Skip to main content
NCBI home page
Glycobiology logoLink to Glycobiology
. 2022 Nov 12;33(2):126–137. doi: 10.1093/glycob/cwac077

Structural remodeling of SARS-CoV-2 spike protein glycans reveals the regulatory roles in receptor-binding affinity

Yen-Pang Hsu 1,✉, Martin Frank 2, Debopreeti Mukherjee 3, Vladimir Shchurik 4, Alexey Makarov 5, Benjamin F Mann 6,✉
PMCID: PMC9990995  PMID: 36370046

Abstract

Glycans of the severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2) spike protein are speculated to play functional roles in the infection processes as they extensively cover the protein surface and are highly conserved across the variants. The spike protein has been the principal target for vaccine and therapeutic development while the exact effects of its glycosylation remain elusive. Analytical reports have described the glycan heterogeneity of the spike protein. Subsequent molecular simulation studies provided a knowledge basis of the glycan functions. However, experimental data on the role of discrete glycoforms on the spike protein pathobiology remains scarce. Building an understanding of their roles in SARS-CoV-2 is important as we continue to develop effective medicines and vaccines to combat the disease. Herein, we used designed combinations of glycoengineering enzymes to simplify and control the glycosylation profile of the spike protein receptor-binding domain (RBD). Measurements of the receptor-binding affinity revealed opposite regulatory effects of the RBD glycans with and without sialylation, which presents a potential strategy for modulating the spike protein behaviors through glycoengineering. Moreover, we found that the reported anti-SARS-CoV-(2) antibody, S309, neutralizes the impact of different RBD glycoforms on the receptor-binding affinity. In combination with molecular dynamics simulation, this work reports the regulatory roles that glycosylation plays in the interaction between the viral spike protein and host receptor, providing new insights into the nature of SARS-CoV-2. Beyond this study, enzymatic glycan remodeling offers the opportunity to understand the fundamental role of specific glycoforms on glycoconjugates across molecular biology.

Keywords: COVID-19, Glycobiology, glycoengineering, viral glycosylation

Introduction

Host cell infection of severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2) is initiated by the interaction between the viral spike (S) proteins and the host cellular receptors, angiotensin-converting enzyme 2 (ACE2; Hoffmann et al. 2020; Letko et al. 2020). The S protein recognizes ACE2, mediates protease priming and then promotes virus membrane fusion (Shang et al. 2020a). Because of its critical role in the infection processes, the S protein has become the principal target for vaccine and therapeutic development (Huang et al. 2020; Sternberg and Naujokat 2020). Similar to other β-coronavirus members, the SARS-CoV-2S protein is heavily glycosylated (Watanabe et al. 2020a; Shajahan et al. 2020b). It contains 22 N-linked glycosylation sites with 18 of them conserved from the SARS-CoV spike (Fig. 1A; Walls et al. 2020; Watanabe et al. 2020a). The glycan moieties contribute ~17% molecular weight to the native-state trimetric SARS-CoV-2 S protein, shielding ~40% of the protein surface (Grant et al. 2020).

Fig. 1.

Fig. 1

Open in a new tab

SARS-CoV-2 S protein glycosylation. A) Glycosylation sites are distributed throughout the S protein structures. Star symbols (*) indicate conserved glycosylation sites with SARS-CoV S protein. NTD: N-terminal domain; RBD: receptor-binding domain; FP: fusion peptide; HR: heptad repeat; TM: transmembrane region; and CT: cytoplasmic domain fusion. B) Structure of S protein trimer-ACE2 complex with highlighted RBD glycosylation sites in purple and pink. The glycosylated S trimer is modeled based on the work from Woo et al. (2020). The ACE2 is added based on PDB entry 6m0j. C) Hydrophilic interaction liquid chromatography (HILIC) chromatogram of glycans collected from HEK293-expressed S protein RBD used in this study. D) Composition of S protein RBD glycosylation described in the relative abundance of glycoform(s) with each property. Left: graphic representative of the RBD glycopattern. Upper right: composition of glycan classes. Bottom right: composition of terminal saccharide species. 56.5% of the RBD glycan population contains at least 1 terminal sialic acid; 29.3% of the population exhibits at least 1 terminal galactose without having any sialic acid; 9.6% of the population terminates with GlcNAc without having any galactose and sialic acid.

Glycosylation plays versatile roles in viral pathobiology (Watanabe et al. 2019). For example, glycans are part of the cell entry machinery of human immunodeficiency virus 1 (HIV-1). Deleting certain glycosylation sites from HIV-1 envelope protein reduces virus integrity and results in significantly decreased infectivity (François and Balzarini 2011). A similar phenomenon in SARS-CoV-2 was reported recently by Casalino et al., where they found that the N165 and N234 glycans on the S protein S1 subunit modulate the protein conformational transition required for receptor binding (Casalino et al. 2020). Removal of these 2 glycosylation sites through mutation leads to a 10–40% reduction of ACE2 binding response, as supported by biolayer interferometry (BLI) analyses (Casalino et al. 2020). Another well-known role of viral glycosylation is masking viral immunogenic epitopes from the host immune systems by camouflaging them with host-derived glycans (Zhao et al. 2020; Watanabe et al. 2020b). This has proved to be a powerful strategy to maintain the infectivity in coronavirus S protein, HIV-1 envelope protein, and influenza hemagglutinin (Tate et al. 2014; Stewart-Jones et al. 2016; Walls et al. 2016; Crispin et al. 2018; Yang et al. 2020). In SARS-CoV-2, the surface of the S protein is extensively covered by glycans, indicating effective camouflaging effects against antibody recognition. However, a notable exception was found in the receptor-binding domain (RBD; Grant et al. 2020).

The S protein RBD contains 222 residues (R319 to F541 residues) with 2 glycosylation sites located at N331 and N343 (Fig. 1B). Strong ACE2 interactions were found in the receptor-binding motif (S438 to G504 residues) through hydrogen bonds and salt bridges, as revealed by crystal structures (Benton et al. 2020; Lan et al. 2020). Because of its crucial role in ACE2 binding, the RBD has the lowest coverage of glycan shielding among the entire protein, which makes it vulnerable to immune recognition (Casalino et al. 2020; Watanabe et al. 2020b). One possible explanation for this phenomenon is the existence of the “up and down” conformational change of the spike protein during the cell entry processes, where the RBD remains buried by the heavily glycosylated S1 unit (the “down” state) during trafficking for immune evasion until it engages ACE2 at the infection interface (turning into the “up” state; Cai et al. 2020; Shang et al. 2020a). This strategy minimizes the exposure time of RBD to the surrounding local environment, reducing the probability of immune recognition. However, this explanation makes the roles of the RBD glycans even more intriguing, especially for the glycan at N343 which is very close to the receptor-binding motif (RBM; Lan et al. 2020). We suspect that the RBD glycans could have roles additional to glycan shielding in the S protein-ACE2 interactions.

In-depth probing of glycans’ functions during viral infection is challenging, largely due to the lack of strategies to control glycan structures and minimize their microheterogeneity (Wong 2005; Struwe and Robinson 2019; Guo et al. 2020; Ma et al. 2020). As a result, molecular dynamics (MD) simulation has become the predominant approach for studying glycan biochemistry; yet support from experimental data is in great demand (Feig et al. 2018). In this work, we report the strategies to harmonize the glycans of SARS-CoV-2 S protein RBD into controlled glycoforms, as well as subsequent binding affinity measurement between human ACE2 and the glycoengineered RBD. By the designed combinations of glycoengineering enzymes that we have characterized, we successfully transformed the S protein RBD glycans into (i) glycoforms with harmonized terminal glycan species; and (ii) structure-defined single glycoforms (Hsu et al. 2021). This work reveals the regulatory roles of S protein RBD glycan in receptor binding: A double-edged sword that can either stabilize or destabilize RBD–ACE2 interactions. In combination with their roles in glycan shielding, these insights lay the foundations for modulating the S protein’s nature through glycan remodeling, which may create new strategies for vaccine and therapeutic design.

