Pharmacogenomics and Nutrigenomics: Synergies and Differences

REVIEW Pharmacogenomics and nutrigenomics: synergies and differences

D Ghosh, MA Skinner and WA Laing

The Horticulture and Food Research Institute of New Zealand Ltd, Auckland, New Zealand

The success of the Human Genome Project and the spectacular development of broad genomics tools have catalyzed a new era in both medicine and nutrition. The terms pharmacogenomics and nutrigenomics are relatively new. Both have grown out of their genetic forbears as large-scale genomics technologies have been developed in the last decade. The aim of both disciplines is to individualize or personalize medicine and food and nutrition, and ultimately health, by tailoring the drug or the food to the individual genotype. This review article provides an overview of synergies and differences between these two potentially powerful science areas. Individual genetic variation is the common factor on which both pharmacogenomics and nutrigenomics are based. Each human is genetically (including epigenetics) unique and phenotypically distinct. One of the expectations of both technologies is that a wide range of gene variants and related single-nucleotide polymorphism will be identified as to their importance in health status, validated and incorporated into genotype based strategies for the optimization of health and the prevention of disease. Pharmacogenomics requires rigorous genomic testing that will be regulated and analyzed by professionals and acted on by medical practitioners. As further information is obtained on the importance of the interaction of food and the human genotype in disease prevention and health, pharmacogenomics can provide an opportunity driver for nutrigenomics. As we move from disease treatment to disease prevention, the two disciplines will become more closely aligned. European Journal of Clinical Nutrition (2007) 61, 567–574. doi:10.1038/sj.ejcn.1602590; published online 10 January 2007

Keywords: bioethics; drugs; food; nutrigenomics; pharmacogenomics

Introduction new, being first found in the medical literature in the late 1990s. The related term pharmacogenetics was used in the The success of the Human Genome Project and the powerful 1950s (Motulsky, 2002). This early use of the concept is not tools of molecular biology have ushered in a new era in surprising as it was observed from the start that there is a both medicine and nutrition. The pharmaceutical industry genetic basis to the differences in the way people respond to expects to leverage sequence data from the Human Genome drugs. A recent example of a pharmacogenetics medicine is Project to develop new drugs based on knowledge of drug the targeting of BiDil (Taylor et al., 2004), used for treatment targets, taking into account variation in the genetic makeup of heart failure, towards black African Americans, although of individuals. Likewise, the food industry now has an in this case the use of race as a surrogate for genetics is opportunity to position food and nutritional bioactives to debatable. Genetic variation within and between pheno- promote health and prevent disease based on the genetic typically identified populations is relevant to the provision makeup of the individual consumer. of genetic services and population screening. For example, The term pharmacogenomics (Motulsky, 2002; Steele, social group membership can be used clinically as a surrogate 2005; Eichelbaum et al., 2006; Kalow, 2006) is relatively for a high risk of a particular genetic variation in the case of people of Ashkenazi Jewish descent, a population with a high familial risk of breast cancer. However, in general, in such screening programmes, treatment on the basis of self- Correspondence: Dr D Ghosh, Health and Food, The Horticulture and Food Research Institute of New Zealand Ltd, PB 92169, Auckland 1003, New identification is unlikely to be an effective strategy (Bamshad Zealand. et al., 2004; Duster, 2005). In addition, this kind of race and E-mail: [email protected] ethnicity classification for medical treatment can lead to Guarantor: WA Laing. discrimination (Richards, 2006). Contributors: All contributors made an equal contribution. Received 12 April 2006; revised 30 October 2006; accepted 2 November Pharmacogenomics can be described as an approach to 2006; published online 10 January 2007 pharmacology that takes into account the genotype of the Pharmacogenomics and nutrigenomics D Ghosh et al 568 patient (Motulsky, 2002; Guo et al., 2005; Steele, 2005; individual’s genome is ever important in the response to any Eichelbaum et al., 2006; Kalow, 2006; Phillips and Van biologically active substance, such as drugs and nutrients, Bebber, 2006). It uses genomic techniques to study drug and that both fields will benefit greatly by understanding the functionality and discover new drug targets, genetically other areas. based high throughput screens to discover new drugs and large-scale genotyping to characterize and analyze the research and treatment populations. Thus, pharmaco- Background to individualized medicine and food genomics uses genotyping to screen populations in order to determine which drugs would work best in human Inter-individual variation in drug response among patients is subpopulations in curing or preventing a disease, as well as well known and poses a serious problem in medicine. There using various genomic technologies (e.g. gene expression) to is a need for rapid and simple methods to discriminate understand how the drug is interacting with the genotype. between patients on the basis of their response to a drug. The older term pharmacogenetics differs from pharmaco- There are no biomarkers at present that can predict which genomics only in the technologies used and in the breadth group of patients will or will not respond positively and of scope of the genes studied. which will experience adverse reactions for the same Nutrigenomics or nutritional genomics (Muller and medication and dose (Shastry, 2006). Physicians have to Kersten, 2003; Trayhurn, 2003; Kaput and Rodriguez, 2004; optimize a dosage regimen for an individual patient by a Labadarios and Meguid, 2004; Ordovas and Mooser, 2004; trial-and-error method. This kind of approach may cause van Ommen, 2004; Corthesy-Theulaz et al., 2005; Kauwell, adverse drug reactions in some patients. In more than two 2005; Mutch et al., 2005; Ferguson, 2006; Ordovas, 2006) million cases annually in the US, adverse reactions, includ- appears to be a 21st century term, with few references before ing 100 000 deaths (Lazarou et al., 1998; Daniel et al., 2006) the year 2000, although the term nutritional genomics was have been reported to occur. Similarly, according to a used in 1999 to refer to using plant genomics to improve the German study, about 6% of adverse drug reactions are nutritional quality of plants (Della Penna, 1999). Of course, attributed to