Immunogenicity and efficacy of the COVID-19 candidate vector vaccine MVA-SARS-2-S in preclinical vaccination

Significance The highly attenuated vaccinia virus MVA is licensed as smallpox vaccine; as a vector it is a component of the approved adenovirus-MVA–based prime-boost vaccine against Ebola virus disease. Here, we provide results from testing the COVID-19 candidate vaccine MVA-SARS-2-S, a poxvirus-based vector vaccine that proceeded to clinical evaluation. When administered by intramuscular inoculation, MVA-SARS-2-S expresses and safely delivers the full-length SARS-CoV-2 S protein, inducing balanced SARS-CoV-2–specific cellular and humoral immunity, and protective efficacy in vaccinated mice. Substantial clinical experience has been gained with MVA vectors using homologous and heterologous prime-boost applications, including the immunization of children and immunocompromised individuals. Thus, MVA-SARS-2-S represents an important resource for developing further optimized COVID-19 vaccines.

In vitro characterization of recombinant MVA-SARS-2-S. Genetic identity and genetic stability of vector viruses was confirmed by polymerase chain reaction (PCR) using viral DNA and detection of S-protein synthesis following serial passage at low MOI. For the latter, 95% confluent DF-1 cells were infected at MOI 0.05, incubated for 48h, harvested and used for reinfection. In total, five rounds of low MOI passage were performed. After the fifth passage, sixty virus isolates were obtained and amplified in 24-well DF-1 cultures for further testing. PCR analysis was performed to confirm genetic stability of viral genomes and MVA-and SARS-2-S-specific immunostaining served to monitor recombinant gene expression. The replicative capacity of recombinant MVA was tested in duplicate in multi-step-growth experiments on monolayers of DF-1, HaCat, HeLa or A549 cells grown in 6-well-tissue-culture plates. Viruses were inoculated at MOI 0.05, harvested at 0, 4,8,24,48, and 72 h after infection, and titrated on CEF monolayers to determine infectivities in cell lysates in PFU.

Western Blot analysis of recombinant protein.
To monitor production of the recombinant SARS-2-S protein, DF-1 cells were infected at MOI 10 with recombinant or non-recombinant MVA or remained uninfected (mock). At indicated time points of infection, cell lysates were prepared from infected cells and stored at −80 °C. Proteins from lysates were separated by electrophoresis in a sodium dodecyl sulfate (SDS)-10% polyacrylamide gel (SDS-PAGE; Bio-Rad, Munich) and subsequently transferred to a nitrocellulose membrane by electroblotting. The blots were blocked in a phosphate buffered saline (PBS) buffer containing 5% Bovine Serum Albumin (BSA) (Sigma-Aldrich, Taufkirchen, Germany) and 0.1% Tween-20 (Sigma-Aldrich, Taufkirchen, Germany) and incubated for 60 min with primary antibody, monoclonal anti-HAtag antibody (1:8000; HA Tag mAb 2-2.2.14, Thermo Fisher Scientific, Planegg, Germany) or COVID-19 patient serum (1:200). Next, membranes were washed with 0.1% Tween-20 in PBS and incubated with anti-mouse or anti-human IgG (1:5000; Agilent Dako, Glostrup, Denmark), conjugated to horseradish peroxidase. Blots were washed and developed using SuperSignal® West Dura Extended Duration substrate (Thermo Fisher Scientific, Planegg, Germany). Chemiluminescence was visualized using the ChemiDoc MP Imaging System (Bio-Rad, Munich, Germany). For use of patient serum ethical approval was granted by the Ethics Committee at the Medical Faculty of LMU Munich (vote 20-225 KB) in accordance with the guidelines of the Declaration of Helsinki.
Vaccination experiments in mice. Female BALB/c mice (6 to 10 week-old) were purchased from Charles River Laboratories (Sulzfeld, Germany). Mice were maintained under specified pathogen-free conditions, had free access to food and water, and were allowed to adapt to the facilities for at least one week before vaccination experiments were performed. All animal experiments were handled in compliance with the European and national regulations for animal experimentation (European Directive 2010/63/EU; Animal Welfare Acts in Germany). Immunizations were performed using intramuscular applications with vaccine suspension containing either 10 7 or 10 8 PFU recombinant MVA-SARS2-S, non-recombinant MVA or PBS (mock) into the quadriceps muscle of the left hind leg. Blood was collected on days 0, 18, or 35. Coagulated blood was centrifuged at 1300×g for 5 min in MiniCollect vials (Greiner Bio-One, Alphen aan den Rijn, The Netherlands) to separate serum, which was stored at −20 °C until further analysis.
