Elucidating the Implications of Norovirus N- and O-Glycosylation, O-GlcNAcylation, and Phosphorylation

Norovirus is the most common cause of foodborne gastroenteritis, affecting millions of people worldwide annually. Among the ten genotypes (GI–GX) of norovirus, only GI, GII, GIV, GVIII, and GIX infect humans. Some genotypes reportedly exhibit post-translational modifications (PTMs), including N- and O-glycosylation, O-GlcNAcylation, and phosphorylation, in their viral antigens. PTMs have been linked to increased viral genome replication, viral particle release, and virulence. Owing to breakthroughs in mass spectrometry (MS) technologies, more PTMs have been discovered in recent years and have contributed significantly to preventing and treating infectious diseases. However, the mechanisms by which PTMs act on noroviruses remain poorly understood. In this section, we outline the current knowledge of the three common types of PTM and investigate their impact on norovirus pathogenesis. Moreover, we summarize the strategies and techniques for the identification of PTMs.


Introduction
The most prevalent causes of foodborne outbreaks are noroviruses, which account for approximately 50% of all occurrences worldwide [1]. They cause 20% of all gastrointestinal illnesses worldwide, resulting in 200,000 deaths and 700 million infections annually [2,3]. Norovirus infections are characterized by emesis, acute watery diarrhea, nausea, low-grade fever, and abdominal cramps. Norovirus infections are usually transient, with symptoms vanishing within 12-72 h [4]. Although norovirus-associated disease is usually self-limiting, exposure to norovirus leaves neonates, the elderly, and immunocompromised patients vulnerable to chronic severe or life-threatening symptoms [5,6]. Norovirus infections are common among people of all ages and cause substantial health and economic burdens in developed and developing countries [6].
Noroviruses are single-stranded positive-sense ribonucleic acid (RNA) viruses that belong to the Caliciviridae family. Genomic RNA is covalently coupled to a viral protein (VPg) at the 5 end and polyadenylated at the 3 end [7]. Most norovirus genomes are structured into three open reading frames (ORFs), whereas murine noroviruses have four ORFs [8]. ORF1 encodes nonstructural proteins NS1/2 to NS7. Among these, NS7, an RNA-dependent RNA polymerase (RdRp), plays a crucial role in genome replication. ORF2 encodes the major capsid (VP1), which has a shell (S) and protruding (P) domain. The

Effects of Post-Translational Modifications on Protein Function
Protein PTMs provide crucial insights into various cellular functions [23]. PTMs are typically formed by enzymatic processes that add functional groups to the side chains of amino acids [24]. These modifications are reversible and essential for biological functions. Over 620 modifications have been discovered that occur after protein synthesis [25]. Nand O-glycosylation, O-GlcNAcylation, and phosphorylation are the most common modifications that increase protein solubility, conformation, interactions, signaling, and degradation, which are all critical for cell growth [26]. Nand O-glycosylation facilitates receptor binding and alters the structure of the secreted proteins [27,28]. In contrast, O-GlcNAcylation and phosphorylation are competitive processes implicated in many signal transduction pathways [29,30].

Effects of Post-Translational Modifications on Viruses
Many pathogens, including viruses and bacteria, can utilize post-translational modifications to enhance interactions with host proteins crucial to infection ( Table 1). As viruses rely on the protein synthesis machinery of host cells to support their replication, most viral proteins are subjected to PTMs [31]. This improves viral replication, assembly, release, and immune escape during infection, thereby promoting virus propagation. Furthermore, PTMs improve solubility and antigenicity, which enhances virulence [29]. For example, phosphorylation of the dengue virus (DENV) type 2 regulates interactions between viral replication proteins [22]. In this review, we elucidate the effects of PTMs on noroviruses. Here, we focus on the mechanism of Nand O-glycosylation, O-GlcNAcylation, and phosphorylation and discuss the mechanisms by which these modifications affect norovirus pathogenesis.

