Cellular DNAJA3, a Novel VP1-Interacting Protein, Inhibits Foot-and-Mouth Disease Virus Replication by Inducing Lysosomal Degradation of VP1 and Attenuating Its Antagonistic Role in the Beta Interferon Signaling Pathway

This study pioneeringly determined the antiviral role of DNAJA3 in FMDV. DNAJA3 was found to interact with FMDV VP1 and trigger its degradation via the lysosomal pathway. In addition, this study is also the first to clarify the mechanism by which VP1 suppressed IFN-β signaling pathway by inhibiting the phosphorylation, dimerization, and nuclear translocation of IRF3. Moreover, DNAJA3 significantly abrogated VP1-induced inhibitive effect on the IFN-β signaling pathway. These data suggested that DNAJA3 plays an important antiviral role against FMDV by both degrading VP1 and restoring of IFN-β signaling pathway.

ABSTRACT

DnaJ heat shock protein family (Hsp40) member A3 (DNAJA3) plays an important role in viral infections. However, the role of DNAJA3 in replication of foot-and-mouth-disease virus (FMDV) remains unknown. In this study, DNAJA3, a novel binding partner of VP1, was identified using yeast two-hybrid screening. The DNAJA3-VP1 interaction was further confirmed by coimmunoprecipitation and colocalization in FMDV-infected cells. The J domain of DNAJA3 (amino acids 1 to 168) and the lysine at position 208 (K208) of VP1 were shown to be critical for the DNAJA3-VP1 interaction. Overexpression of DNAJA3 dramatically dampened FMDV replication, whereas loss of function of DNAJA3 elicited opposing effects against FMDV replication. Mechanistical study demonstrated that K208 of VP1 was critical for reducing virus titer caused by DNAJA3 using K208A mutant virus. DNAJA3 induced lysosomal degradation of VP1 by interacting with LC3 to enhance the activation of lysosomal pathway. Meanwhile, we discovered that VP1 suppressed the beta interferon (IFN-β) signaling pathway by inhibiting the phosphorylation, dimerization, and nuclear translocation of IRF3. This inhibitory effect was considerably boosted in DNAJA3-knockout cells. In contrast, overexpression of DNAJA3 markedly attenuated VP1-mediated suppression on the IFN-β signaling pathway. Poly(I⋅C)-induced phosphorylation of IRF3 was also decreased in DNAJA3-knockout cells compared to that in the DNAJA3-WT cells. In conclusion, our study described a novel role for DNAJA3 in the host’s antiviral response by inducing the lysosomal degradation of VP1 and attenuating the VP1-induced suppressive effect on the IFN-β signaling pathway.

IMPORTANCE This study pioneeringly determined the antiviral role of DNAJA3 in FMDV. DNAJA3 was found to interact with FMDV VP1 and trigger its degradation via the lysosomal pathway. In addition, this study is also the first to clarify the mechanism by which VP1 suppressed IFN-β signaling pathway by inhibiting the phosphorylation, dimerization, and nuclear translocation of IRF3. Moreover, DNAJA3 significantly abrogated VP1-induced inhibitive effect on the IFN-β signaling pathway. These data suggested that DNAJA3 plays an important antiviral role against FMDV by both degrading VP1 and restoring of IFN-β signaling pathway.

INTRODUCTION

Foot-and-mouth disease virus (FMDV) is the etiological agent of foot-and-mouth disease (FMD), which is highly contagious in domestic and wild cloven-hoofed animals (1–3). Seven known serotypes of FMDV (A, O, Asia1, C, SAT1, SAT2, and SAT3) exist and consist of numerous subtypes (4, 5). The FMDV genome is approximately 8.5 kb in size, positive sense, single-stranded RNA that encodes an unique polyprotein (P1-P2-P3). The polyprotein is processed by the leader protein (L^pro), 2A, and 3C protease (3C^pro) to produce four structural proteins (VP1, VP2, VP3, and VP4), as well as eight nonstructural proteins (L, 2A, 2B, 2C, 3A, 3B, 3C, and 3D) (6, 7). The outer capsid surface is composed of 60 copies of three structural proteins, VP1 to VP3, while a fourth structural protein, VP4, is located inside the capsid. The capsid of FMDV has a prominent loop feature on its surface, formed between the G and H β-strands of VP1, which is known to include a major neutralizing antigenic site (8), and also contains the highly conserved arginine-glycine-aspartate (RGD) residues (9, 10), which comprise part of the motif that is involved in the attachment of the virus to integrin receptor molecules on the surfaces of susceptible cells (11). Several integrins (αvβ1, αvβ6, αvβ8, and αvβ3), which are heterodimers of α and β chains, are known to bind to this RGD motif in FMDV (12–16). In addition, VP1 could trigger apoptosis via the Akt signaling pathway after binding to the cells (17) and could also suppress the type I interferon response by interacting with sorcin (18). However, knowledge of the FMDV VP1-interacting host proteins and how they regulate FMDV replication remains limited.

Some host proteins have been found to interact with FMDV proteins. For example, DCTN3 can interact with 3A and the interaction appears critical for virus replication in cattle (19). Vimentin is a specific host binding partner for 2C, and the interaction between 2C and vimentin is essential for virus replication (20). Beclin1 is also a natural ligand of 2C and that its involvement in the autophagy pathway is important for FMDV replication (21). Sam68 colocalizes and coprecipitates with FMDV nonstructural proteins 3C and 3D, highlighting the importance of this host factor to the progression of FMDV infection (22). VP1-induced suppression of type I interferon (IFN) is mediated by interactions with sorcin, a protein that appears to regulate cell response to viral infections (18), and 2B interacts with RIG-I and induces the reduction of RIG-I (23). L^pro cleaves translation initiation factor eIF4G to shut off host mRNA translation (24) and also suppresses the induction of IFN-α/β (25), such as causing the degradation of NF-κB subunit of p65/RelA (26), decreasing the levels of interferon regulatory factor 3 (IRF3) and IRF7 (27), possessing deubiquitinase (DUB) and deISGylase activities to manipulate IFN signaling pathways (28), and cleaving G3BP1 and G3BP2 to inhibit stress granule formation (29).

To gain more insight into the possible cellular factors that could interact with VP1 and to understand how they regulate FMDV replication, we used a yeast two-hybrid approach. We identified a host cellular DNAJA3 as a specific binding partner of VP1. DNAJA3 is a member of the DNAJ/Hsp40 protein family, which is an evolutionarily conserved family of proteins that performs multiple functions, including the regulation of transcription, translation, and protein folding (30).

