PTPRN2 (IA-2beta, phogrin) is a single-pass type I transmembrane protein of the IA-2 family of receptor-like protein-tyrosine phosphatases, expressed mainly in brain, pituitary and pancreatic islets. It resides in the membrane of neuroendocrine dense-core secretory granules and, in neurons, synaptic vesicles; after granule exocytosis it is transiently present at the plasma membrane and is recycled by clathrin/AP-2-dependent endocytosis. The precursor is processed in the granule pathway to a mature IA-2beta60 transmembrane form with a short luminal segment and a cytoplasmic PTP-like domain. That domain carries substitutions at conserved catalytic positions, and the protein has no meaningful activity on phosphoprotein substrates; instead, the purified transmembrane form dephosphorylates phosphoinositides (PI(3)P, PI(5)P and PI(4,5)P2) in a catalytic-cysteine-dependent manner and lowers plasma membrane PI(4,5)P2 when expressed in cells. Loss of PTPRN2 in mice reduces the number and half-life of dense-core vesicles, insulin content and glucose-stimulated insulin secretion, and (together with PTPRN) reduces brain monoamine content and neurotransmitter release. PTPRN2 is a major autoantigen in type 1 diabetes, and in breast cancer cells its PI(4,5)P2-lowering activity promotes cofilin-dependent migration.
| GO Term | Evidence | Action | Reason |
|---|---|---|---|
| GO:0004725 protein tyrosine phosphatase activity | IEA GO_REF:0000120 | REMOVE | Summary: Domain-based electronic assignment of protein tyrosine phosphatase activity to a PTP-like domain whose conventional catalytic activity is lost. Reason: The PTP-like domain of PTPRN2 carries substitutions at conserved catalytic positions. Direct biochemical work on the orthologous rat protein (same PANTHER subfamily PTHR46106:SF5) found no significant activity on phosphoproteins or pNPP, the knockout study describes IA-2beta as enzymatically inactive, and the human cell study states it has no activity against phosphoprotein substrates. UniProt also notes that the weak acidic-pH activity reported at cloning is of questionable in vivo relevance. The measurable catalytic-cysteine-dependent activity is toward phosphoinositides, which is captured by the separate phosphatidylinositol phosphate phosphatase activity term. The InterPro/EC mapping is therefore demonstrably wrong for this protein. Supporting Evidence: PMID:26620550 PTPRN2 belongs to the protein tyrosine phosphatase family, but does not exhibit activity against phosphoprotein substrates due to several critical amino acid variations in the PTP domain (Magistrelli et al, 1996). PMID:20097759 By sequence, phogrin is a member of the family of receptor-like protein-tyrosine phosphatases, but it contains substitutions in conserved catalytic sequences, and no significant enzymatic activity for phogrin has ever been reported. file:human/PTPRN2/PTPRN2-uniprot.txt CC -!- CAUTION: Has no tyrosine-protein phosphatase activity at mild acidic |
| GO:0005001 transmembrane receptor protein tyrosine phosphatase activity | TAS PMID:8798755 Cloning and characterization of islet cell antigen-related p... | REMOVE | Summary: Author statement from the cloning paper that recombinant IAR has phosphatase activity, extended to receptor PTP activity. Reason: The 1996 cloning abstract says only that recombinant IAR "has phosphatase activity"; UniProt records that this was weak activity at pH 4.5, absent at pH 5.5, and of questioned in vivo relevance. Subsequent direct assays found no significant activity on phosphoprotein or pNPP substrates, and later human work states PTPRN2 is inactive against phosphoproteins. Beyond lacking protein-tyrosine phosphatase activity, there is no ligand or receptor signalling activity for PTPRN2; it is a granule membrane protein whose luminal domain is largely removed during maturation. Receptor protein tyrosine phosphatase activity is contradicted, not merely unsupported. This is a TAS statement, not an experimental annotation, and it conflicts with the later direct evidence. Supporting Evidence: PMID:8798755 Recombinant IAR polypeptide has phosphatase activity. PMID:26620550 PTPRN2 belongs to the protein tyrosine phosphatase family, but does not exhibit activity against phosphoprotein substrates due to several critical amino acid variations in the PTP domain (Magistrelli et al, 1996). PMID:15220191 enzymatically inactive member of the transmembrane protein tyrosine phosphatase family |
