CSGALNACT1

Summary: CSGALNACT1 is a type II Golgi membrane protein, so the term is correct. The cited paper, however, describes chondroitin synthase (CHSY1), not CSGALNACT1.

Reason: The Golgi membrane location is right for CSGALNACT1. It is a single-pass type II membrane protein (PMID:11788602), Reactome places it in the Golgi membrane, and UniProt curates the Golgi stack membrane. We accept the term on that evidence. The reference is wrong: PMID:11514575 cloned the 802-aa chromosome-15 chondroitin synthase (CHSY1, Q86X52), not the 532-aa chromosome-8 CSGALNACT1. The defect is recorded below.

Summary: Reactome places the CSGALNACT-catalysed linker GalNAc transfer at the Golgi membrane.

Reason: This is the expected site of chondroitin chain initiation by a type II Golgi glycosyltransferase. It agrees with the UniProt curated location and the transmembrane topology.

Summary: Reactome places CSGALNACT-catalysed chain GalNAc transfer at the Golgi membrane.

Reason: This agrees with the curated Golgi stack membrane location and type II topology.

Summary: Phylogenetic inference of Golgi localization for the chondroitin synthase family.

Reason: This is correct and consistent with the Golgi membrane location. The broader Golgi apparatus term from the family-level IBA is acceptable. The more specific Golgi membrane term is supported separately.

Summary: Electronic Golgi apparatus annotation.

Reason: Correct, though broader than Golgi membrane. Consistent with the IBA row and the curated location.

Summary: CSGALNACT1 is a GalNAc transferase, but this general term should be refined. The cited paper is about chondroitin synthase (CHSY1), not CSGALNACT1.

Reason: The claim is true of CSGALNACT1 on independent evidence (PMID:11788602, PMID:12163485). It is less specific than the evidence supports. The enzyme transfers beta1,4-GalNAc to GlcA on the linkage tetrasaccharide (GO:0047237, GalNAcT-I) and on growing chondroitin chains (GO:0047238, GalNAcT-II). The gene already carries both terms with experimental support. The cited PMID:11514575 describes the 802-aa chromosome-15 chondroitin synthase (CHSY1). The GalNAc transfer measured there is CHSY1's activity, so a reference defect is recorded.

Proposed replacements: glucuronylgalactosylproteoglycan 4-beta-N-acetylgalactosaminyltransferase activity glucuronosyl-N-acetylgalactosaminyl-proteoglycan 4-beta-N-acetylgalactosaminyltransferase activity

Summary: ARBA-based general GalNAc-transferase annotation.

Reason: Correct but general. The specific activities, GalNAcT-I (GO:0047237) and GalNAcT-II (GO:0047238), are experimentally supported and already annotated.

Proposed replacements: glucuronylgalactosylproteoglycan 4-beta-N-acetylgalactosaminyltransferase activity glucuronosyl-N-acetylgalactosaminyl-proteoglycan 4-beta-N-acetylgalactosaminyltransferase activity

Summary: General GalNAc-transferase annotation citing the paper that cloned CSGALNACT1 itself.

Reason: PMID:11788602 directly expressed and assayed CSGALNACT1 (532 aa, chromosome 8). It showed GalNAc transfer both to polymer chondroitin (GalNAcT-II) and to a linkage-region acceptor (GalNAcT-I). The general term should be refined to those two specific activities. The row is an ISS citing CSGALNACT2 (Q8N6G5) as WITH, although the paper is direct evidence for Q8TDX6. That is recorded as an evidence-code query.

Proposed replacements: glucuronylgalactosylproteoglycan 4-beta-N-acetylgalactosaminyltransferase activity glucuronosyl-N-acetylgalactosaminyl-proteoglycan 4-beta-N-acetylgalactosaminyltransferase activity

Summary: Wrong term and wrong gene. GO:0008955 is the bacterial lipid II (GlcNAc-MurNAc pentapeptide) cell-wall polymerase reaction. The cited paper characterised human chondroitin synthase (CHSY1), not CSGALNACT1.

