Cellulose synthase-like D (CSLD) glycosyltransferase of peanut (Arachis hypogaea; ORF Ahy_B05g078721). It belongs to the CESA/CSL superfamily of family 2 glycosyltransferases. The 1139-residue product is full length and has the CESA-type architecture: an N-terminal RING-type zinc-finger region, two N-terminal transmembrane helices, a large cytosolic glycosyltransferase domain, and five C-terminal transmembrane helices. The catalytic aspartates, the nucleotide-sugar binding residues and the QxxRW motif of processive beta-glycan synthases are all conserved. In a neighbour-joining tree the protein falls in the angiosperm CSLD2/CSLD3 clade, and its closest relatives are Arabidopsis CSLD3 (KOJAK) and CSLD2 (about 85% identity). Arabidopsis CSLD3 is a UDP-glucose-dependent beta-1,4-glucan synthase. It accumulates at the apical plasma membrane of tip-growing root hairs and is needed to build a cellulose-like wall there. Earlier mutant and microsomal studies linked CSLD2, CSLD3 and CSLD5 to mannan synthesis, but this assignment is not supported by the later work on purified protein. By orthology, the peanut protein is expected to synthesise a beta-1,4-glucan wall polymer from UDP-glucose. It probably passes through the ER and Golgi on its way to the plasma membrane, where it acts during polarised wall deposition. There is no direct experimental study of this peanut protein.
Summary: Golgi membrane comes from a UniProt subcellular-location rule. Arabidopsis CSLD3 has been found in the Golgi by organelle proteomics, and CSLD proteins are multi-pass membrane proteins that move through the secretory pathway. The polarised plasma-membrane domain in tip-growing cells is where CSLD3 activity has been shown, so the Golgi is a real location but not the main site of function.
Reason: The localisation is plausible for an ortholog of AtCSLD3, which LOPIT proteomics placed in the Golgi. The previous review accepted this term as core. That is changed here, because the evidence on where CSLD3 acts points to the apical plasma membrane of root hairs (PMID:21765420), not the Golgi.
Summary: Cytoplasm is true by GO part_of inheritance, because Golgi membrane is part of the cytoplasm, but it says nothing useful about a multi-pass membrane glycosyltransferase.
Reason: This is an ARBA-derived term. The Golgi membrane annotation in this review already entails it. It is not wrong, so it is not removed.
Summary: In Arabidopsis, CSLD2 and CSLD3 expression rises at low temperature, and growth at reduced temperature partly rescues the single-mutant root-hair phenotypes. That is temperature-dependent expression and phenotype, not evidence that the glycosyltransferase takes part in a cold-response pathway.
Reason: This is an ARBA-derived term. Its likely basis is the Arabidopsis expression data (PMID:18768911, which TAIR curated to AtCSLD3 as acts_upstream_of_or_within). Expression induction and conditional rescue of a phenotype show that the gene's activity is temperature sensitive. They do not show that the protein does any of the work of responding to cold. Transferring the term from Arabidopsis expression data to a peanut paralog-group member is weaker still.
Summary: Endomembrane system is correct but broad. The Golgi membrane annotation already entails it.
Reason: This is an ARBA-derived term. Golgi membrane is part of the endomembrane system, and Golgi membrane is retained in this review. The general term adds no information.
Summary: The protein is an integral membrane protein with seven predicted transmembrane helices. The specific membrane where the ortholog acts is the plasma membrane.
Reason: Membrane is correct but general. The supported specific location is the plasma membrane: AtCSLD3 accumulates in an apical plasma-membrane domain of root hairs, and PAINT node PTN001410221 propagates plasma membrane to AtCSLD3. That node is ancestral to the CSLD2/CSLD3 clade that contains this protein. Plasma membrane is also proposed below as a NEW annotation.
Summary: GO:0016760 means the transfer of glucose from UDP-glucose onto a (1,4)-beta-D-glucan chain. That is the activity shown for Arabidopsis CSLD3, this protein's closest characterised relative: purified, reconstituted CSLD3 is a UDP-glucose-dependent beta-1,4-glucan synthase. A CSLD3 chimera carrying a CESA6 catalytic domain also complements the csld3 mutant. The catalytic residues and the QxxRW motif are conserved in the peanut protein.
