Universal subunit vaccine protects against multiple SARS-CoV-2 variants and SARS-CoV

Characterization of the recombinant protein subunit vaccines

Four recombinant S plasmids containing the conserved backbone of S protein of SARS-CoV-2 Omicron BA.1 (a then dominant subvariant) were constructed. These include the backbone only by truncating its RBD (S-6P-no-RBD), or respectively fusing the RBD of 1) a Delta variant (S-6P-Delta-RBD), 2) an ancestral SARS-CoV-2 wild-type (S-6P-WT-RBD), or 3) an Omicron BA.5 variant (S-6P-BA5-RBD) into the above S backbone construct (Fig. 1a). The ancestral S-6P protein (without RBD truncation) was included as a control (Fig. 1a). The recombinant protein vaccines purified from transfected HEK293F cell culture supernatants had high purity and maintained strong thermal and pH stability (Supplementary Figs. 1 and 2), and they were used for analysis or subsequent immunization of mice. Cryo-EM analysis of the S-6P-no-RBD protein indicated that the S particles (with truncated RBD) formed an open conformation, and the side and top views showed that they were missing the densities corresponding to the RBD regions (Fig. 2a, b and Supplementary Fig. 3). These data demonstrate that S protein without the RBD was capable of forming conformational structures for insertion of a heterologous RBD.

Vaccine-induced neutralizing antibodies and T-cell responses

To evaluate the ability of subunit vaccines to induce humoral immune responses, K18-human ACE2-transgenic (K18-hACE2, C57BL/6 background) mice were immunized with the respective proteins, a cocktail (combined S-6P-Delta-RBD and S-6P-BA5-RBD proteins), or PBS control, plus Aluminum Hydroxide Gel (Alum for short) and Monophosphoryl Lipid A (MPL) adjuvants, and boosted twice at 3-week intervals. Sera collected 10 days after the 3^rd immunization were tested for SARS-CoV-2 S/RBD-specific IgG antibodies (Fig. 1b). All vaccines induced IgG antibodies against the test proteins with varying levels, and S-6P-no-RBD induced the lowest titer of such antibodies against WT-RBD and Delta-RBD proteins (Supplementary Fig. 4a–d).

The neutralizing antibodies of related mouse sera against pseudoviruses expressing S protein of multiple SARS-CoV-2 strains were examined (Fig. 1b). S-6P-Delta-RBD elicited broad and potent neutralizing antibodies against SARS-CoV-2 WT and all VOCs tested, including Alpha, Beta, Gamma, Delta, and five Omicron subvariants (Fig. 2c–l). Notably, S-6P and S-6P-BA5-RBD induced relatively low-titer neutralizing antibodies against SARS-CoV-2 WT, and Alpha, Beta, Gamma, and Delta VOCs (Fig. 2c–g); however, S-6P-BA5-RBD elicited relatively high-titer neutralizing antibodies against the Omicron subvariants tested, particularly BA.2, BA.2.75, BA.4.6, and BA.5 (Fig. 2h–l). The neutralizing antibody titers induced by S-6P-Delta-RBD against SARS-CoV-2 WT, and Alpha, Beta, Gamma, and Delta VOCs were significantly higher (P < 0.001) than those induced by S-6P and S-6P-BA5-RBD (Fig. 2c–g), and were significantly higher (P < 0.001), or higher, than those induced by S-6P-WT-RBD against Omicron BA.1, BA.2, BA.2.75, BA.4.6, and BA.5 subvariants (Fig. 2h–l). Unlike other subunit proteins, S-6P-no-RBD induced low-titer neutralizing antibodies against SARS-CoV-2 WT (50% neutralizing antibody titer (NT₅₀) < 1:500), Alpha, Delta (NT₅₀ < 1:100), Gamma, or Omicron-BA.2.75 variants, but not against the other SARS-CoV-2 pseudoviruses tested (Fig. 2c–l). Of note, combined S-6P-Delta-RBD and S-6P-BA5-RBD (i.e., the cocktail) induced high levels of potent neutralizing antibodies against all SARS-CoV-2 strains tested (Fig. 2c–l). However, mice injected with adjuvant/PBS control produced no (or undetectable) neutralizing antibodies against these pseudoviruses (Fig. 2c–l).

