AP3M2 is the mu3B cargo-recognition subunit of the heterotetrameric AP-3 membrane-coat adaptor complex. Its C-terminal mu-homology domain recognizes tyrosine-based sorting motifs in membrane-protein tails. AP-3 selects cargo for transport through endosomal compartments to lysosomes, lysosome-related organelles and neuronal synaptic vesicles. Mammalian AP-3 can associate with clathrin and also functions through clathrin-independent routes. In neurons, mu3B-containing AP-3 promotes formation of a subset of synaptic vesicles from endosomes and supports their cargo composition; mouse loss of mu3B reduces vesicle abundance and impairs hippocampal GABA release. Human expression and non-neuronal vertebrate phenotypes indicate functions beyond the historical neuron-specific designation.
| GO Term | Evidence | Action | Reason |
|---|---|---|---|
| GO:0005737 cytoplasm | IEA GO_REF:0000117 | ACCEPT | Summary: Cytoplasmic localization is integral to AP-3 coat recruitment; the historical ARBA-rule reproducibility concern is separate from this biological assessment. Reason: AP3M2 is the neuronal mu subunit of a cytoplasmic coat adaptor. PMID:9151686 describes recruitment of the neuronal p47/beta-NAP complex from brain cytosol onto membranes; its redistribution experiment assays delta in nonneuronal AP-3, not AP3M2 directly. This complex-level evidence supports cytoplasm as a core location alongside specific endosomal sites. The earlier ARBA00026971 condition-set audit remains a provenance question, not evidence against cytoplasmic localization, and is retained below with its version limitation. Propagation Review Root cause: SOURCE WEAK OR INFERRED Failure modes: SOURCE EVIDENCE WEAK Sources checked: ARBA:ARBA00026971 · ARBA rule conferring GO:0005737 SOURCE WEAK OR INFERRED Historical audit recorded in AP3M2-notes.md section 4.7: the then-served rule had 2,388 condition sets; the two sets matching AP3M2 signatures additionally required either Saccharomyces or sigma-subunit signatures absent from human mu adaptin. That snapshot did not reproduce the source row. The API lacked a meaningful version, so this does not establish that the original annotation was generated incorrectly. Preserve this provenance question independently of the accepted biological location. Supporting Evidence: PMID:9151686 Using a heterologous in vitro system we showed that the p47/β-NAP complex can be recruited from brain cytosol onto membranes prepared from tissue culture cells or liver fractions. |
| GO:0005769 early endosome | NAS PMID:23247405 Cell type-specific Rab32 and Rab38 cooperate with the ubiqui... | ACCEPT | Summary: Early endosome, asserted by ComplexPortal for the neuronal AP-3 complexes CPX-5053/CPX-5055 of which AP3M2 is the mu subunit. This is the compartment at which AP-3 actually works. Reason: Independent immuno-electron microscopy settled a long-running controversy in favour of the endosome: AP-3 decorates budding profiles on tubular sorting endosomes and routes LAMP-1/CD63 from there to lysosomes (PMID:15051738). The cited ComplexPortal reference makes the same placement in pigment cells, and GO's own definition of GO:0030123 says the complex 'is found associated with endosomal membranes'. This is the core site of action and is accepted as such. Supporting Evidence: PMID:23247405 Rab32 and Rab38 interact with AP-1, AP-3 and BLOC-2 on early/recycling endosome tubules, where cargo such as tyrosinase and Tyrp1 are loaded into vesicles or transport intermediates. PMID:15051738 Here, we show by immuno-electron microscopy that AP-3 is associated with budding profiles evolving from a tubular endosomal compartment that also exhibits budding profiles positive for AP-1. |
| GO:0005794 Golgi apparatus | IEA GO_REF:0000044 | KEEP AS NON CORE | Summary: A minor or conditional Golgi-associated AP-3 pool is compatible with its major endosomal role. Reason: Full primary PMID:15051738 reports a small TGN-associated fraction rather than zero Golgi localization, and discusses earlier dual-localization studies. Retain the inherited/complex-level localization as contextual; predominant tubular-endosome localization does not establish exclusion from Golgi. Supporting Evidence: PMID:15051738 This quantitative approach revealed low levels (4%) of staining in the TGN |