Results

Dissecting the glycoforms of SARS-CoV-2 S protein RBD

Glycans in the S protein RBD comprise N-glycan species (Watanabe et al. 2020a; Shajahan et al. 2020b). Their biosynthesis occurs in the endoplasmic reticulum (ER) and continues in the Golgi in a species-, cell-, protein-, and site-specific manner. N-glycans share a common core structure made of 2 N-acetylglucosamine (GlcNAc) and 3 mannose (Man) residues. The core further extends into a myriad of glycoforms through the activity of glycosidases and glycosyltransferases. It has been known that different glycoforms could lead to different regulatory effects on protein–protein or cell–cell interactions (Reily et al. 2019).

Given that glycan formation is sensitive to the expression conditions, we first analyzed the glycoforms of the recombinant RBD (expressed from HEK293, GenScript Z03483) used in this study by isolating the glycans using PNGase F and profiling them by liquid-chromatography mass spectrometry (LC–MS) analysis (Fig. 1C). Our substrate RBD exhibits heterogeneous glycoforms composed of complex-type species (95.4%), with 3 antennae the highest number we observed, and a small amount of high-mannose (1%) and hybrid-type (3.6%) species (SI-Glycan profile). As summarized in Fig. 1D, GlcNAc, galactose, and sialic acid (SA) appear as the terminal monosaccharide at the non-reducing ends of N-glycans with relative abundances of 9.6%, 29.3%, and 56.5%, respectively. Over 99% of the RBD glycans contain the core fucose and 18.6% of the glycoforms have additional fucose located at the complex-type glycan antennae, as supported by glycosidase treatment experiments (Supplementary Fig. S1, see online supplementary material for a color version of this figure). N-Acetylgalactosamine (GalNAc), the epimer of GlcNAc, likely exists in the RBD glycans since N-acetylglucosaminidase (GlcNAcase) alone was not able to remove all the N-acetylhexosamine (HexNAc) residues, unless GalNAcase was also added (Supplementary Fig. S2, see online supplementary material for a color version of this figure). In addition, retention time comparison using glycan standards suggested that a small amount of bisecting N-glycan species exist, as indicated by FA2B glycoform ([Man]3[GlcNAc]5[Fuc]1; Supplementary Fig. S3, see online supplementary material for a color version of this figure). We note that our analyses based on isolated glycans did not provide site-specific information for the glycosidic linkages. The reported numbers here are averaged results from the glycan populations at N331 and N343.

RBD glycans stabilize RBD–ACE2 interaction

We used BLI to study the interaction between the S protein RBD and ACE2 expressed from HEK293. The RBD with native glycoforms binds to ACE2 with an equilibrium dissociation constant (KD) of 99 ± 12 nM, similar to earlier reports (Shang et al. 2020b). The terminal saccharides of mature glycans often regulate the biochemical properties and functions of glycoconjugates. An example comes from the human blood group antigens that are classified by their terminal residue species (Cummings 2017). Therefore, we aimed to harmonize the RBD glycan termini and investigate their potential effect on the RBD–ACE2 interaction. First, we created glycoengineered RBD with all N-glycans ending in terminal HexNAc by incubating the native RBD with α2–3/6/8 neuraminidase, β1–4 galactosidase, and α1–2,3/4 fucosidases (Figs. 2 and 3A, Supplementary Table SI). Interestingly, we found that the resulting substrate, named tHexNAc-RBD, had improved binding affinity to ACE2 with a KD value of 47 ± 8 nM (Fig. 3B). Direct comparison of ACE2 binding curves showed a 20% increase in binding response in tHexNAc-RBD compared with the native RBD (Supplementary Fig. S4, see online supplementary material for a color version of this figure). Furthermore, we prepared glycoengineered RBD bearing (i) the core glycan ([Man]3[GlcNAc]2[Fuc]1, core-RBD), and (ii) species terminating with galactose (tGal-RBD) by introducing N-acetylglucosaminidase/N-acetylhexosaminidase cocktail and Galactosyltransferase to the reactions, respectively. Similar to the tHexNAc-RBD, having glycans terminating with mannose and galactose slightly improved the ACE2 binding affinity relative to the native RBD. This result suggests that certain glycoforms can facilitate RBD–ACE2 interactions.

Fig. 2.

Fig. 2

Open in a new tab

Scheme of the S protein RBD glycan remodeling routes. A) Enzymes used in this work and their corresponding saccharide targets. B) Preparation of the S protein RBD with controlled glycoforms using enzymatic reactions. The yield indicates estimated populations of the desired glycoform(s).

Fig. 3.

Fig. 3

Open in a new tab

Glycans of the S protein RBD affect ACE2 binding affinity. A) Chromatograms of glycans collected from glycoengineered RBD in this study. Left: the RBD with harmonized terminal glycan species. Right: the RBD substrates with single mono-antennary glycoforms. B) Binding affinity measurement (BLI) between human ACE2 and the S protein RBD with controlled glycoforms. The numbers located at the bottom of each data point indicate the R2 value of curve fitting. Error bars: standard error. C) The structure of the RBD-ACE2 complex revealed potential interaction between the RBD N343 glycan and ACE2 glycans/amino acid backbone (PDB: 6m0j). Insertion displays the electrostatic potential of the ACE2 surface that potentially interacts with RBD N343 glycan. Red: negatively charged; blue: positively charged.

Recently published simulation studies by Mehdipour et al. provided a plausible explanation for our observation (Mehdipour and Hummer 2021). In addition to the S protein RBD, ACE2 is also a heavily glycosylated protein (Shajahan et al. 2020a). Both the RBD glycan at N343 (N343RBD) and the ACE2 glycan at N322 (N322ACE2) are close to the binding interface with a distance of only 31 Å between the 2 glycosylation sites (Asn side chains, Fig. 3C). Mehdipour and co-workers analyzed the ACE2 glycan dynamics and found that the N322ACE2 glycans interact with the RBD protein backbone as well as the N343RBD glycans (Mehdipour and Hummer 2021). This simulation result was supported by our BLI experiments where we measured the ACE2 binding affinity of deglycosylated RBD prepared by endoglycosidase F2 treatment (Trimble and Tarentino 1991; Tarentino and Plummer Jr 1994; Supplementary Fig. S5, see online supplementary material for a color version of this figure). An apparent reduction in binding response was found with a KD value of 110 ± 3.3 nM, a weaker affinity than that of native RBD (Fig. 3B). Moreover, when the ACE2 N-glycans were removed, a dramatic decrease in binding affinity was observed (Supplementary Fig. S6, see online supplementary material for a color version of this figure). Along with the published simulation study, our result suggested that the RBD glycans can stabilize ACE2 binding, likely through the interactions with the N322ACE2 glycans or the ACE2 backbone.

Glycans with terminal sialic acid destabilize RBD–ACE2 interactions

Sialic acids are often involved in biomolecular interactions because they are strongly (negatively) charged and usually found at glycan non-reducing end termini (Matrosovich et al. 2015). By incubating α2–6 sialyltransferase with the tGal-RBD, we introduced sialic acids to the terminus of the RBD glycans. The resulting tSA-RBD contains heterogeneous glycoforms with over 99% of the population bearing at least 1 terminal sialic acid (Fig. 3A, SI-Glycan Profile). Unexpectedly, BLI indicated a lower binding affinity of tSA-RBD to ACE2 (KD = 130 ± 6 nM) compared with the Native and deglycosylated RBD. We reasoned this result to the electrostatic repulsion as computational calculation has shown that the ACE2 surface is predominantly negative, including the area that sialylated N343RBD glycan may engage (Xie et al. 2020; Fig. 3C, Supplementary Fig. S7, see online supplementary material for a color version of this figure).

