A protein associated with Parkinson’s disease can move from one brain cell to another. Yale researchers have now identified an interacting pair of proteins that appears to help it cross the cell membrane and contribute to neuronal damage.
The proteins, mGluR4 and NPDC1, are found on the surfaces of dopamine-producing neurons in the substantia nigra, the brain region whose deterioration causes many of Parkinson’s characteristic movement problems. In laboratory cells and mouse models, both proteins bound and internalized fibrils of misfolded α-synuclein.
The findings do not show that Parkinson’s can be halted in people, and they do not mean the disease is infectious. They do, however, identify a specific molecular interaction that researchers may eventually be able to disrupt with a drug.
The study in Nature Communications was conducted by a Yale School of Medicine team led by co-first authors Azucena Perez-Canamas and Mingming Chen, with neurologist and neuroscientist Stephen Strittmatter as the corresponding author.
A protein that can spread from cell to cell
Parkinson’s is best known as a movement disorder marked by tremor, stiffness, slowed motion and problems with balance. These symptoms are closely tied to the loss of dopamine-producing neurons in the substantia nigra.
Inside affected neurons, the protein α-synuclein can misfold and collect into abnormal structures. Evidence accumulated over the past two decades suggests that misfolded α-synuclein can also be released, taken up by neighbouring cells and encourage additional proteins to misfold.
Researchers often call this “prion-like” propagation because of the way one malformed protein can seed further abnormalities. That comparison describes a molecular process, not an infection: no bacterium, virus or other pathogen is involved, and Parkinson’s is not considered contagious.
A central question has therefore been what allows extracellular α-synuclein fibrils to bind to neurons and enter them.
A screen of 4,401 membrane proteins
The Yale team screened 4,401 genes encoding membrane-associated proteins. The researchers engineered cells to display individual proteins and then tested whether misfolded α-synuclein fibrils would bind to them.
The screen identified 16 previously unrecognized binding proteins. The researchers concentrated on mGluR4 and NPDC1 because both are expressed by the nigral dopamine neurons that are particularly vulnerable in Parkinson’s disease.
Further experiments suggested that each protein can bind and help internalize α-synuclein. The evidence also indicated that mGluR4 and NPDC1 interact with one another and that the resulting complex may be especially important to the process.
That makes “two independent doorways” an overly simple description. The study points instead to two components of a molecular entry system whose effects overlap and interact.
What the mouse experiments actually showed
The researchers next examined genetically modified mice in two experimental models.
In one model, α-synuclein fibrils were injected into the striatum, setting off changes that eventually damage dopamine-producing neurons. Removing either Grm4, the gene encoding mGluR4, or Npdc1 protected a substantial number of those neurons.
Importantly, the abnormal phosphorylated α-synuclein deposits still accumulated near the injection site. The gene deletions therefore did not simply prevent every visible protein deposit from forming. They appeared to weaken the connection between the deposits and subsequent neuronal loss.
In a separate model involving mice that produced a disease-associated form of human α-synuclein, the results varied by gene. Removing Grm4 improved survival and some measures of movement. Removing Npdc1 improved certain motor outcomes but did not produce a statistically significant improvement in survival or in every behavioural test.
The broadest benefit appeared when the researchers reduced both parts of the system. Combined partial or complete depletion of mGluR4 and NPDC1 improved survival, preserved spinal motor neurons and reduced several movement impairments.
Why the result matters for treatment
Most current Parkinson’s treatments manage symptoms rather than stopping the underlying loss of neurons. Levodopa remains one of the most effective ways to replenish dopamine and improve movement, but its benefits can fluctuate, and long-term treatment can be accompanied by involuntary movements known as dyskinesia.
Experimental strategies are attacking the disease process from several directions. Some aim to keep α-synuclein from misfolding or aggregating. Others attempt to clear abnormal proteins, influence immune activity or alter biological pathways outside the brain.
For example, a 2026 study in Parkinsonian rats examined whether nardosinone could affect levodopa-induced dyskinesia through changes in the microbiota–gut–brain axis. That work was also preclinical and does not establish that a gut-directed intervention modifies Parkinson’s progression in people.
The Yale study targets a different stage: the interaction that helps extracellular α-synuclein enter vulnerable neurons.
mGluR4 may be the more immediately familiar drug target because it belongs to a well-studied family of glutamate receptors. The study’s authors note that a previous mGluR4-modulating compound reached a phase 2 trial and appeared acceptably safe, although its short-term symptomatic benefit was not statistically significant and its potential to alter disease progression was not tested.
Deleting a gene in a mouse is also very different from safely blocking a protein in a person. A viable therapy would need to disrupt the harmful α-synuclein interaction without interfering excessively with the proteins’ normal functions.
The scale of the problem
The Parkinson’s Foundation estimates that more than 1.1 million people in the United States currently live with Parkinson’s disease and that nearly 90,000 new US cases are diagnosed each year. Worldwide, the total exceeds 10 million.
Age is the strongest established risk factor, so the number of patients is expected to increase as populations grow older. That makes the absence of a proven disease-modifying treatment increasingly consequential.
What comes next
The next challenge is to determine whether the mGluR4–NPDC1 interaction can be disrupted pharmacologically rather than genetically. Researchers will also need to establish which parts of the complex should be targeted, whether a drug can reach the relevant brain tissue and what happens when the pathway is altered for long periods.
Human Parkinson’s disease is considerably more complicated than either of the mouse models used in the study. Many therapies that protect neurons in animals have failed in clinical trials, particularly when treatment begins after substantial damage has already occurred.
Even so, the study narrows an important question. Instead of referring broadly to the cell-to-cell spread of α-synuclein, researchers now have two interacting surface proteins they can manipulate and investigate.
A more precise view of Parkinson’s progression
The finding is part of a broader attempt to understand neurodegenerative disease as a process that unfolds not only inside individual cells but also through interactions between cells.
Misfolded proteins associated with Alzheimer’s, Huntington’s disease, ALS and Parkinson’s can move through tissue and promote further misfolding. Researchers are consequently paying close attention to the receptors, transport pathways and extracellular interactions that make that movement possible.
mGluR4 and NPDC1 are not yet proven treatment targets in humans. But by identifying a plausible entry mechanism and showing that altering it can protect neurons in mice, the Yale team has provided a more concrete place for the next round of Parkinson’s research to begin.