Back

Bone Morphogenetic Protein Pathway Modulates Parkinson's Disease Genetic Risk and Promotes Motor Recovery

Rajkovic, A.; Zhang, M.; Sahu, R.; Liu, L.; Mishra, M.; Komkov, D.; Rosenstein, I.; Kahn, J.; Friedel, R. H.; Gan-Or, Z.; Brodski, C.

2026-06-11 neuroscience
10.64898/2026.06.08.730818 bioRxiv
Show abstract

Progress toward disease-modifying Parkinsons disease (PD) therapies is hindered by limited understanding of pathogenic drivers and lack of therapies that restore the damaged dopaminergic (DA) neurons driving motor deficits. We investigated the role of the bone morphogenetic protein (BMP) pathway in PD using common and rare genetic variants across large-scale datasets. Single-variant analyses identified nominal associations with PD risk and onset age. A BMP polygenic risk score (PRS) was significantly associated with PD risk (OR = 1.21, empirical P = 1.0 x 10-), a result replicated in proxy-case analyses (OR = 1.14, empirical P = 1.0 x 10-) and remained significant after sensitivity testing. To test the results functional relevance, we genetically and pharmacologically inhibited BMP signaling in mice, which induced motor deficits and PD-like neuropathology. In a -synuclein preformed fibril (PFF) mouse model, BMP5 and BMP7 (BMP5/7) demonstrated neuroprotective effects when administered concurrently with -synuclein PFFs. Importantly, when delivered after PFFs-induced motor symptoms onset, BMP5/7 demonstrated neurorestorative effects, ameliorating both motor impairments and neuropathology. These findings provide first evidence that BMP signaling variations contribute to polygenic PD risk, identify a physiological role for this pathway in safeguarding against PD-related pathology, and demonstrate BMPs therapeutic potential for disease modification.

Matching journals

The top 1 journal accounts for 50% of the predicted probability mass.

50% of probability mass above

"Similar papers" are the closest papers from that journal in the model's embedding space. They show what the match is built on, but the ranking comes mostly from a classifier over the whole training set, not from these examples alone.