Nutrigenomics

4.57

Nutrigenomics: A Possible Road to Personalized Nutrition

LE Cahill and A El-Sohemy, University of Toronto, Toronto, ON, Canada © 2011 Elsevier B.V. All rights reserved.

4.57.1 4.57.2 4.57.3 4.57.4 4.57.5 4.57.6 4.57.6.1 4.57.6.2 4.57.6.3 4.57.7 References

Introduction Alcohol Caffeine Dietary Fat Fruits and Vegetables Micronutrients Choline Vitamin C Folate Conclusions

Glossary allele One of the variant forms of a gene at a particular location on a chromosome. human genome The complete genetic makeup of a human. gene The functional and physical unit of heredity passed from parent to offspring. Genes are pieces of DNA that contain the information for making a specific protein. genotype The genetic identity of an individual for a genetic site, determined from the combination of maternal and paternal chromosomes, which does not necessarily show as outward characteristics. haplotype A group of alleles that are inherited together. linkage disequilibrium A situation where alleles at specific sites of DNA are associated more or less frequently than would be expected randomly.

703 704 705 705 707 708 708 709 709 709 710

nutrigenomics Both the examination of how nutrients affect genes (i.e., influence gene expression and function) and how genes affect diet (i.e., what an individual eats and how an individual responds to nutrients), with the latter sometimes being referred to as ‘nutrigenetics’. Nutrigenomics can include the full spectrum of research strategies from basic cellular and molecular biology to clinical trials, epidemiology, and population health. phenotype An observable trait in an individual such as hair color, high blood sugar concentrations, or the presence of a disease. Individuals with the same genotype may have different phenotypes in different environments. single nucleotide polymorphism (SNP) A genetic variation caused by a change in a single DNA nucleotide that occurs in at least 1% of the population.

4.57.1 Introduction Variability between individuals in response to dietary intervention is a well-known phenomenon in nutrition research and practice [1]. For example, the effect of dietary changes on phenotypes such as blood cholesterol, body weight, and blood pressure can differ significantly between individuals [2]. There are many factors influencing response to diet, such as age, sex, physical activity, smoking, and genetics. The goal of personalized nutrition is to identify individuals who benefit from a particular nutrition intervention (responders), and identify alternatives for those who do not (non-responders). Individuals will no longer be subjected to unnecessary diets they find unpleasant and ineffective when there may be an alternate dietary approach that is more effective. Personalized nutrition could be useful in both the prevention and treatment of chronic diseases by tailoring dietary advice to an individual’s unique genetic profile. The daily ingestion, absorption, digestion, transport, metabolism, and excretion of nutrients and food bioactives by humans involve many proteins such as enzymes, receptors, transporters, ion channels, and hormones. Variations in genes encoding proteins that affect any of these processes can alter both the amount of the protein produced as well as how well that protein functions. If a genetic variation leads to altered production or function of these proteins, then nutritional status might be affected. The study of the relationship between genes and diet is called nutrigenomics (or nutritional genomics), which is an umbrella term for two complimentary approaches: how nutrients affect gene function and how genetic variation affects nutrient response. The latter is sometimes referred to as nutrigenetics [3], and includes the study of how genetic variations affect food intake and eating behaviors [4, 5]. Although the term ‘nutrigenomics’ is relatively new, the concept has been around for some time. Perhaps the most familiar example is lactose intolerance, which is a condition resulting from an inadequate production of lactase in the small intestine due to

703

704

Intergration of Biotechnologies

genetic variation in the lactase gene [6]. Individuals with lactose intolerance are unable to efficiently break down the primary milk sugar (lactose) from dairy products. Consequently, the dietary recommendation is to limit lactose-containing foods or to use lactase supplements or lactose-free dairy products to prevent gastrointestinal discomfort [7]. Phenylketonuria (PKU) is an inborn error of metabolism, which represents another classic example of nutrigenomics. PKU can result from a genetic variation in phenylalanine hydroxylase (the enzyme needed to convert phenylalanine to tyrosine), which leads to a decrease in phenylalanine hydroxylase activity [8]. Individuals with PKU can develop neurological damage (severe mental retardation and seizures) from excess pheny­ lalanine [9] unless they follow the recommended low-phenylalanine diet [10]. Lactose intolerance and PKU are examples that involve a single genetic defect along with a single dietary exposure. A major challenge of nutrigenomics, however, is to identify gene–nutrient interactions that affect complex polygenic disorders such as obesity, diabetes, cancer, and cardiovascular disease that take several years – or even decades – to develop and have multi­ factorial etiologies. As such, it is often difficult to determine the role of specific dietary factors and gene variants in the development of these diseases. The concept of diet–gene interactions involves a genetic variant modulating the effect of a dietary factor on a specific phenotype or health outcome measure such as serum lipid concentrations, high blood glucose, or obesity. Conversely, diet–gene interactions can refer to the dietary modification of the effect of a genetic variant on the phenotype or health outcome measure [1]. A major goal of nutrigenomics is the prevention of the onset and progression of chronic disease. Research strategies currently contribute to this goal by building the body of evidence linking nutrients to metabolic pathways that affect disease outcomes. The incorporation of genetics into nutritional epidemiologic studies aims to improve their consistency. Current research could lead to the development of personalized nutrition guidelines for individuals and specific subpopulations, which could decrease the risk of chronic diseases. Using examples of specific dietary factors such as alcohol, caffeine, fat, fruits, vegetables, and micronutrients, this article will highlight the importance of incorporating genetics into the field of nutrition and also address some of the questions and methods associated with the development of personalized nutrition.