| GO:0005886 plasma membrane | TAS PMID:8798755 Cloning and characterization of islet cell antigen-related p... | KEEP AS NON CORE | Summary: PTPRN2 is transiently present at the plasma membrane after granule fusion before being endocytosed and recycled. Reason: The steady-state location is the dense-core granule (and synaptic vesicle) membrane. Plasma membrane residence is real but transient, arising from exocytosis and followed by AP-2/clathrin-mediated retrieval; in breast cancer cells the protein is also detected at the plasma membrane where it lowers PI(4,5)P2. Correct, but peripheral to the core granule location. Supporting Evidence: PMID:26620550 As a transmembrane protein, PTPRN2 shuttles between secretory vesicles and the plasma membrane. PMID:26620550 Both PTPRN2 and PLCΞ²1 also demonstrated localization to the plasma membrane in breast cancer cells, in addition to some cytoplasmic localization (Fig EV3A). |
| GO:0005886 plasma membrane | TAS Reactome:R-HSA-6800426 | KEEP AS NON CORE | Summary: Reactome places PTPRN2 at the plasma membrane as the product of granule-membrane exocytosis. Reason: Consistent with the other plasma membrane row, the protein reaches the plasma membrane after secretory granule fusion. The Reactome context (neutrophil ficolin-1-rich granules) is doubtful for this neuroendocrine protein, but the location itself is supported independently by the human cell data; it remains non-core. Supporting Evidence: PMID:26620550 As a transmembrane protein, PTPRN2 shuttles between secretory vesicles and the plasma membrane. |
| GO:0006470 protein dephosphorylation | TAS PMID:8798755 Cloning and characterization of islet cell antigen-related p... | REMOVE | Summary: Process counterpart of the inferred protein tyrosine phosphatase activity. Reason: This TAS row follows from the assumption that PTPRN2 is an active protein tyrosine phosphatase, which later direct assays contradict. No study shows PTPRN2 dephosphorylating, or being required for the dephosphorylation of, a specific phosphoprotein. The UniProt by-similarity note that it may regulate active PTPs such as PTPRA through dimerization is speculative and would at most be regulation, not participation. The demonstrated catalytic substrate class is phosphoinositides. Supporting Evidence: PMID:26620550 PTPRN2 belongs to the protein tyrosine phosphatase family, but does not exhibit activity against phosphoprotein substrates due to several critical amino acid variations in the PTP domain (Magistrelli et al, 1996). PMID:20097759 phogrin is able to dephosphorylate specific inositol phospholipids, including phosphatidylinositol (PI) 3-phosphate and PI 4,5-diphosphate but not PI 3,4,5-trisphosphate. |
| GO:0007269 neurotransmitter secretion | IEA GO_REF:0000107 | KEEP AS NON CORE | Summary: Ensembl Compara transfer from mouse Ptprn2 of a role in neurotransmitter secretion. Reason: The donor is the mouse ortholog in the same PANTHER subfamily (PTHR46106:SF5). In mice lacking both Ptprn and Ptprn2, brain norepinephrine, dopamine and serotonin are reduced, synaptosomal release of dopamine, GABA and glutamate falls, and synaptic vesicle number decreases; Ptprn2 itself co-fractionates with synaptic vesicles. This supports a role for PTPRN2 in the vesicle supply that sustains neurotransmitter release, but the decisive data come from the double knockout, and the PAINT curators placed only insulin secretion and the general regulation of secretion on the family node, not neurotransmitter secretion. The term is therefore kept as non-core. Supporting Evidence: PMID:19361477 IA-2beta, but not IA-2, co-purifies with fractions rich in synaptic vesicles (SV) PMID:19361477 the secretion of dopamine, GABA and glutamate from the synaptosomes of the DKO mice was significantly decreased as was the number of SV |
