Dabie bandavirus Nonstructural Protein Interacts with Actin to Induce F-Actin Rearrangement and Inhibit Viral Adsorption and Entry

ABSTRACT

Dabie bandavirus (DBV) is an emerging Bandavirus that causes multiorgan failure with a high fatality rate in humans. While many viruses can manipulate the actin cytoskeleton to facilitate viral growth, the regulation pattern of the actin cytoskeleton and the molecular mechanisms involved in DBV entry into the host cells remain unclear. In this study, we demonstrate that expression of nonstructural protein (NSs) or infection with DBV induces actin rearrangement, which presents a point-like distribution, and this destruction is dependent on inclusion bodies (IBs). Further experiments showed that NSs inhibits viral adsorption by destroying the filopodium structure. In addition, NSs also compromised the viral entry by inhibiting clathrin aggregation on the cell surface and capturing clathrin into IBs. Furthermore, NSs induced clathrin light chain B (CLTB) degradation through the K48-linked ubiquitin proteasome pathway, which could negatively regulate clathrin-mediated endocytosis, inhibiting the viral entry. Finally, we confirmed that this NSs-induced antiviral mechanism is broadly applicable to other viruses, such as enterovirus 71 (EV71) and influenza virus, A/PR8/34 (PR8), which use the same clathrin-mediated endocytosis to enter host cells. In conclusion, our study provides new insights into the role of NSs in inhibiting endocytosis and a novel strategy for treating DBV infections.

IMPORTANCE Dabie bandavirus (DBV), a member of the Phenuiviridae family, is a newly emerging tick-borne pathogen that causes multifunctional organ failure and even death in humans. The actin cytoskeleton is involved in various crucial cellular processes and plays an important role in viral life activities. However, the relationship between DBV infection and the actin cytoskeleton has not been described in detail. Here, we show for the first time the interaction between NSs and actin to induce actin rearrangement, which inhibits the viral adsorption and entry. We also identify a key mechanism underlying NSs-induced entry inhibition in which NSs prevents clathrin aggregation on the cell surface by hijacking clathrin into the inclusion body and induces CLTB degradation through the K48-linked ubiquitination modification. This paper is the first to reveal the antiviral mechanism of NSs and provides a theoretical basis for the search for new antiviral targets.

INTRODUCTION

Tick-borne infectious diseases pose an increasing threat to public health (1). Severe fever with thrombocytopenia syndrome is a newly emerging tick-borne infectious disease caused by Dabie bandavirus (DBV; formerly known as severe fever with thrombocytopenia syndrome virus [SFTSV]) (2). Patients with DBV infection often have acute fever, thrombocytopenia, and leukocytopenia, among other symptoms. In severe cases, it can lead to multifunctional organ failure and even death (2–4). The virus was first isolated in China in 2010 and then spread in South Korea, Japan, and Vietnam (5–8). It remains a threat to public safety, and a deeper understanding of the pathogenesis of DBV is required.

DBV is a negative-stranded RNA virus with segmented genomes which belongs to the family Phenuiviridae of the order Bunyavirales (9). DBV consists of three RNA segments: the L segment encodes an RNA-dependent RNA polymerase (RdRp), the M segment encodes the two viral envelope glycoproteins (GPs), Gn and Gc, and the S segment encodes a nuclear protein (NP) and a nonstructural protein (NSs). As the most important virulence factor of DBV, NSs plays an important role in antagonizing innate immune responses (10, 11). Chaudhary et al. found that NSs inhibited interferon signal transduction by inhibiting STAT1 phosphorylation and activation of STAT1 (12). Hong et al. reported that NSs directly interacts with and sequesters interferon regulatory factor 7 (IRF7) into the inclusion bodies (IBs), and phospho-IRF7 (p-IRF7) is trapped in the IBs, resulting in the reduction of the interferon-α2 (IFN-α2) and -α4 induction and therefore enhanced viral replication (10). Studies have shown that NSs, by interacting with retinoic acid-inducible gene I (RIG-I) and tripartite motif containing 25 (TRIM25), isolates these proteins in viral-induced IBs and prevents them from performing their functions, thus escaping immunity from interferon (13–15). In addition, there are also some studies on the interaction between NSs and host protein, revealing the potential molecular mechanism of viral infection. Liu et al. found that NSs interacts with CDK1 to induce G₂/M arrest in DBV-infected cells, thereby promoting viral replication (16). Choi et al. found that the interaction between NSs and TRIM21 induced the activation of the p62-Keap1-Nrf2 antioxidant reaction, which would create a strong environment for the life cycle and pathogenesis of the virus (17).

Virus entry into the host cell is the first step of infection and is also a key target of virus infection prevention and treatment. It has been proposed that virus entry and cytoplasmic transport are directly or indirectly dependent on actin or the microtubule cytoskeleton (18). A variety of animal viruses, including adenoviruses, herpesviruses, and influenza viruses, rely on microtubules. Although the dynamic interactions between specific viruses and host cytoskeletal proteins vary greatly, microtubule networks always provide a directional transport mode for invading pathogens (19). Actin is the most abundant cytoskeletal protein, mostly localized in the cytoplasm, and forms microfilaments (20). Actin is present in cells in two forms, G-actin and F-actin (21), which plays an important role in virus adsorption, entry, assembly, and release (20, 22, 23). Herpes simplex virus 1 (HSV-1) mediates the assembly of F-actin in the early stage of infection, which is followed by a continuous decrease in F-actin during the later stages of infection. The biphasic F-actin dynamics play a key role in the process of HSV-1 infection, suggesting that the virus promotes its replication process by regulating the structure of the actin skeleton network (24). In addition, it has been reported that viral proteins interact with actin to regulate viral replication. Respiratory syncytial virus matrix (RSV M) interacts with actin, and destabilization of microfilaments leads to relocalization of M within the cells, which facilitates the transport of viral components (25). Although Liu et al. found that the intracellular transport of the DBV virion is dependent on actin filaments (26), the overall scenario by which DBV enters host cells remains to be explored, especially actin’s intermodulatory mechanism during virus entry.

