key: cord-0964663-v5ke0lc2
authors: Nguyen, Long C.; Renner, David M.; Silva, Diane; Yang, Dongbo; Medina, Kaeri M.; Nicolaescu, Vlad; Gula, Haley; Drayman, Nir; Valdespino, Andrea; Mohamed, Adil; Dann, Christopher; Wannemo, Kristin; Robinson-Mailman, Lydia; Gonzalez, Alan; Stock, Letícia; Cao, Mengrui; Qiao, Zeyu; Moellering, Raymond E.; Tay, Savas; Randall, Glenn; Beers, Michael F.; Rosner, Marsha R.; Oakes, Scott A.; Weiss, Susan R.
title: SARS-CoV-2 diverges from other betacoronaviruses in only partially activating the IRE1α/XBP1 ER stress pathway in human lung-derived cells
date: 2021-12-30
journal: bioRxiv
DOI: 10.1101/2021.12.30.474519
sha: 1155f84dee174d5b2a41d8e01ac10aec31554680
doc_id: 964663
cord_uid: v5ke0lc2

Despite the efficacy of vaccines, severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2) has killed over 5 million individuals worldwide and continues to spread in countries where the vaccines are not yet widely available or its citizens are hesitant to become vaccinated. Therefore, it is critical to unravel the molecular mechanisms that allow SARS-CoV-2 and other coronaviruses to infect and overtake the host machinery of human cells. Coronavirus replication triggers endoplasmic reticulum (ER) stress and activation of the unfolded protein response (UPR), a key host cell pathway widely believed essential for viral replication. We examined the activation status and requirement of the master UPR sensor IRE1α kinase/RNase and its downstream transcription factor effector XBP1s, which is processed through an IRE1α-mediated mRNA splicing event, in human lung-derived cells infected with betacoronaviruses. We found human respiratory coronavirus OC43 (HCoV-OC43), Middle East respiratory syndrome coronavirus (MERS-CoV), and the murine coronavirus (MHV) all induce ER stress and strongly trigger the kinase and RNase activities of IRE1α as well as XBP1 splicing. In contrast, SARS-CoV-2 only partially activates IRE1α whereby it autophosphorylates, but its RNase fails to splice XBP1. Moreover, IRE1α was dispensable for optimal replication in human cells for all coronaviruses tested. Our findings demonstrate that IRE1α activation status differs upon infection with distinct betacoronaviruses and is not essential for efficient replication of any of them. Our data suggest that SARS-CoV-2 actively inhibits the RNase of autophosphorylated IRE1α through an unknown mechanism, perhaps as a strategy to eliminate detection by the host immune system. Importance SARS-CoV-2 is the third lethal respiratory coronaviruses after MERS-CoV and SARS-CoV to emerge this century, causing millions of deaths world-wide. Other common coronaviruses such as OC43 cause less severe respiratory disease. Thus, it is imperative to understand the similarities and differences among these viruses in how each interacts with host cells. We focused here on the inositol-requiring enzyme 1α (IRE1α) pathway, part of the host unfolded protein response to virus-induced stress. We found that while MERS-CoV and OC43 fully activate the IRE1α kinase and RNase activites, SARS-CoV-2 only partially activates IRE1α, promoting its kinase activity but not RNase activity. We propose that SARS-CoV-2 prevents IRE1α RNase activation as a strategy to eliminate detection by the host immune system.

Despite the efficacy of vaccines, severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2) has killed over 5 million individuals worldwide and continues to spread in countries where the vaccines are not yet widely available or its citizens are hesitant to become vaccinated. Therefore, it is critical to unravel the molecular mechanisms that allow SARS-CoV-2 and other coronaviruses to infect and overtake the host machinery of human cells. Coronavirus replication triggers endoplasmic reticulum (ER) stress and activation of the unfolded protein response (UPR), a key host cell pathway widely believed essential for viral replication. We examined the activation status and requirement of the master UPR sensor IRE1a kinase/RNase and its downstream transcription factor effector XBP1s, which is processed through an IRE1a-mediated mRNA splicing event, in human lung-derived cells infected with betacoronaviruses. We found human respiratory coronavirus OC43 (HCoV-OC43), Middle East respiratory syndrome coronavirus (MERS-CoV), and the murine coronavirus (MHV) all induce ER stress and strongly trigger the kinase and RNase activities of IRE1a as well as XBP1 splicing. In contrast, SARS-CoV-2 only partially activates IRE1a whereby it autophosphorylates, but its RNase fails to splice XBP1. Moreover, IRE1a was dispensable for optimal replication in human cells for all coronaviruses tested. Our findings demonstrate that IRE1a activation status differs upon infection with distinct betacoronaviruses and is not essential for efficient replication of any of them. Our data suggest that SARS-CoV-2 actively inhibits the RNase of autophosphorylated IRE1a through an unknown mechanism, perhaps as a strategy to eliminate detection by the host immune system.