new hospital admissions (Dormann et al., nutritional genetics, the study of the interaction of nutrition 2004). This inter-individual variability in drug response with human genotypes, is a much older term (Ordovas and could be because of multiple factors such as disease Mooser, 2004). Nutrigenomics is an approach to nutrition determinants, environmental factors and genetic based and human health that takes into account and studies the variability in drug target response (pharmacodynamic re- effect of genetic differences in human response to foods. It sponse) and metabolism or idiosyncratic responses. These also uses technologies from genomics and other related areas factors affect drug absorption, distribution, action, metabo- to study how food has an impact on gene expression, lism and excretion and have a strong genotype basis (Zheng biochemistry, metabolism and promotion of health (Elliott et al., 2004). An understanding of the variability in efficacy and Ong, 2002; Stover, 2006). The use of -omics research to and toxicity of the same doses of medications in the human analyze the influence of nutrients on human health is thus population, therefore, may provide safer and more efficient based on two observations: (1) the nutritional environment drug therapy. modifies the expression of genes, and (2) depending upon It is now clear that nutrients also effect metabolism by the genotype of an individual, the metabolism of nutrients exerting their effects at various genetic levels of biological may vary and ultimately result in a different health status complexity, such as gene transcription, RNA processing, (Corthesy-Theulaz et al., 2005). Thus, nutrigenomics treats mRNA stability and post-translational modifications (Young, food as a major environmental factor in the genetics– 2002) as well as their direct effects on cell metabolism. environment interaction. In parallel with the pharmaco- Nutritional requirements vary between different physiologi- genetics/-genomics split, the extension from nutritional cal groups, including age and gender and also among genetics to nutrigenomics thus lies in the use of genomics apparently similar individuals. Various genetic polymorph- technologies. Like pharmacogenomics, nutrigenomics also isms of importance to nutrition have been identified, for includes the study of the effect of dietary compounds on example folate metabolism, iron homeostasis, bone health, gene expression as a tool to understand how a nutrigenomic lipid metabolism, immune function, phenylketonuria and food acts on a particular genotype. gut inflammation (Baudin, 2000; Stewart and Ralston, 2000; Both pharmacogenomics and nutrigenomics have grown Ye and Kwiterovich, 2000; Mathew, 2001; Moyers and Bailey, out of their genetic forbears as large-scale genomics tech- 2001). Foods that are known to cause specific adverse nologies have been developed in the last decade (Blum et al., reactions in some genotypes are also well documented, and 2006). The aim of both disciplines is to individualize or include gluten in Celiac disease, which is partially linked to personalize medicine and food and nutrition, and ultimately HLA DQ2 and DQ8 (Evans et al., 2001), and lactose health, by tailoring the drug or the food to the individual intolerance, but the specific genetic defect that underlies genotype. This review article provides an overview of the intolerance is often unknown. As more such links synergies and differences between these two potentially between polymorphisms and disease conditions are char- powerful science areas. Our purpose is to show that an acterized, and genotype information on individuals is

European Journal of Clinical Nutrition Pharmacogenomics and nutrigenomics D Ghosh et al 569 accumulated, the scope for targeting dietary information allowed on the market. One could speculate that this may be and recommendations to specific subpopulations based on the case with the COX 2 inhibitors (Chen et al., 2004). More genotype will increase. life-threatening toxicities would probably justify the merit of such testing with new medications that have unique pharmacological properties. Pharmacogenomics and personalized drug therapy The integration of sophisticated genomic tests (e.g., high- throughput sequencing, single-nucleotide polymorphism The science of pharmacogenomics, which aims to define the (SNP) chips, DNA and protein microarrays, and bioinfor- gene-based determinants of drug efficacy, has evolved over matics) with extensive phenotypic characterization of uni- the past 50 years. At present, the best recognized and formly treated patients is essential if we are to define the completely developed examples of genetic polymorphisms inherited nature of most drug effects and the identification that alter drug response in humans are monogenic (single of drugs that are effective in genetically defined sub gene) traits that affect drug metabolizing enzymes, drug populations. A recent example is seen in the use of transporters and drug targets. One such example is that of microarrays to characterize tumors on the basis of their thiopurine S-methyltransferase (TPMT) and its influence on activated oncogenic pathway and to associate this with the thiopurine drugs mercaptopurine and azathiopurine in therapeutic agents (Bild et al., 2006). patients who inherit nonfunctional TPMT alleles (Evans et al., 2001). Other examples are the inter-individual differences in response to the antihypertensive drug debri- Nutrigenomics and personalized nutrition soquine and in the metabolism of the oxytocic drug sparteine, which led to family studies revealing that the Like pharmacogenomics, nutrigenomics also offers the pharmacodynamic effects of these drugs are inherited (Evans promise of personalizing nutrition to the genotype of the and Relling, 2004). consumer, based on a knowledge of variations in the genes of Pharmacogenomics may enhance drug discovery and nutrient metabolism and in downstream genes and proteins development by allowing genotype-based subpopulation- that interact with nutrients (nutrient targets) (Mutch et al., specific drug development. An example of major importance 2005). Although both fields of endeavor are in their infancy, will be the use of pharmacogenomics to identify genetic there are a number of new nutritional bioactives that offer polymorphisms that predispose patients to adverse drug glimpses of opportunities to come. Nutrigenomics could effects. Although these may occur in only a small subset of provide a framework for the development of genotype- the people treated with a new medication, they are dependent novel foods that will promote health and prevent sufficiently toxic to jeopardize further development of the and help manage chronic diseases. For example, phenylala- drug for all patients (Kirchheiner et al., 2005; Phillips and nine-restricted tyrosine-supplemented diets and galactose- Van Bebber, 