DDX3X Is Hijacked by Snakehead Vesiculovirus Phosphoprotein To Facilitate Virus Replication via Stabilization of the Phosphoprotein

ABSTRACT

Asp-Glu-Ala-Asp (DEAD) box helicase 3 X-linked (DDX3X) plays important regulatory roles in the replication of many viruses. However, the role of DDX3X in rhabdovirus replication has seldomly been investigated. In this study, snakehead vesiculovirus (SHVV), a kind of fish rhabdovirus, was used to study the role of DDX3X in rhabdovirus replication. DDX3X was identified as an interacting partner of SHVV phosphoprotein (P). The expression level of DDX3X was increased at an early stage of SHVV infection and then decreased to a normal level at a later infection stage. Overexpression of DDX3X promoted, while knockdown of DDX3X using specific small interfering RNAs (siRNAs) suppressed, SHVV replication, indicating that DDX3X was a proviral factor for SHVV replication. The N-terminal and core domains of DDX3X (DDX3X-N and DDX3X-Core) were determined to be the regions responsible for its interaction with SHVV P. Overexpression of DDX3X-Core suppressed SHVV replication by competitively disrupting the interaction between full-length DDX3X and SHVV P, suggesting that full-length DDX3X-P interaction was required for SHVV replication. Mechanistically, DDX3X-mediated promotion of SHVV replication was due not to inhibition of interferon expression but to maintenance of the stability of SHVV P to avoid autophagy-lysosome-dependent degradation. Collectively, our data suggest that DDX3X is hijacked by SHVV P to ensure effective replication of SHVV, which suggests an important anti-SHVV target. This study will help elucidate the role of DDX3X in regulating the replication of rhabdoviruses.

IMPORTANCE Growing evidence has suggested that DDX3X plays important roles in virus replication. In one respect, DDX3X inhibits the replication of viruses, including hepatitis B virus, influenza A virus, Newcastle disease virus, duck Tembusu virus, and red-spotted grouper nervous necrosis virus. In another respect, DDX3X is required for the replication of viruses, including hepatitis C virus, Japanese encephalitis virus, West Nile virus, murine norovirus, herpes simplex virus, and severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2). Because DDX3X has rarely been investigated in rhabdovirus replication, this study aimed at investigating the role of DDX3X in rhabdovirus replication by using the fish rhabdovirus SHVV as a model. We found that DDX3X was required for SHVV replication, with the mechanism that DDX3X interacts with and maintains the stability of SHVV phosphoprotein. Our data provide novel insights into the role of DDX3X in virus replication and will facilitate the design of antiviral drugs against rhabdovirus infection.

INTRODUCTION

Asp-Glu-Ala-Asp (DEAD) box helicase 3 (DDX3) belongs to the DEAD box helicase family and is divided into two closely related forms, i.e., DDX3X and DDX3Y (1). DDX3Y is exclusively located on the Y chromosome (2), and DDX3X is located on the X chromosome (3) and usually named DDX3 (4). DDX3X, which is ubiquitously expressed in most tissues (5), is an RNA-binding protein that participates in all aspects of mRNA metabolism (6–9) and cellular processes, such as cellular stress responses (10) and cancer progression (11). In addition, DDX3X plays an important role in virus infection (12–14).

DDX3X is known to affect innate antiviral immunity by acting as a viral RNA sensor or mediator in the type I interferon (IFN) pathway (15–17). Numerous studies have also revealed that DDX3X acts as an essential cellular factor for the replication of different viruses. To date, DDX3X has been reported to be involved in the replication of viruses of 19 species from 12 families (13, 18). Some viruses hijack DDX3X for efficient replication either by promoting viral mRNA translation or by blocking the host innate antiviral immunity. These viruses comprise DNA viruses, including herpes simplex virus 1 (HSV-1) (19) and vaccinia virus (VACV) (20, 21), and RNA viruses, including human immunodeficiency virus type 1 (HIV-1) (13), hepatitis C virus (HCV) (13), Norwalk-like virus (NLV) (22), severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2) (23), West Nile virus (WNV) (13), Zika virus (ZIKV) (12), Japanese encephalitis virus (JEV) (24), Junín virus (JUNV) (25), and lymphocytic choriomeningitis virus (LCMV) (25). In contrast, several studies have demonstrated that DDX3X exerts antiviral activity against some viruses (26–31). DDX3X promotes the production of IFN-β to inhibit the replication of viruses, including Newcastle disease virus (NDV), duck Tembusu virus (DTMUV), and red-spotted grouper nervous necrosis virus (RGNNV) (29–31). DDX3X regulates stress granule (SG) formation through interaction with NS1 and NP proteins, thus inhibiting influenza A virus (IAV) replication (28). In addition, DDX3X inhibits infectious hematopoietic necrosis virus (IHNV) replication, although the specific mechanism remains unclear (18). Based on the important role of DDX3X in virus replication, increasing evidence suggests that DDX3X can be considered a potential broad-spectrum antiviral target (13, 32). However, the exact role of DDX3X in rhabdovirus replication remains to be elucidated.

