Functional annotation report: *C. elegans dpm-1* (UniProt U4PF58) Falcon Edison Scientific Literature 23 citations 1 artifacts 2026-09-10T14:44:41.565231

The research report should be a detailed narrative explaining the function, biological processes, and localization of the gene product. Citations should be given for all claims.

You should prioritize authoritative reviews and primary scientific literature when conducting research. You can supplement
this with annotations you find in gene/protein databases, but these can be outdated or inaccurate.

We are specifically interested in the primary function of the gene - for enzymes, what reaction is catalyzed, and what is the substrate specificity? For transporters, what is the substrate? For structural proteins or adapters, what is the broader structural role? For signaling molecules, what is the role in the pathway.

We are interested in where in or outside the cell the gene product carries out its function.

We are also interested in the signaling or biochemical pathways in which the gene functions. We are less interested in broad pleiotropic effects, except where these elucidate the precise role.

Include evidence where possible. We are interested in both experimental evidence as well as inference from structure, evolution, or bioinformatic analysis. Precise studies should be prioritized over high-throughput, where available.

Functional annotation report: C. elegans dpm-1 (UniProt U4PF58)

Executive conclusion

The supplied identifiers are internally consistent: U4PF58 corresponds to the Caenorhabditis elegans gene dpm-1, ORF/locus Y66H1A.2 (CELE_Y66H1A.2), annotated as dolichol-phosphate mannosyltransferase subunit 1 and containing a DPM1-like domain (InterPro IPR039528). However, the literature retrieved here contains no gene-specific biochemical characterization of U4PF58/Y66H1A.2. In particular, no purified-enzyme assay, mutant rescue, endogenous localization experiment, or pathway-resolved glycomic analysis was found for this exact worm protein. Comparative analyses recognize a conserved nematode DPM1 sequence, but do not experimentally characterize the C. elegans locus (haeuptle2009congenitaldisordersof pages 8-9, haeuptle2009congenitaldisordersof pages 7-8).

Accordingly, the most defensible annotation is that U4PF58 is the worm ortholog of dolichol-phosphate mannose synthase catalytic subunit DPM1, predicted to catalyze:

GDP-α-D-mannose + dolichol phosphate → GDP + dolichol-phosphate mannose (Dol-P-Man).

This assignment is strongly supported by the supplied DPM1-like domain and the conserved biochemistry of DPM1/DPMS proteins, but remains an orthology/domain-based functional inference rather than a reaction demonstrated directly with U4PF58 (maeda2008dolicholphosphatemannosesynthase pages 3-4, banerjee2017dolicholphosphatemannose pages 1-3).