4.57.2 Alcohol Incorporating data on genetic variability into a nutrition study can clarify whether or not a specific dietary compound is linked to a particular health outcome. This diet–disease connection is made by the observation of whether the association between the disease and the dietary factor is influenced by functional variants in genes involved in the metabolism of the dietary factor. For example, moderate alcohol consumption has been associated with a lower risk of heart disease. The primary mechanism proposed for this association is the higher levels of high-density lipoprotein (HDL) cholesterol found among moderate drinkers. However, it is possible that the protective effect of moderate alcohol consumption is due to some confounding lifestyle factor associated with moderate alcohol intake that differs from abstainers or heavy drinkers. The enzyme alcohol dehydrogenase 1C (ADH1C), also known as alcohol dehydrogenase type 3 (ADH3), oxidizes alcohol to acetaldehyde. It has two polymorphic forms with distinct kinetic properties. The ADH1C*1 allele produces γ1 and the ADH1C*2 allele produces γ2. The rate of alcohol metabolism (ethanol oxidation) in γ1γ1 individuals is more than twice as high as in γ2γ2 individuals, with heterozygous individuals metabolizing alcohol at an intermediate rate. Therefore, if there is a causal protective effect of moderate alcohol on risk of heart disease, this effect would be expected to be stronger in individuals with the slow ADH1C genotype compared to those with fast metabolism. An early gene–diet interaction study examined data from the Physicians’ Health Study and the Nurses’ Health Study to determine whether the effect of moderate alcohol consumption on HDL levels and the risk of myocardial infarction (MI) vary according to ADH3 genotype [11]. The study reported evidence of a dose–response decrease in heart disease as alcohol metabolism slows according to genotype, for a marked 86% reduction in heart disease risk in subjects with the slow-alcohol-metabolizing genotype who consumed at least 1 alcoholic drink per day [11]. The polymorphism was also associated with higher levels of HDL [11]. The finding of an effect of the functional ADH1C polymorphism on the relation between alcohol consumption and the risk of heart disease suggests that the protective effect is due to the alcohol and not some associated lifestyle factor since those with the slow-alcohol­ metabolizing genotype who did not consume alcohol were not at a lower risk. Subsequent studies have replicated the interaction with risk of heart disease [12–14]. For example, the British prospective Second Northwick Park Heart Study reported that subjects with moderate alcohol consumption (equivalent to 8–24 g of ethanol per week) and the γ2γ2 genotype exhibited a significant 78% reduction in coronary heart disease risk compared to the γ1γ1 genotype [13]. A cross-sectional analysis of the Framingham Offspring Study of adult men and women reported a borderline significant coronary heart disease risk reduction among participants with the γ2γ;2 genotype who reported consum­ ing alcohol [14]. The HDL cholesterol results of the first study [11] led to the hypothesis that ADH1C genotype modifies the effect of alcohol on HDL cholesterol levels. A slower rate of alcohol clearance would allow more time to increase HDL cholesterol and slow metabolizers would be expected to show higher HDL cholesterol levels. However, the studies replicated the interaction with risk of heart disease, but not HDL levels [12–14], concluding that the interaction may not be mediated through effects on HDL. A step in the road toward personalized nutrition is to identify the subset(s) of the population that have different reactions to dietary factors. In the case of alcohol, such subsets, based on genetics, have been identified. The extent to which this knowledge is being or will be incorporated into nutrition practice remains unknown.

Nutrigenomics: A Possible Road to Personalized Nutrition

705

4.57.3 Caffeine Knowledge gained by incorporating genetic variation into a nutrition study not only provides a more rational basis for giving personalized dietary advice, but will also improve the quality of evidence used for making population-based dietary recommenda­ tions for the prevention of specific diseases. This can be illustrated by considering recent studies that explore the effects of caffeine on certain health outcomes [15]. These studies have helped pinpoint caffeine as the bioactive in coffee that affects the risk of various chronic diseases. These studies have also identified how a response to coffee by a specific genotype can be beneficial or detrimental depending on the disease being examined. Prior to the incorporation of genetic data into the study of coffee and risk of MI [15], the association between coffee intake and MI had been controversial, and it was not known if it was caffeine, some other compound in coffee, or another environmental factor associated with coffee consumption that may have been responsible for the association between coffee and cardiovascular disease. Distinguishing between the effects of caffeine and other compounds found in coffee has been difficult given the strong association between caffeine and coffee intake in many populations. Caffeine is metabolized primarily by the cytochrome P450 1A2 (CYP1A2) enzyme, and a polymorphism in the CYP1A2 gene determines whether individuals are rapid caffeine metabolizers (those who are homozygous for the −163 A allele) or slow caffeine metabolizers (carriers of the −163 C allele). The hypothesis of the study was that, if caffeine is involved in the development of MI, then CYP1A2 genotype should modify the association between intake of caffeinated coffee and risk of MI. The findings showed that intake of coffee was associated with an increased risk of MI only among individuals with slow caffeine metabolism [15], suggesting that caffeine increases risk of MI since it is the only major compound in coffee that is known to be detoxified by CYP1A2. Furthermore, a protective effect of moderate coffee consumption was observed among the fast metabolizers, suggesting that other components of coffee might be protective and their effects are unmasked in those who eliminate caffeine rapidly. This coffee–CYP1A2 genotype interaction has since been supported by a prospective study investigating the effect of coffee intake on the risk of developing hypertension in individuals stratified by CYP1A2 genotype [16]. The study found that the risk of hypertension associated with coffee intake varies according to CYP1A2 genotype. Slow caffeine metabolizers were at increased risk of hypertension from drinking coffee, whereas individuals with the fast caffeine metabolizing genotype appeared to drink coffee without risk [16]. This study also measured epinephrine and norepinephrine in the urine of the subjects, because these chatecho­ lamines have been shown to increase after caffeine administration in humans. Urinary epinephrine was significantly higher in coffee drinkers than abstainers, but only among slow caffeine metabolizers, which is of interest because increased sympathetic activity is considered a main mechanism through which caffeine raises blood pressure. There is no evidence to date that giving up coffee will lower blood pressure in individuals with established hypertension. A similar concept was utilized in an observational study of coffee and breast cancer [17]. By dividing subjects into CYP1A2 genotypes, this study aimed to determine if caffeine was the compound in coffee that explains the protective effect that had previously been observed between coffee and risk of breast cancer [18, 19]. Indeed, a diet–gene interaction was present. However, unlike the studies on risk of myocardial infarction (MI) and hypertension in which slow caffeine metabolizers were at increased risk from drinking coffee, in this study coffee was associated with a lower risk of breast cancer among slow metabolizers [17]. No protective effect was observed among fast metabolizers, implicating caffeine as the protective component of coffee. This is consistent with findings from animal studies showing that caffeine inhibits the development of mammary tumors [20, 21]. The study is of additional interest because the subjects were BRCA1 mutation carriers and, therefore, genetically predisposed to breast cancer. Personalized nutrition will consider genetic response to food, and also more specifically, the genetic response to food in relation to the particular disease(s) that an individual is genetically most at risk for.

4.57.4 Dietary Fat Considerable research efforts have been devoted to determining the optimal diet to prevent obesity and chronic diseases such as cardiovascular disease and type 2 diabetes. Dietary fat has been particularly studied, especially in relation to biomarkers of chronic disease such as concentrations of blood lipids. There is growing evidence that the optimal amount and type of dietary fat intake depends, in part, on an individual’s unique genetic profile [1]. Apolipoproteins (apo) A-I and A-II, coded by the APOA1 and APOA2 genes, respectively, play an important role in lipid metabolism. A recent study investigated an interaction between a common functional APOA2 polymorphism (−265T>C), saturated fat intake, and body mass index (BMI) [22]. The interaction was examined in three populations in the United States: the Framingham Offspring Study (1454 Caucasians), the Genetics of Lipid Lowering Drugs and Diet Network Study (1078 Caucasians), and the Boston-Puerto Rican Centers on Population Health and Health Disparities Study (930 Hispanics of Caribbean origin). The prevalence of the CC genotype in the three populations ranged from 10.5% to 16.2% [22]. Statistically significant interactions between the APOA2 −265T>C and saturated fat intake on BMI were identified in all three populations. Estimations of obesity for individuals with the CC genotype were almost twofold compared with carriers of the T allele in the subjects with high intake of saturated fat, but no association was detected in the low-saturated fat group. This study represents the first time that a diet–gene interaction influencing risk of obesity has been strongly and consistently replicated in three independent populations, making this a noteworthy report for the field of nutrigenomics. The replication increases the robustness of the finding

