Glycan Masking of Epitopes in the NTD and RBD of the Spike Protein Elicits Broadly Neutralizing Antibodies Against SARS-CoV-2 Variants

Glycan-masking the vaccine antigen by mutating the undesired antigenic sites with an additional N-linked glycosylation motif can refocus B-cell responses to desired epitopes, without affecting the antigen’s overall-folded structure. This study examined the impact of glycan-masking mutants of the N-terminal domain (NTD) and receptor-binding domain (RBD) of SARS-CoV-2, and found that the antigenic design of the S protein increases the neutralizing antibody titers against the Wuhan-Hu-1 ancestral strain and the recently emerged SARS-CoV-2 variants Alpha (B.1.1.7), Beta (B.1.351), and Delta (B.1.617.2). Our results demonstrated that the use of glycan-masking Ad-S-R158N/Y160T in the NTD elicited a 2.8-fold, 6.5-fold, and 4.6-fold increase in the IC-50 NT titer against the Alpha (B.1.1.7), Beta (B.1.351) and Delta (B.1.617.2) variants, respectively. Glycan-masking of Ad-S-D428N in the RBD resulted in a 3.0-fold and 2.0-fold increase in the IC-50 neutralization titer against the Alpha (B.1.1.7) and Beta (B.1.351) variants, respectively. The use of glycan-masking in Ad-S-R158N/Y160T and Ad-S-D428N antigen design may help develop universal COVID-19 vaccines against current and future emerging SARS-CoV-2 variants.


INTRODUCTION
Severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2) is a coronavirus that can evolve within an infected host as a quasi-species with many mutations (within-host variants) arising for virus survival (1). Multiple SARS-CoV-2 variants have already emerged and are circulating globally, four of which are a cause of serious public health concern, also known as variants of concern: (i) Alpha (B.  (4)(5)(6). In particular, the RBD mutations K417, L452, E484 and N501 may directly form a new interaction with the human angiotensin-converting enzyme 2 (ACE2) receptor (7,8). Thus, the emerging SARS-CoV-2 B.1.351 variant can evade antibodymediated immunity without affecting the virus fitness and disease, as recently reported using convalescent plasma, vaccine-induced sera, and monoclonal antibodies (9)(10)(11)(12)(13)(14)(15). Whether the current authorized or approved vaccines can still provide effective protection against current and future emerging SARS-CoV-2 variants remains unclear.
Glycan-masking the vaccine antigen by mutating the undesired antigenic sites with an additional N-linked glycosylation motif can refocus the B-cell responses to the desired epitopes, without affecting the antigen's overall-folded structure (16,17). This antigen design strategy has been used to develop vaccines against human immunodeficiency virus (HIV)-1 (18)(19)(20)(21)(22)(23), influenza virus (24)(25)(26)(27), dengue and Zika viruses (28), and Middle East respiratory syndrome coronavirus (29). In this study, we used an adenovirus (Ad) vector encoding the fulllength S gene of the SARS-CoV-2 Wuhan-Hu-1 isolate with a series of site-specific glycan-masking mutations on the Nterminal domain (NTD) and RBD in a mouse immunization model, and then investigated the breadth of neutralizing antibodies elicited against SARS-CoV-2 and its Alpha (B.1.1.

Cell Lines
HEK293A and HEK293T cells were obtained from the Bioresource Collection and Research Center (BCRC), Taiwan. These cells were grown in Dulbecco's modified Eagle medium (DMEM) (Thermo Scientific) supplemented with 10% fetal bovine serum (FBS) (Gibco) and 100 units/ml penicillin/ streptomycin (P/S), and maintained in an incubator at 37°C with 5% CO 2 .

Selection of Glycan-Masking Sites From the SARS-CoV-2 S Protein Structure
Selection of the glycan-masking sites was based on the 3Dstructure of SARS-CoV-2 S protein (PDB ID:7C2L). The exposed loops or the protruding sites of the exposed loops on the NTD and RBD of the S protein were examined using PyMol (The PyMol Molecular Graphics System, version 4.0; Schrödinger, LLC). Glycan-masking sites that are less than 5 Å from the native glycans and RBD were discarded.