To evaluate the ability of subunit vaccines to induce cellular immune responses, immunized C57BL/6 mice were analyzed for SARS-CoV-2 S-specific CD4⁺ and CD8⁺ T cells by flow cytometry in splenocytes 4 months after the 3^rd immunization (Fig. 1b). CD4⁺ and CD8⁺ T cells from all vaccination groups produced IFN-γ or TNF-α cytokines after stimulation with S-6P-no-RBD protein; cytokine levels were overall higher than those produced by splenocytes from the PBS control group (Supplementary Fig. 5a–d). Specifically, S-6P induced relatively high numbers of IFN-γ and TNF-α-secreting CD4⁺ and CD8⁺ T cells; in particular, it induced significantly higher numbers of TNF-α-secreting CD8⁺ T cells (P < 0.01 or P < 0.001) than the other immunization groups; notably, the protein cocktail further increased secretion of IFN-γ by CD8⁺ T cells (Supplementary Fig. 5a–d).

The above data demonstrate that subunit vaccines induce favorable immune responses in immunized mice. Among the single proteins tested, S-6P, particularly S-6P-BA5-RBD, alone only induced high-titer neutralizing antibodies against Omicron subvariants. By contrast, S-6P-WT-RBD, particularly S-6P-Delta-RBD, alone elicited a balanced, potent, and broad neutralizing antibody response against all SARS-CoV-2 pseudoviruses tested, with neutralizing antibody titers comparable to those induced by S-6P-Delta-RBD plus S-6P-BA5-RBD (the cocktail). Therefore, these subunit vaccines, including S-6P-Delta-RBD, have potential for development as universal COVID-19 vaccines against current and future SARS-CoV-2 VOCs.

Protective efficacy of the vaccines against SARS-CoV-2 Delta variant

To access the protective efficiency of subunit vaccines against lethal challenge with SARS-CoV-2 Delta VOC, immunized K18-hACE2 mice were challenged with a high lethal dose (10⁴ PFU/mouse) of this variant strain 2 weeks after the 3^rd vaccination, and weight loss and overall survival were monitored for 14 days (Fig. 1b). All mice receiving a single protein, or a combination, survived during the monitoring period (Fig. 3a–f). By comparison, control mice injected with adjuvant/PBS lost weight continuously, and all died by Day 10 post-challenge of Delta variant (Fig. 3g, h). In particular, no weight loss was observed in mice immunized with S-6P-Delta-RBD alone, whereas obvious or significant weight loss was observed in mice immunized with other single proteins, including S-6P, S-6P-no-RBD, S-6P-WT-RBD, and S-6P-BA5-RBD, or in mice immunized with the cocktail of proteins (Fig. 3h). These data suggest that S-6P-Delta-RBD protein alone is sufficient to prevent Delta variant-induced death and weight loss.

Protective efficacy of the vaccines against SARS-CoV-2 Omicron variant

We examined the protective efficacy of subunit vaccines against SARS-CoV-2 Omicron VOC. BA.1, a dominant subvariant during the time of the experiment, was selected for the challenge study. Immunized K18-hACE2 mice were challenged with an optimal infectious dose of Omicron BA.1 2 weeks after the 3^rd immunization. Since the Omicron variant was not lethal to the test mouse model^20,21, the viral titers and viral replication (i.e., N-gene expression) were measured 2 days post-challenge (Fig. 1b). Viral titers were highest in the lungs of control mice receiving adjuvant/PBS, which were significantly higher than those in any of the immunized groups (Fig. 3i). In particular, mice immunized with S-6P-Delta-RBD, and those with other single proteins, including S-6P, S-6P-WT-RBD, and S-6P-no-RBD, had similarly low-level viral titers in the lungs (Fig. 3i). Notably, viral titers in mice receiving combined S-6P-Delta-RBD and S-6P-BA5-RBD (the cocktail) were much higher than those in other vaccination groups (Fig. 3i). In addition, the lungs of mice receiving adjuvant/PBS showed significantly higher expression of viral N-protein than the vaccinated groups (Fig. 3j). By contrast, mice immunized with S-6P-Delta-RBD showed the lowest expression of the N-protein, followed by the mice immunized with S-6P (Fig. 3j). These data indicate that S-6P-Delta-RBD alone prevents infection and replication of SARS-CoV-2 Omicron variant in challenged mice. Therefore, this vaccine can be developed as a universal COVID-19 vaccine against SARS-CoV-2 Omicron and other variants.

