INTRODUCTION
Three highly pathogenic beta-coronaviruses (CoVs), including severe acute respiratory syndrome (SARS)-CoV, Middle East respiratory syndrome (MERS)-CoV, and SARS-CoV-2, which were first identified in 2002, 2012, and 2019, respectively, caused global outbreaks or pandemics (
1–3). Infections with SARS-CoV and MERS-CoV resulted in mortality rates of 10% and ~35%, respectively (
2,
4). According to the World Health Organization, as of 28 April 2024, SARS-CoV-2 has infected more than 775.3 million persons worldwide, with at least 7.04 million persons dying of coronavirus disease 2019 (COVID-19), the disease caused by SARS-CoV-2 (
5). SARS-CoV-2 has been found to undergo constant mutations, with these mutations resulting in significant resistance to vaccines and therapeutic antibodies against the original or earlier variants (
6–8). In addition, MERS-CoV continues to infect humans, and some SARS-like or MERS-like CoVs from bats use the same receptors as SARS-CoV-2 or MERS-CoV for viral entry, having pandemic potential (
9–11). Effective vaccines with broad-spectrum protective efficacy against these CoVs are therefore urgently needed to prevent infection with MERS-CoV, SARS-CoV-2 variants, or other CoVs with pandemic potential.
The surface spike (S) protein of CoVs binds to its respective receptor on host cells, initiating viral entry and the process of infection. The S proteins of CoVs consist of S1 and S2 subunits. The receptor-binding domains (RBDs) of SARS-CoV-2 and SARS-CoV bind to their cellular receptor, angiotensin-converting enzyme 2 (ACE2), whereas the RBD of MERS-CoV binds dipeptidyl peptidase 4 (DPP4) for efficient entry into host cells (
9,
12–15). Thus, the CoV S proteins and the RBDs are important vaccine targets. The RBDs in the S1 subunit of SARS-CoV and MERS-CoV contain critical neutralizing epitopes and induce potent neutralizing antibodies against infection by different strains of these viruses (
2,
16,
17). The RBD of SARS-CoV-2, however, has been found to mutate rapidly, resulting in at least five variants of concern to date (Alpha, Beta, Gamma, Delta, and Omicron) (
18,
19). Currently available SARS-CoV-2 vaccines are relatively ineffective against emerging Omicron subvariants, with the latter found to significantly escape neutralizing immunity or protective efficacy of vaccines targeting the wild-type (WT) strain or earlier variants (
20,
21).
The present study describes the development of a pan-beta-CoV subunit vaccine, called Om-S-MERS-RBD, by fusing the highly neutralizing RBD region of MERS-CoV into an RBD-truncated S protein of SARS-CoV-2 Omicron variant. Its conformational structure, antigenicity, as shown by its ability to bind antibodies against MERS-CoV RBD and SARS-CoV-2 S, and functionality, as shown by its ability to bind the MERS-CoV receptor, human DPP4, were investigated. This protein was tested for its ability to neutralize multiple SARS-CoV-2 strains, SARS-CoV, and MERS-CoV. Moreover, its protective efficacy against the three CoVs (i.e., SARS-CoV-2 Omicron, SARS-CoV, and MERS-CoV) was evaluated in mouse models.
DISCUSSION
A variety of COVID-19 vaccines have been developed, but many of those targeting the S protein of the original strain of SARS-CoV-2 and subsequent variants have lower or no neutralizing activity or protection against the newly emerging SARS-CoV-2 Omicron subvariants (
20,
25). Thus, variant-specific COVID-19 vaccines have to be constantly designed to prevent individuals from infection with these new Omicron subvariants or other future variants with pandemic potential. Developing new vaccines, however, normally requires additional time and costs. Moreover, COVID-19-specific vaccines are not expected to show neutralizing activity or significant protection against pathogenic Merbecovirus, such as MERS-CoV, due to lower homology among their sequences of S and other proteins. These findings indicate a need to develop pan-beta-CoV vaccines, without the need for constant change of vaccine antigens based on newly emerging viral strains. These pan-beta-CoV vaccines should not only protect against infection with newly emerging SARS-CoV-2 variants but should also protect against MERS-CoV and other beta-CoVs, which may infect human hosts and have future pandemic potential.
