Nonstructural proteins 2C and 3D are involved in autophagy as induced by the encephalomyocarditis virus

To identify the EMCV proteins involved in autophagy induction, we first amplified the VP1, VP2, VP3, VP4, 2A, 2B, 2C, 3A, 3B, 3C and 3D genes of the EMCV BJC3, GST or LC3 genes by RT-PCR or PCR and constructed corresponding recombinant plasmids with HA tags or GFP tags. The amplified product for each gene was consistent with the size as expected (Additional file 1: Figure S1A). The expression of each EMCV protein by the recombinant plasmids was examined in BHK-21 cells through transfection and Western blotting analysis. The EMCV proteins were expressed effectively except for the 3B gene (Additional file 1: Figure S1B). The transfected BHK-21 cells were simultaneously used to analyze subsequently the expression levels of LC3-I, LC3-II and p62.

The LC3 is a specific marker protein for monitoring autophagic vesicle formation through its vesicle formation and lipidation reaction [26]. To facilitate the observation of autophagic vesicles by fluorescence microscopy, BHK-21 cells were transfected with a green fluorescent protein-tagged LC3 plasmid (GFP-LC3). The patterns of the GFP-LC3 in BHK-21 cells were diffuse. By contrast, the green fluorescent pattern exhibited puncta morphology upon EMCV infection, resembling the pattern of autophagosome-like vesicles, which suggests that the BHK-21 cells that were transfected and were expressing GFP-LC3 were suitable for studying autophagy (Figure 1A).To assess which viral protein could trigger autophagy, plasmids expressing various HA-tagged EMCV proteins were co-tranfected with pEGFP-LC3 into BHK-21 cells. The fluorescence microscopy was performed to provide a puncta analysis of the co-transfected cells. The results showed that the BHK-21 cells that were co-transfected with the VP1-, VP3-, VP4-, 2B-, 2C-, 3A- or 3D-expressing plasmids and pEGFP-LC3 exhibited a number of positive puncta that were distributed in the cytoplasm, and the VP1, VP3, VP4, 2B, 2C, 3A or 3D colocalized with a subset of GFP-LC3 puncta (Figure 1B, 1D, 1E, 1G, 1H, 1I, and 1L), whereas the BHK-21 cells that were co-transfected with the VP2-, 2A- or 3C-expressing plasmids and pEGFP-LC3 and mock-transfected with the pEGFP-LC3 and pCMV-HA displayed no positive puncta (Figure 1C, 1F, 1K, and 1M). In addition, theexpression of 3B was not observed in the BHK-21 cells co-transfected with the 3B--expressing plasmid and pEGFP-LC3 (Figure 1J), which was consistent with the result of Western blotting analysis.It was necessary to rule out the possibility that positive puncta are elicited by VP1, VP3, VP4, 2B, 2C, 3A and 3D due to the overloaded protein expression. In comparison with the cells expressing VP1, VP3, VP4, 2B, 2C, 3A and 3D, the cells that were expressing glutathione S-transferase (GST) as a irrelevant protein control exhibited no positive puncta (Figure 1N).

Upon autophagosome formation, LC3 is known to be converted from its cytosolic form (LC3-I) to the lipidated, autophagosome-associated form (LC3-II), which exhibits faster mobility on SDS-polyacrylamide gels. The increased level of LC3-II has been used to reflect the degree of autophagosome formation in each sample [26]. P62 is a link protein between LC3 and ubiquitinated substrates, and it is specifically degraded by autolysosomes, which is considered a marker for autophagic degradation activity [25, 27]. We further analyzed the expression levels of LC3-I, LC3-II and p62 in the BHK-21 cells transfected with the recombinant plasmid expressing each EMCV protein by Western blotting. The results demonstrated that the LC3-II level significantly increased and p62 clearly degraded in the BHK-21 cells transfected with 2C-or 3D-expressing plasmids (Figure 2A). The densitometry ratio of LC3-IIand β-actin in 2C- or 3D-expressing cells was much higher than it was in the other EMCV protein-expressing cells and mock-transfected BHK-21 cells, suggesting that a clear increase in autophagosome formation occurred in the BHK-21 cells that were transfected with a 2C- or 3D-expressing plasmid. The relative p62/β-actin ratios were quantified and significantly decreased in the BHK-21 cells expressing 2C- or 3D, leading to the enhanced autophagic activity.