The DNAJA3-VP1 interaction was further confirmed by coimmunoprecipitation (Co-IP) and colocalization in FMDV-infected cells. We further deciphered that DNAJA3 decreased VP1 protein abundance and negatively regulated FMDV replication. Pharmacological inhibition of the proteasome, lysosome, and caspase pathway by their inhibitor, as well as Co-IP assays, revealed that DNAJA3 induced lysosomal degradation of VP1 and enhanced the lysosomal pathway by interacting with LC3. Furthermore, VP1 suppressed the IFN-β signaling pathway by inhibiting the phosphorylation, dimerization, and nuclear translocation of IRF3, and such inhibition effect was weakened by DNAJA3. Therefore, these studies demonstrated the antiviral role of DNAJA3 against FMDV and explored a novel antagonistic mechanism employed by host cells to combat FMDV infection.

RESULTS

DNAJA3 interacts with FMDV VP1 protein.

A yeast two-hybrid system was used to screen host cellular proteins that potentially interact with FMDV VP1 protein (31). Several host protein candidates were shown to specifically interact with VP1 (Table 1). DNAJA3 (Gene ID 100774313) was selected for further study, since it has been shown to be involved in the replication of several viruses (32–35). Yeast cotransformation was performed to exclude possible self-activation. We observed that only the BD-VP1/AD-DNAJA3-cotransformed Y2HGold strain formed round colonies in SD/–4/X/Aba medium (Fig. 1A).

TABLE 1.

Host proteins that interact with FMDV VP1 protein as determined by yeast two-hybrid analyses

Prey	Gene ID	No. of hits
DnaJ homolog, subfamily A, member 3 (DNAJA3)	100774313	14
Kinesin family member 5B (KIF5B)	100774173	4
Guanine nucleotide binding protein (G protein) alpha 12 (Gna12)	100760944	2
Ribonucleotide reductase M2 (RRM2)	100752542	2
Protein arginine methyltransferase 3 (Prmt3)	100761738	4
Serine hydrolase-like protein (SHLP)	100752244	1
Reticulon 4 interacting protein 1 (Rtn4ip)	100767145	1
Sorting nexin 6	100757927	1

FIG 1.

VP1 interacts with DNAJA3. (A) Yeast two-hybrid assay. The indicated plasmids were cotransformed into Y2HGold yeast strain. (B) Co-IP analysis of DNAJA3 and VP1. HEK293T cells were transfected with the indicated plasmids, and cell lysates were immunoprecipitated with an Flag or Myc antibodies, followed by immunoblotting with Myc and Flag antibodies. HC, heavy chain; LC, light chain. (C) Colocalization of VP1 with endogenous DNAJA3. PK-15 cells were infected with or without FMDV. Cells were then analyzed using immunofluorescence staining with anti-VP1 (green), anti-DNAJA3 (red) and DAPI (blue) and microscopy.

To further validate the DNAJA3-VP1 interaction, a Co-IP assay was conducted by cotransfection of Flag-VP1 and Myc-DNAJA3 plasmids into HEK293T cells. After immunoprecipitation with either anti-Flag or anti-Myc antibody, Flag-VP1 was confirmed to coprecipitate with the Myc-DNAJA3 (Fig. 1B). The colocalization of VP1 and DNAJA3 in FMDV-infected cells was analyzed by double-label immunofluorescence using a confocal microscope. The results showed clear colocalization of VP1 and DNAJA3 in cells (Fig. 1C). Collectively, the data indicated that the FMDV VP1 specifically interacted with host DNAJA3.

The K208 of VP1 is critical for its interaction with DNAJA3.

To identify the domains of DNAJA3 and VP1 that were responsible for the specific interaction, a series of Myc-tagged DNAJA3 or Flag-tagged VP1 mutants were constructed (Fig. 2A, C, E, and G), and the interaction was investigated by Co-IP assays. As shown in Fig. 2B, DNAJA3 mutants containing the amino acid (aa) 168 to 296 and aa 235 to 453 regions failed to interact with VP1. Only, the protein containing the aa 1 to 235 region interacted with VP1. These data suggest that the J domain (amino acids 1 to 168) of DNAJA3 is responsible for the interaction between DNAJA3 and VP1. In addition, we observed that the VP1 mutant with the aa 1 to 115 and aa 37 to 188 regions failed to interact with DNAJA3, but the aa 115 to 211 region interacted with DNAJA3. Therefore, the data suggest that the C-terminal (aa 188 to 211 region) of VP1 was essential for the DNAJA3-VP1 interaction (Fig. 2D).

FIG 2.

The K208 of VP1 is critical for its interaction with DNAJA3. (A, C, E, and G) Schematic representations of individual DNAJA3 mutants (A) and VP1 mutants (C, E, and G). (B) The D-1-168 (J domain) of DNAJA3 is required for its association with VP1. HEK293T cells were transfected with the indicated plasmids, and cell lysates were immunoprecipitated with Flag antibodies, followed by immunoblotting with Myc and Flag antibodies. Asterisks represent target proteins. (D) V-188-211 (C-terminal domain) of VP1 interacts with DNAJA3. The method is the same as for panel B. (F) The VP1-207-211 alanine mutant lost the ability to interact with DNAJA3. The method is the same as for panel B. (H) The VP1-K208A alanine mutant lost the ability to interact with DNAJA3. The method is the same as for panel B. (I) Co-IP of FMDV VP1 with endogenous DNAJA3. FMDV-infected (+) or mock-infected (–) BHK-21 cells were used for immunoprecipitation with mouse anti-VP1 antibody and immunoblotted with rabbit anti-DNAJA3 antibody. (J) The K208A mutant FMDV lost the ability to interact with endogenous DNAJA3. rO-VP1K208A-infected, rO-WT-infected, or mock-infected BHK-21 cells were used for immunoprecipitation with rabbit anti-DNAJA3 and were immunoblotted with mouse anti-VP1 antibody.

To determine the binding site for DNAJA3 present in the C-terminal (aa 188 to 211 region) of VP1, an alanine scanning mutagenesis approach was used. VP1-207-211 alanine mutant lost the ability to interact with DNAJA3 (Fig. 2F). Using site-directed mutagenesis, further research showed that K208 of VP1 was the key site to interact with DNAJA3 (Fig. 2H). To examine whether VP1 interacted with the DNAJA3 in the process of FMDV infection, the FMDV-infected cell lysates were immunoprecipitated with anti-VP1 antibodies and probed for the presence of DNAJA3. DNAJA3 was pulled down by VP1 in FMDV-infected cells (multiplicity of infection [MOI] of 0.5) (Fig. 2I), indicating that VP1 interacted with endogenous DNAJA3 during FMDV infection. The K208A mutant virus (rO-VP1 K208A) was then successfully rescued by reverse genetics technology, and Co-IP assays showed that rO-VP1 K208A mutant virus lost the ability to interact with DNAJA3 (Fig. 2J), further confirming the prominent role of K208 in the interaction of VP1 with DNAJA3. Together, the J domain of DNAJA3 (aa 1 to 168) and the K208 of VP1 demonstrated significance in DNAJA3-VP1 interaction.

