Cell lines, virus, antibody reagents, and receptors. Vero E6 cells (ATCC, CRL-1586) were maintained in Minimal Eagles Medium (MEM) supplemented with 10% fetal bovine serum, 1% glutamine and 1% penicillin and streptomycin. The SARS-CoV-2 (WA-1) isolated was obtained from the Center for Disease Control (WA-1 strain) and stock virus prepared in Vero E6 cells. Histidine-tagged hACE2 and histidine-DPP4 receptors were purchased from Syno Biologics (Beijing, CN). Rabbit anti-SARS-CoV spike protein was purchased form Biodefense and Emerging Infections Research Resources Repository (catalog no. NR-4569, BEI Resources, Manassas, VA).
SARS-CoV-2 protein expression. SARS-CoV-2 constructs were synthetically produced from the full-length S glycoprotein gene sequence (GenBank MN908947 nucleotides 21563–25384). The full-length S-genes were codon optimized for expression in Spodoptera frugiperda (Sf9) cells and synthetically produced by GenScript (Piscataway, NJ, USA). The QuikChange Lightning site-directed mutagenesis kit (Agilent) was used to produce two spike protein variants: modifications by mutating the S1/S2 cleavage site by mutation of the furin cleavage site (682-RRAR-685) to 682-QQAQ-685 to be protease resistant and designated as BV2365. The single mutant BV2365 was further stabilized by introducing two proline substitutions at positions K986P and V987P (2P) to produce the double mutant, NVX-CoV2373. Full-length S-genes were cloned between the BamHI – HindIII sites in the pFastBac baculovirus transfer vector (Invtrogen, Carlsbad, CA) under transcriptional control of the Autograha californica polyhedron promoter. Recombinant baculovirus constructs were plaque purified and master seed stocks prepared and used to produce the working virus stocks. The baculovirus master and working stock titers were determined using rapid titer kit (Clontech, Mountain View, CA). Recombinant baculovirus stocks were prepared by infectingSf9 cells with a multiplicity of infection (MOI) of ≤ 0.01 plaque forming units (pfu) per cell25–27.
Expression and purification. SARS-CoV-2 S proteins were produced in Sf9 cells expanded in serum-free medium to a density of 2–3 × 106 cell mL− 1 and infected with recombinant baculovirus at MOI of ≤ 0.1 pfu per cell. Cells were cultured at 27 ± 2 °C and harvested at 68–72 hours post-infection by centrifugation (4000 x g for 15 min). Cell pellets were suspended in 25 mM Tris HCl (pH 8.0), 50 mM NaCl and 0.5-1.0% (v/v) TERGITOL NP-9 with leupeptin. S-proteins were extracted from the plasma membranes with Tris buffer containing NP-9 detergent, clarified by centrifugation at 10,000 x g for 30 min. S-proteins were purified by TMAE anion exchange and lentil lectin affinity chromatography. Hollow fiber tangential flow filtration was used to formulate the purified spike protein at 100–150 µg mL− 1 in 25 mM sodium phosphate (pH 7.2), 300 mM NaCl, 0.02% (v/v) polysorbate 80 (PS 80)26. Purified S-proteins were evaluated by 4–12% gradient SDS-PAGE stained with Gel-Code Blue reagent (Pierce, Rockford, IL) and purity was determined by scanning densitometry using the OneDscan system (BD Biosciences, Rockville, MD).
Dynamic light scattering (DLS). Samples were equilibrated at 25 °C and scattering intensity was monitored as a function of time in a Zetasizer NanoZS (Malvern, UK). Cumulants analysis of the scattered intensity autocorrelation function was performed with instrument software to provide the z-average particle diameter and polydispersity index (PDI).