| GO:0007269 neurotransmitter secretion | ISS GO_REF:0000024 | KEEP AS NON CORE | Summary: Curator ortholog transfer from mouse Ptprn2 of a role in neurotransmitter secretion. Reason: Same claim as the Ensembl Compara row, from the same mouse SF5 ortholog. Supported by mouse double-knockout neurotransmitter and synaptosome data, but non-core for the reasons given there. Supporting Evidence: PMID:19361477 IA-2beta, but not IA-2, co-purifies with fractions rich in synaptic vesicles (SV) |
| GO:0030141 secretory granule | IBA GO_REF:0000033 | MODIFY | Summary: PAINT propagation of secretory granule localization from the IA-2 family node (PTN000439769), seeded by mouse, rat and worm members. Reason: The node placement is sound; human PTPRN2 sits in subfamily SF5 under the IA-2 family node with experimentally characterized mouse and rat orthologs, and granule localization is conserved across the family. The term is correct but less specific than the evidence. PTPRN2 is an integral membrane protein of dense-core granules specifically, so dense core granule membrane (a descendant of secretory granule via secretory granule membrane) is the appropriate term. Proposed replacements: dense core granule membrane Supporting Evidence: PMID:20097759 in which it is localized to the membrane of insulin-containing dense-core vesicles file:human/PTPRN2/PTPRN2-uniprot.txt Note=Predominantly found on dense-core secretory granules. |
| GO:0030141 secretory granule | IEA GO_REF:0000120 | MODIFY | Summary: Electronic secretory granule localization, from the mouse ortholog and the IA-2 family signature. Reason: Same claim as the IBA row; correct but general. The protein is a dense-core granule membrane protein. Proposed replacements: dense core granule membrane Supporting Evidence: PMID:20097759 in which it is localized to the membrane of insulin-containing dense-core vesicles |
| GO:0030658 transport vesicle membrane | IEA GO_REF:0000044 | ACCEPT | Summary: UniProt subcellular-location mapping of "secretory vesicle membrane" to transport vesicle membrane. Reason: A broad but correct mapping of UniProt's secretory vesicle membrane location. Transport vesicle membrane is an ancestor of the synaptic vesicle membrane term also carried by the gene, and is acceptable as an electronic location; the specific granule and synaptic vesicle membrane terms carry the precise claim. Supporting Evidence: file:human/PTPRN2/PTPRN2-uniprot.txt CC -!- SUBCELLULAR LOCATION: Cytoplasmic vesicle, secretory vesicle membrane |
| GO:0030667 secretory granule membrane | IEA GO_REF:0000107 | MODIFY | Summary: Ensembl Compara transfer of secretory granule membrane localization from mouse Ptprn2. Reason: Correct; the mouse SF5 ortholog and rat phogrin are dense-core vesicle membrane proteins in beta cells, and loss of Ptprn2 reduces dense-core vesicle number. The specific child term dense core granule membrane is supported. Proposed replacements: dense core granule membrane Supporting Evidence: PMID:21732083 Islet antigen 2 (IA-2) and IA-2Ξ² are dense core vesicle (DCV) |
| GO:0030667 secretory granule membrane | ISS GO_REF:0000024 | MODIFY | Summary: Curator ortholog transfer of secretory granule membrane localization from mouse Ptprn2. Reason: Same claim as the Ensembl Compara row; refine to dense core granule membrane. Proposed replacements: dense core granule membrane Supporting Evidence: PMID:20097759 in which it is localized to the membrane of insulin-containing dense-core vesicles |
| GO:0030672 synaptic vesicle membrane | IEA GO_REF:0000120 | ACCEPT | Summary: Electronic synaptic vesicle membrane localization from the mouse ortholog. Reason: In mouse forebrain, IA-2beta (Ptprn2) co-fractionates with synaptic-vesicle-enriched fractions and is protected from surface biotinylation of synaptosomes, unlike its paralog IA-2 (Ptprn). The donor is the SF5 ortholog, so this location transfers to human PTPRN2. Supporting Evidence: PMID:19361477 IA-2beta, but not IA-2, co-purifies with fractions rich in synaptic vesicles (SV) |
| GO:0030672 synaptic vesicle membrane | ISS GO_REF:0000024 | ACCEPT | Summary: Curator ortholog transfer of synaptic vesicle membrane localization from mouse Ptprn2. Reason: Same claim as the electronic row, supported by the mouse synaptic vesicle fractionation data. Supporting Evidence: PMID:19361477 IA-2beta, but not IA-2, co-purifies with fractions rich in synaptic vesicles (SV) |
