A newly identified receptor rolls out the red carpet for α-synuclein fibrils, ushering them inside neurons and fast-tracking their spread, according to research from Fudan University, Shanghai. In the February 21 Science, scientists led by Cong Liu, Peng Yuan, and Jin-Tai Yu report that FAM171A2—a PD GWAS hit—modulates α-synuclein endocytosis. Boosting FAM171A2 levels in neurons ramped up α-synuclein uptake, while knocking it down suppressed it. FAM171A2 clamps onto α-synuclein’s C-terminus. Curiously, an approved cancer drug blocks this interaction, barring α-synuclein’s entry.

  • New variants in the FAM171A2 gene associated with Parkinson’s. 
  • FAM171A2 helps cells take up α-synuclein fibrils.  
  • The cancer drug bemcentinib blocks α-synuclein-FAM171A2 interactions. 

This is a comprehensive study,” said Michael Henderson of the Van Andel Institute, Michigan. “Starting with genetic risk, then moving to protein changes in humans, then to cell and mouse models, the authors link FAM171A2 to α-synuclein pathology.”

Co-senior author Yu and colleagues had previously linked an SNP in FAM171A2 to higher odds of getting Alzheimer’s (Lambert et al., 2013), Parkinson’s (Lill et al., 2012), or frontotemporal dementia (Rademakers et al., 2008Xu et al., 2020). Here, they home in on PD. In a new meta-GWAS, first author Kai-Min Wu and colleagues tapped genetic data from the U.K. Biobank, Finland’s FinnGen genetic data repository, and a Genentech study of nearly 6,500 PD cases and 300,000 controls (Chang et al., 2017). Wu identified four new SNPs in FAM171A2 that each nudged up a person’s risk of PD by 5 to 10 percent.

The authors next investigated if FAM171A2 protein levels are altered in PD. In cerebrospinal fluid samples from 30 people with PD and an equal number of controls at Huashan Hospital, Shanghai, the former had, on average, more FAM171A2. Likewise, in CSF samples from the Parkinson’s Progression Markers Initiative, FAM171A2 tracked higher in 517 people with PD than in 169 controls. In PPMI datasets, high CSF FAM17A2 also correlated with low CSF α-synuclein. The authors suggest that CSF α-synuclein falls as the protein aggregates in the brain parenchyma, as has been proposed previously (Van Steenoven et al., 2018).

Those aggregates spread through the brain, but scientists don’t fully understand how. To see if FAM171A2 might be involved, Wu and colleagues increased or silenced FAM171A2 in cultured mouse neurons, then exposed the cells to fluorescently labeled α-synuclein fibrils. Overexpressing FAM171A2 caused a spike in fibril uptake, while knocking it down reduced entry. Similarly, six months after injecting preformed fibrils of α-synuclein (PFF) into mouse striatum, animals overexpressing FAM171A2 accumulated more phosphorylated α-synuclein in the striatum and the substantia nigra than did mice with normal FAM171A2 expression. Knockdown of FAM171A2 relieved pathology, leaving fewer phosphorylated deposits in the motor cortex, striatum, amygdala, and substantia nigra (image below).

Slow the Spread. In the mouse brain, phospho-synuclein pathology (green) accumulates (top) in, from left to right, the motor cortex, striatum, amygdala, and the substantia nigra. Knocking down FAM171A2 (bottom) decelerated the spread (bottom row). [Courtesy of Wu et al. Science, 2025.]

“The PFF model is an effective way to assess modifiers of α-synuclein seeding and neuron death,” Henderson told Alzforum, though he cautioned against overinterpreting the results (comment below). “The role of a specific receptor in this pathway is difficult to establish,” he said. Indeed, several receptors have been proposed, yet other studies suggest internalization requires no specific receptor (Holmes et al., 2013). Kelvin Luk, University of Pennsylvania, echoed these sentiments. “There are two main schools of thought: those who believe it does so via specific receptors, and those who think it uses more generic mechanisms, such as through charged membrane interactions, e.g., heparan sulfate proteoglycans,” he said. Luk noted that knocking down any single receptor, including FAM171A2, does not fully stop α-synuclein entry. “That tells you this likely depends on multiple processes,” he said.

If FAM171A2 supports one of those processes, then how does it do that? To find out, Wu and colleagues used total internal reflection fluorescence microscopy to track α-synuclein in cultured N2a cells. They saw PFFs disappearing from the plasma membrane as cells internalized them. Nearly 80 percent of these vanishing acts occurred in puncta containing FAM171A2. Wu found that the first domain of FAM171A2 was critical for this process. Its removal prevented fibrils from being engulfed in HEK293 cells. In an ELISA assay, this domain bound α-synuclein fibrils with nanomolar affinity, approximately 1,000 times more strongly than it bound to monomers.

Focusing on the α-synuclein side of the partnership, NMR spectroscopy revealed that domain 1 of FAM171A2 locks onto α-synuclein’s C terminus. Snipping off the last 40 amino acids of α-synuclein completely abolished this interaction. An AlphaFold-based structural model predicted that the negatively charged α-synuclein C-terminus fits snugly against a positively charged patch on FAM171A2 (image below). Because the C-terminus is partly hidden in monomers but fully exposed in fibrils, the authors think this explains FAM171A2’s preference for the latter.

