Human non-small cell lung cancer (A549) cells, stable knockdown A549 cell lines of viral 2C^pro (shRNA-2C-002), stable knockdown A549 cell lines of viral 3A^pro (shRNA-3A-001), and their negative control A549 cell lines (shRNA-NC) by lentivirus infection were cultured at 37°C under 5% CO₂ in RPMI-1640 (Lanzhou Bailing Biotech) containing 15% new bovine serum (NBS, Lanzhou Minhai Bio-engineering). Baby hamster kidney (BHK-21) cells were cultured at 37°C under 5% CO₂ in Dulbecco’s modified Eagle’s medium (DMEM, Lanzhou Bailing Biotech) containing 10% new bovine serum (NBS, Lanzhou Minhai Bio-engineering). EMCV strain PV21 (GenBank No: X74312) was propagated and titrated in BHK-21 cells.

NP-40 lysis buffer (P0013F), radioimmunoprecipitation assay (RIPA) lysis buffer (P0013K), protein G Agarose (P2009), antibodies against rabbit IgG (A7016), and GRP78 (AF0171) were purchased from Beyotime (Shanghai, China). Lipofectamine™ 2000 (11668019) was purchased from Thermo Fisher Scientific (Waltham, MA, United States). Polyvinylidene fluoride (PVDF) Membrane (IPVH00010) was purchased from Merck (Darmstadt, Germany). Antibodies against HSP90β (ABP54794) and HSP70 (Abm40042) were purchased from Abbkine (Wuhan, Hubei, China). Antibodies against glyceraldehyde-3-phosphate dehydrogenase (GAPDH) (60004-1-Ig), HA Tag (51064-2-AP), Myc Tag (60003-2-Ig), IRF3 (11312-1-AP), HSP27 (18284-1-AP), MDA5 (21775-1-AP), and MAVS (14341-1-AP) were purchased from Proteintech (Wuhan, Hubei, China). Antibodies against p-IRF3 (29047), TBK1 (38066), and p-TBK1 (5483) were purchased from Cell Signaling Technology (Boston, MA, United States). Anti-VP1 antibody was kindly provided by Dr. Juan Bai (Nanjing Agricultural University, Nanjing, China). Peroxidase affiniPure goat anti-rabbit IgG (H + L) (111-035-003) and anti-mouse IgG (H + L) (115-035-003) antibodies were purchased from Jackson ImmunoResearch Laboratories (West Grove, PA, United States).

The empty vectors pCMV-HA and pCMV-Myc were provided by the Biomedical Research Center of Northwest Minzu University. The recombinant plasmids Myc-HSP27, Myc-OAS1, Flag-MDA5, HA-VP1, HA-VP2, HA-2A, HA-2C, HA-3A, Myc-VP3, and Myc-3C were all constructed in the lab. The proteasome inhibitor MG132 (S1748) and caspase inhibitor Z-VAD-FMK (C1202) were obtained from Beyotime (Shanghai, China). The autolysosome inhibitor chloroquine (tlrl-chq) and poly (I:C) (tlrl-pic) were obtained from InvivoGen (San Diego, CA, United States). The small interfering RNA (siRNAs) an analog of double-stranded RNA (dsRNA) targeting homo HSP27 were designed and synthesized by Guangzhou Ribo Biotechnology (China). The siRNA sequences are listed in Table 1. A549 cells in 6-well plates were transfected with siRNAs or various plasmids using Lipofectamine™ 2000 according to the manufacturer’s instructions.

For in vitro virus infection, the untreated A549 cells or A549 cells transfected with different plasmids were infected with EMCV at a multiplicity of infection (MOI) of 0.01 or 0.1 and cultured at 37°C for 2 h. Then cells were washed with phosphate-buffered saline (PBS) three times, replaced with DMEM containing 3% NBS, and incubated at 37°C for the indicated time. The cells and supernatants were repeatedly frozen and thawed to detect TCID₅₀ in BHK-21 cells (using the Reed-Muench method), while the cells collected separately were used for subsequent quantitative real-time polymerase chain reaction (RT-qPCR) or immunoblotting analysis.

