 autocrine loop in human breast cancer cells. J Biol Chem 1996;271:

1016–1021.

Kalechman Y, Longo DL, Catane R, Shani A, Albeck M, Sredni B: Synergistic anti-tumoral effect of

paclitaxel (Taxol)AS101 in a murine model of B16 melanoma: association with ras-dependent

signal-transduction pathways. Int J Cancer 2000;86:281–288.


Krueger JS, Keshamouni VG, Atanaskova N, Reddy KB: Temporal and quantitative regulation of

mitogen-activated protein kinase (MAPK) modulates cell motility and invasion. Oncogene 2001;

20:4209–4218.

Lee SC, Kuan CY, Yang CC, Yang SD: Bioflavonoids commonly and potently induce tyrosine dephos-

phorylation/inactivation of oncogenic proline-directed protein kinase FA in human prostate carci-

noma cells. Anticancer Res 1998;18:1117–1121.

Leighton T, Ginther C, Fluss L, Harter WK, Cansado J, Notario V: Phenolic compounds in food and their

effects on health II. Washington, American Chemical Society, 1992, pp 229–238.

Le Marchand L, Murphy SP, Hankin JH, Wilkens LR, Kolonel LN: Intake of flavonoids and lung cancer.

J Natl Cancer Inst 2000;92:154–160.

Lieu CH, Liu CC, Yu TH, Chen KD, Chang YN, Lai YK: Role of mitogen-activated protein kinase in

taxol-induced apoptosis in human leukemic U937 cells. Cell Growth Differ 1998;9:767–776.

Messina MJ, Persky V, Setchell KD, Barnes S: Soy intake and cancer risk: a review of the in vitro and in

vivo data. Nutr Cancer 1994;21:113–131.

Montesano R, Soriano JV, Hosseini G, Pepper MS, Schramek H: Constitutively active mitogen-activated

protein kinase kinase MEK1 disrupts morphogenesis and induces an invasive phenotype in

Madin-Darby canine kidney epithelial cells. Cell Growth Differ 1999;10:317–332.

Yano T, Pinski J, Groot K, et al: Stimulation by bombesin and inhibition by bombesin/GRP antagonist

RC-3905 of growth of human breast cancer cell lines 1123. Cancer Res 1992;52:4545–4547.

Yoshizumi M, Tsuchiya K, Kirima K, Kyaw M, Suzaki Y, Tamaki T: Quercetin inhibits Shc- and phos-

phatidylinositol 3-kinase-mediated c-Jun N-terminal kinase activation by angiotensin II in cul-

tured rat aortic smooth muscle cells. Mol Pharmacol 2001;60:656–665.

Dr. Huynh Hung

Laboratory of Molecular Endocrinology

Division of Cellular and Molecular Research

National Cancer Centre of Singapore

Singapore 169610 (Singapore)

Tel. 65 436 8347, Fax 65 226 5694, E-Mail [email protected]



Osteoporosis: The Role of Genetics and the Environment

Boonsong Ongphiphadhanakul

Department of Medicine, Faculty of Medicine, Ramathibodi Hospital,

Mahidol University, Bangkok, Thailand

AbstractOsteoporosis is partly genetically determined. The genetics of osteoporosis is poly-

genic in nature with multiple common polymorphic alleles interacting with each other and

environmental factors to determine bone mass. A number of studies have attempted to dissect

the genetic factors responsible for the pathogenesis of osteoporosis using genome-wide

scanning and the candidate gene approach. However, the results of such studies among differ-

ent populations have been mostly inconsistent, suggesting genetic heterogeneity of osteoporo-

sis. It is likely that the cohort of genes indicating predisposition to the risk of osteoporosis

may be different among populations with different ethnic backgrounds. The successful iden-

tification of susceptibility genes for osteoporosis should prove to be helpful in targeting pre-

ventive and therapeutic measures to individuals at higher risk and to render the effort more

cost-effective. Information with regard to genetic variations is also likely to be useful in tar-

geting preventive or therapeutic measures to subjects genetically determined to have better

responsiveness. Intestinal calcium absorption is dependent on vitamin D receptor gene poly-

morphisms. Skeletal responsiveness to estrogen, particularly at lower doses, is related to

polymorphisms in the estrogen receptor-� gene. Recently, circulating homocysteine levels

have been shown to be associated with fracture risk. Folate and vitamin B supplements for

reducing serum homocysteine and fractures in postmenopausal women have not been fully

investigated. However, there is an interaction between folate status and methylenetetrahydro-

folate reductase gene polymorphism on bone phenotypes. Due to recent technological

advances, whole-genome association study is becoming more feasible. Genomic information

with regard to the susceptibility to osteoporosis and the responsiveness to preventive or ther-

apeutic modalities should supplement rather than replace conventional clinical information.

Clinical decision should also take into account the social, health and economic perspectives

in order to balance the benefit of novel clinical strategies against the associated risks and

available resources.


Osteoporosis, Genetics and Environment 159

Genetic Determinants of Osteoporosis

The risk of osteoporosis is partly genetically determined. Studies per-

formed in families demonstrated a resemblance in bone mineral density (BMD)

between child-parent pairs [1, 2]. Moreover, studies carried out in twins also

revealed more resemblance in bone mass among monozygotic twins than dizy-

gotic twins [3, 4]. In multigenerational pedigree studies, a genetic contribution

to bone mass was also demonstrated [5]. The heritability of bone mass is esti-

mated to be 60–80% and its influence can be demonstrated as early as before

puberty [6]. The skeletal quantitative phenotypes for osteoporosis do not con-

form to a simple monogenic model and the genetics of osteoporosis is poly-

genic in nature with multiple common polymorphic alleles interacting with

each other and environmental factors to determine the quantitative bone pheno-

types [7, 8]. A number of studies have attempted to dissect the genetic factors

responsible for the pathogenesis of osteoporosis using genome-wide scanning

and the candidate gene approach. For genome scans, the results of studies in

different populations, however, are still inconsistent, which suggests that osteo-

porosis may be genetically heterogeneous [9, 10]. Association studies which

have examined the association between variants at candidate genetic loci and

osteoporosis have investigated a number of genes in various populations including

genes encoding the �1-chain of type 1 collagen (COLIA1), vitamin D receptor

(VDR), estrogen receptor-� (ESR1) and others.

Role of the VDR GeneVDR was the first gene identified to be associated with an osteoporosis-

related phenotype. The VDR gene contains 11 exons and spans approximately

75 kb. There are at least 3 single nucleotide polymorphisms (SNPs) at the 3� end

of the VDR gene which have originally been studied in relation to osteoporosis.

The SNPs can be identified by BsmI, ApaI and TaqI restriction endonucleases,

respectively. In 1992, it was reported that these SNPs, which are in linkage

disequilibrium, are associated with BMD. The relation is such that the B allele

which denotes the absence of the BsmI restriction site is associated with lower

bone mass. Numerous studies then followed with inconsistent results. According

to meta-analyses, the effect of the VDR gene on osteoporosis appears to be pos-

itive but the magnitude is rather small. Between genotypes, BMD differs by

0.15–0.2 Z score units. Moreover, the effect of the VDR gene is also dependent

on age and menopausal status [11, 12].

Role of the COLIA1 GeneType 1 collagen is one of the major proteins in bone. The type 1 collagen

gene has been investigated as a susceptibility gene for osteoporosis. In 1996,

Ongphiphadhanakul 160

Grant et al. [13] reported that a G-to-T SNP in the promoter of the gene encod-

ing the COLIA1 gene which affects the binding site to the transcription factor

Sp1 is associated with BMD. BMD is higher in the presence of the G allele and

a dose-related relationship is also apparent. Besides BMD, the SNP is also asso-

ciated with osteoporotic fractures. The relationship between the COLIA1 gene

and osteoporosis has been investigated in meta-analyses which suggested that

the COLIA1 polymorphism is associated with BMD [14, 15]. Nevertheless, the

effect is small being 0.15 Z scores per T allele. On the other hand, the effect of

the SNP on fractures is out of proportion to that on BMD and it is likely that the

polymorphism may be more related to bone quality rather than quantity.

Therefore, the increase in the risk of fractures is mainly due to the impairment

in bone quality in subjects with the T allele.

Role of the ESR1 GeneSince estrogen deficiency is the main cause of osteoporosis and the action

of estrogen is mediated through estrogen receptors, a number of studies have

also investigated the relation between the ESR1 gene and osteoporosis. Two of

the more frequently studied SNPs are located in intron 1 of the ESR1 gene

which can be detected by PvuII and XbaI restriction endonucleases, respec-

tively. The results of studies on the ESR1 gene and osteoporosis are inconsis-

tent. However, a meta-analysis showed that the XX allele according to the

detection by XbaI restriction endonuclease was related to less BMD and more

osteoporotic fractures [16]. An effect of the P allele as assessed by PvuII restric-

tion endonuclease was not found.

Role of the Low-Density Lipoprotein Receptor-Related Protein 5 GeneBesides the candidate gene approach, genome scan has also been utilized

to locate genes for osteoporosis without a priori knowledge of the function of

the genes involved. Chromosomal regions which have been found to be related

to bone mass include those on chromosomes 1, 2, 4, 5, 6, and 11. The impor-

tance of the chromosomal region 11q12–13 has been substantiated by studies of

2 rare monogenic bone disorders, namely autosomal dominant high bone mass

and autosomal recessive osteoporosis pseudoglioma syndrome, which were also

mapped to chromosomal region 11q12. The gene responsible in both disorders is

the low-density lipoprotein receptor-related protein 5 (LRP5) gene which was

previously unknown to play a role in bone metabolism. With regard to the rela-

tion between the LRP5 gene and osteoporosis, Ferrari et al. [17] studied the

associations of genetic variants at the LRP5 locus and bone mass in adults, ado-

lescents and children. A number of polymorphisms in LRP5 were found and

associations demonstrated between these variants and peak bone mass as well


as bone size. Subsequent studies also demonstrated a relation between various

SNPs in the LRP5 gene and bone mass in Caucasians [18, 19] as well as in

Asians [20, 21], but not without dispute [22].

Limitations of Studies on the Genetics of OsteoporosisAlthough association study has higher power than conventional linkage

analysis, it can be flawed by population admixture [23]. Using the transmission

disequilibrium test to test linkage and association could be an alternative and

robust approach [24]. Moreover, most studies regarding genetic susceptibility

to osteoporosis have only examined the relation between genetic variations and

bone mass. The genetic correlation between the susceptibility to osteoporotic

fractures and bone mass variation can be low [25] and genetic susceptibility to

low bone mass cannot be readily generalized to osteoporotic fractures. To date,

the COLIA1 gene polymorphism appears to be more consistently related to

BMD and osteoporotic fractures [13, 26, 27] and the functionality of the

polymorphism has been suggested [28]. However, there appear to be mar-

ked differences in allele frequency of the COLIA1 polymorphism in populations

with different ethnicity. In particular, studies performed in Asian populations

revealed that the COLIA1 polymorphism is almost nonexistent which under-

mines its role, if any, in the risk assessment of osteoporosis in Asians [29–31].

Taken together, it is likely that the cohort of genes predisposing to the risk

of osteoporosis can be different among populations with different ethnic

backgrounds.

Genetic Determinants of Responsiveness in Bone-Related Phenotypes

Responsiveness to CalciumCalcium is an essential nutrient for bone health. Epidemiological studies

have demonstrated that calcium intake affects the risk factors for osteoporotic

fractures. Nevertheless, the effect of calcium on bone mass in young adults is

dependent on the duration and continuity of increased calcium intake as well as

possibly the reproductive hormonal milieu. A short-term study demonstrated

increased BMD in prepubertal subjects [32] while another study revealed a ben-

eficial effect only after menarche [33]. Moreover, the beneficial effect of sup-

plemental calcium on bone rapidly dissipated within 1–2 years of stopping the

calcium supplements [34]. On the other hand, long-term calcium supplementa-

tion appears to exert a minimal effect on bone mass. A 7-year study in adoles-

cents showed that the gain in bone mass during the first 4 years of calcium


supplementation diminished at the end of the study at 7 years [35]. In a meta-

analysis of cross-sectional studies, it was found that in young and middle-aged

females, there is a correlation between bone mass and calcium intake. Moreover,

it was observed in a meta-analysis of longitudinal studies that increased cal-

cium intake can prevent 1% bone loss per year [36]. Despite its possible benefi-

cial effects on bone mass in children and young adults, the role of calcium alone

in the prevention of osteoporotic fractures in postmenopausal women is less

clear. Calcium supplementation in women within 5 years postmenopausally

does not appear to possess a beneficial effect on bone mass. Nevertheless, in

elderly women with osteoporosis, calcium supplementation tends to decrease

vertebral fractures [37].

Intestinal absorption of calcium appears to be influenced by genetic fac-

tors, particularly VDR gene polymorphism. The increase in intestinal calcium

absorption during low calcium intake as a normal physiologic adaptation dis-

appears in subjects with the BB genotype but not in subjects with the bb geno-

type [38]. Despite the difference in calcium absorption, the influence of VDRpolymorphisms on changes in bone mass after calcium supplementation has

not been studied, although the effect of calcium or calcitriol in preventing

osteoporotic fractures has been found to depend on VDR gene polymorphisms

[39].

Responsiveness to EstrogenSince estrogen exerts its effect mainly through estrogen receptors, it is

therefore not surprising that ESR1 has been investigated as a target causing

variation in individual response to estrogen. Despite the inconsistency in the

relation between ESR1 polymorphisms and bone mass, there are relatively more

consistent evidences regarding the relation between these polymorphisms and

skeletal responsiveness to estrogen. For example, polymorphisms in ESR1 have

been demonstrated to be associated with the protective effects of estrogen on

BMD [40, 41] and fracture risk [42], although not without dispute [43]. Part of

the reasons for the inconsistent results may be that the influence of ESR1 poly-

morphism on skeletal responsiveness can be more apparent at a lower dose of

estrogen [41]. Despite the association between ESR1 polymorphisms and

response in BMD, the effect on the fracture rate is as yet unknown. Similar to

the studies investigating ESR1 polymorphisms and BMD, the more often stud-

ied polymorphisms are the polymorphic sites identifiable by PvuII or XbaI

restriction endonucleases in intron 1 which may not be directly involved in the

physiology of bone responsiveness. Besides skeletal responsiveness to estrogen,

ESR1 has also been found to influence the increase in high-density lipoprotein

cholesterol after estrogen [44], which further supports the role of nucleotide

variations in the ESR1 gene in tissue responsiveness to estrogen.


Homocysteine and OsteoporosisRecently, it has been found that homocysteine is a novel risk factor for

osteoporotic fractures. It is well established that patients with homocystinuria, a

rare autosomal recessive disorder with markedly elevated circulating homocys-

teine levels, have generalized osteoporosis even at young age. The reason for

early osteoporosis is unclear, but may be related to impairment in collagen

cross-linking. In elderly subjects, 2 independent studies reported concurrently

that high homocysteine levels were associated with fractures. In one of the stud-

ies (from the Netherlands), subjects in the highest quartiles of homocysteine

levels had twice the risk of a nonvertebral fracture compared to those in the

other quartiles [45]. The other study (from the Framingham Study cohort in the

USA) demonstrated that men in the highest quartiles had four times the risk of

hip fractures while in women the risk increased by a factor of 2 relative to those

in the lowest quartile of homocysteine levels [46]. Despite the associations, a

causal relation between homocysteine levels and fracture is less clear. However,

more recently, a prospective study revealed that the risk of fractures was

reduced by half in patients with stroke after folate and mecobalamin supple-

mentation [47]. Nevertheless, the factors more directly responsible for the

reduction in fracture risk are homocysteine or folate and the B vitamin status

cannot be clearly determined.

There is a common allelic polymorphism C677T in the methylenete-

trahydrofolate reductase (MTHFR) gene with changes in the amino acid at

position 222 from alanine to valine. The substitution results in an enzyme

which is more thermolabile and causes higher homocysteine and lower folate

levels. The polymorphism has been found to be associated, albeit inconsis-

tently, with BMD or fractures in a number of studies [48–50]. The inconsis-

tency may be explained by the possible interaction between the MTHFRgenotype and the intake of B vitamins and folate. Subjects with low folate

levels who had the TT genotype tended to have lower broadband ultrasound

attenuation and lower BMD at the Ward’s triangle. Subjects with higher folate

levels and with the TT genotype had higher, rather than lower, hip BMD [51].

In a longitudinal study with a mean of 6.6 years of follow-up, there was no

relation between the MTHFR genotype and BMD at baseline. However,

increasing riboflavin intake correlated with femoral BMD both at baseline

and the end of the follow-up period in subjects homozygous for the T allele in

the MTHFR gene [52]. Taken together, the findings suggest a protective role

of B vitamins in osteoporotic fractures which may depend on the MTHFRgenotype. A long-term interventional study taking into account the nutrient-

gene interaction is clearly needed.

With the current technological advances in the genomic scale analysis, it is

not inconceivable that the cohort of susceptibility genes for osteoporosis as well


as the skeletal responsiveness to drugs and nutrients will soon be elucidated.

The information will be potentially useful in the clinical management of osteo-

porosis and should supplement rather than replace conventional clinical infor-

mation. In the face of limited resources, clinical decision should also take into

account the social, health and economic perspectives in order to balance the

benefit of novel clinical strategies against the associated risks and available

resources.

References

1 Sowers MR, Burns TL, Wallace RB: Familial resemblance of bone mass in adult women. Genet

Epidemiol 1986;3:85–93.

2 Lutz J: Bone mineral, serum calcium, and dietary intakes of mother/daughter pairs. Am J Clin

Nutr 1986;44:99–106.

3 Pocock NA, Eisman JA, Hopper JL, Yeates MG, Sambrook PN, Eberl S: Genetic determinants of

bone mass in adults: a twin study. J Clin Invest 1987;80:706–710.

4 Slemenda CW, Christian JC, Williams CJ, Norton JA, Johnston CC Jr: Genetic determinants of

bone mass in adult women: a reevaluation of the twin model and the potential importance of gene

interaction on heritability estimates. J Bone Miner Res 1991;6:561–567.

5 Deng HW, Chen WM, Conway T, Zhou Y, Davies KM, Stegman MR, Deng H, Recker RR:

Determination of bone mineral density of the hip and spine in human pedigrees by genetic and

life-style factors. Genet Epidemiol 2000;19:160–177.

6 Ferrari S, Rizzoli R, Slosman D, Bonjour JP: Familial resemblance for bone mineral mass is

expressed before puberty. J Clin Endocrinol Metab 1998;83:358–361.

7 Gueguen R, Jouanny P, Guillemin F, Kuntz C, Pourel J, Siest G: Segregation analysis and variance

components analysis of bone mineral density in healthy families. J Bone Miner Res 1995;10:

2017–2022.

8 Sowers MR, Boehnke M, Jannausch ML, Crutchfield M, Corton G, Burns TL: Familiality and

partitioning the variability of femoral bone mineral density in women of child-bearing age. Calcif

Tissue Int 1992;50:110–114.

9 Niu T, Chen C, Cordell H, Yang J, Wang B, Wang Z, Fang Z, Schork NJ, Rosen CJ, Xu X:

A genome-wide scan for loci linked to forearm bone mineral density. Hum Genet 1999;104:

226–233.

10 Koller DL, Econs MJ, Morin PA, Christian JC, Hui SL, Parry P, Curran ME, Rodriguez LA,

Conneally PM, Joslyn G, Peacock M, Johnston CC, Foroud T: Genome screen for QTLs con-

tributing to normal variation in bone mineral density and osteoporosis. J Clin Endocrinol Metab

2000;85:3116–3120.

11 Gong G, Stern HS, Cheng SC, Fong N, Mordeson J, Deng HW, Recher RR: The association of

bone mineral density with vitamin D receptor gene polymorphisms. Osteoporos Int 1999;9:55–64.

12 Thakkinstian A, D’Este C, Eisman J, Nguyen T, Attia J: Meta-analysis of molecular association

studies: vitamin D receptor gene polymorphisms and BMD as a case study. J Bone Miner Res

2004;19:419–428.

13 Grant SF, Reid DM, Blake G, Herd R, Fogelman I, Ralston SH: Reduced bone density and osteo-

porosis associated with a polymorphic Sp1 binding site in the collagen type I alpha 1 gene. Nat

Genet 1996;14:203–205.

14 Mann V, Hobson EE, Li B, Stewart TL, Grant SF, Robins SP, Aspden RM, Ralston SH: A COLIA1

Sp1 binding site polymorphism predisposes to osteoporotic fracture by affecting bone density and

quality. J Clin Invest 2001;107:899–907.

15 Efstathiadou Z, Tsatsoulis A, Ioannidis JP: Association of collagen Ialpha 1 Sp1 polymorphism

with the risk of prevalent fractures: a meta-analysis. J Bone Miner Res 2001;16:1586–1592.


16 Ioannidis JP, Stavrou I, Trikalinos TA, Zois C, Brandi ML, Gennari L, Albagha O, Ralston SH,

Tsatsoulis A; ER-alpha Genetics Meta-Analysis: Association of polymorphisms of the estrogen

receptor alpha gene with bone mineral density and fracture risk in women: a meta-analysis. J Bone

Miner Res 2002;17:2048–2060.

17 Ferrari SL, Deutsch S, Choudhury U, Chevalley T, Bonjour JP, Dermitzakis ET, Rizzoli R,

Antonarakis SE: Polymorphisms in the low-density lipoprotein receptor-related protein 5 (LRP5)

gene are associated with variation in vertebral bone mass, vertebral bone size, and stature in

whites. Am J Hum Genet 2004;74:866–875.

18 Bollerslev J, Wilson SG, Dick IM, Islam FM, Ueland T, Palmer L, Devine A, Prince RL: LRP5

gene polymorphisms predict bone mass and incident fractures in elderly Australian women. Bone

2005;36:599–606.

19 Koller DL, Ichikawa S, Johnson ML, Lai D, Xuei X, Edenberg HJ, Conneally PM, Hui SL,

Johnston CC, Peacock M, Foroud T, Econs MJ: Contribution of the LRP5 gene to normal variation

in peak BMD in women. J Bone Miner Res 2005;20:75–80.

20 Zhang ZL, Qin YJ, He JW, Huang QR, Li M, Hu YQ, Liu YJ: Association of polymorphisms in

low-density lipoprotein receptor-related protein 5 gene with bone mineral density in post-

menopausal Chinese women. Acta Pharmacol Sin 2005;26:1111–1116.

21 Urano T, Shiraki M, Ezura Y, Fujita M, Sekine E, Hoshino S, Hosoi T, Orimo H, Emi M, Ouchi Y,

Inoue S: Association of a single-nucleotide polymorphism in low-density lipoprotein receptor-

related protein 5 gene with bone mineral density. J Bone Miner Metab 2004;22:341–345.

22 Koh JM, Jung MH, Hong JS, Park HJ, Chang JS, Shin HD, Kim SY, Kim GS: Association between

bone mineral density and LDL receptor-related protein 5 gene polymorphisms in young Korean

men. J Korean Med Sci 2004;19:407–412.

23 Recker RR, Deng HW: Role of genetics in osteoporosis. Endocrine 2002;17:55–66.

24 Deng HW, Shen H, Xu FH, et al: Tests of linkage and/or association of genes for vitamin D recep-

tor, osteocalcin, and parathyroid hormone with bone mineral density. J Bone Miner Res 2002;17:

678–686.

25 Deng HW, Mahaney MC, Williams JT, et al: Relevance of the genes for bone mass variation to

susceptibility to osteoporotic fractures and its implications to gene search for complex human dis-

eases. Genet Epidemiol 2002;22:12–25.

26 Efstathiadou Z, Tsatsoulis A, Loannidis JP: Association of collagen I alpha 1 Sp1 polymorphism

with the risk of prevalent fractures: a meta-analysis. J Bone Miner Res 2001;16:1586–1592.

27 Uitterlinden AG, Burger H, Huang Q, Yue F, McGuigan FE, Grant SF, Hofman A, van Leeuwen JP,

Pols HA, Ralston SH: Relation of alleles of the collagen type Ialpha1 gene to bone density

and the risk of osteoporotic fractures in postmenopausal women. N Engl J Med 1998;338:

1016–1021.

28 Garnero P, Borel O, Grant SF, Ralston SH, Delmas PD: Collagen Ialpha1 Sp1 polymorphism,

bone mass, and bone turnover in healthy French premenopausal women: the OFELY study. J Bone

Miner Res 1998;13:813–817.

29 Han KO, Moon IG, Hwang CS, Choi JT, Yoon HK, Min HK, Han IK: Lack of an intronic Sp1

binding-site polymorphism at the collagen type I alpha1 gene in healthy Korean women. Bone

1999;24:135–137.

30 Nakajima T, Ota N, Shirai Y, Hata A, Yoshida H, Suzuki T, Hosoi T, Orimo H, Emi M: Ethnic dif-

ference in contribution of Sp1 site variation of COLIA1 gene in genetic predisposition to osteo-

porosis. Calcif Tissue Int 1999;65:352–353.

31 Lambrinoudaki I, Kung AW: Absence of high-risk ‘s’ allele associated with osteoporosis at the

intronic SP1 binding-site of collagen I alpha1 gene in Southern Chinese. J Endocrinol Invest

2001;24:499–502.

32 Johnston CC Jr, Miller JZ, Slemenda CE, Reister TK, Hui S, Christian JC, Peacock M: Calcium

supplementation and increases in bone mineral density in children. N Engl J Med 1992;327:

82–87.

33 Nowson CA, Green RM, Hopper JL, Sherwin AJ, Young D, Kaymakci B, Guest CS, Smid M,

Larkins RG, Wark JD: A co-twin study of the effect of calcium supplementation on bone density

during adolescence. Osteoporos Int 1997;7:219–225.


34 Chevalley T, Rizzoli R, Hans D, Ferrari S, Bonjour JP: Interaction between calcium intake and

menarcheal age on bone mass gain: an eight-year follow-up study from prepuberty to postmenar-

che. J Clin Endocrinol Metab 2005;90:44–51.

35 Matkovic V, Goel PK, Badenhop-Stevens NE, Landoll JD, Li B, Llich JZ, Skugor M, Nagode LA,

Mobley SL, Ha EJ, Hangartner TN, Clairmont A: Calcium supplementation and bone mineral den-

sity in females from childhood to young adulthood: a randomized controlled trial. Am J Clin Nutr

2005;81:175–188.

36 Welten DC, Kemper HC, Post GB, van Staveren WA: A meta-analysis of the effect of

calcium intake on bone mass in young and middle aged females and males. J Nutr 1995;125:

2802–2813.

37 Shea B, Wells G, Cranney A, Zytaruk N, Robinson V, Griffith L, Ortiz Z, Peterson J, Adachi J,

Tugwell P, Guyatt G; Osteoporosis Methodology Group and The Osteoporosis Research Advisory

Group: Meta-analyses of therapies for postmenopausal osteoporosis. 7. Meta-analysis of calcium

supplementation for the prevention of postmenopausal osteoporosis. Endocr Rev 2002;23:

552–559.

38 Dawson-Hughes B, Harris SS, Finneran S: Calcium absorption on high and low calcium intakes in

relation to vitamin D receptor genotype. J Clin Endocrinol Metab 1995;80:3657–3661.

39 Morrison NA, George PM, Vaughan T, Tilyard MW, Frampton CM, Gilchrist NL: Vitamin D

receptor genotypes influence the success of calcitriol therapy for recurrent vertebral fracture in

osteoporosis. Pharmacogenet Genomics 2005;15:127–135.

40 Ongphiphadhanakul B, Chanprasertyothin S, Payattikul P, et al: Association of a T262C transition

in exon 1 of estrogen-receptor-alpha gene with skeletal responsiveness to estrogen in post-menopausal

women. J Endocrinol Invest 2001;24:749–755.

41 Ongphiphadhanakul B, Chanprasertyothin S, Payatikul P, et al: Oestrogen-receptor-alpha gene

polymorphism affects response in bone mineral density to oestrogen in post-menopausal women.

Clin Endocrinol (Oxf) 2000;52:581–585.

42 Salmen T, Heikkinen AM, Mahonen A, Kroger H, Komulainen M, Saarikoski S, Honkanen R,

Maenpaa PH: The protective effect of hormone-replacement therapy on fracture risk is modulated

by estrogen receptor alpha genotype in early postmenopausal women. J Bone Miner Res 2000;15:

2479–2486.

43 Han KO, Moon IG, Kang YS, Chung HY, Min HK, Han IK: Nonassociation of estrogen receptor

genotypes with bone mineral density and estrogen responsiveness to hormone replacement ther-

apy in Korean postmenopausal women. J Clin Endocrinol Metab 1997;82:991–995.

44 Herrington DM, Howard TD, Hawkins GA, et al: Estrogen-receptor polymorphisms and effects of

estrogen replacement on high-density lipoprotein cholesterol in women with coronary disease.

N Engl J Med 2002;346:967–974.

45 van Meurs JB, Dhonukshe-Rutten RA, Pluijm SM, van der Klift M, de Jonge R, Lindemans J,

de Groot LC, Hofman A, Witteman JC, van Leeuwen JP, Breteler MM, Lips P, Pols HA,

Uitterlinden AG: Homocysteine levels and the risk of osteoporotic fracture. N Engl J Med

2004;350: 2033–2041.

46 McLean RR, Jacques PF, Selhub J, Tucker KL, Samelson EJ, Broe KE, Hannan MT, Cupples LA,

Kiel DP: Homocysteine as a predictive factor for hip fracture in older persons. N Engl J Med

2004;350:2042–2049.

