Integrative genomic, transcriptomic, and proteomic analyses to investigate sex-specific differences in Alzheimer's Disease

点击翻译按钮获取中文摘要

Alzheimer's disease (AD) affects 35 million people worldwide. However, there is no effective treatments to slow or halt the underlying neurodegeneration of AD. Strikingly, women are affected by AD about twice as much as men. Why women are disproportionately affected by AD is not well understood. Here, we hypothesize that there is an interaction between biological sex and brain gene expression that predisposes women to have a higher risk for AD. This hypothesis builds on our work investigating the genetic control of gene expression in the human brain. In those studies, we have identified genetic sites that are associated with variation in gene expression at the transcript and protein level in the human brain. Identifying genetic sites associated with natural variation in brain gene expression are incredibly useful to resolve genetic signals identified by genome-wide association study (GWAS) to specific causal transcripts and proteins. We have successfully used brain gene transcript and protein expression results with GWAS results to identify novel proteins for AD (Wingo et al, Nat Genet, 2021) and depression (Wingo et al, Nat Neurosci, 2021). These analyses implicate the causes of inherited genetic risk, which are presumably among the most important early contributors to disease, which make them highly relevant to resolving sex-specific disease risk. To test our hypothesis, we will first perform a sex-specific brain expression analysis for transcripts and proteins (Aim 1). To do this, we will leverage existing brain transcripts and protein expression data generated by national resources, and we propose a novel approach to cost-effectively increase the depth of proteomes to improve power. We expect to generate sex-specific quantitative trait loci (QTL) for brain transcripts (N=1655, 61% women) and proteins (N=1584, 66% women), and each gene will then be tested for evidence of interaction with sex. These results are likely to be of general interest to the neuroscience and genetics communities, and the data and results will be made widely available to the science community. In Aim 2, we will perform a sex-specific GWAS of AD, and we will identify genetic variants that interact with sex. To do this, we will perform sex-stratified GWAS using 1) case-control status for AD in participants from US studies (N=27,580), 2) AD-proxy case-control in participants from the UK Biobank (N=431,000), and iii) a meta- analysis of 1 and 2. In Aim 3, we will identify brain transcripts and proteins contributing to AD pathogenesis that are specific to men or women, respectively. To do this, we will integrate the sex-specific brain QTLs with the sex-specific GWAS results to resolve GWAS signals to proteins for women and men, separately. For AD genetic signals with evidence for sex interaction, we will test whether sex-specific gene expression accounts for the differences in AD risk between the sexes using causal inference approaches (e.g., Mendelian randomization). Findings from this project are highly likely to provide novel mechanistic insights into sex differences in AD and promising new targets for further sex-specific mechanistic and therapeutic studies of AD.