Abstract
Background
Sex has increasingly been recognized as an important factor contributing to the heterogeneity in Alzheimer’s disease (AD).
Method
We performed a sex‐specific meta‐analysis of two large epigenome‐wide association studies (ADNI, AIBL) in the blood, with a total of 633 and 651 samples in females and males, respecitively. For each dataset, we adjusted covariate variables age, sex, batch, and immune cell‐type proportions. The Inverse‐variance fixed‐effects meta‐analysis model was then used to prioritize the most consistent DNAm differences across datasets.
Result
In female samples, 2 CpGs, mapped to the PRRC2A and RPS8 genes, reached the 5% false discovery rate (FDR). No CpGs reached 5% FDR in male samples analysis. At a more relaxed significance threshold of P < 1×10−5, an additional 22 CpGs and 5 CpGs were identified in female samples and male samples, respectively. For these 29 AD‐associated CpGs, the odds ratios ranged from 0.843 to 1.335 in females and 0.993 to 1.058 in males. The majority of these CpGs were hypermethylated in AD subjects (23 CpGs), located outside CpG islands or shores (27 CpGs), or in distal regions located greater than 2k bp from the TSS (27 CpGs). Only 2 of these 29 CpGs were located in gene promoters at SLC5A8 and C16orf89.
At 5% Sidak corrected P‐value, our DMR analysis using comb‐p software identified 49 distinct differentially methylated regions (DMRs) in female and male samples each. The median numbers of CpGs in these DMRs are 5 CpGs in females and 4 CpGs in males. Among the top 10 most significant DMRs, about half of them (6 in females, 5 in males) were hypermethylated in AD. In females, 5 out of the 10 DMRs were mapped to promoter regions of the NNAT, CCDC169‐SOHLH2, ARHGEF15, LPAR5, and ZNF595 genes. In males, 4 out of the 10 DMRs were mapped to promoter regions of the MCCC1, PM20D1, PRR19, TTC23, and LRRC28 genes.
Conclusion
As sex is a strong factor underlying phenotypic variability in AD, the results of our study are particularly relevant for a better understanding AD pathophysiology. They also provide a valuable resource for sex‐specific biomarkers and drug targets in AD.