SARS-CoV-2 is the third lethal respiratory coronaviruses after MERS-CoV and SARS-CoV to emerge this century, causing millions of deaths world-wide. Other common coronaviruses such as OC43 cause less severe respiratory disease. Thus, it is imperative to understand the similarities and differences among these viruses in how each interacts with host cells. We focused here on the inositol-requiring enzyme 1α (IRE1α) pathway, part of the host unfolded protein response to virus-induced stress. We found that while MERS-CoV and OC43 fully activate the IRE1α kinase and RNase activites, SARS-CoV-2 only partially activates IRE1α, promoting its kinase activity but not RNase activity. We propose that SARS-CoV-2 prevents IRE1α RNase activation as a strategy to eliminate detection by the host immune system. stress." The presence of misfolded proteins in the ER are sensed by three transmembrane sentinel proteins--activating transcription factor 6 (ATF6), PKR-like ER kinase (PERK), and inositol-requiring enzyme 1α (IRE1α)--which trigger an intracellular signaling pathway called the unfolded protein response (UPR). In an effort to restore proteostasis, activation of these sensors induces transcription factors that turn on genes encoding chaperones, oxidoreductases, and ERassociated decay (ERAD) components (8) . The UPR also inhibits Cap-dependent translation, thus decreasing the load on the ER and giving it extra time to fold proteins already in production (9, 10) . If successful, these adaptive UPR programs restore ER homeostasis.

The most ancient UPR pathway is controlled by IRE1a-an ER transmembrane bifunctional kinase/endoribonuclease (RNase) that employs auto-phosphorylation to control its catalytic RNase function (11, 12) . In response to ER stress, IRE1a undergoes auto-phosphorylation and dimerization to allosterically activate its RNase domain to excise a 26nt non-conventional intron in XBP1 mRNA; re-ligation of spliced XBP1 shifts the open reading frame, and its translation produces the homeostatic transcription factor XBP1s (s=spliced) (13, 14) . Once synthesized, XBP1s upregulates genes that expand the ER and its protein folding machinery (15) . IRE1a can additionally lead to apoptosis and inflammation via JUN N-terminal kinase (JNK) and p38 mitogen-activated protein kinase (MAPK) signaling (16) . Prolonged ER stress can induce regulated IRE1-dependent decay (RIDD), promoting the cleavage of additional targets beyond XBP1 mRNA, such as secretory protein and ER-localized mRNAs (17) . In the short term RIDD may promote adaptation through further reducing translation and protein burden on the ER; however, prolonged RIDD leads to the depletion of vital ER resident enzymes and structural components to exacerbate ER stress and hasten cell death (11, 18) .

There is a large body of evidence that viral replication of mammalian cells can trigger ER stress and UPR activation in infected cells (19) , and numerous studies report that the UPR is activated upon infection of host cells by coronavirus family members (6, 7, (20) (21) (22) (23) . Coronaviruses induce stress in ER in several ways. First, conserved replicase encoded, non structural proteins nsp3, nsp4 and nps6 are embedded into the ER membrane, and along with unknown host factors, promote membrane curvature to form double memebrane vesicles (DMVs), the site of viral replication and transcription centers. In addition to taking over the ER, coronaviruses further condition infected cells by shifting translation away from host mRNAs and instead to viral mRNAs.

