HADH (Q16836) review notes

Gene: HADH (HGNC:4799); aka SCHAD, HAD1, HADHSC, M/SCHAD.
Protein: Hydroxyacyl-coenzyme A dehydrogenase, mitochondrial (HCDH_HUMAN). 314 aa precursor;
12-aa mitochondrial transit peptide; mature 13-314. EC 1.1.1.35.

Core enzymatic function (SCHAD, EC 1.1.1.35)

HADH catalyzes the third step of the mitochondrial fatty-acid beta-oxidation spiral: the
NAD+-dependent oxidation of L-(3S)-3-hydroxyacyl-CoA to 3-ketoacyl-CoA.

• UniProt FUNCTION: "Mitochondrial fatty acid beta-oxidation enzyme that catalyzes the third
  step of the beta-oxidation cycle for medium and short-chain 3-hydroxy fatty acyl-CoAs (C4 to
  C10)" [Q16836 UniProt, ECO:0000269|PubMed:10231530, 11489939, 16725361].
• Reaction (Rhea:22432): "a (3S)-3-hydroxyacyl-CoA + NAD(+) = a 3-oxoacyl-CoA + NADH + H(+)";
  EC=1.1.1.35 [Q16836 UniProt CATALYTIC ACTIVITY].
• Substrate preference is short/medium chain. PMID:10231530
• Homodimer; crystal structures solved with NAD+/substrate. PMID:10231530
• Catalytic mechanism: His158 general base, Glu170 electrostatic assist PMID:10231530.
• Conformational interdomain shifting on cofactor/substrate binding PMID:10840044.

Distinct from HADHA (the long-chain, membrane-bound MTP alpha subunit). HADH is a soluble
matrix homodimer.

Subcellular location

Mitochondrion matrix. [Q16836 UniProt SUBCELLULAR LOCATION "Mitochondrion matrix
{ECO:0000305|PubMed:8687463}."] Confirmed by high-confidence mitochondrial proteomics
[PMID:34800366 mitochondrial proteome, HTP]. Note a legacy LIFEdb IDA "cytoplasm" annotation
(GO:0005737) likely reflects an overexpressed GFP-fusion artifact, not the native location.

Tissue expression

Expressed in liver, kidney, pancreas, heart and skeletal muscle. PMID:8687463 Notably expressed in islets of Langerhans PMID:21990309.

Moonlighting / regulatory role: GDH inhibition and insulin secretion (HHF4 / HHF7)

Loss-of-function HADH mutations cause hyperinsulinaemic hypoglycaemia. This was the FIRST
fatty-acid-oxidation defect associated with hyperinsulinism, revealing a link between FAO and
insulin secretion.

• First patient / P258L: PMID:11489939 The P258L variant abolishes catalytic activity PMID:11489939; <5% residual activity in
  fibroblast mitochondria PMID:11489939.
• All early patients had hyperinsulinism, suggesting a regulatory role PMID:16725361.
• MECHANISM (protein-protein, partly enzyme-independent): HADH binds and inhibits glutamate
  dehydrogenase 1 (GLUD1); loss of this inhibition de-represses GDH-driven amino-acid-stimulated
  insulin secretion. UniProt records this from mouse ortholog (Q61425): [Q16836 UniProt
  FUNCTION "Plays a role in the control of insulin secretion by inhibiting the activation of
  glutamate dehydrogenase 1 (GLUD1), an enzyme that has an important role in regulating amino
  acid-induced insulin secretion (By similarity)."] and [Q16836 UniProt SUBUNIT "Interacts with
  GLUD1; this interaction inhibits the activation of glutamate dehydrogenase 1 (GLUD1) (By
  similarity)."] This GDH-inhibition mechanism is the basis for it being "partly independent of
  beta-oxidation activity" — the protein-protein interaction is a separate molecular function
  (enzyme inhibitor activity) from the dehydrogenase catalysis.
• Mouse Hadh-/- KO: elevated insulin, low glucose PMID:21990309; islets hypersecrete insulin PMID:21990309

Disease (UniProt): HADH deficiency [MIM:231530] (hypoglycemia, hepatoencephalopathy, myopathy/
cardiomyopathy, sudden death) and Hyperinsulinemic hypoglycemia familial 4 (HHF4) [MIM:609975].
(OMIM also uses HHF7 in some sources; UniProt entry uses HHF4.)