Transduction of vaccinated mice with Ad_ACE2-mCherry and challenge infection with SARS-CoV-2. All animal experiments were performed in accordance with Animal Welfare Acts in Germany and were approved by the regional authorities. Vaccinated mice were housed under pathogen-free conditions and underwent intratracheal inoculation with 5x10 8 PFU Adenovirus-ACE2-mCherry (cloned at ViraQuest Inc., North Liberty, IA, USA) under ketamine/xylazine anesthesia. Three days post transduction, mice were infected via the intranasal route with 1.5x104 tissue culture infectious dose 50 (TCID50) SARS-CoV-2 (BavPat1/2020 isolate, European Virus Archive Global # 026V-03883). Mice were sacrificed four days post infection and serum as well as lung tissue samples were taken for analysis of virus loads.
Quantitative real-time reverse transcription PCR to determine SARS-CoV-2 or mCherry RNA. Tissue samples of immunized and challenged mice were excised from the left lung lobes and homogenized in 1 ml DMEM. SARS-CoV-2 titres in supernatants (in TCID50 per ml) were determined on VeroE6 cells. RNA isolation was performed with the RNeasy minikit (Qiagen) according to the manufacturer's instructions. The RNA amount was measured using the NanoDrop ND-100 spectrophotometer. Total RNA was reverse transcribed and quantified by realtime PCR using the OneStep RT-PCR kit (Qiagen) as described previously (6) with the primer pair upE-Fwd and upE-Rev and the probe upE-Prb on a StepOne high-throughput fast real-time PCR system (ThermoFisher). Additionally, for every tissue sample from transduced and infected mice, evidence for successful ACE2 transduction was determined by real-time RT-PCR for mCherry mRNA with the OneStep RT-PCR kit (Qiagen). All samples for mCherry analysis were evaluated in one RT-PCR run. Quantification was carried out using a standard curve based on 10-fold serial dilutions of appropriate control RNA ranging from 10 2 to 10 5 copies.
Histopathological examination of lung tissue. Lungs were collected on day 4 post challenge with SARS-CoV-2 and processed for histological analysis. Briefly, tissue was fixed in formalin and embedded in paraffin. Four µm sections were cut with a microtome (RM2255, Leica Biosystems) and stained with hematoxylin and eosin (HE). To investigate the presence of viral RNA in lung tissue by in situ hybridization, the RNAscope® 2.5 HD Assay -RED Kit from Bio-Techne (Cat. No. 322360) was used according to the manufacturer's instructions. Briefly, mounted slides were incubated at 60 °C, deparaffinized with xylene and 100% ethanol and pretreated with RNAscope® Pretreatment Reagents (Cat. No. 322300 and 322000) to enable access to the target RNA. Subsequently, a RNA-specific probe, targeted against the S gene of the SARS-CoV-2 (Cat. No. 848561), was hybridized to the RNA. Afterwards, signal amplification was performed and alkaline-phosphatase-labeled probes were used in combination with Fast Red substrate allowing signal detection. The slides were counterstained with Gill's Hematoxylin I. A RNAscope® Negative Control Probe (Cat. No. 310043) was used in parallel to monitor background staining.
