Interferon regulatory factor 3 mediates effective antiviral responses to human coronavirus 229E and OC43 infection

Seasonal human coronaviruses OC43, 229E, and NL63 induce cell surface modulation of entry receptors and display host cell-specific viral replication kinetics

The emergence of the COVID-19 pandemic prompted an increased interest in seasonal human coronaviruses. OC43, 229E, NL63, and HKU1 are endemic seasonal coronaviruses that cause the common cold and are associated with generally mild respiratory symptoms. In this study, we identified cell lines that exhibited cytopathic effects (CPE) upon infection by three of these coronaviruses and characterized their viral replication kinetics and the effect of infection on host surface receptor expression. We found that NL63 produced CPE in LLC-MK2 cells, while OC43 produced CPE in MRC-5, HCT-8, and WI-38 cell lines, while 229E produced CPE in MRC-5 and WI-38 by day 3 post-infection. We observed a sharp increase in nucleocapsid and spike viral RNA (vRNA) from day 3 to day 5 post-infection for all viruses; however, the abundance and the proportion of vRNA copies measured in the supernatants and cell lysates of infected cells varied considerably depending on the virus-host cell pair. Importantly, we observed modulation of coronavirus entry and attachment receptors upon infection. Infection with 229E and OC43 led to a downregulation of CD13 and GD3, respectively. In contrast, infection with NL63 and OC43 leads to an increase in ACE2 expression. Attempts to block entry of NL63 using either soluble ACE2 or anti-ACE2 monoclonal antibodies demonstrated the potential of these strategies to greatly reduce infection. Overall, our results enable a better understanding of seasonal coronaviruses infection kinetics in permissive cell lines and reveal entry receptor modulation that may have implications in facilitating co-infections with multiple coronaviruses in humans.IMPORTANCESeasonal human coronavirus is an important cause of the common cold associated with generally mild upper respiratory tract infections that can result in respiratory complications for some individuals. There are no vaccines available for these viruses, with only limited antiviral therapeutic options to treat the most severe cases. A better understanding of how these viruses interact with host cells is essential to identify new strategies to prevent infection-related complications. By analyzing viral replication kinetics in different permissive cell lines, we find that cell-dependent host factors influence how viral genes are expressed and virus particles released. We also analyzed entry receptor expression on infected cells and found that these can be up- or down-modulated depending on the infecting coronavirus. Our findings raise concerns over the possibility of infection enhancement upon co-infection by some coronaviruses, which may facilitate genetic recombination and the emergence of new variants and strains.

SARS-CoV-2-infected human airway epithelial cell cultures uniquely lack interferon and immediate early gene responses caused by other coronaviruses

Objectives

Severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2) is a member of a class of highly pathogenic coronaviruses. The large family of coronaviruses, however, also includes members that cause only mild symptoms, like human coronavirus-229E (HCoV-229E) or OC43 (HCoV-OC43). Unravelling how molecular (and cellular) pathophysiology differs between highly and low pathogenic coronaviruses is important for the development of therapeutic strategies.

Methods

Here, we analysed the transcriptome of primary human bronchial epithelial cells (PBEC), differentiated at the air-liquid interface (ALI) after infection with SARS-CoV-2, SARS-CoV, Middle East Respiratory Syndrome (MERS)-CoV and HCoV-229E using bulk RNA sequencing.

Results

ALI-PBEC were efficiently infected by all viruses, and SARS-CoV, MERS-CoV and HCoV-229E infection resulted in a largely similar transcriptional response. The response to SARS-CoV-2 infection differed markedly as it uniquely lacked the increase in expression of immediate early genes, including FOS, FOSB and NR4A1 that was observed with all other coronaviruses. This finding was further confirmed in publicly available experimental and clinical datasets. Interfering with NR4A1 signalling in Calu-3 lung epithelial cells resulted in a 100-fold reduction in extracellular RNA copies of SARS-CoV-2 and MERS-CoV, suggesting an involvement in virus replication. Furthermore, a lack in induction of interferon-related gene expression characterised the main difference between the highly pathogenic coronaviruses and low pathogenic viruses HCoV-229E and HCoV-OC43.