This study aimed to investigate the biological significance of the interaction between a viral protein and actin. We report that the interaction between NSs and actin leads to actin rearrangement, which inhibits virus adsorption and entry. We further detail the molecular mechanisms of NSs-inhibited virus entry. Finally, we explore the role of NSs in other viral infections also relying on clathrin-mediated endocytosis to enter the cell. The new findings and the explanation of the mechanism provide a theoretical basis for improving the infection mechanism of DBV and searching for potential antiviral targets.

RESULTS

DBV NSs interacts with actin.

In recent years, proteomics-based mass spectrometry has been widely used in virology research (27–29). To gain insight into the molecular mechanism of DBV infection, we used affinity tag-purification mass spectrometry (AP-MS) to identify the host proteins that interact with DBV NSs. As shown in Fig. 1A, 293T cells were transfected with VR1012 vector plasmid (as a control group) and VR1012-NSs-HA recombinant plasmid. After 48 h, the cells were collected, and antihemagglutinin (anti-HA) affinity matrix was used to enrich the interaction protein. Then, trypsin digestion, dimethyl labeling, desalination, and mass spectrometry were performed as described in Materials and Methods. We identified 122 high-confidence protein-protein interactions (PPIs) with fold change values greater than 10 based on relative protein abundance. To understand the cellular processes and pathways in which NSs is involved, we performed Gene Ontology (GO) analysis. The biological processes identified mainly include structural constituents of ribosome, cell adhesion molecule binding, and actin filament binding. Among these, NSs had strong interaction with proteins such as RPS3A, RPS12, ACTN4, HNRNPK, GSN, and ACTN4 (Fig. 1B). In order to more intuitively observe the host proteins interacting with NSs, we also generated an NSs protein-human protein interaction map and classified and displayed 18 kinds of various actin cytoskeleton-related proteins (Fig. 1C, yellow circles). Actin is the most abundant protein in the cytoskeleton, which regulates various stages of the viral life cycle (30–32). To independently validate the interaction, we performed a coimmunoprecipitation (co-IP) experiment. Endogenous actin coprecipitated only in the presence of NSs, and reverse coimmunoprecipitation also confirmed the interaction between them, which was consistent with our AP-MS results (Fig. 1D and E). Moreover, the proximity ligation assay (PLA) of DBV-infected HeLa cells also showed an interaction between the virally produced NSs protein and endogenous actin (Fig. 1F). PLA was used to detect the interaction between low-expression proteins and did not need overexpression protein and higher specificity than co-IP. As previously described (33), adjacent red spots indicate an interaction between the two proteins. These results show that DBV NSs interacts with actin.

FIG 1.

NSs interacts with actin. (A) A schematic representation of the AP-MS approach for identifying DBV-human PPI in 293T cells. (B) GO of proteins after mass spectrometry data screening in the NSs group. The ribbons indicate which gene belongs to which category. The middle box represents logarithm from fold change (LogFC). The color of each LogFC bar corresponds with the LogFC value. (C) Host protein network map of NSs interactions. The blue represents the interacting host proteins, the red represents NSs, and the yellow represents actin filament-related protein. (D) Coimmunoprecipitation of NSs-HA with endogenous actin. 293T cells were transfected with 6 μg NSs-HA, and whole-cell lysates were pulled down by an anti-HA affinity matrix, followed by immunoblotting with the indicated antibody. (E) 293T cells were transfected with 3 μg NSs-HA and 3 μg actin-Flag, and whole-cell lysates were pulled down by an anti-Flag affinity matrix, followed by immunoblotting with the indicated antibody. (F) HeLa cells infected with DBV at an MOI of 1 for 24 h were examined by a Duolink proximity ligation assay. Each bright red spot represents an interaction, and the nucleus was stained with DAPI (bar = 20 μm).

NSs disrupts the filamentous structure of actin.

Actin assumes various structural forms in cells, such as filopodia, lamellae, and stress fibers, which regulate the virus life cycle (22, 34). To explore the relationship between DBV and actin, we first evaluated the level of F-actin in the cells by flow cytometry. 293T cells were infected with DBV at a multiplicity of infection (MOI) of 1. At 3 h, 6 h, 12 h, 24 h, 36 h, and 48 h after infection, the cells were collected, permeated, blocked, and stained with fluorescein isothiocyanate (FITC)-phalloidin, which only binds to F-actin. The fluorescence intensity was used to characterize the amount of F-actin. We found that DBV infection did not affect the content of F-actin in 293T cells (Fig. 2A). Many viruses induce a rearrangement of the actin network. For example, Lv et al. found that the porcine hemagglutinating encephalomyelitis virus (PHEV) causes an actin cytoskeletal rearrangement to promote its invasion (35). In addition, Zheng et al. showed that HSV-1 infection induces the cytoskeleton rearrangement, which facilitates virus entry into neuronal cells (36). To determine whether DBV infection induces the cytoskeleton rearrangement, HeLa cells were transfected with green fluorescent protein (GFP)-actin recombinant plasmids and infected with DBV (MOI = 1) for 48 h and then fixed, permeabilized, blocked, probed with specific antibodies, and detected by confocal microscopy. Compared with mock groups, the morphological structure of actin in DBV-infected cells was a point-like distribution. In addition, the colocalizations of the virally produced NSs protein and GFP-actin were confirmed using ImageJ software (Fig. 2B).

FIG 2.