Antigen-specific IgG ELISA. SARS-2-S-specific serum IgG titres were measured by enzymelinked immunosorbent assay (ELISA) as described previously (7). Flat bottom 96-well ELISA plates (Nunc MaxiSorp Plates, Thermo Fisher Scientific, Planegg, Germany) were coated with 50 ng/well recombinant 2019-nCoV (COVID-19) S protein (Full Length-R683A-R685A-HisTag, ACROBiosystems, Newark, USA) overnight at 4 °C. Plates were washed and then blocked for 1 h at 37 °C with blocking buffer containing 1% BSA (Sigma-Aldrich, Taufkirchen, Germany) and 0.15M sucrose (Sigma-Aldrich, Taufkirchen, Germany) dissolved in PBS. Mouse sera were serially diluted three-fold down the plate in PBS containing 1% BSA (PBS/BSA), starting at a dilution of 1:100. Plates were then incubated for 1 h at 37 °C. After incubating and washing, plates were probed with 100 µl/well of goat anti-mouse IgG HRP (1:2000; Agilent Dako, Denmark) diluted in PBS/BSA for 1 h at 37 °C. After washing, 100 µl/well of 3´3´, 5´5´-Tetramethylbenzidine (TMB) Liquid Substrate System for ELISA (Sigma-Aldrich, Taufkirchen, Germany) was added until a colour change was observed. The reaction was stopped by adding 100µl/well of Stop Reagent for TMB Substrate (450 nm, Sigma-Aldrich, Taufkirchen, Germany). Absorbance was measured at 450 nm. The absorbance of each serum sample was measured at 450 nm with a 620 nm reference wavelength. ELISA data were normalized using the positive control. The cut-off value for positive mouse serum samples was determined by calculating the mean of the normalized OD 450nm values of the PBS control group sera plus 6 standard deviations (mean + 6 SD).
RBD-specific IgG ELISA. RBD-specific serum IgG titres were measured by enzyme-linked immunosorbent assay (ELISA). RBD ELISA was performed by coating 96-well microtitre plates with SARS-CoV-2 RBD protein in PBS overnight, as previously described (26). After blocking, serum dilutions (diluted 1:100) were incubated at 37°C for 1 h. Antigen-specific antibodies were detected by using peroxidase-labeled rabbit anti-human IgG (Dako) and 3,3′,5,5′tetramethylbenzidine as a substrate. Absorbance was measured at 450 nm. A cutoff was set at an optical density of 0.5.

Surrogate virus neutralization assay (sVNT).
To test for the presence of neutralizing anti-SARS-CoV-2-S serum antibodies we used surrogate virus neutralization test as described before with slight modifications (8). Briefly, 6 ng of SARS-CoV-2 S RBD (Trenzyme) was pre-incubated for 1 hour at 37 °C with heat-inactivated test sera at final dilutions between 1:20 to 1:540, as indicated on the graphs. Afterwards, SARS-CoV-2 S RBD-serum mixtures were loaded onto MaxiSorp 96F plates (Nunc) coated with 200 ng/well ACE2 [produced in-house as described in Bosnjak et al. (8)] and blocked with 2% bovine serum albumin/2% mouse serum (Invitrogen) and incubated for additional 1 h at 37 °C. As controls we used SARS-CoV-2-S-RBD pre-incubated only with buffer and non-specific mouse serum (Invitrogen). Plates were extensively washed with phosphate-buffered saline/0.05% Tween-20 (PBST), followed by incubation for 1 h at 37 °C with an HRP-conjugated anti-His-tag antibody (1.2 µg/ml; clone HIS 3D5). After appropriate washing, colorimetric signals were developed by addition of the chromogenic substrate 3,3',5,5'tetramethylbenzidine (TMB; TMB Substrate Reagent Set, BD Biosciences) and stopped by addition of equal volume of 0.2 M H2SO4. The optical density values measured at 450 nm and 570 nm (SpectraMax iD3 microplate reader, Molecular Devices) were used to calculate percentage of inhibition after subtraction of background values as inhibition (%) = (1 -Sample OD value/Average SARS-CoV-2 S RBD OD value) x100. To remove background effects, the mean percentage of inhibition from non-specific mouse serum (Invitrogen) was deducted from sample values and neutralizing anti-SARS-CoV-2-S antibodies titres were determined as serum dilution that still had binding reduction > mean + 2 SD of values from sera of vehicle-treated mice.