To further study the glycan spatial orientation and flexibility, we performed MD simulations on isolated S RBD with sialylated N-glycans. Distance measurement between the terminal sialic acids and the RBD Thr500 residue, a binding-motif residue located at the RBD–ACE2 interaction interface, provides an estimate to what extent the glycans could intrude into the negative electrostatic field of ACE2. The data from the accumulated 3-μs simulation suggests that the glycans are highly flexible (Fig. 4A). A significant number of conformations exists in the ensemble where one of the negatively charged termini of glycan N343RBD would be located close to the ACE2 binding interface. Next, we introduced ACE2 to the simulation and performed (accumulated) 1 μs of the complex. In agreement with published work, we observed significant glycanACE2–glycanRBD interactions based on an atom–atom contact analysis (Fig. 4B and C, Supplementary Fig. S8, see online supplementary material for a color version of this figure). A comparison of the sialic acid-T500 distance distributions between the free RBD and the RBD-ACE2 complex revealed a shift towards larger values for the latter, especially for the sialic acid residue at the α1–3 mannose antenna (Fig. 4D, Supplementary Fig. S9, see online supplementary material for a color version of this figure). This is possibly an effect of the repulsion of the negative electrostatic field of ACE2. Together, sialylated glycoforms, which are negatively charged, could cause electrostatic repulsion and destabilize RBD–ACE2 interactions. Given that the native RBD has a heterogeneous glycan profile, the existence of the sialylated glycan species possibly negates the stabilizing effects resulting from other glycoforms, which explains the weaker ACE2 binding affinity in native RBD over non-sialylated glycoforms. Recent reports of site-specific glycan mapping have also indicated that both native N343RBD and N322ACE2 glycosylation sites have a low content of sialylated glycan species, further implying that sialylation may not be favored for RBD–ACE2 interactions (Shajahan et al. 2020a; Watanabe et al. 2020a).

Fig. 4.

Fig. 4

Open in a new tab

Molecular dynamic simulation indicates N343RBD glycan interactions with ACE2 glycans at the binding interface. A) Analysis of the distances between the 2 sialic acid residues of N343RBD glycan and the Thr500 residue based on the accumulated 3 μs MD simulation. The distance correlation plot is colored by free energy calculation derived from the Boltzmann equation. Lower energy indicates a higher probability. The mean, minimum and maximum distance is summarized in the table. B) Atom–atom contact analysis based on MD simulation. The numbers show the averaged counts of atom–atom contacts between the units. C) Snapshot from MD simulation showing intensive interactions between N343RBD and N322ACE2 glycans. Terminal sialic acid residues are highlighted in purple. D) Comparison of sialic acid-Thr500 distances between free RBD and RBD-ACE2 complex.

Another possible explanation for the reduced ACE2 binding affinity found in tSA-RBD could be the increased steric hindrance caused by sialylation. Having over-constructed glycan structures nearby the receptor-binding motif could prevent the S protein from approaching ACE2. To test this possibility, we constructed harmonized mono-antennary glycans on RBD, as confirmed by LC–MS analysis (Fig. 3A, Supplementary Fig. S10, see online supplementary material for a color version of this figure, SI-Glycan Profile). The FA1G1-RBD bears an octa-saccharide with only 1 antenna connected to the α1-3-linked core mannose. This was achieved by incubating the core-RBD with N-acetylglucosaminyltransferase I (also known as MGAT1) and galactosyltransferase in one pot. Similarly, the FA1G1S1-RBD contains an additional sialic acid and was prepared by introducing sialyltransferase to the reaction. BLI analyses showed that the FA1G1-RBD has improved ACE2 binding affinity compared with native RBD; by contrast, FA1G1S1-RBD showed weaker binding than native RBD. The results are consistent with the data we got earlier, suggesting that steric hindrance is not the major cause of the ACE2-binding affinity reduction found in the tSA-RBD. Instead, the electrostatic repulsion is more likely the reason. Our results, together, suggest that the RBD glycans have regulatory effects on the RBD–ACE2 interactions: non-sialylated glycoforms reinforce the binding, whereas sialylated glycoforms destabilize it.

RBD glycan fucosylation does not have apparent impacts on ACE2 binding

Glycosidase treatment provided direct evidence that the core structures of our RBD glycans are fucosylated (Fig. 3A, Supplementary Fig. S1, see online supplementary material for a color version of this figure). The core fucose linked to the reducing end GlcNAc has been a hot target of interest for pharmaceutical research because of its regulatory effects on protein–protein interactions, such as immunoglobulin G (IgG) and its receptors (Mizushima et al. 2011). To investigate the role of fucosylation in the RBD–ACE2 interaction, we prepared partially defucosylated RBD using α1–6 fucosidase that targets the core fucose. A 10-day reaction converted ~45% of native glycans into non-fucosylated forms (Supplementary Fig. S11, see online supplementary material for a color version of this figure). It is known that the activity of α1–6 fucosidase is glycoform-dependent, where lower structural complexity leads to higher enzyme activity (Hsu et al. 2021). Therefore, we used the core-RBD as the substrate for the reaction and successfully remove fucose from over 90% of the core glycans (Supplementary Fig. S11, see online supplementary material for a color version of this figure). BLI measurement showed that both defucosylated RBD substrates have no significant difference in ACE2 binding affinity compared with their parent species (Fig. 3B). This result is not surprising because the core fucose is shielded close to the RBD peptide backbone, which minimizes its probability of interacting with ACE2 or ACE2 glycans. In addition to the core fucose, ~18.6% of RBD glycans contain α1-2 and α1-3/4 fucose, which can be removed by corresponding fucosidases (Supplementary Fig. S1, see online supplementary material for a color version of this figure). Their removal did not result in a significant difference in ACE2 binding affinity (KD = 110 ± 14 nM).

Monoclonal Antibody S309 neutralizes the regulatory effects of the RBD glycans

Neutralizing antibodies have presented effective therapeutics for viral infection treatment (Corti et al. 2016). An antibody, named S309, isolated from a SARS-CoV-infected patient was reported to have strong cross-neutralization activity on SARS-CoV-2 with the ability to induce antibody-dependent cell cytotoxicity (ADCC) and antibody-dependent cellular phagocytosis (ADCP) (Pinto et al. 2020). S309 has a distinct RBD binding interface from that of ACE2. It recognizes a SARS-CoV-2 S protein RBD epitope consisting of 2 regions, residues 356–361 and residues 440–444. Notably, the S309 binding interface sandwiches the N343RBD glycosylation site and shows potential interaction with the glycan structure, especially the core fucose (Pinto et al. 2020; Fig. 5A). To investigate this interaction, we measured the binding affinity between S309 and the RBD with differently engineered glycan structures. BLI analyses showed strong RBD–S309 interaction with KD values at the nanomolar scale (Fig. 5B). However, no significant difference was found when the RBD glycans bore different terminal saccharide species, so to the removal of the core structure. Indeed, whether or not the RBD glycans, especially the core fucoses, are involved in S309 interaction remains unclear. Available cryo-EM structures (PDB entry 6wps) showed that the core fucose establishes a significant number of favorable interactions with S309, whereas high-resolution crystal structures such as PDB entry 7r6w (1.83 Å) provided opposite indications (Pinto et al. 2020; Starr et al. 2021). Overall, our results suggest that the N343RBD glycan is not essential for the RBD-S309 binding. The RBD–S309 interaction is likely established mainly on the amino acid epitope of the RBD.

Fig. 5.

