Upload
arun-das
View
213
Download
1
Embed Size (px)
DESCRIPTION
about nutrigenomics and other genoms and diet interaction
Citation preview
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.
Nutrition and Genomics 5
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.
Nutrition and Genomics 7
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 and Health
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
Nutrition and Genomics 9
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
Nutrition and Genomics 11
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
Nutrition and Genomics 13
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
Nutrition and Genomics 15
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
Nutrition and Genomics 17
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
Nutrition and Genomics 19
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,
Nutrition and Genomics 21
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.
Nutrition and Genomics 23
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.
52 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.
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]
Tai ES, Gillies PJ (eds): Nutrigenomics – Opportunities in Asia.
Forum Nutr. Basel, Karger, 2007, vol 60, pp 25–30
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.
Copyright © 2007 S. Karger AG, Basel
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.
2 Ordovas JM, Corella D: Nutritional genomics. Annu Rev Genomics Hum Genet 2004;5:71–118.
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)
Tel. �1 416 946 5776, Fax �1 416 978 5882, E-Mail [email protected]
Tai ES, Gillies PJ (eds): Nutrigenomics – Opportunities in Asia.
Forum Nutr. Basel, Karger, 2007, vol 60, pp 31–41
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.
Copyright © 2007 S. Karger AG, Basel
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
Epigenomics and Nutrition 35
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
Epigenomics and Nutrition 37
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
Epigenomics and Nutrition 39
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.
Epigenomics and Nutrition 41
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)
Tel. �61 8 8204 4645, Fax �61 8 8204 6406, E-Mail [email protected]
Tai ES, Gillies PJ (eds): Nutrigenomics – Opportunities in Asia.
Forum Nutr. Basel, Karger, 2007, vol 60, pp 42–48
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.
Copyright © 2007 S. Karger AG, Basel
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
Nutrition and Epigenetics 45
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
Nutrition and Epigenetics 47
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]
Tai ES, Gillies PJ (eds): Nutrigenomics – Opportunities in Asia.
Forum Nutr. Basel, Karger, 2007, vol 60, pp 49–65
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).
Copyright © 2007 S. Karger AG, Basel
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.
Nutrition and Genome Health 53
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.
Nutrition and Genome Health 55
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].
Nutrition and Genome Health 57
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
Nutrition and Genome Health 59
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
Nutrition and Genome Health 61
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.
Nutrition and Genome Health 63
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.
Nutrition and Genome Health 65
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)
Tel. �61 8 8303 8880, Fax �61 8 8303 8899, E-Mail [email protected]
Tai ES, Gillies PJ (eds): Nutrigenomics – Opportunities in Asia.
Forum Nutr. Basel, Karger, 2007, vol 60, pp 66–79
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.
Copyright © 2007 S. Karger AG, Basel
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
Nutrition: Ethics and Social Implications 69
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
Slamet-Loedin/Jenie 70
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
Nutrition: Ethics and Social Implications 71
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
Slamet-Loedin/Jenie 72
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
Nutrition: Ethics and Social Implications 73
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
Slamet-Loedin/Jenie 74
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
Nutrition: Ethics and Social Implications 75
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
Slamet-Loedin/Jenie 76
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
Nutrition: Ethics and Social Implications 77
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].
Slamet-Loedin/Jenie 78
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.
Nutrition: Ethics and Social Implications 79
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]
Tai ES, Gillies PJ (eds): Nutrigenomics – Opportunities in Asia.
Forum Nutr. Basel, Karger, 2007, vol 60, pp 80–90
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.
Copyright © 2007 S. Karger AG, Basel
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.
43 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.
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]
Tai ES, Gillies PJ (eds): Nutrigenomics – Opportunities in Asia.
Forum Nutr. Basel, Karger, 2007, vol 60, pp 91–96
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.
Copyright © 2007 S. Karger AG, Basel
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)
Tel. �61 8 8204 4964, Fax �61 8 8204 3943, E-Mail [email protected]
Tai ES, Gillies PJ (eds): Nutrigenomics – Opportunities in Asia.
Forum Nutr. Basel, Karger, 2007, vol 60, pp 97–101
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.
Copyright © 2007 S. Karger AG, Basel
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)
Tel. �82 2 364 4700, Fax �82 2 364 4778, E-Mail [email protected]
Tai ES, Gillies PJ (eds): Nutrigenomics – Opportunities in Asia.
Forum Nutr. Basel, Karger, 2007, vol 60, pp 102–109
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
Nutrigenomics and Health
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.
Genes, Diet and Nutrition 105
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
Ordovas/Corella/Kaput 106
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.
