Genome-wide analysis of polymorphism × sodium interaction effect on blood pressure identifies a novel 3′-BCL11B gene desert locus

Excessive sodium intake is a global risk factor for hypertension. Sodium effects on blood pressure vary from person to person; hence, high-risk group targeting based on personal genetic information can play a complementary role to ongoing population preventive approaches to reduce sodium consumption. To identify genetic factors that modulate sodium effects on blood pressure, we conducted a population-based genome-wide interaction analysis in 8,768 Japanese subjects, which was >3 times larger than a similar previous study. We tested 7,135,436 polymorphisms in the discovery cohort, and loci that met suggestive significance were further examined in an independent replication cohort. We found that an interaction between a novel 3′-BCL11B gene desert locus and daily sodium consumption was significantly associated with systolic blood pressure in both discovery and replication cohorts under the recessive model. Further statistical analysis of rs8022678, the sentinel variant of the 3′-BCL11B gene desert locus, showed that differences in mean systolic blood pressure between high and low sodium consumption subgroups were 5.9 mm Hg (P = 8.8 × 10−12) in rs8022678 A carriers and −0.3 mm Hg (P = 0.27) in rs8022678 A non-carriers, suggesting that the rs8022678 genotype can classify persons into sodium-sensitive (A carriers) and sodium-insensitive (A non-carriers) subgroups. Our results implied that rs8022678 A carriers may receive a greater benefit from sodium-lowering interventions than non-carriers.

. Quantile-quantile plots of genome-wide interaction analyses for DBP after applying genomic control correction.
The x-axis indicates the expected -log10 P-values under the null hypothesis. The y-axis shows the observed -log10 P-values calculated from genome-wide interaction analyses.
The black line represents y = x, which corresponds to the null hypothesis. The gray shaded area shows 95% confidence interval of the null hypothesis. The genomic inflation factor (lambda) is the median of the observed test statistics divided by the median of the expected test statistics. Figure S6. Genome-wide interaction signals for DBP. The x-axis represents chromosomal positions and the y-axis represents -log10 P-values of interaction test. The P-values after applying genomic control correction was shown. The grey dotted horizontal lines indicate the suggestive significance level (P = 1 × 10 -5 ). Variants, whose P-value was lower than the suggestive significance, were shown in red, whereas colors for other variants indicate chromosomes.  shows age, sex and BMI-adjusted blood pressure (SBP for top panels and DBP for bottom panels). We analyzed measured SBP and DBP rather than imputed SBP and DBP. Plots for rs8022678 A non-carriers (denoted as 'GG homo') are shown in left-side panels and plots for rs8022678 A carriers (denoted as 'A carriers') are shown in right-side panels. The red lines represent regression lines and the slope parameters are shown at the top of each panel in red. discovery cohort with excluding subjects taking antihypertensive medication.
The x-axis indicates daily sodium consumption level. The y-axis shows age, sex and BMI-adjusted blood pressure (SBP for top panels and DBP for bottom panels). Plots for rs8022678 A non-carriers (denoted as 'GG homo') are shown in left-side panels and plots for rs8022678 A carriers (denoted as 'A carriers') are shown in right-side panels.
The red lines represent regression lines and the slope parameters are shown at the top of each panel in red.
replication cohort with excluding subjects taking antihypertensive medication.
The x-axis indicates daily sodium consumption level. The y-axis shows age, sex and BMI-adjusted blood pressure (SBP for top panels and DBP for bottom panels). Plots for rs8022678 A non-carriers (denoted as 'GG homo') are shown in left-side panels and plots for rs8022678 A carriers (denoted as 'A carriers') are shown in right-side panels.
The red lines represent regression lines and the slope parameters are shown at the top of each panel in red.
Figure S13. Sodium effect on blood pressure stratified by rs8022678 genotype in discovery cohort with excluding hypertensive subjects. The x-axis indicates daily sodium consumption level. The y-axis shows age, sex and BMI-adjusted blood pressure (SBP for top panels and DBP for bottom panels). Plots for rs8022678 A non-carriers (denoted as 'GG homo') are shown in left-side panels and plots for rs8022678 A carriers (denoted as 'A carriers') are shown in right-side panels. The red lines represent regression lines and the slope parameters are shown at the top of each panel in red.
Figure S14. Sodium effect on blood pressure stratified by rs8022678 genotype in replication cohort with excluding hypertensive subjects. The x-axis indicates daily sodium consumption level. The y-axis shows age, sex and BMI-adjusted blood pressure (SBP for top panels and DBP for bottom panels). Plots for rs8022678 A non-carriers (denoted as 'GG homo') are shown in left-side panels and plots for rs8022678 A carriers (denoted as 'A carriers') are shown in right-side panels. The red lines represent regression lines and the slope parameters are shown at the top of each panel in red.               We analyzed measured SBP and DBP rather than imputed SBP and DBP. Note that measured and We analyzed measured SBP and DBP rather than imputed SBP and DBP. Note that measured and imputed BPs were exactly same for all subjects included in the analysis for this Table. This association analysis was based on the combined dataset of the discovery and replication cohorts and was performed with a logistic regression model with adjustment for age, sex and cohort. The subgroups of rs8022678 A non-carriers was considered as reference.