Reason: GO:0008955 is defined as polymerisation of GlcNAc-(1->4)-Mur2Ac-pentapeptide-undecaprenyl diphosphate (lipid II) into peptidoglycan (term checked in OLS; not obsolete). This bacterial cell-wall reaction is carried out by GT51 transglycosylases and SEDS proteins (rule FRR:0001). CSGALNACT1 is a Golgi GT7 enzyme that transfers GalNAc from UDP-GalNAc to glucuronic acid in chondroitin. It has no peptidoglycan substrate, and vertebrates make no peptidoglycan. The 2003 curator most likely confused "peptidoglycan" with the proteoglycan/chondroitin polymer acceptor; the right terms for the measured reactions are GO:0047238 (GalNAc) and GO:0050510 (GlcA). The paper also did not assay CSGALNACT1. It cloned the 802-aa chromosome-15 chondroitin synthase (CHSY1, Q86X52), and that paper is correctly cited for CHSY1 under GO:0047238 and GO:0050510. CSGALNACT1 (532 aa, chromosome 8) was cloned a year later by homology to that synthase. The REMOVE rests on the term: peptidoglycan glycosyltransferase activity is false for any vertebrate Golgi GT7 enzyme. The wrong-gene citation is recorded separately as a source defect, as for the other rows from this paper. CSGALNACT1's real GalNAc-transferase activities are already captured with valid evidence by the GO:0047237 and GO:0047238 rows (IDA PMID:12163485, EXP PMID:21160489/PMID:27599773, IBA, TAS).

Summary: CSGALNACT1 does not synthesise heparan sulfate. It diverts linkage-region acceptors away from heparan sulfate toward chondroitin sulfate. The cited paper is about CHSY1.

Reason: PMID:11514575 only draws an analogy between chondroitin synthase and the heparan sulfate polymerases. It makes no claim that either enzyme makes heparan sulfate, and it is about CHSY1 anyway. CSGALNACT1 adds GalNAc, a residue absent from heparan sulfate. Overexpressing it replaces heparan sulfate with chondroitin sulfate on a hybrid core protein. In patient cells, the heparan sulfate increase after its loss comes from unused linker acceptor, not from any heparan sulfate synthetic step performed by CSGALNACT1. The gene does none of the work of heparan sulfate synthesis.

Summary: Glucuronosyltransferase activity belongs to chondroitin synthase (CHSY1), the protein studied in the cited paper, not to CSGALNACT1.

Reason: PMID:11514575 showed that the 802-aa chondroitin synthase (CHSY1) has both beta1,3-GlcUA-transferase and beta1,4-GalNAc-transferase activities. CSGALNACT1 is a separate 532-aa protein with only GalNAc-transferase activity. Every biochemical study of CSGALNACT1 describes it as transferring GalNAc (PMID:11788602, PMID:12163485, PMID:12446672). UniProt curates only EC 2.4.1.174 for it. There is no evidence that CSGALNACT1 transfers GlcA. The activity is CHSY1's (GO:0050510 on Q86X52).

Summary: Using UDP-GalNAc as a sugar donor is not a UDP-GalNAc metabolic process. The cited paper is about CHSY1.

Reason: The row records only that the enzyme studied in PMID:11514575 used UDP-GalNAc as a donor substrate. Donor consumption by a glycosyltransferase is captured by its molecular function, not by a nucleotide-sugar metabolic process. The enzyme studied was CHSY1, not CSGALNACT1. GOA labels this term obsolete; we do not rely on that, because the removal stands on the biology alone.

Summary: General proteoglycan biosynthesis. Correct but should be refined to chondroitin sulfate proteoglycan biosynthesis.

Reason: CSGALNACT1 initiates and elongates chondroitin/dermatan sulfate chains on proteoglycan core proteins. GO:0050650 is the precise, experimentally supported descendant, and the gene already carries it. See the ISS evidence-code note on the GO:0008376 ISS row from the same paper.

Proposed replacements: chondroitin sulfate proteoglycan biosynthetic process

Summary: CSGALNACT1 does not make heparin. The cited paper is about CHSY1.

Reason: Heparin chains contain GlcNAc and hexuronic acid, not GalNAc. CSGALNACT1's GalNAc addition to the linker commits the site to chondroitin sulfate and away from heparin/heparan sulfate (PMID:12446672). PMID:11514575 is about CHSY1 and mentions heparan sulfate polymerases only as an analogy.

Summary: Electronic mapping from the UniProt "Golgi stack membrane" subcellular location.

Reason: This is consistent with the curated UniProt location (Golgi stack membrane) and the type II signal-anchor topology. It is a reasonable refinement of Golgi membrane.

Summary: Wrong gene and inappropriate term. CSGALNACT1 does not use UDP-GlcA. The CHSY1 enzyme in the cited paper uses it only as a donor.

Reason: PMID:11514575 measured GlcUA transfer from UDP-GlcUA by chondroitin synthase (CHSY1). Donor use is not a UDP-glucuronate metabolic process, and CSGALNACT1 has no glucuronosyltransferase activity.

Summary: Divalent metal binding expected for a GT7 UDP-sugar glycosyltransferase. Not directly shown in the cited abstract.