Snakehead vesiculovirus (SHVV), which is a member of the family Rhabdoviridae, was isolated from diseased hybrid snakehead fish in China (33). The disease caused by SHVV has resulted in huge economic losses in snakehead fish culture (34). However, there are currently no specific antiviral drugs or vaccines available for SHVV infection. The genome of SHVV is composed of a negative-sense single-stranded RNA of approximately 11.5 kb that encodes five proteins, including nucleoprotein (N), phosphoprotein (P), matrix protein (M), glycoprotein (G), and large protein (L) (35). The P protein and the genomic RNA, N protein, and L protein form a ribonucleoprotein complex (NC) that functions in the transcription and replication of the viral genome. In this study, a combination of coimmunoprecipitation (Co-IP) and mass spectrometric analyses was applied to identify host factors that interact with SHVV P protein. We identified DDX3X as an interacting partner of SHVV P protein. Modulation of the expression of DDX3X via overexpression and knockdown strategies showed that DDX3X positively regulated SHVV replication. Moreover, the N-terminal and core regions of DDX3X were identified as the regions responsible for the DDX3X-P interaction, and overexpression of the core region of DDX3X inhibited SHVV replication by competing with the full-length DDX3X to interact with the P protein. Further studies revealed that DDX3X acted as a positive regulator of SHVV replication not by inhibiting IFN expression but by maintaining the stability of P protein to avoid its degradation via the autophagy-lysosome pathway. Our findings provide novel insights into the role of DDX3X in virus replication.

RESULTS

DDX3X interacts with SHVV P protein.

To identify cellular factors that interact with SHVV P, a Co-IP assay was performed with pEGFP-P-transfected channel catfish ovary (CCO) cells using rabbit anti-enhanced green fluorescent protein (EGFP) antibody. CCO cells transfected with pEGFP-N1 (empty vector) were used as control. The eluted samples were analyzed by Western blotting and silver staining (Fig. 1A). Further mass spectrometric analysis identified 25 potential interacting partners of SHVV P, which have been reported to be involved in virus replication (Fig. 1B). Among the 25 potential partners, DDX3X has been reported to regulate the replication of many viruses with different mechanisms (4, 12–14, 36), presenting an attractive target for the development of antiviral drugs (11, 32, 36). Moreover, the role of DDX3X in rhabdovirus replication has rarely been investigated. Therefore, DDX3X was selected for investigation of its role in SHVV replication in this study. To verify the interaction between DDX3X and SHVV P, a Co-IP assay was conducted. As shown in Fig. 1C, hemagglutinin (HA)-DDX3X was coimmunoprecipitated with Myc-P. Further immunofluorescence results demonstrated that DDX3X colocalized with SHVV P (Fig. 1D). Taken together, these results indicate that SHVV P interacts with DDX3X.

FIG 1.

DDX3X interacts with SHVV P protein. (A) CCO cells were transfected with pEGFP-N1 (control) or pEGFP-P, and the cells were collected at 24 h posttransfection. An IP assay was performed using anti-EGFP antibody. The eluted samples were detected by Western blotting (immunoblotting [IB]) and silver staining. (B) The samples from the experiment shown in panel A were analyzed by liquid chromatography-tandem mass spectrometry (LC-MS/MS), and the previously reported potential interacting proteins involved in virus replication were determined. (C) 293T cells were transfected with pCMV-Myc and pCMV-HA-DDX3X or pCMV-Myc-P and pCMV-HA-DDX3X. The cells were collected at 24 h posttransfection, followed by a Co-IP assay with anti-HA antibody. (D) 293T cells were transfected with pEGFP-P and pDsRED-N1, pEGFP-N1 and pDsRED-DDX3X, or pEGFP-P and pDsRED-DDX3X. pDsRED-N1 and pEGFP-N1 were empty vectors used as controls. At 24 h posttransfection, the cells were fixed and counterstained with the nuclear stain DAPI. The samples were analyzed using a confocal microscope.

SHVV infection upregulates DDX3X at an early stage of infection.

To determine the response of cellular DDX3X during SHVV infection, CCO cells were infected with SHVV, and the mRNA and protein levels of DDX3X were determined at different time points postinfection. The mRNA level of DDX3X was increased ~2-fold at 3 h postinfection (hpi), compared with that at 0 hpi, and gradually decreased to the normal level at 24 hpi (Fig. 2A). The protein level of DDX3X increased from 6 hpi, reached a peak at 12 hpi, and decreased to the normal level at 24 hpi (Fig. 2B). To confirm these results, we investigated the expression of DDX3X at different time points after SHVV infection using an immunofluorescence assay with anti-DDX3X antibody. The endogenous DDX3X expression increased gradually to a peak at 12 hpi and returned to the normal level at 24 hpi (Fig. 2C). Collectively, these data suggest that DDX3X is upregulated by SHVV infection at an early stage, followed by returning to the normal level at a later stage.

