SARS-CoV-2 Orf9b suppresses type I interferon responses by targeting TOM70

COVID-19 is caused by SARS-CoV-2. As of July 16th, 2020, there were 13,579,581 diagnosed cases and 584,782 deaths attributed to COVID-19 reported globally (https://coronavirus.jhu.edu/map. html). Unfortunately, there is still no effective drug or vaccine for treating this disease. To accelerate drug development, there is an urgent need to identify critical molecular targets and the role they play in infection. Herein, we reported that Orf9b localizes on the membrane of mitochondria and suppresses type I interferon (IFN-I) responses through association with TOM70, and TOM70 overexpression could largely rescue this inhibition. Our results suggest the potential of targeting Orf9b-TOM70 interaction as a novel therapeutic strategy of COVID-19. Induction of IFN-I is a central event of the immune defense against viral infection. Upon exposure to RNA viruses, an intracellular antiviral response is initiated by activation of RIG-I like receptors. In particular, when RIG-I/MDA5 detects viral RNA, they trigger a signaling complex on the mitochondrial outer membrane, including the adapter proteins MAVS/TRAF3/TRAF6/TOM70, which ultimately leads to IFN-β production and induction of a host antiviral state. Recent studies have shown that the most prominent feature of SARS-CoV-2, in terms of immune responses as compared to that of other viruses such as influenza A, is that it triggers a very low level of IFN-I. In addition, it has also been found that the chemical, Liquiritin, can inhibit SARS-CoV-2 by mimicking IFN-I. Thus, understanding how SARS-CoV-2 suppresses IFN-I responses may be a particularly promising approach to devise therapeutic strategies to counteract SARS-CoV-2 infections. Previous studies have shown that SARS-CoV Orf9b, an alternative open reading frame within the nucleocapsid (N) gene, can significantly inhibit IFN-I production as a result of targeting mitochondria. In addition, antibodies against Orf9b were present in the sera of convalescent SARS-CoV. or SARS-CoV-2 patients. Therefore, we speculate that SARS-CoV-2 Orf9b may play a critical role in coronavirus-host interactions, particularly via an effect on IFN-I production. To explore the role of Orf9b in host–pathogen interaction, we employed a biotin-streptavidin affinity purification mass spectrometry approach to identify the human proteins that physically interact with Orf9b (Supplementary Fig. 1a). We found that TOM70 scored the highest among all of the identified interactions (Supplementary Table 1). To validate this interaction, we performed co-immunoprecipitation (co-IP) and found that HA-TOM70 coprecipitated with Orf9b (Fig. 1a) and Orf9b could be pulled down with biotinylated TOM70 (Supplementary Fig. 1b). To quantify the binding strength of this interaction, we performed Biolayer Interferometry (BLI) and found that the Kd is indeed relatively low (44.9 nM) (Fig. 1b). Considering the high homology of Orf9b in SARS-like coronaviruses (Fig. 1c), we also tested whether SARS-CoV Orf9b interacts with TOM70. Interestingly, we found that SARS-CoV Orf9b exhibits a similar binding strength as SARS-CoV-2 Orf9b, indicating that the interaction may be conserved across the SARS-like coronavirus family (Supplementary Fig. 1c). To further pinpoint the region of TOM70 that is required for the interaction with Orf9b, TOM70 was divided into individual domains according to the known functions of the regions (Fig. 1d). We found that only the construct consisting of residues 235–608 (TOM70235-608) that contained both the core and C-terminal domains precipitated with biotinylated Orf9b, and this interaction was comparable with that of the fulllength TOM70 (Fig. 1e, Supplementary Fig. 1d). This suggests that the core and C-terminal domains of TOM70 are essential for this interaction, while the transmembrane and clamp domains are not required. Since TOM70 is located in the outer membrane of mitochondria, we hypothesized that SARS-CoV-2 Orf9b may also localize to the outer membrane of mitochondria through interaction with TOM70. Indeed, immunostaining of Orf9b-Flag expressing HEK 293T cells revealed that both SARS-CoV and SARS-CoV-2 Orf9b localize to the membrane of mitochondria (Supplementary Fig. 2a) and colocalize with TOM70 (Fig. 1f). Further, we expressed TOM70ΔTM, a construct without the N-terminal transmembrane domain of TOM70, to investigate whether it would change the mitochondria localization of Orf9b. Despite the presence of endogenous TOM70 in the cells, TOM70ΔTM overexpression indeed partially disrupted the association of SARS-CoV or SARS-CoV-2 Orf9b with mitochondria (Fig. 1g, Supplementary Fig. 2b). Considering the critical role of mitochondria and TOM70 in IFN-I responses, we next investigated whether Orf9b impacted antiviral IFN-I signaling. We monitored human interferon-β (IFN-β) promoter activity in the presence or absence of SARS-CoV-2 Orf9b using a dual luciferase reporter assay. We observed that Orf9b significantly reduced the activation of IFN-β as compared to that of the vehicle controls. The vehicle controls were prepared by co-transfecting with poly(I:C) (Fig. 1h) or MAVS overexpression (Fig. 1i). Next, we examined whether overexpression of TOM70 can counteract the Orf9b-mediated inhibition of IFN-I responses. We observed that