NSs disrupts the formation of F-actin. (A) 293T cells infected with DBV (MOI = 1). The cells were collected at 3 h, 6 h, 12 h, 24 h, 36 h, and 48 h and stained with FITC-phalloidin, and the fluorescence intensity was analyzed by flow cytometry. (B) HeLa cells were infected with DBV (MOI = 1), and then the cells were fixed, permeabilized, blocked, and probed with specific antibodies at the indicated times of infection. The actin morphological structure was detected by confocal microscopy. (C) 293T cells were transfected with plasmids encoding NSs-HA and collected at 12 h, 24 h, and 48 h, and F-actin was detected as described for panel A. (D) HeLa cells were transfected with NSs-HA recombinant plasmids and detected by confocal microscopy. DAPI was used to label nuclei. The colocalization of NSs and actin in the white line region was analyzed using ImageJ software. The blue arrows indicate the colocation peak formed by NSs IBs and actin (P, Pearson correlation coefficient value of the whole picture; bar = 20 μm, n.s.).

Next, we investigated the effect of NSs overexpression on actin quantity and arrangement. 293T cells and HeLa cells were transfected with NSs-HA or with an empty vector as a control for 12 h, 24 h, and 48 h, and then fixed, stained with FITC-phalloidin, and analyzed by flow cytometry. We found no significant change in the amount of F-actin in 293T (Fig. 2C) and HeLa (data not shown) cells. In addition, HeLa cells were transfected with NSs-HA recombinant plasmids. At 6 h,12 h, and 24 h after transfection, the changes in the actin cytoskeleton were assessed by confocal microscopy. The immunofluorescence assay results showed that the morphological structure of actin in the NSs overexpression group was characterized by a point-like distribution state at different times, which was significantly different from actin forms in the control group. In addition, we analyzed the colocalization of NSs and actin using ImageJ software, which was consistent with previous results (Fig. 2D). These data reveal that DBV and NSs induced changes of actin morphological structure but did not affect the content of F-actin in cells.

The C terminus of NSs binds to actin.

Moriyama et al. found that the NSs-P66/69A mutant and NSs-V21/23A mutant resulted in the absence of IB formation in the cytoplasm (37). Liu et al. found that the N-terminal truncations NSs-△N35-HA and NSs-△N45-HA no longer formed IBs (16). To further explore whether NSs-induced actin rearrangement is IB-dependent, HeLa cells were cotransfected with VR1012, NSs-WT-HA, NSs mutants, N-terminal truncations, and C-terminal truncation recombinant plasmids. The immunofluorescence assay results showed that although mutant and truncated NSs still colocate with actin, there is no significant change in the morphological structure of actin in the mutants compared with the control group. Consistent with the previous results, wild-type NSs altered the actin morphological structure, indicating that the destruction of actin morphological structure is IB-dependent. In addition, the colocalizations of these proteins and actin were analyzed using ImageJ software, and the black arrow indicates that the colocalization of NSs-ΔC50-HA and actin was weakened (Fig. 3A).

FIG 3.

The C terminus of NSs interacts with actin. (A) HeLa cells were cotransfected with the indicated plasmids. After 24 h, the cells were fixed, permeabilized, blocked, and probed with specific antibodies. The morphology structure of actin was observed by confocal microscopy. The colocalization of NSs, mutants, and truncates with actin was analyzed with ImageJ software. (B to D) 293T cells were transfected with the indicated plasmids (3 μg) and GFP-actin (3 μg) recombinant plasmids and collected after 48 h. Interactions between these proteins and actin were detected by immunoprecipitation (bar = 20 μm).

Subsequently, we detected whether the interaction between NSs and actin was IB-dependent using coimmunoprecipitation. 293T cells were transfected with VR1012, NSs-HA, NSs-P66/69A-HA, NSs-V21/23A-HA, and GFP-actin recombinant plasmids. We found that these two mutants still interacted with actin, suggesting that the interaction between NSs and actin was not IB-dependent (Fig. 3B). To clarify the interaction regions between NSs-HA and actin, we used coimmunoprecipitation to analyze the interaction between NSs truncations and actin. We found that the N-terminally truncated NSs variants still interacted with actin (Fig. 3C), whereas, the C-terminally truncated NSs variant interaction with actin was significantly reduced, indicating that the C terminus of NSs may be an important binding region of actin (Fig. 3D).

NSs inhibits virus adsorption and entry.

Liu et al. revealed that actin is necessary to transport DBV to the cell surface and for entry (26). To evaluate the effect of NSs on the adsorption of DBV, HeLa cells were transfected with VR-1012, NSs-HA, NSs-P66/69A-HA, and NSs-ΔC50-HA. At 24 h after transfection, the cells were infected with DBV (MOI = 1) and adsorbed for 30 min at 4°C, and the levels of viral RNA in cells were measure by reverse transcription-quantitative PCR (qRT-PCR). Compared with the control group, the abundance of RNA was reduced by 50% in the NSs expression group and by about 70% in the NSs-P66/69A-HA group. The RNA abundance in the NSs-ΔC50-HA group was reduced by about 60% (Fig. 4A). To determine whether the inhibition of DBV adsorption on the cell surface was related to NSs-induced actin dysfunction, the adsorption capacity of DBV was detected after cotransfection of NSs-HA and GFP-actin. We found that, compared with the control group, the abundance of DBV was reduced by 30% after the overexpression of NSs and actin. Compared with the overexpression of NSs only, the RNA abundance of DBV was increased. In other words, overexpression of actin can reverse the inhibition of DBV adsorption caused by NSs, suggesting that the inhibition of DBV adsorption is indeed caused by the interaction with actin (Fig. 4B). To evaluate the effect of NSs on the entry of DBV, HeLa cells were transfected with VR-1012, NSs-HA, NSs-P66/69A-HA, and NSs-ΔC50-HA. Compared with the control group, the RNA abundance of DBV in the NSs-P66/69A-HA group decreased by about 70%, and that in the NSs-ΔC50-HA decreased by about 55% (Fig. 4C). Similarly, after cotransfection of NSs-HA- and GFP-actin-carrying plasmids, we found that, compared with the overexpression of the NSs group, the RNA abundance of DBV was increased by 20% in the NSs and actin group, suggesting that the DBV entry inhibition is also caused by the interaction with actin (Fig. 4D). We found that mutated and truncated NSs had a stronger inhibition on the adsorption and entry of virus. As expected, the IBs played an important role in the function of NSs. We suspect that this may be due to changes in the regulatory function of mutant NSs on proteins related to the actin cytoskeleton or other virus adsorption and entry of regulatory genes. This may also be because mutated NSs leads to the formation of a state that was not conducive to virus infection, thus enhancing the inhibition of virus adsorption and entry. This still needs to be explored.

