key: cord-0809078-yxqi7m2o
authors: Saccon, Elisa; Chen, Xi; Mikaeloff, Flora; Rodriguez, Jimmy Esneider; Szekely, Laszlo; Vinhas, Beatriz Sá; Krishnan, Shuba; Byrareddy, Siddappa N.; Frisan, Teresa; Végvári, Ákos; Mirazimi, Ali; Neogi, Ujjwal; Gupta, Soham
title: Cell type resolved quantitative proteomics map of interferon response against SARS-CoV-2
date: 2021-04-20
journal: iScience
DOI: 10.1016/j.isci.2021.102420
sha: ccf2d84bff94c0d2128139c6a95288cd1d7effbc
doc_id: 809078
cord_uid: yxqi7m2o

The commonly used laboratory cell lines are the first line of experimental models to study the pathogenicity and performing antiviral assays for emerging viruses. Here, we assessed the tropism and cytopathogenicity of the first Swedish isolate of SARS-CoV-2 in six different human cell lines, compared their growth characteristics and performed quantitative proteomics for the susceptible cell lines. Overall, Calu-3, Caco2, Huh7, and 293FT cell lines showed a high to moderate level of susceptibility to SARS-CoV-2. In Caco2 cells the virus can achieve high titers in the absence of any prominent cytopathic effect. The protein abundance profile during SARS-CoV-2 infection revealed cell-type-specific regulation of cellular pathways. Type-I interferon signaling was identified as the common dysregulated cellular response in Caco2, Calu-3 and Huh7 cells. Together, our data shows cell-type specific variability for cytopathogenicity, susceptibility and cellular response to SARS-CoV-2 and provide important clues to guide future studies.

Severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2), the causative agent of 45 coronavirus disease 2019 (COVID-19) pandemic, is a highly pathogenic coronavirus that has 46 created a global public health challenge (Hu et al., 2020). The virus primarily attacks the lung 47 and Calu-3 (lung) that were seeded for 72h prior to infection showed the highest virus production 106 with >4 log10 RNA copies by 48hpi (p<0.001) and thereafter marginal increase till 120hpi. It 107 was interesting to note that Calu-3 cells, which were infected after 72h of seeding and showed 108 tightly closed together cells with polygonal or cuboidal features and defined boundary, had a 109 higher susceptibility to SARS-CoV-2 compared to Calu-3 cells that were infected after 24h of 110 seeding (round and isolated) (Supplementary Figure S1A) . Immunofluorescent staining for β-111 catenin and β-actin that are essential for organization of polarized epithelium and cell to cell 112 contact showed a defined co-localization along the cell margin after 72h of incubation indicating 113 polarization of the Calu-3 cells (Supplementary Figure S1B) . This leads us to speculate that 114 enhanced susceptibility of Calu-3 with longer incubation of 72h prior to infection was possibly 115 due to polarization of the cells (Foster et al., 2000) , as it was reported previously for SARS-CoV 116 (Tseng et al., 2005) . 293FT (kidney; p<0.01) and Huh7 (liver; p<0.02) showed moderate virus 117 production with >1 log10 viral RNA copies in the supernatant by 120hpi. 16HBE (lung) and 118 A549 (lung) cells showed very poor virus production with <0.6 log10 RNA copies. Interestingly, 119 other than Vero-E6, viral-induced cytotoxicity was only observed in Calu-3 cells with a viability 120 of ≤50% by 48hpi and ≤80% by 72hpi. None of the other cell lines showed any apparent 121 cytotoxicity (viability>85%) ( Figure 1B) . 122

We observed that the virus production in susceptible cell lines reached saturation by 48hpi. 123 Therefore, we investigated the changes caused in the cell surface of Calu-3, Caco2, Huh7, and 124 293FT cells during virus production at 48hpi (moi 0.1) using scanning electron microscopy 125 (SEM). We did not observe any significant changes in the morphology of the cell surface in the 126 mock-infected cells. In SARS-CoV-2 infected Calu-3 and Caco2 cells, numerous virus-like 127 particles corresponding to the size of SARS-CoV-2 (approx. 70nM) were observed to be attached 128

Since we observed differential susceptibility to SARS-CoV-2 infection in different cell lines 170 originating from the human lung (Calu-3), intestine (Caco2), liver (Huh7) and kidney (293FT), 171 we investigated how the cellular proteins are regulated during infection in these cell lines. To this 172 end, we either infected or mock-infected polarized Calu-3 cells, Caco2 cells, Huh7 cells and 173

293FT cells with SARS-CoV-2 (moi 1) in triplicates. For proteomics we used 24h of infection 174 since, at this time point despite maintaining high viability all the cell lines showed exponential 175 virus production, that plateaued after 24hpi. The cells were harvested 24hpi, lysed and equal 176 concentration of the protein was used to perform quantitative proteomics using a TMT-labeling 177 strategy as previously described (Appelberg et al., 2020). The unprocessed and processed raw 178 data is presented in supplementary tables (Table S4 -Table S11 ). Among the four cell lines, 179

