Nanoparticle Vaccines Based on the Receptor Binding Domain (RBD) and Heptad Repeat (HR) of SARS-CoV-2 Elicit Robust Protective Immune Responses

Abstract

Various vaccine strategies have been proposed in response to the global COVID-19 pandemic, each with unique strategies for eliciting immune responses. Here, we developed nanoparticle vaccines by covalently conjugating the self-assembled 24-mer ferritin to the receptor binding domain (RBD) and/or heptad repeat (HR) subunits of the Severe Acute Respiratory Syndrome Coronavirus-2 (SARS-CoV-2) spike (S) protein. Compared to monomer vaccines, nanoparticle vaccines elicited more robust neutralizing antibodies and cellular immune responses. RBD and RBD-HR nanoparticle vaccinated hACE2 transgenic mice vaccinated with RBD and/or RBD-HR nanoparticles exhibited reduced viral load in the lungs after SARS-CoV-2 challenge. RBD-HR nanoparticle vaccines also promoted neutralizing antibodies and cellular immune responses against other coronaviruses. The nanoparticle vaccination of rhesus macaques induced neutralizing antibodies, and T and B cell responses prior to boost immunization; these responses persisted for more than three months. RBD- and HR-based nanoparticles thus present a promising vaccination approach against SARS-CoV-2 and other coronaviruses.

Keywords: COVID-19; HR; RBD; SARS-CoV-2; nanoparticle vaccine.

Graphical abstract

Figure 1

Construction and Purification of RBD- and HR-based Nanoparticle Vaccine (A) Schematic of vaccine components which were 6 × His-tagged SC-Ferritin, ST-RBD and ST-HR. SC: SpyCatcher. SP: secretory signal peptide. ST: SpyTag. (B) The procedure of nanoparticle vaccine production. SC-Ferritin was expressed and purified from Escherichia coli (E.coli). ST-RBD and ST-HR were expressed and purified from CHO-S cells. SC-Ferritin and ST-RBD or ST-HR were mixed and incubated in Tris buffer to facilitate ST/SC irreversible conjugation. The conjugated nanoparticles were separated with SEC and concentrated by ultrafiltration device. (C) Schematic illustration of Ferritin-based RBD and RBD-HR nanoparticles. RBD nanoparticle contained 100% RBD-Ferritin. RBD-HR nanoparticle contained 70% RBD-Ferritin and 30% HR-Ferritin. (D) Coomassie blue staining and western blotting of Ferritin, RBD, HR, RBD nanoparticle, and RBD-HR nanoparticle. Both His and Ferritin antibodies were used to confirm the expression and purity of each protein. (E) SEC of Ferritin core, RBD-Ferritin, and RBD-HR-Ferritin. The ultraviolet absorptions at 280 were shown. The retention volume represented peaks of each nanoparticles. (F) TEM images and two-dimensional (2D) reconstruction of each nanoparticle. See also Figure S1.

Figure 2

Humoral Immune Responses in Nanoparticles Vaccinated BALB/c Mice (A) Schematic of BALB/c vaccination. Six mice from each group were prime/boost-vaccinated with different vaccines at week 0 and week 4. Serum was collected every two weeks. All mice were euthanized at week 10. (B–E) SARS-CoV-2 RBD- and HR-specific IgG titers of immunized BALB/c mice at each time point were detected by ELISA. IgG antibody titers of serum which collected at week 6 were determined by serial dilution, and represented as the reciprocal of the endpoint serum dilution (B and C) (n = 6). RBD and HR-specific IgG titers in each week were calculated and plotted as time-course curve (D and E). (F) FRNT50 of nAbs of each vaccine group was determined by FRNT and represented as half-maximal inhibitory concentrations (IC50), which was the reciprocal of half-maximal neutralizing dilution (n = 3). The right panel of (F) showed the representatives of FRNTspot wells within 1:100 and 1:1000 dilution groups. (G) FRNTspots of serum of each time point in 1:100 and 1:1000 dilutions. (H and I) The percentages of RBD-specific IgG1⁺ and IgG2b⁺ MBCs (CD19⁺B220⁺CD38⁺) within spleen of each vaccine group (n = 4). Experiments were conducted independently in triplicates. Data represented as mean ± SEM. Adjusted p values were calculated by one-way ANOVA with Tukey’s multiple comparisons test. ^∗p < 0.05, ^∗∗p < 0.01, ^∗∗∗p < 0.001, ^∗∗∗∗p < 0.0001. See also Figures S2 and S3.