According to manufacturers’ protocol, total RNA was extracted from cells using RNAiso Plus (Takara, 9109). Then, 1 μg of total RNA was reverse transcribed using Evo M-MLV RT kit with gDNA clean for qPCR II (Accurate biology, AG11711) to synthesis cDNA. Relative RT-qPCR was performed using the TransStart^® Top Green qPCR SuperMix (+Dye II) (Transgen, AQ132-24), and target gene expression was calculated by normalization to GAPDH using the 2^–ΔΔCt method. The copy number of the polymerase 3D gene was detected by absolute RT-qPCR using the Premix Ex Taq Probe qPCR (Takara, RR390A). The probe and primer sequences used in this study are described in Table 2.

The transfected cells were lysed in RIPA or NP-40 lysis buffer containing a protease and phosphatase inhibitor cocktail (Beyotime, P1045) for 30 min on ice. And then, the cell lysates were centrifuged at 12,000 × g for 30 min at 4°C to remove the debris. For Immunoprecipitation, the cell supernatant was incubated with corresponding antibodies at 4°C overnight, then added Protein G Agarose (Beyotime, P2009) and incubated at 4°C for another 3 h. The bead complexes were washed five times with ice-cold phosphate-buffered saline-tween (PBST) and subjected to immunoblotting, target proteins were fractionated on 10% SDS-PAGE gels and transferred to PVDF membranes (Merck, ISEQ00010) separately, and these membranes were blocked and incubated with the corresponding primary antibodies and peroxidase affinipure goat anti-rabbit or mouse IgG (H + L). Finally, the bound protein bands were detected using Western lightning Plus-ECL kit (PerkinElmer, NEL105001EA) in membranes and visualized with Amersham Imager 600 imaging system (GE).

Data are displayed as the mean ± standard deviation (SD) of three independent experiments. Statistical significance was determined with one-way analysis of variance (ANOVA) or Student’s t-test (*P < 0.05; ^**P < 0.01; ^***P < 0.001).

Lung epithelial A549 cells were infected with EMCV (MOI = 0.1) to explore the expression profile of HSPs in EMCV-infected cells, and the dynamics of HSPs expression were evaluated over a 24 h period. Out of all HSPs tested, HSP27 expression decreased only at the protein level from 9 h post-infection (hpi) onward, suggesting that HSP27 protein expression is lost, presumably via degradation during an EMCV infection (Figures 1A–C). We hypothesize that EMCV proteins are responsible for HSP27 degradation. A549 cells were transfected with increasing doses of HA or Myc-labeled EMCV viral proteins, and endogenous HSP27 protein expression was assessed. Only viral proteins 2C^pro and 3A^pro reduced HSP27 protein levels (Figures 2A,B), whereas VP1^pro, VP2^pro, VP3^pro, 2A^pro, and 3C^pro had no effect (Supplementary Figures 1A–E). We further verified this in stable A549 cell lines engineered to express short hairpin RNA (shRNA) targeted toward viral 2C^pro and 3A^pro. Here, we showed that both shRNA-2C-002 and shRNA-3A-001 significantly reversed the loss of HSP27 expression during EMCV infection (Figures 2C,D). Moreover, when we attenuated the expression of 2C^pro and 3A^pro, respectively, the EMCV VP1 protein and viral titers were significantly reduced (Figures 2C–F), indicating that the expression of 2C^pro and 3A^pro are significantly positively correlated with EMCV proliferation. To see whether the proteasomal, lysosomal, or caspase-dependent apoptosis pathways play a role in 2C^pro and 3A^pro-mediated HSP27 degradation, we used the proteasome inhibitor MG132, the autolysosome inhibitor chloroquine (CQ), or the pan-caspase inhibitor, Z-VAD-FMK, to evaluate the inhibitory effects of 2C^pro and 3A^pro on HSP27 protein expression, respectively. Nearly all tested inhibitors failed to reverse the 2C^pro-mediated reduction of HSP27 (Figure 2G), suggesting that the inhibitory effect of 2C^pro on HSP27 was independent of these pathways. However, we found that the reduction of HSP27 protein expression by 3A^pro could be reversed upon treatment with Z-VAD-FMK (Figure 2H), indicating that 3A^pro-mediated HSP27 degradation is dependent on caspase activity. We then wondered whether HSP27 directly interacts with 2C^pro and 3A^pro. Indeed, co-immunoprecipitation (Co-IP) assays confirmed an interaction between HSP27 and 2C^pro or 3A^pro (Figures 3A,B). Collectively, these results show that 2C^pro and 3A^pro directly bind and degrade HSP27.