DNAJA3 negatively modulates FMDV replication.

DNAJA3 interacted with FMDV VP1, but how it influences FMDV replication remained unknown. To investigate the effects of DNAJA3 on FMDV replication, gain-of-function and loss-of-function assays were performed. CHO-677 cells were transfected with increasing amounts of Myc-DNAJA3 plasmids, and the empty vector was used in the transfection process to ensure that the cells received the same amounts of total plasmids. Cells were then infected with equal amounts of FMDV (MOI of 0.5) for 24, 36, or 48 h. At 24 h postinfection (hpi), viral RNA levels and viral protein abundance were compared. At 36 or 48 hpi, viral titers were again compared. As shown in Fig. 3A, overexpression of DNAJA3 significantly suppressed FMDV replication. Real-time PCR (RT-PCR), Western blotting, and viral titer analysis indicated that DNAJA3 combated FMDV replication in a dose-dependent manner.

FIG 3.

DNAJA3 negatively modulates FMDV replication. (A) Overexpression of DNAJA3 suppresses FMDV replication in a dose-dependent manner. CHO-677 cells were transfected with the indicated plasmids and infected with FMDV (MOI of 0.5). At 24 hpi, we collected cell precipitation and then detected the expression of DNAJA3 and viral proteins or mRNA by Western blotting or RT-PCR. At 36 and 48 hpi, we collected the cell supernatant and determined the viral titers by 50% tissue culture infective dose (TCID₅₀) assay. The effect of empty vector on the viral replication was also evaluated as a control. (B) Evaluation of the efficiency of NC or DNAJA3 siRNA in silencing DNAJA3 expression. CHO-677 cells were transfected with 150 nM NC or DNAJA3 siRNA, and the expression of DNAJA3 mRNA or protein was detected by RT-PCR or Western blotting. (C) Downregulation of DNAJA3 promotes FMDV replication. CHO-677 cells were transfected with NC siRNA or DNAJA3 siRNA and subsequently infected with equal amounts of FMDV for different time points. The expression of DNAJA3 and viral mRNA or protein was detected by RT-PCR or Western blotting. The viral titers were determined by TCID₅₀ assay. The relative fold change in the abundance of viral proteins was determined by densitometric analysis. (D) The levels of FMDV RNA copy numbers were determined in BHK-21 and PK-15 cells at different time points by RT-PCR. The standard curve function [y = –3.416 log(x) + 42.85] was used to calculate copy numbers of FMDV to obtain a virus replication curve. (E) Confirmation of successful knockout of DNAJA3 in DNAJA3-KO cell line by Western blotting. DNAJA3-WT and DNAJA3-KO PK-15 cells were infected with FMDV for 24 or 48 h, and the expression of DNAJA3 and viral proteins or viral mRNA was detected by Western blotting or RT-PCR at 24 hpi. The viral titers were determined by TCID₅₀ assay at 48 hpi. (F) Viral titers of DNAJA3-WT or DNAJA3-KO cells infected with rO-WT or rO-VP1K208A mutant virus. DNAJA3-WT and DNAJA3-KO PK-15 cells were infected with rO-WT or rO-VP1K208A mutant virus for 48 h, and the viral titers were determined by TCID₅₀ assay.

Next, we assessed FMDV replication in DNAJA3-downregulated cells. DNAJA3 was knocked down in CHO-677 cells using RNAi. Three DNAJA3 small interfering RNAs (siRNAs) were designed and synthesized, and their silencing efficiencies were evaluated by RT-PCR and Western blotting. SiRNA-2 was proved to exert the highest efficiency in decreasing DNAJA3 expression (Fig. 3B). CHO-677 cells were transfected with negative-control (NC) siRNA or siRNA-2 and then infected with equal amounts of FMDV. The siRNA knockdown efficiency was confirmed by Western blotting, and the viral RNA, viral proteins, and viral titers in the siRNA-2 cells were compared to NC siRNA cells at the indicated time points following virus infection. The levels of FMDV replication were higher in DNAJA3 siRNA cells than control cells (Fig. 3C), suggesting that FMDV replication was significantly enhanced in the DNAJA3 knockdown cells. In addition, we observed the same trend in BHK-21 and PK-15 cells (Fig. 3D). These data confirmed that DNAJA3 played an universal antiviral role against FMDV replication in different cells.

Since porcine cells are the natural host of FMDV, DNAJA3-knockout (DNAJA3-KO) PK-15 cells were subsequently established using the CRISPR/Cas9 system to further confirm the antiviral role of DNAJA3 against FMDV. Successful knockout of DNAJA3 in the established cell line was verified by Western blotting (Fig. 3E). DNAJA3 wild-type (DNAJA3-WT) cells were also obtained in parallel and used as a control. Equal amounts of FMDV (MOI of 0.5) were incubated in DNAJA3-WT or DNAJA3-KO cells, and the viral RNA levels and viral titers were compared. The expression of viral RNA, as well as the viral titers, was significantly higher in the DNAJA3-KO cells compared to that in the DNAJA3-WT cells (Fig. 3E). The virus titer of the rO-VP1K208A mutant virus was significantly higher than that of the rO-WT virus in DNAJA3-WT cells, and the virus titer had no change in DNAJA3-KO cells regardless of whether the cells were infected with rO-VP1 K208A mutant virus or rO-WT virus. The results further prove that the K208 of VP1 is critical for reducing virus titer induced by DNAJA3 (Fig. 3F). Taken together, the gain- and loss-of-function experiments indicated that DNAJA3 negatively regulated FMDV replication.

DNAJA3 induces FMDV VP1 protein degradation in a lysosome-dependent manner.