Genes, Diet and Nutrition 107
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
Ordovas/Corella/Kaput 108
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,
Genes, Diet and Nutrition 109
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.
17 Ordovas JM, Corella D: Nutritional genomics. Annu Rev Genomics Hum Genet 2004;5:71–118.
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)
Tel. �1 617 556 3102, Fax �1 617 556 3103, E-Mail [email protected]
Tai ES, Gillies PJ (eds): Nutrigenomics – Opportunities in Asia.
Forum Nutr. Basel, Karger, 2007, vol 60, pp 110–117
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.
Copyright © 2007 S. Karger AG, Basel
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
The Genetics of Lipoprotein Metabolism and Heart Disease 113
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.
The Genetics of Lipoprotein Metabolism and Heart Disease 115
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.
The Genetics of Lipoprotein Metabolism and Heart Disease 117
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]
Tai ES, Gillies PJ (eds): Nutrigenomics – Opportunities in Asia.
Forum Nutr. Basel, Karger, 2007, vol 60, pp 118–126
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.
Copyright © 2007 S. Karger AG, Basel
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
Gene-Environment Interaction in Diabetes 121
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.
Mohan/Sudha/Radhika/Radha/Rema/Deepa 122
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
Gene-Environment Interaction in Diabetes 123
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.
Mohan/Sudha/Radhika/Radha/Rema/Deepa 124
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
� visible fat (�median)
Family history negative
� visible fat (�median)
Family history positive
� visible fat (�median)
Fig. 2. Synergistic effect of heritability and visible fat on the prevalence of glucose
intolerance – the CURES.
Gene-Environment Interaction in Diabetes 125
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.
Mohan/Sudha/Radhika/Radha/Rema/Deepa 126
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)
Tel. �91 44 4396 8888, Fax �91 44 2835 0935, E-Mail [email protected]
Tai ES, Gillies PJ (eds): Nutrigenomics – Opportunities in Asia.
Forum Nutr. Basel, Karger, 2007, vol 60, pp 127–139
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.
Copyright © 2007 S. Karger AG, Basel
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
Takeda/Arai/Muto/Matsuo/Sakuma/Fukaya/Yamanaka-Okumura/Yamamoto/Taketani 130
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.
Gene Expression in Low Glycemic Index Diet 131
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.
Takeda/Arai/Muto/Matsuo/Sakuma/Fukaya/Yamanaka-Okumura/Yamamoto/Taketani 132
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.
Gene Expression in Low Glycemic Index Diet 133
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).
Takeda/Arai/Muto/Matsuo/Sakuma/Fukaya/Yamanaka-Okumura/Yamamoto/Taketani 134
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-�
Gene Expression in Low Glycemic Index Diet 135
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].
Takeda/Arai/Muto/Matsuo/Sakuma/Fukaya/Yamanaka-Okumura/Yamamoto/Taketani 136
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.
Gene Expression in Low Glycemic Index Diet 137
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.
Takeda/Arai/Muto/Matsuo/Sakuma/Fukaya/Yamanaka-Okumura/Yamamoto/Taketani 138
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.
Gene Expression in Low Glycemic Index Diet 139
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)
Tel. �81 88 633 7093, Fax �81 88 633 7094, E-Mail [email protected]
Tai ES, Gillies PJ (eds): Nutrigenomics – Opportunities in Asia.
Forum Nutr. Basel, Karger, 2007, vol 60, pp 140–145
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.
Copyright © 2007 S. Karger AG, Basel
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/
Genetic Polymorphisms and Risk of Gastroesophageal Cancers 143
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.
Genetic Polymorphisms and Risk of Gastroesophageal Cancers 145
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)
Tel. �86 10 8778 8491, Fax �86 10 6772 2460, E-Mail [email protected]
Tai ES, Gillies PJ (eds): Nutrigenomics – Opportunities in Asia.
Forum Nutr. Basel, Karger, 2007, vol 60, pp 146–157
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.
Copyright © 2007 S. Karger AG, Basel
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.
Quercetin and Lung Cancer 149
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.
Quercetin and Lung Cancer 151
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.
Quercetin and Lung Cancer 153
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.
Quercetin and Lung Cancer 155
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.
Quercetin and Lung Cancer 157
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]
Tai ES, Gillies PJ (eds): Nutrigenomics – Opportunities in Asia.
Forum Nutr. Basel, Karger, 2007, vol 60, pp 158–167
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.
Copyright © 2007 S. Karger AG, Basel
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
Osteoporosis, Genetics and Environment 161
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
Ongphiphadhanakul 162
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.