706

Intergration of Biotechnologies

and is noted because few attempts have been made previously to replicate findings of diet–gene interactions. Some gene–disease association research has shown that certain genes are influenced by epistatic (gene–gene) interactions that differ between ancestral groups [23]. The finding that the results for the diet–APOA2 polymorphism interaction effect was similar in genetically diverse populations (i.e., European Americans and Hispanic-Americans) [22] suggests that this gene may interact directly with saturated fat intake to influence energy balance across distinct populations. Peroxisome proliferator-activated receptor-gamma (PPARγ) is another gene that influences lipid metabolism and common variations have been linked to metabolic disorders. PPARγ is a nuclear receptor that is involved in many processes, including lipid metabolism and adipocyte differentiation. Because the natural ligands for PPARγ include fatty acids, it is possible that the effect of the common PPARγ variant Pro12Ala may be altered by the type of fat in the diet, particularly the ratio of dietary polyunsaturated to saturated fat. Results of an earlier study showed that when the dietary polyunsaturated fat to saturated fat ratio is low, the BMI of carriers of the Ala allele was greater than that of noncarriers, but when the dietary ratio is high, the opposite effect is observed [24]. This observation was made during a time when it became evident that genetic association studies, including those linking PPARγ to cardiometabolic diseases, were yielding inconsistent findings from one population to the next. The observation that the effect of PPARγ genotype is strongly influenced by the type of dietary fat suggested that genetic association studies needed to consider the dietary habits of the populations studied. Genetic epidemiologists, however, have been slow to adopt this knowledge. This is partly due to the challenges of assessing dietary intake in large-scale population-based studies. There has been ongoing interest in the potential protective effects of HDL cholesterol in the prevention of cardiometabolic disease because HDL cholesterol has been shown to have anti-atherogenic properties [25]. HDL cholesterol is inversely associated with risk of cardiovascular disease [26], and low plasma concentrations are a major characteristic of diabetic dyslipidemia [27]. HDL concentrations are determined by both environmental and genetic factors. Several polymorphisms have been shown to affect HDL cholesterol through interactions with dietary fat. For example, cholesterol ester transfer protein (CETP) plays an important role in the regulation of HDL metabolism, and the TaqIB polymorphism of the CETP gene has been shown to interact with dietary fat to influence HDL concentrations in men with type 2 diabetes [28]. HDL concentrations were significantly higher in men with the B2B2 or B1B2 genotype than in those with the B1B1 genotype [28]. This inverse association of the B1 allele with plasma HDL concentrations existed for those with a high consumption of animal fat, saturated fat, and monounsaturated fat (MUFA), but not for polyunsaturated (PUFA) or vegetable fat [28]. Another candidate polymorphism of interest is hepatic lipase, an enzyme that is considered to be an important determinant of HDL-cholesterol metabolism. A polymorphism has been identified in the hepatic lipase gene (LIPC), where the –514T allele is associated with significantly greater HDL-cholesterol concentrations only in subjects consuming <30% of energy from fat [29]. When total fat intake was greater than or equal to 30% of energy, mean HDL-cholesterol concentrations were lowest among those with the TT genotype, and no differences were observed between CC and CT individuals. These diet–gene interactions were seen for saturated fat and MUFA intakes, but not for PUFA [29]. Another study reported a similar interaction between the polymorphism and saturated fat on HDL cholesterol, but that was dependent on ethnocultural background [30]. In contrast, a study of men with type 2 diabetes reported that the T allele was associated with higher HDL-cholesterol concentrations only in men who had higher saturated fat intake [31]. These findings suggest that some diet–gene interactions may be modified by other environmental or genetic factors. Dietary PUFA can also alter HDL-cholesterol concentrations in humans [32], although the effects of both n-3 and n-6 PUFA have not been consistent [33]. For example, Ordovas et al. showed that a common genetic polymorphism in the APOA1 gene (–74G>A) modifies the association between dietary PUFA intake and plasma HDL-cholesterol concentrations among women in the Framingham Offspring Study [34]. Dietary PUFA has also been shown to interact with other genes to influence HDL-cholesterol concentrations. Tumor necrosis factor (TNF)-α is a major pro-inflammatory cytokine that can be modulated by PUFA [35], and is a candidate gene of interest. TNF-α has been shown to alter gene expression and activity of various proteins that tightly regulate the reverse cholesterol transport [36], and anti-TNF therapy administered to humans increases circulating levels of HDL cholesterol [36]. High concentrations of TNF-α in humans have been associated with low concentrations of circulating HDL cholesterol [37]. Two SNPs in the TNF-α gene, one at position -238G>A (rs361525), and another at position -308G>A (rs1800629), are associated with altered transcription and production of the cytokine [38, 39]. In subjects with type 2 diabetes, no effect was seen with saturated fat or MUFA, but PUFA intake was positively associated with serum HDL cholesterol in carriers of the –238A allele, and inversely associated in those with the –238GG genotype [40]. PUFA intake was inversely associated with HDL cholesterol in carriers of the – 308A allele, but not in those with the –308GG genotype [40]. An even stronger diet–gene interaction was observed when the polymorphisms at the two positions (–238/–308) were combined [40]. This interaction effect was also observed in a population of young and healthy adults [41]; however, the effects observed were in the opposite direction. These findings suggest that the health status or age of the population examined could also modify the effects of diet–gene interactions, particularly on modifiable biomarkers such as blood lipids. Nevertheless, these observations show that genetic variation influencing inflammation interacts with dietary fat to influence HDL-cholesterol concentrations. Indeed, a common polymorphism in the nuclear factor kappa B (NF-κB) NFKB1 gene (–94Ins/Del ATTG) also interacts with PUFA intake to affect HDL cholesterol [42]. NF-κB is a major regulator of inflammation and is known to regulate TNF-α. Among individuals with the Ins/Ins genotype, each 1% increase in energy from PUFA was associated with an increase in HDL cholesterol of 0.06±0.02 mmol l–1 and 0.03±0.01 among subjects with and without diabetes, respectively. An inverse relationship was observed among those with the Del/Del genotype, which was significantly different from that of the Ins/Ins groups in both populations and no effects were observed for the Ins/Del genotype in either population.

