Abstract

Progressive supranuclear palsy (PSP) is the second most common neurodegenerative Parkinsonian disorder after Parkinson’s disease, and is characterized as a primary tauopathy. Leveraging the considerable clinical and neuropathologic heterogeneity associated with PSP, we measured tau neuropathology as quantitative traits to perform a genome-wide association study (GWAS) within PSP to identify genes and biological pathways that underlie the PSP disease process. In 882 PSP cases, semi-quantitative scores for phosphorylated tau-immunoreactive coiled bodies (CBs), neurofibrillary tangles (NFTs), tufted astrocytes (TAs), and tau threads were documented from 18 brain regions, and converted to latent trait (LT) variables using the R ltm package. LT analysis utilizes a multivariate regression model that links categorical responses to unobserved covariates allowing for a reduction of dimensionality, generating a single, continuous variable to account for the multiple lesions and brain regions assessed. We first tested for association with PSP LTs and the top PSP GWAS susceptibility loci. Significant SNP/LT associations were identified at rs242557 (MAPT H1c sub-haplotype) with hindbrain CBs and rs1768208 (MOBP) with forebrain tau threads. Digital microscopy was employed to quantify phosphorylated tau burden in midbrain tectum and red nucleus in 795 PSP cases and tau burdens were used as quantitative phenotypes in GWAS. Top associations were identified at rs1768208 with midbrain tectum and red nucleus tau burden. Additionally, we performed a PSP LT GWAS on an initial cohort, a follow-up SNP panel (37 SNPs, P < 10–5) in an extended cohort, and a combined analysis. Top SNP/LT associations were identified at SNPs in or near SPTBN5/EHD4, SEC13/ATP2B2, EPHB1/PPP2R3A, TBC1D8, IFNGR1/OLIG3, ST6GAL1, HK1, CALB1, and SGCZ. Finally, testing for SNP/transcript associations using whole transcriptome and whole genome data identified significant expression quantitative trait loci at rs3088159/SPTBN5/EHD4 and rs154239/GHRL. Modeling tau neuropathology heterogeneity using LTs as quantitative phenotypes in a GWAS may provide substantial insight into biological pathways involved in PSP by affecting regional tau burden.

Highlights

  • Progressive supranuclear palsy (PSP) is an atypical Parkinsonian disorder where patients typically exhibit early unexplained falls, vertical gaze palsy, axial rigidity, and levodopa unresponsive parkinsonism

  • The greatest information is provided by the items midbrain tectum, red nucleus, pontine base, ventral thalamus, and motor cortex which is consistent with what we observe in PSP tau neuropathology

  • We have taken an innovative approach to summarize tau neuropathology in PSP by applying a statistical model principally used in psychometric testing

Read more

Summary

Introduction

Progressive supranuclear palsy (PSP) is an atypical Parkinsonian disorder where patients typically exhibit early unexplained falls, vertical gaze palsy, axial rigidity, and levodopa unresponsive parkinsonism. PSP is a primary tauopathy because on neuropathologic examination there is predominately hyperphosphorylated, aggregated tau protein. PSP neuropathologic features include tau-immunoreactive neuronal and glial lesions in the basal ganglia, diencephalon, and brainstem with variable involvement of the neocortex. Macroscopic examination in PSP reveals pigment loss in the substantia nigra and atrophy of multiple brain regions including superior cerebellar peduncle, subthalamic nucleus, hilus of the cerebellar dentate, and midbrain with dilation of the aqueduct of Sylvius. Neuropathologic diagnostic criteria for PSP require the presence of tau neurofibrillary tangles in the most affected nuclei which are the globus pallidus, subthalamic nucleus, and substantia nigra [19, 31]. Tufted astrocytes (TA) are astrocytic tau lesion found consistently in the motor cortex and striatum in PSP [34, 38] and oligodendroglial coiled bodies (CB), often accompanied by neuropil threads in the white matter of the diencephalon, brainstem, and cerebellum [5]

Methods
Results
Conclusion
Full Text
Published version (Free)

Talk to us

Join us for a 30 min session where you can share your feedback and ask us any queries you have

Schedule a call