FIG 4.

NSs inhibits virus adsorption and entry. (A and C) HeLa cells were transfected with VR1012, NSs-HA, NSs-P66/69A-HA, and NSs-ΔC50-HA. After 24 h, the cells were infected with DBV (MOI = 1) for 30 min at 4°C. The cells were washed with ice-cold PBS three times to remove unbound viruses. Then, some of the cells were used to extract the total RNA to evaluate the effect of NSs on virus adsorption. The remaining cells were cultured in fresh serum-free DMEM and subsequently shifted to 37°C to allow virus internalization. After 1 h, the cells were washed with ice-cold PBS three times to remove the noninternalized virions. RNA was extracted for qRT-PCR. The protein expressions of NSs-HA, NSs-P66/69A-HA, and NSs-ΔC50-HA were tested by WB. (B and D) HeLa cells were transfected with VR1012, NSs-HA, green fluorescent protein (GFP)-actin, and NSs-HA +GFP-actin encoding plasmids, and infected with DBV (MOI = 1), and the cells were adsorbed at 4°C for 30 min. The unadsorbed virus particles were eluted with PBS and then transferred to a 37°C incubator for further culturing for 1 h; RNA was extracted for qRT-PCR. The expression of NSs-HA and GFP-actin in cells was detected by WB. (E) 293T cells were transfected with NSs-HA plasmid, infected with DBV, and collected. The expression of NP was detected by WB. The grayscale value of NP was analyzed with ImageJ and Prism software. (F) HeLa cells were transfected with NSs-HA plasmid and infected with DBV. The fluorescence intensity of NP was detected by confocal microscopy. (G) 293T cells and HeLa cells were transfected with NSs plasmid or infected with DBV, and CCK8 detected the viability of the cells (bar = 20 μm, *, P < 0.05; **, P < 0.01; ***, P < 0.001).

In addition, to prove the effect of NSs on the viral protein expression of DBV, 293T cells were transfected with NSs-HA plasmid for 24 h and then infected with DBV (MOI = 1) for 24 h or 48 h. Western blotting (WB) detected the expression of NP of DBV. The results showed that NP expression was reduced by 30% at 24 h and 40% at 48 h in the overexpressed NSs group compared with the control group. Thus, NSs reduced viral protein expression in a time-dependent manner, indicating that NSs inhibited virus life cycle processes (Fig. 4E). In addition, immunofluorescence was used to detect the expression of DBV protein in the overexpressed NSs cells, and the results showed that the fluorescence intensity of NP in the overexpressed NSs cells was significantly lower than that in the nontransfected NSs cells, suggesting that the overexpression of NSs could inhibit the expression of DBV protein (Fig. 4F). The state of the cell will also affect the entry and replication of the virus. Next, the cell proliferation and cytotoxicity (CCK-8) kit was used to detect whether overexpression of NSs and DBV infection affected cell viability. The results showed that neither the overexpressed protein nor the virus infection affected the cell activity in 24 h (Fig. 4G).

NSs inhibits virus entry by regulating clathrin-mediated endocytosis.

Liu et al. found that DBV infection leads to clathrin recruitment on the cell surface, which is regulated by actin (26). Therefore, we further explored the relationship between NSs and clathrin. We first synthesized the recombinant plasmid of clathrin light chain B (CLTB)-GFP. CLTB is a light chain of the clathrin complex, which can be used to track clathrin. HeLa cells were cotransfected with VR1012 and CLTB-GFP or NSs-HA and CLTB-GFP recombinant plasmids to further explore the relationship between NSs and clathrin aggregation. After transfection (24 h), the cells were infected with quantum dot-labeled DBV (QD-DBV) (the QD-DBV appears as red fluorescent dots in Fig. 5A). After 1 h, the accumulation of CLTB (green point) on the cell surface was detected by confocal microscopy. The results of immunofluorescence and aggregate counting showed that the fluorescence amount of CLTB-GFP on the cell surface was significantly reduced in the NSs overexpression group compared with the control group, suggesting that NSs could reduce the clathrin aggregation induced by DBV infection, thereby inhibiting the entry of the virus (Fig. 5A and B).

FIG 5.

NSs hijacks CLTB and inhibits clathrin aggregation on the cell surface. (A) HeLa cells were transfected with VR1012, NSs-HA, and CLTB-GFP recombinant plasmids and infected with QD-DBV (red point). After incubation at 37°C for 1 h, the accumulation of clathrin (the orange arrow) on the cell surface was detected by confocal microscopy (The white arrow indicates the IBs structure formed by the colocation of CLTB and NSs). (B) Statistical analysis of the number of CLTB-GFP aggregates. (C) HeLa cells were transfected with VR1012 or NSs-HA recombinant plasmids, and the colocalization of NSs-HA and CLTB was observed by confocal microscopy. The colocalizations of NSs-HA and CLTB were analyzed by ImageJ software. (D) HeLa cells were transfected with a CLTB-GFP plasmid. At 24 h after transfection, the cells were infected with DBV (MOI = 1). The colocalization of NSs and CLTB was observed by confocal microscopy. The colocalizations of the virally produced NSs protein and CLTB in the white line region was analyzed using ImageJ software. The blue arrow indicates the colocation peak formed by NSs IBs (P: Pearson correlation coefficient value of the whole picture, bar = 20 μm, **, P < 0.01).

