CHD is a multifactorial disease, caused by both genetic and environmental factors. The inherited 'defective' genes will vary from individual to individual, and any single mutation is likely to be making only a small contribution to risk. The context dependency, i.e. the importance of environmental factors in influencing genetic risk, is now becoming evident. Thus, a mutation may have a modest effect on risk in individuals who maintain a low environmental risk, but a major effect in a high-risk environment. Methods of analysing gene–environment interactions on CHD risk will be discussed and illustrated with several examples. APOE has three common alleles, ɛ2, ɛ3 and ɛ4. The ɛ4 allele has consistently been associated with CHD risk, which has been confirmed by meta-analysis. However, when the effect of genotype on risk was considered in smokers and non-smokers separately, risk in non-smokers was similar in all APOE genotypes. By comparison, in the smokers, ɛ3 homozygotes, as expected, had an approximately 2-fold higher risk, while for ɛ4 carriers there was a significantly greater than additive effect of genotype and smoking on risk (P ≫0.007). Thus, the impact of the ɛ4 allele on CHD risk appears to be confined to current smokers, an effect that has been confirmed in several studies. Another example is the interaction between the alcohol dehydrogenase 3 gene variant and alcohol consumption on CHD risk (P ≫0.001), showing the context dependency of the effect. Thus, the importance of considering environmental factors as potential genotype-risk modifiers has major public health implications.

Fe homeostasis is maintained by regulation of Fe absorption to balance largely unregulated body Fe losses. The majority of human subjects maintain relatively constant Fe stores; however, Fe deficiency and Fe overload are common conditions. Fe overload is frequently associated with mutations in genes of Fe metabolism. The present paper summarises present knowledge of these mutations as well as indicating other genes that animal studies have implicated as candidates for influencing body Fe stores.

Mammalian members of the cation diffusion facilitator (CDF) and zrt-, irt-like protein (ZIP) families of Zn transporters, initially identified in Saccharomyces cerevisiae and Arabidopsis thalania spp., have been cloned during the last 8 years and have been classified as families SLC30 and SLC39 respectively. The cloning of human Zn transporters ZnT-like transporter 1 (hZTL1)/ZnT5 (SLC30A5) and hZIP4 (SLC39A4) were major advances in the understanding of the molecular mechanisms of dietary Zn absorption. Both transporters are localised at the enterocyte apical membrane and are, therefore, potentially of fundamental importance in dietary Zn uptake. hZTL1 mediates Zn uptake when expressed in Xenopus laevis oocytes and hZIP4 is mutated in most cases of the inherited Zn deficiency disease acrodermatitis enteropathica. Localisation of hZTL1/ZnT5 at the apical membrane of the placental syncytiotrophoblast indicates a fundamental role in the transfer of Slc30 Zn to the foetus. Observations in rodent models indicate that in the intestine increased Zn availability increases expression of Zn transporters. Human intestinal Caco-2 cells show a similar response to increasing the Zn2+ concentration of the nutrient medium in relation to the expression of mRNA corresponding to several Zn transporters and that of ZnT1 (SLC30A1) and hZTL1/ZnT5 proteins. In the human placental cell line JAR, however, expression at the mRNA level of a number of Zn transporters is not modified by Zn availability, whilst ZnT1 and hZTL1/ZnT5 proteins are reduced under Zn-supplemented conditions. These differences between Caco-2 and JAR cells in Zn transporter gene responses to Zn supply may reflect the different extracellular Zn concentrations encountered by the corresponding cell types in vitro.

