The shortest version of this story: one missing working copy of a gene called SRRM2 seems to scramble how brain cells edit RNA, tune synapses, and wrap wiring in a mouse model of schizophrenia. The interesting version takes a bit longer.

Meet SRRM2, the Cellular Copy Editor

SRRM2 helps organize nuclear speckles, little work zones inside the cell nucleus where RNA messages get processed before they become proteins [2]. Think of RNA splicing as movie editing: cut the boring footage, stitch together the useful scenes, and pray the final cut is not scenes 4, 19, and 3 duct-taped together.

In this Cell Reports study, Aryal and colleagues used mice with only one working copy of Srrm2, a situation called haploinsufficiency. That is genetic bean-counting for "half the usual dose is not enough." Rare loss-of-function variants in human SRRM2 have been tied to schizophrenia risk and neurodevelopmental disorders, so the team asked the obvious question: what goes wrong in the brain when this RNA-splicing scaffold is short-staffed? [1,3,4]

The Synapse Group Chat Got Weird

The mice did not show one tiny molecular hiccup. They showed a whole Slack outage.

Across several brain regions, neuronal and glial cells changed gene-expression programs linked to synapses, translation, mitochondria, and DNA-binding. At the synapse, the team found a drop in the gamma isoform of SynGAP, a protein that helps regulate how strongly neurons respond to incoming signals. Synapses are neuronal text messages, and SynGAP helps decide whether a message deserves "lol," "seen," or "please stop firing at 2 a.m."

They also found altered splicing and abundance of Agap3, a SynGAP-interacting protein involved in synaptic signaling. Even better, in the scientifically annoying but useful sense of "better," SRRM2-deficient human iPSC-derived neurons showed conserved AGAP3 splicing problems. That does not prove the mouse findings map neatly onto people, but it does mean the weirdness is not just a mouse-only side quest.

The Cable Crew Was Missing Too

Then came the glia. The Srrm2+/- mice had fewer oligodendrocytes, especially in the striatum, plus lower levels of myelin-related RNAs and proteins. Oligodendrocytes are the brain's cable-insulation crew. They wrap axons in myelin so electrical signals move quickly and cleanly. If neurons are the rock stars, oligodendrocytes are the exhausted roadies making sure the amps do not burst into flames.

This matters because schizophrenia research has been circling myelin and glial biology for years. Genetics keeps pointing to synapses, while brain-imaging, transcriptomic, and postmortem studies also point to altered support cells and wiring insulation. The new Srrm2 mouse data links those worlds: RNA splicing, synaptic proteins, glial changes, and circuit readouts in one genetic model.

Even Sleep Had Receipts

The mice also moved less, had impaired startle responses, and showed reduced sleep spindles on EEG. Sleep spindles are brief bursts of brain rhythm during non-REM sleep, the kind of tiny electrical drumroll your brain performs while you are unconscious and, ideally, not thinking about grant deadlines. Human schizophrenia studies have reported reduced spindle density too, making a neat bridge from molecules to measurable brain activity [5].

Look. Mice are not tiny people with lab coats and committee meetings. A mouse model cannot capture schizophrenia's full human complexity, and this study does not hand anyone a diagnostic test or treatment. But it does something useful: it turns a rare genetic clue into a chain of testable biology.

If these findings hold up and expand, they could help researchers ask sharper questions. Are specific splicing errors early drivers of disease risk? Could SynGAP-gamma or AGAP3 isoforms serve as molecular readouts? Do oligodendrocyte and myelin defects worsen circuit instability? Could sleep EEG mark some biology-defined subtypes? That is the good stuff: not "one gene explains schizophrenia," because no, absolutely not, put that headline back in the drawer. More like "one rare gene gives us a flashlight in a very cluttered basement."

And in neuroscience, a flashlight is not nothing.

References

1. Aryal S, et al. Srrm2 haploinsufficiency drives SynGAP-gamma reduction, Agap3 mis-splicing, and oligodendrocyte deficits in a genetic mouse model of schizophrenia. Cell Reports. 2026;45(6):117436. doi:10.1016/j.celrep.2026.117436. PMID: 42189682.
2. Xu S, et al. SRRM2 organizes splicing condensates to regulate alternative splicing. Nucleic Acids Research. 2022;50:8599-8614. doi:10.1093/nar/gkac669. PMCID: PMC9410892.
3. Farsi Z, Sheng M. Molecular mechanisms of schizophrenia: Insights from human genetics. Current Opinion in Neurobiology. 2023;81:102731. doi:10.1016/j.conb.2023.102731. PMID: 37245257.
4. Liu D, et al. Schizophrenia risk conferred by rare protein-truncating variants is conserved across diverse human populations. Nature Genetics. 2023;55:369-376. doi:10.1038/s41588-023-01305-1. PMCID: PMC10011128.
5. Kozhemiako N, Wang J, Jiang C, et al. Non-rapid eye movement sleep and wake neurophysiology in schizophrenia. eLife. 2022;11:e76211. doi:10.7554/eLife.76211. PMCID: PMC9113745.

Disclaimer: The image accompanying this article is for illustrative purposes only and does not depict actual experimental results, data, or biological mechanisms.