47 Sato Y, Honda Y, Iwamoto J, Kanoko T, Satoh K: Effect of folate and mecobalamin on hip fractures

in patients with stroke: a randomized controlled trial. JAMA 2005;293:1082–1088.

48 Miyao M, Morita H, Hosoi T, Kurihara H, Inoue S, Hoshino S, Shiraki M, Yazaki Y, Ouchi Y:

Association of methylenetetrahydrofolate reductase (MTHFR) polymorphism with bone mineral

density in postmenopausal Japanese women. Calcif Tissue Int 2000;66:190–194.

49 Jorgensen HL, Madsen JS, Madsen B, Saleh MM, Abrahamsen B, Fenger M, Lauritzen JB:

Association of a common allelic polymorphism (C677T) in the methylenetetrahydrofolate reduc-

tase gene with a reduced risk of osteoporotic fractures. A case control study in Danish post-

menopausal women. Calcif Tissue Int 2002;71:386–392.

50 Abrahamsen B, Madsen JS, Tofteng CL, Stilgren L, Bladbjerg EM, Kristensen SR, Brixen K,

Mosekilde L: A common methylenetetrahydrofolate reductase (C677T) polymorphism is associated


with low bone mineral density and increased fracture incidence after menopause: longitudinal data

from the Danish osteoporosis prevention study. J Bone Miner Res 2003;18:723–729.

51 McLean RR, Karasik D, Selhub J, Tucker KL, Ordovas JM, Russo GT, Cupples LA, Jacques PF,

Kiel DP: Association of a common polymorphism in the methylenetetrahydrofolate reductase

(MTHFR) gene with bone phenotypes depends on plasma folate status. J Bone Miner Res

2004;19:410–418.

52 Macdonald HM, McGuigan FE, Fraser WD, New SA, Ralston SH, Reid DM: Methylenetetra-

hydrofolate reductase polymorphism interacts with riboflavin intake to influence bone mineral

density. Bone 2004;35:957–964.

Dr. Boonsong Ongphiphadhanakul

Department of Medicine, Faculty of Medicine

Ramathibodi Hospital, Mahidol University

Bangkok (Thailand)

Tel. �66 2 201 2416, ext. 1590, Fax �66 2 201 2416, E-Mail [email protected]



Application of Nutrigenomics in Eye Health

Cécile Delcourt

Inserm, U593 ‘Epidemiology, Public Health and Development’;

Université Victor Segalen Bordeaux 2, Bordeaux, France

AbstractThis paper reviews recent findings on the implication of nutritional and genetic factors

in age-related eye diseases: age-related macular degeneration (AMD; a degenerative disease

of the retina) and cataract (opacification of the lens). Because of direct exposure to light, the

eye is particularly sensitive to oxidative stress. Antioxidants, such as vitamin E, C or zinc,

clearly have a protective effect in AMD and probably in cataract. In addition, two carotenoids,

lutein and zeaxanthin, may play a more specific role in the eye: they accumulate in the retina,

where they form the macular pigment, and in the lens. Their role is probably to filter out pho-

totoxic blue light and to quench singlet oxygen. Finally, docosahexaenoic acid (an ��3

polyunsaturated fatty acid) is particularly important for the retina, where it exerts structural,

functional and protective actions. Besides, these diseases are strongly influenced by genetics,

as demonstrated by familial and twin studies. The apolipoprotein E4 allele is associated

with a reduced risk of AMD, while an association of AMD with complement factor H poly-

morphism has recently been demonstrated. Nutrigenomics, by studying the interactions

between genetic variability and nutritional factors, represents a new challenge in order to

account for interindividual variations in disease susceptibility. Such potential interactions are

presented.


There is growing evidence for a major implication of nutrition and genet-

ics in the etiology of age-related eye diseases [age-related macular degeneration

(AMD), cataract and glaucoma], which are the major causes of blindness

worldwide [1]. However, the interest in nutritional risk factors for these diseases

and the identification of the associated genes are still recent. As such, the inter-

actions between nutritional and genetic factors have not yet been studied. Some

hypotheses can be drawn from the known interactions between nutrition and

specific biological mechanisms.

Nutrigenomics and Eye Health 169

AMD is a degeneration of the central retina, known as the macula. It is

associated with extracellular deposits forming yellow spots on the retina,

named drusen. These deposits are probably related to decreased degradation

and elimination of cellular components during the process of renewal of the

photoreceptors. Late-stage AMD is characterized by the development of

choroidal neovascularization (exudative AMD) or by the disappearance of pho-

toreceptors and underlying retinal pigment epithelium (atrophic AMD).

Cataract is an opacification of the lens, which focuses the light on the retina.

Glaucoma is a neuropathy of the optic nerve, leading to a gradual loss of the

peripheric visual field and leading finally to total blindness. The prevalence of

these diseases increases sharply with age. They are multifactorial, with both

genetic and environmental factors. Some risk factors have been clearly identi-

fied, such as apolipoprotein E, complement factor H polymorphisms and smok-

ing for AMD; light exposure, smoking, diabetes and oral corticosteroid use for

cataract, and intraocular pressure for glaucoma.

Oxidative stress plays an important role in eye ageing. The retina is particu-

larly susceptible to oxidative stress because of its high content of easily peroxidiz-

able long-chain polyunsaturated fatty acids (PUFA), in particular docosahexaenoic

acid (DHA; an ��3 PUFA) [2]. Its susceptibility is also due to the high level of in

situ reactive oxygen species production, due in particular to light exposure and

high metabolic activity [2]. Opacification of the lens is due to oxidation of the

structural proteins of the lens, inducing their aggregation [3].

Three types of nutritional factors offer or may offer protection against eye

ageing: antioxidants, such as vitamins C and E or zinc; lutein and zeaxanthin,

two carotenoids which accumulate specifically in the retina and lens; ��3

PUFA, and in particular DHA, which have important structural and protective

functions in the retina. Initial epidemiological observations, showing that high

vitamin E plasma levels may protect against AMD [4], have been confirmed by

a large randomized clinical trial performed in the United States. In this study,

performed on nearly 5,000 subjects, supplementation for 6 years with high

doses of antioxidants (vitamins E and C, and �-carotene) and zinc significantly

reduced the risk of developing advanced AMD by 34% in subjects with early

AMD [5]. In parallel, numerous studies have evidenced a 20–50% reduction of

the risk for nuclear cataract (one of the subtypes of cataract, based on the local-

ization of the opacities) in subjects with high dietary intakes or high plasma

concentrations of vitamins C and E [6]. However, in several large randomized

clinical trials, the risk for cataract was not reduced with antioxidant supplemen-

tation [7–9]. Only the REACT study showed an effect of supplementation with

vitamins C and E, and �-carotene on cortical cataract [10].

A more recent research domain regards the role of two carotenoids, lutein

and zeaxanthin, for the protection of the retina and the lens. These carotenoids

Delcourt 170

accumulate in the macula, where they are known as the macular pigment [11],

and they are also the only carotenoids found in the lens [12]. Besides their

antioxidant properties, they probably act as a filter against the phototoxic

effects of blue light [11]. Two clinical studies have shown that eyes at risk of

AMD have a lower density of the macular pigment [13, 14]. Epidemiological

studies also suggest that a high intake or high plasma levels of lutein and zeax-

anthin could protect against AMD and cataract [15–22]. Although all these

studies have yielded results in the direction of a protective effect, they were not

always significant due to small sample sizes. A small randomized study showed

improvement of near visual acuity with lutein supplementation, in subjects with

atrophic AMD [23].

Finally, DHA is a major component of the photoreceptors, where it exerts

structural (membrane fluidity, interaction with rhodopsin) and protective func-

tions [24]. The protective functions include the systemic anti-inflammatory,

antiangiogenic and antiapoptotic functions, but also specific actions such as

increase in lysosomal acid lipase, leading to increased lipid degradation in the

retinal pigment epithelium [24]. Few epidemiological studies are available

concerning the associations of AMD with fat. In two cross-sectional studies

[25, 26], weekly fish consumption, which is the main source of DHA, was

associated with a 50–60% reduction in the risk for AMD, after multivariate

adjustment. In the Eye Disease Case-Control Study, a high dietary intake of

��6 PUFA was significantly associated with a 2-fold increased risk for exuda-

tive AMD, after multivariate adjustment. Consumption of ��3 PUFA was sig-

nificantly associated with a 31% reduction in the risk for AMD after age and

gender adjustment, but not after multivariate adjustment. Results were similar

for fish intake [27]. In a pooled analysis of the Nurses’ Health and Health

Professionals’ Cohort Studies, subjects consuming fish had a reduced risk of

developing AMD, after multivariate adjustment [28]. High DHA intakes were

also associated with a reduced risk of AMD, whereas, surprisingly, high intakes

of �-linolenic acid were associated with an increased risk. Finally, in a study on

261 patients, initially presenting early AMD, total fat intake, and more specifi-

cally, intakes of vegetable fat, monounsaturated fatty acids and PUFA (mainly

due to ��6 PUFA) were positively associated with the risk of developing late

AMD [29]. Fish intake was associated with a decreased risk of late AMD only

in those with a low dietary intake of linoleic acid. Globally, these results sug-

gest that excessive intake of ��6 PUFA, and low intake of ��3 PUFA may be

associated with an increased risk for AMD. Recent studies suggest that ��3

and ��6 PUFA may also be implicated in other eye conditions, such as glau-

coma [30] or dry eye syndrome [31].

Besides the nutritional dimension of AMD and cataract, these diseases are

strongly influenced by genetics, as demonstrated by familial and twin studies


[32–37]. The apolipoprotein E4 allele is associated with a 50% reduced risk of

AMD [38–43]. Recently, three independent teams simultaneously demonstrated a

significant association between the Y402H polymorphism of the complement fac-

tor H gene and AMD in North American subjects [44–46], immediately followed

by three other corroborating papers in North American populations [47–49], one

study from France [50] and one from Iceland [51]. Y402H is a common variant,

with about 30% of the general population bearing the minor (C) allele (at least in

Caucasians). In these studies, subjects heterozygotes for C have a 2.5- to 4-fold

increased risk for AMD, while subjects homozygotes for C have a 3.5- to 7.5-fold

increased risk for AMD [50]. Complement factor H is a key regulator of the com-

plement system of innate immunity [52]. Histologic observations are consistent

with inappropriate activation of the complement system in AMD [53].

Finally, several linkage studies show an association of AMD with chromo-

some 10p26 [54]. With respect to cataract, a recent linkage study identified a

major locus on chromosome 6p12–q12 for cortical cataract [55].

Interactions between genetic variability and nutritional factors represent a

new challenge in order to account for interindividual variations in disease sus-

ceptibility. While some properties of nutritional factors rely on direct effects

(such as antioxidant properties, or structural functions of DHA), many nutri-

tional factors also have cellular effects and interact with genes. Nutrigenomics

in eye health therefore potentially includes all genes implicated in the metabo-

lism or activities of nutritional factors associated with eye diseases, and all

nutrients implicated in the activities of genes associated with eye diseases,

thereby opening a vast research domain. Since the genes identified to date are

from the lipid metabolism (apolipoprotein E) and innate immunity (comple-

ment factor H), interactions with lipids and antioxidants are particularly

expected. Recently, in an animal model, the combination of the apolipoprotein

E4 allele with a high-fat diet induced modifications of the retina that mimic the

pathology associated with human AMD [56]. It is also well known that PUFA

and zinc interact with genes of inflammation and immunity [57, 58]. Whether

the risk for AMD may be modified by interactions of PUFA and zinc with the

complement factor H gene remains to be determined. Zinc has recently been

implicated in the binding of complement factor H with its target complement

factor (C3b) [59]. In the field of carotenoids, the Pi isoform of the glutathione

S-transferase (GSTP1) has recently been identified as a membrane-bound bind-

ing protein for zeaxanthin in the macula [60]. The same authors have shown that

GSTP1 and zeaxanthin act in synergy for the prevention of membrane lipid per-

oxidation [61]. Interestingly, GSTP1 polymorphism was associated with the

risk of cortical cataract in an Estonian population [62]. These data globally sug-

gest that interaction of the GSTP1 gene with dietary zeaxanthin may be impli-

cated in AMD and cataract.

Delcourt 172

In conclusion, age-related eye diseases, which are the major causes of blind-

ness worldwide, are strongly influenced by nutrition and genetics. Nutrigenomics,

by studying the interactions of nutritional and genetic factors, opens a new

research avenue. Understanding the interaction of nutrients with genes may help

target susceptible individuals for nutritional prevention of eye diseases.

References

1 Resnikoff S, Pascolini D, Etya’ale D, Kocur I, Pararajasegaram R, Pokharel GP, Mariotti SP:

Global data on visual impairment in the year 2002. Bull World Health Organ 2004;82:844–851.

2 Beatty S, Koh H, Phil M, Henson D, Boulton M: The role of oxidative stress in the pathogenesis of

age-related macular degeneration. Surv Ophthalmol 2000;45:115–134.

3 Boscia F, Grattagliano I, Vendemiale G, Micelli Ferrari T, Altomare E: Protein oxidation and lens

opacity in humans. Invest Ophthalmol Vis Sci 2000;41:2461–2465.

4 Delcourt C, Cristol JP, Tessier F, Leger CL, Descomps B, Papoz L: Age-related macular degenera-

tion and antioxidant status in the POLA study. POLA Study Group. Pathologies Oculaires Liées à

l’Âge. Arch Ophthalmol 1999;117:1384–1390.

5 Age-Related Eye Disease Study Group: A randomized, placebo-controlled, clinical trial of high-

dose supplementation with vitamins C and E, beta carotene, and zinc for age-related macular

degeneration and vision loss: AREDS report No 8. Arch Ophthalmol 2001;119:1417–1436.

6 Taylor A, Hobbs M: 2001 assessment of nutritional influences on risk for cataract. Nutrition

2001;17:845–857.

7 Age-Related Eye Disease Study Group: A randomized, placebo-controlled, clinical trial of high-

dose supplementation with vitamins C and E and beta carotene for age-related cataract and vision

loss: AREDS report No 9. Arch Ophthalmol 2001;119:1439–1452.

8 McNeil JJ, Robman L, Tikellis G, Sinclair MI, McCarty CA, Taylor HR: Vitamin E supplementa-

tion and cataract: randomized controlled trial. Ophthalmology 2004;111:75–84.

9 Teikari JM, Virtamo J, Rautalahti M, Palmgren J, Liesto K, Heinonen OP: Long-term supplemen-

tation with alpha-tocopherol and beta-carotene and age-related cataract. Acta Ophthalmol Scand

1997;75:634–640.

10 Chylack LT Jr, Brown NP, Bron A, Hurst M, Kopcke W, Thien U, Schalch W: The Roche European

American Cataract Trial (REACT): a randomized clinical trial to investigate the efficacy of an oral

antioxidant micronutrient mixture to slow progression of age-related cataract. Ophthalmic

Epidemiol 2002;9:49–80.

11 Schalch W, Dayhaw-Barker P, Barker FM: The carotenoids of the human retina; in Taylor A (ed):

Nutritional and Environmental Influences of the Eye. Boca Raton, CRC Press, 1999, pp 215–250.

12 Yeum KJ, Shang FM, Schalch WM, Russell RM, Taylor A: Fat-soluble nutrient concentrations in

different layers of human cataractous lens. Curr Eye Res 1999;19:502–505.

13 Beatty S, Murray IJ, Henson DB, Carden D, Koh H, Boulton ME: Macular pigment and risk for

age-related macular degeneration in subjects from a Northern European population. Invest

Ophthalmol Vis Sci 2001;42:439–446.

14 Bone RA, Landrum JT, Mayne ST, Gomez CM, Tibor SE, Twaroska EE: Macular pigment in donor

eyes with and without AMD: a case-control study. Invest Ophthalmol Vis Sci 2001;42:235–240.

15 Eye Disease Case-Control Study Group: Antioxidant status and neovascular age-related macular

degeneration. Arch Ophthalmol 1993;111:104–109.

16 Seddon JM, Ajani UA, Sperduto RD, Hiller R, Blair N, Burton TC, Farber MD, Gragoudas ES,

Haller J, Miller DT, et al: Dietary carotenoids, vitamins A, C, and E, and advanced age-related

macular degeneration. Eye Disease Case-Control Study Group. JAMA 1994;272:1413–1420.

17 Snellen EL, Verbeek AL, Van Den Hoogen GW, Cruysberg JR, Hoyng CB: Neovascular age-

related macular degeneration and its relationship to antioxidant intake. Acta Ophthalmol Scand

2002;80:368–371.


18 Gale CR, Hall NF, Phillips DI, Martyn CN: Lutein and zeaxanthin status and risk of age-related

macular degeneration. Invest Ophthalmol Vis Sci 2003;44:2461–2465.

19 Chasan-Taber L, Willett WC, Seddon JM, Stampfer MJ, Rosner B, Colditz GA, Speizer FE,

Hankinson SE: A prospective study of carotenoid and vitamin A intakes and risk of cataract

extraction in US women. Am J Clin Nutr 1999;70:509–516.

20 Brown L, Rimm EB, Seddon JM, Giovannucci EL, Chasan-Taber L, Spiegelman D, Willett WC,

Hankinson SE: A prospective study of carotenoid intake and risk of cataract extraction in US men.

Am J Clin Nutr 1999;70:517–524.

21 Lyle BJ, Mares-Perlman JA, Klein BE, Klein R, Greger JL: Antioxidant intake and risk of incident

age-related nuclear cataracts in the Beaver Dam Eye Study. Am J Epidemiol 1999;149:801–809.

22 Jacques PF, Chylack LT Jr, Hankinson SE, Khu PM, Rogers G, Friend J, Tung W, Wolfe JK,

Padhye N, Willett WC, Taylor A: Long-term nutrient intake and early age-related nuclear lens

opacities. Arch Ophthalmol 2001;119:1009–1019.

23 Richer S, Stiles W, Statkute L, Pulido J, Frankowski J, Rudy D, Pei K, Tsipursky M, Nyland J:

Double-masked, placebo-controlled, randomized trial of lutein and antioxidant supplementation in

the intervention of atrophic age-related macular degeneration: the Veterans LAST study (Lutein

Antioxidant Supplementation Trial). Optometry 2004;75:216–230.

24 SanGiovanni JP, Chew EY: The role of omega-3 long-chain polyunsaturated fatty acids in health

and disease of the retina. Prog Retin Eye Res 2005;24:87–138.

25 Smith W, Mitchell P, Leeder SR: Dietary fat and fish intake and age-related maculopathy. Arch

Ophthalmol 2000;118:401–404.

26 Heuberger RA, Mares Perlman JA, Klein R, Klein BE, Millen AE, Palta M: Relationship of

dietary fat to age-related maculopathy in the Third National Health and Nutrition Examination

Survey. Arch Ophthalmol 2001;119:1833–1838.

27 Seddon JM, Rosner B, Sperduto RD, Yannuzzi L, Haller JA, Blair NP, Willett W: Dietary fat and

risk for advanced age-related macular degeneration. Arch Ophthalmol 2001;119:1191–1199.

28 Cho E, Hung S, Willett WC, Spiegelman D, Rimm EB, Seddon JM, Colditz GA, Hankinson SE:

Prospective study of dietary fat and the risk of age-related macular degeneration. Am J Clin Nutr

2001;73:209–218.

29 Seddon JM, Cote J, Rosner B: Progression of age-related macular degeneration: association with

dietary fat, transunsaturated fat, nuts, and fish intake. Arch Ophthalmol 2003;121:1728–1737.

30 Kang JH, Pasquale LR, Willett WC, Rosner BA, Egan KM, Faberowski N, Hankinson SE: Dietary

fat consumption and primary open-angle glaucoma. Am J Clin Nutr 2004;79:755–764.

31 Miljanovic B, Trivedi KA, Dana MR, Gilbard JP, Buring JE, Schaumberg DA: Relation between

dietary n–3 and n–6 fatty acids and clinically diagnosed dry eye syndrome in women. Am J Clin

Nutr 2005;82:887–893.

32 Gorin MB, Breitner JC, De Jong PT, Hageman GS, Klaver CC, Kuehn MH, Seddon JM: The

genetics of age-related macular degeneration. Mol Vis 1999;5:29.

33 Congdon N, Broman KW, Lai H, Munoz B, Bowie H, Gilbert D, Wojciechowski R, West SK: Cortical,

but not posterior subcapsular, cataract shows significant familial aggregation in an older population

after adjustment for possible shared environmental factors. Ophthalmology 2005;112:73–77.

34 Congdon N, Broman KW, Lai H, Munoz B, Bowie H, Gilber D, Wojciechowski R, Alston C, West

SK: Nuclear cataract shows significant familial aggregation in an older population after adjust-

ment for possible shared environmental factors. Invest Ophthalmol Vis Sci 2004;45:2182–2186.

35 Hammond CJ, Duncan DD, Snieder H, De Lange M, West SK, Spector TD, Gilbert CE: The heri-

tability of age-related cortical cataract: the twin eye study. Invest Ophthalmol Vis Sci 2001;42:

601–605.

36 Hammond CJ, Snieder H, Spector TD, Gilbert CE: Genetic and environmental factors in age-

related nuclear cataracts in monozygotic and dizygotic twins. N Engl J Med 2000;342:1786–1790.

37 Seddon JM, Cote J, Page WF, Aggen SH, Neale MC: The US twin study of age-related macular

degeneration: relative roles of genetic and environmental influences. Arch Ophthalmol 2005;123:

321–327.

38 Souied EH, Benlian P, Amouyel P, Feingold J, Lagarde JP, Munnich A, Kaplan J, Coscas G,

Soubrane G: The epsilon 4 allele of the apolipoprotein E gene as a potential protective factor for

exudative age-related macular degeneration. Am J Ophthalmol 1998;125:353–359.

Delcourt 174

39 Klaver CCW, Kliffen M, Vanduijn CM, Hofman A, Cruts M, Grobbee DE, Vanbroeckhoven C,

Dejong PTVM: Genetic association of apolipoprotein E with age-related macular degeneration.

Am J Hum Genet 1998;63:200–206.

40 Simonelli F, Margaglione M, Testa F, Cappucci G, Manitto MP, Brancato R, Rinaldi E:

Apolipoprotein E polymorphisms in age-related macular degeneration in an Italian population.

Ophthalmic Res 2001;33:325–328.

41 Schmidt S, Klaver C, Saunders A, Postel E, De La Paz M, Agarwal A, Small K, Udar N, Ong J,

Chalukya M, Nesburn A, Kenney C, Domurath R, Hogan M, Mah T, Conley Y, Ferrell R, Weeks D,

De Jong P, Van Duijn C, Haines J, Pericak Vance M, Gorin M: A pooled case-control study of the

apolipoprotein E (APOE) gene in age-related maculopathy. Ophthalmic Genet 2002;23:209–223.

42 Zareparsi S, Reddick AC, Branham KE, Moore KB, Jessup L, Thoms S, Smith Wheelock M,

Yashar BM, Swaroop A: Association of apolipoprotein E alleles with susceptibility to age-related

macular degeneration in a large cohort from a single center. Invest Ophthalmol Vis Sci

2004;45:1306–1310.

43 Baird PN, Guida E, Chu DT, Vu HT, Guymer RH: The epsilon2 and epsilon4 alleles of the

apolipoprotein gene are associated with age-related macular degeneration. Invest Ophthalmol Vis

Sci 2004;45:1311–1315.

44 Klein RJ, Zeiss C, Chew EY, Tsai JY, Sackler RS, Haynes C, Henning AK, Sangiovanni JP, Mane

SM, Mayne ST, Bracken MB, Ferris FL, Ott J, Barnstable C, Hoh J: Complement factor H poly-

morphism in age-related macular degeneration. Science 2005;308:385–389.

45 Haines JL, Hauser MA, Schmidt S, Scott WK, Olson LM, Gallins P, Spencer KL, Kwan SY,

Noureddine M, Gilbert JR, Schnetz Boutaud N, Agarwal A, Postel EA, Pericak Vance MA:

Complement factor H variant increases the risk of age-related macular degeneration. Science

2005;308:419–421.

46 Edwards AO, Ritter R III, Abel KJ, Manning A, Panhuysen C, Farrer LA: Complement factor H

polymorphism and age-related macular degeneration. Science 2005;308:421–424.

47 Conley YP, Thalamuthu A, Jakobsdottir J, Weeks DE, Mah T, Ferrell RE, Gorin MB: Candidate

gene analysis suggests a role for fatty acid biosynthesis and regulation of the complement system

in the etiology of age-related maculopathy. Hum Mol Genet 2005;14:1991–2002.

48 Zareparsi S, Branham KE, Li M, Shah S, Klein RJ, Ott J, Hoh J, Abecasis GR, Swaroop A: Strong

association of the Y402H variant in complement factor H at 1q32 with susceptibility to age-related

macular degeneration. Am J Hum Genet 2005;77:149–153.

49 Hageman GS, Anderson DH, Johnson LV, Hancox LS, Taiber AJ, Hardisty LI, Hageman JL,

Stockman HA, Borchardt JD, Gehrs KM, Smith RJ, Silvestri G, Russell SR, Klaver CC,

Barbazetto I, Chang S, Yannuzzi LA, Barile GR, Merriam JC, Smith RT, Olsh AK, Bergeron J,

Zernant J, Merriam JE, Gold B, Dean M, Allikmets R: A common haplotype in the complement

regulatory gene factor H (HF1/CFH) predisposes individuals to age-related macular degeneration.

Proc Natl Acad Sci USA 2005;102:7227–7232.

50 Souied EH, Leveziel N, Richard F, Dragon-Durey MA, Coscas G, Soubrane G, Benlian P,

Fremeaux-Bacchi V: Y402H complement factor H polymorphism associated with exudative age-

related macular degeneration in the French population. Mol Vis 2005;11:1135–1140.

51 Magnusson KP, Duan S, Sigurdsson H, Petursson H, Yang Z, Zhao Y, Bernstein PS, Ge J, Jonasson

F, Stefansson E, Helgadottir G, Zabriskie NA, Jonsson T, Bjornsson A, Thorlacius T, Jonsson PV,

Thorleifsson G, Kong A, Stefansson H, Zhang K, Stefansson K, Gulcher JR: CFH Y402H confers

similar risk of soft drusen and both forms of advanced AMD. PLoS Med 2005;3:e5.

52 Rodriguez de Cordoba S, Esparza-Gordillo J, Goicoechea de Jorge E, Lopez-Trascasa M,

Sanchez-Corral P: The human complement factor H: functional roles, genetic variations and dis-

ease associations. Mol Immunol 2004;41:355–367.

53 Anderson DH, Mullins RF, Hageman GS, Johnson LV: A role for local inflammation in the forma-

tion of drusen in the aging eye. Am J Ophthalmol 2002;134:411–431.

54 Fisher SA, Abecasis GR, Yashar BM, Zareparsi S, Swaroop A, Iyengar SK, Klein BE, Klein R,

Lee KE, Majewski J, Schultz DW, Klein ML, Seddon JM, Santangelo SL, Weeks DE, Conley YP,

Mah TS, Schmidt S, Haines JL, Pericak-Vance MA, Gorin MB, Schulz HL, Pardi F, Lewis CM,

Weber BH: Meta-analysis of genome scans of age-related macular degeneration. Hum Mol Genet

2005;14:2257–2264.


55 Iyengar SK, Klein BE, Klein R, Jun G, Schick JH, Millard C, Liptak R, Russo K, Lee KE, Elston

RC: Identification of a major locus for age-related cortical cataract on chromosome 6p12–q12 in

the Beaver Dam Eye Study. Proc Natl Acad Sci USA 2004;101:14485–14490.

56 Malek G, Johnson LV, Mace BE, Saloupis P, Schmechel DE, Rickman DW, Toth CA, Sullivan PM,

Rickman CB: Apolipoprotein E allele-dependent pathogenesis: a model for age-related retinal

degeneration. Proc Natl Acad Sci USA 2005;102:11900–11905.

57 Calder PC, Grimble RF: Polyunsaturated fatty acids, inflammation and immunity. Eur J Clin Nutr

2002;56(suppl 3):S14–S19.

58 Calder PC, Kew S: The immune system: a target for functional foods? Br J Nutr 2002;88(suppl

2):S165–S177.

59 Blom AM, Kask L, Ramesh B, Hillarp A: Effects of zinc on factor I cofactor activity of C4b-bind-

ing protein and factor H. Arch Biochem Biophys 2003;418:108–118.

60 Bhosale P, Larson AJ, Frederick JM, Southwick K, Thulin CD, Bernstein PS: Identification and

characterization of a Pi isoform of glutathione S-transferase (GSTP1) as a zeaxanthin-binding

protein in the macula of the human eye. J Biol Chem 2004;279:49447–49454.

61 Bhosale P, Bernstein PS: Synergistic effects of zeaxanthin and its binding protein in the prevention

of lipid membrane oxidation. Biochim Biophys Acta 2005;1740:116–121.

62 Juronen E, Tasa G, Veromann S, Parts L, Tiidla A, Pulges R, Panov A, Soovere L, Koka K,

Mikelsaar AV: Polymorphic glutathione S-transferases as genetic risk factors for senile cortical

cataract in Estonians. Invest Ophthalmol Vis Sci 2000;41:2262–2267.