Nutrigenomics: A Possible Road to Personalized Nutrition

707

A major focus has been placed on the epidemiological diet–gene interaction studies, but clinical dietary trials that examine the difference in response to treatment among genotypes have also started to be conducted based on the findings from observational studies. For example, plasma omega 3 fatty acid response to an omega 3 fatty acid supplement was found to be modulated by apoE ε4, but not by the common PPAR-α L162V polymorphism [43]. After supplementation, only noncarriers of E4 had increased omega­ 3 in their plasma [43]. After 3 months of supplementation with 3.6 g of omega 3 fatty acids/day (containing 2.4 g of EPA and DHA) or placebo capsules containing olive oil, changes in cholesterol, including HDL cholesterol concentrations, were similar among the PPAR-α L162V genotypes [44]. In another study, subjects followed a low-fat diet for 8 weeks and then were supplemented daily with 5 g of fish oil for 6 weeks, and the decrease in plasma triglyceride concentrations was comparable for both PPAR-α L162V genotype groups, although a significant diet–gene interaction was observed for plasma C-reactive protein [45]. Such clinical studies are often limited by small sample sizes and short durations to assess physiological changes. However, such limitations could be overcome with partnerships and collaborations among researchers [46]. The replication of diet–gene interactions is an important step toward personalized nutrition because it strengthens the evidence that will be used to design clinical trials and subsequent nutrition recommendations by genotype. It is widely recognized that nutrigenomics provides industry with an incentive to develop functional foods and novel nutritional products [47]. Concern exists that functional foods created for prevention of chronic diseases may incorporate bioactive com­ pounds, such as PUFAs, without considering the interaction with genetic polymorphisms [48]. Even where diet–gene interactions are well established, there may be a lag between the science and the evidence-based development and promotion of functional foods optimized to certain genotypes. For example, the consumption of a food containing added PUFA could lead to a range of responses in individuals, from significant benefit to an adverse effect [48].

4.57.5 Fruits and Vegetables Many physiologically plausible and complementary mechanisms have been suggested to explain why high fruit and vegetable intake would be associated with a reduced risk of disease [49]. For example, fruits and vegetables can affect multiple pathways and biological processes in the human body because they are sources of water, fiber, vitamins, minerals, and phytochemicals. Phytochemicals, such as carotenoids, phenolics, isoflavonoids, isothiocyanates (ITCs), and indoles are bioactive non-nutrient chemical compounds that demonstrate a range of physiological properties. In particular, phytochemicals have been reported to induce detoxification enzymes, scavenge free radicals, reduce inflammation, inhibit malignant transformation, stimulate immune functions, and regulate the growth of cancer cells. Evidence derived from epidemiological studies has suggested that high fruit and vegetable intake is associated with a reduced risk of a variety of cancers [49] as well as cardiovascular disease [50]. However, this same protective association has not been established conclusively in clinical studies. This may be due to general technical difficulties in conducting long-term clinical trials, or to differences in dietary treatments and controls given. However, human genetic variation may also be one of the factors modifying response to plant foods and constituents. It has been observed that dietary intake of a phytochemical or its precursor may not necessarily equate with exposure at the tissue level [51]. This variation in subject response may explain, in part, the observed heterogeneity across different study populations. Characterizing how genetic factors modify the effects of a high plant-based diet or specific components in plant foods on human health outcome may clarify how plant foods influence disease risk and help identify populations that may particularly benefit from high fruit and vegetable intake. Variation in genes affecting phytochemical absorption, distribution, utilization, and excretion potentially influences nutrient exposure at the tissue level [51]. For example, the protein expression and activity of many nutrient metabolism enzymes are modulated by several compounds, including the substrates they act on. Therefore, genetic variation in the pathways within which these compounds act could alter biological response to dietary factors. Although few studies have addressed these potential genetic differences [51], one area of fruit and vegetable research that has received some attention in relation to genetic variation is the effect of polymorphisms in biotransformation enzymes such as glutathione S-transferases (GSTs). The GSTs are a family of enzymes that catalyze the transfer of glutathione to a variety of substrates and are involved in the detoxification of various carcinogens. GSTs are induced by ITCs and glucosinolates, phytochemicals that may protect against coronary heart disease and of which cruciferous vegetables (broccoli, cabbage, cauliflower, Brussels sprouts, and mustard greens) are a major dietary source. Although recent comprehensive reviews have indicated a protective role of cruciferous vegetables on cancer risk, the evidence suggests that cruciferous vegetable consumption may reduce the risk only of certain types of cancers, such as gastric and lung cancers [49]. GSTM1 and GSTT1 are isoforms of the mu and theta class of GSTs, respectively. Homozygosity for a common deletion of the GSTM1 gene (GSTM1*0) results in an absence of GSTM1 activity and is considered to be a nonfunctional genotype [52]. Similarly, a deletion polymorphism in GSTT1 leads to lack of enzyme activity and has two alleles denoted GSTT1*0 (nonfunctional) and GSTT1*1 (functional) [53]. It has been hypothesized that individuals who have nonfunctional GSTM1 and GSTT1 genotypes, and who therefore conjugate and excrete ITCs less readily, would have greater amounts of ITCs at the tissue level and hence experience a greater protective effect of ITCs consumption (51). Investigators have hypothesized that the GSTM1 null genotype might be associated with increased CYP1A2 activity among frequent consumers of cruciferous vegetables [54]. The rationale behind this is that the excretion of CYP1A2 inducers contained in cruciferous vegetables is reduced in the absence of the GSTM1 enzyme. In a cross-sectional study, subjects who consumed cruciferous vegetables at least once a week had a 16% higher CYP1A2 activity if they had the GSTM1-null genotype than if they

708

Intergration of Biotechnologies

were GSTM1-positive individuals [54]. The greatest effect was seen for broccoli, for within frequent (at least once a week) broccoli consumers, GSTM1-null individuals had a 21% higher CYP1A2 activity than did GSTM1-positive individuals [54]. A similar result was not observed in a controlled feeding study designed to test the effect of GSTM1 genotype on response to a diet high in cruciferous vegetables, where increased CYP1A2 activity in individuals on a cruciferous vegetable-containing diet was not altered by GSTM1 genotype [55]. However, serum GST-alpha and -mu concentrations increased significantly in response to vegetable consumption in GSTM1-null individuals only [56]. In addition to determining how an individual responds to an ingested amount of a specific dietary bioactive, genetic variation can also influence our food preferences and ingestive behaviors. Taste is an important determinant of how much a food is liked or disliked and subsequently eaten or not. Common polymorphisms in genes involved in taste perception may account for differences in food preferences and dietary habits [4]. This variability could affect food choices, which may influence health status and the risk of chronic disease [5]. For example, there are several vegetables that are disliked by many because they are experienced as tasting extremely bitter, yet many of these vegetables are rich sources of nutrients that have been associated with improved health outcomes. Bitter taste receptors are encoded by the family of T2R taste receptors, which consists of approximately 25 members [57]. One receptor is encoded by the TAS2R38 gene, which is characterized by three SNPs (A49P, V262A, and I296V) that make up two common haplotypes (AVI and PAV) [58, 59]. The TAS2R38 protein is known to detect two bitter substances called phenylthio­ carbamide (PTC) and 6-n-propylthiouracil (PROP) [60, 61], which are not found in foods, but have structural similarities to bitter compounds in certain foods. Carriers of the PAV haplotype have been classified as super-tasters because they have a higher sensitivity to PTC or PROP in comparison to individuals homozygous for AVI [62, 63]. TAS2R50 has been associated with risk of MI, which has been hypothesized to be due to differences in dietary preferences for bitter foods that appear to offer cardiovascular protection [64]. However, the dietary ligands that bind to this specific receptor remain unknown. Establishing a genetic basis for food likes and dislikes may partially explain some of the inconsistencies among epidemiologic studies relating diet to risk of chronic diseases because the genetic makeup of a group of individuals could be a confounder to the nutritional exposure of interest. Additionally, polymorphisms strongly associated with taste perception may potentially be useful as surrogate markers of dietary exposure in gene–disease association studies where information on dietary habits is not available. This approach is referred to as Mendelian randomization. Finally, an understanding of the genetics of taste perception may lead to the development of realistic personal and public health dietary recommendations. Fruit and vegetables are generally regarded as healthy foods to be consumed by all. However, it may be beneficial for certain individuals, based on genetics, to focus on specific fruits and vegetables. For example, genotypes associated with more favorable metabolism of carcinogens may be associated with less favorable metabolism of phytochemicals [51]. Research findings to date suggest a complex association between consumption of several vegetables and biotransformation enzyme activities in humans [55]. Genetic variation in pathways affecting nutrient absorption, transport, utilization and excretion, taste preference, and food tolerance all potentially influence the effect of plant-based diets on risk of disease.

