A CTB-SARS-CoV-2-ACE-2 RBD Mucosal Vaccine Protects Against Coronavirus Infection

Abstract

Mucosal vaccines protect against respiratory virus infection by stimulating the production of IgA antibodies that protect against virus invasion of the mucosal epithelium. In this study, a novel protein subunit mucosal vaccine was constructed for protection against infection by the beta coronavirus SARS-CoV-2. The vaccine was assembled by linking a gene encoding the SARS-CoV-2 virus S1 angiotensin converting enzyme receptor binding domain (ACE-2-RBD) downstream from a DNA fragment encoding the cholera toxin B subunit (CTB), a mucosal adjuvant known to stimulate vaccine immunogenicity. A 42 kDa vaccine fusion protein was identified in homogenates of transformed E. coli BL-21 cells by acrylamide gel electrophoresis and by immunoblotting against anti-CTB and anti-ACE-2-RBD primary antibodies. The chimeric CTB-SARS-CoV-2-ACE-2-RBD vaccine fusion protein was partially purified from clarified bacterial homogenates by nickel affinity column chromatography. Further vaccine purification was accomplished by polyacrylamide gel electrophoresis and electro-elution of the 42 kDa chimeric vaccine protein. Vaccine protection against SARS-CoV-2 infection was assessed by oral, nasal, and parenteral immunization of BALB/c mice with the CTB-SARS-CoV-2-ACE-2-RBD protein. Vaccine-induced SARS-CoV-2 specific antibodies were quantified in immunized mouse serum by ELISA analysis. Serum from immunized mice contained IgG and IgA antibodies that neutralized SARS-CoV-2 infection in Vero E6 cell cultures. In contrast to unimmunized mice, cytological examination of cell necrosis in lung tissues excised from immunized mice revealed no detectable cellular abnormalities. Mouse behavior following vaccine immunization remained normal throughout the duration of the experiments. Together, our data show that a CTB-adjuvant-stimulated CTB-SARS-CoV-2-ACE-2-RBD chimeric mucosal vaccine protein synthesized in bacteria can produce durable and persistent IgA antibodies in mice that neutralize the SARS-CoV-2 subvariant Omicron BA.1.1.

Keywords: COVID-19; SARS-CoV-2; angiotensin converting enzyme (ACE-2) receptor binding domain (RBD); beta coronaviruses; cholera toxin B subunit; mucosal immunization; protein subunit vaccine; sIgA; virus neutralization.

Figure 1

Construction of a CTB-SARS-CoV-2 mucosal vaccine: A 309 bp fragment of a gene encoding the cholera toxin B subunit (CTB) mucosal adjuvant was linked to the 5′ end of a DNA fragment encoding the SARS-CoV-2-ACE-2-RBD coronavirus S1 antigen. A Pro–Gly–Pro–Gly amino acid hinge region (H) was inserted between the CTB and ACE-2-RBD DNA fragments to allow molecular flexibility and to enhance ACE-2-RBD and CTB protein epitope availability for binding by antigen-presenting cells. The CTB-SARS-CoV-2-ACE-2-RBD fusion gene was inserted into the multiple cloning site (MCS) of a His-tag-based E. coli expression vector for nickel affinity column isolation of the vaccine fusion protein from bacterial homogenates.

Figure 2

Identification of the CTB-ACE-2-RBD fusion protein in E. coli BL-21 cells: Following IPTG amplification of the vaccine protein in recombinant E. coli BL-21 cells, the cells were lysed, and the centrifuged homogenate proteins (H) were resuspended and separated by electrophoresis on 12% polyacrylamide gels. (Panel A) Lanes from left to right contain: (Lane 1) Protein molecular weight standards. (Lane 2) An approximate 42 kDa protein band excised from the homogenate gel separation and predicted to contain the CTB-SARS-CoV-2-ACE-2-RBD fusion protein was re-electrophoresed on a new gel and stained with Coomassie blue. (Lane 3) Molecular weight marker proteins negatively stained on the corresponding immunoblot. (Lane 4) Immunoblot detection of the ACE-2-RBD protein in the 42 kDa homogenate band identified by reaction with a primary human antibody (Ab¹) made against the ACE-2-RBD protein followed by binding a mouse anti-human IgG (Ab²), conjugated to alkaline phosphatase. (Panel B) Confirmation of vaccine protein synthesis in transformed E. coli BL-21 cells: From left to right, (Lane 1) contains prestained protein molecular weight markers. (Lanes 2 and 3) Electro-elution of bacterial proteins in the MW range approximating 50 kDa (H-T), designated Homogenate Top band to 37 kDa (H-B) and Homogenate Bottom band. The protein bands were visualized by staining with Coomassie Blue. (Lane 4) Control untransformed BL-21 cell proteins in the same molecular weight range (50–37 kDa). (Lane 5) Negatively stained protein molecular weight markers on immunoblot of identical samples shown in the first 4 lanes. (Lanes 6–7) Immunoblot from identical proteins in lanes 2 and 3 probed with Ab¹ antibodies specific for the human ACE-2-RBD protein. (Lane 8) Immunoblot of the control untransformed E. coli BL-21 cell proteins identical to lane #4 and probed with human ACE-2-RBD antibodies (Ab¹). (Panel C) is identical to (Panel B), with the exception that the Coomassie Blue-stained homogenate lanes were overlayed with the corresponding immunoblot to demonstrate more precisely the location of the vaccine protein in homogenate proteins isolated from the transformed bacterial cells, in addition to demonstrating the absence of vaccine protein in the homogenate of untransformed bacterial cells.