| GO:0005802 trans-Golgi network | IBA GO_REF:0000033 | KEEP AS NON CORE | Summary: The ancestral TGN location remains plausible as a minor or conditional AP-3 pool. Reason: Actual PTHR10529 root PTN000055849 is an ancestor of exact AP3M2 leaf PTN002540363 through AP-3 node PTN002237676. The recovered TGN IRD belongs to another branch. Full mammalian AP-3 immuno-EM reports low but nonzero TGN labeling (4%); major endosome residence does not imply exclusivity. Retain contextual inherited localization while the focused report examines isoform-specific evidence. Propagation Review Root cause: NO FAILURE NON CORE Sources checked: PANTHER:PTN000055849 SUPPORTS TRANSFER Actual target ancestry verified. TGN loss at PTN000242370 is outside this lineage. Primary AP-3 data support a minor TGN pool, with mu3B-specific extent uncertain. Supporting Evidence: PMID:15051738 This quantitative approach revealed low levels (4%) of staining in the TGN |
| GO:0006886 intracellular protein transport | IEA GO_REF:0000002 | ACCEPT | Summary: A broader term correctly captures the core AP-3 trafficking role. Reason: AP3M2 is a cytoplasmic coat-adaptor subunit that contributes cargo selection and vesicle transport. Axon cytoplasm includes membrane-bound carriers and their peripheral coats, not only freely soluble proteins. More specific endosomal/AP-3 terms do not invalidate these broader core categories. Supporting Evidence: PMID:11588176 However, only the neuronal form of AP-3 can produce synaptic vesicles from endosomes in vitro. |
| GO:0006896 Golgi to vacuole transport | IBA GO_REF:0000033 | ACCEPT | Summary: AP-3-mediated endosomal sorting contributes to Golgi-to-vacuole/lysosome transport. Reason: GO:0006896 specifies directed movement from Golgi to vacuole and does not require a direct route without an endosomal intermediate. AP-3 sorts biosynthetic lysosomal membrane cargo at endosomes, a step within that route. The verified ancestral node lies on the AP3M2 lineage; a yeast direct route and mammalian endosomal step do not make the broad process incompatible. Propagation Review Root cause: NO FAILURE CORE Sources checked: PANTHER:PTN000055849 SUPPORTS TRANSFER Root transport IBD is on the target path; the recovered transport IRD PTN002575694 is not. Term does not exclude an endosomal intermediate. Supporting Evidence: PMID:15051738 Based on these data, we propose that AP-3 defines a novel pathway by which lysosomal membrane proteins are transported from tubular sorting endosomes to lysosomes. |
| GO:0008021 synaptic vesicle | IEA GO_REF:0000107 | KEEP AS NON CORE | Summary: Synaptic vesicle, transferred from the rat orthologue. The rat annotation comes from a deep synaptic-vesicle proteome whose authors name AP-3 as an endosomal-related visitor of the SV fraction rather than a resident. Reason: The proximate source is rat Ap3m2 (P53678), which QuickGO shows carries GO:0008021 with EXP and IDA from PMID:33376223, an ultra-deep SV-fraction proteome. That paper explicitly separates residents from visitors and puts AP-3 in the second group, and its per-protein dataset is not in our cached text, so the strength of the call for Ap3m2 itself cannot be checked here. There is genuine independent support for AP-3 on synaptic vesicles - neuronal AP-3 binds purified synaptic vesicles in vitro - but that is coat recruitment to a membrane the complex is acting on, which makes 'is_active_in synaptic vesicle' a transient association rather than residency. Kept for that reason, and kept non-core because the mu3B-specific activity is upstream, at the endosome. Propagation Review Root cause: SOURCE WEAK OR INFERRED Failure modes: SOURCE EVIDENCE WEAK Sources checked: UniProtKB:P53678 · Ap3m2 / p47B (Rattus norvegicus), the 1:1 orthologue SOURCE WEAK OR INFERRED Rat Ap3m2, Swiss-Prot, PTHR10529. Its GO:0008021 rows are EXP and IDA from PMID:33376223; the source paper classifies AP-3 among likely SV visitors, so the donor annotation rests on co-fractionation rather than on demonstrated residency. ensembl:ENSRNOP00000025675 · Ensembl protein for rat Ap3m2 SUPPORTS TRANSFER Orthology identifier only. Ensembl Compara projects the annotation mechanically once the orthology call is made; the judgement therefore rests entirely on the donor annotation, not on this identifier. Supporting Evidence: PMID:33376223 This may explain the presence of endosomal-related proteins (e.g., Stx7, AP3) or proteins of the AZ (e.g., Piccolo, Bassoon) in the SV proteome. PMID:11588176 In the presence of GTPgammaS both ubiquitous and neuronal forms of AP-3 can bind to purified synaptic vesicles. |
