African green monkey kidney cells (Vero) and human embryonic kidney (HEK293T) were obtained from Lanzhou Veterinary Research Institute (LVRI), Lanzhou, China. All cells were sub-cultured and maintained in Dulbecco's Modified Eagle's medium (DMEM) supplemented with 10% fetal bovine serum (FBS) treated with 1% penicillin–streptomycin (Gibco, Grand Island, NY, USA). Infected or transfected cells were maintained in DMEM supplemented with 2% FBS treated with 1% penicillin–streptomycin. All cells were grown or maintained at 37°C supplied with 5% of CO₂. A vaccine strain of PPRV Nigeria 75/1 (accession No. X74443) at a titer of 10^4.6 TCID₅₀was obtained from the frozen stock (−80°C) of LVRI, China. During experiments, Vero cells were infected with PPRV at the multiplicity of infection (MOI) = 3, or mock-infected with free culture media and incubated at 37°C supplied with 5% CO₂ at indicated times according to the experimental designs. At suitable times, PPRV-infected or mock-infected cells were either treated or left untreated, and then harvested and prepared for either real-time quantitative polymerase chain reaction (RT-qPCR) or immunoblotting, or stained for immunofluorescence assays (IFAs). A mouse monoclonal antibody against N-protein of PPRV was obtained from Dr. Xuelian Meng, LVRI, China. The commercial antibodies and reagents used are as follows: Rabbit anti-PERK/EIF2AK3 polyclonal antibody (24390-1-AP), Rabbit anti-EIF2S1 polyclonal antibody (11170-1-AP), Rabbit anti-GADD34 polyclonal antibody (10449-1-AP), Rabbit anti-ATF4 polyclonal antibody (10835-1-AP), Mouse anti-His-Tag monoclonal antibody (66005-1-Ig), Rabbit anti-G3BP1 (13057-2-AP), Rabbit anti-TIA-1 (12133-2-AP), and Mouse anti-β-actin. Monoclonal antibody (60008-1-ig) was purchased from Proteintech; Rabbit anti-phospho-PERK (Thr982) polyclonal antibody (DF7576) was purchased from Affinity Biosciences; Rabbit anti-EIF2S1(phosphoS51) antibody (ab32157), Rabbit anti-mouse IgG H&L (HRP) (ab6728), Goat anti-rabbit IgG H&L (HRP) (ab205718), Rabbit Anti-DDDDK tag antibody (equivalent to FLAG antibodies from Sigma) (ab1162), Mouse anti-ATF6 (ab11909), Rabbit anti-Bip/GRP78 (ab21685), Rabbit anti-GFP (ab6556), Goat anti-rabbit IgG H&L (Alexa Fluor 488) (ab150077), and Goat anti-mouse IgG H&L (Alexa Fluor 647) (ab150115) were purchased from Abcam. Halt Protease and Phosphatase Inhibitor cocktail (78442) and Rabbit anti-GADD34 (PA1-139) polyclonal antibody were purchased from Thermo Fisher Scientific, USA. Rabbit anti-CHOP (D46F1) monoclonal antibody was purchased from Cell Signaling. RIPA cell lysis buffer (P0013B) was purchased from Beyotime Biotech.

The full-length ORFs of the PPRV P and PPRV N genes were PCR amplified from cDNA synthesized from Vero cells infected with PPRV. A His-tag was added to the C-terminal of the P gene by PCR cloning and ligated via MluI and NotI cloning sites of the pCI mammalian expression vector purchased from Promega (Madison, WI, USA), under the control of T7polymerase promoter to generate a pCI-P His plasmid expressing a His-tagged PPR P protein. The ORF of N genes was inserted by fusion with Flag tag into pCMV-tag 2 mammal expression vector (Clontech) between Srf I and EcoRI to generate a Flag-pCMV-N protein-expressing Flag-tagged PPRV N protein under the control of the CMV promoter. All plasmid constructs were used to transform DH5α competent cells (Takara). A stock of purified His-tagged pCI-P and Flag-tagged pCMV-N plasmids was aliquoted and kept at −20°C until use. Inserts of plasmid constructs were confirmed by DNA sequencing. A Flag-tagged GADD34 plasmid (pXJ40 GADD34 Wt) (1–674aa) was previously constructed and published (26). Primer sequences used for plasmid construction in this study are listed in Table 1.

