Dietary fat intake is correlated with increased insulin resistance (IR). However, it is unknown whether gene–diet interaction modulates the association. This study estimated heritability of IR measures and the related genetic correlations with fat intake, and tested whether dietary fat intake modifies the genetic influence on type 2 diabetes (T2D)-related traits in Chinese child twins. We included 622 twins aged 7–15 years (n 311 pairs, 162 monozygotic (MZ), 149 dizygotic (DZ)) from south-eastern China. Dietary factors were measured using FFQ. Structural equation models were fit using Mx statistical package. The intra-class correlation coefficients for all traits related to T2D were higher for MZ twins than for DZ twins. Dietary fat and fasting serum insulin (additive genetic correlation (rA) 0·20; 95 % CI 0·08, 0·43), glucose (rA 0·12; 95 % CI 0·01, 0·40), homoeostasis model of assessment-insulin resistance (Homa-IR) (rA 0·22; 95 % CI 0·10, 0·50) and the quantitative insulin sensitivity check index (Quicki) (rA −0·22; 95 % CI −0·40, 0·04) showed strong genetic correlations. Heritabilities of dietary fat intake, fasting glucose and insulin were estimated to be 52, 70 and 70 %, respectively. More than 70 % of the phenotypic correlations between dietary fat and insulin, glucose, Homa-IR and the Quicki index appeared to be mediated by shared genetic influence. Dietary fat significantly modified additive genetic effects on these quantitative traits associated with T2D. Analysis of Chinese twins yielded high estimates of heritability of dietary fat intake and IR. Genetic factors appear to contribute to a high proportion of the variance for both insulin sensitivity and IR. Dietary fat intake modifies the genetic influence on blood levels of insulin and glucose, Homa-IR and the Quicki index.

Blood lipid response to a given dietary intervention could be determined by the effect of diet, gene variants or gene–diet interactions. The objective of the present study was to investigate whether variants in presumed nutrient-sensitive genes involved in lipid metabolism modified lipid profile after weight loss and in response to a given diet, among overweight European adults participating in the Diet Obesity and Genes study. By multiple linear regressions, 240 SNPs in twenty-four candidate genes were investigated for SNP main and SNP–diet interaction effects on total cholesterol, LDL-cholesterol, HDL-cholesterol and TAG after an 8-week low-energy diet (only main effect), and a 6-month ad libitum weight maintenance diet, with different contents of dietary protein or glycaemic index. After adjusting for multiple testing, a SNP–dietary protein interaction effect on TAG was identified for lipin 1 (LPIN1) rs4315495, with a decrease in TAG of − 0·26 mmol/l per A-allele/protein unit (95 % CI − 0·38, − 0·14, P= 0·000043). In conclusion, we investigated SNP–diet interactions for blood lipid profiles for 240 SNPs in twenty-four candidate genes, selected for their involvement in lipid metabolism pathways, and identified one significant interaction between LPIN1 rs4315495 and dietary protein for TAG concentration.

Nutritional genomics is a new and promising science area which can broadly be defined as the application of high throughput genomics (transcriptomics, proteomics, metabolomics/metabonomics) and functional genomic technologies to the study of nutritional sciences and food technology. First utilised in the food industry by plant biotechnologists to manipulate plant biosynthetic pathways, the use of genomic technologies has now spread within the agriculture sector, unleashing a host of new applications (e.g. approaches for producing novel, non-transgenic plant varietals; identification of genetic markers to guide plant and animal breeding programmes; exploration of diet–gene interactions for enhancing product quality and plant/animal health). Beyond agriculture, genomic technologies are also contributing to the improvement of food processing, food safety and quality assurance as well as the development of functional food products and the evolution of new health management concepts such as ‘personalised nutrition’, an emerging paradigm in which the diet of an individual is customised, based on their own genomic information, to optimise health and prevent disease. In this review the relevance of nutritional genomics to the food industry will be considered and examples given on how this science area is starting to be leveraged for economic benefits and to improve human nutrition and health.

Genetic and nutritional factors interact together and modulate the plasma lipid profile. We identified variations in the gene encoding the liver X receptor α (LXRα) and investigated their effects on the plasma lipoprotein/lipid profile. We also examined whether the association between cholesterol intake and plasma lipid profile was modulated by LXRα variants. The LXRα gene was sequenced in thirty-five French-Canadian men with high plasma total cholesterol (>5·0 mmol/l) and LDL-cholesterol (>3·5 mmol/l) concentrations. Dietary cholesterol was obtained from a food-frequency questionnaire. The LXRα c.-115G>A, c.-840C>A and c.-1830T>C genotypes were determined by direct sequencing in 732 subjects. Molecular screening of the LXRα gene revealed sixteen variants. Genotypes c.-115G>A, c.-840C>A and c.-1830T>C (rare allele frequency of 14·3 %, 14·2 % and 11·0 %, respectively) were analysed further. Plasma total cholesterol concentrations were higher in carriers of the -115A, -840A and -1830C allele, compared with the -115G/G, -840C/C and -1830T/T homozygotes (P ≤ 0·05). In a model including the c.-115G>A polymorphism, cholesterol intake, the interaction term c.-115G>A × cholesterol intake (mg/d) and covariates, LXRα-115G>A explained 1·8 % and 2·1 % of the variance in total cholesterol and LDL-cholesterol concentrations (P = 0·02 and P = 0·01), whereas the interaction term explained 2·9 % (P = 0·002) and 2·8 % (P = 0·005), respectively. When subjects were divided into four groups according to the median of cholesterol (290·8 mg) and -115G>A genotypes, high cholesterol intake was associated with higher cholesterol levels in -115A carriers. Similar results were observed for c.-840C>A and c.-1830T>C. These results suggest that cholesterol intake interacts with LXRα variants to modulate the plasma lipid profile.