BCAP31 is best curated as an ER-resident membrane protein-sorting factor with three linked roles:
The ERAD role is directly supported by the Cell paper showing that BAP31 binds Sec61beta and TRAM, associates with CFTRDeltaF508, and promotes retrotranslocation and proteasomal degradation [PMID:18555783 "BAP31 is an endoplasmic reticulum protein-sorting factor"; PMID:18555783 "promotes its retrotranslocation from the ER and degradation by the cytoplasmic 26S proteasome system"]. This supports keeping the positive regulation annotations and adding the direct ERAD pathway annotation (GO:0036503).
The ER-mitochondria contact role is supported by the Tom40 study: BAP31 forms an ER-mitochondria bridging complex and stimulates NDUFS4 translocation to mitochondria for complex I assembly [PMID:31206022 "BAP31 acts as a key factor in mitochondrial homeostasis"; PMID:31206022 "interacts with mitochondria-localized proteins, including Tom40"]. This supports the MAM localization and protein localization to mitochondrion annotations as core/non-incidental.
The primary localization is ER membrane. The early BAP31 paper identifies p28 Bap31 as a polytopic ER protein PMID:9334338, and UniProt also states ER membrane plus ERGIC/cis-Golgi shuttling [UniProt:P51572 "May shuttle between the ER and the intermediate compartment/cis-Golgi complex"].
ERGIC/Golgi membrane annotations are acceptable as trafficking-related localizations, but ER membrane is the core site. Lipid droplet annotation is treated as non-core because the cited study is a lipid droplet fraction proteomics experiment in HuH7 cells, not a mechanistic BCAP31 localization paper PMID:14741744. Plasma membrane remains undecided because the original surface-antigen study supports antibody-accessible surface expression PMID:8706661, but later work establishes ER/ERGIC residency as the primary biology.
BCAP31 is a caspase substrate and ER-mitochondria apoptosis platform component, but broad apoptosis annotations overstate the intact protein's core function. The original Bap31 apoptosis paper shows caspase cleavage and the p20 fragment, not a constitutive apoptotic regulator function for intact BCAP31 [PMID:9334338 "cleavage of p28 at the two caspase recognition sites"; PMID:9334338 "p20 fragment induces apoptosis"]. The Fis1 paper similarly frames the event as Fis1 facilitating Bap31 cleavage into p20Bap31 PMID:21183955.
Therefore:
Generic protein binding rows should be removed. BCAP31 has many real interactors, including Sec61beta, TRAM, Derlin-1, CFTR, BCL2, CASP8, Fis1, TOMM40, and viral SH protein, but GO:0005515 does not capture the function. The biologically informative terms are ERAD pathway/regulation, retrograde ER-to-cytosol transport, ER-to-Golgi transport, MHC class I protein binding, MAM localization, and mitochondrial protein localization.
GO:0036503 ERAD pathway is justified because the strongest mechanistic evidence places BCAP31 directly in the ERAD substrate-handling pathway rather than only upstream regulation. PMID:18555783 shows BCAP31 associating with the misfolded CFTRDeltaF508 substrate, Sec61 translocon components, and Derlin-1, and reducing BAP31 reduces proteasomal degradation [PMID:18555783 "Depletion of BAP31 reduces the proteasomal degradation of DeltaF508"; PMID:18555783 "associates physically and functionally with the Derlin-1 protein disclocation complex"].
GO:0005886) call referenced under "Localization Calls" was resolved as MARK_AS_OVER_ANNOTATED in PR #317. The PMID:8706661 surface-antigen detection is real but represents a trafficking intermediate / overexpression artifact in transfected cells; later work establishes ER/ERGIC residency as the primary biology, consistent with the cytoplasmic KKXX retrieval motif.GO:0030136) IEA was resolved as MARK_AS_OVER_ANNOTATED in the same PR — Ensembl rat transfer with no human-specific support, and BCAP31 traffics via COPI/COPII machinery rather than the clathrin pathway.GO curators raised a PAINT issue (PANTHER:PTN000294723) questioning the
involved_in GO:0006888 endoplasmic reticulum to Golgi vesicle-mediated transport
IBA on human BCAP31 and its orthologs, plus GO:0070973 protein localization to ER
exit site on the S. pombe ortholog SPAC9E9.04. The curator argument is that
BCAP31 acts upstream of the ER exit site as a quality-control / translocation
chaperone, so the anterograde-transport IBA is an erroneous over-propagation:
Curator M. Feuermann (GO_Central PAINT) commented (2026-05-21) that the family
lacked the right terms and recommends MF GO:0140388 protein translocation
chaperone activity; he is re-annotating the family. Curator V. Wood notes the
core biology is BAP31 interacting with Sec61 translocons and promoting
retrotranslocation of CFTRΔF508 via the Derlin-1 complex (PMID:18555783).