| GO:0008089 anterograde axonal transport | IEA GO_REF:0000107 | KEEP AS NON CORE | Summary: Anterograde axonal transport, from mouse Ap3m2. The mouse annotation is an IMP citing a study whose only AP-3 allele is mocha (Ap3d1), which removes both AP-3A and AP-3B. Reason: Retain a contextual role for the neuronal AP-3 cargo-sorting complex in delivery of vesicle membrane proteins into axons. Full PMID:21998198 perturbs AP-3 delta (mocha), so it supplies complex-level evidence rather than an AP3M2-specific knockout. The mu3B neuronal assembly and budding studies make participation plausible; AP3M2 need not itself be the motor to contribute cargo selection. This distinction limits evidence specificity without negating the process. Propagation Review Root cause: NO FAILURE NON CORE Sources checked: UniProtKB:Q8R2R9 SUPPORTS TRANSFER Mouse source is supported at AP-3 complex level by PMID:21998198; mu3B identity supported separately by PMID:15492041/11588176. No assertion of a target-specific axonal motor. Supporting Evidence: PMID:21998198 Our findings indicate a novel vesicle transport mechanism requiring BLOC-1 and AP-3 complexes for cargo sorting from neuronal cell bodies to neurites and nerve terminals. PMID:21998198 PI4KIIα is present in synapses, and its presynaptic levels are decreased in AP-3–null brains (Ap3d1mh/mh). file:human/AP3M2/AP3M2-uniprot.txt cell bodies for delivery into neurites and nerve terminals. |
| GO:0008089 anterograde axonal transport | ISS GO_REF:0000024 | KEEP AS NON CORE | Summary: Anterograde axonal transport, from mouse Ap3m2. The mouse annotation is an IMP citing a study whose only AP-3 allele is mocha (Ap3d1), which removes both AP-3A and AP-3B. Reason: Retain a contextual role for the neuronal AP-3 cargo-sorting complex in delivery of vesicle membrane proteins into axons. Full PMID:21998198 perturbs AP-3 delta (mocha), so it supplies complex-level evidence rather than an AP3M2-specific knockout. The mu3B neuronal assembly and budding studies make participation plausible; AP3M2 need not itself be the motor to contribute cargo selection. This distinction limits evidence specificity without negating the process. Propagation Review Root cause: NO FAILURE NON CORE Sources checked: UniProtKB:Q8R2R9 SUPPORTS TRANSFER Mouse source is supported at AP-3 complex level by PMID:21998198; mu3B identity supported separately by PMID:15492041/11588176. No assertion of a target-specific axonal motor. Supporting Evidence: PMID:21998198 Our findings indicate a novel vesicle transport mechanism requiring BLOC-1 and AP-3 complexes for cargo sorting from neuronal cell bodies to neurites and nerve terminals. PMID:21998198 PI4KIIα is present in synapses, and its presynaptic levels are decreased in AP-3–null brains (Ap3d1mh/mh). file:human/AP3M2/AP3M2-uniprot.txt cell bodies for delivery into neurites and nerve terminals. |
| GO:0016183 synaptic vesicle coating | NAS PMID:15537701 Genetic analysis of the neuronal and ubiquitous AP-3 adaptor... | MODIFY | Summary: Synaptic vesicle coating, asserted by ComplexPortal for the neuronal AP-3 complex. The GO term is defined as clathrin-coated-pit formation in the presynaptic endocytic zone, which is the AP-2 pathway, not the AP-3 one. Reason: The cited neuronal AP-3 studies support sorting and budding of synaptic vesicles from endosomal membranes. GO:0016183 specifically describes clathrin-coated pits at the presynaptic plasma membrane, while GO:0048488 describes retrieval of membrane constituents from that surface. Endosomal vesicle reformation is more directly represented by GO:0016182. This source-specific refinement does not assert that AP3M2 can never participate in another endocytic context. Proposed replacements: synaptic vesicle budding from endosome Supporting Evidence: PMID:11588176 However, only the neuronal form of AP-3 can produce synaptic vesicles from endosomes in vitro. |
| GO:0016192 vesicle-mediated transport | IEA GO_REF:0000002 | ACCEPT | Summary: A broader term correctly captures the core AP-3 trafficking role. Reason: AP3M2 is a cytoplasmic coat-adaptor subunit that contributes cargo selection and vesicle transport. Axon cytoplasm includes membrane-bound carriers and their peripheral coats, not only freely soluble proteins. More specific endosomal/AP-3 terms do not invalidate these broader core categories. Supporting Evidence: PMID:11588176 However, only the neuronal form of AP-3 can produce synaptic vesicles from endosomes in vitro. |
| GO:0030119 AP-type membrane coat adaptor complex | TAS PMID:8076832 Two rat homologs of clathrin-associated adaptor proteins. | ACCEPT | Summary: A broader term correctly captures the core AP-3 trafficking role. Reason: AP3M2 is a cytoplasmic coat-adaptor subunit that contributes cargo selection and vesicle transport. Axon cytoplasm includes membrane-bound carriers and their peripheral coats, not only freely soluble proteins. More specific endosomal/AP-3 terms do not invalidate these broader core categories. Supporting Evidence: PMID:11588176 However, only the neuronal form of AP-3 can produce synaptic vesicles from endosomes in vitro. |