Preparation of Ad Vectors Expressing SARS-CoV-2 S Gene and Glycan-Masking Mutants
The human codon-optimized S gene of SARS-CoV-2 (Wuhan-Hu-1 isolate, accession number MN908947.3) was obtained from GenScript. Site-directed mutagenesis was used to produce the glycan-masking S mutant genes, with the addition of an N-linked glycosylation motif at the S protein residues 135N/N137T, R158N/Y160T, N354/K356T, N370/A372T, G413N, D428N and H519N/P521T. Wild-type S and glycan-masking S genes were first cloned into the pENTR1A vector (Invitrogen), and then cloned into the adenoviral plasmid pAd/CMV/V5-DEST (Invitrogen) using LR ClonaseTM II Enzyme Mix (Invitrogen) to produce the Ad plasmid expressing SARS-CoV-2 S gene. To obtain Ad particles, the Ad plasmids were cleaved with Pac I restriction enzyme to expose the inverted terminal repeats and then transfected into 293A cells separately using TurboFect transfection reagent (Fermentas). After 10-15 d, once the cytopathic effects were visible, the transfected cells and culture media were collected. The cells were disrupted by means of three freeze-thaw cycles to release the intracellular viral particles, and the supernatants of the cell lysates were collected by centrifugation (3000 rpm, 15 min, 4°C) to obtain the Ad vectors expressing the SARS-Co-V-2 S proteins. To prepare higher titers, the virus was concentrated using a 30-kDa Amicon Ultra-15 Centrifugal Filter (Millipore). The viral stocks were stored at -80°C. To determine the Ad titers, HEK293A cells were seeded into 6-well plates at a density of 10 6 cells/well and incubated at 37°C overnight. The 10-fold serially diluted Ad stocks were then added to each well at 37°C for 24 h. Next, the media containing the diluted Ad vectors were removed, and 3 mL/well of DMEM containing 0.4% agarose and 100 U/ml P/S was added to the 6-well plates. The plaques were visibly quantified 7-10 d after the cells were infected with Ad vectors, and the pfu count was noted.

Mouse Immunization
Groups of female BALB/c mice (6 to 8 weeks old) (n=5 per group) were obtained from the National Laboratory Animal Center, Taipei , and allowed to coat 96-well plates at a concentration of 2mg/ mL in coating buffer (100mL/well) overnight at 4°C. Coating buffers were aspirated and washed three times with PBS containing 0.05% Tween 20 (PBST). Each well was blocked with 200mL blocking buffer (1% BSA in PBST) at room temperature for 2 h. Heat-inactivated serum samples were prediluted 1:1000, followed by 2-fold serial dilution in dilution buffer (0.05% tween 20 + 1% BSA in PBST). The plates were washed three times with 300 µL PBST (PBS with 0.05% Tween- 20), and then blocked with 200 µL PBS buffer plus 1% BSA for 2 h at room temperature, followed by three additional washes with 300 µL PBST. Following this, the plates were incubated with 100 µl of HRP) conjugated anti-mouse IgG antibody (1:30000 in dilution buffer) for 1 h at room temperature. After three additional washes with 300 µL PBST, 100 µL of TMB substrate (BioLegend) was added to each well and incubated in the dark for 15 min. The reaction was stopped by the addition of 100 µL of 2 N H 2 SO 4 . The optical density at 450 nm was measured using a TECAN spectrophotometer. The end-point titration values were calculated in terms of a final serial dilution higher than 0.2 optical density value.

SARS-CoV-2 Pseudotyped Lentivirus Neutralization Assay
To produce SARS-CoV-2 pseudoviruses, a plasmid expressing the full-length S protein (Wuhan-Hu-1, B.1.1.7, or B.1.351) of SARS-CoV-2 was co-transfected into HEK293T cells with packaging and reporter plasmids pCMVD 8.91 and pLAS2w.FLuc.Ppuro (RNAi Core, Academia Sinica), using TransIT-LT1 transfection reagent (Mirus Bio). The medium was harvested and concentrated at 48 h post-transfection, followed by estimation of the pseudovirus titer in terms of the luciferase activity of SARS-CoV2-Spp transduction. Serum samples were serially diluted and incubated with 1,000 TU of SARS-CoV-2-pseudotyped lentivirus in DMEM (supplemented with 1% FBS and 100 U/mL P/S) for 1 h at 37°C. The mixture was then inoculated with an equal volume of 10,000 HEK-293T cells stably expressing the ACE2 gene in 96-well plates, which were seeded one day before infection. The culture medium was replaced with fresh complete DMEM (supplemented with 10% FBS, 100 U/mL P/S) at 16 h post-infection and the cells were then continuously cultured for another 48 h before being subjected to a luciferase assay (Promega Bright-GloTM Luciferase Assay System). The percentage of inhibition was calculated as the ratio of the loss of luciferase readout (RLU) in the presence of serum to that of the no serum control. The formula used for the calculation was (RLU Control -RLU Serum)/RLU Control. Neutralization titers (IC-50) were measured as the reciprocal of the serum dilution required to obtain a 50% reduction in RLU compared to a control containing the SARS-CoV-2 S-pseudotyped lentivirus only. Neutralization curves and IC-50 values were analyzed using the GraphPad Prism 5 Software.

