Characterization and immunogenicity assessment of MERS-CoV pre-fusion spike trimeric oligomers as vaccine immunogen

Cloning, expression, and purification of MERS-CoV prefusion spike ectodomain from Expi293F cells

A human-codon-optimized gene encoding MERS-CoV spike (England 1 strain, YP_009047204.1) from residues 16–1291, flanked by a signal peptide at the N-terminal and with a T4 fibritin trimerization domain and 8×His-tag at C-terminal, was synthesized and subcloned into the eukaryotic expression vector pDNA 3.1 with restriction sites for BamHI and XbaI.¹⁹ The S1/S2 furin-recognition site 748-RSVR-751 was mutated to ASVG to produce a single-chain S₀ protein resembling the prefusion molecule. Along with this, two proline mutations at residues V1060 and L1061 were introduced for stabilization in the recombinant construct. Plasmids encoding human codon-optimized MERS-CoV spike were synthesized by Thermo Fisher Scientific, USA (Life Technologies Corporation, USA).

For suspension cell culture, Expi293F cells were maintained in suspension cultures in Expi Expression Medium ((#A1435101, Thermo Fisher Scientific, USA) and transfection was conducted using the ExpiFectamine 293 Transfection Kit (#A14524, Thermo Fisher Scientific) according to the manufacturer’s instructions. The supernatant was harvested after 5–7 days of post-transfection or when the viability reached < 60%. Cultures were centrifuged and supernatant filtered to remove cellular debris before IMAC protein purification using Ni-NTA resin (#30210, QIAGEN, India) and an automated AKTA purification system (Akta Pure 25, Cytiva lifesciences). The eluted protein was then dialyzed against PBS at 4°C and thereafter concentrated with a 50 ml amicon filter with a 30K cutoff to get the final protein. The purity of purified M-CoV-S protein was analyzed by 12% resolving gel on SDS-PAGE and further staining by Coomassie blue stain.

Western blotting and blue native gel analysis

The purified M-CoV-S protein was resolved on a 12% SDS-PAGE and transferred to a Millipore polyvinylidene fluoride (PVDF) membrane. The membrane was blocked in blocking buffer for 2 hours at 37°C followed by incubation with anti-MERS-CoV-S mouse polyclonal sera (1:5000) and commercial MERS-RBD (#10621-CV, R&D Systems) (1:1000) at 4°C for overnight. The membrane was washed with washing buffer (0.1% Tween 20 in PBS) and further developed with anti-mouse/anti-human HRP conjugated secondary antibody (#115-035-003, Jackson Immuno Research, USA) at a dilution of 1:5000 was added and then the mixture was incubated for 1 hour at 37°C. Later, the membrane was washed twice with wash solution and lastly washed with PBS. The blot was then developed using an ECL reagent (#786–003, G-Biosciences) and imaged using Bio rad Chemidoc.

For Native-PAGE About 60 µg of protein was diluted in commercial 4× Native dye (no heating) and loaded onto a precast native gel (4–15%) (Invitrogen™ and BioRad pre cast gel). 0.05% Coomassie G250 in Native PAGE running buffer was used as cathode buffer. The protein was resolved at 120 V for 3 hours. The gel was stained in Coomassie stain for 1 hour followed by de-staining.

Size exclusion chromatography

The purified M-CoV-S protein was run on the Superdex 200 Increase column for size exclusion chromatography. The column was equilibrated with 3 CV (column volume) of equilibration buffer (10 mM PBS). The M-CoV-S protein (3 mg/ml) was injected manually using a capillary loop. 1.5 CV of 10 mM PBS (flow rate of 0.750 ml/min) was used to elute the protein. UNICORN 7.1 software was used to analyze the chromatogram.

Differential scanning colorimetry

The melting temperature of M-CoV-S protein was estimated using Nano-DSC (TA instruments). M-CoV-S protein in 10 mM PBS at a concentration of 0.5 mg/ml was loaded onto the instrument against 10 mM PBS as a reference. A scanning rate of 1 °C/min under 3.0 atmospheres of pressure was used to perform thermal melting. Data collected was then analyzed using NanoAnalyze software, 3.6.0 (TA instruments), and the graph plotted using Graph Pad Prism.

