Activation of the PKR/eIF2α signaling cascade inhibits replication of Newcastle disease virus

HeLa cells were infected with NDV strain LaSota or Herts/33 at a multiplicity of infection (MOI) of 1. Cell pellets were collected at 6, 12, and 24 h post-infection (hpi), respectively. As shown in Figure 1A and B, infection by NDV strains LaSota or Herts/33 infection strongly induced PKR activation at 12 and 24 hpi. Furthermore, phosphorylation of eIF2α, the substrate of phosphorylated PKR, was observed at 12 and 24 hpi. To further verify this finding, we explored the NDV-induced phosphorylation of PKR and eIF2α at a dose of 0.5, 1, or 2 MOI for 12 h. Notably, western blot analysis results showed that PKR and eIF2α could be phosphorylated by NDV strains LaSota (Figure 1C) and Herts/33 (Figure 1D) in a dose-dependent manner.

To investigate whether viral replication is required for NDV-induced PKR and eIF2α activation, ultraviolet (UV) light-inactivated NDV Herts/33 was used for the experiment. The results showed that neither PKR nor eIF2α was phosphorylated at all timepoints post-infection (Figure 1E) and, furthermore, indicated that PKR activation is induced by NDV infection and not by interacting cells with non-infecting virus particles. Interestingly, total PKR protein content was not increased, suggesting that IFN-induced responses were not fully activated in the early phase of NDV infection.

Since PKR is considered a PRR for viral dsRNA, we deduced that NDV generates dsRNA during replication. To verify this deduction, monoclonal antibody (mAb) J2 recognizing the dsRNA molecules of more than 40 bp was used to detect intracellular dsRNA using an immunofluorescence assay. As shown in Figure 2, viral dsRNA particles were identified as bright dots distributed around the nuclei of the HeLa cells infected with NDV strain Herts/33 or LaSota, while no positive signals were found in cells treated with UV-inactivated viruses.

To determine the role of PKR in NDV replication, we transfected HeLa cells with PKR-specific siRNA to knockdown PKR and transfected other cells with non-targeting siRNA as a negative control. The results showed that siPKR efficiently impaired endogenous PKR expression (Figure 3A, B). Using this approach, we successfully inhibited PKR expression, which was followed by an apparent reduction of eIF2α phosphorylation. The reduction of endogenous PKR resulted in an obvious increase in viral NP synthesis (Figure 3A). Consistent with the increase of viral NP, the increase of virus titer in the supernatant of PKR knockdown cells was also analyzed and measured. The results showed that the reduction in PKR protein levels resulted in a 1.7-fold increase in extracellular virus yields (Figure 3C).

To further probe the role of PKR, we established HeLa cells overexpressing PKR through transfection of pcDNA-PKR plasmids. PKR protein levels in cells overexpressing PKR were obviously increased compared to that of normal cells (Figure 3D, E). PKR overexpression resulted in a decrease of viral NP synthesis (Figure 3D). Consistently, extracellular viral yields were decreased by 4.9-fold in cells overexpressing PKR infected with NDV strain Herts/33 (Figure 3F). Collectively, these data suggested that PKR plays an antiviral role in NDV-infected cells.