We repeatedly observed that overexpression of DNAJA3 resulted in a decreased level of VP1 when examining the interaction between DNAJA3 and VP1 (Fig. 1B). We thus hypothesized that DNAJA3 might regulate the stability of VP1 protein or inhibit its expression. To test this hypothesis, the plasmids encoding Flag-VP1 and HA-VP1 were both used to confirm DNAJA3-induced reduction of VP1. For immunoblotting analysis, Myc-DNAJA3 with HA-VP1 or Flag-VP1 was cotransfected into HEK293T cells. It was observed that VP1 protein abundance was reduced by the overexpression of DNAJA3 in a dose-dependent manner (Fig. 4A). However, there was no major decrease in VP1 mRNA levels (Fig. 4B). To explore whether DNAJA3 degraded other FMDV proteins, Myc-DNAJA3 plasmid or vector plasmid and various plasmids expressing Flag-tagged viral proteins (VP0, L, 3C^pro, 2B, VP3, 3D^pol, VP2, or 3A) were cotransfected into HEK293T cells. As shown in Fig. 4C, DNAJA3 did not degrade other FMDV proteins. These data suggested that DNAJA3 specifically downregulated FMDV VP1 expression at the translational level instead of at the transcriptional level. To further confirm the suppressive activity of DNAJA3 on VP1 expression, the HEK293T cells were cotransfected with DNAJA3 or vector plasmids with VP1 plasmids. The transfected cells were maintained in the culture medium in the presence of actinomycin D or CHX. Actinomycin D is a specific inhibitor of cellular transcription and cycloheximide (CHX) is a specific inhibitor of protein synthesis. The half-life of VP1 protein was determined by Western blotting in the presence or absence of DNAJA3. As shown in Fig. 4D, overexpression of DNAJA3 significantly accelerated degradation of VP1 in cells treated with actinomycin D or CHX, indicating that DNAJA3 controls half-life of VP1. To explore whether the proteasome-, lysosome-, or caspase-dependent pathways were involved in DNAJA3-induced reduction of VP1, the proteasome inhibitor MG132, the lysosome inhibitor NH₄Cl, and the general caspase inhibitor Z-VAD-FMK were used to evaluate the inhibitory effects. Myc-DNAJA3 and Flag-VP1 plasmids were cotransfected into HEK293T cells, and the cells were cultured with or without inhibitors. As shown in Fig. 4E, incubation of NH₄Cl restored VP1 levels in DNAJA3-overexpressed cells, while MG132 and Z-VAD-FMK had no such effect on VP1 restoration. Moreover, both the lysosome inhibitor chloroquine diphosphate (CQ) and 3-methyladenine (3-MA) blocked DNAJA3-induced VP1 degradation, which provided more evidence that DNAJA3 induced VP1 degradation in a lysosome-dependent manner.

FIG 4.

DNAJA3 induced FMDV VP1 protein degradation in a lysosome-dependent manner. (A) DNAJA3 induces the reduction of FMDV VP1 proteins in a dose-dependent manner. HEK293T cells were transfected with Flag-VP1- or HA-VP1-expressing plasmid, along with increasing quantities of Myc-DNAJA3-expressing plasmid. The expression of Myc-DNAJA3 and Flag-VP1 or HA-VP1 was detected by Western blotting. (B) The expression of VP1 mRNA was determined by RT-PCR analysis. (C) HEK293T cells were transfected with the indicated plasmids. The expression of Myc-DNAJA3- and Flag-tagged viral proteins was detected by Western blotting. (D) DNAJA3 regulates the half-life of the VP1 protein. HEK293T cells were cotransfected with Myc-DNAJA3 and Flag-VP1 plasmids. After 24 h, the cells were treated with CHX (100 μg/ml) or actinomycin D (10 μg/ml) before immunoblot analysis was performed. The relative fold change in the abundance of VP1 was determined by densitometric analysis. (E) HEK293T cells were cotransfected with Flag-VP1 plasmids and empty vector or Myc-DNAJA3 plasmids and maintained in the presence or absence of MG132 (2 or 20 μM), NH₄Cl (10 or 20 mM), Z-VAD-FMK (20 or 50 μM), CQ (50 or 100 μM), or 3-MA (5 or 10 mM) for 36 h. The expression of Myc-DNAJA3 and Flag-VP1 proteins was detected by Western blotting. The relative fold change in abundance of VP1 was determined by densitometric analysis. (F) HEK293T cells were cotransfected with Flag-VP1 plasmids and empty vector or Myc-DNAJA3 or truncated Myc-DNAJA3 plasmids. The expression of these proteins was detected by Western blotting.

The cytotoxicity of the inhibitors against HEK293T cells was determined using an MTS assay. All doses of the inhibitors used in the experiments showed no detectable cytotoxicity (data not shown). To explore which domain of DNAJA3 played a key role in inducing VP1 degradation, Flag-VP1 and truncated Myc-DNAJA3 plasmids or empty vector were cotransfected into HEK293T cells. As shown in Fig. 4F, the aa 1 to 235 and aa 235 to 453 regions of DNAJA3 degraded Flag-VP1, but the aa 168 to 296 region failed, indicating that the J domain (aa 1 to 168) and the C terminus (aa 296 to 453) of DNAJA3 were involved in inducing VP1 degradation. These data suggested that the DNAJA3 protein induced FMDV VP1 degradation via the lysosomal pathway, with the J domain and the C terminus of DNAJA3 being essential for this process.

DNAJA3 interacts with LC3 to enhance the lysosomal pathway.

DNAJA3 induced FMDV VP1 degradation in a lysosomal pathway-dependent manner. We subsequently examined the role of DNAJA3 in regulating the lysosomal pathway. Consistent with the previous report that tumorous imaginal disc 1 (TID1), a human DNAJ protein, induced macroautophagy (36), our data showed that DNAJA3 induced a significant increase of LC3B-II (Fig. 5A). This indicated that the upregulation of DNAJA3 induced autophagy. We further established ATG5, ATG12, P62, and LC3 Flag-tagged plasmids, which are key mediators of autophagy. These plasmids were cotransfected with Myc-DNAJA3 into HEK293T cells. The results showed that DNAJA3 interacted with LC3 (Fig. 5B). HEK293T cells were cotransfected with truncated Myc-DNAJA3 and Flag-LC3, and immunoblotting analyses indicated that the interaction between LC3 and DNAJA3 required the C terminus of DNAJA3 (Fig. 5C). In order to explore which domain of DNAJA3 played a key role in regulating the lysosomal pathway, truncated Myc-DNAJA3 plasmids or the empty vector were transfected into HEK293T cells. Only the aa 1 to 235 and aa 235 to 453 plasmids induced a significant increase of LC3B-II, whereas aa 168 to 296 and vector did not (Fig. 5D). These data indicated that the J domain (aa 1 to 168) and C terminus (aa 296 to 453) of DNAJA3 played a key role in inducing autophagy and VP1 degradation. These data confirmed that the C terminus of DNAJA3 could increase the levels of LC3-II through interaction with LC3.

