B3GALNT2 is a glycosyltransferase (EC 2.4.1.313) whose canonical protein has a single-pass type-II membrane topology. It transfers N-acetylgalactosamine from UDP-GalNAc to terminal beta-linked N-acetylglucosamine, forming a beta-1,3 linkage. Its established physiological role is the assembly of the core M3 O-mannosyl glycan of alpha-dystroglycan in the endoplasmic reticulum. It extends the product of POMGNT2, producing the trisaccharide recognized by the mannose kinase POMK and enabling later glycan elaboration required for extracellular-matrix binding. Cultured-cell studies also support modification of N-glycans on intracellular glycoproteins, although the endogenous substrate range and physiological significance remain incompletely resolved. Biallelic loss of B3GALNT2 function causes dystroglycanopathy with muscle and, in severe cases, brain and eye abnormalities.
Definition: Catalysis of transfer of an N-acetylgalactosaminyl residue from UDP-GalNAc to the 3-position of the terminal beta-1,4-linked GlcNAc of a protein O-mannosyl glycan, forming GalNAc-beta1,3-GlcNAc-beta1,4-Man-O-protein.
Justification: Candidate for ontology-curator assessment of the reaction established in PMID:14724282 and PMID:23929950 and assigned EC2.4.1.313/RHEA37667. GO:0008376 covers broader oligosaccharide GalNAc transfer; the core-M3 acceptor and beta1,3 product linkage provide a chemically specific refinement. The complete returned AmiGO immediate-child list and indexed searches did not identify this reaction, but the displayed ontology load date was 2026-08-06 and direct OLS/QuickGO descendant access was unavailable. These bounded checks do not establish exhaustive absence from the current GO release. Confirm whether an equivalent term exists before creating one; no GO identifier is asserted, no NEW annotation is added, and the existing activity remains GO:0008376 pending that check.
Parent term: acetylgalactosaminyltransferase activity
Mappings:
| GO Term | Evidence | Action | Reason |
|---|---|---|---|
| GO:0000139 Golgi membrane | IBA GO_REF:0000033 | KEEP AS NON CORE | Summary: A curated ancestral inference supports Golgi membrane localization; ER is the directly supported core compartment. Reason: The positive PAINT assertion at PTN000130249 supports inherited Golgi membrane localization, and curated UniProt independently records Golgi membrane by similarity. ER localization is directly demonstrated for the human protein and is used in the core summary. The partial-Golgi statement in PMID:30898876 cites PMID:23453667; a complete PAINT tree/alignment and an independent target Golgi-imaging experiment were not inspected. |
| GO:0006493 protein O-linked glycosylation | IBA GO_REF:0000033 | MODIFY | Summary: B3GALNT2 directly extends the protein O-mannosyl glycan in alpha-dystroglycan core M3 synthesis. Reason: The O-glycan assertion is correct; specify protein O-linked glycosylation via mannose because B3GALNT2 directly extends the POMGNT2-generated glycan in core M3. This is a target-specific refinement supported by 23929950, not a claim that the original IBD was wrong. PTN002257389 is legitimately grounded by Q8NCR0 itself. Proposed replacements: protein O-linked glycosylation via mannose |
| GO:0008194 UDP-glycosyltransferase activity | IBA GO_REF:0000033 | MODIFY | Summary: The experimentally defined UDP-sugar reaction is acetylgalactosaminyl transfer. Reason: Specify the transferred GalNAc residue rather than broad UDP-glycosyltransferase activity. Positive PTN002257389 inheritance and target self-grounding are valid; direct human enzyme assays establish the more specific activity. Proposed replacements: acetylgalactosaminyltransferase activity |
| GO:0000139 Golgi membrane | IEA GO_REF:0000044 | KEEP AS NON CORE | Summary: The UniProt subcellular-location mapping records Golgi membrane as a secondary location. Reason: The UniProt SL-0134 mapping carries a positive curated Golgi membrane assignment supported by similarity (ECO:0000250). Retain that location as non-core alongside the experimentally supported ER compartment. PMID:30898876 cites the earlier localization study and is not an independent imaging measurement. |