Several pan-CoV vaccines have been designed based on the nanoparticles or mRNAs, displaying a mosaic RBD or S protein of SARS-CoV-2 or other beta-CoVs on their surfaces (
26–29). For example, a chimeric mRNA vaccine expressing S protein (with different NTD, RBD, and S2 regions) of SARS-CoV-2 or SARS-related CoVs was found to induce neutralizing antibodies against several SARS-CoV-2 variants and to protect mice from infection with SARS-CoV-2 (WT strain and B.1.351 variant), SARS-CoV, Bt-CoV RsSHC014, and Bt-CoV WIV-1 (
26). In addition, a mosaic-8 nanoparticle vaccine, which presents the RBD of SARS-CoV-2 Beta variant and seven animal Sarbecoviruses, was found to elicit neutralizing antibodies against SARS-CoV-2 Omicron subvariants (BA.1, BA.2, BA.2.12.1, or BA.4/5), protecting animals from challenge with SARS-CoV and the Beta and Delta variants of SARS-CoV-2 (
27).
In this study, a different approach was applied to design the pan-beta-CoV subunit vaccines. Since the RBD of SARS-CoV-2 Omicron is highly variable, it was removed during the vaccine design, and replaced by the RBD of MERS-CoV or SARS-CoV, which shows low variations among different strains but potent neutralizing activity against multiple viral infections (
2,
16,
30). The RBD of SARS-CoV-2 is divergently mutated, but the other region of the S protein has high homology among all SARS-CoV-2 strains, including the Omicron variant (
31–33). Therefore, the RBD-truncated region was used to fuse with the RBD of MERS-CoV or SARS-CoV during the vaccine design.
Om-S-MERS-RBD protein (which expresses the MERS-CoV RBD and lacks the SARS-CoV-2 RBD) formed a conformational structure, allowing it to strongly bind MERS-CoV receptor DPP4, and antibodies specific to SARS-CoV-2 S protein and MERS-CoV RBD, respectively. This subunit vaccine elicited potent or effective neutralizing antibodies against MERS-CoV, SARS-CoV, and SARS-CoV-2 WT strain. Its ability to neutralize SARS-CoV-2 Omicron subvariants was significantly improved by priming with Om-S-BA5-RBD protein. Further studies will be conducted to evaluate the neutralizing activities and protective efficacy of this vaccine after priming with a SARS-CoV-2 S protein from the ancestral strain, or a variant of current or future emerging strains, and compared for the improved neutralizing antibodies and protection against infection of different SARS-CoV-2 strains and other CoVs. Of note, vaccination with Om-S-MERS-RBD simultaneously protected animals from infection with SARS-CoV-2 (Omicron-BA.5), SARS-CoV, and MERS-CoV. In addition, Om-S-SARS-RBD (which expresses the SARS-CoV RBD and lacks the SARS-CoV-2 RBD) induced favorable neutralizing antibodies and protection against SARS-CoV-2 and SARS-CoV, rather than MERS-CoV, potentially due to the lack of homologous sequences between the MERS-CoV and SARS-CoV RBDs.
