In well-characterized adults with pre- or Stage 1 hypertension, studied during controlled dietary intake, we found that genotype at two angiotensinogen SNPs, two β2-adrenergic receptor SNPs, and the Q121E SNP of the kallikrein gene were associated with the extent to which BP fell with reduced sodium intake. We found that the ADBR2_C79G SNP was significantly associated with the extent to which BP fell with DASH dietary pattern, with a qualitatively similar association for the ADBR2_A46G SNP. The difference in effect between favorable and unfavorable genotypes was as much as a two- to threefold difference in the extent of BP-lowering. In both dietary interventions, there was a trend for racial differences in these relationships. Although many of these associations were not statistically significant in a secondary analysis with a fairly conservative adjusted for multiple comparisons, such an adjustment may have led to insufficient power. Given the exploratory nature of this study, statistically significant associations in the unadjusted analysis warrant further consideration.
The particular genes implicated in this study are important in normal and abnormal regulation of BP, so it is interesting that they may also be involved in BP response to environmental (i.e., dietary) factors. For example, several previous studies have demonstrated associations between BP phenotypes and polymorphisms of the angiotensinogen gene. This gene has been associated with elevated levels of circulating angiotensinogen1–3
with the presence of essential hypertension,2–5
with family history of hypertension,8
and possibly with hypertension in pregnancy.6
Thus, genotype at the angiotensinogen locus appears to confer susceptibility to hypertension. In the Trials of Hypertension Prevention, AGT genotype was associated with the impact of reduced sodium intake on the incidence of hypertension in whites.1
Our data extend these observations to identify two polymorphisms that are associated with BP response to reduced sodium intake.
The role of AGT genotype in modulating response to DASH is less clear. In this study, there is no significant association between genotype at two AGT SNPs and BP response to DASH. However, in a previous analysis of DASH participants only, the BP response to the DASH diet trended greater in the AA genotype of the G-6A AGT SNP and least in the GG genotype.10
The lack of consistency between the two analyses raise the possibility that the association may be spurious or complex, but other data suggest that DASH lowers BP through effects on the renin–angiotensin–aldosterone system,21
raising the likelihood that genetic variation influences the effect of DASH on renin– angiotensin–aldosterone system, perhaps at other loci. Future research is needed to clarify this relationship.
-adrenergic receptor (β2-ADR) gene is also highly implicated in normal and abnormal BP regulation. We and others have previously reported a significant association between β2-ADR genotype and the presence of hypertension in unrelated individuals, families, and twins.7
Further, in 109 African-American sib-pairs we found preliminary evidence of linkage of β2-ADR gene and salt sensitivity.9
In this study, the observed association between two β2-ADR SNPs and BP response to reduced sodium intake strongly suggests that this locus modulates dietary sodium sensitivity. These findings are consistent with those of Pojoga et al.
who reported greater salt sensitivity associated with the A allele of A46G and the C allele of C79G.25
In addition, this study confirms the previously reported association between β2-ADR genotype and response to DASH dietary pattern.10
The role of the kallikrein gene in BP regulation is less well-established. Renal kallikrein acts on the substrate kininogen to form lysyl-bradykinin (kallidin), a potent vasodilator that also promotes sodium excretion. Kallikrein excretion is decreased with hypertension26–29
and increases with higher potassium intake.27,30
Thus, the relationship between BP and dietary sodium/potassium ratio may be mediated through effects on the kallikrein system. Our data suggest that the BP response to sodium intake, but not (potassium-rich) DASH is associated with genotype at the kallikrein locus, but the significance of this relationship is unclear.
Although we did not explore the relationship between genotype and biochemical response to diet in this analysis, it is known that both sodium reduction and DASH dietary pattern increase plasma renin activity and serum aldosterone,31
and that genotype can influence the magnitude of this effect.25
Further research is needed to determine the mechanism(s) through which genotype (and presumably gene product) influences BP response to diet.
It is common to consider the extent to which genotype modulates BP response to a particular intervention. This is a useful approach for understanding mechanisms and identifying targets of drug therapy. However, it is also useful to consider the extent to which dietary intervention can mitigate or enhance a genetic predisposition, with the goal of targeting an intervention to those genetically predisposed to respond to it. For example, current guidelines recommend that all individuals with prehypertension or hypertension eat the DASH dietary pattern, reduce sodium intake, and lose weight if overweight or obese.16
Changing behavior is difficult, and it may be that changing several behaviors is more difficult than focusing on a single behavior change. Identifying which interventions are likely to have the greatest effect on BP for a given individual may lead to a specific lifestyle prescription that is more likely to be successfully adopted. In this context, this study lays groundwork for targeting behavioral interventions to genotype, a strategy that may become more feasible and attractive as genetic screening becomes more commonplace.
Our study is consistent with data from other clinical conditions. For example, a genotype of the transcription factor 7-like 2 gene confers excess risk of type 2 diabetes mellitus, but the difference in risk between deleterious genotype and favorable genotype is eliminated by a lifestyle intervention.32
Similarly, a genetic variant of the 5-lipoxygenase gene promoter is associated with atherosclerosis, as indicated by carotid artery intimal–medial thickness, but the expression of excess risk is modulated by dietary intake of both arachidonic and linoleic acid.33
Thus, investigation of gene–environment interactions, when the environment refers to dietary intake, may prove useful in developing individualized, targeted behavioral interventions.
The strengths of this study derive mainly from the study design. Previous studies of gene–diet association generally depend on inferences about dietary intake from food records, 24-h excretion data or response to dietary supplement or infusion of saline. Most data are restricted to the effect of sodium rather than other nutrients or overall dietary pattern, and the populations studied are small and not racially diverse.34
In contrast, our sample size was large, the population diverse and well-characterized, BP response was measured using precise, standardized, unbiased methods, and the interventions were provided in the context of controlled feeding studies. Nutrient intake was controlled and quantified, and confounders were mitigated (e.g., body weight remained stable). A potential limitation of this study is that we focused on several candidate genes rather than screening for all possible polymorphisms associated with BP response to diet. However, although genome-wide association studies can detect genes not previously suspected of having a role in the disease under study, they are highly prone to false positive results.35
Our candidate gene approach increases confidence in the validity of the results. In addition, as so much is known about systems involved in BP regulation, a focus on genes regulating these systems could lead more efficiently to deeper understanding. Nonetheless, to better understand the universe of genes associated with BP response to DASH and sodium reduction, future research should include genome-wide association studies. In addition, our results should be verified in independent populations. Although we focus here on BP effects, the possibility that the same genetic polymorphisms might also affect the relationship between BP and clinical outcomes should also be investigated.
In summary, our study demonstrates a relationship between angiotensinogen, β2-adrenergic receptor and kallikrein geno-type and the effect of sodium intake and the DASH dietary pattern on BP. These findings have implications for understanding the mechanism(s) through which diet affects BP, the heterogeneity of these effects, and the extent to which dietary interventions can modulate genetic predisposition.