key: cord-0982488-zutbsx7j
authors: Lokugamage, Kumari G.; Hage, Adam; de Vries, Maren; Valero-Jimenez, Ana M.; Schindewolf, Craig; Dittmann, Meike; Rajsbaum, Ricardo; Menachery, Vineet D.
title: Type I Interferon Susceptibility Distinguishes SARS-CoV-2 from SARS-CoV
date: 2020-11-09
journal: J Virol
DOI: 10.1128/jvi.01410-20
sha: 64f53306c89c24bd7be42e901beddea04b31551d
doc_id: 982488
cord_uid: zutbsx7j

SARS-CoV-2, a novel coronavirus (CoV) that causes COVID-19, has recently emerged causing an ongoing outbreak of viral pneumonia around the world. While distinct from SARS-CoV, both group 2B CoVs share similar genome organization, origins to bat CoVs, and an arsenal of immune antagonists. In this report, we evaluate type I interferon (IFN-I) sensitivity of SARS-CoV-2 relative to the original SARS-CoV. Our results indicate that while SARS-CoV-2 maintains similar viral replication to SARS-CoV, the novel CoV is much more sensitive to IFN-I. In Vero E6 and in Calu3 cells, SARS-CoV-2 is substantially attenuated in the context of IFN-I pretreatment, whereas SARS-CoV is not. In line with these findings, SARS-CoV-2 fails to counteract phosphorylation of STAT1 and expression of ISG proteins, while SARS-CoV is able to suppress both. Comparing SARS-CoV-2 and influenza A virus in human airway epithelial cultures, we observe the absence of IFN-I stimulation by SARS-CoV-2 alone but detect the failure to counteract STAT1 phosphorylation upon IFN-I pretreatment, resulting in near ablation of SARS-CoV-2 infection. Next, we evaluated IFN-I treatment postinfection and found that SARS-CoV-2 was sensitive even after establishing infection. Finally, we examined homology between SARS-CoV and SARS-CoV-2 in viral proteins shown to be interferon antagonists. The absence of an equivalent open reading frame 3b (ORF3b) and genetic differences versus ORF6 suggest that the two key IFN-I antagonists may not maintain equivalent function in SARS-CoV-2. Together, the results identify key differences in susceptibility to IFN-I responses between SARS-CoV and SARS-CoV-2 that may help inform disease progression, treatment options, and animal model development. IMPORTANCE With the ongoing outbreak of COVID-19, differences between SARS-CoV-2 and the original SARS-CoV could be leveraged to inform disease progression and eventual treatment options. In addition, these findings could have key implications for animal model development as well as further research into how SARS-CoV-2 modulates the type I IFN response early during infection.

the results were not statistically different. By 48 h, replication of both viruses had plateaued, and significant cytopathic effect was observed for both SARS-CoV and SARS-CoV-2 infections. Together, the results indicated that SARS-CoV and SARS-CoV-2 replicate with similar replication kinetics in Vero E6 cells.

We next evaluated the susceptibility of SARS-CoV-2 to IFN-I pretreatment. Treatment with IFN-I (recombinant gamma interferon [IFN-␥]) has been attempted as an antiviral approach for a wide variety of pathogens, including hepatitis B and C viruses, as well as HIV (20) . During both the SARS and the MERS-CoV outbreaks, IFN-I was used with limited effect (21, 22) . In this study, we pretreated Vero E6 cells with 1,000 U/ml of recombinant IFN-I (IFN-␣) 18 h prior to infection. Vero E6 cells lack the capacity to produce IFN-I but are able to respond to exogenous treatment (23) . After pretreatment with IFN-I, SARS-CoV infection has a modest reduction in viral titer of 1.5 log 10 PFU compared to untreated controls 24 h postinfection (Fig. 1A) . However, by 48 h, SARS-CoV has nearly equivalent viral yields as the untreated conditions (7.2 log 10 PFU versus 7.5 log 10 PFU). In contrast, SARS-CoV-2 shows a significant reduction in viral replication following IFN-I treatment. At both 24 and 48 h postinfection (hpi), SARS-CoV-2 showed massive 3-log 10 (24 hpi) and 4-log 10 (48 hpi) decreases in viral titer compared to control untreated cells. Together, the results demonstrate a clear sensitivity to a primed IFN-I response in SARS-CoV-2, which is not observed with SARS-CoV.

