ALDH7A1 (antiquitin, alpha-aminoadipic semialdehyde dehydrogenase) is the horse ortholog of the mammalian ALDH7 family aldehyde dehydrogenase. Its conserved role is the NAD+-dependent oxidation of alpha-aminoadipic semialdehyde (in equilibrium with delta1-piperideine-6-carboxylate) to alpha-aminoadipate in L-lysine catabolism; the orthologous human enzyme also oxidizes betaine aldehyde and lipid-peroxidation-derived aldehydes. The enzyme is a homotetramer of the aldehyde-dehydrogenase fold with a catalytic glutamate and cysteine. Mammalian ALDH7A1 is found mainly in mitochondria and the cytosol, the two pools arising from transcripts with and without an amino-terminal mitochondrial targeting peptide, with smaller nuclear and membrane-associated pools. A0A9L0RRL6 is one of several Ensembl-predicted products of the horse ALDH7A1 gene: it starts at the mitochondrial-precursor methionine but skips one 54-nt coding exon, removing 18 residues (human 65-82, two beta strands at the start of the aldehyde-dehydrogenase domain). Catalytic and NAD-binding residues are retained, but whether this exon-skipped product is expressed or active has not been tested.
Summary: Condition-dependent Golgi-membrane pool of ALDH7A1, shown for the human enzyme and propagated by PAINT to the vertebrate ALDH7A1 clade.
Reason: Human ALDH7A1 associates with Golgi membranes and generates NADH there during hypoxia and starvation (PMID:31492851). PAINT curators placed GO:0000139 at node PTN000192435, whose seeded descendants include human, mouse and zebrafish ALDH7A1, so the horse ortholog (PANTHER PTHR43521:SF1, together with mouse, rat and bovine ALDH7A1) descends from the node carrying this assertion. This is a peripheral, stress-regulated pool rather than the main site of lysine catabolism, so it is kept as non-core. The row is a gene-level inference; the exon-skipped A0A9L0RRL6 product itself has not been localized.
Summary: Broad NAD+-dependent aldehyde dehydrogenase activity, demonstrated for human ALDH7A1 and asserted by PAINT at a deep node of the ALDH7 family.
Reason: Purified human ALDH7A1 oxidizes betaine aldehyde, lipid-peroxidation aldehydes such as 4-HNE, and medium-chain aldehydes in addition to AASA (PMID:20207735, PMID:21338592). PAINT placed GO:0004029 at node PTN000192422, seeded by both Arabidopsis ALDH12A1 (AT5G62530, a P5C dehydrogenase) and human ALDH7A1, i.e. at a node ancestral to both the ALDH7 and ALDH12 branches, so broad aldehyde dehydrogenase activity is an ancestral property that the horse ortholog inherits. A0A9L0RRL6 retains the catalytic Glu (horse 278 = human 296) and Cys (horse 312 = human 330) and all mapped NAD-binding residues. Caveat recorded in notes; the 18-residue exon skip in this particular gene model has an untested effect on folding, so the annotation is accepted as a property of the horse ALDH7A1 gene product rather than a measured property of this splice form.
Summary: The defining ALDH7A1 activity, oxidation of alpha-aminoadipic semialdehyde to alpha-aminoadipate; core function of the horse ortholog.
Reason: Human ALDH7A1 oxidizes AASA as a purified enzyme (PMID:20207735), and pathogenic ALDH7A1 variants abolish its P6C/AASA dehydrogenase activity (PMID:16491085). PAINT curators placed GO:0004043 at node PTN000192435 with human, mouse and zebrafish seeds, a vertebrate (or deeper) node from which horse ALDH7A1 descends; horse is in PANTHER subfamily PTHR43521:SF1 with the mouse, rat and bovine orthologs. The horse gene product is therefore expected to carry this activity. A0A9L0RRL6 is a non-canonical Ensembl model (ENSECAT00000119202) lacking the second coding exon; its catalytic and NAD-binding residues are intact, but activity of this specific splice form is unverified (see identity notes).
Summary: Minor nuclear pool reported for mammalian ALDH7A1; gene-level, non-core.
Reason: ALDH7A1 protein was detected in nuclei as well as in cytosol and mitochondria (PMID:20207735), and PAINT asserted GO:0005634 at node PTN000192435, from which the horse ortholog descends. No nuclear function is established, so the location is kept as non-core.
Summary: Mitochondrial ALDH7A1 is the precursor form that A0A9L0RRL6 represents; core location.
Reason: Mammalian ALDH7A1 has mitochondrial and cytosolic transcripts (PMID:20207735), and PAINT asserted GO:0005739 at node PTN000192435 with human and mouse seeds. A0A9L0RRL6 begins at the methionine corresponding to the start of the human mitochondrial precursor and pairs with all 26 residues of the human transit peptide, so this product carries the targeting signal. Import has not been measured in horse.
Summary: Cytosolic ALDH7A1 is a real gene-level pool, but in mammals it comes from the transcript lacking the transit peptide, which this product retains.
Reason: ALDH7A1 protein is found in the cytosol (PMID:20207735) and PAINT asserted GO:0005829 at node PTN000192435, so the horse ALDH7A1 gene product is expected to have a cytosolic pool. In human that pool corresponds to the variant lacking residues 1-28 (isoform 2). A0A9L0RRL6 carries the mitochondrial transit peptide, so cytosolic localization is not the expected fate of this particular product. It is kept as a gene-level, non-core location rather than accepted as core for this accession.
