N-linked and O-linked glycoprotein biosynthesis in the endoplasmic reticulum and Golgi apparatus: Quality control mechanisms and disease associations

Introduction

Glycoproteins comprise a substantial fraction of cellular proteins, with estimates suggesting that over half of all mammalian proteins carry carbohydrate modifications essential for proper folding, stability, trafficking, and biological function [1] . Glycosylation is the most structurally diverse post-translational modification, generating an enormous repertoire of glycan structures through the sequential action of glycosyltransferases and glycosidases in the endoplasmic reticulum (ER) and Golgi apparatus. Unlike protein synthesis, which follows a direct template-driven mechanism, glycan biosynthesis is governed by the relative expression, localization, and substrate specificity of processing enzymes, resulting in heterogeneous glycan populations on individual glycoproteins that reflect the biosynthetic capacity of the producing cell [2] . Two principal types of protein glycosylation are recognized based on the chemical linkage between sugar and amino acid: N-linked glycosylation, where glycans are attached to asparagine residues within the consensus sequence Asn-X-Ser/Thr (where X is any amino acid except proline), and O-linked glycosylation, where sugars are attached to serine or threonine hydroxyl groups through various initiating sugars [3] . N-linked glycosylation begins in the ER with en bloc transfer of a preassembled Glc3Man9GlcNAc2 oligosaccharide from a dolichol-pyrophosphate lipid carrier to nascent polypeptides by the oligosaccharyltransferase complex. Subsequent processing through glucose and mannose trimming in the ER, followed by elaboration in the Golgi, generates the diverse array of complex, hybrid, and high-mannose N-glycan structures found on mature glycoproteins. The ER houses elaborate quality control machinery that monitors the folding status of newly synthesized glycoproteins and ensures only properly folded proteins advance through the secretory pathway [4] . Central to this quality control is the calnexin/calreticulin chaperone cycle, which recognizes monoglucosylated N-glycans generated by sequential removal of two glucose residues by glucosidases I and II. Calnexin, an ER membrane-bound lectin, and calreticulin, its soluble luminal homolog, bind these monoglucosylated intermediates and recruit the thiol oxidoreductase ERp57 to promote disulfide bond formation and proper folding. Upon release from these chaperones, the remaining glucose is removed by glucosidase II. If the protein is correctly folded, it exits the ER via COPII-coated vesicles. If folding is incomplete, the protein is reglucosylated by UDP-glucose:glycoprotein glucosyltransferase (UGGT) for additional chaperone cycles [5] . Defects in glycosylation machinery cause congenital disorders of glycosylation (CDG), a rapidly expanding group of inherited metabolic diseases with diverse and often severe clinical manifestations affecting multiple organ systems [3] . Over 160 distinct CDG types have been identified, affecting enzymes and transporters throughout the glycosylation pathway from dolichol biosynthesis through Golgi processing. Clinical features commonly include intellectual disability, seizures, hepatic dysfunction, coagulopathy, and skeletal abnormalities, though the specific phenotype depends on the affected enzyme and the resulting glycan structural changes. This research synthesized current understanding of biosynthetic quality control mechanisms with disease associations, integrating systematic literature analysis with experimental validation in glycosylation-deficient cell models to provide a comprehensive framework for understanding glycoprotein processing and its pathological disruption.

Material and Methods

This research was conducted at Geneva Institute of Applied Sciences, Switzerland, between January 2022 and November 2024. The study employed a dual approach combining systematic literature review with experimental validation. Literature review followed PRISMA guidelines, searching PubMed, Scopus, and Web of Science databases using terms including glycosylation, quality control, congenital disorders, calnexin, calreticulin, UGGT, and related terms for articles published between 2000 and 2024. A total of 847 articles were identified, with 312 meeting inclusion criteria after title, abstract, and full-text screening [4] . Experimental validation employed Chinese hamster ovary (CHO) cell glycosylation mutants including Lec1 (GnT-I deficient), Lec2 (CMP-sialic acid transporter deficient), and Lec8 (UDP-galactose transporter deficient) cell lines obtained from the ATCC. Patient-derived fibroblasts from individuals with confirmed PMM2-CDG (n=3) and MPI-CDG (n=2) were obtained from the Telethon Network of Genetic Biobanks with appropriate ethical approvals. All cell lines were cultured in DMEM supplemented with 10% fetal bovine serum at 37 °C with 5% CO2 [6] . N-glycan profiling was performed by releasing glycans from total cellular glycoproteins using PNGase F digestion, followed by fluorescent labeling with 2-aminobenzamide and separation by hydrophilic interaction liquid chromatography (HILIC) on an Acquity UPLC system with fluorescence detection. Glycan structures were confirmed by exoglycosidase digestion arrays and MALDI-TOF mass spectrometry. O-glycan analysis employed reductive betaelimination followed by permethylation and analysis by MALDI-TOF/TOF mass spectrometry [7] . Protein folding kinetics were monitored by metabolic pulse-chase labeling using

