Back

Identification of RNA binding proteins that mediate a quality control mechanism of splicing

Arafat, M.; Gotea, V.; Umar, M. I.; Muharram, A.; Kamhi, E.; Agami, R.; Sperling, J.; Hafner, M.; Elnitski, L.; Sperling, R.

2025-07-23 genomics
10.1101/2025.07.20.665773 bioRxiv
Show abstract

Accurate splicing, which involves the controlled removal of non-coding sequences (introns) from precursor messenger RNAs (pre-mRNAs), is essential for producing correct mature mRNAs that encode functional proteins. Within pre-mRNAs, latent splice sites (LSSs) resemble proper splice sites but are usually not used because their activation can introduce in-frame STOP codons. The nuclear suppression of splicing (SOS) mechanism prevents the use of LSSs. Although the SOS mechanism is not fully understood, recent studies have identified initiator-tRNA and the NCL protein as key components. To discover additional regulators, we performed a genetic screen targeting RNA-binding proteins (RBPs) with an siRNA library and a luminescence reporter for latent splice site activation. This identified five RBPs -- ALYREF (THOC4), PPIE, DDX41, DHX38, and HNRNPA2B1 -- whose knockdown significantly increased LSS usage in the reporter. RNA-Seq analysis after knocking down each of these RBPs confirmed these results, showing widespread LSS activation in hundreds of mRNAs. Among these, we focused on ALYREF, a conserved protein involved in mRNA export and splicing. Using fPAR-CLIP, we found that U5 snRNA is ALYREFs main binding partner. Overexpressing ALYREF deletion mutants activated latent splicing, and affinity purification confirmed its interaction with U5 snRNA. These mutants exhibited different binding properties, highlighting the importance of specific structural elements within ALYREF in SOS regulation. Our findings reveal that nuclear RBPs play a key role in suppressing LSS activation and suggest that ALYREF has a novel role in maintaining splicing accuracy within the spliceosome, advancing our understanding of the SOS mechanism.

Matching journals

The top 2 journals account for 50% of the predicted probability mass.

50% of probability mass above

"Similar papers" are the closest papers from that journal in the model's embedding space. They show what the match is built on, but the ranking comes mostly from a classifier over the whole training set, not from these examples alone.