Translation of viral mRNAs casues the ER, already depleted of membranes, to be flooded with heavily glycosylated viral structural proteins (e.g., spike (S), membrane (M) and envelope (E)), challenging the organelle's folding capacity and overall integrity. Finally, cell membranes are depleted as enveloped virus particles are asembled into new virions in the ER-Golgi intermediate compartment before budding from the infected cell (1). Thus, coronaviruses as well as other enveloped viruses promote a massive ER expansion and modification necessary to replicate their genomes, transcribe mRNAs, and finally to process and package their protein products into viral particles.

In this study, we have compared the activation status and requirement of the IRE1a/XBP1 arm of the UPR in well-characterized human lung epithelial cell lines and in induced pluripotent stem cell (iPSC)-dervied type II alveolar (iAT2) cells. following infection with four betacoronaviruses representing three distinct lineages. We find that infection with MERS-CoV, HCoV-OC43 and MHV leads to phosphorylation of IRE1a and the consequent production of spliced XBP1 transcription factor. Surprisingly, while we observed phosphorylation of IRE1a in SARS-CoV-2 infected cells, there was notable absence of XBP1s, suggesting SARS-CoV-2 inhibits downstream signaling of the IRE1a/XBP1 arm of the UPR.

To determine whether betacoronaviruses activate IRE1a, we first examined the level of phosphorylated IRE1α after viral infection of the A549 human lung carcinoma cell line. We used A549 cells stably expressing the following receptors to facilitate optimal entry for each of the viruses: carcinoembryonic antigen cell adhesion molecule (CEACAM)1a or MHVR (MHV), dipeptidyl peptidase DPP4 (MERS-CoV), or angiotensin converting enzyme (ACE)2 (SARS-CoV-2). HCoV-OC43 can infect parental A549 or cells expressing ACE2 (3). Consistent with previous reports that embeco lineage coronaviruses MHV (20, 24) and OC43 (25) induce ER stress, we observed a significant increase in phospho-IRE1a (p-IRE1a) during infection by either HCoV-OC43 (24 or 48hpi) or MHV (24hpi) (Figure 2A-C) . To confirm the specificity of the p-IRE1α band, we pretreated cells prior to infection with KIRA8, a highly selective kinase inhibitor of IRE1α inhibitor known to inhibit both autophosphorylation and consequently RNase activity. As expected, KIRA8 significantly inhibited the induction of p-IRE1α by OC43 and MHV ( Figure 2A&C ). Thapsigargin (Tg) and tunicamycin (Tm), both inducers of ER stress, were used as further controls ( Figure 2B ,D&E). Robust induction of p-IRE1α was observed with 1 hour of Tg (1μM) treatment; while no activation of p-IRE1α was observed after 8 hours of treatment with Tm (1μg/ mL), consistent with the negative feedback regulation observed with extended Tm treatment (26) . We also observed robust phosphorylation of IRE1a in A549-DDP4 cells and A549-ACE2 cells infected by MERS-CoV and SARS-CoV-2, respectively at 24 and 48 hpi ( Figure 2D -F and S1A&B). As with OC43 and MHV, IRE1α phosphorylation during SARS-CoV-2 infection was inhibited by KIRA8 ( Fig 2F) . These results are not limited to a single cell type as we observed similar induction of p-IRE1a in Calu-3 cells, which can be productively infected with both MERS-CoV or SARS-CoV-2 ( Figure 2G ). These results demonstrate that MERS-CoV and SARS-CoV-2, like OC43 and MHV, activate the host IRE1α kinase after infection. Figure 3D&H ). In agreement with the RNA results, OC43, but not SARS-CoV-2, infection induced XBP1s protein levels ( Figure 3I&J ).

To determine how different coronaviruses impact the UPR at the transcriptional level, we performed RNA-sequencing of A549-DPP4 cells infected with MERS-CoV for 24 and 36 hours and compared the results to published RNA-seq data sets (24, 27) To support the results of the gel electrophoresis splicing assays for XBP1 mRNA that distinguished SARS-CoV-2 infection from that of the other betacoronaviruses (Figure 3 ), we further utilized the RNA sequencing results to quantitatively measure XBP1 mRNA splicing by these coronaviruses. Through RNA-seq, we visualized both the unspliced and spliced XBP1 mRNA reads based on whether they contain the 26 nucleotide non-conventional intron that is removed as a result of RNase activity of IRE1a ( Figure 4B&C ). MERS-CoV infection resulted in significant XBP1 mRNA splicing, in contrast with no difference detected in SARS-CoV-2 infected versus mock-infected cells ( Figure 4B&C ). We further quantified total XBP1 spliced vs unspliced reads, which consistently showed a substantial increase in the percent expression of the XBP1s reads when normalized to total XBP1 reads for MERS-CoV at both 24 and 36 hours post-infection but not for SARS-CoV-2 infected cells ( Figure 4D&E ), consistent with significant upregulation of DNAJB9 and total XBP1 during infection with MERS-CoV but not SARS-CoV-2 ( Figure 4F -I).

