crppa (formerly ispd) encodes the zebrafish ortholog of vertebrate ISPD/CRPPA, a cytosolic member of the IspD/TarI cytidylyltransferase family. The enzyme condenses CTP with D-ribitol 5-phosphate to form the nucleotide sugar CDP-ribitol, the donor used by the Golgi ribitol-phosphate transferases Fktn and Fkrp to build the ribitol-phosphate-containing core M3 O-mannosyl glycan (matriglycan) on alpha-dystroglycan, which alpha-dystroglycan needs to bind laminin and other extracellular matrix ligands. Vertebrates lack the non-mevalonate (MEP) isoprenoid pathway in which bacterial IspD acts, so the vertebrate enzyme is dedicated to dystroglycan glycosylation. Knockdown of zebrafish ispd/crppa causes alpha-dystroglycan hypoglycosylation, loss of sarcolemma integrity with muscle fiber degeneration, hydrocephalus and reduced eye size, modelling human Walker-Warburg syndrome, and sub-effective ispd, fktn and fkrp morpholinos act synergistically.
| GO Term | Evidence | Action | Reason |
|---|---|---|---|
| GO:0005829 cytosol | IBA GO_REF:0000033 | ACCEPT | Summary: PAINT propagation of cytosolic localization from the ISPD ancestral node, seeded by the human ortholog (UniProtKB:A4D126). Cytosolic localization was shown experimentally for human ISPD, and CDP-ribitol must be made in the cytosol before it is used in the Golgi by Fktn/Fkrp. No zebrafish-specific localization data exist, but nothing in the zebrafish sequence suggests a different compartment. Reason: The human ortholog was localized to the cytosol in the study that solved its structure and demonstrated cytidyltransferase activity [PMID:26687144], and UniProt records the same location for the zebrafish protein by similarity. A cytosolic nucleotide-sugar synthase feeding Golgi transferases is the expected topology for this pathway, so the node placement is sound and zebrafish crppa falls inside the clade that inherited it. Supporting Evidence: PMID:26687144 Functional studies demonstrated cytosolic localization of hISPD, and cytidyltransferase activity toward pentose phosphates file:DANRE/crppa/crppa-uniprot.txt SUBCELLULAR LOCATION: Cytoplasm, cytosol |
| GO:0035269 protein O-linked glycosylation via mannose | IBA GO_REF:0000033 | ACCEPT | Summary: PAINT propagation of the O-mannosyl glycosylation process role. The zebrafish gene itself is one of the descendant evidences (ZFIN gene in WITH/FROM), alongside human ISPD. crppa supplies the CDP-ribitol donor that Fktn/Fkrp use to add ribitol-phosphate to the core M3 O-mannosyl glycan of alpha-dystroglycan, so it performs a step of the pathway rather than merely being required for it. Reason: Participation is direct - ISPD synthesizes CDP-ribitol, and FKTN/FKRP transfer ribitol phosphate from that donor onto alpha-dystroglycan [PMID:27194101]. In zebrafish, ispd knockdown reduces IIH6-reactive (glycosylated) alpha-dystroglycan and sub-effective ispd morpholino doses synergize with fktn and fkrp morpholinos [PMID:22522421], which is the in vivo counterpart of the biochemistry. The zebrafish gene appearing in its own WITH/FROM reflects that experimental grounding and is expected. This is a core function. Supporting Evidence: PMID:22522421 Compared with control embryos, glycosylated Ξ±DG was reduced in ispd MO1-injected zebrafish embryos PMID:22522421 These results support a cooperative interaction between ispd and fktn/fkrp in Ξ±DG glycosylation. PMID:27194101 we demonstrate that isoprenoid synthase domain-containing protein (ISPD) synthesizes CDP-ribitol, present in muscle, and that both recombinant fukutin (FKTN) and fukutin-related protein (FKRP) can transfer a ribitol phosphate group from CDP-ribitol to Ξ±-dystroglycan file:DANRE/crppa/crppa-uniprot.txt Cytidylyltransferase required for protein O-linked mannosylation (PubMed:22522421) |