Identification of Post-Translational Modifications
Previously, studies on PTMs were limited owing to the requirement for laborious biochemical approaches, including radioactive-isotope-labeled substrates, antibody-based Western blot analysis, and peptide and protein arrays [57][58][59]. However, these procedures are inefficient owing to the difficulty in identifying modified proteins using their corresponding weakly radioactive-isotope-labeled substrates. Furthermore, creating antibodies that recognize the minor structural motifs of particular PTMs using antibody-based Western blot analysis is also challenging [59]. Over the last decade, mass spectrometry (MS) has been demonstrated to be a powerful technique for identifying modified proteins and mapping PTM locations. Sites containing Nand O-glycosylation, O-GlcNAcylation, and phosphorylation modifications were enriched and successfully identified using liquid  [59][60][61]. In addition, enzymes or inhibitors may be used to investigate PTMs (Table 1). Sections 3.1, 4.1 and 5.1 in this article comprehensively describe these mechanisms.

N-and O-Glycosylation on Proteins
Glycosylation is a multienzymatic process that produces various glycoconjugates covalently bound to lipids or proteins [62]. Glycosylation is classified as N-linked or Olinked [30] (Figure 1). The first step in N-linked glycosylation is the attachment of an N-linked glycan to the asparagine residue of a nascent polypeptide chain by oligosaccharyltransferase (OST) in the endoplasmic reticulum (ER). Other enzymes subsequently construct glycans, resulting in a diverse spectrum of glycan structures such as oligomannose, hybrid, and complex-type N-glycan structures [30,45,62]. In O-linked glycosylation, N-acetylgalactosamine (GalNAc) is covalently linked to the hydroxyl group of serine or threonine residues in the Golgi apparatus [45,63]. This process is mediated by 20 different GalNAc transferases, each of which may produce unique mucin-type O-glycan core structures [45]. Nand O-glycosylation can alter protein characteristics, such as stability, solubility, protease resistance, and biological activity [64]. For example, glycans can be structurally integrated into the protein fold and exhibit significant glycan-protein interactions to stabilize the protein [30].
Traditional N-glycosylation detection methods include mutagenesis of anticipated glycosylation sites and enzymes to cleave glycans from protein substrates, which aids in distinguishing between terminally and core-glycosylated N-glycans. The peptide-Nglycosidase F (PNGase F) specifically cleaves the linkage between the innermost GlcNAc and asparagine. In addition, endoglycosidase H cleaves within the chitobiose core of high mannose and some hybrid oligosaccharides from N-linked glycoproteins. The molecular weight and functionalities of an N-glyco protein can be altered after incubation with these N-glycan removal enzymes (Table 1). Tunicamycin, for example, acts as an analog of uridine diphosphate N-acetylglucosamine (UDP-GlcNAc) to inhibit dolichol phosphatedependent N-acetylglucosamine 1-phospho-transferase (DPAGT1), thus preventing the first step in N-glycoprotein biosynthesis [65]. In addition to the approaches stated above, gel electrophoresis and immunoblotting with antibodies are practical since PTM with glycan moieties changes the electrophoretic mobility of the protein [63].
Enrichment and MS technologies are used in modern strategies for detecting glycoproteins. Most N-glycoprotein enrichment procedures are based on hydrazide chemistry, which involves oxidation of the carbohydrate side chain and conjugating glycopeptides to hydrazide resin [66]. The isolated glycopeptides are then released using a glycan-specific enzyme, such as peptide-N-glycosidase F (PNGase F), followed by MS identification, thereby facilitating a comprehensive analysis of the N-glycosylated proteome [59,63,67]. Using the GlycoStore database, approximately 850 unique glycan structures of glycoproteins and glycolipids can be determined [68]. Alternatively, the use of data-independent collection mode mass spectrometry (MS E ) and ProteinLynx Global Server (PLGS) software (Waters Corporation, Milford, MA, USA) may help identify short glycopeptides [28,69]. As for the identification of complicated N-glycans, PNGase F can liberate glycans and subsequently label them with the fluorophore 2-aminobenzoic acid (2-AA). The 2-AA tagged glycans can be purified by solid-phase extraction and detected using a fluorescence detector or mass spectrometer [70].