Om-S-MERS-RBD protein elicited low-titer neutralizing antibodies against the pseudotyped Omicron-BA.5 variant, but it induced high-level RBD-truncated S-specific IL-4 cytokine-associated T-cell responses in the immunized mice. Therefore, although vaccine-induced neutralizing antibodies may play a critical role in the ability of Om-S-MERS-RBD to protect against SARS-CoV and MERS-CoV infections, its ability to protect against infection with the SARS-CoV-2 Omicron variant would likely be due to the vaccine-induced T-cell responses. Notably, vaccination with Om-S-MERS-RBD or other proteins did not elicit a significant level of S-specific Th1 cytokines, such as TNF-α, and IFN-γ, partially due to the use of Alum adjuvant, which is a strong Th2 inducer (
34,
35). Future studies will be performed to optimize the ratio of Alum and MPL (Th1 inducer) adjuvant combination or use other adjuvants, such as SMQ or a combination of CpG and MF59, which tend to induce Th1 cytokines or a balanced T-cell response (
34–36).
Overall, the present study demonstrated that a pan-beta-CoV subunit vaccine with protective efficacy against three highly pathogenic CoVs (i.e., SARS-CoV-2, SARS-CoV, and MERS-CoV), and a pan-sarbecovirus subunit vaccine with protective efficacy against SARS-CoV-2 and SARS-CoV could be designed. These vaccines were characterized, and their ability to protect against viral infections was confirmed. Further development of these vaccines may protect against highly pathogenic CoVs and other beta-CoVs with pandemic potential.
MATERIALS AND METHODS
Cell lines
HEK293F, 293T, 293T expressing human ACE2 receptor (hACE2/293T), Huh-7, Vero, Vero E6, and Vero E81 cells were cultured at 37°C in the presence of 5% CO2. The 293F cells were cultured in an ESF serum-free medium (Expression Systems). The other cells were cultured in Dulbecco’s Modified Eagle Medium (DMEM) containing 10% fetal bovine serum (FBS) (R&D Systems) and 1% Gibco Penicillin-Streptomycin (Thermo Fisher Scientific).
Plasmids and viruses
DNA sequences encoding S protein of SARS-CoV-2 WT strain (GenBank accession number
QHR63250.2), Omicron BA.1 variant (GISAID accession number EPI_ISL_6795835), SARS-CoV Tor2 strain (GenBank accession number
AY274119), or MERS-CoV EMC2012 strain (GenBank accession number
JX869059.2) were inserted into pcDNA3.1/V5-His-TOPO vector to construct the respective recombinant plasmids. SARS-CoV-2 BA.2 (GISAID accession number EPI_ISL_12030355), BA.5 (GISAID accession number EPI_ISL_12043290), BQ.1.1 (GISAID accession number EPI_ISL_15370776), and XBB (GISAID accession number EPI_ISL_15341139) recombinant plasmids containing the respective RBD mutant residues were constructed based on the above BA.1-S recombinant plasmid using multi-site-directed mutagenesis kit. Live SARS-CoV-2 Omicron-BA.5 variant, MERS-CoV (EMC2012 strain, GenBank accession number
JX869059.2), and SARS-CoV (MA15 strain) were used in this study.
Vaccine preparation
This was prepared as described below (
37). Briefly, the codon-optimized RBD of MERS-CoV (GenBank accession number
JX869059.2) or SARS-CoV (GenBank accession number
AY274119) was fused with the RBD-truncated, codon-optimized S extracellular domain sequence of SARS-CoV-2 Omicron variant (GISAID accession number EPI_ISL_6795835), and inserted into a pLenti expression vector using ClonExpress MultiS One Step Cloning kit (Cellagen Technology). Each recombinant plasmid contains an N-terminal tPA signal sequence, HexaPro sequence, a C-terminal foldon trimerization motif, and a His
6 tag. The sequence-confirmed recombinant plasmids were respectively transfected into HEK293F cells, and the culture supernatant was collected for purification of the respective proteins using Ni-NTA Superflow (Qiagen).