Fig. 5

Open in a new tab

Antibody S309 neutralizes the regulatory effect of the RBD glycans. A) The N343RBD glycan is located at the RBD-S309 binding interface. The RBD–S309 interaction hinders the accessibility of the N343RBD glycan to ACE2 and ACE2 glycans. The model was extracted from the MD simulation of RBD-ACE2-S309 complex built on PDB entries 6wps and 6m0j. Glycan sialic acids are highlighted in purple. B) RBD glycans are not involved in S309 binding. No significant difference was observed in the binding affinity analysis. C) The RBD-S309 binding neutralizes the regulatory effect of the RBD glycan on ACE2 interaction. The numbers located at the bottom of each data point indicate the R2 value of curve fitting. Error bars: standard error.

From another perspective, the presence of the S309 mAb may limit the accessible conformational space of the N343RBD glycan and possibly restrict its interaction with ACE2 and/or ACE2 glycans. This is supported by MD simulation showing that S309-RBD interactions “lock” the orientation of the N343RBD glycan, pushing it away from the ACE2 binding interface (Fig. 5A, Supplementary Fig. S12, see online supplementary material for a color version of this figure). To test this hypothesis, we measured the ACE2 affinity of the RBD samples that were bound to S309 already. We observed similar KD values (≈ 90 nM) on the native RBD, tHexNAc-RBD and tSA-RBD that were saturated with S309 (Fig. 5C). Namely, S309 neutralizes the regulatory effect of the RBD glycan. Given that the native RBD glycans tend to stabilize ACE2 interaction owing to the low content of sialylated glycoforms, the RBD-S309 binding could compromise this stabilizing effect and reduce the infectivity of the virus, which presents a new inhibitory mechanism in S309 neutralization.

Discussions

The COVID-19 pandemic has caused numerous deaths worldwide. Although the vaccination has relieved the immediate suffering, chronic symptoms could have persisted in many who have contracted the virus (Baig 2021). With the rapid evolution of SARS-CoV-2, a comprehensive understanding of the virus pathology is in urgent demand for developing vaccines and effective therapeutics to cease the spread of the virus variants (Cele et al. 2021, Hirotsu et al. 2021, Leung et al. 2021).

Glycans play versatile structural and functional roles in protein biochemistry, but it has remained historically challenging to elucidate the phenotypic roles of discrete glycoforms. Given the proximity of the RBD glycans to the binding interface with ACE2, investigating their roles in biological properties could provide insights into the infection mechanism (Walls et al. 2020; Watanabe et al. 2020a). In this work, we have used an enzymatic remodeling approach to remove the native glycosylation microheterogeneity and decorate the S protein RBD with specific glycan structures to study their impact on receptor binding (Hsu et al. 2021). We found that the neutrally charged RBD glycans can stabilize ACE2 binding, likely through the interaction with ACE2 glycans, as suggested by MD simulations presented here and elsewhere (Mehdipour and Hummer 2021; Huang et al. 2022). Remarkably, this effect is reversed when negatively charged sialic acids are present on the RBD glycans, a destabilizing force possibly caused by electrostatic repulsion. The dissociation constants measured herein varied ~3-fold, based solely on the displayed N-glycan structures. The results suggest that SARS-CoV-2 infectivity could be altered by manipulating the content of sialic acid on the RBD glycans. For example, neuraminidase inhibitors could mitigate the viral infection by halting the removal of sialic acid from the spike protein. Although anti-influenza neuraminidase inhibitors have proven to be inactive for SARS-CoV-2 treatment, strategies that regulate the neuraminidase and sialyltransferase activities on the RBD glycans are still worth investigating (Wang et al. 2020).

In addition to ACE2, cell surface lectins are also known to interact with the spike protein (Geijtenbeek and Gringhuis 2009). Their interactions improve viral adhesion to the cell surface, facilitating the entry processes. DC-SIGN (CD209) was characterized as a SARS-CoV-1 spike protein receptor and has been reported to bind to the SARS-CoV-2 spike protein RBD with a KD at a micro-molar scale (Marzi et al. 2004; Hoffmann et al. 2021; Lempp et al. 2021; Thépaut et al. 2021). We studied the impact of RBD glycosylation on their interactions using BLI. However, we did not see a significant difference between differently glycoengineered RBDs, unless the RBD N-glycans are fully removed (Supplementary Fig. S13, see online supplementary material for a color version of this figure). This could attribute to DC-SIGN’s substrate selectivity for oligomannose N-glycans, instead of the complex-type N-glycans found on the RBD (van Liempt et al. 2006). Finally, we found that the binding of S309 neutralizes the regulatory effects of the RBD glycan, preventing the potential stabilization of ACE2 interaction through the glycan structure. Given that the glycosylation sites of the RBD are highly conserved across its evolution reported to date, glycans could provide new handles for developing strategies to combat the emerging virus variants (Supplementary Fig. S14, see online supplementary material for a color version of this figure). For example, glycan remodeling could be applied to protein subunit vaccines to maximize the mimicry of viral glycoproteins and improve the immune response.

Generally speaking, this work demonstrates that screening the biological behavior of discrete protein glycoforms on glycoproteins can be achieved by enzymatic glycoengineering. However, challenges are remaining in the current method. For example, site-specific glycoengineering is still not available. As glycosylation at different sites could have distinct bio-functions, enabling the remodeling of targeted sites while preserving the nature of the rest would allow us to dissect the particular roles they play. Because of this limit, we were not able to study RBD glycosylation in this work by using the full-length spike protein trimer as the substrate, a better mimicry of the viral infection machinery that contains 66 N-glycosylation sites. To address this challenge, evolving glycoengineering enzymes toward higher site-specificity presents a potential solution (Arnold 2018). Having an in-depth understanding of how glycosylation alters protein behaviors, we believe, will greatly reinforce our ability to design targeted strategies to defeat diseases.

Materials and methods

Please refer to electronic supplementary information (ESI) for supplementary tables and figures.

The Spike protein RBD glycan remodeling

Please see Fig. 2 and Supplementary Fig. S1 for reaction conditions and remodeling routes. Reactions were started by mixing the S protein RBD and glycoengineering enzyme(s) in a 10 K MWCO tube for buffer exchange into the corresponding reaction buffer. The reaction solution was then collected into a 1.7-mL microtube and incubated in a temperature-controlled shaker until the reaction was complete, as confirmed by glycan analysis (see below). Standard Ni-NTA resin purification protocol was used to purify the crude product (an empty solid-phase extraction, or SPE, column and vacuum manifold were used for the purification). Finally, the purified product was transferred into 10 K MWCO tubes for buffer exchange into either Octet buffer (25 mM Tris–HCl pH 7.4, 50 mM NaCl, 0.02% Tween 20) for biolayer interferometry analysis or the reaction buffer for the subsequent reaction. The enzyme and RBD concentrations were determined by NanoDrop. The glycan remodeling processes can be simplified by combining certain enzymes into one-pot reactions (for more details see Fig. 2).

The reaction buffer was prepared as follows: For glycosidase reactions: 50 mM sodium acetate at pH 5.5 with 5 mM CaCl2. For glycosyltransferase reactions: 25 mM Tris–HCl pH 7.4 with 50 mM NaCl and 10 mM cations (MnCl2, MgCl2, and CaCl2). The buffer for Endoglycosidase reactions contains 50 mM sodium acetate at pH 4.5 for Endo F2 and Endo F3; 50 mM sodium acetate at pH 6 for Endo H and Endo M; 50 mM sodium acetate at pH 5.5 and 5 mM CaCl2 for Endo S.

