NTR: How to annotatate intron specific splicing

[ValWood]

I have a splicing paper and I don't know which GO term to use.

PMID:36095128

Title | Splicing of branchpoint-distant exons is promoted by Cactin, Tls1 and the ubiquitin-fold-activated Sde2. Authors | Anil AT, Choudhary K, Pandian R, Gupta P, Thakran P, Singh A, Sharma M, Mishra SK Publication date | 2022 09 12

Basically, the authors show very clearly that a set of 3 conserved proteins are required for intron-specific splicing. Sde2 (A ubiquitin fusion protein, ubiquitin is removed from the active form) (human SDE2) Tls1 (human C9orf78/TLS1) and Cay1 (human cactin)

In their absence constitutive splicing is fine, but introns with a longer branch site to acceptor distance are spliced inefficiently.

In humans TLS1 has been shown to facilitate alternative splicing but in fission yeast, it is probably only required to excise these "weak" introns with longer branch to acceptor distance. It is possible that this could be regulating splicing in some way but this isn't clear to me if it is regulation in the "GO" sense.

So which term would I use to describe this process? Just

GO:0045292 | mRNA cis splicing, via spliceosome doesn't seem sufficient. But I don't see anything suitable under here.

Any suggestions?

@pgaudet @vanaukenk @krchristie

Requirement is quite precise: This result indicated that Sde2-C becomes critical for using the 3′ss at ≥21 nt from the BP (Figure 3C).

[pgaudet]

Note that @Antonialock annotated human TLS1, maybe she has some insight into this?

[Antonialock]

Hmm yes, I only annotated to the general term "mRNA cis splicing, via spliceosome".

The UniProt text descriptions ended up looking slightly different for pombe and human, but perhaps "mRNA 3'-splice site recognition " would fit both descriptions, do you agree?

POMBE Plays a role in pre-mRNA splicing by facilitating excision of introns featuring long spacing between the branchpoint and 3'-splice site (PubMed:36095128). Assists the splicing of several components involved in chromatin organization, such as several shelterin complex subunits (PubMed:25245948, PubMed:36095128).

HUMAN Plays a role in pre-mRNA splicing by promoting usage of the upstream 3'-splice site at alternative NAGNAG splice sites; these are sites featuring alternative acceptor motifs separated by only a few nucleotides (PubMed:35241646). May also modulate exon inclusion events (PubMed:35241646). Plays a role in spliceosomal remodeling by displacing WBP4 from SNRNP200 and may act to inhibit SNRNP200 helicase activity (PubMed:35241646). Binds U5 snRNA (PubMed:35241646). Required for proper chromosome segregation (PubMed:35167828). Not required for splicing of shelterin components (PubMed:35167828).

[ValWood]

OK , I will annotate to GO:0045292 | mRNA cis splicing, via spliceosome for now. I have phenotypes for the branch distance part so it would be easy to update if new terms are added later.

[krchristie]

Is this term appropriate? The abstract mentions alternative splicing a couple times, so it seems likely to be a good term.

[Term] id: GO:0000380 name: alternative mRNA splicing, via spliceosome namespace: biological_process def: "The process of generating multiple mRNA molecules from a given set of exons by differential use of exons from the primary transcript(s) to form multiple mature mRNAs that vary in their exon composition." [GOC:krc, PMID:12110900] comment: Note that this process most commonly occurs in cis, selecting or skipping exons from the same primary transcript, but it has also been observed to occur in trans at low frequency, at least in some mammals. synonym: "alternative nuclear mRNA splicing, via spliceosome" EXACT [GOC:vw] synonym: "splice site selection" BROAD [] is_a: GO:0000398 ! mRNA splicing, via spliceosome

[ValWood]

Hmm, I think it could be used for human, but there isn't any evidence in pombe that there is alternative splicing controlled by this process. The targets appear to be genes involved in heterochromatin assembly, and seems likely that there is a regulatory role (controlled processing of sde2 to initiate correct splicing when these are components are required. The mispliced proteins are short and truncated truncated). For now I will stick with just splicing.

For the alternative splicing I noticed that it isn't under cis splicing which may explain why it was missed. There is a comment "Note that this process most commonly occurs in cis, selecting or skipping exons from the same primary transcript, but it has also been observed to occur in trans at low frequency, at least in some mammals."

Does anyone know the source of that comment and if it definitely results in functional proteins? I know that sequencing can occasionally produce chimeras with look like trans-spliced products, and it seems that by definition alternative splicing must be cis ( alternative trans-splicing would be just 'trans splicing' to as different exon.

Alternative splicing is a cellular process in which exons from the same gene are joined in different combinations, leading to different, but related, mRNA transcripts.