Cu is an essential nutrient that is required for a broad range of cellular and molecular processes. Mammals have efficient systems to control Cu homeostasis that operate at the level of controlling uptake, distribution, sequestration and excretion of Cu. The study of diseases associated with disturbed Cu homeostasis has greatly enhanced our understanding of the molecular mechanisms involved in Cu metabolism. In man the liver is responsible for excreting excess Cu from the body by means of biliary secretion. Wilson disease is a severe human disorder characterized by Cu accumulation in the liver as a result of a deficiency in biliary Cu secretion. This disorder is caused by mutations in the gene that encodes a Cu-transporting P-type ATPase (ATP7B). The MURR1 gene was identified recently, and it was hypothesized that this gene is also essential for biliary Cu excretion and is presumed to act downstream of ATP7B. MURR1 is mutated in canine Cu toxicosis, a disorder with phenotypic characteristics similar to those of Wilson disease. MURR1 encodes a protein that is of unknown function and is without detectable sequence homology to known proteins. MURR1 is readily detected in all tissues and cell types, suggesting that it may exhibit a pleiotropic function in different organs, which may or may not be exclusively linked to Cu homeostasis. The use of genetic, biochemical and genomic tools, as well as the development of appropriate models in organisms other than dog, will allow the elucidation of the molecular and cellular function of MURR1 in relation to hepatic Cu homeostasis and biliary Cu excretion.

The nutritional biochemistry of trivalent Cr has been a poorly understood field of study; investigations of the biochemistry of the other essential transition metals have not proven as problematic. Despite over four decades of endeavour, only recently has a picture of the role of Cr potentially started to be defined. The biologically-relevant form is the trivalent ion. Cr3+ appears to be required for proper carbohydrate and lipid metabolism in mammals, although fortunately Cr deficiency is difficult to achieve. Conditions that increase circulating glucose and insulin concentrations increase urinary Cr output. Cr is probably excreted in the form of the oligopeptide chromodulin. Chromodulin may be the key to understanding the role of Cr at a molecular level, as the molecule has been found to bind to activated insulin receptor, stimulating its kinase activity. A mechanism for the action of chromodulin has recently been proposed; this mechanism can serve as a potential framework for further studies to test the role of Cr in metabolism. An examination of the nutritional supplement chromium picolinate illustrates some of the difficulties associated with these biochemical studies.

Alcohol dehydrogenase (ADH) and mitochondrial aldehyde dehydrogenase (ALDH2) are responsible for metabolizing the bulk of ethanol consumed as part of the diet and their activities contribute to the rate of ethanol elimination from the blood. They are expressed at highest levels in liver, but at lower levels in many tissues. This pathway probably evolved as a detoxification mechanism for environmental alcohols. However, with the consumption of large amounts of ethanol, the oxidation of ethanol can become a major energy source and, particularly in the liver, interferes with the metabolism of other nutrients. Polymorphic variants of the genes for these enzymes encode enzymes with altered kinetic properties. The pathophysiological effects of these variants may be mediated by accumulation of acetaldehyde; high-activity ADH variants are predicted to increase the rate of acetaldehyde generation, while the low-activity ALDH2 variant is associated with an inability to metabolize this compound. The effects of acetaldehyde may be expressed either in the cells generating it, or by delivery of acetaldehyde to various tissues by the bloodstream or even saliva. Inheritance of the high-activity ADH β2, encoded by the ADH2*2 gene, and the inactive ALDH2*2 gene product have been conclusively associated with reduced risk of alcoholism. This association is influenced by gene–environment interactions, such as religion and national origin. The variants have also been studied for association with alcoholic liver disease, cancer, fetal alcohol syndrome, CVD, gout, asthma and clearance of xenobiotics. The strongest correlations found to date have been those between the ALDH2*2 allele and cancers of the oro-pharynx and oesophagus. It will be important to replicate other interesting associations between these variants and other cancers and heart disease, and to determine the biochemical mechanisms underlying the associations.

Chronic alcohol consumption is a major risk factor for cancer of upper aero-digestive tract (oro-pharynx, hypopharynx, larynx and oesophagus), the liver, the colo-rectum and the breast. Evidence has accumulated that acetaldehyde is predominantly responsible for alcohol-associated carcinogenesis. Acetaldehyde is carcinogenic and mutagenic, binds to DNA and protein, destroys the folate molecule and results in secondary cellular hyper-regeneration. Acetaldehyde is produced by mucosal and cellular alcohol dehydrogenase, cytochrome P450 2E1 and through bacterial oxidation. Its generation and/or its metabolism is modulated as a result of polymorphisms or mutations of the genes responsible for these enzymes. Acetaldehyde can also be produced by oral bacteria. Smoking, which changes the oral bacterial flora, also increases salivary acetaldehyde. Cigarette smoke and some alcoholic beverages, such as Calvados, contain acetaldehyde. In addition, chronic alcohol consumption induces cytochrome P450 2E1 enxyme activity in mucosal cells, resulting in an increased generation of reactive oxygen species and in an increased activation of various dietary and environmental carcinogens. Deficiencies of riboflavin, Zn, folate and possibly retinoic acid may further enhance alcohol-associated carcinogenesis. Finally, methyl deficiency as a result of multiple alcohol-induced changes leads to DNA hypomethylation. A depletion of lipotropes, including methionine, choline, betaine and S-adenosylmethionine, as well as folate, results in the hypomethylation of oncogenes and may lead to DNA strand breaks, all of which are associated with increased carcinogenesis.

