Glial-Specific Genes Are Strongly Associated With Alzheimer's Disease by Gene-based Polygenic Risk Score Analysis

Methods: Gene-based PRSs were constructed in AD cases and controls within each gene of the glial cells, according to the GWAS summary statistics of European ancestry. In detail, gene-based PRSs were first calculated for each glial cell type-specific gene for AD cases and controls in the discovery dataset (ADc1234ADA) using PRSet software. A meta-analysis with a fixed model was performed when the signal in both datasets was in the same direction. Bonferroni corrections for multiple testing (at α = .05) were used to determine significance within each of the three glial groups. Forest plots were used to visualize the results of meta-analyses. Results: Two genes (Bcl-3 and RelB) from microglia were identified to be significantly associated with AD when adjusted with PC1 and PC2. However, when we adjusted with additional covariates age, sex, and APOE ε4 count, the association between BCL-3 and AD turned out to be non-significant. Similarly, the association between RelB and AD was also statistically non-significant.One gene (BIN1) from oligodendrocyte was significant in PC1 and PC2 and also significant when we included the sex, age, APOE ε4 count covariates. None of the genes from astrocytes were significantly associated with AD. Conclusion: Our findings emphasize the importance of considering cell-type-specific genetic contributions to AD, providing novel insights into the disease's complex biological mechanisms for the disease. In this study, we identified two genes (Bcl-3 and RelB) from microglia and one gene (BIN1) from oligodendrocytes that might play an important role in the development of AD.