To determine further whether autophagy is triggered in the BHK-21 cells that were transfected with a 2C- or 3D-expressing plasmid, TEM was performed to provide an ultrastructural analysis of the transfected cells. Previous studies have shown that the VP1 of EMCV can colocalize with LC3 during viral infection [25]. Our present finding showed that GFP-LC3-positive puncta could be simultaneously induced and colocalized with VP1 in the BHK-21 cells that were transfected with pCMV-HA-VP1 (Figure 1B). Thus, in our study, the BHK-21 cells expressing EMCV VP1 that did not induce autophagy were used as a control to analyze the autophagosome-like vesicle formation ability as induced by 2C or 3D. As shown in Figure 2B, the BHK-21 cells that were transfected with the 2C- or 3D-expressing plasmid had significantly increased double- or single-membrane vesicles in the cytoplasm in comparison with the cells that were transfected with the pCMV-HA-VP1 plasmid (as a control of viral protein that did not induce autophagy) or pCMV-HA plasmid (as a control) and mock-transfected cells; meanwhile, the recognizable cytoplasmic contents or degraded organelles seemed to be sequestered in most of the well-defined vesicles with morphologically typical autophagic vacuoles within the cells. Further quantitative analyses indicated that there was a significant increase in the number of autophagosome-like vesicles in the cytoplasm of the cells transfected with 2C- or 3D-expressing plasmid (Figure 2C).

To verify the possibility that LC3 modification was caused by autophagic signalling instead of reflecting an EMCV 2C- or 3D-induced membrane alteration, the knockdown effect of Beclin1 (one of the crucial factors for autophagosome formation) on the induced LC3 and p62 levels was assessed [28]. We used specific siRNA to silence the Beclin1 gene in BHK-21 cells. The cells treated with specifically targeted Beclin1 siRNA showed a dose-dependent reduction in the Beclin1 protein level (Figure 2D), and 40 pmol of siRNA was then used for further study. The increased levels of LC3-II bands and degraded p62 levels in response to 2C or 3D over-expression were significantly inhibited upon Beclin1 silencing when compared with the control groups (Figure 2E), indicating that EMCV 2C and 3D can induce autophagy signalling.

Taken together, we found that EMCV 2C or 3D protein exhibited clear abilities to induce the conversion of LC3, degrade p62 and increase the number of autophagosome-like vesicles in BHK-21 cells in comparison with control cells expressing other proteins. Thus, our further studies of autophagy mechanisms primarily focused on analyzing 2C and 3D proteins.

Because many picornaviruses can induce ER stress [29, 30] and this ER stress induces autophagy [31‐33], EMCV might cause ER stress and therefore induce autophagy. To test this hypothesis, we analyzed whether the marker molecules of the ER stress were regulated in BHK-21 cells by EMCV infection, 2C or 3D protein expression or thapsigargin (Tg) treatment (as a positive control), which is a known ER stress inducer [34]. As expected, when compared with the control cell groups, western blotting analyses showed that EMCV infection, 2C or 3D protein expression or Tg treatment could enhance the levels of ER stress marker proteins GRP78 and GRP94 (Figure 3A). Moreover, we also found that 2C or 3D altered the ER homeostasis, which was reflected in the induction of the ER stress marker, such as DNA damage-inducible gene 153 (GADD153/CHOP), ER chaperones GRP78 and GRP94, calreticulin and the activating transcription factor (ATF4), according to the luciferase assays. As shown in Figure 3B, the relative luciferase activity of these protein promoters were significantly activated in the BHK-21 cells that were transfected with 2C- or 3D-expressing plasmid compared with the transfected cells with control plasmid pCMV-HA, suggesting that EMCV 2C and 3D can trigger the up-regulation of ER stress molecules at the transcriptional level.To detect whether the activation of ER stress induces autophagy in BHK-21 cells, we analyzed the autophagy level in BHK-21 cells treated with Tg, an ER stress inducer (Figure 3C). The conversion of LC3 was significantly enhanced by Tg in comparison with the mock-treated cells, indicating that ER stress can induce the autophagy signaling of BHK-21 cells. The above results demonstrated that EMCV and 2C or 3D protein could induce ER stress, and the activation of ER stress induced by Tg could enhance the autophagy signaling. Therefore, we inferred that EMCV and 2C or 3D protein might induce autophagy via activating ER stress in BHK-21 cells.

In response to ER stress, the UPR pathway is activated and serves to minimize ER malfunction. To dissect the mechanism of autophagy induction by EMCV infection and expression of EMCV 2C or 3D protein, we examined whether EMCV and 2C or 3D protein can induce the PERK pathway. As shown in Figure 4A, EMCV infection and 2C or 3D protein could activate the PERK pathway, as indicated by the levels of PERK, p-PERK, eIF2α and p-eIF2α protein. The cells infected with EMCV and transfected with 2C- or 3D-expressing plasmid exhibited higher expressions of the p-PERK and p-eIF2α proteins in comparison with the mock-infected cells or the cells that were transfected with pCMV-HA and mock cells; nevertheless, the PERK and eIF2α protein levels remained unchanged.

