Live-attenuated vaccine sCPD9 elicits superior mucosal and systemic immunity to SARS-CoV-2 variants in hamsters

Ethics statement

In vitro and animal work were conducted under appropriate biosafety conditions in a BSL-3 facility at the Institut für Virologie, Freie Universität Berlin, Germany. All animal experiments were performed in compliance with relevant institutional, national and international guidelines for the care and humane use of animals and approved by the competent state authority, Landesamt für Gesundheit und Soziales, Berlin, Germany (permit number 0086/20).

Cells

Vero E6 (obtained from ATCC, CRL-1586), Vero E6-TMPRSS2 (obtained from the National Institute for Biological Standards and Control (NIBSC), 100978) and Calu-3 (obtained from ATCC, HTB-55) cells were cultured in minimal essential medium (MEM) containing 10% fetal bovine serum, 100 IU ml⁻¹ penicillin G and 100 µg ml⁻¹ streptomycin at 37 °C and 5% CO₂. In addition, the cell culture medium for Vero E6-TMPRSS2 cells contained 1,000 µg ml⁻¹ geneticin (G418) to ensure selection for cells expressing the genes for neomycin resistance and TMPRSS2.

Viruses

The modified live-attenuated SARS-CoV-2 mutant sCPD9 and SARS-CoV-2 variants B.1 (BetaCoV/Munich/ChVir984/2020; B.1, EPI_ISL_406862), Beta (B.1.351; hCoV-19/Netherlands/NoordHolland_20159/2021) and Delta (B.1.617.2; SARS-CoV-2, Human, 2021, Germany ex India, 20A/452R (B.1.617)) were propagated on Vero E6-TMPRSS2 cells. Omicron BA.1 (B.1.1.529.1; hCoV-19/Germany/BE-ChVir26335/2021, EPI_ISL_7019047) was propagated on CaLu-3 cells. All virus stocks were whole genome sequenced before infection experiments to confirm genetic integrity in the majority of the population, specifically at the furin cleavage site. Before experimental infection, virus stocks were stored at −80 °C.

Animal husbandry

Nine- to 11-week-old Syrian hamsters (Mesocricetus auratus; breed RjHan:AURA) were purchased from Janvier Labs and were housed in groups of 2 to 3 animals in individually ventilated cages. The hamsters had free access to food and water. They were allowed to get used to the housing conditions for 7 d before vaccination. For both experiments, the cage temperatures were maintained at a constant range of 22 to 24 °C with a relative humidity between 40 and 55%.

Vaccination and infection experiments

For infection experiments, Syrian hamsters were randomly assigned to groups, with 50–60% of the animals in each group being female. In the first experiment, 15 hamsters were mock-vaccinated or vaccinated with live-attenuated sCPD9 virus, Ad2-spike or mRNA. Vaccination with sCPD9 was applied by intranasal instillation under anaesthesia (1 × 10⁵ focus-forming units (f.f.u.), 60 µl)⁵³. Ad2-spike (5 × 10⁸ infectious units, 200 μl) and mRNA vaccine (5 μg mRNA, 100 μl) were applied intramuscularly. Mock-vaccinated hamsters were vaccinated by intranasal instillation with sterile cell culture supernatant obtained from uninfected Vero E6-TMPRSS2 cells. At 21 d after vaccination, hamsters were challenge-infected with SARS-CoV-2 Delta variant (1 × 10⁵ plaque-forming units (p.f.u.), 60 µl) by intranasal instillation under anaesthesia. In the second experiment, 10 hamsters were either mock-vaccinated or vaccinated with one of the three vaccines (see above) followed by a booster vaccination 21 d later. At 14 d after booster vaccination, the hamsters were challenged as described above.

Transmission experiments

To determine onward transmission of challenge virus in vaccinated individuals, we vaccinated 3 animals per group in a prime-boost regimen. To this end, hamsters received either 1 × 10⁴ f.f.u. sCPD9delFCS in 60 µl MEM intranasally, 5 μg BNT162b2 mRNA in 100 μl normal saline (0.9% NaCl in sterile water) intramuscularly or 60 µl plain MEM intranasally (mock). Vaccination was boosted using the same vaccines for each respective group 21 d after initial vaccination.

Vaccinated hamsters were challenge-infected with 1 × 10⁵ p.f.u. SARS-CoV-2 variant B.1 as described above. At 24 h after infection, infected vaccinated hamsters were brought into contact with naïve animals and co-habitated to monitor transmission for 6 consecutive days. Daily oral swabs were obtained from each animal to monitor virus shedding and transmission.