2005). free diets are prescribed for the nutritional treatment of type The hope at the outset is that no such toxicities would 1 phenylketonuria and galactosemia (galactose-1-phosphate emerge for new agents under development. Should severe uridyltranferase deficiency), respectively, based on routine toxicity occur in a very small percentage of patients who genetic tests (Chen, 2001; Longo, 2001). In these cases, the could be prospectively identified based on genotype, then relevant genes encode proteins directly involved in the an otherwise efficacious new drug might be ‘saved’ from metabolism of the nutrients. It is important to note that not abandonment during development, or from withdrawal after all genes that are critical to clinical outcome are directly approval and widespread use. For example, patients at high involved in the pathogenesis of the disease or nutritional risk of abacavir hypersensitivity can be identified on the benefit. For example, polymorphisms in apolipoprotein E basis of their HLA-B genotypes (Martin et al., 2005), but it (APO E4) modify the clinical effectiveness of Cognex, which remains unclear whether prospective screening of HLA inhibits acetylcholine breakdown in Alzheimer’s patients genotypes is cost-effective with current technology. With (Poirier et al., 1995), and also the potential dietary benefits of the statins (Schmitz and Langmann, 2006), severe adverse vitamin E in Alzheimer’s disease (Morris et al., 2002). The side effects can occur in a minority of patients. Although study has shown that vitamin E from food may be associated relevant candidate genes for predicting adverse effects of with a reduced risk of Alzheimer’s disease, but this associa- lipid-lowering drugs, including the statins, have been tion was observed only among individuals without the APO identified, no reports have yet been made analyzing genetic E4 allele. variants in the genes related to clinical complications. Nutrient/diet–gene interactions may also explain why Presumably, this will be the added benefit of pharmaco- some individuals respond more favorably to dietary inter- genomics in the future, in which, as well as targeting ventions than others. For example, blood pressure is appropriate genotypes for drug effect, those with genotypes controlled in part by angiotensin, a vasoconstrictor. An associated with adverse effects can be identified and alter- SNP in the gene that encodes the precursor form of this native treatments selected. It may also mean that drugs polypeptide, angiotensinogen (ANG), results in a glycine to with adverse effects in only a proportion of patients may be arginine substitution (G6A) in the gene. The AA genotype for

European Journal of Clinical Nutrition Pharmacogenomics and nutrigenomics D Ghosh et al 570 the ANG GÀ6A polymorphism has been associated with larly, both nutritional status and diet can affect the action higher levels of circulating ANG and hypertension (Svetkey of drugs by altering their metabolism and function, and et al., 2001). Interestingly, the results of a sub-study of also importantly, various dietary components can have subjects who participated in the Dietary Approaches to Stop pharmacological activity under certain circumstances Hypertension (DASH) trial revealed that subjects with the AA (Schmidt and Dalhoff, 2002; Lila and Raskin, 2005). A well genotype were more responsive to the DASH diet than those publicized example of drug/nutrient interaction is grape- with the GG genotype (Svetkey et al., 2001). fruit, which can interact with several drugs such as Oxidative stress and genetic factors are potent determi- amiodarone, buspirone, carbamazepine, cyclosporine and nants of proinflammatory cytokine production (Grimble, nifedipine. There can also be dangerous interactions 2002). One method of reducing inflammatory stress would between monoamine oxidase inhibitors such as phenelcy- be to feed nutrients that either suppress proinflammatory promine, tranylcypromine and isocarboxazid with tyra- cytokine production (e.g. fish oil) (Endres et al., 1989) or act mine, a substance found in some foods and beverages as antioxidants (e.g. vitamin E) (Mol et al., 1997). At present (e.g. all aged and mature cheeses, fermented soy products it is not known whether antioxidants interact with gene and air-dried sausages) (Marcason, 2005). Adverse interac- variants associated with oxidative stress and inflammation in tion between blood-thinning warfarin and vitamin K is also the same way as anti-inflammatory nutrients, for example reported (Camilo et al., 1998). High doses of aluminum or n-3 polyunsaturated fatty acid. Proteomic studies have shown magnesium hydroxide antacids can cause phosphate deple- a link of iNOS and SOD with the NRAMP1 gene (Kovarova tion, leading to muscle weakness, anorexia and even et al., 2001). Thus, the high level of oxidant molecules that congestive heart failure. Thiazide and furosemide diuretics enhances pro-inflammatory cytokine production is owing to can cause sodium, potassium and magnesium depletion, the influence of variations in the NRAMP1 gene. Specific resulting in anorexia and muscle weakness (Knauf et al., ketogenic diets are already used for the treatment of patients 2006). Commonly used folate antagonists include metho- with intractable epilepsy. Balanced diets with essential fatty trexate, a cancer-chemotherapeutic agent; triamterene, a acids are paramount as the background diet for patients diuretic; trimethoprim, an antibacterial agent; phenytoin, with chronic inflammatory diseases, such as arthritis, an anticonvulsant; and sulphasalazine, an anti-inflamma- asthma, ulcerative colitis, lupus, etc. (Simopoulos, 2002). tory agent (Nozaki et al., 2004). Sulphasalazine and As our understanding of the human genome progresses, we phenytoin are competitive inhibitors of folate transport in can envisage genotype/food interactions being discovered in addition to being folate antagonists. Folate deficiency can other areas of health and disease such as athlete training, lead to weight loss and anorexia. Penicillamine induces zinc mood and depression and coping with the aging processes. depletion, which may cause a loss of taste acuity and Thus, the benefits associated with personalized nutrition will possibly decreased food intake (Floersheim et al., 1984). include those linked to our state of health and disease Alcohol abuse also commonly results in deficiencies of