Fluorescence spectroscopy

A Varian Fluorescence spectrophotometer was used to measure the tryptophan fluorescence of the M-CoV-S. M-CoV-S was excited at 280 nm wavelength and emission spectra were recorded from 290 nm to 500 nm with a slit width of 5 nm. Three independent spectra were recorded and temperature was maintained by the attached Peltier temperature controller. Further, for thermal analysis the temperature was increased,^20,21 and emission spectra were recorded for each temperature.

FAR UV circular dichroism

Circular dichroism (CD) was used to monitor the secondary structure. All the major secondary structures of protein have a characteristic CD spectrum. M-CoV-S prefusion spike at a concentration of 200 µg/ml was used for CD spectroscopy using JASCO 815 spectrometer, Japan. The instrument was flushed with nitrogen for 2 minutes and the lamp was allowed to warm for 10 minutes. The temperature controller was set to 4°C. Spectrum was recorded at a bandwidth of 1 nm with a step size of 1 nm. Three measurements were made and then Yang’s reference spectra were used to estimate the relative proportions of different secondary structures using Spectra Manager software from JASCO. Further, for thermal analysis, the temperature was increased to respective temperature,²⁰ ,²¹ and spectra were recorded.

Size and zeta potential

M-CoV-S spike was characterized for size, polydispersity index, and zeta potential using Zetasizer Nano ZS (Malvern Instruments Ltd., U.K.). 1 mg/ml of protein in disposable sizing plastic cuvette was used for measurement and disposable folded capillary cell for zeta potential. Reading was represented as the average of three independent measurements.

Ethics statement and mice immunization

All animal experimentation was conducted in compliance with the ethical considerations and guidelines issued by the Committee for Control and Supervision of Experiments on Animals (CPSEA) of the Government of India and with the approval of the Institutional Animal Ethics Committee (IAEC, THSTI) regarding laboratory animals (IAEC Approval number: IAEC/THSTI/256). For the immunization study, 6–8 weeks old female BALB/c mice weighing 18–25 g and inbred in the institute’s (THSTI) experimental animal facility (EAF) were used. 29 mice were randomly divided into five groups having six mice in each group. All the immunizations administered were investigator blindfolded. One group was immunized with M-CoV-S + AddaVax^TM (#vac-adx-10, Invivogen) as adjuvant mixed in a 1:1 ratio containing 25 μg of protein. The second group (Addavax control group) was treated with PBS + AddaVax^TM mixed in a 1:1 ratio. The third group was immunized with M-CoV-S + Imject adjuvant (#77161, Thermo Fisher Scientific) mixed in a 1:1 ratio containing 25 μg of protein. The fourth group (Imject control group) was treated with PBS + Imject mixed in a 1:1 ratio. Fifth group was treated with PBS only. Immunization was performed via the intramuscular route (cranial thigh muscles) twice at an interval of 4 weeks in a prime-boost regimen. The mice were observed daily and weighed twice a week over the course of the immunization period. Blood sample from each mouse was collected at day 0 (pre-immune sera), day 14 (sera after priming), day 14 and day 21 (sera after the boost). The serum was separated from the blood, heat inactivated at 56°C for 30 min and stored at − 20 ◦C for future use.

Enzyme Linked immunosorbent assay (ELISA)

The ELISA was performed to characterize the binding of the IgG and its subtypes from immunized sera collected from MERS-CoV prefusion spike protein groups. The M-CoV-S protein was coated on Maxisorp plates (#M9410, Nunc, Sigma Aldrich, India) with 100 μl of protein at 2 μg/ml concentration in 1×carbonate/bicarbonate buffer, pH 9.6 for overnight at 4°C. Next day, the plate was blocked by adding 200 μl/well of PBS containing 5% skimmed milk. Immunized mice sera collected from all six mice from each group were added separately in three-fold dilutions to the wells and incubated for 1 hr at RT. The wells were washed and HRP-conjugated anti-mouse secondary antibody (#115-035-003, Jackson Immuno Research, USA) in 1:2000 dilutions were used for developing ELISA. Similarly, for IgG subtyping, in other round of experiments secondary antibodies against IgG1(#115-035-205), IgG2a (#115-035-206), IgG2b (#115-035-207), and IgG3 (#115-035-209) were used as secondary antibodies, the plates were developed with tetramethylbenzidine (TMB) (#002023, Sigma-Aldrich, India) substrate, and the reaction was allowed to develop for 5-min, and then stopped with 1N sulfuric acid. The absorbance of plates was recorded at 450 nm on an ELISA reader (Bio-Rad). The endpoint titer was determined for IgG and each subtype which is the reciprocal of the highest test serum dilution giving a reading above the cutoff. In each assay, the sera from PBS treated mice was considered as negative control to calculate the cutoff value. The cutoff value was calculated by multiplying the average OD value of negative control samples + 3 times SD of OD value of negative control samples. The test result was considered positive when signal to cutoff ratio >/ = 1.

