Near-Duplicate Lipoic Acid Transport Reactions

[Devlin-Moyer]

Background

There are currently three reactions representing sodium-dependent import of lipoic acid via SLC5A6 (ENSG00000138074):
MAR01483: ATP [c] + H2O [c] + lipoic acid [e] + Na+ [e] --> ADP [c] + H+ [c] + lipoic acid [c] + Na+ [c] + Pi [c]
MAR06406: ATP [c] + H2O [c] + lipoic acid [e] + 2 Na+ [e] --> ADP [c] + H+ [c] + lipoic acid [c] + 2 Na+ [c] + Pi [c]
MAR09608: lipoic acid [e] + 2 Na+ [e] --> lipoic acid [c] + 2 Na+ [c]

Problems

SLC5A6 is a sodium-dependent transporter, and while the sodium concentration gradient it depends on to bring its various substrates into cells is established by ATP-dependent sodium transporters, SLC5A6 does not itself hydrolyze ATP (sources: 1, 2). None of the references currently associated with MAR01483 or MAR06406 found SLC5A6 to be ATP-dependent (many aren’t even relevant to this reaction), and most directly state that the stoichiometry is 2 Na⁺ to 1 substrate, which is consistent with MAR09068 but not MAR01483. MAR01483 and MAR06406 should be removed, since neither accurately represent the function of SLC5A6, and MAR09606 does.

Proposed changes

• Remove MAR01483 and MAR06406
• Merge annotations for MAR01483 and MAR06406 in reactions.tsv with those for MAR09608
• Set the references for MAR09608 to PMID:9516450;PMID:10329687;PMID:15561972

[haowang-bioinfo]

the provided evidence for SLC5A6, as well as its Uniprot annotation, support the removal of MAR01483 and MAR06406, please move on to implementation.

additionally, would suggest to associate MAR01483 with Rhea id RHEA:73379/RHEA:73380

for the references, it seems PMID:20980265 (source 2) is about iodide(I−) uptake, not lipoic acid?

[Devlin-Moyer]

oh yea I should probably not add that to the list of references for MAR09608, since it only provides evidence that SLC5A6 isn't ATP-dependent, which is justification to remove the other two reactions, rather than evidence supporting the existence of MAR09608; I'll update the proposed changes accordingly

[haowang-bioinfo]

fixed in #706