To explore differences in IFN-I antagonism between SARS-CoV and SARS-CoV-2, we examined both STAT1 activation and IFN stimulated gene (ISG) expression following IFN-I pretreatment and infection. Upon examining Vero E6 cell protein lysates, we found that IFN-I-treated cells infected with SARS-CoV-2 induced phosphorylated STAT1 by 48 h postinfection (Fig. 1B) . SARS-CoV had no evidence of STAT1 phosphorylation in either IFN-I-treated or untreated cells, illustrating robust control over IFN-I signaling pathways. In contrast, SARS-CoV-2 is unable to control signaling upon IFN-I treatment. Examining further, IFITM1, a known ISG (17) , had increased protein expression in the context of SARS-CoV-2 infection following IFN-I pretreatment compared to SARS-CoV under the same conditions (Fig. 1B) . Basal STAT1 levels are reduced during SARS-CoV infection relative to uninfected control and, to a lesser extent, during SARS-CoV-2 infection, likely due to the mRNA targeting activity of nonstructural protein 1 (NSP1) (24) . However, IFN-I treatment results in augmented protein levels for IFITM1 following SARS-CoV-2 infection compared to untreated SARS-CoV-2. In contrast, IFN-I-treated SARS-CoV had no significant increase in IFITM1 relative to control infection. Together, the STAT1 phosphorylation, ISG production, and viral protein levels indicate that SARS-CoV-2 lacks the same capacity to modulate the IFN-I stimulated response as the original SARS-CoV.

While capable of responding to exogenous IFN-I, Vero E6 cells lack the capacity to produce IFN-I following infection, which likely plays a role in supporting robust replication of a wide range of viruses (25) . To evaluate SARS-CoV-2 in an IFN-I responsive cell type, we infected Calu3 2B4 cells, a lung epithelial cell line sorted for ACE2 expression and previously used in coronavirus and influenza research (26) . Using an MOI of 1, we examined the viral replication kinetics of SARS-CoV-2 relative to SARS-CoV in Calu3 cells. We found that both SARS-CoV and SARS-CoV-2 replicate with similar overall kinetics, peaking 24 h postinfection ( Fig. 2A) . However, SARS-CoV-2 replication is slightly attenuated relative to SARS-CoV at 24 h postinfection (0.82-log 10 reduction). The attenuation in viral replication expands at 48 h (1.4-log 10 reduction), indicating a significant change in total viral titers between SARS-CoV and SARS-CoV-2. Notably, no similar attenuation was observed in untreated Vero E6 cells (Fig. 1A) , suggesting possible immune modulation of SARS-CoV-2 infection in the respiratory cell line due to differential sensitivity to secreted IFN-I during infection. We next evaluated the susceptibility of SARS-CoV-2 to IFN-I pretreatment in Calu3 cells. When pretreating cells with 1,000 U/ml of recombinant IFN-I 18 h prior to infection, SARS-CoV infection has a modest reduction in viral Each point on the line graph represents the group mean (n Ն 3). All error bars represent the SD. A two-tailed Student t test was used to determine P values (***, P Ͻ 0.001). (B) Calu3 cell protein lysates from IFN-I-treated and untreated cells were probed at 48 h postinfection by Western blotting for phosphorylated STAT1 (Y701), STAT1, IFITM1, SARS spike, and actin. (C) Western blot quantification for phosphorylated STAT1, total STAT1, and IFITM1. A two-tailed Student t test was used to determine P values (*, P Ͻ 0.05; **, P Ͻ 0.01; ***, P Ͻ 0.001).

titer of ϳ0.8 log 10 PFU compared to untreated control at both 24 and 48 h postinfection ( Fig. 2A ). Similar to Vero E6 cell results, SARS-CoV-2 shows a significant reduction in viral replication following IFN-I treatment in Calu3 cells. At both 24 and 48 h postinfection, SARS-CoV-2 showed 2.65-log 10 (24 hpi) and 2-log 10 (48 hpi) decreases in viral titer, respectively, compared to control untreated Calu3 cells. Together, the results demonstrate a clear sensitivity to a primed IFN-I response in SARS-CoV-2, which is not observed with SARS-CoV.

To further evaluate activation by IFN-I, we examined both STAT1 phosphorylation and ISG expression following infection of Calu3 2B4 cells at 48 h. Probing cell protein lysates, we found that untreated cells infected with SARS-CoV or SARS-CoV-2 induced phosphorylated STAT1 by 48 h postinfection ( Fig. 2B and C) . However, the level of STAT1 is markedly diminished in SARS-CoV infection compared to SARS-CoV-2, suggesting that the original epidemic strain disrupts the expression of the ISG. The diminished STAT1 levels in SARS-CoV correspond to robust spike expression in untreated cells. In contrast, SARS-CoV-2 spike is reduced compared to SARS-CoV, consistent with the lower replication observed in untreated Calu3 cells ( Fig. 2A) . The noticeable IFITM1 expression levels observed in untreated Calu3 cells infected with SARS-CoV-2 that were not observed in Vero E6 cells may signify higher baseline ISG levels in Calu3 cells during SARS-CoV-2 infection. Although IFITM1 is not significantly increased following infection or IFN-I treatment, its presence at baseline may contribute to SARS-CoV-2 attenuation relative to SARS-CoV.