| GO:0047349 D-ribitol-5-phosphate cytidylyltransferase activity | IBA GO_REF:0000033 | ACCEPT | Summary: PAINT propagation of the specific catalytic activity (CTP + D-ribitol 5-phosphate to CDP-ribitol + diphosphate) from the vertebrate ISPD node, seeded by human ISPD. The activity was demonstrated with recombinant mammalian ISPD and the product identified by mass spectrometry and NMR; the zebrafish protein retains the ISPD cytidylyltransferase domain and morphant phenotypes match loss of this activity. Reason: Recombinant mammalian ISPD shows a clear CDP-ribitol pyrophosphorylase activity with D-ribitol-5-P as by far the best substrate and no activity on the MEP-pathway substrate 2-C-methyl-D-erythritol-4-P [PMID:27194101]; the human enzyme structure has a canonical cytidyltransferase domain and makes CDP-pentoses in vitro [PMID:26687144]. The zebrafish protein has not been assayed directly, but it is 65% similar to human, its knockdown phenocopies loss of dystroglycan glycosylation [PMID:22522421], and there is no sign of catalytic-residue loss, so inheriting the activity is the correct phylogenetic call. This is the core molecular function. Supporting Evidence: PMID:27194101 However, we clearly observed a CDP-ribitol pyrophosphorylase activity (Fig. 1b). PMID:27194101 No activity was observed in the presence of UTP, or in the presence of CTP and 2-C-methyl-D-erythritol-4-P, which is the substrate utilized in related enzymes in the mevalonate pathway. PMID:26687144 cytidyltransferase activity toward pentose phosphates, including ribulose 5-phosphate, ribose 5-phosphate, and ribitol 5-phosphate file:DANRE/crppa/crppa-uniprot.txt Reaction=D-ribitol 5-phosphate + CTP + H(+) = CDP-L-ribitol + diphosphate |
| GO:0003824 catalytic activity | IEA GO_REF:0000002 | MODIFY | Summary: InterPro2GO mapping from the IspD/TarI cytidylyltransferase domain (IPR018294) to the root molecular-function term. True but uninformative; the specific activity is D-ribitol-5-phosphate cytidylyltransferase (GO:0047349), which the gene already carries by IBA, ISS and RHEA/EC mapping. Reason: The domain mapping is correct in kind but far too general. The reaction catalysed by vertebrate ISPD is known [PMID:27194101] and is already annotated as GO:0047349, so this row should be expressed at that level. Proposed replacements: D-ribitol-5-phosphate cytidylyltransferase activity Supporting Evidence: PMID:27194101 However, we clearly observed a CDP-ribitol pyrophosphorylase activity (Fig. 1b). file:DANRE/crppa/crppa-uniprot.txt Catalyzes the formation of CDP-ribitol |
| GO:0005829 cytosol | IEA GO_REF:0000044 | ACCEPT | Summary: UniProt subcellular-location keyword mapping (SL-0091, cytosol). The UniProt location is itself transferred by similarity from human ISPD, for which cytosolic localization was demonstrated experimentally. Reason: Same evidence chain as the IBA cytosol row - human ISPD is cytosolic [PMID:26687144] and the zebrafish protein has no targeting signals or transmembrane segments that would argue otherwise; cytosolic synthesis of the nucleotide sugar precedes its use in the Golgi. Supporting Evidence: file:DANRE/crppa/crppa-uniprot.txt SUBCELLULAR LOCATION: Cytoplasm, cytosol PMID:26687144 Functional studies demonstrated cytosolic localization of hISPD |
| GO:0008299 isoprenoid biosynthetic process | IEA GO_REF:0000002 | REMOVE | Summary: InterPro2GO mapping from the IspD domain (IPR018294) to isoprenoid biosynthesis, reflecting the bacterial/plant role of IspD in the MEP pathway. Vertebrates do not have the MEP pathway, and mammalian ISPD has no activity on the MEP substrate; the vertebrate enzyme makes CDP-ribitol for dystroglycan glycosylation. Reason: This is a demonstrably wrong domain-to-process transfer. The MEP pathway is absent from vertebrates [PMID:22522421], and recombinant mammalian ISPD showed no activity with CTP and 2-C-methyl-D-erythritol-4-P while readily forming CDP-ribitol [PMID:27194101]. The enzyme retained the fold and the cytidylyltransferase chemistry but was repurposed to a different pathway, so the process term is wrong in vertebrates. Supporting Evidence: PMID:22522421 In plants, protozoa and some bacteria, ISPD belongs to the non-mevalonate isoprenoid biosynthesis (MEP) pathway, which is absent in vertebrates PMID:27194101 No activity was observed in the presence of UTP, or in the presence of CTP and 2-C-methyl-D-erythritol-4-P, which is the substrate utilized in related enzymes in the mevalonate pathway. file:DANRE/crppa/crppa-uniprot.txt Not involved in dolichol production file:DANRE/crppa/crppa-deep-research-falcon.md vertebrate CRPPA/ISPD function is in **CDP-ribitol synthesis for glycosylation**, not the microbial **MEP isoprenoid pathway** |
