SARS-CoV-2 Variants Show Different Host Cell Proteome Profiles With Delayed Immune Response Activation in Omicron-Infected Cells

The ancestral SARS-CoV-2 strain that initiated the Covid-19 pandemic at the end of 2019 has rapidly mutated into multiple variants of concern with variable pathogenicity and increasing immune escape strategies. However, differences in host cellular antiviral responses upon infection with SARS-CoV-2 variants remain elusive. Leveraging whole-cell proteomics, we determined host signaling pathways that are differentially modulated upon infection with the clinical isolates of the ancestral SARS-CoV-2 B.1 and the variants of concern Delta and Omicron BA.1. Our findings illustrate alterations in the global host proteome landscape upon infection with SARS-CoV-2 variants and the resulting host immune responses. Additionally, viral proteome kinetics reveal declining levels of viral protein expression during Omicron BA.1 infection when compared to ancestral B.1 and Delta variants, consistent with its reduced replication rates. Moreover, molecular assays reveal deferral activation of specific host antiviral signaling upon Omicron BA.1 and BA.2 infections. Our study provides an overview of host proteome profile of multiple SARS-CoV-2 variants and brings forth a better understanding of the instigation of key immune signaling pathways causative for the differential pathogenicity of SARS-CoV-2 variants.

Proteome Discoverer (PD) 2.4 software (ThermoFisher Scientific. Proteomics data was analysed further using PERSEUS (24) 1.6.15.0 software. Statistics determined using unpaired two-sided students T. test were additionally FDR corrected and values q < 0.05 were considered significant. Molecular assays were performed in three independent experiments with three replicates each. Analysis as well as graphical representation was conducted with Graph Pad Prism (version 9.3.1, GraphPad Software). Statistical significance was calculated by one or two-way ANOVA (ns-not significant, * p < 0.05, ** p < 0.01, *** p < 0.005, **** p < 0.0001) Error bars are described in the figure legends.
Virus variants and propagation SARS-CoV-2 virus strains were propagated as described elsewhere (25). In brief, SARS-CoV-2 was grown on Caco-2 cells and the Tissue Culture Infection Dose50 (TCID50) was calculated according to Spearman and Kaerber by titration of supernatants on 96-well plates of confluent Caco-2 (26,27). Virus stocks were stored at -80 °C. The variants used in this study were as

Virus infection
For infection assays, cells were washed with PBS and maintained in medium with reduced FCS concentration (1 %). Cells were infected with SARS-CoV-2 at indicated MOI for 2 h and supplied with fresh medium after the incubation period. Virus containing samples were inactivated using validated protocols (29). All experiments with viable SARS-CoV-2 were performed under Biosafety Level-3. Immunofluorescence J o u r n a l P r e -p r o o f Cells were fixed with 3 % PFA for 30 min and subsequently permeabilised with 0.5 % Triton-X for 30 min. After blocking for 1 h with 1 % BSA, primary antibody was diluted in 0.5 % BSA and the solution incubated over night at 4 °C. The secondary antibody mixed with DAPI (0.02 mg/ml) was incubated for 1 h at RT. Primary antibodies used were as followed: mouse anti SARS-CoV-2 Spike (GeneTex #GTX632604 1:1000), rabbit anti SARS-CoV-2 Spike (SinoBiological #40150-R007 1:1000), rabbit anti IRF-3 (Cell Signaling #4302 1:1000), rabbit anti NF-κB (Cell Signaling #8242 1:1000). Secondary antibodies used were as followed: goat anti mouse Alexa 488 (Invitrogen #A11001 1:1000), goat anti rabbit Alexa 488 (Invitrogen #A11008 1:1000), goat anti rabbit Alexa 647 (Invitrogen, #A21244 1:1000). Images were acquired using the Operetta CLS High Content Analysis System (PerkinElmer) followed by image analysis with Harmony (PerkinElmer).