The translation initiation factor eIF2α, which has known involvement in protein translation by composing the ternary complex eIF2α-GTP-^MettRNA_i, was identified as a PKR substrate [16]. Recently, PKR-induced eIF2α phosphorylation followed by translation inhibition has been shown to play a role in inhibition of viral replication [17]. To investigate whether eIF2α was involved in NDV replication, we established eIF2α knockdown cells by transfecting siRNA specific to eIF2α or non-targeting siRNA into HeLa cells, followed by infection with NDV strain Herts/33. The efficiency of siRNA and the content of eIF2α and NP proteins were analyzed by western blot analysis. Transfection of sieIF2α resulted in a reduction of eIF2α protein level by 4.5 folds in mock-infected cells (Figure 4A, B). As shown in Figure 4A, viral NP protein was decreased by approximately 80% in eIF2α knockdown cells compared to that of control cells. Consistent with this finding, eIF2α knockdown resulted in a 2.6-fold increase in extracellular virus yield, indicating that eIF2α knockdown significantly impaired NDV replication (Figure 4C). Okadaic acid (OA) was defined as a protein phosphatase inhibitor, promoting eIF2α phosphorylation [18]. To further confirm the function of eIF2α in NDV replication, we augmented eIF2α phosphorylation in HeLa cells through OA treatment. PKR and eIF2α phosphorylation were significantly increased in OA-treatment cell, compared with that in mock cells (Figure 4D-E). To intuitively confirm whether eIF2α phosphorylation plays a role in NDV infection, green fluorescent protein (GFP)-expressing NDV ZJ1 strain (ZJ1-GFP) was utilized [19]. As shown in Figure 4G, in contrast to untreated cells, in the presence of OA, the percentage of GFP-positive cells was significantly decreased. Analysis of the spotlights under the fluorescence microscope showed that interfering with eIF2α dephosphorylation pharmacologically led to an approximate 3-fold decrease in intracellular ZJ1-GFP replication. Above all, these results indicated that eIF2α is required for NDV replication and eIF2α phosphorylation by active PKR induces an antiviral effect in NDV-infected cells.

PKR may serve as a PRR that mediates IFN production during WNV, SFV, and TMEV infection. On the other hand, HCV suppressed RIG-I translation thereby inducing IFN-β protein synthesis via PKR activation. To evaluate the role of PKR in type I IFN production after NDV infection, we first infected with HeLa cells with NDV strain Herts/33 and assessed the phosphorylation of PKR and eIF2α. Meanwhile, polyinosinic-polycytidylic acid [poly (I:C)] was used as a positive control. As shown in Figure 5B, IFN-β mRNA levels were increased by 25- and 451-fold in cells treated with poly(I:C) and NDV strain Herts/33, respectively, compared with mock-infected cells. The dramatic elevation of IFN-β mRNA levels in NDV-treated cells coincided with phosphorylation of PKR and eIF2α (Figure 5A). To further identify the role of PKR in IFN-β production, Hela cells were transfected with either siPKR or scrambled siRNA and then infected with NDV strain Herts/33. As expected, PKR knockdown significantly suppressed IFN-β production. These results demonstrated that PKR was involved in type I IFN production during NDV infection.

NDV lentogenic strain LaSota was obtained from China Institute of Veterinary Drug Control (Beijing, China). The velogenic strain Herts/33 and recombinant NDV ZJ1 strain expressing GFP (ZJ1-GFP) were kindly provided by Xiufan Liu (Yangzhou University, Jiangsu, China) [19]. To obtain incomplete viral replication, viral suspensions were irradiated with UV-C at 75 mWs/cm² using a low-pressure mercury vapor discharge lamp [38]. The absence of infective viruses after UV treatment was confirmed by the lack of replication in 9-day-old chicken embryonated eggs and in DF1 monolayer cultures. The human cervical cancer cell line (HeLa) and galline embryonic fibroblast cell line (DF-1) was purchased from ATCC (Manassas, VA, USA). Both cell lines were maintained in Dulbecco’s modified Eagle’s medium (DMEM) supplemented with 10% fetal bovine serum (FBS; Invitrogen, Carlsbad, CA, USA), 0.1 mg/mL of streptomycin (Sigma-Aldrich, St. Louis, MO, USA), and 100 U/mL of penicillin (Sigma-Aldrich) at 37°C in an atmosphere of 5% CO₂.