Figure 3

FPLC purification of the CTB-SARS-CoV-2-ACE-RBD mucosal vaccine fusion protein: Electro-elution of a 42–50 kDa protein band following acrylamide gel electrophoresis of transformed E. coli BL-21 cell homogenates may contain, in addition to the vaccine protein, bacterial proteins of a similar molecular weight. To further purify the vaccine protein, the electroeluted proteins were separated by fast protein liquid chromatography (FPLC) on two tandem Agilent SEC-5 columns (5 µm, 300 A, 4.6 × 100 mm) packed with 5 µm silica particles coated with a neutral, hydrophilic layer to aid in protein separation, connected in series and run in PBS buffer, pH 7.4, at 0.4 mL/min on a GE Healthcare FPLC system. The column eluate was monitored by UV spectrophotometry at 280 nm, and protein fractions (0.25 mL) were collected manually. The vaccine fusion protein was identified by ELISA assay in the 5.0–6.0 mL column fractions. In the assay, individual column fraction samples were bound to the wells of an Immulon 1B plate (Thermo Fisher Scientific Inc.). The vaccine protein was identified with an anti-His-Tag Mouse Monoclonal primary antibody (Thermo Fisher Scientific Inc. #37-2900) and a goat anti-mouse (H&L) peroxidase secondary antibody (Thermo Fisher Scientific Inc. #62-6520). Slower-moving molecules and molecular debris were eluted between 7.5 and 8.5 mL.

Figure 4

ELISA detection of SARS-CoV-2-ACE-2-RBD-specific antibodies in immunized mouse serum: This ELISA permits quantitative assessment of vaccine-induced antibodies that specifically neutralize the SARS-CoV-2 coronavirus. Briefly: (1) The wells of a 96-well microplate were coated with glycosylated ACE-2-RBD protein from the SARS-CoV-2 Omicron variant. (2) The ACE-2-RBD coated wells were incubated with serum dilutions from vaccine immunized mice, allowing (3) anti-RBD antibodies present in the serum to bind the ACE-2-RBD protein. (4) The plate was washed with PBS to remove excess serum, and (5) the remaining free ACE-2-RBD molecules were detected by incubation with recombinant ACE-2 biotinylated antigen and streptavidin-horseradish peroxidase (HRP) conjugate. (6) Addition of TMB + H₂O₂ substrates to each well initiated an oxidation reaction by HRP to convert colorless TMB to a blue color. The reaction was terminated by acidification with H₂SO_4, and the now-yellow color was measured at OD 450 nm. If there are few or no antibodies in the serum to bind to the ACE-2-RBD, the color in the well will be bright yellow. However, if anti-RBD antibodies available in the serum have blocked the ability of biotinylated ACE-2 to bind the ACE-2-receptor binding domain, a color reduction or no color will be detected.

Figure 5

Optimization of SARS-CoV-2-ACE-2-RBD specific antibody levels. A sensitive virus neutralization ELISA assay (DIA.PRO ACE-2-RBD) was used to compare several immunization protocols for optimum production of SARS-CoV-2-ACE-2-RBD antibodies in coronavirus-vaccinated mice. Mouse anti-virus antibody inhibition of biotinylated ACE-2 protein binding to the virus ACE-2 receptor binding domain (OD₄₅₀ reduction) is presented on the Y axis. All vaccination protocols produced significant ACE-2-RBD-specific antibody levels in comparison with the unimmunized control. 4x-refers to delivery of four separate immunizations to the mice at intervals of 3–4 weeks between immunizations. Statistical analysis of the data from each experimental group was compared with the control group by a nonparametric Wilcoxon test where probability values ranged from * = p-value < 0.05, ** = p-value < 0.01, *** = p-value < 0.001, ns = p-value > 0.05. The error bars represent SEM.