Reason: The previous review changed this term to mannan synthase activity. That decision is reversed. The InterPro2GO mapping from IPR005150 is broad, but for a member of the CSLD2/CSLD3 clade it matches the strongest available biochemistry (PMID:32169960, PMID:21765420). Clade membership was confirmed by the alignment and tree in the bioinformatics folder. Whether the CSLD3 product counts as cellulose in the microfibril sense is still open, but the GO term is defined by the reaction, and the reaction is the one shown.
Summary: Arabidopsis CSLD3 synthesises the beta-1,4-glucan of the cellulose-like polysaccharide in root-hair tips, and new synthesis of that polymer is needed for tip growth. The peanut protein is an ortholog in the same clade and has an intact catalytic site, so it is expected to do the same polymerisation step.
Reason: The previous review changed this to cell wall organization on the assumption that CSLDs make mannan. Reconstitution of CSLD3 and rescue with a CSLD3-CESA6 chimera contradict that assumption. The more specific GO:0052324 (plant-type cell wall cellulose biosynthetic process) would assume that the CSLD product is wall cellulose proper. The cited work calls it cellulose-like, so the general term is kept as the most precise one the evidence supports.
Summary: Mannan synthase activity (GDP-mannose donor) rests on Arabidopsis csld2/csld3/csld5 mutants, which have less mannan epitope, and on assays of CSLD proteins expressed in heterologous membranes (PMID:21471331, PMID:21904114). Purified, reconstituted CSLD3 instead behaves as a UDP-glucose-dependent beta-1,4-glucan synthase (PMID:32169960). Mannan synthase is not the established activity of the CSLD2/CSLD3 clade.
Reason: This is an ARBA-derived term. TAIR holds an experimental mannan-synthase annotation for AtCSLD3 (PMID:21904114), so the claim is not removed. The evidence conflicts, though. A reduced mannan epitope in a mutant is necessity evidence, and microsomal activity can come from co-purifying CSLA enzymes. The defined-protein reconstitution supports glucan synthesis. The previous review made this term the core activity, and that is reversed: the term is kept as a disputed, non-core claim.
ISS PMID:21765420 A role for CSLD3 during cell-wall synthesis in apical plasma...
NEW
Summary: The ortholog AtCSLD3 localises to a polarised plasma-membrane domain at root-hair tips, where it supplies beta-1,4-glucan synthase activity. PAINT node PTN001410221, which is ancestral to the CSLD2/CSLD3 clade, carries plasma membrane.
Reason: This is the main site of activity for the clade and is missing from this protein's annotations. GOA has no IBA rows for this peanut entry, which explains the gap. The protein's placement inside the AtCSLD2/AtCSLD3 clade was confirmed by alignment and tree, so it descends from the PAINT node whose annotations AtCSLD3 inherits.
the target groups with the AtCSLD2 + AtCSLD3 pair, then with OsCSLD2 and OsCSLD1
Core Functions
Processive UDP-glucose-dependent beta-1,4-glucan synthase of the CSLD2/CSLD3 clade. It is expected to polymerise a cellulose-like wall glucan at the plasma membrane during polarised wall deposition, such as tip growth.
Q: Do CSLD2/CSLD3-clade proteins make any mannan in vivo, or do the reduced mannan epitopes in csld mutants come from indirect effects on CSLA-made mannan?
Q: Is the CSLD3 beta-1,4-glucan product crystalline cellulose, which would justify GO:0052324, or a different wall glucan?
Suggested Experiments
Experiment: Express and purify the peanut protein, reconstitute it in proteoliposomes, and supply UDP-glucose or GDP-mannose to test donor specificity and the linkage of the product.
Experiment: Express a fluorescent fusion in peanut root hairs or pollen tubes to test whether the protein accumulates at the apical plasma membrane, as AtCSLD3 does.
External Prediction Reviews
These computational predictions are reviewed separately from the GOA annotation set used for this review. The assessments below are from this project and do not constitute official GO annotations or endorsement by GO/UniProt. They are not included in the existing annotation review above.
ProtNLM2 made no predictions (GO or subcellular location) for this peanut protein and returned only the protein name "Cellulose synthase-like protein D3". That name is consistent with the gene review, which places the protein in the angiosperm CSLD2/CSLD3 clade of the CESA/CSL glycosyltransferase family and infers, by orthology to Arabidopsis CSLD3, a UDP-glucose-dependent beta-1,4-glucan synthase acting at the plasma membrane during polarised wall deposition. With no GO or location output there is nothing to score; the absence of predictions is recorded as assessed.