Cell culture

Human embryonic kidney (HEK) 293T (CRL-11268; ATCC) and Madin-Darby canine kidney (MDCK) (CCL-34; ATCC) cells were cultured in Dulbecco’s modified Eagle’s medium (Sigma-Aldrich) supplemented with 10% heat-inactivated fetal calf serum and 1% penicillin/streptomycin (Thermo Fisher Scientific).

Plasmids

The open reading frame (ORF) sequence of the GP of the CCHFV Hoti strain was synthesized (Genewiz). The synthesized GP sequence was codon-optimized for high expression in human cells and inserted into pCAGGS (pCAGGS/CCHFV GP). The expression plasmid for CCHFV GPΔNSm was constructed by removal of the NSm coding region from pCAGGS/CCHFV GP using a KOD-Plus mutagenesis kit (Toyobo). The ORF encoding the NSm from the pCAGGS/CCHFV GP was amplified by polymerase chain reaction (PCR) and inserted into pCAGGS with a FLAG or His tag using an In-Fusion HD cloning kit (TaKaRa). Hemagglutinin (HA)-tagged Severe fever with thrombocytopenia virus (SFTSV) NSs, His-tagged STAT1, His-tagged STAT2, and FLAG-tagged STAT2 expression plasmids were prepared as described previously [10]. To construct the green fluorescent protein (GFP)–fused STAT1 or STAT2 expression plasmids, the ORF encoding STAT1 or STAT2 was inserted into a pAC-GFP N1 plasmid (Clontech). Expression plasmids for a series of STAT2 deletion mutants were constructed using a KOD-Plus mutagenesis kit (Toyobo). Expression plasmids for STAT1(2SH) and STAT2(1SH) were constructed using the In-Fusion HD cloning kit (TaKaRa). The expression plasmid for HA-tagged ubiquitin with only K48 and other lysines mutated to arginines, pRK5-HA-Ubiquitin-K48, was a gift from Ted Dawson (Addgene plasmid #17605; http://n2t.net/addgene:17605; RRID:Addgene_17605) [11].

Virus

The vesicular stomatitis Indiana virus (VSIV) was a kind gift from Dr. H. Kida (Hokkaido University). Virus stocks were prepared from the culture supernatants of MDCK cells.

Reporter gene assay

The reporter gene assay was performed as previously described [10,12]. Briefly, HEK293T cells (2 × 10⁵ cells/well in 12-well plates) were transfected with the ISRE reporter plasmid containing the ISRE promoter upstream of the firefly luciferase gene (500 ng; Promega), and the pRL-TK plasmid (Renilla luciferase control plasmid driven by the constitutively active herpes simplex virus thymidine kinase promoter; 50 ng; Promega) as an internal control. Transfections were performed using TransIT-LT1 transfection reagent (Mirus Bio, Cat# MIR 2300) according to the manufacturer’s instructions, with or without the indicated amount of NSm-FLAG expression plasmid. At 24 h post-transfection, the cells were treated or untreated with IFN-αA/D (500 U/mL) (PBL Assay Science) for 18 h. Luciferase activities were then measured using a Dual-Luciferase Reporter Assay kit (Promega) and a TriStar LB941 system (Berthol). Firefly luciferase activity was normalized to Renilla luciferase activity to account for transfection efficiency.

Co-localization assay

All transfections were performed using TransIT-LT1 transfection reagent (Mirus Bio, Cat# MIR 2300) according to the manufacturer’s instructions. To investigate STAT1 and STAT2 localizations, the expression plasmids for His tagged STAT1 and STAT2 were transfected into HEK293T cells with pDsRed2-ER (Clontech) or pDsRed-Monomer-Golgi (Clontech) and the expression plasmid for NSm-FLAG. The cells were subjected to an indirect immunofluorescence assay (IFA) at 48 h post-transfection. pDsRed2-ER contains a coding sequence for red fluorescent protein (RFP) tagged with an N-terminal signal peptide and a C-terminal ER retention sequence (K-D-E-L). pDsRed-Monomer-Golgi encodes a fusion protein of an RFP, while the N-terminal 81 amino acids of human β-1,4-galactosyltransferase contain the membrane-anchoring signal peptide utilized in trans-Golgi network (TGN) localization.

