Molecular pathways associated with blood pressure and hexadecanedioate levels

Abstract

The dicarboxylic acid hexadecanedioate is associated with increased blood pressure (BP) and mortality in humans and feeding it to rats raises BP. Here we aim to characterise the molecular pathways that influence levels of hexadecanedioate linked to BP regulation, using genetic and transcriptomic studies. The top associations for hexadecanedioate in a genome-wide association scan (GWAS) conducted on 6447 individuals from the TwinsUK and KORA cohorts were tested for association with BP and hypertension in the International Consortium for BP and in a GWAS of BP extremes. Transcriptomic analyses correlating hexadecanedioate with gene expression levels in adipose tissue in 740 TwinsUK participants were further performed. GWAS showed 242 SNPs mapping to two independent loci achieving genome-wide significance. In rs414056 in the SCLO1B1 gene (Beta(SE) = -0.088(0.006)P = 1.65 x 10-51, P < 1 x 10-51), the allele previously associated with increased risk of statin associated myopathy is associated with higher hexadecanedioate levels. However this SNP did not show association with BP or hypertension. The top SNP in the second locus rs6663731 mapped to the intronic region of CYP4Z2P on chromosome 1 (0.045(0.007), P = 5.49x10-11). Hexadecanedioate levels also correlate with adipose tissue gene-expression of the 3 out of 4 CYP4 probes (P<0.05) and of alcohol dehydrogenase probes (Beta(SE) = 0.12(0.02); P = 6.04x10-11). High circulating levels of hexadecanedioate determine a significant effect of alcohol intake on BP (SBP: 1.12(0.34), P = 0.001; DBP: 0.70(0.22), P = 0.002), while no effect is seen in the lower hexadecanedioate level group. In conclusion, levels in fat of ADH1A, ADH1B and CYP4 encoding enzymes in the omega oxidation pathway, are correlated with hexadecanedioate levels. Hexadecanedioate appears to regulate the effect of alcohol on BP.

Fig 1. Flowchart of the study design.

Fig 2. Manhattan plot showing genome-wide P values from association analysis of hexadecanedioate in the TwinsUK-KORA meta-analysis.

The y axis shows the −log10 P values of SNPs, and the x axis shows their chromosomal positions.

Fig 3. Association plot of the genomic region around (i) SLCO1B1, (ii) CYP4B1 and (iii)ADH1B showing both type and imputed SNPs with location of genes and recombination rate.

P values of SNPs are plotted (as −log₁₀ P) against their physical position on the chromosomes (NCBI Build 36). Estimated recombination rates from the HapMap CEU population show the local LD structure. The color of each SNP indicates linkage disequilibrium with the index SNP (rs4149056 or rs6663731 or rs1693457 respectively) based on pairwise r² values from HapMap CEU data.