GO terms referencing EC:4.2.1.134 (very-long-chain (3R)-3-hydroxyacyl-CoA dehydratase)

[sjm41]

There are 5 GO terms using EC:4.2.1.134 as an xref and/or definition attribution.

https://enzyme.expasy.org/EC/4.2.1.134 Name: very-long-chain (3R)-3-hydroxyacyl-CoA dehydratase. Reaction: a very-long-chain (3R)-3-hydroxyacyl-CoA <=> a very-long-chain (2E)-enoyl-CoA + H2O

The xref on GO:0102158 (shown below) looks correct, although:

• [x] the definition ought to be changed to use "a very-long-chain (2E)-enoyl-CoA" in the right hand side (to match EC/Rhea)
• [x] the '(3R)' bit should be included in the term name:

id: GO:0102158 name: very-long-chain 3-hydroxyacyl-CoA dehydratase activity [9658 annotations, no EXP] namespace: molecular_function def: "Catalysis of the reaction: a very-long-chain (3R)-3-hydroxyacyl-CoA = H2O + a very-long-chain trans-2,3-dehydroacyl-CoA." [GOC:pz, RHEA:45812] xref: EC:4.2.1.134 xref: MetaCyc:RXN-11750 xref: RHEA:45812 is_a: GO:0016836 ! hydro-lyase activity

---

I think the other 4 (shown below) sh/could be obsoleted - the EC xref/attribution isn't accurate, none have any EXP annotations, and the all the non-EXP annotations are the same IEAs with EC2GO via EC:4.2.1.134. But if they are retained, then the EC references should be removed, and they should become children (rather than sisters) of GO:0102158. The first one also appears to have a discrepancy in the substrate involved in the term name vs definition.

id: GO:0102343 name: 3-hydroxy-arachidoyl-CoA dehydratase activity [6165 annotations, no EXP] namespace: molecular_function def: "Catalysis of the reaction: (R)-3-hydroxyicosanoyl-CoA <=> trans-2-icosenoyl-CoA + H2O." [EC:4.2.1.134, GOC:pz] xref: EC:4.2.1.134 xref: MetaCyc:RXN-13302 is_a: GO:0016836 ! hydro-lyase activity

id: GO:0102344 name: 3-hydroxy-behenoyl-CoA dehydratase activity [6165 annotations, no EXP] namespace: molecular_function def: "Catalysis of the reaction: (R)-3-hydroxybehenoyl-CoA <=> trans-2-docosenoyl-CoA + H2O." [EC:4.2.1.134, GOC:pz] xref: EC:4.2.1.134 xref: MetaCyc:RXN-13303 is_a: GO:0016836 ! hydro-lyase activity

id: GO:0102345 name: 3-hydroxy-lignoceroyl-CoA dehydratase activity [6165 annotations, no EXP] namespace: molecular_function def: "Catalysis of the reaction: (R)-3-hydroxylignoceroyl-CoA(4-) <=> trans-2-tetracosenoyl-CoA + H2O." [EC:4.2.1.134, GOC:pz] xref: EC:4.2.1.134 xref: MetaCyc:RXN-13304 is_a: GO:0016836 ! hydro-lyase activity

id: GO:0102346 name: 3-hydroxy-cerotoyl-CoA dehydratase activity [0 annotations] namespace: molecular_function def: "Catalysis of the reaction: (R)-3-hydroxycerotoyl-CoA(4-) <=> trans-2-hexacosenoyl-CoA(4-) + H2O." [EC:4.2.1.134, GOC:pz] xref: MetaCyc:RXN-13305 is_a: GO:0016836 ! hydro-lyase activity

[sjm41]