NSs can form IBs in the cytoplasm, and we speculated whether NSs and CLTB colocalize in IBs, affecting the function of clathrin. To verify this hypothesis, HeLa cells were transfected with VR1012 or NSs-HA recombinant plasmids. After 24 h, the colocation between NSs and CLTB was detected by confocal microscopy. We found that NSs and CLTB colocalized in IBs, which hijacked CLTB into IBs, preventing clathrin aggregation on the cell surface and inhibiting viral entry (Fig. 5C). We also examined the association between DBV infection and CLTB and found that NSs proteins produced by viral infection colocalize with CLTB in IBs, consistent with expectations (Fig. 5D). These data suggest that NSs inhibits clathrin aggregation on the cell surface, which inhibits viral entry.

NSs downregulates the expression of CLTB through the K48-linked ubiquitin-proteasome pathway.

To investigate the relationship between the changes in protein expression levels of NSs and CLTB, HEK293T cells and HeLa cells were infected with DBV or transfected with NSs-HA or an empty vector as a control. After 24 h, the level of CLTB was detected by WB. The protein level of CLTB was significantly decreased by 50% in DBV-infected and overexpressed NSs 293T cells (Fig. 6A and B) and HeLa cells (data not shown). These data revealed that DBV infection and the NSs could downregulate the expression of CLTB. Fan et al. found that mortalin downregulates CLTC through the proteasome degradation pathway (38). Therefore, the proteasome inhibitors MG132 and PS341 were employed to determine whether DBV infection and NSs degraded CLTB through the proteasomal-dependent pathways. We found that the protein level of CLTB almost did not change in the MG132 and PS341 pretreatment group but was significantly reduced in the virus-infected and NSs overexpression groups (Fig. 6C and D). Taken together, these results indicated that the degradation of CLTB protein was blocked by treatment with the proteasome inhibitor and that NSs could downregulate CLTB through the protease degradation pathway. Furthermore, NSs can form IBs in the cytoplasm, and we speculate whether the degradation of CLTB by NSs was IB-dependent. To verify this hypothesis, 293T cells were transfected with VR1012, NSs-HA, NSs-P66/69A-HA, and NSs-ΔC50-HA. The level of CLTB was detected by WB. We found that the expression of NSs-P66/69A-HA and NSs-ΔC50-HA weakened the inhibitory effect on CLTB; in particular, the overexpression of NSs-ΔC50-HA had almost no effect on the expression of CLTB compared with the overexpressed NSs group, indicating that the degradation of CLTB by NSs was IB-dependent (Fig. 6E).

FIG 6.

NSs downregulates CLTB through the proteasomal degradation pathway. (A) 293T cells were infected with DBV, and the expression of CLTB was detected by WB. (B) 293T cells were transfected with VR1012 and NSs-HA plasmid, and the expression of CLTB was detected by WB. (C) 293T cells were infected with DBV, and MG132 and PS341 inhibitors were added 12 h before sample collection. The expression of CLTB was detected by WB. (D) 293T cells were transfected with NSs-HA plasmid, and MG132 and PS341 inhibitors were added 12 h before cell collection. Cells were collected for 24 h, and the expression of CLTB was detected by WB. (E) 293T cells were transfected with VR1012, NSs-HA, NSs-P66/69A-HA, and NSs-ΔC50-HA encoding plasmids, and the expression of CLTB was detected by WB. The gray value of CLTB was analyzed using ImageJ and Prism software. (F) 293T cells were cotransfected with 4 μg VR1012 + 2 μg Ub-K48-Myc, 2 μg VR1012 + 2 μg CLTB-Flag + 2 μg Ub-K48-Myc, and 2 μg CLTB-Flag + 2 μg Ub-K48-Myc + 2 μg NSs-HA/NSs-P66/69A-HA recombinant plasmids, and coimmunoprecipitation was used to detect whether the degradation of CLTB depended on the ubiquitination of K48 (*, P < 0.05; **, P < 0.01; ***, P < 0.001).

Different methods of ubiquitination are involved in the regulation of different biological processes. The ubiquitination of K48 and K11 plays an important role in protein degradation. To illustrate which type of ubiquitin is involved in NSs-induced degradation of CLTB, 293T cells were transfected with either K48- or K11-linked ubiquitin plasmids followed by NSs-HA and NSs-P66/69A-HA overexpression. We found that the K48 ubiquitination of CLTB can be detected after overexpression of NSs, while the ubiquitination was not detected after overexpression of NSs-P66/69A-HA, which are results consistent with previous studies (Fig. 6F). The K11 ubiquitination of CLTB was not detected after overexpression of NSs-HA and NSs-P66/69A-HA (data not shown). Collectively, our results demonstrate that NSs triggers CLTB degradation through K48-linked ubiquitination modification.

NSs interferes with other virus adsorption and entry mediated by clathrin.