LC–MS analysis of S protein RBD glycans

This protocol is adapted from the Glycoworks manual provided by Waters. (i) Glycan isolation: 15 μg of the S protein RBD (in either reaction buffer or octet buffer), 6 μL RapiGest SF (50 mg/mL, in Glycoworks buffer), and water were mixed in a 1.5 mL microtube to a final volume of 25 μL. The mixture was then incubated at 90 °C for 5 min to denature the substrates. After the samples were cooled down to room temperature, 1.2 μL Rapid PNGase F was added to the tube, followed by another incubation at 50 °C for 10 min. (ii) Glycan labeling: After the PNGase F digestion, 12 μL RapiFluor-MS solution (70 mg/mL, in DMF) was added directly to the solution. The mixture was gently vortexed and then incubated at room temperature for 20 min without any light exposure. RapiFluor-MS labeling gives higher signal-to-noise (S/N) ratio in MS analysis. A larger amount of analyte might be required without the labeling step. After labeling, the samples (~40 μL) were diluted with 360 μL acetonitrile and were ready for purification. (iii) Glycan purification: Oasis SPE μPlate from Waters (along with the use of μPlate extraction manifold) was then used for the first solid-phase extraction purification, and Discovery SPE from Sigma-Aldrich (along with the use of 20-wells SPE vacuum manifold) was used for the second purification to ensure high signal-to-noise ratio. The SPE columns/μPlate were first washed with water (1 column volume) and then conditioned by water-acetonitrile solution (10:90 v/v, 1 column volume). The glycan samples (in ~90% acetonitrile solution) were then charged to the column/μPlate, followed by washing with washing buffer (formic acid/water/acetonitrile 1:9:90 v/v/v, 2 column volume). The glycans were then eluted with 80 μL elution buffer (200 mM ammonium acetate in 5% acetonitrile). (iv) HILIC–MS analysis: The purified glycan samples were injected into ultraperformance liquid chromatography equipped with ACQUITY BEH Glycan column (130 Å, 1.7 μm, 2.1 × 150 mm) in tandem with IMQ-TOF MS (Ion-Mobility Quadrupole Time-of-Flight Mass Spectrometry) for glycan profile analysis. The method provided by Waters for glycan chromatography was used in this work (Supplementary Table SII).

The instrument setting for the collection of mass spectrometry data was as follows: Source type: Dual AJS ESI with the gas temperature at 300 °C; Nebulizer pressure at 35 PSIG; Sheath gas Temperature at 325 °C; Nozzle voltage at 2 kV; and Fragmentor voltage: 175 V. The scan range was 600 – 3,200 m/z. Only intact mass data was collected. System tuning was performed using Agilent tune mix (G1969-8501, G1969-8503).

Glycan analyses

The N-glycans on the S protein RBD were determined by LC–MS analysis. MassHunter Software from Agilent was used to acquire and analyze the data. (i) Peak area integration: The area of each chromatographic peak was determined by drawing the peak baseline manually in order to include the glycoforms with low abundance. Peaks with an S/N ratio of < 1.2 were excluded. (ii) Glycan assignment: The collected MS data from each peak was imported into the GlycanMass and GlycoMod analysis tools operated by Expasy to reveal the saccharide composition of the glycans (Cooper et al. 2001; Gasteiger et al. 2003). Given that the majority of RBD glycans belong to complex-type species, glycoforms can be predicted by matching the saccharide compositions to the complex-type glycan database (GlycoMod). Glycan standards (Waters) were used to confirm certain glycoforms, e.g. FA2 glycan, based on retention time comparison, in addition to the accurate mass measurement. (iii) Glycoform population: After peak assignment, the relative abundance of glycoform(s) of interest was calculated based on the UV peak area. Detailed glycan structure characterization using NMR or MS fragmentation is beyond the scope of this study and was not included.

Binding affinity analysis using biolayer interferometry (Octet)

Ni-NTA Octet sensors were used for the measurement of the S protein RBD-ACE2 binding affinity. The sensors were hydrated in Octet buffer (25 mM Tris–HCl pH 7.4, 50 mM NaCl, 0.02% Tween 20) for 30 min before the experiments. To begin the measurement, the sensors were loaded with 2 μg/mL (50 nM) His-tagged S protein RBD for 400 s, followed by baseline equilibration for 10 min. The association of recombinant human ACE2 was performed at 0, 10, 20, 30, 40, 60, 80, and 120 μg/mL (0-1 μM) for 400 s. Dissociation was measured for 30 min. The equilibrium dissociation constant (KD) values were calculated using a 1:1 global fit model in the Octet data analysis software. The Octet buffer was used in all the steps. Measurement for the RBD-S309 binding affinity was carried out using a similar protocol with the antibody concentration at 0, 0.2, 0.4, 0.8, 1.6, 3.2, 6.4, and 12.8 μg/mL (0–85 nM). The RBD-DC-SIGN binding affinity study was performed using the same protocol with the DC-SIGN concentration range to be 0.1 to 20 μM.

The measurement of ACE2 binding affinity with the S309-bound RBD was carried out using adjusted protocols: An incubation with S309 at a high concentration (12 μg/mL) was introduced to the BLI measurement process after the Ni-NTA sensors were loaded with the RBD. The binding response was monitored in real-time until it reached a plateau (1,200 s) to ensure S309 binding was saturated. The ACE2 association and dissociation measurements were then carried out using the same protocol for the KD calculation. All the measurements were triplicated using separately prepared RBD substrates, buffers, and unused sensors. A consistent trend was obtained and the data set with the highest R2 value was reported.

MALDI-TOF analysis of intact glycoengineered S protein RBD

The MALDI-MS datasets for the glycoengineered RBD were acquired using a Bruker rapifleX MALDI-TOF instrument (Bruker Daltonics, Billerica, MA, United States) that was set to an average of 10,000 laser shots per sample at a laser energy value of 90% with a scan range of 20–220 kDa. The instrument was calibrated prior to measurements using: (i) the Bruker Protein Standard II calibration kit (Bruker Daltonics, Billerica, MA, United States), and (ii) a 0.2 mg/mL solution of Bovine Serum Albumin (Sigma Aldrich, MO) in 50:50 acetonitrile:water. The mass errors for the calibration data were confirmed by the differences between observed and theoretical masses for each protein, and the error values were found to be <100 ppm. Sinapic acid (Sigma Aldrich, MO) dissolved in 50:50 acetonitrile:water, at a concentration of 10 mg/mL with 0.1% 3-Nitrobenzylalcohol (Sigma Aldrich, MO) to aid in ionization, was used as a MALDI matrix for both instrument calibration and sample measurement purposes. The MALDI-TOF data was collected using a linear mode of detection with either a detector gain factor of 1.9X or a higher gain value of 10X.

Molecular dynamics simulations

Starting structures of the molecular systems were built using the graphical interface of YASARA (Krieger and Vriend 2014, 2015). The starting structures were assembled based on PDB entries 6vsb, 6m0j, and 6wps. In general, the systems were solvated in 0.9% NaCl solution (0.15 M) and simulations were performed at 310 K using periodic boundary conditions using the AMBER14 force field, which includes GLYCAM06 parameters for the carbohydrates (Kirschner et al. 2008; Maier et al. 2015). The box size was rescaled dynamically to maintain a water density of 0.996 g/mL. Simulations were performed using YASARA with GPU (Graphic Processing Unit) acceleration in “fast mode” (4 fs time step; Krieger and Vriend 2015) on “standard computing boxes” equipped e.g. with one 12-core i9 CPU and NVIDIA GeForce GTX 1080 Ti. Conformational Analysis Tools (CAT, http://www.md-simulations.de/CAT/) was used for the analysis of trajectory data, general data processing and generation of scientific plots. VMD was used to generate molecular graphics (Humphrey et al. 1996).

Plotting and graphic

Data plotting and curve fitting were done with GraphPad Prism 8. Figures were created by Adobe Illustrator. Protein 3-dimensional (3D) structures were constructed in PyMOL with the use of the published crystal models, PDB 6m0J and 7a92 (Benton et al. 2020; Lan et al. 2020). The protein surface electrostatic potential map was generated by the APBS Electrostatics plugin of PyMOL (Baker et al. 2001; Jurrus et al. 2018).