| GO:0030123 AP-3 adaptor complex | IBA GO_REF:0000033 | ACCEPT | Summary: AP3M2 is the mu3B subunit of AP-3. Reason: Actual PTHR10529 topology places the target below AP-3 IBD PTN002237676; complex assembly and neuronal vesicle-budding experiments independently agree. Pairwise identity is corroborative, not the phylogenetic test. Propagation Review Root cause: NO FAILURE CORE Sources checked: PANTHER:PTN002237676 SUPPORTS TRANSFER Current IBD and exact target lineage verified. The AP-3 node is ancestral to AP3M2. Supporting Evidence: PMID:11588176 However, only the neuronal form of AP-3 can produce synaptic vesicles from endosomes in vitro. |
| GO:0030123 AP-3 adaptor complex | NAS PMID:9151686 Characterization of the adaptor-related protein complex, AP-... | ACCEPT | Summary: AP-3 adaptor complex, asserted by ComplexPortal from the paper that named and completed the complex. Reason: Simpson et al. identified the delta and sigma3 subunits and showed by co-immunoprecipitation that they belong to the same heterotetramer as mu3 and beta3. ComplexPortal encodes the neuronal variants CPX-5053 and CPX-5055 with AP3M2 as the mu subunit, and UniProt states the same composition. Duplicating the IBA row with a literature-anchored assertion is appropriate, not redundant. Supporting Evidence: PMID:9151686 Antibodies raised against recombinant delta and sigma3 show that they are the other two subunits of the adaptor-like complex. file:human/AP3M2/AP3M2-uniprot.txt AP3B2), a medium adaptin (mu-type subunit AP3M1 or AP3M2) and a small |
| GO:0030131 clathrin adaptor complex | IEA GO_REF:0000002 | KEEP AS NON CORE | Summary: AP-3 can participate in clathrin-associated coat assemblies. Reason: The actual definition is a membrane coat adaptor complex that links clathrin to a membrane. Mammalian AP-3 beta3 binds clathrin and AP-3 is observed on clathrin-coated membranes (PMID:9545220, PMID:15051738). The structural study PMID:39705307 explicitly discusses both clathrin-dependent and independent functions. AP-3 being outside the named ontology children is not biological evidence of impossibility; retain a contextual complex-level capacity while mu3B-specific deployment is examined. Supporting Evidence: PMID:15051738 We conclude from these data that AP-3 does localize with clathrin, but to a lesser extent than does AP-1 PMID:39705307 AP-3 has a clathrin-binding motif in the β3 linker (22) and partially colocalizes with clathrin in vivo, although clathrin knockdown or inhibition does not fully disrupt traffic of AP-3-dependent cargoes (25). |
| GO:0030659 cytoplasmic vesicle membrane | IEA GO_REF:0000044 | ACCEPT | Summary: Cytoplasmic vesicle membrane, from the UniProt subcellular-location vocabulary. This is where a coat adaptor sits: on the cytoplasmic face of the budding vesicle. Reason: UniProt records the location as a peripheral membrane protein on the cytoplasmic side, forming part of the coat surrounding coated vesicles. The recent cryo-EM work makes the same point structurally, resolving membrane-inserting amphipathic helices on the delta and mu3 subunits of membrane-bound AP-3. Accepted as a core location. Propagation Review Root cause: NO FAILURE CORE Sources checked: UniProtKB-SubCell:SL-0089 · UniProt subcellular location: Cytoplasmic vesicle membrane SUPPORTS TRANSFER Reproduces the UniProt SUBCELLULAR LOCATION line, which further specifies peripheral membrane protein on the cytoplasmic side - exactly the topology of a coat adaptor. Supporting Evidence: file:human/AP3M2/AP3M2-uniprot.txt cytoplasmic face of coated vesicles located at the Golgi complex. PMID:39705307 Compared to the mGL control, both of the AH fusions robustly bound to SLBs |