Proteasome inhibitor treatment

To analyse the effect of MG132, a proteasome inhibitor, on STAT2 degradation, HEK293T cells were transfected with or without NSm-FLAG expression plasmid. At 24 h post-transfection, the cells were treated with or without MG132 (0.7 μM; Sigma-Aldrich) for 16 h prior to harvesting.

Quantitative real-time RT-PCR

Quantitative real-time RT-PCR was performed as previously described using the primers corresponding to the OAS1 gene (forward primer, 5′- CATCCGCCTAGTCAAGCACTG-3′; reverse primer, 5′- CACCACCCAAGTTTCCTGTAG-3′), GAPDH gene (forward primer, 5′- TGTTGCCATCAATGACCCCTT-3′; reverse primer, 5′- CTCCACGACGTACTCAGCG-3′) or STAT2 gene (forward primer, 5′- CAGGTCACAGAGTTGCTACAGC-3′; reverse primer, 5′- CGGTGAACTTGCTGCCAGTCTT-3′) [10,12]. Real-time RT-PCR was performed using the One Step TB Green PrimeScript PLUS RT-PCR Kit (TaKaRa). Relative mRNA levels were calculated using the 2^−ΔΔCT method with GAPDH mRNA used as an internal control and are shown as relative fold-changes normalized to untreated control samples.

VSIV infection

HEK293T cells transfected with NSm-FLAG or SFTSV NSs-HA were treated or untreated with IFN-αA/D (500 U/mL), which is universal IFN-I, (PBL Assay Science) for 24 h. The treated and untreated cells were infected with VSIV at a multiplicity of infection (MOI) of 10. At 6 h post-inoculation, the cells were lysed in 200 μL of sodium dodecyl sulphate sample buffer.

Western blotting

Western blotting was performed as previously described using the following antibodies [10,12,13]: anti-HA (clone 3F10; Roche), anti-His (clone 9F2; Fujifilm Wako), anti-FLAG (catalog no. PM020; MBL), anti-VSIV M (clone 23H12; Kerafast), anti-VSIV G (clone P5D4; Sigma-Aldrich), anti-STAT2 (catalog no.: D9J7L; Cell Signaling Technology), anti-STAT1 (catalog no.: D19KY; Cell Signaling Technology), anti-STAT2 (phosphor Y690) (catalog no.: D3P2P; Cell Signaling Technology), anti-STAT1 (phosphor Y701) (catalog no.: D4A7; Cell Signaling Technology), anti-CCHFV Gn (in-house production) or anti-β-actin (catalog no. AC-15; Sigma-Aldrich). The band intensities of VSIV-G, VSIV-M, STAT1, pSTAT1, STAT2, pSTAT2, NSm-FLAG, and β-actin were quantified using ImageJ software (National Institutes of Health). The expression level of VSIV-G, VSIV-M, STAT1, pSTAT1, STAT2, and pSTAT2 ware adjusted by the amounts of β-actin. The STAT1 or STAT2 levels in the IP eluents were adjusted according to the amount of NSm-FLAG.

Indirect IFA

To observe the co-localization between host and viral proteins, an indirect IFA was performed as previously described [10,12]. All transfected cells were fixed with 4% paraformaldehyde in phosphate-buffered saline (Wako). The fixed cells were incubated in cold methanol for permeabilization and in blocking buffer (5% goat serum (Gibco) and 0.3% Triton X-100 (Wako) in phosphate-buffered saline (Shimadzu Diagnostics)) for blocking. The cells were then treated with primary antibodies overnight at 4°C and stained with secondary antibodies for 2 h at room temperature with 4′,6-diamidino-2-phenylindole (DAPI)(Roche) to visualize the nuclei. Images were acquired using an LSM780 confocal-microscope (Carl Zeiss). Distribution of signals were measured by ImageJ software (National Institutes of Health).