Statistical Analyses
Statistical tests for multiple comparison were performed for all groups (except for the PBS control) in case of the ELISA data. The results were analyzed using the nonparametric Kruskal-Wallis test, with corrected Dunn's multiple comparison test, using GraphPad Prism v6.01. Statistical significance has been expressed as follows: *p < 0.05; **p < 0.01; and ***p < 0.001. Neutralization curves were fitted based on the equation of nonlinear regression log (inhibitor) vs. normalized responsevariable slope using GraphPad Prism v6.01. The IC-50 values of the neutralization were obtained from the fitting curves using GraphPad Prism v6.01

Design of Engineered Glycan-Masking S Antigens in the NTD and RBD for Ad Vector Immunization
The S protein of SARS-CoV-2 is trimeric, and each monomer comprises of S1 and S2 subunits (30)(31)(32). The S1 subunit contains NTD and RBD. To design glycan-masking S antigen(s) for immunization, we used an Ad vector encoding the full-length S gene of the SARS-CoV-2 Wuhan-Hu isolate, by introducing a series of N-linked glycosylation motifs into the S1 region of the S protein, to refocus the antibody responses to the RBD ( Figure 1A). The sites of glycan-masking were introduced not only in the RBD, but also in the NTD, as RBD and NTD may spatially interact with each other in the quaternary structure of the intact trimeric S protein ( Figure 1B). The exposed loops or the protruding sites of the exposed loops on the NTD and RBD of the 3-D S protein structure (PDB ID: 7C2L) were chosen for the addition on the glycan-masking sites. Seven glycan-masking N-glycan sites were engineered in the Ad-S vector: Figure 1B). To characterize the glycan-masking mutations on the S protein, the lysates of HEK293A cells infected with each Ad-S vector were analyzed using 8% SDS-PAGE gels, followed by western blotting with an S1-specific polyclonal antibody. The results indicated the presence of S and S1 in the cell lysates of HEK293A cells infected with Ad-S, Ad-S-F135N/N137T, Ad-S-R158N/Y160T, Ad-S-N354/K356T, Ad-S-N370/A372T, Ad-S-G413N, Ad-S-D428N, and Ad-S-H519N/P521T ( Figure 2).
Anti-S, Anti-RBD, and Pseudo-Neutralizing Antibody Titers Against the Wuhan-Hu-1 Ancestral Strain To investigate the antibody responses elicited by these Ad-S vectors, groups of BALB/c mice (n=5 per group) were intramuscularly immunized with two doses of each Ad vector at 5 x 10 7 pfu per dose for the first set of immunization experiments and at 1 x 10 8 pfu per dose for the second set of immunization experiments all in a 3-week interval; PBSimmunized animals were used as control. Sera from these immunized groups were collected after 2 weeks of the second dose and analyzed for the levels of anti-S, anti-RBD, and neutralizing antibodies against the original SARS-CoV-2 Wuhan-Hu-1 isolate. Two separate sets of immunization experiments were conducted in this study: (i) with Ad-S, Ad-S-  Figure 4C), with an approximately 2.8-fold increase in the neutralization IC-50 titer, as compared to that of the