obs: tRNA-pseudouridine synthase activity?

[gocentral]

Hi,

I was wondering if there is really a good reason to have a separate term for 'tRNA-pseudouridine synthase activity' (GO:0016439) in the function ontology. There isn't really anything molecularly different between tRNA, rRNA, snRNA, snoRNA, etc and it's known that some of the enzymes that do this act on multiple 'types' of RNAs (PMID:10022901). So, for the function ontology, it seems to me that just having the parent term 'pseudouridine synthesis' (GO:0001522) would be sufficient. All the various granular terms in the process ontology to indicate specific substrates makes sense to me, but I wonder if it might not be a good idea in function.

If we do want to keep the tRNA specific term, then we should probably add terms for the corresponding 'rRNA-...' and 'snRNA...', and maybe eventually other specific terms for other types of RNAs.

-Karen

[Term] id: GO:0009982 name: pseudouridine synthase activity namespace: molecular_function def: "Catalysis of the reaction: RNA uridine = RNA pseudouridine. Conversion of uridine in an RNA molecule to pseudouridine by rotation of the C1'-N-1 glycosidic bond of uridine in RNA to a C1'-C5." [EC:5.4.99.12, GOC:mah] comment: Note that this term should not be confused with 'pseudouridylate synthase activity ; GO:0004730', which refers to the formation of free pseudouridine from uracil and ribose-5-phosphate. subset: gosubset_prok xref: EC:5.4.99.- is_a: GO:0016866 ! intramolecular transferase activity

[Term] id: GO:0016439 name: tRNA-pseudouridine synthase activity namespace: molecular_function def: "Catalysis of the reaction: tRNA uridine = tRNA pseudouridine. Conversion of uridine in a tRNA molecule to pseudouridine by rotation of the C1'-N-1 glycosidic bond of uridine in RNA to a C1'-C5." [EC:5.4.99.12] comment: Note that this term should not be confused with 'pseudouridylate synthase activity ; GO:0004730', which refers to the formation of free pseudouridine from uracil and ribose-5-phosphate. subset: gosubset_prok synonym: "tRNA-pseudouridine synthase I activity" NARROW [] xref: EC:5.4.99.12 xref: MetaCyc:TRNA-PSEUDOURIDINE-SYNTHASE-I-RXN is_a: GO:0009982 ! pseudouridine synthase activity

Massenet S, Motorin Y, Lafontaine DL, Hurt EC, Grosjean H, Branlant C (1999) Pseudouridine mapping in the Saccharomyces cerevisiae spliceosomal U small nuclear RNAs (snRNAs) reveals that pseudouridine synthase pus1p exhibits a dual substrate specificity for U2 snRNA and tRNA. Mol Cell Biol 19(3):2142-54 PMID:10022901

Reported by: krchristie

Original Ticket: "geneontology/ontology-requests/4283":https://sourceforge.net/p/geneontology/ontology-requests/4283

[gocentral]

Logged In: YES user_id=546388 Originator: NO

This has come up before because people don't want to "lose the information". Actually, I seem to remember that there was a request somewhere for splitting this into several (reaction at U55, reaction in the anticodon (middle and other, etc.). A process term might not be appropriate because the reaction is one step (unlike the rRNA, etc. where a template is needed, so that one could conceivably think of the process as template binding, then catalysis by enzyme, etc.).

However, I agree that the actual reaction is the swap of a N-C bond for a C-C bond by "rotating" the base (not what literally happens, but you get the jist). THe specificites get really confusing

The only entry in EC is 5.4.99.12 tRNA-pseudouridine synthase I. (AN: tRNA pseudouridylate synthase I. tRNA-uridine isomerase.)

The "I" modifies uridylate residues at positions 38, 39 and 40 of many tRNAs, and also 27/28 of others (yeast), aka Pus1. However, it is also known that "Pus1" also modifies U2

But there is synthase II, ( Pus3 Pus4, which does the U55 to PseudoU55 ... Pus 5-8??? Pus9 Mitochondrial tRNA pseudouridine synthase involved in pseudouridylation of mitochondrial tRNAs at position 32

I would opt for just indicating that the modification occurs to the base incorporated into the RNA chain to distinguish is from the other EC entry 4.2.1.70 Pseudouridylate synthase. (AN: Pseudouridine synthase. Uracil hydrolyase.)

Uracil + D-ribose 5-phosphate <=> pseudouridine 5'-phosphate + H(2)O

Original comment by: hdrabkin

[gocentral]

Logged In: YES user_id=473890 Originator: YES

Harold,

I'm not really sure what you're recommending. Could you clarify?

thanks,

-Karen

Original comment by: krchristie

[gocentral]

Logged In: YES user_id=546388 Originator: NO

I think we should try to keep some specificity (ie, tRNA, rRNA) if we can. Perhaps we can specify even "dual specificity" RNA pseudouridine synthase activity to handle some. I know it's the same reaction, but we have differentiated between different substrates in other enzyme activites. The EC is just not complete.

Original comment by: hdrabkin

[gocentral]

• assigned_to: nobody --> hjdrabkin

Original comment by: mah11

[gocentral]

Some but not all; for example, the one that does 55 in a tRNA (called TruB in E.coli: I think it's called Psu 2 in eukaryotes). We have other places in the ontology where the reaction is the same but the substrates are different and thus get a separate GO term. As you indicate, most likely the basis of the substrate selection is local RNA structure.

As processes, these are 1 step things, so I think turning them into processes would be in error.

We DO need to distinguish between the synthesis on an RNA molecule vs the formation of free pseudouridine from uracil and ribose-5-phosphate.

see M Del Campo, Y Kaya and J Ofengand Identification and site of action of the remaining four putative pseudouridine synthases in Escherichia coli. RNA 2001 7: 1603-1615 For a good summary.

Original comment by: hdrabkin

[gocentral]

REgarding "We DO need to distinguish between the synthesis on an RNA molecule vs the formation of free pseudouridine from uracil and ribose-5-phosphate." I was definitely NOT suggesting that those things should be merged.

-Karen

Original comment by: krchristie

[gocentral]

I don't think it's so much that the EC is not complete, as that it's not completely logical. It is often based on the specific experimental assay that was done, so if the researcher tested one particular substrate and named the activity by that substrate, that's what the EC represents. If someone else tests a different substrate, the same enzyme might get a second name. I don't necessarily think that's a good model for GO. I don't really mind having separate process terms for the different substrates, e.g. tRNA, rRNA, snoRNA, etc, but having multiple function terms for a single activity bugs me. I think it would make more sense to have one function term and it can have links to multiple process terms to represent that it is active in tRNA modification, rRNA modification, etc.

-Karen

Original comment by: krchristie

[gocentral]

I would also prefer to annotate to single function and different process in these cases, based on the fact that process is the level that most researchers need to be able to make a biological distinction and that it isn't so useful (and therefore necessary) for analyses to split function terns in this way. I can expand on this if necessary but I'm probably stating the obvious. (Functions are the only case where I wold advocate less granularity) Val

Original comment by: ValWood

[gocentral]

The other thing you'll be able to do Real Soon Now(tm) is put a substrate designation, such as a SO ID, in column 16 for the function annotation.

m

Original comment by: mah11

[gocentral]

• status: open --> closed-out-of-date

Original comment by: paolaroncaglia

[gocentral]

comment: just need to make sure to distinguis the ones that occur when the uridine is in an RNA vs the process/activity that occurs in some organisms at the nucleoside/tide leve (I have the info buried somewhere for that one I think)

Original comment by: hdrabkin