Coimmunoprecipitation assay

The co-immunoprecipitation (co-IP) assay was performed as previously described [10,12,13]. The cells were mixed with magnetic beads conjugated to anti-FLAG (FLA-1; MBL) or anti-HA (5D8; MBL) monoclonal antibodies and incubated at 4°C for 3 h or overnight. The magnetic beads were then washed with lysis buffer (25 mM Tris-HCl (Wako), 150 mM NaCl (Wako), 1 mM ethylenediaminetetraacetic acid [EDTA] (Wako), and 1% Triton X-100 (Wako)) and wash buffer (50 mM Tris-HCl (Wako), 1% NP-40 (Wako), 0.25% deoxycholic acid sodium salt (Wako), 150 mM NaCl (Wako), and 1 mM EDTA (Wako)) and analysed by western blotting.

Statistical analyses

The data are expressed as mean and standard deviation (SD), and statistically significant differences were determined using Student’s t test (Fig 2C), one-way ANOVA with Tukey’s (Figs 2B and 6E), or Dunnett’S (Figs 1A–1C and 6F), or two-way ANOVA with Tukey’s (Fig 4) multiple comparison test. All statistical analyses were performed using Prism GraphPad 9 software (GraphPad Software, Inc.).

Download:

• PPT
  PowerPoint slide
• PNG
  larger image
• TIFF
  original image

Fig 1. Function of NSm as an IFN-I antagonist in human cell line.

(A) The reporter plasmids were transfected into HEK293T cells with or without the expression plasmid for CCHFV GP, CCHFV GPΔNSm, or NSm-FLAG (2μg). At 24 h post-transfection, the cells were treated with IFN-αA/D (500 U/mL) or left untreated for 18 h and then lysed for the measurement of luciferase activity and detection of protein expression using immunoblotting. Relative light units in transfected cells with each plasmid in the absence of IFN-αA/D were set as 1. Fold-change activation by IFN-αA/D is indicated. (B) NSm dose-dependent inhibition of ISRE activation was examined using different amounts of the NSm expression plasmid in HEK293T cells. (C) NSm-FLAG expression plasmid was transfected into HEK293T cells. At 24 h post-transfection, the cells were treated with IFN-αA/D (500 U/mL) for 10 h or left untreated. Expression levels of Oas1 mRNAs in each cell line were measured by quantitative real-time reverse transcription polymerase chain reaction. The mRNA expression levels of OAS1 in untreated cells were set as 1. Fold-change activation by IFN-αA/D is indicated. (D) The expression plasmids for NSm-FLAG or SFTSV NSs–hemagglutinin were transfected into HEK293T cells. At 24 h post-transfection, the cells were treated with IFN-αA/D (500 U/mL) for 24 h or left untreated and then infected with VSIV at a multiplicity of infection of 10. At 6 h post-infection, the protein expressions of VSIV M and G were detected by western blotting. The numbers indicate the relative band intensities. The relative expression rate of VSVI G and M in untreated cells with IFN-I was set at 100%. These assays were independently performed in triplicate. The data represent means and standard deviations. *P < 0.05; **P < 0.01 versus empty. Blots have been cropped; full uncropped blots are available as S1Fig.CCHFV, Crimean-Congo hemorrhagic fever virus; GP, glycoprotein precursor; IFN-I, type I interferon; VSIV, vesicular stomatitis Indiana virus.

https://doi.org/10.1371/journal.pntd.0013695.g001

Download:

• PPT
  PowerPoint slide
• PNG
  larger image
• TIFF
  original image

Fig 2. NSm degrades STAT2 via a ubiquitin/proteasome-dependent pathway.