| GO:0005783 endoplasmic reticulum | IEA GO_REF:0000044 | ACCEPT | Summary: ER localization is consistent with the primary wild-type localization experiment. Reason: ER localization agrees with wild-type human construct colocalization with ERp72 in C2C12 cells. The broad organelle term remains correct for the core location; the core summary can use only the more precise membrane term. |
| GO:0009101 glycoprotein biosynthetic process | IEA GO_REF:0000120 | ACCEPT | Summary: B3GALNT2 participates directly in glycoprotein biosynthesis with established O-glycan and qualified N-glycan substrate scope. Reason: The enzyme catalyzes glycoprotein glycan synthesis. Retain this broad electronic assertion because its biology includes established O-mannosyl-glycan extension and qualified cultured-cell N-glycan modification; replacing its entire scope by the O-mannose branch would discard supported substrate breadth. WITH/FROM contains both ARBA00027560 and IPR002659. No new N-glycan process term is needed. Supporting Evidence: PMID:30898876 identified the glycoproteins that carry B3GALNT2-generated type-I LDN in their N-glycans. |
| GO:0016020 membrane | IEA GO_REF:0000002 | MODIFY | Summary: The membrane annotation is correct; ER membrane is the supported specific location. Reason: General membrane localization is sound but less informative than ER membrane, supported by the human ER enzyme model and primary localization. Generality alone is not overannotation; retain the original electronic source fields and cite the independent specific evidence. Proposed replacements: endoplasmic reticulum membrane |
| GO:0016758 hexosyltransferase activity | IEA GO_REF:0000002 | MODIFY | Summary: The broad hexosyltransferase annotation can be refined to acetylgalactosaminyltransferase activity. Reason: Hexosyltransferase is a valid ancestor but the transferred residue is experimentally identified as GalNAc. Refine to acetylgalactosaminyltransferase; do not infer galactose, GlcNAc-transfer or kinase activity from family names. Proposed replacements: acetylgalactosaminyltransferase activity |
| GO:0005515 protein binding | IPI PMID:32296183 A reference map of the human binary protein interactome. | UNDECIDED | Summary: TMBIM1 is the recorded interaction partner, but the exact target experiment and construct remain unverified in this audit. Reason: The original IPI assertion names TMBIM1/Q969X1 and is corroborated as a curated pair by immutable UniProt. General HuRI methods were read, but the exact primary target row/construct and pair-specific validation could not be inspected: IntAct direct/API access failed. This is an evidence-access limit, not an unknown partner, disproven interaction, or generic-binding policy removal. No more informative molecular role is established by the accessed material. |
| GO:0006493 protein O-linked glycosylation | TAS Reactome:R-HSA-8932505 | MODIFY | Summary: The cited Reactome pathway identifies the protein O-mannosyl core M3 branch. Reason: The cited Reactome pathway specifically concerns core M3 O-mannosyl glycans. B3GALNT2 performs GalNAc addition within that pathway, supporting the more precise O-mannose-linked process rather than only its parent. Proposed replacements: protein O-linked glycosylation via mannose Supporting Evidence: Reactome:R-HSA-8932505 Core M3 glycans are initiated by beta-1,4-linked GlcNAc extension of O-mannose |
| GO:0008376 acetylgalactosaminyltransferase activity | TAS Reactome:R-HSA-8931648 | ACCEPT | Summary: The human Reactome reaction assigns GalNAc transfer to B3GALNT2. Reason: The cited human Reactome reaction explicitly assigns UDP-GalNAc transfer to B3GALNT2. Direct human enzyme experiments corroborate the molecular activity. |