Actions taken in this review update:
- GO:0006888 and GO:0070973: ACCEPT → MARK_AS_OVER_ANNOTATED.
- core_functions entry 1: MF changed GO:0140597 protein carrier activity →
GO:0140388 protein translocation chaperone activity; directly_involved_in
changed GO:0006888 → GO:0036503 ERAD pathway; "cargo receptor" framing
removed in favour of translocon-associated chaperone framing.
- description and several annotation reason fields reworded to drop the
"cargo receptor" characterization.
The normal projection contains 61 distinct source assertions from 64 raw rows, including eight previously omitted partner records and three exact duplicate raw records. All source identities, qualifiers and partners are retained. Three old authored NEW entries are assessed separately; their proposed withdrawal removes no GOA assertion.
This reassessment supersedes the earlier substrate-only apoptosis rationale. Intact BCAP31 helps assemble the FIS1/procaspase-8 signaling platform, and a cleavage product can transmit a signal. The carrier role is grounded in client binding/delivery, while an intrinsic ATP-dependent ratchet has not been established by the available evidence. The two disputed PAINT trafficking assertions remain unresolved; the real curator issue and positive MHC-I experiments must both be represented.
The rat clathrin transfer is contradicted by the cited donor fractionation experiment. In contrast, the experimental cell-surface and lipid-droplet annotations are retained as secondary contexts without speculative artifact explanations. Mouse spermatogenesis remains unresolved because the traced evidence is an expression survey whose BCAP31-specific body evidence was not inspected.
Reactome's cytosolic cleavage entity is the terminal 238–246 fragment, not p20. The two machine-derived UniProt alternative products are a separate splicing record and do not imply RefSeq-numbering equivalence or isoform-specific function.
Supported generic interactions are kept as non-core under the user-supplied ActionEnum. The CFTR client and the separate FIS1/CASP8 platform rows receive evidence-specific refinements; machinery, combined-partner and high-throughput records are not indiscriminately relabeled. No NEW process assertion or new direct quotation is proposed.
The existing Falcon report and all raw/cached sources are preserved. Normal cache availability and selected external primary reading are recorded separately. Full-paper, figure-pixel, supplemental-pair and PAINT-node reconstruction are not claimed where not performed. Five of six additional normal sources were subsequently imported through Source92 and reassessed before this integrated proposal. PMID:19342655 remained unrecovered; its provisional extra citations are omitted, while its earlier bounded web-reading evidence and the actual normal-fetch failure remain recorded. The precise recovered-source read scopes are recorded in each reference review.
The integrated proposal retains 61 source annotations: 18 ACCEPT, 36 KEEP_AS_NON_CORE, three MODIFY, three UNDECIDED and one REMOVE. Three old authored NEW entries are withdrawn separately. It incorporates the two machine-derived UniProt alternative products and three reviewed core functions, without assigning function by isoform number. No new direct quotation is added.
The revised authored pathway supersedes the earlier substrate-only exclusion of apoptosis, an intrinsic ATP-translocation mechanism, and universal cytoplasmic coiled-coil client recognition. ER-associated p20 is distinguished from the cytosolic terminal fragment. HACD2 is stabilized in its reported client context; it is not described as a BAP31-promoted degradation target. TOMM40-associated localization and turnover remain bounded to the experiments. The provider report and all raw or cached sources are unchanged.
Source92 recovered five normal records: PMID:9396746, PMID:14581517, PMID:15187134, PMID:17056546 and PMID:23967155. Their canonical bytes match the staged bytes used for the bounded reading recorded here. PMID:19342655 returned exit1 without a timeout or quarantined output and remains absent; this establishes a failed normal retrieval only, not retraction, inaccessibility or absence of biological evidence. Earlier official abstract/selected web reading is preserved in the original consultation. Its provisional extra citations are removed from rows1,2,9 and the revised pathway; no original GOA source assertion is removed. The two detailed trafficking inferences remain UNDECIDED and the broad vesicle-transport context remains non-core, supported by the available15187134/17056546 material. New sources do not alter the reviewed annotation actions or three core functions. The human MHC-I experiment is context dependent; the isolated-domain coiled-coil study does not disprove full-length BAP29/BAP31 association.
The applied review retains all 61 distinct source assertions from 64 raw GOA rows, restores eight omitted assertions and 32 supporting-entity lists, and withdraws only three old authored NEW entries. The two alternative products are derived from the unchanged UniProt record without assigning RefSeq or functional equivalence. The pathway now reflects the reviewed client-handling, signaling-platform and selective mitochondrial-localization evidence.