[krchristie]

For the alternative splicing I noticed that it isn't under cis splicing which may explain why it was missed. There is a comment "Note that this process most commonly occurs in cis, selecting or skipping exons from the same primary transcript, but it has also been observed to occur in trans at low frequency, at least in some mammals." does anyone know the source of that comment and if it definitely results in functional proteins. I know that sequencing can occasionally produce chimeras with look like trans-spliced products, and it seems that by definition alternative splicing must be cis ( alternative trans splicing woulfd be just 'trans splicing' to as different exon.

Considering the GO ID and the fact that I am cited as a definition dbxref, I am probably the source of that comment. It has been a very long time since I worked on those terms, so I no longer remember why I wrote it, but I'm sure there must have been indications in the literature that alternative splicing was occasionally observed in trans-splicing.

[ukemi]

PMID:14522953?

[krchristie]

@ValWood - The authors of the paper you're annotating seem to think that they are talking about alternative splicing:

Here, we report that Sde2 facilitates excision of introns with longer spacing between the BP and 3′ss (referred to as BP-distant 3′ss). This activity requires the cleavage of Sde2UBL by DUB and the free lysine of processed KSde2-C. Using splicing reporters made with the ura4 gene split by introns of varying distances between the BP and 3′ss, we searched for S. pombe deletion mutants of putative splicing factors. We identified Cactin/Cay1 and Tls1 as additional regulators of BP-distant splicing. These three splicing regulators are absent in budding yeast but are conserved from the fission yeast S. pombe to metazoans. Furthermore, by altering the spacing between the BP and 3′ss in selected targets, we show that introns with longer spacing between the BP and 3′ss require Sde2, Cay1 and Tls1 for efficient splicing. They regulate gene expression and alternative splicing of various heterochromatin factors, including the telomeric factor Rap1.

[ValWood]

Thanks! So there should probably be 2 terms alternative cis-splicing and alternative trans-splicing?

[ValWood]

I suppose it is "alternative splicing". But it seems a bit of a stretch because really it's just "intron retention" producing a nonfunctional product? Would we call that alternative splicing? That seems to happen all the time for "weak" introns. They get picked up by the mRNA surveillance machinery and degraded. I never considered those to be alternatively spliced products before, I always assumed that both products would need to be functional?

[krchristie]

I think alternative splicing just means changing the exon choice. I don't think there is a requirement that all the possible products be functional. I think that choosing between a splicing pattern that produces a full length functional product and one that is degraded by NMD because it has an in-frame stop codon is still alternative splicing. If I remember correctly, I've seen this kind of alternative splicing happens in mammals as well as the more commonly thought of alternative splicing that generates multiple products of different lengths, sometimes in the same gene.

[ValWood]

OK, in that case I will go with alternative splicing, but since I try to keep all our splicing annotation under "cis splicing" is there any objection if I add alternative cis-splicing and alternative trans-splicing?

Seems that it could be useful to separate these. The Drosophila example appears to use the allele on the opposite strand which is very funky and most likely involves different machinery?

[krchristie]

OK, in that case I will go with alternative splicing, but since I try to keep all our splicing annotation under "cis splicing" is there any objection if I add alternative cis-splicing and alternative trans-splicing?

Seems that it could be useful to separate these. The Drosophila example appears to use the allele on the opposite strand which is very funky and most likely involves different machinery?

I don't have any objection to having separate terms for alternative cis-splicing and alternative trans-splicing. However, if they occur by different mechanisms, then there probably isn't any reason to have a grouping term. A grouping term seems like it might just get mis-used because some annotators might not notice the more specific terms, even when they are appropriate.

I don't have any thoughts about whether it is true that they use different mechanisms, though I do recall there is a lot of commonality between the mechanisms of cis- and trans- splicing.

[ValWood]

I'll do a bit of digging and look at the existing annotations before I add anything.

[ValWood]

there is already a descendant GO:0000366 intergenic mRNA trans splicing

[krchristie]

there is already a descendant GO:0000366 intergenic mRNA trans splicing

This isn't explicitly stated to be alternative splicing, only a different type of trans splicing than its sibling term "mRNA trans splicing, SL addition" (GO:0045291)

[ValWood]

It (GO:0000366 intergenic mRNA trans splicing) is a descendant of alternative splicing though...so implicitly it is...(maybe shouldn't be)?

[krchristie]

It (GO:0000366 intergenic mRNA trans splicing) is a descendant of alternative splicing though...so implicitly it is...(maybe shouldn't be)?

One of the papers cited for the definition calls it trans-splicing in the abstract (the other is a review). I's suggest checking that paper and make sure that this term is appropriate for your needs or whether you need a new term.

[ValWood]

Some digging in this ticket so I will close this one as duplicate https://github.com/geneontology/go-ontology/issues/27611

[ValWood]

It seems I meant to close this as a duplicate but didn't