Authors’ contributions

Method development for glycan remodeling, binding affinity measurements, LC–MS analysis, and data analysis was performed by Y.-P.H. MALDI-ToF analysis was performed by D.M. and V.S. MD simulations were performed and analyzed by M.F. Y.-P.H and B.F. M. conceptualized the study. All the authors were involved in the design of the research and manuscript drafting.

Supplementary Material

Hsu_et_al_ESI_Final_GLYCO-2021-00158_cwac077
Click here for additional data file. (5.4MB, docx)
Hsu_et_al_SI-Glycan_Profile_GLYCO-2021-00158_cwac077
Click here for additional data file. (617.5KB, xlsx)

Acknowledgment

We thank the Merck Research Laboratory (MRL), Merck Postdoctoral Research Fellows Program, and Analytical Research and Development (AR&D) department for the financial support. We are also thankful to Erik Regalado and Jimmy DaSilva (Method Screening and Purification group) for technical assistance and advice; Jeffrey Moore and David Row (Protein Engineering group) for their kind suggestions on the experimental designs. We also thank Caroline McGregor and Ian Mangion for their support of this study in the AR&D department.

Contributor Information

Yen-Pang Hsu, Merck & Co., Inc., Merck Research Laboratories, Discovery Biologics, 320 Bent St., Cambridge, MA 02141, United States.

Martin Frank, Biognos AB, 417 05 Gothenburg, Sweden.

Debopreeti Mukherjee, Merck & Co., Inc., Merck Research Laboratories, Analytical Research and Development, 90 E. Scott Ave., Rahway, NJ 07065, United States.

Vladimir Shchurik, Merck & Co., Inc., Merck Research Laboratories, Analytical Research and Development, 90 E. Scott Ave., Rahway, NJ 07065, United States.

Alexey Makarov, Merck & Co., Inc., Merck Research Laboratories, Analytical Research and Development, 90 E. Scott Ave., Rahway, NJ 07065, United States.

Benjamin F Mann, Merck & Co., Inc., Merck Research Laboratories, Analytical Research and Development, 90 E. Scott Ave., Rahway, NJ 07065, United States.

Funding

Merck Research Laboratories & Merck Postdoctoral Research Fellows Program.

Conflict of interest statement

The authors declare no competing interests.

Data availability

All data are available upon request to the corresponding authors.

