Atrial Fibrillation and Primary Cilia-Associated Genes: The Role of CEP68

Recent studies have demonstrated that primary cilia not only play a role in cardiovascular development, but also in the progression of acquired heart disease. Their role in atrial fibrillation (AF) is incompletely understood. We hypothesize that there is a causal link between primary cilia genes and the occurrence of AF. We integrated AF GWAS data with various multi-omic datasets-including data on gene expression, DNA methylation, and protein expression quantitative trait loci (eQTL, mQTL, and pQTL)-from human left atrial appendage (LAA) tissues and blood. Genetic variants linked to primary cilia-related genes were used as instrumental variables to explore their causal links to AF, through summary-data-based Mendelian randomization (SMR) and Bayesian colocalization. Single-cell sequencing data were used to analyze the expression of the selected genes across different cell types. The mechanisms by which the selected genes exert their effects were explored using RNA sequencing data, clinical indicators, and immunohistochemical markers from 22 patients without AF from the PREDICT-AF cohort, and 21 patients with paroxysmal AF and 19 patients with persistent AF from the MARK-AF cohort. Through SMR analyses, we established significant associations between predicted CEP68 expression and AF in both blood (OR 1.25; 95% CI 1.18-1.33; false discovery rate (FDR) = 1.81 × 10^-9) and LAA tissue (OR 1.12; 95% CI 1.08-1.16; FDR = 6.18 × 10^-9). Moreover, predicted methylation of CEP68 showed an inverse relationship with AF risk (OR 0.87; 95% CI 0.84-0.90; FDR = 2.55 × 10^-15). Colocalization results for CEP68 in both blood and the LAA indicated strong evidence of a shared causal variant. Within single-cell data, compared to the control group, AF patients had higher levels of CEP68 in fibroblasts (p = 0.046). In bulk RNA-seq data, CEP68 expression showed no significant differences among the no AF, paroxysmal AF, and persistent AF groups. CEP68 was positively correlated with the cardiac remodeling marker Thrombospondin-2 in 22 patients without AF from the PREDICT-AF cohort (r = 0.45, p = 0.03). In AF patients from the MARK-AF study, CEP68 was also positively associated with LAVI (r = 0.34, p = 0.03). Collectively, our results support a model in which genetically predicted CEP68 regulation is linked to AF liability and is consistent with fibroblast activation and remodeling-related pathways as potential mediators.