N-and O-Glycosylation on Viruses
Flaviviruses, severe acute respiratory syndrome-associated coronavirus (SARS-CoV2), influenza viruses, and rotaviruses exhibit viral protein Nand O-glycosylation (Table 1), which aids in viral entry, assembly, transmission potential, virulence, and pathogenicity [30]. RNA viruses manufacture their envelopes and surface glycoproteins using the host ER/Golgi system. Additionally, N-linked glycans can facilitate the folding and trafficking of viral glycoproteins via host ER quality control [71]. Viruses are often highly glycosylated on their surfaces, which increases the attachment of viral proteins to cells and facilitates infection [64]. Furthermore, they can mask or modify antibody-mediated recognition of antigenic epitopes, helping them evade the immune system of the host [64,71].
For example, glycosylation of the DENV NS1 aids in protein secretion by forming hexamers that bind to lectin pathway proteins such as C1s, C4, C4b-binding protein (C4BP), and mannose-binding lectin (MBL). This modification assists immune evasion by limiting lectin complement activation and DENV neutralization, controlling pathogenesis, and contributing to virulence [30,36,72]. In SARS-CoV-2, the spike protein, as well as M and E proteins, are glycosylated and responsible for membrane fusion, invasion, and immune escape [35,43]. The spike protein is attached by N-glycan, which facilitates its entry into the host cells and protects the epitopes to evade the immune response [43,44]. N-linked and O-linked glycosylation of the M protein facilitates viral particle assembly and budding. The E protein is involved in many viral processes, including membrane construction and interactions with other membrane proteins, and has two glycosylation sites, N48 and N66 [35].

Glycosylation on Noroviruses
HBGA is essential for norovirus infection. Fucosyltransferase 2 (FUT2), an enzyme that catalyzes 1,2-fucosylation of terminal galactose, regulates the production of HBGAs in intestinal epithelial cells. Individuals who lack FUT2s do not express HBGAs on their epithelial cells, rendering them exceptionally resistant to the gastroenteritis caused by certain norovirus strains, such as GII genotype 4 ( Figure 1) [75][76][77]. Despite the fact that HBGA glycosylation significantly alters binding affinity, little is revealed about the Nand O-glycosylation of norovirus capsid protein VP1, which merits additional exploration. On the other hand, the deamidation of Asn373 and the formation of isoD373 on the norovirus capsid protein VP1 impair its recognition of HBGAs [11] (Figure 1). Asn373 is found in the antigenic loop next to the HBGA binding site. Asn373 interacts with the glycan ligand through two direct hydrogen bonds; when converted to isoD373, only one hydrogen bond remains [78]. In addition, the peptides that contain isoD373 in the P dimmer domain do not show elevated flexibility [79]. Thus, the formation of isoD373 decreases the binding affinity of the P protein for HBGAs.

O-GlcNAcylation on Proteins
O-GlcNAcylation is a type of noncanonical glycosylation whereby O-linked N-acetylglucosamine (O-GlcNAc) is coupled to the hydroxyl groups of serine or threonine residues in proteins [82,83]. The hexosamine biosynthetic pathway, which incorporates glucose, amino acids, fatty acids, and nucleotide metabolism, produces the donor sugar for O-GlcNAcylation, UDP-GlcNAc. O-GlcNAc transferase (OGT) and O-GlcNAcase (OGA) catalyze the addition and removal of O-GlcNAc, respectively [84] (Figure 2). These two enzymes are found in all multicellular organisms and are substantially conserved from worms to humans [85]. In contrast to glycosylation, which is stable and localizes mainly at the ER and Golgi apparatus, O-GlcNAcylation is reversible and occurs in the cytoplasm [82]. O-GlcNAcylation has been implicated in several biological activities, including transcription, translation, metabolism, signal transmission, and apoptosis [84].