FIG 2.

SHVV infection upregulates DDX3X at an early stage of infection. CCO cells were infected with SHVV and harvested at 3, 6, 12, and 24 hpi. (A) The cellular mRNA levels for DDX3X were measured using qRT-PCR. (B) The protein levels of DDX3X were determined by Western blotting. β-Actin was used as the internal control. Quantification of DDX3X bands in the Western blot shown in panel B was performed using Image-Pro Plus v6.0, with normalization to β-actin levels. The value of the DDX3X protein band in cells at 0 hpi was set as 100. (C) CCO cells were infected with SHVV, fixed at 0, 3, 6, 12, and 24 hpi, and subjected to immunofluorescence analysis using confocal microscopy. DDX3X was detected with anti-DDX3X antibody. **, P < 0.01.

DDX3X is a proviral factor for SHVV replication.

To explore the role of DDX3X in SHVV replication, CCO cells were transfected with pCMV-HA (control) or pCMV-HA-DDX3X, followed by SHVV infection. The mRNA and protein levels of DDX3X were assessed at 24 hpi. The results showed that DDX3X was overexpressed in pCMV-HA-DDX3X-transfected cells, compared to the level in pCMV-HA-transfected cells (Fig. 3A). Further study suggested that overexpression of DDX3X promoted SHVV replication, through detection of the viral P protein in cells and viral titers in supernatants (Fig. 3B). To confirm the effect of DDX3X on SHVV replication, we synthesized three small interfering RNAs (siRNAs) for DDX3X (siDDX3X-22, siDDX3X-647, and siDDX3X-1337). The CCO cells were transfected with siRNA for NC (siNC) or three siRNAs targeting DDX3X, followed by SHVV infection. The knockdown efficiency of the three siRNAs was verified, and siDDX3X-647 exhibited the greatest knockdown efficiency (Fig. 3C). Moreover, our results showed that knockdown of DDX3X with siDDX3X-647 significantly suppressed SHVV replication (Fig. 3D). Collectively, these data indicate that DDX3X is a proviral factor for SHVV replication.

FIG 3.

DDX3X is a proviral factor for SHVV replication. CCO cells were transfected with pCMV-HA (empty vector) or pCMV-HA-DDX3X for 24 h. At 24 h posttransfection, CCO cells were infected with SHVV. (A) The mRNA and protein levels of DDX3X were detected by qRT-PCR and Western blotting at 24 hpi. (B) The viral titer in the supernatants and P protein levels in cells were measured using TCID₅₀ assays or Western blotting, respectively. (C and D) CCO cells were transfected with siNC, siDDX3X-22, siDDX3X-647, or siDDX3X-1337, followed by SHVV infection. The mRNA and protein levels of DDX3X were detected by qRT-PCR and Western blotting at 24 hpi (C). The viral titer in the supernatants and P protein levels in cells were measured using TCID₅₀ assays or Western blotting, respectively (D). **, P < 0.01.

The N-terminal and core domains of DDX3X are the regions responsible for its interaction with SHVV P.

As an RNA helicase of the DEAD box family, DDX3X is composed of a core helicase domain flanked by N- and C-terminal domains. The core helicase domain of human DDX3X comprises amino acids 168 to 582, while the flanking N- and C-terminal domains comprise amino acids 1 to 167 and amino acids 583 to 662, respectively (37). Sequence alignment showed that the amino acid sequence of DDX3X from channel catfish (Ictalurus punctatus) (GenBank accession number XP_017325032.1) exhibited 70.12% identity with the DDX3X of humans (Homo sapiens) (GenBank accession number O00571), indicating that DDX3X is a relatively conserved protein (Fig. 4A). To identify which domain of DDX3X is responsible for its interaction with SHVV P, a series of plasmids expressing truncated channel catfish DDX3X, i.e., DDX3X-N (N-terminal domain [amino acids 1 to 209]), DDX3X-Core (core domain [amino acids 210 to 627]), and DDX3X-C (C-terminal domain [amino acids 628 to 706]), were generated (Fig. 4B). A Co-IP assay indicated that both DDX3X-N and DDX3X-Core interacted with SHVV P, suggesting that DDX3X N-terminal and core domains were the regions responsible for the DDX3X-P interaction (Fig. 4C). To confirm this result, HA-tagged DDX3X-Core and DDX3X-N were coexpressed with Myc-tagged SHVV P in 293T cells. By using a Co-IP assay, we found that both DDX3X-N and DDX3X-Core were able to interact with the P protein (Fig. 4D). These results suggest that the N-terminal and core domains of DDX3X are the regions responsible for its interaction with SHVV P.

