In my exploration of nutrigenomics, I came across this interesting recent study  that looked at the interaction between a particular gene polymorphism (rs4977574 - available on 23andMe) and cardiovascular disease (CVD), as mediated by consumption of either vegetables or wine.
The researchers followed 24,000 people for 15 years, during which time about 3000 of them developed cardiovascular disease. So it was a pretty big cohort, with lots of people experiencing the outcome in question - cardiovascular disease.
Across the entire population, eating more veggies and drinking more wine resulted in less CVD - not too surprising given previous research on the health benefits of these foods.
Things got more interesting when the researchers looked at polymorphisms of SNP rs4977574. Having one or two of the risk alleles (G worse than A) for this SNP on chromosome 9 has been previously shown to be associated with an increased risk of CVD . For example, study  found for every G allele one carries, one has about a 13% increased risk of CVD. Study  was similar - in the 20-25% of the population that carry two G alleles for this SNP, risk of CVD was increased 30-40% relative to people with AA for rs4977574. This new study  found something very similar - for each additional G allele at rs4977574, people were 16% more likely to develop CVD. And these polymorphisms are quite common, ~30% of the study population were AA for rs4977574, 50% were AG, and 20% were GG.
But things got really interesting when they looked at how vegetable and wine consumption influenced with the link between this polymorphism and CVD.
For people with two 'normal' alleles for rs4977574 (AA), increasing vegetable intake was associated with lower CVD, just like in the population as a whole - no surprise. But for people with either one or two G's for rs4977574, higher vegetable intake didn't help! In other words, compared with high consumers of vegetables who carried two A's for rs4977574 (the reference group), people with AG for rs4977574 were 20-30% more likely to develop CVD, and people with GG for rs4977574 were 40-50% more likely to develop CVD, regardless of how many vegetables they ate!
The opposite was true for wine. Wine didn't help reduce risk of CVD in AA carriers for rs4977574, but it did reduce risk in AG and GG carriers, by ~30% and ~40% respectively! In fact, drinking wine appeared to nearly entirely counteract the baseline increased risk of CVD in AG and GG carriers, relative to AA carriers.
Here is the relevant table of results from the paper for those interested in the precise details:
In summary, this study suggests that if you have one or (especially) two G alleles for rs4977574, you are at higher risk for cardiovascular disease, and that consuming wine, but not vegetables, can help lower your risk.
FYI, 23andMe shows I've got one G allele for rs4977574 - which is a bummer since I love veggies and don't drink alcohol. :(
Of course its only one study, and one gene locus, so the results should be taken with a grain of salt. I don't plan to eat fewer veggies or take up drinking as a result of this study, particularly since alcoholism runs in my family. I figure my good cholesterol numbers and healthy diet/lifestyle make it unlikely I'll die of CVD anyway.
But it seems like another interesting example how genes and diet/lifestyle can interact to influence health in significant and sometimes surprising ways.
 BMC Med Genet. 2014 Dec 31;15(1):1220. [Epub ahead of print]
The chromosome 9p21 variant interacts with vegetable and wine intake to influence the risk of cardiovascular disease: a population based cohort study.
Hindy G, Ericson U, Hamrefors V, Drake I, Wirfält E, Melander O, Orho-Melander M.