prevention and amelioration of many of the ‘lifestyle’ thiamin, folate, vitamin B6, vitamin A and zinc (Bowden diseases. On the other hand, personalized drug prescription, et al., 1994). Nutrients are sometimes used in unusually by the very nature of the process, is mainly applicable, with a high doses for their pharmacologic effect. Niacin, for few exceptions, to the treatment of disease or at least the example, is used pharmacologically to reduce blood prevention of further deterioration (e.g. osteoporosis). cholesterol levels (Morgan et al., 2004). Retinoid derivatives of vitamin A have been used successfully to treat severe acne and other conditions (Thielitz et al., 2006). None of Relationship between food and drugs these examples is known to include a genotype component, but individuals who are born with deficiencies in the genes Humans evolved in the presence of a wide variety of different need to produce key functional enzymes, and may require foods, and human metabolism is well adapted to the great amounts of certain nutrients greatly in excess of those variety of foods derived from compounds found in nature. required by most people. Such inborn metabolic errors have Man-made pharmaceuticals are a modern invention in been identified for enzymes necessary for absorption, terms of human evolutionary history. However, one aspect metabolism or storage of nearly all of the vitamins. Thus, of human physiology is suited to the metabolism of both as the relationships between foods and drugs become more food chemicals and pharmaceuticals, namely the detoxifica- fully understood, so will the linkages between pharmaco- tion mechanisms predominantly carried out by the liver. genomics and nutrigenomics. These detoxification mechanisms are mainly aimed at plant derived xenobiotics (Liska, 1998), but many drugs are also metabolized by the same systems. This may reflect the Shifting the paradigm from treatment to number of drugs that are derived from natural compounds prevention in nutrition and pharmacology (Chin et al., 2006). Prescription drugs may interfere with nutrient absorp- The concept of the link between nutrition and health is best tion, digestion, metabolism, utilization or excretion. Simi- appreciated if we move from the biomedical approach of

European Journal of Clinical Nutrition Pharmacogenomics and nutrigenomics D Ghosh et al 571 treating disease, and awareness of maintaining health. An Genomics tools used by both approaches awareness that multiple minor changes in metabolism and its biochemical regulation contribute to the onset of chronic In order to implement personalized nutrition and pharma- nutrition-related disorders emphasizes the importance of cology approaches, it is critical that the individuals be nutrition in health maintenance. Examples where nutrition genotyped. The current data indicate that approximately can promote health, and not merely treat disease, include 30 000 genes are in the human genome (Guttmacher and obesity, diabetes type 2, cardio vascular disease (CVD), Collins, 2002). Only about half of 30 000 genes, have osteoporosis and chronic inflammatory syndromes, all of recognizable DNA sequence patterns that suggest possible which show genetic based differences in susceptibility functions. Mutations known to cause disease have been (Child, 1988). identified in approximately 1000 genes although many more It is absolutely imperative that human nutritional inter- SNPs have been detected. However, it is likely that nearly all vention studies take advantage of nutrigenomics develop- human genes are capable of causing disease if they are ments by moving on from focussing on disease or damage, altered substantially. Owing to the dynamic nature of DNA and apply these genetic based nutrigenomics tools toward expression and also through the mechanism of alternative health promotion and the prevention of disease. Knowledge splicing, more than 100 000 proteins can be derived from of the genotype will allow selection of health promoting these B30 000 genes. In addition, a number of ‘epigenetic’ food and extracts for individuals. Fortified foods, functional phenomena, such as methylation and histone modification, foods, and nutraceuticals are bioactive natural compounds can alter the effect of a gene (Feinberg, 2001; Verma et al., that have health promoting or disease preventing properties. 2003). Maternal nutritional status, in particular, can alter For example, the natural angiotensin converting enzyme the epigenetic state of the fetal genome and resultant (ACE) inhibitor activity in milk peptides may be used in subtle phenotypic differences observed in identical twins certain genotypes to control hypertension (FitzGerald et al., (Dennis, 2003). 2004) instead of using synthetically produced ACE inhibitor The commonest type of genetic variability is the SNP, drugs, whose activity depends on particular gene poly- a single base substitution within the DNA sequence morphisms. Obesity is one area where the application of (Chakravarty, 2001), although other structural genetic both genomic and nutritional approaches has already been changes also occur (Feuk et al., 2006). These SNPs occur highly successful (Defeat Diabetes Association, 2003). roughly once every 1000–2000 nucleotides in the human Another example is colon cancer, where an estimated 60% genome. Research indicates that, just as SNPs are associated of the cancers are related to nutrition (Stierum et al., 2001). It with responsiveness to pharmaceuticals, they may also may be more effective in the long run to investigate the influence the responsiveness of individuals to changes maintenance of optimal gut health instead of the applica- in nutrient intake. For example SNPs may be responsible tion of nutrition against the recurrences of adenomas for variations in the lipeamic response to dietary lipids after surgical intervention. Likewise, prevention of chronic (Minihane et al., 2000), influence the interrelationship metabolic stress through optimal nutrition may be more between plasma vitamin B12, folate and homocysteine effective than dietary therapy of atherosclerosis with anti- (Andreassi et al., 2003) and modulate the ability of fish oil oxidant vitamins. to reduce production of the proinflammatory cytokine Pharmaceuticals are available that can be viewed as aiding tumor necrosis factor (TNF) (Grimble et al., 2002). Recent in disease prevention/health maintenance. For example, the development of extensive genetic polymorphism data- 3-hydroxymethlglutaryl coenzyme A reductase inhibitors or bases and high throughput genetic screening now make statins are used to prevent CVD by lowering cholesterol meaningful study of inter-individual variation not only (Shepherd et al., 1995). Similarly, the consumption of plant possible but also critical for the future of nutrition and sterol functional foods as part of the diet has been reported clinical research. SNPs may be screened using a range of to have similar overall effects in cholesterol reduction technologies. However, microarray chips, with the ability to (Goldberg et al., 2006). There is evidence that subsets of screen for huge numbers of SNPs simultaneously, have the individuals respond better than others to the cholesterol potential to revolutionize population and individual geno- lowering effects of both statin drugs and plant sterol typing along with the development of hapmaps (Interna- functional foods, and this has been linked to particular tional Hapmap consortium) (Gabriel et al., 2002; Lin et al., genotypes (Schmitz and Langmann, 2006). 