Summary: Stress-regulated membrane pool shown for human ALDH7A1 and propagated by PAINT; gene-level, non-core.
Reason: Human ALDH7A1 is recruited to cellular membranes, where it generates NADH that supports FSP1 and protects against ferroptosis (PMID:40233740, PMID:31492851). PAINT asserted GO:0005886 at node PTN000192435, from which horse ALDH7A1 descends. The pool is condition-dependent and peripheral to the lysine-catabolic role, so it is kept as non-core.
Summary: Betaine-aldehyde oxidation is demonstrated for human ALDH7A1 and propagated by PAINT, but is a secondary activity.
Reason: Purified human ALDH7A1 oxidizes betaine aldehyde and ALDH7A1 expression protects cells against hyperosmotic stress (PMID:20207735). PAINT asserted GO:0008802 at node PTN000192435. The physiological contribution relative to the dedicated betaine-aldehyde dehydrogenase ALDH9A1 is not established, so this is kept as a true but non-core activity, consistent with the human ALDH7A1 review.
Summary: Correct but uninformative InterPro2GO term; the specific aldehyde-dehydrogenase activities are supported.
Reason: The aldehyde-dehydrogenase domain (IPR015590) and the glutamate active-site signature (IPR029510) support oxidoreductase chemistry, but the activity is more precisely the NAD+-dependent oxidation of AASA and other aldehydes, both of which this review accepts. Replace with those descendants.
Summary: ALDH7A1 catalyses the AASA-to-alpha-aminoadipate step of lysine degradation; core process.
Reason: ALDH7A1 performs a catalytic step of lysine catabolism, the oxidation of AASA/P6C to alpha-aminoadipate, which is shared by the saccharopine and pipecolate routes. Loss of human ALDH7A1 activity causes accumulation of AASA/P6C (PMID:16491085, PMID:36225138), and PAINT curators placed GO:0019477 at node PTN000192435 with human, mouse and zebrafish seeds. Horse ALDH7A1 descends from that node. This is participation, since the enzyme catalyses the step, not merely a requirement.
NAD+-dependent oxidation of alpha-aminoadipic semialdehyde to alpha-aminoadipate, the ALDH7A1-catalysed step of L-lysine catabolism, in mitochondria. Inferred for the horse ALDH7A1 gene product from the PAINT node PTN000192435 (human, mouse and zebrafish seeds) and human biochemistry; A0A9L0RRL6 is an exon-skipped gene model whose own activity is untested.
Broad NAD+-dependent oxidation of reactive and medium-chain aldehydes (lipid-peroxidation products such as 4-HNE), an ancestral activity asserted by PAINT at PTN000192422, the node joining the ALDH7 and ALDH12 branches.
Q: Is the horse ALDH7A1 transcript ENSECAT00000119202, which skips the 54-nt second coding exon (human residues 65-82), supported by RNA-seq junction reads, or is it an annotation artefact?
Q: Should GO/UniProt annotation of ALDH7A1 in horse be anchored on the full-length Ensembl canonical product A0A9L0R3Y5 rather than on alternative gene models such as A0A9L0RRL6?
Suggested Experiments
Experiment: Express full-length horse ALDH7A1 and the exon-skipped A0A9L0RRL6 product in E. coli, then compare solubility, oligomeric state (size-exclusion chromatography) and NAD+-dependent AASA and 4-HNE oxidation kinetics.
Hypothesis: Removing the two beta strands encoded by the skipped exon disrupts folding or tetramerization of ALDH7A1.
External Prediction Reviews
These computational predictions are reviewed separately from the GOA annotation set used for this review. The assessments below are from this project and do not constitute official GO annotations or endorsement by GO/UniProt. They are not included in the existing annotation review above.
The exact glyceraldehyde-3-phosphate substrate prediction cannot be established from ALDH7A1 family membership or from experiments on other aldehydes. The selected horse model also has an internal deletion of unresolved functional consequence.
Prediction method: ProtNLM2 · Version: UniProt API snapshot 2026-09-08
Review rationale: Human ALDH7A1 is an alpha-aminoadipic-semialdehyde dehydrogenase with experimentally demonstrated activity on several additional aldehydes. Those results establish ALDH chemistry but do not establish glyceraldehyde-3-phosphate oxidation. The selected horse sequence retains catalytic residues but has an internal 18-residue deletion, further limiting an exact substrate transfer. No target GOA overlap or direct glyceraldehyde-3-phosphate assay was found in the inspected evidence, and PubMed searches for ALDH7A1 or antiquitin with glyceraldehyde(-3-phosphate) returned no reports. The kinetic survey of purified human ALDH7A1 found only low affinity for the related short-chain aldose glyceraldehyde, and non-phosphorylating glyceraldehyde-3-phosphate dehydrogenase (GAPN) activity is characteristic of a different ALDH family. The different canonical substrate alone does not prove the predicted reaction is impossible for a broad-specificity ALDH, so the prediction remains uncertain rather than receiving a confident paralog-error label.
Supporting Evidence:
PMID:20207735: "including the osmolyte precursor, betaine aldehyde, lipid peroxidation-derived aldehydes, and the intermediate lysine degradation product, alpha-aminoadipic semialdehyde."
PMID:20207735: "ALDH7A1 exhibited low affinity for benzaldehyde, an aromatic aldehyde; propanal, a short-chain aliphatic aldehyde; and glyceraldehyde, a short-chain aldose."