Following IFN-I pretreatment in Calu3 cells, the differences in STAT1 activation and ISG induction are less prominent compared to the Vero E6 cell experiments; this is likely due to an already-induced IFN-I response to the virus infection. However, we still found a reduction in STAT1 phosphorylation and ISG induction in IFN-I-pretreated SARS-CoV infection versus SARS-CoV-2 infection. The total STAT1 levels were augmented in IFN-I-treated cells following SARS-CoV infection to values similar to control IFN-I-treated cells. In contrast, the SARS-CoV-2 STAT1 levels remained amplified relative to SARS-CoV but were similar to untreated SARS-CoV-2 infections. Although marginal, IFITM1 expression was slightly increased in IFN-I pretreatment SARS-CoV-2 relative to SARS-CoV. Importantly, the spike protein levels show a significant impact of IFN-I on SARS-CoV infection consistent with the titer. For SARS-CoV-2, the spike blot demonstrates the massive attenuation in the presence of IFN-I. Overall, the results in Calu3 cells are consistent with Vero E6 cell findings and indicate a significant sensitivity of SARS-CoV-2 to IFN-I pretreatment.

SARS-CoV-2 blocks IFN-I signaling in polarized human airway epithelial cultures. Having established IFN-I sensitivity in Vero E6 cells and Calu3 respiratory cell lines, we next sought to evaluate SARS-CoV-2 in polarized human airway epithelial cell (HAEC) cultures. These cultures provide both the complexity of different cell types and the architecture of the human respiratory epithelium. Previous work with SARS-CoV had already established its capacity to control the IFN-I response in human airway cultures (27) . Therefore, to examine the impact on overall level of IFN-I signaling in HAECs, we compared SARS-CoV-2 infection to influenza A virus (H1N1) infection, which is known to induce robust innate immune responses in these cultures. Briefly, HAECs were pretreated with IFN-␣ on the basolateral side prior to and during infection (Fig. 3A) . Apical washes were collected at multiple time points and analyzed for viral infectious titers by plaque assay. In addition, culture lysates at endpoint were examined for STAT1 phosphorylation by Western blotting. Both influenza A and SARS-CoV-2 viruses triggered different levels of immune stimulation (Fig. 3B ). Influenza A virus infection alone induced robust expression of both total and phosphorylated STAT1 by 48 h postinfection ( Fig. 3C to E). In stark contrast, SARS-CoV-2 infection alone indicated no increase in STAT1 or phosphorylated STAT1. IFN-I pretreatment resulted in robust induction of STAT1 and pSTAT1 in all cultures. Despite the distinct immune stimulation profiles upon infection alone, both viruses robustly replicated in the HAECs, with influenza A virus achieving higher viral yields compared to SARS-CoV-2 ( Fig. 3F and G). Pretreatment reduced influenza virus infection by ϳ1.5 log 10 at both 24 and 48 h postinfection, a finding consistent with previous reports (28) . In contrast, IFN-I pretreatment nearly ablated SARS-CoV-2 with 2-log 10 and 4-log 10 reductions in viral titers at 24 and 48 h, respectively, relative to untreated controls. Together, the results highlight the capacity of SARS-CoV-2 to prevent IFN-I signaling but also demonstrate the sensitivity of SARS-CoV-2 to IFN-I pretreatment.

Having established that SARS-CoV-2 cannot overcome a preinduced IFN-I state, we next evaluated the impact of IFN-I treatment postinfection. Vero E6 cells were infected with SARS-CoV or SARS-CoV-2 at an MOI of 0.01 and subsequently treated with 1,000 U/ml IFN-␣ at 4 h postinfection (Fig.  4A ). For SARS-CoV, treatment postinfection had no significant impact on viral yields at either 24 or 48 h postinfection, consistent with findings from pretreatment experiments. In contrast, SARS-CoV-2 had a substantial 2-log 10 reduction in viral titers at 24 h postinfection relative to the control. However, by 48 h, SARS-CoV-2 replication had achieved similar level to untreated controls, indicating that posttreatment had only a transient impact in Vero E6 cells. We subsequently performed the posttreatment experiment utilizing Calu3 respiratory cells at an MOI of 1 (Fig. 4B) . Similar to what we observed in Vero E6 cells, SARS-CoV infection posttreatment resulted in no significant changes at 24 h and a modest decrease (ϳ0.5 log 10 ) at 48 h, illustrating its resistance to IFN-I. In contrast, IFN-I posttreatment of SARS-CoV-2 infection resulted in a Ͼ2-log 10 reduction in viral titers at 24 h and expanded reduction at 48 h postinfection (ϳ4-log 10 reduction). The results indicate that in Calu3 cells, SARS-CoV-2 is unable to prevent inhibitory effects of IFN-I even after establishing initial infection. Overall, the pre-and posttreatment data highlight distinct differences between SARS-CoV and SARS-CoV-2 in modulation of IFN-I pathways.