Figure 3

T Cell Immune Responses in Nanoparticles Vaccinated BALB/c Mice (A–C) BALB/c mice were euthanized at six weeks post boost vaccination. Splenocytes were incubated with SARS-CoV-2 S peptides pool. The percentages of IFN-γ⁺, IL-2⁺ and TNF-α⁺ CD8⁺ T cells were determined by ICCS. (D and E) Splenocytes were stimulated with S peptide pool as in (A). The percentages of IFN-γ⁺ and IL-4⁺ CD4⁺ T cells were determined by ICCS. (F and G) Splenocytes were stimulated with S peptide pool. ELISpot assays were conducted for IFN-γ and IL-4 secretion in splenocytes. Experiments were conducted independently in triplicates. Data represented as mean ± SEM (n = 4). Adjusted p values were calculated by one-way ANOVA with Tukey’s multiple comparisons test. ^∗p < 0.05, ^∗∗p < 0.01, ^∗∗∗p < 0.001. See also Figure S3.

Figure 4

Potent Antigen Presentation and T-B Coordination Induced by Nanoparticle Vaccines (A) Schematic of in vitro DC and macrophage antigen internalization experiment. Six C57BL/6 mice were subcutaneously injected with equal moles of RFP-tagged RBD and RFP-tagged RBD-Ferritin, both proteins were adjuvanted with SAS adjuvant. 4 h post injection, inguinal lymph nodes from both sides were obtained and proceeded to FCM analysis to determine the percentages of RFP-positive DCs (B220^-CD11c^hiMHC-II⁺) and macrophages (B220^-CD11b⁺F4/80^-CD169⁺). (B-E) The FCM results of RFP-positive DCs and macrophages. (B) and (D) represented the typical FCM figures. (C) and (E) represented the statistical graphs of the FCM results (n = 8 for DCs, n = 10 for macrophages). (F) Cryosections of inguinal lymph nodes were immunostained with antibodies against CD11b. RFP-positive cells indicated RFP-tagged RBD-Ferritin nanoparticles. The blue staining indicated DAPI-stained nuclei. Scale bars in the upper panel represented 200 μm. Scale bars in the lower panel represented 100 μm. (G) Schematic of in vitro DC antigen presentation experiments. PBMCs were harvested from three healthy individuals and proceeded to monocyte isolation. Mature DCs were induced and then loaded with RBD or RBD-Ferritin antigens, followed by co-culture with autologous CD8⁺ T cells. ELISpot assays were conducted for IFN-γ CD8⁺ T cells. (H) ELISpot results of in vitro DC antigen presentation experiment in three healthy donors (n = 3 for Donor1 and Donor 2, n = 4 for Donor 3). (I and J) BALB/c mice were immunized with different vaccines. Ten days post immunization, mice were euthanized. The percentages of Tfh cells (CD4⁺CXCR5⁺PD-1⁺) and GC B cells (CD19⁺B220⁺CD95⁺ GL7⁺) were determined by FCM (n = 3). Experiments were conducted independently in triplicates. Data represented as mean ± SEM. P values in (C), (E), and (H) were calculated by Student’s t test. Adjusted p values in (I) and (J) were calculated by one-way ANOVA with Tukey’s multiple comparisons test. ^∗p < 0.05, ^∗∗p < 0.01, ^∗∗∗p < 0.001. See also Figure S3.

Figure 5

Cross Reactivity of Nanoparticle Vaccine-induced Antibodies and T Cell Responses (A) Six BALB/c mice within each group were prime-boost-immunized with Ferritin core, HR monomer and HR-Ferritin nanoparticle, respectively. Serum was collected two weeks post boost vaccination and incubated with authentic SARS-CoV-2, followed by incubating with Vero E6 cells. The FRNTspots in 1:100 dilution were plotted for each group (n = 6). (B–F) Cross-neutralization of serum of immunized BALB/c was detected with pseudotyped-CoVs which contained SARS-CoV, MERS-CoV, HCoV-229E, HCoV-OC43, and RATG13 (n = 3). Neutralizations at dilution of 1: 30 of serum were shown. (G and H) Splenocytes of RBD and RBD-HR nanoparticle vaccines-immunized mice were incubated with hCoV-OC43 S peptides pool and hCoV-229E S peptides pool, respectively. ELISpot assay was conducted for IFN-γ secretion in splenocytes. Experiments were conducted independently in triplicates. Data represented as mean ± SEM. Adjusted p values in (A–F) were calculated by one-way ANOVA with Tukey’s multiple comparisons test. P values in (G) and (H) were calculated by Student’s t test. ^∗∗p < 0.01, ^∗∗∗p < 0.001, ^∗∗∗∗p < 0.0001.