FIG 4.

The N-terminal and core domains of DDX3X are the regions responsible for its interaction with SHVV P protein. (A) Sequence alignment analysis between DDX3X from Ictalurus punctatus and DDX3X from Homo sapiens was conducted using ClustalW software (51). (B) The schematic shows the plasmids expressing different DDX3X truncations that were constructed to identify the domains essential for its interaction with SHVV P. (C and D) The plasmid pMyc-P was cotransfected with one of the plasmids expressing EGFP-tagged DDX3X truncations (C) or HA-tagged DDX3X truncations (D) into 293T cells. At 24 h posttransfection, cell lysate was collected and subjected to Co-IP with anti-EGFP or anti-HA antibody. The immunoprecipitates and cell lysates were then analyzed with anti-HA, anti-Myc, and anti-EGFP antibodies.

The DDX3X-P interaction is required for SHVV replication.

To determine whether DDX3X-N or DDX3X-Core could inhibit SHVV replication by competitively disrupting the interaction between the full-length DDX3X and SHVV P, CCO cells were transfected with pCMV-HA, pCMV-HA-DDX3X, pCMV-HA-DDX3X-Core, or pCMV-HA-DDX3X-N, followed by SHVV infection. The results showed that overexpression of the full-length DDX3X and DDX3X-N promoted SHVV replication, while overexpression of DDX3X-Core suppressed SHVV replication (Fig. 5A and B). Furthermore, the effects of DDX3X-Core and DDX3X-N on the interaction between the full-length DDX3X and SHVV P were investigated. We found that, in the presence of DDX3X-Core, P protein coimmunoprecipitated with less full-length DDX3X than in the absence of DDX3X-Core, while DDX3X-N did not significantly disrupt the interaction between the full-length DDX3X and SHVV P (Fig. 5C), suggesting that DDX3X-Core competed with the full-length DDX3X to interact with the P protein. Our results indicate that DDX3X-Core inhibited SHVV replication by competitively disrupting the interaction between the full-length DDX3X and SHVV P, demonstrating that DDX3X-P interaction is required for SHVV replication. Based on the important role of the DDX3X core domain in the inhibition of SHVV replication, we hypothesized that a fragment of DDX3X-Core might be used as an anti-SHVV peptide. To test this hypothesis, we tried to pinpoint the specific area of DDX3X-Core for the interaction with SHVV P; the core domain was divided into DDX3X-Core_210–463 and DDX3X-Core_464–627, based on functional region analysis by SMART software (38). By using a Co-IP assay, we found that both DDX3X-Core_210–463 and DDX3X-Core_464–627 were able to interact with the P protein (Fig. 5D).

FIG 5.

The DDX3X-P interaction is required for SHVV replication. CCO cells were transfected with pCMV-HA (empty vector), pCMV-HA-DDX3X, pCMV-HA-DDX3X-Core, or pCMV-HA-DDX3X-N, followed by SHVV infection. (A) The mRNA level of SHVV G in cells was measured by qRT-PCR. (B) The viral titer in the supernatants was measured using a TCID₅₀ assay. (C) The plasmids pMyc-P and pCMV-HA-DDX3X were transfected into 293T cells, together with pCMV-HA, pCMV-HA-DDX3X-Core, or pCMV-HA-DDX3X-N. At 24 h posttransfection, cell lysates were collected and subjected to Co-IP with anti-Myc antibody. The immunoprecipitates and cell lysates were then analyzed with the anti-HA and anti-Myc antibodies. (D) The plasmid pMyc-P was cotransfected into 293T cells with one of the plasmids expressing HA-tagged DDX3X-Core truncations. At 24 h posttransfection, cell lysate was collected and subjected to Co-IP with anti-HA antibody. The immunoprecipitates and cell lysates were then analyzed with anti-HA and anti-Myc antibodies. **, P < 0.01.

DDX3X promotes SHVV replication not by inhibiting IFN expression.

To date, four type I IFNs have been identified in channel catfish (CFIFN1 to CFIFN4), but only CFIFN1 and CFIFN2 were remarkably induced by virus infection (39), demonstrating that CFIFN1 and CFIFN2 play essential roles in the host antiviral response. In this study, we found that CFIFN1 and CFIFN2 were remarkably induced upon SHVV infection (Fig. 6A). Consistent with previous reports, we found that the expression of CFIFN3 and CFIFN4 showed limited detection upon SHVV infection (data not shown). Given that DDX3X has been identified as a mediator of IFN expression in mammalian cells (4, 12, 14), we asked whether DDX3X promoted SHVV replication through the inhibition of IFN expression. Our results showed that overexpression of DDX3X promoted, while knockdown of DDX3X suppressed, SHVV replication and SHVV-induced CFIFN1 and CFIFN2 (Fig. 6B and C), indicating that DDX3X-mediated promotion of SHVV replication was not caused by inhibition of IFN expression.

