Hsd11b2 (rat, UniProt P50233) curation notes

Re-review 2026-10-04

What GOA changed

The refreshed Hsd11b2-goa.tsv carries 25 rows and no new or retired rows relative
to the reviewed set (no retired: true rows, no action: PENDING rows were seeded).
The refreshed UniProt record (P50233) is substantially expanded relative to the text
the original review quoted: it now documents the androgen-directed activities
(11beta-hydroxytestosterone to 11-ketotestosterone, 11beta-hydroxyandrostenedione to
11-ketoandrostenedione), the oxysterol activity
[UniProtKB:P50233 "Catalyzes the conversion of 7-beta-hydroxycholesterol into 7-ketocholesterol (7-oxocholesterol), a precursor of 7-keto,27-hydroxycholesterol, which activates smoothened (SMO) and the smoothened signaling pathway"],
the corresponding Rhea reaction (RHEA:68656), a KM of 10 nM for corticosterone, a
microsome/ER subcellular location, the SAME-rat disruption phenotype
[UniProtKB:P50233 "manifesting hypokalaemia by 15 days of age and contributing to salt sensitive hypertension"],
and a CAUTION note
[UniProtKB:P50233 "Rats and mice do not produce appreciable cortisol, because they do not express the 17-alpha hydroxylase (Cyp17a1) enzyme in the adrenals."].

Actions changed

GO:0045880 positive regulation of smoothened signaling pathway (IEA
GO_REF:0000107; ISO GO_REF:0000121): REMOVE -> KEEP_AS_NON_CORE.
The previous reason asserted that any smoothened effect must be an indirect
consequence of glucocorticoid inactivation. The donor evidence contradicts this.
The mouse donor (UniProtKB:P51661 / MGI:MGI:104720) annotation comes from
PMID:30340023, where HSD11B2 is characterized as an oxysterol synthase acting
upstream of SMO
PMID:30340023,
positioned between PTCH1 and SUFU
PMID:30340023,
with rescue by the downstream oxysterols
PMID:30340023.
The glucocorticoid-mediated explanation was tested and excluded in that paper
PMID:30340023.
Kept as non-core (not ACCEPT) because the enzyme catalyses a biosynthetic step
upstream of the SMO ligand rather than acting within the pathway, and because the
effect is confined to cell types expressing the enzyme in Hedgehog-active domains
(cerebellar external granule layer, Hedgehog-associated medulloblastoma).
propagation_review on both rows changed from
PROPAGATION_BAD/ROLE_CONFLATION/SUPPORTS_SOURCE_BUT_NOT_TARGET to
NO_FAILURE_NON_CORE/SUPPORTS_TRANSFER.
GO:0047022 7-beta-hydroxysteroid dehydrogenase (NADP+) activity (IEA
GO_REF:0000120; ISS GO_REF:0000024; ISO GO_REF:0000121):
MARK_AS_OVER_ANNOTATED -> KEEP_AS_NON_CORE.
The previous reason called this a family-level SDR transfer. It is not: the
oxidation is directly demonstrated and isomer-specific
PMID:30340023,
inhibitor-sensitive in cells and cilia
PMID:30340023,
and the rat UniProt record carries both the conversion and RHEA:68656.
Retained as non-core with an explicit caveat: the GO term names NADP(+) as
cofactor, and the cofactor actually used for the 7beta-oxidation by the rat enzyme
has not been established experimentally (the characterized 11beta-dehydrogenase
chemistry is NAD(+)-dependent). The ISS row gained a propagation_review naming
the mouse anchor; the ISO row's block changed from
PROPAGATION_BAD/WRONG_ORTHOLOG_OR_PARALOG+FUNCTIONAL_DIVERGENCE to
NO_FAILURE_NON_CORE/SUPPORTS_TRANSFER.

Other edits (no action changes)

description rewritten as standalone biology. The prior text contained curation
commentary ("The review accepts ...", "Falcon (Edison Scientific) deep research
corroborates ...") and asserted Smoothened signalling was an over-extension, which
is no longer the review's position. The new text adds the KM, the microsome/ER
location, the AME/salt-sensitive-hypertension consequence of loss, and the two
secondary activities (11-ketoandrogen production, oxysterol synthesis).
GO:0005783 endoplasmic reticulum (IEA): kept KEEP_AS_NON_CORE, but the
supported_by quote was the FUNCTION sentence, which says nothing about location;
replaced with the SUBCELLULAR LOCATION line.
GO:0007565 female pregnancy (ISO): kept KEEP_AS_NON_CORE, support replaced with the
UniProt statement that actually bears on pregnancy
[UniProtKB:P50233 "Plays an important role in maintaining glucocorticoids balance during preimplantation and protects the fetus from excessive maternal corticosterone exposure"],
and a propagation_review added for the mouse donor.
GO:0034650 cortisol metabolic process (IEA, ISO): MARK_AS_OVER_ANNOTATED retained;
the UniProt CAUTION note on Cyp17a1 and cortisol in rats and mice was added as
direct support for the species argument.
PMID:30340023 added to references with four findings and a reference_review
(relevance HIGH, correctness VERIFIED; cached full text read).
core_functions unchanged: NAD(+)-dependent glucocorticoid inactivation remains the
single core activity, and the two re-rated rows are explicitly secondary.

Open questions

Which cofactor does the rat enzyme use for the 7beta-hydroxycholesterol oxidation?
GO:0047022 specifies NADP(+); the UniProt reaction (RHEA:68656) is written with
NADP(H), but no rat (or, as far as the cached paper shows, mouse) measurement
establishes the cofactor for this side reaction. If it is NAD(+)-dependent, the
correct term would be a 7beta-hydroxysteroid dehydrogenase (NAD+) activity term,
which does not obviously exist.
The androgen-directed activities on the UniProt record (11beta-hydroxytestosterone
to 11-ketotestosterone, 11beta-hydroxyandrostenedione to 11-ketoandrostenedione) are
"By similarity" and currently have no GO annotation on the rat gene. No NEW term was
proposed: the evidence is transferred, not rat-specific, and GO:0070523
(11-beta-hydroxysteroid dehydrogenase (NAD+) activity) already covers the chemistry
at the general-substrate level.

Stale UniProt quotes (2026-10-10): replaced 7 UniProtKB:P50233 supporting_text quotes (an abridged FUNCTION sentence no longer in the refreshed flat file) with verbatim CATALYTIC ACTIVITY reactions (generic 11beta-hydroxysteroid and corticosterone) and a verbatim FUNCTION sentence on glucocorticoid inactivation from the current Hsd11b2-uniprot.txt.