To confirm our results in a more physiologically relevant cell, we obtained iPSC-dervied type II alveolar (iAT2) cells. We employed the SPC2 line, which expresses tdTomato from the surfactant protein-C (SFTPC) locus as an AT2 marker, that we have previously used to characterize innate immune responses to SARS-CoV-2 infection (3). Type II alveolar cells are a major target during both MERS-CoV and SARS-CoV-2 infection in humans, and their destruction may be a contributing factor to lung pathogenesis in severe cases (28, 29).

Both MERS-CoV and SARS-CoV-2 replicate in these cells and release infectious virus as quantified by plaque assay ( Figure 5A ). Notably, MERS-CoV replicated to higher titers than SARS-CoV-2 in these lung derived cells. This complements our previous findings that SARS-CoV-2 replicates more efficiently than MERS-CoV in upper respiratory derived primary nasal cells

(3) and may reflect that MERS-CoV is better adapted to replicating within the lower respiratory tract while SARS-CoV-2 replicates more efficiently in the upper airway. Despite this difference in replication, both viruses were observed to induce p-IRE1α over the course of infection ( Figure   5B ). In agreement with our results in A549 and Calu-3 cells, SARS-CoV-2 failed to induce XBP1 splicing in iAT2 cells, as measured by RT-qPCR ( Figure 5C ). Contrasting this, MERS-CoV induced XBP1 splicing, albeit to a lower extent than in immortalized cell lines. Lastly, we visualized XBP1 splicing using RT-PCR and agarose gel electrophoresis ( Figure 5D ). Again, our data indicate that SARS-CoV-2 fails to induce XBP1 splicing at either 24 or 48hpi in iAT2 cells, despite inducing p-IRE1α. MERS-CoV, however, induced increasing XBP1 splicing over the course of infection, matching the results in A549 and Calu-3 cells (Figures 2 and 3) . Overall, these results indicate that both SARS-CoV-2 and MERS-CoV induce ER stress as evidenced by IRE1α phosphorylation during infection of primary iAT2 cells, but only MERS-CoV induces the downstream effects of active IRE1α RNase.

We then tested whether SARS-CoV-2 actively inhibits splicing of XBP1 induced by glycosylation inhibitor tunicamycin (TM), a common agent used to chemically induce ER stress. To do so, A549-ACE2 cells were either mock infected or infected with SARS-CoV-2 or OC43 for 24 hours and then treated with TM for 6 hours prior to analysis. Interestingly, while SARS-CoV-2 infection did not completely prevent XBP1 splicing induced by TM, it led to significantly lower XBP1 splicing levels compared with mock infected cells ( Figure 6A ). In contrast, OC43 increased further XBP1 splicing at all used concentration of TM ( Figure 6B ). This result suggests that SARS-CoV-2 actively inhibits RNase activation IRE1α.