(A) HEK293T cells transfected with the expression plasmid for NSm-FLAG and treated with IFN-αA/D (2,000 U/mL) for 30 min or left untreated and then lysed for the detection of expression of each protein by immunoblotting. The numbers indicate the relative band intensities. The relative expression rate of each protein in mock-transfected cells was set at 100%. (B) The expression plasmid for NSm-FLAG (2 µg) was transfected into HEK293T cells. At 24 or 48 h post-transfection, the RNA expression level of STAT2 was determined by quantitative real-time reverse transcription polymerase chain reaction. (C)The transfected HEK293T cells with or without the expression plasmid for NSm-FLAG were treated with MG132 (0.7 μM) for 16 h or left untreated. The expression of each protein was detected by western blotting (upper panel). The expression level of STAT2 was quantified (lower panel). The relative expression rate of STAT2 in mock-transfected cells was set at 100%. (D) HEK293T cells were transfected with STAT2-FLAG, NSm-His, and ubiquitin (K48)-hemagglutinin, and co-immunoprecipitation assays were performed by anti-hemagglutinin antibodies. These assays were independently performed in triplicate. The data are shown as means and standard deviations. *P < 0.05; **P < 0.01 versus empty (B) or no treatment (C) conditions. Blots have been cropped; full uncropped blots are available as S2 and S3 Figs.

https://doi.org/10.1371/journal.pntd.0013695.g002

Download:

• PPT
  PowerPoint slide
• PNG
  larger image
• TIFF
  original image

Fig 3. Interaction of NSm with STAT1 and STAT2.

(A) HEK293T cells were transfected with the expression plasmid for NSm-FLAG, and co-immunoprecipitation assays were performed. (B) HEK293T cells were transfected with the expression plasmid for NSm-FLAG and STAT1-His or STAT2-His. The protein expression levels in cell lysates (upper) and co-immunoprecipitation assays (lower) using an anti-FLAG antibody are shown. Blots have been cropped; full uncropped blots are available as S4 Fig. (C) Co-localization of NSm with STAT1 and STAT2. HEK293T cells were transfected with the expression plasmid for NSm-FLAG (0.1 µg) and STAT1- or STAT2-GFP (0.1 µg). An immunofluorescent assay was also performed with NSm, STAT1 (or STAT2), and the nuclei shown in red, green, and blue, respectively. The distribution of NSm and STAT1 (or STAT2) was shown. Representative results of western blotting (A and B) and immunofluorescence (C) assays are shown. The scale bar represents 20 μm.

https://doi.org/10.1371/journal.pntd.0013695.g003

CCHFV NSm suppresses IFN-I signaling

IFN-I activates the ISRE promoter and upregulates ISG expression. To investigate whether NSm suppresses IFN-I signaling, we examined the effect of NSm on ISRE activation by IFN-I in HEK293T cells, which are IFN-I-competent and susceptible to CCHFV infection [10,14], using a Dual-Luciferase Reporter Assay kit (Promega). In this experiment, the cells were treated or untreated with IFN-αA/D (500 U/ml) for 18 h. Consistent with previous reports [12], IFN-αA/D induced strong ISRE activation in HEK293T cells (Fig 1A). In this condition, the GP and NSm proteins significantly suppressed the ISRE activation driven by IFN-αA/D, while GP-deleted NSm protein did not. Each protein was detected by western blotting (Fig 1A). In cells expressing NSm, two distinct bands were observed (Fig 1A). The upper band may correspond to a post-translationally modified form of the NSm protein, such as glycosylation or phosphorylation. In addition, NSm dose-dependently inhibited ISRE activation (Fig 1B). To investigate the effect of NSm on ISG mRNA induction by IFN-I, the induction of oligoadenylate synthetase 1 (OAS1) mRNA was examined by quantitative real-time RT-PCR. The induction of OAS1 by IFN-αA/D was inhibited by NSm expression in a dose-dependent manner (Fig 1C).

