Hidden exitron splicing creates shorter UBQLN2 proteins linked to ALS
Researchers find human UBQLN2 hides a cryptic exitron whose removal yields unstable, truncated protein isoforms and is elevated in male ALS motor cortex.
Summary
UBQLN2 is an X-linked ALS gene that was thought to be expressed from a single exon with no splicing. This preprint shows that human UBQLN2 mRNA contains a cryptic coding intron, called an exitron, that can be spliced out. The spliced transcript is translated from a downstream methionine (M243), giving an N-terminally truncated protein that is less stable, targets the proteasome poorly, and aggregates differently when ALS mutations are added. The splicing factors SRSF1 and PTBP1 regulate the event by binding near the splice donor, and TDP-43 appears to act as an indirect repressor. In motor cortex RNA-seq from male ALS patients, UBQLN2 splicing was higher and inversely tracked total UBQLN2 expression. The work suggests splicing helps tune UBQLN2 dosage and function, with possible relevance to UBQLN2-linked proteinopathy.
Detailed Summary
Mutations in UBQLN2, a protein-quality-control gene on the X chromosome, cause dominant X-linked ALS and frontotemporal dementia. UBQLN2 is a single-exon gene, so it was assumed to produce one transcript and one protein. Whether hidden regulation of its expression contributes to disease has been unclear, and this question matters because splicing dysregulation is a recognized feature of ALS and other neurodegenerative diseases.
The authors combined long-read sequencing data (PacBio and Nanopore) with RT-PCR, qRT-PCR, Sanger sequencing, splicing reporter constructs, and mutagenesis. They used RNA immunoprecipitation and shRNA knockdown or overexpression of candidate splicing regulators in HeLa and HEK293T cells. They also analyzed iPSC-derived excitatory neurons and reanalyzed Target ALS short-read RNA-seq from medial motor cortex, filtering for RNA integrity (RIN ≥ 5.5) because degradation can bias 5' splicing measurements.
The central finding is an unannotated exitron, a protein-coding intron spanning the 5' UTR and the start of the coding sequence. Removing it yields a spliced transcript (UBQLN2-Sp) in which the normal start codon is lost, so translation begins at M243. The resulting isoform (UBQLN2-M243) is less stable and less able to engage the proteasome. When ALS mutations were introduced, it also showed altered aggregation behavior. Exitron splicing appears to be primate-specific, and the 5' UTR, splice sites, and a series of polypyrimidine tracts act as cis-regulatory elements. SRSF1 and PTBP1 bind motifs near the splice donor and control the event. TDP-43 appeared to act as an indirect repressor, since its knockdown altered splicing without clear direct binding of the same kind.
In medial motor cortex from male ALS-spectrum patients, the spliced isoform fraction was elevated compared with non-neurological controls. It was also inversely correlated with overall UBQLN2 expression, suggesting splicing may lower functional full-length UBQLN2 dosage. Together, the findings propose alternative splicing as a previously unrecognized layer of UBQLN2 regulation that could influence proteostasis and the behavior of mutant protein.
Several caveats apply. This is a preprint, and most mechanistic work was done in cell lines and reporter systems rather than patient tissue or animal models. The patient analysis relies on public bulk short-read RNA-seq with imputed RIN values and subgroup restrictions, so it shows association rather than causation. Whether the truncated isoform is made at meaningful protein levels in human neurons, and whether it contributes to disease, remains to be shown.
Key Findings
- Human UBQLN2 contains a cryptic exitron spanning the 5' UTR and start of the coding sequence; splicing it out yields a distinct transcript.
- The spliced transcript is translated from M243, producing a truncated isoform with reduced stability and weaker proteasome targeting.
- SRSF1 and PTBP1 regulate exitron splicing via splice donor-proximal motifs; TDP-43 acts as an indirect repressor.
- Spliced UBQLN2 was elevated in male ALS medial motor cortex and inversely correlated with total UBQLN2 expression.
- Exitron splicing appears primate-specific and depends on 5' UTR polypyrimidine tracts and splice-site sequences.
Methodology
The study combined long-read and short-read RNA-seq analysis with RT-PCR, qRT-PCR, minigene splicing reporters, site-directed mutagenesis, RNA immunoprecipitation, and shRNA knockdown or overexpression of RNA-binding proteins in human cell lines and iPSC-derived neurons. Patient relevance was assessed by quantifying spliced isoforms with RSEM in Target ALS medial motor cortex RNA-seq, filtered for RNA integrity, and comparing ALS-spectrum and control samples by sex.
Study Limitations
This is a non-peer-reviewed preprint, and much of the mechanistic evidence comes from cell lines and reporters rather than patient neurons or animal models. The patient data are bulk RNA-seq associations with imputed RNA quality metrics and show correlation, not causation. The functional consequences of the M243 isoform in human disease remain unproven.
Enjoyed this summary?
Get the latest longevity research delivered to your inbox every week.
Enter your email to subscribe:
