A massive collaborative effort has shed some light on the mechanisms at work in progressive supranuclear palsy, a neurodegenerative tauopathy that ravages movement, balance, and cognition. In a manuscript posted November 13 on bioRxiv, researchers led by John Crary at the Icahn School of Medicine at Mount Sinai in New York and Adam Naj at the University of Pennsylvania in Philadelphia present the results of the largest genome-wide association study for PSP to date. It turned up six risk loci, confirming five previously identified and one new signal, near the complement C4A gene. Several of the risk polymorphisms appeared to operate in oligodendrocytes, casting the myelinating cells as ground zero for PSP pathogenesis. The researchers spotted injured axons, oligodendrocytes, C4a, and p-tau suspiciously mingling together in postmortem brain samples from people who died with the disease.

  • Largest GWAS for progressive supranuclear palsy conducted to date.
  • Six risk loci came up, of which three tweak gene expression in oligodendrocytes.
  • C4A is the candidate gene for a novel risk locus.

Though rare, PSP is the second-most-common cause of Parkinsonism after PD. Scientists do not know the cause, but studies have found that it can cluster in families. Efforts to understand the genetic risk factors have been hampered by the small number of people affected. Previously, smaller GWAS identified MAPT as the strongest risk gene, along with variants in MOBP, the gene for myelin associated protein, and STX6, which encodes a SNARE protein involved in intracellular trafficking (Jun 2011 news; Chen et al., 2018).

The new GWAS includes 2,779 PSP cases and 5,584 age-matched controls, nearly double the number of cases included in all previous studies. More than 90 percent of cases, and all of the controls, were neuropathologically confirmed at autopsy, and genetic analyses were performed from extracted brain tissue. While burdensome, this neuropathological clarity was critical for strengthening the validity of the findings, Crary told Alzforum.

The tissue samples were collected at more than 40 institutions worldwide, including contributions from the Mayo Clinic in Jacksonville, Florida; University College London; University of California, Los Angeles; University of Pennsylvania in Philadelphia; and the University of Louisville in Kentucky. The National Institute on Aging’s Alzheimer’s disease research centers contributed most of the age-matched, neurologically normal controls. Genotyping was performed on DNA extracted from the tissue samples at Icahn, UPenn, and UCLA, and the datasets harmonized for GWAS analysis.

First author Kurt Farrell and colleagues tied polymorphisms in six loci to PSP. Five—MAPT, STX6, MOBP, RUNX, and SLCO1A2—had cropped up in prior, smaller GWAS. Another, located near the TNXB gene adjacent to the polymorphic metropolis of the HLA locus, was a first-timer. Together, the six loci comprised 218 genes within the vicinities of their lead single nucleotide polymorphisms (SNPs). Referencing annotated maps of promoters, enhancers, and other epigenetic features in different CNS cell types, the researchers found that, more often than can be expected by chance, PSP-associated SNPs were found in oligodendrocyte enhancer regions. This contrasts the scenario in AD, in which risk polymorphisms abound in microglial enhancers, and in PD, where they are enriched in both neuronal and oligodendrocyte promoters.

To learn in which genes, and cell types, the risk loci were active, the scientists referenced gene expression data from the Genotype-Tissue Expression (GTEX) consortium, as well as single-cell transcriptomics data from human brain samples (Bryois et al., 2022). They found that risk variants near STX6 and MOBP associated with elevated expression of these genes, and that the variants aligned with oligodendrocyte-specific enhancers. In the case of RUNX, the risk variants clustered within a region with overlapping microglial and oligodendrocyte enhancers for the gene, raising the possibility that it modulates expression in both cell types. No eQTLs were identified for the previously identified SLCO1A2 locus, and the regulatory complexity of the MAPT locus precluded this type of analysis for the tau gene.

The researchers next investigated the new signals near TNXB. Due to this gene's proximity to the HLA locus, the researchers had to bust out a suite of computational tools to narrow down the list of potential causal genes. Their analyses converged on the complement gene C4A. More than any other gene in the region, its expression tracked with the risk polymorphisms. Previous studies have tied extra copies of the gene to schizophrenia (Sekar et al., 2016). Farrell found that C4A copy number appeared to strengthen risk for PSP, although the study is too small to be sure. In all, the findings suggest that a glut of C4A might somehow tip the balance toward this tauopathy.

Havoc of PSP. In a PSP brain sample, C4A (pink) blankets processes of an oligodendrocyte and the myelin sheath. The oligodendrocyte cell body also contains C4A as well as p-tau (green) as demonstrated by the “coiled body” (pink and green merged: blue) that encircles the oligodendrocyte nucleus (brown). Phospho-tau (green) also accumulated within a tufted astrocyte nearby (wispy green cluster, right of center). [Courtesy of Farrell et al., bioRxiv, 2023.]