To assess whether EMCV and 2C or 3D protein also induces the IRE1-XBP1 pathway to trigger UPR, we analyzed XBP1 mRNA splicing in BHK-21 cells that were infected with EMCV or were overexpressing 2C or 3D proteins. The XBP1 cDNA was amplified by RT-PCR and digested by PstI, for which there is a restriction site located within the 26-nt region of XBP1 cDNA as removed by IRE1-mediated splicing, as previously described [30, 35] (Figure 4B). Tunicamycin (Tu), an inducer of XBP1 splicing (XBP1s), served as a positive control [35]. We noticed that Tu-treated cells appeared to express high XBP1s, as evidenced by the resistance of the spliced product to PstI digestion. When compared with BHK-21 cells infected with EMCV and mock-infected cells, no obvious differences were observed in the levels of XBP1, which was subtly spliced, indicating that EMCV infection is not able to significantly activate the IRE1-XBP1 pathway substantially through splicing XBP1. Although 2C or 3D proteins could cause low levels of XBP1s in BHK-21 cells tranfected with pCMV-HA-2C and pCMV-HA-3D, no significant differences were shown between the cells expressing 2C or 3D protein and the control cells (Figure 4C), indicating that the overexpression of 2C or 3D protein cannot induce the activation of the IRE1-XBP1 pathway.

ATF6a is cleaved by trans-membrane proteases and translocated to the nucleus, where it activates the genes responsible for the ER stress response [36, 37]. ATF6a cleavage was analysed in BHK-21 cells to assess whether EMCV infection or the expression of EMCV 2C or 3D protein activate the ATF6a pathway. As shown in Figure 4D, an intact form of ATF6a (90 kDa) was detected in the control cells, and in EMCV-infected or 2C- or 3D-expressing BHK-21 cells, a lower level of its intact form was uncovered by the Western blotting assay.

BHK-21 cells were originally obtained from the American Type Culture Collection (ATCC) and maintained in Dulbecco’s modified Eagle medium (DMEM) supplemented with 10% heat-inactivated foetal bovine serum (FBS), 100 U/ml of penicillin G, and 100 g/ml streptomycin at 37°C in a humidified 5% CO₂ incubator. An EMCV strain (BJC3) that was isolated in our laboratory was used in this study [53]. Plasmids pEGFP-C1 and pCMV-HA were purchased from Clontech (Mountain View, CA, USA), and plasmids pGL3-Basic and pRL-TK were purchased from Promega (Madison, WI, USA).

The following antibodies were used: anti-GRP78, anti-GRP94, anti-ATF6, anti-PERK, anti-phospho-PERK, anti-eIF2α, and rabbit anti-phospho-eIF2α antibody (Cell Signal Technology, Inc., Danvers, MA, USA). The rabbit antibodies against LC3 and p62 and a mouse monoclonal antibody against HA and β-actin, the secondary antibodies tetramethyl rhodamine isothiocyanate TRITC-conjugated goat anti-mouse and horseradish peroxidase conjugated goat anti-mouse or anti-rabbit were purchased from Sigma-Aldrich (St. Louis, MO, USA). Anti-Beclin1 rabbit antibody was obtained from Santa Cruz Biotechnology (Santa Cruz, CA, USA). Mouse anti-VP1 mAb was prepared in our laboratory. Tunicamycin (Tu) and thapsigargin (Tg) were purchased from Sigma-Aldrich. RIPA lysis buffer was obtained from Beyotime (Jiangsu, China). Lipofectamine LTX and PLUS and RNAiMAX reagent were purchased from Invitrogen (Auckland, NY, USA).

Total RNA from cultured cells was isolated with an RNeasy Mini Kit (Qiagen, Hilden, Germany) by following the manufacturer’s protocol. First-strand cDNAs were reverse-transcribed (RT) with 2 μg of RNA as the template, and the target genes were PCR-amplified by following the procedures described in the one-step RT-PCR kit (Qiagen) according to the manufacturer’s recommendations.

The XBP1 and β-actin genes were amplified by RT-PCR with the specific primers listed in Additional file 1: Table S1, and the primers were designed according to the XBP1 and Actin sequences available in the GenBank database (accession no. NM_001244047 and no. XM_006176094, respectively).