Dr. Cécile Delcourt

Inserm, U593, Université Victor Segalen Bordeaux 2

146, rue Léo Saignat

FR–33076 Bordeaux Cedex (France)

Tel. �33 5 57 57 15 96, Fax �33 5 57 57 14 86, E-Mail [email protected]



Nutrigenomics of Taste – Impact on FoodPreferences and Food Production

Ahmed El-Sohemy, Lindsay Stewart, Nora Khataan, Bénédicte Fontaine-Bisson, Pauline Kwong, Stephen Ozsungur, Marilyn C. Cornelis

Department of Nutritional Sciences, Faculty of Medicine, University of Toronto,

Toronto, Ont., Canada

AbstractFood preferences are influenced by a number of factors such as personal experiences,

cultural adaptations and perceived health benefits. Taste, however, is the most important

determinant of how much a food is liked or disliked. Based on the response to bitter-tasting

compounds such as phenylthiocarbamide (PTC) or 6-n-propylthiouracil (PROP), individuals

can be classified as supertasters, tasters or nontasters. Sensitivity to bitter-tasting compounds

is a genetic trait that has been recognized for more than 70 years. Genetic differences in bit-

ter taste perception may account for individual differences in food preferences. Other factors

such as age, sex and ethnicity may also modify the response to bitter-tasting compounds.

There are several members of the TAS2R receptor gene family that encode taste receptors on

the tongue, and genetic polymorphisms of TAS2R38 have been associated with marked dif-

ferences in the perception of PTC and PROP. However, the association between TAS2R38genotypes and aversion to bitter-tasting foods is not clear. Single nucleotide polymorphisms

in other taste receptor genes have recently been identified, but their role in bitter taste per-

ception is not known. Establishing a genetic basis for food likes/dislikes may explain, in part,

some of the inconsistencies among epidemiologic studies relating diet to risk of chronic dis-

eases. Identifying populations with preferences for particular flavors or foods may lead to the

development of novel food products targeted to specific genotypes or ethnic populations.


Background

Food preferences are determined by a number of factors such as

taste. Individual differences in the perception of sweet, salty, sour, umami or

bitter taste may influence dietary habits, which can affect nutritional status and

risk of chronic diseases [1, 2]. The ability to taste bitter compounds such as

Nutrigenomics – Applications to the Food Industry

Nutrigenomics of Taste 177

phenylthiocarbamide (PTC) or 6-n-propylthiouracil (PROP) is a genetic trait

that is due to the presence of a functional TAS2R38 receptor, which is expressed

in taste cells on the tongue [3] (fig. 1). PTC and PROP are members of a class

of compounds called ‘thioureas’ and carry the chemical group N-C�S, which

is responsible for their characteristic bitter taste [1, 2, 4]. Although these chemi-

cals are not found in foods or beverages, their structural similarities to naturally

occurring chemicals have made them useful tools to study taste preferences and

food aversions. Populations have typically displayed bimodality in sensitivity to

PTC, with approximately 75% of individuals perceiving this compound as

bitter, while the remaining 25% find this compound to be relatively tasteless

[3, 5]. Those who perceive the bitter taste can be further divided into tasters and

supertasters [3–8].

The TAS2R38 gene consists of a single exon that is 1,002 bp long and

encodes a 333-amino-acid 7-transmembrane domain, guanine nucleotide-

binding protein-coupled receptor [9]. Single nucleotide polymorphisms (SNPs)

that result in amino acid substitutions have been identified in the TAS2R38gene, and can be used to predict taster status [9–11]. There are two common

haplotypes that consist of three SNPs (A49P, V262A, and I296V), which show

a strong association with PTC taster status [12]. Tasters have the PAV haplotype

whereas nontasters have the AVI haplotype [11]. The A49P substitution has

been used as a tag SNP to identify taster status [13]. Only a few studies have

explored the relationship between genetic differences in the TAS2R38 gene and

taster phenotype [9, 13, 14]. A recent study examined the association between

TAS2R38 genotype and PROP sensitivity using a series of PROP solutions

given to both children and adults [13]. A gene-dosage effect was observed such

that two copies of the P allele conferred greater PROP sensitivity than a single

copy. A modest heterozygote effect was also apparent in another study where

AP heterozygotes had a higher PTC taste threshold than PP individuals and

were described as slightly less sensitive to PTC [9]. Duffy et al. [14] also found

that PROP bitterness varied significantly across genotypes with AA homozy-

gotes tasting less bitterness than either AP heterozygotes or PP homozygotes.

Using PTC paper, we evaluated the association between taster status and

TAS2R38 genotype with the A49P tag SNP. Subjects (n � 366) were asked to

NH

C

S

NH2

Fig. 1. Chemical structure of PTC.

El-Sohemy/Stewart/Khataan/Fontaine-Bisson/Kwong/Ozsungur/Cornelis 178

rate the bitterness of PTC paper on a scale of 1 (not at all bitter) to 9 (extremely

bitter), and were then categorized as either nontasters (1–3), tasters (4–6) or

supertasters (7–9). DNA was isolated from fasting blood samples and genotyp-

ing was performed using real-time PCR analysis for the A49P polymorphism.

In this multiethnic population, the frequency of the P allele was 55%. As

expected, the presence of the P allele was associated with bitter taste perception

(fig. 2). Among those with the AA genotype, 80% were considered nontasters,

17% tasters and 2% supertasters. For subjects who had the PP genotype, 11%

were nontasters, 40% were tasters and 49% were supertasters. Most heterozygotes

had the intermediate taster phenotype (52%), although some were classified as

either nontasters (26%) or supertasters (22%).

In addition to the TAS2R38 genotype, other factors such as age, sex and

ethnicity have been shown to modify PTC taste perception. Previous studies

have revealed that the frequency of PTC nontasters is 6–23% in China, 30% in

North American Caucasians and 40% in India [2, 5, 15, 16]. However, the mod-

ifying effects of ethnicity that have been reported could be due to population

differences in the frequency of the different TAS2R38 alleles. Indeed, in a multi-

ethnic population, we found the frequency of nontasters to be 12% in Asians,

31% in South Asians and 43% in Caucasians, and this was mainly due to differ-

ences in the frequencies of the TAS2R38 alleles. Nevertheless, other population

or cultural differences might modify the genotype-phenotype association

within each of the ethnicities. Such differences could also be responsible for

any differences in food preferences between different ethnic groups.

The ability to taste PTC or PROP is present in young children and declines

gradually with age [17]. Some individuals who are born sensitive to bitter-tasting

0

20

40

60

80

AA AP PP

Nontaster

Taster

Supertaster

TAS2R38 genotype

Fre

qu

en

cy (

%)

Fig. 2. Association between TAS2R38 genotype and taster status.


compounds may become less sensitive with age because of experience, aging or

disease [18, 19]. In a recent study, the genotype-phenotype association varied

by age with the relationship being stronger among children [13]. Moreover,

genotype modified their food preferences such that those with the AP or PP

genotypes preferred sweeter-tasting foods. Unlike the children, however, there

was no association between TAS2R38 genotypes and sweet preferences in the

adults [13]. Another study involving a population of women aged 60 years and

older found no association between the TAS2R38 genotype and food preference

[16]. The decline in taste sensitivity that occurs with age could have masked any

effects of genotype.

An indirect effect of the ability to taste bitter compounds has been

observed in the consequent avoidance of foods containing these bitter-tasting

substances [4, 7]. Many of these foods also have antioxidants, and thus the tast-

ing ability that leads to their avoidance has been implicated in the etiology of

common disorders. There are different classes of common dietary compounds

that are bitter, and might be particularly bitter to certain individuals, such as:

polyphenols, methylxanthines, isoflavones, flavonoids, glucosinolates and sul-

famides. Previous studies have observed an association between taste respon-

siveness to PROP and food preferences. Drewnowski et al. [20] found an

inverse relationship between PROP sensitivity and acceptance of cruciferous

vegetables and tart citrus fruits in young women. A second observational study

of young women by the same group reported an association between perception

of the bitter taste of PROP and reduced preferences for Brussels sprouts, cab-

bage, spinach, and coffee [21]. They also observed greater sensitivity to

PROP among female breast care patients who had a lower acceptance for bitter

taste [22].

Despite the numerous associations reported between PROP sensitivity and

acceptance of bitter foods, fewer studies have examined the relationship

between PROP sensitivity and food consumption. A cross-sectional study by

Kaminski et al. [23] found no direct relationship between PROP taster status

and the frequency of consumption of 22 bitter food items. However, Basson

et al. [24] observed that men who found PROP to be very bitter consumed fewer

vegetables. In a recent study, vegetable preference was found to be a direct pre-

dictor of intake and those who perceived PROP as very bitter consumed vegeta-

bles less frequently [25]. Thus, the perception of bitter compounds appears to

directly influence dietary habits such as vegetable intake. It has been hypothe-

sized that supertasters eat less of the healthy vegetables due to an increased sen-

sitivity to their bitterness, and eat more sweets or fatty foods that are associated

with an increased risk of cardiovascular disease. However, there is evidence that

individuals who are sensitive to bitter-tasting vegetables are also more sensitive

to sweet foods [8, 26, 27]. Thus, tasters and supertasters may have a stronger


taste acuity in general, and their heightened taste perceptions prevent overcon-

sumption of a variety of foods.

Because the TAS2R38 gene has recently been linked to bitterness taster sta-

tus, only a few studies have examined the link between this genotype and food

preferences [13, 16]. Timpson et al. [16] were the first to examine the effect of

genetic variation in taste on eating behavior and the risk of diet-related chronic

disease. They examined the relationship between TAS2R38 haplotypes, coro-

nary heart disease, coronary heart disease risk factors, and food consumption

behaviors in postmenopausal women. No significant associations were observed

between TAS2R38 haplotypes and either coronary heart disease traits or food

consumption. However, they did observe a marginally lower risk of diabetes

among those with the nontaster genotype. This suggests that these individuals

may have been more likely to consume a diet rich in bitter-tasting vegetables. In

that study, there was no direct association between TAS2R38 genotypes and

food preferences. However, the population consisted of elderly women and the

effect of age on taste perception may have masked an association between

genotype and food preference. We are currently examining the relationship of

PTC taster status and TAS2R38 genotype to food preferences and food intake in

a multiethnic population of subjects 20–29 years of age. The T2R gene family

consists of about 25 different members, each with its unique ability to perceive

the different tastes of diverse dietary compounds [3]. Genetic polymorphisms

in these other taste receptors may also be important determinants of preferences

for particular foods. Recently, an SNP in the TAS2R50 gene was associated with

an increased risk of myocardial infarction suggesting that it may be associated

with adverse dietary habits [28]. Further studies will be needed to determine

whether TAS2R50 genotypes affect food preferences and intake.

Conclusion

Taste is one of the most important determinants of food preferences and

genetic differences in taste perception may explain individual differences in

eating habits. Sensitivity to bitter-tasting compounds has been associated with a

lower preference and consumption of bitter-tasting foods. Although removal of

bitter compounds from certain foods may increase their palatability, it could

also decrease their nutritional benefit since many bitter-tasting phytochemicals

may have health-promoting properties. The bitter taste can also be masked by

the addition of salt, sugar or fat, but this leads to the increased consumption of

dietary factors that have been associated with an increased risk of chronic dis-

ease. The development of ‘bitter blockers’ may be a better strategy to enhance

the taste of some vegetables that are perceived to be bitter by certain individuals


[29]. Identifying populations with preferences for particular flavors or foods

may lead to the development of novel food products targeted to specific geno-

types or ethnic populations.

Acknowledgements

This research was supported by a grant from the Advanced Foods and Materials

Network (M&E-B-4). A. E. holds a Canada Research Chair in Nutrigenomics.

References

1 Drewnowski A, Rock CL: The influence of genetic taste markers on food acceptance. Am J Clin

Nutr 1995;62:506–511.

2 Tepper BJ: 6–n-Propylthiouracil: a genetic marker for taste, with implications for food preference

and dietary habits. Am J Hum Genet 1998;63:1271–1276.

3 Kim UK, Drayna D: Genetics of individual differences in bitter taste perception: lessons from the

PTC gene. Clin Genet 2004;67:275–280.

4 Bartoshuk LM, Duffy VB, Miller IJ: PTC/PROP tasting: anatomy, psychophysics, and sex effects.

Physiol Behav 1994;56:1165–1171.

5 Guo SW, Reed DR: The genetics of phenylthiocarbamide perception. Ann Hum Biol 2001;28:

111–142.

6 Drewnowski A, Henderson SA, Shore AB, Barratt-Fornell A: Nontasters, tasters, and supertasters

of 6–n-propylthiouracil (PROP) and hedonic response to sweet. Physiol Behav 1997;62:649–655.

7 Reed DR, Bartoshuk LM, Duffy V, Marino S, Price RA: Propylthiouracil tasting: determination of

underlying threshold distributions using maximum likelihood. Chem Senses 1995;20:529–533.

8 Chang WI, Chung JW, Kim YK, Chung SC, Kho HS: The relationship between phenylthiocar-

bamide (PTC) and 6–n-propylthiouracil (PROP) taster status and taste thresholds for sucrose and

quinine. Arch Oral Biol 2006;51:427–432.

9 Kim UK, Jorgenson E, Coon H, Leppert M, Risch N, Drayna D: Positional cloning of the human

quantitative trait locus underlying taste sensitivity to phenylthiocarbamide. Science 2003;299:

1221–1225.

10 Wooding S, Kim UK, Bamshad MJ, Larsen J, Jorde LB, Drayna D: Natural selection and molecu-

lar evolution in PTC, a bitter-taste receptor gene. Am J Hum Genet 2004;74:637–646.

11 Drayna D: Human taste genetics. Annu Rev Genomics Hum Genet 2005;6:217–235.

12 Bufe B, Breslin PA, Kuhn C, Reed DR, Tharp CD, Slack JP, et al: The molecular basis of individ-

ual differences in phenylthiocarbamide and propylthiouracil bitterness perception. Curr Biol

2005;15:322–327.

13 Mennella JA, Pepino MY, Reed DR: Genetic and environmental determinants of bitter perception

and sweet preferences. Pediatrics 2005;115:e216–e222.

14 Duffy VB, Davidson AC, Kidd JR, Kidd KK, Speed WC, Pakstis AJ, et al: Bitter receptor gene

(TAS2R38), 6–n-propylthiouracil (PROP) bitterness and alcohol intake. Alcohol Clin Exp Res

2004;28:1629–1637.

15 Tepper BJ, Ullrich NV: Taste, smell, and the genetics of food preferences. Top Clin Nutr 2002;17:

1–14.

16 Timpson NJ, Christensen M, Lawlor DA, Gaunt TR, Day IN, et al: TAS2R38 (phenylthiocar-

bamide) haplotypes, coronary heart disease traits, and eating behavior in the British Women’s

Heart and Health Study. Am J Clin Nutr 2005;81:1005–1011.

17 Whissell-Buechy D: Effects of age and sex on taste sensitivity to phenylthiocarbamide (PTC) in

the Berkeley Guidance sample. Chem Senses 1990;15:39–57.


18 Cowart BJ, Yokomukai Y, Beauchamp GK: Bitter taste in aging: compound-specific decline in

sensitivity. Physiol Behav 1994;56:1237–1241.

19 Bartoshuk LM, Duffy VB, Reed D, Williams A: Supertasting, earaches and head injury: genetics

and pathology alter our taste worlds. Neurosci Biobehav Rev 1996;20:79–87.

20 Drewnowski A, Henderson SA, Shore AB, Barratt-Fornell A: Sensory responses to 6–n-propylth-

iouracil (PROP) or sucrose solutions and food preferences in young women. Ann NY Acad Sci

1998;855:797–801.

21 Drewnowski A, Henderson SA, Levine A, Hann C: Taste and food preferences as predictors of

dietary practices in young women. Public Health Nutr 1999;2:513–519.

22 Drewnowski A, Henderson SA, Hann CS, Berg WA, Ruffin MT: Genetic taste markers and pref-

erences for vegetables and fruit of female breast care patients. J Am Diet Assoc 2000;100:

191–197.

23 Kaminski LC, Henderson SA, Drewnowski A: Young women’s food preferences and taste respon-

siveness to 6–n-propylthiouracil (PROP). Physiol Behav 2000;68:691–697.

24 Basson MD, Bartoshuk LM, Dichello SZ, Panzini L, Weiffenbach JM, Duffy VB: Association

between 6–n-propylthiouracil (PROP) bitterness and colonic neoplasms. Dig Dis Sci 2005;50:

483–489.

25 Dinehart ME, Hayes JE, Bartoshuk LM, Lanier SL, Duffy VB: Bitter taste markers explain vari-

ability in vegetable sweetness, bitterness, and intake. Physiol Behav 2006;87:304–313.

26 Duffy VB: Associations between oral sensation, dietary behaviors and risk of cardiovascular dis-

ease (CVD). Appetite 2004;43:5–9.

27 Lucchina LA, Curtis OF 5th, Putnam P, Drewnowski A, Prutkin JM, Bartoshuk LM: Psychophysical

measurement of 6–n-propylthiouracil (PROP) taste perception. Ann NY Acad Sci 1998;855:

816–819.

28 Shiffman D, Ellis SG, Rowland CM, Malloy MJ, Luke MM, Iakoubova OA, et al: Identification of

four gene variants associated with myocardial infarction. Am J Hum Genet 2005;77:596–605.

29 McGregor R: Taste modification in the biotech era. Food Technol 2004;58:24–30.

Dr. A. El-Sohemy

Department of Nutritional Sciences, Faculty of Medicine

University of Toronto, 150 College St., Room 350

Toronto, Ont. M5S 3E2 (Canada)




Prospects for Improving the NutritionalQuality of Dairy and Meat Products

Shaun G. Coffey

CSIRO Livestock Industries, St. Lucia, Australia

AbstractKnowledge of the function of human and animal genes and their interactions is rapidly

increasing as a result of the completion of sequencing efforts for the human, bovine and other

genomes. Through transcriptomics, proteomics and metabolomics, we have the capacity to

study the health effects of food compounds at the molecular level. The same tools that can

assist the understanding of nutrigenomics in humans can also be applied to producing

animal-derived foods with desired capacities to alter gene expression in humans. This, essen-

tially, represents food taking another major step in value through the personalisation of

health and nutrition. In its own right, nutrigenomics offers the potential to improve animal

production enterprises through major health and productivity gains.


Food derived from animals is an important source of nutrients in diets, and

the demand for high-quality meat and dairy products will rise over the next 20 years,

accompanied by more stringent quality assurance requirements. Nutrigenomic

approaches offer the potential to develop enhanced systems that overcome

some of the existing limitations of production [Andersen et al., 2005; Givens,

2005]. Discoveries associated with sequencing of the human and animal

genomes, and the emergence of related technologies provide an exciting oppor-

tunity to develop a deeper understanding of nutrition at both the molecular and

systems level, thus opening the possibility of major transformation in both

human nutrition and animal production. As advances in animal genomics start

to match pace with human studies, the rate of progress in nutrigenomics can be

expected to increase.

This paper briefly considers the contribution of animals to human diets,

and hence to health and nutrition; outlines recent developments in animal

genomics, proteomics, metabolomics and transcriptomics, and explores the

Coffey 184

possibilities for manipulating the nutrient profile of animal products designed

for human consumption as well as the application of nutrigenomics to improv-

ing animal production itself. If the promise of personalised food [German and

Watzke, 2004] is to be met, a considerable research effort is needed.

Animal Products in Human Diets

Consumption of animal products has grown rapidly over the last 40 years

[WHO/FAO, 2003] and is projected to continue to grow (tables 1, 2), especially

in developing and transition economies [Delgado et al., 1998]. Animal products

are a significant source of high-quality protein, calcium and iron, but have also

Table 1. Per capita global consumption of meat: actual and projected

(kg per capita per year)

Region 1964–1966 1997–1999 2030

World 24.2 36.4 45.3

Developing countries 10.2 25.5 36.7

Transition countries 42.5 46.2 60.7

Industrialised countries 61.5 88.2 100.1

East Asia 8.7 37.7 58.5

South Asia 3.9 5.3 11.7

WHO/FAO [2003].

Table 2. Per capita global consumption of milk: actual and projected

(litres per capita per year)

Region 1964–1966 1997–1999 2030

World 73.9 78.1 89.5

Developing countries 28.0 44.6 65.8

Transition countries 156.7 159.1 178.7

Industrialised countries 185.5 212.2 221.0

East Asia 3.6 10.0 17.8

South Asia 37.0 67.5 106.9

WHO/FAO [2003].

Improving Nutritional Quality of Food 185

been identified as a major source of excessive saturated fat intakes. Animal-

derived food contributes around 30% of the total energy input, over 50% of

which is derived from fat. Milk and dairy products have made a major contribu-

tion to saturated fatty acid consumption, a topic that serves as a useful example

of the possibilities for nutrigenomics.

Dietary Fat and Chronic Disease: Exploring the Possibilities for Nutrigenomics

Givens [2005] has summarised the current knowledge of the relationships

between dietary fatty acids and chronic disease. Concern over some animal-

derived foods, especially saturated fatty acids (SFAs), have led to concerns

about the contribution of these foods to an increased risk of cardiovascular dis-

ease and the metabolic syndrome. Undoubtedly, total consumption of SFAs

needs to be reduced, and significant benefits are to be gained by replacing them

with monounsaturated fatty acids (MUFAs) and polyunsaturated fatty acids

(PUFAs). As animal products currently make a contribution to SFA consump-

tion, this situation needs to change.

Fortunately, fatty acid composition of animal products is highly responsive

to nutrition in the animal, and a sizeable research effort is being made to lower

SFA, and raise MUFA and PUFA levels in milk and meat. This effort aims to

retain the other inherent nutritional values of these foods.

Manipulations possible in milk and meat serve as an illustration of this point.

Fatty acid in milk originates through a complex process that responds to several

nutritional approaches to enable manipulation of fatty acid composition. For

example, supplements of plant oils, or oilseeds (such as canola, soya bean and

sunflower) reduce both short- and medium-chain fatty acids in milk, reducing

SFAs and raising MUFAs and PUFAs [Givens, 2005] (although it should be noted

that PUFAs are not synthesised in any appreciable amounts in ruminant tissues,

and therefore, concentrations in milk reflect levels of PUFAs leaving the rumen).

Conjugated linoleic acid concentrations in milk have been shown to be

higher in milk fat from cows offered fresh forages as compared with conserved

forages, and are also enhanced by the use of supplements of oilseeds and fish

oils. Conjugated linoleic acid is of interest because of its potential anticancer

effects.

In meats derived from ruminant animals, efforts have been made to

increase the ratio of PUFAs to SFAs and enhancing n–3 PUFA, tasks that can be

achieved again through nutrition, using high-forage-based diets or oil supple-

ments rich in PUFAs. The fatty acid profile in non-ruminant meat, in contrast,

is essentially a reflection of that in the diet.

Coffey 186

The evidence is clear that the nutrient profile of animal products is highly

responsive to nutrition. Feeding strategies offer the possibility of changes in

production systems; however, we need a better understanding of the impact of

the diet.

Andersen et al. [2005] illustrate that feeding strategies have a regulatory

effect on the biological processes in muscle, which is directly reflected in the

quality of the meat. The fact that specific diet components, inter alia, regulate

gene expression is well recognised, but few studies have explored the complex

interactions between individual nutrients on the genome. The ongoing mapping

of the genomes of farm animals together with advances in bioinformatics and

molecular biology will undoubtedly accelerate progress in the future.

The Challenges to Animal Production Systems

Two challenges face animal agriculture: first, the unravelling of the

specific interactions of food on health and nutrition at the molecular level

and second, the need to study the constellation of changes that take place in

the nutritional environment where impacts are not necessarily additive (a sys-

tems biology approach). These are equivalent to the challenges in the human

domain.

Nutrigenomic approaches are being explored to ensure higher standards

for the quality of meat and dairy products for consumption. Examples will be

discussed in relation to exploiting basic and empirical understandings of physi-

ological and physical processes. These can then be placed in a systems biology

context as an aid to decision making by producers. A change of focus in pro-

duction systems towards an understanding of how feed influences biological

mechanisms and impacts quality traits, for example, may lead to the production

of a diverse range of products with quite specific nutritional attributes, and tar-

geted market segments.

In animal production systems, this requires a change from classical nutri-

tion research (where all test individuals are treated as genetically identical) to

one where responses to diet are analysed with an awareness of individual, age

and genotype interactions.

Nutrigenomics and systems biology combined offer the opportunity to

transform both the nutritional value and the overall utility of animal-derived

human foods. These will require major changes to farming systems, a topic

beyond the scope of this paper. Suffice it to say that as individual diets are

selected to cater for individual human dispositions, animal agriculture will

experience pressure to continue to move further away from the production of

food as a bulk commodity: a prospect with far-reaching impacts.


General Progress with Animal Genomes and beyond

The advent of significant progress in genomics, proteomics, metabolomics

and transcriptomics also allows a change of focus in our study of animals. The

focus is now shifting towards an understanding of how animal diets, feeding

regimes, production conditions and handling strategies influence biological

mechanisms and the outcome of these in relation to specific meat, milk and

other product quality parameters. In turn, the more fundamental understanding

of, for example, muscle physiological and physical processes, and their interac-

tions in relation to gene expression and environmental constraints will enable

full exploitation of systems biology approaches. These will shape future man-

agement strategies in animal production.

With the completion of the human genome, science has begun to unravel

and understand human biological complexity through the emerging fields of

proteomics and pharmacogenetics [Kauwell, 2005]. Progress in animal genome

studies is not as well advanced [Hendersen et al., 2005] but is making rapid

progress [Dalrymple, 2005]. The mouse, rat and dog genomes are now avail-

able; the bovine genome is close to completion, and several others are at

advanced stages of sequencing (table 3).

Having the genome sequences available is important because they open the

way to address research questions, to undertake genome-wide analyses and to

allow comparative studies across the increasing number of available genome

sequences. This work forms the basic platform on which we can develop

nutrigenomics in the animal production enterprise.

Mammalian Genomics and the Bovine Genome

The Commonwealth Scientific and Industrial Research Organisation

(CSIRO) has taken a leading role in functional genomics involving a number of

Table 3. Progress in sequencing animal genomes

Species Status

Mouse, dog, rat, chicken (red jungle fowl) completed

Bovine, chimpanzee, opossum, rhesus macaque advanced draft

Orangutan, marmoset, elephant, shrew, hedgehog, draft

guinea pig, tenrec, armadillo, rabbit, cat, wallaby, platypus

Dalrymple [2005] and Moore et al. [2005].

Coffey 188

international research organisations in the International Bovine BAC Mapping

Consortium and the International Bovine Genome Project.

A significant reason for our decision to participate in the Bovine Genome

Project was to ensure that the whole sequence of a major production species be

available in the public domain for use by all interested research agencies. The

first draft of the bovine sequence was released in October 2004. This first draft

already provided access to increased numbers of microsatellite markers and

randomly generated single nucleotide polymorphisms (SNPs), facilitating the

identification of genetic polymorphisms associated with valuable traits in

Qualitative Trait Loci and other studies [Barendse, 2005]. The third draft was

due for release in early 2006, and will greatly enhance the quality of informa-

tion available. CSIRO has built a number of databases from this work, some of

which are publicly available (http://www.livestockgenomics.csiro.au).

Significant progress has been made as a result of early access to the very

large amount of ordered sequence covering most of the protein-coding regions

of the genome. Among our early projects was the design and implementation of

an interactive bovine in silico SNP database, an activity started in 2001. This

project identified the need for the production of SNPs based on the bovine

expressed sequence tag collections and thus also for the clustering and annota-

tion of the clustered transcripts. The analysis pipeline and results are described

in detail in Hawken et al. [2004]. Significant progress has been made in tran-

scriptomics [Carninici et al., 2005] in recent times.

CSIRO has also developed the first bovine cDNA-based microarray for the

analysis of muscle and fat. This consisted of 9,600 elements derived from 2,000

expressed sequence tags and 73,000 anonymous cDNA clones [Lehnert et al.,

2004]. This array has been used in a variety of experiments with muscle and fat

samples from cattle of different genotypes or undergoing a range of treatments

[Byrne et al., 2005; Reverter et al., 2003, 2004, 2005a, c; Wang et al., 2005a, b].

Combined, these studies have started us down the track of systems biology

[Reverter et al., 2005b]. We have related work in the ovine [Dalrymple, 2005]

and chicken domains [Moore et al., 2005] although the latter is focussed on dis-

ease and disease management.

Genome Information: Current Limits

The genome sequences of domestic animals and particularly the bovine

sequence are incredibly valuable, as they will provide the complete repertoire of

genes that are present in these species as well as the means to measure the tran-

scriptional activities of these genes in different tissues and under different nutri-

tional regimes.


In addition, it will be increasingly possible to associate individual variation

in phenotype with individual variation in genome sequence in a manner that

allows selection for optimally performing animals in particular production

environments.

That being said, we are far from understanding all that is contained within

a genome sequence. The gene repertoires of most mammals examined to date

are very similar. Intriguingly, the actual protein-encoding genes in a mam-

malian genome only account for approximately 1.5% of the genome sequence.

The remaining 98.5% have not been thoroughly explored but may contain com-

plex gene expression regulatory information. We currently have a poor view of

how gene expression is controlled and how multiple genes contribute to com-

plex phenotypes. So what makes a mouse a mouse and a human a human? The

answer probably lies with species-specific regulatory mechanisms that control

the spatial and temporal expression patterns of genes. A significant component

of this system undoubtedly involves species-specific epigenetic modifications

of DNA and histones.

Perhaps one of the most significant advances in recent years has been the

realization that epigenetic modifications controlling gene expression can be

influenced by nutrition. At a higher level, it is in the interactions between the

gene products that lies hidden-layer regulatory information that we are just

starting to appreciate. This presents significant technical challenges, as at pre-

sent we do not understand the rules governing the non-linear dynamics of these

complex biological systems. This is a fertile ground for fundamental investiga-

tions that may lead to a better understanding of the ways in which nutrition

impacts on gene expression.

