A systematic study of site-specific GalNAc-type O-glycosylation modulating proprotein convertase processing

Site-specific GalNAc-type O-glycosylation is emerging as an important co-regulator of proprotein convertase (PC) processing of proteins. PC processing is crucial in regulating many fundamental biological pathways and O-glycans in or immediately adjacent to processing sites may affect recognition and function of PCs. Thus, we previously demonstrated that deficiency in site-specific O-glycosylation in a PC site of the fibroblast growth factor, FGF23, resulted in marked reduction in secretion of active unprocessed FGF23, which cause familial tumoral calcinosis and hyperostosis hyperphosphatemia. GalNAc-type O-glycosylation is found on serine and threonine amino acids and up to 20 distinct polypeptide GalNAc transferases catalyze the first addition of GalNAc to proteins making this step the most complex and differentially regulated steps in protein glycosylation. There is no reliable prediction model for O-glycosylation especially of isolated sites, but serine and to a lesser extent threonine residues are frequently found adjacent to PC processing sites. In the present study we used in vitro enzyme assays and ex vivo cell models to systematically address the boundaries of the region within site-specific O-glycosylation affect PC processing. The results demonstrate that O-glycans within at least ±3 residues of the RXXR furin cleavage site may affect PC processing suggesting that site-specific O-glycosylation is a major co-regulator of PC processing.     

Building pathway clusters from Random Forests classification using class votes

Background

As one of the most common protein post-translational modifications, glycosylation is involved in a variety of important biological processes. Computational identification of glycosylation sites in protein sequences becomes increasingly important in the post-genomic era. A new encoding scheme was employed to improve the prediction of mucin-type O-glycosylation sites in mammalian proteins.

Results

A new protein bioinformatics tool, CKSAAP_OGlySite, was developed to predict mucin-type O-glycosylation serine/threonine (S/T) sites in mammalian proteins. Using the composition of k-spaced amino acid pairs (CKSAAP) based encoding scheme, the proposed method was trained and tested in a new and stringent O-glycosylation dataset with the assistance of Support Vector Machine (SVM). When the ratio of O-glycosylation to non-glycosylation sites in training datasets was set as 1:1, 10-fold cross-validation tests showed that the proposed method yielded a high accuracy of 83.1% and 81.4% in predicting O-glycosylated S and T sites, respectively. Based on the same datasets, CKSAAP_OGlySite resulted in a higher accuracy than the conventional binary encoding based method (about +5.0%). When trained and tested in 1:5 datasets, the CKSAAP encoding showed a more significant improvement than the binary encoding. We also merged the training datasets of S and T sites and integrated the prediction of S and T sites into one single predictor (i.e. S+T predictor). Either in 1:1 or 1:5 datasets, the performance of this S+T predictor was always slightly better than those predictors where S and T sites were independently predicted, suggesting that the molecular recognition of O-glycosylated S/T sites seems to be similar and the increase of the S+T predictor's accuracy may be a result of expanded training datasets. Moreover, CKSAAP_OGlySite was also shown to have better performance when benchmarked against two existing predictors.

Conclusion

Because of CKSAAP encoding's ability of reflecting characteristics of the sequences surrounding mucin-type O-glycosylation sites, CKSAAP_ OGlySite has been proved more powerful than the conventional binary encoding based method. This suggests that it can be used as a competitive mucin-type O-glycosylation site predictor to the biological community. CKSAAP_OGlySite is now available at http://bioinformatics.cau.edu.cn/zzd_lab/CKSAAP_OGlySite/.     

The location and characterisation of the O-linked glycans of the human insulin receptor

O-linked glycosylation is a post-translational and post-folding event involving exposed S/T residues at beta-turns or in regions with extended conformation. O-linked sites are difficult to predict from sequence analyses compared to N-linked sites. Here we compare the results of chemical analyses of isolated glycopeptides with the prediction using the neural network prediction method NetOGlyc3.1, a procedure that has been reported to correctly predict 76% of O-glycosylated residues in proteins. Using the heavily glycosylated human insulin receptor as the test protein six sites of mucin-type O-glycosylation were found at residues T744, T749, S757, S758, T759, and T763 compared to the three sites (T759 and T763- correctly, T756- incorrectly) predicted by the neural network method. These six sites occur in a 20 residue segment that begins nine residues downstream from the start of the insulin receptor beta-chain. This region which also includes N-linked glycosylation sites at N742 and N755, is predicted to lack secondary structure and is followed by residues 765-770, the known linear epitope for the monoclonal antibody 18-44.     