Reference

  1. Arnold  FH. Directed evolution: bringing new chemistry to life. Angew Chem Int Ed. 2018:57:4143–4148. [DOI] [PMC free article] [PubMed] [Google Scholar]
  2. Baig  AM. Chronic COVID syndrome: need for an appropriate medical terminology for long-COVID and COVID long-haulers. J Med Virol. 2021:93:2555–2556. [DOI] [PubMed] [Google Scholar]
  3. Baker  NA, Sept  D, Joseph  S, Holst  MJ, McCammon  JA. Electrostatics of nanosystems: application to microtubules and the ribosome. Proc Natl Acad Sci. 2001:98:10037–10041. [DOI] [PMC free article] [PubMed] [Google Scholar]
  4. Benton  DJ, Wrobel  AG, Xu  P, Roustan  C, Martin  SR, Rosenthal  PB, Skehel  JJ, Gamblin  SJ. Receptor binding and priming of the spike protein of SARS-CoV-2 for membrane fusion. Nature. 2020:588:327–330. [DOI] [PMC free article] [PubMed] [Google Scholar]
  5. Cai  Y, Zhang  J, Xiao  T, Peng  H, Sterling  SM, Walsh  RM, Rawson  S, Rits-Volloch  S, Chen  B. Distinct conformational states of SARS-CoV-2 spike protein. Science. 2020:369:1586–1592. [DOI] [PMC free article] [PubMed] [Google Scholar]
  6. Casalino  L, Gaieb  Z, Goldsmith  JA, Hjorth  CK, Dommer  AC, Harbison  AM, Fogarty  CA, Barros  EP, Taylor  BC, McLellan  JS  et al.  Beyond shielding: the roles of glycans in the SARS-CoV-2 spike protein. ACS Central Science. 2020:6:1722–1734. [DOI] [PMC free article] [PubMed] [Google Scholar]
  7. Cele  S, Gazy  I, Jackson  L, Hwa  S-H, Tegally  H, Lustig  G, Giandhari  J, Pillay  S, Wilkinson  E, Naidoo  Y  et al.  Escape of SARS-CoV-2 501Y.V2 from neutralization by convalescent plasma. Nature. 2021:593:142–146. [DOI] [PMC free article] [PubMed] [Google Scholar]
  8. Cooper  CA, Gasteiger  E, Packer  NH. GlycoMod--a software tool for determining glycosylation compositions from mass spectrometric data. Proteomics. 2001:1:340–349. [DOI] [PubMed] [Google Scholar]
  9. Corti  D, Misasi  J, Mulangu  S, Stanley  DA, Kanekiyo  M, Wollen  S, Ploquin  A, Doria-Rose  NA, Staupe  RP, Bailey  M  et al.  Protective monotherapy against lethal Ebola virus infection by a potently neutralizing antibody. Science. 2016:351:1339–1342. [DOI] [PubMed] [Google Scholar]
  10. Crispin  M, Ward  AB, Wilson  IA. Structure and immune recognition of the HIV glycan shield. Annu Rev Biophys. 2018:47:499–523. [DOI] [PMC free article] [PubMed] [Google Scholar]
  11. Cummings  PSRD. Structures common to different glycans. In: Varki  A, editors. Essentials of glycobiology. New York: Cold Spring Harbor Laboratory Press; 2017. pp. 161–178. [Google Scholar]
  12. Feig  M, Nawrocki  G, Yu  I, Wang  P-H, Sugita  Y. Challenges and opportunities in connecting simulations with experiments via molecular dynamics of cellular environments. J Phys Conf Ser. 2018:1036:012010. [DOI] [PMC free article] [PubMed] [Google Scholar]
  13. François  KO, Balzarini  J. The highly conserved glycan at asparagine 260 of HIV-1 gp120 is indispensable for viral entry*. J Biol Chem. 2011:286:42900–42910. [DOI] [PMC free article] [PubMed] [Google Scholar]
  14. Gasteiger  E, Gattiker  A, Hoogland  C, Ivanyi  I, Appel  RD, Bairoch  A. ExPASy: the proteomics server for in-depth protein knowledge and analysis. Nucleic Acids Res. 2003:31:3784–3788. [DOI] [PMC free article] [PubMed] [Google Scholar]
  15. Geijtenbeek  TBH, Gringhuis  SI. Signalling through C-type lectin receptors: shaping immune responses. Nat Rev Immunol. 2009:9:465–479. [DOI] [PMC free article] [PubMed] [Google Scholar]
  16. Grant  OC, Montgomery  D, Ito  K, Woods  RJ. Analysis of the SARS-CoV-2 spike protein glycan shield reveals implications for immune recognition. Sci Rep. 2020:10:14991. [DOI] [PMC free article] [PubMed] [Google Scholar]
  17. Guo  J, Tu  H, Atouf  F. Measurement of macro- and micro-heterogeneity of glycosylation in biopharmaceuticals: a pharmacopeia perspective. Future. Drug Des Discov. 2020:2:FDD48. [Google Scholar]
  18. Hirotsu  Y, Omata  M. Discovery of a SARS-CoV-2 variant from the P.1 lineage harboring K417T/E484K/N501Y mutations in Kofu, Japan. J Inf. 2021:82:276–316. [DOI] [PMC free article] [PubMed] [Google Scholar]
  19. Hoffmann  M, Kleine-Weber  H, Schroeder  S, Krüger  N, Herrler  T, Erichsen  S, Schiergens  TS, Herrler  G, Wu  N-H, Nitsche  A  et al.  SARS-CoV-2 cell entry depends on ACE2 and TMPRSS2 and is blocked by a clinically proven protease inhibitor. Cell. 2020:181:271–280.e278. [DOI] [PMC free article] [PubMed] [Google Scholar]
  20. Hoffmann  D, Mereiter  S, Jin  OY, Monteil  V, Elder  E, Zhu  R, Canena  D, Hain  L, Laurent  E, Grünwald-Gruber  C  et al.  Identification of lectin receptors for conserved SARS-CoV-2 glycosylation sites. EMBO J. 2021:40:e108375. [DOI] [PMC free article] [PubMed] [Google Scholar]
  21. Hsu  Y-P, Verma  D, Sun  S, McGregor  C, Mangion  I, Mann  BF. Successive remodeling of IgG glycans using a solid-phase enzymatic platform. Commun Biol. 2022:5:328. [DOI] [PMC free article] [PubMed] [Google Scholar]
  22. Huang  Y, Yang  C, Xu  X-f, Xu  W, Liu  S-W. Structural and functional properties of SARS-CoV-2 spike protein: potential antivirus drug development for COVID-19. Acta Pharmacol Sin. 2020:41:1141–1149. [DOI] [PMC free article] [PubMed] [Google Scholar]
  23. Huang  Y, Harris  BS, Minami  SA, Jung  S, Shah  PS, Nandi  S, McDonald  KA, Faller  R. SARS-CoV-2 spike binding to ACE2 is stronger and longer ranged due to glycan interaction. Biophys J. 2022:121:79–90. [DOI] [PMC free article] [PubMed] [Google Scholar]
  24. Humphrey  W, Dalke  A, Schulten  K. VMD: visual molecular dynamics. J Mol Graph. 1996:14(33-38):27–38. [DOI] [PubMed] [Google Scholar]
  25. Jurrus  E, Engel  D, Star  K, Monson  K, Brandi  J, Felberg  LE, Brookes  DH, Wilson  L, Chen  J, Liles  K  et al.  Improvements to the APBS biomolecular solvation software suite. Protein Sci. 2018:27:112–128. [DOI] [PMC free article] [PubMed] [Google Scholar]
  26. Kirschner  KN, Yongye  AB, Tschampel  SM, González-Outeiriño  J, Daniels  CR, Foley  BL, Woods  RJ. GLYCAM06: a generalizable biomolecular force field. Carbohydrates J Comput Chem. 2008:29:622–655. [DOI] [PMC free article] [PubMed] [Google Scholar]
  27. Krieger  E, Vriend  G. YASARA View - molecular graphics for all devices - from smartphones to workstations. Bioinformatics. 2014:30:2981–2982. [DOI] [PMC free article] [PubMed] [Google Scholar]
  28. Krieger  E, Vriend  G. New ways to boost molecular dynamics simulations. J Comput Chem. 2015:36:996–1007. [DOI] [PMC free article] [PubMed] [Google Scholar]
  29. Lan  J, Ge  J, Yu  J, Shan  S, Zhou  H, Fan  S, Zhang  Q, Shi  X, Wang  Q, Zhang  L  et al.  Structure of the SARS-CoV-2 spike receptor-binding domain bound to the ACE2 receptor. Nature. 2020:581:215–220. [DOI] [PubMed] [Google Scholar]
  30. Lempp  FA, Soriaga  LB, Montiel-Ruiz  M, Benigni  F, Noack  J, Park  Y-J, Bianchi  S, Walls  AC, Bowen  JE, Zhou  J  et al.  Lectins enhance SARS-CoV-2 infection and influence neutralizing antibodies. Nature. 2021:598:342–347. [DOI] [PubMed] [Google Scholar]
  31. Letko  M, Marzi  A, Munster  V. Functional assessment of cell entry and receptor usage for SARS-CoV-2 and other lineage B betacoronaviruses. Nat Microbiol. 2020:5:562–569. [DOI] [PMC free article] [PubMed] [Google Scholar]
  32. Leung  K, Shum  MH, Leung  GM, Lam  TT, Wu  JT. Early transmissibility assessment of the N501Y mutant strains of SARS-CoV-2 in the United Kingdom, October to November 2020. Euro Surveill. 2021:26. [DOI] [PMC free article] [PubMed] [Google Scholar]
  33. Ma  B, Guan  X, Li  Y, Shang  S, Li  J, Tan  Z. Protein glycoengineering: an approach for improving protein properties. Front Chem. 2020:8. [DOI] [PMC free article] [PubMed] [Google Scholar]
  34. Maier  JA, Martinez  C, Kasavajhala  K, Wickstrom  L, Hauser  KE, Simmerling  C. ff14SB: improving the accuracy of protein side chain and backbone parameters from ff99SB. J Chem Theory Comput. 2015:11:3696–3713. [DOI] [PMC free article] [PubMed] [Google Scholar]
