Sex-specific Genetic Determinants of Right Ventricular Structure and Function

Abstract

Rationale: While sex differences in right heart phenotypes have been observed, the molecular drivers remain unknown.

Objectives: To provide biological insights into sex differences in the structure and function of the right ventricle (RV) using common genetic variation.

Methods: RV phenotypes were obtained from cardiac magnetic resonance imaging in 18,156 women and 16,171 men from the UK Biobank. Observational analyses and sex-stratified genome-wide association studies were performed. Candidate female-specific loci were evaluated against invasively measured cardiac performance in 479 female patients with idiopathic or heritable pulmonary arterial hypertension (PAH), recruited to the UK NIHR BioResource Rare Diseases study.

Measurements and main results: Sex was associated with differences in RV volumes and ejection fraction in models adjusting for left heart counterparts, blood pressure, lung function and sex hormone levels. Six genome-wide significant loci (13%) revealed heterogeneity of allelic effects between women and men, and significant sex-by-genotype interaction. These included two sex-specific candidate loci present in women only: a locus for RV ejection fraction in BMPR1A and a locus for RV end-systolic volume near DMRT2. Epigenetic data in RV tissue indicate that variation at the BMPR1A locus likely alters transcriptional regulation. In female patients with PAH, a variant located in the promoter of BMPR1A was significantly associated with cardiac index (effect size 0.16 l/min/m²), despite similar RV afterload.

Conclusions: BMPR1A has emerged as a biologically plausible candidate gene for female-specific genetic determination of RV function, showing associations with cardiac performance under chronically increased afterload in female patients with PAH.

Keywords: Gender; Genetics; Pulmonary arterial hypertension; Right ventricle; Sex.

Figure 1.

Heterogeneity in allelic effects between women and men. (A and B) Six association pairs were highlighted that showed significant heterogeneity between women and men and significant sex-by-genotype interaction. (C) The effects of these six association pairs were distinguished into four sex-differential effects, in which the combined effect was predominantly driven by one sex, and two sex-specific effects, which were present in only one sex. (D and E) Sex-differential allelic effects are depicted in D and sex-specific effects in E. *Indicates that in men rs3851363 (r² 0.36 with lead variant in women) was used because of signal differences between women and men. RV = right ventricle; RVEDV = RV end-diastolic volume; RVEF = RV ejection fraction; RVESV = RV end-systolic volume; RVSV = RV stroke volume.

Figure 2.

Regional association for sex-specific loci. Regional plots show the genomic positions of variants on the x-axis and their association strength (−log10[P value]) on the y-axis. Correlation coefficients (r²) with the lead variants are indicated by color. (A) Locus for RVEF on chromosome 10 is displayed. (B) Locus for RVESV on chromosome 9. RVEF = right ventricle ejection fraction; RVESV = right ventricle end-systolic volume.

Figure 3.

Epigenetic marks in right ventricular (RV) tissue from adult women. Topologically associated domains (TAD) indicate genomic regions that are in close proximity based on the three-dimensional chromatin structure (Hi-C maps). Genome-wide association study (GWAS) variants are displayed that are in high linkage disequilibrium (LD) with the lead variant (r² > 0.9). Lead variants are highlighted in green. DNase I hypersensitivity sites mark chromatin regions that are assessable for transcription factors. Epigenomic modification indicates areas likely to contain active regulatory regions and promoters. H3K4Me1 is often found in enhancers, H3K4Me3 typically marks promoters and H3K27ac is often marking active regulatory regions. (A) Female-specific locus for RV ejection fraction on chromosome 10 is displayed. Here, a variant rs140745739 (highlighted in yellow) that is in high LD with lead variant (r² = 0.97) resides within the active promotor region of BMPR1A. (B) Female-specific locus for RV end-systolic volume on chromosome 9 is displayed. DMRT2 is the closest gene and shares a TAD with the GWAS variants, but no adjacent active regulatory region is found. BMPR1A = bone morphogenetic protein receptor type 1A; DMRTs = doublesex and mab3-related transcription factor 1 to 3; H3K4Me1 = histone H3 lysine 4 monomethylation; H3K4Me3 = histone H3 lysine 4 trimethylation; H3K27ac = H3 lysine 27 acetylation; LDB3 = LIM Domain Binding 3; SMARCA2 = SWI/SNF related, matrix associated, actin dependent regulator of chromatin, subfamily A, member 2.

Figure 4.

BMPR1A (bone morphogenetic protein receptor type 1A) locus in women. The genotypic effects of the BMPR1A promoter variant (rs140745739, minor allele C) align between RVEF in 18,156 female participants of the UKB and invasively measured cardiac index in 479 female patients with idiopathic or hereditary PAH. Data are presented as mean with SD. PAH = pulmonary arterial hypertension; RVEF = right ventricle ejection fraction; UKB = UK Biobank.