| GO:0009101 glycoprotein biosynthetic process | IEA GO_REF:0000117 | MODIFY | Summary: ARBA machine-learned mapping to the general glycoprotein-biosynthesis term. Correct in direction - crppa supplies a donor for alpha-dystroglycan O-mannosyl glycan assembly - but the gene is already annotated with the specific descendant protein O-linked glycosylation via mannose (GO:0035269) by IBA and IMP. Reason: GO:0035269 is a descendant of GO:0009101 and is the term supported by the biochemistry (CDP-ribitol donor for FKTN/FKRP ribitol-phosphate transfer to alpha-dystroglycan [PMID:27194101]) and by the zebrafish knockdown phenotype [PMID:22522421]. The broad term is not wrong, but the specific one is established and should be used. Proposed replacements: protein O-linked glycosylation via mannose Supporting Evidence: PMID:27194101 both recombinant fukutin (FKTN) and fukutin-related protein (FKRP) can transfer a ribitol phosphate group from CDP-ribitol to Ξ±-dystroglycan file:DANRE/crppa/crppa-uniprot.txt Cytidylyltransferase required for protein O-linked mannosylation (PubMed:22522421) |
| GO:0047349 D-ribitol-5-phosphate cytidylyltransferase activity | IEA GO_REF:0000120 | ACCEPT | Summary: Automated mapping from the UniProt catalytic-activity annotation (RHEA:12456, EC 2.7.7.40) to the matching GO activity term. The reaction is the one demonstrated for mammalian ISPD. Reason: The RHEA/EC mapping is exact for the reaction demonstrated with recombinant mammalian ISPD [PMID:27194101] and recorded on the zebrafish entry by similarity. Same core activity as the IBA and ISS rows. Supporting Evidence: file:DANRE/crppa/crppa-uniprot.txt Reaction=D-ribitol 5-phosphate + CTP + H(+) = CDP-L-ribitol + diphosphate PMID:27194101 While our data provide evidence that ISPD serves to produce CDP-ribitol from CTP and D-ribitol-5-P |
| GO:0070567 cytidylyltransferase activity | IEA GO_REF:0000002 | MODIFY | Summary: InterPro2GO mapping from the ISPD-specific domain entry (IPR034683) to the generic cytidylyltransferase activity term. Correct chemistry but less specific than the established D-ribitol-5-phosphate cytidylyltransferase activity. Reason: The parent term is true (ISPD transfers a cytidylyl group to a pentose phosphate acceptor [PMID:26687144]) but the physiologically relevant acceptor, D-ribitol 5-phosphate, is known [PMID:27194101] and GO:0047349 is already annotated; the row should be narrowed to it. Proposed replacements: D-ribitol-5-phosphate cytidylyltransferase activity Supporting Evidence: PMID:26687144 Our combined results indicate that hISPD is a cytidyltransferase file:DANRE/crppa/crppa-uniprot.txt Catalyzes the formation of CDP-ribitol |
| GO:0047349 D-ribitol-5-phosphate cytidylyltransferase activity | ISS GO_REF:0000024 | ACCEPT | Summary: UniProt curator ISS transfer of the specific activity from human ISPD (A4D126). The donor annotation rests on direct enzymology with recombinant mammalian ISPD. Reason: The source annotation is experimentally grounded [PMID:27194101, PMID:26687144] and the zebrafish protein is a one-to-one ortholog with the same domain architecture whose loss produces the dystroglycan-hypoglycosylation phenotype expected from loss of CDP-ribitol supply [PMID:22522421]. Core activity. Supporting Evidence: PMID:27194101 However, we clearly observed a CDP-ribitol pyrophosphorylase activity (Fig. 1b). file:DANRE/crppa/crppa-uniprot.txt Catalyzes the formation of CDP-ribitol |