Immune profiling of the serum cytokines

Th1/Th2 cytokine levels were measured in immunized (M-CoV-S with AddaVax^TM or Imject) and non-immunized control mouse serum collected after the booster dose (day 42) using a Mouse Th1/Th2 CBA kit (#560485, BD Biosciences, USA). The standard dilutions of cytokines and test samples were incubated with cytokine capture beads and PE-detection reagent at room temperature for 3 hours in the dark conditions. After the incubation, beads were washed twice with the wash buffer and resuspended in the wash buffer for data acquisition using BD FACS Canto II Flow Cytometer (BD Biosciences, CA, USA). Instrument setting for sample acquisition was done prior to the experiment as per the manufacturer’s instructions. The data analysis was performed using the software FCAP Array and graphs were plotted on GraphPad Prism 9.0 software.

Protein’s de-glycosylation with PNGase F

De-glycosylation of M-CoV-S protein by PNGase F (#A39245, Gibco) (non-denaturing reaction conditions) was used following the manufacturer’s protocol. Briefly, Master Mix was prepared by combining 4 µL of PNGase F Buffer 10X and 0.5 µL of PNGase F enzyme. The Master Mix was added to 25 µg of purified dialyzed prefusion M-CoV-S protein and reaction mixture was incubated at 50°C for 1 h. Along with PNGase F treated protein untreated protein was also run as control to determine the extent of de-glycosylation. PNGase F treated and untreated samples were run on 12% SDS-PAGE and the extent of de-glycosylation was estimated by the shift in mobility of protein bands.

Immuno-fluorescence analysis

BHK21 cells (40,000/well) (ATCC) were seeded in Minimum Essential medium Eagle (MEM from Himedia) containing 10% FBS and 100 U/ml Penicillium-Streptomycin (P/S from Gibco). Next day, 0.5 μg each of CoV-1 spike full-length plasmid and ACE2 expressing plasmid were co-transfected using FuGENE HD transfecting reagent (#E2311, Promega). After 24 hr of transfection, cells were fixed in 4% paraformaldehyde (PFA) for 15 mins and permeabilized with 0.1% Triton-X100. Nonspecific binding was blocked using 4% Fetal Bovine Serum in PBS for 1 hr at RT. Cells were then incubated at RT for 1 h with heat inactivated pooled serum samples from a terminal bleed of M-CoV-S immunized mice (day 49 after prime immunization) or the control group in 1:200 dilutions. After incubation, cells were washed and incubated for 1 hr at RT with the Alexa Fluor 488 (#115-545-062, Jackson Immuno Research, USA) labeled rabbit anti-mouse IgG secondary antibody (1:1000 dilutions). Cells were washed three times with PBS and their nuclei were counterstained with Hoechst 33,342 (Thermo Scientific) and images were acquired on an Olympus IX-71 fluorescence microscope.

For the immunofluorescence assay with influenza virus, The MDCK-London cells were infected with 10 TCID₅₀ of the A/Guangdong-Maonan/SWL1536/2019 of influenza virus. After 36 hr of post-infection, the cells were fixed in 4% paraformaldehyde for 15 minutes and processed similarly as described above.

Structural modelling of the pre fusion M-CoV-S immunogen

The secondary structure of the pre-fusion M-CoV-S immunogen was determined using the PSIPRED server and SWISS-MODEL-Expasy.²² The PSIPRED server helps in predicting the basic building blocks of protein structures such as beta-strands, alpha helices, and coils by employing a network called feed-forward neural networks.²³ The SWISS-MODEL-Expasy web server aids in performing homology modeling of the query sequence to the target sequence. The same sequence is further analyzed using the MEMSTAT-SVM server to predict the various structural domains present in the sequence of the M-CoV-S protein.