The data from the assays described above indicated that NSs could degrade CLTB via the proteasomal pathway and suppress the entry of DBV mediated by clathrin-mediated endocytosis (CME). We then investigated whether NSs interferes with the adsorption and entry of other viruses, namely, EV7 and PR8, which also rely on CME to enter host cells (39, 40). In terms of adsorption capacity, compared with the control group, the RNA abundance of EV71 and PR8 decreased by about 20% and 5%, respectively, in the NSs overexpression group (Fig. 7A and C). In terms of the entry process, the RNA abundance of EV71 and PR8 was reduced by 60% and 25%, respectively (Fig. 7B and D). The expression of NSs was detected by WB. Influenza viruses can enter host cells through multiple endocytic pathways, including CME, caveolin-mediated endocytosis, and clathrin- or caveolae-independent pathways of micropinocytosis. This versatility may be why NSs has little effect on the adsorption and entry of PR8 (41–44). In general, these results revealed that NSs interferes with the other virus adsorption and entry and that this process is mediated by clathrin.

FIG 7.

Effects of NSs on the other virus adsorption and entry mediated by clathrin. (A and C) RD (A) and A549 (C) cells were transfected with an NSs-HA encoding plasmid. After 24 h, the cells were infected with EV71 or PR8 with an MOI of 1. After 4°C of adsorption for 1 h, the cells were collected, and the amount of viral RNA was detected by qRT-PCR. The expression of NSs-HA was detected by WB. (B and D) RD and A549 cells were transfected with an NSs-HA encoding plasmid. After 24 h, the cells were infected with EV71 or PR8 with and MOI of 1. After adsorption for 1 h at 4°C, the unadsorbed virus particles were eluted with PBS and then transferred to a 37°C incubator for further culturing for 1 h; the cells were collected, and the amount of viral RNA was detected by qRT-PCR. The expression of NSs-HA was detected by WB (n.s., P > 0.05; **, P < 0.01).

DISCUSSION

Actin is a cytoskeletal protein highly conserved among eukaryotes that is involved in the virus life cycle (45–47). DBV is an emerging phlebovirus that causes multiorgan failure with a high fatality rate in humans (2, 48). However, there is a pressing need to understand DBV pathogenesis due to the lack of therapies and vaccines. Although actin has been reported to play a role in DBV infection, the specific mechanisms have remained largely uncertain (26). In this study, we showed that actin is the key host-binding protein of NSs. The interaction between NSs and actin inhibits viral adsorption and entry by inducing rearrangement of the actin network. In addition, NSs inhibits clathrin-mediated endocytosis by downregulating CLTB, inhibiting viral entry (Fig. 8).

FIG 8.

Proposed model for DBV NSs-inhibited viral adsorption and entry. In the left panel, DBV travels along the filopodia to the cell surface (A) and then induces clathrin aggregation on the cell surface at the endocytosis site (B), forming clathrin-coated Pit, which facilitates entry into host cells. In the right image, the arrangement of actin is destroyed in NSs-transfected cells, which causes the decrease of filopodia (A′), and NSs degrades CLTB by the proteasome pathway and inhibits clathrin accumulation on the cell surface (B′), inhibiting viral adsorption and entry.

The first step of viral infection is the binding to molecules on the cell surface and the transport of these complexes to the site of endocytosis, where the virus enters the cell via endocytosis (18, 23, 26). Human papillomavirus type 31 (HPV31), human herpesvirus 8 (HHV-8), and DBV combine with filopodia “surfing” to aggregate on the surface of the cell body, reach the high endocytosis site, and then enter the cell (30, 49). We discovered that infection with DBV or expression of NSs altered the morphological structure of actin, presenting a point-like distribution state at different times (Fig. 2B and D). However, the changes in the morphological structure of actin observed in this study are different from those observed during other viral infections, which may be related to the IB structure induced by NSs. Our data also showed that the punctate distribution of actin is IB-dependent (Fig. 3A). Filopodia were reported to play vital roles in viral infection, including initial viral attachment, cell surfing, internalization, virus release, and spread to other cells in a form that would avoid the host immune system (50). Viruses can activate the formation of filopodia to enhance infection of host cells (30, 51). It has been reported that virus transport along filopodia during the attachment results in virus internalization in many viruses (22, 49, 52). Many signaling pathways are responsible for virus-induced filopodia formation, including tyrosine kinase and phosphoinositide 3 (PI3)-kinase signaling pathways, cdc42 GTPases, and Rac1 pathways (49, 53, 54). It is necessary to further explore the signaling pathways for DBV induced from filopodia formation, which may lead to the development of new antiviral therapies that block the signaling pathways used by the virus to trigger filopodia formation.

Viruses utilize various endocytosis mechanisms to enter host cells, and CME is commonly utilized as the uptake mechanism for the virus, such as Rabies virus (RABV), stomatitis virus (VSV), dengue virus, and influenza A virus (IAV) (31, 55, 56). DBV also utilized the CME machinery to invade host cells. In the process of endocytosis, actin is essential for membrane invagination and rupture (57, 58). In this study, we found that NSs inhibits clathrin accumulation on the cell surface, which is the first step in endocytosis, thereby inhibiting the adsorption of the virus. In addition, NSs traps CLTB in IBs and prevents its function, which may also inhibit the entry of the virus (Fig. 5 and 6). Further studies showed that NSs induced CLTB degradation through the K48-linked ubiquitin-proteasome pathway, which could negatively regulate CME (Fig. 6F). Importantly, this antiviral mechanism is widely applicable to other viruses that use the same CME pathway (38), such as EV71 and PR8 (Fig. 7). Lemasson et al. found that HBZ protein of human T-cell leukemia virus type 1 (HTLV-1) interacts with cell transcription factor CREB to inhibit the transcription of HTLV-1 and HTLV-1 antigen expression, which may be a significant advantage for the virus in infected cells by preventing its detection through a cytotoxic T-lymphocyte response (59). Furthermore, HTLV-1-encoded p30II can restrict its replication by a novel posttranscriptional mechanism and become a transiently dormant state to evade the host’s immune surveillance (60). Thus, the interaction between viral proteins and host proteins can negatively regulate viral infection through various mechanisms, which may be a strategy for viruses to escape the host immune surveillance.

