Investigation of the HLA locus in autopsy-confirmed progressive supranuclear palsy

Objectives: Progressive supranuclear palsy (PSP) is a neurodegenerative disease showing pathological tau accumulation in subcortical neurons and glial cells. The human leukocyte antigen (HLA) locus on chromosome 6 is a polymorphic region with complex linkage patterns that has been implicated in several autoimmune and neurological disorders. The HLA locus has not been systematically examined in PSP. It is unclear whether tau and HLA can interact to induce an autoimmune disease mechanism. Methods: We evaluated an autopsy confirmed PSP cohort (n = 44) and compared allele/haplotype frequencies to those of the reference group of a local deceased Canadian donor pool. We performed HLA-Tau peptide binding prediction and modelling of HLA Class II - Tau Peptide interactions. Findings: Odds ratio was 2.94(95% CI 1.01 to 8.55; p = 0.047) for DQB1*06:01 allele, and 2.59(95% CI 1.39 to 4.83; p = 0.0025) for the narcolepsy-associated haplotype (DRB1*15:01-DQB1*06:02). One patient with 4-repeat tau PSP-type pathology was a carrier of the IgLON5-associated haplotype (DRB1*10:01-DQB1*05:01). HLA-Tau peptide binding prediction and modelling of HLA Class II - Tau Peptide interactions revealed strong-binding tau peptides but not the PSP-protofilament fold for alleles DQA1*01:02-DQB1*06:02 and DQA1*01:03-DQB1*06:01. Conclusion: Our study suggests that epitopes within the tau peptide may bind to HLA alleles that are found in a subset of PSP patients supporting the notion of an autoimmune pathophysiological component. These findings have implications for subtyping and stratifying patients for therapies, including those targeting immune modulation.