Genetic analyses of the electrocardiographic QT interval and its components identify additional loci and pathways

Abstract

The QT interval is an electrocardiographic measure representing the sum of ventricular depolarization and repolarization, estimated by QRS duration and JT interval, respectively. QT interval abnormalities are associated with potentially fatal ventricular arrhythmia. Using genome-wide multi-ancestry analyses (>250,000 individuals) we identify 177, 156 and 121 independent loci for QT, JT and QRS, respectively, including a male-specific X-chromosome locus. Using gene-based rare-variant methods, we identify associations with Mendelian disease genes. Enrichments are observed in established pathways for QT and JT, and previously unreported genes indicated in insulin-receptor signalling and cardiac energy metabolism. In contrast for QRS, connective tissue components and processes for cell growth and extracellular matrix interactions are significantly enriched. We demonstrate polygenic risk score associations with atrial fibrillation, conduction disease and sudden cardiac death. Prioritization of druggable genes highlight potential therapeutic targets for arrhythmia. Together, these results substantially advance our understanding of the genetic architecture of ventricular depolarization and repolarization.

Fig. 1. Annotation of an example ECG signal.

QRS duration and the JT interval approximate the time periods for ventricular depolarization and repolarization on the surface ECG. The entire segment from onset of the Q wave to end of the T wave is the QT interval.

Fig. 2. Workflow of the genetic analyses performed for QT, JT, and QRS.

Workflow including single variant and gene-based meta-analyses, and downstream bioinformatics. VEP (Variant Effect Predictor), CADD (Combined Annotation Dependent Depletion), eQTL (expression Quantitative Trait Locus), GTEx (Genotype-Tissue Expression project), COLOC (Colocalization), GARFIELD (GWAS Analysis of Regulatory and Functional Information Enrichment with LD correction), DEPICT (Data-driven Expression-Prioritized Integration for Complex Traits, GWAS (Genome-Wide Association Study), EA (European Ancestry), PRS (Polygenic Risk Score), AF (Atrial Fibrillation), CAD (Coronary Artery Disease), CD (Conduction Disease), HF (Heart Failure), NICM (Non-Ischemic Cardiomyopathy), VA (Ventricular Arrhythmia), SCD (Sudden Cardiac Death).

Fig. 3. Circular Manhattan plot for QT, JT, and QRS multi-ancestry meta-analyses.

Circular Manhattan plots for QT (outer, yellow), JT (middle, red), and QRS (inner, blue) multi-ancestry GWAS linear regression meta-analyses. The Y-axis has been restricted to -log10 P-value < 30. Two-sided P-values are reported. A Bonferroni-corrected threshold (<5 × 10⁻⁸) was used to declare significance. Overlapping JT and QRS loci are labeled with the most likely candidate at the locus color coded according to a concordant (green) or discordant (purple) direction of effect at a variant level. Direction of effect was compared by comparing the lead JT variant beta with the corresponding direction of effect of the same variant in the QRS GWAS meta-analysis. This plot was produced using the R package Circlize version 0.4.10. Gu, Z. (2014) circlize implements and enhances circular visualization in R. Bioinformatics. 10.1093/bioinformatics/btu393.

Fig. 4. Comparison of co-localized eQTL signals for QT, JT, and QRS in right atrial appendage and left ventricle tissues.

Colocalization analyses performed using data from GTEx (version 8), using the R package COLOC (methods). A posterior probability of >75% was used to declare significance. Boxes are color coded to show either increased (red) or decreased (blue) effect on tissue-specific gene expression. The degree of shading reflects the normalized effect sizes (and therefore no units) from the slope of the linear regression model for the effect allele relative to the non-effect allele (see methods for more information). The direction of effect has been aligned to the ECG trait prolonging allele. Y axis: Transcripts. RAA: Right atrial appendage, LV: Left ventricle.

Fig. 5. Enrichment network visualization of DEPICT GO biological processes.

The first three panels (QT, JT, and QRS) were created using Cytoscape (v3.8.2). Significant GO biological processes (false discovery rate [FDR] < 0.01) from DEPICT pathway analyses (represented as a colored point in the image) were linked together (light orange line) when containing a minimum of 25% overlap of gene members. Orphan pathways or those with less than three edges were excluded. This created discrete “modules” of interlinked pathways, from which common themes could be identified. The final panel shows a bar graph with the most significant GO process members (Y-axis) for JT and QRS from each “common theme”, along with their enrichment P-values (X-axis) and color coded by FDR (see legend). Enrichment P-values are as output by DEPICT which compares z-scores derived from Welch’s t-test again the null hypothesis (see methods for more information). TGF-beta: Transforming growth factor beta, TRPS/TKS: transmembrane receptor protein serine/threonine kinase.

Fig. 6. Odds ratios and confidence intervals for ECG PRS with clinical outcomes in UK Biobank.

Data are presented as odds ratios (OR) and 95% confidence intervals (lower 2.5% and upper 97.5%) for association of each ECG (QT – blue, JT – green, QRS – yellow) polygenic risk score (PRS) with prevalent cases in UK Biobank from logistic regression analyses. Associations are reported as risk per standard deviation increase in the PRS and statistical tests were two sided. To adjust for multiple testing, a Bonferroni-corrected threshold (P < 6.3 × 10⁻³) was used to declare significance. A total of 371,951 individuals of European ancestry were included in this analysis. AF (Atrial Fibrillation), AVB (Atrioventricular block), PPM (Permanent pacemaker), BBB (Bundle branch block), HF (Heart Failure).