Database analysis of O-glycosylation sites in proteins

Statistical analysis was carried out to study the sequential aspects of amino acids around the O-glycosylated Ser/Thr. 992 sequences containing O-glycosylated Ser/Thr were selected from the O-GLYCBASE database of O-glycosylated proteins. The frequency of occurrence of amino acid residues around the glycosylated Ser/Thr revealed that there is an increased number of proline residues around the O-glycosylation sites in comparison with the nonglycosylated serine and threonine residues. The deviation parameter calculated as a measure of preferential and nonpreferential occurrence of amino acid residues around the glycosylation site shows that Pro has the maximum preference around the O-glycosylation site. Pro at +3 and/or -1 positions strongly favors glycosylation irrespective of single and multiple glycosylation sites. In addition, serine and threonine are preferred around the multiple glycosylation sites due to the effect of clusters of closely spaced glycosylated Ser/Thr. The preference of amino acids around the sites of mucin-type glycosylation is found likely to be similar to that of the O-glycosylation sites when taken together, but the acidic amino acids are more preferred around Ser/Thr in mucin-type glycosylation when compared totally. Aromatic amino acids hinder O-glycosylation in contrast to N-glycosylation. Cysteine and amino acids with bulky side chains inhibit O-glycosylation. The preference of certain potential sequence motifs of glycosylation has been discussed.     

Role of peptide sequence and neighboring residue glycosylation on the substrate specificity of the uridine 5'-diphosphate-alpha-N-acetylgalactosamine:polypeptide N-acetylgalactosaminyl transferases T1 and T2: kinetic modeling of the porcine and canine submaxillary gland mucin tandem repeats

A large family of uridine 5'-diphosphate (UDP)-alpha-N-acetylgalactosamine (GalNAc):polypeptide N-acetylgalactosaminyl transferases (ppGalNAc Ts) initiates mucin-type O-glycan biosynthesis at serine and threonine. The peptide substrate specificities of individual family members are not well characterized or understood, leaving an inability to rationally predict or comprehend sites of O-glycosylation. Recently, a kinetic modeling approach demonstrated neighboring residue glycosylation as a major factor modulating the O-glycosylation of the porcine submaxillary gland mucin 81 residue tandem repeat by ppGalNAc T1 and T2 [Gerken et al. (2002) J. Biol. Chem. 277, 49850-49862]. To confirm the general applicability of this model and its parameters, the ppGalNAc T1 and T2 glycosylation kinetics of the 80+ residue tandem repeat from the canine submaxillary gland mucin was obtained and characterized. To reproduce the glycosylation patterns of both mucins (comprising 50+ serine/threonine residues), specific effects of neighboring peptide sequence, in addition to the previously described effects of neighboring residue glycosylation, were required of the model. Differences in specificity of the two transferases were defined by their sensitivities to neighboring proline and nonglycosylated hydroxyamino acid residues, from which a ppGalNAc T2 motif was identified. Importantly, the model can approximate the previously reported ppGalNAc T2 glycosylation kinetics of the IgA1 hinge domain peptide [Iwasaki, et al. (2003) J. Biol. Chem. 278, 5613-5621], further validating both the approach and the ppGalNAc T2 positional weighting parameters. The characterization of ppGalNAc transferase specificity by this approach may prove useful for the search for isoform-specific substrates, the creation of isoform-specific inhibitors, and the prediction of mucin-type O-glycosylation sites.     

Basic amino acids as modulators of an O-linked glycosylation signal of the herpes simplex virus type 1 glycoprotein gC: functional roles in viral infectivity

The herpes simplex virus type 1 (HSV-1) glycoprotein gC-1 is engaged both in viral attachment and viral immune evasion mechanisms in the infected host. Besides several N-linked glycans, gC-1 contains numerous O-linked glycans, mainly localized in two pronase-resistant clusters in the N-terminal domain of gC-1. In the present study we construct and characterize one gC-1 mutant virus, in which two basic amino acids (114K and 117R) in a putative O-glycosylation sequon were changed to alanine. We found that this modification did not modify the N-linked glycosylation but increased the content of O-linked glycans considerably. Analysis of the O-glycosylation capacity of wild-type and mutant gC-1 was performed by in vitro glycosylation assays with synthetic peptides derived from the mutant region predicted to present new O-glycosylation sites. Thus the mutant peptide region served as a better substrate for polypeptide GalNAc-transferase 2 than the wild-type peptide, resulting in increased rate and number of O-glycan attachment sites. The predicted increase in O-linked glycosylation resulted in two modifications of the biological properties of mutant virus-that is, an impaired binding to cells expressing chondroitin sulfate but not heparan sulfate on the cell surface and a significantly reduced plaque size in cultured cells. The results suggested that basic amino acids present within O-glycosylation signals may down-regulate the amount of O-linked glycans attached to a protein and that substitution of such amino acid residues may have functional consequences for a viral glycoprotein involving virus attachment to permissive cells as well as viral cell-to-cell spread.     