Conclusion

Our results demonstrate a previously unknown suppression of a host response gene set by SARS-CoV-2 and confirm a difference in interferon-related gene expression between highly pathogenic and low pathogenic coronaviruses.

Human Betacoronavirus OC43 Interferes with the Integrated Stress Response Pathway in Infected Cells

Viruses evolve many strategies to ensure the efficient synthesis of their proteins. One such strategy is the inhibition of the integrated stress response-the mechanism through which infected cells arrest translation through the phosphorylation of the alpha subunit of the eukaryotic translation initiation factor 2 (eIF2α). We have recently shown that the human common cold betacoronavirus OC43 actively inhibits eIF2α phosphorylation in response to sodium arsenite, a potent inducer of oxidative stress. In this work, we examined the modulation of integrated stress responses by OC43 and demonstrated that the negative feedback regulator of eIF2α phosphorylation GADD34 is strongly induced in infected cells. However, the upregulation of GADD34 expression induced by OC43 was independent from the activation of the integrated stress response and was not required for the inhibition of eIF2α phosphorylation in virus-infected cells. Our work reveals a complex interplay between the common cold coronavirus and the integrated stress response, in which efficient viral protein synthesis is ensured by the inhibition of eIF2α phosphorylation but the GADD34 negative feedback loop is disrupted.

SARS-CoV-2 infection in immunosuppression evolves sub-lineages which independently accumulate neutralization escape mutations

One mechanism of variant formation may be evolution during long-term infection in immunosuppressed people. To understand the viral phenotypes evolved during such infection, we tested SARS-CoV-2 viruses evolved from an ancestral B.1 lineage infection lasting over 190 days post-diagnosis in an advanced HIV disease immunosuppressed individual. Sequence and phylogenetic analysis showed two evolving sub-lineages, with the second sub-lineage replacing the first sub-lineage in a seeming evolutionary sweep. Each sub-lineage independently evolved escape from neutralizing antibodies. The most evolved virus for the first sub-lineage (isolated day 34) and the second sub-lineage (isolated day 190) showed similar escape from ancestral SARS-CoV-2 and Delta-variant infection elicited neutralizing immunity despite having no spike mutations in common relative to the B.1 lineage. The day 190 isolate also evolved higher cell-cell fusion and faster viral replication and caused more cell death relative to virus isolated soon after diagnosis, though cell death was similar to day 34 first sub-lineage virus. These data show that SARS-CoV-2 strains in prolonged infection in a single individual can follow independent evolutionary trajectories which lead to neutralization escape and other changes in viral properties.

Evolution and neutralization escape of the SARS-CoV-2 BA.2.86 subvariant

Omicron BA.2.86 subvariant differs from Omicron BA.2 as well as recently circulating variants by over 30 mutations in the spike protein alone. Here we report on the isolation of the live BA.2.86 subvariant from a diagnostic swab collected in South Africa which we tested for escape from neutralizing antibodies and viral replication properties in cell culture. We found that BA.2.86 does not have significantly more escape relative to Omicron XBB.1.5 from neutralizing immunity elicited by either Omicron XBB-family subvariant infection or from residual neutralizing immunity of recently collected sera from the South African population. BA.2.86 does have extensive escape relative to ancestral virus with the D614G substitution (B.1 lineage) when neutralized by sera from pre-Omicron vaccinated individuals and relative to Omicron BA.1 when neutralized by sera from Omicron BA.1 infected individuals. BA.2.86 and XBB.1.5 show similar viral infection dynamics in the VeroE6-TMPRSS2 and H1299-ACE2 cell lines. We also investigate the relationship of BA.2.86 to BA.2 sequences. The closest BA.2 sequences are BA.2 samples from Southern Africa circulating in early 2022. Similarly, many basal BA.2.86 sequences were sampled in Southern Africa. This suggests that BA.2.86 potentially evolved in this region, and that unobserved evolution led to escape from neutralizing antibodies similar in scale to recently circulating strains of SARS-CoV-2.