FIG 6.

DDX3X promotes SHVV replication not by inhibiting IFN expression. (A) CCO cells were infected with SHVV, and the mRNA levels of CFIFN1 and CFIFN2 were detected at 24 hpi. (B and C) CCO cells were transfected with the indicated plasmids (B) or siRNAs (C), followed by SHVV infection. At 24 hpi, cells were collected for the detection of mRNA levels of SHVV G, CFIFN1, and CFIFN2. **, P < 0.01.

DDX3X maintains the stability of SHVV P to avoid its degradation in an autophagy-lysosome-dependent manner.

To determine whether DDX3X affected the stability of SHVV P, CCO cells were transfected with plasmid pCMV-Myc-P together with pCMV-HA or pCMV-HA-DDX3X. The cells were cultured with or without the protein synthesis inhibitor cycloheximide (CHX) (25 μg/mL). The results showed that overexpression of DDX3X restored the degradation of the P protein (Fig. 7A). To confirm the effect of DDX3X on SHVV P, RK-33 (40), a kind of DDX3X inhibitor that abrogated DDX3X activity and reduced DDX3X expression (7), was used. The cell viability experiment showed that RK-33 at 3, 5, 10, or 20 μM exhibited no significant effects on cell viability (Fig. 7B). Because DDX3X is an RNA-dependent ATPase (8), ATPase activity was detected in RK-33-treated cells. The results showed that RK-33 at 10 or 20 μM significantly reduced the DDX3X ATPase activity (Fig. 7C). To determine the time and dose of RK-33 needed for it to effectively exert its inhibitory effect, CCO cells were treated with different concentrations of RK-33 for 4 h or 8 h. The results showed that RK-33 at 10 μM for 8 h significantly reduced the levels of DDX3X (Fig. 7D). To test whether DDX3X regulates the stability of the P protein during SHVV infection, the cells were infected with SHVV. At 16 hpi, cells were cultured with CHX, with or without RK-33 (10 μM), for 8 h. As shown in Fig. 7E, treatment with CHX reduced the levels of viral P and N proteins, while the presence of RK-33 promoted the reduction of the P protein but not the N protein, indicating that DDX3X specifically maintained the stability of SHVV P. To explore which pathway was involved in the degradation of SHVV P, MG132 (a proteasome inhibitor), NH₄Cl (a lysosome inhibitor), and 3-methyladenine (3-MA) (an autophagy inhibitor) were used. The results showed that NH₄Cl and 3-MA effectively restored the degradation of SHVV P (Fig. 7F). Our data suggest that DDX3X maintains the stability of SHVV P to avoid its degradation in an autophagy-lysosome-dependent manner.

FIG 7.

DDX3X maintains the stability of SHVV P to avoid its degradation in an autophagy-lysosome-dependent manner. (A) Plasmid pCMV-Myc-P was transfected into CCO cells, together with pCMV-HA or pCMV-HA-DDX3X. At 24 h posttransfection, the cells were treated with CHX for 5 h or 7 h and then harvested to detect the indicated proteins using Western blotting. β-Actin was used as the internal control. (B) CCO cells were treated with the indicated concentrations of RK-33 for 24 h. The cell viability was determined and calculated as a percentage of the viability of cells treated with dimethyl sulfoxide (DMSO). (C) CCO cells were treated with 10 μM or 20 μM RK-33 for 8 h, and DMSO was used as a control. The ATPase level was examined with a malachite green phosphate detection kit (Beyotime). (D) CCO cells were treated with the indicated concentrations of RK-33 for 4 h or 8 h. The cell lysates were collected and subjected to Western blotting with an anti-DDX3X antibody. (E and F) CCO cells were infected with SHVV. At 16 hpi, the cells were treated with CHX and RK-33, with or without the indicated inhibitors (5 μM MG132, 15 μM NH₄Cl, or 60 μM 3-MA) for 8 h. The cells were then harvested for Western blotting using the indicated antibodies. The integrated optical densities of the protein bands were measured using Image-Pro Plus v6.0. The values for the DDX3X, N, and P protein bands were normalized to that of β-actin. *, P < 0.05; **, P < 0.01.