Thermogenesis / body weight (mouse; ISS in human)

Mouse KO data implicate Hadh in adaptive thermogenesis and body-weight regulation.
PMID:21990309 and
PMID:21990309 The GO term positive regulation of cold-induced thermogenesis
(GO:0120162) is ISS from mouse — a downstream physiological consequence of impaired FAO, not a
distinct molecular function; treat as non-core.

Spermatogenesis (By similarity, mouse)

UniProt: [Q16836 UniProt FUNCTION "Plays a role in the maintenance of normal spermatogenesis
through the reduction of fatty acid accumulation in the testes (By similarity)."] Keyword-derived
GO:0007283 (IEA from UniProtKB-KW). Downstream/By-similarity; non-core.

Annotation-review reasoning summary

• Core MF: GO:0003857 (3S)-3-hydroxyacyl-CoA dehydrogenase (NAD+) activity — strongly supported
  by EXP/IDA (PMID:10231530, 11489939, 16725361), IBA, TAS. ACCEPT (core).
• NAD+ binding GO:0070403 (IDA PMID:10840044) — ACCEPT; molecular detail of the catalytic MF.
• identical protein binding GO:0042802 (IDA PMID:10231530) — homodimer; ACCEPT but non-core
  (descriptive of quaternary structure; the homodimer crystal structure supports it).
• GO:0016509 long-chain (3S)-3-hydroxyacyl-CoA dehydrogenase activity (IEA RHEA) — HADH prefers
  short/medium chain; long-chain is HADHA. UniProt does list a long-chain Rhea reaction "By
  similarity" (ECO:0000250|UniProtKB:P00348, pig). MODIFY/over-annotation: the defining,
  preferred function is short/medium-chain; long-chain is at best minor. Mark as over-annotated /
  keep non-core rather than core. Use MARK_AS_OVER_ANNOTATED (does not contradict but misrepresents
  preference).
• GO:0016491 oxidoreductase activity, GO:0016616 (CH-OH donor, NAD acceptor) — correct but
  generic parents of GO:0003857; KEEP_AS_NON_CORE.
• GO:0016740 transferase activity (UniProtKB-KW, in DR lines) — NOT in GOA tsv, so not a row to
  review; the "Transferase" keyword is spurious for HADH (it is an oxidoreductase). Not in scope.
• GO:0006631 fatty acid metabolic process (IEA InterPro), GO:0006635 fatty acid beta-oxidation
  (IBA, IDA, IEA) — beta-oxidation is the specific correct BP. ACCEPT GO:0006635 (core BP);
  GO:0006631 is the generic parent -> KEEP_AS_NON_CORE.
• mitochondrion GO:0005739 (IBA is_active_in, IDA HPA, IEA, HTP, TAS) and mitochondrial matrix
  GO:0005759 (IEA SubCell, TAS Reactome x7) — matrix is correct specific location. ACCEPT matrix
  (core); mitochondrion = generic parent, KEEP_AS_NON_CORE.
• cytoplasm GO:0005737 (IDA LIFEdb GO_REF:0000054) — overexpressed fusion-protein localization;
  HADH is matrix. MARK_AS_OVER_ANNOTATED (not native).
• Insulin-secretion terms: regulation of insulin secretion GO:0050796 (ISS UniProtKB GO_REF:0000024
  from mouse Q61425; also IEA Ensembl) — well-supported moonlighting role; KEEP_AS_NON_CORE
  (genuine but secondary). negative regulation of insulin secretion GO:0046676 (IEA Ensembl) is
  more precise and matches the GDH-inhibition mechanism -> KEEP_AS_NON_CORE.
• response to insulin GO:0032868, response to hormone GO:0009725, response to activity GO:0014823,
  response to xenobiotic GO:0009410 (all IEA Ensembl from rat ortholog) — generic transcriptional/
  physiological responses, weakly informative; KEEP_AS_NON_CORE (not contradicted, low value).
• positive regulation of cold-induced thermogenesis GO:0120162 (ISS x2) — mouse KO phenotype;
  KEEP_AS_NON_CORE.
• Reactome mitochondrial protein degradation rows (R-HSA-9838035/9838081/9838093/9838289) are
  CLPXP/LONP1 "binds/degrades matrix proteins" reactions — they annotate matrix LOCATION via
  HADH being a substrate, NOT a degradation function. The located_in GO:0005759 they assert is
  correct. ACCEPT as location evidence (non-core).

