biotin-protein ligase activity terms - merge

[sjm41]

Below is the current hierarchy for 'biotin-protein ligase activity' and children, together with current xrefs):

 biotin-protein ligase activity (EC:6.3.4.-, MetaCyc:BIOTINLIG-RXN)
    |__biotin-[acetyl-CoA-carboxylase] ligase activity (EC:6.3.4.15, MetaCyc:BIOTINLIG-RXN, RHEA:11756)
    |__biotin-[methylcrotonoyl-CoA-carboxylase] ligase activity (EC:6.3.4.11, MetaCyc:6.3.4.11-RXN, RHEA:24376)
    |__biotin-[methylmalonyl-CoA-carboxytransferase] ligase activity (EC:6.3.4.9, MetaCyc:6.3.4.9-RXN, RHEA:23668)
    |__biotin-[propionyl-CoA-carboxylase (ATP-hydrolyzing)] ligase activity (EC:6.3.4.10, MetaCyc:6.3.4.10-RXN, RHEA:11204)
    |__biotin-[pyruvate-carboxylase] ligase activity (EC:6.3.4.-)

Issues here:

• EC:6.3.4.- is a grouping class for 'Other carbon--nitrogen ligases' so shouldn't be an xref on either of two terms above
• MetaCyc:BIOTINLIG-RXN is given as an xref to both the parent and a child term.
• "biotin-[pyruvate-carboxylase] ligase activity" has no specific EC/RHEA/MetaCyc xref
• The name and def of "biotin-[acetyl-CoA-carboxylase] ligase activity" (GO:0004077) don't match those of its EC:6.3.4.15/RHEA:11756 xrefs: EC name is now "biotin--[biotin carboxyl-carrier protein] ligase" and def is "ATP + biotin + L-lysyl-[protein] <=> AMP + diphosphate + H(+) + N(6)-biotinyl-L-lysyl-[protein]"

PMID: 37997490 is a 2023 review about biotin ligases. It's intro says: The discovery of biotin protein ligase (BPL) by Lane and colleagues[4]-[7] followed the realization that protein-bound biotin was involved in mammalian carboxylation reactions...In early work, these enzymes were sometimes named after the enzyme-substrate used to assay BPL activity (e.g., holopyruvate carboxylase synthetase) in the belief that each biotin-dependent enzyme was modified by a specific ligase. However, genetic studies in microorganisms and humans indicate each organism generally encodes a single BPL that modifies each of the biotin-requiring enzymes.

Consistent with that view, direct EXP annotations to these GO terms are all very similar: biotin-protein ligase activity: yeast BPL1, fly Hcs, human HLCS, mouse Hlcs, bacterial birA biotin-[acetyl-CoA-carboxylase] ligase activity: yeast BPL1, bacterial birA, Arabidopsis HCS1 biotin-[methylcrotonoyl-CoA-carboxylase] ligase activity: Arabidopsis HCS1 biotin-[methylmalonyl-CoA-carboxytransferase] ligase activity: [no annotations] biotin-[propionyl-CoA-carboxylase (ATP-hydrolyzing)] ligase activity: human HLCS, rat Hlcs biotin-[pyruvate-carboxylase] ligase activity: yeast BPL1

And EC annotations to Swiss-Prot entries are also very similar: EC:6.3.4.15: mammalian HLCS, yeast BPL1, Arabidopsis HCS1, bacterial birA EC:6.3.4.11: mammalian HLCS, yeast BPL1, Arabidopsis HCS1 EC:6.3.4.9: mammalian HLCS, yeast BPL1 EC:6.3.4.10: mammalian HLCS, yeast BPL1

And the counts of UniProt annotations to the corresponding RHEA IDs are: RHEA:11756: 18,325 proteins RHEA:24376: 5 proteins RHEA:23668: 4 proteins RHEA:11204: 4 proteins

So, taking all this together, I think all the terms mentioned above need merging into a single term based on GO:0004077/EC:6.3.4.15/RHEA:11756, that would look like this:

name: biotin--[biotin carboxyl-carrier protein] ligase activity def: "Catalysis of the reaction: ATP + biotin + L-lysyl-[protein] = AMP + diphosphate + H+ + N(6)-biotinyl-L-lysyl-[protein]." [RHEA:11756] synonym: "biotin-apoprotein ligase activity" EXACT [] etc xref: EC:6.3.4.15 [exactMatch] xref: MetaCyc:BIOTINLIG-RXN [exactMatch] xref: RHEA:11756 [exactMatch] xref: EC:6.3.4.9 [narrowMatch] xref: MetaCyc:6.3.4.9-RXN [narrowMatch] xref: RHEA:23668 [narrowMatch] xref: EC:6.3.4.10 [narrowMatch] xref: MetaCyc:6.3.4.10-RXN [narrowMatch] xref: RHEA:11204 [narrowMatch] xref: EC:6.3.4.11 [narrowMatch] xref: MetaCyc:6.3.4.11-RXN [narrowMatch] xref: RHEA:24376 [narrowMatch] is_a: GO:0016879 ! ligase activity, forming carbon-nitrogen bonds is_a: GO:0140096 ! catalytic activity, acting on a protein

[raymond91125]

Should they be merges at EC and connected at RHEA? @kaxelsen Thanks.

[pgaudet]

Usually RHEA keeps the specific substrates, but GO doesn't. We can go ahead with the obsoletions.