Viruses are a product of coevolution with their hosts, which requires assistance from host cells to infect and transfer viral genes into host cells (18). To prevent other viruses from entering the cell and consuming the enzymes, proteins, and other substances needed for DBV infection, the NSs interacts with actin to play an antiviral role and prevent other viruses from infecting the host cell again. In summary, we studied the interaction function and molecular mechanism between NSs and actin and revealed the key role of actin and CME in the virus life cycle, which may help identify new antiviral targets to inhibit viral replication and also reveal the actin key role in the immune response.

MATERIALS AND METHODS

Reagents and chemicals.

Anti-HA affinity matrix (catalog no. 11815016001) was purchased from Roche. Anti-Flag affinity matrix (catalog no. A2220) and Duolink proximity ligation assay (PLA) (catalog no. DUO92006) were purchased from Sigma. Primary antibodies to anti-HA (catalog no. AE008) were purchased from ABclonal Technology. The primary antibody to anti-Flag (catalog no. 80010-1-RR), anti-actin (catalog no. 66009-1-Ig), and anti-CLTB (catalog no. 10455-1-AP) were purchased from Proteintech Group. DAPI (4′,6-diamidino-2-phenylindole) (catalog no. H-1200) and the CCK-8 cell proliferation and cytotoxicity assay kit (catalog no. CA1210) were acquired from Solarbio Life Science. The enhanced chemiluminescence (ECL) reagent (catalog no. SQ201) was obtained from EpiZyme. The sulfo-NHS-LC-biotin (catalog no. A8003) was obtained from APE×BIO, and the SA-QDs 605 (catalog no. QS605) was purchased from Wuhan Jiayuan Quantum Dots Corporation (China).

Cells and viruses.

HeLa cervical carcinoma cells (HeLa cells) and HEK293T human embryonic kidney 293 cells (293T cells) were obtained from the American Type Culture Collection (ATCC) and cultured at 37°C in 5% CO₂. Cells were maintained in Dulbecco’s modified Eagle’s medium (DMEM) supplemented with 10% fetal bovine serum (FBS) and 1% penicillin-streptomycin (100 IU/mL). DBV strain HB29 was obtained from the Dexin Li (Chinese Center for Disease Control and Prevention). The virus was propagated in African green monkey kidney cells (Vero cells) for 7 days and obtained by collecting culture supernatant. Human enterovirus 71 (EV71) and influenza virus A/PR8/34 (PR8) were obtained from the Fei Guo (Chinese Academy of Medical Sciences). The EV71 was propagated in human malignant embryonic rhabdomyoma cells (RD cells) for 48 h before the cells and supernatant were collected, repeatedly frozen and thawed 3 times, and centrifuged at 3,000 rpm for 10 min. The precipitate was discarded. The PR8 was cultured in Madin-Darby canine kidney cells (MDCK cells) for 72 h and obtained by collecting the culture supernatant. The titers of all viruses were determined by 50% tissue culture infective dose (TCID₅₀), and the subpackaged viruses were stored at −80°C.

Preparation of biotin quantum dot-labeled DBV.

A total of 10⁷ PFU/mL DBV was incubated with sulfo-NHS-LC-biotin reagent (0.5 mg/mL) at 37°C for 2 h to produce biotin-labeled DBV (Bio-DBV). The Vero cells were infected with Bio-DBV at 4°C for 25 min and then washed 3 times with precooled 0.1% bovine serum albumin (BSA) to remove the nonadsorbed virus. Then, 5 × 10⁻⁹ M SA-QDs 605 was added to the Vero cells, incubated at 4°C for 15 min, and washed 3 times with 0.1% BSA to produce quantum dot-labeled DBV (QD-DBV).

Coimmunoprecipitation.

First, 3 μg VR1012, NSs-HA, NSs-P66/69A-HA, NSs-V21/23A-HA, NSs-ΔN35-HA, NSs-Δ N45-HA, and NSs-ΔC50-HA, and 3 μg GFP-actin plasmid with 18 μL 1 mg/mL polyethylenimine (PEI) reagent were cotransfected into 293T cells (5 × 10⁶ cells/mL). After 48 h, the cells were suspended with phosphate-buffered saline (PBS) and centrifuged at 6,000 rpm for 5 min to collect cell precipitate. The cells were lysed in 800 μL lysis buffer (50 mM Tris-HCl, pH = 7.5, 150 mM NaCl, and 0.5% Triton-100), and the lysates were incubated with anti-HA affinity matrix or anti-Flag M2 affinity gel at 4°C for 6 h. The lysates were washed six times with PBS, eluted with glycine buffer (pH = 2), boiled in SDS loading buffer, and analyzed by Western blotting (WB). All co-IP experiments were performed in 293T cells with a total amount of 6 μg transfection plasmid. Unless otherwise specified, each plasmid was transfected with 3 μg.

Western blotting.

At the indicated times of infection and transfection, cells were washed three times with PBS and lysed in RIPA buffer. The protein samples were separated by SDS-PAGE, transferred to nitrocellulose, probed with the indicated primary antibodies, and exposed to species-specific horseradish peroxidase (HRP)-conjugated secondary antibodies. Immunoreactive bands were visualized by enhanced chemiluminescence. The gray value of protein was analyzed by ImageJ software and Prism. Tubulin was used as a loading control.

DBV NSs–host proteome interaction data acquisition and analysis.