Annotation Best-supported conclusion Evidence level Key caveat
Identity dpm-1, ORF Y66H1A.2 / CELE_Y66H1A.2, encodes the Caenorhabditis elegans protein U4PF58, annotated as dolichol-phosphate mannosyltransferase subunit 1 and containing a DPM1-like domain (IPR039528). Direct database identity The submitted UniProt annotation is computationally imported rather than supported by a retrieved gene-specific biochemical study.
Primary molecular function/reaction Most likely catalyzes GDP-α-D-mannose + dolichol phosphate → GDP + dolichol-phosphate mannose (Dol-P-Man). This is the conserved reaction of DPM1/DPMS enzymes (maeda2008dolicholphosphatemannosesynthase pages 3-4, banerjee2017dolicholphosphatemannose pages 1-3). Strong orthology-mechanistic inference Catalytic activity has not been directly demonstrated for purified U4PF58 or in C. elegans extracts.
Substrates/products Predicted physiological donor: GDP-mannose; lipid acceptor: dolichol phosphate; products: GDP and Dol-P-Man. Other prenyl-phosphate or lipid alcohol acceptors can be used by some DPMS enzymes in vitro, but U4PF58 specificity and preferred dolichol-chain length are unknown (banerjee2017dolicholphosphatemannose pages 1-3). Strong orthology-mechanistic inference Broad in-vitro acceptor permissiveness observed in other species should not be assigned to the worm enzyme.
Enzyme family/mechanism Expected to be a CAZy GT2, GT-A-fold, inverting glycosyltransferase with a conserved acidic DXD/DAD metal-binding motif; comparative models implicate metal-assisted nucleotide-sugar binding and direct transfer to the Dol-P phosphate (maeda2008dolicholphosphatemannosesynthase pages 3-4, banerjee2017dolicholphosphatemannose pages 1-3, gandini2026crystallographicdatafor pages 6-9). Strong orthology-mechanistic inference The fold, catalytic residues, metal requirement, and product stereochemistry have not been experimentally established for U4PF58.
Cellular localization Most plausibly associated with the endoplasmic-reticulum membrane, with synthesis occurring on the cytosolic face before Dol-P-Man is translocated for lumenal use (packer2022functionalcharacterisationof pages 51-55, packer2022functionalcharacterisationofa pages 51-55). Strong orthology-mechanistic inference Worm-specific imaging, membrane topology, and dependence on DPM2/DPM3-like partners have not been established.
Downstream pathways Dol-P-Man is expected to supply mannose for lumenal N-glycan precursor assembly, GPI-anchor biosynthesis, and protein O- and C-mannosylation (banerjee2017dolicholphosphatemannose pages 1-3, koff2023proteinomannosylationone pages 1-2). Strong orthology-mechanistic inference The relative contribution of C. elegans dpm-1 to each branch has not been measured; individual pathways may differ among lineages.
Worm-specific experimental evidence The retrieved literature did not provide a U4PF58/Y66H1A.2-specific enzyme assay, mutant-rescue study, localization experiment, expression analysis, or defined phenotype. Comparative literature recognizes a conserved nematode DPM1 sequence but does not characterize this worm locus directly (haeuptle2009congenitaldisordersof pages 8-9, haeuptle2009congenitaldisordersof pages 7-8). Not established in worm Database annotation and homology currently carry most of the functional assignment.
Recent 2023–2024 comparative developments A 2023 reconstitution study found Dpm1 activity sensitive to membrane composition—enhanced by phosphatidylethanolamine/fluidity and inhibited by ergosterol or ceramide (kikul2023theinfluenceof pages 87-89). A 2024 human keratinocyte/3D-epidermis study linked DPM1 depletion to reduced glycans, impaired desmosomal adhesion, and altered differentiation (rathod2024dpm1modulatesdesmosomal pages 2-5, rathod2024dpm1modulatesdesmosomal pages 23-26). Liposomal mannose-1-phosphate improved glycosylation in several patient-derived CDG fibroblast models in 2024 (budhraja2024liposomeencapsulatedmannose1phosphatetherapy pages 13-15). Strong comparative evidence; not direct worm evidence These studies used reconstituted membranes or human cells and do not demonstrate equivalent regulation, phenotypes, or therapeutic responses in C. elegans.
Key unresolved questions Priorities are direct enzymatic validation; GDP-mannose and Dol-P kinetic measurements; dolichol-chain preference; metal dependence; ER topology; complex composition; tissue/developmental expression; loss-of-function phenotype; and pathway-resolved glycomic analysis. Not established in worm Definitive annotation requires targeted genetics, subcellular imaging, membrane biochemistry, complementation, and lipid-linked glycan profiling.

Table: Compact functional annotation of C. elegans dpm-1/U4PF58, explicitly separating database identity and conserved mechanistic inference from findings not yet demonstrated in the worm.

1. Identity verification and ambiguity control

The target should be referred to as C. elegans dpm-1 / Y66H1A.2 / U4PF58, not simply “DPM1.” The symbol is used for orthologous proteins in humans, fungi, plants, protists, and other organisms, whose membrane topology and subunit organization can differ substantially. The protein description and DPM1-like domain supplied by UniProt align with the conserved dolichol-phosphate mannose synthase family. Literature on human DPM1, yeast Dpm1, plant DPMS1, or archaeal DPMS is therefore used below only as comparative mechanistic evidence—not as direct evidence about the worm protein (maeda2008dolicholphosphatemannosesynthase pages 3-4, haeuptle2009congenitaldisordersof pages 7-8).

No contradictory literature indicating that Y66H1A.2 encodes a different protein was found. The principal uncertainty is not target identity, but the absence of direct functional experiments on this exact gene product.

2. Primary molecular function

Reaction and physiological substrates

DPM1-family enzymes transfer mannose from the nucleotide sugar GDP-mannose to the lipid-phosphate acceptor dolichol phosphate, releasing GDP and producing Dol-P-Man. DPMS is described as a dolichylphosphate β-D-mannosyltransferase and an inverting glycosyltransferase; thus, transfer from GDP-α-D-mannose generates the opposite anomeric configuration in the glycosidic product (banerjee2017dolicholphosphatemannose pages 1-3, gandini2026crystallographicdatafor pages 6-9).