FIG 5.

DNAJA3 increases the levels of LC3-II through interaction with LC3. (A) Myc-DNAJA3 plasmids were transfected into HEK293T cells in a dose-dependent manner, and total protein lysates were collected and analyzed with appropriate antibodies. β-Actin was used as protein loading control. (B) The HEK293T cells were transfected with the indicated plasmids, and cell lysates were immunoprecipitated with a Flag antibody, followed by immunoblotting with Myc and Flag antibodies. (C) HEK293T cells were transfected with the indicated plasmids, and cell lysates were immunoprecipitated with a Flag antibody, followed by immunoblotting with Myc and Flag antibodies. (D) Truncated Myc-DNAJA3 plasmids were transfected into HEK293T cells, and total protein lysates were collected and analyzed with the appropriate antibodies. β-Actin was used as protein loading control.

FMDV VP1 inhibits the SeV-induced IFN-β signaling pathway.

The innate immune response plays a critical role in controlling FMDV in the course of infection (37), and type I IFN is particularly important for host defense against FMDV infection (38). It was reported that VP1 was responsible for inhibiting the type I IFN response (18), but the precise mechanism has not been elucidated. We performed luciferase reporter assays and demonstrated that overexpression of VP1 inhibited the IFN-β promoter reporter activation induced by SeV infection (Fig. 6A). Various components are involved in the IFN-β signaling pathway. To investigate the potential mechanism of the VP1-mediated suppressive role on IFN-β signaling pathway, HEK293T cells were cotransfected with Flag-VP1 and the adapter molecules RIG-I, VISA, TBK1, IRF3, or IRF7, along with a luciferase reporter plasmid. The expression of VP1 inhibits activation of the IFN-β promoter induced by overexpression of upstream components such as RIG-I, VISA, TBK1, and IRF3, but not the downstream component IRF7 (Fig. 6B). These results suggested that VP1 might negatively regulate the activation of IFN-β signaling by targeting IRF3. In addition, the expression of VISA, MDA5, TBK1, p-TBK1, IRF3, and p-IRF3 in SeV-infected VP1-overexpressing cells was detected by Western blotting. The phosphorylation of IRF3 was significantly inhibited in VP1-overexpressing cells compared to that in vector-transfected cells (Fig. 6C). Further research using native-polyacrylamide gel electrophoresis (native-PAGE) and confocal microscopy assay showed that FMDV VP1 inhibits dimerization and nuclear translocation of IRF3 after stimulation with SeV in HEK293T cells (Fig. 6D and E). These results indicated that VP1 suppressed the IFN-β signaling pathway by inhibiting the phosphorylation, dimerization, and nuclear translocation of IRF3.

FIG 6.

FMDV VP1 protein inhibits the SeV-induced IFN-β signaling pathway. (A) The expression of VP1 inhibits SeV-induced activation of IFN-β promoters. HEK293T cells were transfected with Flag-VP1 or empty vector, together with IFN-β luciferase reporter. At 24 hpt, the cells were mock infected or infected with SeV for 12 h before luciferase assays were performed using a dual-specific luciferase assay kit. (B) VP1 negatively regulates SeV-induced activation of IFN-β at the IRF3 level. HEK293T cells were transfected with IFN-β reporter, pRL-TK, expression plasmids for Flag-VP1, and the indicated protein. Luciferase assays were performed with a dual-specific luciferase assay kit. Protein expression was analyzed by Western blotting. (C, D, and E) FMDV VP1 inhibits the phosphorylation, dimerization, and nuclear translocation of IRF3 after SeV stimulation. HEK293T cells were transfected with the indicated plasmids for 24 h. Cells were infected with SeV at various time points and then harvested for analysis by Western blotting (C) or native-PAGE (D). (E) Cells were stained with the indicated antibodies and imaged by confocal microscopy.

DNAJA3 restores the VP1-induced inhibitory effect on the IFN-β signaling pathway.

Host cellular proteins exploit various strategies to impair or block the function or antagonistic roles of viral proteins. To determine whether DNAJA3 affected the VP1-mediated inhibitory effect on the IFN-β response, we transfected VP1 and DNAJA3 or vector into HEK293T cells. As shown in Fig. 7A, overexpression of DNAJA3 protein significantly restored IFN-β signaling pathway activation in VP1-overexpressing cells in a dose-dependent manner. Poly(I⋅C) is a synthetic analog of viral dsRNA. We found that VP1 potently inhibited poly(I⋅C)-triggered activation of the IFN-β promoter in DNAJA3-KO cells, indicating that endogenous DNAJA3 also impaired the VP1-mediated antagonistic effect (Fig. 7B). Results from RT-PCR experiments indicated that knockout of DNAJA3 enhanced the inhibition of poly(I⋅C)-induced expression of IFN-β and its downstream IFN-stimulating gene (ISG) mRNA by VP1. The ISGs, including ISG15, MX1, ISG54, and GBP1, were downregulated in DNAJA3-KO cells overexpressing VP1 compared to those of DNAJA3-WT cells (Fig. 7C). Next, we tested the influence of endogenous DNAJA3 on the VP1 inhibitory phosphorylation of IRF3. DNAJA3-WT and DNAJA3-KO cells were seeded in six-well culture plates and transfected with HA-VP1 or vector for 24 h. Cells were transfected by Lipofectamine 2000 with or without poly(I⋅C) (50 ng/ml) for 12 h. The protein levels of phosphorylation IRF3, IRF3, DNAJA3, and HA-VP1 were detected by Western blotting. IRF3 phosphorylation was downregulated in DNAJA3-KO cells with VP1 overexpression compared to that of DNAJA3-WT cells (Fig. 7D). These results suggest that DNAJA3 had a negative regulatory role on the inhibition of VP1 on the IFN-β signaling pathway.

FIG 7.