4.57.6 Micronutrients Another step in the road to personalized nutrition is to identify the subset(s) of the population that may have different micronutrient requirements. Dietary intake of a micronutrient is not necessarily associated with certain concentrations at the blood or tissue level, potentially because of genetic differences in pathways affecting micronutrient metabolism. To date, there are no dietary reference intake (DRI) values for nutrients based on genotype. However, the micronutrients choline, vitamin C, and folate offer examples of micronutrients for which certain individuals may have different dietary requirements based on their individual genetic profile.

4.57.6.1

Choline

Choline is a nutrient needed for normal function of all cells and memory development, and can be synthesized endogenously as well as be obtained from dietary sources. Some humans deplete quickly when fed a low-choline diet, whereas others do not, potentially because of differing amounts of endogenous production of choline. A series of studies were carried out using a set study design in which adult men and women were fed diets containing adequate choline, followed by a diet containing almost no choline, until they were clinically determined to be choline deficient. When deprived of dietary choline, organ dysfunction, a sign of choline deficiency, developed in 77% of men, 80% of postmenopausal women, and 44% of premenopausal women [65]. It was concluded that sex and menopausal status influence human dietary requirements for choline. These differences are likely because estrogen induces phosphatidylethanolamine N-methyltransferase (PEMT) gene expression, which allows premenopausal women to make more choline endogenously. It was also noticed that genetic polymorphisms could limit the availability of methyltetrahy­ drofolate and thereby increase use of choline as a methyl donor because individuals with the common 5,10­ methylenetetrahydrofolate dehydrogenase –1958A allele were seven times more likely than noncarriers to develop signs of choline deficiency [66]. The most remarkable finding was the strong association in premenopausal women with the allele who had 15 times increased susceptibility to developing organ dysfunction on a low-choline diet [66]. These findings indicate that nutrient needs can vary greatly for different individuals and that different choline requirement recommendations could be made for three groups: young women, men and older women, and individuals with the 5,10-methylenetetrahydrofolate dehydrogenase –1958A allele.

Nutrigenomics: A Possible Road to Personalized Nutrition

4.57.6.2

709

Vitamin C

Vitamin C (ascorbic acid) inhibits oxidative damage, recycles other antioxidants, facilitates iron absorption, aids in the conversion of cholesterol to bile acids and is also an essential cofactor for enzymatic reactions required for the synthesis of carnitine, collagen, norepinephrine, and epinephrine. As would be expected from the functions of vitamin C, there is evidence for an inverse relation­ ship between serum ascorbic acid concentrations and risk of chronic diseases such as cardiovascular disease [67, 68], and diabetes [69, 70]. However, the findings of studies into dietary vitamin C and the prevention of chronic diseases remain inconsistent and controversial [71]. This inconsistency could be due to individual variability in serum ascorbic acid concentrations since studies report variable dose–response curves between dietary vitamin C and serum ascorbic acid [71, 72]. Several determinants of serum ascorbic acid have been identified, but they explain only a small portion of the large individual variability [73]. After the ingestion of vitamin C, ascorbic acid concentrations in the blood are established through a combination of overlapping mechanisms, offering several possibilities of proteins in which a functional change, due to genetic variation, could affect serum ascorbic acid response to dietary vitamin C. Significant diet–gene interactions have been observed between dietary vitamin C and GST (theta and mu classes) genotypes on blood concentrations of ascorbic acid [74]. GSTs are a family of detoxifying enzymes that contribute to the glutathione– ascorbic acid antioxidant cycle in the human body. The omega class of GST is directly able to reduce dehydroascorbic acid back to ascorbic acid, but it is not known whether any of the other classes of GST can carry out this function as well. The mu and theta classes of GSTs each contain a common genetic polymorphism that results in a nonfunctional enzyme. The association between the GST genotype and serum ascorbic acid concentrations depended on the amount of dietary vitamin C consumed. Individuals with the GST nonfunctional mu and theta genotypes were found to have increased risk of serum ascorbic acid deficiency if they did not meet the recommended dietary allowance (RDA) for vitamin C [74]. For example, the odds ratio (95% confidence interval) for serum ascorbic acid deficiency (<11 μmol l–1) was 2.17 (1.10, 4.28) for subjects with the functional GSTT1*1 allele who did not meet the RDA of vitamin C compared to those who did. However, the odds ratio was 12.28 (4.26, 33.42) for individuals with the nonfunctional GSTT1*0/*0 genotype. The corresponding odds ratios were and 2.29 (0.96, 5.45) and 4.03 (2.01, 8.09) for the functional GSTM1*1 allele and nonfunctional GSTM1*0/*0 genotype, respectively. This study demonstrates that the RDA for vitamin C is protective against serum ascorbic acid deficiency, regardless of genotype. However, some individuals appear to be more vulnerable to deficiency if not obtaining recommended daily amounts of dietary vitamin C, and so meeting dietary requirements could be emphasized in such populations. Additionally, because the functional version of the gene appears to protect against deficiency, these findings suggest a novel biological role for the GST-theta and -mu enzymes in sparing serum ascorbic acid, possibly by aiding in the reduction of dehydroascorbic acid to ascorbic acid.

4.57.6.3

Folate

Adequate folate intake appears to be important in the prevention of several major chronic diseases [75]. For example, deficient folate status is associated with the elevation of blood homocysteine levels, which may increase risk of cardiovascular disease [76]. Deficient folate status can also lead to abnormal DNA synthesis, methylation, and repair, which may increase cancer risk [77]. Indeed, higher total folate intake is inversely associated with the risk of some cancers such as colorectal cancer [78], although the effects of folate on cancer might be a ‘double-edged sword’ since it may actually increase the development of established tumors [79]. Methylenetetrahydrofolate reductase (MTHFR) is a key enzyme involved in the metabolism of folate. The MTHFR 677C/T polymorphism is a nonsynonymous single nucleotide polymorphism that results in impaired enzyme activity. This polymorphism is associated with a reduced risk of some forms of cancer; however, the protective effect of this folate-related polymorphism is dependent on adequate folate status [78]. Therefore, cancer risk may be increased in individuals with the homozygous genotype for the MTHFR 677C/T polymorphism who have low status of methyl-related nutrients including folate [78]. Modeling analyses have recently been developed to project RDAs for nutrients based on genotype, as used in a recent assessment of the effect of the MTHFR 677C/T polymorphism on the current folate RDA for American adults (400 μg d–1) by race and ethnicity [80]. This assessment concluded that the MTHFR 677C/T variant, the most widely studied genotype affecting folate metabolism, may not necessarily affect individual requirements because the projected RDA for each genotype was similar to the current RDA [80]. Parallel analyses could be conducted for determining upper limits for nutrients since genetic variations could also impact the levels of nutrients that cause toxicity.

