Merge L-2-aminoadipate N-acetyltransferase activity into GO:0043869 alpha-aminoadipate acetyltransferase activity

[pgaudet]

Cleaning up IMG: cross references #18908

Hi @hdrabkin

Are those 2 terms the same ?

id: GO:0043741 name: L-2-aminoadipate N-acetyltransferase activity namespace: molecular_function def: "Catalysis of the reaction: L-2-aminoadipate + acetyl-CoA = N2-acetyl-L-aminoadipate + CoA." [IMG:01447] xref: MetaCyc:RXN-5181 is_a: GO:0008080 ! N-acetyltransferase activity

=> no annotations

Merge into GO:0043869 alpha-aminoadipate acetyltransferase activity : Catalysis of the reaction: alpha-aminoadipate + acetyl-CoA = N2-acetyl-alpha-aminoadipate + coenzyme A." [MetaCyc:RXN-5181,

=> 1 ZFIN => Metacyc https://biocyc.org/META/NEW-IMAGE?type=NIL&object=RXN-5181 has the substrate for GO:0043741.

Thanks, Pascale

[pgaudet]

BTW both GO terms have the same MetaCyc xref MetaCyc:RXN-5181

[hdrabkin]

Yes I see the same metacyc. I'd merge the second into the first (ie, keep the one with the L-2-aminoadipate substrate. I can request since the metacyc rxn has all the chebi ids I need, but they will require a PMID also.

[hdrabkin]

Use the zfin annotation PMID?

[pgaudet]

sure

[pgaudet]

Should I do it or do you want to ?

[hdrabkin]

You can do the merge and mark it RHEA pending I just found the ref: PMID:29053735 to add to the def.

[hdrabkin]

Rhea requested for merged term. [help #181278] in the subject line of all future correspondence about this issue.

[hdrabkin]

I would leave it open if we are waiting for RHEA..

[hdrabkin]

Hi @pgaudet Here is RHEA's response I had supplied the PMID that zfin used to annotate the less specific term for GO:0043869 alpha-aminoadipate; the only PMID we had.

Bottom line: No evidence for the reaction GO:0043741 Maybe we want to obsolete it.

Dear Harold,

From PMID:29053735 "ALDH7A1 encodes alpha-aminoadipic semialdehyde dehydrogenase (EC 1.2.1.31) in the lysine catabolic pathway and the enzyme deficiency leads to accumulation of alpha-aminoadipic acid semialdehyde (alpha-AASA) and piperideine 6-carboxylate (P6C), the latter inactivates pyridoxal-5-phosphate."

In UniProt: https://www.uniprot.org/uniprot/P49419 P49419 (AL7A1_HUMAN) Alpha-aminoadipic semialdehyde dehydrogenase has a sequence highly similar to the 2 Zebrafish TrEMBL entries F1QR17_DANRE (https://www.uniprot.org/uniprot/F1QR17) Q803R9_DANRE (https://www.uniprot.org/uniprot/Q803R9)

One of the reaction catalyzed by AL7A1_HUMAN is RHEA:12304 (S)-2-amino-6-oxohexanoate + H2O + NADP(+) = 2 H(+) + L-2-aminoadipate + NADPH (https://www.rhea-db.org/reaction?id=12304) Note that CHEBI:58321 (S)-2-amino-6-oxohexanoate == L-2-aminoadipate semi-aldehyde == alpha-AASA Whereas RXN-5181 (L-2-aminoadipate + acetyl-CoA = N2-acetyl-L-aminoadipate + CoA) is an N-acetyl transferase reaction, not mentioned in the article. If I remember well, I think RXN-5181 was a reaction supposed to protect the amino group of L-aminoadipate in some L-lysine biosynthesis pathway. It was discovered later that this amino group was in fact protected by binding to LysW (the reaction catalyzed by LysX I mentioned in my previous mail). Note that the MetaCyc reaction (https://metacyc.ai.sri.com/META/NEW-IMAGE?object=RXN-5181) is neither linked to an article nor to a known enzyme.

In conclusion, we can't create the requested reaction as long as we don't have experimental evidence for it.

I hope it make sense for you and you will understand our concern.

Cheers, Anne

On Wed Feb 26 20:25:06 2020, harold.drabkin@jax.org wrote:

Hi Ann

It's a zfin gene that used it PMID:29053735 Characterization of the first knock-out aldh7a1 zebrafish model for pyridoxine-dependent epilepsy using CRISPR-Cas9 technology.

We are merging two terms GO:0043869 into GO:0043741 Both reference the same metacyc id MetaCyc:RXN-5181, and the metacyc is specific for the L-2-aminoadipate (43869 merely says 2-aminoadipate

So I thought I'd get a RHEA to match the metacyc.

I have attached the pdf for PMID;

On 2/26/20, 2:14 PM, "Anne Morgat via RT" rhea-reaction@sib.swiss wrote:

  Hi Harold,
  I have some trouble to find the requested reaction in PMID:29053735
  What are the genes annotated with GO:0043741 term?

  I can find some TrEMBL entries (unreviewed) still annotated as
  "L-2-aminoadipate N-acetyltransferase"
  Ex: https://www.uniprot.org/uniprot/A0A1G7D5W5
  But A0A1G7D5W5 share 96.1% seq identity with Q5SH23 (LYSX_THET8)
  "Alpha-aminoadipate--LysW ligase LysX"
  RHEA:41940 (EC 6.3.2.43)
  [2-aminoadipate-carrier protein]-C-terminal-L-glutamate + ATP +
  L-2-aminoadipate = [2-aminoadipate-carrier
  protein]-C-terminal-gamma-(L-2-aminoadip-2-yl)-L-glutamate + ADP + H(+) +
  phosphate

  Cheers, Anne

  On Wed Feb 26 19:11:22 2020, harold.drabkin@jax.org wrote:

    (sent by harold.drabkin@jax.org)
    To be used for
    d: GO:0043741
    name: L-2-aminoadipate N-acetyltransferase activity
    namespace: molecular_function
    def: "Catalysis of the reaction: L-2-aminoadipate + acetyl-CoA = N2-acetyl-L-aminoadipate + CoA."
    Metacyc:RXN-5181

    L-2-aminoadipate + acetyl-CoA = N2-acetyl-L-aminoadipate + CoA."
    All of these have CHEBI ids

    PMID:29053735

-- Anne Morgat SIB | Swiss Institute of Bioinformatics CMU - 1, rue Michel Servet - 1211 Geneva 4 t: +41 22 379 49 25 - f: +33 4 76 61 54 08 Anne.Morgat@sib.swiss - http://www.sib.swiss

[pgaudet]

I think the merge is OK.

[hdrabkin]

yes as long as GO:0043869 is the winning term

[pgaudet]

Ah ! no. I'll swap them.

[hdrabkin]

Because the less general term (43869) is the one with the annotation and reference and the one with the L-2-aminoadipate N-acetyltransferase activity was the one RHEA could not find sufficient evidence for .

[pgaudet]

Sure, but this is not 'visible' once everything is merged ?