Analysis of post-translational modification of mycobacterial proteins using a cassette expression system

A recombinant expression system was developed to analyse sequence determinants involved in O-glycosylation of proteins in mycobacteria. By expressing peptide sequences corresponding to known glycosylation sites within a chimeric lipoprotein construct, amino acids flanking modified threonine residues were found to have an important influence on glycosylation. The expression system was used to screen mycobacterial sequences selected using a neural network (NetOglyc) trained on eukaryotic O-glycoproteins. Evidence of glycosylation was obtained for eight of 11 proteins tested. The results suggest that sites involved in O-glycosylation of mycobacterial and eukaryotic proteins share similar structural features.     

Identification and Functional Characterization of Glycosylation of Recombinant Human Platelet-Derived Growth Factor-BB in Pichia pastoris

Yeast Pichia pastoris is a widely used system for heterologous protein expression. However, post-translational modifications, especially glycosylation, usually impede pharmaceutical application of recombinant proteins because of unexpected alterations in protein structure and function. The aim of this study was to identify glycosylation sites on recombinant human platelet-derived growth factor-BB (rhPDGF-BB) secreted by P. pastoris, and investigate possible effects of O-linked glycans on PDGF-BB functional activity. PDGF-BB secreted by P. pastoris is very heterogeneous and contains multiple isoforms. We demonstrated that PDGF-BB was O-glycosylated during the secretion process and detected putative O-glycosylation sites using glycosylation staining and immunoblotting. By site-directed mutagenesis and high-resolution LC/MS analysis, we, for the first time, identified two threonine residues at the C-terminus as the major O-glycosylation sites on rhPDGF-BB produced in P. pastoris. Although O-glycosylation resulted in heterogeneous protein expression, the removal of glycosylation sites did not affect rhPDGF-BB mitogenic activity. In addition, the unglycosylated PDGF-BB^ΔGly mutant exhibited the immunogenicity comparable to that of the wild-type form. Furthermore, antiserum against PDGF-BB^ΔGly also recognized glycosylated PDGF-BB, indicating that protein immunogenicity was unaltered by glycosylation. These findings elucidate the effect of glycosylation on PDGF-BB structure and biological activity, and can potentially contribute to the design and production of homogeneously expressed unglycosylated or human-type glycosylated PDGF-BB in P. pastoris for pharmaceutical applications.     

AMP-BERT: Prediction of antimicrobial peptide function based on a BERT model

Antimicrobial resistance is a growing health concern. Antimicrobial peptides (AMPs) disrupt harmful microorganisms by nonspecific mechanisms, making it difficult for microbes to develop resistance. Accordingly, they are promising alternatives to traditional antimicrobial drugs. In this study, we developed an improved AMP classification model, called AMP-BERT. We propose a deep learning model with a fine-tuned bidirectional encoder representations from transformers (BERT) architecture designed to extract structural/functional information from input peptides and identify each input as AMP or non-AMP. We compared the performance of our proposed model and other machine/deep learning-based methods. Our model, AMP-BERT, yielded the best prediction results among all models evaluated with our curated external dataset. In addition, we utilized the attention mechanism in BERT to implement an interpretable feature analysis and determine the specific residues in known AMPs that contribute to peptide structure and antimicrobial function. The results show that AMP-BERT can capture the structural properties of peptides for model learning, enabling the prediction of AMPs or non-AMPs from input sequences. AMP-BERT is expected to contribute to the identification of candidate AMPs for functional validation and drug development. The code and dataset for the fine-tuning of AMP-BERT is publicly available at https://github.com/GIST-CSBL/AMP-BERT .     