Plaque reduction neutralization test 50 (PRNT50). We tested serum samples for their neutralization capacity against SARS-CoV-2 (German isolate; GISAID ID EPI_ISL 406862; European Virus Archive Global #026V-03883) by using a previously described protocol (9). We 2fold serially diluted heat-inactivated samples in Dulbecco modified Eagle medium supplemented with NaHCO3, HEPES buffer, penicillin, streptomycin, and 1% foetal bovine serum, starting at a dilution of 1:10 in 50 μL. We then added 50 μL of virus suspension (400 plaque-forming units) to each well and incubated at 37°C for 1 h before placing the mixtures on VeroE6 cells (ATCC CRL1586). After incubation for 1 h, we washed, cells supplemented with medium, and incubated for 8 h. After incubation, we fixed the cells with 4% formaldehyde/phosphate-buffered saline (PBS) and stained the cells with polyclonal rabbit anti-SARS-CoV antibody (Sino Biological, https://www.sinobiological.com) and a secondary peroxidase-labeled goat anti-rabbit IgG (Dako, https://www.agilent.com). We developed the signal using a precipitate forming 3,3′,5,5′tetramethylbenzidine substrate (True Blue; Kirkegaard and Perry Laboratories, https://www.seracare.com) and counted the number of infected cells per well by using an ImmunoSpot Image Analyzer (CTL Europe GmbH, https://www.immunospot.eu). The serum neutralization titre is the reciprocal of the highest dilution resulting in an infection reduction of >50% (PRNT50). We considered a titre >20 to be positive.

SARS-CoV-2 virus neutralization test (VNT100).
The neutralizing activity of mouse serum antibodies was investigated based on a previously published protocol (10). Briefly, samples were serially diluted in 96-well plates starting from a 1:16 serum dilution. Samples were incubated for 1 h at 37°C together with 100 50% tissue culture infectious doses (TCID50) of SARS-CoV-2 (BavPat1/2020 isolate, European Virus Archive Global # 026V-03883). Cytopathic effects (CPE) on VeroE6 cells (ATCC CRL1586) were analyzed 4 days after infection. Neutralization was defined as the absence of CPE compared to virus controls. For each test, a positive control (neutralizing COVID-19 patient plasma) was used in duplicates as an inter-assay neutralization standard. Ethical approval was granted by the Ethics Committee at the Medical Faculty of LMU Munich (vote 20-225 KB) in accordance with the guidelines of the Declaration of Helsinki.

Prediction and generation of synthetic SARS-2-S peptides.
The sequence of the SARS-CoV-2 S protein (NCBI ID: QHD43416.1, Uniprot ID: P0DTC2 (SPIKE_SARS2)) served for epitope prediction, and probable CD8+ and CD4+ T cell determinants were examined with the Immune Epitope Database and Analysis Resource (IEDB, https://www.iedb.org/). For identification of potential CD8+ T cell determinants, the MHC-I Binding Prediction and MHC-I Processing Prediction tools (11,12) were used and projections for 9-11mer peptides spanning the entire SARS-2-S protein sequence were obtained. The inputs selected for the search included the Prediction Method 'IEDB recommended 2.22', the MHC source species 'Mouse' and the MHC class I alleles H2-K d , H2-D d and H2-L d . The output was restricted to a percentile rank cut-off of 10.0. After lists of peptides were generated, all peptides with an IC50 score of 500nM or less were selected for inclusion in the top 5% list. All the peptides in this list were further analyzed using the MHC-I Processing Prediction tool 'Proteasomal cleavage/TAP transport/MHC class I combined predictor'. All peptides with an IC50 score of 500nM or less and a high total score were chosen and subsequently included in the top peptides list. To confirm that these peptides were potential binders of MHC class I alleles H2-K d , H2-D d and H2-L d , they were further screened for MHC I binding using the RankPep server (13). Peptides that were found to bind to any of the above alleles were selected for synthesis and testing. For the identification of potential CD4+ T cell determinants, the MHC-II Binding Prediction tool (13) served to obtain 15mer peptides spanning the entire SARS-2-S protein sequence. The inputs for the analysis included the Prediction Method 'IEDB recommended 2.22', the MHC source species 'Mouse' and the MHC class II alleles H2-IA d and H2-IE d . Peptides with percentile rank of 10.0 or less and an IC50 score of 1000 nM or less were further tested for MHC class II binding using the RankPep server. Peptides bound to any of the above MHC class II alleles were selected for synthesis and testing. All peptides were obtained from Thermo Fisher Scientific (Planegg, Germany) as crude material (<50% purity) at a 1-4 mg scale, dissolved in PBS or DMSO to 2 mg/ml, aliquoted and stored at -20 ⸰ C.