4.57.7 Conclusions Dietary intake of a nutrient does not necessarily result in the same concentrations in the blood or tissue because substantial individual variability in the absorption, distribution, metabolism, and elimination can exist. The mechanisms responsible for the between-person differences in dietary response are very complex and have been poorly understood. Research to date has indicated that diet–gene interactions play a significant role in this between-person variability, and has clarified some of these genetic differences.

710

Intergration of Biotechnologies

The interaction between genetic and dietary influences can result in a higher risk of disease in certain individuals and populations. Currently, diet–gene association studies are revealing evidence on which to base gene-specific dietary intervention trials to confirm results. Replication of current findings and further research in the form of genotype-specific nutritional intervention studies are necessary. The future of nutrigenomics research promises to provide additional knowledge of biological function and individual response to diet. This iterative approach to health research is paving the road toward personalized nutrition. Nutrigenomic practices for specific monogenic conditions such as PKU are currently being successfully used in healthcare interventions in the form of newborn screening programs. The application of nutrigenomics by healthcare professionals for the prevention and treatment of complex chronic diseases, however, has not yet been widely adopted. Whether such practice will be feasible for the population-at-large in the immediate future remains to be determined, but the principles and tools of nutrigenomics are expected to soon allow for earlier and more targeted interventions than currently exist [81]. As the current research in nutrigenomics often focuses on how diet–gene interactions influence phenotypes found to be predictive biomarkers of disease [82], it is likely that the path from research to applications will proceed into clinical practice using these markers of chronic disease as outcome measures. Recent developments in genome-wide approaches have already identified many susceptibility alleles for common complex diseases [83], including several previously unknown etiologic pathways in disease pathogenesis [84], and have the potential to identify novel targets for prevention or treatment with dietary factors. Diet is an important environmental factor that interacts with the genome to modulate disease risk. A clear understanding of these interactions has the potential to support disease prevention through optimization of dietary recommendations. The extent to which nutrigenomics will be incorporated in nutrition therapy and promotion remains unknown. However, nutrigenomics has emerged as a rapidly developing research area with great potential to yield findings that could change the way dietary guidelines for populations and recommendations for individuals are established and advised in the future.

References [1] Ordovas JM (2008) Genotype–phenotype associations: Modulation by diet and obesity. Obesity (Silver Spring) 16(supplement 3): S40–S46. [2] Ordovas JM, Kaput J, and Corella D (2007) Nutrition in the genomics era: Cardiovascular disease risk and the Mediterranean diet. Molecular Nutrition Food Research 51: 1293–1299. [3] El-Sohemy A (2007) Nutrigenetics. Forum Nutrition 60: 25–30. [4] Eny KM, and El-Sohemy A (2010) The Genetic Determinants of Ingestive Behaviour: Sensory, Energy Homeostasis and Food Reward aspects of ingestive behaviour. In: Dube L, Bechara A, DagherA, Drewnowski A, LeBel J, James P, Yada RY, and Laflamme-Sanders M.-C. Obesity Prevention: the role of brain and society on individual behavior: 149–160. Elsevier. [5] Garcia-Bailo B, Toguri C, Eny KM, and El-Sohemy A (2009) Genetic variation in taste and its influence on food selection OMICS 13: 69–80. [6] Swallow DM (2003) Genetics of lactase persistence and lactose intolerance. Annual Review of Genetics 37: 197–219. [7] Swagerty DL, Jr., Walling AD, and Klein RM (2002) Lactose intolerance. American Family Physician 65: 1845–1850. [8] DiLella AG, Kwok SC, Ledley FD, et al. (1986) Molecular structure and polymorphic map of the human phenylalanine hydroxylase gene. Biochemistry 25: 743–749. [9] Surtees R and Blau N (2000) The neurochemistry of phenylketonuria. European Journal of Pediatrics 159(supplement 2): S109–S113. [10] (2001) National Institutes of Health Consensus Development Conference Statement: Phenylketonuria: Screening and management, October 16–18, 2000. Pediatrics 108: 972–982. [11] Hines LM, Stampfer MJ, Ma J, et al. (2001) Genetic variation in alcohol dehydrogenase and the beneficial effect of moderate alcohol consumption on myocardial infarction. New England Journal of Medicine 344: 549–555. [12] Heidrich J, Wellmann J, Doring A, et al. (2007) Alcohol consumption, alcohol dehydrogenase and risk of coronary heart disease in the MONICA/KORA-Augsburg cohort 1994/ 1995–2002. European Journal of Cardiovascular Prevention and Rehabilitation 14: 769–774. [13] Younis J, Cooper JA, Miller GJ, et al. (2005) Genetic variation in alcohol dehydrogenase 1C and the beneficial effect of alcohol intake on coronary heart disease risk in the Second Northwick Park Heart Study. Atherosclerosis 180: 225–232. [14] Djousse L, Levy D, Herbert AG, et al. (2005) Influence of alcohol dehydrogenase 1C polymorphism on the alcohol-cardiovascular disease association (from the Framingham Offspring Study). American Journal of Cardiology 96: 227–232. [15] Cornelis MC, El-Sohemy A, Kabagambe EK, and Campos H (2006) Coffee, CYP1A2 genotype, and risk of myocardial infarction. Journal of the American Medical Association 295: 1135–1141. [16] Palatini P, Ceolotto G, Ragazzo F, et al. (2009) CYP1A2 genotype modifies the association between coffee intake and the risk of hypertension. Journal of Hypertension 27: 1594–1601. [17] Kotsopoulos J, Ghadirian P, El-Sohemy A, et al. (2007) The CYP1A2 genotype modifies the association between coffee consumption and breast cancer risk among BRCA1 mutation carriers. Cancer Epidemiology, Biomarkers and Prevention 16: 912–916. [18] Nkondjock A, Ghadirian P, Kotsopoulos J, et al. (2006) Coffee consumption and breast cancer risk among BRCA1 and BRCA2 mutation carriers. International Journal of Cancer 118: 103–107. [19] Baker JA, Beehler GP, Sawant AC, et al. (2006) Consumption of coffee, but not black tea, is associated with decreased risk of premenopausal breast cancer. Journal of Nutrition 136: 166–171. [20] Wolfrom DM, Rao AR, and Welsch CW (1991) Caffeine inhibits development of benign mammary gland tumors in carcinogen-treated female Sprague-Dawley rats. Breast Cancer Research and Treatment 19: 269–275. [21] Yang H, Rouse J, Lukes L, et al. (2004) Caffeine suppresses metastasis in a transgenic mouse model: A prototype molecule for prophylaxis of metastasis. Clinical and Experimental Metastasis 21: 719–735. [22] Corella D, Peloso G, Arnett DK, et al. (2009) APOA2, dietary fat, and body mass index: Replication of a gene-diet interaction in 3 independent populations. Archives of Internal Medicine 169: 1897–1906. [23] Helgadottir A, Manolescu A, Helgason A, et al. (2006) A variant of the gene encoding leukotriene A4 hydrolase confers ethnicity-specific risk of myocardial infarction. Nature Genetics 38: 68–74. [24] Luan J, Browne PO, Harding AH, et al. (2001) Evidence for gene–nutrient interaction at the PPARgamma locus. Diabetes 50: 686–689. [25] Barter PJ, Nicholls S, Rye KA, et al. (2004) A ntiinflammatory properties of HDL. Circulation Research 95: 764–772. [26] Chirovsky DR, Fedirko V, Cui Y, et al. (2009) Prospective studies on the relationship between high-density lipoprotein cholesterol and cardiovascular risk: A systematic review. European Journal of Cardiovascular Prevention and Rehabilitation 16: 404–423.