Cryo-EM grid preparation and data acquisition
The isolated spike (4 µL at 0.96 µM) was applied to freshly glow-discharged Quantifoil R1.2/1.3 300-mesh copper grids (EM Sciences), and then blotted for 4 s at 22°C under 100% chamber humidity and plunge-frozen in liquid ethane using a Vitrobot Mark IV (FEI). Cryo-EM data were collected using EPU (Thermo Fisher Scientific) on a Titan Krios electron microscope (Thermo Fisher Scientific) equipped with a K3 direct electron detector with a Biocontinuum energy filter (Gatan) in CDS mode at the Hormel Institute, University of Minnesota. The movies were collected at a nominal magnification of 81,000× (corresponding to 1.1 Å per pixel), a 20 eV slit width, a dose rate of 25 e– per Å2 per second, and a total dose of 50 e−/Å2. The statistics of cryo-EM data collection are summarized in Table S1.
Image processing
Cryo-EM data were processed using cryoSPARC v4.0.3 (
38), and the data processing procedures are outlined in Fig. S1. Dose-fractionated movies were first subjected to Patch motion correction and Patch CTF estimation with MotionCor2 (
39) and CTFFIND-4.1.13 (
40), respectively. Images with defocus values outside of −1.0 to −2.0 µm or the CTF fit resolutions worse than 6 Å were excluded from the further steps. Particles were picked using both Blob picker and Template picker accompanied by removing duplicate particles. Three rounds of 2D classifications were applied to remove junk particles and particles (213,507) extracted from the good 2D classes were used for Ab-initio Reconstruction of four maps and then for the heterogeneous refinements. The good 3D class (192,374 particles) was finally subjected to further homogeneous, non-uniform, and CTF refinements to generate a 3.45 Å resolution final map. Map resolution was determined by gold-standard Fourier shell correlation (FSC) at 0.143 between the two half-maps. Local resolution variation was estimated from the two half-maps in cryoSPARC v4.0.3.
Model building and refinement
Initial model building of the Om-S-MERS-RBD was performed in Coot-0.8.9 using PDB 7TGW as the starting model (
41). Several rounds of refinement in Phenix-1.16 (
42) and manual building in Coot-0.8.9 were performed until the final reliable models were obtained. The final model has good stereochemistry by evaluation in MolProbity (
43). The statistics of 3D reconstruction and model refinement are shown in Table S1. Figures were generated using UCSF Chimera X v0.93 (
44).
Enzyme-linked immunoassay
Enzyme-linked immunoassay (ELISA) was performed to test the binding of proteins to CoV S/RBD-specific antibodies (
45,
46). Briefly, ELISA plates were coated with each protein (1 µg/mL) at 4°C overnight and blocked with a blocking buffer [2% non-fat milk in PBS containing 0.05% Tween-20 (PBST)] at 37°C for 2 h. After washing five times with PBST, the plates were incubated with serially diluted SARS-CoV-2 S or MERS-CoV RBD protein-immunized mouse sera (pAb) (
17,
37,
47), SARS-CoV-2 RBD-targeting nanobody (Nb) (
46), or SARS-CoV RBD-targeting mouse monoclonal antibody (mAb, 33G4) (
45), which was followed by incubation with horseradish peroxidase (HRP)-conjugated anti-mouse IgG (Fab-specific 1:10,000, Sigma; for mouse sera or mAb) or anti-Camelid VHH antibody (1:10,000, GenScript; for Nb) antibody at 37°C for 1 h. The plates were washed and then incubated sequentially with substrate TMB (3,3′,5,5′-Tetramethylbenzidine) (Sigma), and H
2SO
4 (1 N). Absorbance at 450 nm (A450 value) was measured using Cytation 7 Microplate Multi-Mode Reader (BioTek Instruments).
ELISA was also performed to test the binding of proteins to CoV receptors (
17,
37,
47). Briefly, ELISA plates were coated with each protein, and blocked as described above, followed by the addition of diluted human DPP4 (MERS-CoV receptor) or human ACE2 (SARS-CoV-2 or SARS-CoV receptor) protein (Laboratory stock) for incubation at 37°C for 1 h. After washes, the plates were sequentially incubated with goat anti-hDPP4 or anti-hACE2 antibody (1:1,000, R&D Systems) and HRP-conjugated rabbit-anti-goat IgG antibody (1:5,000, Abcam). Other steps were the same as described above.