| GO:0035773 insulin secretion involved in cellular response to glucose stimulus | IBA GO_REF:0000033 | ACCEPT | Summary: PAINT propagation from the vertebrate IA-2/IA-2beta node (PTN000439772), seeded by mouse Ptprn and Ptprn2. Reason: The node covers the vertebrate PTPRN/PTPRN2 clade, and the curators restricted this term to it rather than the deeper family node used for granule location and regulation of secretion, a sensible placement since insulin is a vertebrate beta-cell product. Mouse Ptprn2 knockout alone impairs glucose-stimulated insulin secretion and reduces dense-core vesicle number and insulin content, and in rat phogrin the catalytic-cysteine-dependent PIPase activity correlates with effects on glucose-stimulated insulin secretion. PTPRN2 contributes as a dense-core granule membrane component that maintains granule number and membrane phosphoinositides, which is participation in the secretory process rather than an indirect phenotype. Supporting Evidence: PMID:15220191 The IA-2beta(-/-) mice, in contrast to wild-type controls, showed mild glucose intolerance and impaired glucose-stimulated insulin secretion. PMID:21732083 Islets from single and DKO mice all showed a significant decrease in insulin content, insulin secretion and the number and half-life of DCV PMID:20097759 The PIPase activity of phogrin depended on the catalytic site cysteine and correlated with effects on glucose-stimulated insulin secretion. |
| GO:0035773 insulin secretion involved in cellular response to glucose stimulus | IEA GO_REF:0000107 | ACCEPT | Summary: Ensembl Compara transfer from mouse Ptprn2. Reason: Same claim as the IBA row, from the SF5 mouse ortholog with knockout evidence. Supporting Evidence: PMID:15220191 The IA-2beta(-/-) mice, in contrast to wild-type controls, showed mild glucose intolerance and impaired glucose-stimulated insulin secretion. |
| GO:0035773 insulin secretion involved in cellular response to glucose stimulus | ISS GO_REF:0000024 | ACCEPT | Summary: Curator ortholog transfer from mouse Ptprn2. Reason: Same claim as the IBA row, from the SF5 mouse ortholog with knockout evidence. Supporting Evidence: PMID:15220191 The IA-2beta(-/-) mice, in contrast to wild-type controls, showed mild glucose intolerance and impaired glucose-stimulated insulin secretion. |
| GO:0043235 signaling receptor complex | IDA PMID:23382219 Structural basis for endosomal trafficking of diverse transm... | UNDECIDED | Summary: Direct-assay annotation from a structural study of SNX17/SNX27/SNX31 PX-FERM binding to NPxY sorting motifs of diverse transmembrane cargos. Reason: The study surveyed binding of PX-FERM sorting nexins to sorting motifs of many putative cargos, and PTPRN2 carries a tyrosine-based internalization motif, so it was plausibly one of the tested cargo peptides. The cached record does not mention PTPRN2, and neither the cached text nor any other evidence identifies a signalling receptor complex containing PTPRN2. A cargo-motif binding assay would more naturally support a sorting-nexin binding function than complex membership. Without the relevant figure or table the experimental basis cannot be checked, so the annotation is neither accepted nor removed. |
| GO:0045202 synapse | IBA GO_REF:0000033 | MODIFY | Summary: PAINT propagation of synapse localization from the IA-2 family node (PTN000439769). Reason: The placement is sound (mouse and rat Ptprn2 and worm IDA-1 are neuronal vesicle proteins), but synapse is a general term. For PTPRN2 the supported location within the synapse is the synaptic vesicle membrane, which the gene already carries and which is part of the presynapse. Proposed replacements: synaptic vesicle membrane Supporting Evidence: PMID:19361477 IA-2beta, but not IA-2, co-purifies with fractions rich in synaptic vesicles (SV) |