Full Text: http://www.ncbi.nlm.nih.gov/pmc/articles/PMC4331503/pdf/12881_2014_Article_138.pdf
AbstractBackgroundChromosome 9p21 variants are associated with cardiovascular disease (CVD) but not with any of its known risk markers. However, recent studies have suggested that the risk associated with 9p21 variation is modified by a prudent dietary pattern and smoking. We tested if the increased risk of CVD by the 9p21 single nucleotide polymorphism rs4977574 is modified by intakes of vegetables, fruits, alcohol, or wine, and if rs4977574 interacts with environmental factors on known CVD risk markers.MethodsMultivariable Cox regression analyses were performed in 23,949 individuals from the population-based prospective Malmö Diet and Cancer Study (MDCS), of whom 3,164 developed CVD during 15 years of follow-up. The rs4977574 variant (major allele: A; minor allele: G) was genotyped using TaqMan® Assay Design probes. Dietary data were collected at baseline using a modified diet history method. Cross-sectional analyses were performed in 4,828 MDCS participants with fasting blood levels of circulating risk factors measured at baseline.ResultsEach rs4977574 G allele was associated with a 16% increased incidence of CVD (95% confidence interval (CI), 1.10¿1.22). Higher vegetable intake (hazard ratio (HR), 0.95 [CI: 0.91¿0.996]), wine intake (HR, 0.91 [CI: 0.86¿0.96]), and total alcohol consumption (HR, 0.92 [CI: 0.86¿0.98]) were associated with lower CVD incidence. The increased CVD incidence by the G allele was restricted to individuals with medium or high vegetable intake (Pinteraction¿=¿0.043), and to non- and low consumers of wine (Pinteraction¿=¿0.029). Although rs4977574 did not associate with any known risk markers, stratification by vegetable intake and smoking suggested an interaction with rs4977574 on glycated hemoglobin and high-density lipoprotein cholesterol (Pinteraction¿=¿0.015 and 0.049, respectively).ConclusionsOur results indicate that rs4977574 interacts with vegetable and wine intake to affect the incidence of CVD, and suggest that an interaction may exist between environmental risk factors and rs4977574 on known risk markers of CVD.
 Science. 2007 Jun 8;316(5830):1488-91. Epub 2007 May 3.
A common allele on chromosome 9 associated with coronary heart disease.
McPherson R1, Pertsemlidis A, Kavaslar N, Stewart A, Roberts R, Cox DR, Hinds DA, Pennacchio LA, Tybjaerg-Hansen A, Folsom AR, Boerwinkle E, Hobbs HH, Cohen JC.
AbstractCoronary heart disease (CHD) is a major cause of death in Western countries. We used genome-wide association scanning to identify a 58-kilobase interval on chromosome 9p21 that was consistently associated with CHD in six independent samples (more than 23,000 participants) from four Caucasian populations. This interval, which is located near the CDKN2A and CDKN2B genes, contains no annotated genes and is not associated with established CHD risk factors such as plasma lipoproteins, hypertension, or diabetes. Homozygotes for the risk allele make up 20 to 25% of Caucasians and have a approximately 30 to 40% increased risk of CHD.
 J Intern Med. 2013 Sep;274(3):233-40. doi: 10.1111/joim.12063. Epub 2013 Mar 25.
Chromosome 9p21 genetic variation explains 13% of cardiovascular disease incidence but does not improve risk prediction.
Gränsbo K1, Almgren P, Sjögren M, Smith JG, Engström G, Hedblad B, Melander O.
AbstractOBJECTIVES:To evaluate the proportion of cardiovascular disease (CVD) incidence that is explained by genetic variation at chromosome 9p21 and to test whether such variation adds incremental information with regard to CVD prediction, beyond traditional risk factors. DESIGN, SETTING AND PARTICIPANTS:rs4977574 on chromosome 9p21 was genotyped in 24 777 subjects from the Malmö Diet and Cancer study who were free from CVD prior to the baseline examination. Association between genotype and incident CVD (n = 2668) during a median follow-up of 11.7 years was evaluated in multivariate Cox proportional hazard models. Analyses were performed in quartiles of baseline age, and linear trends in effect size across age groups were estimated in logistic regression models. RESULTS:In additive models, chromosome 9p21 significantly predicted CVD in the entire population (hazard ratio 1.17 per G allele, 95% confidence interval 1.11-1.23, P < 0.001). Effect estimates increased from the highest (Q4) to the lowest quartile (Q1) of baseline age, but this trend was not significant. The overall population attributable risk conferred by chromosome 9p21 in fully adjusted models was 13%, ranging from 17% in Q1 to 11% in Q4. Addition of chromosome 9p21 to traditional risk factors only marginally improved predictive accuracy. CONCLUSION:The high population attributable risk conferred by chromosome 9p21 suggests that future interventions interfering with downstream mechanisms of the genetic variation may affect CVD incidence over a broad range of ages. However, variation of chromosome 9p21 alone does not add clinically meaningful information in terms of CVD prediction beyond traditional risk factors at any age.