search

geneontology / go-ontology

Source ontology files for the Gene Ontology
http://geneontology.org/page/download-ontology
Creative Commons Attribution 4.0 International
220 stars 40 forks source link

NTR: carbon-nitrogen ligase activity on lipid II #16422

Closed vanessadoan closed 3 years ago

vanessadoan commented 5 years ago

New Term: carbon-nitrogen ligase activity on lipid II Definition: Catalysis of the transfer of the amide nitrogen of glutamine to a variety of substrates. GATases catalyze two separate reactions at two active sites, which are located either on a single polypeptide chain or on different subunits. In the glutaminase reaction, glutamine is hydrolyzed to glutamate and ammonia, which is added to an acceptor substrate in the synthase reaction. To amidate lipid II Aspect: Function Relationships: child of GO:0016884 carbon-nitrogen ligase activity, with glutamine as amido-N-donor References: PMID 22291598 @sandyl27

ukemi commented 5 years ago

Not sure about this one.

sandyl27 commented 5 years ago

@jimhu-tamu

pgaudet commented 5 years ago

@ukemi Why don't you like it ? Because it's mentioning the substrate ?

ukemi commented 5 years ago

Yes. We need to determine if we can define it in terms of ChEBI and get a Rhea cross ref for it.

jimhu-tamu commented 5 years ago

Are we talking about tweaking the definition or something else?

This activity is also hydrolyzes ATP, presumably to make a phosphorylated intermediate (hypothesized but not demonstrated in the paper). In vitro the enzyme works on multiple substrates - lipid I, lipid II, and lipid II-gly5, but the authors assert evidence that the in vivo substrate is the lipid II-gly5.

lipid ii is CHEBI:27692. What's kind of weird at ChEBI is that the only annotated biological role is as a signal for innate immunity in mice. There's nothing for peptidoglycan biosynthesis.

pgaudet commented 5 years ago

@amorgat Is there a Rhea reaction for this ?

pgaudet commented 5 years ago

@jimhu-tamu @vanessadoan @sandyl27 ~1. Should we call this activity 'UDP-N-acetylglucosamine--N-acetylmuramyl-(pentapeptide) pyrophosphoryl-undecaprenol N-acetylglucosamine transferase' to match the UniProt entry for that gene? https://www.uniprot.org/uniprot/P17443~

  1. Your definition talks about 2 reactions: "Catalysis of the transfer of the amide nitrogen of glutamine to a variety of substrates. GATases catalyze two separate reactions at two active sites, which are located either on a single polypeptide chain or on different subunits. In the glutaminase reaction, glutamine is hydrolyzed to glutamate and ammonia, which is added to an acceptor substrate in the synthase reaction. To amidate lipid II"

Does that mean that we need to activities ? Anyways as it is, it's not a single activity.

Thanks, Pascale

jimhu-tamu commented 5 years ago

@pgaudet

  1. No. That entry is for E. coli MurG. The paper is for the S. aureus GatD/MurT complex. This has related domains but the reaction is different. MurG attaches GlcNac to MurNac in the sugar part of the peptidoglycan. GatD/MurT amidates the D-Glu in the pentapeptide.

So, while I'm not sure, I think the lipid II part is OK. Based on looking at some BRENDA pages, I'm thinking: undecaprenyldiphospho-N-acetyl-(N-acetylglucosaminyl)muramoyl pentapeptide amidotransferase (glutamine-hydrolyzing) activity.

This is consistent, I think, with GO:0004088 carbamoyl-phosphate synthase (glutamine-hydrolyzing) activity, but since I am not calling it a synthase, that might not be appropriate.

  1. We debated 1 vs 2 activities based on the enzyme based on the molecular function ontology documentation

If the intermediates are known to be released and used in other biological processes, function terms should be added to represent the steps of the reaction, and a biological process term added to represent the sum of these functions. If the separate activities generating and acting on the molecular intermediates can be associated with different subunits of the enzyme, two or more functions should probably be made. If neither of these conditions holds, a single function term can be made.

Here, if I'm reading the paper correctly, the intermediate is not released but the activities are on separate subunits. "probably" leaves wiggle room, so we looked at GO:0004088 - carbamoyl-phosphate synthase (glutamine-hydrolyzing) activity, which I view as a paradigm for this class of enzymes, including the intermediate tunneling.

I believe this is a single activity term where subunits have separate active sites.

pgaudet commented 5 years ago

Can we improve the definition ? "Catalysis of the transfer of the amide nitrogen of glutamine to a variety of substrates. GATases catalyze two separate reactions at two active sites, which are located either on a single polypeptide chain or on different subunits. In the glutaminase reaction, glutamine is hydrolyzed to glutamate and ammonia, which is added to an acceptor substrate in the synthase reaction. To amidate lipid II"

  1. "amide nitrogen of glutamine to a variety of substrates" is pretty vague - can we specify something about these substrates ?
  2. The last sentence ends abruptly.