Andrew West, Duke University, Durham, North Carolina, told Alzforum that more work is needed to detail the types of α-synuclein conformer FAM171A2 interacts with. This is important because, in people, the synuclein C-terminus is a hotbed of posttranslational modification and truncation, which may be poorly represented in the mouse models used.

Hand-in-Glove. Predicted electrostatic surface of α-synuclein residues 100–140, in complex with domain 1 of FAM171A2, as modeled using AlphaFold-Multimer. Red indicates negative charge; blue represents positive charge. Negatively charged α-synuclein’s C-terminus—amino acids 118 to 126 (red, right)—aligns with the positively charged surface of FAM171A2 domain 1—amino acids 109, to 118 (blue, right). [Image courtesy of Wu et al., Science, 2025.]

What if the binding of α-synuclein and FAM171A2 could be disturbed? In search of a small molecule that could do that, Wu et al. screened, in silico, 7,173 FDA-approved drugs. This bubbled up seven that might bind with high affinity. Of these, the cancer drug bemcentinib, an AXL kinase inhibitor, was predicted to be most potent.

Testing this empirically, Wu found that pretreating N2a cells overexpressing FAM171A2 with bemcentinib cut α-synuclein uptake by a third. In mice, bemcentinib did not readily cross the blood-brain barrier, so the scientists injected it daily into the lateral ventricle for a week, then injected fluorescent synuclein fibrils into the striatum. The small molecule reduced internalization of α-synuclein into neurons of the substantia nigra.

Bruno Benitez, Harvard Medical School, Boston, called this an interesting discovery but wondered how it might work in practice. “The possible use of AXL receptor tyrosine kinase inhibitors in Parkinson's disease may need further investigation because of other on- and off-target effects,” he said (comment below). Nine different kinase inhibitor drugs are, or have been, in repurposing trials for Parkinson’s, Alzheimer’s, and related neurodegenerative diseases (Therapeutics).

The authors acknowledge that while bemcentinib itself may not be clinically viable, their structural analysis approach could pave the way for more effective drugs. “Altogether, we identified a pathway potentially involved in the spread of α-syn[uclein] fibrils between brain areas and suggest a compelling strategy for impeding this process in individuals with PD,” they wrote.—George Heaton

George Heaton is a freelance writer in Durham, North Carolina.

Comments

  1. This is a comprehensive study. Starting with genetic risk and moving to protein changes in humans, to various cell culture and mouse experiments, Wu and colleagues show a relationship between FAM171A2 and α-synuclein pathology. The authors identified FAM171A2 as a genetic risk factor for Parkinson's disease with an odds ratio of 1.05-1.10. Genetic risk factors with low odds ratios are difficult to investigate since they are expected to have small effects. Indeed, the authors show small effects of the risk allele on increasing FAM171A2 proteins levels in PD brain and CSF. They then show effects in the PFF mouse and neuron models and suggest that these are through a direct interaction of α-synuclein with FAM1717A2 and that FAM1717A2 acts as a receptor for α-synuclein fibrils.

    The PFF model is an effective way to assess modifiers of α-synuclein seeding and neuron death. However, the role of a specific receptor in this pathway is more difficult to establish. Several potential receptors for α-synuclein have been proposed, but other studies have also demonstrated that α-synuclein internalization does not require a specific receptor. It is unclear from this study how the role that FAM1717A2 plays in PFF internalization relates to other potential receptors or bulk endocytosis.

    The study provides human genetic and modeling data to support a role for targeting FAM171A2, but since variants in FAM171A2 have small effect sizes, and FAM171A2 may have other important roles, substantial additional research will be required to understand to what extent FAM1717A2 could be targeted therapeutically.

  2. The GWAS signal in this locus on Chr17 affecting PD risk has been reported previously (Nalls et al., 2019; Kim et al., 2024). However, at the time of those publications, and using additional criteria such as functional genomics and QTL analysis, the nominated genes were PGRN, and ITGA2B. This highlights the importance of unbiased fine-mapping approaches and the complexities and difficulties in defining the "causal" gene from GWAS hits.

    Here, Wu et al. performed a meta-analysis using summary statistics from three independent cohorts, focusing on an extended analysis of 12 additional variants in the FAM171A2 gene, resulting in a highly significant association with PD risk. However, a meta-analysis including all the SNPs in LD with the sentinel SNP across the entire locus beyond FAM171A2 would have been more compelling. It is not clear if an SNP previously identified at this locus is driving the association here (Chang et al., 2017). Nonetheless, the data shown here, with overexpression and knockdown of FAM171A2 having opposite effects on the p-α-syn pathology, substantia nigra cell death, and behavioral changes in mice, make a compelling argument for the potential role of neuronal FAM171A2 in synuclein pathology.

    Interestingly, the same group had previously reported that SNPs in the FAM171A2 gene are genetic modifiers of the CSF levels of PGRN. They also noted that the FAM171A2 gene is considerably enriched in vascular endothelium and microglia (Xu et al., 2020). These results emphasize the intricate relationships among these genes at this locus and their potential impact on different cell types and brain pathologies.