2005), which allow efficient mapping of associated SNPs Unlike pharmacological interventions in which drug not in linkage disequilibrium. access, is tightly controlled, nutritional intervention Other technologies are being developed that will also be of with functional foods is generally unrestricted in terms major significance for both pharmacogenomics and nutrige- of access and professional advice may not be involved. nomics. These allow the simultaneous determination of the However, in both approaches, individuals have total control expression of many thousands of genes at the mRNA over whether to take the food/drug or not, and such (transcriptomics) (Gail, 2005) and protein (proteomics) decisions in the future will be based on knowledge of the (Nebert and Vesell, 2004) level, as well as the determination genotype. of the majority of the cellular metabolites (metabolomics)

European Journal of Clinical Nutrition Pharmacogenomics and nutrigenomics D Ghosh et al 572 (German et al., 2003). Not only is genotyping of importance From the negative viewpoint, the new sciences of in determining variation in response, but other ‘phenotypic’ ‘nutrigenomics’ and the idea of ‘personalised nutrition’ have approaches are critical in detecting differences in genetically been challenged recently (e.g. GeneWatch, 2006). The based responses to drugs and food. challenge advocated that, in most cases, genetic differences One thing is clear; pharmacogenomics will be the driver appear to make only small and subtle differences to a for application of genomics technologies to pharmacology, person’s risk of diet-related disease and hence very little and nutrigenomics will piggyback on the advances made by difference to the foods they should eat. We do not believe medicine. This is because of the expense and technical this is borne out by the information and examples presented demands of genotyping and related disciplines. Pharmaceu- in this review. tical companies will be driven to genotype populations in order to minimize risk and avoid inappropriate or ineffective drug based therapies, and medical personnel will subse- Conclusions and expectations of the future quently genotype their patients in order to take advantage of knowledge of gene/drug interactions. Few will be genotyped Individual genetic variation is the common factor on solely to maintain or improve their health through good which both pharmacogenomics and nutrigenomics are nutrition. However, because the individual genotype infor- based. One of the expectations of both technologies is mation will be available through pharmacogenomics, this that a wide range of modified genes and related SNPs will knowledge could be applied in developing individualized be identified, validated and incorporated into strategies for nutrition or nutrigenomics. the optimization of health and the prevention of disease through drugs and nutrition. This will introduce a massive set of new data and increase the biological complexity of Ethical, legal and social implications of developing drugs and food-based dietary guidelines that nutrigenomics and pharmacogenomics will shape the future of medicine and health. Pharmaco- genomics requires rigorous genomic testing that will be The ethical, legal and social implications arising from regulated and analyzed by professionals and acted on advances in genomic research have been well recognized by medical practitioners. As recognition grows of the and have strong implications for both pharmacogenomics importance of food in disease prevention and health, and nutrigenomics (Kaput et al., 2005; Bergmann et al., pharmacogenomics can provide an opportunity driver for 2006). Major issues are privacy and fairness in the use and nutrigenomics. In addition, as our understanding of the interpretation of genetic information, clinical integration of action of a wide range of animal, plant and microbial new genetic technologies, issues surrounding genetic re- nutrients develops, we may find that the continuum from search and public and professional education (http:// pharmaceuticals to nutraceuticals through to food-based www.genome.gov/10001754#what). Disclosure of genetic bioactive compounds (often secondary plant products information to others, intentional or otherwise, may be which may be pharmaceuticals in their own right) will followed by social stigmatization and discriminatory prac- bring the disciplines of pharmacogenomics and nutrige- tices by employers, insurers or schools. Patenting and nomics closer together. Likewise, as we move from disease licensing issues related to human genes add to the complex- treatment to disease prevention, the two disciplines may be ity and conceivably the expense of genetic testing (Guttma- more closely aligned. cher and Collins, 2002). Disparity in access to genomic In the light of the new era of pharmacogenomics and medicine in economically poor regions of the world must nutrigenomics, policymakers and regulators can expect to also be addressed. Genetic literacy among health care wrestle with the complex issue of subpopulations as the providers is also an important issue. Accurate original simplicity of ‘one-size-fits-all’ erodes in the full light of new research, interpretation of patient genotype and application knowledge. Similarly, the drug and food industries may of the integrated information is critical to avoiding errors expect to encounter challenging new business demands as and could have far-reaching legal implications in medical mass-marketing moves toward mass-customization. Finally, malpractice. Again, nutrigenomics is likely to follow the lead as society moves to adopt these practices, it may see a of pharmacogenomics in this regard, with emphasis on bioethical debate ensue over the impact of a growing patient–doctor confidentiality, and the use of professional ‘genomic divide’ in personal and global health equity advice in medical as well as nutritional decision-making (Editorial, 2001; Gillies, 2003; Guttmacher and Collins, by the individual. Although companies may offer services in 2003; William et al., 2004). genotyping, it is likely that implementation of genotypic The future for both disciplines looks bright, and as new advice will be on an individual basis for both disciplines. It is knowledge is obtained, the quote from Jean Anthelme also likely that both disciplines will have an impact on drug Brillat-Savarin (1755–1826) ‘tell me what you eat, and I will and food labelling, although it may be some time before tell you what you are’ will need to become ‘tell me what you a food is labelled with a statement indicating the genotypes eat, what drugs you take, and your genotype, and I will tell it is aimed at. you what you are’!