Rabbit polyclonal anti-total PKR (Santa Cruz Biotechnology, Inc., Dallas, TX, USA), rabbit polyclonal anti-phospho-PKR threonine 451 (Epitomics, Inc., Burlingame, CA, USA), rabbit polyclonal anti-eIF2α, rabbit polyclonal anti-phospho-eIF2α (Cell Signaling Technology, Inc., Danvers, MA, USA), and mouse monoclonal anti-β-actin antibody (Sigma-Aldrich) were used in the current study. Mouse monoclonal anti-NP mAb was generated using the NP of LaSota strain as an antigen expressed in a prokaryotic system. Briefly, sequences encoding the full length of NP gene was amplified by reverse transcription polymerase chain reaction and cloned into the bacterial expression vector pET32a (Novagen, Madison, WI, USA) in order to express the histidine-tagged protein heterologously in Eshcerichia coli strain BL21. The expression was induced by treatment with 0.5 mM isopropyl-β-D-thiogalactopyranoside at 37°C for 4 h, and the expression product was purified using a His Band kit (Novagen) according to the manufacturer’s instructions. BALB/c mice were immuned by purified proteins for 4 times and sacrificed. The spleen of BALB/c mice was obtained for monoclonal antibodies preparation as previously [39].

HeLa cells were infected with NDV at an MOI of 1 at 37°C. For dose-dependent assay, HeLa cells were infected with different doses (0.5 MOI, 1 MOI, and 2 MOI) of NDV. After a 1-h absorption period, the cells were washed three times with cold phosphate-buffered saline (PBS) to remove unattached viruses and then cultured in DMEM containing 1% FBS. At 6, 12 and 24 hpi, the cell pellets and supernatant were harvested and stored at −80°C for further use. The replication of NDV in the cells was determined as the expression of NDV NP protein with monoclonal antibody against NDV NP protein using western blotting. Extracellular viral yields in cell supernatant were determined by estimating the median tissue culture infective dose (TCID₅₀) in DF1 cells using the Reed and Muench method [40]. In the transfection assay, cells were transfected with eukaryotic expression plasmids or siRNA using lipofectamine 2000 transfection reagent (Invitrogen). 24 h (plasmid transfection) or 48 h (siRNA transfection) post transfection, cells were infected with NDV at an MOI of 1 and harvested at 12 hpi. For pharmacological experiments, HeLa cells were infected with ZJ1-GFP at an MOI of 1 or mock infected for 9 h in the presence of OA (10 nM). The number of GFP-positive cells was quantified in 10 randomly chosen fields (magnification × 100) using Image J software (http://​imagej.​nih.​gov/​ij).

HeLa cells were lysed in lysis buffer (50 mM Tris–HCl, pH 8.0, 150 mM NaCl, 20 mM NaF, 1 mM ethylenediaminetetraacetic acid, 1% Triton, 1 mM phenylmethylsulfonyl fluoride, 1 mM ethylene glycol tetra-acetic acid, 20 mM Na₄P₂O₇, 1 mM Na₃VO₄) containing the protease inhibitors leupeptin (0.5 μg/mL), aprotinin (0.5 μg/mL), and pepstatin (0.7 μg/mL). Cells were sonicated for 1 second using a Vibra Cell VCX130 sonicator (Sonics Vibra cell; Sonics & Material, Newtown, CT, USA), boiled for 5 min, and then cleared by centrifugation for 10 min at 12,000 g at 4°C. The lysates were further denatured by incubation for 5 min at 95°C in SDS-PAGE sample loading buffer (Beyotime, Nantong, China). The samples were further separated on 10% polyacrylamide gel (Bio-Rad Laboratories, Inc., Hercules, CA, USA), and transferred to a nitrocellulose membrane at 250 mA for 90 min, and then blocked using Tris buffered saline (TBS) solution containing 0.1% Tween 20 and 5% non-fat milk for 2 h. The membrane was incubated overnight at 4°C with the indicated antibodies. After washing three times with TBS/0.1% Tween 20, the membrane was exposed to horseradish peroxidase-conjugated anti-rabbit or anti-mouse IgG antibody (Sigma-Aldrich) at a dilution of 1:10000 for 2 h at room temperature. The protein bands were visualized with an enhanced chemiluminescent reagent (Pierce Biotechnology, Rockford, IL, USA) and quantified using Image J software.