Passive protective efficacy of the vaccines against SARS-CoV-2 Delta variant

To evaluate whether serum neutralizing antibodies generated by immunized mice play a critical role in preventing viral infection and virus-induced pathological effects, we injected naive K18-hACE2 mice with sera from each group of vaccinated mice, challenge them with SARS-CoV-2 Delta variant, and then monitored weight loss, viral titers, and pathological changes in the lungs (Fig. 1b). Neutralizing antibody titers in pooled sera were measured prior to injection into naive mice. Sera from S-6P-Delta-RBD-immunized mice showed the highest neutralizing antibody titers, followed by those from the S-6P-WT-RBD- and cocktail-immunized sera (Fig. 4a). Viral titers in mice receiving S-6P-Delta-RBD immune sera were significantly lower than those in the other groups, and no body weight loss was observed in these mice post-challenge (Fig. 4b, c). Of note, serum neutralizing antibodies generated by S-6P-Delta-RBD-immunized mice effectively protected naive recipient mice from virus-induced edema, without any obvious pathological changes (Fig. 4d, f). By contrast, S-6P-no-RBD-immunized mouse sera generated low or background neutralizing antibody titers (Fig. 4a). Thus, naive mice receiving these sera had significantly higher viral titers in the lungs, lost weight consistently after challenge, and suffered more edema than mice receiving other immune sera (Fig. 4b–d, i). Overall, the higher the neutralizing antibody level in transferred sera, the lower viral titer and the fewer pathological changes were observed in the lungs of recipient naive mice (Fig. 4a–k). These data demonstrate that serum neutralizing antibodies generated by S-6P-Delta-RBD-immunized mice are critical for protecting mice from subsequent challenge with the Delta variant, and that the protective efficacy correlated positively with the neutralizing antibody titers.

Protective efficacy of the vaccines against SARS-CoV

To evaluate the protective efficacy of subunit vaccines against SARS-CoV, immunized K18-hACE2 mice were challenged with a lethal dose (200 PFU/mouse) of SARS-CoV MA15 strain 2 weeks after the 3^rd immunization, and overall survival and body weight changes were monitored for 13 days post-challenge (Fig. 1b). All mice immunized with a single protein, including S-6P, S-6P-Delta-RBD, S-6P-BA5-RBD, S-6P-WT-RBD, or S-6P-no-RBD (Fig. 5a–e), and all mice receiving the two combined proteins (the cocktail) (Fig. 5f), survived without weight loss (Fig. 5h). By contrast, the control mice receiving adjuvant/PBS had consistent weight loss and 80% of them died by Day 10 post-challenge (Fig. 5g, h).

To elucidate the protective mechanisms of these subunit vaccines, splenocytes and serum samples were harvested from immunized C57BL/6 mice 4 months after the 3^rd immunization, and analyzed for SARS-CoV RBD-specific CD4⁺ and CD8⁺ T cells and neutralizing antibodies against pseudotyped SARS-CoV (Fig. 1b). S-6P-Delta-RBD protein alone elicited the highest level of such neutralizing antibodies, which were significantly higher (P < 0.001) than those induced by the other vaccination groups, including S-6P, S-6P-BA5-RBD, S-6P-WT-RBD, or S-6P-no-RBD alone, and the cocktail proteins (Fig. 5i). Whereas S-6P-Delta-RBD, S-6P-BA5-RBD, and the cocktail induced similar levels of IFN-γ or TNF-α-producing CD4⁺ T cells (Fig. 5j, k), S-6P elicited higher or significantly higher levels of IFN-γ and TNF-α-producing CD8⁺ T cells than the other vaccination groups (Fig. 5l, m), potentially explaining why mice maintained increased body weight post-SARS-CoV challenge (Fig. 5h). These data indicate that single proteins, or the combined S-6P-Delta-RBD and S-6P-BA5-RBD proteins, completely protected vaccinated mice against SARS-CoV-induced death and weight loss, and that both neutralizing antibodies and T cell responses likely play a role in protection against SARS-CoV infection. Thus, these subunit vaccines, particularly S-6P-Delta-RBD, have potential for development as universal vaccines against SARS-CoV-2 and SARS-CoV.