Vaccine preparations

sCPD9 was grown on Vero E6-TMRSS2 cells and titrated on Vero E6 cells as described previously; final titres were adjusted to 2 × 10⁶ f.f.u. ml⁻¹ in MEM. Recombinant Ad2-spike was generated, produced on 293T cells and purified as previously described²³. BNT162b2 was obtained as a commercial product (Comirnaty) and handled exactly as recommended by the manufacturer, except that the final concentration of mRNA was adjusted to 50 µg ml⁻¹ (100 µg ml⁻¹ is the recommended concentration for use in humans) by adding injection-grade saline (0.9% NaCl in sterile water) immediately before use.

To increase genetic stability of the sCPD9 construct, the furin cleavage site (FCS) of the spike protein was deleted to create sCPD9delFCS. This FCS-deleted vaccine virus was only used for the transmission study of this paper (Extended Data Fig. 5). Importantly, all vaccines used in this study contain the same SARS-CoV-2 spike antigen derived from the ancestral B.1 (Wuhan) sequence.

Vaccination

sCPD9 was administered intranasally under general anaesthesia (0.15 mg kg⁻¹ medetomidine, 2.0 mg kg⁻¹ midazolam and 2.5 mg kg⁻¹ butorphanol) at a dose of 1 × 10⁵ f.f.u. per animal in a total volume of 60 µl MEM. For transmission experiments (Extended Data Fig. 5), 1 × 10⁴ f.f.u. sCPD9delFCS was applied in the same way. Ad2-spike was injected intramuscularly at 5 × 10⁸ infectious units in 200 µl injection buffer (3 mM KCl, 1 mM MgCl₂, 10% glycerol in PBS). BNT162b2 was injected intramuscularly at a dose of 5 µg mRNA per animal in 100 µl physiological saline (0.9% NaCl in sterile water).

Nasal washes

Nasal washes were obtained from each hamster in this study. To this end, the skull of each animal was split slightly paramedian, such that the nasal septum remained intact on one side of the nose. Subsequently, a 200 µl pipette tip was carefully slid underneath the nasal septum and 150 µl wash fluid (PBS with 30 µg ml⁻¹ ofloxacin and 10 µg ml⁻¹ voriconazole) was applied. The wash was collected through the nostril and the washing procedure was repeated twice; approximately 100 µl of sample was recovered after the third wash.

Nasal washes obtained from the prime-only vaccination trial were subjected to enzyme-linked immunosorbent assay (ELISA) analysis of SARS-CoV-2 spike-specific IgA antibodies. Nasal washes obtained from the prime-boost vaccination trial were used for microneutralization assay to assess their capacity to neutralize the SARS-CoV-2 ancestral variant B1.

Plaque assay

For quantification of replication-competent virus, 50 mg of lung tissue were used. Serial 10-fold dilutions were prepared after homogenizing the organ samples in a bead mill (Analytic Jena). The dilutions were plated on Vero E6 cells grown in 24-well plates and incubated for 2.5 h at 37 °C. Subsequently, cells were overlaid with MEM containing 1.5% carboxymethylcellulose sodium (Sigma Aldrich) and fixed with 4% formaldehyde solution 72 h after infection. To count the plaque-forming units, plates were stained with 0.75% methylene blue.

Histopathology, immunohistochemistry and in situ hybridization

Lungs were processed as previously described⁵³. After careful removal of the left lung lobe, tissue was fixed in PBS-buffered 4% formaldehyde solution (pH 7.0) for 48 h. For conchae preparation, parts of the left skull half were fixed accordingly. Afterwards, lungs or conchae were gently removed from the nasal cavity and embedded in paraffin, cut at 2 µm thickness and stained with hematoxylin and eosin (H&E). In situ hybridization on lungs was performed as previously described⁵⁴ using the ViewRNA ISH Tissue Assay kit (Invitrogen by Thermo Fisher) according to the manufacturer’s instructions, with minor adjustments. For SARS-CoV-2 RNA localization, probes detecting N gene sequences (NCBI database NC_045512.2, nucleotides 28,274–29,533, assay ID: VPNKRHM) were used. Sequence-specific binding was controlled by using a probe for detection of pneumolysin. Immunohistochemistry on conchae was performed as described earlier⁵⁵ (details in Supplementary Methods).

Blinded microscopic analysis was performed by a board-certified veterinary pathologist (J.B.).

SARS-specific Ig measurement by ELISA from serum and nasal washes

An in-house ELISA was performed to investigate SARS-specific IgG levels in serum and SARS-specific IgA levels in nasal washes after vaccination (details in Supplementary Methods).