Differential scanning calorimetry (DCS). Samples and corresponding buffers were heated from 4 °C to 120 °C at 1 °C per minute and the differential heat capacity change was measured in a NanoDSC (TA Instruments, New Castle, DE). A separate buffer scan was performed to obtain a baseline, which was subtracted from the sample scan to produce a baseline-corrected profile. The temperature where the peak apex is located is the transition temperature (Tmax) and the area under the peak provides the enthalpy of transition (ΔHcal).
Transmission electron microscopy (TEM) and 2D class averaging. Electron microscopy was perform by NanoImaging Services (San Diego, CA) with a FEI Tecani T12 electron microscope, operated at 120 keV equipped with a FEI Eagle 4 k x 4 k CCD camera. SARS-CoV-2 S proteins were diluted to 2.5 µg mL− 1 in formulation buffer. The samples (3 µL) were applied to nitrocellulose-supported 400-mesh copper grids and stained with uranyl format. Images of each grid were acquired at multiple scales to assess the overall distribution of the sample. High-magnification images were acquired at nominal magnifications of 110,000 × (0.10 nm/pixel) and 67,000 × (0.16 nm/pixel). The images were acquired at a nominal under focus of -1.4 µm to -0.8 µm (110,000x) and electron doses of ~ 25 e/Å2.
For class averaging, particles were identified at high magnification prior to alignment and classification. The individual particles were selected, boxed out, and individual sub-images were combined into a stack to be processed using reference-free classification. Individual particles in the 67,000x high magnification images were selected using an automated picking protocol17. An initial round of alignments was performed for each sample, and from the alignment class averages that appeared to contain recognizable particles were selected for additional rounds of alignment. These averages were used to estimate the percentage of particles that resembled single trimers and oligomers. A reference-free alignment strategy based on XMIPP processing package was used for particle alignment and classification18.
Kinetics of SARS-CoV-2 S binding to hACE2 receptor by BLI. S-protein receptor binding kinetics was determined by bio-layer interferometry (BLI) using an Octet QK384 system (Pall Forté Bio, Fremont, CA). Hist-tagged human ACE2 (2 µg mL− 1) was immobilized on nickel-charged Ni-NTA biosensor tips. After baseline, SARS-CoV-2 S protein containing samples were 2-fold serially diluted over a range 4.7 nM to 300 nM range were allowed to associate for 600 sec followed by dissociation for an additional 900 sec. Data was analyzed with Octet software HT 101:1 global curve fit.
Specificity of SARS-CoV-2 S binding to hACE2 receptor by ELISA. Ninety-six well plates were coated with 100 µL SARS-CoV-2 spike protein (2 µg mL− 1) overnight at 4 °C. Plates were washed with phosphate buffered saline with 0.05% Tween (PBS-T) buffer and blocked with TBS Startblock blocking buffer (ThermoFisher, Scientific). Histidine-tagged hACE2 and hDPP4 receptors were 3-fold serially diluted (5-0.0001 µg mL− 1) and added to coated wells for 2 hours at room temperature. The plates were washed with PBS-T. Optimally diluted horseradish peroxidase (HRP) conjugated anti-histidine was added and color developed by addition of and 3,3’,5,5’-tetramethylbenzidine peroxidase substrate (TMB, T0440-IL, Sigma, St. Louis, MO, USA). Plates were read at an OD of 450 nm with a SpectraMax Plus plate reader (Molecular Devices, Sunnyvale, CA, USA) and data analyzed with SoftMax software. EC50 values were calculated by 4-parameter fitting using GraphPad Prism 7.05 software
Animal ethics statement. The mouse immunogenicity studies were performed by Noble Life Sciences (Sykeville, MD). Noble Life Sciences is accredited by the Association for Assessment and Accreditation of Laboratory Animal Care (AAALACC International). All animal procedures were in accordance with NRC Guide for the Care and Use of Laboratory Animals, the Animal Welfare Act, and the CDC/NIH Biosafety in Microbiological and Biomedical Laboratories. The olive baboon (Papio cynocephalus anubis) study was performed at the University of Oklahoma Health Science Center (OUHSC). OUHSC is accredited by AAALACC International. Baboons were maintained and treated according to the Institutional Biosafety Committee guidelines. Baboon experiments were approved by the Institutional Animal Care and Use Committee (IACUC) and the Institutional Biosafety Committee of OUHSC. Studies were conducted in accordance with the National Institutes of Health Guide for Care and Use of Laboratory Animals (NIH publication 8023, Revised 1978).