For U4PF58, GDP-mannose and dolichol phosphate are therefore the predicted physiological substrates. The worm enzyme’s preferred dolichol chain length, apparent Km values, turnover number, metal preference, and ability to use alternative polyprenyl phosphates have not been measured. Some non-worm DPMS preparations accept undecaprenol, citronellol, or retinol derivatives in vitro, but such assay permissiveness should not be assigned to U4PF58 without testing (banerjee2017dolicholphosphatemannose pages 1-3).

Enzyme family and catalytic mechanism

Comparative DPM1 proteins are classified as CAZy family GT2, GT-A-fold glycosyltransferases. They contain a conserved acidic DXD/DAD motif involved in metal-dependent nucleotide-sugar binding. Structural and mechanistic work supports binding of GDP-mannose and a divalent metal before dolichol phosphate, followed by attack of the acceptor phosphate on mannose C1 and inversion of stereochemistry. Comparative modeling has implicated residues including Tyr12, Asp44, Asp97, and Arg212 in some DPM1 homologs, but residue numbering and functional importance must be verified in the U4PF58 sequence before transfer to the worm protein (maeda2008dolicholphosphatemannosesynthase pages 3-4, banerjee2017dolicholphosphatemannose pages 1-3, gandini2026crystallographicdatafor pages 6-9).

The DPM1-like domain IPR039528 therefore aligns well with the proposed catalytic role. Direct confirmation should include motif inspection, mutagenesis of the worm DXD-like sequence, and activity assays with GDP-mannose and defined dolichol-phosphate species.

3. Cellular localization and topology

Across eukaryotic systems, Dol-P-Man synthesis occurs at the cytosolic face of the endoplasmic-reticulum membrane. This orientation is chemically logical because GDP-mannose is available to the cytosolic catalytic domain. Dol-P-Man must subsequently become available on the ER lumenal side to supply lumen-facing mannosyltransferases (packer2022functionalcharacterisationof pages 51-55, packer2022functionalcharacterisationofa pages 51-55).

Thus, U4PF58 is most plausibly an ER-associated enzyme acting on the cytosolic membrane leaflet. This remains inferred: no endogenous U4PF58 imaging, microsomal protease-protection assay, or membrane-topology experiment was retrieved.

DPM systems differ structurally among lineages. Saccharomyces cerevisiae uses a roughly 267-residue Dpm1 with a membrane anchor, whereas mammalian DPM1 is an approximately 260-residue, largely cytosolic catalytic subunit associated with transmembrane partners DPM2 and DPM3. In mammalian systems, removal of DPM2 can increase the reaction Km approximately tenfold, illustrating that accessory proteins affect lipid-substrate coordination and complex stability (bananadube2018investigationofaa pages 71-74, packer2022functionalcharacterisationof pages 51-55). It is currently unsafe to assign either architecture to C. elegans without sequence-topology analysis and interaction experiments.

4. Biochemical pathways

Dol-P-Man is not generally the final glycan product. It is a membrane-activated mannose donor supplying several ER-lumen pathways:

  1. N-linked glycosylation. Dol-P-Man donates lumenal mannose residues during maturation of the dolichol-linked oligosaccharide precursor subsequently transferred to Asn residues of secretory and membrane proteins.
  2. GPI-anchor biosynthesis. Mannose residues in glycosylphosphatidylinositol anchors derive from Dol-P-Man.
  3. Protein O-mannosylation. ER O-mannosyltransferase families use Dol-P-Man to modify selected Ser/Thr residues.
  4. Protein C-mannosylation. Dol-P-Man also supplies mannose for appropriate tryptophan C-mannosylation reactions (maeda2008dolicholphosphatemannosesynthase pages 3-4, banerjee2017dolicholphosphatemannose pages 1-3, koff2023proteinomannosylationone pages 1-2).

The 2023 review by Koff and colleagues identifies animal ER O-mannosylation machinery—including POMT1/2, TMTC1–4, and TMEM260 families—as Dol-P-Man-dependent. It emphasizes that substrate recognition and the full substrate repertoire remain incompletely understood. This is broad animal evidence and does not demonstrate which branches specifically depend on C. elegans dpm-1 (published August 2023; https://doi.org/10.1093/glycob/cwad067) (koff2023proteinomannosylationone pages 1-2).