Given the presumed importance of IRE1a/XBP1s to expand the ER and maintain protein folding during viral replication, and the interesting differences we observed between SARS-CoV-2 and the other betacoronaviruses, we next explored the consequences for its inhibition on the replication of each. To determine whether IRE1α activity is required for replication and propagation of MHV, HCoV-OC43, MERS-CoV or SARS-CoV-2, we utilized CRISPR/Cas9 gene editing to knock out IRE1α in A549 cell lines expressing receptors for each coronavirus ( Figure   S3 A-F). Surprisingly, we did not observe any significant differences in the capability of all tested coronaviruses to replicate in cells lacking IRE1α ( Figure 6C -F). These results suggest IRE1a is not essential for coronavirus replication. Since SARS-CoV-2 does not lead to IRE1α-mediated XBP1 splicing, we also tested replication of SARS-CoV-2 and OC43 in XBP1 KO cells ( Figure   6C&D and S3G). Consistently, there was no detectable effect of XBP1 KO on SARS-CoV-2 or OC43 replication in A549-ACE2. Together, these results demonstrate that none of the coronaviruses tested require the activation IRE1a/XBP1 pathway for replication. MERS-CoV was observed to also induce IRE1a/XBP1 activation in iAT2 cells ( Figure 5 ). In contrast, while SARS-CoV-2 also promoted autophosphorylation of IRE1a, there was no evidence of XBP1s, indicating that the pathway was only partially activated and suggesting that the IRE1a kinase activity was active while the RNase activity was not. The differential splicing of XBP1 during SARS-CoV-2 and MERS-CoV infection was also observed in iPSC-derived AT2 cells, confirming the results in a more physiologically relevant system ( Figure 5 ). The difference among these viruses is suprising as the viral replicase encoded proteins nsp3, 4 and 6 required for formation of replication factories in the DMV as well as three membrane associated glycoproteins that flood the ER would be expected to induce similar stress on the ER and lead to UPR activation.

Indeed, our data suggest that that SARS-CoV-2 actively prevents XBP1 splicing ( Figure 6A ,B &G). Consistent with this idea, a recombinant SARS-CoV lacking the E protein (rSARS-CoV-ΔE) was reported to induce more XBP1 splicing as well as induction of UPR genes compared to parental wild type virus (32).

Multiple coronavirus family members, including HCoV-OC43 (25) , MHV (20) , IBV (33) , have been reported to activate IRE1a [reviewed in (6) ]. In the few instances where its functional role in coronavirus infection has been tested, IRE1a has been reported to be critical for viral replication and cytotoxicity. For example, knockdown and overexpression experiments demonstrated that IRE1α protects infected cells from IBV-induced apoptosis (33) . In addition, over-expression of spike proteins from MHV, SARS-CoV, SARS-CoV-2, and HKU1 all induce ER stress and UPR activation in cultured cells (34) . Other SARS-CoV-encoded accessory proteins, including proteins 3a, 6 and 8ab, as well as SARS-CoV-2 Orf8 protein (22, 35) , have also been reported to induce ER stress when over-expressed in cells (36) (37) (38) . However, caution must be taken in interpreting effects of overexpressed proteins and a role for these in infection has not been confirmed.

Most current evidence suggests that virus-induced hyperactivation of IRE1α/XBP1 is critical to massively increase ER capacity in host cells in order to accommodate the inordinate stress of increased viral protein production. Indeed, inhibition of IRE1α has been reported to be detrimental to influenza A virus replication (39) . Of the three arms of the UPR, IRE1α is the most important for signaling ER expansion. IRE1α-mediated XBP1 splicing results in production of the Xbp1s transcription factor whose target genes include many involved in lipid biogenesis and vesicular trafficking (40) . Forced expression of Xbp1s is sufficient to induce synthesis of phosphatidylcholine, the primary phospholipid of the ER membrane, and massively increase the surface area and volume of the rough ER (41) . As such, activation of IRE1α/Xbp1s causes the ER to expand in size and function. Moreover, given its extensive communication with other key signaling pathways (mitogen-activated protein kinase pathways, inflammatory responses, apoptosis, autophagy and innate immunity), the UPR may coordinate multiple anti-viral responses and constitute a critical coronavirus-host interaction (6) . In summary, hyperactivation of the UPR, and its IRE1α-/XBP1 branch in particular, is often considered a general feature of infection by coronaviruses.

To investigate the importance of IRE1α for CoV replication, we evaluated replication of each of the betacoronaviruses in IRE1α KO A549 cells. In contrast to influenza, all of the betacoronaviruses examined were able to replicate efficiently in the absence of IRE1α signaling.

This raises interesting possibilities for the role of IRE1α during CoV infection. As previously stated, IRE1α can produce both cytoprotective (through XBP1s) and destructive responses (via RIDD and JNK/p38 signaling) depending on the extent of the encountered stress. It seems likely that CoV infection would induce extensive and prolonged ER stress, which may push IRE1α beyond the initial pro-recovery responses and towards a pro-apoptitic response. Indeed, our data reveal that, at least with MERS-CoV and SARS-CoV-2 infection, IRE1α phosphorylation is readily detectable by 24hpi and remains steady throughout the course of infection ( Figure S1A&B ).