The deleterious health effects of high alcohol consumption are numerous and well recognized; however, the effect of moderate alcohol consumption on overall health continues to be a debated issue. Among the more prevalent diseases in Westernized countries, epidemiological research suggests that alcohol in moderation substantially reduces the risk of CHD, while it modestly increases the risk for certain cancers, such as breast and colon cancer. Despite the overwhelming data supporting the beneficial effect of moderate alcohol consumption on the cardiovascular system, some researchers are not convinced. Sceptics argue that the reduction in risk is attributed to a favourable lifestyle factor associated with moderate alcohol consumption, or that it may be attributed to constituents of alcoholic beverages other than ethanol, such as the antioxidants in the grapes. In order to promote overall health for the general public, it is necessary to elucidate these issues. One approach is to study population differences in alcohol metabolic efficiency, which is likely to contribute to an individual's susceptibility to alcohol-associated diseases. Among the population there is substantial variability in the efficiency to metabolize alcohol. Genetic variation among the alcohol-metabolizing genes is known to produce isoenzymes with distinct kinetic properties. Studying genetic differences that potentially influence disease susceptibility among populations may provide insight into the mechanism(s) for the relationship between risk factor and disease, such as alcohol and CHD.

In view of the developments in health care relating to the increased prevalence and incidence of chronic diseases and the continuing increase in health-care expenditure, more attention should be paid to health maintenance and disease prevention. Any strategy that can influence health maintenance is of interest, especially lifestyle factors such as nutrition, exercise or stress control. Alcohol has an important place in the daily life of many healthy as well as sick individuals. Alcohol has three major characteristics; it is a nutrient (energy source), a psycho-active drug and a toxin. Each consumer has the choice of which of the characteristics of alcohol he/she wants to utilise. Thus, alcohol represents one of the most important self-implemented disease modifiers in our modern society. The major determinants of the health effects of alcohol are the absolute amount consumed, the consumption frequency, associated lifestyle factors (e.g. smoking, nutrient intake, substrate composition, physical activity pattern) and last, but not least, the genetic background. There are few known disease conditions that have not already been associated positively or negatively with alcohol consumption. The list of diseases includes atherosclerosis, dementia, diabetes, obesity and conditions relating to Zn metabolism. Obesity represents the most important disease modifier in the world and the prevalence rates are increasing rapidly. Evidence suggests that alcohol represents a risk factor for overweight and obesity as a result of specific effects on energy metabolism and substrate metabolism. The potential role of alcohol as an important modulator for the postprandial lipidaemia and its role in the pathogenesis of modern diseases will be discussed.

Over the last 25 years, the effects of fatty acids on the immune system have been characterized using in vitro, animal and human studies. Advances in fatty acid biochemistry and molecular techniques have recently suggested new mechanisms by which fatty acids could potentially modify immune responses, including modification of the organization of cellular lipids and interaction with nuclear receptors. Possibilities for the clinical applications of n-3 PUFA are now developing. The present review focuses on the hypothesis that the anti-inflammatory properties of n-3 PUFA in the arterial wall may contribute to the protective effects of n-3 PUFA in CVD, as suggested by epidemiological and secondary prevention studies. Studies are just beginning to show that dietary n-3 PUFA can be incorporated into plaque lipid in human subjects, where they may influence the morphology and stability of the atherosclerotic lesion.