The likely proximal consequence of losing worm DPM-1 would therefore be depletion of Dol-P-Man, followed by defects in one or more of these pathways. Precise branch-specific effects cannot currently be ranked from direct worm evidence.

5. Evidence status in C. elegans

The retrieved evidence does not establish for Y66H1A.2/U4PF58:

Consequently, claims such as embryonic lethality, neurological dysfunction, altered epidermal adhesion, or ER stress must not be presented as worm phenotypes. Those observations arise from other organisms or human disease systems. The current worm annotation is best described as high-confidence predicted molecular function, but low direct experimental validation (haeuptle2009congenitaldisordersof pages 8-9, haeuptle2009congenitaldisordersof pages 7-8).

6. Recent developments, 2023–2024

Membrane control of enzyme activity

Kikul’s January 2023 defined-liposome study found that reconstituted Dpm1 activity varied with bilayer composition. Phosphatidylethanolamine and a 3:1 unsaturated-to-saturated lipid ratio enhanced activity, whereas more saturated 1:1 mixtures reduced it; ergosterol and ceramide were inhibitory under the tested conditions. Comparisons used at least three replicates, but the artificial liposome system, batch variability, and incomplete organismal validation limit interpretation (https://doi.org/10.11588/heidok.00034034) (kikul2023theinfluenceof pages 87-89). This work supports the expert view that DPMS activity is not determined by enzyme abundance alone: access to the hydrophobic Dol-P acceptor and membrane physical properties may be regulatory variables. It does not show that the same lipid dependencies occur in C. elegans.

Human epithelial adhesion and differentiation

Rathod et al. used CRISPR-mediated DPM1 depletion in HaCaT cells, primary human keratinocytes, and three-dimensional reconstructed epidermis. DPM1 loss reduced oligomannose and complex glycans, weakened dispase-measured cell adhesion, altered desmosomal proteins, and disrupted epidermal stratification. In reconstructed epidermis, the cornified layer thickened while other layers thinned, although total epidermal thickness did not change. Several experiments used three to five biological replicates—for example, N=4 HaCaT adhesion assays and N=5 primary-keratinocyte assays. SERPINB5 and altered desmoplakin phosphorylation were implicated downstream (published March 2024; https://doi.org/10.1083/jcb.202305006) (rathod2024dpm1modulatesdesmosomal pages 2-5, rathod2024dpm1modulatesdesmosomal pages 23-26).

This is a real-world mechanistic application of DPM1 biology to epithelial tissue organization, but it is human-cell evidence. It neither identifies a comparable worm tissue phenotype nor proves that SERPINB5 is a conserved DPM-1 effector in nematodes.

Experimental glycosylation therapy

Budhraja et al. treated patient-derived congenital-disorder-of-glycosylation fibroblasts with liposome-encapsulated mannose-1-phosphate (GLM101). Glycosylation improved significantly in PMM2-CDG and ALG2-CDG cells and mildly in ALG11-CDG cells, although exact effect sizes were not available in the retrieved excerpt. The intervention acts upstream by increasing mannose-phosphate/GDP-mannose precursor availability; it is not a direct DPM1 replacement. Translation remains preliminary and requires clinical validation (published June 2024; https://doi.org/10.1016/j.ymgme.2024.108487) (budhraja2024liposomeencapsulatedmannose1phosphatetherapy pages 13-15).

7. Applications and wider significance

DPM1/DPMS biology currently has three principal applications:

A further host-pathogen application is suggested by human CRISPR screens in which the DPMS complex supports flavivirus RNA replication and glycoprotein folding. Such findings make the pathway a potential antiviral target, although systemic DPMS inhibition is likely to have substantial toxicity because several essential host glycosylation pathways share Dol-P-Man.

The strongest present conclusion is that DPM-1 synthesizes the activated lipid-linked mannose donor Dol-P-Man at the ER, thereby occupying a metabolically upstream position shared by several glycosylation pathways. The precise reaction assignment is much stronger than any proposed worm phenotype or tissue role.