Timing of Production Interventions

Perhaps one of the most significant advances in nutrigenomics over the last

decade has been the realization that nutritional restrictions on an animal or fetus

during critical developmental periods can induce permanent change in gene

expression programs and hence the phenotype even on removal of the restric-

tion. One of the best examples of this phenomenon is the observation that nutri-

tional constraint in a fetus in utero can result in a predisposition toward obesity

of that individual as an adult according to the Barker hypothesis [Kwong et al.,

2004; Wilson, 1999]. Presumably the physiological ‘set points’ of the fetus are

altered in utero (fetal programming) to reflect the expected harsh environment

that will meet the newborn animal. The animal’s metabolism is therefore geared

toward accumulation and storage of energy and in the absence of the harsh envi-

ronment this results in the accumulation of adipose tissue.

Coffey 190

Another example involves nurturing. A newborn rat that is not sufficiently

licked, groomed and nurtured by its mother shows epigenetic alterations of the

gene encoding the glucocorticoid receptor. This protein plays a key role in the

response of the rat to stress. In the poorly nurtured pups, the gene is partially

silenced by these epigenetic modifications resulting in pups with increased lev-

els of stress hormones and a decreased ability to explore new environments.

These effects are permanent [Weaver et al., 2004].

This persistence of phenotypic traits related to diet, variously described as

nutritional ‘imprinting’ [Levin, 2000] or nutritional programming [Singal

et al., 2003], points to the fact that biological organisms have a variety of

mechanisms to respond to the environment to which they are exposed. They

adapt through a variety of structural, biochemical and regulatory processes,

and respond in a complex manner to factors such as physical forces, blood sup-

ply, extracellular matrix macromolecules and growth factors [Harper and

Pethick, 2004].

For production animals, these findings offer a significant potential for

altering production systems to enable specific qualities of animal-derived

foods if there are suitable epigenetic pathways for this to happen. By interven-

ing at critical points in the growth/life cycle of the animal, the phenotype

might be controlled to the extent that the ultimate product can be specified

more closely.

Prospective Targets for Nutrigenomics in Animal Production

Nutritional genomics has many prospects for applications in animal pro-

duction. The full range of applications goes well beyond just altering the nutri-

ent profile of food for human consumption. Animal scientists are already

responding to the consumer demands for high standards of safety, nutritional

value, texture and flavour, tenderness, water-holding capacity, colour and lipid

content, lipid composition, oxidative stability and uniformity. Included within

the CSIRO portfolio, for example, are:

• assessment of food safety and food quality [see for example Bettleheim

et al., 2005; Gilbert et al., 2005];

• developing new biomarkers of nutrient exposure and disease risk

[Donaldson et al., 2005; Ingham et al., 2005; Wang et al., 2005a, 2005b;

Strandberg et al., 2005];

• studies of cellular effects of nutrients and bioactive compounds [Krause

et al., 2005; Lai et al., 2005; Pan et al., 2005];

• developing better methods to examine data sets related to genomics and

nutrigenomics [Reverter et al., 2005b];


• understanding the effects of functional foods [Dunshea et al., 2005; Liu

and Eady, 2005], and methods to manipulate food composition [Kitessa

et al., in press];

• understanding the biology and development of the muscle [Harper and

Pethick, 2004; Pethick et al., 2004; Vuocolo et al., 2005];

• using gene expression profiling to examine stress and meat quality [Kerr

and Hines, 2005; Moser et al., 2004];

• an expanded understanding of the gut through the application of metage-

nomics to enable a holistic and mechanistic analysis of the genetic and meta-

bolic potential of entire microbial communities [Morrison et al., 2005].

These, and other areas of applications, will develop quickly over the next

3–5 years largely as a result of the current investment in postgenomic technolo-

gies. A significant challenge also lies in developing a more detailed and fuller

knowledge of the full range and role of bioactives so that we might understand,

for example, the apparent protective effect of increased consumption of milk

against cardiovascular disease [Givens, 2005].

Transgenic Animals

One avenue to changing the nutrient profile of animal products is through

the use of transgenesis to transfer to food animal species traits from other

species. Seymour et al. [2004] consider that transgenics is unlikely to be part of

animal production in the near future, but is likely to be important in the longer

term. Consumer acceptance currently is a major issue, but may dissipate because

of the significant consumer benefit deriving from its use. Table 4 outlines some

of the possible targets and benefits including those of nutrigenomic interest.

As our targeting of benefits improves, we can be more confident that

transgenesis will become increasingly accepted for food products, as it already

is in the production of therapeutic drugs. Many groups, including our own [see

Adams and Briegel, 2005], have developed and studied transgenic animals, so

we know that the technology is assessable. Harper et al. [2003] survey existing

knowledge on transgenic food animals and provide an excellent source of infor-

mation on this topic.

Conclusions

The integrative nature of the discipline of nutrition has long distinguished

the field of study. The advent of genomics, and the growing capacity of our new

research tools to enable a deeper understanding of function at both the subcellular

Coffey 192

Table 4. Survey of transgenic livestock species, existing or predicted, that should be of interest to Australian

or New Zealand animal food producers in the next 5 years

Animal Genes introduced or deleted Performance criteria (consumer benefit)

Bovine � and � casein increased expression of casein proteins (improved protein

content of milk)

Bovine intestinal lactase reduction of lactose in the milk (lactose-intolerant people)

Bovine lysostaphin mastitis resistance (reduced use of antibiotics)

Bovine �-lactoglobulin increased production of this protein in milk, as well as

increased growth and disease resistance in calves feeding

on the milk (reduced antibiotic use and improved health

benefits)

Ovine growth hormone increased growth rates, increased feed conversion efficiency,

decreased carcass fatness, and increased lactation (leaner meat)

Ovine myostatin reduced myostatin expression and increased muscle in sheep

(leaner meat)

Porcine insulin-like growth factor 1 increased growth rate and reduced carcass fatness (leaner meat)

Porcine bovine �-lactalbumin increased growth rate and improved health of piglets (unknown

consumer value)

Porcine spinach stearoyl-CoA modified lipid composition (increased unsaturated fats)

desaturase

Porcine phytase expressed in saliva utilisation of phosphorus bound to phytate by the pig, and

hence a reduction in waste phosphorous (environmental

impact reduced)

Caprine lysostaphin cure or prevention of Staphylococcus aureus mastitis

(reduced antibiotic use)

Caprine rat stearoyl-CoA desaturase modified milk fat composition (increased unsaturated fatty

acid proportion)

Caprine human lysozyme modified milk fat composition and enhanced immune

responses (reduced antibiotic use and modified milk)

Common carp growth hormone increased growth rate (cheaper fish)

Chinook salmon growth hormone increased growth rate (cheaper fish)

Silver sea bream growth hormone increased growth rate (cheaper fish)

Japanese abalone growth hormone increased growth rate (cheaper fish)

Rainbow trout growth hormone increased growth rate (cheaper fish)

Seymour et al. [2004].

and the whole-system level challenge both animal and human research to

understand the complex relationship between genome and diet.

Application of nutritional genomics in agriculture is a rapidly growing

field and offers the prospect of designing effective dietary and management

regimes to improve animal production and better manage disease status. The


same principles can be applied to design and produce animal-derived food for

specific functions in the human diet.

As a still-emerging endeavour, nutrigenomics requires a significant

research effort, especially at the more basic end of the spectrum. Within animal

production, however, excellent potential exists, and early results support the

belief that animal products will have an increasingly significant impact on

‘designer’ human health and nutrition.

Acknowledgements

I would like to recognise Ross Tellam, Alan Brownlee, Gene Wijffels, Chris Prideaux,

Soressa Kitessa, Peter Willadsen, Bill Barendse and Greg Harper for many useful discus-

sions. I also acknowledge our partners in the International Bovine Genome Project, and espe-

cially Steve Kappes of the USDA. My many colleagues in CSIRO Livestock Industries

continue to pursue the best in bioscience and technology, and are a continuous source of

inspiration.

The partners in the Bovine Genome Project are: the National Human Genome Research

Institute (USA); CSIRO (Australia); the US Department of Agriculture; the State of Texas;

Genome Canada; Agritech Investments Limited (New Zealand); Dairy Insight Ltd. (New

Zealand); AgResearch Ltd. (New Zealand); the Kleberg Association (USA); the National,

Texas and South Dakota Beef Councils (USA).

The sequencing of the bovine genome is being undertaken by the Baylor College of

Medicine’s Human Genome Sequencing Center (USA) and the British Columbia Cancer

Agency (Vancouver, Canada).

References

Adams NR, Briegel JR: Multiple effects of an additional growth hormone gene in adult sheep. J Anim

Sci 2005;83:1868–1874.

Anderson HJ, Oksbjerg N, Young JF, Therkildsen M: Feeding and meat quality – a future approach. Meat

Sci 2005;70:543–554.

Barendse W: The transition from quantitative trait loci to diagnostic test in cattle and other livestock.

Aust Jo Exp Agric 2005;45:831–836.

Bettleheim KA, Kuzevski A, Gilbert RA, Krause DO, McSweeney CS: The diversity of Escherichia coliserotypes and biotypes in cattle faeces. J Appl Microbiol 2005;98:699–709.

Bryne KA, Wang YH, Lehnert SA, Harper GS, McWilliam SM, Bruce HL, Reverter A: Gene expression

profiling of muscle tissue in Brahman steers during nutritional restriction. J Anim Sci 2005;83:

1–12.

Carninci P, et al: The transcriptional landscape of the mammalian genome. Science 2005;309:1559–1563.

Dalrymple BP: Harnessing the bovine genome sequence for the Australian cattle and sheep industries.

Aust J Exp Agric 2005;45:1011–1016.

Delgado CL, Courbois CC, Rosegrant ML: Global Food Demand and the Contribution of Livestock as

We Enter the New Millennium. Washington, International Food Policy Research Institute, 1998.

Donaldson L, Vuocolo T, Gray C, Strandberg Y, Reverter A, McWilliam SM, Wang Y, Byrne K,

Tellam R: Construction and validation of a Bovine Innate Immune Microarray. BMC Genomics

2005;6:135.

Coffey 194

Dunshea FR, D’Souza DN, Pethick DW, Harper GS, Warner RD: Effects of dietary factors and metabolic

modifiers on quality and nutritional value of meat. Meat Sci 2005;71:8–38.

German JB, Watzke HJ: Personalizing foods for health and delight. Compr Rev Food Sci Food Saf

2004;3:145–151.

Gilbert RA, Tomkins N, Padmanabha J, Gough JM, Krause DO, McSweeney CS: Effect of finishing

diets on Escherichia coli populations and prevalence of enterohaemorrhagic E. coli virulence

genes in cattle faeces. J Appl Microbiol 2005;99:885–894.

Givens DI: The role of animal nutrition in improving the nutritive value of animal-derived foods in rela-

tion to chronic disease. Proc Nutr Soc 2005;64:395–402.

Harper GS, Brownlee A, Hall TE, Seymour R, Lyons R, Ledwith P: Global progress toward transgenic

food animals: a survey of publicly available information. A report commissioned by Food

Standards Australia and New Zealand as an authoritative overview of the transgenic animal field.

2003. Available at www.foodstandards.gov.au.

Harper GS, Pethick DW: Viewpoint: how might marbling begin? Aust J Exp Agric 2004;44:653–662.

Hawken RJ, Barris WC, McWilliam SM, Dalrymple BP: An interactive bovine in silicio SNP database

(IBISS). Mamm Genome 2004;15:819–827.

Henderson D, Thomas M, Yang D: Conference review: bovine genomics from academia to industry.

Comp Funct Genomics 2005;6:174–180.

Ingham AB, Sproat KW, Tizard MLV, Moore RJ: A versatile system for the expression of nonmodified

bacteriocins in Escherichia coli. J Appl Microbiol 2005;98:676–683.

Kauwell GPA: Emerging concepts in nutrigenomics: a preview of what is to come. Nutr Clin Pract

2005;20:75–87.

Kerr CA, Hines BM: The potential of new genetic technologies in selecting for stress resistance in pigs.

Aust J Exp Agric 2005;45:775–782.

Kitessa SM, Peake D, Bencini RC, Vercoe PE, Williams AJ, Murray K: Change in conjugated linoleic acid

content of milk fat during lactation in different breeds of dairy sheep. Aust J Exp Agric, in press.

Krause DO, Smith WJM, Brooker JD, McSweeney CS: Tolerance mechanisms of streptococci to

hydrolysable and condensed tannins. Anim Feed Sci Technol 2005;121:59–75.

Kwong WY, Osmond C, Fleming TP: Support for Barker hypothesis upheld in rat model of material

under nutrition during the preimplantation period: application of integrated ‘random effects’ sta-

tistical model. Reprod Biomed Online 2004;8:574–576.

Lai YP, Peng YF, Zuo Y, Li J, Huang J, Wang LF, Wu ZR: Functional and structural characterization

of recombinant dermcidin-1L, a human antimicrobial peptide. Biochem Biophys Res Commun

2005;328:243–250.

Lehnert SA, Wang YH, Byrne KA: Development and application of a bovine cDNA microarray for

expression profiling of muscle and adipose material. Aust Jo Exp Agric 2004;44:1127–1133.

Levin BE: Metabolic imprinting on genetically predisposed neural circuits perpetuates obesity. Nutrition

2000;16:909–915.

Liu SM, Eady SJ: Glutathione: its implications for animal health, meat quality and health benefits of

consumers. Aust J Agric Res 2005;56:775–780.

Moore RJ, Doran TJ, Wise TG, Riddell S, Granger K, Crowley TM, Jenkins KA, Karpala AJ, Bean AGD,

Lowenthal JW: Chicken functional genomics. Aust J Exp Agric 2005;45:459–556.

Morrison M, Adams SE, Nelson KE, Attwood GT: Metagenomic analysis of the microbiomes in rumi-

nants and other herbivores; in Makkar HPS, McSweeney CS (eds): Methods in Gut Microbial

Ecology for Ruminants. Dordrecht, Springer Science, 2005.

Moser RJ, Reverter A, Kerr CA, Beh KJ, Lehnert SA: A mixed-model approach for the analysis of cDNA

microarray gene expression data from extreme-performing pigs after infection with Actinobacilluspleuropneumoniae. J Anim Sci 2004;82:1261–1271.

Pan Y, Wan J, Roginski H, Lee A, Shiell BJ, Michalski WP, Coventry J: Effect of chemical modification on

anti-microbial and anti-viral properties of milk proteins. Aust J Dairy Technol 2005;60: 149–151.

Pethick DW, Harper GS, Oddy VH: Growth, development and nutritional manipulation of marbling in

cattle: a review. Aust J Exp Agric 2004;44:705–715.

Reverter A, Barris W, McWilliam SM, Bryne KA, Wang YH, Tan SH, Hudson N, Dalrymple BP:

Validation of alternative methods of data normalization in gene co-expression studies.

Bioinformatics 2005a;21:1112–1120.


Reverter A, Barris W, Moreno-Sanchez N, McWilliam SM, Wang YH, Harper GS, Lehnert SA,

Dalrymple BP: Construction of gene interaction and regulatory networks in bovine skeletal mus-

cle from expression data. Aust J Exp Agric 2005b;45:821–829.

Reverter A, Byrne KA, Bruce HL, Wang YH, Dalrymple BP, Lehnert SA: A mixture model-based clus-

ter analysis of cDNA mircoarray gene expression data on Brahman and Brahman composite steers

fed high-, medium- and low-quality diets. J Anim Sci 2003;81:1900–1910.

Reverter A, McWilliam SM, Barris W, Dalrymple BP: A rapid method for computationally inferring

transcriptome coverage and microarray sensitivity. Bioinformatics 2005c;21:80–89.

Reverter A, Wang YH, Byrne KA, Tan SH, Harper GS, Lehnert SA: Joint analysis of multiple cDNA

microarray studies via multivariate mixed-models applied to genetic improvement of beef cattle.

J Anim Sci 2004;82:3430–3439.

Seymour RB, Lyons RE, Hall TE, Brownlee A, Ledwith P, Harper GS: Transgenic food animals for

Australia and New Zealand. Animal production in Australia. Proc Aust Soc Anim Prod 2004;25:

204–207.

Shingal A, Wells J, Cole TJ, Fewtrell M, Lucas A: Programming of lean body mass: a link between birth

weight, obesity, and cardiovascular disease. Am J Clin Nutr 2003;77:726–730.

Strandberg Y, Gray C, Vuocolo T, Donaldson L, Broadway M, Tellam R: Lipopolysaccharide and lipotei-

choic acid induce different innate immune responses in bovine mammary epithelial cells. Cytokine

2005;31:72–86.

Vuocolo T, Cockett N, Dalrymple BP, Tellam RL: Expression of imprinted genes surrounding the cal-

lipyge mutation in ovine skeletal muscle. Aust J Exp Agric 2005;45:879–892.

Wang YH, Byrne KA, Reverter A, Harper GS, Taniguchi M, McWilliam SM, Mannen H, Oyama K,

Lehnert SA: Comparison of gene expression profiles of muscle tissue in Japanese Black and

Holstein cattle. Mamm Genome 2005a;16:201–210.

Wang YH, Reverter A, Mannen H, Taniguchi M, Harper GS, Oyama K, Byrne KA, Oka A, Tsuji S,

Lehnert SA: Transcriptional profiling of adult muscle in Japanese black cattle to identify genes

involved with the development of intramuscular fat. Aust J Exp Agric 2005b;45:809–820.

Weaver ICG, Cervoni N, Champagne FA, D’Alessio AC, Sharma S, Seckl JR, Dymov S, Szyf M,

Meaney MJ: Epigenetic programming by maternal behaviour. Nat Neurosci 2004;7:847–854.

WHO/FAO: Diet, Nutrition and the Prevention of Chronic Diseases. Report of a Joint WHO/FAO Expert

Consultation. Geneva, World Health Organization, 2003.

Wilson J: The Barker hypothesis. An analysis. Aust NZ J Obstet Gynaecolo 1999;39:1–7.

Mr. Shaun G. Coffey

Industrial Research Ltd

69 Gracefield Rd

Lower Hutt 5040 (New Zealand)

Tel. �64 4931 3000, Fax �64 4566 6004, E-Mail [email protected]



Functionality of Probiotics – Potential for Product Development

James Dekker, Michael Collett, Jaya Prasad, Pramod Gopal

Fonterra Co-operative Group, Palmerston North, New Zealand

AbstractIt is becoming increasingly accepted by consumers that live lactic acid bacteria do exert

health benefits when eaten. In addition, it is also becoming recognised that not all probiotic

bacteria are equal. It is now no longer just a question of providing sufficient numbers of

viable bacteria in a product; industry must also provide proof of efficacy for each strain. In

the early 1990s, Fonterra embarked on a programme to develop proprietary probiotic strains,

and as a result, commercialised two strains, Bifidobacterium lactis HN019 and Lactobacillusrhamnosus HN001. Over the past decade, Fonterra has developed a significant body of peer-

reviewed published reports around these strains, including studies showing safety in ani-

mal and human trials, protection against pathogens such as Salmonella typhimurium and

Escherichia coli O157:H7, modulation of human and animal immune markers at realistic

dose rates, and efficacy in human clinical trials. Based on this work, HN019 and HN001

have been applied to several functional foods both by Fonterra (under the DR10™ and

DR20™ brands, respectively) and by third parties (e.g. under the HOWARU™ brand by

Danisco). While the ‘gold standard’ of proof of efficacy is a phase III clinical trial, ethical

considerations as well as expense preclude the use of clinical trials as screening tools for pro-

biotics. Therefore, biomarkers have to be employed to identify strains with probiotic utility,

and to define the different positive health benefits of existing probiotic strains. However, as

the mechanisms by which most probiotic bacteria exert their health benefits remain unclear,

the question of which biomarkers accurately reflect efficacy in vivo remains unresolved.

With recent technological advances, and the shift toward probiotics targeted to specific con-

ditions, researchers are beginning to tease out how probiotic bacteria work, and it is this

knowledge that will inform biomarker development and improve the ability to offer the mar-

ket safe and effective probiotic functional foods.


If nutrigenomics involves the relationship between nutrition and host

gene expression, then a vital component of this interaction is the gut micro-

biota. It has been estimated that the human gastro-intestinal tract (GIT)

Functionality of Probiotics 197

contains around 1014 micro-organisms, outnumbering human cells by a factor

of 10:1 [1]. The gut microbial community is not only large, but also extremely

diverse, with at least 400 different bacterial strains inhabiting the GIT at any

one time. It also appears that the make-up of this microbial community is

unique to each individual, in terms of the actual species present and their rel-

ative proportions [2, 3]. It has become apparent from microbial ecology stud-

ies that both the environment and host genotype influence the composition of

gut microbiota.

Thus, within the intestinal tract, food substances will come into contact

with a range of metabolically active bacterial cells. These bacterial cells are

known to metabolize or modify various nutrients and may even play vital roles

in human nutrition. However, it is also known that gut bacteria interact with

host gut epithelial and immune cells, and interact with other micro-organisms,

including pathogens, whereby the growth or impact of other strains may be

enhanced or diminished. Given the huge diversity of the gut microbiota, it is

obvious that individual strains will differ in their ability to participate in such

interactions.

Until recently, medical science has largely considered the members of the

gut microbial community as falling into one of two camps, either pathogenic

micro-organisms (including opportunistic pathogens) that can do harm to the

host, or non-pathogenic micro-organisms that are considered benign or

neutral. However, there is now increasing evidence that a third class exists,

whereby some bacterial strains offer a positive health impact. This concept

underlies the definition of probiotic bacteria as being ‘live microorganisms

that, when administered in adequate amounts, confer a health benefit on the

host’ [4].

Development of Fonterra Probiotic Strains

Fonterra is the world’s leading exporter of dairy products and is responsi-

ble for a third of international dairy trade. It is a co-operative company owned

by more than 12,000 New Zealand dairy farmers, and exports a wide range of

dairy-based commodity and specialty products to over 140 countries. In the mid

1990s, Fonterra decided to develop its own proprietary probiotic strains to

specifically target gut health and immune enhancement. A strategy was devel-

oped to isolate probiotic bacterial strains useful as functional food ingredients.

Strains were screened for their ability to survive in the human GIT, to be safe

for human consumption, to produce a positive impact on the human microbiota,

to show efficacy in providing a health benefit, and for their suitability for appli-

cation to consumer products.

Dekker/Collett/Prasad/Gopal 198

Survival in the GITAs part of an initial screen, over 2,000 bacterial strains of human or dairy

origin were tested for the ability to survive passage through the GIT and to

colonise the gut epithelium [5, 6]. As the use of in vivo human trials for this was

obviously impractical, a series of in vitro assays were applied to predict survival

in the GIT. Strains were first tested for survival at low pH and high bile acid

concentrations, and then acid- and bile-tolerant strains were tested for adher-

ence to the human intestinal epithelial cell lines Caco-2, HT-29, and HT-29

MTX (a mucus-producing variant of HT-29) as an assay for gut colonisation.

Candidate probiotic strains were selected according to the results of these

screening tests and preliminary immune efficacy data from mouse studies.

Following further efficacy and safety trials, two strains were commercialised:

Lactobacillus rhamnosus HN001, which was trademarked as DR20™, and

Bifidobacterium lactis HN019, trademarked as DR10™. These strains have

also been licensed to Danisco, and are marketed under the HOWARU™ brand.

SafetyLike other functional foods, probiotic bacteria need to provide a health

benefit that is essentially risk-free to the consumer. That is, the probiotic bacte-

rial strain must pose no threat to the consumer even when consumed in large

quantities or by those with relatively low immune status. Although lactobacilli

and bifidobacteria are normal inhabitants of the human gut, and are generally

considered as ‘safe’ bacteria, there is still the requirement to show that each

specific strain is non-pathogenic.

Several in vitro and animal models were employed to provide science-

based evidence that the candidate probiotics were safe for human consumption.

Three mouse feeding trials were conducted using groups of BALB/c mice

orally administered DR10 or DR20 over a range of dose rates, daily for 7–28

days [7–9]. The mice were assessed for various haematological, histological,

and growth parameters. The results indicated that even at the highest dose of

2.5 � 1012 CFU/kg body weight/day, given for 28 days, neither probiotic strain

was associated with alterations in food and water intake, live body weight,

haematological parameters (including red blood cell and leucocyte counts),

blood biochemistry (including levels of plasma total protein, albumin, choles-

terol, and glucose), and mucosal histology (such as epithelial cell height,

mucosal thickness, and villus height) [8]. In addition, there was no evidence

that probiotic bacteria had translocated to the blood, liver, spleen, kidney, or

mesenteric lymph nodes.

Further in vitro studies have examined specific safety issues. For instance,

the mucus layer of the gut is an important defence mechanism that protects the

gut epithelia from attack from pathogens. Experiments have shown that DR10


and DR20 do not degrade mucin, the principal component of the mucus layer,

and so are unlikely to encourage pathogenic infection [10]. It has also been

shown that DR10 and DR20 cannot aggregate or activate platelets [11], which

is a positive finding as platelet aggregation by bacteria in the blood can lead to

various pathologies, and neither DR10 nor DR20 exhibited undesirable or atyp-

ical antibiotic resistance profiles [12]. Recently, another aspect of safety has

been examined in an animal model. Given that probiotic bacteria are thought to

modulate the host immune system, there remains a theoretical possibility

that probiotic bacteria could exacerbate immune-related disorders, such as an

autoimmune disease. However, in a murine model of experimental autoimmune

thyroiditis, feeding with DR10 or DR20 for extended periods did not accelerate

disease progression [13].

It was concluded from all these studies that DR10 and DR20 are safe for

human consumption.

Microbial EcologyScreening procedures provided in vitro evidence that DR10 and DR20

could survive within the human GIT. Further studies have been conducted to

see whether these bacteria colonised the gut in vivo, and their impact on resi-

dent gut microflora was assessed. One dietary intervention study involved feed-

ing 10 healthy human volunteers with a total daily dose of 1.6 � 109 CFU

DR20 using a three-stage study design over a 15-month period [3]. For the first

6 months (run-in period), volunteers consumed a daily portion of low-fat milk.

For the next 6 months (treatment period), this milk contained DR20, and for the

final 3 months of the trial, the volunteers reverted back to the milk-only por-

tion. Using a variety of detection methods, the study showed that the range of

total lactobacillus counts from faeces was less than 102 to 2 � 108 CFU/g dur-

ing the run-in phase. Upon the consumption of HN001, this range increased to

105 to 108 CFU/g, which could not be explained by merely the amount of DR20

consumed. Therefore, it appeared that DR20 enhanced the growth of other lac-

tobacilli in the healthy adult gut. Although in most cases DR20 became the

dominant lactobacillus strain in the GIT during the feeding period, it did not

generally persist in the faeces beyond the treatment period. Thus, consumption

of DR20 transiently altered the total lactobacillus and enterococcal counts in

most cases, without affecting biochemical or other bacteriological factors.

DR10 was the subject of a randomised, double-blind, and placebo-

controlled study to examine its ecological impact as well as the effect of galacto-

oligosaccharides (GOS) on the GIT microbiota. GOS are known to act as a

‘prebiotic’, a food compound that supports the growth of healthy bacteria. GOS

have been shown to enhance the growth in the GIT of lactobacilli and bifidobac-

teria species including DR10 and DR20 [14]. In the study, 30 subjects were


randomly assigned into three groups [15]. After a two-week pre-intervention

(run-in) period, subjects in group 1 consumed milk containing the GOS prebi-

otic, group 2 subjects received a total daily dose of 3 � 1010 CFU DR10 in milk,

while group 3 subjects were given milk without supplementation for a 4-week

period. This was followed by a 2-week wash-out period with no dietary interven-

tion. Faecal samples were collected from all subjects at weekly intervals and

analysed for bacterial content. The results indicated that while subjects consum-

ing milk supplemented with either GOS or DR10 exhibited significantly increased

bifidobacteria faecal counts (p � 0.0002), subjects that received the control milk

showed no significant change compared with the run-in period. Interestingly,

both groups 1 and 2 also exhibited increased lactobacilli faecal counts. Thus,

similar to DR20, the study indicated that DR10 at least transiently colonised the

human gut, and had positive impacts on gut microflora [4].

The question as to how many probiotic bacteria were required to observe

the beneficial changes in gut microflora was addressed in a recent randomised,

double-blind, placebo-controlled study involving 80 healthy ‘elderly’ (�60

years old) subjects [16]. The subjects were divided into 4 groups that received

either a high dose of DR10 (5 � 109 CFU/day), a medium dose (1.0 � 109

CFU/day), a low dose (6.5 � 107 CFU/day), or vehicle only. The results

indicated that even at the lowest dose, consumption of DR10 significantly

increased bifidobacterial counts, but reduced enterobacterial counts, compared

with vehicle-only controls.

In summary, results of microbial ecology studies showed that both DR10

and DR20 can survive within the gut environment and appeared to exert benefi-

cial effects on gut health through the enhancement of other lactobacilli and

bifidobacteria populations, genera associated with gastro-intestinal health in

humans. Neither strain permanently colonised the adult human GIT, and there-

fore had to be taken on a regular basis to maintain significant numbers.

EfficacyThe defining characteristic of a probiotic strain, apart from its viability in

the GIT, is the provision of a health benefit to its host. Based on Fonterra’s aim

of producing probiotic strains with immune-enhancing effects, immune assays

were used to explore efficacy. The mammalian immune system is large and

multifaceted, and is generally considered to consist of two major divisions:

first, elements that are ‘hard wired’ to respond to certain immune challenges

(the innate immune system) and second, elements that can adapt through vari-

ous mechanisms to the bewildering array of antigens that the immune system is

exposed to (the adaptive immune system). To explore whether DR10 or DR20

could provide a health benefit through immune enhancement, both the adaptive

and innate arms of the immune system were tested.