Enteral tube feeding is common in both the hospital and community environment; however, patients can suffer alterations in faecal output that can have serious clinical sequelae. Problems associated with accurate characterisation of faecal output and definition of diarrhoea impede the comparison of research studies and prevent standardised assessment of therapeutic interventions in clinical practice. The colonic microflora may protect the patient against diarrhoea by preventing enteropathogenic infection and by producing SCFA that stimulate colonic water absorption. However, studies in healthy volunteers suggest that the composition of the enteral formula may have a negative impact on the microflora and SCFA concentrations. The addition of fructo-oligosaccharides to the enteral formula may partially prevent negative alterations to the microflora, although conclusive data from studies in patients are not yet available. Modification of the microflora with probiotics and prebiotics may hold potential in prophylaxis against diarrhoea during enteral tube feeding.

Indian Asians living in the UK have a 50% higher CHD mortality rate compared with the indigenous Caucasian population, which cannot be attributed to traditional risk factors. Instead, features of the metabolic syndrome, including raised plasma triacylglycerol, reduced HDL-cholesterol (HDL-C) and an increased proportion of small dense LDL particles, together with insulin resistance and central obesity, are prevalent among this population. The present review examines evidence to support the hypothesis that an imbalance in dietary PUFA intake, specifically a higher intake of n-6 PUFA in combination with the lower intake of the long-chain (LC) n-3 PUFA, plays an important role in the prevalence of the metabolic syndrome observed in Indian Asians. Data are presented to illustrate the impact of manipulation of the background n-6 PUFA intake (moderate or high n-6 PUFA) and the subsequent response to supplementation with LC n-3 PUFA on blood lipids and insulin action in a group of Indian Asian volunteers. The results demonstrate that supplementation with LC n-3 PUFA had no impact on insulin action in those subjects consuming either the moderate- or high-n-6 PUFA diet. In the postprandial phase reductions in plasma triacylglycerol concentrations were greater in those consuming the high-n-6 PUFA background diet subsequent to fish oil supplementation. The present study concludes that, contrary to the central hypothesis, the prevalence of metabolic abnormalities in Indian Asians compared with Caucasians may not be attributable to differences in intakes of n-6 and n-3 PUFA.

The somatotrophic axis is the main endocrine system regulating postnatal growth; however, prenatal growth is independent of growth hormone (GH). Fetal development relies on the coordinated actions of a range of hormones, including insulin-like growth factors (IGF), and prolactin (PRL), in the control of differentiation, growth and maturation. In the sheep the abundance peaks for liver IGF-II and PRL receptors occur during late gestation while that for IGF-I receptor occurs at birth. All receptors, with the exception of GH receptor subsequently decrease by age 6 months. It has been proposed that maternal undernutrition during gestation regulates the maturation of the fetal hypothalmic–pituitary–adrenal axis and endocrine sensitivity. Critically, the timing of the nutritional insult may affect the magnitude of reprogramming. Maternal malnutrition during early to mid-gestation (3·2–3·8 MJ/d (60% total metabolisable energy requirements) v. 8·7–9·9 MJ/d (150% total metabolisable energy requirements) between 28 and 80 d of gestation) had no effect on body or liver weight. Nutrient-restricted (NR) fetuses sampled at 80 d (mid-gestation) showed up-regulation of hepatic PRL receptor, but following refeeding the normal gestational rise in PRL and GH receptors did not occur. Hepatic IGF-II receptor was down regulated in NR fetuses at both mid- and late gestation. Conversely, 6-month-old offspring showed no difference in the abundance of either GH receptor or PRL receptor, while IGF-II mRNA was increased. Offspring of ewes malnourished during late gestation (9·1 MJ/d (60% total metabolisable energy requirements) v. 12·7 MJ/d (100% total metabolisable energy requirements) from 110 d of gestation to term) showed reduced abundance of hepatic GH and PRL receptor mRNA. In conclusion, maternal undernutrition during the various stages of gestation reprogrammed the PRL–GH–IGF axis. Nutritional regulation of cytokine receptors may contribute to altered liver function following the onset of GH-dependent growth, which may be important in regulating endocrine adaptations during subsequent periods of nutritional deprivation.