Human PKR cDNA was amplified from HeLa cell mRNA and inserted into the HindIII/XhoI sites of pcDNA3.1(+) to construct pcDNA3.1-PKR. The purified plasmid DNA was prepared using the HiSpeed plasmid maxi kit (Qiagen, Hilden, Germany). The transfection procedure was performed essentially as described in the manufacturer’s instructions, but with minor modifications. Briefly, HeLa cells (1 × 10⁶/well) were seeded into each well of a 6-well tissue culture plate and cultured at 37°C overnight. On the following day, 2 μg of plasmid was incubated for 45 min with 5 μL of Lipofectamine 2000 transfection reagent (Invitrogen) and then the transfection was allowed to occur overnight before the media was replaced with fresh DMEM. At 24 h post-transfection (hpt), cells were infected with Herts/33 strain at an MOI of 1 and harvested at 12 hpi.

Short interfering RNA oligos against PKR or eIF2α targeted to specific sequences (siPKR: 5′-GCGAGAAACUAGACAAAGU -3′ [41]; eIF2α: 5′-GAGAGGCUUGAAAGAGAAA-3′ [42]; non-targeting siRNA oligos: 5′-UUCUCCGAACGUGUCACGU-3′) were custom synthesized by GenePharma (Shanghai) Co., Ltd. (Shangai, China). HeLa cells were cultured to 60–70% confluency in a 6-well plate at 37°C overnight. PKR-specific siRNA, eIF2α-specific siRNA, and nontargeting siRNA as a negative control (100 pmol/well) were transfected with Lipofectamine 2000 according to the manufacturer’s instructions. At 6 hpt, the culture media was replaced with DMEM supplemented with 1% FBS (v/v). After a 48-h incubation period, the transfected cells were infected with NDV strain Herts/33 at an MOI of 1 and harvested at 12 hpi.

Total RNA of HeLa cells infected with NDV strain Hert/33 or PKR-silenced HeLa cells were extracted using the RNeasy kit (Qiagen, Valencia, CA, USA) and then reverse transcribed to cDNA using the SuperScript® III First-Strand Synthesis System kit (Invitrogen) with oligo dT primers (Invitrogen) [43]. To determine mRNA levels of IFN-β and β-actin, real-time PCR was performed using the SYBR Premix Ex Taq kit (TaKaRa Biotechnology (Dalian) Co., Ltd., Dalian, China) according to the instruction manual. The specific primers targeting IFN-β (forward: 5′-ACGACAGCTCTTTCCATGA-3′; reverse: 5′-AGCCAGTGCTCGATGAATCT-3′) and β-actin (forward: 5′-GATCTGGCACCACACCTTCT-3′; reverse: 5′-GGGGTGTTGAAGGTCTCAAA-3′) used in this study were described previously [43, 44]. The expression levels of target mRNA were normalized to that of β-actin and calculated by the comparative C_T method using the formula: fold change = 2^−ΔΔCT, where ΔΔCT = [(CT _{target mRNAs} –CT _β-actin) in HeLa cells infected with NDV – (CT _{target mRNAs} – CT _β-actin) in mock cells] [45].

HeLa cells were grown on coverslips overnight and infected with NDV strain Herts/33 at an MOI of 1 or stimulated with poly (I:C) for 12 h. Cells were fixed in 3% paraformaldehyde, permeabilized with 0.5% Triton X-100 for 10 min, incubated in blocking buffer (3% BSA/PBS), and then stained with mouse mAb J2 (Scicons, Hungary) followed by Cy3-labeled goat anti-mouse IgG (Beyotime). Nuclei were stained with 4’,6-diamidino-2-phenylindole (DAPI) (Thermo Fisher Scientific Inc., Rockford, IL, USA) at a dilution of 1:500 for 10 min, and then the coverslips were mounted on slide glasses and visualized using a fluorescence microscope (Nikon Eclipse 80i; Nikon, Tokyo, Japan).