European Journal of Clinical Nutrition Pharmacogenomics and nutrigenomics D Ghosh et al 573 Acknowledgements Dormann H, Neubert A, Criegee-Rieck M, Egger T, Radespiel-Troger M, Azaz-Livshits T et al. (2004). Readmissions and adverse drug We are members of the Nutrigenomics NZ (NuNZ) collabora- reactions in internal medicine: the economic impact. J Int Med 255, 653–663. tion (www.nutrigenomics.org.nz), but this review represents Duster T (2005). Race and reification in science. Science 307, our own views and not the views of NuNZ. The writing of 1050–1051. this review was supported in part by funding provided by the Editorial (2001). Defining a new bioethic. Nat Genet 28, 297–298. Foundation for Research Science and Technology (FRST) Eichelbaum M, Ingelman-Sundberg M, Evans WE (2006). Pharmaco- genomics and individualised drug therapy. Annu Rev Med 57, (CO2X0403). We thank David Stevenson and Denis Lauren 119–137. for careful reviews of this manuscript. Elliott R, Ong TJ (2002). Nutritional genomics. BMJ 324, 1438–1442. Endres S, Ghorbani R, Kelley VE, Georgilis K, Lonnemann G, van der Meer JW et al. (1989). The effect of dietary supplementation with References n-3 polyunsaturated fatty acids on the synthesis of interleukin-1 and tumor necrosis factor by mononuclear cells. New Engl J Med 320, 265–271. Andreassi MG, Botto N, Cocci F, Battaglia D, Antonioli E, Masettiet S Evans WE, Hon YY, Bomgaars L, Coutre S, Holdsworth M, Janco R et al. (2003). Methylenetetrahydrofolate reductase gene C677T et al. (2001). Preponderance of thiopurine-S-methyltransferase polymorphism, homocysteine, vitamin B12, and DNA damage in deficiency and heterozygosity among patients intolerent to coronary artery disease. Hum Genet 112, 171–177. mercaptopurine or azathioprine. J Clin Oncol 19, 2293–2301. Bamshad M, Wooding S, Salisbury BA, Claibourne SJ (2004). Evans WE, Relling MV (2004). Moving towards individualized Deconstructing the relationship between genetics and race. Nat medicine with pharmacogenomics. Nature 429, 464–468. Rev Genet 5, 598–609. Feinberg AP (2001). Cancer epigenetics takes center stage. Proc Natl Baudin B (2000). Angiotensin I-converting enzyme gene polymor- Acad Sci USA 98, 392–394. phism and drug response. Clin Chem Lab Med 38, 853–856. Ferguson LR (2006). Nutrigenomics: integrating genomic approaches Bergmann MM, Bodzioch M, Bonet ML, Defoort C, Lietz G, Mathers into nutrition research. Mol Diagn Ther 10, 101–108. JC (2006). Bioethics in human nutrigenomics research: European Feuk L, Carson AR, Scherer SW (2006). Structural variation in the Nutrigenomics Organisation workshop report. Br J Nutr 95, human genome. Nat Rev Genet 7, 85–97. 1024–1027. FitzGerald RJ, Murray BA, Walsh DJ (2004). Hypotensive peptides Bild AH, Yao G, Chang JT, Wang Q, Potti A, Chasse D et al. (2006). from milk proteins. JNutr134, 980S–988S. Oncogenic pathway signatures in human cancers as a guide to Floersheim GL, Bianchi L, Probst A, Chiodetti N, Honegger CG targeted therapies. Nature 439, 353–357. (1984). Influence of zinc, D-penicillamine and oxygen on poison- Blum K, Chen TJH, Blum SH, Downs BW, Braverman ER, Mengucci JE et al. (2006). Nutrigenomics and pharmacogenomics: a scientific ing with Amanita phalloides. Zinc accelerates liver regeneration wonderland. Soc Sci Inform 45, 35–52. and prevents the depletion of brain noradrenaline caused by the Bowden S, Bardenhagen F, Ambrose M, Whelan G (1994). Alcohol, mushroom. Agents Actions 14, 24–30. thiamin deficiency, and neuropsychological disorders. Alcohol Gabriel SB, Schaffner SF, Nguyen H, Moore JM (2002). The structure Alcohol Suppl 2, 267–272. of haplotype blocks in the human genome. Science 296, Camilo ME, Paiva SA, O’Brien ME, Booth SL, Davidson KW, Sokoll LJ 2225–2229. et al. (1998). The interaction between vitamin K nutriture and Gail PAK (2005). Emerging concepts in nutrigenomics: a preview of warfarin administration in patients with bacterial overgrowth due what is to come. Nutr Clin Practice 20, 75–87. to atrophic gastritis. J Nutr Health Aging 2, 73–78. GeneWatch (UK) (2006). Your diet tailored to your genes: preventing Chakravarty A (2001). Single nucleotide polymorphisms: to a future diseases or misleading marketing?. http://www.genewatch.org/ of genetic medicine. Nature 409, 822–823. uploads/f03c6d66a9b354535738483c1c3d49e4/Nutrigenomics.pdf. Chen JC, Huang KC, Wingerd B, Wu WT, Lin WW (2004). HMG-CoA German JB, Roberts MA, Watkins SM (2003). Genomics and reductase inhibitors induce COX-2 gene expression in murine metabolomics as markers for the interaction of diet and health: macrophages: role of MAPK cascades and promoter elements for lessons from lipids. J Nutr 133, 2078S–2083S. CREB and C/EBPaˆ. Exp Cell Res 301, 305–319. Gillies PJ (2003). Nutrigenomics: the rubicon of molecular nutrition. Chen YT (2001). Glycogen storage diseases and other inherited J Am Diet Assoc 103, S50–S55. disorders of carbohydrate