Homology modelling

The homology modeling of the MERS spike protein was carried out using the Prime module within the Schrödinger Release 2023–3 software. The protein sequence of MERS was subjected to NCBI BLASTp against the PDB database, and PDB ID: 6NB3²⁴ was identified as the most suitable template due to its high sequence identity. The three-dimensional crystal structure was retrieved from the RCSB PDB database and utilized as the template in the Schrödinger Prime modeling interface. Subsequent refinement involved loop refinement and energy minimization using the OPLS4 force field in Schrödinger Prime. The resulting minimized model underwent validation through Ramachandran plot and by energy minimization analysis.²⁵

Protein preparation and docking

For protein preparation and docking, the structure with PDB ID 5GMQ representing MERS-CoV-RBD (resolution: 2.70 Å) was processed using the Protein Preparation Wizard of Schrödinger. The MERS-CoV-RBD region was extracted from the structure, and the MCA 1 antibody was pre-processed in the Protein Preparation Wizard. This involved the addition of hydrogens and bond orders, filling missing side-chains and loops using Prime, and capping at N- and C-terminals. Hydrogen bond optimization and restrained minimizations were performed for the systems using the OPLS4 force field,^25–28 The prepared antibody structure was then docked to the modeled spike protein using PIPER module of Schrodinger, with non-CDR regions of the antibody being masked during the process. The CDR region was defined according to the Chothia definition.

Statistical analysis

All immunization experiments were carried out with (n = 6) BALB/c mice per group. ELISA was used for determining the antibody generated and was expressed as endpoint titer measured at a wavelength of 450 nm. Each mouse was treated as an individual sample and the endpoint titer was expressed as group mean. For ELISA, the statistical significance of group means and standard deviation (SD) was assessed by an unpaired students t-test where M-CoV-S + Addavax was compared with Addavax control and M-CoV-S + Imject was compared with Imject control groups. For analyzing serum cytokine levels, one-way ANOVA analysis was performed using the Tukey-Kramer test for multiple comparisons, followed by post hoc tests inbuilt in GraphPad Prism 9 (GraphPad Software Inc., San Diego, CA, USA) to test for normal distribution of the samples by the Shapiro-Wilk normality test. Statistical analysis was then performed using multiple comparison tests where necessary. Where, *p < .05, **p < .01, ***p < .001, and ****p < .0001 were considered significant.

Comprehensive biophysical characterization of M-CoV-S

The properties of transiently expressed soluble M-CoV-S proteins were further assessed by size exclusion chromatography (SEC) using the Superdex 200 column. Size exclusion chromatogram of the recombinant protein showed a single elution peak at 9 ml which indicates the formation of higher order oligomeric structure (Figure 2a). The combined data from the blue native-PAGE and size exclusion chromatography suggests the formation of a homogeneous dimer of trimers. Next, the size distribution profiling of a M-CoV-S protein was determined using dynamic light scattering (DLS) at room temperature with light scattering at 90° for individual measurement and observed to be 33.96 ± 1.20 nm with a PDI value of 0.265 (Figure 2b). The hydrodynamic volume measured using DLS for SARS-CoV-2 trimeric spike protein from different strains varies from 13–18 nm. The 34 nm size seen in case of M-CoV-S further suggests formation of dimer of trimers.³¹ Also, a PDI value of 0.25 indicates a uniform distribution of particle sizes of the synthesized M-CoV-S proteins in suspension. Further, the apparent zeta potential of a M-CoV-S protein was determined in the 10 mM PBS (pH 7.5). The magnitude of the zeta potential is significantly influenced by the ionic strength and pH of the dispersant. The purified M-CoV-S proteins were negatively charged and the apparent zeta potential was depicted to be ‒7.3 ± 2.20 mV and remained constant for repeated batches suggesting moderate to good colloidal stability of the synthesized protein.