A mouse trial was conducted in which mice were fed DR10 or DR20 at

total doses of 109 CFU/day, and various immune parameters were compared

with control mice [17]. The results indicated that aspects of both the innate and

acquired arms of the immune system were positively affected by the presence of

the probiotic bacteria. One measure involved examining the activity of phago-

cytes. Phagocytes act as scavenger cells, patrolling the body looking for invad-

ing or foreign cells, and engulfing these cells whole. The engulfed cells are then

broken down inside the phagocytes and the antigens presented to immune

responder cells to drive the immune response. Compared with control mice,

mice fed either DR10 or DR20 showed increased phagocytic activity by both

blood leucocytes and peritoneal macrophages. Another cell type with a surveil-

lance role is the natural killer (NK) cell. These cells roam the body looking for

virally infected or tumour cells, whereupon they kill the affected cell. An assay

examining the activity of NK cells against tumour cells in vitro showed that NK

cells from DR10- and DR20-fed mice exhibited enhanced tumour cell killing

compared with NK cells from control mice. As an example of the adaptive

immune response, the production of specific antibodies by mice inoculated

with tetanus toxoid and cholera antigens showed that DR10 and DR20 exhib-

ited adjuvant activity, that is, the presence of the bacteria enhanced the produc-

tion of specific antibodies in response to antigen challenge. Interestingly, it was

also shown that the immune-enhancing effects of DR10 and DR20 did not per-

turb the overall leucocyte cell counts or the relative levels of lymphocyte sub-

populations. A related study [18] explored the effect of probiotic dose on

phagocytic activity and antibody production, and found that the increased

phagocytosis observed was not due to increased numbers of phagocytic cells,

but rather to increased activity of phagocytic cells already present. While blood

phagocyte activity improved even at the lowest dose (107 CFU DR20/day),

higher doses (109 or 1011 CFU DR20/day) were associated with greater improve-

ment in phagocytic activity.

Mouse models have also been used to investigate possible mechanisms by

which DR10 and DR20 mediate immune enhancement. A previous study [17]

showed that immune cells taken from the spleens of mice fed DR20 and then

stimulated to non-specifically activate T lymphocytes (immune responder cells)

tended to produce more of a cytokine called interferon-� than the same stimu-

lated cells taken from control mice. Cytokines are chemical messengers or

‘immune hormones’ that are used to signal other immune cells nearby or at a

distance. Current models propose that T helper (Th) lymphocytes, in reacting to

antigens presented by antigen-presenting cells, can promote either an immune

response that favours immune cells such as phagocytes (a Th1 response) or an

immune response that favours antibody production (a Th2 response). This is

achieved through expression of characteristic cytokines, and it is believed that


the balance between Th1 and Th2 responses can have profound impacts on how

the immune system reacts to potential disease threats. For instance, one theory

suggests that allergic responses may be due at least in part to a skewing of the

immune response to the allergen towards a Th2 response [19, 20]. In an experi-

mental model in which mice are immunised with ovalbumin protein, spleen

immune cells taken from these mice will respond to the addition of ovalbumin

peptides by producing the cytokines interleukin (IL)-4 and IL-5, which are con-

sidered indicative of a Th2 response. In mice fed DR20 prior to and during oval-

bumin immunisation, isolated stimulated spleen cells produced increased levels

of interferon-� (a pro-Th1 cytokine) without affecting IL-4 and IL-5 levels [21].

A similar effect on Th1/Th2 balance was observed for peritoneal macrophages.

Thus, it seems that although DR20 appears to promote a Th1-type response, it

does not do so at the expense of an ongoing Th2 response. As the balance

between Th1 and Th2 responses may have consequences for normal immune

function, this finding was particularly significant.

Apart from enhanced immune parameters, another potential beneficial

effect of probiotic bacteria is the protection against infection by pathogenic

micro-organisms. Both DR10 and DR20 have been shown to be effective

against pathogens in several animal models. For instance, mice fed DR20

showed reduced morbidity, reduced bacterial translocation, and increased mark-

ers of innate and acquired immunity toward the pathogenic Escherichia colistrain O157:H7 [22], as well as impressive protection against mortality from

Salmonella typhimurium infection [23]. Likewise, DR10 was shown to be

effective in protecting weaning piglets against diarrhoea associated with

rotavirus or E. coli infection [24], and protected mice against pathogenic E. coliO157:H7 [25] and against pathogenic S. typhimurium challenges designed to

mimic both a single exposure to high pathogen levels (as might be encountered

after consumption of heavily contaminated food prepared under poor hygiene)

or chronic exposure to pathogens over a period of time [26].

Given the positive animal trial results, the probiotic efficacy of DR10 and

DR20 was examined in human clinical trials. For DR20, several 3-stage pre- to

post-intervention trials have been performed on groups of healthy middle-aged

or elderly subjects [27–29], with 3-week run-in periods to establish baseline

data, 3-week intervention periods using milk supplemented with DR20 (5 � 109

or 5 � 1010 CFU/day), followed by 3-week wash-out periods. In all the studies,

analysis of blood samples taken from the subjects showed significant increases

in the phagocytic activities of blood polymorphonuclear leucocytes and mono-

cytes, as well as increased tumouricidal activity by NK cells.

Similar benefits for human immune function have been shown for DR10,

including improved phagocytic capacity [28, 30, 31], interferon-� production

[31], and NK cell activity [30, 31]. Interestingly, one trial noted that subjects


with relatively poor phagocytic responses prior to treatment tended to exhibit

the largest relative increases in activity in response to DR10 [32]. This finding

suggested that DR10 may be effective in protecting against immune senescence

in the elderly [33], although this hypothesis requires further study.

Although the experimental results so far represent strong evidence that

DR10 and DR20 exhibit physiological efficacy, the gold standard of proof

remains the demonstration of actual health benefits in a phase III clinical trial.

Such a trial has recently been completed for DR10 that investigated the health

effects of supplemented milk powder in children [Sazawal et al., in preparation,

2005]. The trial was designed to meet the standards used by the pharmaceutical

industry, and consisted of a randomised, double-blind and placebo-controlled trial

involving more than 600 children aged 12–36 months. The study was conducted

in a lower- to middle-class residential area in Sangam Vihar, India. Healthy chil-

dren with no chronic illnesses were enrolled in the study, and randomly assigned

to either treatment or control groups (n � 312 each). The treatment group

received milk fortified with DR10 and GOS, while the control group received

non-supplemented milk, twice daily for a period of 12 months, followed by a

2-year follow-up period. As the trial was double-blinded to prevent possible bias,

neither the subjects nor researchers knew which group a child was in during the

trial period. The results showed convincing and credible evidence that the group

fed the fortified milk displayed significant health benefits over those that con-

sumed the normal milk. For instance, children fed DR10 and GOS were 22% bet-

ter protected against dysentery, 16% better protected against the burden of severe

illness (non-diarrhoeal disease), 32% better protected against sickness with high

temperature, 7% better protected against ear infection, and 6% less likely to need

antibiotics. All of these findings are consistent with the view that DR10 has

immune-enhancing properties. The study also found that children fed the fortified

milk displayed a 35% reduction in iron deficiency anaemia as well as signifi-

cantly improved growth rates more in line with growth charts published by the

National Center for Health Statistics, USA, compared with the control group.

Another recently completed double-blind, randomised, placebo-controlled

trial showed that DR10 and DR20 may be beneficial against allergy [34]. This

trial, based in Wellington, New Zealand, studied children aged 1–10 years with a

history of atopic dermatitis (eczema). The study subjects (n � 59) all exhibited a

positive skin reaction to at least one common allergen using the skin prick test, and

were randomly assigned to control or treatment groups. The control group received

a placebo supplement, while the test group consumed a supplement containing

DR10 and DR20 daily for 12 weeks. The parents of the subjects (also blinded to

whether the child received the probiotic bacteria or placebo) kept symptom diaries.

Symptom severity was then tracked using the scoring atopic dermatitis (SCORAD)

index. While the treatment group showed significant improvement in eczema


symptoms compared with baseline SCORAD index values, this improvement was

slightly below statistical significance when compared with the placebo effect in

the control group. Nonetheless, in a subgroup of children that exhibited at least

one positive skin prick test to a food allergen (n � 37), the group fed DR10 and

DR20 had significantly decreased SCORAD values compared with the placebo

group. The results indicated that DR10 and DR20 exerted a positive health impact

at least on children with allergic dermatitis who exhibit a food allergy. Based on

this trial, a second much larger study is now in progress to examine the impact of

DR10 and DR20 on childhood allergy.

ProductsA large body of science-based evidence now shows that DR10 and DR20

are safe for human consumption, are able to survive in the human GIT, and exert

health benefits on the host both in terms of gut health and immune enhancement.

Based on these findings, Fonterra’s probiotic bacterial strains have been applied

to various products, including milk powders, yoghurts, cheese, fermented milk

drinks, and powdered supplements, both under Fonterra’s own DR10 and DR20

brands, and licensed to third parties, for example under Dansico’s HOWARU

brands.

Further Research

Although previous studies have shown that DR10 and DR20 mediate

immune-enhancing effects in vitro, and offer demonstrable health benefits to

the host, the mechanisms by which these benefits occur remain to be resolved.

Likewise, it remains unclear how in vitro assays reflect immune mechanisms

in vivo. Knowledge into how probiotic bacteria actually work can be applied to

a number of areas, such as the selection of new strains targeted to specific

effects (e.g. the ability to stimulate interferon-� production or to increase NK

cell activity), or to be active against specific diseases, such as inflammatory

bowel disease. Mechanistic information can also be used to support the probi-

otic strains in the market and assist decision-making by regulatory authorities.

Therefore, even though DR10 and DR20 have already been successfully com-

mercialised, they remain the subjects of ongoing research programmes at

Fonterra and other research organisations. The research is centred around five

main areas which are presented in what follows.

GenomicsA draft genome sequence of DR20 has been obtained. This has allowed the

interrogation of the genome for genes involved in probiotic mechanisms. For


instance, using the sequence data, the DR20 genome has been screened for

genes containing the known cell-wall-anchoring motif ‘LPxTG’ to identify pro-

teins expressed on the cell wall. As these proteins interact with the external

environment of the bacterium, it is possible that these proteins are involved in

mediating probiotic effects. Some of these LPxTG-containing proteins are now

the subjects of further study into immune mechanisms. Genome data can also

be used to answer specific issues. For example, analysis of genome data

strongly suggested that the vancomycin resistance exhibited by DR20 (a trait

typical of Lactobacillus species) was encoded on the chromosome, meaning

that it cannot be easily passed on to other bacteria. Genome data can also be

used to assist proteomic studies, and to reconstruct metabolic pathways in silico

to predict how DR20 might operate in different environments.

Genetic ToolsKnowledge of the DR20 genome has allowed the development of genetic

tools (plasmid vectors) that enable the genetic manipulation of DR20. Using

our suite of genetic tools, we are now able to overexpress as well as disrupt

genes of interest, allowing investigation into the roles of specific genes in dif-

ferent probiotic attributes.

In vitro AssaysAssays based on cell lines or ex vivo leucocyte populations offer many

advantages over the use of animal models and clinical trials in terms of cost,

ease of use, and numbers of experimental samples that can be tested. To explore

aspects of probiotic function, we have developed several in vitro assays of

immune interaction that can be used to identify bacterial components and/or

genes important in mediating probiotic effects, and perhaps even identify host

receptor molecules. Other in vitro strategies, such as microarrays and single

nucleotide polymorphism screening, may be useful in investigating the role of

human gene expression and gene polymorphisms in determining immune

responses to probiotic bacteria, allowing the application of nutrigenomic

approaches to probiotics research.

While each in vitro assay may provide a useful result, it will be the ability to

integrate data from across multiple assays and combine with data from animal

studies and human clinical trials that will provide the greatest benefit. An inte-

grated approach will also help identify in vitro assays that best reflect the mech-

anisms of in vivo health impacts observed in human trials of DR10 and DR20.

Animal ModelsThe main advantages of animal trials are their lower cost (compared with clin-

ical trials) and their ability to mimic human physiological conditions (compared


with in vitro assays). Animal trials can be used to test specific product formats

or encapsulation techniques to ensure that efficacy is maintained, and to pro-

vide material to explore probiotic mechanisms that would be unavailable from

human trials. Also, there exist a wide variety of animal models of human dis-

ease that can be used to identify whether DR10 and DR20 have any activity

against specific pathologies or conditions. Finally, DR10 and DR20 may also

prove to be effective as probiotics in companion and/or production animals.

FermentationFonterra has extensive expertise in the large-scale production and handling

of dairy cultures and their properties in various formats, including the mainte-

nance of viability during processing [35]. As probiotic bacteria such as DR10

and DR20 can grow, or at least survive, in dairy-based products such as yoghurt

and cheese, knowledge of their sensory impacts is vital. DR20 is part of a

flavour biotechnology research programme at Fonterra, and has yielded at least

one enzyme that may have utility in controlling bitterness in fermented dairy

products [36].

Conclusions

The development of a successful functional food is a long and complex

process. Such foods must be completely safe for the consumer, yet offer a

demonstrable health benefit without unwanted side-effects or risk of overdose.

To this end, Fonterra has embarked on a long-term programme to prove that its

probiotic bacterial strains DR10 and DR20 are safe and effective using science-

based studies that withstand the rigour of external peer review. The application

of further experimental research programmes will close the gap between health

ingredients with proven health benefits and the understanding of the mecha-

nisms responsible for these health benefits at the molecular level.

References

1 Savage DC: Microbial ecology of the gastrointestinal tract. Annu Rev Microbiol 1977;31:

107–133.

2 Zoetendal EG, von Wright A, Vilpponen-Salmela T, Ben-Amor K, Akkermans AD, de Vos WM:

Mucosa-associated bacteria in the human gastrointestinal tract are uniformly distributed along the

colon and differ from the community recovered from feces. Appl Environ Microbiol 2002;68:

3401–3407.

3 Tannock GW, Munro K, Harmsen HJ, Welling GW, Smart J, Gopal PK: Analysis of the fecal

microflora of human subjects consuming a probiotic product containing Lactobacillus rhamnosusDR20. Appl Environ Microbiol 2000;66:2578–2588.


4 Joint Food and Agriculture Organization of the United Nations and World Health Organization

Working Group: Joint FAO/WHO Working Group report on drafting guidelines for the evaluation

of probiotics in Food. FAO/WHO, 2002.

5 Prasad J, Gill HS, Smart J, Gopal PK: Selection of characterisation of Lactobacillus and

Bifidobacterium strains for use as probiotics. Int Dairy J 1998;8:993–1002.

6 Gopal PK, Prasad J, Smart J, Gill HS: In vitro adherence properties of Lactobacillus rhamnosusDR20 and Bifidobacterium lactis DR10 strains and their antagonistic activity against an entero-

toxigenic Escherichia coli. Int J Food Microbiol 2001;67:207–216.

7 Zhou JS, Shu Q, Rutherfurd KJ, Prasad J, Gopal PK, Gill HS: Acute oral toxicity and bacterial

translocation studies on potentially probiotic strains of lactic acid bacteria. Food Chem Toxicol

2000;38:153–161.

8 Zhou JS, Shu Q, Rutherfurd KJ, Prasad J, Birtles MJ, Gopal PK, Gill HS: Safety assessment of

potential probiotic lactic acid bacterial strains Lactobacillus rhamnosus HN001, Lb. acidophilusHN017, and Bifidobacterium lactis HN019 in BALB/c mice. Int J Food Microbiol 2000;56: 87–96.

9 Shu Q, Zhou JS, Rutherfurd K, Birtles MJ, Prasad J, Gopal PK, Gill HS: Probiotic lactic acid bac-

teria (Lactobacillus acidophilus HN017, Lactobacillus rhamnosus HN001 and Bifidobacteriumlactis HN019) have no adverse effects on the health of mice. Int Dairy J 1999;9:831–836.

10 Zhou JS, Gopal PK, Gill HS: Potential probiotic lactic acid bacteria Lactobacillus rhamnosus(HN001), Lactobacillus acidophilus (HN017) and Bifidobacterium lactis (HN019) do not degrade

gastric mucin in vitro. Int J Food Microbiol 2001;63:81–90.

11 Zhou JS, Rutherfurd KJ, Gill HS: Inability of probiotic bacterial strains Lactobacillus rhamnosusHN001 and Bifidobacterium lactis HN019 to induce human platelet aggregation in vitro. J Food

Prot 2005;68:2459–2464.

12 Zhou JS, Pillidge CJ, Gopal PK, Gill HS: Antibiotic susceptibility profiles of new probiotic

Lactobacillus and Bifidobacterium strains. Int J Food Microbiol 2005;98:211–217.

13 Zhou JS, Gill HS: Immunostimulatory probiotic Lactobacillus rhamnosus HN001 and Bifidobacteriumlactis HN019 do not induce pathological inflammation in mouse model of experimental autoim-

mune thyroiditis. Int J Food Microbiol 2005;103:97–104.

14 Gopal PK, Sullivan PA, Smart JB: Utilisation of galacto-oligosaccharides as selective substrates

for growth by lactic acid bacteria including Bifidobacterium lactis DR10 and Lactobacillus rhamnosusDR20. Int Dairy J 2001;11:19–25.

15 Gopal PK, Prasad J, Gill HS: Effects of the consumption of Bifidobacterium lactis HN019

(DR10™) and galacto-oligosaccharides on the microflora of the gastrointestinal tract in human

subjects. Nutr Res 2003;23:1313–1328.

16 Ahmed M, Prasad J, Gill H, Stevenson L, Gopal P: Impact of consumption of different levels of

Bifidobacterium lactis HN019 on the intestinal microflora of elderly human subjects. J Nutr

Health Aging 2007;11:26–31.

17 Gill HS, Rutherfurd KJ, Prasad J, Gopal PK: Enhancement of natural and acquired immunity by

Lactobacillus rhamnosus (HN001), Lactobacillus acidophilus (HN017) and Bifidobacterium lactis(HN019). Br J Nutr 2000;83:167–176.

18 Gill HS, Rutherfurd KJ: Viability and dose-response studies on the effects of the immunoenhanc-

ing lactic acid bacterium Lactobacillus rhamnosus in mice. Br J Nutr 2001;86:285–289.

19 Cross ML, Gill HS: Can immunoregulatory lactic acid bacteria be used as dietary supplements to

limit allergies? Int Arch Allergy Immunol 2001;125:112–119.

20 Holt PG: Key factors in the development of asthma: atopy. Am J Respir Crit Care Med 2000;161:

172S–175S.

21 Cross ML, Mortensen RR, Kudsk J, Gill HS: Dietary intake of Lactobacillus rhamnosus HN001

enhances production of both Th1 and Th2 cytokines in antigen-primed mice. Med Microbiol

Immunol (Berl) 2002;191:49–53.

22 Shu Q, Gill HS: Immune protection mediated by the probiotic Lactobacillus rhamnosus HN001

(DR20) against Escherichia coli O157:H7 infection in mice. FEMS Immunol Med Microbiol

2002;34:59–64.

23 Gill HS, Shu Q, Lin H, Rutherfurd KJ, Cross ML: Protection against translocating Salmonellatyphimurium infection in mice by feeding the immuno-enhancing probiotic Lactobacillus rhamnosusstrain HN001. Med Microbiol Immunol (Berl) 2001;190:97–104.


24 Shu Q, Qu F, Gill HS: Probiotic treatment using Bifidobacterium lactis HN019 reduces weanling

diarrhea associated with rotavirus and Escherichia coli infection in a piglet model. J Pediatr

Gastroenterol Nutr 2001;33:171–177.

25 Shu Q, Gill HS: A dietary probiotic Bifidobacterium lactis (HN019) reduces the severity of

Escherichia coli O157:H7 infection in mice. Med Microbiol Immunol 2001;189:147–152.

26 Shu Q, Lin H, Rutherfurd KJ, Fenwick SG, Prasad J, Gopal PK, Gill HS: Dietary Bifidobacteriumlactis (HN019) enhances resistance to oral Salmonella typhimurium infection in mice. Microbiol

Immunol 2000;44:213–222.

27 Sheih YH, Chiang BL, Wang LH, Liao CK, Gill HS: Systemic immunity-enhancing effects in

healthy subjects following dietary consumption of the lactic acid bacterium Lactobacillus rhamnosusHN001. J Am Coll Nutr 2001;20:149–156.

28 Gill HS, Cross ML, Rutherfurd KJ, Gopal PK: Dietary probiotic supplementation to enhance cel-

lular immunity in the elderly. Br J Biomed Sci 2001;58:94–96.

29 Gill HS, Rutherfurd KJ: Probiotic supplementation to enhance natural immunity in the elderly:

effects of a newly characterized immunostimulatory strain Lactobacillus rhamnosus HN001

(DR20™) on leucocyte phagocytosis. Nutr Res 2001;21:183–189.

30 Chiang BL, Sheih YH, Wang LH, Liao CK, Gill HS: Enhancing immunity by dietary consumption

of a probiotic lactic acid bacterium (Bifidobacterium lactis HN019): optimization and definition

of cellular immune responses. Eur J Clin Nutr 2000;54:849–855.

31 Arunachalam K, Gill HS, Chandra RK: Enhancement of natural immune function by dietary con-

sumption of Bifidobacterium lactis (HN019). Eur J Clin Nutr 2000;54:263–267.

32 Gill HS, Rutherfurd KJ, Cross ML, Gopal PK: Enhancement of immunity in the elderly by dietary

supplementation with the probiotic Bifidobacterium lactis HN019. Am J Clin Nutr 2001;74:

833–839.

33 Gill HS, Darragh AJ, Cross ML: Optimizing immunity and gut function in the elderly. J Nutr

Health Aging 2001;5:80–91.

34 Sistek D, Kelly R, Wickens K, Stanley T, Fitzharris P, Crane J: Is the effect of probiotics on atopic

dermatitis confined to food sensitized children? Clin Exp Allergy 2006;36:629–633.

35 Prasad J, McJarrow P, Gopal P: Heat and osmotic stress responses of probiotic Lactobacillusrhamnosus HN001 (DR20) in relation to viability after drying. Appl Environ Microbiol 2003;69:

917–925.

36 Christensson C, Bratt H, Collins LJ, Coolbear T, Holland R, Lubbers MW, O’Toole PW, Reid JR:

Cloning and expression of an oligopeptidase, PepO, with novel specificity from Lactobacillusrhamnosus HN001 (DR20). Appl Environ Microbiol 2002;68:254–262.

Dr. James Dekker

Fonterra Co-operative Group

Palmerston North (New Zealand)




Developing the Promise of Nutrigenomicsthrough Complete Science andInternational Collaborations

Jim Kaput

Laboratory of Nutrigenomic Medicine, University of Illinois, Chicago, Ill., and

NCMHD Center of Excellence in Nutritional Genomics, University of California,

Davis, Calif., USA; the European Nutrigenomics Organisation (http://www.nugo.org)

AbstractFood is economically available to 4 billion of the world’s 6 billion people, a situation

that resulted from dramatically improved methods for producing, storing, and distributing

food on a mass scale during the last 100 years. Nevertheless, almost 2 billion people are mal-

nourished through either over-consumption of fats and calories or lack of adequate calories

and micronutrients. Malnourishment results in chronic diseases, immune dysfunction, and

early death. Analyzing and understanding gene – nutrient interactions is therefore a neces-

sary step for designing and producing foods for maintaining the health of populations and

individuals. Nutrigenomics is the study of how constituents of the diet interact with genes,

and their products, to alter phenotype and conversely, how genes and their products metabo-

lize these constituents into nutrients, antinutrients, and bioactive compounds. However,

defining causal gene X nutrient interactions involved in maintaining optimum health are

more challenging because of the (i) chemical complexity of food, (ii) genetic heterogeneity

of humans, and (iii) the complexity of physiological responses to nutrient intakes in health

and disease. Three significant developments will allow progress in nutrition and nutrige-

nomics: the development of high throughput omic (genomic, transcriptomic, proteomic, and

metabolomic) technologies, improved experimental designs, and the development of

research collaborations to study complex biological processes. The practical applications of

nutrigenomics are the possibility of delivering the right micronutrients in the optimum

amount to the food insecure and developing novel foods which are more nutritious, flavour-

ful, storable, and health promoting than many of the products manufactured today.


Humans have dramatically improved methods for economically producing,

storing, and distributing food on a mass scale during the last 100 years. The

Conclusion

Kaput 210

result of those advances is a world population of over 6 billion people covering

almost all ecological niches. In no other era has more food been available for

more people. Nevertheless, almost 2 billion people are malnourished, through

either overconsumption of fats and calories or underconsumption of calories

and micronutrients. While these malnourishment problems could theoretically

be solved by personal responsibility (eating less) in lands of plenty or distribut-

ing food equitably and inexpensively to the undernourished, the realization of

the genetic uniqueness of individuals may confound simple-minded solutions

of distributing the same types and amounts of foods to all. That is, food pro-

duced or formulated for one population may not be optimum for every member

of that population or for individuals in populations because of the genetic het-

erogeneity of the human species. Analyzing and understanding gene-nutrient

interactions is therefore a necessary step for designing and producing foods for

maintaining the health of populations and individuals.

Nutrigenomics is the study of how constituents of the diet interact with

genes, and their products, to alter phenotypes and conversely, how genes and

their products metabolize these constituents into nutrients, antinutrients, and

bioactive compounds. As is well illustrated by the contributions to this volume,

the methodologies for analyzing genes (genomics), their RNA products (tran-

scriptomics), proteins (proteomics), and metabolites (metabolomics) are being

applied to the study of nutrition in cell culture, animal models, and humans.

New analytical methods [Dawson et al., 2005; Moore, 2004; Roweis and Saul,

2000; Tenenbaum et al., 2000] are being developed and applied to these com-

plex, high-dimensional data sets to identify associations among food intake,

genetic makeup, and physiological responses in an individual. Focusing on the

individual as the unit of analyses for nutrient intakes rather than the population

is a profound transformation for the field of nutrition and for the food industry.

Dietary guidelines have historically been derived from associating some physi-

ological markers (e.g. cholesterol) with dietary intake as determined from pop-

ulation studies. Although perhaps unintended, dietary guidelines imply that all

individuals are genetically, culturally, and physiologically identical. The same

foods with the same nutrient compositions can be produced for everyone.

Nutrigenomics, on the other hand, may determine the optimal nutrient intakes

for an individual as opposed to a population. The concepts of nutritional genomics

are therefore a profound transformation from past practices in science and its

applications to the food industry.

Nutrigenomics has been demonstrated by monogenic diseases such

as phenylketonuria (http://www.ncbi.nlm.nih.gov/entrez/dispomim.cgi?id �261600) where a mutation causes a susceptibility to phenylalanine. Reduce the

amount of phenylalanine in the diet, and many with this genetic condition

survive. However, defining causal gene-nutrient interactions involved in

Developing the Promise of Nutrigenomics 211

maintaining optimum health is more challenging because of (a) the che-

mical complexity of food, (b) the genetic heterogeneity of humans, and

(c) the complexity of physiological responses to nutrient intakes in health and

disease.

The Diversity Challenges

Chemicals in FoodsNumerous studies have demonstrated that chemicals classified as

macronutrients, such as certain fatty acids, and micronutrients, such as vita-

mins, regulate gene expression directly (reviewed in Schuster [2006]) or

through changes in signal transduction pathways [Guo and Sonenshein, 2006].

Although many of these studies used single nutrients added to cell culture

media, laboratory animal food, or to human diets, most dietary chemicals that

are consumed are found in complex mixtures. Oil extracted from corn contains

at least 9 fatty acids, 13 different triglycerides, 9 sterols, more than 12 fatty acid

sterols, and 5 tocols [Kaput and Rodriguez, 2004]. While many of these nutri-

ents are metabolized for energy, a number of them can be direct (i.e., nonme-

tabolized) activators of transcription factors (e.g. �-sitosterol [Cantafora et al.,

2003; Orzechowski et al., 2002]) or be metabolized to activators (certain fatty

acids to 15-deoxy-� 12,14-PGJ2 [Kliewer et al., 1995]). Hence, nutrigenomic

research requires detailed knowledge of the nutrient composition of foods and

effects of cooking [Malfatti and Felton, 2006]. Food composition databases

have been or are being developed by the Food and Agriculture Organization

of the United Nations (http://www.fao.org/infoods/publications_en.stm), the

International Life Sciences Institute (ILSI Crop Composition Database:

http://www.cropcomposition.org/), the International Network of Food Data

Systems (http://www.fao.org/infoods/directory_en.stm), USDA (Food

Composition Database: http://www.nal.usda.gov/fnic/foodcomp/Data/SR18/

sr18.html), and the European Food Information Resource Network (EuroFIR:

http://www.eurofir.net/). Other national governments have compiled food com-

position tables for their countries, but data must often be extracted from

unlinked flat files or from publications. Some of these databases may be incom-

plete, particularly for minor components that may be bioactive.

Identifying the bioactive chemicals and the amounts that alter expression

of genetic information and physiological processes is a critical part of under-

standing how diet alters molecular pathways. Nutrient intake assessments are

challenging since free-living humans do not regard daily life as a science exper-

iment where the amount and type of food is accurately and precisely recorded.

It is generally accepted in the nutrition community that food recall tools are less

Kaput 212

than ideal [Rutishauser, 2005] but are likely to be the most convenient for large-

scale studies. Ultimately, quantitative analytical methods need to be developed

for analyzing nutrient intakes and for gene-nutrient-phenotype studies [Corella

and Ordovas, 2005; Ordovas and Corella, 2004].