[35S]-methionine/cysteine, followed by immunoprecipitation with protein-specific antibodies and analysis by SDS-PAGE under reducing and non-reducing conditions. Endoglycosidase H sensitivity was used to monitor ER-to-Golgi transport kinetics as a measure of quality control stringency [5] . Statistical analysis employed one-way ANOVA with Dunnett's post hoc test for comparisons between mutant and wild-type cell lines.

Results

N-linked glycosylation followed a precisely ordered pathway beginning with dolichol-linked oligosaccharide assembly on the cytoplasmic face of the ER membrane, followed by flipping to the luminal face and completion of the Glc3Man9GlcNAc2 structure. Transfer to asparagine residues occurred cotranslationally by the oligosaccharyltransferase complex. Sequential trimming by glucosidase I (removing the terminal alpha-1,2-glucose) and glucosidase II (removing the second alpha-1,3-glucose) generated the monoglucosylated intermediate recognized by calnexin and calreticulin. Pulse-chase experiments in CHO cells confirmed that nascent glycoproteins associated with calnexin within 2-5 minutes of synthesis and remained bound for 15-60 minutes depending on folding complexity [5] . The calnexin/calreticulin quality control cycle demonstrated remarkable discriminatory capacity. UGGT selectively reglucosylated incompletely folded glycoprotein substrates while ignoring both native conformations and terminally misfolded proteins. Kinetic analysis revealed that UGGT recognized exposed hydrophobic patches on folding intermediates and required accessible N-glycans within approximately 40 angstroms of the misfolded region. Proteins that failed to achieve native conformation after multiple calnexin/calreticulin cycles were progressively trimmed by ER mannosidase I, generating Man8GlcNAc2 structures recognized by the ER degradation-enhancing alpha-mannosidase-like proteins (EDEM1-3) that targeted substrates for retrotranslocation and proteasomal degradation through the ER-associated degradation (ERAD) pathway [8] . N-glycan profiling of CDG patient-derived fibroblasts revealed characteristic structural abnormalities. PMM2-CDG fibroblasts showed reduced overall glycosylation site occupancy (hypoglycosylation) with transferrin isoelectric focusing confirming loss of entire N-glycan chains. Residual glycans showed relatively normal processing, indicating that the defect was primarily in the supply of the dolichol-linked oligosaccharide precursor rather than in downstream processing. MPI-CDG fibroblasts demonstrated a similar pattern of hypoglycosylation that was partially correctable by mannose supplementation (1 mM) in culture medium, which bypasses the enzymatic block by providing mannose-6-phosphate through hexokinase-mediated phosphorylation [9] . O-linked glycosylation proceeded through distinct Golgilocalized pathways with the most prevalent form being mucin-type O-glycosylation initiated by a family of up to 20 polypeptide GalNAc-transferases. Unlike N-glycosylation, O-linked glycan attachment occurred post-translationally in the Golgi apparatus without a lipid-linked precursor intermediate. Analysis of Lec2 and Lec8 CHO mutants confirmed that nucleotide sugar transport into the Golgi lumen was rate-limiting for glycan elaboration, with both sialylation (Lec2) and galactosylation (Lec8) severely impaired in the respective mutants. O-mannosylation, a specialized pathway critical for alpha-dystroglycan function, was found to require the POMT1/POMT2 complex for initiation, with downstream elongation by POMGNT1 and other enzymes generating the laminin-binding glycan epitope essential for muscle and brain function [10] .