Calu-3 showed major changes in protein abundance upon infection, with 6462 proteins 180 differentially expressed in infected cells than the mock, followed by Caco2 with a significant 181 difference in 177 proteins. No change in the global protein abundance was observed in Huh7 182 (only four proteins differentially expressed) and 293FT (no proteins differentially expressed) at 183 24hpi ( Figure 3A , Supplemental Table S12 ). The PCA plot showing the sample to sample 184 relationship and the volcano plot showing the differentially altered protein abundance in SARS-185

CoV-2 infected cells as compared to the mock infected cells is shown in Supplementary Figure  186 S2. The virus secretion in the cell culture supernatant at 3hpi and 24hpi is shown in Figure 3B . 187

The viral protein abundance in the cells is shown in Figure 3C . The proteins that were detected 188 are ORF1ab, ORF3a, ORF6, ORF7a, ORF8, M, N, S, nsp4, nsp8 and nsp10. The higher 189 abundance of viral proteins detected in Calu-3 correlated with higher level of virus production 190 and the change in host protein abundance. 191 192

Since Calu-3 and Caco2 were the only cell lines that showed substantial protein regulation upon 195 infection at 24h, we compared the significantly regulated proteins in those two cell lines. As 196 shown in the Venn diagram in Figure 4A , there were 132 proteins that were commonly similarly upregulated (44 proteins) and downregulated (44 proteins) in both the cell lines 199 (Supplemental figure S3). Reactome pathway analysis on the 132 significantly altered proteins  200   common to both cell lines showed a strong enrichment of type-I and type-II interferon Table S13 ). At 24hpi, both 293FT and Huh7 did not show any differentially 210 regulated protein belonging to IFN-signaling pathways (109 detected proteins; Supplemental 211 Table S13 ) and the heat maps are shown in Supplemental Figure S4 . As shown in the heatmaps, 212 among 129 detected proteins belonging to these pathways, 105 were differentially regulated in Table S13 ). The protein-protein interaction network of the significantly altered 216 proteins showed two definite clusters in Calu-3: one including proteins associated with RIG-I 217 (DDX58) and type-I/II signaling complex and another majorly including components of 218 nucleoporin complex that were down-regulated. The karyopherin family, and a single cluster 219 (RIG-I (DDX58) and type-I signaling complex) were upregulated in Caco2 (Supplemental 220 Figure S5A and S5B). In general, we observed an interferon stimulation in SARS-CoV-2 221 infected Calu-3 and Caco2 cells and SARS-CoV-2 receptor ACE2 has been considered to be an 222 interferon stimulatory gene (Ziegler et al., 2020) . However, we did not observe any significant 223 differences in the protein levels of ACE2 or TMPRSS2 upon infection in our proteomics data, 224 rather ACE2 was down-regulated in SARS-CoV-2 infected cells (Supplemental Figure S6) . Figure 7C . We observed discordance in 4 proteins, where TTR and IFI35 were 322 upregulated in their study but downregulated in ours and ITGB4 and LYPD3 were 323 downregulated in their study but were upregulated in ours. We also specifically looked into the 324 proteins related to IFN-signaling pathways and observed several nuclear transporters to be 325 upregulated in the Bojkova et al. study. Among the ISGs, only ISG15 showed an upregulation in 326 both the data (Supplemental Figure S11 ). Of note, unlike others, the Frankfurt strain was the only 327 strain that was isolated and adapted in Caco2 that could have possibly led to higher susceptibility ISGs like IFIT1, ISG15 and DDX58 were upregulated in all the three cell lines (Figure 6) . 381

Induction of type I and type II IFN that was proportional to the viral load was also noted in a 382 proteomics study performed in autopsy lung material of fatal COVID-19 cases and lung tissues obtained from SARS-CoV-2 infected non-human primates in a longitudinal manner (Kalocsay et 384 al., 2020) . Another recent study performed on autopsy tissues from varies organs observed that 385 COVID-19 patients with coronary heart disease had upregulation of multiple protein belonging 386 to RIG-I signaling pathway and overall, interferon gamma receptor 1 (IFNGR1) was 387 dysregulated in all the major organs except for thyroid and testes (Nie et al., 2021) . These 388 findings suggest that even in presence of cell-type specific diversity in cellular responses, there 389 are common pathways that could be efficiently targeted to inhibit SARS-CoV-2. This is 390 particularly relevant since SARS-CoV-2 can infect different organs of the body (Mallapaty, 391 2020; Trypsteen et al., 2020) . 392

In conclusion, we identify some cell lines of human origin that could be used to study the 394 biological properties of SARS-CoV-2. In addition, we observed that type-I interferon is 395 commonly regulated during infection in cell lines originating from lungs, colon and liver and, 396 thus deserving more mechanistic studies to identify factors that could be utilized to control the 397 infection. 398

We acknowledge some limitations of our study. Foremost, is that the analysis is restricted to cell 401 lines, which may not be physiologically representative of the human tissue, like other ex vivo 402 systems such as organoids, However the use of cell lines can still provide an overview of the 403 complexity and variability of the interaction between SARS-CoV-2 and the human cellular 404 targets. Furthermore, we restricted our proteomics study to 24hpi and more detailed time kinetics 405 experiments are required to elucidate better the dynamic changes occurring during infection.