Quantification of intracellular virus RNA
Infected cells were lysed with RLT buffer (QIAGEN) and RNA was isolated using the RNeasy  Table 1 and 2). Samples were collected in lysis buffer (100 mM TRIS, 2 % SDS, 10 mM TCEP, 40 mM 2-CAA) and sample lysates for total assays were performed as described in (30,31). Briefly, sample lysates were methanol/chloroform precipitated and resuspended in buffer containing 8 M Urea and 100 mM Tris (PH.8). Protein concentration was determined by Bradford assay and 300 µg of protein per sample was used for digestion after dilution to 1M Urea and 100mM Tris. Samples were digested with 1:50 wt/wt LysC and 1:100 wt/wt Trypsin overnight at 37 °C. Digested samples were acidified with TFA and peptides were purified using Waters Oasis Prime HLB 30mg columns according to manufacturer`s instructions. Dried peptide samples were resuspended in TMT labelling buffer containing 200 mM EPSS and 10 % acetonitrile and incubated at 37 °C for 10 min. Peptide concentration was determined by µBCA and 100 µg of peptides per sample were used for TMT labelling (TMTpro-16) by one-hour incubation at room temperature (RT) using a 1:2.5 peptide/TMT-ratio. The J o u r n a l P r e -p r o o f reaction was quenched by addition of 1:10 (vol) 5 % hydroxylamine solution at RT for 15 min.
TMT labelling quality was verified by mixing equimolar ratios of each TMT channel followed by single injection measurement by LC-MS/MS. Samples were pooled acidified using 20 % TFA and purified using SepPak (Waters Oasis Prime HLB 30mg columns). For whole cell proteome pooled peptides were used for High pH Reverse phase fractionation by Dionex Ultimate 3000 analytical HPLC (30,31). The eluted peptides were collected for 96 fractions and cross concatenated into 24 fractions and dried for processing. Liquid chromatography and mass spectrometry were performed as described previously in (30,31).
Mass spectrometry raw data analysis was performed using Proteome Discoverer (PD) 2.4 software (ThermoFisher Scientific). Default settings were used for selection of spectra.
SequestHT node was opted for database searches against trypsin digested Homo sapiens reference proteome (Taxonomy ID 9606) downloaded from UniProt (12-March-2020; "One Sequence Per Gene", 20531 entries) and SARS-CoV-2 (UniProt pre-release, 10-February-2020, Taxonomy ID 2697049; 14 entries). Precursor mass tolerance of 7 ppm and a fragment mass tolerance of 0.5 Da was set in the database search. Static modifications were set as TMTpro at the N terminus and carbamidomethyl at cysteine residues. The following dynamic modifications were taken into account: Oxidation (M), and Acetyl (Protein N terminus). False discovery rates were controlled using Percolator < 0.01 FDR at peptide and < 0.05 FDR at protein level. For whole cell proteomics quantification all PSMs were summed intensity normalized, followed by IRS (32) normalization. Further data analysis was performed using PERSEUS (24) 1.6.15.0 software. Significance was tested using unpaired two-sided students T. test and values were further FDR corrected. Values q < 0.05 were considered significant.

Immunoblot analysis
Confluent cells in 6-well plates were lysed using Triton-X lysis buffer supplemented with protease and phosphatase inhibitors (Roche) and left on ice for 30 min. Samples were mixed with an equal volume of Laemmli buffer (Sigma) supplemented with 5 % ß-mercaptoethanol.

Replication kinetics of SARS-CoV-2 variants in Calu-3 lung cells
The SARS-CoV-2 VOC Omicron BA.1 carries several mutations within the Spike (S) protein resulting in impaired cell entry ( Fig 1A). To identify resulting changes in cellular replication, we

Induction of cellular immune response pathways upon SARS-CoV-2 variant infection
To attain insight into variant-specific host cell changes, we next examined host cellular proteins that significantly changed across SARS-CoV-2 variants in comparison to mock-infected cells (Supplementary Table 4

Interferon and NF-B signalling induced by infection with SARS-CoV-2 variants
Since the proteome analysis revealed evident increases of host cell immune signalling proteins corresponding to antiviral pathways including IFN and NF-B signalling, we analysed phosphorylation patterns of IRF-3 and NF-B using immunoblots. IRF-3 is known to induce expression of type I and III interferons and NF-B to induce expression of pro-inflammatory cytokines (22,23). We also included the formerly dominating variant BA.2, which was not circulating at the time of proteome analyses. Interestingly, viral levels of Omicron BA.2 were even lower than those of Omicron BA.1 ( Figure S4).

Nuclear translocation of key transcription factors and INF release upon SARS-CoV-2 variant infection
Phosphorylated IRF-3 and NF-B accumulate in the nucleus, where they act as transcription factors inducing expression of type I and III interferons and pro inflammatory cytokines, respectively (20-23). Thus, we next examined nuclear localisation of IRF-3 and NF-B by quantitative fluorescence microscopy ( Figure 7A). We used a lower MOI of 0.01, in order to gain a higher temporal resolution. Notably, due to the strong CPE induction of B.1 and Delta, In summary, by different orthogonal assays, we observed a deferred immune response upon Omicron infection when compared to the other variants examined in this study.

Discussion
In comparison to other SARS-CoV-2 VOCs, Omicron variants are particularly outstanding for their immune escape and altered pathogenicity (4-6, 11, 12). An unbiased assessment of the host cellular immune response upon infection of an Omicron variant was therefore essential.
In this study, we performed unbiased proteomics combined with molecular assays to compare      Table 3). Statistics were determined as described in proteomics method section, FDR q-value <0.05 was considered significant (n = 3 biologically independent replicates), error bar show ± s.e.m. * q<0.05, ** q<0.02, ns: not significant