Additionally, unlike what has been observed with chemically-induced ER stress (26, 42), IRE1α phosphorylation does not appear to attenuate at any point during CoV infection, again suggesting a hyperactive and destructive outcome. As stated above, destruction of cells, in particular AT2 cells in the lung, may contribute to pathogenesis during CoV infection. However, SARS-CoV-2 appears to limit the downstream consequences of IRE1α activation, most notably XBP1 splicing via its RNase activity and thus may be protected from this destructive phenotype. MERS-CoV may induce apoptosis redundantly in the UPR, as it has been reported that MERS-CoV induces and benefits from apoptisis mediated by the PERK arm of the UPR (21, 43) . Overall, despite the lack of apparent virus replication defects with IRE1α deficinecy, further characterization of the repertoire of betacoronavirus induced IRE1α signaling is warranted, including contributions to cytokine production, apoptosis, and pro-inflammatory responses. While we have initially investigated this pathway from the perspective of the impact on virus replication, future studies will examine effects of IRE1α activation on the host, including inflammation and cell death through the JNK and p38 MAPK signaling scaffolded by IRE1α (16) 

SARS-CoV-2 and MERS-CoV infections and plaque assays were performed as previously described (1, 5). In brief, A549 cells were seeded at 3x10 5 cells per well in a 12-well plate for infections. Calu-3 cells were seeded similarly onto rat tail collagen type I coated plates (Corning #356500). Cells were washed once with PBS before infecting with virus diluted in serum free media -RPMI for A549 cells or DMEM for Calu-3 cells. Virus was absorbed for 1 hour (A549

At the indicated timepoints, 200μL of media was collected to quantify released virus by plaque assay, and stored at -80 degrees Celsius. Infections for MHV growth curves were performed similarly in BSL-2 conditions. For OC43 infections, similar infection conditions and media were used, however virus was absorbed and the infections incubated at 33 degrees Celsius rather than 37 degrees.

Plaque assays were performed using VeroE6 cells for SARS-CoV-2 and OC43; VeroCCL81 cells for MERS-CoV; and L2 cells for MHV. SARS-CoV-2 and MERS-CoV plaque assays were performed in 12-well plates at 37 degrees Celsius. OC43 and MHV plaque assays were performed in 6-well plates at 33 degrees Celsius and 37 degrees Celsius, respectively. In all cases, virus was absorbed onto cells for one hour at the indicated temperatures before overlay was added. For SARS-CoV-2, MERS-CoV, and OC43 plaque assays, a liquid overlay was used (DMEM with 2% FBS, 1x sodium pyruvate, and 0.1% agarose). A solid overlay was used for MHV plaque assays (DMEM plus 2% FBS, 1x HEPES, 1x glutamine, 1x Fungizone, and 0.7% agarose).

Cell monolayers were fixed with 4% paraformaldehyde and stained with 1% crystal violet after the following incubation times: SARS-CoV-2 and MERS-CoV, 3 days; OC43, 5 days; MHV, 2 days.

All plaque assays were performed in biological triplicate and technical duplicate.

KIRA8 was purchased at >98% purity from Chemveda Life Sciences India Pvt. Ltd. For use in tissue culture, KIRA8 stock solution was prepared by dissolving in DMSO. Tunicamycin(cat. #T7765) and thapsigargin (cat. #T9033) were purchased at >98% purity from Sigma. For use in tissue culture, Tunicamycin stock solution was prepared by dissolving in DMSO.