DISCUSSION

DDX3X was initially known as a key factor in RNA metabolism processes. Recently, accumulating evidence suggested that DDX3X promotes the replication of many viruses, such as murine norovirus (MNV), WNV, HCV, Venezuelan equine encephalitis virus (VEEV), HIV-1, and human parainfluenza virus 3 (HPIV-3) (12). Conversely, several studies reported that DDX3X inhibits the replication of viruses, including hepatitis B virus (HBV) (27), IAV (28), NDV (29), DTMUV (30), IHNV (18), and RGNNV (31). These studies indicate that DDX3X is a double-edged sword for viral replication. To our knowledge, however, the role of DDX3X in rhabdovirus replication has seldomly been investigated. In the present study, the P protein of the fish rhabdovirus SHVV was used as bait to capture host interacting factors, and DDX3X was identified as an interacting partner of the P protein. Because DDX3X was previously reported to be a proviral or antiviral factor, here we focused on investigating whether DDX3X is involved in SHVV replication via interaction with the P protein, as well as the underlying mechanisms.

During viral infection, DDX3X has been regarded as a viral RNA sensor or mediator of the innate immune response to trigger antiviral responses (4, 14). DDX3X functions in the recognition of abortive HIV-1 RNA transcripts, thus activating the transcription factor IFN regulatory factor 3 (IRF3) and the NF-κB pathway (41). DDX3X also serves as a downstream adaptor to participate in human cytomegalovirus (HCMV)-induced IFN-β expression (42). It was demonstrated that several viruses could block the DDX3X-mediated antiviral innate immune response. HBV polymerase escapes host antiviral immunity by blocking the interaction between DDX3X and IκB kinase ε (IKKε)/TANK-binding kinase 1 (TBK1) (43). The VACV K7 protein disrupts IRF3 activation by binding to the N-terminal region of DDX3X (44). Unlike previous studies, we found that DDX3X functioned as a positive factor for SHVV replication and did not inhibit SHVV-induced IFN expression. We speculated that DDX3X promoted SHVV replication through a mechanism independent of the IFN pathway.

To further explore the mechanism by which DDX3X promotes SHVV replication, we focused our attention on the effect of DDX3X on the SHVV P expression level. We found that DDX3X functioned to maintain the stability of SHVV P through avoidance of the degradation of the P protein in the autophagy-lysosome pathway. As a multifunctional protein, however, DDX3X has been reported to maintain protein stability only in cancer cell proliferation and metastasis (45, 46) and never in virus infection. Therefore, to our knowledge, this is the first time to report that DDX3X functions to maintain viral protein stability, thus promoting virus replication.

DDX3X is composed of a highly conserved core region and hypervariable C- and N-terminal regions (37, 47). The helicase core region is composed of two RecA-like domains involved in ATP binding, RNA binding, ATP hydrolysis, and RNA strand unwinding (36). The functions of the variable N- and C-terminal regions are mainly responsible for protein-protein interactions (12, 48). For example, the interaction between ribosomal protein L13 and the N-terminal region of DDX3X promotes internal ribosome entry site (IRES)-driven translation in foot-and-mouth disease virus (FMDV) (7). The C-terminal region of DDX3X interacts with IAV NS1 and NP proteins to inhibit virus replication through the regulation of SG formation. In addition, the regulatory effect of DDX3X on RGNNV- and VACV-induced IFN responses is dependent on its N-terminal region. In our study, we found that both N-terminal and core regions of DDX3X were the regions responsible for the DDX3X-P interaction. Moreover, we found that overexpression of the core region of DDX3X inhibited SHVV replication by competitively disrupting the interaction between the full-length DDX3X and SHVV P protein. This prompted us to speculate that a fragment of DDX3X-Core might be used as an anti-SHVV peptide. Further research found that both DDX3X-Core_210–463 and DDX3X-Core_464–627 were able to interact with the P protein. In the future, the small peptides from the DDX3X-Core participating in the interaction with SHVV P will be determined to assess the anti-SHVV effects.

In conclusion, we reported for the first time that cellular DDX3X interacted with SHVV P and positively regulated SHVV replication by maintaining P protein stability through avoidance of the degradation of the P protein in an autophagy-lysosome-dependent manner (Fig. 8). Our data provided novel insights into the role of DDX3X in regulating the replication of rhabdoviruses. Moreover, our study revealed that DDX3X could be considered a potential target to interrupt SHVV replication.

FIG 8.

Diagram showing the mechanism used by DDX3X to regulate SHVV replication. Following SHVV infection, P protein is synthesized. DDX3X interacts with the P protein and prevents the degradation of SHVV P through the autophagy-lysosome pathway.

MATERIALS AND METHODS

Cells and viruses.

CCO cells were grown in minimum essential medium (MEM) (Gibco) containing 10% fetal bovine serum (FBS) (Gibco) and 1% penicillin-streptomycin (100 g/mL), and 293T cells were grown in Dulbecco's modified Eagle's medium (DMEM) (Gibco) containing 10% FBS (Gibco) and 1% penicillin-streptomycin (100 g/mL). SHVV (GenBank accession number KP876483.1) was isolated from diseased hybrid snakehead fish and stored at −80°C (33).