Also, looks like GO:0102158 very-long-chain (3R)-3-hydroxyacyl-CoA dehydratase activity should be made a child of this term:

id: GO:0080023 name: 3R-hydroxyacyl-CoA dehydratase activity [5 EXP annotations] namespace: molecular_function def: "Catalysis of the reaction: 3R-hydroxyacyl-CoA = 2E-enoyl-CoA + H2O." [PMID:16982622] xref: Reactome:R-HSA-5676637 "PTPLs dehydrate VLC3HA-CoA to VLCTDA-CoA" is_a: GO:0016836 ! hydro-lyase activity

[sjm41]

Further (!), are these two terms referring to the same thing?:

id: GO:0080023 name: 3R-hydroxyacyl-CoA dehydratase activity [5 EXP annotations] namespace: molecular_function def: "Catalysis of the reaction: 3R-hydroxyacyl-CoA = 2E-enoyl-CoA + H2O." [PMID:16982622] xref: Reactome:R-HSA-5676637 "PTPLs dehydrate VLC3HA-CoA to VLCTDA-CoA" is_a: GO:0016836 ! hydro-lyase activity

id: GO:0018812 name: 3-hydroxyacyl-CoA dehydratase activity [11 EXP annotations] namespace: molecular_function def: "Catalysis of the reaction: alkene-CoA + H2O = alcohol-CoA. Substrates are crotonoyl-CoA (producing 3-hydroxyacyl-CoA) and 2,3-didehydro-pimeloyl-CoA (producing 3-hydroxypimeloyl-CoA)." [UM-BBD_ruleID:bt0291] xref: Reactome:R-HSA-8957389 "RPP14 (HTD2) dehydrates 3HA-CoA to t2E-CoA" is_a: GO:0016836 ! hydro-lyase activity

I don't know if "3R-hydroxyacyl..." is different from "3-hydroxyacyl..."??

[pgaudet]

Dear all,

The proposal has been made to obsolete GO:0102343 3-hydroxy-arachidoyl-CoA dehydratase activity GO:0102344 3-hydroxy-behenoyl-CoA dehydratase activity GO:0102345 3-hydroxy-lignoceroyl-CoA dehydratase activity GO:0102346 3-hydroxy-cerotoyl-CoA dehydratase activity

The reason for obsoletion is that these correspond to specific substrates of GO:0102158 very-long-chain 3-hydroxyacyl-CoA dehydratase activity. There are no EXP annotations to these terms. There is one mapping, EC:4.2.1.134, which corresponds to the more general activity described by GO:0102158. These terms are not present in any subsets.

You can comment on the ticket: https://github.com/geneontology/go-ontology/issues/24738

Thanks, Pascale

[pgaudet]

• [x] GO:0080023 3R-hydroxyacyl-CoA dehydratase activity == RHEA:26526 & EC:4.2.1.119

[pgaudet]

• [x] GO:0018812 : added to the definition: "The stereospecifity of this reaction is not specified."

[sjm41]

Thanks Pascale! Given GO:0018812 (3-hydroxyacyl-CoA dehydratase activity) is agnostic of the stereoisomer, should GO:0080023 (3R-hydroxyacyl-CoA dehydratase activity) be a child of it?

[pgaudet]

Yes, I've added this, thanks

We now have

. '3-hydroxyacyl-CoA dehydratase activity' . . '(3R)-3-hydroxyacyl-CoA dehydratase activity' . . . '(3R)-3-hydroxybutyryl-CoA dehydratase activity' . . . 'very-long-chain (3R)-3-hydroxyacyl-CoA dehydratase activity'

[sjm41]

Sorry, looks like there's some additional nesting to do here. I think the tree needs to be arranged as shown below (terms not discussed so far in this ticket are shown in bold):