metabolism. In: Braunwald E, Goldberg AC, Ostlund Jr RE, Bateman JH, Schimmoeller L, McPher- Fausi AS, Kasper DL, Hauser SL, Longo DL, Jameson JL (eds). son TB, Spilburg CA (2006). Effect of plant stanol tablets on low- Harrison’s Principles of Internal Medicines. McGraw-Hill: New York. density lipoprotein cholesterol lowering in patients on statin pp 2288–2289. drugs. Amer J Cardiol 97, 376–379. Child B (1988). Genetic variation and nutrition. Am J Clin Nutr 48, Grimble RF (2002). Inflammatory status, insulin resistance. Curr Opin 1500–1504. Clin Nutr Metab Care 5, 551–559. Chin YW, Balunas MJ, Chai HB, Kinghorn AD (2006). Drug discovery Grimble RF, Howell WM, O’Reilly G, Turner SJ, Markovic O, Hirrell S from natural sources. AAPS J 8, E239–E253. et al. (2002). The ability of fish oil to suppress tumor necrosis Corthesy-Theulaz I, den Dunnen JT, Ferre P, Geurts JM, Muller M, factor-alpha production by peripheral blood mononuclear cells van Belzen N et al. (2005). Nutrigenomics: the impact of biomics in healthy men is associated with polymorphisms in genes technology on nutrition research. Ann Nutr Metab 49, 355–365. which influence TNF-alpha production. Am J Clin Nutr 76, Daniel SB, Daniel AP, Kelly NW, Aaron BM, Thomas JS, Joseph LA 454–459. (2006). National surveillance of emergency department visits for Guo Y, Shafer S, Weller P, Usuka J, Peltz G (2005). Pharmacogenomics outpatient adverse drug events. JAMA 296, 1858–1866. and drug development. Pharmacogenomics 6, 857–864. Defeat Diabetes Association (2003). Food portions may explain the Guttmacher AE, Collins FS (2002). Genomic medicine: a primer. ‘French Paradox’ of rich foods and a thin population. (posted 28/ N Engl J Med 347, 1512–1520. 08/03),http://www.defeatdiabetes.org/Articles/diet030828.htm. Guttmacher AE, Collins FS (2003). Welcome to the genomic era. Della Penna D (1999). Nutritional genomics: manipulating plant N Engl J Med 349, 996–998. micronutrients to improve human health. Science 285, 375–379. Kalow W (2006). Pharmacogenetics and pharmacogenomics: origin, Dennis C (2003). Epigenetics and disease: altered states. Nature 421, status, and the hope for personalized medicine. Pharmacogenomics 686–688. J 6, 162–165.

European Journal of Clinical Nutrition Pharmacogenomics and nutrigenomics D Ghosh et al 574 Kaput J, Ordovas JM, Ferguson L, Ommen Bv, Rodriguez RL, Allen L Nozaki Y, Kusuhara H, Endou H, Sugiyama Y (2004). Quantitative et al. (2005). The case for strategic international alliances to evaluation of the drug-drug interactions between methotrexate harness nutritional genomics for public and personal health. Br J and nonsteroidal anti-inflammatory drugs in the renal uptake Nutr 94, 1–12. process based on the contribution of organic anion transporters Kaput, Rodriguez RL (2004). Nutritional genomics: the next frontier and reduced folate carrier. J Pharmacol Exp Ther 309, 226–234. in the postgenomic era. Physiol Genomics 16, 166–177. Ordovas JM (2006). Nutrigenetics, plasma lipids, and cardiovascular Kauwell GP (2005). Emerging concepts in nutrigenomics: a preview risk. J Am Dietetic Assoc 106, 1074–1081. of what is to come. Nutr Clin Pract 20, 75–87. Ordovas JM, Mooser V (2004). Nutrigenomics and nutrigenetics. Kirchheiner J, Fuhr U, Brockmoller J (2005). Pharmacogenetics-based Curr Opin Lipidol 15, 101–108. therapeutic recommendations-ready for clinical practice? Nat Rev Phillips KA, Van Bebber SL (2005). Measuring the value of 4, 639–647. pharmacogenomics. Nat Rev 4, 500–510. Knauf H, Bailey MA, Hasenfuss G, Mutschler E (2006). The influence Phillips KA, Van Bebber SL (2006). Regulatory perspectives on of cardiovascular and antiinflammatory drugs on thiazide-induced pharmacogenomics: a review of the literature on key issues faced hemodynamic and saluretic effects. Eur J Clin Pharmacol 10.1007/ by the United States Food and Drug Administration. Med Care Res s00228-006-0190-3 [epub ahead of print]. Rev 63, 301–326. Kovarova H, Necasova R, Porkertova S, Radzioch D, Macela A (2001). Poirier J, Delisle MC, Quirion R, Aubert I, Farlow M, Lahiri D et al. Natural resistance to intracellular pathogens: modulation of (1995). Apolipoprotein E4 allele as a predictor of cholinergic macrophage signal transduction related to the expression of the deficits and treatment outcome in Alzheimer’s disease. Proc Natl Bcg locus. Proteomics 1, 587–596. Acad Sci USA 92, 12260–12264. Labadarios D, Meguid MM (2004). Nutrigenomics: unraveling man’s Richards MPM (2006). Genetic diversity and science communica- constitution in relation to food. Nutrition 20, 2–3. tion-some issues of ‘translation’ and language. Clin Genet 70, Lazarou J, Pomeranz BH, Corey PN (1998). Incidence of adverse drug 451–454. reactions in hospitalized patients: a meta-analysis of prospective Schmidt LE, Dalhoff K (2002). Food–drug interactions. Drugs 62, studies. JAMA 279, 1200–1205. 