Analysis of O-glycosylation site occupancy in bovine kappa-casein glycoforms separated by two-dimensional gel electrophoresis

The ability of two-dimensional gel electrophoresis (2-DE) to separate glycoproteins was exploited to separate distinct glycoforms of kappa-casein that differed only in the number of O-glycans that were attached. To determine where the glycans were attached, the individual glycoforms were digested in-gel with pepsin and the released glycopeptides were identified from characteristic sugar ions in the tandem mass spectrometry (MS) spectra. The O-glycosylation sites were identified by tandem MS after replacement of the glycans with ammonia / aminoethanethiol. The results showed that glycans were not randomly distributed among the five potential glycosylation sites in kappa-casein. Rather, glycosylation of the monoglycoform could only be detected at a single site, T152. Similarly the diglycoform appeared to be modified exclusively at T152 and T163, while the triglycoform was modified at T152, T163 and T154. While low levels of glycosylation at other sites cannot be excluded the hierarchy of site occupation between glycoforms was clearly evident and argues for an ordered addition of glycans to the protein. Since all five potential O-glycosylation sites can be glycosylated in vivo, it would appear that certain sites remain latent until other sites are occupied. The determination of glycosylation site occupancy in individual glycoforms separated by 2-DE revealed a distinct pattern of in vivo glycosylation that has not been recognized previously.     

O-GLYCBASE: a revised database of O-glycosylated proteins.

O-GLYCBASE is a comprehensive database of information on glycoproteins and their O-linked glycosylation sites. Entries are compiled and revised from the SWISS-PROT and PIR databases as well as directly from recently published reports. Nineteen percent of the entries extracted from the databases needed revision with respect to O-linked glycosylation. Entries include information about species, sequence, glycosylation site and glycan type, and are fully referenced. Sequence logos displaying the acceptor specificity for the GaINAc transferase are shown. A neural network method for prediction of mucin type O-glycosylation sites in mammalian glycoproteins exclusively from the primary sequence is made available by E-mail or WWW. The O-GLYCBASE database is also available electronically through our WWW server or by anonymous FTP.     

The influence of flanking sequences on O-glycosylation

The influence of flanking sequences on O-glycosylation of serine and threonine residues was explored by comparison of known acceptor sites. Positions -6, -1 and +3 relative to the site were identified as particularly significant. To test the hypothesis that O-glycosylation could be affected by amino acid sequence, a series of test peptides was made containing substitutions at the sensitive positions. In vitro glycosylation of the peptides confirmed that the acceptor status of threonine was markedly influenced by the residues present at positions -6, -1 and +3. Circular dichroism indicated that peptides which had random structure were glycosylated, except when they contained a charged residue at position -1.     

Mucin core O-glycosylation is modulated by neighboring residue glycosylation status. Kinetic modeling of the site-specific glycosylation of the apo-porcine submaxillary mucin tandem repeat by UDP-GalNAc:polypeptide N-acetylgalactosaminyltransferases T1 and T2

The influence of peptide sequence and environment on the initiation and elongation of mucin O-glycosylation is not well understood. The in vivo glycosylation pattern of the porcine submaxillary gland mucin (PSM) tandem repeat containing 31 O-glycosylation sites (Gerken, T. A., Gilmore, M., and Zhang, J. (2002) J. Biol. Chem. 277, 7736-7751) reveals a weak inverse correlation with hydroxyamino acid density (and by inference the density of glycosylation) with the extent of GalNAc glycosylation and core-1 substitution. We now report the time course of the in vitro glycosylation of the apoPSM tandem repeat by recombinant UDP-GalNAc:polypeptide alpha-GalNAc transferases (ppGalNAc transferase) T1 and T2 that confirm these findings. A wide range of glycosylation rates are found, with several residues showing apparent plateaus in glycosylation. An adjustable kinetic model that reduces the first-order rate constants proportional to neighboring glycosylation status, plus or minus three residues of the site of glycosylation, was found to reasonably reproduce the experimental rate data for both transferases, including apparent plateaus in glycosylation. The unique, transferase-specific, positional weighting constants reveal information on the peptide/glycopeptide recognition site for each transferase. Both transferases displayed high sensitivities to neighboring Ser/Thr glycosylation, whereas ppGalNAc T2 displayed additional high sensitivities to the presence of nonglycosylated Ser/Thr residues. This is the first demonstration of the ability to model mucin O-glycosylation kinetics, confirming that under the appropriate conditions neighboring glycosylation status can be a significant factor modulating the first step of mucin O-glycan biosynthesis.     