Next, the intrinsic fluorescence of M-CoV-S was measured using Varian Fluorescence spectrophotometer. The intrinsic fluorescence of a protein is a representative of the cumulative fluorescence due to all the fluorescence amino acids (mainly tryptophan and partially tyrosine residues). The Trp fluorescence maximum and intensity are highly influenced by the polarity of its micro-environment, hydrogen bonding and other non-covalent interactions that are present in a particular conformation. M-CoV-S pre-fusion spike contains 8 Trp and 75 Tyr residues. The emission maximum of fluorescence was recorded at a wavelength of 336 nm when taken at 4 degrees (Figure 2c). For the SARS-CoV-2 spike protein the reported emission maximum is at 325 nm. SARS-CoV-2 has 12 Trp and 55 Tyr residues while M-CoV-S has 8 Trp and 75 Tyr residues. The red shift in case of M-CoV-S may be accounted for the lower number of Trp and higher number of Tyr residues.³¹ Nevertheless, a good intensity of emission suggests that the overall M-CoV-S spikes’ structure is folded.³²

M-CoV-S was further characterized for secondary structure by far UV circular dichroism (CD) spectroscopy. CD exploits the ability of optically active moieties present in proteins to absorb circularly polarized light differentially. The CD signature of M-CoV-S, which is very similar to that of SARS-CoV-2 spike protein,³¹ suggests equal percentage of both alpha-helix and beta-sheet structure, with slightly higher content of alpha-helix as modest negative peak can be seen at 222 nm and 208 nm. The similarity in CD signature of SARS-CoV-2 and MERS-spike protein may be attributed to similar geometry and content of the secondary structures in both the spike proteins. We used Spectra Manager Suite (Jasco) to deduce the secondary structure. It was observed that M-CoV-S showed a secondary structure consisting of 24% alpha-helical, 12% beta-sheet, and 35% random coil (Figure 2d).

M-CoV-S immunogen shows promising structural stability

The in-silico prediction of secondary structure derived from the M-CoV-S protein sequence using the PSIPRED server showed the presence of three building blocks of a protein structure such as helices, coils, and strands. In the M-CoV-S sequence of 1318 amino acids, in-silico prediction depicted a percentage of 14.32, 20.62, and 39.97 for alpha-helices, beta-strands, and coils, respectively (Supplementary Figure S2a). The sequence plot generated via MEMSTAT-SVM evaluated three different domains naming two extracellular domains (one from 1–260 amino acids and another from 950–1318 amino acids), one cytoplasmic domain (278–931 amino acids), and two short membrane-interacting domains (one from 261–277 amino acids and another from 932–949 amino acids) in the sequence of 1318 amino acids for the purified M-CoV-S proteins (Supplementary Figure S2b).

To experimentally assess the thermal stability and thermal unfolding of the spike protein, a differential scanning colorimetry (DSC) experiment was performed. DSC is a technique that measures the change in heat required to unfold a protein as a function of temperature. Thus, it is used to estimate the enthalpy for the process and thereby, the melting temperature (Tm) estimation of the protein. The M-CoV-S showed a single transition peak at 58°C (Figure 3a). Reports for stable forms of the SARS-CoV-2 spike proteins also recorded a single transition.^21,33 We hypothesize that formation of dimer of trimers of M-CoV-S might have stabilize the structure and noncooperative unfolding among the different species (monomer, trimer, dimer-of-trimer) has obscured the transitional T_m of each subunit as a single transitional peak with T_m at 58°C. Additionally, the thermal stability of tertiary conformation was evaluated with SDS_PAGE and intrinsic tryptophan fluorescence as a function of temperature. The protein was stored at different temperatures (4°C, 25°C and 37°C) and assessed by SDS-PAGE. A single band was observed indicating that protein did not undergo any significant conformational change till one week at a temperature of 37°C (Figure 3b). Next, we observed, with increasing temperature, the fluorescence gradually decreased, however, the maxima of 336 nm were retained till a temperature of 37°C. At 55°C occurrence of a second maxima along with significant decrease in intensity was observed (Figure 3c). This indicates destabilization of protein structure was accompanied by tryptophan being exposed to a more hydrophilic environment, however, as prefusion spike contains 75 tyrosine along with 8 tryptophan, it may be due to self-quenching or tendency to retain the native structure. A gradual decrease in intensity but retention of maxima suggests that though temperature caused destabilization but protein had a tendency to retain its folded structure. We next used the far UV-CD spectra to study the temperature-dependent changes in the secondary structure of M-CoV-S protein. The CD spectrum remained intact till a temperature of 55°C. Analysis showed that M-CoV-S possessed 20–24% helix, 10–14% beta, and 30–35% random coil conformation up till a temperature of 37°C. Beyond 55°C there was a gain of 200 and 219 peaks indicating beta structure occurrence (Figure 3d).