DNAJA3 restores VP1-induced inhibitory effect on IFN-β signaling pathway. (A) DNAJA3 restores the inhibition of VP1 on IFN-β signaling pathway in a dose-dependent manner. The HEK293T cells were transfected indicated plasmids and then mock infected or infected with SeV for 12 h before luciferase assays were performed. (B and C) VP1 inhibits the IFN-β or its downstream ISGs more efficiently on DNAJA3-KO PK-15 cells than on WT cells. DNAJA3-WT and DNAJA3-KO PK-15 cells were transfected with Flag-VP1 or vector. At 24 hpt, the cells were transfected by Lipofectamine 2000 with or without poly(I⋅C) at 50 ng/ml for 12 h. (B) Luciferase assays were performed using a dual-specific luciferase assay kit. (C) The IFN-β, ISG15, MX1, ISG54, and GBP1 mRNA levels were detected by RT-PCR. **, P < 0.01; *, P < 0.05. (D) VP1 inhibits IRF3 phosphorylation more efficiently on DNAJA3-KO PK-15 cells than on WT cells. DNAJA3-WT and DNAJA3-KO PK-15 cells were seeded in six-well culture plates, transfected with the indicated plasmids, and then transfected with or without poly(I⋅C) at 50 ng/ml for 12 h. The p-IRF3, IRF3, DNAJA3, and Flag-VP1 protein levels were detected by Western blotting. The relative fold change in p-IRF3 protein was determined by densitometric analysis.

DISCUSSION

FMD is a highly infectious and economically important disease that impacts domestic cloven-hoofed animals (39). FMDV manipulates host cellular expression for its benefit and simultaneously its life cycle is also influenced by different host factors (18). The potential mechanisms by which FMDV proteins interact with host cell proteins are not fully understood. Here, for the first time, we determined by using the yeast two-hybrid system that the FMDV structural protein VP1 interacts with host cellular DNAJA3. In concordance with this result, we also demonstrated the occurrence of coimmunoprecipitation and colocalization of DNAJA3 with VP1 in cells infected with FMDV.

DNAJA3 is a member of the DNAJ/Hsp40 protein family (40), which plays an important role in the life cycles of various viruses, such as influenza A virus (35, 41, 42), human immunodeficiency virus type 2 (43), measles virus (44), and hepatitis B virus (45). Although it is commonly perceived that viruses hijack chaperone pathways to facilitate their propagation in host cells (42, 46–49), it is interesting to note that cellular chaperones are associated with antiviral activity in some cases. Previous studies reported that Hsp40 proteins could interact with the hepatitis B virus core protein and inhibit viral replication through the destabilization of viral core and X proteins (32). HspBP1 acted as an endogenous negative regulator of HIV-1 gene expression and replication by suppressing NF-κB-mediated activation of viral transcription (50). Host factor DNAJC14, an Hsp40 chaperone protein family member, inhibits yellow fever virus and hepatitis C virus at the step of viral RNA replication (51). Notably, we demonstrated that host cellular DNAJA3 proteins bind to FMDV VP1 and inhibit the replication of FMDV. The K208 of VP1 and the J domain of DNAJA3 were shown to be critical for the interaction. The overexpression of DNAJA3 resulted in decrease in viral yields, and knockdown or knockout of DNAJA3 significantly enhanced FMDV replication (Fig. 3). This indicated that DNAJA3 proteins function as an antiviral factor to inhibit the replication of FMDV; however, the precise mechanisms by which this occurs are not fully understood. Therefore, further study is needed to understand the molecular mechanisms of DNAJA3-mediated regulation of virus replication.

Interestingly, we routinely observed that overexpression of DNAJA3 specifically resulted in decreased protein levels of VP1 but had no effect on mRNA levels of VP1, indicating that DNAJA3 regulates the stability of VP1. Protein stability is regulated by ubiquitin/proteasome-mediated degradation, caspase-mediated degradation, or autophagy/lysosome-mediated degradation (52–54). In the present study, we found that DNAJA3-mediated degradation of VP1 was inhibited by the autophagy/lysosome inhibitors NH₄Cl, CQ, and 3-MA, but not by the proteasome inhibitor MG132 or the caspase inhibitor Z-VAD-FMK. In addition, we found that DNAJA3 possessed two domains that were important for the induction of autophagy: the J domain that mediated the binding of DNAJA3 to VP1 and the C terminus that mediated the binding of DNAJA3 to LC3. These data suggested that DNAJA3 interacted with VP1 and degraded VP1 through the autophagy/lysosome pathway. Previous studies reported that DNAJA3, also known as TID1, mediated macroautophagy by interacting with the Beclin1-containing autophagy protein complex through the J domain of DNAJA3 (36). Notably, our results found that DNAJA3 can induce autophagy in cells through the C terminus of DNAJA3 by interacting with LC3. Considering this difference, we speculate that it is possibly due to the complexity of DNAJA3-mediated autophagy via different pathways under different conditions. Autophagy is a double-edged sword which either plays its original antiviral role or is hijacked by virus to facilitate their replication. The antiviral role of autophagy in our study is consistent ATG5-ATG12 positively regulation antiviral effect of NF-κB and IRF3 signaling during FMDV infection (55), while contradictory to FMDV capsid protein VP2-induced autophagy via interaction with HSPB1 facilitating viral replication (56–58). Various cell types or virus strains used in different studies may cause this contradiction.

Innate immunity is the first line of host defense against pathogenic infections. During evolution, viruses have acquired numerous strategies to evade the host immune system. Previous studies have shown that FMDV VP3 inhibits the expression of the key adaptor molecule VISA as a strategy for evading innate antiviral immunity during the signal transduction phase (59). FMDV 3A was identified as a negative regulator of virus-triggered IFN-β signaling pathway by inhibiting the expression of RIG-I, MDA5, and VISA (60). FMDV VP3 degrades JAK1 to inhibit IFN-γ signal transduction pathways (61). FMDV 3C protease cleaves NEMO to impair innate immune signaling (62). FMDV L protease targets LGP2 helicase for cleavage, resulting in lower levels of IFN-β and antiviral activity (63). In line with previous work (18), our study shows that expression of VP1 inhibited the reporter activities of IFN-β promoters following SeV infection. However, the precise mechanism of VP1 inhibition of the IFN-β signaling pathway was not clarified in the Li et al. study. To explore the mechanisms by which VP1 regulates virus-triggered IFN-β induction, we first examined the molecular order of the involvement of VP1 in this process and found that VP1 regulated IFN-β activation at the level of IRF3. IRF3 is essential for induction of IFN-β during viral infection and its activation requires signal-dependent phosphorylation, dimerization, and nuclear translocation. Our results further indicated that VP1 could inhibit phosphorylation, dimerization, and nuclear translocation of IRF3 during SeV infection. However, DNAJA3 can significantly reduce the VP1-induced inhibitive effect on the IFN-β signaling pathway, and this reduction occurred in a dose-dependent manner. The induction of hundreds of ISGs through the activation of the type I IFN signaling pathway performs numerous antiviral effector functions. GBP1, ISG54, ISG15, and MX1 are classical antiviral ISGs that contribute to antiviral responses. We further demonstrated that VP1 inhibits various ISG expression, and this inhibition was reinforced in DNAJA3-KO cells. Moreover, the VP1-induced inhibitive effect on the phosphorylation of IRF3 was also enhanced in DNAJA3-KO cells. These findings suggest a negative regulatory role for DNAJA3 in the inhibition of VP1 on the IFN-β signal pathway.

