Rare pathogenic variants in G-protein coupled receptor genes involved in gut-to-host communication are associated with cardiovascular disease risk

• Genomics
• Rare variants
• G-protein couple receptors
• Short-chain fatty acids
• Hypertension
• Major adverse cardiac events

Gut microbial metabolites called short-chain fatty acids (SCFAs) confer protection against cardiovascular disease (CVD) and its primary risk factor, hypertension.¹ Fermentable dietary fibers reach the large intestine undigested, where they are metabolized by fermentative microbes, releasing SCFAs as by-products. These metabolites lower blood pressure in humans and protect against cardiac hypertrophy and fibrosis in mice.^{1, 2} Proposed mechanisms include anti-inflammatory signalling activated by the binding of SCFAs to G-protein-coupled receptors (GPCR, particularly FFAR3/GPR41, FFAR2/GPR43, and HCAR2/GPR109A), as suggested by previous mouse studies using knockout strains.^{1, 3} Here, we conducted a large-scale population-based study aimed at determining if rare pathogenic variants (RPVs) affecting SCFA-sensing GPCR genes increase the risk in humans of hypertension and major adverse cardiac events (MACE), including acute coronary syndrome, heart failure, and ischemic stroke.

We used the population healthcare, lifestyle, and whole-exome sequencing (WES) data from the UK Biobank, available for 393,649 participants of white-European genetic ancestry (Figure A). Rare (minor allele frequency <1%) variants in the GPCR genes FFAR3, FFAR2 and HCAR2 and their genotypes were extracted from quality-controlled WES data, based on genomic coordinates. We identified a subset of 158 RPVs with predicted high-impact functional consequences or missense-pathogenic likelihood scores (Figure A). Annotations were based on the Ensembl Variant Effect Predictor tool (http://www.ensembl.org/info/docs/tools/vep), as performed previously.⁴ Cases and controls for hypertension and MACE subtypes were identified based on diagnoses from hospital admissions (recorded as ICD-10 – International Classification of Disease 10^th version - codes), self-reported medical histories from health-related questionnaires, surgical records and death registries (Figure A). The characteristics of the study cohort are detailed in Figure B. Additional information is provided in the online Supplementary Material.

• Download figure
• Open in new tab

Figure Rare pathogenic variants (RPVs) in short-chain fatty acid (SCFA)-sensing G-protein coupled receptors (GPCR) genes are associated with poor cardiovascular outcomes.

(A) Study design flowchart, including phenotype and RPVs definitions. RPVs were identified based on two classifications: (1) ‘high-impact functional consequence’, comprising stop gained/lost, frameshift, and splicing alterations based on Ensembl Variant Effect Predictor annotations; (2) ‘pathogenic missense variants’, comprising missense variants with pathogenic likelihood scores >0.564 from AlphaMissense, >0.025 from Mendelian Clinically Applicable Pathogenicity, or >20 from Combined Annotation Dependent Depletion. (B) Study cohort demographic and clinical characteristics, stratified as cases and controls for each phenotype of interest, encompassing hypertension (HTN), major adverse cardiac events (MACE) and MACE subtypes (acute coronary syndrome (ACS), heart failure (HF), and ischemic stroke). (C) Significant associations between GPCR RPVs and phenotypes’ risk. (D) Increased cardiomyopathy prevalence in hypertensive cases carrying GPCR RPVs. (E) Increased hypertension prevalence in GPCR RPV carriers among adequate-fiber diet consumers. Statistics: Multiple comparisons were corrected by false discovery rate using the Benjamini-Hochberg method (threshold q<0.05). Panels C-E: Odds-ratio (OR) and P values are derived from multivariable logistic regression analyses, adjusted for: age and sex (model 1); age, sex, and BMI (model 2); age, sex, BMI, and standard modifiable cardiovascular risk factors (SMurFs), including smoking (current/past), hypertension, diabetes, and hypercholesterolemia (model 3; only applied for MACE and subtypes). Genetic ancestry (10-top genetic principal components) was also included as covariate when testing the association between GPCR RPVs and phenotypes’ risk (Panel C). All OR forest plots are in a log₁₀ scale.

Multiple and single GPCR gene-based analyses were conducted to test whether the presence of RPVs that disrupt gene function is associated with increased phenotype risk, using multivariable-adjusted regression models. Age- and sex-adjusted models, either excluding or including BMI as a covariate (model 1 and model 2, respectively), revealed the prevalence of RPV carriers affecting any GPCR gene of interest was significantly higher in patients with hypertension and MACE compared to controls (Figure C). Given the occurrence of MACE is recognized to be associated with standard modifiable cardiovascular risk factors (SMuRFs) – including hypertension, diabetes, hypercholesterolemia, and smoking – we further adjusted MACE analyses for these additional covariates (model 3). SMuRFs-adjusted analyses confirmed a significantly increased risk of MACE among GPCR RPV carriers, independent of these additional risk factors (Figure C). Finally, when GPCR genes were tested individually, carrying RPVs in the HCAR2 (GPR109A) gene was associated with a markedly increased risk of heart failure, while carrying RPVs in the FFAR2 (GPR43) gene was associated with an increased risk of hypertension, with associations remaining significant after correction for multiple comparisons (accounting for 3 GPCR genes and distinct models) (Figure C).

Next, to identify other diseases and conditions that are more common among hypertension and MACE patients carrying GPCR RPVs (compared to non-carriers) and could contribute to this association, we performed a comorbidity analysis by retrieving hospital inpatient records for 1,259 ICD-10 codes (Chapters I–XIV) across 39,563,559 data-points (243 data-fields covering 162,813 cases for hypertension/MACE). After filtering for ICD-10 diseases with a prevalence greater than 1% and adjusting for multiple comparisons (297 ICD-10 diseases and two models), we found cardiomyopathy (ICD-10: I42) was more than twice as common in hypertensive patients carrying GPCR RPVs (Figure D). Consistently, cardiac hypertrophy and fibrosis have been observed in experimental mouse models that lack SCFA-sensing GPCR genes.^{1, 3}

Since SCFAs are produced primarily by the gut microbiota as a response to dietary fiber intake, to further understand the complex relationship between fiber, gut-to-host communication, and cardiovascular health, we investigated whether carrying GPCR RPVs increases the risk of hypertension and MACE among individuals consuming an adequate-fiber diet. This would support the hypothesis that fiber’s beneficial cardiovascular effects depend upon GPCR signalling. To estimate daily fiber intake, we studied a subset of UK Biobank participants with dietary data, assessed by a 24-hour dietary recall of food intake on the previous day in five different instances (n=168,142 with dietary data available for at least one instance). Based on nutritional recommendations,⁵ an adequate-fiber diet was defined as an average intake of ≥25 g/day for women and ≥30 g/day for men, leaving 13,858 participants for subsequent analyses. Consistent with our hypothesis, the prevalence of hypertension was significantly higher among GPCR RPVs carriers (Figure E), suggesting that cardiovascular benefits of fiber intake are diminished in the presence of genetic impairments in SCFA-related GPCR signalling. No significant associations were observed for MACE, likely due to the smaller sample size.

Overall, our findings show that lack of signalling via SCFA-sensing GPCRs, even when fiber intake is adequate, is associated with human hypertension and MACE in a large population-based cohort from the UK. This underscores the potential of targeting gut-to-host mechanisms through SCFA-sensing GPCRs as a therapeutic strategy for hypertension and CVD.