Vaccinating mice against synuclein

The research record

This is the project's file for this landmark, as written and checked. Sources are listed at the end; anything not yet confirmed is marked "Not yet verified".

Who

Eliezer Masliah (UC San Diego) and 11 co-authors, including E. Rockenstein, P. Seubert, D. Games and D. Schenk (the earlier text names Elan Pharmaceuticals scientists; their affiliations were not re-read) (Neuron, June 2005). The abstract lists NIA grant support.

What happened

Mice that overproduce human alpha-synuclein were vaccinated with the protein (Neuron 46:857–868). In mice that made high-affinity antibodies, aggregated synuclein built up less in nerve cell bodies and endings and there was less nerve damage. The antibodies recognised abnormal synuclein on the neuronal membrane and promoted breakdown of aggregates, probably through lysosomes. An applied labelled synuclein antibody had similar effects.

Integrity concern. Alzforum lists this paper among the Masliah papers flagged in a 2024 dossier (seven flags about western blots and microscopy images). In September 2024 NIH said it had found falsification or fabrication of figure panels in two Masliah publications and did not name them. Alzforum does not tie this paper to that finding, but Neuron has since attached an expression of concern to it, dated 5 Nov 2025 (DOI 10.1016/j.neuron.2025.10.012; seen in Crossref and the GetFTR record, rechecked 23 Oct 2026; the notice text was not readable). I found no retraction or correction. Other Masliah synuclein vaccine and antibody papers have since been retracted (the 2011 passive-immunisation paper in landmark 204 and an npj Vaccines DNA-vaccine paper, retracted March 2025). The mouse result has not been independently re-checked here, so treat the 2005 data with caution; the human trials stand on their own results.

The vaccines tested in people:

• PD01A (AFFiRiS), a 24-person safety trial reported in 2020 (landmark 324).
• ACI-7104 (AC Immune), a Phase 2 with safety and antibody data in 2026 (landmark 325).
• UB-312 (Vaxxinity) (landmark 183).

Why it moved us toward a cure

A vaccine would make the body produce the antibody continuously, which could be cheaper and longer lasting than infusions like prasinezumab (landmark 63). It rests on alpha-synuclein being the main protein of Lewy bodies (landmark 5) and on mouse models that overproduce it (landmark 31).

Outcome

Advanced to human trials, though the founding paper is now under a cloud (see above). No vaccine has shown it slows Parkinson's. Passive antibodies have had mixed results (64).

Sources

• Masliah et al., Neuron 2005;46(6):857–868, DOI 10.1016/j.neuron.2005.05.010, PMID 15953415 (abstract via PubMed) · Alzforum dossier list · ScienceDaily
• Alzforum, "Data fabrication ousted NIA neuroscience director Eliezer Masliah" (NIH finding on two unnamed papers; dossier of 132 papers)
• Expression of concern on Masliah et al., Neuron 2005, DOI 10.1016/j.neuron.2025.10.012 (dated 5 Nov 2025 in Crossref; notice text not read)
• Not yet verified: which two papers the NIH finding covers; what the 2005 paper's expression of concern says, and whether the paper will be corrected or retracted; the NIH statement itself (the NIH page blocked automated reading).