| GO:0035615 clathrin-cargo adaptor activity | IBA GO_REF:0000033 | UNDECIDED | Summary: Clathrin-cargo coupling by mu3B requires contextual adjudication. Reason: AP3M2 is a cargo-binding subunit of AP-3, and mammalian AP-3 has both clathrin-associated and independent functions. Thus independence alone does not justify removing the ancestral clathrin claim. GO:0035615 additionally specifies formation of endocytic vesicles; whether the neuronal endosomal route and its clathrin-associated states satisfy that scope remains unresolved. Actual root IBD is ancestral to the target; donor names alone do not demonstrate bad propagation. Propagation Review Root cause: UNRESOLVED Sources checked: PANTHER:PTN000055849 UNRESOLVED Current clathrin-cargo IBD is at the actual family-root ancestor of AP3M2. Biological clathrin association is credible, but exact target endocytic-coupling scope remains unresolved. Supporting Evidence: PMID:15051738 We conclude from these data that AP-3 does localize with clathrin, but to a lesser extent than does AP-1 PMID:39705307 AP-3 has a clathrin-binding motif in the β3 linker (22) and partially colocalizes with clathrin in vivo, although clathrin knockdown or inhibition does not fully disrupt traffic of AP-3-dependent cargoes (25). PMID:11588176 However, only the neuronal form of AP-3 can produce synaptic vesicles from endosomes in vitro. |
| GO:0035651 AP-3 adaptor complex binding | IEA GO_REF:0000107 | UNDECIDED | Summary: The mouse-derived AP-3-complex-binding assertion requires source-level clarification. Reason: GO:0035651 does not state that the binding protein must be external to the complex. Constitutive subunit status therefore does not by itself disqualify AP3M2. The actual mouse donor Q8R2R9 has IDA PMID:19010779; its abstract confirms purification of cross-linked AP-3 assemblies, but the target-specific interaction topology could not be verified in accessible full text. Retain uncertainty rather than invent an external-ligand restriction. Propagation Review Root cause: UNRESOLVED Sources checked: UniProtKB:Q8R2R9 UNRESOLVED Ortholog IDA source PMID:19010779 verified, but full target-level cross-linking/MS evidence unavailable after official and PMC attempts. No assertion that the curated source is wrong. |
| GO:0035654 clathrin-coated vesicle cargo loading, AP-3-mediated | NAS PMID:9545220 Association of the AP-3 adaptor complex with clathrin. | ACCEPT | Summary: AP-3 cargo loading can occur in clathrin-associated trafficking. Reason: Source PMID:9545220 directly reports mammalian AP-3/clathrin interaction and colocalization, corroborated by full PMID:15051738 and PMID:39705307. The term accommodates AP-3 linking cargo to clathrin in relevant organisms/contexts. Retain conserved cargo loading without claiming every AP-3 carrier is clathrin coated or that the term is globally erroneous. Supporting Evidence: PMID:15051738 We conclude from these data that AP-3 does localize with clathrin, but to a lesser extent than does AP-1 PMID:39705307 AP-3 has a clathrin-binding motif in the β3 linker (22) and partially colocalizes with clathrin in vivo, although clathrin knockdown or inhibition does not fully disrupt traffic of AP-3-dependent cargoes (25). |
| GO:0036465 synaptic vesicle recycling | NAS PMID:15537701 Genetic analysis of the neuronal and ubiquitous AP-3 adaptor... | ACCEPT | Summary: Synaptic vesicle recycling, asserted by ComplexPortal for the neuronal AP-3 complex. The correct parent for the endosomal limb of the synaptic vesicle cycle, which is where AP-3B acts. Reason: GO:0016182, the term this review proposes on two other rows, is a descendant of GO:0036465, so this row is the true parent of the specific claim rather than a competing one. The cited study shows that neuronal and ubiquitous AP-3 make non-redundant contributions to the protein content of synaptic vesicles, and the mu3B knockout shows the vesicles themselves are fewer and smaller. Accepted. Supporting Evidence: PMID:15537701 Consistently, beta3B deficiency compromised synaptic zinc stores assessed by Timm's staining and the synaptic vesicle targeting of membrane proteins involved in zinc uptake (ZnT3 and ClC-3). PMID:15492041 Thus, these results indicate that AP-3B is involved in the biogenesis of synaptic vesicles in hippocampus in vivo. |
| GO:0048488 synaptic vesicle endocytosis | IEA GO_REF:0000107 | MODIFY | Summary: Synaptic vesicle endocytosis, transferred from a SynGO annotation on mouse Ap3m2. The donor paper is the mu3B knockout, and its own conclusion is endosomal budding, which GO expresses with a different term. Reason: The cited neuronal AP-3 studies support sorting and budding of synaptic vesicles from endosomal membranes. GO:0016183 specifically describes clathrin-coated pits at the presynaptic plasma membrane, while GO:0048488 describes retrieval of membrane constituents from that surface. Endosomal vesicle reformation is more directly represented by GO:0016182. This source-specific refinement does not assert that AP3M2 can never participate in another endocytic context. Propagation Review Root cause: TERM SCOPING PROBLEM Failure modes: COMPARTMENT OR COMPLEX MISMATCH Sources checked: UniProtKB:Q8R2R9 · Ap3m2 (Mus musculus), the 1:1 orthologue SUPPORTS