Spectral flow cytometry.
Spleens from immunized or mock-immunized groups of BALB/c mice (n=6) were obtained at day 14 after the last inoculation. Single cell suspensions of splenocytes in RPMI-10 were prepared by meshing the organs through 40 µm cell strainers. All samples were subjected to erythrocyte lysis. Non-specific antibody binding was blocked by incubating samples in 10% rat serum at 4 °C for 15 min. Without washing, cells were incubated with a mix of antibodies for additional 15 min at on 37 °C. The full list of antibodies and staining reagents is shown in Table S1. After washing, cells were acquired on Cytek Aurora spectral flow cytometer (Cytek) equipped with five lasers operating at 355 nm, 405 nm, 488 nm, 561 nm and 640 nm.
All flow cytometry data were analysed using FCS Express V7 (Denovo) and Graphpad Prism 7 (GraphPad). All samples were individually pre-gated (Fig. S8A) and subsequently concatenated into one file. The concatenated file was then pregated to exclude NK and NKT cells (Fig. S8B). Data processing included scaling, normalisation, weighted density downsampling to 200.000 cells and dimensionality reduction (UMAP). Heatmaps were used to identify cell clusters (Fig. S8C). Statistical analysis. Data were prepared using GraphPad Prism version 5 (GraphPad Software Inc., San Diego CA, USA) and expressed as mean ± standard error of the mean (SEM). Data were analyzed by unpaired, two-tailed t-tests to compare two groups and one-way ANOVA to compare three or more groups. P < 0.05 was used as the threshold for statistical significance. Fig. S1. Molecular analysis of the MVA-SARS-2-S genome.   Local inflammation of the myofiber interstitium and the adjacent adipose tissue was observed and interpreted as part of the physiological immune reaction to the vaccine virus as a consequence of the treatment procedure. The degree and extent of inflammation, myodegeneration and necrosis was in accordance with the ratio of inoculum volume in relation to the administration site. The lymphoid hyperplasia observed in draining lymph nodes is interpreted as a sign of immune competence of the animals and is characteristic for any early response to inflammation at a draining site. In conclusion, we observed no evidence for a potential toxicity of the full human dose of MVA-SARS-2-S in BALB/c mice. The repeated vaccination was well tolerated and caused no adverse events and no relevant macroscopic or histopathological changes. The observed reactions were comparable to previous experiments using non-recombinant MVA or other recombinant MVA vaccine constructs and are considered to be part of the pharmacodynamic principle of MVA-based vaccination (16).      Graphs show the frequency and absolute number of IFN-γ+ TNF-α+ CD8+ T cells. Differences between groups were analyzed by one-way ANOVA and Tukey post-hoc test. Asterisks represent statistically significant differences between two groups. * p < 0.05, ** p < 0.01.    Table S1). After washing, cells were acquired on a Cytek Aurora spectral flow cytometer. (A) All samples were individually pregated and subsequently concatenated into one file. (B) The concatenated file was then gated to exclude NK and NKT cells. (C) Heatmaps were used to identify cell clusters. (D) Data from selected T cell subsets show high levels of CD8+ effector memory T cells (CD8eff/m) and reduced levels of naïve CD4+ T cells (CD4naive) in MVA-S and MVA immunized animals, whereas, comparable populations of T helper cell populations were detected. (E) IFN-γ production measured by ICS and conventional FACS analysis. Stimulation with SARS-CoV-2-specific peptide S268-276 (S1; GYLQPRTFL) shows the specific detection of IFN-γ+ CD8+ T cells in splenocytes from MVA-SARS-2-S immunized mice. Differences between groups were analyzed by one-way ANOVA and Tukey post-hoc test. Asterisks represent statistically significant differences between two groups. *** p < 0.001, **** p < 0.0001.  (Table S2). IFN-γ spot-forming cells were counted by ELISPOT. About two weeks after the last immunization the mice were sensitized with an adenovirus expressing hACE2 and mCherry and infected with SARS-CoV-2 three days after transduction. (A) Body weight change was monitored daily, (B) spontaneous behavior, and general condition which were summarized in a clinical score, (C) four days after SARS-CoV-2 infection mice were sacrificed, lungs were isolated and mCherry mRNA copies were evaluated.