Mouse study designs. Female BALB/c mice (7–9 weeks old, 17–22 grams, N = 10 per group) were immunized by intramuscular (IM) injection with a single dose (study day 14) or two doses spaced 14-days apart (study day 0 and 14) containing a dose range (0.01, 0.1, 1.0, or 10 µg) of NVX-CoV2373 with 5 µg saponin-based Matrix-M™ adjuvant (Novavax, AB, Uppsala, SE). A separate group (n = 10) received two doses of 10 µg NVX-CoV2373 without adjuvant. A placebo group served as a non-immunized control. Serum was collected for analysis on study days − 1, 13, 21, and 28. Vaccinated and control animals were intranasally challenged with SARS-CoV-2 42-days following one or two immunizations (study day 56).
To assess the T cell response mediated by Matrix-M adjuvant, groups of female BALB/c mice (N = 6 per group) were immunized IM with 10 µg NVX-CoV2373 with and without 5 µg Matrix-M adjuvant in 2 doses spaced 21-days apart. Spleens were collected 7-days after the second immunization (study day 28). A non-vaccinated group (N = 3) served as a control.
Baboon study design. Ten adult baboons (10–16 years of age) were randomized into 4 groups of 2–3/group and immunized by IM injection with 1, 5 or 25 µg NVX-CoV2373 with 50 µg Matrix-M adjuvant. A separate group was immunized with 25 µg NVX-CoV2373 without adjuvant. Animals were vaccinated with 2-doses spaced 21-days apart. Serum was collected on study days 0, 21, 28 and 35. For T cell analysis, peripheral blood mononuclear cells (PBMCs) were collected 7-days after the second immunization (study day 28). Subsequent to the start of the study one animal tested positive for STLV and was therefore dropped from the study.
SARS-CoV-2 challenge in mice. Mice were transduced intranasally with 2.5 × 108 pfu Ad/CMVhACE2 (VVC-McCray-7580, University of Iowa Vector Core) 38-days after the second vaccination. At four days post infection, mice were anaesthetized by intraperitoneal injection 50 µL of a mix of xylazine (0.38 mg/mouse) and ketamine (1.3 mg/mouse) diluted in phosphate buffered saline (PBS). Mice were intranasally inoculated with 1.5 × 105 pfu of SARS-CoV-2 in 50 µL divided between nares. Challenged mice were weighed on day of infection and daily for up to 7 days post infection. At days 4- and 7-days post infection, 5 mice were sacrificed from each vaccination and control group, and lungs were harvested to determine for titer by a plaque assay and prepared for histological scoring.
SARS-CoV-2 plaque assay. SARS-CoV-2 lung titers were quantified by homogenizing harvested lungs in PBS (Quality Biological Inc.) using 1.0 mm glass beads (Sigma Aldrich) and a Beadruptor (Omini International Inc.). Homogenates were added to Vero E6 near confluent cultures and SARS-CoV-2 virus titers determined by counting plaque forming units (pfu) using a 6-point dilution curve.