M-CoV-S in combination with either AddaVax^TM or imject adjuvants induces high titer humoral immune response

Adjuvants play an important role in stimulating and enhancing the immune responses mainly in protein-based vaccine formulation.^34,35 Hence, to evaluate the effect of M-CoV-S antigen to elicit immune responses in the presence of adjuvants, we included two key adjuvants that are licensed for human use; a) AddaVax^TM, which is MF59 squalene-like adjuvant, known to induce a balanced Th1/Th2 responses, and b) Imject an Alum adjuvant showed to have enhanced immune responses and induces Th2 type responses.³⁶ As described in materials and methods BALB/c mice were immunized intramuscularly in prime-boost regimen with 25 µg of M-CoV-S antigen either with AddaVax^TM or Imject adjuvant (Figure 4a). It was observed that after one immunization of prime dose, the immunized mice with M-CoV-S antigen + AddaVaxTM elicited moderately higher (206550) whole IgG titer than the M-CoV-S antigen + Imject immunized group (6750) (Figure 4c) p < .0048. However, after boosting, both the immunized groups elicited high antibody responses as measured at 14 days (mean of 984,150 for both the M-CoV-S adjuvanted groups) and 21 days post-boosting immunized sera (mean of 984,150 for M-CoV-S addavax and 874,800 for M-CoV-S Imject group, p < .3409) (Figure 4c-e) as compared to the control mice immunized with only adjuvants. The assessment of subtypes of IgG was estimated for the day 14 boost serum. It was observed that both the adjuvants generated similar IgG1 and IgG2b titers (Figure 4f). Although, the immunized AddaVax^TM group showed higher IgG2a responses whereas Imject immunized group showed higher IgG3 titer, suggesting that Addavax^TM adjuvant induced a more balanced Th1/Th2 response compared to the Imject adjuvant in combination with MERS pre-fusion spike protein.

Next, we assessed the immune profiling of the serum collected from the MERS-CoV spike immunized mice after 14 days of booster dose using the Th1/Th2-Mouse CBA kit. We determined the serum concentration of seven different cytokines i.e., IFN-γ, TNF-α, IL-5, IL-4, IL-17a, IL-10, and IL-2 (Figure 4g). The serum concentration of IFN-γ, TNF-α, and IL-2 cytokines in M-CoV-S immunized with AddaVax^TM estimated to be 240.8 pg/ml, 160.4 pg/ml, and 122.2 pg/ml, respectively, whereas the serum concentration of IL-5, IL-4, IL-17a, and IL-10 cytokines appeared to be 68.4 pg/ml, 48.5 pg/ml, 55.4 pg/ml, and 61.8 pg/ml, respectively. In the M-CoV-S and Imject immunized group the peak serum concentration was shown only for IFN-γ, IL-2, and IL-10 cytokines and the titers were 151.4 pg/ml, 92.7 pg/ml, and 46.1 pg/ml, respectively. The serum cytokine profiling of the immunized mice suggests induction of strong Th1 polarized immune responses in AddaVax^TM adjuvanted animals, and the elicitation of IFN-γ and TNF-α also indicates the significant contribution in induction of humoral and cellular responses, however a balanced Th1/Th2 responses as compared to Imject adjuvanted immunized sera.

