Pseudouridine synthases modify human pre-mRNA co-transcriptionally and affect splicing

Background

Pseudouridylation is a wide-spread RNA modification, found in all kingdoms of life. This modification involves isomerization of uridine ring to pseudouridine (𝛹) which is catalyzed by pseudouridine synthases (Pus) enzymes. While four stand-alone PUS enzymes in yeast and humans have been reported to catalyze pseudouridylation of mRNA in response cellular stresses, such as heat shock, the timing and relevance of these modifications on the fate of mRNA are not known. This study used a combination of sequencing-based and in vitro and approaches to reveal known and novel 𝛹 sites on pre-mRNA and their effects on pre-mRNA processing.

Major findings

Pre-mRNA is co-transcriptionally psuedouridylated

While all previous studies have studied 𝛹 sites on poly(A)+ mRNA, the authors in this study enriched for chromatin associated unspliced pre-mRNA and subjected 11 replicates from HepG2 cells to Pseudo-seq. Briefly, Pseudo-seq involves treating the samples with the chemical chemical N- cyclohexyl-N′-beta-(4-methylmorpholinium) ethylcarbodiimide p-tosylate (CMC) which forms a bulky covalent adduct with 𝛹 sites. These modified sites create a block during reverse transcription which can then be detected as premature stops by reverse transcription. Pseudo-seq data revealed that the majority of the pseudouridines mapped onto introns, and several sites were found on chromatin-binding non-coding RNA like the small nucleolar RNAs (snoRNAs), small cajal body-specific RNAs (scaRNAs) and long non-coding RNAs lncRNAs.

Pre-mRNA 𝛹 sites are enriched near splice sites and overlap with RBP sites

To investigate the functional relevance of presence of 𝛹 sites on introns, the authors compared 𝛹 reads with RNA biding sites of 103 RBPs (RNA binding proteins) using existing eCLIP data, and found that several eCLIP clusters (RBP sites) overlapped with 𝛹 sites, of which majority were splicing factors. Moreover, the intronic 𝛹 sites were significantly enriched within 500nt of splice sites. Collectively, these observations prompted the authors to investigate the potential these intronic 𝛹 sites in regulating pre-mRNA splicing.

PUS1, RPUSD3 and PUS7 regulate global alternative splicing in vitro and in vivo.

Next, the authors employed an in vitro approach to demonstrate that PUS can pseudouridylate pre-mRNA and regulate alternative splicing. For this, they synthesized a pool of RNA containing pseudouridine sites flanked by 130nt of endogenous sequences and incubated it with one of the human PUS enzymes, PUS1, PUS7, PUS7L, PUS10, RPUSD2, RPUSD4, TRUB1, TRUB2 or HepG2 nuclear extract to catalyze in vitro pseudouridylation. They then subjected this pool to Pesudo-seq and identified PUS1, RPUSD4 and PUS7 as pre-dominant pre-mRNA pseudouridylating enzymes in vitro. Next, they individually depleted the levels of PUS1, PUS7 or RPDS4 by CRISPR/Cas9 knockout or siRNA mediated knockdown and found significant changes in alternative splicing of several genes.

Site-specific pseudouridylation by PUS7 directly affects slicing in vitro

In order to account of possible indirect effects stemming from knockout/knockdown, the authors developed an in vitro splicing reporter assay. They treated a chimeric two-exon transcript with intronic sequences containing PUS7 target splice sites under splicing conditions and assed percentage of splicing via RT-PCR gel electrophoresis. They reported that the presence of PUS7-sensitive 𝛹 site upstream of 3’ splice site enhanced splicing as compared to the unmodified site.

Conclusions

This preprint for the for the time reveals the presence of 𝛹 sites in introns, and the importance of these intronic 𝛹 sites and PUS enzymes in regulating alternative splicing. The presence of pseudo sites disposes the pre-mRNA towards a more stabilized secondary structure, alters RNA-RNA and/or RNA-protein interactions, and thus regulate alternative splicing of the target pre-mRNA.

Questions for the authors

1. Are the intronic 𝛹 sites dynamically regulated, e.g., in response environmental stimuli, as in the case of mRNA pseudouridylation?
2. Does differential expression of PUS proteins across various tissues correlate with the expression levels of their target isoforms?

 

Posted on: 25 September 2020

doi: https://doi.org/10.1242/prelights.24934

Read preprint (No Ratings Yet)