Experimental design and vaccine preparation

The subunit vaccines were prepared as described below¹⁸. Specifically, the DNA sequence of SARS-CoV-2 Omicron BA.1 variant containing a C-terminal Foldon trimerization domain and a His₆ tag (S-6P) was amplified by PCR based on a codon-optimized plasmid expressing the S protein of Omicron BA.1 variant of SARS-CoV-2 (GISAID accession number EPI_ISL_6795835) with HexaPro sequences (a mutated furin cleavage site and 6 proline substitutions). The amplified PCR fragment was inserted into a pLenti mammalian cell expression vector. The S-6P-no-RBD was constructed by truncating the RBD fragment of the above S-6P sequence, and ligating using ClonExpress MultiS One Step Cloning kit (Cellagen Technology). The S-6P-Delta-RBD, S-6P-BA5-RBD, and S-6P-WT-RBD were constructed by inserting the respective RBD fragment of SARS-CoV-2 Delta variant (GISAID accession number EPI_ISL_7178410), Omicron BA.5 subvariant (GISAID accession number EPI_ISL_12043290), or the ancestral SARS-CoV-2 strain (GenBank accession number QHR63250.2) using the above Cloning kit. Each recombinant plasmid was confirmed for correct sequences by sequencing analysis, and transfected into HEK293F cells, followed by purification of the related proteins from the cell culture supernatants using Ni-NTA Superflow (Qiagen).

SDS-PAGE of the purified proteins

The purified proteins were analyzed by SDS-PAGE. Specifically, the proteins were mixed with Laemmli SDS-PAGE Sample buffer (Bio-Rad), and separated by 8% SDS-PAGE Gel in the presence of Tris-Glycine running buffer, followed by staining using SimplyBlue SafeStain buffer (Thermo Fisher Scientific).

Thermal shift assay of the purified proteins

The thermal stability of the expressed proteins at variable pH values was performed using the protein gel stain reagent and real-time PCR machine based on the manufacturer’s instruction. Specifically, 10 µl of each protein was added into a 96-well PCR plate, followed by addition of 2.5 µl of 50× SYPRO Ruby Protein Gel Stain (Sigma-Aldrich). In total, 12.5 µl of Tris buffers at specific pH values was then added to each well. The melt temperature of the proteins in each well was measured by CFX Opus 96 Real-Time PCR System instrument (Bio-Rad).

Cryo-EM grid preparation and data acquisition

The isolated S-6P-no-RBD protein (4 µl at 3.78 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 4 °C under 100% chamber humidity and plunge-frozen in liquid ethane using a Vitrobot Mark IV (FEI). Cryo-EM data were collected using Latitude-S (Gatan) 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 ×130,000 (corresponding to 0.664 Å per pixel), a 20 eV slit width, a dose rate of 21 e– per Ų per second, and a total dose of 42 e − /Ų. The statistics of cryo-EM data collection are summarized in Supplementary Table 1.