Cells were washed once with ice-cold PBS and lysates harvested at the indicated times post infection with lysis buffer (1% NP-40, 2mM EDTA, 10% glycerol, 150mM NaCl, 50mM Tris HCl, pH 8.0) supplemented with protease inhibitors (Roche complete mini EDTA-free protease inhibitor) and phosphatase inhibitors (Roche PhosStop easy pack). After 5 minutes, lysates were incubated on ice for 20 minutes, centrifuged for 20 minutes at 4°C and supernatants mixed 3:1 with 4x Laemmli sample buffer (Bio-rad 1610747). Samples were heated at 95°C for 5 minutes, then separated on SDS-PAGE, and transferred to PVDF membranes. Blots were blocked with 5% nonfat milk or 5% BSA and probed with antibodies (table below) diluted in the same block buffer. Primary antibodies were incubated overnight at 4°C or for 1 hour at room temperature. All secondary antibody incubation steps were done for 1 hour at room temperature. Blots were visualized using Thermo Scientific SuperSignal chemiluminescent substrates (Cat #: 34095 or 34080). (53) . The resulting BAM files were counted by featureCounts 1.6.4 to count the number of reads for each gene (54) . Differential expression between mock, 24hpi, and 36hpi experimental conditions were analyzed using the raw gene counts files by DESeq2 1. 22 .1 (55) . A PCA plot of RNA-seq samples and a normalized gene expression matrix were also generated by DESeq2. 

All statistical analyses and plotting of data were performed using GraphPad Prism software. RT-qPCR data were analyzed by Student t test. Plaque assay data were analyzed by two-way ANOVA with multiple comparisons correction. Displayed significance is determined by p-value (P), where * = P < 0.05; ** = P < 0.01; *** = P < 0.001; **** = P < 0.0001; ns = not significant. 

RT-qPCR was used to quantify the relative expression of the spliced version of XBP1 (XBP1s) by using specific pairs of primers for human alternatively spliced XBP1 and total XBP1 (primer sequences are described above) as previously described (58) . The relative percentage of alternative splicing of XBP1 (%XBP1s) was indicated by calculating the ratio of signals between XBP1s and total XBP1. were quantified by RT-qPCR, calculated, and displayed as described above. Values are means ± SD (error bars). Statistical significance (infected compared to mock) were determined using two-tailed, paired Student t-test. Displayed significance is determined by p-value (P), where * = P < 0.05; ** = P < 0.01; *** = P < 0.001; **** = P < 0.0001; ns = not significant. (D) RT-PCR was performed using extracted RNA and primers crossing the XBP1 splicing site. The product was run an on an agarose gel to visualize XBP1 splicing. Tunicamycin treatment (1μg/mL for 6 hours) was used as a positive control for RT-(q)PCR, while DMSO treatment served as a vehicle control.

Data shown are from one representative experiment from at least two independent experiments. control A549-ACE2 were treated with tunicamycin (Tm, 1μg/mL) for 8 hours. Total RNA was harvested and amplied for XBP1. XBP1 cDNA product was assayed on agarose gel to visualize

Antagonism of dsRNA-Induced Innate Immune Pathways by NS4a and NS4b Accessory Proteins during MERS Coronavirus Infection

Antagonism of the interferon-induced OAS-RNase L pathway by murine coronavirus ns2 protein is required for virus replication and liver pathology

Coronavirus-induced ER stress response and its involvement in regulation of coronavirus-host interactions

Coronavirus infection, ER stress, apoptosis and innate immunity

Functional and genomic analyses reveal an essential coordination between the unfolded protein response and ER-associated degradation

The unfolded protein response: a stress signaling pathway critical for health and disease

Transcriptional and translational control in the Mammalian unfolded protein response

IRE1alpha kinase activation modes control alternate endoribonuclease outputs to determine divergent cell fates

Allosteric Inhibition of the IRE1alpha RNase Preserves Cell Viability and Function during Endoplasmic Reticulum Stress

IRE1 couples endoplasmic reticulum load to secretory capacity by processing the XBP-1 mRNA

XBP1 mRNA is induced by ATF6 and spliced by IRE1 in response to ER stress to produce a highly active transcription factor

XBP-1 regulates a subset of endoplasmic reticulum resident chaperone genes in the unfolded protein response

The unfolded protein response: controlling cell fate decisions under ER stress and beyond

IRE1alpha cleaves select microRNAs during ER stress to derepress translation of proapoptotic Caspase-2

New insights into the role of RNase L in innate immunity

Viruses, endoplasmic reticulum stress, and interferon responses

Coronavirus infection modulates the unfolded protein response and mediates sustained translational repression

Unfolded Protein Response Inhibition Reduces Middle East Respiratory Syndrome Coronavirus-Induced Acute Lung Injury