Our diets are also greatly influenced by food produced by industry. Most

individuals in developed countries can inexpensively obtain enough calories and

most macro- and micronutrients. However, many foods with varying combinations

of macro- and micronutrient compositions were developed primarily for taste

and economy without the sophisticated knowledge of how individual humans

respond to nutrients. Hence, it is not surprising that our current national and

international food systems have not produced the most healthful foods. New

knowledge created by nutrition and nutrigenomic research should enable food

companies to design and produce healthy and tasty food as economically as is

done now. Perhaps most importantly, increased nutrigenomic knowledge may

also address a problem resulting from globalization: humanitarian food aid

often supplies enough calories but with inadequate and unbalanced nutrient

content [Hughes and Lawrence, 2005]. Urban dwellers in developing countries

often adopt Western foods and habitats resulting in increases in chronic dis-

eases such as type 2 diabetes mellitus (T2DM) [Epping-Jordan et al., 2005;

James et al., 2001; Lieberman, 2003]. A better understanding of the nutrient

needs of individuals in genetically diverse populations is needed for targeting

the right nutrients to the right individuals.

Genetic DiversityDuring the stepwise migrations from East Africa and subsequent popula-

tion centers [Jorde and Wooding, 2004; Tishkoff and Verrelli, 2003], groups

migrating to new locations carried a subset of the genetic diversity. Migration

and subsequent population expansion resulted in individual humans sharing

99.9% of their genomic sequences. More detailed genetic analyses indicated

that it is possible to place individuals in groups based on ancestry considering

differences in the 0.1% variation in sequence. On average, there is a 12–14%

difference between geographically distinct populations [Jorde and Wooding,

2004] – for example, between Africa and Asia. The majority of human varia-

tion (estimated range of 86–88%) occurs within a geographic population (e.g.

Europe). These differences among ancestral groups resulted from genetic drift

and selective pressures from the environment, including food availability in

the new environments. An example to illustrate the role of food is lactose tol-

erance. A mutation in the promoter of the lactase gene provided a selective

advantage to individuals living in northern climates (in this case Europe)

because a new food source, other mammals’ milk, could be consumed in

adulthood [Enattah et al., 2002; Harvey et al., 1998]. Other examples of how


genetic heterogeneity affects health outcomes can be found in differences in

efficacy of drugs in different ancestral populations. Alleles of CYP3A4 differ

in metabolism of many prescription drugs and the allele frequencies vary by

ethnicity [Xie et al., 2001]. Similar differences are likely to occur in other

metabolic pathways, some of which will be involved in metabolism of nutri-

ents and food-derived bioactives. However, one cannot assume the presence of

an allele based on ancestry because alleles may be present in a population but

at differing frequencies than in another population – that is, one still has a

chance to inherit an allele identical to one that predominates in another popu-

lation.

The implications of the genetic similarity of all humans as a species and

diversity among individuals are significant for nutrigenomic researchers.

Metabolism is a set of interconnected biochemical pathways. Interactions

between genes or their proteins are called gene-gene interactions (or epistasis)

[Carlborg and Harley, 2004; Chiu et al., 2006; Moore, 2004] and at the

protein level ‘biochemical buffering’ [Hartman et al., 2001; Hartman, 2006].

Since allele frequencies of many genes differ among ancestral populations,

the chance of inheriting a specific set of gene variants varies depending upon

the genetic history of one’s family line. Some polymorphisms, such as single

nucleotide polymorphisms (SNPs), alter gene-gene interactions and therefore

a single gene variant is not deterministic. Rather, it expresses itself within the

background of the individual’s total DNA sequence. The terms genome struc-

ture or architecture encapsulate the total set of variants (SNPs and other

sequence differences) within one’s genome. Hence, SNP analyses of individ-

ual genes must be accompanied by analyses of ancestral chromosomal

regions inherited from one’s parents. Several examples of this effect have

been found.

• The HapK haplotype (a collection of SNPs inherited as a unit) in the

leukotriene A4 hydrolase gene (LTA4H). This haplotype confers a relative

increased risk of 1.36 of myocardial infarctions plus cardiovascular disease

(CVD) in European Americans [Helgadottir et al., 2006] but the risk is

almost 5-fold in African Americans (fig. 1). About 27% of the European

Americans in the control group inherited at least one copy of HapK but

that haplotype was present in only 6% of African American controls. Since

preliminary analyses indicated that HapK is very rare in Africa, its occur-

rence in African Americans is due to the presence of European admixture.

Synergistic interactions between the HapK haplotype (European derived)

in LTA4H and other genes of African origin (gene-gene interactions or

epistasis) may increase risk. Another possibility could be different

responses to environmental factors (gene-environment interactions) which

manifest themselves in African-European admixtures.

Kaput 214

• The association of the APOE4 allele with Alzheimer’s disease [Kang et al.,

1997]. Even though the APOE4 variant is twice as common in African

Americans than Asians, Asians develop Alzheimer’s disease five times

more frequently than African Americans [Ntais et al., 2004].

• The association between variants in the calpain-10 gene (CAPN10) and

T2DM is strongest in Mexican Mestizo and Mexican American popula-

tions [del Bosque-Plata et al., 2004].

African American

European American

12

Ancestral group

0.22

0.09

PAR

4.94

1.35

Rel. riskFrequency of HapK

0.18

0.19

MI � CVD

0.04

0.15

control

LTAH4 HapK

Fig. 1. Effect of ancestral inheritance on the effect of a haplotype. Inheriting LTAH4 hap-

lotype K increases the relative risk of myocardial infarction (MI) and other CVD symptoms

depending upon the presence of non-LTAH4 genomic sequences. European-derived regions are

open regions of chromosomes and African-derived are hatched. These other genomic regions

are inherited differently depending upon the ancestral background altering the risk of disease

because of gene-gene interactions. Each individual from an admixed population may have dif-

ferent chromosomal segments from different ancestral groups. Frequency of HapK � The fre-

quency of the allele in the cases (MI � CVD) and controls; Rel. risk � the relative risk of

inheriting the HapK allele; PAR � the population-attributable risk. Although PAR is popula-

tion based, it is often used as a measure of the effect of the allele on the total disease risk. The

values of 0.09 and 0.22 indicate that other genes and environmental factors contribute to the

risk in each population. Adapted from Helgadottir et al. [2006].


Since chromosomal regions are often rearranged during meiosis, each

individual may inherit a unique patchwork of chromosomal regions from differ-

ent ancestral backgrounds, particularly in cultures where several ancestral

groups intermarry. Population stratification has been shown to occur in a seem-

ingly homogeneous population such as European Americans [Campbell et al.,

2005]. Genetic homogeneity is assumed because of unrelated phenotypic

similarities such as skin color, which is a poor substitute for genetic analyses

[Kittles and Weiss, 2003; Shriver et al., 2005]. The challenge for geneticists and

nutrigenomics researchers, however, is to develop analytical approaches for

calculating the effect of gene variants in different genomic (ancestral) back-

grounds. Nonlinear dimensionality reduction or other algorithms [Dawson et

al., 2005; Moore, 2004] hold some promise for analyzing the complex data sets

generated by nutritionists, geneticists, physiologists, and nutrigenomics

researchers.

Complexity of Health and Disease ProcessesHealth is often thought of as the absence of symptoms of disease and, as

importantly, as an all or none phenomenon, i.e. one is either healthy or sick.

This dichotomous view of health and disease does not accurately describe

biological processes. Complex traits, such as health, or the chronic diseases

obesity, diabetes, cardio- or cerebrovascular diseases, Alzheimer’s disease, and

certain cancers are caused by the contribution of many genes interacting with

multiple environmental factors (reviewed in Kaput [2004, 2006]). Indeed, the

list of candidate genes for different chronic diseases attests to the molecular

heterogeneity: a large number of genes in different molecular pathways have

been linked with T2DM [Parikh and Groop, 2004], CVD [Ordovas and Corella,

2005], and, indeed, all chronic diseases including cancers (http://condor.bcm.

tmc.edu/oncogene.html). While the identity and combination of contributing

genes (or their variants) to disease in any one individual have not been ascer-

tained for most chronic diseases, one can conclude that the physiological

complexity of the same disease in different individuals results from many genes

interacting with each other and with environmental factors [Ordovas and

Corella, 2005]. These combinations differ among individuals because of human

genetic variation.

Addressing the Diversity Challenges

Advances in understanding disease processes, the molecular and genetic

workings of biological systems, and increase in health and longevity in

Kaput 216

many countries since the 1950s are a direct result of the worldwide interest in

and support for biomedical research. The United States government appropriated

USD 334.9 billion to the National Institutes of Health (NIH) for the 55-year

period ending in 2005 (http://officeofbudget.od.nih.gov/ui/Appropriations

HistoryByIC.htm). Austria, Belgium, Denmark, Finland, France, Germany,

Greece, Ireland, Italy, Luxembourg, the Netherlands, Portugal, Spain, Sweden,

and the UK (the original European Union countries) provided EUR 536.3

million (approximately USD 0.9 billion) from 1993 to 2003 (NIH appropriations

were USD 137.8 billion in that period). While gains in health measures have

been impressive, it is still not possible to describe in full how disease is initiated

or how to identify genetic or nutrigenomic susceptibility (that is, diet-related

susceptibility to disease). The majority of research funds were earmarked for

disease research – i.e., understanding the etiology of disease processes and link-

ing intake of specific nutrients to disease at the population level and substan-

tially less to disease prevention. Only 4% of the US NIH budget is for nutrition

research (http://www.scnrc.org/docs/starke-reed_files/frame.htm and http://

hnrim.nih.gov/Report/ Nih02_rpt.pdf). Hence, while our progress in biomedical

research has been substantial, it is not possible to develop nutritional recommen-

dations for maintaining optimal health for each individual nor for personalized

medicine. Three significant developments will allow progress in nutrition and

nutrigenomics: the development of high-throughput ‘omic’ (genomic, transcrip-

tomic, proteomic, and metabolomic) technologies, improved experimental

designs, and the development of research collaborations to study complex bio-

logical processes.

TechnologiesThe human genome project provided more than just the sequence of

human DNA: it propelled the development of high-throughput DNA analysis

methods and instruments. The success of DNA and RNA methods spawned

similar advances in high-throughput analyses of proteins and metabolites.

Proteomics and metabolomics provide the additional necessary data sets to

understand genes and their expression patterns.

While the technical advances were of great importance, the human genome

project also initiated a change in the molecular and genetic research culture: the

ability and willingness to analyze complex biological processes rather than only

the reductionist approach of studying single genes and proteins. While reduc-

tionist biology will continue to contribute important insights, whole organisms

can now be studied. Coupled with systems biology approaches of monitoring

macromolecules (DNA, RNA, and protein) and metabolites, these methodolog-

ical approaches will provide a more complete analytical picture of complex bio-

logical processes.


Experimental DesignThe relative ease of analyzing gene variants and transcripts with microar-

rays (e.g. Affymetrix), beads (e.g. Illumina technology), and quantitative real-time

PCR (e.g. Applied Biosystem) allowed the analyses of many candidate gene-

disease and gene-nutrient-phenotype association studies. Many of the initial

studies, however, could not be replicated. This led experts in genetic epidemiol-

ogy to a critical analysis of association study designs. The most common design

flaws are sample sizes that lack appropriate statistical power, control groups

that are not appropriately matched to cases, population stratification that occurs

because of genetic admixtures among study participants, and overinterpretation

of data [Cardon and Bell, 2001; Colhoun et al., 2003; Hopper et al., 2005;

Ioannidis, 2005; Lander and Kruglyak, 1995; Newton-Cheh and Hirschhorn,

2005; Risch, 1997; Tabor et al., 2002].

As a means to improve genetic association studies, several international

collaborations have been established. HuGENet™ (http://www.cdc.gov/genomics/

hugenet/default.htm) [Khoury, 2004], P3G (the Public Population Project in

Genomics, http://www.p3gconsortium.org/), and a ‘network of investigator net-

works’ [Ioannidis et al., 2006] have all been formed for improving human

genetic research by sharing best practices, tools, and methods for analyses of

associations between genetic variation and common diseases (http:// www.cdc.gov/

genomics/hugenet/default.htm).

Nutrigenomic researchers [Kaput, 2004, 2005; Ordovas and Corella, 2004]

added that (1) chronic disease may result from multiple molecular pathways

that may obscure gene-disease or gene-nutrient-phenotype association analy-

ses, (2) the physiological response to the presence of a disease may alter expres-

sion of genetic information, (3) genotype-environment interactions are rarely

taken into account in nutritional or genetic epidemiological experiments, and

these interactions are known to affect the expression of genetic information in

response to different environments, (4) ancestral background should be

included because of epistasis (interaction of genes that are not alleles, espe-

cially the suppression of the effect of one gene by another), and (5) laboratory

animals and cultured cells may not account for genetic or nutritional variations

found in humans. Recommendations to address these issues were proposed as a

means to improve study designs and the reliability of conclusions and are based

on the biological fact that expression of genetic information results from inter-

action of DNA (through the organism) and its environment.

The International EffortRecognizing and acknowledging the limitations of current nutrition,

genetic, and nutrigenomic research designs and strategies, 89 scientists from

22 countries called for strategic international alliances to harness nutrigenomic

Kaput 218

research for personal and public health [Kaput et al., 2005, 2006]. The goals

outlined were to (1) create a federation for sharing data from cell culture exper-

iments, laboratory animal studies, and in particular human nutritional interven-

tion and cohort (prospective and retrospective) studies, (2) develop more highly

powered human studies, (3) improve analyses and consistency of phenotypes,

(4) develop better measurements of food intake, (5) introduce controls for pop-

ulation stratification, (6) analyze a wider array of genetic makeup by recruiting

individuals from different ethnic groups, (7) include other environmental vari-

ables that alter expression of genetic information, and (8) promote interactions

between academia and industry to convert knowledge for the public good.

International collaborations have a rich tradition in scientific research and

many arose through serendipity and personal contacts. The goal of the interna-

tional nutrigenomic research community [Kaput et al., 2005] is to provide the

means for making these collaborative efforts easier to initiate, maintain, and

sustain. The development of best practices and sharing of data will produce

more reliable results than can be obtained through individual efforts.

The Next Steps

While many perceive that nutrigenomics has its ‘home’ exclusively in the

discipline of nutrition, it encompasses the concepts and technologies of numer-

ous research and application fields including genetics, molecular biology, phys-

iology, food science, agriculture, behavioral science, anthropology, ethics, the

food industry, and health care. The integration of these disciplines necessary for

analyzing and understanding nutrient (and environment)-gene interactions for

health and disease processes, that is, into a biological whole, requires intensive

and extensive collaborations. The international research network will aid in

focusing the talents and resources of individuals with diverse expertise in con-

cepts and technologies.

The initial efforts at forming productive collaborations among researchers

are being done through an international network initiated by the European

Nutrigenomics Organisation (NuGO) and called the Nutrigenomics Society.

NuGO is the ideal foundational organization for this effort since it is the only

nutrigenomics organization whose mission is multinational and not explicitly

linked to a specific research program. It is funded by the European Union and

consists of 23 partner institutions in 10 European countries. The mission of

NuGO is to develop, integrate, and facilitate genomic technologies, infrastruc-

ture and research for nutritional science, to train a new generation of nutrige-

nomics scientists, in order to improve the impact of nutrition in health

promotion and disease prevention. The mission of NuGO is consistent with the


many diverse efforts of nutrigenomic centers and researchers around the world,

and is essentially identical to that of the international Nutrigenomics Society.

Since the NuGO was developed and operated to foster communication

among institutes, laboratories, and individuals in Europe, the NuGO website

and resources have been made available to the Nutrigenomics Society.

Members of the international society are collaborating with NuGO to develop a

nutrigenomics information portal for scientists, health care workers, and the

public [Kaput et al., 2006], to participate in the development of best practices,

and collaborate on the research and applications of the science.

The Role of Nutrigenomics Researchers and Businesses in Asia

Asia has about 25% of the globe’s land mass and about 60% of the world’s

population. The ecosystems and human genetic ancestries are highly diverse, pro-

viding a vital resource for analyzing nutrient-gene interactions. Comparing the

nutrigenomic results across genetic ancestries and food intake habits within Asia

and between Asia and other groups throughout the world may illuminate important

pathways for maintaining health and preventing disease among all ancestral popu-

lations. A prime example is the appearance of obesity-related metabolic disorders

that occur at a lower body mass index in Asian (and Mexican [Sanchez-Castillo

et al., 2005]) populations compared to European populations [WHO Expert

Consultation, 2004]. The genetic and molecular explanations for the differences

among Mexicans, Asians, Europeans, and other populations may identify the path-

ways which contribute most to these disorders in different ancestral backgrounds.

Since food types and amounts vary among these populations, nutrigenomics

research is likely to provide the information for improving and maintaining health

of individuals in Asia and the Pacific region, as well as for others in the world.

Nutrigenomics may also provide fundamental and valuable data for

implementing food policies. Genetic analyses have demonstrated that all

humans belong to one species yet differ among individuals and ancestral

groups [Hinds et al., 2005; Jorde and Wooding, 2004; Shriver et al., 2005].

Differential responses to the same food have been observed at the macro level:

certain populations that adopt Western diets and activity levels have higher

levels of obesity and other chronic diseases than European ancestral popula-

tions. One of the best-known examples is type 2 diabetes, which occurs rarely

in Pima Indians living in Mexico while the incidence is approximately 50% for

those whose ancestors migrated to the USA. The incidence of T2DM in the

USA is around 7% (http://diabetes.niddk.nih.gov/dm/pubs/statistics/index.

htm#7). Hence, analyzing the responses of different ancestral groups to foods,

particularly to manufactured foods, is a public health imperative.

Kaput 220

Economic progress and humanitarian aid has reduced the proportion of

undernourished throughout the world, and particularly in China (http://

www.fao.org/docrep/007/y5650e/y5650e03.htm#P1_33). However, at the same

time, type 2 diabetes is projected to double in the next 20 years but most of the

increases will occur in India, China, Pakistan, Mexico, Brazil, and developing

countries [Yach et al., 2006]. Certain ancestral groups appear to be more sensi-

tive to increased calorie consumption and/or physical inactivity. With these sig-

nificant differences in response to calories and macronutrients, one might also

question whether the standard recommended daily intakes of micronutrients is

appropriate for all populations. While it is known that micronutrient intakes dif-

fer among ethnic groups in developed countries [Arab et al., 2003; Hamrosi et

al., 2005; Huang et al., 2002; Manav et al., 2004; Satia-About et al., 2003],

whether different ancestral groups require different amounts of micronutrients

is, apparently, not known. Nutrigenomics research may provide the answers to

these questions and provide the foundation for evidence-based decisions on the

types and amounts of foods needed for individuals in developed and developing

countries.

The Future of Nutrigenomic Science and Applications

Nutrigenomics researchers are advocating that an organism cannot be

completely, or even adequately, analyzed or understood in isolation from its

environment. That is, the reductionist view that genes or their variants alone

will inform us about health or disease processes is incomplete at best, and

perhaps provides erroneous results and conclusions. This strong statement is

justified: all organisms respond at the molecular level to nutrients in their envi-

ronment – if they did not, they would not survive that environment. Hence,

nutrigenomic research is leading the development of a more complete scientific

method, one which includes analyses of genes and the environmental variables

with which they interact.

The practical applications of nutrigenomics are immense: from delivering

the right micronutrients in the optimum amount to the food insecure, to devel-

oping novel foods which are more nutritious, flavorful, storable, and health pro-

moting than many of the products manufactured today.

Acknowledgements

The preparation of the manuscript was supported by the National Center for Minority

Health and Health Disparities Center of Excellence in Nutritional Genomics, Grant MD

00222 and by the European Union, EU FP6 NoE Grant, Contract No. CT2004-50594.


References

Arab L, Carriquiry A, Steck-Scott S, Gaudet MM: Ethnic differences in the nutrient intake adequacy of

premenopausal US women: results from the Third National Health Examination Survey. J Am

Diet Assoc 2003;103:1008–1014.

Campbell CD, Ogburn EL, Lunetta KL, Lyon HN, Freedman ML, Groop LC, Altshuler D, Ardlie KG,

Hirschhorn JN: Demonstrating stratification in a European American population. Nat Genet

2005;37:868–872.

Cantafora A, Blotta I, Rivabene R, Pisciotta L, Bertolini S: Evaluation of RNA messengers involved in

lipid trafficking of human intestinal cells by reverse-transcription polymerase chain reaction with

competimer technology and microchip electrophoresis. Electrophoresis 2003;24:3748–3754.

Cardon LR, Bell JI: Association study designs for complex diseases. Nat Rev Genet 2001;2:91–99.

Carlborg O, Haley CS: Epistasis: too often neglected in complex trait studies? Nat Rev Genet 2004;5:

618–625.

Chiu S, Diament AL, Fisler JS, Warden CH: Gene-gene epistasis and gene-environment interactions

influence diabetes and obesity; in Kaput J, Rodriguez RL (eds): Nutritional Genomics.

Discovering the Path to Personalized Nutrition. Hoboken, Wiley, 2006, pp 135–152.

Colhoun HM, McKeigue PM, Smith GD: Problems of reporting genetic associations with complex out-

comes. Lancet 2003;361:865–872.

Corella D, Ordovas JM: Single nucleotide polymorphisms that influence lipid metabolism: interaction

with dietary factors. Annu Rev Nutr 2005;25:341–390.

Dawson K, Rodriguez RL, Malyj W: Sample phenotype clusters in high-density oligonucleotide

microarray data sets are revealed using Isomap, a nonlinear algorithm. BMC Bioinformatics 2005;

6:195.

del Bosque-Plata L, Aguilar-Salinas CA, Tusie-Luna MT, et al: Association of the calpain-10 gene with

type 2 diabetes mellitus in a Mexican population. Mol Genet Metab 2004;81:122–126.

Enattah NS, Sahi T, Savilahti E, Terwilliger JD, Peltonen L, Jarvela I: Identification of a variant associ-

ated with adult-type hypolactasia. Nat Genet 2002;30:233–237.

Epping-Jordan JE, Galea G, Tukuitonga C, Beaglehole R: Preventing chronic diseases: taking stepwise

action. Lancet 2005;366:1667–1671.

Guo S, Sonenshein G: Green tea polyphenols and cancer prevention; in Kaput J, Rodriguez RL (eds):

Nutritional Genomics. Discovering the Path to Personalized Nutrition. Hoboken, Wiley, 2006,

pp 177–206.

Hamrosi MA, Wallace EM, Riley MD: Iodine status in pregnant women living in Melbourne differs by

ethnic group. Asia Pac J Clin Nutr 2005;14:27–31.

Hartman JL, Garvik B, Hartwell L: Principles for the buffering of genetic variation. Science 2001;291:

1001–1004.

Hartman JL 4th: Genetic and molecular buffering of phenotypes; in Kaput J, Rodriguez RL (eds):

Nutritional Genomics. Discovering the Path to Personalized Nutrition. Hoboken, Wiley, 2006,

pp 105–134.

Harvey CB, Hollox EJ, Poulter M, Wang Y, Rossi M, Auricchio S, Iqbal TH, Cooper BT, Barton R,

Sarner M, Korpela R, Swallow DM: Lactase haplotype frequencies in Caucasians: association

with the lactase persistence/non-persistence polymorphism. Ann Hum Genet 1998;62:215–223.

Helgadottir A, Manolescu A, Helgason A, et al: A variant of the gene encoding leukotriene A4 hydrolase

confers ethnicity-specific risk of myocardial infarction. Nat Genet 2006;38:68–74.

Hinds DA, Stuve LL, Nilsen GB, Halperin E, Eskin E, Ballinger DG, Frazer KA, Cox DR: Whole-genome

patterns of common DNA variation in three human populations. Science 2005;307:1072–1079.

Hopper JL, Bishop DT, Easton DF: Population-based family studies in genetic epidemiology. Lancet

2005;366:1397–1406.

Huang MH, Schocken M, Block G, Sowers M, Gold E, Sternfeld B, Seeman T, Greendale GA: Variation

in nutrient intakes by ethnicity: results from the Study of Women’s Health Across the Nation

(SWAN). Menopause 2002;9:309–319.

Hughes RG, Lawrence MA: Globalization, food and health in Pacific Island countries. Asia Pac J Clin

Nutr 2005;14:298–306.

Kaput 222

Ioannidis JP: Why most published research findings are false. PLoS Med 2005;2:e124.

Ioannidis JP, Gwinn M, Little J, et al: A road map for efficient and reliable human genome epidemiology.

Nat Genet 2006;38:3–5.

James PT, Leach R, Kalamara E, Shayeghi M: The worldwide obesity epidemic. Obes Res 2001;9(suppl 4):

228S–233S.

Jorde LB, Wooding SP: Genetic variation, classification and ‘race’. Nat Genet 2004;36(suppl 1):

S28–S33.

Kang DE, Saitoh T, Chen X, et al: Genetic association of the low-density lipoprotein receptor-related

protein gene (LRP), an apolipoprotein E receptor, with late-onset Alzheimer’s disease. Neurology

1997;49:56–61.

Kaput J: Diet-disease gene interactions. Nutrition 2004;20:26–31.

Kaput J: An introduction and overview of nutritional genomics: application to type 2 diabetes and inter-

national nutrigenomics; in Kaput J, Rodriguez RL (eds): Nutritional Genomics. Discovering the

Path to Personalized Nutrition. Hobeken, Wiley, 2006, pp 1–36.

Kaput J, Astley S, Renkema M, Ordovas J, van Ommen B: Harnessing nutrigenomics: development of

web-based communication, databases, resources, and tools. Genes Nutr 2006;1:5–12.

Kaput J, Ordovas JM, Ferguson L, et al: The case for strategic international alliances to harness nutri-

tional genomics for public and personal health. Br J Nutr 2005;94:623–632.

Kaput J, Rodriguez RL: Nutritional genomics: the next frontier in the postgenomic era. Physiol Genomics

2004;16:166–177.

Khoury MJ: The case for a global human genome epidemiology initiative. Nat Genet 2004;36:

1027–1028.

Kittles RA, Weiss KM: Race, ancestry, and genes: implications for defining disease risk. Annu Rev

Genomics Hum Genet 2003;4:33–67.

Kliewer SA, Lenhard JM, Willson TM, Patel I, Morris DC, Lehmann JM: A prostaglandin J2 metabolite

binds peroxisome proliferator-activated receptor gamma and promotes adipocyte differentiation.

Cell 1995;83:813–819.

Lander E, Kruglyak L: Genetic dissection of complex traits: guidelines for interpreting and reporting

linkage results. Nat Genet 1995;11:241–247.

Lieberman LS: Dietary, evolutionary, and modernizing influences on the prevalence of type 2 diabetes.

Annu Rev Nutr 2003;23:345–377.

Malfatti, MA, Felton JS: Susceptibility to exposure to heterocyclic amines from cooked food: role of

UDP-glucuronosyltransferases; in Kaput J, Rodriguez RL (eds): Nutritional Genomics. Discovering

the Path to Personalized Nutrition. Hoboken, Wiley, 2006, pp 331–352.

Manav M, Su J, Hughes K, Lee HP, Ong CN: Omega-3 fatty acids and selenium as coronary heart dis-

ease risk modifying factors in Asian Indian and Chinese males. Nutrition 2004;20:967–973.

Moore JH: Computational analysis of gene-gene interactions using multifactor dimensionality reduc-

tion. Expert Rev Mol Diagn 2004;4:795–803.

Newton-Cheh C, Hirschhorn JN: Genetic association studies of complex traits: design and analysis

issues. Mutat Res 2005;573:54–69.

Ntais C, Polycarpou A, Ioannidis JP: Meta-analysis of the association of the cathepsin D Ala224Val gene

polymorphism with the risk of Alzheimer’s disease: a HuGE gene-disease association review. Am

J Epidemiol 2004;159:527–536.

Ordovas JM, Corella D: Nutritional genomics. Annu Rev Genomics Hum Genet 2004;5:71–118.

Ordovas JM, Corella D: Genetic variation and lipid metabolism: modulation by dietary factors. Curr

Cardiol Rep 2005;7:480–486.

Orzechowski A, Ostaszewski P, Jank M, Berwid SJ: Bioactive substances of plant origin in food –

Impact on genomics. Reprod Nutr Dev 2002;42:461–477.

Parikh H, Groop L: Candidate genes for type 2 diabetes. Rev Endocr Metab Disord 2004;5:151–176.

Risch N: Evolving methods in genetic epidemiology. 2. Genetic linkage from an epidemiologic perspec-

tive. Epidemiol Rev 1997;19:24–32.

Roweis ST, Saul LK: Nonlinear dimensionality reduction by locally linear embedding. Science 2000;290:

2323–2326.

Rutishauser IH: Dietary intake measurements. Public Health Nutr 2005;8:1100–1107.


Sanchez-Castillo CP, Velasquez-Monroy O, Lara-Esqueda A, et al: Diabetes and hypertension increases

in a society with abdominal obesity: results of the Mexican National Health Survey 2000. Public

Health Nutr 2005;8:53–60.

Satia-Abouta J, Galanko JA, Martin CF, Potter JD, Ammerman A, Sandler RS: Associations of micronu-

trients with colon cancer risk in African Americans and whites: results from the North Carolina

Colon Cancer Study. Cancer Epidemiol Biomarkers Prev 2003;12:747–754.

Schuster GU: Nutrients and gene expression; in Kaput J, Rodriguez RL (eds): Nutritional Genomics.

Discovering the Path to Personalized Nutrition. Hoboken, Wiley, 2006, pp 153–176.