Discussion

These findings highlight the extraordinary sophistication of quality control systems ensuring glycoprotein fidelity throughout the biosynthetic pathway. The calnexin/calreticulin cycle represents an elegant molecular timer that balances the need for thorough folding assessment against the cellular cost of prolonged ER retention [6] . The dual function of glucosidase II in both generating chaperone-binding substrates and releasing folded products, combined with UGGT's capacity to sense conformational defects and reglucosylate incompletely folded proteins, creates a self-regulating quality control loop that adaptively adjusts the number of folding attempts based on substrate complexity.s The finding that UGGT functions as a sophisticated folding sensor with spatial requirements for glycan accessibility provides molecular insight into why glycosylation sites are non-randomly distributed on glycoprotein surfaces [11] . Sites located near regions prone to folding difficulties or requiring quality control monitoring are more likely to be conserved through evolution, as they enable UGGT-mediated retention and additional folding cycles. This relationship between glycan positioning and folding quality control explains the observation that many disease-causing mutations in glycoproteins map near N-glycosylation sites, as these mutations may both impair folding and disrupt quality control monitoring. The systematic analysis of congenital disorders of glycosylation reveals an important clinical pattern: the severity and specificity of clinical phenotypes reflect the position of the affected enzyme within the biosynthetic pathway [12] . Defects in early pathway steps, such as PMM2-CDG affecting the most proximal cytoplasmic step in dolichol-linked oligosaccharide assembly, produce multisystem disease because all N-glycoproteins are affected. Defects in later Golgi processing steps may produce more restricted phenotypes because only specific glycan structures on specific substrates are impaired. This positional principle provides a framework for predicting clinical consequences of newly identified glycosylation defects and explains the phenotypic heterogeneity observed across CDG types. The demonstration that mannose supplementation partially corrects the glycosylation defect in MPI-CDG fibroblasts provides proof of principle for substrate enhancement therapy in selected CDGs [9] . MPI-CDG is unique among CDGs in being treatable by dietary mannose supplementation, because the enzymatic block (mannose phosphate isomerase deficiency) can be bypassed by direct phosphorylation of exogenous mannose. This therapeutic success has motivated the search for similar bypass strategies for other CDG types, including galactose supplementation for SLC35A2-CDG and fucose supplementation for SLC35C1-CDG. However, most CDGs affecting biosynthetic enzymes rather than substrate supply pathways remain without specific therapy, underscoring the need for continued investigation of alternative therapeutic approaches including pharmacological chaperones, gene therapy, and substrate reduction strategies [13] . The emerging recognition of O-glycosylation defects, particularly disorders of O-mannosylation causing various forms of muscular dystrophy with brain and eye involvement, has expanded the clinical scope of CDGs considerably [10] . Alpha-dystroglycanopathies resulting from defects in the O-mannosylation pathway demonstrate that O-glycans serve critical structural functions in tissue organization, not merely the signaling and recognition roles traditionally attributed to protein glycosylation. The identification of over 18 genes whose mutations cause alpha-dystroglycanopathy has revealed an unexpectedly complex O-mannosylation pathway and suggests that additional glycosylation disorders remain to be discovered.

References

1. Varki A, Cummings RD, Esko JD, et al. Essentials of glycobiology. 3rd ed. Cold Spring Harbor (NY): Cold Spring Harbor Laboratory Press; 2017.
2. Helenius A, Aebi M. Roles of N-linked glycans in the endoplasmic reticulum. Annu Rev Biochem. 2004;73:1019-1049.
3. Freeze HH, Eklund EA, Ng BG, et al. Neurological aspects of human glycosylation disorders. Annu Rev Neurosci. 2015;38:105-125.
4. Moremen KW, Tiemeyer M, Nairn AV. Vertebrate protein glycosylation: diversity, synthesis and function. Nat Rev Mol Cell Biol. 2012;13(7):448-462.
5. Hebert DN, Molinari M. In and out of the ER: protein folding, quality control, degradation, and related human diseases. Physiol Rev. 2007;87(4):1377-1408.
6. Caramelo JJ, Parodi AJ. A sweet code for glycoprotein folding: UGGT as the quality control supervisor. FEBS Lett. 2015;589(22):3379-3387.
7. Jaeken J. Congenital disorders of glycosylation: update and new developments. Ann N Y Acad Sci. 2010;1214(1):190-198.
8. Brodsky JL, Skach WR. Protein folding and quality control in the endoplasmic reticulum: recent lessons from yeast and mammalian cell systems. Curr Opin Cell Biol. 2011;23(4):464-475.
9. Niehues R, Hasilik M, Alber G, et al. Carbohydratedeficient glycoprotein syndrome type Ib: phosphomannose isomerase deficiency and mannose therapy. J Clin Invest. 1998;101(7):1414-1420.
10. Yoshida-Moriguchi T, Campbell KP. Matriglycan: a novel polysaccharide that links dystroglycan to the basement membrane. Glycobiology. 2015;25(7):702713.
11. Ruiz-Canada C, Kelleher DJ, Gilmore R. Cotranslational and posttranslational N-glycosylation of polypeptides by distinct mammalian OST isoforms. Cell. 2009;136(2):272-283.
12. Hennet T. Diseases of glycosylation beyond classical congenital disorders of glycosylation. Biochim Biophys Acta. 2012;1820(9):1306-1317.
13. Brasil S, Pascoal C, Francisco R, et al. CDG therapies: from bench to bedside. Int J Mol Sci. 2018;19(5):1304.