O-GlcNAcylation on Proteins
O-GlcNAcylation is a type of noncanonical glycosylation whereby O-linked N-acetylglucosamine (O-GlcNAc) is coupled to the hydroxyl groups of serine or threonine residues in proteins [82,83]. The hexosamine biosynthetic pathway, which incorporates glucose, amino acids, fatty acids, and nucleotide metabolism, produces the donor sugar for O-GlcNAcylation, UDP-GlcNAc. O-GlcNAc transferase (OGT) and O-GlcNAcase (OGA) catalyze the addition and removal of O-GlcNAc, respectively [84] (Figure 2). These two enzymes are found in all multicellular organisms and are substantially conserved from worms to humans [85]. In contrast to glycosylation, which is stable and localizes mainly at the ER and Golgi apparatus, O-GlcNAcylation is reversible and occurs in the cytoplasm [82]. O-GlcNAcylation has been implicated in several biological activities, including transcription, translation, metabolism, signal transmission, and apoptosis [84].
O-GlcNAcylation can be detected using various techniques such as lectins, antibodies, or click chemistry-based approaches. Lectins, such as Concanavalin A wheat germ agglutinin (WGA), are primarily used for binding to sialic acids and terminal β-GlcNAc on complex glycans [86][87][88]. Metabolic or chemical labeling followed by conjugation to an affinity linker, such as biotin or streptavidin, is a valuable method for detecting O-GlcNAcylation when combined with MS. In addition, some specific enzymes, such as galactosyltransferase, can selectively label the modified sites with a ketone-containing galactose analog, which also helps to identify this modification [87]. O-GlcNAcylation can be detected using various techniques such as lectins, antibodies, or click chemistry-based approaches. Lectins, such as Concanavalin A wheat germ agglutinin (WGA), are primarily used for binding to sialic acids and terminal β-GlcNAc on complex glycans [86][87][88]. Metabolic or chemical labeling followed by conjugation to an affinity linker, such as biotin or streptavidin, is a valuable method for detecting O-Glc-NAcylation when combined with MS. In addition, some specific enzymes, such as galactosyltransferase, can selectively label the modified sites with a ketone-containing galactose analog, which also helps to identify this modification [87].

O-GlcNAcylation on Viruses
Unlike N-and O-glycosylation, which appears on the surface of viruses, O-GlcNAcylation occurs on proteins surrounding the nucleic acid components of viruses [85]. For example, multiple sites on the basic phosphoprotein of human cytomegalovirus are O-Glc-NAcylated (Table 1) [50]. Furthermore, O-GlcNAcylation occurs in rotaviruses, where O-GlcNAc has been detected in RNA polymerase II transcription factors [47]. This modification is also present in adenoviruses and insect viruses, such as baculoviruses [48]. The implications of O-GlcNAcylation include playing a regulatory function, stabilizing multiprotein complexes, and conferring proteolytic resistance [82]. There have been few investigations of viral protein O-GlcNAcylation; nevertheless, the roles of O-GlcNAcylation in viruses are worth investigating further.

O-GlcNAcylation on Noroviruses
Most enveloped viruses have glycosylated surface proteins; however, only a few nonenveloped viruses have glycoproteins in their capsids. Noroviruses fall within the latter category. In 2022, several potential modification sites were discovered to be adjacent to

O-GlcNAcylation on Viruses
Unlike Nand O-glycosylation, which appears on the surface of viruses, O-GlcNAcylation occurs on proteins surrounding the nucleic acid components of viruses [85]. For example, multiple sites on the basic phosphoprotein of human cytomegalovirus are O-GlcNAcylated (Table 1) [50]. Furthermore, O-GlcNAcylation occurs in rotaviruses, where O-GlcNAc has been detected in RNA polymerase II transcription factors [47]. This modification is also present in adenoviruses and insect viruses, such as baculoviruses [48]. The implications of O-GlcNAcylation include playing a regulatory function, stabilizing multiprotein complexes, and conferring proteolytic resistance [82]. There have been few investigations of viral protein O-GlcNAcylation; nevertheless, the roles of O-GlcNAcylation in viruses are worth investigating further.

O-GlcNAcylation on Noroviruses
Most enveloped viruses have glycosylated surface proteins; however, only a few nonenveloped viruses have glycoproteins in their capsids. Noroviruses fall within the latter category. In 2022, several potential modification sites were discovered to be adjacent to the amino acid of the S domain (Thr65, Ser67) and P domain (Thr238, Ser519 in the P1 domain, and Thr350, Thr369, Thr371, Thr381 in the P2 domain), which may be relevant for receptor interactions [46] (Figure 2). The modifications were obtained by MALDI-MS of ethylaminylated peptides from the noroviral VP1 or by LC-MS2 sequencing on the native glycopeptides. Using immunoassays with lectins and antibodies, the authors confirmed the O-GlcNAcylation on VP1 protein. Based on this research, we speculate that O-GlcNAcylation may affect the binding affinity of noroviruses on the HBGAs. Several studies have revealed an interaction between O-GlcNAcylation and phosphorylation [85,89]. However, we lack sufficient studies on O-GlcNAcylation, although there are sufficient reports on norovirus phosphorylation.