    Another interesting discovery is the effect of bemcentinib binding to FAM171A2, inhibiting α-synuclein preformed fibril uptake and potentially blocking the spread of α-synuclein pathology. However, the potential use of AXL receptor tyrosine kinase inhibitors in Parkinson's disease needs further investigation because of the likelihood of other on- and off-target effects of such inhibitors.

    Many more questions will need to be addressed, including whether FAM171A2 is expressed in susceptible substantia nigra neurons and whether it is safe for neuronal, endothelial, or microglial survival to lower their levels to ~50 percent. What is the level of FAM171A2 expression in neurons with Lewy body pathology? Are circulating levels of FAM171A2 in CSF or plasma potential biomarkers for PD? Is FAM171A2's involvement in α-synuclein uptake/endocytosis mediated by interactions with other retromer or endolysosomal proteins, such as PGRN? Are individuals with coding (rare) variants in FAM171A2, the binding site for α-synuclein, more or less susceptible to a-synuclein pathology?

    References:

    . Identification of novel risk loci, causal insights, and heritable risk for Parkinson's disease: a meta-analysis of genome-wide association studies. Lancet Neurol. 2019 Dec;18(12):1091-1102. PubMed.

    . Multi-ancestry genome-wide association meta-analysis of Parkinson's disease. Nat Genet. 2024 Jan;56(1):27-36. Epub 2023 Dec 28 PubMed.

    . A meta-analysis of genome-wide association studies identifies 17 new Parkinson's disease risk loci. Nat Genet. 2017 Oct;49(10):1511-1516. Epub 2017 Sep 11 PubMed.

    . The FAM171A2 gene is a key regulator of progranulin expression and modifies the risk of multiple neurodegenerative diseases. Sci Adv. 2020 Oct;6(43) Print 2020 Oct PubMed.

Make a Comment

To make a comment you must login or register.

References

Paper Citations

  1. . Meta-analysis of 74,046 individuals identifies 11 new susceptibility loci for Alzheimer's disease. Nat Genet. 2013 Dec;45(12):1452-8. Epub 2013 Oct 27 PubMed.
  2. . Comprehensive research synopsis and systematic meta-analyses in Parkinson's disease genetics: The PDGene database. PLoS Genet. 2012;8(3):e1002548. Epub 2012 Mar 15 PubMed.
  3. . Common variation in the miR-659 binding-site of GRN is a major risk factor for TDP43-positive frontotemporal dementia. Hum Mol Genet. 2008 Dec 1;17(23):3631-42. PubMed.
  4. . The FAM171A2 gene is a key regulator of progranulin expression and modifies the risk of multiple neurodegenerative diseases. Sci Adv. 2020 Oct;6(43) Print 2020 Oct PubMed.
  5. . A meta-analysis of genome-wide association studies identifies 17 new Parkinson's disease risk loci. Nat Genet. 2017 Oct;49(10):1511-1516. Epub 2017 Sep 11 PubMed.
  6. . α-Synuclein species as potential cerebrospinal fluid biomarkers for dementia with lewy bodies. Mov Disord. 2018 Nov;33(11):1724-1733. Epub 2018 Nov 15 PubMed.
  7. . Heparan sulfate proteoglycans mediate internalization and propagation of specific proteopathic seeds. Proc Natl Acad Sci U S A. 2013 Aug 13;110(33):E3138-47. Epub 2013 Jul 29 PubMed.

Other Citations

  1. Therapeutics

Further Reading

Papers

  1. . Meta-analysis of 74,046 individuals identifies 11 new susceptibility loci for Alzheimer's disease. Nat Genet. 2013 Dec;45(12):1452-8. Epub 2013 Oct 27 PubMed.
  2. . Comprehensive research synopsis and systematic meta-analyses in Parkinson's disease genetics: The PDGene database. PLoS Genet. 2012;8(3):e1002548. Epub 2012 Mar 15 PubMed.
  3. . Common variation in the miR-659 binding-site of GRN is a major risk factor for TDP43-positive frontotemporal dementia. Hum Mol Genet. 2008 Dec 1;17(23):3631-42. PubMed.
  4. . The FAM171A2 gene is a key regulator of progranulin expression and modifies the risk of multiple neurodegenerative diseases. Sci Adv. 2020 Oct;6(43) Print 2020 Oct PubMed.
  5. . α-Synuclein species as potential cerebrospinal fluid biomarkers for dementia with lewy bodies. Mov Disord. 2018 Nov;33(11):1724-1733. Epub 2018 Nov 15 PubMed.
  6. . Heparan sulfate proteoglycans mediate internalization and propagation of specific proteopathic seeds. Proc Natl Acad Sci U S A. 2013 Aug 13;110(33):E3138-47. Epub 2013 Jul 29 PubMed.

Primary Papers

  1. . Neuronal FAM171A2 mediates α-synuclein fibril uptake and drives Parkinson's disease. Science. 2025 Feb 21;387(6736):892-900. Epub 2025 Feb 20 PubMed.