When HEK293T or Vero cells were grown at 70–80% confluence, the cells were transfected or co-transfected with appropriate amount of either pCI-P His/Flag-pCMV-N or pXJ40 GADD34 plasmids, or mock-transfected with the same amount of empty pCI/ pCMV vector or free transfection reagent as a negative control using Attractene Transfection Reagent (301007) (Qiagen) according to the manufacturer's instructions. At the indicated time, cells were lysed and analyzed by SDS-PAGE followed by Western blotting or stained for IFA analysis.

When Vero cells were grown at 70–80% confluence, the cells were infected with PPRV at MOI = 3 or mock-infected. After 2-h adsorption, infection media was changed to fresh media treated with either different concentration of Salubrinal (CAS405060-95-9) or 200 nM integrated stress inhibitor (ISRIB) (CAS1597403-47-8) (Santa Cruz Biotechnology, Santa Cruz, CA, USA) at indicated times. Cells were harvested and total RNA was extracted for RT-qPCR or lysed for SDS-PAGE followed by immunoblotting analysis, respectively, for PPRV expression. Vero cells infected with PPRV or mock-infected were incubated and then treated with 2 mM dithiothreitol (DTT) (CAS27565-41-9) (Santa Cruz) or 1 μM Thapsigargin (TG) (HY-13433) (MedChemExpress) for 2 h up to 48 h.p.i. for DTT or 12 h up to 60 h.p.i. for TG, respectively, to induce ER stress. Cells were harvested and lysed for SDS-PAGE analysis followed by immunoblotting of target proteins.

Infected or transfected cells were washed twice with phosphate-buffered saline (PBS) and lysed in appropriate lysis buffer supplemented with a protease and phosphatase inhibitor cocktail according to the manufacturer's instruction and centrifuged at 16,000 × g for 10 min. The supernatant was collected and mixed with appropriate protein loading buffer and boiled at 95°C for 5 min. An equal amount of protein was loaded and separated by SDS-PAGE and transferred to polyvinylidene difluoride (PVDF) membrane according to the standard protocols. Target proteins were then probed with specific primary antibodies overnight, followed by incubation for 1 h with an appropriate horseradish peroxidase-conjugated (HRP) secondary antibody at proper dilutions. Corresponding signals bands were detected using Millipore Immobilon Western Chemiluminescent HRP Substrate (WBKLS0100).

Vero or HEK293T cells grown directly on glassware dishes for 12 h at 70–80% confluence were infected/transfected/ or mock-infected/transfected at indicated times. Cells were washed thrice for 5 min with TBS, fixed with 4% paraformaldehyde (PFA) for 30 min at room temperature, and permeabilized with 0.2% Triton X-100-TBS for 5 min. To avoid non-specific bindings, cells were blocked using 5% BSA-TBS for 1 h at room temperature. A proper dilution of primary antibodies was incubated overnight in 1% BSA-TBST at 4°C followed by three washes and incubation with appropriate secondary antibodies for 45 min at room temperature. Stained cells were mounted with 4,6-Diamidino-2-Phenylindole (DAPI) and scanned with confocal fluorescence microscopy TCS SP8 (Leica Microsystems).

Viral RNA was extracted from samples using the RNeasy Mini Kit (74104) (Qiagen) according to the manufacturer's instruction. RNA was eluted in nuclease-free water, quantified by Nanovue (GE Life science, USA), aliquoted, and stored at −80°C until use. The RT-qPCR assay used in this study was performed in a one-step RT-qPCR system as previously described (27, 28) with some modifications. Briefly, a single tube of 25-μl reaction was prepared with 12.5 μl of Superscript III/Platinum Taq One-step RT-qPCR reaction mix, 1 μl of Superscript III/Platinum Taq One-step RT-qPCR enzyme mix, 5 pmol Taqman probes, 10 pmol of forward/reverse primers each, and 2 μl of total RNA as a template. Cycling conditions were set as follows: 50°C for 15 min (1 cycle), 95°C for 10 min (1 cycle), and 95°C for 15 s followed by 60°C for 1 min (40 cycles). The reaction was run into the Agilent Technologies Stratagene Mx3005P thermocycler (Life Technologies, USA). The standard RNA used for PPRV quantification in RT-qPCR assay was prepared as previously described (27) using full-length ORF of PPRV N protein cloned into the pCMV-Tag2 vector as described in section Plasmid Construction.