Conservation of IFN-I antagonists across SARS-CoV and SARS-CoV-2. Previous work has established several key IFN-I antagonists in the SARS-CoV genome, including NSP1, NSP3, ORF3b, ORF6, and others (29) . Considering SARS-CoV-2's sensitivity to IFN-I, we next sought to evaluate conservation of IFN-I antagonist proteins encoded by SARS-CoV-2, SARS-CoV, and several bat SARS-like viruses, including WIV16-CoV (30), SHC014-CoV (31), and HKU3.1-CoV (32) . Using sequence analysis, we found several genetic differences to SARS-CoV-2 that potentially contribute to IFN-I sensitivity (Fig. 5) . For SARS-CoV structural proteins, including the nucleocapsid (N) and matrix (M) proteins, a high degree of sequence homology (Ͼ90% amino acid [aa] identity) suggests that their reported IFN-I antagonism is likely maintained in SARS-CoV-2 and other SARS-like viruses. Similarly, the ORF1ab polyprotein retains high sequence identity in SARS-CoV-2, and several known IFN-I antagonists contained within the polyprotein (NSP1, NSP7, and NSP14 to NSP16) are highly conserved relative to SARS-CoV. One notable exception is the large papain-like protease, NSP3, which is only 76% conserved between SARS-CoV and SARS-CoV-2. However, SARS-CoV-2 does maintain a deubiquitinating domain thought to confer IFN-I resistance (33) . For SARS-CoV ORF3b, a 154-aa protein known to antagonize IFN-I responses by blocking IRF3 phosphorylation (34) , sequence alignment indicates that the SARS-CoV-2 equivalent ORF3b contains a premature stop codon resulting in a truncated 24-aa protein. Similarly, HKU3.1-CoV also has a premature termination resulting in a predicted 39-aa protein. Both WIV16-CoV and SHC014-CoV, the bat viruses most closely related to SARS-CoV, encode longer a 114-aa truncated protein with Ͼ99% homology with SARS-CoV ORF3b, suggesting that IFN-I antagonism might be maintained in these specific group 2B CoV strains. In addition, SARS-CoV ORF6 has been shown to be an IFN-I antagonist that disrupts karyopherin-mediated transportation of transcription factors such as STAT1 (34, 35) . In contrast to ORF3b, all five surveyed group 2B CoVs maintain ORF6; however, SARS-CoV-2 had only 69% homology with SARS-CoV, while the other three group 2B bat CoVs had Ͼ90% conservation. Importantly, SARS-CoV-2 has a 2-aa truncation in its ORF6; previous work has found that alanine substitution in this C terminus of SARS-CoV ORF6 resulted in ablated antagonism (35) . Together, the sequence homology analysis suggests that differences in NSP3, ORF3b, and/or ORF6 may be key drivers of SARS-CoV-2 IFN-I susceptibility.

With the ongoing outbreak of COVID-19 caused by SARS-CoV-2, viral characterization remains a key factor in responding to the emergent novel virus. In this report, we describe differences in the IFN-I sensitivity between SARS-CoV-2 and the original SARS-CoV. While both viruses maintain similar replication in untreated Vero E6 cells, SARS-CoV-2 shows a significant decrease in viral protein and replication following IFN-I pretreatment. The decreased SARS-CoV-2 replication correlates with phosphorylation of STAT1 and augmented ISG expression largely absent following SARS-CoV infection despite IFN-I pretreatment. Infection of IFN-I competent Calu3 2B4 cells resulted in reduced SARS-CoV-2 replication relative to SARS-CoV; IFN-I pretreatment also corresponded to increased sensitivity for SARS-CoV-2 in Calu3 cells. However, SARS-CoV-2 fails to induce IFN-I pathways during infection of unprimed polarized HAEC compared to influenza A virus, suggesting a capacity to control IFN-I signaling. However, pretreatment nearly ablates replication of SARS-CoV-2, highlighting a robust sensitivity to IFN-I pathways once activated. Finally, SARS-CoV-2, unlike SARS-CoV, responded to postinfection IFN-I treatment with reduced viral yields in both Vero E6 and Calu3 cells. Analysis of viral proteins shows that SARS-CoV-2 has several genetic differences that potentially impact its capacity to modulate the IFN-I response, including loss of an equivalent ORF3b and a short truncation of ORF6, both known as IFN-I antagonists for SARS-CoV (34) . Together, our results suggest SARS-CoV and SARS-CoV-2 have differences in their ability to antagonize the IFN-I response once initiated and that this may have major implications for COVID-19 disease and treatment.