Focused validation and pathway PMID checks passed. The 23 advisories comprise 18 supported generic protein-binding annotations retained as non-core under the supplied ActionEnum; two UNDECIDED IBA annotations whose ancestral nodes were not reconstructed; one unchanged Falcon report not used as direct annotation evidence; one ERAD core process absent from the source annotation block; and one ontology-label version discrepancy. GO:0140597 retains the current official label protein carrier activity verified during the review, whereas the local validator expected protein carrier chaperone. No additional NEW assertion is introduced to silence these advisories. Rendering and generated Codex EDIT history validation are recorded separately.
Five of six requested additional normal references were recovered and reviewed within the documented reading limits. PMID:19342655 remains absent after the recorded normal retrieval failure; this is neither a retraction claim nor a claim that its publication is inaccessible. Its provisional extra citations were omitted without changing the associated decisions. No repository-wide validation result is claimed.
This entry supersedes the earlier counts and the statement that both trafficking IBAs remain unresolved. All 61 source assertions and both UniProt alternative products remain intact. The current decisions are 18 ACCEPT, 37 KEEP_AS_NON_CORE, three MODIFY, two UNDECIDED and one REMOVE. No NEW annotation is added. Older notes above are retained as a journal of prior interpretations.
GO issue 6385 contains a genuine recommendation from M. Feuermann on 21 May and his completion message on 3 June. The previous reasons incorrectly dismissed the whole discussion as copied AI text. His proposed GO:0140388 is considered explicitly: its current definition requires an ATP-dependent translocation ratchet. The CFTR study supports client delivery to the Derlin-1 degradation pathway but does not establish that motor mechanism PMID:18555783. GO:0140597 retains the current official label, protein carrier activity; protein carrier chaperone is an official synonym and the label expected by the older validator ontology.
Mouse MHC-I association and reduced mSec31 colocalization after combined Bap29/Bap31 loss support exit-site localization as a non-core function PMID:15187134. The vesicle-transport IBA remains UNDECIDED because direct participation in that step and the revised ancestral assertion were not resolved here. Conditional or redundant human export effects are positive evidence with limits; lack of a surface-MHC-I decrease after depletion does not prove absence of function PMID:17056546.
The ERAD core now uses the accepted positive-regulation term GO:1904294. The mitochondrial-localization core records an explicit molecular-function knowledge gap: crosslinked association with TOMM40 and precursor proteins, together with fractionation and turnover effects, does not by itself isolate the proposed capture-and-handoff reaction PMID:31206022. This gap is preserved rather than filled with an unsupported carrier or motor assertion.
The clathrin donor is rat/RGD; MGI supplies the cross-species comparison display. The primary rat-liver fractionation result remains decisive: PMID:9396746. Reference assessments now emphasize each paper's contribution and reading limits, without internal recovery identifiers or repeated procedural sentences.
The final independent science peer passed before applying this follow-up. just validate human BCAP31 exited 0 with 21 warnings: 18 retained non-core protein-binding records, one unresolved IBA lacking an independently inspected PAINT node, one uncited provider-output warning, and one ontology-version label warning. The current official label for GO:0140597 is protein carrier activity; the local validator expects its older synonym protein carrier chaperone. Supported interaction records remain non-core under the supplied action definitions; no unsupported replacement activity or propagation diagnosis was invented to suppress warnings. The separate pkg_resources deprecation notice is an environment warning. The pathway PMID check passed. No repository-wide validation is claimed.
just render human BCAP31 succeeded. A new Codex/gpt-6 history session was scaffolded for PR #3605, with its validation recorded separately. The prior history and pathway, immutable source files, and all 30 publication/Reactome dependencies are preserved unchanged. This entry supersedes the earlier 23-warning validation count.
This entry supersedes the earlier statement that the revised PAINT node was not inspected. The cached PAINT table contains four IBDs at PTN000294723: GO:0005789 (20260528), GO:0140388 (20260603), GO:0030970 and GO:2000060 (both 20260521). Neither GO:0006888 nor GO:0070973 remains. The two historical GOA IBAs, both dated 20170228, are now MARK_AS_OVER_ANNOTATED. Their source node and mouse donors are retained unchanged as historical provenance. The client-specific mouse and human results remain valid; this change follows the retired phylogenetic assertions rather than interpreting an incomplete phenotype as absence of function [PMID:15187134; PMID:17056546].