Priority experiments for definitive annotation are:

  1. Express and purify U4PF58, or prepare tagged worm microsomes, and assay incorporation from radiolabeled or LC–MS-traceable GDP-mannose into Dol-P-Man.
  2. Determine Km and kcat for GDP-mannose and multiple native-length Dol-P species, with Mg²⁺/Mn²⁺ titration.
  3. Mutate the predicted DXD/DAD motif and test loss of activity plus rescue by wild-type dpm-1.
  4. Establish ER localization and catalytic sidedness using endogenous tagging, colocalization, microsomal protease protection, and selective membrane permeabilization.
  5. Identify complex partners by affinity purification–mass spectrometry and test candidate DPM2/DPM3-like proteins.
  6. Generate a conditional or tissue-specific loss-of-function allele, followed by rescue and targeted analysis of Dol-P-Man, lipid-linked N-glycan precursors, GPI-anchored proteins, and O-/C-mannosylated substrates.

Final annotation

Recommended functional statement: C. elegans dpm-1 (Y66H1A.2; UniProt U4PF58) encodes a DPM1-like dolichol-phosphate mannosyltransferase predicted to catalyze transfer of mannose from GDP-mannose to dolichol phosphate, yielding GDP and Dol-P-Man at the cytosolic face of the ER. Dol-P-Man is expected to supply lumenal N-glycan assembly, GPI-anchor synthesis, and protein O- and C-mannosylation. The reaction, ER localization, topology, complex composition, substrate-chain preference, and physiological pathway contributions have not yet been demonstrated directly for U4PF58.

References

  1. (haeuptle2009congenitaldisordersof pages 8-9): Micha A. Haeuptle and Thierry Hennet. Congenital disorders of glycosylation: an update on defects affecting the biosynthesis of dolichol‐linked oligosaccharides. Human Mutation, 30:1628-1641, Dec 2009. URL: https://doi.org/10.1002/humu.21126, doi:10.1002/humu.21126. This article has 231 citations and is from a domain leading peer-reviewed journal.

  2. (haeuptle2009congenitaldisordersof pages 7-8): Micha A. Haeuptle and Thierry Hennet. Congenital disorders of glycosylation: an update on defects affecting the biosynthesis of dolichol‐linked oligosaccharides. Human Mutation, 30:1628-1641, Dec 2009. URL: https://doi.org/10.1002/humu.21126, doi:10.1002/humu.21126. This article has 231 citations and is from a domain leading peer-reviewed journal.

  3. (maeda2008dolicholphosphatemannosesynthase pages 3-4): Yusuke Maeda and Taroh Kinoshita. Dolichol-phosphate mannose synthase: structure, function and regulation. Biochimica et biophysica acta, 1780 6:861-8, Jun 2008. URL: https://doi.org/10.1016/j.bbagen.2008.03.005, doi:10.1016/j.bbagen.2008.03.005. This article has 152 citations.

  4. (banerjee2017dolicholphosphatemannose pages 1-3): Dipak K. Banerjee, Zhenbo Zhang, Krishna Baksi, and Jesús E. Serrano-Negrón. Dolichol phosphate mannose synthase: a glycosyltransferase with unity in molecular diversities. Glycoconjugate Journal, 34:467-479, Jun 2017. URL: https://doi.org/10.1007/s10719-017-9777-4, doi:10.1007/s10719-017-9777-4. This article has 25 citations and is from a peer-reviewed journal.

  5. (gandini2026crystallographicdatafor pages 6-9): Rosaria Gandini, Markus M. Keskitalo, Tom Reichenbach, Dayanand C. Kalyani, and Christina Divne. Crystallographic data for pyrococcus furiosus dolichylphosphate mannose synthase suggest that the enzyme could flip its glycolipid product. Scientific Reports, Mar 2026. URL: https://doi.org/10.1038/s41598-026-44343-5, doi:10.1038/s41598-026-44343-5. This article has 2 citations and is from a peer-reviewed journal.

  6. (packer2022functionalcharacterisationof pages 51-55): RD Packer. Functional characterisation of novel polyprenyl phosphomannose synthases for biocatalysis and medicine. Unknown journal, 2022.

  7. (packer2022functionalcharacterisationofa pages 51-55): RD Packer. Functional characterisation of novel polyprenyl phosphomannose synthases for biocatalysis and medicine. Unknown journal, 2022.

  8. (koff2023proteinomannosylationone pages 1-2): Melissa Koff, Pedro Monagas-Valentin, Boris Novikov, Ishita Chandel, and Vladislav Panin. Protein o-mannosylation: one sugar, several pathways, many functions. Glycobiology, 33:911-926, Aug 2023. URL: https://doi.org/10.1093/glycob/cwad067, doi:10.1093/glycob/cwad067. This article has 22 citations and is from a peer-reviewed journal.