Plasmids.

The plasmids pEGFP-P and pMyc-P were constructed by amplifying and cloning the SHVV P gene from the previously constructed plasmid pCDNA-P (49, 50) into pEGFP-N1 or pCMV-Myc with the primers listed in Table 1. The coding region of DDX3X was amplified by using total RNAs extracted from CCO cells with the primers listed in Table 1 and was cloned into pCMV-HA and pDsRed-N1 to construct plasmids pCMV-HA-DDX3X and pDsRed-DDX3X, respectively. The truncated DDX3X plasmids were constructed by amplifying the corresponding region of DDX3X using the plasmid pCMV-HA-DDX3X as a template with the primers listed in Table 1, followed by cloning into pCMV-HA and pEGFP-N1.

TABLE 1.

Primers used in this study

Application and primer	Sequence (5′ to 3′)
Expression
HA-DDX3X-FW	CTTATGGCCATGGAGGCCCGAATGAGTCATGTGGCCGTCGA
HA-DDX3X-BW	TCGAGAGATCTCGGTCGACCGTTAGTTGCCCCACCAGTCCA
HA-DDX3X-N-FW	CTTATGGCCATGGAGGCCCGAATGAGTCATGTGGCCGTCGA
HA-DDX3X-N-BW	TCGAGAGATCTCGGTCGACCGTTACACAGGAATGTCATCATACT
EGFP-DDX3X-N-FW	CGGAATTCTGATGAGTCATGTGGCCGTCGACGGTC
EGFP-DDX3X-N-BW	CGGGATCCCGCACAGGAATGTCATCATACTTCTCA
HA-DDX3X-CORE-FW	CTTATGGCCATGGAGGCCCGAATGGAGGCCACTGGCAGCAACTG
HA-DDX3X-CORE-BW	TCGAGAGATCTCGGTCGACCGTTAGTTCTTGTGCTGGTGCTCAA
EGFP-DDX3X-CORE-FW	CGGAATTCTGATGGAGGCCACTGGCAGCAACTGCC
EGFP-DDX3X-CORE-BW	CGGGATCCCGGTTCTTGTGCTGGTGCTCAAAGGCCA
HA-DDX3X-C-FW	CTTATGGCCATGGAGGCCCGAATGAACACACGAGGCCGCTCCAA
HA-DDX3X-C-BW	TCGAGAGATCTCGGTCGACCGTTAGTTGCCCCACCAGTCCA
EGFP-DDX3X-C-FW	CGGAATTCTGATGAACACACGAGGCCGCTCCAAGA
EGFP-DDX3X-C-BW	CGGGATCCCGGTTGCCCCACCAGTCCACTTGGGTG
DsRed-DDX3X-FW	GATCTCGAGCTCAAGCTTCGAATGAGTCATGTGGCCGTCGA
DsRed-DDX3X-BW	CGCGGTACCGTCGACTGCAGGTTGCCCCACCAGTCCACTT
HA-DDX3X210–463-FW	TCTTATGGCCATGGAGGCCCAATGGTAACATAACTCTGAGCCGC
HA-DDX3X210–463-BW	CTCGAGAGATCTCGGTCGACCTTACTTCTGGGTAATGTTCTCTG
HA-DDX3X464–627-FW	CTTATGGCCATGGAGGCCCGAATGGTGGTCTGGGTTGAAGAG
HA-DDX3X464–627-BW	TCGAGAGATCTCGGTCGACCGTTAGTTCTTGTGCTGGTGCTC
MYC-P-FW	CTTATGGCCATGGAGGCCCGAATGGCAAAACCAGTTTTCCA
MYC-P-BW	CGAGAGATCTCGGTCGACCGGAACAGCACCATTTGCTGAA
EGFP-P-FW	GATCTCGAGCTCAAGCTTCGAATGGCAAAACCAGTTTTCCA
EGFP-P-BW	CGCGGTACCGTCGACTGCAGTATTCATCAGTACACTCCAT
qRT-PCR
Q-DDX3X-FW	TTCTGCTCCCGGTGTTGAGTC
Q-DDX3X-BW	ATGGGATACTGCTTACGGCGG
Q-IFN1-FW	TCGGTGTCGATTGTCCCGCA
Q-IFN1-BW	GATCTGCTTCGTGGCCTGAGC
Q-IFN2-FW	CAATGACAGGAAACACCTCT
Q-IFN2-BW	GCACTGAAGAGAATGATGAT
Q-G-FW	ACACCATACATGCCAGAGGC
Q-G-BW	GCCTCGCTGGGTATCCAAAT
Q-β-actin-FW	CACTGTGCCCATCTACGAG
Q-β-actin-BW	CCATCTCCTGCTCGAAGTC

Antibodies and reagents.

