In:
Science, American Association for the Advancement of Science (AAAS), Vol. 377, No. 6608 ( 2022-08-19)
Abstract:
An estimated 6 million people worldwide are affected by Parkinson’s disease (PD), for which there are no disease-modifying therapies. Since 2005, genome-wide association studies (GWASs) have aimed to find common variant risk loci for PD. These PD GWASs have now uncovered 90 genome-wide significant risk signals. However, target genes and mechanisms remain largely unknown, hampering the field’s ability to develop downstream therapeutic approaches from these genomic studies. RATIONALE We dissected a chromosome 7 locus linked to risk for PD by GWASs to find the target gene and understand its role in PD pathophysiology. We first performed colocalization analyses of expression quantitative trait locus (eQTL) and PD risk signals, confirming eQTL effects with allele-specific expression (ASE) studies of candidate target genes in the human brain. We then characterized the consequences of manipulating target gene expression in human induced pluripotent stem cell–derived neurons (iPSC-Ns) and other cell models. Finally, we determined whether levels of the target gene or protein were associated with disease state and severity in humans. RESULTS We linked the GWAS-derived chromosome 7 PD risk locus [sentinel single-nucleotide polymorphism (SNP) rs199347] to GPNMB —which encodes the transmembrane protein glycoprotein nonmetastatic melanoma protein B (GPNMB)—with a 94% posterior probability of the PD risk and caudate GPNMB eQTL signals sharing a causal variant. In human brain samples across multiple brain regions (the caudate, cerebellum, and cingulate cortex) from both PD cases ( n = 4) and neurologically normal controls ( n = 2), ASE studies confirmed that the PD risk–associated haplotype was associated with threefold higher GPNMB expression. Having linked GPNMB expression levels to PD risk, we generated iPSC-N lines with normal levels of GPNMB [wild-type (WT)], heterozygous loss of GPNMB (Het), or homozygous loss of GPNMB [knockout (KO); two different lines generated] by CRISPR-Cas9 genome editing. Analyzing these iPSC-Ns using confocal microscopy, we found that loss of GPNMB was accompanied by a marked reduction in α-synuclein (aSyn) at the synapse—a result confirmed by immunoblotting of synaptosomal preparations for aSyn. Culturing the iPSC-Ns for an additional week, to a state of greater maturity, only exaggerated the differences in synaptic aSyn when comparing WT versus all other lines. Moreover, transcriptomic profiling of these genome-edited iPSC-N lines revealed that aSyn served as a key protein-protein interaction hub for genes with dysregulated expression in GPNMB Het and KO states. Because aSyn is the protein that accumulates in the hallmark pathological inclusions of PD, and because progression of PD may result from the uptake and cell-to-cell spread of pathological forms of aSyn, we further explored the relationship between GPNMB and aSyn. In HEK293 and HeLa cells, we found that GPNMB and aSyn colocalized and coimmunoprecipitated. In iPSC-Ns, we observed minimal uptake of fibrillar forms of aSyn in GPNMB Het and KO iPSC-Ns, in contrast to rapid and obvious uptake of fibrillar aSyn by WT iPSC-Ns. Additionally, when iPSC-Ns were cultured for 14 days after a one-time addition of fibrillar aSyn to the culture medium, WT iPSC-Ns developed hyperphosphorylated, insoluble inclusions of aSyn, but these forms of aSyn pathology were minimal in GPNMB Het and KO iPSC-Ns. These experiments, together, demonstrated that GPNMB was necessary for internalization of fibrillar aSyn and subsequent development of aSyn pathology. Meanwhile, in cell lines with low GPNMB expression that do not normally internalize fibrillar aSyn, exogenous expression of GPNMB was sufficient to confer ability to take up aSyn. Finally, using samples from 731 PD and 59 neurologically normal control individuals, we found that plasma GPNMB is elevated in PD and that PD individuals with higher plasma GPNMB levels have more severe disease. CONCLUSION Computational, cell biological, and human tissue–based studies establish GPNMB as a GWAS-derived risk gene for PD, with higher expression levels mediating pathogenicity through interactions with aSyn. As a transmembrane protein that is expressed at the cell surface, with an extracellular, soluble form generated by cleavage, GPNMB is a candidate for biomarker development and therapeutic targeting in PD. GPNMB and aSyn in PD. Haplotypes associated with PD result in increased expression of GPNMB , encoding GPNMB. GPNMB interacts with aSyn, with effects on aSyn expression at the synapse, cellular uptake of fibrillar aSyn, and development of aSyn pathology in neurons. Together, these cellular processes mediate risk for development and progression of PD. Created with BioRender.com. PFF, preformed fibril.
Type of Medium:
Online Resource
ISSN:
0036-8075
,
1095-9203
DOI:
10.1126/science.abk0637
Language:
English
Publisher:
American Association for the Advancement of Science (AAAS)
Publication Date:
2022
detail.hit.zdb_id:
128410-1
detail.hit.zdb_id:
2066996-3
detail.hit.zdb_id:
2060783-0
SSG:
11
Permalink