  9. (kikul2023theinfluenceof pages 87-89): Frauke Kikul. The influence of membrane lipids on glycosylation processes: activity of dpm1 in different lipid environments. Text, Jan 2023. URL: https://doi.org/10.11588/heidok.00034034, doi:10.11588/heidok.00034034. This article has 0 citations and is from a peer-reviewed journal.

  10. (rathod2024dpm1modulatesdesmosomal pages 2-5): Maitreyi Rathod, Henriette Franz, Vivien Beyersdorfer, Marie-Therès Wanuske, Karen Leal-Fischer, Pauline Hanns, Chiara Stüdle, Aude Zimmermann, Katarzyna Buczak, Camilla Schinner, and Volker Spindler. Dpm1 modulates desmosomal adhesion and epidermal differentiation through serpinb5. The Journal of Cell Biology, Mar 2024. URL: https://doi.org/10.1083/jcb.202305006, doi:10.1083/jcb.202305006. This article has 10 citations.

  11. (rathod2024dpm1modulatesdesmosomal pages 23-26): Maitreyi Rathod, Henriette Franz, Vivien Beyersdorfer, Marie-Therès Wanuske, Karen Leal-Fischer, Pauline Hanns, Chiara Stüdle, Aude Zimmermann, Katarzyna Buczak, Camilla Schinner, and Volker Spindler. Dpm1 modulates desmosomal adhesion and epidermal differentiation through serpinb5. The Journal of Cell Biology, Mar 2024. URL: https://doi.org/10.1083/jcb.202305006, doi:10.1083/jcb.202305006. This article has 10 citations.

  12. (budhraja2024liposomeencapsulatedmannose1phosphatetherapy pages 13-15): Rohit Budhraja, Silvia Radenkovic, Anu Jain, Irena J.J. Muffels, Moulay Hicham Alaoui Ismaili, Tamas Kozicz, Akhilesh Pandey, and Eva Morava. Liposome-encapsulated mannose-1-phosphate therapy improves global n-glycosylation in different congenital disorders of glycosylation. Jun 2024. URL: https://doi.org/10.1016/j.ymgme.2024.108487, doi:10.1016/j.ymgme.2024.108487. This article has 20 citations and is from a peer-reviewed journal.

  13. (bananadube2018investigationofaa pages 71-74): P Banana-Dube. Investigation of a polyprenyl/dolichyl phosphate mannose synthase involved in bacterial protein mannosylation. Unknown journal, 2018.

  14. (banerjee2017dolicholphosphatemannose pages 8-9): Dipak K. Banerjee, Zhenbo Zhang, Krishna Baksi, and Jesús E. Serrano-Negrón. Dolichol phosphate mannose synthase: a glycosyltransferase with unity in molecular diversities. Glycoconjugate Journal, 34:467-479, Jun 2017. URL: https://doi.org/10.1007/s10719-017-9777-4, doi:10.1007/s10719-017-9777-4. This article has 25 citations and is from a peer-reviewed journal.

Artifacts

Citations

  1. banerjee2017dolicholphosphatemannose pages 1-3
  2. kikul2023theinfluenceof pages 87-89
  3. koff2023proteinomannosylationone pages 1-2
  4. haeuptle2009congenitaldisordersof pages 8-9
  5. haeuptle2009congenitaldisordersof pages 7-8
  6. maeda2008dolicholphosphatemannosesynthase pages 3-4
  7. gandini2026crystallographicdatafor pages 6-9
  8. packer2022functionalcharacterisationof pages 51-55
  9. packer2022functionalcharacterisationofa pages 51-55
  10. bananadube2018investigationofaa pages 71-74
  11. banerjee2017dolicholphosphatemannose pages 8-9
  12. https://doi.org/10.1093/glycob/cwad067
  13. https://doi.org/10.11588/heidok.00034034
  14. https://doi.org/10.1083/jcb.202305006
  15. https://doi.org/10.1016/j.ymgme.2024.108487
  16. https://doi.org/10.1002/humu.21126,
  17. https://doi.org/10.1016/j.bbagen.2008.03.005,
  18. https://doi.org/10.1007/s10719-017-9777-4,
  19. https://doi.org/10.1038/s41598-026-44343-5,
  20. https://doi.org/10.1093/glycob/cwad067,
  21. https://doi.org/10.11588/heidok.00034034,
  22. https://doi.org/10.1083/jcb.202305006,
  23. https://doi.org/10.1016/j.ymgme.2024.108487,