CoV pseudovirus generation and neutralization assay
The pseudoviruses were prepared as described below (
48–50). Briefly, each recombinant plasmid DNA encoding the respective S protein of SARS-CoV-2, SARS-CoV, or MERS-CoV was transfected, in the presence of PS-PAX2 and pLenti-CMV-luciferase plasmids, into 293T cells using the PEI transfection reagent. The culture supernatant containing each pseudovirus was collected 72 h post-transfection, and incubated with immunized mouse sera at 37°C for 1 h. The mixture of virus and sera was incubated in 293T cells expressing SARS-CoV-2 or SARS-CoV receptor, human ACE2 (hACE2/293T) (for SARS-CoV-2 or SARS-CoV pseudovirus), or Huh-7 cells expressing MERS-CoV receptor human DPP4 (for MERS-CoV pseudovirus), for 24 h. The cells were further cultured for 48 h after the addition of fresh medium. The lysed cell supernatant was incubated with luciferase substrate (Promega), and relative luciferase activity was measured using Cytation 7 Microplate Multi-Mode Reader and Gen5 software. Neutralizing antibody activity against pseudovirus infection was reported as 50% neutralizing antibody titer (NT
50).
Multiplex immunoassay
Immunized mice were evaluated for specific T-cell responses by Multiplex immunoassay. Specifically, PBMCs were isolated by gradient centrifugation using Histopaque-1083 solution (Sigma-Aldrich). Residual red blood cells were removed using Red Blood Cell Lysis Buffer (Sigma-Aldrich). The isolated PBMCs were resuspended in RPMI 1640 cell culture medium containing 10% FBS, penicillin (100 U/mL), streptomycin (100 µg/mL), β-mercaptoethanol (55 µM), non-essential amino acids, sodium pyruvate (1 mM), and mouse IL-2 (1 ng/mL), and stimulated with RBD-truncated SARS-CoV-2 Omicron-S protein (5 µg/mL). 48 hours later, the cells were re-stimulated with the same protein (5 µg/mL) for an additional 24 hours. The supernatant was collected by centrifugation, and the cytokines in the supernatant were measured using a Bio-Plex Pro Mouse Cytokine Th1/Th2 Assay kit (Bio-Rad). The results were analyzed using the Bio-Plex 200 System (Bio-Rad).
Mouse immunization and sample collection
Female BALB/c mice (6- to 8-week-old) were used in the study based on the preliminary studies, and they were randomly assigned to the indicated vaccination groups. Three separate immunizations were performed as described below (
37,
51). The mice were intramuscularly (i.m.) immunized with the following proteins (10 µg/mouse) for three doses at a 3-week interval: (i) Om-S-BA5-RBD, for three doses; (ii) Om-S-SARS-RBD, for three doses; (iii) Om-S-MERS-RBD, for three doses; (iv) Om-S-BA5-RBD for the 1st dose and Om-S-SARS-RBD for the 2nd and 3rd doses; and (v) Om-S-BA5-RBD for the 1st dose and Om-S-MERS-RBD for the 2nd and 3rd doses. An optimal combination of Alum (500 µg/mouse) and MPL (10 µg/mouse) (InvivoGen) adjuvants was thoroughly mixed with the respective proteins before use. PBS plus the above adjuvants were included as control. Sera were collected 10 days after the last dose to test neutralizing antibodies against pseudotyped CoV infection, and PBMCs were isolated 5 months post-last immunization to detect T-cell responses, as described above. The immunized mice were proceeded to the subsequent challenge studies.
Challenge of mice with SARS-CoV-2 Omicron variant
This was performed as described below (
37,
51). The immunized mice described above were intranasally (i.n.) challenged with an Omicron-BA.5 variant of SARS-CoV-2 at an optimal dose of 10
5 plaque-forming unit (PFU)/mouse (50 µL/mouse) 3 weeks after the last vaccination. The lungs of challenged mice were collected 2 days after the challenge, and tested for viral titers and viral replication by plaque assay and qPCR method, respectively, as described below.
