glutamate catabolic process

[ValWood]

I could use either of these terms, do we need both?

GO:0006539 glutamate catabolic process via 2-oxoglutarate Definition The chemical reactions and pathways resulting in the breakdown of glutamate, via the intermediate 2-oxoglutarate

this one sounds a little vague?

GO:0019551 glutamate catabolic process to 2-oxoglutarate Definition (GO:0019551 GONUTS page) The chemical reactions and pathways resulting in the breakdown of glutamate into other compounds, including 2-oxoglutarate.

[ukemi]

In the first term, the entire pathway would describe an end point that is after the formation of 2-oxoglutarate in the pathway. In the second term, the 2-oxoglutarate would be the endpoint of the pathway. The catabolism needs to be looked at carefully to investigate how the pathway through the 2-oxoglutarate can branch, but on the surface, it looks like GO:0006539 should have has_part GO:0019551 and GO:0019551 should be part_of GO:0006539. The definition of GO:0019551 also looks like it needs to be refined. With this structure, GO:0019551 can be used as a module for the upstream process that is the beginning f downstream processes that branch off of it.

[ukemi]

Looking at this closer, the synonyms of GO:0019551 seem a bit odd, but the logical definition seems ok. We also need to investigate how alpha-ketoglutarate etc relate to 2-oxoglutarate metabolically.

[ukemi]

The logical definitions of both terms look fine as I described them above. In GO:0019551, 2-oxoglutarate is the output and in GO:0006539, 2-oxoglutarate is an intermediate.

[deustp01]

2-oxoglutarate (aka alpha-ketoglutarate) is a TCA cycle intermediate, so it makes sense to make it an endpoint for a glutamate-specific pathway in the same way that glycolysis ends at pyruvate, not because that's a physiological endpoint but because it's a junction where metabolism unique to glucose feeds into multiple other pathways.

If this reasoning makes sense, then GO:0019551 looks good as-is and GO:0006539 looks over-broad. However, the only textbook path from glutamate to 2-oxoglutarate requires only a single transamination reaction, so GO:0019551 doesn't meet the suggestion (requirement?) that processes involve more than one reaction / molecular function.

[pgaudet]

@deustp01 @ukemi

Is there an action here, or are those terms correct ? Looking at genes annotated to this term, most are 'glutamic-oxaloacetic transaminase', which according to Uniprot catalyze 2-oxoglutarate + L-aspartate = L-glutamate + oxaloacetate - so this annotation would be wrong anyway ?

Thanks, Pascale

[deustp01]

2-oxoglutarate (aka alpha-ketoglutarate) is a TCA cycle intermediate, so it makes sense to make it an endpoint for a glutamate-specific pathway in the same way that glycolysis ends at pyruvate, not because that's a physiological endpoint but because it's a junction where metabolism unique to glucose feeds into multiple other pathways.

I've changed my mind on this (apologies for the youthful folly). Reactions that inter-convert glutamate and 2-oxo-glutarate are not necessarily part of glutamate catabolism just as reactions that convert ATP to ADP are not necessarily part of ATP catabolism. For both ADP and 2-oxo-glutarate, one possible fate is entry into a process that leads to complete catabolism so GO:0006539 glutamate catabolic process via 2-oxoglutarate now seems exactly right for such a fate, but other important processes can regenerate ATP and glutamate, so GO:0019551 glutamate catabolic process to 2-oxoglutarate seems risky. The only way I know to get from glutamate to 2-oxoglutarate involves a single reaction so a process term may not be possible. If one is wanted for the glutamate recycling (of which this reaction is one part), that could work.

[pgaudet]

OK, if I understand - the conclusion is to propose to obsolete 'GO:0019551 glutamate catabolic process to 2-oxoglutarate'

[deustp01]

the conclusion is to propose to obsolete 'GO:0019551 glutamate catabolic process to 2-oxoglutarate'

That is my latest conclusion, yes. But let's wait for agreement (or not) from @ukemi and @hdrabkin

[pgaudet]

@ukemi and @hdrabkin Thoughts ?

[hdrabkin]

I'm really having a hard time following these arguments. however, if this 'GO:0019551 glutamate catabolic process to 2-oxoglutarate is a single step, then maybe @ukemi is correct

[pgaudet]

As far as I can see glatamate degradation is catalyzed by glutamate dehydrogenase in a single step

Are there other possible pathways ? @amorgat ???

[pgaudet]

This looks like an interconversion, not a catabolic process.

[deustp01]

Now (as also 2 weeks ago but not last year), yes, this looks like an interconversion. If we pretend that reactions have purposes, it's hard to see how the "purpose" here is clearly catabolism, i.e., breakdown to waste products and energy - that's one possible fate of the reaction products but interconversion is another, at least in mammals. Prototrophic unicellular organisms may be different - the molecule may be committed to an irreversible catabolic fate earlier in the process, I don't know.

[ValWood]

Any action? close?

[hdrabkin]

propose to obsolete 'GO:0019551 glutamate catabolic process to 2-oxoglutarate'; 75 experimental annotations. Give notice Obsoletion reason: suggests a single step.

[pgaudet]

Where do you see 75 EXP ? I only see 6

[hdrabkin]

I saw six first myself, but then http://amigo.geneontology.org/amigo/search/annotation?q=GO:0019551 and filter on experimental.

[pgaudet]

But this includes ALL annotations to any gene product ALSO annotated to the term you looked for. See GO terms, you have annotations to

  | eag | ether a go-go |   | voltage-gated potassium channel activity |   | FlyBase | PMID:9497369

[pgaudet]

We should use 'consider' GO:0004069 L-aspartate:2-oxoglutarate aminotransferase activity Annotations should be easy to fix.

[ValWood]

It seems this one is ready to implement

obsolete 'GO:0019551 glutamate catabolic process to 2-oxoglutarate'; 6 EXP reason single step 'consider' GO:0004069 L-aspartate:2-oxoglutarate

[ValWood]

I'll take this....

[sjm41]

tagging @rozaru

[ValWood]

• [ ] fix MetaCyc mappings