Thanks, Pascale

pgaudet commented 5 years ago

@jimhu-tamu @ukemi Does this seem OK ?

~+id: GO:0140277~ GO:0140282 +name: undecaprenyldiphospho-N-acetyl-(N-acetylglucosaminyl)muramoyl pentapeptide amidotransferase (glutamine-hydrolyzing) activity +namespace: molecular_function +def: "Catalysis of the transfer of the amide nitrogen of glutamine to a variety of substrates. GATases catalyze two separate reactions at two active sites, which are located either on a single polypeptide chain or on different subunits. In the glutaminase reaction, glutamine is hydrolyzed to glutamate and ammonia, which is added to an acceptor substrate in the synthase reaction." [PMID:22291598] +comment: Created new term as requested in https://github.com/geneontology/go-ontology/issues/16422 +is_a: GO:0016884 ! carbon-nitrogen ligase activity, with glutamine as amido-N-donor

Thanks, Pascale

ukemi commented 5 years ago

It is ok, but whenever we create a new enzyme reaction we should be coordinating with EC, Rhea and ChEBI.

pgaudet commented 5 years ago

Thanks @ukemi

I have done very little with enzymes. Do we have instructions for that ?

ukemi commented 5 years ago

As you know, the coordination of enzymes with eternal resources is an ongoing project that we will address at the Geneva ontology developers' meeting. One of the in things we should accomplish there is to create a standard data flow. One thing we know for sure is that reactome will be the gatekeeper for reactome xrefs. Here is the existing documentation, but it will need review as we proceed: http://wiki.geneontology.org/index.php/Enzymes_and_EC_mappings http://wiki.geneontology.org/index.php/Curator_Guide:_Enzymes_and_Reactions

pgaudet commented 5 years ago

OK - having another look at this.

The paper cites the following SwissProt AC:

pgaudet commented 5 years ago

@jimhu-tamu Can you tell me the EC, or the Brenda entry?

jimhu-tamu commented 5 years ago

I couldn't find an EC for this activity. What I mentioned above was to look at activities with similar chemistry and activities that work on lipid II to suggest possible names. But getting feedback from EC sounds like a good thing.

pgaudet commented 5 years ago

@kaxelsen Can you help ?

kaxelsen commented 5 years ago

Several things: You write: 'GATases catalyze two separate reactions at two active sites'. This is wrong. GATase (aka glutamine amidotransferase [this name is misleading since the subunit is not a transferase]) is the name of the glutaminase subunit, so it is not the complex that is called GATase. The "GATase" subunit is GatD (SA1707). The resulting NH3 from the hydrolysis of glutamine is channeled to the active site in MurT (SA1708), where it is attached to the lipid II glutamate residue.

A GATase subunit is actually as mentioned above a glutaminase, EC 3.5.1.2, but in similar complexes is only active when it is in the complex.

CHEBI: CheBI molecule used in Rhea is CHEBI:60033 (The charged version of chebi:27692). This is based on PMIDs 22291598 and 30093673. PMID 30154570 shows another variant, lipid II-gly-5

Rhea: We do not have this reaction in Rhea.

EC: There is no EC number for this activity. A simple name could be: Lipid II amidase. Another more correct name for the heterodimer complex could be glutamine--lipid II amidotransferase

pgaudet commented 5 years ago

Definition for GO:0016884 carbon-nitrogen ligase activity, with glutamine as amido-N-donor

Current: Catalysis of the transfer of the amide nitrogen of glutamine to a variety of substrates. GATases catalyze two separate reactions at two active sites, which are located either on a single polypeptide chain or on different subunits. In the glutaminase reaction, glutamine is hydrolyzed to glutamate and ammonia, which is added to an acceptor substrate in the synthase reaction.

Proposed by @kaxelsen : Catalysis of the transfer of the amide nitrogen of glutamine to a substrates. Usually composed of two subunits or domains, one that first hydrolyzes glutamine, and then transfers the resulting ammonia to the second subunit (or domain), where it acts as a source of nitrogen.

pgaudet commented 5 years ago

New definition GO:0140282: Catalysis of the transfer of the amide nitrogen of glutamine to lipid II.

pgaudet commented 5 years ago

Now we're only missing the EC number. @kaxelsen let us know when it is ready :)

kaxelsen commented 4 years ago

The EC number is now available as EC 6.3.5.13. The associated Rhea reaction is RHEA:57928. RHEA:57928 is a multi-step reaction composed of three parts: RHEA:15889 + RHEA:59488 + RHEA:57932