. '3-hydroxyacyl-CoA dehydratase activity' (GO:0018812) . . 'enoyl-CoA hydratase activity' (GO:0004300, EC:4.2.1.17) [34 EXP annotations] . . 'long-chain-enoyl-CoA hydratase activity' (GO:0016508, EC:4.2.1.74) [8 EXP annotations] . . '3-hydroxypropionyl-CoA dehydratase activity' (GO:0043956, EC:4.2.1.116) [1 EXP annotation] . . '(3R)-3-hydroxyacyl-CoA dehydratase activity' (GO:0080023, EC:4.2.1.119) . . . '(3R)-3-hydroxybutyryl-CoA dehydratase activity' (GO:0003859, 4.2.1.55) . . . 'very-long-chain (3R)-3-hydroxyacyl-CoA dehydratase activity' (GO:0102158, EC:4.2.1.134)

Additional changes to those new terms:

• [x] id: GO:0004300 name: enoyl-CoA hydratase activity def: "Catalysis of the reaction: (3S)-3-hydroxyacyl-CoA = trans-2-enoyl-CoA + H2O." [EC:4.2.1.17] xref: EC:4.2.1.17 xref: MetaCyc:ENOYL-COA-HYDRAT-RXN xref: RHEA:16105 xref: UM-BBD_enzymeID:e0014 => Change definition to match EC? => But maps to 3 different RHEAs: RHEA:16105, RHEA:20724, RHEA:34863

• [ ] id: GO:0016508 name: long-chain-enoyl-CoA hydratase activity [8 EXP annotations] def: "Catalysis of the reaction: a long-chain (3S)-3-hydroxyacyl-CoA = a long-chain trans-2-enoyl-CoA + H2O. A long-chain acyl-CoA is an acyl-CoA thioester where the acyl chain contains 13 to 22 carbon atoms." [EC:4.2.1.74] synonym: "long-chain enoyl coenzyme A hydratase activity" RELATED [EC:4.2.1.74] synonym: "long-chain-(3S)-3-hydroxyacyl-CoA hydro-lyase activity" RELATED [EC:4.2.1.74] xref: EC:4.2.1.74 xref: MetaCyc:LONG-CHAIN-ENOYL-COA-HYDRATASE-RXN => name and def in EC is "medium-chain" rather than "long-chain" (though "long-chain" is a synonym!) => add RHEA: 71647

• [x] id: GO:0043956 name: 3-hydroxypropionyl-CoA dehydratase activity def: "Catalysis of the reaction: 3-hydroxypropionyl-CoA = acrylyl-CoA + H2O." [GOC:jl, PMID:11821399] comment: Note that this function is one of the activities of the trifunctional enzyme propionyl-coenzyme A synthase. See PMID:11821399. xref: EC:4.2.1.116 xref: MetaCyc:RXN-6383 xref: RHEA:26518 => change product from 'acrylyl-CoA' to 'acryloyl-CoA' to match EC/RHEA => EC maps to RHEA:26558 (as well as RHEA:26518)

Do you want a separate ticket for any of this??

[pgaudet]

GO:0004300 enoyl-CoA hydratase activity

• [x] Changed definition: -def: "Catalysis of the reaction: (3S)-3-hydroxyacyl-CoA = trans-2-enoyl-CoA + H2O." [EC:4.2.1.17] +def: "Catalysis of the reaction: a 3-hydroxy-fatty acyl-CoA = a enoyl-CoA + H2O. This reaction usually occurs in the reverse direction, leading to the reduction of the double bound of enoyl-CoA in position 2 or 3. Specific reactions catalyzed include: a 4-saturated-(3S)-3-hydroxyacyl-CoA = a (3E)-enoyl-CoA + H2O and a (3S)-3-hydroxyacyl-CoA = a (2E)-enoyl-CoA + H2O." [EC:4.2.1.17]

• [x] xrefs: added as skos:narrowMatch

• RHEA:16105 a (3S)-3-hydroxyacyl-CoA = a (2E)-enoyl-CoA + H2O 19,054 protein(s) (UniProtKB)

• RHEA:20724 a 4-saturated-(3S)-3-hydroxyacyl-CoA = a (3E)-enoyl-CoA + H2O 19,049 protein(s) (UniProtKB)