With a similar genome organization and disease symptoms in humans, the SARS-CoV-2 outbreak has drawn insights from the closely related SARS-CoV. However, the differences in sensitivity to IFN-I pretreatment illustrate a clear distinction between the two CoVs. Coupled with a novel furin cleavage site (36) , robust upper airway infection (8) , and transmission prior to symptomatic disease (37) , the differences between SARS-CoV and SARS-CoV-2 could prove important in disrupting the ongoing spread of COVID-19. For SARS-CoV, in vitro studies have consistently found that wild-type SARS-CoV is indifferent to IFN-I pretreatment (38, 39) . Similarly, in vivo SARS-CoV studies have found that the loss of IFN-I signaling had no significant impact on disease (40) , suggesting that this virus is not sensitive to the antiviral effects of IFN-I. However, more recent reports suggest that host genetic background may majorly influence this finding (41) . For SARS-CoV-2, our results suggest that IFN-I pretreatment produces a 3-to 4-log 10 decrease in viral titer and is consistent with reports from other groups (42) . This level of sensitivity is similar to MERS-CoV and suggests that the novel CoV lacks the same capacity to escape a primed IFN-I response as SARS-CoV (43, 44) . Notably, the sensitivity to IFN-I does not completely ablate viral replication; unlike SARS-CoV 2=-Omethyltransferase mutants (38) , SARS-CoV-2 is able to replicate to low, detectable levels even in the presence of IFN-I. This finding could help explain positive test results in patients with minimal symptoms and the range of disease observed. The results also correspond to severe COVID-19 disease in patients with impaired IFN-I responses (45) . In addition, while SARS-CoV-2 is sensitive to IFN-I pretreatment, both SARS-CoV and MERS-CoV employ effective means to disrupt virus recognition and downstream signaling until late during infection (26) . While SARS-CoV-2 may employ a similar mechanism early during infection, STAT1 phosphorylation and reduced viral replication are observed in IFN-I primed and posttreatment conditions, indicating that the novel CoV does not block IFN-I signaling as effectively as the original SARS-CoV.

For SARS-CoV-2, the sensitivity to IFN-I indicates a distinction from SARS-CoV and suggests differential host innate immune modulation between the viruses. The distinct ORF3b and truncation and/or differences in ORF6 could signal a reduced capacity of SARS-CoV-2 to interfere with IFN-I responses. For SARS-CoV ORF6, the N-terminal domain has been shown to have a clear role in its ability to disrupt karyopherinmediated STAT1 transport (35) ; in turn, the loss or reduction of ORF6 function for SARS-CoV-2 would likely render it much more susceptible to IFN-I pretreatment as activated STAT1 has the capacity to enter the nucleus and induce ISGs and the antiviral response. In these studies, we have found that following IFN-I pretreatment, STAT1 phosphorylation is induced following SARS-CoV-2 infection. The increase in ISG proteins (STAT1 and IFITM1) suggests that SARS-CoV-2 ORF6 does not effectively block nuclear transport as well as SARS-CoV ORF6. For SARS-CoV ORF3b, the viral protein has been shown to disrupt phosphorylation of IRF3, a key transcriptional factor in the production of IFN-I and the antiviral state (34) . While its mechanism of action is not clear, the ORF3b absence in SARS-CoV-2 infection likely impacts its ability to inhibit the IFN-I response and eventual STAT1 activation. Similarly, while the NSP3 deubiquitinating domain remains intact, SARS-CoV-2 has a 24-aa insertion upstream of this deubiquitinating domain that could potentially alter that function (33) . While other antagonists are maintained with high levels of conservation (Ͼ90%), single point mutations in key locations could modify function and contribute to increased IFN-I sensitivity.

Overall, the sequence analysis suggests that differences between SARS-CoV and SARS-CoV-2 viral proteins may drive attenuation in the context of IFN-I pretreatment.

The increased sensitivity of SARS-CoV-2 suggests utility in treatment using IFN-I. While IFN-I has been used in response to chronic viral infection (46) , previous examination of SARS-CoV cases found inconclusive effects for IFN-I treatment (47) . However, the findings from the SARS-CoV outbreak were complicated by combination therapy of IFN-I with other treatments, including ribavirin/steroids and lack of a regimented protocol. While IFN-I has been utilized to treat MERS-CoV-infected patients, no conclusive data yet exist to determine efficacy (48) . However, in vivo studies with MERS-CoV have found that early induction with IFN-I can be protective in mice (49) ; importantly, the same study found that late IFN-I activation can be detrimental for MERS-CoV disease (49) . Similarly, early reports have described treatments using IFN-I in combination for SARS-CoV-2 infection, though the efficacy of these treatments and the parameters of their use are not known (50) . Overall, sensitivity data suggest that IFN-I treatment may have utility for treating SARS-CoV-2 if the appropriate parameters can be determined. In addition, use of type III IFN, which is predicted to have utility in the respiratory tract, could offer another means for effective treatment for SARS-CoV-2.

