key: cord-1014164-ahpkmffy
authors: Blanco-Melo, Daniel; Nilsson-Payant, Benjamin E.; Liu, Wen-Chun; Møller, Rasmus; Panis, Maryline; Sachs, David; Albrecht, Randy A.; tenOever, Benjamin R.
title: SARS-CoV-2 launches a unique transcriptional signature from in vitro, ex vivo, and in vivo systems
date: 2020-03-24
journal: bioRxiv
DOI: 10.1101/2020.03.24.004655
sha: 817fe8b5b1f43b114b5802f168d6294607b3f825
doc_id: 1014164
cord_uid: ahpkmffy

One of the greatest threats to humanity is the emergence of a pandemic virus. Among those with the greatest potential for such an event include influenza viruses and coronaviruses. In the last century alone, we have observed four major influenza A virus pandemics as well as the emergence of three highly pathogenic coronaviruses including SARS-CoV-2, the causative agent of the ongoing COVID-19 pandemic. As no effective antiviral treatments or vaccines are presently available against SARS-CoV-2, it is important to understand the host response to this virus as this may guide the efforts in development towards novel therapeutics. Here, we offer the first in-depth characterization of the host transcriptional response to SARS-CoV-2 and other respiratory infections through in vitro, ex vivo, and in vivo model systems. Our data demonstrate the each virus elicits both core antiviral components as well as unique transcriptional footprints. Compared to the response to influenza A virus and respiratory syncytial virus, SARS-CoV-2 elicits a muted response that lacks robust induction of a subset of cytokines including the Type I and Type III interferons as well as a numerous chemokines. Taken together, these data suggest that the unique transcriptional signature of this virus may be responsible for the development of COVID-19.

Coronaviruses are a diverse group of single-stranded positive-sense RNA viruses infecting a wide range of vertebrate hosts 1 . These viruses are thought to generally cause mild upper respiratory tract illnesses in humans such as the common cold 2 .

However, in the past two decades, three highly pathogenic human coronaviruses have emerged from zoonotic viruses: severe acute respiratory syndrome-related coronavirus (SARS-CoV-1) infecting ~8,000 people worldwide with a case-fatality rate of ~10% in 2002-2003, Middle East respiratory syndrome-related coronavirus (MERS-CoV) infecting ~2,500 people with a case-fatality rate of ~36%, and now severe acute respiratory syndrome-related coronavirus 2 (SARS-CoV-2) which causes Coronavirus Disease-2019 (COVID-19) whose global mortality rate remains to be determined 3, 4 .

Infection with these highly pathogenic coronaviruses can result in acute respiratory distress syndrome (ARDS) and acute lung injury (ALI), often leading to reduction of lung function and even death 3 .

The current pandemic of COVID-19 represents an acute and rapidly developing global health crisis. In an effort to better understand the molecular basis of the disease and identify putative markers for COVID-19, we compared the transcriptional response of SARS-CoV-2 to that of seasonal influenza A virus (IAV) and human orthopneumovirus (commonly known as human respiratory syncytial virus (RSV)), two common recurring respiratory viruses. Comparing the transcriptional response in both primary human lung epithelium and transformed lung alveolar cells revealed that SARS-CoV-2 elicits a unique transcriptional response as compared to IAV and RSV. In a homogenous cell population, the transcriptional response to SARS-CoV-2 infection shows a significant lack of Type I and III interferon (IFN-I and IFN-III) expression as compared to IAV and RSV. Moreover, while a core number of cytokines that comprise the antiviral host defense are shared amongst all three viruses, an equal number are notably absent in response to SARS-CoV-2. Lastly, the only genes that appear to be unique to SARS-CoV-2 infection are secreted peptides implicated in diseases of the airways. This unique response to SARS-CoV-2 could also be recapitulated in vivo comparing influenza infection to COVID-19 in ferrets. Taken together, these data reveal that the interaction between virus and host as it relates to SARS-CoV-2 may be responsible for its unusual high morbidity and mortality.

In order to identify similarities and differences between the host response to SARS-CoV-2 and seasonal respiratory infections, we sought to elucidate the transcriptome of lung epithelial cells during viral infection. To achieve this, we infected human alveolar adenocarcinoma cells with SARS-CoV-2, RSV and IAV and performed mRNA-seq analysis (Supplementary Table 1) .

Despite undetectable levels of ACE2 and TMPRSS2, the putative receptor and protease for SARS-CoV-2 entry 5 , these cells were able to support viral replication, evidenced by full genome coverage in total RNA samples extracted from infected cells (Figure 1a and Extended Data Fig.1a ). Our analysis identified 120 differentially expressed genes (DEGs, qval < 0.05) of which more than 80% were up-regulated (Figure 1b and Supplementary Table 2 Figure 1c ). This response however is lacking the robust induction of antiviral genes commonly observed following IFN-I/-III signaling 6 . This muted response however does show a positive correlation in the host antiviral response overall as determined by gene set enrichment analysis (GSEA) (Extended Data Fig. 1b) . Taken together our analyses indicate that replication of SARS-CoV-2 in alveolar cells results in a limited antiviral response.

In an effort to determine if the apparent modest response to SARS-CoV-2 infection was the result of low receptor expression, a low multiplicity of infection (MOI), or due to the immortalized nature of the cell line, we next infected primary human bronchial epithelial (NHBE) cells (Fig. 2a) (Fig. 2b) . Taken together, these results suggest that the cellular response to SARS-CoV-2 is relatively uniform and void of IFN-I and IFN-III expression.