Phosphorylation vs. O-GlcNAcylation
Protein phosphorylation is a well-known primary reversible switch for cell signaling control that plays an essential role in various cellular processes. Unlike O-GlcNAcylation, which OGT and OGA control, phosphorylation uses a myriad of protein kinases to transfer γ-phosphate from adenosine triphosphate (ATP) to the amino acid residue in the substrate protein ( Figure 3). The phosphorylation of substrate proteins may occur at one or more sites. Nine amino acids are used as phosphate acceptors, including serine, threonine, and tyrosine (which contain hydroxyl groups (-OH), basic histidine, arginine, lysine, and acidic aspartic acid, glutamic acid, and cysteine. O-phosphorylation of serine, threonine, or tyrosine residues forms a phosphodiester (P-O) link between the -OH and the γ-phosphate of ATP. N-phosphorylation of the histidine, arginine, or lysine residues forms a phosphoramidite (P-N) link between the -NH and the γ-phosphate of ATP. O-phosphorylation is stable. In contrast, N-phosphorylation is acid-labile and, consequently, difficult to detect [90]. Adding a phosphate group to an amino acid residue substantially alters the protein structure. Phosphorylation affects protein characteristics such as enzymatic activity, stability, subcellular localization, and interaction with binding partners [22,91].
domain, and Thr350, Thr369, Thr371, Thr381 in the P2 domain), which may be relevant for receptor interactions [46] (Figure 2). The modifications were obtained by MALDI-MS of ethylaminylated peptides from the noroviral VP1 or by LC-MS2 sequencing on the native glycopeptides. Using immunoassays with lectins and antibodies, the authors confirmed the O-GlcNAcylation on VP1 protein. Based on this research, we speculate that O-GlcNAcylation may affect the binding affinity of noroviruses on the HBGAs. Several studies have revealed an interaction between O-GlcNAcylation and phosphorylation [85,89]. However, we lack sufficient studies on O-GlcNAcylation, although there are sufficient reports on norovirus phosphorylation.

Phosphorylation vs. O-GlcNAcylation
Protein phosphorylation is a well-known primary reversible switch for cell signaling control that plays an essential role in various cellular processes. Unlike O-GlcNAcylation, which OGT and OGA control, phosphorylation uses a myriad of protein kinases to transfer γ-phosphate from adenosine triphosphate (ATP) to the amino acid residue in the substrate protein (Figure 3). The phosphorylation of substrate proteins may occur at one or more sites. Nine amino acids are used as phosphate acceptors, including serine, threonine, and tyrosine (which contain hydroxyl groups (-OH), basic histidine, arginine, lysine, and acidic aspartic acid, glutamic acid, and cysteine. O-phosphorylation of serine, threonine, or tyrosine residues forms a phosphodiester (P-O) link between the -OH and the γ-phosphate of ATP. N-phosphorylation of the histidine, arginine, or lysine residues forms a phosphoramidite (P-N) link between the -NH and the γ-phosphate of ATP. O-phosphorylation is stable. In contrast, N-phosphorylation is acid-labile and, consequently, difficult to detect [90]. Adding a phosphate group to an amino acid residue substantially alters the protein structure. Phosphorylation affects protein characteristics such as enzymatic activity, stability, subcellular localization, and interaction with binding partners [22,91].  One of the most common approaches for detecting phosphorylation is to use radioactiveisotope-labeled substrates such as 32 P orthophosphate. Furthermore, Western blot analysis and arrays have been used to detect phosphorylation. However, these techniques cannot provide information about the phosphorylation sites [59]. Mass spectrometry, coupled with enrichment techniques, has proven to be more robust in identifying PTM substrates and mapping PTM locations [59,63]. Because the overall proteome contains only a small fraction of phosphorylated proteins/peptides, enrichment is an essential step in MS detection of phosphorylation. Antibody-based affinity enrichment and ionic-interaction-based enrichment are the two enrichment procedures [59,92]. The use of pan-PTM antibodies to identify PTM peptides has been proven effective for tyrosine phosphorylation [88]. Furthermore, the interaction between the phosphate group and immobilized metal ions or titanium dioxide (TiO 2 ) is the most common enrichment technique for analyzing phosphorylated peptides by MS, which recognizes over 3000 unique phosphopeptides [59,[93][94][95].
As the density of co-occurring PTMs on proteins is high, several PTMs can affect the action of another via a process termed PTM crosstalk [83]. The most documented form is the PTM crosstalk between phosphorylation and O-GlcNAcylation ( Figure 3). As they occur primarily on the same amino acid residues (serine and threonine), these two PTMs undergo crosstalk. Crosstalk may happen in various ways, including competition for the same site/residue (reciprocal crosstalk) and modifications affecting each other (proximal or distal to the peptide sequences) [83]. Furthermore, interruption of phosphorylation events alters the GlcNAcylation levels and vice versa. These findings demonstrate crosstalk between the two modifications [83,89].