In addition to treatment, the sensitivity to IFN-I may also have implications for animal model development. For SARS-CoV, mouse models that recapitulate human disease were developed through virus passage in immunocompetent mice (51) . Similarly, mouse models for MERS-CoV required adaptation in mice that had genetic modifications of their dipeptidyl-peptidase 4 (DPP4), the receptor for MERS-CoV (52, 53) . However, each of these MERS-CoV mouse models still retained full immune capacity. In contrast, SARS-CoV-2 sensitivity to IFN-I may signal the need to use an immune-deficient model to develop relevant disease. While initial work has suggested incompatibility to SARS-CoV-2 infection in mice based on receptor usage (8) , the IFN-I response may be a second major barrier that needs to be overcome. Similar to the emergent Zika virus outbreak, the use of IFN-I receptor knockout mice or IFN-I receptor blocking antibody may be necessary to develop a useful SARS-CoV-2 animal models for therapeutic testing (54) .

Overall, our results indicate that SARS-CoV-2 has a much higher sensitivity to type I IFN than the previously emergent SARS-CoV. This augmented type I IFN sensitivity is likely due to genetic differences in viral proteins between the two epidemic CoV strains. Moving forward, these data could provide important insights for both the treatment of SARS-CoV-2, as well as developing novel animal models of disease. In this ongoing outbreak, the results also highlight a distinction between the highly related viruses and suggest insights from SARS-CoV must be verified for SARS-CoV-2 infection and disease.

Viruses and cells. SARS-CoV-2 USA-WA1/2020 was provided by the World Reference Center for Emerging Viruses and Arboviruses (WRCEVA) or BEI Resources and was originally obtained from the U.S. Centers of Disease Control as described previously (55) . SARS-CoV-2 and mouse-adapted recombinant SARS-CoV (MA15) (51) were titrated and propagated on Vero E6 cells, grown in Dulbecco modified Eagle medium (DMEM) with 5% fetal bovine serum and 1% antibiotic/antimycotic (Gibco). Calu3 2B4 cells were grown in DMEM with 10% defined fetal bovine serum, 1% sodium pyruvate (Gibco), and 1% antibiotic/ antimycotic (Gibco). Standard plaque assays were used for SARS-CoV and SARS-CoV-2 (27, 56) . All experiments involving infectious virus were conducted at the University of Texas Medical Branch (UTMB; Galveston, TX) or New York University School of Medicine (New York, NY) in approved biosafety level 3 laboratories with routine medical monitoring of staff.

Infection and type I IFN pre-and posttreatment. Viral replication studies in Vero E6 and Calu3 2B4 cells were performed as previously described (38, 57) . Briefly, cells were washed two times with phosphate-buffered saline (PBS) and inoculated with SARS-CoV or SARS-CoV-2 at an MOI of 0.01 for 60 min at 37°C. After inoculation, the cells were washed three times, and fresh medium was added to signify time zero. Three or more biological replicates were harvested at each described time. No blinding was used in any sample collections, nor were samples randomized. For type I IFN pretreatment, the experiments were completed as previously described (38) . Briefly, Vero E6 cells were incubated with 1,000 U/ml of recombinant type I IFN-␣ (PBL Assay Sciences) 18 h prior to infection (38) . Cells were infected as described above, and type I IFN was not added back after infection.

Generation of polarized HAEC cultures. hTert-immortalized human normal airway tracheobronchial epithelial cells, BCi.NS1.1 (58) , were maintained in ExPlus growth media (StemCell). To generate HAEC, BCi.NS1.1 cells were plated (7.5E4 cells/well) on rat-tail collagen type 1-coated permeable transwell membrane supports (6.5 mm; Corning Inc.), immersed in ExPlus growth media in both the apical and the basal chambers. Upon reaching confluence, the medium in the apical chamber was removed (airlift), and the medium in the basal chamber was changed to Pneumacult ALI maintenance media (StemCell). Pneumacult ALI maintenance medium was changed every 2 days for approximately 6 weeks to form differentiated, polarized cultures that resemble in vivo pseudostratified mucociliary epithelium.