| GO:0005829 cytosol | ISS GO_REF:0000024 | ACCEPT | Summary: UniProt curator ISS transfer of cytosolic localization from human ISPD, where it was shown experimentally. Reason: Cytosolic localization of human ISPD was demonstrated in the structure and activity study [PMID:26687144]; the zebrafish protein lacks any organellar targeting feature, and the pathway logic (cytosolic donor synthesis, Golgi transfer by Fktn/Fkrp) is conserved. Supporting Evidence: PMID:26687144 Functional studies demonstrated cytosolic localization of hISPD file:DANRE/crppa/crppa-uniprot.txt SUBCELLULAR LOCATION: Cytoplasm, cytosol |
| GO:0042803 protein homodimerization activity | ISS GO_REF:0000024 | KEEP AS NON CORE | Summary: ISS transfer from human ISPD of a homodimeric quaternary structure, recorded by UniProt as a SUBUNIT comment. Homodimerization is a structural property of the cytidylyltransferase fold rather than a function in its own right; the zebrafish protein has not been characterized structurally. Reason: The claim is plausible and consistent with the human crystal structure lineage of the UniProt comment, but the only support available here is the by-similarity SUBUNIT line; it says nothing about what the enzyme does, so it is retained as a non-core structural descriptor rather than a core function. Supporting Evidence: file:DANRE/crppa/crppa-uniprot.txt SUBUNIT: Homodimer. |
| GO:0070567 cytidylyltransferase activity | ISS GO_REF:0000024 | MODIFY | Summary: UniProt ISS transfer of the generic cytidylyltransferase activity from human ISPD. Correct chemistry but the specific D-ribitol-5-phosphate cytidylyltransferase activity is established and already annotated. Reason: Human ISPD is a cytidyltransferase acting on pentose phosphates [PMID:26687144], with D-ribitol 5-phosphate as the physiological substrate [PMID:27194101]; the specific child GO:0047349 should replace this generic parent, as for the IEA row from IPR034683. Proposed replacements: D-ribitol-5-phosphate cytidylyltransferase activity Supporting Evidence: PMID:26687144 Our combined results indicate that hISPD is a cytidyltransferase file:DANRE/crppa/crppa-uniprot.txt Catalyzes the formation of CDP-ribitol |
| GO:0010559 regulation of glycoprotein biosynthetic process | IMP PMID:22522421 Mutations in ISPD cause Walker-Warburg syndrome and defectiv... | MODIFY | Summary: ZFIN IMP annotation from the ispd MO1 morphant (ZDB-MRPHLNO-120629-1), which shows reduced IIH6-reactive alpha-dystroglycan on western blot. The experimental finding is sound, but it shows that crppa is a component of the glycosylation pathway (it makes the CDP-ribitol donor), not a regulator that modulates the rate of glycoprotein biosynthesis. Reason: GO:0010559 is defined as a process that modulates the rate or extent of glycoprotein biosynthesis. Loss of a pathway enzyme reduces the product, but that is participation, not regulation - ISPD synthesizes the CDP-ribitol that FKTN/FKRP transfer onto alpha-dystroglycan [PMID:27194101]. The same paper is the basis for the UniProt IMP annotation to protein O-linked glycosylation via mannose (GO:0035269), which is the correct process term for this evidence (alpha-dystroglycan hypoglycosylation in ispd morphants [PMID:22522421]). Proposed replacements: protein O-linked glycosylation via mannose Supporting Evidence: PMID:22522421 Compared with control embryos, glycosylated Ξ±DG was reduced in ispd MO1-injected zebrafish embryos PMID:27194101 both recombinant fukutin (FKTN) and fukutin-related protein (FKRP) can transfer a ribitol phosphate group from CDP-ribitol to Ξ±-dystroglycan |
