Source of four IEA rows on AMDHD1, from signatures of differing specificity: IPR005920 (imidazolonepropionase, family-specific) yields the correct process term GO:0006548, whereas IPR006680 (Amidohydro_1) and IPR011059 (metal-dependent hydrolase composite domain) are superfamily-level matches and yield only the generic hydrolase-activity ancestors GO:0016787 and GO:0016810. The split explains why one InterPro2GO row is accepted and two are modified.

Basis for the GO:0003674 root molecular_function placeholder assigned by BHF-UCL in 2012. By the ND convention the placeholder records that no molecular-function data were available at that time; it is superseded now that GO:0050480 is annotated by both IBA and IEA, which is why that row is removed rather than kept as non-core.

GO_Central phylogenetic inference assigns AMDHD1 imidazolonepropionase activity (GO:0050480) and the L-histidine catabolic process (GO:0006548) from node PANTHER:PTN000505272, with the characterized bacterial enzyme UniProtKB:Q9HU91 among the descendant evidences. Human AMDHD1 retains the substrate-binding and metal-ligating residues of the clade, so it sits inside the group that inherited the function.

Combined multi-method electronic annotation supplies four AMDHD1 rows: the specific catalytic term GO:0050480 from the RHEA:23660 / EC 3.5.2.7 mapping, its cyclic-amide parent GO:0016812 from ARBA plus IPR005920, the process term GO:0006548, and the cytoplasm location from UniProtKB-SubCell SL-0086. Because the same pipeline emits both GO:0050480 and its ancestor, the ancestor is modified rather than kept.

BioPlex 2.0 is a proteome-scale affinity-purification mass-spectrometry network covering more than 25% of human protein-coding genes and over 56,000 candidate interactions. AMDHD1 appears only as one node of that screen, with KLHL23 (Q8NBE8) as its reported partner; the paper does not characterise AMDHD1 or assign it a function.

BioPlex 3.0 plus an HCT116 network independently reproduce the AMDHD1-KLHL23 (Q8NBE8) association in a second cell background. This replication is what makes the interaction itself credible, and is why removing the bare protein-binding term is a statement about the term rather than about the measurement.

Reactome curates the AMDHD1 reaction as cytosolic and names substrate and product explicitly, which supports both the cytosol location and the use of the specific GO:0050480 term over the cyclic-amide parent. The same entry states that the existence of the human enzyme is inferred only from high-throughput screening (Yamada et al. 2004) rather than from a direct assay - the clearest statement in this review of why a human enzymological characterization is still outstanding.

The major route of histidine catabolism proceeds in four steps to glutamate, converting one tetrahydrofolate to 5-formiminotetrahydrofolate. AMDHD1 catalyses the third of these steps, between urocanate hydratase (UROC1) and glutamate formimidoyltransferase (FTCD), which places AMDHD1 inside the process rather than merely adjacent to it.

AMDHD1 catalyses hydrolytic cleavage of the carbon-nitrogen bond in imidazolone-5-propanoate to form N-formimidoyl-L-glutamate, the third step of the universal histidine degradation pathway (EC 3.5.2.7, RHEA:23660). The catalytic activity, EC number and pathway position are all assigned by similarity to UniProtKB:P42084, which is why the entry name is "Probable imidazolonepropionase".

The enzyme binds one zinc or ferric iron ion per subunit, coordinated by residues 260 and 334, which each carry a BINDING line for either metal; substrate-binding residues are 159, 192 and 263, while 159 and 336 bind the product N-formimidoyl-L-glutamate. It belongs to the metallo-dependent hydrolases superfamily, HutI family. This is the basis for the proposed GO:0046872 metal ion binding annotation, which is asserted at the parent level because UniProt gives Zn(2+) and Fe(3+) as an explicit either/or at equal evidential strength.

Subcellular location is Cytoplasm, with experimental evidence (ECO:0000269|PubMed:39143229); the same report is the source of the SMAD2/3/4 interaction and the TGF-beta tumor-suppressor activity recorded in the FUNCTION line. That paper is not cached in this repository, so the moonlighting role informs the description only and is not annotated.

The only interaction recorded in the entry is with KLHL23 (Q8NBE8, NbExp=3), matching the two proteome-scale screens; no function is attached to it.