Reason: UniProt predicts two divalent-metal binding residues (360 and 477) by sequence analysis, as expected for a GT7 enzyme. The cited abstract does not mention a metal requirement, and no CSGALNACT1-specific metal data are in the cached sources. The general term is as precise as the available evidence allows: the identity of the metal (likely Mn2+) is unverified here. The activity is a cofactor property, not the gene's core function.

Summary: GalNAcT-I (chain-initiation) activity. Disease-associated missense mutants lose activity.

Reason: GalNAcT-I, the transfer of the first GalNAc to the linkage-region GlcA, is the defining core activity of CSGALNACT1. The cached abstract reports loss of GalNAcT-II activity in the H234R and M509R mutants. The UniProt curators cite this paper for the linker-acceptor reaction (RHEA:23464), presumably from full-text data. We defer to them, since the GalNAcT-I activity is independently established (PMID:11788602, PMID:12163485).

Summary: GalNAcT-I activity of CSGALNACT1, reduced by the patient P384R variant.

Reason: CSGALNACT1 initiates CS chains on the linkage tetrasaccharide. The P384R skeletal-dysplasia variant reduces this GalNAc-transferase activity.

Summary: Recombinant CSGALNACT1 transfers GalNAc to the GlcA of the linkage tetrasaccharide.

Reason: Direct assay with the linkage tetrasaccharide acceptor, which was one of the best acceptors. Gotoh et al. also built GalNAc-GlcA-Gal-Gal-Xyl-peptide in vitro with this enzyme.

Summary: Electronic mapping from RHEA:23464 / EC 2.4.1.174.

Reason: The mapping matches the experimentally supported GalNAcT-I activity.

Summary: Phylogenetic inference of GalNAcT-II (chain elongation) activity across the chondroitin synthase family.

Reason: CSGALNACT1 transfers GalNAc to the GlcA of growing chondroitin chains, as shown directly in PMID:11788602 and PMID:12163485. Q8TDX6 appearing among the IBA sources reflects its own experimental grounding. Its elongation activity is weaker than CSGALNACT2's (PMID:12446672) but is real, so the term is accepted.

Summary: Recombinant CSGALNACT1 transfers GalNAc to chondroitin poly- and oligosaccharides.

Reason: Direct assay of elongation activity.

Summary: Ensembl Compara transfer from mouse Csgalnact1.

Reason: Consistent with direct human evidence (PMID:11788602, PMID:12163485).

Summary: This Reactome reaction is the linker (GalNAcT-I) step. GalNAc goes onto the GlcA-Gal-Gal-Xyl tetrasaccharide, which GO:0047237 describes more exactly.

Reason: The GalNAcT-II term is true of CSGALNACT1 on direct evidence, and rows sharing a term must agree, so the term is accepted. The reaction this Reactome entry describes (GalNAc onto the linker GlcA) is GO:0047237, which the gene already carries. This is noted as a suggestion for Reactome-to-GO mapping rather than an error in the claim.

Summary: Reactome chain-elongation step, GalNAc onto the terminal GlcA of the CS chain.

Reason: Matches the GalNAcT-II activity shown directly for CSGALNACT1.

Summary: Phylogenetic inference of CS proteoglycan biosynthesis.

Reason: Core process. CSGALNACT1 catalyses the chain-initiating and elongating GalNAc transfers of CS chains.

Summary: Core process for CSGALNACT1. The cited paper, however, is about chondroitin synthase (CHSY1).

Reason: The process is correct for CSGALNACT1 on independent evidence (PMID:11788602, PMID:12163485, PMID:12446672, PMID:17145758). The cited PMID:11514575 describes the 802-aa chromosome-15 chondroitin synthase (CHSY1) and did not assay CSGALNACT1. A reference defect is recorded.

Summary: In vitro synthesis of the first chondroitin pentasaccharide on a xylose-peptide with CSGALNACT1 and the linker enzymes.

Reason: Direct demonstration of the CS-initiating step on a peptide acceptor.

Summary: Electronic annotation to CS proteoglycan biosynthesis.

Reason: Consistent with experimental evidence.

Summary: CS proteoglycan biosynthesis, citing the CSGALNACT1 cloning paper.

Reason: The process is correct. The paper is direct evidence for Q8TDX6 (see the evidence-code note on the GO:0008376 ISS row).

Summary: Reactome CS-GAG biosynthesis pathway.

Reason: CSGALNACT1 catalyses the GalNAc transfer steps of this pathway.

Summary: Dermatan sulfate chains are built on the same GalNAc-GlcA backbone that CSGALNACT1 initiates.

Reason: Dermatan sulfate arises by epimerisation of chondroitin backbones. CSGALNACT1 catalyses the chain-initiating GalNAc transfer for both CS and DS chains. Patient fibroblasts show fewer CS/DS chains.