A prenylated dsRNA sensor protects against severe COVID-19
Résumé
INTRODUCTION
Interferons (IFNs) are cytokines that are rapidly deployed in response to invading pathogens. By initiating a signaling cascade that stimulates the expression of hundreds of genes, IFNs create an antiviral state in host cells. Because IFNs heavily influence COVID-19 outcomes, and severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2) replication can be inhibited by the antiviral state, it is important to understand how the individual antiviral effectors encoded by IFN-stimulated genes (ISGs) inhibit SARS-CoV-2.
RATIONALE
We hypothesized that IFN-stimulated antiviral effectors can inhibit SARS-CoV-2, and that variation at the loci encoding these defenses underlies why some people are more susceptible to severe COVID-19.
RESULTS
We used arrayed ISG expression screening to reveal that 2′-5′-oligoadenylate synthetase 1 (OAS1) consistently inhibited SARS-CoV-2 in different contexts. Using CRISPR-Cas9, we found that endogenous OAS1 makes a substantial contribution to the antiviral state by recognizing short stretches of double-stranded RNA (dsRNA) and activating RNase L. We globally mapped where OAS1 binds to SARS-CoV-2 viral RNAs and found that OAS1 binding is remarkably specific, with two conserved stem loops in the SARS-CoV-2 5′-untranslated region (UTR) constituting the principal viral target.
OAS1 expression was readily detectable at the sites of infection in individuals who died of COVID-19, and specific OAS1 alleles are known to be associated with altered susceptibility to infection and severe disease. It had previously been reported that alleles containing a common splice-acceptor single nucleotide polymorphism in OAS1 (Rs10774671) were associated with less severe COVID-19. We determined that people with at least one allele with a G at this position could express a prenylated form of OAS1 (p46), whereas other individuals could not. Using a series of mutants, we found that C-terminal prenylation was necessary for OAS1 to initiate a block to SARS-CoV-2. Furthermore, confocal microscopy revealed that prenylation targeted OAS1 to perinuclear structures rich in viral dsRNA, whereas non-prenylated OAS1 was diffusely localized and unable to initiate a detectable block to SARS-CoV-2 replication.
The realization that prenylation is essential for OAS1-mediated sensing of SARS-CoV-2 allowed us to examine the transcriptome of infected patients and investigate whether there was a link between the expression of prenylated OAS1 and SARS-CoV-2 disease progression. Analysis of the OAS1 transcripts from 499 hospitalized COVID-19 patients revealed that expressing prenylated OAS1 was associated with protection from severe COVID-19.
Because prenylated OAS1 was so important in human cases, we wanted to determine whether horseshoe bats, the likely source of SARS-CoV-2, possessed the same defense. When we examined the genomic region where the prenylation signal should reside, retrotransposition of a long terminal repeat sequence had ablated this signal, preventing the expression of prenylated anti-CoV OAS1 in these bats.
CONCLUSION
C-terminal prenylation targets OAS1 to intracellular sites rich in viral dsRNA, which are likely the SARS-CoV-2 replicative organelles. Once in the right place, OAS1 binds to dsRNA structures in the SARS-CoV-2 5′-UTR and initiates a potent block to SARS-CoV-2 replication. Thus, the correct targeting of OAS1 and the subsequent inhibition of SARS-CoV-2 likely underpins the genetic association of alleles containing a G at Rs10774671 with reduced susceptibility to infection and severe disease in COVID-19. Moreover, the conspicuous absence of this antiviral defense in horseshoe bats potentially explains why SARS-CoV-2 is so sensitive to this defense in humans.