Single-cell map of the Parkinson's midbrain

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

Semra Smajić, Carlos Prada-Medina, Zied Landoulsi and colleagues at the Luxembourg Centre for Systems Biomedicine (University of Luxembourg) and the University of Lübeck. The corresponding authors are Malte Spielmann (Lübeck, last author) and Anne Grünewald (Luxembourg). Funding came mainly from the Luxembourg National Research Fund (FNR) and a donation from Le Foyer Assurances Luxembourg, plus German research grants for Spielmann.

What happened

Using single-nucleus RNA sequencing, which reads each cell's active genes, the team profiled 41,435 nuclei from the midbrains of six people with idiopathic Parkinson's disease and five age- and sex-matched controls. They saw: a cluster of neurons with high CADPS2 and low tyrosine hydroxylase present only in Parkinson's, which laser-microdissection suggested were dysfunctional dopamine neurons; more numerous, more amoeboid (activated-looking) microglia with a pro-inflammatory profile including IL1B and GPNMB; fewer oligodendrocytes; and stress markers in astrocytes (S100B, CD44). Parkinson's risk variants mapped onto both glial and neuronal expression patterns. Brain 145(3):964–978, April 2022 (DOI 10.1093/brain/awab446; PMID 34919646). The authors saw no significant loss of dopamine neurons in their data and call the CADPS2 cluster small, needing a larger sample.

Why it moved us toward a cure

It took the old observation of activated microglia near dying neurons (landmark 52) and the inflammation pathways (landmark 55, landmark 53, landmark 38) into human cell-by-cell data, and offered candidate markers and targets for immune-modulating therapies. A separate human atlas (landmark 99) also reported expanded GPNMB-expressing microglia, according to Alzforum's summary.

Outcome

Open. The study is small (11 donors) and postmortem, and no correction or retraction is recorded (Crossref, Europe PMC), so it shows association, not whether inflammation causes neuron loss or follows it. The authors call for studying immune-modulating treatments, but none has been shown to slow Parkinson's. Whether the Parkinson-specific neuron state is a stage before death or an artefact remains to be tested.

Sources

• Smajić et al., Brain 2022;145:964–978, DOI 10.1093/brain/awab446 (full text PMC9050543) · Luxembourg repository · Alzforum citation entry
• medRxiv preprint
• Alzforum on Kamath et al. (GPNMB microglia)
• Not yet verified: per-cell-type counts; whether other single-cell studies replicated the CADPS2 neuron state (Alzforum's piece on Kamath et al. says Smajić had too few dopamine neurons to resolve subsets).