Anti-SARS-CoV-2 spike IgG by ELISA. An ELISA was used to determine anti-SARS-CoV-2 S IgG titers. Briefly, 96 well microtiter plates (ThermoFischer Scientific, Rochester, NY, USA) were coated with 1.0 µg mL− 1 of SARS-CoV-2 spike protein. Plates were washed with PBS-T and blocked with TBS Startblock blocking buffer (ThermoFisher, Scientific). Mouse, baboon or human serum samples were serially diluted (10− 2 to 10− 8) and added to the blocked plates before incubation at room temperature for 2 hours. Following incubation, plates were washed with PBS-T and HRP-conjugated goat anti-mouse IgG or goat anti-human IgG (Southern Biotech, Birmingham, AL, USA) added for 1 hour. Plates were washed with PBS-T and 3,3’,5,5’-tetramethylbenzidine peroxidase substrate (TMB, T0440-IL, Sigma, St Louis, MO, USA) was added. Reactions were stopped with TMB stop solution (ScyTek Laboratories, Inc. Logan, UT). Plates were read at OD 450 nm with a SpectraMax Plus plate reader (Molecular Devices, Sunnyvale, CA, USA) and data analyzed with SoftMax software. EC50 values were calculated by 4-parameter fitting using SoftMax Pro 6.5.1 GxP software. Individual animal anti-SARS-CoV-2 S IgG titers and group geometric mean titers (GMT) and 95% confidence interval (± 95% CI) were plotted GraphPad Prism 7.05 software.
ACE2 receptor blocking antibodies. ACE2 receptor blocking antibodies were determined by ELISA. Ninety-six well plates were coated with 1.0 µg mL− 1 SARS-CoV-2 S protein overnight at 4 °C. Serially diluted serum from groups of immunized mice, baboons or humans were and added to coated wells and incubated for 2 hours at room temperature. After washing, 30 ng mL− 1 of histidine-tagged hACE or hDPP4 was added to wells for 1 hour at room temperature. HRP-conjugated anti-histidine IgG was added followed by washing prior to addition of TMB substrate. Plates were read at OD 450 nm with a SpectraMax plus plate reader (Molecular Devices, Sunnyvale, CA, USA) and data analyzed with SoftMax software. Serum dilution resulting in a 50% inhibition of receptor binding was used to calculate titer determined using 4-parameter fitting with GraphPad Prism 7.05 software.
SARS-CoV-2 neutralization assay. SARS-CoV-2 was handled in a select agent ABSL3 facility at the University of Maryland, School of Medicine. Mouse or baboon sera were diluted 1:20 in Vero E6 cell growth media and further diluted 1:2 to 1:40960. SARS-CoV-2 (MOI of 0.01 pfu per cell) was added and the mixture incubated for 60 min at 37 °C. Vero E6 media was used as negative control. The inhibitory capacity of each serum dilution was assessed for cytopathic effect (CPE). The endpoint titer was reported as the dilution that blocked 100% of CPE at 3 days post infection.
ELISPOT assay. Murine IFN-γ and IL-5 ELISPOT assays were performed following manufacturer’s procedures for mouse IFN-γ and IL-5 ELISPOT kit (Mabtech, Cincinnati, OH). Briefly, 3 × 105 splenocytes in a volume of 200 µL were stimulated with NVX-CoV2373 protein or peptide pools (PP) of 15-mer peptides with 11 overlapping amino acids covering the entire spike protein sequence (JPT, Berlin, Germany) in plates that were pre-coated with anti-IFN-γ or anti-IL-5 antibodies. Each stimulation condition was carried out in triplicate. Assay plates were incubated overnight at 37ºC in a 5% CO2 incubator and developed using BD ELISPOT AEC substrate set (BD Biosciences, San Diego, CA). Spots were counted and analyzed using an ELISPOT reader and Immunospot software (Cellular Technology, Ltd., Shaker Heights, OH). The number of IFN-γ or IL-5 secreting cells was obtained by subtracting the background number in the medium controls. Data shown in the graph are the average of triplicate wells.
Similarly, Baboon IFN-γ and IL-4 assays were carried out using NHP IFN-γ and Human IL-4 assay kit from Mabtech using PBMC collected at day 7 following the second immunization (day 28).