293T cells (5 × 10⁷ cells/mL) were transfected with VR1012 vector plasmid (25 μg, as control group) or VR1012-NSs-HA recombinant plasmid (25 μg). After 48 h, the cells were collected and anti-HA affinity matrix was used to enrich the interaction protein. The 230-μL co-IP eluent was added to 270 μL 12 M urea (dissolved in 100 mM NaHCO3), followed by a final concentration of 100 mM dithiothreitol (DTT), and incubated at room temperature for 1 h. We then added 4 μL 1 M iodoacetamide (IAA), mixed it, and incubated the mixture in dark for 1 h. To terminate the reaction, 4 μL 1 M DTT was added, mixed, and incubated for 45 min. The solution was transferred to a 10-kDa concentrator, and 50 mM NaHCO₃ was added, centrifuged at 12,000 rpm for 20 min, and washed 3 to 4 times until the solution volume in the concentrator remained at 50 μL, and 1 μL trypsin was added for digestion at 37°C overnight. For the dimethyl treatment, 2.5 μL 1 M NaCNBH3 was added to the solution, followed by 0.5 μL CH₂O in the control group and 0.92 μL CD₂O in the NSs-expressing group, which were incubated at 37°C for 1 h, and 2 μL 500 mM NH₄HCO₃ was added for 10 min to stop the reaction. For the desalting treatment, we added 2.5 μL formic acid. Two groups of samples were mixed and divided into two parts and added into the activated ZipTip for repeated blowing 50 times. The product was washed twice with wash solution (0.1% trifluoroacetic acid [TFA]); 10 μL elution solution (0.1% TFA/50% acetonitrile (ACN)) was used to obtain the sample. Then, affinity tag-purification mass spectrometry (AP-MS) was used to identify the host proteins that interact with DBV NSs protein, and three biological replicates were used for each sample. The PPI map was drawn using Cytoscape. Gene Ontology (GO) function annotation was analyzed using Goplot.

Immunofluorescence staining and confocal microscopy.

At the indicated times of infection and transfection, HeLa cells were washed three times with PBS, fixed in 4% paraformaldehyde (PFA) for 15 min, permeabilized with 0.5% Triton X-100 for 10 min, blocked with 1% bovine serum albumin (BSA) for 20 min, and then probed with specific antibodies (dilution ratio of primary antibody, 1:100; secondary antibody, 1:500). Then, 10 μg/mL DAPI was used to label nuclei. Images were captured with a confocal laser scanning microscope system. The colocalization of signals from two channels was detected using ImageJ software.

qRT-PCR of viral RNA.

Total RNA from the cells was extracted using a QIAzol lysis reagent. The RNA was reverse-transcribed into cDNA using TransScript first-strand cDNA synthesis supermix according to the manufacturer’s instruction. S-segment-specific primers were used to determine the quantities of viral mRNAs. The relative amount of target gene mRNA was normalized to that of GAPDH mRNA in the same sample.

The primers’ sequences are as follows: DBV forward, 5′-GGG TCC CTG AAG GAG TTG TAA A-3′; DBV reverse, 5′-TGC CTT CAC CAA GAC TAT CAA TGT-3′; Probe, 5′-TexasRed-TTC TGT CTT GCT GGC TCC GCG C-BHQ-2-3′; EV71 forward, 5′-AAG CAC TTC TGT TTC CC-3′; EV71 reverse, 5′-ATT CAG GGG CCG GAG GA-3′; PR8 forward, 5′-TGC TTC AAA ACA GCC AAG TG-3′; PR8 reverse, 5′-GCC CAG TAC CTG CTT CTC AG-3′; GAPDH forward, 5′-GAA GGT GAA GGT CGG AGT C-3′; GAPDH reverse, 5′-GAA GAT GGT GAT GGG ATT TC-3′.

Assay of viral adsorption and entry.

HeLa cells were transfected with plasmids encoding NSs, NSs mutants, and NSs truncates. After 24 h, the cells were incubated with DBV at an MOI of 1 for 30 min at 4°C to allow virus attachment without internalization (61). The cells were washed with precooled PBS three times, and the unbound viruses were removed. Then, some of the cells were used to extract the total RNA to evaluate the effect of NSs on virus adsorption. The remaining cells were cultured in fresh serum-free DMEM and subsequently shifted to 37°C to allow virus internalization. After 1 h, the cells were washed with precooled PBS three times to remove the noninternalized virions.

Flow cytometry.

HeLa cells and 293T cells were washed with PBS and fixed in 4% PFA for 10 min. After being washed with PBS, the cells were permeabilized with 0.1% Triton X-100 for 10 min, washed again with PBS, and stained with 10 μM fluorescein isothiocyanate (FITC)-phalloidin-PBS for 40 min, and the unbound phalloidin conjugate was removed by several washes with PBS. The fluorescence was analyzed with a flow cytometer.

Proximity ligation assay.

HeLa cells were infected or mock-infected with DBV (MOI = 1) for 24 h, fixed with 4% PFA, permeabilized with 0.5% Triton X-100, and blocked with blocking buffer for 1 h. The cells in the mock-infection group were stained with anti-NSs and anti-actin primary antibody, and the infected cells were incubated with anti-NSs or anti-NSs and anti-actin antibodies (dilute 1:100 in 1% BSA). Then, all the cells were incubated with anti-mouse minus and anti-rabbit plus PLA reagents followed by ligation and amplification mix. Images were captured with a laser confocal laser scanning microscope system.

CCK-8 cell proliferation and cytotoxicity assay.

293T cells and HeLa cells (10⁵ cells/mL) were transfected with 200 ng VR1012 vector plasmid, transfected with 200 ng NSs-HA recombinant plasmid, or infected/mock-infected with DBV (MOI = 1). After 24 h, 10 μL CCK-8 reagent was added to the cells, the mixture was incubated at 37°C for 30 min, and the absorbance at 450 nm was measured using a microplate reader according to the manufacturer’s instructions.

Statistical analysis.

All results were analyzed with GraphPad Prism and are presented as means ± standard deviations (SD). Statistical significance was determined using Student’s two-tailed t test (not significant [n.s.], *, P < 0.05; **, P < 0.01; ***, P < 0.001).