question exonuclease/phosphodiesterase

[gocentral]

How do we correctly describe:

Uniprot 3'-5' exoribonuclease that posttranscriptionally removes uridine and adenosine nucleosides from the 3' end of spliceosomal U6 small nuclear RNA (snRNA), directly catalyzing terminal 2', 3' cyclic phosphate formation

SGD Putative phosphodiesterase specific for U6 snRNA 3' end modification; trims the 3' poly(u) tract to leave a terminal 3' phosphate; human homolog, hUSB1 (aka C16orf57) produces a 2',3' cyclic phosphate;

I see it variously described as:

GO:0000175 3'-5'-exoribonuclease activity http://www.uniprot.org/uniprot/Q9BQ65 USB1 human

GO:0008081 phosphoric diester hydrolase activity SGD USB1

Should there be a more specific child term to capture both of these terms?

Should there be a grouping term here for nucleotide phosphodiesterases? http://www.ebi.ac.uk/QuickGO/GSearch?q=phosphodiesterase

Reported by: ValWood

Original Ticket: geneontology/ontology-requests/11346

[gocentral]

This is related, there should be a realtionship between these 2 terms, but I think that would be addressed by the above GO:0008081 phosphoric diester hydrolase activity GO:1990174 phosphodiesterase decapping endonuclease activity

Original comment by: ValWood

[gocentral]

Hi Val,

Yes, we may need more links or terms. I’ll ask Rhea curators to comment, as these specific nucleotide phosphodiesterase activities are not covered by the EC or Rhea yet (except for this one: http://www.chem.qmul.ac.uk/iubmb/enzyme/EC3/1/16/1.html).

Thanks,

Paola

Original comment by: paolaroncaglia

[gocentral]

Note for self: Rhea ticket here: https://sourceforge.net/p/rhea-ebi/reaction-requests-updates/60/

Original comment by: paolaroncaglia

[gocentral]

Kristian writes:

I have been looking at your request and the available evidence (that I could find), and I have come up with the following:

You requested a:

• a broad entry to define nucleotide phosphodiesterases as child of EC 3.1.4 Phosphoric Diester Hydrolases In Rhea, we only create reactions that are actually catalysed by biochemically characterized enzymes, so the generic reactions you do find, are catalysed by enzymes with a broad specificity. So the notion of parent reactions and child reactions are not implemented in Rhea, and I can thus not create this reaction.
• a more specific entry to define nucleotide endonucleases As above, I can only create the reaction if you can point me to literature describing the characterisation of such a reaction. You should be aware that EC 3.1.21 do cover endonucleases, but these EC numbers describe the catalysed reactions using free text which we cannot model in Rhea.
• a more specific entry to define 3'-5'-nucleotide phosphodiesterases (exonucleases) I have created one reaction covering the activity of USB1 and MPN1 (from S. pombe). After carefully reading the available literature I do not believe the yeast USB1 is different form the human and pombe enzymes, so I have not created a specific reaction for this enzyme producing a free 3' OH group, but only the reaction producing 2',3'-cyclic phosphate. This reaction will become public in the May release tentatively scheduled for May 25. It has the ID RHEA:46052 for the master reaction and RHEA:46052 for the LR, RHEA:46054 for the RL and RHEA:46055 for the BI reactions.

Original comment by: paolaroncaglia

[ValWood]

My query in this one was fixed independently by Rob Nash yesterday.

Hi Val,

I took a look at your annotation and the paper. It seems that the older paper we had used came to a different conclusion based on the evidence that had at the time. However, I think the evidence presented in the paper you used was solid and I think your choice of term was good, so I have removed our current function annotation, and replaced if with an ISS to mpn1. I have also added the paper to our db and updated our description. I don’t think the two terms should be related, so I don’t think one should be an ancestor of the other.

Cheers, Rob