Surface and intracellular cytokine staining. For surface staining, murine splenocytes were first incubated with an anti-CD16/32 antibody to block the Fc receptor. To characterize T follicular helper cells (Tfh), splenocytes were incubated with the following antibodies or dye: BV650-conjugated anti-CD3, APC-H7-conjugated anti-CD4, FITC-conjugated anti-CD8, Percp-cy5.5-conjugated anti-CXCR5, APC-conjugated anti-PD-1, Alexa Fluor 700-conjugated anti-CD19, PE-conjugated anti-CD49b (BD Biosciences, San Jose, CA) and the yellow LIVE/DEAD® dye (Life Technologies, NY). To stain germinal center (GC) B cells, splenocytes were labeled with FITC-conjugated anti-CD3, PerCP-Cy5.5-conjugated anti-B220, APC-conjugated anti-CD19, PE-cy7-conjugated anti-CD95, and BV421-conjugated anti-GL7 (BD Biosciences) and the yellow LIVE/DEAD® dye (Life Technologies, NY).
For intracellular cytokine staining (ICCS) of murine splenocytes, cells were cultured in a 96-well U-bottom plate at 2 × 106 cells per well. The cells were stimulated with NVX-CoV2373 or pools of a 15-mer peptide pool (PP) as described above (JPT, Berlin, Germany). The plate was incubated 6 h at 37 °C in the presence of BD GolgiPlug™ and BD GolgiStop™ (BD Biosciences). Cells were labeled with murine antibodies against CD3 (BV650), CD4 (APC-H7), CD8 (FITC), CD44 (Alexa Fluor 700), and CD62L (PE) (BD Pharmingen, CA) and the yellow LIVE/DEAD® dye. After fixation with Cytofix/Cytoperm (BD Biosciences), cells were incubated with PerCP-Cy5.5-conjugated anti-IFN-γ, BV421-conjugated anti-IL-2, PE-cy7-conjugated anti-TNF-α, and APC-conjugated anti-IL-4 (BD Biosciences). All stained samples were acquired using a LSR-Fortessa flow cytometer (Becton Dickinson, San Jose, CA) and the data were analyzed with FlowJo software version Xv10 (Tree Star Inc., Ashland, OR).
For ICCS, baboon PBMCs were collected 7 days after the second immunization (day 28) and stimulated as described above with NVX-CoV2373. Cells were labelled with human/NHP antibodies BV650-conjugated anti-CD3, APC-H7-conjugated anti-CD4, APC-conjugated anti-CD8, BV421-conjugated anti-IL-2, PerCP-Cy5.5-conjugated anti-IFN-γ, PE-cy7-conjugated anti-TNF-α (BD Biosciences), and the yellow LIVE/DEAD® dye.
Histopathology. Mice were euthanized at 4- and 7-days following SARS-CoV-2 challenge. The lungs were fixed with 10% formalin, and sections were stained with H&E for histological examination. Slides were examined in a blinded fashion for total inflammation, periarteriolar, and peribronchiolar inflammation and epithelial cell denuding.
COVID-19 convalescent serum. Convalescent serum samples were provided by Dr. Pedro A Piedra (Baylor College of Medicine, Houston, TX, USA). Samples were collected from COVID-19 patients 18–79 years of age 4–6 weeks after testing positive for SARS CoV-2. Symptoms ranged from asymptomaic, mild to moderate symptoms, to severe symptoms requiring hospitalization. Sera were analyzed for anti-SARS-CoV-2 S IgG and hACE2 receptor inhibiting antibody levels.
Statistical analysis. Statistical analyses were performed with GraphPad Prism 7.05 software (La Jolla, CA). Serum antibody titers were plotted for individual animals and the geometric mean titer (GMT) and 95% confidence interval (95% CI) or the means ± SEM as indicated in the figure. T-test was used to determine differences between paired groups. Weight change between immunized and placebo groups was determined for each day using a t-test. P-values ≤ 0.05 were considered as statistically significant.