Phosphorylation on Viruses
Many intracellular obligatory pathogens require phosphorylation to initiate a productive infection cycle [22]. The phosphorylation of viral proteins affects viral-host interactions, which substantially impact viral infection, replication, and cytotoxicity [30]. The PTMs of viral proteins, particularly RdRps, are common. Hepatitis C virus (HCV) RdRp, for example, is phosphorylated by protein kinase C-related kinase 2 (PRK2), which is essential for effective viral replication [91,96]. The viral RdRp enzyme is the main enzyme involved in viral RNA genome replication in plus-strand RNA viruses, including noroviruses [97]. RdRp phosphorylation has been proposed to be functionally connected to viral replication [16,98].

Phosphorylation on Noroviruses
Phosphorylation is widely recognized for directly regulating viral protein activity and acting as a molecular signal for a binding partner. Norovirus RdRp and FCV VPg proteins are reportedly phosphorylated. RdRp is phosphorylated at a position (Thr33) at the interface of the RdRp finger and thumb domains. This modification is exclusive to the most common norovirus genotypes, including GII.4 and GII.b [16]. The phosphorylation sites of the FCV VPg protein are threonine at position 80 and serine at position 107. These polymerases and virus-encoded proteins are required for viral function. RdRp, found in viral particles, is responsible for viral genome transcription and replication. VPg interacts with NS7 to help viral RNA synthesis, whereas its interaction with eIF4A triggers viral protein synthesis and VP1 functioning in viral encapsidation [15,99]. Both RdRp and VPg play essential roles in viral evolution and fitness. Consequently, the phosphorylation of RdRp and VPg may provide a mechanism for noroviruses and FCV, respectively, to regulate the viral life cycle and impact viral pathogenicity.

Conclusions
PTMs enable viruses to regulate molecular functions by maintaining stability, interacting with receptors, and suppressing the immune system. Many RNA viruses feature PTMs, as shown by previous studies; however, the understanding of norovirus PTMs remains poorly elucidated. Research suggests that phosphorylation promotes norovirus pathogenicity; however, information on glycosylation and O-GlcNAcylation is limited. Investigating the factors behind this conclusion is worthwhile because phosphorylation and O-GlcNAcylation frequently interact. One reason for this might be that only a few nonenveloped viruses exhibit glycosylation and O-GlcNAcylation. Furthermore, the potential for O-GlcNAcylation varies from cell to cell. The host cell type should be considered when analyzing viral glycosylation, and sophisticated mass spectrometry tools can aid the investigation. Given that glycosylation, O-GlcNAcylation, and phosphorylation are all associated with viral pathogenicity, identifying the modification sites on noroviruses will aid in developing future vaccines and treatments.

Data Availability Statement:
The data that support the findings of this study are available in the material of this article.

Conflicts of Interest:
The authors declare no conflict of interest.