Challenge of mice with MERS-CoV
This was performed as described below (
23,
49). Three weeks after the last immunization, the vaccinated mice described above were i.n. transduced with an optimal dose (2.5 × 10
8 focus-forming unit: FFU) of Ad5-human DPP4 (Ad5-hDPP4: Ad5CMV/hDPP4-myc-flag; UI Viral Vector Core Web) suspended in 75 µL/DMEM culture medium, and then i.n. challenged with an optimal dose of MERS-CoV (EMC2012 strain; 10
5 PFU/mouse, 50 µL/mouse) 5 days after transduction. Lungs were collected 3 days after challenge, and measured for viral titers using plaque assay as described below.
Challenge of mice with SARS-CoV
This was performed as described below (
52). Briefly, 5 months after the last immunization, the vaccinated mice described above were i.n. challenged with an optimal dose of SARS-CoV (MA15 strain; 200 PFU/mouse, 50 µL/mouse), and lungs were collected 2 days after the challenge to measure viral titers using plaque assay as described below.
Plaque assay
This was performed as described below (
16,
53). Lungs from mice challenged with SARS-CoV-2 Omicron-BA.5, SARS-CoV, or MERS-CoV were homogenized in PBS. The tissue homogenate supernatant was serially diluted in a DMEM cell culture medium. Vero E6 cells (for SARS-CoV and SARS-CoV-2 wild-type), Vero in the presence of ACE2 and TMPRSS2 (for SARS-CoV-2 Omicron subvariants), or Vero E81 (for MERS-CoV) cells were plated in 12-well plates and cultured at 37°C for 1 h in 5% CO
2 with gentle rocking every 15 min. The medium was removed 1 h later, and the plates were overlaid with 0.6% agarose, which were removed after 3 days. The plaques were visualized by staining with 0.1% crystal violet. Viral titers were quantified as PFU/mL of lung tissues.
qRT-PCR
This was performed as described below (
54). Lungs from mice challenged with the SARS-CoV-2 Omicron-BA.5 variant were homogenized in Trizol buffer by Invitrogen, and RNA was extracted according to the manufacturer’s protocol (Thermo Fisher Scientific). Total RNA (1 µg) was used as a template for the first strand of cDNA, which was subjected to amplification of selected genes by real-time quantitative PCR (qRT-PCR) using Power SYBR Green PCR Master Mix (Applied Biosystems). The nucleocapsid (N) gene of Omicron-BA.5 was detected using nCOV_N1 primer (IDT, Cat# 10007031). The expression levels were normalized to glyceraldehyde-3-phosphate dehydrogenase (GAPDH) by the following threshold cycle (CT) equation: ΔCT = CT of the gene of interest − CT of GAPDH. All results are shown as a ratio to GAPDH calculated as 2
−ΔCT.
Statistical analysis
Statistical software (GraphPad Prism 9) was used to determine statistical significance among different groups. Ordinary one-way ANOVA Tukey’s multiple comparison test was applied for statistical analysis. *, **, ***, and **** indicate P < 0.05, P < 0.01, P < 0.001, and P < 0.0001, respectively.
ACKNOWLEDGMENTS
This work was supported by National Institutes of Health grants (R01AI157975, R01AI139092, R01AI137472, and R01AI110700).
L.D. conceived the study. G.W. constructed the vaccines and prepared pseudoviruses. G.W. and X.G. characterized the vaccines and tested immune responses. G.W., A.K.V., X.G., and A.E.O immunized animals and performed challenge studies. F.B., F.L., and B.L. analyzed and supervised the cryo-EM structures. S.P. and L.D. supervised the study, wrote, and revised the paper with input from all authors.