For interferon pretreatment of human airway epithelial cultures, 1,000 U/ml of universal recombinant IFN-␣ was added to the basolateral chamber 2 h prior to viral infections. For viral infections, cultures were washed apically with 50 l of prewarmed PBS three times for 30 min at 37°C. Virus inoculates (1.3E6 PFU for influenza A/California/07/2009 [H1N1] virus and 1.35E5 PFU for SARS-CoV-2 USA/WA/1/2020 in 50 l of PBS Mg/Ca) were added apically for 2 h. Inoculates were then removed, and cultures were washed three times with PBS; the third wash was collected, and then the cultures were incubated for the indicated times. Progeny virus was collected in apical washes with 50 l of PBS Mg/Ca for 30 min at 8, 24, and 48 hpi (endpoint). At the endpoint, the membranes were collected and lysed in 300 l of LDS lysis buffer (Life Technologies). Protein levels were measured by Western blotting. Briefly, 5-l portions of cell lysates were loaded into an SDS-PAGE gel and transferred to a nitrocellulose membrane. The membranes were blocked for 1 h at room temperature with 5% skim milk and then incubated overnight at 4°C with appropriate primary antibody dilution: anti-STAT1 (1:1,000; Cell Signaling), anti-pSTAT1 (1:1,000; Cell Signaling), or anti-beta actin (1:5,000; Invitrogen). Then, the membranes were incubated with the recommended dilution of horseradish peroxidase (HRP)-conjugated secondary antibody (goat anti-rabbit IgG HRP [1:10,000; Invitrogen] and goat antimouse IgG HRP [1:10,000; Invitrogen]) at room temperature for 1 h. A Super Signal West Dura kit (Thermo Scientific) was used for signal development according to manufacturer's recommendations, and the chemiluminescence was detected with a ChemiDoc imager (Bio-Rad).

Phylogenetic tree and sequence identity heat map. Heat maps were constructed from a set of representative group 2B coronaviruses using alignment data paired with neighbor-joining phylogenetic trees built in Geneious (v.9.1.5) and were derived from the following accession numbers: AY278741.1 (SARS-CoV Urbani), NC_045512.2 (SARS-CoV-2), DQ022305 (HKU3-1-CoV), KC881005.1 (RsSHC014-CoV), and KT444582.1 (WIV16-CoV). Sequence identity was visualized using EvolView (http://evolgenius.info/) and utilized SARS-CoV Urbani as the reference sequence (AY278741.1). The phylogenetic tree shows the degree of genetic similarity of SARS-CoV-2 and SARS-CoV across selected group 2B coronaviruses.

Immunoblot analysis and antibodies. Viral and host protein analysis were evaluated as previously described (55, 59) . Briefly, cell lysates were resolved on 7.5% Mini-Protean TGX SDS-PAGE gels and then transferred to polyvinylidene difluoride membranes using a Trans-Blot Turbo transfer system (Bio-Rad). Membranes were blocked with 5% (wt/vol) nonfat dry milk in TBST (TBS plus 0.1% [vol/vol] Tween 20) for 1 h and then probed with the indicated primary antibody in 3% (wt/vol) bovine serum albumin in TBST at 4°C overnight. After overnight incubation, the membranes were probed with the following secondary antibodies in 5% (wt/vol) nonfat dry milk in TBST for 1 h at room temperature: anti-rabbit or anti-mouse IgG-HRP-conjugated antibody from sheep (both 1:10,000; GE Healthcare). Proteins were visualized using ECL or SuperSignal West Femto chemiluminescence reagents (Pierce) and detected by autoradiography. The following primary antibodies were used: anti-pSTAT1 Y701 (1:1,000; Cell Signaling Technologies, 9171L), anti-STAT1 D1K9Y (1:1,000; Cell Signaling Technologies, 14994P), anti-IFITM1 (1:1,000; Invitrogen, PA5-20989), anti-SARS-CoV/CoV-2 Spike 1A9 (1:1,000; GeneTex, GTX632604), and anti-␤-actin (1:1,000; Abcam, ab8227).

Statistical analysis. All statistical comparisons in the manuscript involved the comparison between two groups, the SARS-CoV-infected or SARS-CoV-2-infected groups, under equivalent conditions. Thus, significant differences in viral titer were determined by an unpaired two-tailed Student t test.

Return of the coronavirus: 2019-nCoV

Coronaviridae Study Group of the International Committee on Taxonomy of Viruses. 2020. The species SEVERE acute respiratory syndrome-related coronavirus: classifying 2019-nCoV and naming it SARS-CoV-2

China Novel Coronavirus I Research Team. 2020. A novel coronavirus from patients with pneumonia in China

Clinical features of patients infected with 2019 novel coronavirus in Wuhan

Characteristics of and important lessons from the coronavirus disease 2019 (COVID-19) outbreak in China: summary of a report of 72,314 cases from the Chinese Center for Disease Control and Prevention

Evolution of the novel coronavirus from the ongoing Wuhan outbreak and modeling of its spike protein for risk of human transmission

Functional assessment of cell entry and receptor usage for SARS-CoV-2 and other lineage B betacoronaviruses

