Abstract
BACKGROUND: Epidemiological evidence supports the association between periodontitis and cardiovascular diseases (CVD); however, their shared genetic mechanisms remain inadequately defined. This study elucidates their genetic architecture by identifying shared risk loci and associated genes. METHODS: This study employs Mendelian randomization (MR) to investigate bidirectional causal relationships between periodontitis and five types of CVD based on genome-wide association study (GWAS) summary data. Cross-trait analyses were applied to examine genetic correlations across trait pairs, identifying pleiotropic loci and associated genes. Functional annotation and tissue-specificity analyses elucidate their biological functions. RESULTS: Bidirectional and multivariable MR analyses confirmed that the association between CVD and periodontitis is not driven by a direct causal relationship. Additionally, the genetic correlation between these disorders underscores the importance of investigating their shared genomic architecture. Colocalization analysis identified significant shared causal variants at loci 4p14 and 15q25.1. At the gene level, seven unique pleiotropic genes (e.g., CD151, POLR2L, and HLA-DQA1) were annotated. Pathway analysis revealed that these genetic architectures likely mediate cross-disease interactions through an inflammation-metabolism regulatory axis (Inflammatory Response and Cholesterol Metabolism Pathway). Tissue enrichment analyses demonstrated that pleiotropic signals, from SNP to gene levels, were significantly enriched in immune-related tissues and disease-relevant sites like the heart. CONCLUSION: This study reveals a shared genetic basis between periodontitis and five types of CVD, suggesting potential underlying mechanisms. However, based on summary-level data, it remains unclear whether this association represents direct biological genetic determinants or indirect pathways mediated by shared environmental or behavioral risk factors. Future studies utilizing individual-level data with covariate adjustments are needed to further investigate these relationships.