The open reading frame fragment of the Atg8 homolog known as the microtubule-associated protein light chain 3 (LC3) gene from BHK-21 cells was amplified by RT-PCR, and the specific primers were designed in accordance with the Atg8 gene sequence in the GenBank database (accession no. XM_003495104) (Additional file 1: Table S1). The amplified cDNA was then subcloned into pEGFP-C1. Each gene of the EMCV protein was amplified by RT-PCR with the primers listed in Additional file 1: Table S1 and cloned into pCMV-HA to generate the following expression plasmids: pCMV-HA-VP1, pCMV-HA-VP2, pCMV-HA-VP3, pCMV-HA-VP4, pCMV-HA-2A, pCMV-HA-2B, pCMV-HA-2C, pCMV-HA-3A, pCMV-HA-3B, pCMV-HA-3C and pCMV-HA-3D. The glutathione S-transferase (GST) gene was amplified by PCR from cloning vector pGEX-6P-1 (accession no. U78872) with the specific primers listed in Additional file 1: Table S1 and then cloned into pCMV-HA to generate recombinant plasmid pCMV-HA-GST. The pGL3-GRP78-luc, pGL3-GRP94-luc, pGL3-calreticulin-luc, pGL3-ATF4-luc, pGL3-CHOP-luc-carrying mouse GRP78, GRP94, calreticulin, ATF4 and CHOP promoters were cloned into the pGL3-basic vector [54]. The plasmids were sequenced to confirm that each amplified product had no errors introduced as a result of PCR amplification.

In accordance with the requirements of different experiments, the BHK-21 cells were infected with either EMCV BJC3, or they were mock-infected with phosphate-buffered saline (PBS). Following a 1 h absorption period, unattached viruses were removed by aspiration. The cells were then washed thrice with PBS and cultured in complete medium at 37°C for the indicated time points until different samples had been harvested for further experiments.

A whole cell lysate of BHK-21 cells was prepared at the indicated time points after transfecting with the RIPA lysis buffer according to the manufacturer’s protocol. The supernatant was stored at −80°C for western blotting. Twenty micrograms of each extract was subjected to electrophoresis on an SDS-polyacrylamide gel and transferred onto an Immobilon-P^SQ transfer membrane (Millipore, Billerica, MA, USA). The membrane was probed with the indicated primary antibodies and appropriate secondary antibodies, which were detected by using a chemiluminescence detection kit (Thermo Scientific, Inc., Waltham, MA, USA) and then exposed to a chemiluminescence apparatus (Proteinsimple, Santa Clara, CA, USA).

BHK-21 cells were grown to 70-80% confluence in 24-well culture plates (Corning Inc., NY) and transfected with firefly luciferase reporter vectors along with the internal control Renilla luciferase reporter construct, pRL-TK (firefly luciferase reporter construct and pRL-TK in a ratio of 20:1 for BHK-21 cells) and the indicated EMCV protein recombinant plasmids. Forty-eight hours after the transfection, the cells were harvested and assayed for luciferase activity by using the Dual-Luciferase assay system according to the manufacturer^’s instructions from Promega (E1910).

BHK-21 cells that had grown to approximately 70-80% confluence in 24-well culture plates were co-transfected with each EMCV protein-expressing vector, GST protein-expressing plasmid (as a control) and pEGFP-LC3 for 48 h, or they were infected with EMCV after transfecting with pEGFP-LC3 plasmid for 12 h. The cells were fixed with pre-cooled 4% paraformaldehyde at the indicated times. The cells were then washed three times with PBS (pH 7.4), and a mouse monoclonal antibody against HA was allowed to incubate with the cells for 2 h at 37°C. The cells were washed three times with PBS and then incubated for 2 h at 37°C with a 1:200 dilution of goat anti-mouse secondary antibodies conjugated to TRITC. Finally, the cells were washed with PBS and directly visualized under a Nikon TE-2000E confocal immunofluorescence microscope (Nikon Instruments, Inc., Melville, NY, USA).

Sub-confluent monolayers of BHK-21 cells grown on 6-well culture plates (Corning Inc.) were transfected with pCMV-HA-2C, pCMV-HA-3D, pCMV-HA-VP1 or pCMV-HA and mock-transfected cells for 48 h. The cell samples were then processed as previously described [39]. Ultrathin sections were prepared, examined and imaged under a Hitachi H-7500 transmission electron microscope (Hitachi Ltd; Japan).

siRNAs designed by the GenePharma Company (Shanghai, China) were used to knock down Beclin1 in BHK-21 cells. The siRNA sequences included siBeclin1 (sense, 5′-CGGGAAUACAGUGAAUUUATT-3′; antisense, 5′-UAAAUUCACUGUAUUCCCGTT-3′) and negative control siRNA (as a control) (sense, 5′-UUCUCCGAACGUGUCACGUTT-3′; antisense, 5′-ACGUGACACGUUCGGAGAATT-3′). The BHK-21 cells were dissociated and mixed with different concentrations of siRNA by using Lipofectamine RNAiMAX reagent as recommended by the manufacturer’s protocol. The cells were harvested for further analysis after 48 h.