| GO:0005789 endoplasmic reticulum membrane | TAS Reactome:R-HSA-8931648 | ACCEPT | Summary: The human Reactome catalyst is associated with ER membrane. Reason: The exact cited Reactome catalyst is associated with ER membrane, consistent with type-II topology and the primary ER localization experiment. |
| GO:0008376 acetylgalactosaminyltransferase activity | IDA PMID:23929950 SGK196 is a glycosylation-specific O-mannose kinase required... | ACCEPT | Summary: A soluble human B3GALNT2 construct directly transfers GalNAc to the POMGNT2-generated glycopeptide. Reason: The cited main experimental narrative directly tests soluble human B3GALNT2dTM with UDP-GalNAc and the GlcNAc-Man glycopeptide; mass spectrometry identifies GalNAc addition. Supplementary Methods were not inspected; the engineered deletion construct is not automatically either natural alternative product. Supporting Evidence: PMID:23929950 B3GALNT2 could transfer a GalNAc residue |
| GO:0006493 protein O-linked glycosylation | IDA PMID:23929950 SGK196 is a glycosylation-specific O-mannose kinase required... | MODIFY | Summary: The directly assayed glycopeptide belongs to the protein O-mannosyl glycosylation pathway. Reason: The same cited enzyme experiment directly extends an O-mannosyl glycopeptide, making the mannose-linked branch a supported refinement. This includes glycan elongation rather than only initial mannose transfer. Proposed replacements: protein O-linked glycosylation via mannose |
| GO:0005783 endoplasmic reticulum | IDA PMID:23453667 Mutations in B3GALNT2 cause congenital muscular dystrophy an... | ACCEPT | Summary: Wild-type human B3GALNT2 colocalizes with an ER marker in transfected C2C12 myoblasts. Reason: Wild-type human B3GALNT2-V5 colocalizes with ERp72 in transfected mouse C2C12 myoblasts in the indexed original Methods/Results/Figure 4 caption. This distinguishes human construct identity from host species and wild-type location from mutant redistribution. Supporting Evidence: PMID:23453667 B3GALNT2 localized to the endoplasmic reticulum |
| GO:0009101 glycoprotein biosynthetic process | IMP PMID:23453667 Mutations in B3GALNT2 cause congenital muscular dystrophy an... | MODIFY | Summary: Patient and zebrafish perturbation evidence supports the alpha-dystroglycan glycosylation pathway. Reason: The cited patient muscle/fibroblast and zebrafish experiments concern defective functional alpha-dystroglycan glycosylation. Specify the O-mannosyl pathway with the independently demonstrated core-M3 reaction from 23929950 as additional support; do not describe the 23453667 perturbation study as a purified enzyme assay or use it as evidence for all N-glycan substrates. Proposed replacements: protein O-linked glycosylation via mannose |
Loading supporting contentβ¦
Download this section (compressed HTML)Q: Which endogenous N-glycan substrates carry a chemically confirmed beta-1,3 GalNAc linkage attributable to B3GALNT2 under unperturbed conditions?
Q: How do the extensive sequence changes of Q8NCR0-2 affect catalytic activity, localization and protein stability?
Q: Which exact B3GALNT2 and TMBIM1 constructs support the HuRI interaction, and can the interaction be reproduced under conditions that preserve their native membrane topology?
Experiment: Compare endogenous B3GALNT2 loss and matched rescue in cells or tissues using linkage-resolved glycopeptide analysis, distinguishing type-I LacdiNAc from type-II LacdiNAc and measuring both alpha-dystroglycan and intracellular N-glycan substrates.
Experiment: Compare Q8NCR0-1 and Q8NCR0-2 at matched expression levels for stability, subcellular localization and GalNAc-transfer activity on defined glycopeptide and N-glycan acceptors, with catalytic controls.
Experiment: Extend the already demonstrated POMGNT2-B3GALNT2-POMK reaction sequence with quantitative substrate and kinetic comparisons, rather than treating its basic reconstitution as an unperformed experiment.
Loading supporting contentβ¦
Download this section (compressed HTML)Loading supporting contentβ¦
Download this section (compressed HTML)Loading supporting contentβ¦
Download this section (compressed HTML)