A Strategy for O-Glycoproteomics of Enveloped Viruses—the O-Glycoproteome of Herpes Simplex Virus Type 1

Glycosylation of viral envelope proteins is important for infectivity and interaction with host immunity, however, our current knowledge of the functions of glycosylation is largely limited to N-glycosylation because it is difficult to predict and identify site-specific O-glycosylation. Here, we present a novel proteome-wide discovery strategy for O-glycosylation sites on viral envelope proteins using herpes simplex virus type 1 (HSV-1) as a model. We identified 74 O-linked glycosylation sites on 8 out of the 12 HSV-1 envelope proteins. Two of the identified glycosites found in glycoprotein B were previously implicated in virus attachment to immune cells. We show that HSV-1 infection distorts the secretory pathway and that infected cells accumulate glycoproteins with truncated O-glycans, nonetheless retaining the ability to elongate most of the surface glycans. With the use of precise gene editing, we further demonstrate that elongated O-glycans are essential for HSV-1 in human HaCaT keratinocytes, where HSV-1 produced markedly lower viral titers in HaCaT with abrogated O-glycans compared to the isogenic counterpart with normal O-glycans. The roles of O-linked glycosylation for viral entry, formation, secretion, and immune recognition are poorly understood, and the O-glycoproteomics strategy presented here now opens for unbiased discovery on all enveloped viruses.     

Structural requirements for additional N-linked carbohydrate on recombinant human erythropoietin

N-Linked glycosylation is a post-translational event whereby carbohydrates are added to secreted proteins at the consensus sequence Asn-Xaa-Ser/Thr, where Xaa is any amino acid except proline. Some consensus sequences in secreted proteins are not glycosylated, indicating that consensus sequences are necessary but not sufficient for glycosylation. In order to understand the structural rules for N-linked glycosylation, we introduced N-linked consensus sequences by site-directed mutagenesis into the polypeptide chain of the recombinant human erythropoietin molecule. Some regions of the polypeptide chain supported N-linked glycosylation more effectively than others. N-Linked glycosylation was inhibited by an adjacent proline suggesting that sequence context of a consensus sequence could affect glycosylation. One N-linked consensus sequence (Asn123-Thr125) introduced into a position close to the existing O-glycosylation site (Ser126) had an additional O-linked carbohydrate chain and not an additional N-linked carbohydrate chain suggesting that structural requirements in this region favored O-glycosylation over N-glycosylation. The presence of a consensus sequence on the protein surface of the folded molecule did not appear to be a prerequisite for oligosaccharide addition. However, it was noted that recombinant human erythropoietin analogs that were hyperglycosylated at sites that were normally buried had altered protein structures. This suggests that carbohydrate addition precedes polypeptide folding.     

Alterations in protein secretion caused by metabolic engineering of glycosylation pathways in fungi

Due to its natural properties, Trichoderma reesei is commonly used in industry-scale production of secretory proteins. Since almost all secreted proteins are O-glycosylated, modulation of the activity of enzymes of the O-glycosylation pathway are likely to affect protein production and secretion or change the glycosylation pattern of the secreted proteins, altering their stability and biological activity. Understanding how the activation of different components of the O-glycosylation pathway influences the glycosylation pattern of proteins and their production and secretion could help in elucidating the mechanism of the regulation of these processes and should facilitate creation of engineered microorganisms producing high amounts of useful proteins. In this review we focus on data concerning Trichoderma, but also present some background information allowing comparison with other fungal species.     

Differential O-glycosylation of a conserved domain expressed in murine and human ZP3

Murine sperm initiate fertilization by binding to the zona pellucida (mZP), the specialized extracellular matrix of their homologous eggs. O-Glycans occupying two highly conserved vicinal glycosylation sites (Ser-332 and Ser-334) on the mZP glycoprotein designated mZP3 were previously implicated in this interaction. However, recent biophysical analyses confirm that neither site is occupied, implying that an alternate O-glycosylation domain may be operational in native mZP3. Since human ZP3 (huZP3) can substitute for mZP3 in rescue mice to mediate sperm binding, the site specificity of O-glycosylation in both native mZP3 and huZP3 was analyzed using ultrasensitive mass spectrometric techniques. Two O-glycosylation sites in native mZP3, one at Thr-155 and the other within the glycopeptide at positions 161-168 (ATVSSEEK), are conserved in huZP3 derived from transgenic mice. Thus, there is a specific O-glycosylation domain within native mZP3 expressing two closely spaced O-glycans that is very well conserved in an evolutionarily related glycoprotein. In native mZP3, core 2 O-glycans predominate at both sites. However, in huZP3 derived from rescue mice, the O-glycans associated with Thr-156 (analogous to Thr-155 in mZP3) are exclusively core 1 and related Tn sequences, whereas core 2 O-glycans predominate at the other conserved site. This unique restriction of O-glycan expression suggests that sequence differences in the conserved O-glycosylation domains of mZP3 and huZP3 affect the ability of core 2 N-acetylglucosaminyltransferase(s) to extend the core 1 sequence. However, this difference in O-glycosylation at Thr-156 does not affect the fertility or the sperm binding phenotype of eggs derived from female huZP3 rescue mice.     

