SARS-CoV-2 non-structural protein 13 (nsp13) hijacks host deubiquitinase USP13 and counteracts host antiviral immune response

Abstract

Dear Editor, COVID-19 (Coronavirus Disease-2019), a respiratory disease caused by the novel virus strain, SARS-CoV-2 (severe acute respiratory syndrome coronavirus 2), an enveloped, positivesense, single-stranded RNA betacoronavirus of the family Coronaviridae, has spread worldwide. Notably, SARS-CoV-2 infection led to poor induction of type I interferon response, and the impaired type I IFN responses have been shown to be associated with severe COVID-19 disease. However, molecular mechanisms by which SARS-CoV-2 suppresses type I IFN production, and how host cells respond to the inhibition of type I IFN response during SARSCoV-2 infection, remain largely unknown. Here, we found that SARS-CoV-2 non-structural protein 13 (nsp13) has an inhibitory role in regulating type I interferon production. nsp13 overexpression suppressed IFN-β levels induced by RNA virus mimics (3p-hpRNA, Poly: IC) and influenza virus (Fig. 1a–c and Supplementary Fig. S1). We next examined the interaction of nsp13 with core components of host antiviral immune signaling initiated by RNA virus, and found that nsp13 interacts with TBK1 but not others (Fig. 1d and Supplementary Fig. S2a, b), implying that nsp13 might suppress type I IFN production through interacting with TBK1. TBK1 has to be recruited to MAVS that serves as a scaffold to bring IRF3 and TBK1 into proximity, thereby facilitating IRF3 by TBK1. Thus, we asked whether nsp13 suppresses the recruitment of TBK1 to MAVS, and consequently disrupts TBK1-mediated IRF3 phosphorylation. As expected, the interaction of MAVS and TBK1 was impaired in the presence of nsp13 (Fig. 1e). TRAF family has been shown to be responsible for the recruitment of TBK1 to MAVS, thus we asked whether nsp13 affects the interaction of TBK1 and TRAFs. nsp13 overexpression impeded the interaction between TBK1 and TRAF2/3/6 (Fig. 1e). Consistently, we found that the scaffold dimerization domain (SDD) of TBK1 that is responsible for the TRAFs–TBK1 interaction, is required for the interaction of nsp13 and TBK1 (Supplementary Fig. S2c), suggesting that nsp13 competes with TRAFs to bind to TBK1, which could explain the reduced interaction of TBK1 and TRAFs in the presence of nsp13 (Fig. 1e). These results suggest that nsp13 suppresses type I IFN production by disrupting the association of TBK1 with MAVS. Next, we studied how host cells respond to the inhibitory role of nsp13 in regulating type I IFN response. The ubiquitination and deubiquitination system play critical roles in regulating multiple signaling pathways, including protein degradation, interaction, and activation. We hypothesized that host cells may employ the ubiquitination system to target nsp13 for degradation and thereby defending against virus infection, while nsp13 in turn might hijack the host deubiquitination system to prevent itself from degradation. Therefore, we performed a screen assay to identify potential deubiquitinases (DUBs) that interact with nsp13, which might be explored for therapeutic use. USP13 was identified to interact with nsp13 (Fig. 1f and Supplementary Fig. S3a, b). We examined nsp13 protein levels in USP13 knockdown cells, and found that nsp13 levels decreased in the absence of USP13, and the decrease of nsp13 levels could be reversed by the addition of proteasome inhibitor MG132 (Fig. 1g and Supplementary Fig. S3c, d), suggesting that USP13 regulates nsp13 levels in a proteasome-dependent manner. Notably, treatment with USP13 inhibitor spautin-1, which could block the deubiquitination activity of USP13, also led to the reduction of nsp13 levels (Fig. 1h and Supplementary Fig. S3e), suggesting that the enzymatic activity of USP13 may be required for its regulation of nsp13. Consistently, overexpression of wild type (WT), but not the catalytic-inactive (CA) mutant of USP13 with a mutation at the core enzymatic domain, could rescue the decreased nsp13 levels caused by USP13 depletion (Fig. 1i and Supplementary Fig. S3f), suggesting that USP13 can regulate nsp13 levels most likely by deubiquitinating and consequently stabilizing nsp13. As expected, we found that USP13 regulates nsp13 ubiquitination in cells. Loss of USP13 led to the increase of ubiquitinated nsp13 (Supplementary Fig. S3g). Overexpression of WT, but not the CA mutant of USP13, decreased the ubiquitination of nsp13 (Supplementary Fig. S3h). These results suggest that nsp13 can hijack the host deubiquitinase USP13 to prevent itself from degradation, thereby suppressing type I IFN production and helping the virus to survive in host cells. Next, we asked whether USP13 affects the inhibitory role of nsp13 in regulating type I IFN production, and whether USP13 inhibitor could be used to target nsp13 for degradation and thereby promoting type I IFN production and consequently suppressing virus replication in host cells. We knocked down USP13 in nsp13 expressing cells, and detected IFN-β levels. As shown in Supplementary Fig. S4a, nsp13 overexpression led to decrease of IFN-β levels, and loss of USP13 can reverse the decrease of IFN-β levels caused by nsp13 overexpression. Reexpression of WT, but not the CA mutant of USP13 in USP13depleted cells, could restore the inhibition of IFN-β levels caused by nsp13 (Supplementary Fig. S4b). Intriguingly, treatment with USP13 inhibitor spautin-1 also reversed the decreased IFN-β levels caused by nsp13 overexpression (Supplementary Fig. S4c). Above results prompted us to further evaluate the effects of USP13 and nsp13 on viral infection. Currently, we could not perform experiments using live SARS-CoV-2 in the lab. We thus used influenza virus as a model. This could also be relevant for potential interaction between SARS-CoV-2 and influenza virus infection. As shown in Fig. 1j, k and Supplementary Fig. S4d, much more viruses were detected in nsp13 overexpression cells than that in control cells. Depletion of USP13 led to the reduction of viruses caused by