SOURCE BUT NOT TARGET Mouse Ap3m2 carries GO:0048488 with both IMP and IDA from SynGO, citing PMID:15492041 - the mu3B knockout, the single strongest piece of evidence on this gene family. The donor gene and the donor experiment are exactly right; only the branch of the ontology is not, since the paper's own reading places AP-3B on endosomal budding rather than presynaptic retrieval. ensembl:ENSMUSP00000147967 · Ensembl protein for mouse Ap3m2 SUPPORTS TRANSFER Orthology identifier only. Ensembl Compara projects the annotation mechanically once the orthology call is made; the judgement therefore rests entirely on the donor annotation, not on this identifier. Proposed replacements: synaptic vesicle budding from endosome Supporting Evidence: PMID:11588176 However, only the neuronal form of AP-3 can produce synaptic vesicles from endosomes in vitro. |
| GO:0048490 anterograde synaptic vesicle transport | IEA GO_REF:0000107 | KEEP AS NON CORE | Summary: Anterograde synaptic vesicle transport, from mouse Ap3m2. Same donor annotation and same mocha-allele caveat as the anterograde axonal transport rows. Reason: Retain a contextual role for the neuronal AP-3 cargo-sorting complex in delivery of vesicle membrane proteins into axons. Full PMID:21998198 perturbs AP-3 delta (mocha), so it supplies complex-level evidence rather than an AP3M2-specific knockout. The mu3B neuronal assembly and budding studies make participation plausible; AP3M2 need not itself be the motor to contribute cargo selection. This distinction limits evidence specificity without negating the process. Propagation Review Root cause: NO FAILURE NON CORE Sources checked: UniProtKB:Q8R2R9 SUPPORTS TRANSFER Mouse source is supported at AP-3 complex level by PMID:21998198; mu3B identity supported separately by PMID:15492041/11588176. No assertion of a target-specific axonal motor. Supporting Evidence: PMID:21998198 Our findings indicate a novel vesicle transport mechanism requiring BLOC-1 and AP-3 complexes for cargo sorting from neuronal cell bodies to neurites and nerve terminals. file:human/AP3M2/AP3M2-uniprot.txt cell bodies for delivery into neurites and nerve terminals. |
| GO:0048490 anterograde synaptic vesicle transport | ISS GO_REF:0000024 | KEEP AS NON CORE | Summary: Anterograde synaptic vesicle transport, from mouse Ap3m2. Same donor annotation and same mocha-allele caveat as the anterograde axonal transport rows. Reason: Retain a contextual role for the neuronal AP-3 cargo-sorting complex in delivery of vesicle membrane proteins into axons. Full PMID:21998198 perturbs AP-3 delta (mocha), so it supplies complex-level evidence rather than an AP3M2-specific knockout. The mu3B neuronal assembly and budding studies make participation plausible; AP3M2 need not itself be the motor to contribute cargo selection. This distinction limits evidence specificity without negating the process. Propagation Review Root cause: NO FAILURE NON CORE Sources checked: UniProtKB:Q8R2R9 SUPPORTS TRANSFER Mouse source is supported at AP-3 complex level by PMID:21998198; mu3B identity supported separately by PMID:15492041/11588176. No assertion of a target-specific axonal motor. Supporting Evidence: PMID:21998198 Our findings indicate a novel vesicle transport mechanism requiring BLOC-1 and AP-3 complexes for cargo sorting from neuronal cell bodies to neurites and nerve terminals. file:human/AP3M2/AP3M2-uniprot.txt cell bodies for delivery into neurites and nerve terminals. |
| GO:0098978 glutamatergic synapse | IEA GO_REF:0000107 | KEEP AS NON CORE | Summary: AP3M2-associated machinery acts at glutamatergic as well as inhibitory synapses. Reason: The mouse mu3B study reports vesicle morphological changes in excitatory and inhibitory terminals. Lack of a glutamate-release defect in the tested assay does not refute a cellular-component annotation; localization does not require the same physiological phenotype in both synapse types. Supporting Evidence: PMID:15492041 Morphological abnormalities in excitatory and inhibitory presynaptic terminals |
