Approval for the experiment including SARS-CoV-2 614G Gamma and Delta variants was given by the Dutch authorities (Project license/working protocol (WP) number: 27700202114492-WP12). The Omicron study was conducted in Hannover at the University of Veterinary Medicine, Foundation with the approval of the Niedersächsisches Landesamt für Verbraucherschutz und Lebensmittelsicherheit (LAVES file number 21/3755). The animals were under veterinary observation during the experiment and all efforts were made to minimize distress. Eight to ten weeks old male Syrian golden hamsters were kept for ten days under BSL-3 conditions prior to the experiment for acclimatization. Hamsters (n=22) were divided in 4 groups of 5-6 hamsters each. They were infected by administrating a suspension containing 104 TCID50 with either, Gamma (n=5), Delta (n=5), Omicron (n=6) VOCs or SARS-CoV-2 614G (n=6) respectively. For four days the animals were monitored and weighed twice daily until euthanized at day 4 post infection using an overdose of Ketamine and Medetomidin followed by exsanguination. Immediately after death, necropsies were performed, the right lung lobe and nasal swabs were collected and frozen at -700C for virological analyses. Subsequently the left lung lobe, nasal turbinates, and tracheas were fixed in 10% buffered formalin (Chemie Vertrieb GmbH & Co Hannover KG, Hannover, Germany). In addition, brain, liver, spleen, kidney, adrenal gland, stomach, small and large intestine, pancreas and testicle from hamsters infected with the VOC Omicron were also collected. Lungs were pre-fixed by injections of 10% buffered formalin as recommended by Meyerholz and colleagues39 to ensure an optimal histopathological evaluation. Nasal samples, following formalin fixation, were decalcified for about 14 days prior routine tissue processing.
SARS-CoV-2 614G (isolate Bavpat-1; European Virus Archive Global #026 V-03883) was grown to passage 3 on VeroE6 cells, and VOCs were grown to passage 3 on Calu-3 cells. For stock production, infections were performed at a multiplicity of infection (moi) of 0.01 and virus was collected at 72 hours post-infection, clarified by centrifugation and stored at -80°C in aliquots. Stock titers were determined as described below. All work with infectious SARS-CoV-2 was performed in a Class II Biosafety Cabinet under BSL-3 conditions at Viroclinics Xplore.
Viral genome sequences were determined using Illumina deep-sequencing as described before 40. The 614G virus contained a spike S686G change in 48% of reads compared with the passage 1 (kindly provided by Dr. Christian Drosten) and no other variants >40%. The VOC Gamma, Delta and Omicron variant passage 3 sequences were identical to the original respiratory specimens and no minor variants >40% were detected. For VOC Omicron, the S1 region of spike was not covered well, due to primer mismatches. Therefore, the S1 region of the original respiratory specimen and passage 3 virus were confirmed to be identical by Sanger sequencing. VOC Gamma contained the following spike changes: L18F, T20N, P26S, D138Y, R190S, K417T, E484K, N501Y, D614G, H655Y, T1027I, V1176F. VOC Delta contained the following spike changes: T19R, G142D, del156-157, R158G, A222V, L452R, T478K, D614G, P681R and D950N. VOC Omicron contained the following spike mutations: A67VS, del69-70, T95I, G142-, del143-144, Y145D, del211, L212I, ins215EPE, G339D, S371L, S373P, S375F, K417N, N440K, G446S, S477N, T478K, E484A, Q493R, G496S, Q498R, N501Y, Y505H, T547K, D614G, H655Y, N679K, P681H, N764K, D796Y, N856K, Q954H, N969K, L981F. The VOCs gamma, delta and omicron sequences are available on Genbank under accession numbers OM442897, OM287123, and OM287553, respectively
Virus infectivity titration
Virus infectious titers were determined in Vero cells for the SARS-CoV-2 614G, VOCs Gamma and Delta variants. Due to the limited replication of VOC Omicron in Vero cells, this variant was propagated and titrated in Calu-3 cells. Cells were seeded in 96 well plate and incubated at 37°C. Then 24h after seeding, cell culture media was replaced by DMEM + 2% FBS in the case of Vero cells that were infected with 10 fold serial dilutions of lung or nasal turbinate homogenate tissue samples. Plates were further incubated in a humidified atmosphere at 37°C, 5 % CO2. Five days after infection, cytopathic effect was evaluated In the case of VOC Omicron titration, culture media was replaced for MEM + 2% FBS, 5dpi cells were fixed and stained using anti-SARS-CoV-2 Nucleocapsid antibody (Sinobiological). Viral titers (TCID50/ml) were calculated using the “Spearman-Kärber method” 41.
Formalin-fixed paraffin embedded samples were cut into 2 μm thick serial sections and stained with hematoxylin and eosin (H&E). Sections of the nasal turbinates, trachea, and lung were scanned using an Olympus VS200 Digital slide scanner (Olympus Deutschland GmbH, Hamburg, Germany) and evaluated in a blinded manner with a semi-quantitative scoring system with special emphasis on inflammation, degeneration and regeneration as previously described, with minor modification17. Histopathological semi-quantitative evaluations were performed by veterinary pathologists (GB, FA, MC, LA) and subsequently confirmed by a European board certified veterinary pathologist (WB). Nasal turbinates were evaluated on a full length longitudinal section of the nose including respiratory and olfactory epithelium. Trachea was evaluated on cross- and longitudinal sections along the entire length of the organ. Finally, the lung was evaluated on one cross section (at the level of the entry of the main bronchus) and one longitudinal section (along the main bronchus) of the entire left lung lobe. The applied scoring systems are provided in details in Supplementary table 5.
Immunohistochemistry of SARS-CoV-2 NP was performed using the Dako EnVision+ polymer system (Dako Agilent Pathology Solutions) and 3,3´-Diaminobenzidine tetrahydrochloride (Sigma-Aldrich, St. Louis, MO, United States) as previously described32,33. Monoclonal mouse primary antibody against SARS-CoV-2 NP (Sino Biological, Peking, China-40143-MM05; dilution 1 :16000 ) was applied overnight at 4◦C.
Digital image analysis
For the quantification of immunolabelled cells in nasal turbinates as well as in tracheal and pulmonary tissue, slides were digitized using the Olympus VS200 (Olympus Deutschland GmbH, Hamburg, Germany) slide scanner. Image analysis was performed using the open source software package QuPath for digital pathology image analysis42. For all animals nasal turbinates and tracheal whole slides images as well as one longitudinal section (along the main bronchus) of the entire left lung lobe were evaluated. In brief, regions of interest (ROI), in the nasal turbinates (respiratory and olfactory mucosa) and the trachea (tracheal epithelum and subepithelial layer) were indicated by a veterinary pathologist. In the lung total tissue was detected automatically through digital thresholding and additional ROIs for conductive airways, including bronchi, bronchioles and terminal bronchioles were subsequently indicated by a veterinary pathologist. Lung parenchyma (alveolar and vascular compartments) was then obtained by subtraction of conductive airways from total lung tissue. The total numbers of immunolabelled and non-labelled cells was determined by automated cell detection in all ROIs, based on marker and tissue specific thresholding.
Statistical analyses and graph design were performed using GraphPad Prism (GraphPad Software, San Diego, CA, USA) for Windows™. Data was tested for significant differences using Kruskal-Wallis tests. Pairwise comparisons among groups were obtained by two-tailed Mann-Whitney-U tests. Statistical significance was accepted at exact p-values of ≤ 0.05 (*), ≤ 0.01 (**) and ≤ 0.001 (***), respectively.