Nutrigenomics: A Possible Road to Personalized Nutrition

711

[27] Nesto RW (2005) Beyond low-density lipoprotein: Addressing the atherogenic lipid triad in type 2 diabetes mellitus and the metabolic syndrome. American Journal of Cardiovascular Drugs 5: 379–387. [28] Li TY, Zhang C, Asselbergs FW, et al. (2007) Interaction between dietary fat intake and the cholesterol ester transfer protein TaqIB polymorphism in relation to HDL-cholesterol concentrations among US diabetic men. American Journal of Clinical Nutrition 86: 1524–1529. [29] Ordovas JM, Corella D, Demissie S, et al. (2002) 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 106: 2315–2321. [30] Tai ES, Corella D, Deurenberg-Yap M, et al. (2003) Dietary fat interacts with the −514C → T polymorphism in the hepatic lipase gene promoter on plasma lipid profiles in a multiethnic Asian population: The 1998 Singapore National Health Survey. Journal of Nutrition 133: 3399–3408. [31] Zhang C, Lopez-Ridaura R, Rimm EB, et al. (2005) Interactions between the −514C → T polymorphism of the hepatic lipase gene and lifestyle factors in relation to HDL concentrations among US diabetic men. American Journal of Clinical Nutrition 81: 1429–1435. [32] Wijendran V and Hayes KC (2004) Dietary n–6 and n–3 fatty acid balance and cardiovascular health. Annual Review of Nutrition 24: 597–615. [33] Russo GL (2009) Dietary n–6 and n–3 polyunsaturated fatty acids: From biochemistry to clinical implications in cardiovascular prevention. Biochemical Pharmacology 77: 937–946. [34] Ordovas JM, Corella D, Cupples LA, et al. (2002) Polyunsaturated fatty acids modulate the effects of the APOA1 G-A polymorphism on HDL-cholesterol concentrations in a sex-specific manner: The Framingham Study. American Journal of Clinical Nutrition 75: 38–46. [35] Grimble RF and Tappia PS (1998) Modulation of pro-inflammatory cytokine biology by unsaturated fatty acids. Zeitschrift für Ernährungswissenschaft 37(supplement 1): 57–65. [36] Popa C, van den Hoogen FH, Radstake TR, et al. (2007) Modulation of lipoprotein plasma concentrations during long-term anti-TNF therapy in patients with active rheumatoid arthritis. Annals of the Rheumatic Diseases 66: 1503–1507. [37] Svenungsson E, Gunnarsson I, Fei GZ, et al. (2003) Elevated triglycerides and low levels of high-density lipoprotein as markers of disease activity in association with up-regulation of the tumor necrosis factor alpha/tumor necrosis factor receptor system in systemic lupus erythematosus. Arthritis and Rheumatism 48: 2533–2540. [38] Kaluza W, Reuss E, Grossmann S, et al. (2000) Different transcriptional activity and in vitro TNF-alpha production in psoriasis patients carrying the TNF-alpha 238A promoter polymorphism. Journal of Investigative Dermatology 114: 1180–1183. [39] Kroeger KM, Carville KS, and Abraham LJ (1997) The –308 tumor necrosis factor-alpha promoter polymorphism effects transcription. Molecular Immunology 34: 391–399. [40] Fontaine-Bisson B, Wolever TM, Chiasson JL, et al. (2007) Genetic polymorphisms of tumor necrosis factor-alpha modify the association between dietary polyunsaturated fatty acids and fasting HDL-cholesterol and apo A-I concentrations. American Journal of Clinical Nutrition 86: 768–774. [41] Fontaine-Bisson B and El-Sohemy A (2008) Genetic polymorphisms of tumor necrosis factor-alpha modify the association between dietary polyunsaturated fatty acids and plasma high-density lipoprotein-cholesterol concentrations in a population of young adults. Journal of Nutrigenetics and Nutrigenomics 1: 215–223. [42] Fontaine-Bisson B, Wolever TM, Connelly PW, et al. (2009) NF-kappaB –94Ins/Del ATTG polymorphism modifies the association between dietary polyunsaturated fatty acids and HDL-cholesterol in two distinct populations. Atherosclerosis 204: 465–470. [43] Plourde M, Vohl MC, Vandal M, et al. (2009) Plasma n–3 fatty acid response to an n–3 fatty acid supplement is modulated by apoE epsilon4 but not by the common PPAR-alpha L162V polymorphism in men. British Journal of Nutrition 102: 1121–1124. [44] Lindi V, Schwab U, Louheranta A, et al. (2003) Impact of the Pro12Ala polymorphism of the PPAR-gamma2 gene on serum triacylglycerol response to n–3 fatty acid supplementation. Molecular Genetics and Metabolism 79: 52–60. [45] Caron-Dorval D, Paquet P, Paradis AM, et al. (2008) Effect of the PPAR-alpha L162V polymorphism on the cardiovascular disease risk factor in response to n–3 polyunsaturated fatty acids. Journal of Nutrigenetics and Nutrigenomics 1: 205–212. [46] McCabe-Sellers B, Lovera D, Nuss H, et al. (2008) Personalizing nutrigenomics research through community based participatory research and omics technologies. OMICS 12: 263–272. [47] Kaput J (2007) Developing the promise of nutrigenomics through complete science and international collaborations. Forum of Nutrition 60: 209–223. [48] Ferguson LR (2009) Nutrigenomics approaches to functional foods. Journal of the American Dietetic Association 109: 452–458. [49] Kim MK and Park JH (2009) Conference on “Multidisciplinary approaches to nutritional problems.” Symposium on “Nutrition and health.” Cruciferous vegetable intake and the risk of human cancer: Epidemiological evidence. Proceedings of the Nutrition Society 68: 103–110. [50] He FJ, Nowson CA, and MacGregor GA (2006) Fruit and vegetable consumption and stroke: Meta-analysis of cohort studies. Lancet 367: 320–326. [51] Lampe JW (2009) Interindividual differences in response to plant-based diets: Implications for cancer risk. American Journal of Clinical Nutrition 89: 1553S–1557S. [52] Xu S, Wang Y, Roe B, and Pearson WR (1998) Characterization of the human class mu glutathione-S-transferase gene cluster and the GSTM1 deletion. Journal of Biological Chemistry 273: 3517–3527. [53] Bruhn C, Brockmoller J, Kerb R, et al. (1998) Concordance between enzyme activity and genotype of glutathione S-transferase theta (GSTT1). Biochemical Pharmacology 56: 1189–1193. [54] Probst-Hensch NM, Tannenbaum SR, Chan KK, et al. (1998) Absence of the glutathione S-transferase M1 gene increases cytochrome P4501A2 activity among frequent consumers of cruciferous vegetables in a Caucasian population. Cancer Epidemiology, Biomarkers and Prevention 7: 635–638. [55] Lampe JW, King IB, Li S, et al. (2000) Brassica vegetables increase and apiaceous vegetables decrease cytochrome P450 1A2 activity in humans: Changes in caffeine metabolite ratios in response to controlled vegetable diets. Carcinogenesis 21: 1157–1162. [56] Lampe JW, Chen C, Li S, et al. (2000) Modulation of human glutathione S-transferases by botanically defined vegetable diets. Cancer Epidemiology, Biomarkers and Prevention 9: 787–793. [57] Behrens M and Meyerhof W (2006) Bitter taste receptors and human bitter taste perception. Cellular and Molecular Life Sciences 63: 1501–1509. [58] Bufe B, Breslin PA, Kuhn C, et al. (2005) The molecular basis of individual differences in phenylthiocarbamide and propylthiouracil bitterness perception. Current Biology 15: 322–327. [59] Mennella JA, Pepino MY, and Reed DR (2005) Genetic and environmental determinants of bitter perception and sweet preferences. Pediatrics 115: 216–222. [60] El-Sohemy A, Stewart L, Khataan N, et al. (2007) Nutrigenomics of taste – impact on food preferences and food production. Forum of Nutrition 60: 176–182. [61] Kim UK, Jorgenson E, Coon H, et al. (2003) Positional cloning of the human quantitative trait locus underlying taste sensitivity to phenylthiocarbamide. Science 299: 1221–1225. [62] Wooding S, Kim UK, Bamshad MJ, et al. (2004) Natural selection and molecular evolution in PTC, a bitter-taste receptor gene. American Journal of Human Genetics 74: 637–646. [63] Drayna D (2005) Human taste genetics. Annual Review of Genomics and Human Genetics 6: 217–235. [64] Shiffman D, Ellis SG, Rowland CM, et al. (2005) Identification of four gene variants associated with myocardial infarction. American Journal of Human Genetics 77: 596–605. [65] Fischer LM, daCosta KA, Kwock L, et al. (2007) Sex and menopausal status influence human dietary requirements for the nutrient choline. American Journal of Clinical Nutrition 85: 1275–1285. [66] Kohlmeier M, da Costa KA, Fischer LM, and Zeisel SH (2005) Genetic variation of folate-mediated one-carbon transfer pathway predicts susceptibility to choline deficiency in humans. Proceedings of the National Academy of Sciences of the United States of America 102: 16025–16030. [67] Boekholdt SM, Meuwese MC, Day NE, et al. (2006) Plasma concentrations of ascorbic acid and C-reactive protein, and risk of future coronary artery disease, in apparently healthy men and women: The EPIC-Norfolk prospective population study. British Journal of Nutrition 96: 516–522. [68] Jacob RA and Sotoudeh G (2002) Vitamin C function and status in chronic disease. Nutrition in Clinical Care 5: 66–74. [69] Paolisso G, D’Amore A, Balbi V, et al. (1994) Plasma vitamin C affects glucose homeostasis in healthy subjects and in non-insulin-dependent diabetics. American Journal of Physiology 266: E261–E268. [70] Sinclair AJ, Taylor PB, Lunec J, et al. (1994) Low plasma ascorbate levels in patients with type 2 diabetes mellitus consuming adequate dietary vitamin C. Diabetic Medicine 11: 893–898.