| GO:0051046 regulation of secretion | IBA GO_REF:0000033 | ACCEPT | Summary: PAINT propagation from the IA-2 family node (PTN000439769), seeded by fly and worm IA-2 orthologs. Reason: Regulation of dense-core-vesicle-dependent secretion is the conserved role across the IA-2 family, from invertebrate IDA-1/IA-2 to mammalian PTPRN and PTPRN2, and human PTPRN2 sits within that clade. In vertebrate beta cells, PTPRN2 knockout reduces dense-core vesicle number and insulin secretion, and PTPRN2 overexpression inhibits glucose-stimulated insulin secretion in insulinoma cells. This general term is as precise as the cross-family evidence allows. Supporting Evidence: PMID:21732083 Islets from single and DKO mice all showed a significant decrease in insulin content, insulin secretion and the number and half-life of DCV PMID:16418280 Administration of ghrelin or overexpression of IA-2beta, but not overexpression of IA-2, inhibited glucose-stimulated insulin secretion in MIN6 insulinoma cells |
| GO:0101003 ficolin-1-rich granule membrane | TAS Reactome:R-HSA-6800426 | MARK AS OVER ANNOTATED | Summary: Reactome lists PTPRN2 among neutrophil ficolin-1-rich granule membrane proteins. Reason: The Reactome neutrophil degranulation event lists ficolin-1-rich granule membrane proteins drawn from high-throughput neutrophil granule proteomics. PTPRN2 is a neuroendocrine dense-core granule protein expressed mainly in brain and pancreas, and there is no targeted evidence of PTPRN2 in neutrophils or their granules. The assignment may reflect proteomic detection rather than a functional granule component, and it does not describe the biology of the protein. Supporting Evidence: file:human/PTPRN2/PTPRN2-uniprot.txt CC -!- TISSUE SPECIFICITY: Highest levels in brain and pancreas |
| GO:0052866 phosphatidylinositol phosphate phosphatase activity | ISS PMID:20097759 The neurosecretory vesicle protein phogrin functions as a ph... | NEW | Summary: PTPRN2 is a transmembrane phosphatidylinositol phosphate phosphatase; its catalytic-cysteine-dependent activity is on phosphoinositides, not phosphoproteins. Reason: Purified transmembrane rat phogrin, the PTHR46106:SF5 ortholog of human PTPRN2, dephosphorylates PI(3)P and PI(4,5)P2 (but not PI(3,4,5)P3) with a specific activity comparable to other PIPases, and this activity is lost on mutation of the catalytic cysteine. Human PTPRN2 knockdown and overexpression raise and lower plasma membrane PI(4,5)P2 respectively, and its pro-migratory effect requires the catalytic cysteine (C945A). The general phosphatidylinositol phosphate phosphatase term is used because the phosphate position removed from PI(4,5)P2 was not determined. Supporting Evidence: PMID:20097759 phogrin is able to dephosphorylate specific inositol phospholipids, including phosphatidylinositol (PI) 3-phosphate and PI 4,5-diphosphate but not PI 3,4,5-trisphosphate. PMID:20097759 The PIPase activity of phogrin depended on the catalytic site cysteine and correlated with effects on glucose-stimulated insulin secretion. PMID:26620550 Both PTPRN2 and PLCΞ²1 are enzymes with previously demonstrated activities for the substrate PI(4,5)P2 (Rebecchi & Pentyala, 2000; Caromile et al, 2010). PMID:26620550 We found that while wildβtype PTPRN2 overexpression was sufficient to increase invasion, migration, and metastatic lung colonization in mice, these effects were dependent on PTPRN2's catalytic activity as equivalent overexpression of PTPRN2C945A failed to enhance these phenotypes (Fig 2DβF). |
Loading supporting contentβ¦
Download this section (compressed HTML)Q: Which phosphate (4- or 5-position) does PTPRN2 remove from PI(4,5)P2, and does the dense-core granule membrane PI(4,5)P2 pool, rather than the plasma membrane pool, mediate its effect on insulin secretion?
Q: Does PTPRN2 bind the FERM domain of SNX17 or SNX27 through its tyrosine-based internalization motif, and does that binding contribute to its recycling back to dense-core granules?
Experiment: Assay purified full-length human PTPRN2 and its C945S mutant against PI(4,5)P2 with product identification by mass spectrometry to define positional specificity, and rescue Ptprn2-null beta cells with wild-type versus catalytic-dead protein to separate the lipid-phosphatase function from the structural granule role.
Loading supporting contentβ¦
Download this section (compressed HTML)Loading supporting contentβ¦
Download this section (compressed HTML)Loading supporting contentβ¦
Download this section (compressed HTML)Loading supporting contentβ¦
Download this section (compressed HTML)