Summary

Genetic ablation of the Unc13a cryptic exon preserved memory in a mouse model of TDP-43 dysfunction, supporting UNC13A as a therapeutic target for related dementias.

Contents

The molecular problem

TDP-43 is an RNA-binding protein that normally prevents the splicing machinery of a cell from including so-called "cryptic exons" — segments of DNA that lie within genes but are not meant to appear in mature messenger RNA. When TDP-43 loses its function, whether because the protein aggregates into insoluble clumps or because the gene that encodes it is disrupted, these cryptic exons are aberrantly spliced into mRNA. The resulting proteins are frequently truncated or non-functional.

One of the most consequential cryptic targets of TDP-43 is UNC13A. The UNC13A protein is essential for synaptic vesicle release, the process by which neurons expel neurotransmitters to communicate with one another. When TDP-43 dysfunction causes a cryptic exon to be included in UNC13A mRNA, the resulting protein cannot support normal synaptic transmission, and communication between neurons is impaired.

This molecular pathology is thought to underlie frontotemporal dementia and limbic-predominant age-related TDP-43 encephalopathy (LATE), two neurodegenerative dementias that currently have no effective therapy.

What the researchers did

To test whether blocking cryptic splicing could protect cognitive function, the researchers used a mouse model in which the TDP-43 gene was selectively deleted from forebrain neurons. This model, the authors report, recapitulates features of TDP-43 dysfunction observed in the early stages of human disease.

In a subset of these mice, the team additionally removed the cryptic exon of Unc13a by genetic engineering, preventing the abnormal splice from occurring while leaving the rest of the gene intact. Memory was then assessed in three groups: normal control mice, TDP-43 knockout mice, and TDP-43 knockout mice in which the Unc13a cryptic exon had been ablated.

What they found

TDP-43 knockout mice displayed memory deficits relative to normal controls. However, when the Unc13a cryptic exon was additionally removed in the same animals, memory performance was preserved to levels comparable with control mice. The authors state that genetic ablation of the Unc13a cryptic exon "solely" in TDP-43 knockout mice was sufficient to preserve cognition.

The study also examined a broader intervention. When cryptic splicing of multiple TDP-43 targets was prevented — not limited to UNC13A — animals showed additional protection against neuron loss beyond what was achieved by targeting UNC13A alone. This suggests that while blocking UNC13A cryptic splicing may be sufficient to preserve memory function, suppressing cryptic splicing more broadly may be required to protect neuronal survival.

Why this matters

The results provide experimental evidence that a relatively focused molecular intervention — preventing the splicing of a single cryptic target — can maintain cognitive function in the setting of TDP-43 pathology. For therapeutic development, this implies that strategies aimed specifically at blocking UNC13A cryptic splicing might preserve memory even if other consequences of TDP-43 dysfunction are not fully corrected.

The authors conclude that their findings "strongly support strategies designed to repress cryptic splicing of multiple targets of TDP-43, including UNC13A" for optimal outcomes, because preserving memory may not require addressing every downstream effect of TDP-43 loss, but protecting neuronal survival may.

Evidence and limitations

The work was conducted in a mouse model with forebrain-neuron-specific TDP-43 knockout. The study is a preprint posted on bioRxiv on September 15, 2026, and has not yet undergone peer review. Specific experimental details, including sample sizes and statistical methods, are not described in the available abstract. The authors declare no competing interests. The research was supported by National Institutes of Health grants NS095969, NS115608, NS115161, and NS129878.

Sources