712

Intergration of Biotechnologies

[71] Loria CM, Whelton PK, Caulfield LE, et al. (1998) Agreement among indicators of vitamin C status. American Journal of Epidemiology 147: 587–596. [72] Levine M, Conry-Cantilena C, Wang Y, et al. (1996) Vitamin C pharmacokinetics in healthy volunteers: Evidence for a recommended dietary allowance. Proceedings of the National Academy of Sciences of the United States of America 93: 3704–3709. [73] Block G, Mangels AR, Patterson BH, et al. (1999) Body weight and prior depletion affect plasma ascorbate levels attained on identical vitamin C intake: A controlled-diet study. Journal of the American College of Nutrition 18: 628–637. [74] Cahill LE, Fontaine-Bisson B, and El-Sohemy A (2009) Functional genetic variants of glutathione S-transferase protect against serum ascorbic acid deficiency. American Journal of Clinical Nutrition 90: 1411–1417. [75] Jiang R, Hu FB, Giovannucci EL, et al. (2003) Joint association of alcohol and folate intake with risk of major chronic disease in women. American Journal of Epidemiology 158: 760–771. [76] Boushey CJ, Beresford SA, Omenn GS, and Motulsky AG (1995) A quantitative assessment of plasma homocysteine as a risk factor for vascular disease. Probable benefits of increasing folic acid intakes. Journal of the American Medical Association 274: 1049–1057. [77] Blount BC, Mack MM, Wehr CM, et al. (1997) Folate deficiency causes uracil misincorporation into human DNA and chromosome breakage: Implications for cancer and neuronal damage. Proceedings of the National Academy of Sciences of the United States of America 94: 3290–3295. [78] Bailey LB (2003) Folate, methyl-related nutrients, alcohol, and the MTHFR 677C T polymorphism affect cancer risk: Intake recommendations. Journal of Nutrition 133: 3748S–3753S. [79] Smith AD, Kim YI, and Refsum H (2008) Is folic acid good for everyone? American Journal of Clinical Nutrition 87: 517–533. [80] Robitaille J, Hamner HC, Cogswell ME, and Yang Q (2009) Does the MTHFR 677C T variant affect the recommended dietary allowance for folate in the US population? American Journal of Clinical Nutrition 89: 1269–1273. [81] DeBusk R (2009) Diet-related disease, nutritional genomics, and food and nutrition professionals. Journal of the American Dietetic Association 109: 410–413. [82] Kaput J, Noble J, Hatipoglu B, et al. (2007) Application of nutrigenomic concepts to type 2 diabetes mellitus. Nutrition, Metabolism and Cardiovascular Diseases 17: 89–103. [83] Office of Population Genomics, National Human Genome Research Institute, National Institutes of Health. A Catalog of Published Genome-Wide Association Studies. [84] Ding K and Kullo IJ (2009) Genome-wide association studies for atherosclerotic vascular disease and its risk factors. Circulation: Cardiovascular Genetics 2: 63–72.