Neutralization assays from nasal washes

To assess the capacity of nasal washes obtained from the prime-boost vaccination trial to neutralize authentic SARS-CoV-2 (B.1), nasal washes were diluted 1:1 in 2× MEM containing 50 µg ml⁻¹ enrofloxacin and 10 µg ml⁻¹ voriconazole. Subsequent serial dilutions were performed in MEM containing 25 mg ml⁻¹ enrofloxacin, 5 µg ml⁻¹ voriconazole and 1% FBS. SARS-CoV-2 (50 p.f.u.) were added to the nasal wash dilutions and dilutions from 1:2 to 1:256 were plated on near-confluent Vero E6 cells seeded in 96-well cell culture plates. At 3 d after inoculation, cells were fixed and stained with methylene blue. To identify virus-neutralizing dilutions, the integrity of the cell monolayer was assessed by comparison with control wells that contained either no nasal wash or no virus. The last dilution with no evidence of virus-induced cytopathic effect was considered the neutralizing titre for the respective sample.

Serum neutralization assay

Serum samples were tested for their ability to neutralize different SARS-CoV-2 variants. Day 0 samples of the prime-boost trial could not be tested for neutralizing antibodies against B.1.351 (Beta) due to lack of material. Sera were inactivated at 56 °C for 30 min. Twofold serial dilutions (1:8 to 1:1,024) were plated on 96-well plates and 200 p.f.u. SARS-CoV-2 were pipetted into each well. After an incubation time of 1 h at 37 °C, the dilutions were transferred to 96-well plates containing sub-confluent Vero E6 cells and incubated for 72 h at 37 °C (B.1, Beta, Delta) or for 96 h at 37 °C (Omicron). The plates were fixed with 4% formaldehyde solution and stained with 0.75% methylene blue. Wells that showed no cytopathic effect were considered neutralized.

IFN-γ ELISpot analysis

Hamster IFN-γ ELISpot analysis was performed as described previously⁵⁶. In brief, the hamster IFN-γ ELISpot^BASIC kit (MABTECH) was used to detect IFN-γ secretion by 5 × 10⁵ isolated splenocytes, each in co-culture with different stimuli. Medium-treated splenocytes served as negative control and recombinant ovalbumin (10 mg ml⁻¹) was used as negative protein control stimulus. General stimulation of T cells was achieved using 5 μg ml⁻¹ concanavalin A (ConA, Sigma Aldrich). Recombinant SARS-CoV-2 (2019-nCoV) spike protein (S1 + S2 ECD, His tag; 10 mg ml⁻¹; Sino Biological Europe) or 10 mg ml⁻¹ recombinant SARS-CoV-2 (2019-nCoV) nucleocapsid protein (N) (Sino Biological Europe) were used to re-stimulate SARS-CoV-2-specific T cells. Spots were counted using an Eli.Scan ELISpot scanner (AE.L.VIS) and the analysis software ELI.Analyse v5.0 (AE.L.VIS).

RNA extraction and qPCR

To quantify genomic copies in oropharyngeal swabs and 25 mg homogenized lung tissue, RNA was extracted using innuPREP Virus DNA/RNA kit (Analytic Jena) according to the manufacturer’s instructions. qPCR was performed using the NEB Luna Universal Probe One-Step RT–qPCR kit (New England Biolabs) with cycling conditions of 10 min at 55 °C for reverse transcription, 3 min at 94 °C for activation of the enzyme, and 40 cycles of 15 s at 94 °C and 30 s at 58 °C on a qTower G3 cycler (Analytic Jena) in sealed qPCR 96-well plates. Primers and probes were used as previously reported⁵⁷. Oligonucleotides (Sequence (5’-3’)): E_Sarbeco_F: ACAGGTACGTTAATAGTTAATAGCGT;

E_Sarbeco_R: ATATTGCAGCAGTACGCACACA;

E_Sarbeco_P1: FAM-ACACTAGCCATCCTTACTGCGCTTCG-BBQ.

Mesocricetus auratus genome annotation

For quantification of gene expression, we used the MesAur 2.0 genome assembly and annotation available via the NCBI genome database ( The GFF file was converted to GTF using gffread 0.12.7⁵⁸. Where no overlaps were produced, 3’-UTRs in the annotation were extended by 1,000 bp as described previously⁵⁹. Further polishing steps for the GTF file are described on the GitHub page accompanying this paper ( The final gtf file used for the analysis is available through GEO (https://www.ncbi.nlm.nih.gov/geo/query/acc.cgi?acc=GSE200596).