We used a VSIV infection assay to monitor the antiviral state because its replication is efficiently limited by IFN-I treatment [15]. HEK293T cells transfected with the expression plasmid for NSm-FLAG or SFTSV NSs-HA, which is known to function as an IFN-I signaling inhibitor in HEK293T cells [10,12], were treated with IFN-αA/D. At 24 h post-treatment, the cells were infected with VSIV. At 6 h post-infection, the VSIV G and M protein expressions were confirmed by western blotting. As shown in Fig 1D, the treatment of IFN-αA/D reduced VSIV G and M protein expression by 0% and 14%, respectively. In the presence of NSm or SFTSV NSs, the relative expression level of VSIV G and M protein was 8.4% (NSm) or 21.6% (SFTSV NSs) and 78.9% (NSm) or 77.3% (SFTSV NSs), respectively, indicating that the expression level of VSIV G and M proteins was partially recovered by NSm and SFTSV NSs. These results suggest that CCHFV NSm interferes with IFN-I signaling in human cells.

CCHFV NSm degrades STAT2 via a ubiquitin/proteasome-dependent pathway

To examine the protein expression levels of STAT1 and STAT2 in the presence of NSm, HEK293T cells were transfected with or without NSm-FLAG expression plasmid. Regardless of the NSm concentration, the STAT1 protein expression levels did not change (Fig 2A). However, the STAT2 protein expression level decreased in an NSm concentration–dependent manner (Fig 2A). In addition, to investigate the STAT1 and STAT2 phosphorylation induced by IFN-I in the presence of NSm, transfected cells were treated or untreated with IFN-αA/D (2000 U/ml) for 30 min. As shown in Fig 2A, NSm downregulated STAT2 but not STAT1 phosphorylation. Moreover, to investigate whether NSm reduce the STAT2 mRNA expression level, HEK293T cells were transfected with or without the expression plasmid of NSm (2.0 µg). The quantitative real-time RT-PCR results showed an increase in STAT2 mRNA levels in the presence of NSm (Fig 2B). These results suggest that NSm downregulates STAT2 at the protein but not mRNA level.

We then examined whether NSm induces STAT2 degradation via a ubiquitin/proteasome-dependent pathway. HEK293T cells were transfected with or without the NSm-FLAG expression plasmid and then treated with proteasome inhibitor MG132 (0.7 μM) for 16 h. NSm reduced the amount of STAT2 by 50% (Fig 2C). In contrast, MG132 restored the amount of STAT2 to the control levels (Fig 2C). K48-linked ubiquitination is a process in which a chain of ubiquitin molecules is specifically linked via the lysine 48 (K48) residue of ubiquitin and added to a target protein [16]. This modification primarily functions as a signal for protein degradation by the proteasome. Therefore, we performed a co-IP assay using the lysates of HEK293T cells transfected with K48-linked ubiquitin, STAT2-FLAG, or NSm-His expression plasmids and treated with proteasome inhibitor MG132 (0.7 μM) for 16 h. The co-IP assay using anti-HA antibodies indicated that the amount of K48 ubiquitin-modified STAT2 was increased by NSm. These results indicate that NSm induced STAT2 degradation via a ubiquitin/proteasome-dependent pathway.