Molecular modelling and docking of M-CoV-S protein

Homology modeling was used to generate model of M-CoV-S protein of the given sequence using PDB-ID: 5X5C prefusion MERS spike conformation. The Ramachandran plot (Figure S3b) indicates the quality of the modeled protein (Figure 5). The superimposition of the modeled protein with respect to template was 0.4 Å, further indicating that other than loops the major structural portion of the model is overlaid nicely, therefore, we proceed for molecular docking study. From the assessed docking results, the best poses were selected based on their cluster size, with a total of nine clusters considered. Interacting residues were analyzed and depicted in the corresponding figure (Figure 5a). In molecular docking the energetically favorable pose was chosen to compare with co-crystal and for qualitative analysis. One of the limitations in this study is unavailability of the MERS live or pseudovirus to evaluate the neutralizing potential of the immunized sera against the M-CoV-S protein. Hence, we utilized the energy minimization using molecular dynamics simulation study to assessed the binding of MCA1 neutralizing antibody with M-CoV-S protein.¹⁸ The binding of MCA1 mAb found to be in the similar binding area as it has been reported in PDB-ID: 6GMQ (interacting with RBD), the paratope residues are common such as Tyr31, Ile49, Ile51, Leu52, Gly98, Asp99, Tyr100, Tyr101, Asp102 and Arg103. Furthermore, the type of interactions was mapped in which several common paratope residues are involved in establishing the similar type of interactions such as hydrogen bonds, salt-bridges and non-bonded interactions (Figure S3c and S3d). The hydrogen bonds were formed between Asp99 (paratop)@Arg525 (epitope), Tyr101@Arg525, Tyr100@Arg525, Asn56@Glu519, Arg103@Glu519 and Asp102@Trp518. The salt-bridges were formed between Asp99@Arg525, Arg47@Asp522, Arg103@Asp522, and Arg103@Glu519. Furthermore, the non-bonded contacts were noticed between Asp99@Trp536, Asp99@Arg525, Tyr100@Tyr523, Tyr100@Tyr524, Arg103@Trp518 and Asp99@Trp536. From interaction fingerprinting it was clearly visible that not only common paratope residues but also the similar property holding residues of epitopes are similar as well, confirming that the antibody binds tightly with modeled spike protein. Moreover, the thermodynamic quantification was also carried out using MM-GBSA module. The docked structure exhibits a binding affinity of −60.03 kcal/mol, while the co-crystal structure demonstrates a binding affinity of −66.17 kal/mol. The comparable binding free energy corroborate well with quantitative analysis as it matched nicely with reported co-crystal structure (PDB-ID: 6GMQ). Overall, the modeling and docking data indicate that antibody MCA1 binds with suitable pose similar to reported co-crystal structure.

The MERS and other Corona viruses or enveloped viruses like Influenza and HIV-1 contain the N glycosylation sites and are highly glycosylated. Previous studies were shown the sialic acid present in the virus envelope protein N glycans are cross-reactive.^37–39 In order to confirm the cross binding due to N-glycans, N-Glycans present in M-CoV-S protein was removed after de-glycosylation with PNGase-F enzyme as described in Materials and Methods section. We found shifting in the molecular weight upon de-glycosylation but the serum specific to SARS-CoV-1-RBD (Supplementary Fig. S4a, Lane 6), SARS-CoV-2-Spike (Supplementary Fig. S4a, Lane 8) and Inf A/Guangdong-Maonan/SWL 1536/2019 (Supplementary Fig. S4a, Lane 11) could still recognized the PNGase-F treated M-CoV-S protein. In order to rule out the nonspecific binding toward mouse polyclonal serum, the M-CoV-S protein was probed with anti-mouse serum collected from normal mice. Protein showed no affinity toward normal mouse serum (Supplementary Fig. S4b, Lane 10). Suggesting the cross binding is not specific only to N-Glycans, but to the existence of some similarity of overlapping sequences of the three viruses’ glycoproteins. However, further studies are needed to confirm the cross reactiveness. In another experiment, BHK21 cells were co-transfected with a plasmid containing a CoV-1 full-length spike and ACE2. Transfection resulted in BHK-21 cells expressing CoV-1 spike over the surface which were incubated with M-CoV-S sera and stained with Alexa 488 dye. The sera showed binding to the CoV-1 spike in the immunofluorescence assay (Supplementary Fig. S4c, e). Similar, cross reactivity also seen in binding of M-CoV-S polyclonal serum with SARS-CoV-2 ancestral Wuhan-Hu-1 pseudo virus infected Vero E6 cells (Supplementary Fig. S4c, e). Further, MDCK-London cells infected with A/Guangdong-Maonan/SWL1536/2019 (H1N1) virus also showed the cross reactivity by binding with M-CoV-S polyclonal sera. Binding was as similar to the polyclonal serum of A/Guangdong-Maonan/SWL 1536/2019 taken as a positive control (Supplementary Fig S4f).