Shriver MD, Mei R, Parra EJ, Sonpar V, et al: Large-scale SNP analysis reveals clustered and continuous

patterns of human genetic variation. Hum Genomics 2005;2:81–89.

Tabor HK, Risch NJ, Myers RM: Candidate-gene approaches for studying complex genetic traits: prac-

tical considerations. Nat Rev Genet 2002;3:391–397.

Tenenbaum JB, de Silva V, Langford JC: A global geometric framework for nonlinear dimensionality

reduction. Science 2000;290:2319–2323.

Tishkoff SA, Verrelli BC: Patterns of human genetic diversity: implications for human evolutionary his-

tory and disease. Annu Rev Genomics Hum Genet 2003;4:293–340.

WHO Expert Consultation: Appropriate body-mass index for Asian populations and its implications for

policy and intervention strategies. Lancet 2004;363:157–163.

Xie HG, Kim RB, Wood AJ, Stein CM: Molecular basis of ethnic differences in drug disposition and

response. Annu Rev Pharmacol Toxicol 2001;41:815–850.

Yach D, Stuckler D, Brownell KD: Epidemiologic and economic consequences of the global epidemics

of obesity and diabetes. Nat Med 2006;12:62–66.

Dr. Jim Kaput

University of Illinois Chicago

MC 958 840 South Wood St.

Chicago, IL 60611 (USA)




ILSI’s First International Conference onNutrigenomics: Opportunities in Asia

Rodolfo F. Florentino

Philippine Association of Nutrition, Metro Manila, Philippines

AbstractILSI’s first international conference on nutrigenomics that was held in Singapore in

December 2005 highlighted the tremendous opportunities of nutrigenomics and the fast

growing ‘omics’ sciences in improving human health. A wide array of topics starting with an

overview of genomics and its application to nutritional science, to the influence of genetic

control and metabolic programming in chronic disease, and to the implications of nutrige-

nomics to individuals and populations, was discussed in nine plenary sessions. The confer-

ence concluded that the future of nutrigenomics in Asia is bright, given strong support in

human resource development, logistical resources, and the participation of the private sector.

Two post-conference symposia followed, dealing with the use of genomics technology in

nutrition research and the application of nutrigenomics in nutritional food science.


The International Life Sciences Institute’s (ILSI) first international confer-

ence on nutrigenomics was held at the Raffles City Convention Center in

Singapore on December 7–9, 2005, with the theme of ‘opportunities in Asia’.

The conference was organized by the ILSI and its Southeast Asia region branch,

in collaboration with the Commonwealth Scientific and Industrial Research

Organization (CSIRO) of Australia. The organizers also received support and

collaboration from other ILSI branches, the National Institutes of Health, USA,

and the Genome Institute of Singapore, Singapore.

The aim of the 3-day conference was to promote the understanding of this

new frontier in nutritional science with the key objective of stimulating research

on the application of nutrigenomics in health promotion and disease prevention.

Attended by close to 400 participants from 33 countries, the conference brought

together over 50 internationally renowned experts and regional scientists to

share their knowledge and the latest developments in the ‘omics’ technologies.

Executive Summary

ILSI’s First International Conference on Nutrigenomics 225

Their insightful presentations were key to the success of the conference, and

many speakers have kindly contributed papers to this book. In appreciation of

the contributions by the other speakers, a summary of their presentations is

shared here with the readers.

Summary of Plenary Presentations

Plenary Session 1: New Perspectives on Genes, Nutrients and HealthThe opening plenary session 1, focusing on new perspectives on genes,

nutrients and health, was co-chaired by Dr. Sushila Chang of Ngee Ann

Polytechnic, Singapore, and Dr. Graeme Young of Flinders University, Australia.

Dr. John Milner, Chief of the Nutritional Science Research Group, National

Cancer Institute, National Institutes of Health, USA, was one of the two keynote

speakers. He gave a comprehensive overview of genomics and its application to

nutritional science, particularly its capacity to offer unprecedented opportunities

to achieve genetic potential, increase physical and mental productivity, and

reduce the risk and consequences of disease. He pointed out that every individ-

ual is different, responding differently to food and food components, as deter-

mined by the phenotype. Thus, the appeal of nutrigenomics, as it elucidates the

interaction between bioactive components in foods and cellular processes, is its

potential for an individualized approach to health and nutrition, to recognize cul-

tural and ethnic differences, and its potential to open new market niches for food

and pharmaceutical companies. The study of nutrigenetics could clarify the role

of genetic differences in response to various doses of a nutrient, just as the study

of transcriptomics could provide clues about molecular targets of specific food

components. Apart from nutrigenetics, epigenetic profiling could give us clues

as to individuals who might benefit from intervention. On the other hand, pro-

teomics and metabolomics are fresh and exciting areas that could also elucidate

the varied responses to bioactive food components. Dr. Milner concluded that we

still do not know enough. We need to identify and validate biomarkers for effect

and susceptibility; we need to effectively communicate these omics information

to the health care community and consumers, and at the same time we need to

work within a responsible bioethical framework.

The other keynote speaker, Dr. Edison Liu, Executive Director of the

Genome Institute of Singapore, Singapore, spoke on how concepts of functional

genomics are very relevant today, as illustrated in their studies in the area of

cancer. For example, with the modern genomic and computational technologies

available today, one can study classes of tumors and collections of genes in a hier-

archical cluster at the same time, and answer the question – ‘Can the genetic expres-

sion of a tumor determine the expression pattern of the tumor?’ Experiments in

Florentino 226

mice showed that expression profile provides a footprint of the molecular history

of the tumor organized along biochemical pathways: the primacy of pathways

over individual genes. This principle is apparently applicable in humans as well.

In fact, the p53 gene deficiency classifier by microarray technology could out-

perform the gold standard of DNA sequencing and in addition has prognostic

significance in specific therapeutic subgroups. One could also conclude that the

p53 signature may reflect the operational configuration of this pathway in breast

cancer. The impact of environmental exposure to tumor phenotype was shown by

studies on the outcome of breast cancer in patients given hormone replacement

therapy. Microarray technology could give a definitive imprint of what tumor

will arise as a result of exogenous exposure. In conclusion, Dr. Liu stated that

nutritional science will have to move from an epidemiologic science to a thera-

peutic science and from case-control to cohort studies, with the aid of very com-

prehensive genomic studies. Furthermore, considering the complexity of the

human organism, we need to employ a systems approach, enlisting a combina-

tion of clinical, technological, and mathematical skills.

Plenary Session 2: Nutritional Influences on Molecular Epidemiology and DiseasesThis plenary session was co-chaired by Dr. Choon Nam Ong of the National

University of Singapore, Singapore, and Dr. Dieter Söll of Yale University, USA.

The first speaker was Dr. Carl Keen, Chair of the Department of Nutrition,

University of California, Davis, USA, who gave an overview of the broad con-

cepts of the interrelationships among food, genetics and chronic disease. The

changing expectations of a healthy diet towards ‘optimal nutrition’, from preven-

tion of nutritional deficiencies to reduction of risk of age-related diseases, has

changed the concept of recommended dietary allowances to dietary recom-

mended intakes in some countries, incorporating the concept of disease preven-

tion. At the same time, there is a growing demand to take into account the

individual’s or population’s genetic background, lifestyle habits and physical envi-

ronment in defining nutritional needs. We have seen the influence of nutrition in

obesity, diabetes, cardiovascular disease, hyperlipidemia and even some cancers,

but the issue to be addressed is the influence of genetic control and metabolic pro-

gramming in all of these factors. This is where the field of nutritional genomics,

which studies the molecular interactions between nutritional stimuli and the

genome, and how these interactions promote health and prevent disease, can con-

tribute. Dr. Keen enumerated five key tenets of nutritional genomics: (1)

improper diets are risk factors for diseases; (2) dietary chemicals alter gene

expression and/or genome structure; (3) influence of diet on health depends upon

an individual’s genetic makeup; (4) genes regulated by diet play a role in chronic

diseases, and (5) diets based upon genotype, nutritional requirements and status


can prevent or mitigate chronic disease. The goal then is individualized nutrition

and genome-based dietary recommendations to achieve and maintain optimal

health, and to prevent, mitigate and treat disease. This would entail, among others,

a technology for low-cost, high-throughput single nucleotide polymorphism

analysis, sequencing and gene expression profiling; access to data sets for large,

diverse human populations; bioinformatic tools and theories for the visualization

of large, complex data sets, and a ‘systems biology’ approach to investigating the

relationships between nutrition and disease.

The genetic influence on obesity was discussed by Dr. Jong Ho Lee, Full

Professor of Food and Nutrition at the College of Human Ecology, Yonsei

University, Korea. Dr. Lee explained that interindividual variation in obesity

results from action of multiple genes and environmental factors. Two types of

human studies are used to identify the specific variants that affect obesity: link-

age analysis and association studies. Linkage analysis has been successful in

mapping genes responsible for single gene disorders, but these conditions are

very rare. Most cases of obesity are polygenic, arising from multiple genes with

a small contribution to the obesity phenotype, and for these, association studies

are required. Candidate genes are selected on the basis of their function in bio-

chemical pathways related to the regulation of energy balance or to the adipose

tissue biology. Currently, positive associations with obesity phenotypes have

been reported for more than 70 genes. Dr. Lee illustrated these studies with

their studies on adinopectin and perilipin. Their data have confirmed that

the �276G/T polymorphism of the adiponectin gene modulates circulating

adiponectin and insulin resistance, particularly in obese states independently

from common environmental factors. They also found that the perilipin locus is

a determinant of coronary artery disease risk in Koreans. In addition, the per-

ilipin gene may be involved in lipid metabolism and systemic inflammation in

coronary artery disease patients who have high visceral fat accumulation.

Dr. Lee concluded that genetic markers of obesity may help in the identification

of individuals who are at greater risk of obesity and its comorbidities, so that

personalized nutrigenetic programs may help people make the lifestyle and

dietary changes best suited to their needs.

The topic of healthy aging and its relation to genetics, plasma lipids and diet

was discussed by Dr. José Ordovas, Director of the Nutrition and Genomics

Laboratory at the Jean Mayer USDA Human Nutrition Research Center on Aging

at Tufts University, USA. The most explored gene in terms of associations with

longevity has been the apolipoprotein E (APOE) gene. It has been observed that

the presence of the APOE4 allele is associated with a decreased life span in

Caucasians but not in Asians. These differences suggest either ethnic-specific

gene-gene or gene-environment interactions that modulate the effects of the

APOE4 allele on life span. Studies on other lipid candidate genes such as APOA1,

Florentino 228

APOA4, APOB, and CETP have also shown significant gene-diet interactions,

suggesting that life span could be affected by the interactions between our genetic

makeup and dietary habits. Dr. Ordovas suggested that large prospective studies

need to be designed, supported by extensive genotyping and analytical capacities,

in order to fully benefit from the contribution of genetics to longevity.

Other speakers in this plenary session were:

• Dr. E. Shyong Tai, Consultant and Clinical Specialist in the Department of

Endocrinology at the Singapore General Hospital, Singapore, who focused

on the genetics of lipoprotein metabolism and heart disease.

• Dr. Mohan Viswanathan, President and Director of the Madras Diabetes

Research Foundation, India, who discussed polymorphisms and dietary

influences on diabetes.

• Dr. Boonsong Ongphiphadhanakul, Professor in the Division of Endo-

crinology and Metabolism at the Faculty of Medicine, Mahidol University,

Thailand, who discussed the genetic perspectives on osteoporosis.

Plenary Session 3: Nutrient-Gene InteractionsOn the second day of the conference, the third plenary session was co-

chaired by Dr. Shuhei Kobayashi of the University of Human Arts and Sciences,

Japan, and Dr. Michael Fenech of CSIRO Human Nutrition, Australia.

Dr. Carl Keen, Professor and Chair of the Department of Nutrition at the

University of California, Davis, USA, discussed nutrient-gene interactions, par-

ticularly in early development. It is well known that a common factor contribut-

ing to pregnancy complications is suboptimal nutrition during embryonic and

fetal development. Recent studies now appear to establish the link between

nutrition and gene expression during these stages of development. As an exam-

ple, Dr. Keen cited the effect of maternal dietary manganese intake on the phe-

notypic expression of the pallid gene or other strain variations in the developing

conceptus, resulting in ataxia, otolith development and other defects in various

animals, and may perhaps be responsible for various syndromes in humans

including epilepsy. Other examples he cited are the effects of maternal dietary

deficiency in copper (as in Menke’s syndrome), iron (both short-term and long-

term effects), and zinc (developmental defects). Other micronutrient deficien-

cies during prenatal development can result in behavioral, immunological and

biochemical abnormalities. In some cases, these may be secondary to epigenetic

or development changes in DNA methylation patterns which may persist into

adulthood, influencing the individual’s risk to certain chronic diseases including

hypertension, diabetes and cardiovascular disease. On the other hand, there is

evidence that certain nutrient-gene interactions during early development may

decrease the risk for some chronic diseases. According to Dr. Keen, the chal-

lenge for the next decade will be the determination of the epigenetic or persistent


consequences associated with mild micronutrient deficiencies during early

development, and the extent to which the persistent effects contribute to the risk

for age-related chronic diseases.


• Dr. John Mathers, Professor of Human Nutrition in the School of Clinical

Medical Science at the University of Newcastle, UK, who presented an

extensive review of the impact of epigenetics on early nutrition.

• Dr. Eiji Takeda, Professor in the Department of Clinical Nutrition at the

Institute of Health Biosciences, University of Tokushima Graduate School,

Japan. Dr. Takeda presented their study on the effect of Inslow, a low

glycemic index liquid formula, in carbohydrate, protein and fatty acid

metabolism and in gene expression.

• Dr. Huynh The Hung, Associate Professor at the Division of Cellular and

Molecular Research at the National Cancer Center, Singapore, who spoke

on the effects of dietary quercetin and kaempferol, which are bioactive

phytochemicals in some vegetables and fruits, in the regulation of cell pro-

liferation and apoptosis.

Plenary Session 4: Nutrigenomics in Cancer Risk ReductionThe fourth plenary session was co-chaired by Dr. Edison Liu of the

Genome Institute of Singapore, Singapore and Dr. Lynnette Ferguson of the

University of Auckland, New Zealand.

One of the speakers was Dr. Mimi Yu, a cancer epidemiologist and

McKnight Presidential Professor at the Cancer Center of the University of

Minnesota, USA. Dr. Yu reviewed studies on tea, particularly green tea and its

protective effect against breast cancer. It has been shown that green tea polyphe-

nols are antioxidants, capable of upregulating phase II enzymes and downregu-

lating phase I enzymes. It has also been shown that either black or green tea

extracts containing catechins can inhibit chemically induced mammary tumors

in animals, and suppress growth of human breast cancer cell lines. More recent

epidemiological data have consistently shown green tea (as opposed to black tea)

drinkers to be associated with breast cancer protection, primarily women pos-

sessing the COMT low-activity genotype. COMT is a key enzyme in tea cate-

chin excretion. The protective effect of green tea against breast cancer is primarily

seen in women possessing the angiotensin-converting enzyme high-activity

genotype. Angiotensin II is a potent angiogenic factor. Thus, anti-angiogenesis

may be a mechanistic pathway behind green tea/breast cancer protection. On the

other hand, recent epidemiologic studies have shown no evidence of a protective

effect of black tea. In fact, cohort studies such as the Singapore Chinese Health

Study suggest a modest increase in the risk of breast cancer with intake of black

tea, possibly because of the increase in circulating estrogen.

Florentino 230


• Dr. Graeme Young, Professor of Gastroenterology at Flinders University

and Head of Gastrointestinal Services at Flinders Medical Center,

Australia. Dr. Young gave a comprehensive discussion of the linkage

between dietary factors and genomic stability, particularly as related to

cancer development.

• Dr. Dongxin Lin, Director and Professor in the Department of Cancer Etiology

and Carcinogenesis at the Cancer Institute and Hospital of the Chinese

Academy of Medicinal Sciences, China, who discussed folate-metabolizing

enzymes and their relationship with gastroesophageal cancer risk.

Plenary Session 5: Nutrigenomics and Molecular ImmunomodulationThe fifth plenary session was co-chaired by Dr. Lee Yuan Kun of the

National University of Singapore and Dr. Nico van Belzen of ILSI Europe.

The first speaker, Dr. Lynnette Ferguson, Head of the Department of

Nutrition and the Center for Mutagen Testing in Auckland, New Zealand,

discussed polymorphisms in genes as they affect immune response and suscepti-

bility to disease. It has been advanced that inflammatory processes including the

release of proinflammatory cytokines and the formation of reactive oxygen and

nitrogen species underlie many chronic diseases including cancer. Dr. Ferguson

particularly focused her discussion on the manner in which polymorphisms in

genes affecting selenium status interact with dietary selenium to affect the risk

of prostate cancer in a high-risk population. Their studies in Auckland among

adults 50–75 years of age suggest that high levels of DNA damage in white

blood cells could predict a high prostate cancer risk and therefore provide a use-

ful biomarker, and that there is increased white blood cell DNA damage and an

increased risk of prostate cancer in GPX1 variants carrying the TT allele, which

could be reversed by selenium supplementation. On the other hand, they showed

that there was decreased white blood cell DNA damage and a decreased risk of

prostate cancer in GPX4 variants carrying the TT allele, which could be

increased by selenium supplementation. Dr. Ferguson concluded that selenium

requirements should be justified on an individual basis in order to reduce the risk

of chronic inflammation and its attendant effect on reactive oxygen species.

The next speaker was Dr. Nina Rautonen, Director of Bioscience at

Danisco Innovation, Finland, who discussed the role of pre- and probiotics in

immune regulation as they interact with the intestinal mucosa, particularly on

their interaction at the molecular level. Using the Caco-2 cell model, they stud-

ied the effects of pre- and probiotics on cyclooxygenase (COX) expression.

COX are enzymes that generate prostaglandins which mediate inflammatory

responses: COX-1 is expressed constitutively, while the expression of COX-2 is


induced in inflammatory reactions and in various cancers. Overexpression of

COX-2 has been linked with early stages of colon cancer development. Results

of their studies showed that specific probiotic strains can induce different COX

expression patterns. For example, Bif420 causes a significant decrease in COX-2

expression, while Lactobacillus acidophilus causes a significant increase.

Using an automated continuous multistage simulator, they also showed that pre-

biotics could alter COX expression in the colonocytes via microbial fermenta-

tion. Dr. Rautonen concluded that the use of different in vitro models could be

a useful tool for studying the molecular mechanisms of the interactions between

pre- and probiotics and the intestinal mucosa.

Dr. Michael Müller, Full Professor and Chair of Nutrition, Metabolism and

Genomics in the Division of Human Nutrition, Wagenigen University, The

Netherlands, spoke on the role of polyunsaturated fatty acids as modulators of

the immune system partly mediated by peroxisome proliferator-activated receptor-�(PPAR-�). Dr. Müller reviewed their attempts to characterize the immunemodu-

latory and anti-inflammatory functions of PPAR-� by means of whole-genome

microarray analysis. He explained that the anti-inflammatory role of PPAR-�has mainly been attributed to its capacity to downregulate proinflammatory

genes and to modulate the NF-�B pathway involved in translocation and activa-

tion of the proinflammatory transcription factor. Dr. Müller concluded that

PPAR-� has an important role in the cellular adaptation to changes in free fatty

acid levels during fasting and feeding, in particular by modulating the innate

immune response, and that downregulation of proinflammatory genes by

polyunsaturated fatty acids is a major part of this PPAR-mediated process.

Plenary Session 6: Nutrigenomics in Other Physiological States and Well-BeingThe sixth plenary session was co-chaired by Dr. Junshi Chen of the ILSI

Focal Point in China, and Dr. John Matthers of the School of Clinical Medical

Sciences, University of Newcastle, UK.

The first speaker was Dr. Lin He, Professor and Head of the Laboratory of

Nutrigenomics, Institute of Nutritional Sciences at the Shanghai Institute of

Biological Sciences of the Chinese Academy of Medical Sciences. Dr. He first

discussed their study on iodine deficiency disorders. It is well known that thyroid

hormones, besides their trophic effect on growth and energy metabolism, also

affect and regulate over 100 enzymes, affecting transcription of many genes by

binding to mitochondria, nuclear and cell membrane receptors. They investigated

the interaction between genetic factors and iodine-deficient physical environment

in determining the overall risk of iodine deficiency disorders in a iodine-deficient

region in China with a high prevalence of iodine deficiency disorders. Haplo-

type analysis revealed a positive association between the APOE gene and the

Florentino 232

borderline mental retardation group but not the mental retardation group. They

also found an association between APOE and metabolic disorders of the thyroid

hormones, T3 and T4, and/or the thyroid-stimulating hormone. Dr. He proceeded

to discuss the results of their studies in the Anhui region in China, one of the

provinces most affected by severe famine during the period from 1959 to 1961, on

the rates of schizophrenia before, during and after the famine years. They

observed that among the exposed birth cohorts, there was a highly significant

twofold increase in the risk of developing schizophrenia in later life. The fact that

the proportion of familial to sporadic cases remained the same in the exposed and

unexposed birth cohorts suggests that the increased risk may be due to lowering

of the genetic risk threshold for schizophrenia and/or an increase in survival of

carriers of schizophrenia risk alleles. Dr. He concluded that genome variation is

likely to play a major role. Their investigation in this area is presently continuing.

The next speaker was Dr. Katsuya Nagai, Vice Director and Professor,

Institute of Protein Research at Osaka University, Japan, who discussed genes

and circadian rhythm, and their relationship with metabolism and food regula-

tion. He briefly described the mechanism of circadian rhythm in mammals as

originating from a master circadian clock in the hypothalamic suprachiasmatic

nucleus which integrates circadian rhythms of brain functions, autonomic

nerves and the endocrine glands, thus generating circadian rhythms of enzyme

activities, hormonal levels and behavior. He discussed the positive-negative

feedback loop on transcriptions of clock-related genes such as period 1, 2 and 3

and cryptochrome 1 and 2, and novel genes such as the suprachiasmatic nucleus

circadian oscillatory protein and the period-1-interacting protein of the suprachi-

asmatic nucleus. Apart from involvement in circadian rhythm, Dr. Nagai

obtained evidence that the suprachiasmatic nucleus is involved in the mecha-

nism of homeostasis through the control of autonomic activities. Their findings

suggest that the molecular mechanism of the circadian clock and histamine

nervous system is involved in the maintenance of homeostasis such as energy

metabolism, blood glucose and blood pressure.

The other speaker in this plenary session was:

• Dr. Cecile Delcourt, Researcher at the National Institute of Health and

Medical Research (INSERM), Laboratory of Epidemiology, Public Health

and Development, France, who discussed the application of nutrigenomics

in eye health, particularly in relation to age-related macular degeneration,

cataract and glaucoma.

Plenary Session 7: Use and Impact of Genomics in the Food SupplyThis plenary session was co-chaired by Dr. William Padolina of the

International Rice Research Institute in the Philippines, and Dr. Lynne Cobiac

of CSIRO Preventative Health Flagship, Australia.


The first speaker was Dr. Peter Weber, Corporate Scientist at the Human

Nutrition and Health Department of DSM Nutritional Products. Dr. Weber gave

some examples of how DSM is using the nutrigenomic approach in human

nutrition, such as in the identification of new bioactive ingredients; understand-

ing the mode of action of new as well as of established ingredients; establishing

biomarkers; developing safe bioactive ingredients, and, for the future, develop-

ing personalized nutritional solutions. As a take-home message, Dr. Weber

emphasized that nutrigenomic technology should not be looked at as a ‘stand-

alone’ tool. Rather, this technology should complement the established (classi-

cal) tools required to provide the scientific evidence to identify, understand and

develop bioactive food ingredients.

The third speaker in this session was Dr. Gerard Barry who is the Harvest

Plus Rice Crop Team Leader and Head of the International Rice Research

Institute’s Intellectual Property Management Unit in the Philippines. Dr. Barry

discussed the work of Harvest Plus with the primary goal of breeding crops for

better human nutrition. The biofortification effort is seen as an additional option

for alleviating micronutrient deficiency, and currently efforts are focused on

increasing the iron, zinc and provitamin A content of major crops, including

maize, rice, beans, cassava, wheat and sweet potato. In addition, preliminary work

has started on sorghum, peanut and other crops. The majority of approaches rely

on germ plasm screening and breeding, but transgenic approaches are also being

pursued in cases where no sufficient variation could be found in the crop. Dr. Barry

illustrated this with the work on maize where �-carotene-rich varieties are being

screened; orange-fleshed sweet potatoes which are already being promoted as

weaning food in some countries; low-phytate maize to improve the bioavailability

of zinc, and the well-known Golden Rice with its high level of �-carotene and

other provitamin A carotenoids. In fact, UK scientists have developed a new

genetically modified strain of the Golden Rice, producing more �-carotene.

Genomic and transgenic approaches are being used to raise the levels of iron and

zinc of cereal crops. Recent feeding trials with the high-iron rice (IR68144) have

shown an impact on improving total body iron among nonanemic iron-deficient

subjects. Dr. Barry reported that Harvest Plus coverage has been increasing, with

the involvement of more centers around the world. Other supporters have come in

such as the Gates Foundation, which is supporting work on bananas, cassava,

sorghum and rice, all with the aim of increasing the micronutrient density of sta-

ple crops for the alleviation of micronutrient deficiency.


• Dr. Ahmed El-Sohemy, Assistant Professor at the Department of

Nutritional Sciences at the Faculty of Medicine, University of Toronto,

Canada, who discussed the nutrigenomics of taste and its impact on food

preferences and food production.

Florentino 234

• Dr. Shaun Coffey, Chief of CSIRO Livestock Industries, Australia, who

discussed how nutrigenomics approaches are being explored to improve

the quality of meat and dairy products for consumption.

Plenary Session 8: Nutrigenomics – Individual and Population ImplicationsThe eighth plenary session was co-chaired by Dr. Heng Leng Chee of the

National University of Singapore, Singapore, and Dr. Aman Wirakartakasumah of

the United Nations Educational, Scientific and Cultural Organization (UNESCO).

The first speaker was Dr. Ben van Ommen of the TNO Quality of Life and

the European Nutrigenomics Organization, The Netherlands. Dr. van Ommen

stated that, driven by the unravelling of the human genome and its related tech-

nological developments, the so-called omics sciences have advanced insights

into the influence of genetic polymorphisms on nutritional metabolism.

Systems biology has provided new insights into the broad molecular action of

nutrients, as it captures patterns, profiles and complex data sets arising from

complex interactions, rather than of single ‘target’ gene responses. Dr. van

Ommen cited systemic inflammation, now appearing to be a central component

in body homeostasis, and its possible involvement in chronic diseases such as

atherosclerosis and the comorbidities of obesity. Thus, we now know the many

processes involved in cholesterol metabolism, suggesting points of intervention

in the prevention of atherosclerosis, using not just pharmaceuticals but nutrition

and diet that could act on these same points of intervention. Using metabolomics

in the study of lipid metabolism, we can discern gene expression changes after

exposure to dietary cholesterol stress. This becomes clearer using ‘knowledge

network’ charts and bioinformatics, enabling us to look at the big picture, in

order to understand, e.g., how cholesterol gene expression changes are related

to inflammation through their effect on macrophage activation. With this

information, we could make a network-based analysis of systemic inflammation.

Likewise, we can use metabolomics to understand the effect of inflammation in

disorders associated with obesity such as insulin resistance, hyperlipidemia,

diabetes, and hypertension. For example, differences in gene expression

changes are seen when rats are given a high-carbohydrate or high-protein diet,

and these are reflected in the balance between immune effector cells and regu-

latory cells in the gut. It means that the cellular response to infection is a very

complex matter, involving a multitude of enzymes, receptors, adaptors, and

transcriptors. Dr. van Ommen concluded that systems biology is needed to

merge nutrition research with physiology and biomedical research. This is nec-

essary in order to understand both the complexity of the many processes

involved in ‘keeping us healthy’, and to study the many ‘overarching’ processes

like inflammation, both in the whole organism (‘systemic’) and in specific/


local processes (e.g. atherosclerosis, gut health). The goal is to be able to subtly

modulate these processes with food bioactives and thus establish a new concept

of a balanced nutrition targeted to tailor-made health.

Dr. Umar Jenie, Head of the Indonesian Institute of Sciences and Professor

of Organic Medicinal Chemistry at the Gadjah Mada University, Indonesia,

spoke on the ethical and social implications of nutrigenomics. While nutrige-

nomics offers great promise in benefiting human health, there is the concern that

genomic data could be misused with the ultimate result of creating inequity.

Dr. Jenie discussed some ethical issues related to nutrigenomics, such as consent

prior to collection of data; access of information; public and individuality issues;

prevention of inequity, and regulatory oversight. He reviewed the International

Declaration of Human Genomic Data issued by UNESCO in 2003, which forms

the basis for ensuring respect for human rights in the collection, processing, use

and storage of human genetic data. Dr. Jenie concluded that while there are

many bioethical issues, there is rarely just one ‘right answer’.

The other speaker in this plenary session was:

• Dr. Michael Fenech, Principal Research Scientist and Project Leader for

Nutrigenomics and Genome Health, CSIRO Human Nutrition, Australia,

who spoke on genome health as a nutrigenomics approach to setting

dietary recommendations.

Plenary Session 9: Nutrigenomics – The FutureThe last plenary session of the conference was co-chaired by Dr. Rodolfo

Florentino of the Nutrition Foundation of the Philippines, and Dr. Suzanne

Harris of ILSI.