In conclusion, DNAJA3 is a novel binding partner of VP1 and induces VP1-lysosomal degradation by interacting with LC3 to enhance the lysosomal pathway. In addition, VP1 suppresses IFN-β signaling pathway at IRF3 level by inhibiting IRF3 phosphorylation, dimerization, and nuclear translocation. However, DNAJA3 attenuated the inhibition effect of VP1 on IFN-β signaling pathway (Fig. 8). The findings in this study reveal a novel role for DNAJA3 in the host’s antiviral response and provide insights contributing to the development of therapeutics targeting viral infections.

FIG 8.

Schematic representation shows the model of DNAJA3 involvement in FMDV replication. DNAJA3 interacts with FMDV VP1 to induce lysosomal degradation of VP1 and restores the VP1-induced inhibitory effect on the IFN-β signaling pathway.

MATERIALS AND METHODS

Cells and viruses.

Porcine kidney (PK-15) cells and baby hamster kidney-21 (BHK-21) cells were cultured in minimum essential medium (HyClone). Human embryonic kidney 293T (HEK293T) cells were cultured in Dulbecco modified Eagle medium (Gibco). Heparan sulfate-deficient Chinese hamster ovary (CHO-677) cells were cultured in Ham F-12 medium (HyClone). All mediums were supplemented with 10% heat-inactivated fetal bovine serum (HyClone), 1% streptomycin (0.2 mg/ml), and penicillin (200 U/ml). All cells were maintained at 37°C with 5% CO₂. FMDV type O strain FMDV/O/ZK/93 was previously described (64). The K208A mutant virus (rO-VP1 K208A) was successfully rescued by reverse genetics technology, as described previously by our laboratory (65). Viral infection experiments were carried out as described previously (64). Sendai virus (SeV) was a negative-sense, single-stranded RNA virus of the family Paramyxoviridae, as described previously (66).

Reagents and antibodies.

The MG132 was purchased from Merck & Co (Germany). Chloroquine diphosphate (CQ), NH₄Cl, benzyloxy carbonyl (Cbz)-l-Val-Ala-Asp (OMe)-fluoromethylketone (Z-VAD-FMK), and poly(I⋅C) were purchased from Sigma-Aldrich (St. Louis, MO). 3-Methyladenine (3-MA) was purchased from MedChemExpress (MCE; Monmouth Junction, NJ). CHX and actinomycin D were purchased from MCE. Antibodies against Flag (sc-166355), Myc (sc-47694), and β-actin (sc-47778) were obtained from Santa Cruz Biotechnology (Dallas, TX). Antibody against DNAJA3 (ab69402) was purchased from Abcam. Antibodies against LC3 (catalog no. 4108) and P62 (catalog no. 8025) were purchased from Cell Signaling Technology (Danvers, MA). Antibody against hemagglutinin (HA) tag (catalog no. 901513) was purchased from BioLegend (San Diego, CA). Mouse anti-VP1 antibody was obtained from the Lanzhou Veterinary Research Institute. Fluorescein isothiocyanate-conjugated goat anti-mouse IgG antibody (F8521) and TRITC (tetramethyl rhodamine isothiocyanate)-conjugated goat anti-rabbit IgG (T6778) were purchased from Sigma-Aldrich.

Plasmid constructs.

A Myc-tagged DNAJA3 plasmid was generated through inserting full-length DNAJA3 cDNA fragment into a pcDNA3.1/myc-HisA vector (Invitrogen, Carlsbad, CA). Genes encoding viral structural and nonstructural proteins were amplified from the FMDV genome and cloned into the FLAG-CMV-7.1 vector (Sigma-Aldrich) to construct plasmids expressing Flag-tagged viral proteins by using standard molecular biology techniques. A series of Flag-tagged truncated VP1 and Myc-DNAJA3 constructs were generated by site-directed mutagenesis PCR. The PCR primer pairs are shown in Table 2. All constructed plasmids were analyzed and verified by DNA sequencing.

TABLE 2.

Primers used in this study

Primer	Sequences (5′–3′)	Target gene
VP1-FMDV-F	ATGACTACCACCACTGGCGAGTCCG	FMDV VP1 gene
VP1-FMDV-R	TGACAAAGTCTGTTTCTCAGGTGCAATGA
DNAJA3-F	ATGGCTGCGCGGTGTTCCCCACG	DNAJA3 gene
DNAJA3-R	GTTTCCAGTTGACCGTTTTCCA
V-1-115-F	ACTACCACCACTGGCGAG	VP1-1-115 gene
V-1-115-R	CAAGCGGGTGATAGGCTGCT
V-115-211-F	TTGGCACTCCCCTACACCGC	VP1-115-211 gene
V-115-211-R	CAAAGTCTGTTTCTCAG
V-37-188-F	GACAGGTTCGTAAAACTC	VP1-37-188 gene
V-37-188-R	GGGCCTGGGGCAGTATG
D-1-235-F	GCTGCGCGGTGTTCCCCACGC	DNAJA3-1-235 gene
D-1-235-R	GGTGTCCATGATGTTCAC
D-235-453-F	ACCTGTGAGCGCTGTGATGG	DNAJA3-235-453 gene
D-235-453-R	GTTTCCAGTTGACCGTTT
D-168-296-F	AGCTCTGGGCAGAGCTACTG	DNAJA3-168-296 gene
D-168-296-R	TCCTGCTCCTCTGCAAAC

Yeast two-hybrid screening.

The pGBKT7-VP1 (BD-VP1) construct was used as bait for hybridization with a CHO-677 cDNA library. Transformants were screened by growth on the plates containing synthetically defined medium lacking Leu, Trp, His, and Ade (SD/–4) for 4 to 6 days. Positive colonies were further identified by growth on SD/–4 medium containing X-Gal (5-bromo-4-chloro-3-indolyl-β-d-galactopyranoside) and aureobasidin A (SD/–4/X/Aba). The positive colonies were cultured in the SD/–4 medium and verified by sequencing. To validate the interaction between VP1 and DNAJA3, the Y2HGold yeast strain was cotransformed with the plasmids BD-VP1 and pGADT7-DNAJA3 (AD-DNAJA3) using a yeast transformation system 2. The transformants were screened on plates containing synthetically defined medium lacking Leu and Trp (SD/–2), SD/–4, or SD/–4/X/Aba. Cotransformation with pGBKT7-p53 (BD-p53)/pGADT7-T (AD-T) (coding for simian virus 40 large T antigen), pGBKT7-Lamin (BD-Lamin; encoding human lamin C protein)/AD-T, and BD/pGADT7 (AD) served as positive, negative, and blank controls, respectively.