wild-type Ad-S-immunized group ( Figure 4D). In the second set of immunization experiments in the Ad-S, Ad-S-N354/K356T, Ad-S-G413N, and Ad-S-D428N-immunized groups, no significant differences were observed in the anti-S1 and anti-RBD titers among these four Ad immunization groups (Figures 4E, F). However, the Ad-S-D428N-immunized group displayed more potent neutralization against the Alpha (B.1.1.7) variant than the three other groups ( Figure 4G), resulting in a 3.0-fold increase in the neutralization IC-50 titer, as compared to that of the wildtype Ad-S-immunized group ( Figure 4H). Therefore, the glycanmasking Ad-S-R158N/Y160T in NTD and glycan-masking Ad-S-D428N in RBD were found to elicit increased titers of neutralizing antibodies against the Alpha (B.  2.8-fold higher in the glycan-masking Ad-S-R158N/Y160T and Ad-S-N370/A372T-immunized groups, respectively ( Figure 5D).
In the second set of immunization experiments in the Ad-S, Ad-S-N354/K356T, Ad-S-G413N, and Ad-S-D428N-immunized groups, no significant differences were observed in the anti-S1 and anti-RBD titers among these four Ad immunization groups ( Figures 5E, F). The Ad-S-D428N-immunized group was more potent in neutralizing the Beta (B.1.351) variant than the other three groups (Figure 5G), resulting in a 2.0-fold increase in the neutralization IC-50 titer, as compared to that of the wild-type Ad-S-immunized group ( Figure 5H). Therefore, immunization with the glycan-masking Ad-S-R158N/Y160T in NTD and glycan-masking Ad-S-N370/A372T and Ad-S-D428N in RBD were more potent than that with the wild-type Ad-S in eliciting neutralizing antibodies against the Beta (B.1.351) variant.
In the case of Delta (B.1.617.2) variant, we found that the anti-S1 IgG titers in the Ad-S-F135N/N137T-immunized group were lower than the wild type Ad-S and Ad-S-R158N/Y160T,immunized groups ( Figure 6A). No significant differences in anti-RBD IgG titers were observed among these Ad-immunized groups in the first set of experiments ( Figure 6B). Both the Ad-S-F135N/N137T and Ad-S-R158N/Y160T-immunized groups elicited more potent pseudo-neutralization against the Delta (B.1.167.2) variant, as compared to the Ad-S, Ad-S-N370/ A372T, and Ad-S-H519N/P521T-immunized groups, in the first set of immunization experiments ( Figure 6C); an approximately 3.7-fold and 4.6-fold increase in the neutralization IC-50 titer was found for the glycan-masking Ad-S-F135N/N137T and Ad-S-R158N/Y160T groups, respectively ( Figure 6D). In the second set of immunization experiments among the Ad-S, Ad-S-N354/K356T, Ad-S-G413N, and Ad-S-D428N-immunized groups, the pseudo-neutralization curves of the three glycan-masking groups against the Delta (B.167.2) variant were less potent than that of the wild-type Ad-S-immunized group ( Figure 6G) with a reduced IC-50 titers to 0.46, 0.23, and 0.46 fold, respectively ( Figure 6H). Therefore, only the glycan-masking Ad-S-F135N/N137T and Ad-S-R158N/ Y160T in NTD elicited more potent neutralizing antibody titers against the Delta (B.1.617.2) variant.