Is there any evidence of oligodendrocyte and/or complement-driven shenanigans in the brains of people who died with PSP? Lo and behold, immunohistochemistry of 10 PSP frontal cortex samples revealed a nexus of tau pathology, complement activation, and oligodendrocyte mayhem. The scientists spotted telltale neuropathological characteristics of PSP, i.e., accumulation of phosphorylated tau in neurons and in tufted astrocytes, as well as in oligodendrocyte coiled bodies. Strikingly, they saw stunted axons littered with C4A near these defunct oligodendrocytes. C4A expression was higher in the brain, and blood, of people with PSP than controls. To Crary, these findings place oligodendrocyte malfunction, axonal injury, and complement-mediated inflammation at the heart of PSP. Mechanistic details remain to be ironed out in future studies.

Crary believes complement-mediated inflammation could be important in tauopathies besides PSP. Complement has been implicated in AD and other neurodegenerative diseases, where it targets synapses for destruction by overzealous microglia (Jun 2008 news; Apr 2016 news). In PSP, pathways involving oligodendrocytes might instigate the complement cascade, Crary suggested.

The scientists hope to gather more PSP case samples worldwide to beef up the size and ethnoracial diversity of their GWAS. Doing so will allow them to uncover rarer disease-associated variants, and to understand how risk plays out on different genetic backgrounds, Farrell told Alzforum.

James Rowe of Cambridge University, U.K., called the study an important step forward. It adds to findings from previous GWAS that pointed to tau proteostasis, the unfolded protein response, and inflammation in PSP pathogenesis. “Given the complexity of the HLA region, the histological evidence supporting immune pathways and a unique oligodendrocyte signature in PSP is of immediate therapeutic interest,” he wrote.—Jessica Shugart


  1. Good science is built on replication and validation. This new paper makes an importation contribution to both.

    It is unclear why the heritability of PSP is so low, in contrast to the closely related disorders of frontotemporal dementia. This is good news for families affected by PSP, but a challenge for research into the causes and modifiable mechanisms of PSP.

    The paucity of autosomal dominance in PSP does not, however, mean there are no genetic signals to elucidate its etiology. International collaboration has been key to the assembly of increasingly large cohorts of people with PSP with whom to conduct genome-wide association, revealing variants in MAPT, STX6, EIF2AK3, MOBP, RUNX2, SLCO1A, SLC2A13 (close to LRRK2), KANSL1 (adjacent to MAPT) in one or more smaller GWAS. These genetic variants imply potential therapeutic strategies via tau proteostasis, unfolded protein response, inflammation, and more.  

    Farrell et al. build impressively on this former work, in three ways:

    • First, through a larger cohort, of ~2,700 people with PSP and ~5,500 controls. GWAS and eQTL confirms five of the susceptibility loci—MAPT, MOBP, STX6, RUNX2, SLCO1A2—beyond doubt.
    • Second, by neuropathological evidence in  more than 93 percent of cases. This diagnostic accuracy is especially important for a rare and heterogeneous disorder with phenotypic mimicry like PSP.
    • Third, by molecular and histological validation studies in human brain tissue. For example, the relevance of inflammation is suggested by the microglial expression of RUNX2, and the new susceptibility locus C4A is underscored by elevated C4A expression in brain tissue, with co-localization of tau aggregates in oligodendrocytic C4.

    Given the complexity of the HLA region, the histological evidence supporting immune pathways and a unique oligodendrocyte signature in PSP is of immediate therapeutic interest.

    Looking ahead, there remains the need to analyze the genetic influences on phenotypic variance and survival. And functional genomics, in vivo models, replication and experimental medicine studies are to be expected. To this end, Farrell et al represent an important step forward.

Make a Comment

To make a comment you must login or register.


News Citations

  1. GWAS Fingers Tau and Other Genes for Parkinsonian Tauopathy
  2. Complement: AD Friend or Foe? New Work Tips Balance to Former
  3. Paper Alert: Microglia Mediate Synaptic Loss in Early Alzheimer’s Disease

Paper Citations

  1. . Joint genome-wide association study of progressive supranuclear palsy identifies novel susceptibility loci and genetic correlation to neurodegenerative diseases. Mol Neurodegener. 2018 Aug 8;13(1):41. PubMed.
  2. . Cell-type-specific cis-eQTLs in eight human brain cell types identify novel risk genes for psychiatric and neurological disorders. Nat Neurosci. 2022 Aug;25(8):1104-1112. PubMed.
  3. . Schizophrenia risk from complex variation of complement component 4. Nature. 2016 Feb 11;530(7589):177-83. Epub 2016 Jan 27 PubMed.

Further Reading


  1. . Genetic determinants of survival in progressive supranuclear palsy: a genome-wide association study. Lancet Neurol. 2021 Feb;20(2):107-116. Epub 2020 Dec 17 PubMed.

Primary Papers

  1. . Genetic, transcriptomic, histological, and biochemical analysis of progressive supranuclear palsy implicates glial activation and novel risk genes. 2023 Nov 13 10.1101/2023.11.09.565552 (version 1) bioRxiv.