In light of the observation that EMCV infection and its viral proteins 2C^pro and 3A^pro could lead to the degradation of HSP27 suggests that HSP27 is involved in dampening EMCV infection. To evaluate this, we designed two siRNAs targeting HSP27 (Figure 4A). Combining both siRNAs enhanced viral growth (Figures 4B,C), corresponding to elevated VP1 expression (Figure 4D). To further confirm the antiviral role of HSP27, A549 cells were transfected with increasing doses of an HSP27 expressing plasmid and then infected with EMCV over 24 h. We also overexpressed 2′,5′-oligoadenylate synthase 1 (OAS1) as a positive control as OAS1 is known to inhibit EMCV proliferation (Kristiansen et al., 2010). Overexpression of either OAS1 or HSP27 reduced VP1 levels in EMCV infected cells (Figure 4E), consistent with lower EMCV copy number and titers (Figure 4F and Supplementary Figure 2), suggesting that HSP27 can dampen EMCV proliferation. Collectively, our data show that HSP27 plays a negative regulatory role in the proliferation of EMCV.

Given that HSP27 can dampen EMCV infectivity, we wanted to see whether this is due to HSP27 affecting EMCV-triggered IFN response. Indeed, siRNA knockdown of HSP27 reduced EMCV-triggered IFN-α and IFN-β (Figure 5A). Consistent with this, HSP27 overexpression reversed this effect (Figure 5B). Since HSP27 can positively regulate IFN-I production triggered by EMCV, we predicted that 2C^pro and 3A^pro would also impact IFN-I production. Indeed, overexpression of 2C^pro and 3A^pro significantly blocked type I IFN production activated by poly (I:C) stimulation, synthetic dsRNA mimic (Figures 5C,D). Altogether, these findings suggest that HSP27 interacts with the type I IFN signaling cascade.

Given that EMCV mainly activates the IFN-I via the MDA5 signaling pathway (Gitlin et al., 2006; Kato et al., 2006), we assessed whether the positive regulatory effect of HSP27 on EMCV-triggered IFN response is due to its assistance to MDA5 signaling pathway components. Indeed, HSP27 knockdown led to a parallel loss of EMCV-triggered MDA5, MAVS, TBK1 protein expression levels (Figure 6A). Though the effect of HSP27 loss on EMCV-triggered IRF3 expression and its phosphorylation were not evident (Figure 6A), HSP27 overexpression did increase the expression of IRF3 and its phosphorylation, as well as the expression of other adaptor molecules in the MDA5 signaling pathway (Figure 6B). Consistent with the inhibitory role of HSP27 on EMCV infectivity (Figure 4), loss of HSP27 increased EMCV viral load by enhanced VP1 expression (Figure 6A). In summary, HSP27 positively regulates EMCV-triggered RLR/MDA5 signaling cascade. Given that MDA5, MAVS, TBK1, and IRF3 are activated by the same promoter, we then overexpressed HSP27 to verify the changes in the protein expression of endogenous MDA5, MAVS, TBK1, and IRF3 to reveal the specific targets of HSP27. Interestingly, we found that over-expression of HSP27 only significantly increased the protein expression of endogenous MDA5 and had no notable effect on the protein expression of endogenous MAVS, TBK1, and IRF3 (Figure 7A), indicating that HSP27 specifically stabilizes MDA5 protein expression. Co-IP assays also revealed that MDA5 directly binds to HSP27 (Figure 7B). In conclusion, we show for the first time that HSP27, as a molecular chaperone, positively regulates EMCV-triggered RLR/MDA5 signal pathway by stabilizing the expression of MDA5.