| GO:0098982 GABA-ergic synapse | IEA GO_REF:0000107 | ACCEPT | Summary: GABA-ergic synapse, from a SynGO annotation on mouse Ap3m2 citing the mu3B knockout. This is the synapse type where loss of the protein produces a functional defect, and it is the defining phenotype of the gene. Reason: Deleting mu3B selectively impairs potassium-evoked GABA release while leaving glutamate release intact, reduces the vesicular GABA transporter in hippocampal synaptosomes while other synaptic vesicle proteins are unchanged, shrinks the vesicles of inhibitory terminals, and causes spontaneous epileptic seizures. Among all the compartment terms on this gene this is the one with direct functional support, and the donor is the gene's own orthologue annotated from that very study. Accepted as core. Propagation Review Root cause: NO FAILURE CORE Sources checked: UniProtKB:Q8R2R9 · Ap3m2 (Mus musculus), the 1:1 orthologue SUPPORTS TRANSFER Mouse Ap3m2, annotated by SynGO with IMP and IDA from PMID:15492041 - the knockout of this very gene, not of a partner subunit. The 1:1 orthology is unambiguous (both proteins are the sole mu3B in their genome and both are classified into PTHR10529), so the transfer is as strong as an ortholog transfer gets. ensembl:ENSMUSP00000147967 · Ensembl protein for mouse Ap3m2 SUPPORTS TRANSFER Orthology identifier only. Ensembl Compara projects the annotation mechanically once the orthology call is made; the judgement therefore rests entirely on the donor annotation, not on this identifier. Supporting Evidence: PMID:15492041 However, the K+-evoked release of GABA, but not of glutamate, was impaired in μ3B−/−ΔNeo mice at 8 wk old or over PMID:15492041 mu3B-/- mice suffered from spontaneous epileptic seizures. file:human/AP3M2/AP3M2-deep-research-affinage.md Ap3m2 knockout mice develop spontaneous epileptic seizures and show altered alcohol-related behaviors, linking the protein to GABAergic transmission |
| GO:1904115 axon cytoplasm | IEA GO_REF:0000108 | ACCEPT | Summary: A broader term correctly captures the core AP-3 trafficking role. Reason: AP3M2 is a cytoplasmic coat-adaptor subunit that contributes cargo selection and vesicle transport. Axon cytoplasm includes membrane-bound carriers and their peripheral coats, not only freely soluble proteins. More specific endosomal/AP-3 terms do not invalidate these broader core categories. Supporting Evidence: PMID:11588176 However, only the neuronal form of AP-3 can produce synaptic vesicles from endosomes in vitro. |
Loading supporting content…
Download this section (compressed HTML)Q: Is the 'neuron-specific' description of mu3B still correct for the human protein, given that human expression atlases report low tissue specificity and the strongest single-cell signal in melanocytes?
Suggested experts: Hiroshi Ohno, Victor Faundez
Q: Does any cargo require mu3B rather than mu3A? The YxxPhi preferences of the two differ, but no cargo has been shown to depend on that difference, and the best-characterised AP-3B-dependent protein, VGAT, appears to use a dileucine signal read by sigma3 instead.
Suggested experts: Hiroshi Ohno, Juan S. Bonifacino
Q: The reconstituted human structure 9C5B uses AP3M1, as the existing sequence/structure mapping establishes. Does the AP3M2-containing complex show the same cargo-coupled conformational transitions and context-dependent clathrin recruitment?
Suggested experts: Richard W. Baker
Q: Is mu3B destabilised in AP3B2-null neurons, as beta3B is in mu3B-null brain? This determines whether DEE48 should be understood as loss of the beta3B subunit or of the entire neuronal AP-3B complex.
Suggested experts: Caroline W. Beck, Victor Faundez
Experiment: Knock out AP3M2 and, separately, AP3M1 in a human melanocytic line (e.g. MNT-1) and score melanosome maturation by electron microscopy together with tyrosinase and TYRP1 localisation. A phenotype in the AP3M2 line would show the paralogue is used outside neurons and would connect the human melanocyte expression signal to the zebrafish iridophore requirement.
Hypothesis: Human AP3M2 acts in lysosome-related organelle biogenesis in pigment cells, not only in neurons.
Type: CRISPR knockout with organelle ultrastructure and cargo localisation readout
Experiment: Reconstitute human AP-3 cores differing only in the mu subunit and compare their binding to a library of cytoplasmic-tail peptides, then validate the top differential hits by rescuing mu3B-null neurons with wild-type mu3B, mu3A, or a mu3B carrying the mu3A residues at the nine cargo-contacting positions identified in PDB 9C5B.
Hypothesis: mu3B and mu3A confer different cargo repertoires on otherwise identical AP-3 complexes.
Type: In vitro reconstitution and peptide-array binding with knockout rescue
Experiment: In cultured mu3B-null hippocampal neurons, separate the two vesicle-formation routes with pHluorin-tagged VGAT and VGLUT1: measure plasma-membrane retrieval kinetics after stimulation, and in parallel follow endosomal budding using a temperature-block and release protocol. AP-3B loss should leave retrieval kinetics intact while reducing the endosome-derived vesicle pool.