Osteoporosis, Genetics and Environment 163
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
Ongphiphadhanakul 164
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.
Osteoporosis, Genetics and Environment 165
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.
Ongphiphadhanakul 166
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
Osteoporosis, Genetics and Environment 167
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]
Tai ES, Gillies PJ (eds): Nutrigenomics – Opportunities in Asia.
Forum Nutr. Basel, Karger, 2007, vol 60, pp 168–175
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.
Copyright © 2007 S. Karger AG, Basel
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
Nutrigenomics and Eye Health 171
[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.
Nutrigenomics and Eye Health 173
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.
Nutrigenomics and Eye Health 175
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]
Tai ES, Gillies PJ (eds): Nutrigenomics – Opportunities in Asia.
Forum Nutr. Basel, Karger, 2007, vol 60, pp 176–182
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.
Copyright © 2007 S. Karger AG, Basel
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.
Nutrigenomics of Taste 179
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
El-Sohemy/Stewart/Khataan/Fontaine-Bisson/Kwong/Ozsungur/Cornelis 180
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
Nutrigenomics of Taste 181
[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.
El-Sohemy/Stewart/Khataan/Fontaine-Bisson/Kwong/Ozsungur/Cornelis 182
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)
Tel. �1 416 946 5776, Fax �1 416 978 5882, E-Mail [email protected]
Tai ES, Gillies PJ (eds): Nutrigenomics – Opportunities in Asia.
Forum Nutr. Basel, Karger, 2007, vol 60, pp 183–195
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.
Copyright © 2007 S. Karger AG, Basel
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.
Improving Nutritional Quality of Food 187
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.
Improving Nutritional Quality of Food 189
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];
Improving Nutritional Quality of Food 191
• 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
Improving Nutritional Quality of Food 193
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.
Improving Nutritional Quality of Food 195
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]
Tai ES, Gillies PJ (eds): Nutrigenomics – Opportunities in Asia.
Forum Nutr. Basel, Karger, 2007, vol 60, pp 196–208
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.
Copyright © 2007 S. Karger AG, Basel
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
Functionality of Probiotics 199
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
Dekker/Collett/Prasad/Gopal 200
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.
Functionality of Probiotics 201
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
Dekker/Collett/Prasad/Gopal 202
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
Functionality of Probiotics 203
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
Dekker/Collett/Prasad/Gopal 204
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
Functionality of Probiotics 205
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
Dekker/Collett/Prasad/Gopal 206
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.
Functionality of Probiotics 207
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.
Dekker/Collett/Prasad/Gopal 208
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)
Tel. �64 6 350 6323, Fax �74 6 350 4658, E-Mail [email protected]
Tai ES, Gillies PJ (eds): Nutrigenomics – Opportunities in Asia.
Forum Nutr. Basel, Karger, 2007, vol 60, pp 209–223
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.
Copyright © 2007 S. Karger AG, Basel
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
Developing the Promise of Nutrigenomics 213
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].
Developing the Promise of Nutrigenomics 215
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.
Developing the Promise of Nutrigenomics 217
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
Developing the Promise of Nutrigenomics 219
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.
Developing the Promise of Nutrigenomics 221
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.
Developing the Promise of Nutrigenomics 223
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)
Tel. �1 312 371 1540, Fax �1 312 996 0669, E-Mail [email protected]
Tai ES, Gillies PJ (eds): Nutrigenomics – Opportunities in Asia.
Forum Nutr. Basel, Karger, 2007, vol 60, pp 224–241
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.
Copyright © 2007 S. Karger AG, Basel
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
ILSI’s First International Conference on Nutrigenomics 227
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
ILSI’s First International Conference on Nutrigenomics 229
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.
Other speakers in this plenary session were:
• 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
Other speakers in this plenary session were:
• 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
ILSI’s First International Conference on Nutrigenomics 231
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.
ILSI’s First International Conference on Nutrigenomics 233
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.
Other speakers in this plenary session were:
• 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/
ILSI’s First International Conference on Nutrigenomics 235
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,
ILSI’s First International Conference on Nutrigenomics 237
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.
ILSI’s First International Conference on Nutrigenomics 239
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
ILSI’s First International Conference on Nutrigenomics 241
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
deficiency and genomic instability
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
deficiency and genomic instability
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
deficiency and genomic instability 53
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
rat studies 132, 133
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
rat studies 132, 133
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
deficiency and genomic instability 53
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
deficiency and genomic instability 53
food content 57
Zeaxanthin, eye protection 169
Zinc
complement interactions 171
deficiency and genomic instability 53
Quercetin (continued)
lung cancer A549 cell effects (continued)