Antibodies for SHVV P, N, and G proteins were prepared by Abiotech (Jinan, China) and stored in our laboratory. Polyclonal antibodies against DDX3X, HA, Myc, His, EGFP, and β-actin were purchased from Proteintech (Wuhan, China). The secondary anti-rabbit IgG and anti-mouse IgG antibodies were purchased from ABclonal Technology (Wuhan, China). RK-33, MG132, 3-MA, NH₄Cl, and CHX were purchased from Selleck Biotechnology (Shanghai, China). siRNAs of DDX3X (siDDX3X) and the control siRNA (siNC) were synthesized by GenePharma (Shanghai, China).

Virus infection and titration.

CCO cells were infected with SHVV at a multiplicity of infection (MOI) of 0.1. After 1 h of adsorption at 28°C, the inoculum was removed, and the cells were washed twice with PBS, followed by the addition of fresh MEM plus 5% FBS. At different time points postinfection, the supernatants were collected for virus titration by the 50% tissue culture infectious dose (TCID₅₀), while the cells were harvested for the detection of viral proteins by Western blotting or viral mRNA by quantitative reverse transcription-PCR (qRT-PCR).

qRT-PCR.

Total RNAs were extracted from cells with TRIzol reagent (Invitrogen) according to the manufacturer’s instructions. For the detection of SHVV G and P mRNAs, cDNA was prepared from 1 μg RNA with the PrimeScript RT reagent kit (TaKaRa), following the manufacturer’s instructions. The qRT-PCRs were conducted in a 20-μL volume containing 10 μL AceQ qPCR SYBR green Master mix (Vazyme, China), 1 μL cDNA template, 0.5 μL forward primer, 0.5 μL backward primer, and 8 μL double-distilled water with the following cycling conditions: 95°C for 5 min, 45 cycles of 95°C for 10 s, 60°C for 10 s, and 72°C for 15 s, and a final step of 95°C at 5°C/s calefactive velocity to make the melt curve. Data were normalized to the level of the β-actin gene in each sample using the 2^-ΔΔCT method.

Co-IP and mass spectrometric assays.

CCO cells were transfected with pEGFP-P or pEGFP-N1 (control). At 24 h post of transfection, the cells were collected and subjected to Co-IP according to the manufacturer’s instructions for ChromoTek GFP-Trap agarose (Proteintech). The bound proteins were eluted by boiling in SDS-PAGE loading buffer for 10 min, separated by SDS-PAGE, and then visualized with the fast silver stain kit (Beyotime, China). The lysates of pEGFP-N1-transfected cells were used as a control. The corresponding protein strips were sent to the BGI mass spectrometry platform for mass spectrometric analysis.

Immunofluorescence and confocal microscopy.

Cells were seeded on 12-mm glass coverslips. On the following day, cells were transfected with the indicated plasmids or infected with SHVV. At the indicated time points, the coverslips were fixed with paraformaldehyde for 20 min and counterstained with the nuclear stain 4′,6-diamidino-2-phenylindole (DAPI). After three washes with PBS, the samples were detected using an N-STORM laser confocal microscope (Nikon, Japan). Image postprocessing was performed with Nikon NIS-Element software.

IP and Western blotting.

Plasmid-transfected cells were collected and subjected to IP using specific antibodies according to the manufacturer’s instructions for protein A/G magnetic beads (MCE). The final solution was separated on 10% SDS-PAGE gels and transferred to nitrocellulose membranes (Biosharp, China). Membranes were blocked for 2 h at room temperature with 5% skim milk dissolved in Tris-buffered saline with 0.1% Tween 20 (TBST) and were incubated with the indicated primary antibodies for 4 h at room temperature. The membranes were washed three times with TBST and incubated with horseradish peroxidase (HRP)-conjugated secondary antibodies for 1 h at room temperature. The signal intensity was then determined using an Amersham Imager 600 system.

siRNA experiments.

CCO cells were transfected with siDDX3X or siNC by using the TransIntro EL transfection reagent (TransGen Biotech, China). The sequences targeting DDX3X were as follows: siDDX3X-22, 5′-GGUCUAGAUCAGCAGCUUUTT-3′; siDDX3X-647, 5′-GCCCUGGCCCUAUUGAUAATT-3′; siDDX3X-1337, 5′-GGGUUGAAGAGAGCGACAATT-3′; siNC, 5′-UUCUCCGAACGUGUCACGUTT-3′.

Statistical analysis.

All statistical analyses were performed using GraphPad Prism v5.0 (GraphPad Software). All of the data are representative of at least two independent experiments, with each determination performed in triplicate (mean ± standard deviation [SD]). The statistical significance of the data was determined by Student's t test, and P values of <0.05 were considered statistically significant.