Abstract
Parkinson’s disease (PD) and progressive supranuclear palsy (PSP) are clinically similar neurodegenerative movement disorders that display unique neuropathological features (i.e. Lewy body pathology and Tau pathology, respectively). While each disorder has distinct clinical and genetic risk factors, both are associated with the MAPT 17q.21.31 locus H1 haplotype. This suggests a pleiotropic effect of this genomic region. To better understand the genetic contribution of this region to these diseases, we fine-mapped the apparent pleiotropy of this locus. Our study indicates that PD and PSP are associated with different sub-haplotypes of the H1 clade. PD-associated sub-haplotypes were associated with altered LRRC37A copy number and expression, which, like other PD risk-associated genes, we hypothesize to be most relevant to astroglial function. In contrast, PSP was associated with grossly altered LD structure across the 17q21.31 locus, and risk-associated variants were found to impact chromatin structure in both neurons and microglia. We conclude that the contribution of the 17q21.31 locus to multiple disorders is a result of its structural and haplotypic complexity, which in turn impacts the regulation of multiple genes and neural cell types. This raises the possibility of novel disease-specific pathogenic mechanisms driven by 17q21.31 structural variation and altered epigenetic regulation that appear to converge on glial function and gene expression. By fine-mapping the association of H1 with PD and PSP, we have begun to untangle the apparent pleiotropy of this locus, and gain better insight into the mechanism of each disease, which will guide future functional analyses and disease models for PD and PSP.