1481–1502. Lila MA, Raskin I (2005). Health-related interactions of phytochem- Schmitz G, Langmann T (2006). Pharmacogenomics of cholesterol- icals. J Food Sci 70, R20–R27. lowering therapy. Vasc Pharmacol 44, 75–89. Lin M, Aquilante C, Johnson JA, Wu R (2005). Sequencing drug Shastry BS (2006). Pharmacogenetics and the concept of individua- response with HapMap. Pharmacogenomics J 5, 149–156. lized medicine. Pharmacogenomics J 6, 16–21. Liska DJ (1998). The detoxification enzyme systems. Altern Med Rev 3, Shepherd J, Cobbe SM, Ford I, Isles CG, Lorimer AR, MacFarlane PW 187–198. et al. (1995). Prevention of coronary heart disease with oravastatin Longo N (2001). Inherited disorders of amino acid metabolism and in men hypercholesterolemia. N Engl J Med 333, 1301–1307. storage. In: Braunwald E, Fausi AS, Kasper DL, Hauser SL, Longo Simopoulos AP (2002). Genetic variation and dietary response: DL, Jameson JL (eds). Harrison’s Principles of Internal Medicines. nutrigenetics/nutrigenomics. Asia Pacific J Clin Nutr 11, S117–S128. McGraw-Hill: New York. pp 2301–2306. Steele FR (2005). Big Pharma’s Commedia. Cell 123, 971–973. Marcason W (2005). What is the bottom line for dietary guidelines Stewart TL, Ralston SH (2000). Review of genetic factors in the when taking monoamine oxidase inhibitors? J Am Diet Assoc 105, pathogenesis of osteoporosis. J Endocrinol 166, 235–245. 163. Stierum R, Burgemeister R, Helvoort AV, Peijnenburg A, Schutze K, Martin AM, Nolan D, Mallal S (2005). HLA-B*5701 typing by Seidelin M et al. (2001). Functional foods against colorectal cancer: sequence-specific amplification: validation and comparison with an example project integrating functional genomics, nutrition sequence-based typing. Tissue Antigens 65, 571–574. and health. Nutr Metab Cardiovasc Dis 4, 1–5. Mathew C (2001). Postgenomic technologies: hunting the genes for Stover PL (2006). Influence of human genetic variation on nutri- common disorders. BMJ 322, 1031–1034. tional requirements. Am J Clin Nutr 83, 436S–442S. Minihane AM, Khan S, Leigh-Firbank EC, Talmud P, Wright JW, Svetkey LP, Moore TJ, Simons-Morton DG, Appel LJ, Bray GA, Sacks Murphy MC et al. (2000). ApoE polymorphism and fish oil FM et al. (2001). Angiotensinogen genotype and blood pressure supplementation in subjects with an atherogenic lipoprotein response in the Dietary Approaches to Stop Hypertension (DASH) phenotype. Arterioscler Thromb Vasc Biol 20, 1990–1997. study. J Hypertens 19, 1949–1956. Mol M, de Rijke Y, Demacker P, Stalenhoef A (1997). Plasma levels of Taylor AL, Ziesche S, Yancy C, Carson P, D’Agostino Jr R, Ferdinand K lipid and cholesterol oxidation products and cytokines in diabetes et al. (2004). Combination of isosorbide dinitrate and hydralazine mellitus and cigarette smoking: effects of vitamin E treatment,. in blacks with heart failure. N Eng J Med 351, 2049–2057. Journal Title: Atherosclerosis 129, 169–176. Thielitz A, Krautheim A, Gollnick H (2006). Update in retinoid Morgan JM, Carey CM, Lincoff A, Capuzzi DM (2004). The effects therapy of acne. Dermatol Ther 19, 272–279. of niacin on lipoprotein subclass distribution. Prev Cardiol 7, Trayhurn P (2003). Nutritional genomics – nutrigenomics. Br J Nutr 182–187. 89, 1–2. Morris MC, Evans DA, Bienias JL, Tangney CC, Bennett DA, Aggarwal van Ommen B (2004). Nutrigenomics: exploiting systems biology in N et al. (2002). Dietary intake of antioxidant nutrients and the risk the nutrition and health arena. Nutrition 20, 4–8. of incident Alzheimer’s disease in a biracial community study. Verma M, Dunn BK, Ross S, Jain P, Wang W, Hayes R et al. (2003). JAMA 287, 3230–3237. Early detection and risk assessment: proceedings and recommen- Motulsky AG (2002). From pharmacogenetics and ecogenetics to dations from the workshop on epigenetics in cancer prevention. pharmacogenomics. Med Secoli 14, 683–705. Ann NY Acad Sci 983, 298–319. Moyers S, Bailey LB (2001). Fetal malformations and folate William E, Evans E, Relling MV (2004). Moving towards individua- metabolism: review of recent evidence. Nutr Rev 59, 215–235. lized medicine with pharmacogenomics. Nature 429, 464–468. Muller M, Kersten S (2003). Nutrigenomics: goals and strategies. Ye SQ, Kwiterovich PO (2000). Influence of genetic polymorphisms Nat Rev Genet 4, 315–322. on responsiveness to dietary fat and cholesterol. Am J Clin Nutr 72, Mutch DM, Wahli W, Williamson G (2005). Nutrigenomics and 12755–12845. nutrigenetics: the emerging faces of nutrition. FASEB J 19, Young VR (2002). Human nutrient requirements: the challenge of 1602–1616. the post-genome era. J Nutr 132, 621–629. Nebert DW, Vesell ES (2004). Advances in pharmacogenomics and Zheng CJ, Sun LZ, Han LY, Ji ZL, Chen X, Chen YZ (2004). Drug individualized drug therapy: exciting challenges that lie ahead. ADME associated protein data base as a resource for facilitating Eur J Pharmacol 500, 267–280. pharmacogenomics research. Drug Dev Res 62, 134–142.

European Journal of Clinical Nutrition