  35. Marzi  A, Gramberg  T, Simmons  G, Möller  P, Rennekamp  AJ, Krumbiegel  M, Geier  M, Eisemann  J, Turza  N, Saunier  B  et al.  DC-SIGN and DC-SIGNR interact with the glycoprotein of Marburg virus and the S protein of severe acute respiratory syndrome coronavirus. J Virol. 2004:78:12090–12095. [DOI] [PMC free article] [PubMed] [Google Scholar]
  36. Matrosovich  M, Herrler  G, Klenk  HD. Sialic acid receptors of viruses. Top Curr Chem. 2015:367:1–28. [DOI] [PMC free article] [PubMed] [Google Scholar]
  37. Mehdipour  AR, Hummer  G. Dual nature of human ACE2 glycosylation in binding to SARS-CoV-2 spike. Proc Natl Acad Sci. 2021:118:e2100425118. [DOI] [PMC free article] [PubMed] [Google Scholar]
  38. Mizushima  T, Yagi  H, Takemoto  E, Shibata-Koyama  M, Isoda  Y, Iida  S, Masuda  K, Satoh  M, Kato  K. Structural basis for improved efficacy of therapeutic antibodies on defucosylation of their Fc glycans. Genes Cells. 2011:16:1071–1080. [DOI] [PMC free article] [PubMed] [Google Scholar]
  39. Pinto  D, Park  Y-J, Beltramello  M, Walls  AC, Tortorici  MA, Bianchi  S, Jaconi  S, Culap  K, Zatta  F, De Marco  A  et al.  Cross-neutralization of SARS-CoV-2 by a human monoclonal SARS-CoV antibody. Nature. 2020:583:290–295. [DOI] [PubMed] [Google Scholar]
  40. Reily  C, Stewart  TJ, Renfrow  MB, Novak  J. Glycosylation in health and disease. Nat Rev Nephrol. 2019:15:346–366. [DOI] [PMC free article] [PubMed] [Google Scholar]
  41. Shajahan  A, Archer-Hartmann  S, Supekar  NT, Gleinich  AS, Heiss  C, Azadi  P. Comprehensive characterization of N- and O- glycosylation of SARS-CoV-2 human receptor angiotensin converting enzyme 2. Glycobiology. 2020a:31:410–424. [DOI] [PMC free article] [PubMed] [Google Scholar]
  42. Shajahan  A, Supekar  NT, Gleinich  AS, Azadi  P. Deducing the N- and O-glycosylation profile of the spike protein of novel coronavirus SARS-CoV-2. Glycobiology. 2020b:30:981–988. [DOI] [PMC free article] [PubMed] [Google Scholar]
  43. Shang  J, Wan  Y, Luo  C, Ye  G, Geng  Q, Auerbach  A, Li  F. Cell entry mechanisms of SARS-CoV-2. Proc Natl Acad Sci. 2020a:117:11727–11734. [DOI] [PMC free article] [PubMed] [Google Scholar]
  44. Shang  J, Ye  G, Shi  K, Wan  Y, Luo  C, Aihara  H, Geng  Q, Auerbach  A, Li  F. Structural basis of receptor recognition by SARS-CoV-2. Nature. 2020b:581:221–224. [DOI] [PMC free article] [PubMed] [Google Scholar]
  45. Starr  TN, Czudnochowski  N, Liu  Z, Zatta  F, Park  Y-J, Addetia  A, Pinto  D, Beltramello  M, Hernandez  P, Greaney  AJ  et al.  SARS-CoV-2 RBD antibodies that maximize breadth and resistance to escape. Nature. 2021:597:97–102. [DOI] [PMC free article] [PubMed] [Google Scholar]
  46. Sternberg  A, Naujokat  C. Structural features of coronavirus SARS-CoV-2 spike protein: targets for vaccination. Life Sci. 2020:257:118056–118056. [DOI] [PMC free article] [PubMed] [Google Scholar]
  47. Stewart-Jones  GB, Soto  C, Lemmin  T, Chuang  GY, Druz  A, Kong  R, Thomas  PV, Wagh  K, Zhou  T, Behrens  AJ  et al.  Trimeric HIV-1-Env structures define glycan shields from clades n and G. Cell. 2016:165:813–826. [DOI] [PMC free article] [PubMed] [Google Scholar]
  48. Struwe  WB, Robinson  CV. Relating glycoprotein structural heterogeneity to function - insights from native mass spectrometry. Curr Opin Struct Biol. 2019:58:241–248. [DOI] [PMC free article] [PubMed] [Google Scholar]
  49. Tarentino  AL, Plummer  TH  Jr. Letter to the glyco-forum: substrate specificity of Flavobacterium meningosepticum Endo F2 and Endo F3: purity is the name of the game. Glycobiology. 1994:4:771–773. [DOI] [PubMed] [Google Scholar]
  50. Tate  MD, Job  ER, Deng  YM, Gunalan  V, Maurer-Stroh  S, Reading  PC. Playing hide and seek: how glycosylation of the influenza virus hemagglutinin can modulate the immune response to infection. Viruses. 2014:6:1294–1316. [DOI] [PMC free article] [PubMed] [Google Scholar]
  51. Thépaut  M, Luczkowiak  J, Vivès  C, Labiod  N, Bally  I, Lasala  F, Grimoire  Y, Fenel  D, Sattin  S, Thielens  N  et al.  DC/L-SIGN recognition of spike glycoprotein promotes SARS-CoV-2 trans-infection and can be inhibited by a glycomimetic antagonist. PLoS Pathog. 2021:17:e1009576. [DOI] [PMC free article] [PubMed] [Google Scholar]
  52. Trimble  RB, Tarentino  AL. Identification of distinct endoglycosidase (endo) activities in Flavobacterium meningosepticum: endo F1, endo F2, and endo F3. Endo F1 and endo H hydrolyze only high mannose and hybrid glycans. J Biol Chem. 1991:266:1646–1651. [PubMed] [Google Scholar]
  53. van  Liempt  E, Bank CM, Mehta  P, Garciá-Vallejo  JJ, Kawar  ZS, Geyer  R, Alvarez  RA, Cummings  RD, Kooyk  Y, van  Die  I. Specificity of DC-SIGN for mannose- and fucose-containing glycans. FEBS Lett. 2006:580:6123–6131. [DOI] [PubMed] [Google Scholar]
  54. Walls  AC, Tortorici  MA, Frenz  B, Snijder  J, Li  W, Rey  FA, DiMaio  F, Bosch  B-J, Veesler  D. Glycan shield and epitope masking of a coronavirus spike protein observed by cryo-electron microscopy. Nat Struct Mol Biol. 2016:23:899–905. [DOI] [PMC free article] [PubMed] [Google Scholar]
  55. Walls  AC, Park  Y-J, Tortorici  MA, Wall  A, McGuire  AT, Veesler  D. Structure, function, and antigenicity of the SARS-CoV-2 spike glycoprotein. Cell. 2020:181:281–292.e286. [DOI] [PMC free article] [PubMed] [Google Scholar]
  56. Wang  X, Cao  R, Zhang  H, Liu  J, Xu  M, Hu  H, Li  Y, Zhao  L, Li  W, Sun  X  et al.  The anti-influenza virus drug, arbidol is an efficient inhibitor of SARS-CoV-2 in vitro. Cell Discovery. 2020:6:28. [DOI] [PMC free article] [PubMed] [Google Scholar]
  57. Watanabe  Y, Bowden  TA, Wilson  IA, Crispin  M. Exploitation of glycosylation in enveloped virus pathobiology. Biochim Biophys Acta Gen Subj. 2019:1863:1480–1497. [DOI] [PMC free article] [PubMed] [Google Scholar]
  58. Watanabe  Y, Allen  JD, Wrapp  D, McLellan  JS, Crispin  M. Site-specific glycan analysis of the SARS-CoV-2 spike. Science. 2020a:369:330–333. [DOI] [PMC free article] [PubMed] [Google Scholar]
  59. Watanabe  Y, Berndsen  ZT, Raghwani  J, Seabright  GE, Allen  JD, Pybus  OG, McLellan  JS, Wilson  IA, Bowden  TA, Ward  AB  et al.  Vulnerabilities in coronavirus glycan shields despite extensive glycosylation. Nat Commun. 2020b:11:2688. [DOI] [PMC free article] [PubMed] [Google Scholar]
  60. Wong  CH. Protein glycosylation: new challenges and opportunities. J Org Chem. 2005:70:4219–4225. [DOI] [PubMed] [Google Scholar]
  61. Woo  H, Park  S-J, Choi  YK, Park  T, Tanveer  M, Cao  Y, Kern  NR, Lee  J, Yeom  MS, Croll  TI  et al.  Developing a fully glycosylated full-length SARS-CoV-2 spike protein model in a viral membrane. J Phys Chem B. 2020:124:7128–7137. [DOI] [PMC free article] [PubMed] [Google Scholar]
  62. Xie  Y, Karki  CB, Du  D, Li  H, Wang  J, Sobitan  A, Teng  S, Tang  Q, Li  L. Spike proteins of SARS-CoV and SARS-CoV-2 utilize different mechanisms to bind with human ACE2. Front Mol Biosci. 2020:7. [DOI] [PMC free article] [PubMed] [Google Scholar]
  63. Yang  T-J, Chang  Y-C, Ko  T-P, Draczkowski  P, Chien  Y-C, Chang  Y-C, Wu  K-P, Khoo  K-H, Chang  H-W, Hsu  S-TD. Cryo-EM analysis of a feline coronavirus spike protein reveals a unique structure and camouflaging glycans. Proc Natl Acad Sci. 2020:117:1438–1446. [DOI] [PMC free article] [PubMed] [Google Scholar]
  64. Zhao  P, Praissman  JL, Grant  OC, Cai  Y, Xiao  T, Rosenbalm  KE, Aoki  K, Kellman  BP, Bridger  R, Barouch  DH  et al.  Virus-receptor interactions of glycosylated SARS-CoV-2 spike and human ACE2 receptor. Cell Host & Microbe. 2022:28:586–601. [DOI] [PMC free article] [PubMed] [Google Scholar]

Associated Data

This section collects any data citations, data availability statements, or supplementary materials included in this article.

Supplementary Materials

Hsu_et_al_ESI_Final_GLYCO-2021-00158_cwac077
Click here for additional data file. (5.4MB, docx)
Hsu_et_al_SI-Glycan_Profile_GLYCO-2021-00158_cwac077
Click here for additional data file. (617.5KB, xlsx)

Data Availability Statement

All data are available upon request to the corresponding authors.

Articles from Glycobiology are provided here courtesy of Oxford University Press

RESOURCES