CCHFV NSm interacts with STAT2

To examine the interaction of NSm with STAT1 and STAT2, we performed co-IP assays using the lysates of HEK293T cells transfected with the NSm-FLAG expression plasmid. As shown in Fig 3A, co-IP of STAT2 with NSm was observed, whereas co-IP of STAT1 with NSm was not. In addition, strong binding of NSm to STAT2 was observed after MG132 treatment (Fig 3A). To understand the difference in the binding ability of NSm to STAT1 and STAT2, we transfected the expression plasmid for NSm-FLAG with the expression plasmids for STAT1-His or STAT2-His into HEK293T cells and then performed a co-IP assay. In an experiment demonstrating the interaction of NSm with endogenous STAT2 (Fig 3A), NSm efficiently bound to STAT2, whereas its interaction with exogenous STAT1 was very weak (Fig 3B). The interactions of NSm with STAT1 and STAT2 were further investigated by the subcellular co-localization of NSm, STAT1, and STAT2. The NSm-FLAG expression plasmid was co-transfected with the STAT1- or STAT2-GFP expression plasmid into HEK293T cells, and the subcellular localization of NSm, STAT1, and STAT2 was observed by indirect IFA. NSm was co-localized with STAT1 and STAT2 (Fig 3C). To further confirm that STAT2 was the target protein of NSm, the anti-IFN-I signaling activity of NSm was evaluated in STAT2-overexpressing HEK293T cells (Fig 4). As done in Fig 1A and 1B, the transfected cells were treated or untreated with IFN-αA/D (500 U/ml) for 18 h. The expression of each protein was observed by immunoblotting. In this condition, the inhibition of ISRE activity by NSm was restored by STAT2 overexpression. Therefore, we suggest that the interaction between NSm and STAT2 inhibits IFN-I signaling.

NSm alters subcellular STAT2 localization

NSm was previously reported to localize to the endoplasmic reticulum (ER) and Golgi apparatus [14]. Therefore, here we hypothesized that the subcellular localization of STAT2 is altered by its binding to NSm. To address this possibility, STAT1-His or STAT2-His were co-expressed with fluorescent marker proteins that localize in the ER or TGN. Consistent with a previous report [17], NSm co-localized with both ER and TGN RFP marker proteins (Fig 5B, 5D, 5F, and 5H). In the absence of NSm, STAT2 was localized in the cytoplasm and ER but not in the TGN (Fig 5E and 5G). In contrast, STAT2 was partially localized to the TGN in the presence of NSm (Fig 5H). Although we did not perform quantitative co-localization analysis, representative images showing the spatial distribution of each signal were provided. Based on these observations, we described STAT2 as being partially localized to the TGN under NSm-expressing conditions. STAT1 was localized in the nucleus, cytoplasm, and ER regardless of NSm expression (Fig 5A–5D). These findings indicate that NSm alters the subcellular localization of STAT2 but not STAT1.

STAT2 SH domain is important for NSm binding

To investigate the region of STAT2 binding to NSm, we prepared a series of STAT2 deletion mutants (Fig 6A) and examined their binding using a co-IP assay. As shown in Fig 6B, the expression level of Δ5 was lower than that of the others. The cause of this is thought to be the instability of protein due to deletion. In this condition, NSm bound to STAT2Δ1, Δ2, Δ3, and Δ4, but not Δ5 (Fig 6B). However, it was still unclear whether the binding of NSm and STAT2 was mediated by SH domain, since the expression level of Δ5 was extremely low. Therefore, we also prepared two chimeric proteins between STAT1 and STAT2: STAT1(2SH) and STAT2(1SH) (Fig 6C) and performed a co-IP assay to examine the interaction between NSm and both chimeric proteins (Fig 6D). Although the expression level of the STAT1(2SH) mutant was much lower than that of STAT1 and STAT2, co-immunoprecipitation of STAT1(2SH) with NSm was observed (Fig 6D). Contrary to our expectations, we found that STAT2(1SH) also interacted with NSm (Fig 6D). Therefore, to quantitatively evaluate these results, we normalized the expression levels of STAT1, STAT2, and both chimeric proteins in the co-IP solution to those in the whole-cell lysates. As shown in Fig 6E, the binding ability of STAT1(2SH) to NSm was much stronger than that of STAT1 to NSm, while that of STAT2(1SH) to NSm was weaker than that of STAT2 to NSm.

Next, we quantitatively evaluated changes in protein expression levels in whole cells using NSm expression. In the presence of NSm, the STAT2 and STAT1(SH) expression levels decreased, whereas the STAT1 and STAT2(1SH) expression levels did not (Fig 6F). These results suggest that the STAT2 SH domain is important for interaction with and degradation by NSm.