Coimmunoprecipitation assay.

HEK293T cells were cultured in 10-cm² dishes, and the monolayer cells were cotransfected with various indicated plasmids. The transfected cells were cultured and lysed in 1 ml of lysis buffer (20 mM Tris [pH 7.5], 150 mM NaCl, 1% Triton X-100, 1 mM EDTA, 10 mg/ml aprotinin, 10 mg/ml leupeptin, and 1 mM phenylmethylsulfonyl fluoride). For each sample, 0.5 ml of lysate was incubated with 0.5 mg of suitable antibody or control IgG and 40 μl of protein G-Sepharose in 20% ethanol (GE Healthcare) for 12 h. The Sepharose beads were washed three times with 1 ml of lysis buffer containing 500 mM NaCl. The precipitates were analyzed by immunoblotting assay. For virus infection assays, BHK-21 cells were infected with FMDV at an MOI of 0.5 at 37°C. The FMDV-infected or mock-infected cells were prepared as described above.

Immunofluorescence microscopy.

HEK293T cells were grown on Nunc glass-bottom dishes. At 24 h posttransfection (hpt), the cells were fixed with 4% paraformaldehyde for 30 min and permeabilized with 0.1% Triton X-100 for 15 min. Cells were then incubated in 5% bovine serum albumin at 4°C for 4 h, followed by incubation with appropriate primary antibody and Alexa Fluor 488- or 594-conjugated secondary antibodies, respectively. The images were acquired with a laser-scanning confocal microscope (Leica SP8; Leica, Solms, Germany). For virus infection assays, PK-15 cells were infected with FMDV/O/ZK/93 (MOI of 0.5) at 37°C for 12 h. The FMDV-infected or mock-treated cells were prepared as described above.

Establishment of a DNAJA3-KO PK-15 cell line using the CRISPR/Cas9 system.

The design of the gRNAs was based on recommendations on the Zhang laboratory website (http://crispr.mit.edu/). To construct the gRNA expression plasmid, complementary oligonucleotides encoding gRNA (5′-AGTGGCGCTCGTGTGAAAGA-3′) were annealed and cloned into BsmBI (Fermentas) sites in lentiCRISPRv2 (Addgene). Lentiviral particles were generated by transfection of HEK293T cells with lentiCRISPRv2-gRNA construct psPAX2 and pMD2.G (Addgene) at a ratio of 4:3:1, respectively. Viral supernatants were collected 48 to 72 h after transfection and concentrated using a centrifugal filter (Merck Millipore) according to the manufacturer’s protocol. Cells were seeded at ∼40% confluence and transduced with lentivirus in 12-well plates. At 24 hpi, the cells were detached with trypsin (Beyotime) and subcultured at a low density. After 3 h of incubation, selective pressure was added. Cell colonies formed after 2-week selection with the indicated concentrations of puromycin (Beyotime). The knockout efficiency was confirmed by Western blotting with wild-type (WT) PK-15 cells as a control.

Western blotting.

For Western blotting, target proteins were resolved by SDS-PAGE and transferred onto an Immobilon-P membrane (Millipore). The membrane was blocked and incubated with appropriate primary antibodies and secondary antibodies. The antibody-antigen complexes were visualized using enhanced chemiluminescence detection reagents (Thermo) (23).

RNA extraction and RT-PCR.

Total RNAs were extracted using TRIzol reagent (Invitrogen). The cDNA was synthesized from the extracted RNA samples, using the Moloney murine leukemia virus reverse transcriptase (Promega) and random hexamer primers (TaKaRa, Japan). The generated cDNA was used as the template to detect expression of FMDV RNA and host cellular mRNA. The Mx3005P QPCR system (Agilent Technologies) and SYBR Premix Ex Taq reagents (TaKaRa) were used in an RT-PCR experiment to quantify the abundance of various RNAs. The glyceraldehyde-3-phosphate dehydrogenase (GAPDH) gene was used as an internal control. Relative expression of mRNA was calculated using the comparative cycle threshold (2^–ΔΔCT) method. All the experiments were repeated three times with duplicates for each experiment.

Pharmacological inhibition of proteasome, lysosome, and caspase inhibitor pathway by their specific inhibitors.

HEK293T cells were cultured in 6-well plates to a confluence of 80% and then cotransfected with Flag-VP1 plasmid and Myc-DNAJA3 plasmid or empty vector using Lipofectamine 2000 (Invitrogen), and maintained in the presence or absence of MG132 (2 or 20 μM), NH₄Cl (10 or 20 mM), Z-VAD-FMK (20 or 50 μM), CQ (50 or 100 μM), or 3-MA (5 or 10 mM) for 36 h. The collected cells were then subjected to Western blotting. An MTS [3-(4, 5-dimethylthiazol-2-yl)-5-(3-carboxyme-thoxyphenyl)-2-(4-sulfophenyl)-2H-tetrazolium] assay was performed to evaluate the cytotoxicity of MG132, NH₄Cl, Z-VAD-FMK, CQ, and 3-MA in HEK293T cells.

RNA interference.

Small interfering RNA (siRNA) used in the RNA interference (RNAi) assay was chemically synthesized by GenePharma (Shanghai, China). The knockdown of endogenous DNAJA3 was carried out by transfection of DNAJA3 siRNA (5′-CCC GGA CAC AAA UAA GGA UTT-3′) into CHO-677 cells using Lipofectamine 2000. Nontargeting siRNA (NC siRNA) (5′-UUC UCC GAA CGU GUC ACG UTT-3′) was used as a negative control.

Luciferase reporter assays.

The IFN-β promoter luciferase reporter plasmids and various HA-tagged plasmids used in this study were kindly provided by Hong-Bing Shu (Wuhan University, Wuhan, China). HEK293T cells (1 × 10⁵) were seeded on 24-well and transfected with 100 ng of reporter plasmid, 20 ng of pRL-TK (Promega), and other indicated plasmids. At 24 hpt, the cells were mock infected or infected with SeV for 12 h; the luciferase activity was then measured by using a dual-specific luciferase assay kit (Promega).

Statistical analysis.

All measured values are represented as means with standard errors from three independent experiments. A Student t test is used to analyze the significance (*, P < 0.05 [considered significant]; **, P < 0.01 [considered highly significant]).