Bulk RNA extraction

To perform RNA bulk sequencing, RNA was isolated from lung tissue using Trizol reagent according to the manufacturer’s instructions (Ambion, Life Technologies). Briefly, 1 ml Trizol was added to the homogenized organ sample and vortexed thoroughly. After an incubation time of 20 min, 200 µl of chloroform were added. The samples were vortexed again and incubated for 10 min at room temperature. Subsequently, tubes were centrifuged at 12,000 × g for 15 min at 4 °C and 500 µl of the aqueous phase were transferred into a new tube containing 10 µg GlycoBlue. Isopropanol (500 µl) was added, followed by vortexing, incubating and centrifuging the samples as described above. Thereafter, isopropanol was removed and 1 ml of ethanol (75%) was applied. The tubes were inverted shortly and centrifuged at 8,000 × g for 10 min. After freeing the pellet from ethanol, RNA was resuspended in 30 µl of RNase-free water and stored at −80 °C.

Cell isolation from blood and lungs

White blood cells were isolated from EDTA-blood as previously described; steps included red blood cell lysis and cell filtration before counting. Lung cells (caudal lobe) were isolated as previously described^26,60; steps included enzymatic digestion, mechanical dissociation and filtration before counting in trypan blue. Buffers contained 2 µg ml⁻¹ actinomycin D to prevent de novo transcription during the procedures.

Cell isolation from nasal cavities

To obtain single-cell suspensions from the nasal mucosa of SARS-CoV-2-challenged hamsters, the skull of each animal was split slightly paramedian so that the nasal septum remained intact on the left side of the nose. The right side of the nose was carefully removed from the cranium and stored in ice-cold 1× PBS with 1% BSA and 2 µg ml⁻¹ actinomycin D until further use. Nose parts were transferred into 5 ml Corning Dispase solution supplemented with 750 U ml⁻¹ Collagenase CLS II and 1 mg ml⁻¹ DNase, and incubated at 37 °C for 15 min. For preparation of cells from the nasal mucosa, the conchae were carefully removed from the nasal cavity and re-incubated in digestion medium for 20 min at 37 °C. Conchae tissue was dissociated by pipetting and pressing through a 70 µm filter with a plunger. Ice-cold PBS with 1% BSA and 2 µg ml⁻¹ actinomycin D was added to stop the enzymatic digestion. The cell suspension was centrifuged at 400 × g at 4 °C for 15 min and the supernatant discarded. The pelleted nasal cells were resuspended in 5 ml red blood cell lysis buffer and incubated at room temperature for 2 min. Lysis reaction was stopped with 1× PBS with 0.04% BSA and cells centrifuged at 400 × g at 4 °C for 10 min. Pelleted cells were resuspended in 1× PBS with 0.04% BSA and 40 µm-filtered. Live cells were counted in trypan blue and viability rates determined using a counting chamber. Cell concentration for scRNA-seq was adjusted by dilution.

Single-cell RNA sequencing

Isolated cells from blood, lungs and nasal cavities of Syrian hamsters were subjected to scRNA-seq using the 10× Genomics Chromium Single Cell 3’ Gene Expression system with feature barcoding technology for cell multiplexing (details in Supplementary Methods).

Analysis of single-cell RNA sequencing data

Sequencing reads were initially processed using bcl2fastq 2.20.0 and the multi command of the Cell Ranger 6.0.2 software. For the cellplex demultiplexing, the assignment thresholds were partially adjusted (for details, see the GitHub page at Further processing was done in R 4.0.4 Seurat R 4.0.6 package⁶¹, as well as R packages ggplot2 3.3.5, dplyr 1.0.7, DESeq2 1.30.1, lme4 1.1–27.1 and dependencies, and in Python 3.9.13 as well as Python packages scanpy 1.9.1, scvelo 0.2.4 and dependencies. In the next step, cells were filtered by a loose quality threshold (minimum of 250 detected genes per cell) and clustered. Cell types were then annotated per cluster and filtered using cell type-specific thresholds (cells below the median or in the lowest quartile within a cell type were removed). The remaining cells were processed using the SCT/integrate workflow⁶² and cell types again annotated on the resulting Seurat object. All code for downstream analysis is available on GitHub at https://github.com/Berlin-Hamster-Single-Cell-Consortium/Live-attenuated-vaccine-strategy-confers-superior-mucosal-and-systemic-immunity-to-SARS-CoV-2.

Statistics and reproducibility

Details on statistical analysis of sequencing data including pre-processing steps are described in the individual Methods section. Analyses of virological, histopathological, ELISA, cell frequencies and cell number statistics were performed with GraphPad Prism 9. Statistical details are provided in respective figure legends. No statistical method was used to predetermine sample size. Data distribution was assumed to be normal but this was not formally tested. No data were excluded from the analyses. All experiments involving live animals were randomized, other experiments were not randomized. The investigators were blinded to allocation of hamsters during animal experiments and primary outcome assessment (clinical development, virus titrations, qPCR, ELISpot, serology and histopathology). Investigators were not blinded to allocation in other experiments and analyses.

Reporting summary

Further information on research design is available in the Nature Portfolio Reporting Summary linked to this article.