• RHEA:34863 (Z)-2,3-didehydroadipyl-CoA + H2O = 3-hydroxyadipyl-CoA ==> No mappings; very specific substrate ===> do not map

[pgaudet]

We cannot put enoyl-CoA hydratase activity as a child of '3-hydroxyacyl-CoA dehydratase activity' (GO:0018812) but we can do the other way around:

. 'enoyl-CoA hydratase activity' (GO:0004300, EC:4.2.1.17) [34 EXP annotations] . . '3-hydroxyacyl-CoA dehydratase activity' (GO:0018812)

• 3-hydroxyacyl-CoA dehydratase activity' Catalysis of the reaction: a 3-hydroxy-fatty acyl-CoA = a (2E)-enoyl-CoA + H2O.
• enoyl-CoA hydratase activity: (new def to match EC: ) Catalysis of the reaction: a 3-hydroxy-fatty acyl-CoA = a enoyl-CoA + H2O. This reaction usually occurs in the reverse direction, leading to the reduction of the double bound of enoyl-CoA in position 2 or 3. Specific reactions catalyzed include: a 4-saturated-(3S)-3-hydroxyacyl-CoA = a (3E)-enoyl-CoA + H2O and a (3S)-3-hydroxyacyl-CoA = a (2E)-enoyl-CoA + H2O.

[pgaudet]

Hi @sjm41

@marcfeuermann and I looked at these terms this morning, and we propose to obsolete

GO:0080023 (3R)-3-hydroxyacyl-CoA dehydratase activity -> 6 EXP

• very-long-chain (3r)-3-hydroxyacyl-coa dehydratase pthr11035
• peroxisomal multifunctional enzyme type 2-related pthr13078

GO:0003859 (3R)-3-hydroxybutyryl-CoA dehydratase activity -> 1 EXP

GO:0102158 very-long-chain (3R)-3-hydroxyacyl-CoA dehydratase activity -> 0 EXP, all IEAs RHEA/EC

GO:0016508 long-chain-enoyl-CoA hydratase activity -> 7 EXP

• peroxisomal multifunctional enzyme type 2-related pthr13078
• trifunctional enzyme subunit alpha pthr43612

with 'replaced by' GO:0018812 3-hydroxyacyl-CoA dehydratase activity, as this appears to be the degree of specificity of the enzymes. These are all involved in beta-oxidation, and the exact composition of the substrate does not seem to be crucial. Moreover, if we keep these terms, we need to add way more to make the hierarchy complete - see for example, all the reactions in this entry: https://www.uniprot.org/uniprotkb/P51659/entry

What do you think ?

Thanks, Pascale

[sjm41]

Sounds good to me - this would greatly simplify this complicated set of terms! So, after these changes, I think the local tree would look like this, right?: . 'enoyl-CoA hydratase activity' (GO:0004300, EC:4.2.1.17) . . '3-hydroxyacyl-CoA dehydratase activity' (GO:0018812) . . '3-hydroxypropionyl-CoA dehydratase activity' (GO:0043956, EC:4.2.1.116)

Maybe GO:0043956 should be obsoleted too? (Has a single EXP annotation according to QuickGO)

[pgaudet]

Good point - in fact GO:0043956 3-hydroxypropionyl-CoA dehydratase activity is not involved in beta oxidation, see UniProt:A4YI89

Plays a role in autotrophic carbon fixation via the 3-hydroxypropionate/4-hydroxybutyrate cycle. Catalyzes the reversible dehydration of 3-hydroxypropionyl-CoA to form acryloyl-CoA, and the reversible dehydration of (S)-3-hydroxybutyryl-CoA to form crotonyl-CoA. Inactive towards (R)-3-hydroxybutyryl-CoA

That seems a more specific reaction. OK ?