Image processing

Cryo-EM data were processed using cryoSPARC v4.0.3⁴¹, and the data processing procedures are outlined in Supplementary Fig. 3. Dose-fractionated movies were first subjected to Patch motion correction and Patch CTF estimation with MotionCor2⁴² and CTFFIND-4.1.13⁴³, respectively. Images with the defocus values outside of −0.6 to −2.8 μm or the CTF fit resolutions worse than 5 Å were excluded from the further steps. Particles were picked using both Blob picker and Template picker accompanied with removing duplicate particles. Three rounds of 2D classifications were applied to remove junk particles and particles (319, 289) 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 (171, 788 particles) was finally subjected to further homogeneous, non-uniform and CTF refinements to generate a 2.8 Å resolution final map with applying C3 symmetry. 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 S-6P-no-RBD was performed in Coot-0.8.9⁴⁴ using PDB 7TGW without RBD domains as the starting model. Several rounds of refinement in Phenix-1.16⁴⁵ and manually building in Coot-0.8.9 were performed until the final reliable models were obtained. The final model has good stereochemistry by evaluation in MolProbity⁴⁶. The statistics of 3D reconstruction and model refinement are shown in Supplementary Table 1. Figures were generated using UCSF Chimera X v0.93⁴⁷.

Construction of recombinant plasmids

The recombinant plasmids were constructed as described below¹⁸. Specifically, the recombinant plasmids encoding the S protein of SARS-CoV-2 ancestral wild-type (WT) strain (GenBank accession number QHR63250.2), SARS-CoV original strain (GenBank accession number AY274119), and SARS-CoV-2 Alpha variant (GISAID accession number EPI_ISL_718813) were constructed by inserting each DNA sequence into a pcDNA3.1/V5-His-TOPO vector. SARS-CoV-2 Omicron BA.1 (GISAID accession number EPI_ISL_6795835), BA.2 (GISAID accession number EPI_ISL_12030355), BA.2.75 (GISAID accession number EPI_ISL_14384334), BA.4.6 (GISAID accession number EPI_ISL_14288784), and BA.5 (GISAID accession number EPI_ISL_12043290), as well as recombinant plasmids expressing each S protein of Beta, Gamma, and Delta variants, were constructed using a multi-site-directed mutagenesis kit, and the mutations in the RBD region were included in each construct. The sequence-confirmed plasmids were used for generation of pseudoviruses as described below.

Pseudovirus generation and neutralization assay

The pseudoviruses were generated as described below^48–50. Specifically, each of the above recombinant plasmid was co-transfected with the pLenti-CMV-luciferase and PS-PAX2 plasmids into 293T cells using a PEI transfection method. Pseudovirus-containing culture supernatants were collected 72 h after transfection, and incubated with serially diluted mouse sera (2–4-fold serial dilutions for serum neutralizing antibodies against different strains of SARS-CoV-2 pseudovirus, and two-fold serial dilutions for serum neutralizing antibodies against SARS-CoV pseudovirus) at 37 °C for 1 h. The virus-serum mixture was then added to 293T cells expressing human ACE2 receptor (hACE2/293 T). 24 h later, fresh medium was added to the cells, which were further cultured for 48 h. The cells were then sequentially incubated with cell lysis buffer and luciferase substrate (Promega), and measured for relative luciferase activity using Cytation 7 Microplate Multi-Mode Reader and Gen5 software. 50% pseudovirus neutralization was calculated as NT₅₀.

Ethics statement

Male and female K18-hACE2 transgenic mice (6–8-week-old) and C57BL/6 mice (8–10-week-old) were used in this study, and they were randomly assigned to each group. The animal protocols were approved by the Institutional Animal Care and Use Committees (IACUC) of Georgia State University and University of Iowa. All mouse-related experiments were conducted by strictly following our approved protocols and the Guidelines for the Care and Use of Laboratory Animals of National Institutes of Health. Mice reaching 25% weight loss with significant clinical symptoms, or 30% weight loss, were humanely euthanized by cervical dislocation under anesthesia.