Manipulation of the unfolded protein response: A pharmacological strategy against coronavirus infection

The Unfolded Protein Response and Autophagy on the Crossroads of Coronaviruses Infections

Coronavirus Endoribonuclease and Deubiquitinating Interferon Antagonists Differentially Modulate the Host Response during Replication in Macrophages

A human coronavirus OC43 variant harboring persistence-associated mutations in the S glycoprotein differentially induces the unfolded protein response in human neurons as compared to wild-type virus

Coordination between Two Branches of the Unfolded Protein Response 32

Severe acute respiratory syndrome coronavirus envelope protein regulates cell stress response and apoptosis

The endoplasmic reticulum stress sensor IRE1alpha protects cells from apoptosis induced by the coronavirus infectious bronchitis virus

Modulation of the unfolded protein response by the severe acute respiratory syndrome coronavirus spike protein

The ORF8 protein of SARS-CoV-2 induced endoplasmic reticulum stress and mediated immune evasion by antagonizing production of interferon beta

The SARS Coronavirus 3a protein causes endoplasmic reticulum stress and induces ligand-independent downregulation of the type 1 interferon receptor

The 8ab protein of SARS-CoV is a luminal ER membrane-associated protein and induces the activation of ATF6

A SARS-CoV protein

ER stress and JNK-dependent apoptosis

Influenza A viral replication is blocked by inhibition of the inositolrequiring enzyme 1 (IRE1) stress pathway

XBP1: the last two decades

In vitro study of aldose reductase inhibitor concentrations in the lens and inhibitory effect on sugar alcohol accumulation

Multiple ways to die: delineation of the unfolded protein response and apoptosis induced by Surfactant Protein C BRICHOS mutants

An SFTPC BRICHOS mutant links epithelial ER stress and spontaneous lung fibrosis

Fusion-defective mutants of mouse hepatitis virus A59 contain a mutation in the spike protein cleavage signal

Derivation of self-renewing lung alveolar epithelial type II cells from human pluripotent stem cells

Masitinib is a broad coronavirus 3CL inhibitor that blocks replication of SARS-CoV-2

Role of spike protein endodomains in regulating coronavirus entry

STAR: ultrafast universal RNA-seq aligner

featureCounts: an efficient general purpose program for assigning sequence reads to genomic features

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

Integrative genomics viewer

Cannabidiol Inhibits SARS-CoV-2 Replication and Promotes the Host Innate Immune Response

Real-time PCR quantification of spliced X-box binding protein 1 (XBP1) using a universal primer method

SARS-COV-2 inhibits IRE1a-medicated XBP1 splicing under ER stress and does not require IRE1α for replication

A549-ACE2 cells were mock infected or infected (in triplicate) with SARS-CoV-2 (MOI=3) (A) or OC43 (MOI=1) (B) for 24 hours prior to treatment with low doses of tunicamycin

Total RNA was harvested and used to quantify the relative %XBP1s and XBP1s expression by RT-qPCR. CT values were normalized to 18S rRNA and expressed as foldchange over mock displayed as 2 −Δ(ΔCt)

01; *** = P < 0.001; ns = not significant). (C-F) Infection of CRISPR/Cas9-edited IRE1a KO A549 cells with different coronaviruses. Experiments were performed using sgControl or IRE1a KO or XBP1 KO (where indicated) A549 cells stably expressing viral receptors

Cells were infected (in triplicate) with SARS-CoV-2, MERS-CoV, OC43, or MHV at a MOI of 1. At the indicated times, supernatants were collected and infectious virus quantified by plaque assay

Statistical significance was determined by two-way ANOVA (* = P < 0

01; ns = not significant). Data shown are from one representative of at least two independent experiments. (G) Model of MHV, OC43 and MERS-CoV infection activating

At the indicated time points, total RNA was collected. RT-PCR was performed using primers crossing the XBP1 splicing site. The product was analyzed on an agarose gel to visualize XBP1 splicing

CRISPR/Cas9 gene edited control or XBP1 KO A549-ACE2 were treated with DMSO or tunicamycin (Tm, 1μg/mL) for 6 hours. Lysates were then immunblotted for XBP1s to confirm knockout efficiency

We thank Alejandra Fausto for help with OC43 propagation and titration and