SARS-CoV-2 diverges from other betacoronaviruses in only partially activating the IRE1a/XBP1 ER stress pathway in human lung-derived cells

Severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2) has killed over 6 million individuals worldwide and continues to spread in countries where 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 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 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 through autophosphorylation, but its RNase activity fails to splice XBP1. Moreover, while IRE1a was dispensable for replication in human cells for all coronaviruses tested, it was required for maximal expression of genes associated with several key cellular functions, including the interferon signaling pathway, during SARS-CoV-2 infection. Our data suggest that SARS-CoV-2 actively inhibits the RNase of autophosphorylated IRE1a, perhaps as a strategy to eliminate detection by the host immune system. A549 cells expressing the MERS-CoV receptor DPP4 were cultured in 10% FBS RPMI media. At 70% cell confluence, cells were washed once with PBS before being mock infected or infected with MERS-CoV (EMC/2012) at MOI = 1. Virus was absorbed for 1 hour at 37 degrees Celsius in serum-free RPMI media. After one hour, virus was removed, cells washed three times with PBS, and 2% FBS RPMI was added. The cells were incubated for another 24 hours or 36 hours, then washed once with PBS and lysed using RLT Plus lysis buffer before genomic DNA removal and total RNA extraction using the Qiagen RNeasy Plus Mini Kit (Qiagen 74134). Three independent biological replicates were performed per experimental condition. RNA sample quality check, library construction, and sequencing were performed by GeneWiz following standard protocols. All samples were sequenced by an Illumina HiSeq sequencer to generate paired-end 150bp reads. Read quality was assessed using FastQC v0.11.2 as described by Andrews, S. (2010) “FastQC: A Quality Control Tool for High Throughput Sequence Data” (http://www.bioinformatics.babraham.ac.uk/projects/fastqc/). Raw sequencing reads from each sample were quality and adapter trimmed using BBDuk 38.73 as described by Bushnell, B at "BBTools software package" (http://sourceforge.net/projects/bbmap). The reads were mapped to the human genome (hg38 with Ensembl V98 annotation) using RNA STAR 2.7.1a. The resulting BAM files were counted by featureCounts 1.6.4 to count the number of reads for each gene. For SARS-CoV-2 and OC43 infections, ACE2-A549 control or IRE1 KO cells were cultured in 10% FBS RPMI to 70% confluence. Cells were washed once with PBS before being mock infected or infected with each virus at MOI = 1 for one hour in serum-free RPMI at 33C. Cells were then washed three times with PBS before 2% FBS RPMI was added. At 48 hours post infection, cells were lysed with RLT Plus lysis buffer before genomic DNA removal and total RNA extraction using the Qiagen RNeasy Plus Mini Kit (Qiagen 74134). Three independent biological replicates were performed per experimental condition. RNA sample quality check, library construction, and sequencing were performed by the University of Chicago Genomics Facility following standard protocols. All samples were sequenced in two runs by a NovaSeq 6000 sequencer to generate paired-end 100bp reads. For each sample, the reads from two flow cells were combined before downstream processing. Quality and adapter trimming were performed on the raw sequencing reads using Trim Galore! 0.6.3 (https://github.com/FelixKrueger/TrimGalore). The reads were mapped to the human genome (UCSC hg19 with GENCODE annotation).