and 100 μM D-biotin before incubating cells overnight at 16 °C. For the preparation of other proteins, the recombinant proteins were expressed in E. coli BL21 by growing cells in 200 mL LB medium to A600 = 0.6 at 37 °C. Protein expression was induced by the addition of 0.2 mM isopropyl-β-dthiogalactoside (IPTG) before incubating cells overnight at 16 °C. For the purification of 6xHistagged proteins, cell pellets were re-suspended in lysis buffer containing 50 mM Tris-HCl pH 8.0, 500 mM NaCl, 20 mM imidazole (pH 8.0), then lysed by a high-pressure cell cracker (Union-biotech, Shanghai, CHN). Cell lysates were centrifuged at 12,000 rpm for 20 mins at 4℃. Supernatants were purified with Ni 2+ Sepharose beads (Smart-lifesciences, China), then washed with lysis buffer (For the purification of biotinylated protein, 1 mM biotin is added to the wash buffer) and eluted with buffer containing 50 mM Tris-HCl pH 8.0, 500 mM NaCl and 300 mM imidazole, pH 8.0. For the purification of GST-tagged proteins, cells were harvested and lysed by a high-pressure cell cracker in lysis buffer containing 50 mM Tris-HCl, pH 8.0, 500 mM NaCl, 1 mM DTT. After centrifugation, the supernatant was incubated with GST-Sepharose beads (Smart-lifesciences, China). The target proteins were washed with lysis buffer twice and eluted with 50 mM Tris-HCl, pH 8.0, 500 mM NaCl, 1 mM DTT, 40 mM glutathione.

Biotin-streptavidin affinity purification mass spectrometry
Cells were harvested and lysed in M-PER ® Mammalian Protein Extraction (Thermo) supplemented with 1 mM PMSF. Cells were incubated with Biotinylated Orf9b (Heavy medium) or control (light medium) and streptavidin Dynabeads (Invitrogen) for overnight at 4℃. The beads were then mixed at 1:1 ratio. The mixture was extensively washed with PBS buffer with 0.1% Triton X-100 for six times. Then the mixture was washed with 100 mM ammonium carboxylate. Trypsin (Promega) was used for on-bead trypsin digestion overnight at 37 °C. The digested peptides were collected and cleaned with ZipTips (Millipore) before mass spectrometry analysis. The tryptic peptide digests of the proteins were analyzed on an EASY-nL 1200 system coupled online to a Q Exactive plus mass spectrometer (Thermo Scientific, Bremen, Germany). The peptide sequences were determined by searching MS/MS spectra against the Protein database using the Protein Discoverer (version 2.4, Thermo Scientific) software suite with a precursor ion mass tolerance of 10 ppm and fragment ion mass tolerance of 0.02 Da. Carbamidomethyl (C) was set as the fixed modification, oxidation (M) and deamidated(NQ) were set as the variable modification.

Co-immunoprecipitation and immunoblot analysis
Cells were harvested at about 48 h after plasmid transfection and then lysed in M-PER ® Mammalian Protein Extraction (Thermo) supplemented with 1 mM PMSF. Then cell lysates with/ without Orf9b-Flag were used for immunoprecipitation with Flag Antibody (Millipore) and Protein G beads (Invitrogen). Generally, 1 μg primary antibody was added to 200 μL lysates with 50 μL Protein G Dynabeads (Invitrogen) for 4 h at 4℃. The beads were extensively washed with PBS buffer with 0.1% Triton X-100 and eluted with SDS loading buffer by boiling for 5 min. For immunoblot analysis, the samples were electrophorized using SDS-PAGE and transferred to a nitrocellulose membrane (Millipore). The membrane was probed with indicated antibody and scanned by a LI-COR Odyssey scanner (LI-COR Bioscience).

In vitro Streptavidin Pull-down Assays
Biotinylated bait protein was incubated with candidate binding partners or BSA and streptavidin coated Dynabeads (Invitrogen) in PBS buffer for 4 h at 4℃. The beads were extensively washed with PBST buffer (0.1% Triton X-100) three times and boiled with SDS loading buffer for 5 min. Alexa-647 conjugated anti-rabbit antibody (Thermo) (1:1000) for 1 h. Then 4′,6-diamidino-2phenylindole (DAPI) was used to localize nucleus. Images were acquired with a confocal microscope (Nikon & A1Si).
Mean Kon, Koff, KD values were calculated by a 1:1 global fit model using the Data Analysis software (ForteBio). All the curves were processed using Prism software (GraphPad Prism 6.0).  with anti-flag antibody (green). The mitochondria were stained with TOM20 (red) and the nuclei were stained in blue using DAPI. Scale bar, 10 µm . b. Confocal microscopy of HEK 293T cells transfected by SARS-CoV Orf9b-Flag and HA-TOM70 ∆TM , which were stained with anti-flag antibody (green) and anti-HA antibody (magenta). The mitochondria were stained with MitoTracker® Orange CMTMRos (Red) and the nuclei were stained in blue using DAPI. Scale bar, 10 µm . c, IFN-β reporter gene assays using HEK 293T cells expressing Flag or HA-TOM70 and induced by transfection of poly(I:C). Luciferase activity is shown as fold induction. Data are representative of three replicates (mean and s.e.m. of n=3 samples), n.s., not significant (two-tailed unpaired ttest). E, HEK 293T cells expressing Flag.