Animal immunization and serum collection

The K18-hACE2 mice were i.m. immunized with each protein (10 μg/mouse) or PBS control plus Alum (500 μg/mouse) and MPL (10 μg/mouse) adjuvants (InvivoGen). The selection of the above adjuvants and the doses of the proteins was based on our previously optimized schedules^18,37,38. The cocktail is the combination of S-6P-Delta-RBD and S-6P-BA5-RBD proteins (5 μg for each protein, 10 μg/mouse). These mice were further boosted twice with the same immunogen and adjuvants every 3 weeks. The sera were collected from each mouse 10 days after the 3^rd immunization and measured for neutralizing antibodies against SARS-CoV-2 and SARS-CoV (as described above), and IgG antibody responses (as described below), or pooled for passive transfer to naive mice for subsequent challenge studies (as described below). In a separate experiment, C57BL/6 mice were immunized as described above, and sera and splenocytes were collected 4 months after the 3^rd immunization for detection of neutralizing antibodies (as described above) and T cell responses (as described below).

ELISA

ELISA was carried out to test S/RBD-specific IgG antibodies in immunized mouse sera^38,48. Specifically, ELISA plates were coated with the respective SARS-CoV-2 S or RBD protein (1 μg/ml) at 4 °C overnight, and blocked at 37 °C for 2 h, with 2% fat-free milk in PBS containing 0.05% Tween-20 (PBST). After washing with PBST for five times, the plates were sequentially incubated with 3-fold serially diluted sera, and horseradish peroxidase (HRP)-conjugated anti-mouse IgG (Fab specific) antibody (1:60,000 dilution; Sigma, #A9917) at 37 °C for 1 h. The plates were washed again as described above, incubated with TMB (3,3’,5,5’-Tetramethylbenzidine) substrate (Sigma, #T4319), and stopped for reaction by H₂SO₄ (1 N). The absorbance at 450 nm (A450) was measured using Cytation 7 Microplate Multi-Mode Reader.

Flow cytometry analysis

Flow cytometry was carried out to analyze SARS-CoV-2 or SARS-CoV-specific CD4⁺ and CD8⁺ T cell responses in the mouse splenocytes^51,52. Briefly, splenocytes (1 × 10⁶ cells/ml, 200 µl/well) were incubated with 5 μg/ml of S-6P-no-RBD (for SARS-CoV-2) or a Fc-fused SARS-CoV RBD (for SARS-CoV)⁵⁰ protein diluted in RPMI 1640 medium containing 10% FBS and mouse IL-2 (R&D Systems, 1 ng/ml) at 37 °C. After 41 h, the cells were restimulated as above for 1 h, followed by incubation with Brefeldin A (5 μg/ml; Biolegend, #420601) and BD GolgiStop™ Protein Transport Inhibitor (containing Monensin) (1:1500 dilution; BD Biosciences, #554724) for another 6 h. After washing with PBS, the cells were stained with Fixable Viability Dye eFluor™ 660 (1:1000 dilution; eBioscience, #65-0864-14) to exclude dead cells. After incubation with TruStain FcX plus (1:200 dilution; Bioligand, #156604) to block Fc receptors, the cells were then stained with anti-mouse CD45-PE-Cy7 (1:333 dilution; BD Biosciences, #552848) and anti-mouse-CD8-FITC (1:200 dilution; BD Biosciences, #553031) or anti-mouse-CD4-FITC (1:200 dilution; Biolegend, #100406) antibodies. After fixation and permeabilization (Fixation/Permeabilization Kit, BD Biosciences, #554714), the cells were stained with anti-mouse-IFN-γ-PE (1:100 dilution; Biolegend, #505808) or anti-mouse-TNF-α-BV421 (1:100 dilution; Biolegend, #506328) antibody, and analyzed by CytoFLEX flow cytometer (Beckman Coulter Life Sciences). The gating strategy and representative plot of flow cytometry are shown in Supplementary Figs. 6 and 7.