core_functions plan

1. (3S)-3-hydroxyacyl-CoA dehydrogenase activity (GO:0003857) / fatty acid beta-oxidation
   (GO:0006635) / mitochondrial matrix (GO:0005759). PRIMARY.
2. Enzyme inhibitor activity (GO:0004857) toward GLUD1 / negative regulation of insulin
   secretion (GO:0046676). SECONDARY moonlighting (By similarity / ISS — flag full_text basis).

2026-11-11 re-review corrections

The dated corrections here supersede the earlier action plan while retaining its original record. See HADH-deep-research-manual.md for source-by-source evidence and access limits.

• Cytoplasm (GO:0005737) includes organelles; the earlier fusion/import artifact explanation was unsupported. Retain the broad compartment.
• Pig enzyme data (PMID:2817332) support long-chain activity despite medium-chain preference. The human Rhea C16 assertion is an orthology transfer; retain it as non-core rather than calling preference a contradiction.
• Trace the four rat response transfers to PMIDs 12176671, 18191600, 8231758 and 7050060. Their measured responses support contextual non-core retention without claims of direct human experiments.
• Accept valid broader core-reaction and mitochondrial-location terms. Accept the insulin-regulatory role consistently with its core_functions entry while stating the mouse mechanistic evidence. No NEW term is proposed.
• His158/Glu170 and D45G/Y214H in their source papers use mature-chain numbering (precursor His170/Glu182 and D57G/Y226H). P258L already uses precursor numbering. Preserve source quotations.
• PMID:10231530 solved an NAD+-bound structure and modeled substrate placement; PMID:10840044 reports substrate/ternary structures. The inspected evidence establishes a homodimer, without the former unsupported qualifier obligate.
• PMID:20670938 and PMID:21990309 report different glucose-stimulated islet responses; retain their contexts separately. The former cached extraction omits methods/results despite its full-text flag.
• PMID:34800366 remains GOA-curated proteomics support, but the specific HADH supplementary row has not been independently checked. A title quotation does not establish gene-specific detection.

Falcon failed with HTTP403/missingJWT and Perplexity fallback with HTTP401/quota; manual primary-source research is explicitly labeled. Candidate: 41 contexts, 33 ACCEPT and 8 KEEP_AS_NON_CORE, no NEW/PENDING. Canonical application, validation and PR delivery remain pending.

PR #6312 review follow-up — 2026-11-11

Refined the two generic molecular-function ancestors (GO:0016491 and GO:0016616) to MODIFY → the already supported GO:0003857, while retaining the pathway-level fatty acid metabolic process and valid compartment annotations. The long-chain row now uses MARK_AS_OVER_ANNOTATED for uncertain physiological relevance, without denying the measured pig activity or calling it assay noise. Similar Km values alone do not establish comparable catalytic efficiency. Added the unresolved isozyme-selectivity limit to all four rat tissue-response rows, distinguishing RGD attribution from independent evidence inspection.

Recorded the PMID:10231530 UniProt RP discrepancy as a PROBABLE/CHECK source query because only the primary abstract was accessible. Removed the PMID:34800366 title-only quote, retaining its evidence-limit finding; all biological quotations are retained. Omitted the optional full_text_unavailable flag for the partially extracted PMID:20670938 and retained the explicit missing-Methods/Results caveat. Restored GLUD1, tissue expression and the supported catalytic-pair mechanism in the biological description; pig-to-human annotation provenance stays in the annotation rationale.

This follow-up supersedes the earlier candidate accounting: 41 native contexts, 31 ACCEPT, 7 KEEP_AS_NON_CORE, 2 MODIFY and 1 MARK_AS_OVER_ANNOTATED; no NEW/PENDING. A new append-only history record links PR #6312; the prior history record remains unchanged. The project checkbox remains unset until verified merge and delivery.