A pneumonia outbreak associated with a new coronavirus of probable bat origin

Prophylactic and therapeutic remdesivir (GS-5734) treatment in the rhesus macaque model of MERS-CoV infection

Remdesivir and chloroquine effectively inhibit the recently emerged novel coronavirus (2019-nCoV) in vitro

Comparative therapeutic efficacy of remdesivir and combination lopinavir, ritonavir, and interferon beta against MERS-CoV

Broad-spectrum antiviral GS-5734 inhibits both epidemic and zoonotic coronaviruses

Preliminary identification of potential vaccine targets for the COVID-19 coronavirus (SARS-CoV-2) based on SARS-CoV immunological studies

Innate immunity: the virtues of a nonclonal system of recognition

Toll-like receptors and RNA helicases: two parallel ways to trigger antiviral responses

TLR signaling

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

Mechanisms of type-I-and type-II-interferonmediated signalling

Viral evasion mechanisms of early antiviral responses involving regulation of ubiquitin pathways

Interferons: success in anti-viral immunotherapy

Middle East respiratory syndrome

From SARS to MERS, thrusting coronaviruses into the spotlight

Homozygous deletion of the ␣-and ␤1-interferon genes in human leukemia and derived cell lines

Coronavirus nonstructural protein 1: common and distinct functions in the regulation of host and viral gene expression

Defectiveness of interferon production and of rubella virus interference in a line of African Green monkey kidney cells (Vero)

Pathogenic influenza viruses and coronaviruses utilize similar and contrasting approaches to control interferon-stimulated gene responses

Release of severe acute respiratory syndrome coronavirus nuclear import block enhances host transcription in human lung cells

Pandemic H1N1 2009 influenza A virus induces weak cytokine responses in human macrophages and dendritic cells and is highly sensitive to the antiviral actions of interferons

SARS coronavirus pathogenesis: host innate immune responses and viral antagonism of interferon

Isolation and characterization of a novel bat coronavirus closely related to the direct progenitor of severe acute respiratory syndrome coronavirus

Isolation and characterization of a bat SARS-like coronavirus that uses the ACE2 receptor

Severe acute respiratory syndrome coronaviruslike virus in Chinese horseshoe bats

Deubiquitinating and interferon antagonism activities of coronavirus papain-like proteases

Severe acute respiratory syndrome coronavirus open reading frame (ORF) 3b, ORF 6, and nucleocapsid proteins function as interferon antagonists

Severe acute respiratory syndrome coronavirus ORF6 antagonizes STAT1 function by sequestering nuclear import factors on the rough endoplasmic reticulum/Golgi membrane

The spike glycoprotein of the new coronavirus 2019-nCoV contains a furin-like cleavage site absent in CoV of the same clade

Potential presymptomatic transmission of SARS-CoV-2

Attenuation and restoration of severe acute respiratory syndrome coronavirus mutant lacking 2=-Omethyltransferase activity

Interferon and cytokine responses to SARScoronavirus infection

SARS-CoV pathogenesis is regulated by a STAT1 dependent but a type I, II and III interferon receptor-independent mechanism

Dysregulated type I interferon and inflammatory monocyte-macrophage responses cause lethal pneumonia in SARS-CoVinfected mice

Inhibition of SARS-CoV-2 by type I and type III interferons

Middle East respiratory syndrome coronavirus nonstructural protein 16 is necessary for interferon resistance and viral pathogenesis

Inhibition of novel beta coronavirus replication by a combination of interferon-␣2b and ribavirin

Impaired type I interferon activity and inflammatory responses in severe COVID-19 patients

The use of interferon-alpha in virus infections

SARS: systematic review of treatment effects

SARS and MERS: recent insights into emerging coronaviruses

IFN-I response timing relative to virus replication determines MERS coronavirus infection outcomes

Potential rapid diagnostics, vaccine and therapeutics for 2019 novel coronavirus (2019-nCoV): a systematic review

A mouse-adapted SARS-coronavirus causes disease and mortality in BALB/c mice

A mouse model for MERS coronavirus induced acute respiratory distress syndrome

Mouse-adapted MERS coronavirus causes lethal lung disease in human DPP4 knockin mice

A mouse model of Zika virus pathogenesis

Severe acute respiratory syndrome coronavirus 2 from patient with 2019 novel coronavirus disease, United States

Cell host response to infection with novel human coronavirus EMC predicts potential antivirals and important differences with SARS coronavirus

Pathways of cross-species transmission of synthetically reconstructed zoonotic severe acute respiratory syndrome coronavirus

Generation of a human airway epithelium derived basal cell line with multipotent differentiation capacity

VAMP8 contributes to the TRIM6-mediated type I interferon antiviral response during West Nile virus infection

This research was supported by NIH grants from the National Institute on Aging