In an effort to compare the response to SARS-CoV-2 with other respiratory viruses, we next infected the lung alveolar carcinoma cell line with either RSV or seasonal IAV.

Bulk RNA sequencing of independent biological replicates revealed that the transcriptional response to SARS-CoV-2 infection is similar in magnitude to that of IAV Fig. 2) , resulting in the large overlap of DEGs between these two viruses, compared to IAV (Fig. 3a) . These responses do not reflect the replication dynamics of these viruses, illustrated by the amount of viral reads recovered from these infections (Fig. 3b ). In an effort to better illustrate the unique responses to SARS-CoV- IL6, CXCL11, IFNG, IL7, CXCL9, and CXCL10 ( Fig. 4b-c ). Taken together with both our in vitro and ex vivo data, which also implicated a diminished cytokine response to infection, these results suggest that the overall response to SARS-CoV-2 is relatively moderate when compared to other respiratory viruses.

In the present study we focus on defining the transcriptional response to SARS-CoV-2, influenza A virus, and respiratory syncytial virus. In general, these data find that the overall transcriptional footprint to infection is greatest for RSV and the lowest for IAV with SARS-CoV-2 representing an intermediate response. A probable explanation for the lack of a response to IAV in general is the potent antagonistic activity of the virus.

However, it is noteworthy that in vivo, the response to IAV exceeds that of SARS-CoV-2

where the activity of NS1 is unable to inhibit some of the pattern recognition receptors at play during a physiological response.

Despite the apparent muted induction of antiviral genes in response to SARS-CoV-2, we do observe a significant up regulation of well-characterized ISGs including: IFIT1-3, ISG15, DDX58, and others. Amongst the genes that are uniquely present when comparing SARS-CoV-2 to other respiratory viruses are EDN1 and TNFSF15 -two putative biomarkers that may contribute to COVID-19 pathology. However, it should be noted that we did not observe these genes in ferrets despite the fact that they could be detected in both cell culture models. Regardless of whether these two unique genes may serve as biomarkers, the overall signature of SARS-CoV-2, RSV, and IAV can serve as a broader map to developing novel diagnostic strategies.

The general induction of ISGs in response to SARS-CoV-2, albeit modest, is particularly interesting as we were not able to detect any reads mapping to IFN-I of IFN-III members

in contrast to what is observed for RSV or IAV. These data likely indicate that induction of IFN-I and IFN-III is very low but sufficient to induce at least a subset of ISGs in response to SARS-CoV-2. It is also noteworthy that we observe comparable replication between primary human bronchial epithelium and a transformed cell line despite the fact that we are unable to detect both the putative receptor (ACE2) or the required protease (TMPRSS2) in the latter cell line. While further work remains to ascertain how viral entry is mediated in this model system, we did note high expression of BSG which has also been suggested to compensate for ACE2 (Wang et al, BioRxv unpublished and Extended Data Fig. 1a) . Moreover, it should be noted that the viral replication profiles as deduced by both total RNA and polyA RNA sequencing were comparable between in vitro, ex vivo, and in vivo samples.

What makes the SARS-CoV-2 distinct from the RSV and IAV strains used in this study is the propensity to selectively induce morbidity and mortality in older populations 8 . The physiological basis for this morbidity is believed to be the selective death of Type II pneumocytes that results in both loss of air exchange and fluid leakage into the lungs 9, 10 . While it remains to be determined whether this moderate cell response to SARS-CoV-2 is responsible for the abnormally high lethality in the older populations, it does explain why the virus is generally asymptomatic in young people with healthy and robust immune systems 11 . Given the results here, it is tempting to speculate that perhaps in the aging population, the immune response itself is muted and thus prevents successful 

All procedures are described in our previous study 17 

Origin and evolution of pathogenic coronaviruses

Coronavirus pathogenesis and the emerging pathogen severe acute respiratory syndrome coronavirus

SARS and MERS: recent insights into emerging coronaviruses

A new coronavirus associated with human respiratory disease in China

SARS-CoV-2 Cell Entry Depends on ACE2 and TMPRSS2 and Is Blocked by a Clinically Proven Protease Inhibitor

A diverse range of gene products are effectors of the type I interferon antiviral response

Influenza A virus lacking the NS1 gene replicates in interferondeficient systems

Novel Coronavirus Pneumonia Emergency Response Epidemiology, T. [The epidemiological characteristics of an outbreak of

Pathological findings of COVID-19 associated with acute respiratory distress syndrome

Innate immune response of human alveolar type II cells infected with severe acute respiratory syndrome-coronavirus

SARS-CoV-2 Infection in Children

Glycosaminoglycan sulfation requirements for respiratory syncytial virus infection

Moderated estimation of fold change and dispersion for RNA-seq data with DESeq2

STRING v11: protein-protein association networks with increased coverage, supporting functional discovery in genome-wide experimental datasets

A penalized matrix decomposition, with applications to sparse principal components and canonical correlation analysis

Gene set enrichment analysis: a knowledge-based approach for interpreting genome-wide expression profiles

Sequential Immunization With Live-Attenuated Chimeric Hemagglutinin-Based Vaccines Confers Heterosubtypic Immunity Against Influenza A Viruses in a