Challenge of immunized mice with SARS-CoV-2 or SARS-CoV

Three separate challenge experiments were performed 2 weeks after the 3^rd immunization as described below^18,38,53. (1) The immunized mice were intranasally (i.n.) infected with SARS-CoV-2 Delta variant (a high lethal dose of 10⁴ plaque forming units (PFU)/mouse, 50 μl/mouse), and observed for body weight changes and survival for 14 days after challenge. (2) The immunized mice were i.n. infected with SARS-CoV-2 Omicron variant (BA.1 subvariant, an optimal infectious dose of 10⁵ PFU/mouse, 50 μl/mouse). The mice were then sacrificed two days after infection, and the lungs were collected for measurement of viral titers and replication. (3) The immunized mice were i.n. infected with SARS-CoV MA15 strain (a lethal dose of 200 PFU/mouse, 50 μl/mouse), and observed for body weight changes and survival for 13 days after infection (note: the reason that these mice were not observed for 14 days post-SARS-CoV challenge, as for post-SARS-CoV-2 Delta variant challenge, was due to closure of the BSL-3 laboratory for yearly maintenance).

Challenge of naive mice receiving immune sera with SARS-CoV-2 Delta variant

Pooled mouse sera (200 μl/mouse) from the respective immunization groups were intraperitoneally (i.p.) administered to naive mice⁵⁴. Six hours after serum administration, the mice were i.n. infected with SARS-CoV-2 Delta variant (5 × 10³ PFU/mouse: a dose optimized for evaluating viral titers and pathological changes in the lungs; 50 μl/mouse). Four days after the viral infection, the mice were euthanized for lung collection. Half lungs were collected in PBS for measurement of viral titers by plaque assay, and the other half were collected in Zinc formalin for pathological analysis, as described below.

Viral titer detection

Lungs collected from SARS-CoV-2 Omicron or Delta variant-challenged mice were measured for viral titers by a standard plaque assay as described below⁵³. Specifically, lung tissues were homogenized and centrifuged, and the supernatants were diluted in DMEM cell culture medium, followed by incubation with Vero E6 cells at 37 °C for 1 h. The inoculation was removed, and the cells were then overlaid with 0.6% agarose and cultured for three days. After removing the overlays, the cells were stained with 0.1% crystal violet to show plaques. Viral titers were calculated as PFU/ml of mouse lung tissues.

RNA isolation and qRT-PCR

Lungs collected from SARS-CoV-2 Omicron variant-challenged mice were also measured for viral replication as described below⁵³. Specifically, RNA was extracted from virus-infected mouse lungs using Invitrogen TRIzol reagent (Thermo Fisher Scientific) according to the manufacturer’s instructions. In all, 1 μg of total RNA was used as template for the first strand of cDNA. The resulting cDNA was subjected to amplification of selected genes by real-time quantitative PCR (qRT-PCR) using Power SYBR Green PCR Master Mix (Applied Biosystems). Viral nucleocapsid (N) gene was detected using nCOV_N1 primer (Integrated DNA Technologies, 10007031). The expression levels were normalized to glyceraldehyde-3-phosphate dehydrogenase (GAPDH) using the following threshold cycle (CT) equation: ΔCT = CT of the gene of interest − CT of GAPDH. The results are expressed as a ratio to GAPDH calculated as 2^−ΔCT.

Pathological analysis of lung tissues

Lungs collected from SARS-CoV-2 Delta variant-challenged mice were analyzed for pathological effects. Specifically, paraffin-embedded lung tissue sections were stained using the hematoxylin and eosin (H&E) method, and the relevant slides were examined for pathological effects using a grouped masking approach⁵⁵. Edema distribution in the lungs were ordinally numbered⁵⁶, with the scores being reported as 0 (none), 1 (< 25%), 2 (26–50%), 3 (51–75%), and 4 (> 75%) of tissue fields. High resolution images were taken using a BX53 Microscope (DP73 digital camera) and CellSens Dimension software (Olympus).

Statistical analysis

GraphPad Prism 9.0.2 statistical software was used to calculate statistical significances among various vaccination groups. Ordinary one-way ANOVA (Dunnett’s multiple comparison test) was performed to assess statistical significances between S-6P-Delta-RBD and other groups, and Ordinary one-way ANOVA (Tukey’s multiple comparison test) was performed to assess statistical significances among different groups. The respective statistical tests are shown in the related figure legends. *, **, and *** designate P < 0.05, P < 0.01, and P < 0.001, respectively.