An inhibitor of O-glycosylation induces apoptosis in NIH3T3 cells and developing mouse embryonic mandibular tissues

The family of UDP-GalNAc:polypeptide N-acetylgalactosaminyltransferases (ppGaNTases) is responsible for initiating mucin-type O-linked glycosylation in higher eukaryotes. To begin to examine the biological role of O-linked glycosylation, mammalian cells were treated with a small molecule inhibitor (designated 1-68A, Ref. 15) of ppGaNTase activity. NIH3T3 cells exposed to the inhibitor were shown to undergo a significant reduction in cell surface O-glycosylation as detected by staining with jacalin and peanut agglutinin lectins after 30 min of treatment; no reduction in staining using antibodies to O-linked N-acetylglucosamine or the lectin concanavalin A was detected. Apoptosis was also observed in treated cells after 45 min of exposure, ostensibly following the O-glycosylation reduction. Overexpression of several different ppGaNTase isoforms restored cell surface O-glycosylation and rescued inhibitor-induced apoptosis. Additionally, mouse embryonic mandibular organ cultures exposed to 1-68A developed abnormally, presumably because of epithelial and mesenchymal apoptosis that followed a reduction in jacalin and peanut agglutinin staining. Our studies suggest that mucin-type O-linked glycosylation may be required for normal development and that ppGaNTases may play a role in the regulation of apoptosis.     

The AATPAP sequence is a very efficient signal for O-glycosylation in CHO cells

The peptide signal sequence for protein O-glycosylation is not fully characterized, although a recent in vitro study proposed that the sequence motif, XTPXP, serves as a signal for mucin-type O-glycosylation. Here, we show that the AATPAP sequence acts as an efficient O-glycosylation signal, in vivo. A secreted fibroblast growth factor (secFGF) was used as a model to analyze glycosylation and its effects on the biological activity of FGF. Two constructs encoding [AATPAP]secFGF in which AATPAP was introduced at the N- or C-terminus of secFGF were constructed in a eukaryotic expression vector. [AATPAP]secFGF proteins were then expressed in Chinese hamster ovary (CHO) cells and secreted into the surrounding medium, primarily as modified forms sensitive to sialidase but not to peptide N-glycosidase F. The modifying groups were not seen when the AATPAP sequence was converted to AAAPAP or when [AATPAP]secFGF was expressed in mutant cells incapable of UDP-GalNAc biosynthesis. The results indicate that the modifying groups were mucin-type O-glycans and that the AATPAP served as an efficient O-glycosylation signal sequence. The O-glycosylated forms of [AATPAP]secFGF were as mitogenic toward human vascular endothelial cells as unmodified secFGF, suggesting that introduction of the signal into biologically active polypeptides is a promising approach with which O-glycosylation may be achieved without affecting original activity.     

Versatile characterization of glycosylation modification in CTLA4-Ig fusion proteins by liquid chromatography-mass spectrometry

CTLA4-Ig is a highly glycosylated therapeutic fusion protein that contains multiple N- and O-glycosylation sites. Glycosylation plays a vital role in protein solubility, stability, serum half-life, activity, and immunogenicity. For a CTLA4-Ig biosimilar development program, comparative analytical data, especially the glycosylation data, can influence decisions about the type and amount of animal and clinical data needed to establish biosimilarity. Because of the limited clinical experience with biosimilars before approval, a comprehensive level of knowledge about the biosimilar candidates is needed to achieve subsequent development. Liquid chromatography-mass spectrometry (LC–MS) is a versatile technique for characterizing N- and O-glycosylation modification of recombinant therapeutic proteins, including 3 levels: intact protein analysis, peptide mapping analysis, and released glycans analysis. In this report, an in-depth characterization of glycosylation of a candidate biosimilar was carried out using a systematic approach: N- and O-linked glycans were identified and electron-transfer dissociation was then used to pinpoint the 4 occupied O-glycosylation sites for the first time. As the results show, the approach provides a set of routine tools that combine accurate intact mass measurement, peptide mapping, and released glycan profiling. This approach can be used to comprehensively research a candidate biosimilar Fc-fusion protein and provides a basis for future studies addressing the similarity of CTLA4-Ig biosimilars.