patched (ptc), Drosophila melanogaster — curation notes

UniProt: P18502 (PTC_DROME) · FlyBase: FBgn0003892 (CG2411) · 1286 aa · PANTHER PTHR46022 ("PROTEIN PATCHED")

Journal for the GO annotation review in ptc-ai-review.yaml. All assertions carry an inline
citation; [PMID:x "quote"] means the quoted string is verbatim in publications/PMID_x.md.

1. What the protein is

Twelve-pass integral membrane protein of the RND (resistance-nodulation-division) permease
superfamily, with a sterol-sensing domain (SSD) spanning TM2–TM6 (UniProt features: DOMAIN 428..583 /note="SSD"),
two large extracellular loops, and a ~180-residue disordered cytoplasmic C-terminal tail
(REGION 1116..1237 /note="Disordered"). TCDB classifies it under the RND superfamily
(DR TCDB; 2.A.6.6.2). It is the founding member of the Patched family; human PTCH1 (Q13635)
and PTCH2 (Q9Y6C5) sit in the same PANTHER family PTHR46022, which is the structural evidence
that the Hedgehog reception tier is conserved from insects to vertebrates
(see modules/hedgehog_signaling.yaml, annoton patched_receptor).

Two things are true simultaneously and are the frame for everything below:

2. Genetics that established the brake

Ingham, Taylor & Nakano showed that ptc acts permissively, repressing wingless transcription, and
that hh antagonises it — and proposed, before any biochemistry, that Ptc is the Hh receptor:
PMID:1653906
and PMID:1653906

Chen & Struhl separated the two jobs explicitly — reception/transduction and ligand sequestration:
PMID:8898207
and PMID:8898207

Denef et al. added the reciprocal trafficking logic, and — importantly for the mechanism question
below — argued for indirect regulation:
PMID:10966113,
PMID:10966113,
PMID:10966113

3. The receptor is Ptc plus a co-receptor

Ihog/Boi are required for pathway activation and bind Ptc directly:
PMID:20048000
The same paper is candid that direct Hh–Ptc binding had not been demonstrated biochemically in the fly:
PMID:20048000, and
PMID:20048000.

Camp et al. later dissociated the two roles genetically and put the ligand-binding job back on Ptc:
PMID:25231763, and
PMID:25231763.

So: Ptc binds Hh (GO:0097108) and is the Hh receptor (GO:0008158); Ihog/Boi raise the affinity and
are required for transduction but are not required for sequestration. Both GO terms are justified
on fly evidence alone.

Ihog also binds Hh itself PMID:16630821,
and the ptc IDA to GO:0097108 from this paper rests on the cell-binding assays
PMID:16630821.

4. Where it acts

5. The mechanism question: how does Ptc restrain Smo?

This is an open knowledge gap, and modules/hedgehog_signaling.yaml records it as such
("The biochemical mechanism by which ligand-free Patched restricts Smoothened remains undetermined").
The important point for this review is that the Drosophila record does not support a cholesterol-transport
molecular function for ptc. What the fly record does contain:

  1. Arguments against direct Ptc–Smo inhibition. PMID:26863604,
    with the stoichiometric argument spelled out in the same sentence — inhibition persists with Smo in
    50-fold molar excess, and substoichiometric Ptc represses Smo, i.e. the inhibition behaves catalytically.
    Denef et al. reached the same conclusion from trafficking PMID:10966113.
    The only fly physical evidence is complex-level co-fractionation
    PMID:17284519,
    which does not distinguish direct from indirect association.

  2. Ptc binds PI(4)P through its SSD. PMID:26863604,
    and Hh shifts the pool PMID:26863604.
    Loss of Ptc raises PI(4)P PMID:26863604.

  3. Ptc binds phosphatidic acid through its cytoplasmic tail, and this is required for Smo inhibition.
    PMID:37847757,
    PMID:37847757,
    PMID:37847757

  4. Ptc is a lipoprotein receptor and moves lipid in bulk, independently of Hh signalling.
    PMID:18198278,
    PMID:18198278,
    and the loss-of-function control PMID:18198278,
    with the authors' own separation of the two jobs PMID:18198278.

Decision taken in this review. The human PTCH1 framing — molecular function GO:0140303 intramembrane lipid carrier activity, cholesterol as the transported species — is grounded in
vertebrate structures and transport assays and in the curated human GO-CAMs. Drosophila UniProt
carries the same claim but flags it By similarity from Q13635, and the fly GOA contains no
GO:0140303 annotation and no cholesterol-transport annotation of any kind. The three fly lipid
observations that do exist point at three different lipids (PI(4)P, phosphatidic acid, lipophorin
cargo), two of which (PI(4)P, PA) are read as Ptc sequestering a Smo activator rather than
exporting a Smo inhibitor, and one of which (lipophorin) the authors explicitly uncouple from
signalling. On that record it would be an unsupported import to assert an intramembrane cholesterol
carrier activity for fly ptc. The review therefore grounds ptc's molecular function on what the fly
data show — GO:0008158 hedgehog receptor activity, GO:0097108 hedgehog family protein binding,
and the two lipid-binding activities — and raises the transport question in suggested_questions
rather than asserting it. No NEW molecular-function term is proposed.

This is also why GO:0005119 smoothened binding (IBA) is marked as an over-annotation rather than
accepted: the PAINT node placement is reasonable at family level, but in the fly the best evidence
argues the restraint is not by direct association, and the term invites reading a binding claim as
the mechanism.

6. Downstream / pleiotropic annotations

Every one of these is a consequence of ptc restraining Smo in a particular tissue; none is a
separate molecular activity, so they are kept as non-core (or, where the paper itself says the
effect is non-autonomous, marked as over-annotated).

Process Paper Note
Bolwig's organ morphogenesis PMID:10704398 abstract foregrounds hh/eya/so; a Ci- and Fu-independent branch
eye-antennal disc morphogenesis PMID:11934850 authors argue a positive, non-canonical Ptc–Smo–Babo route
ovarian somatic stem cells PMID:11279500 GOA term is the vague regulation of mitotic cell cycle; the phenotype is stem-cell proliferation
follicle cell development PMID:29093440 ptc(S2) FSC clones give supernumerary stalk cells
hippo signalling PMID:28667262 Hh acts upstream of Hippo; indirect
apoptosis PMID:23018595 explicitly non-cell-autonomous
axon guidance PMID:15754211 the paper's own title says the effect is via neural identity
tracheal migration PMID:24651658 ptc mutants used as a Hh gain-of-function tool
gonad development PMID:21377458 ptc is screen group com4
genital disc primordium PMID:15893978
wing shape PMID:16648592 quantitative morphometrics on heterozygous insertions; weak as evidence, but the term is independently sound

7. Protein-binding rows

Three GO:0005515 rows, all removed as uninformative (the interactions themselves are not disputed):

8. Loose ends