Comparison of Neutralization IC-50 Titers Elicited by Glycan-Masking Ad-S Mutants Against the Wuhan-Hu-1 Ancestral Strain
To compare these results, the neutralizing IC-50 titers from the two separate sets of immunization experiments were normalized to the titer elicited by the wild-type Ad-S against the Wuhan-Hu-1 ancestral strain from. In the first set of immunization experiments, the neutralization IC-50 titers elicited in the glycan-masking Ad-S-R158N/Y160T-immunized group showed a 2.5-fold increase against the Wuhan-Hu-1 ancestral strain, a 1.

DISCUSSION
This study reported the use of glycan-masking antigen design by selective incorporation of an N-linked glycosylation motif on the NTD and RBD in the SARS-CoV-2 S gene encoded by the Ad vector, to develop universal COVID-19 vaccines. It has been recently confirmed that the S protein of SARS-CoV-2 is heavily glycosylated, with approximately 22 potential N-glycan sites on each S monomer (30,32,33). We introduced seven separate Nlinked glycosylation sites into the S glycoprotein, S-F135N/ N137T, S-R158N/Y160T, Ad-S-N354/K356T, S-N370/A372T, Ad-S-G413N, Ad-S-D428N and S-H519N/P521T, in the NTD and RBD. However, we were unable to demonstrate the addition of a single N-glycosylation site for these glycan-masking Ad-S mutants using SDS-PAGE gel in western blots. It is possible that SDS-PAGE in the conditions used in this work does not have the resolution to show the addition of a single N-glycosylation site when there are already 22 sites. A more advanced use of the selected mutants would warrant thorough site-specific characterization by mass spectrometry. It is also possible that glycan-masking mutations may also affect the S protein stabilization for cell surface expression, S/S1 cleavage, and surface S expression. Thus, there is a need for further characterization of these glycan-masking mutants, particularly Ad-S-R158N/Y160T and Ad-S-D428N expressed S proteins.
Our results showed that the glycan-masking Ad-S-R158N/ Y160T at the N3 loop in the NTD and the glycan-masking Ad-S-N354/K356T at the C-3 loop in the RBD (Supplementary Figure 1) elicited a potent neutralizing antibody response against the Wuhan-Hu-1 ancestral strain (Figure 3). Selection of these glycan-masking sites in this investigation was based on visual inspection of the 3-D S protein structure (PDB ID: 7C2L) for the exposed loops or the protruding sites of the exposed loops in NTD and RBD of the S1 subunit (Supplementary Figure 1). The increased IC-50 NT titers against the Wuhan-Hu NTD were found to increase the neutralization titers against the Delta (B.1.617.2) variant ( Figure 6D). Therefore, only the glycanmasking Ad-S-R158N/Y160T in NTD elicited broadly neutralizing antibody titers against Alpha (B.1.1.7), Beta (B.1.351), and Delta (B.1.617.2) variants. It is possible that the refocused antibodies using the glycan-masking Ad-S-R158N/Y160T antigen may target the NTD neutralizing epitopes in the N3 and N5 loops as recently reported (34,35). One NTD neutralizing epitope has been shown to associate with the hydrophilic interactions of K147 and K150, the hydrophobic and/or p-p interactions of Y145 and W152, and the Nlinked glycosylation site of N149 on the N3 loop of the NTD (34). It is also possible that the glycan-masking R158N/Y160T in the NTD interacts spatially with the RBD of another S1 monomer to affect the RBD up and down conformational structures (36,37). The C-type lectins such as L-SIGN and DC-SIGN have been shown to function as attachment receptors by enhancing ACE2-mediated infection, and monoclonal antibodies to NTD or the RBD conserved site can effectively block lectin-facilitated infection (38). Additionally, the unique N-glycan on N149 of NTD can directly bind to the L-SIGN/ DC-SIGN lectins as a non-ACE2 receptor for SARS-CoV-2 virus infection (39). Another report was shown for the N92 glycan on NTD that can enhance the binding to the L-SIGN lectin to interact with the ACE2 receptor to further facilitate SARS-CoV-2 virus entry (40). It is possible that the glycan-masking Ad-S-R158N/ Y160T site which is nearby the N3 loop on NTD (Supplementary Figure 1) can enhance targeting these epitopes to elicit neutralizing antibodies to block the L-SIGN/DC-SIGN receptor binding and/or the interaction between the L-SIGN/DC-SIGN lectin co-receptor with the ACE2 receptor binding for SARS-CoV-2 infection. Our present findings demonstrated that the glycan-masking Ad-S-R158N/Y160T in NTD resulted in a 2.8-fold, 6.5-fold, and 4.6-fold increase, respectively, in the IC-50 titers against the Alpha (B.  Figure 1). For the Alpha (B.1.1.7) and Beta (B.1.351) variants, the del 69-70, del 144, and del 242-244 deletions in NTD and the K417N/T, E484K, and N501Y mutations in RBD have been shown to increase ACE2 binding affinity and evade antibody-mediated immunity (7)(8)(9)(10)(11)(12)(13)(14)(15). It is likely that selective pressures on the NTD and RBD epitopes of the S1 subunit may ultimately result in immune-evasion variants. Our present findings demonstrated the use of glycanmasking mutations in the neutralization-sensitive NTD and RBD epitopes of S1 subunit can refocus antibody responses to the broadly neutralizing epitope domains to overcome the immuneevasion variants. Therefore, glycan-masking the site-specific NTD and RBD epitopes may help develop universal COVID-19 vaccines against current and future emerging SARS-CoV-2 variants.

DATA AVAILABILITY STATEMENT
The raw data supporting the conclusions of this article will be made available by the authors, without undue reservation.

ETHICS STATEMENT
All experiments were conducted in accordance with the guidelines of the Laboratory Animal Center of the National Tsing Hua University (NTHU). Animal use protocols were reviewed and approved by the NTHU Institutional Animal Care and Use Committee (approval no. 109047).