| GO:0010559 regulation of glycoprotein biosynthetic process | IGI PMID:22522421 Mutations in ISPD cause Walker-Warburg syndrome and defectiv... | MODIFY | Summary: ZFIN IGI annotation from co-injection of sub-effective ispd MO1 with a fktn morpholino (ZDB-MRPHLNO-110418-1), which increased hydrocephalus and further reduced glycosylated alpha-dystroglycan. The genetic interaction places crppa in the same pathway as the ribitol-phosphate transferase Fktn, which is exactly what the later biochemistry showed. Reason: As for the IMP row, the evidence supports participation in alpha-dystroglycan O-mannosyl glycan assembly rather than regulation of glycoprotein biosynthesis. The synergy with fktn is explained by ISPD supplying the CDP-ribitol donor that FKTN uses [PMID:27194101]. Replace with GO:0035269 while retaining the IGI evidence. Proposed replacements: protein O-linked glycosylation via mannose Supporting Evidence: PMID:22522421 Co-injection of sub-effective doses of ispd MO1 with fktn or fkrp MO showed a marked increase in the proportion of embryos with hydrocephalus PMID:22522421 These results support a cooperative interaction between ispd and fktn/fkrp in Ξ±DG glycosylation. |
| GO:0048854 brain morphogenesis | IMP PMID:22522421 Mutations in ISPD cause Walker-Warburg syndrome and defectiv... | KEEP AS NON CORE | Summary: ZFIN IMP annotation from the ispd MO1 morphant, which develops hydrocephalus and incomplete brain folding by 48 hpf, modelling the cobblestone lissencephaly and hydrocephalus of Walker-Warburg syndrome. The brain phenotype is a downstream consequence of alpha-dystroglycan hypoglycosylation, which disrupts basement-membrane attachment of neural cells; crppa does no work in brain morphogenesis itself. Reason: The phenotype is real and highly penetrant with two independent morpholinos [PMID:22522421], and it is biologically important as the zebrafish WWS model, so the row should stay. But the gene product acts upstream of brain morphogenesis (it makes a nucleotide sugar in the cytosol); the qualifier acts_upstream_of_or_within already expresses that, and the term is not a core function. Supporting Evidence: PMID:22522421 High doses of ispd MO1 (7 ng) caused hydrocephalus and incomplete brain folding in 82% of embryos (n=88) by 48 hours post fertilization (h.p.f.) PMID:22522421 Injection of ispd MO2 (3 ng) caused similar morphological abnormalities, assuring the specificity of both MOs |
| GO:0055001 muscle cell development | IMP PMID:22522421 Mutations in ISPD cause Walker-Warburg syndrome and defectiv... | KEEP AS NON CORE | Summary: ZFIN IMP annotation from the ispd MO1 morphant, which shows Evans blue dye uptake into myotendinous-junction-anchored fibers by 48 hpf (compromised sarcolemma) followed by muscle fiber degeneration at 72 hpf. The defect is a dystroglycanopathy - loss of matriglycan-dependent laminin anchorage - and is downstream of the enzyme's role in alpha-dystroglycan glycosylation. Reason: The muscle phenotype is well documented and mechanistically explained in the paper (hypoglycosylated alpha-dystroglycan, sarcolemma leakage preceding degeneration, with laminin and dystrophin still in place) [PMID:22522421]. crppa does not itself execute muscle cell development; it supplies the glycan donor on which muscle-fiber attachment depends. Keep as a non-core, upstream annotation. Supporting Evidence: PMID:22522421 loss of Ispd function in zebrafish results in Ξ±DG hypoglycosylation and compromised sarcolemma integrity, preceding muscle fiber degeneration PMID:22522421 ispd MO1-injected embryos showed muscle fiber degeneration by 72 h.p.f. |
| GO:0035269 protein O-linked glycosylation via mannose | IMP PMID:22522421 Mutations in ISPD cause Walker-Warburg syndrome and defectiv... | ACCEPT | Summary: UniProt IMP annotation from the zebrafish ispd knockdown, in which IIH6-reactive (functionally O-mannosylated) alpha-dystroglycan is reduced. This is the process term that correctly captures what the morphant experiment shows, and it matches the subsequently defined biochemical role of ISPD as the CDP-ribitol synthase of the matriglycan pathway. Reason: Direct zebrafish evidence that crppa is needed for alpha-dystroglycan O-mannosyl glycan maturation [PMID:22522421], combined with the biochemical demonstration that ISPD performs a step of that pathway by making the CDP-ribitol donor used by FKTN/FKRP [PMID:27194101]. Core process. Supporting Evidence: PMID:22522421 Compared with control embryos, glycosylated Ξ±DG was reduced in ispd MO1-injected zebrafish embryos PMID:27194101 we demonstrate that isoprenoid synthase domain-containing protein (ISPD) synthesizes CDP-ribitol, present in muscle, and that both recombinant fukutin (FKTN) and fukutin-related protein (FKRP) can transfer a ribitol phosphate group from CDP-ribitol to Ξ±-dystroglycan file:DANRE/crppa/crppa-uniprot.txt Cytidylyltransferase required for protein O-linked mannosylation (PubMed:22522421) |