Although dietary factors have been extensively studied in many chronic diseases, the role of diet in the epidemiology of rheumatoid arthritis (RA) has received little attention. Fruit and vegetables and dietary antioxidants are thought to play a protective role in the pathogenesis of CVD and some cancers, but few studies have investigated these dietary components in the aetiology of RA. Fish oil supplementation has consistently been shown to have a beneficial effect on the symptoms of established RA, but it is not known whether the PUFA present in fish oils can reduce the risk of developing the disease. There is evidence that RA is less severe in the southern Mediterranean countries, such as Italy and Greece, where oil-rich fish, fruit, vegetables and olive oil are consumed in greater amounts than in many other countries. Overall, the evidence for a role of diet in the aetiology of RA is limited to a small number of observational studies of very different designs. Recently, it was demonstrated that lower intakes of fruit and vegetables and dietary vitamin C are associated with an increased risk of developing inflammatory polyarthritis in a free-living population in Norfolk, UK. These findings provide further evidence for a role of diet in the development of inflammatory arthritis, although the mechanisms involved are uncertain.

Nutrigenetics and nutrigenomics are promising multidisciplinary fields that focus on studying the interactions between nutritional factors, genetic factors and health outcomes. Their goal is to achieve more efficient individual dietary intervention strategies aimed at preventing disease, improving quality of life and achieving healthy aging. Our studies, and those of many other investigators, using population-based and intervention studies have found evidence for interactions between dietary factors, genetic variants and biochemical markers of CVD. Now, the characterization of individuals who may respond better to one type of dietary recommendation than another can be begun. Thus, a low-fat low-cholesterol strategy may be particularly efficacious in lowering the plasma cholesterol levels of those subjects carrying the apoE4 allele at the APOE gene. HDL-cholesterol (HDL-C) levels are also modulated by dietary, behavioural and genetic factors. It has been reported that the effect of PUFA intake on HDL-C concentrations is modulated by an APOA1 genetic polymorphism. Thus, subjects carrying the A allele at the –75 G/A polymorphism show an increase in HDL-C with increased intakes of PUFA, whereas those homozygotes for the more common G allele have the expected lowering of HDL-C levels with increased intake of PUFA. Variability at the hepatic lipase gene is also associated with interactions between intake of fat and HDL-C concentrations that could shed some light on the different abilities of certain ethnic groups to adapt to new nutritional environments. This knowledge should lead to successful dietary recommendations partly based on genetic factors that may help to reduce cardiovascular risk more efficiently than the current universal recommendations.

Differences in whole-body lipid metabolism between men and women are indicated by lower-body fat accumulation in women but more marked accumulation of fat in the intra-abdominal visceral fat depots of men. Circulating blood lipid concentrations also show gender-related differences. These differences are most marked in premenopausal women, in whom total cholesterol, LDL-cholesterol and triacylglycerol concentrations are lower and HDL-cholesterol concentration is higher than in men. Tendency to accumulate body fat in intra-abdominal fat stores is linked to increased risk of CVD, metabolic syndrome, diabetes and other insulin-resistant states. Differential regional regulation of adipose tissue lipolysis and lipogenesis must underlie gender-related differences in the tendency to accumulate fat in specific fat depots. However, empirical data to support current hypotheses remain limited at the present time because of the demanding and specialist nature of the methods used to study adipose tissue metabolism in human subjects. In vitro and in vivo data show greater lipolytic sensitivity of abdominal subcutaneous fat and lesser lipolytic sensitivity of femoral and gluteal subcutaneous fat in women than in men. These differences appear to be due to fewer inhibitory α adrenergic receptors in abdominal regions and greater α adrenergic receptors in gluteal and femoral regions in women than in men. There do not appear to be major gender-related differences in rates of fatty acid uptake (lipogenesis) in different subcutaneous adipose tissue regions. In visceral fat rates of both lipolysis and lipogenesis appear to be greater in men than in women; higher rates of lipolysis may be due to fewer α adrenergic receptors in this fat depot in men. Fatty acid uptake into this depot in the postprandial period is approximately 7-fold higher in men than in women. Triacylglycerol concentrations appear to be a stronger cardiovascular risk factor in women than in men, with particular implications for cardiovascular risk in diabetic women. The increased triacylglycerol concentrations observed in women taking hormone-replacement therapy (HRT) may explain the paradoxical findings of increased rates of CVD in women taking HRT that have been reported from recent primary and secondary prevention trials of HRT.