The GO:0140388 IBD is seeded by UniProtKB:P51572, BCAP31 itself. This is expected when target experimental evidence grounds a PAINT ancestral assertion; it is not circular. The new node has been considered explicitly. The broader GO:0140597 carrier description is retained while a specific knowledge gap records how the narrower ATP-dependent ratchet assignment maps to the client-delivery experiments. The review neither removes that current IBD nor claims to have inspected its originating experimental annotation. Reconciliation with the curator requires the underlying assay, not a donor-count argument PMID:18555783.
The official AmiGO record for GO:0140597, read on 2026-09-30, displays these fields:
Accession: GO:0140597
Name: protein carrier activity
Synonyms: protein carrier chaperone, protein chaperone
Last file loaded: 2026-08-06
Thus the local validator's alternative label reflects a different ontology version. The GO:0140657 parent page lists GO:0140388 under ATP-dependent activity, and the FlyBase GO record describes ATP-dependent binding cycles that drive membrane translocation. These sources make the label and mechanistic distinction checkable without changing source-derived IDs.
Short verbatim anchors now support the signaling and mitochondrial-localization cores [PMID:21183955; PMID:31206022]. PMID21183955's retained normal cache is abstract-only, contrary to the PR comment's full-text characterization; the previously documented selected external Results remain separate. The mitochondrial quote supports the reported localization phenotype and does not establish an autonomous import motor. No NEW annotation is added, and supported generic interactions retain their existing non-core decisions under the supplied action definitions.
The two historical trafficking IBAs now record SOURCE_STALE_OR_MISSING for PANTHER:PTN000294723. This classification means that their respective transferred terms no longer appear on the inspected node. It does not infer the reason for the historical source withdrawal or a target-specific loss of function. No biological failure subtype is assigned. The current BCAP31-seeded translocation-chaperone IBD remains explicitly acknowledged above.
The independently reviewed follow-up and narrowly reviewed propagation metadata pass focused validation (20 warnings), history validation, and rendering. Eighteen warnings concern supported generic binding retained as non-core under the supplied action definitions; one records reliance on direct primary/database sources rather than the unchanged generated report. The remaining warning reflects the older local label for GO:0140597; the current official name, protein carrier activity, was checked against AmiGO. The two prior missing-propagation warnings are resolved with an explicit retired-source classification, without inferring a biological cause. All 61 source objects and two products are preserved. No new global validation pass is claimed.
This entry supersedes earlier explanations that attributed supported generic-binding retention to the action definitions alone. The governing authority is the user's explicit ClinGen project instruction on published main, which states:
retain a supported, biologically correct
GO:0005515(protein binding) annotation asKEEP_AS_NON_COREwhen no evidence-backed, more specific replacement has been established.
The same instruction calls for a more informative replacement when the evidence supports one and an unresolved decision when the relevant evidence cannot be adjudicated. It does not verify an uninspected interaction. For this review, 18 generic-binding assertions remain non-core; the three source-specific refinements are the CFTR client-carrier assertion and the separate CASP8 and FIS1 signaling-adaptor assertions. TOMM40 association is not one of those three refinements. Machinery association, combined-partner records and proteomic proximity do not acquire a narrower molecular activity merely from a partner name. The current action counts remain 18 ACCEPT, 36 KEEP_AS_NON_CORE, three MODIFY, two MARK_AS_OVER_ANNOTATED, one UNDECIDED and one REMOVE across all 61 source assertions. Both alternative products and all three core-function records are retained. No new annotation is added.
The unchanged UniProt P51572 record describes a 246-residue protein with three transmembrane helices (residues 7–27, 44–64 and 103–123) and a cytoplasmic C-terminal region. It contains no nucleotide-binding or ATPase activity annotation. This is a boundary on the available annotation, not proof that BCAP31 cannot couple directly or through another protein to ATP-dependent work. The existing open knowledge gap now makes that distinction explicit; it neither assigns an intrinsic motor nor dismisses complex-dependent participation.
The inspected PTHR12701 PAINT table retains an IBD for GO:0030970, retrograde protein transport, ER to cytosol, at PTN000294723, dated 20260521 and seeded by UniProtKB:P51572. This is current curator support for transport of unfolded or misfolded protein from the ER to the cytosol through the translocon; it does not specify which component supplies ATP-dependent work. The observation supplies context for the carrier/ERAD discussion and the retained GO:0140388 IBD. It does not restore the two different retired source terms, establish an autonomous BCAP31 ratchet, or create a NEW annotation. The term name and definition were checked against the QuickGO ontology API on 2026-10-09.
This follow-up adds no scientific quotation and changes no source identity, qualifier, partner, annotation action, reference assessment or existing supporting-text value. Earlier primary-source reading limits remain in force. It is a clarification of project authority and of the existing open mechanistic question.