Figure 6

Protection of Nanoparticle Vaccines against SARS-CoV-2 in hACE2 Mice (A) Schematic of hACE2 mice vaccination. Six mice within each group were prime/boost-vaccinated with different vaccines at week 0 and week 4. Two weeks post boost, mice were challenged with authentic SARS-CoV-2. All mice were euthanized one week post challenge. Serum was collected at week 0, 2, 4, 6, and 7. (B and C) RBD-specific and HR-specific IgG antibodies titers of serum which collected at week 6 were determined by serial dilution and represented as the reciprocal of the endpoint serum dilution (n = 6). (D) Serum of each mice was 10-fold serially diluted and incubated with 500 FFU of authentic SARS-CoV-2, followed by incubating with Vero E6 cells. The FRNTspots of each well were counted. FRNT50 of nAbs of each vaccine group was determined by FRNT and represented as IC50 which was the reciprocal of half-maximal neutralizing dilution. The right panel showed the representatives of FRNTspot wells within 1:100 and 1:1000 dilution groups. (E) Viral RNA copies in lung of each mice were determined by qRT-PCR and plotted as log10 copies per ml. (F) HE staining and IHC against N proteins were evaluated in lungs of each mice. (G) Serum of each vaccination groups was 10-fold serially diluted and mixed with pseudotyped SARS-CoV-2, followed by incubating with mFcγRI- and mFcγRII-expressing HEK293T cells. The ADE was evaluated by measuring luciferase units of each samples. (H) Sera of naive Ifnar1^−/− mice and ZIKV-infected Ifnar1^−/− mice were 10-fold serially diluted and mixed with pseudotyped ZIKV. Serum/virus mixtures were incubated with mFcγRI/II-HEK293T cells. The luciferase units of each samples were measured 48 h post infection to evaluate ADE. Scale bar in (F) represented 50 μm. Experiments were conducted independently in triplicates. Data represented as mean ± SEM. Adjusted p values in (B–E) were calculated by one-way ANOVA with Tukey’s multiple comparisons test. Data in (G) were analyzed by two-way ANOVA with Tukey’s multiple comparisons test. Adjusted p values in (H) were calculated by two-way ANOVA with Sidak’s multiple comparisons test. ^∗p < 0.05, ^∗∗p < 0.01, ^∗∗∗p < 0.001, ^∗∗∗∗p < 0.0001. See also Figures S4 and S5.

Figure 7

The Immunogenicity of Nanoparticle Vaccines in Rhesus Macaques (A) Schematic of rhesus macaque vaccination. Twelve monkeys were prime-boost-vaccinated with RBD-HR, RBD nanoparticle and RBD-HR nanoparticle, respectively. Serum was collected every two weeks. (B and C) RBD-specific and HR-specific IgG antibodies titers of serum which collected at week 6 were determined by serial dilution (n = 4). (D) Serum of each monkey was 10-fold serially diluted and incubated with 500 FFU of authentic SARS-CoV-2, followed by incubating with Vero E6 cells. The FRNTspots of each well were counted. FRNT50 of nAbs of each vaccine group was determined by FRNT and represented as IC50 which was the reciprocal of half-maximal neutralizing dilution (n = 4). The right panel showed the representatives of FRNTspot wells within 1:100 and 1:1000 dilution groups (n = 4). (E) FRNTspots of serum at each time point in 1:100 and 1:1000 dilutions. (F and G) PBMCs of 12 vaccinated monkeys and 4 even-aged mock monkeys were collected at week 10 and incubated with S peptides pool. ELISpot assay for IFN-γ and IL-4 in PBMCs. (H) Serum of each vaccinated monkey was 10-fold serially diluted and mixed with pseudotyped SARS-CoV-2, followed by incubating with monkey FcγRI- and monkey FcγRII-expressing HEK293T cells. The ADE of each serum was evaluated by measuring the luciferase units of each sample. Experiments were conducted independently in triplicates. Data represented as mean ± SEM. Adjusted p values in (B–D) and (F) and (G) were calculated by one-way ANOVA with Tukey’s multiple comparisons test. Data in (H) were analyzed by two-way ANOVA with Tukey’s multiple comparisons test. ^∗p < 0.05, ^∗∗p < 0.01, ^∗∗∗p < 0.001, ^∗∗∗∗p < 0.0001. See also Figures S6 and S7.