Lead contact 417

Further information and requests for resources and reagents should be directed to and will be 418 fulfilled by the Lead Contact, Soham Gupta, soham.gupta@ki.se. 419

No new reagents were created in this study. 421

The mass spectrometry proteomics data (raw MS files and search files) have been deposited to Table S4 : 293FT and Huh7 TMTpro Raw Peptide groups abundances, related to Figure  552 3. Peptide list containing raw TMTpro-labeling abundances across the samples. Peptide 553 identifications were obtained from 12 HpH-RPLC fractions. All entries were considered with 1% 554 FDR. 555 556 Table S5 : 293FT and Huh7 TMTpro Raw protein groups abundances, related to Figure  557 3. Master protein list calculated from peptide identifications. Each entry contains raw 558 abundances for quantification across the samples. All entries were considered with 1% FDR. 559 Table S6 : Calu-3 and Caco2 TMTpro Raw Peptide groups abundances, related to Figure Table S7 : Calu-3 and Caco2 TMTpro Raw protein groups abundances, related to Figure  566 3. Master proteins list calculated from peptide identifications. Each entry contains raw 567 abundances for quantification across the samples. All entries were considered with 1% FDR. 568 Akt/mTOR/HIF-1 signaling identified by proteo-transcriptomics of SARS-CoV-2 infected cells.

Proteomics of SARS-CoV-2-infected host cells reveals therapy targets

Type-I interferon signatures in SARS

Comparative tropism, replication kinetics, and cell damage profiling of

SARS-CoV-2 and SARS-CoV with implications for clinical manifestations, transmissibility, and 602 laboratory studies of COVID-19: an observational study

COVID-19: organoids go viral

Detection of 2019 novel coronavirus 607 (2019-nCoV) by real-time RT-PCR

Characterization of the

Calu-3 cell line as a tool to screen pulmonary drug delivery

β-Coronaviruses Use Lysosomes for Egress 616 Instead of the Biosynthetic Secretory Pathway

A SARS-CoV-2 protein interaction map 619 reveals targets for drug repurposing

Shotgun proteomics analysis of SARS-CoV-2-infected 622 cells and how it can optimize whole viral particle antigen production for vaccines

Tissue distribution of ACE2 protein, the functional receptor for SARS coronavirus. A first step 626 in understanding SARS pathogenesis

TMPRSS2 and ADAM17 cleave ACE2 differentially and only proteolysis by TMPRSS2 629 augments entry driven by the severe acute respiratory syndrome coronavirus spike protein

Tropism, replication competence, and innate 638 immune responses of the coronavirus SARS-CoV-2 in human respiratory tract and conjunctiva: 639 an analysis in ex-vivo and in-vitro cultures

Propagation, Inactivation, and Safety Testing 641 of SARS-CoV-2

Multiplexed proteomics and 644 imaging of resolving and lethal SARS-CoV-2 infection in the lung. bioRxiv

Tracking Changes in SARS-CoV

Evidence that D614G Increases Infectivity of the COVID-19 Virus

Implications of central 652 carbon metabolism in SARS-CoV-2 replication and disease severity. bioRxiv

Virus-Host Interactome and Proteomic Survey Reveal Potential Virulence Factors

Influencing SARS-CoV-2 Pathogenesis

Distinct infection process of SARS-CoV-2 in human bronchial epithelial cells line

Mini organs reveal how the coronavirus ravages the body

Enhanced isolation of SARS-CoV-2 by TMPRSS2-667 expressing cells

Multi-organ proteomic landscape of COVID-19 autopsies

E6 cells: replication kinetics, rapid adaptation and cytopathology

Differential immune activation profile of SARS-CoV-2 and SARS-CoV 681 infection in human lung and intestinal cells: Implications for treatment with IFN-β and IFN 682 inducer

A Perspective on Organoids for Virology Research

Multi-level proteomics reveals host-perturbation 687 strategies of SARS-CoV-2 and SARS-CoV. bioRxiv

In silico and in vitro studies reveal complement system drives coagulation 690 cascade in SARS-CoV-2 pathogenesis. Computational and structural biotechnology journal 18

On the whereabouts of SARS-CoV-2 in the human body: A systematic review

Apical entry 696 and release of severe acute respiratory syndrome-associated coronavirus in polarized Calu-3 lung 697 epithelial cells

Data, Reagents, Assays and Merits of Proteomics for

SARS-CoV-2 Research and Testing

SARS-CoV-2 Receptor ACE2 Is an

Interferon-Stimulated Gene in Human Airway Epithelial Cells and Is Detected in Specific Cell 706 Subsets across Tissues