A recent study published in Nature Medicine demonstrated genetic heterogeneity of blood pressure (BP) effects between sexes.
Study: Sex-specific genetic architecture of blood pressure. Image Credit: Chompoo Suriyo/Shutterstock.com
Background
Environmental exposures and interactions influence BP traits with a complex and well-defined genetic architecture, with about 30% to 50% heritability.
Hypertension is highly heritable and a significant risk factor for cardiovascular diseases (CVDs). Hypertension and multiple CVDs show sex disparities in clinical presentation, outcomes, and prevalence.
Identifying targets for BP regulation to reduce CVD risks is essential to enhancing strategies for cardiovascular health improvements. Although sex-based effects on gene expression are defined, their relationship with BP remains unknown.
Thus, defining genetic effects by sex can improve the understanding of BP/hypertension genes and their regulation and clinical relevance.
The study and findings
In the present study, researchers performed combined-sex and sex-stratified (female- and male-only) genome-wide association studies (GWAS) of BP traits using the United Kingdom (UK) Biobank (UKB) resource.
Equal female and male samples were created to define sex differences in genetic associations. Specifically, females from the original UKB sample were selected based on ancestry and age to match males at a 1:1 ratio.
Two replication cohorts were included – the Genetic Epidemiology Research on Adult Health and Aging (GERA) cohort and the Michigan Genomics Initiative (MGI) resource.
Sex-stratified and combined-sex GWAS of BP traits revealed 510, 310, and 555 loci associated with pulse pressure (PP), systolic BP (SBP), and diastolic BP (DBP) traits, respectively.
Twenty-nine novel trait–single nucleotide polymorphism (SNP) associations were uncovered. Despite equal sample sizes, 1.8-fold more loci were identified in female-only analyses than in male-only analyses.
Overall, 142 male-specific and 412 female-specific trait–SNP associations were identified. These sex-specific analyses revealed 2.9-fold more female-specific loci than male-specific loci.
Next, sexually dimorphic effects (SDEs) were defined (as the difference in directions and effect sizes of BP associations at any locus by sex) and compared with sex-specific associations.
The researchers observed more directional differences for loci discovered in the SDE analysis than for sex-specific loci associated with only one sex in sex-stratified analyses.
Further, in replication analyses, BP traits in GERA and MGI cohorts were analyzed for associations identified in UKB analyses, and a meta-analysis of GERA and MGI association results was conducted.
The correlation of effect sizes between discovery and replication analyses was consistently high. The meta-analysis showed that 54 male-only, 94 female-only, and 278 combined-sex BP loci met the Bonferroni-corrected significance threshold.
Further, top loci from GWAS or SDEs were queried in the genotype-tissue expression (GTEx) transcriptome data from three arterial tissues of combined-sex samples to map their cis-expression quantitative trait loci (eQTL)-associated genes.
In addition, eQTL-associated genes were prioritized by a colocalization analysis. This showed that 24 and 62 genes were colocalized with male- and female-specific BP-associated loci, respectively.
Notably, no genes were simultaneously detected as female- or male-specific for any trait, and only 8.7% of genes were detected in non-sex- and sex-specific categories. This implied a low likelihood of genes colocalizing with non-sex- and sex-specific loci.
Also, colocalization was more likely between sex-specific BP-linked loci and their arterial eQTL genes than non-sex-specific loci. Next, genes were selected as strong sex-biased candidates for nuclear receptor (NR) motif enrichment analyses.
The ‘RAR: RXR(NR), DR0’ motif was enriched in female-biased arterial genes. ‘TR4(NR), DR1’ and ‘Erra(NR)’ motifs associated with testicular NR and estrogen-related receptors were enriched in male-biased genes.
Non-sex specific BP loci-linked genes were enriched for ‘Nur77(NR)’ and ‘HNF4a(NR), DR1’ motifs associated with macrophage mitochondrial metabolism and hepatocyte nuclear factor regulation, respectively.
Next, transcription factor binding site enrichment was evaluated. Estrogen receptor 1 (ESR1) showed the highest differential TFBS enrichment.
Furthermore, polygenic scores (PGSs) based on top loci identified in sex-stratified or combined-sex GWAS were tested for associations with hypertension in the MGI resource.
Prediction of hypertension using PGSs in combined-sex or sex-stratified analyses showed that the area under the receiver operating characteristic curve was 0.8, with better performance in female-only analyses.
In addition, PGSDBP and PGSSBP were significantly associated with pre-eclampsia and fibromuscular dysplasia. Next, the PGSs from combined-sex or sex-stratified BP GWAS were assessed for associations with CVDs that have clinically significant sex differences.
The PGSs for PP, SBP, and DBP traits were significantly associated with most CVDs regardless of whether PGSs were generated from combined-sex or sex-stratified GWAS results.
Coronary artery dissection exhibited the most significant associations with PGSs for DBP and SBP, while aortic dimension measurements had the most significant associations with PGSs for PP.
Conclusions
Taken together, the researchers performed a comprehensive GWAS for BP. The findings indicate genetic heterogeneity of BP effects between sexes; the genetic effects were more pronounced in females, affecting genome regulation and CVD risks.
Numerous loci demonstrated sex-specific associations with CVDs through genetic BP links, partly involving sex hormone transcription regulation. Overall, the results enhance understanding of sex disparities in BP and CVDs.