Figure 6

Mucosal vaccination generates virus specific IgA antibodies. Several vaccination regimes were examined by an indirect ELISA assay for generating optimum titers of SARS-CoV-2-RBD-specific IgA antibodies. Microtiter plate wells coated with the ACE-2 receptor protein were incubated with undiluted serum from mice immunized by oral, nasal, and IP injection routes. 4x refers to delivery of four separate immunizations to the animals at an interval of 3–4 weeks between each immunization. Virus-receptor-specific IgA antibody levels were detected by incubation of the wells with goat anti-mouse IgA alpha chain conjugated to horseradish peroxidase (HRP). Peroxidase activity was detected by addition of TMB substrate and H₂O₂ to each well and the color reaction measured at OD₄₅₀ nm. (Increases in OD₄₅₀ = greater amounts of IgA). In box plot format, the IgA titration data for each treatment group was divided into quartiles (25% of the population). The boxes represent the second and third quartiles. The horizontal bars represent the median values for the group, with 25% of the group above and 25% below the median separating the second and third quartiles. The whiskers represent the first and last quartiles. The black dots are individual animal IgA OD₄₅₀ absorbance values. Analysis of the data by Wilcoxon test comparing individual vaccination methods to the naive unvaccinated control indicated that IP injection (brown box) was the most effective immunization protocol for generating maximum IgA levels. Comparison among vaccinated groups by Wilcoxon test indicated one significant difference in IgA antibody levels between the IP injected and Oral immunized groups. The asterisk is used to denote the level of statistical significance associated with a p-value. (* = p-value < 0.05, ** = p-value < 0.01, *** = p-value < 0.001, **** = p-value < 0.0001).

Figure 7

The combination of mucosal and parenteral immunization enhances virus neutralization. Coronavirus neutralizing antibody titers were evaluated microscopically by a virus neutralization assay (VN) in Vero E6 Green Monkey kidney cells. In this box plot graph, the Y axis represents serial dilutions of mouse serum starting from a ¼ dilution of the serum. The X axis displays individual mice in each group that generated virus-neutralizing serum titers (black dots). In the Blue box, the mice were primed and boosted at 4 weeks by IP injection with the Sputnik V DNA vaccine. In the Red box, the mice were orally primed with 15 µg mucosal vaccine and boosted once by IP injection of the Sputnik 5 vaccine followed by 2× oral gavage with 15 µg mucosal vaccine protein at 4 wk intervals. In this graph, the samples are divided into quartiles. The boxes represent the second and third quartiles, with the black horizontal line in each box, the sample median, separating the second and third quartiles. The whiskers represent the first and last quartiles. The Oral + Sputnik V + Oral + Oral group (Red box) showed a minimum effective titer at 1:8 and a maximum at 1:64. The Sputnik V immunization group (Blue box), showed a minimum effective titer at 1:32 and a maximum at 1:64.

Figure 8

Oral vaccine neutralization of the SARS-CoV-2 Omicron variant B.1.1.529. Serial dilutions of serum harvested from oral-vaccine-immunized mice were employed to neutralize a constant amount of SARS-CoV-2 Omicron virus variant. In this assay, vaccine-immunized mouse serum + corona virus samples were added to Vero E6 African green monkey kidney cells in tissue culture plates. After several days’ incubation of cell cultures infected with the virus + serum dilution mixtures, the level of virus neutralization was determined by light microscopy, where the titer was read as the highest serum dilution where the detected cytopathic effect (CPE) of cell necrosis was above 50%. As vaccine titers were decreased, there was observed a concomitant decrease in the number of Vero cells (1:32 serum dilution). Further, as vaccine titers were reduced, increased clusters of apoptotic cells were detected (1:32–1:64 serum dilutions), allowing the virus to infect more cells. Disappearance of the background lawn of normal attached Vero cells was seen to dramatically increase as virus-infected cells became necrotic, clumped together and lysed (1:32–1:128 serum dilutions). Microscope magnification = 100×.

Figure 9

Immunohistochemical detection of SARS-CoV-2 specific antibodies in vaccinated mouse alveolar lung tissues. Representative paraffin-embedded lung alveolar tissue sections from CTB-SARS-CoV-2-ACE-2-RBD protein vaccinated BALB/c mice Panel (A) and unvaccinated mice Panel (B). Slides containing several mouse lung alveolar tissue sections (10µm) were incubated for 1 h with the Hungarian SARS-CoV-2 virus isolate SARS-CoV-2/human/HUN/CMC1/2020, GenBank OQ302121.1 wt., at a virus dilution of 1:1000. After virus incubation, the slides were washed extensively in DI water to remove unbound virus from the sections. After washing, the sections were incubated for 1 h with human anti-SARS-CoV-2 antibodies (1:10 dilution). Following antibody incubation, the sections were washed several times to remove unbound antibodies. Finally, the sections were incubated with anti-human IgG-HRP conjugate (Abcam, 1:1000 dilution) for 1 h at room temp. The sections were then incubated with chromogen stain (EnVision Flex HRP magenta substrate, Dako Omnis Code Nr: GV925). Representative tissue sections were inspected for virus-specific antibody bound to residual virus by light microscopy. In Panel (A), section areas with dark red–purple stain (indicated by arrows), indicate the presence of coronavirus-specific antibodies binding residual coronavirus particles in virus-inoculated lung alveolar tissue sections from immunized mice. In Panel (B), the absence of antibody staining indicates no coronavirus-specific antibodies are present in virus-inoculated lung alveolar tissue sections from unimmunized mice. Microscope magnification = 400×.