| GO:0010559 regulation of glycoprotein biosynthetic process | IGI PMID:22522421 Mutations in ISPD cause Walker-Warburg syndrome and defectiv... | MODIFY | Summary: Second ZFIN IGI row for the same paper, this time for the interaction of sub-effective ispd MO1 with the fkrp morpholino (ZDB-MRPHLNO-110418-2). Fkrp is the second ribitol-phosphate transferase that uses CDP-ribitol, so the synergy reflects shared pathway membership, not regulation. Reason: Same argument as the fktn IGI row - ISPD provides the donor for FKRP's ribitol-phosphate transfer [PMID:27194101], and co-knockdown with fkrp increases hydrocephalus and further lowers glycosylated alpha-dystroglycan [PMID:22522421]. The evidence supports GO:0035269, not a regulatory term. Proposed replacements: protein O-linked glycosylation via mannose Supporting Evidence: PMID:22522421 Co-injection of sub-effective doses of ispd MO1 with fktn or fkrp MO showed a marked increase in the proportion of embryos with hydrocephalus PMID:22522421 These results support a cooperative interaction between ispd and fktn/fkrp in Ξ±DG glycosylation. |
| GO:0055001 muscle cell development | IMP PMID:22522421 Mutations in ISPD cause Walker-Warburg syndrome and defectiv... | KEEP AS NON CORE | Summary: Second ZFIN IMP row for muscle cell development, from the independent ispd MO2 morphant (ZDB-MRPHLNO-120629-2), which reproduces the MO1 phenotype and also shows Evans blue dye infiltration of muscle fibers. Same downstream dystroglycanopathy phenotype as the MO1 row. Reason: The second morpholino confirms specificity of the muscle phenotype [PMID:22522421], but the participation argument is unchanged - crppa supplies the CDP-ribitol donor for alpha-dystroglycan glycosylation and does not itself carry out muscle cell development. Retain as non-core with the same action as the MO1 row. Supporting Evidence: PMID:22522421 Injection of ispd MO2 (3 ng) caused similar morphological abnormalities, assuring the specificity of both MOs PMID:22522421 loss of Ispd function in zebrafish results in Ξ±DG hypoglycosylation and compromised sarcolemma integrity, preceding muscle fiber degeneration |
Loading supporting contentβ¦
Download this section (compressed HTML)Q: Has the CDP-ribitol synthase activity of the zebrafish Crppa protein been measured directly, and does the zebrafish enzyme share the mammalian preference for D-ribitol 5-phosphate over ribose/ribulose 5-phosphate?
Q: Does dietary ribitol supplementation, which partially rescues alpha-dystroglycan glycosylation in ISPD-mutant human fibroblasts, rescue the muscle and brain phenotypes of zebrafish crppa loss-of-function models?
Experiment: Express and purify zebrafish Crppa (both N-terminal isoforms) and assay CDP-ribitol formation from CTP and D-ribitol 5-phosphate by HPLC/LC-MS, comparing kinetics with the mammalian enzyme, to convert the ISS/IBA molecular-function annotations into direct evidence.
Type: BIOCHEMICAL
Experiment: Generate a stable crppa mutant line (CRISPR) and test ribitol supplementation for rescue of alpha-dystroglycan IIH6 reactivity, Evans blue dye uptake and hydrocephalus, to confirm the morphant phenotypes and the CDP-ribitol-dependent mechanism in vivo.
Type: GENETIC
Loading supporting contentβ¦
Download this section (compressed HTML)Loading supporting contentβ¦
Download this section (compressed HTML)Loading supporting contentβ¦
Download this section (compressed HTML)