Issues arising in connection with genes and nutrition policy include both nutrigenomics and nutrigenetics. Nutrigenomics considers the relationship between specifc nutrients or diet and gene expression and, it is envisaged, will facilitate prevention of diet-related common diseases. Nutrigenetics is concerned with the effects of individual genetic variation (single nucleotide polymorphisms) on response to diet, and in the longer term may lead to personalised dietary recommendations. It is important also to consider the surrounding context of other issues such as novel and functional foods in so far as they are related to genetic modification. Ethical issues fall into a number of categories: (1) why nutrigenomics? Will it have important public health benefits? (2) questions about research, e.g. concerning the acquisition of information about individual genetic variation; (3) questions about who has access to this information, and its possible misuse; (4) the applications of this information in terms of public health policy, and the negotiation of the potential tension between the interests of the individual in relation to, for example, prevention of conditions such as obesity and allergy; (5) the appropriate ethical approach to the issues, e.g. the moral difference, if any, between therapy and enhancement in relation to individualised diets; whether the 'technological fix' is always appropriate, especially in the wider context of the purported lack of public confidence in science, which has special resonance in the sphere of nutrition.

Is the information currently available to adjust nutritional recommendations and develop individualized nutrition? No. There is not even the information needed for setting dietary recommendations with confidence now at the group level. Will it be available soon? The answer to this question depends on the drive and will of the nutritional community, the success in recruiting funding to the area, the education of nutritionists and the spawning of great ideas and approaches. The emerging tools of genomics, proteomics and metabolomics are enabling the in-depth study of relationships between diet, genetics and metabolism. The advent of technologies can be compared with the discovery of the microscope and the new dimensions of scientific visualization enabled by that discovery. Nutritionists stand at the crest of new waves of data that can be generated, and new methods for their digestion will be required. To date, the study of dietary requirements has been based largely on a black box approach. Subjects are supplemented or depleted and clinical outcomes are observed. Few recommendations are based on metabolic outcomes. Metabolomics and nutrigenomics promise tools with which recommendations can be refined to meet individual requirements and the potential of individualized nutrition can be explored. As yet, these tools are not being widely applied in nutritional research and are rarely being applied by nutritionists. The result is often interesting research that is frequently nutritionally flawed, resulting in inappropriate conclusions. Nutritional education is needed to put nutritionists at the forefront of the development of applications for these technologies, creating a generation of nutrigenomicists. A new generation of nutritionists should be working interdisciplinarily with geneticists, molecular biologists and bioinformaticians in the development of research strategies. The present paper reviews the current status of nutrigenomic research, the current controversies and limitations, and developments needed to advance nutrigenomics and explore fully the promise of individualized nutritional recommendations.

The unravelling of the human genome has the potential to radically extend many of the strategies used in public health nutrition to improve health and to increase food availability, accessibility and utilization. The present paper divides nutrigenomics into two broad but differing areas in asking about possible public health applications: (1) the increasing mismatch between population growth and global food security, on top of the already approximately 800 million of the world population who are food insecure; (2) possible responses to the rising prevalence of non-communicable diseases as the Western diet becomes increasingly inappropriate to the needs of those consuming it. It is clear that complex interactions of multiple polymorphisms play a role in how individuals and sub-populations respond to dietary interventions. All these applications present public health and ethical challenges, particularly in ensuring that any benefits that do come from nutrigenomics are not restricted to the wealthy minority of only the affluent nations. The present paper concludes that the public health applications of nutrigenomics are probably at least a decade away, especially for developing countries. Clinical applications are likely to be more immediate, probably resulting in ‘designer diets’ for individuals with particular polymorphisms, but unless governments take on the role of ensuring some extent of equity in access, any benefits are most likely to go to those who can afford the screening, tests and treatment. At the same time, greatly increased international efforts are needed towards the continuing, and in some cases worsening, global malnutrition, as genetic manipulation of crops is unlikely to provide more than part of the solution.