Hypothesis: AP3M2 contributes to endosomal synaptic-vesicle reformation; simultaneous assays can determine whether it also affects plasma-membrane retrieval under particular stimulation conditions.
Type: Live-cell pHluorin imaging in knockout neurons
Experiment: Express the AP3M2 139-156 segment as an mGreenLantern fusion and assay binding to supported lipid bilayers with high-salt and detergent washes, as was done for the mu3A helix, then test the effect of replacing its hydrophobic face on AP-3B recruitment to endosomes in neurons.
Hypothesis: The mu3 linker amphipathic helix predicted here contributes to membrane binding by the neuronal complex.
Type: Supported-lipid-bilayer pelleting assay with structure-guided mutagenesis
What is not known — curated, literature-grounded statements of the open unknowns (the inverse of core functions).
Gap: It is not known whether human AP3M2 is functionally a neuronal protein. The 'neuron-specific mu3B' description rests on a 1994 rat Northern blot, while human expression atlases report the gene detected in all tissues with low tissue specificity and its highest single-cell signal in melanocytes and spermatogenic cells, not neurons.
OPEN BIOLOGYCURATION BP_DARK
What is known: The rodent restriction itself is solid, and the mouse knockout phenotype is unambiguously neurological. What is unestablished is that the human orthologue has the same restricted distribution, and therefore whether the human protein's principal role is the neuronal one.
Significance: Every neuronal term on this gene arrives by orthology from rodent. If the human paralogue is broadly expressed, a second, non-neuronal function is being missed - and the zebrafish pigment-cell requirement and the human melanocyte enrichment point the same way.
What would resolve it: Cell-type-resolved protein-level measurement of AP3M2 versus AP3M1 in human tissues, and a test of whether AP3M2 loss impairs melanosome biogenesis in human melanocytes.
Provenance (the field's own admissions):
Gap: There is no experimental GO annotation of any kind on human AP3M2. All 26 GOA rows are propagated - phylogenetic, orthology-based, rule-based, or author-statement - so every claim about the human protein rests on rodent genetics, on the complex, or on the family.
OPEN CURATIONBIOLOGY WHOLLY_DARK
What is known: The protein is not uncharacterised: the complex it belongs to is structurally solved, its mouse orthologue has a clean knockout phenotype, and mu3B itself was assayed for YxxPhi binding. The gap is specifically the absence of any experiment performed on the human gene product.
Significance: It means the whole annotation set is only as good as the propagation chain. Counting the rows rather than the impression: 20 of the 26 GOA rows carry a WITH/FROM and so were given a propagation_review - the six that do not are the five ComplexPortal NAS rows and the ProtInc TAS row, and the 21st propagation_review in the file belongs to the proposed NEW row. Of those 20, 13 came back with a non-NO_FAILURE root cause, through seven distinct causes - the PTHR10529 family-root subfamily leak (3 rows), a mouse IMP citing a study that perturbs Ap3d1 rather than Ap3m2 (4 rows, plus 1 more derived from them by inter-ontology inference), an InterPro2GO clathrin-branch mis-mapping, an ARBA rule whose published conditions cannot fire on this protein, a synaptic-vesicle proteome whose authors call AP-3 a visitor, a subunit annotated as a binder of its own complex, and one SynGO branch choice that the donor paper itself argues against.
What would resolve it: Any direct experiment on human AP3M2 - immunoprecipitation of the human AP-3B complex, a YxxPhi binding assay with the human protein, or a knockout in a human neuronal or melanocytic line.
Provenance (the field's own admissions):
Gap: Whether mu3B can assemble into AP-3 without beta3B, and therefore whether AP3M2 has any function independent of the AP-3B complex, is undetermined. Only one direction of the interdependence has been tested.
OPEN BIOLOGY RESIDUAL_SUBGAP
What is known: Deleting mu3B in mouse leaves beta3B barely detectable, which establishes that beta3B needs mu3B. The converse experiment - mu3B stability and localisation in AP3B2-null neurons - was not found.
Significance: AP3B2 loss of function causes developmental and epileptic encephalopathy 48 in humans. Whether that disease reflects loss of the whole AP-3B complex or of beta3B alone depends on this answer.
What would resolve it: Immunoblot and immunoprecipitation of mu3B in AP3B2-null neurons, and a test of whether mu3B can substitute into beta3A-containing complexes.
Provenance (the field's own admissions):
Loading supporting content…
Download this section (compressed HTML)Loading supporting content…
Download this section (compressed HTML)Loading supporting content…
Download this section (compressed HTML)Loading supporting content…
Download this section (compressed HTML)Loading supporting content…
Download this section (compressed HTML)