The first speaker in this session was Dr. Richard Head, Director of CSIRO

Preventative Health Flagship and Affiliate Professor in the Department of

Clinical and Experimental Pharmacology at the University of Adelaide,

Australia. Dr. Head believed that the availability of analytical tools in the area

of genomics and proteomics, together with the measurement of biological

markers, coupled with appropriate mathematics, have contributed to a better

understanding of the health potential of food. However, there are important

challenges to be faced. In the area of biomarkers, the focus is on the nature

of sampling and the influence of physiological and related influences on

approaches to sampling. In the area of proteomics, the key challenge is to both

characterize and understand the influence of nutrients on protein expression. In

the area of genomics, the challenge is in understanding the role of the influence

of nutrients in gene expression. In all of these issues, the role of mathematical-

based science, including bioinformatics, is critical. Dr. Head illustrated these

issues with their current work in these areas: how to link single nucleotide

polymorphisms to chromosome positioning as in the study of prostate cancer;

Florentino 236

gene profiling in colorectal cancer; 3-dimensional display of mass spectrometer

data, and cell-nutrient interactions with colonic bacteria and their byproducts.

In conclusion, Dr. Head stated that the future of nutrigenomics is interlinked

with addressing the challenges of biomarkers and cell-nutrient interactions,

aided by appropriate mathematics.

The next speaker was Dr. Ben van Ommen of the TNO Quality of Life and

the European Nutrigenomics Organization, The Netherlands, who spoke on the

role of nutrigenomics in global health improvement. Dr. van Ommen said that one

of the major challenges in nutrition and health is the increasing emphasis on dis-

ease prevention. Technology is no longer a limiting factor. There are now many

examples of applications of the omics technologies in nutrition. Many examples

of genetic differences relevant to nutrition and health have been studied, and there

is now a better understanding of the complexity of the interactions between genes

and nutrition. However, there is no suitable strategy for tackling multiple small

genetic differences, and nutritional systems biology still needs to be developed. In

addition, effective communication of this new knowledge to consumers and

health professionals must be developed, and new business models must be identi-

fied by the food industry. Moving beyond cohorts and statistics, a systems biol-

ogy approach is needed to model the nutrigenomics complexity in a single

person. For example, the dynamics of the whole cholesterol metabolism in a

hypothetical person can be modeled using a systems biology markup language.

While consumers may be ready for health food, the key issue is whether sci-

ence and industry are ready to meet the demand. Personalized nutrition might

not be there yet, but it is driving us towards ‘community nutrition’: molecular

epidemiology, biotech foods, and food for primary prevention. However, Dr. van

Ommen concluded that nutrigenomics is too big for any single discipline; multi-

ple primary collaborations involving a merger of analytical, informatics and bio-

logical capacities are required. Fortunately, a number of regional and global

initiatives (such as the European Nutrigenomics Organization) support these

developments, giving a bright future for nutritional systems biology.

The last speaker in this session was Dr. Sakarindr Bhumiratana, President of

the National Science and Technology Development Agency, Thailand, who spoke

on the potential of nutrigenomics, particularly in Asia. He declared that nutrige-

nomics is an exciting science with its myriad implications in improving human

health. With the development in the omics sciences, we have the capacity to better

understand, through systems biology, individual responsiveness to food, microbes

and the environment. This could lead to the development of new foods for opti-

mum health, not to mention high-value foods and ingredients with their huge eco-

nomic potential. Dr. Bhumiratana pointed to the broad diversity in Asia, i.e. the

region’s rich biodiversity, the high ethnic diversity of the Asian population, the wide

usage of traditional foods and herbs with their time-tested functional properties,


and the rapidly growing developments in science and technology, which all make

Asia an extremely fertile area for nutrigenomics research. What is needed is

human resource development in research and development, favorable policies to

encourage research and development in science and technology, together with the

provision of logistical resources and involvement of the private sector. The estab-

lishments of Centers of Excellence, regional and international linkages through

research and development, collaboration between scientists, and mechanisms for

joint funding are also important considerations. In conclusion, Dr. Bhumiratana

foresees a bright future for nutrigenomics in Asia, with increasing numbers of aca-

demics, researchers, research students and small start-ups to take full advantage of

Asian opportunities and niches arising from the area’s biodiversity, ethnic and food

diversity. There are already excellent world-class centers in genomics research in a

number of countries in Asia, and certainly more will be established in the future.

Symposia and Workshop

The conference program also included two symposia, namely ‘Using

Genomics Technologies in Nutrition Research’, held on December 7, 2005, and

‘Nutrigenomics and Functional Foods Science’, held on December 9, 2005. A

workshop on ‘Intellectual Property in New Product Development’ was held on

December 8, 2005.

Symposium on Using Genomics Technologies in Nutrition ResearchThe symposium, which aimed to discuss in greater detail the use and appli-

cation of genomics technologies in nutrition-related research, was chaired by

Dr. Sakarindr Bhumiratana of the National Science and Technology

Development Agency, Thailand, and Dr. David Mitchell of CSIRO, Australia.

Dr. Christopher Wong of the Genome Institute of Singapore, Singapore,

presented the different technologies available to study biological events on a

genome-wide scale. He highlighted the use of gene transcription profiling

technologies, such as microarrays in cancer biology; comparative genomic

hybridization array to study chromosomal aberrations, and pathogen detection

chip to identify unique DNA sequences in all viral genomes, including those

that can be found in the contaminated food supply. He then shared the studies

done on caloric restrictions, lymphocyte gene expression and PPAR-� in

chronic diseases. He also added that the gene identification signature/paired-

end diTag sequencing technology may be used to discover novel proteins, while

high content screening automated imaging approach to cell-based assays for

individual cell data in multiple-cell assays. He stressed the importance of seam-

less information management, if integrated data derived from different experi-

ment techniques are to be achieved.

Florentino 238

Research in nutrigenomics is not a simple solution. One must take into

account the complexity of various factors including nutrients, genotypes,

lifestyle and phenotypes, and available technology. Dr. Michael Muller of

Wageningen University, the Netherlands, shares the perspective of interpreting

the wealth of data generated in nutrigenomics research in a biological lingo.

Gene ontology was set up by the Gene Ontology Consortium in 1998 to provide

a comprehensive, controlled set of terms that can be used to describe genes

in all organisms to assist in the interpretation of the variably regulated gene

expression. Dr. Muller also highlighted another challenge in nutrigenomics data

interpretation – namely pathway analysis. While there are established databases

on known pathways, the challenge lies in the discovery of new pathways, such

as regulatory pathways. The use of modeling in the gene set enrichment analy-

sis to determine a statistically significant set of genes between two biological

states may assist in identifying gene sets involved in a certain metabolic path-

way, signal transduction route, or overexpressed in a specific cancer type. Thus,

while databases are readily available, ingenuity is an essential skill for bioinfor-

maticians to interpret nutrigenomics data.

One of the great challenges of maintaining health in the 21st century

includes the acknowledgement on the use of food-based interventions as a solu-

tion. Dr. Laurent B. Fay of the Nestlé Research Center, Switzerland, began his

presentation by introducing the concept of metabolomics, which is a technology

used in metabolite fingerprinting in the field of metabolomics. Mass spec-

troscopy and nuclear magnetic resonance act as complementary data sets in

metabolomics analysis, as nuclear magnetic resonance provides high throughput

over a wide range of chemical classes, and mass spectroscopy data can provide

high sensitivity and resolution, and allow comparison in large, available identi-

fication databases. Currently, the highest resolving power of all mass spectroscopy

techniques available has Fourier transform mass spectrometry. In a study con-

ducted to correlate chocolate consumption to mood uplifting, it was found that

chocolate cravers have a different metabolism as compared to noncravers.

Using the Fourier transform ion cyclotron resonance mass spectrometry

metabolite profiling model, Dr. Fay was able to show strong response following

the consumption of chocolate in predictive biomarkers in plasma metabolites.

Other speakers in this symposium were:

• Dr. Jong-Eun Lee of DNA Link, Inc., Korea, who shared the technology

developed to address the rapidly expanding field of molecular epidemiology,

which requires a sensitive and specific technique in genotype screening.

• Dr. Visith Thongboonkerd of Mahidol University, Thailand, who shared

gel-based two-dimensional polyacrylamide gel electrophoresis and gel-

free two-dimensional liquid chromatography as techniques for protein

separation.


Symposium on Nutrigenomics and Functional Food ScienceThis symposium aimed to derive the data generated in nutrigenomics

research to substantiate efficacy and safety of functional ingredients, through

identification and validation of biomarkers. The chairs of the symposium were

Dr. E-Siong Tee of the Nutrition Society of Malaysia, Malaysia, and Dr. Richard,

Head of CSIRO, Australia.

Dr. Keiko Abe of the University of Tokyo, Japan, opened the session with an

overview of how nutrigenomics had been used to evaluate functional foods and

presented some case studies on soy protein isolate, hyperlipidemia and athero-

sclerosis; cacao polyphenol, fat deposition and fatty acid metabolism; sesamin

and alcohol metabolism; fructo-oligosaccharides and allergenic responses. She

argued that nutrigenomics can be used to explain the causal relationships of

evoked effects, assess food functionality and safety, identify appropriate bio-

markers, and provide a better understanding of physiological processes. She also

highlighted the establishment of a national project on systems biology and ILSI

Japan’s initiatives on functional food science and nutrigenomics.

Dr. Junshi Chen of ILSI Focal Point in China shared the study results of tea

and oral cancers in animals and humans, using silver-stained nucleolar organ-

izer region, proliferation cell nuclear antigen and epidermal growth factor

receptor as biomarkers. Both the animal model and human subjects showed sig-

nificant protection afforded by mixed tea in oral mucosa cells and leukoplakia

mucosa cells, respectively. These findings demonstrated that the selected bio-

markers met the criteria of effective biomarkers for disease risk as agreed at the

meeting Functional Foods – Scientific and Global Perspectives, ILSI Europe,

2002. The responses of silver-stained nucleolar organizer region, proliferation

cell nuclear antigen and epidermal growth factor receptor may serve as the sub-

stantiation of mixed tea functionality and mechanisms of action that may lead to

the development of functional food products.

In nutritional epidemiology, biomarkers can also be used to substantiate

dose-response relationships and suggest potential mechanisms of action.

Dr. Adeline Seow of the National University of Singapore, Singapore, presented

compelling evidence on the level of sulforaphane metabolites in the plasma cor-

related to the consumption amount of cruciferous vegetables. Isothiocyanates

(ITC) induce phase 2 metabolic enzymes, including glutathione-S-transferase

(GST), but GST can also conjugate – and thus inactivate – ITC. M1 and T1

polymorphisms in the genes coding for GST enzymes had been shown to abol-

ish the activity of GST, as shown in the higher ITC urinary output in subjects

with polymorphisms compared to those with fully functioning GST. At the tis-

sue level, GST-null individuals are afforded greater protection of ITC against

chemical carcinogens. This is apparent in many studies, including those con-

ducted among male smokers in Shanghai and male and female smokers in the

Florentino 240

USA. GST-null individuals were also afforded greater protection from breast

cancer as observed among Shanghai women and from colorectal cancer seen

among Chinese people in Singapore.

Dr. Choon Nam Ong of the National University of Singapore, Singapore,

shared the use of cellular proteomics in nutrition research by presenting the study

on the effects of �-phenylethyl ITC on liver cancer cells. It was shown, through

proteomics technologies, that �-phenylethyl ITC increases reactive oxygen

species production in the liver by 9 differentially expressed proteins – most of

which are related to oxidative stress. He argued that the cellular proteomics

approach may be used as a tool to better understand the molecular targets of nutri-

ents and to obtain substantiation of the functionality and mechanisms of action.

The other speaker in this symposium was:

• Dr. James Dekker of Fonterra, New Zealand, who shared the development

of probiotic strains as functional ingredients.

Workshop on Intellectual Property in New Product DevelopmentThe protection of intellectual property in the sciences is an increasingly

important topic. Facilitated by Dr. Sushila Chang of Ngee Ann Polytechnic,

Singapore, and Mr. Martin Schweiger of Schweiger & Partner, the workshop

introduced the various types of intellectual properties, addressed the Singapore

Copyright and Patent Act, and highlighted that similar regulatory mechanisms

may not be available in other countries.

Dr. Chang opened the workshop with a brief presentation on the differ-

ent definitions and nature of copyrights, patents, trademarks, trade secrets and

registered designs. She also shared the steps required to protect and exploit the

intellectual property relating to a product or a process. Mr. Schweiger explained

the patent process in greater detail and discussed the benefits and values of

patents. Short case studies were also shared at this meeting.

Conclusion and Recommendations

This conference, being ILSI’s first international meeting to focus on the excit-

ing new field of nutrigenomics, highlighted the tremendous potential that nutrige-

nomics and the other so-called omics sciences have in improving human health.

The conference showed how, with the unravelling of the human genome and its

related technological developments, nutrigenomics has widened our understand-

ing of the molecular interactions between nutritional stimuli and the genome, with

the goal of achieving our genetic potential, increasing physical and mental produc-

tivity, and reducing the risk and consequences of disease. Thus, it was shown how

nutrigenomics has increased our understanding of the influence of genetic control


and metabolic programming in chronic diseases including obesity, diabetes, car-

diovascular disease, hyperlipidemia, osteoporosis, and even some forms of cancer.

The conference illustrated the application of omics technologies in discerning the

wide influence of genetic differences in nutrition including physiological respon-

ses to nutrition stimuli and recommended intake of nutrients. A major goal is indi-

vidualized nutrition and genome-based dietary recommendations not only to

achieve and maintain optimum health, but also to prevent and mitigate disease. On

the other hand, the conference highlighted the promise of genomic and transgenic

approaches in improving our food supply. Biofortification efforts in major crops

already hold promise in contributing to the alleviation of micronutrient deficien-

cies that plague populations in many parts of the world, while nutrigenomics app-

roaches have been utilized to improve the quality of meat, dairy products and other

foods for consumption of the general population or for special population needs.

At the same time, the conference emphasized that nutrigenomic technologies

should not be perceived and applied as a stand-alone tool. Aside from the estab-

lished ‘classical’ tools, the role of mathematical-based science including bioinfor-

matics is critical. Nutrigenomics is too big for any single discipline; multiple

collaborations involving a merger of analytical, informatics and biological capaci-

ties are required. Nutrition should move beyond cohorts and statistics. Systems

biology is needed in modeling the complexity of the numerous interlinking

processes in nutritional metabolism in order to capture patterns, profiles, and com-

plex data sets arising from these complex metabolic interactions. Besides provid-

ing a clearer understanding of the influence of nutrients on human health, systems

biology could lead to the development of new foods for optimum health, not to

mention high-value foods and ingredients with their high economic potential.

The conference concluded that the future of nutrigenomics in Asia is bright,

considering Asia’s broad diversity not only in culture but in biodiversity and eth-

nicity, and its wide range of traditional foods and herbs. The rapid developments

in science and technology in most of Asia offer many opportunities for nutrige-

nomics research. What is needed is human resource development in research and

development, favorable science and technology policies including the provision

of sufficient logistics, and the participation of the private sector. Finally, the need

to establish Centers of Excellence as well as international and regional linkages

was expressed. At the same time, it is necessary to look for ways to effectively

communicate these omics information to the health care community and con-

sumers, while working within a responsible bioethical framework.

Dr. Rodolfo F. Florentino

Nutrition Foundation of the Philippines

18 May Street, Congressional Village

Quezon City, 1106 (Philippines)

Tel./Fax �63 2 926 7838, E-Mail [email protected]

242

Arai, H. 127

Cobiac, L. 31

Coffey, S.G. 183

Collett, M. 196

Corella, D. 102

Cornelis, M.C. 176

Deepa, R. 118

Dekker, J. 196

Delcourt, C. 168

El-Sohemy, A. 25, 176

Fenech, M. 49

Florentino, R.F. 224

Fontaine-Bisson, B. 176

Fukaya, M. 127

Gopal, P. 196

Hung, H. 146

Jenie, U.A. 66

Kaput, J. 102, 209

Khataan, N. 176

Kwong, P. 176

Lee, J.-E. 97

Li, H. 140

Lin, D. 140

Mathers, J.C. 42

Matsuo, K. 127

Miao, X. 140

Milner, J.A. 1

Mohan, V. 118

Muto, K. 127

Ongphiphadhanakul, B. 158

Ordovas, J.M. 102

Ozsungur, S. 176

Prasad, J. 196

Radha, V. 118

Radhika, G. 118

Rema, M. 118

Sakuma, M. 127

Slamet-Loedin, I.H. 67

Stewart, L. 176

Sudha, V. 118

Tai, E.S. 110

Takeda, E. 127

Taketani, Y. 127

Tan, W. 140

Thongboonkerd, V. 80

Wang, L. 140

Yamamoto, H. 127

Yamanaka-Okumura, H. 127

Young, G.P. 91

Author Index

243

Adiponectin, single nucleotide

polymorphism 227

Age-related macular degeneration

clinical features 169

genetics 171

oxidative stress and nutrient protection

169, 170

Akt, quercetin-induced apoptosis role in

lung cancer cells 151–154

Alcohol dehydrogenase, single nucleotide

polymorphisms and blood lipid levels

113

Alu, features 6, 7

Apolipoproteins

APOE4 allele 214, 228

APOE haplotypes 232

polymorphisms and variability in dietary

lipid response 105, 111, 114

Apoptosis, quercetin modulation in lung

cancer cells 151–156

Association studies

experimental design 217

limitations 161

Bifidobacterium lactis, see Probiotics

Biomarkers, nutrition response studies

4, 5

Breast cancer, catechin protection 229

Calcium

deficiency and genomic instability

53–55

food content 57

osteoporosis and genetics in response

161, 162

Calpain-10, gene variants 214

Cancer, see specific cancers

�-Carotene

deficiency and genomic instability 55

food content 57

rice engineering 233

Cataract

gene polymorphisms 171

oxidative stress and nutrient protection

169, 170

Catechins, breast cancer protection

229

Chennai Urban Population Study

heritability of diabetes 121

overview 119

physical activity findings 123

Chennai Urban Rural Epidemiology Study

diabetes trends 120

gene-environment interactions in diabetes

123, 124

overview 119, 120

susceptibility gene search 121–123

Cholesterol, see High-density lipoprotein

cholesterol, Low-density lipoprotein

cholesterol

Cholesteryl ester transfer protein

gene-diet interactions 228

single nucleotide polymorphisms and

blood lipid levels 111, 112

Collagen type 1 �1-chain, osteoporosis

candidate gene 159, 160

Subject Index

Subject Index 244

Complement

activation in age-related macular

degeneration 171

zinc interactions 171

Confidentiality, data collection 71–74

Consent, data collection 69–71

Coronary artery disease

blood lipids

genetic variation and ethnic differences

111

nutrient-gene interactions and risk

relevance 112–115

nutrient-gene interactions in lipoprotein

metabolism

interaction mechanisms 103, 104

interventional studies 105, 106

observational studies 105

overview 102

postprandial lipidemia 106

prospects for study 106, 107

pathogenesis 215

postprandial hyperglycemia risks, seePostprandial hyperglycemia

Cyclooxygenase, probiotic effects on

expression 230

Cytochromes P450, polymorphisms 27, 28

Dairy products

animal production system

challenges 186

intervention timing 189, 190

nutrigenomics targets 190, 191

bovine genome sequencing and mapping

187, 188

consumption 184, 185

dietary fat and chronic disease

185, 186

genomics limitations 188, 189

transgenic animal prospects 191, 192

Diabetes mellitus

epidemiology 118, 119

gene-diet interactions 125

gene-environment interactions 123, 124

India, see Chennai Urban Population

Study, Chennai Urban Rural

Epidemiology Study

pathogenesis 215

postprandial hyperglycemia

mechanisms of diabetic complications

oxidative stress 129

serotonin metabolism 129–131

palatinose-based food suppression

human studies 132

metabolism gene expression

response 133–135

preparation of food 131, 132

rat studies 132, 133

vascular disease risks 128

prevention 136, 137

DNA damage, see Genomic instability

DNA methylation

aging 33

dietary factor effects in early life

45, 46

diet effects 14, 15, 33, 34, 45

diseases 33

gene silencing 49, 50

mechanism 33

DNA microarray, genotyping 99

Docosahexaenoic acid

age-related macular degeneration

benefits 170

oxidation 169

DR10, see Probiotics

DR20, see Probiotics

Epigenetics

definition 13, 43

dietary factor effects in early life

45–47

gut microflora effects 37

malleability of markings 44, 45

mechanisms 14, 15, 32–36, 43, 44

Esophageal cancer, see Gastroesophageal

cancer

Estrogen receptor-�osteoporosis candidate gene 160

polymorphism and bone health 162

Ethics, nutrigenomics studies 67–78

Extracellular signal-regulated kinase,

quercetin-induced apoptosis role in lung

cancer cells 151–153, 155, 156

Fatty acid synthetase, palatinose-based food

effects on expression 135

Flavonoids, see Quercetin

Subject Index 245

Folate


53–55

food content 57

functions 140, 141

metabolism 140, 141

Fonterra probiotics, see Probiotics

Food preferences, nutrigenomics of taste

176–181

Gastroesophageal cancer

methylenetetrahydrofolate reductase

polymorphisms in esophageal

squamous cell carcinoma 142

thymidylate synthase polymorphisms

142, 143

Genetic screening

definition 74

inequity protection 74, 75

occupational screening 75, 76

Genomic instability

adverse health outcomes 50–52, 91

diet and DNA repair 92–95

genome health clinic concept 61

genome health nutrigenomics 59–61

nutrient deficiency effects 52–54

nutriome concept 56–58

oncogenesis 91, 92

overview 36

recommended dietary allowance

considerations 58, 59

Genomics

animal genome sequencing 187, 188

Human Genome Project 216

overview 5, 6

Genotyping, see Single nucleotide

polymorphism

Glutathione S-transferase

cancer studies 28

induction 28, 240

isoforms 28

polymorphisms in cataract 171

Glycemic index

low-index food, see Palatinose-based food

overview 127, 128

Gut microflora, see also Probiotics

abundance 197

epigenetic effects 37

Haplotype, see also Single nucleotide

polymorphism

HapMap Project 13, 67, 70, 73, 77, 99,

100

Hepatic lipase, single nucleotide

polymorphisms and blood lipid levels

112, 113

High-density lipoprotein cholesterol

genetic and ethnic variation in blood

levels 111

protective effects 110

Histone

diet effects on modification 35, 36

dietary factor effects in early life 45, 46

modification 15, 34, 35, 44

structure 34

Homocysteine, osteoporosis risk factor 163

Human Genome Project 216

International Declaration of Human Genetic

Data 76, 78

International Life Sciences Institute,

international nutrigenomics conference

goals 224

intellectual property workshop

240, 241

participants 224

plenary sessions

new perspectives on genes, nutrients

and health 225, 226

nutrient-gene interactions 228

nutrigenomics and molecular

immunomodulation 230, 231

nutrigenomics

cancer risk reduction 229, 230

individual and population

implications 234, 235

other physiological states and

well-being 231–233

the future 235–237

nutritional influences on molecular

epidemiology and diseases 226–228

use and impact of genomics in the food

supply 233, 234

recommendations 240, 241

symposia

genomics technologies in nutrition

research 237–239

Subject Index 246

nutrigenomics and functional food

science 239, 240

Intervention trials, design 2, 3, 38

Iron


53

rice engineering 234

Isothiocyanates, phase 2 metabolic enzyme

induction 239

Keap1, diet effects 16

Lactobacillus rhamnosus, see Probiotics

Leukotriene A4 hydrolase, single nucleotide

polymorphisms 213

Low birth weight

adult disease risks 42, 43

epigenetic studies 45, 46

Low-density lipoprotein cholesterol

cardiovascular risks 102, 110

diet and individual response 102

genetic and ethnic variation in blood

levels 111

Low-density lipoprotein receptor-related

protein 5, osteoporosis candidate gene

160, 161

Lung cancer, quercetin effects on A549

cells

apoptosis assay and effects 150–156

materials 147, 148

overview 147

proliferation assay and effects 148, 150

Western blot analysis 149

Lutein, eye protection 169

Magnesium, deficiency and genomic

instability 53, 54

Malnutrition, prevalence 210

Manganese, deficiency and genomic

instability 53

Manganese superoxide dismutase, single

nucleotide polymorphisms 12, 13

Mass Array assay, genotyping 98, 99

Mass spectrometry, proteomics techniques

83–85

Maturity-onset diabetes of the young,

genetic screening 74

Meat

animal genome sequencing 187

animal production system

challenges 186

intervention timing 189, 190

nutrigenomics targets 190, 191

consumption 184, 185

dietary fat and chronic disease

185, 186

genomics limitations 188, 189

transgenic animal prospects 191, 192

Metabolomics, overview 18–20

Methylenetetrahydrofolate reductase

bone health and polymorphisms 163

esophageal squamous cell carcinoma

polymorphisms 142

function 141

pancreatic cancer polymorphisms 143,

144

single nucleotide polymorphisms 10, 11,

60, 141

Micronucleus assay, genomic instability

studies of nutrient deficiency effects

50–52, 55

Milk, see Dairy products

Niacin


food content 57

Nrf2, diet effects 16

Nutrigenetics, overview 31, 32, 68

Nutrigenomics

Asia research and business 219, 220,

224, 225

bioactive foods 9, 10

definition 68

diversity challenges

food chemical diversity 211, 212

genetic diversity 212–215

health and disease diversity 215

ethics and social implications 67–78

funding of research 216

international conference, seeInternational Life Sciences Institute

international research collaborations

217, 218

International Life Sciences Institute,

(continued)

symposia (continued)

Subject Index 247

overview 8, 9, 66, 67, 210

prospects 220

Nutrigenomics Organisation 218, 219

Nutriome, concept 56–58

Osteoporosis

candidate genes

association study limitations 161

collagen type 1 �1-chain 159, 160

estrogen receptor-� 160

low-density lipoprotein receptor-

related protein 5 160, 161

vitamin D receptor 159

genetic determinants of bone response

calcium 161, 162

estrogen 162

homocysteine as risk factor 163

heredity 159

Oxidative stress

eye disease 169, 170

hyperglycemia 129

Palatinose-based food

hyperglycemia suppression

human studies 132

metabolism gene expression response

133–135


preparation 131, 132

Pancreatic cancer, folate-metabolizing

enzyme polymorphisms 143, 144

Peroxisome proliferator-activated

receptors

anti-inflammatory actions 231

function 103

palatinose-based food effects on

expression 134, 135

polymorphisms

diabetes susceptibility 121

regulated genes 103, 104, 114

types 103

Phenylketonuria, nutrigenomics 210

Phenylthiocarbamide, nutrigenomics of

taste 177–181

Plasma cell glycoprotein 1, polymorphisms

and diabetes susceptibility 121, 122

Postprandial hyperglycemia

mechanisms of diabetic complications

oxidative stress 129

serotonin metabolism 129–131

palatinose-based food suppression

human studies 132

metabolism gene expression response

133–135

preparation of food 131, 132


vascular disease risks 128

Postprandial lipidemia, gene-diet

interactions 106

Potassium, nutriproteomics studies 85, 86

Privacy, data collection 71–74

Probiotics

cyclooxygenase expression effects 231

definition 197

Fonterra strains

efficacy 200–204

microbial ecology studies 199, 200

products 204

prospects for study

animal models 205, 206

assays 205

fermentation 206

genetic tools 205

genomics 204, 205

safety 198, 199

viability 198

6-Propylthiouracil, nutrigenomics of taste

177–181

Proteomics

capillary electrophoresis/mass

spectrometry 84

liquid chromatography/tandem mass

spectrometry 83

mass spectrophotometric immunoassay

84, 85

overview 17, 18, 81, 82

potassium nutriproteomics studies 85, 86

protein chip 83, 84

two-dimensional gel electrophoresis 82,

83

Quercetin

flavonoid regulation of carcinogenesis

146, 147

intake 147

lung cancer A549 cell effects

Subject Index 248

apoptosis assay and effects 150–156

materials 147, 148

overview 147

proliferation assay and effects 148,

150

Western blot analysis 149

Recommended dietary allowance, genome

health considerations 58, 59

Retinol, food content 57

RNA interference, applications 16, 17

RNA silencing, overview 14, 32

Selenium


protein modification regulation 18

Serotonin, hyperglycemia effects on

metabolism 129–131

Short tandem repeats, features 7

Single nucleotide polymorphism, see alsospecific genes

abundance 26, 98

association studies using markers

98–100

disease susceptibility 7, 8

features 7, 97, 98

food response effects 10–13

haplotyping 13

HapMap Project 13, 67, 70, 73, 77, 99, 100

high-throughput genotyping 98–100

lipoprotein metabolism and nutrient-gene

interactions 102–107

Superoxide dismutase, see Manganese

superoxide dismutase

Suprachiasmatic nucleus, autonomic control

232

Systems biology, overview 86, 87

TASR38, nutrigenomics of taste 177–181

Taste, see Food preferences

Thymidylate synthase

function 141

gastroesophageal cancer polymorphisms

142, 143

pancreatic cancer polymorphisms 143,

144

Transcriptomics, overview 15–17

Transgenic animal, prospects in meat and

dairy production 191, 192

Universal Declaration on Bioethics and

Human Rights 67, 68, 72, 76–78

Universal Declaration on the Human

Genome and Human Rights 76

Vitamin B12, genome health maintenance

58, 59

Vitamin C, deficiency and genomic

instability 53

Vitamin D receptor

osteoporosis candidate gene 159

single nucleotide polymorphisms 11, 12,

162

Vitamin E


food content 57

Zeaxanthin, eye protection 169

Zinc

complement interactions 171


Quercetin (continued)

lung cancer A549 cell effects (continued)

Documents

nutrigenomics