[alanbridge]

Hi,

may be worth checking with @kaxelsen, @amorgat, and the Rhea team (including Lucila Aimo), as stereospecificity is thought to be important in fatty acid metabolism (I'm a bit rusty on this pathway I'm afraid but here is one example):

e.g. FADB of ECOLI - https://www.uniprot.org/uniprotkb/P21177/entry

Catalyzes the formation of 3-oxoacyl-CoA from enoyl-CoA via L-3-hydroxyacyl-CoA.

that is 2 Rhea reactions

a (3S)-3-hydroxyacyl-CoA + NAD(+) = a 3-oxoacyl-CoA + H(+) + NADH a (3S)-3-hydroxyacyl-CoA = a (2E)-enoyl-CoA + H2O

which when joined make

(2E)-enoyl-CoA -> (3S)-3-hydroxyacyl-CoA -> 3-oxoacyl-CoA

FADB is stereospecific, it doesn't do this

(2E)-enoyl-CoA -> (3R)-3-hydroxyacyl-CoA -> 3-oxoacyl-CoA

there are a fair number of other examples too.

All the best, Alan

[pgaudet]

HI @alanbridge

I asked @kaxelsen about this; the current ontology version is very very incomplete -

There are two options, obsolete these pretty random terms, or add all other possible substrates, which would be a very long list, and not in scope for GO, since these various reactions represents substrates of the same/similar enzymes. FOr GO these would be best captured as GO CAM models.

Thanks for the feedback,

Pascale

[sjm41]

Hi @pgaudet Shall we proceed with the additional obsoletions mentioned in https://github.com/geneontology/go-ontology/issues/24738#issuecomment-1403404076 and https://github.com/geneontology/go-ontology/issues/24738#issuecomment-1403427211?

There's also a couple of tasks in https://github.com/geneontology/go-ontology/issues/24738#issuecomment-1387006246 to finish off.

[pgaudet]

Dear all,

The proposal has been made to obsolete GO:0080023 (3R)-3-hydroxyacyl-CoA dehydratase activity -> 6 EXP GO:0003859 (3R)-3-hydroxybutyryl-CoA dehydratase activity -> 1 EXP GO:0102158 very-long-chain (3R)-3-hydroxyacyl-CoA dehydratase activity -> 0 EXP, all IEAs RHEA/EC GO:0016508 long-chain-enoyl-CoA hydratase activity -> 7 EXP

The reason for obsoletion is that these reactions are all more specific than the enzymes they represent.

These terms will be replaced with GO:0018812 3-hydroxyacyl-CoA dehydratase activity, as this is the specificity of the activity.

There are 10 EXP annotations to these terms; if your annotation tool does automatic replacement, there is no action needed on your part. Impacted groups:

ComplexPortal : 1 EXP MTBBASE : 1 EXP Reactome: 2 EXP RGD: 1 EXP UniProt : 5 EXP

https://github.com/geneontology/go-annotation/issues/4488

You can comment on the ticket: https://github.com/geneontology/go-ontology/issues/24738

Thanks, Pascale

[sjm41]

Thanks @pgaudet I think we were going to add '3-hydroxypropionyl-CoA dehydratase activity' GO:0043956 to the obsoletion list too?

[pgaudet]

Marc and I proposed to keep this - see https://github.com/geneontology/go-ontology/issues/24738#issuecomment-1403431466

[sjm41]

Ah, I interpreted that comment as you were going to obsolete it along with the others! No problem keeping it!

[pgaudet]

This term was obsoleted because it represents a specific substrate of 3-hydroxyacyl-CoA dehydratase activity ; GO:0018812.

[pgaudet]

Just wondering : EC:4.2.1.17 is named 'enoyl-CoA hydratase', with the comment, 'Acts in the reverse direction. ' Other terms in this branch are called 'dehydratase'; should I rename enoyl-CoA hydratase?

[sjm41]

Renaming to dehydratase makes sense to me (given the EC comment), but maybe it's always called a 'enoyl-CoA hydratase' in the field/literature. What do you think @kaxelsen ?

[pgaudet]

Sounds good, most papers refer to enoyl-CoA hydratase, so I'll leave this as is.

https://pubmed.ncbi.nlm.nih.gov/?term=enoyl-CoA+dehydratase

[kaxelsen]

In the EC list, lyase reactions (EC 4) are always written in the direction of dehydratase (for EC 4.2.1), so the name hydratase and the comment about that it "acts in the reverse direction" are supplementary pieces of information.

[deustp01]

If I remember correctly, EC naming follows the convention that all chemical reactions are in principle reversible, so as here names are kept consistent and don't necessarily indicate the physiological direction of a reaction (so that direction information needs to come from somewhere else, not through tweaking the name of the enzyme or (maybe) the names of associated GO MF terms.

[kaxelsen]

@depust01. That is actually not the case case. The "accepted name"s are often the names widely used in the scientific field. If the enzyme has not been described before IUBMB tries to coin a sensible name ( e.g. 5-hydroxybenzimidazole synthase for EC 4.1.99.23 or GTP 3',8-cyclase for EC 4.1.99.22). The systematic name is the one that describes the reaction following formal rules that do not take the physiological direction into account.

[pgaudet]

thanks for the feedback both! To me this was confusing because we have

• enoyl-CoA hydratase activity
• • '3-hydroxyacyl-CoA dehydratase activity'
• • '3-hydroxypropionyl-CoA dehydratase activity'

but I am happy to align with the EC nomenclature.

[sjm41]

@pgaudet

We obsoleted 'very-long-chain 3-hydroxyacyl-CoA dehydratase activity' (GO:0102158) (= EC:4.2.1.134) as part of this clean-up, but I now realise that one should have been retained as it describes the key third activity in Microsomal fatty acyl elongation.

Quoting from the EC:4.2.1.134 (https://enzyme.expasy.org/EC/4.2.1.134) entry: This is the third component of the elongase, a microsomal protein complex responsible for extending palmitoyl-CoA and stearoyl-CoA (and modified forms thereof) to very-long chain acyl CoAs. cf. EC 1.1.1.330, EC 1.3.1.93 and EC 2.3.1.199.

(And we have GO terms corresponding to the first, second and fourth components/ECs of the elongase complex.)

Can we re-instate GO:0102158 as a child of 3-hydroxyacyl-CoA dehydratase activity' (GO:0018812)? Also, the definition ought to be tweaked to match EC/Rhea, and the '(3R)' bit should be included in the term name to match EC.

Here's how the entry used to look: id: GO:0102158 name: very-long-chain 3-hydroxyacyl-CoA dehydratase activity namespace: molecular_function def: "Catalysis of the reaction: a very-long-chain (3R)-3-hydroxyacyl-CoA = H2O + a very-long-chain trans-2,3-dehydroacyl-CoA." [GOC:pz, RHEA:45812] xref: EC:4.2.1.134 xref: MetaCyc:RXN-11750 xref: RHEA:45812 is_a: GO:0016836 ! hydro-lyase activity

[sjm41]

Thanks @pgaudet ! Would you agree the parent of the reinstated GO:0102158 should be '3-hydroxyacyl-CoA dehydratase activity' (GO:0018812)?

[pgaudet]

Although I think we had it differently because the parent's definition states

The stereospecifity of the hydroxyacyl in this reaction is not specified.

while GO:0102158 mentions 3R.

[sjm41]

Oh yes, I see. Maybe that caveat isn't explicitly needed on the parent definition - it's implied anyway by its absence in the stated reaction. Then it won't look odd for the child GO:0102158 term to specify 3R?

[pgaudet]

Maybe that caveat isn't explicitly needed on the parent definition

RIght - and I now remember that @marcfeuermann mentioned that these enzymes are probably non-stereo-specific, but this captures substrates that have been tested. Probably best for the definition to be a bit more vague to avoid having to make a complex hierarchy.