All the data are presented as mean ± standard deviation (SD) from three replicates (n = 3) of at least two independent experiments. The densitometry quantification of the relative intensity band ratio for targeted proteins/β-actin was determined by ImageJ analysis. Two-way ANOVA or multiple t-tests were used for statistical analysis with GraphPad Prism 8.0 software with P-value (^*P < 0.05) considered statistically significant.

Several viruses modulate p-eIF2α levels during replication to ensure viral protein synthesis and avoid cellular stress responses. To examine the phosphorylation status of eIF2α in PPRV infection, Vero cells were infected or mock-infected with PPRV at indicated times. Cell lysates were analyzed by SDS-PAGE followed by Western blot for the expression of target proteins. Viral infection was confirmed by Western blot probed with a mouse monoclonal anti-PPRV N antibody (Figures 1A,B). eIF2α and phosphorylated eIF2α (p-eIF2α) at the serine 51 position were analyzed by Western blot stained with specific antibodies. The relative levels of p-eIF2α were normalized to β-actin as an internal loading control protein. Densitometry quantification of p-eIF2α was calculated by ImageJ analysis for 36, 48, 60, and 72 h post-infection (h.p.i.). As shown in Figures 1C,D, the levels of p-eIF2α were significantly lower in infected cells compared to mock-infected cells, indicating that PPRV infection does potentially repress eIF2α phosphorylation. To further investigate the effect of PPRV infection on cellular protein translation, we also examined the expression of the protein kinase R (PKR-like) ER kinase (PERK), which is a well-characterized ER stress-induced translational control pathway (29). PERK and phosphorylated PERK (p-PERK) at Threonine 982 position were analyzed by Western blot stained with specific antibodies. As shown in Figures 1C,D, the results showed a significant downregulation in p-PERK (Figures 1C,D), indicating that PPRV infection does potentially repress both PERK and eIF2α phosphorylation.

The molecular mechanism of eIF2α dephosphorylation has been previously described on a feedback inhibition of the unfolded protein response (UPR) by the GADD34 (30). To analyze whether the predicted PPRV-mediated eIF2α repression was GADD34 dependent, new sets of Vero cells were infected with PPRV or mock-infected and then the effect of PPRV on GADD34 expression was analyzed. Whole-cell lysates were prepared and analyzed for 36, 60, and 72 h.p.i. by Western blot. As expected, the level of GADD34 was significantly higher after 36 h.p.i. in PPRV-infected compared to mock-infected cells (Figures 2A,B). Analysis by IFA also showed higher GADD34 fluorescence in PPRV-infected Vero cells than in mock-infected cells (Figure 2C). Reciprocally, to analyze the effect of GADD34 on PPRV replication, another set of Vero cells were transiently transfected with a construct expressing Flag-GADD34 plasmid (26) or mock-transfected with a free transfection reagent before infection with PPRV or mock infection and subjected to IFA analysis. In Vero cells transiently expressing Flag-GADD34 before infection (12 h), the fluorescence of Flag-GADD34 was highly enhanced compared to that of mock-infected cells (see Figures 2C,D). It was interesting to note that in cells overexpressing GADD34 prior to infection, this protein was more concentrated in both the cytoplasm and nucleus under the effect of PPRV infection (Figure 2D). These results suggest a strong reciprocal interaction between PPRV replication and GADD34 expression, which may be linked to the aforementioned PPRV-mediated eIF2α dephosphorylation.

To further confirm that PPRV was modulating eIF2α phosphorylation, Vero cells were infected with PPRV or mock-infected followed by ER stress induction through treatment with either 2 mM DTT for 2 h to a total of 48 h.p.i. or 1 μM TG for 12 h to a total of 60 h.p.i. DTT is a strong ER stress inducer that can induce ER stress in a very short time, while TG is a SERCA-pump inhibitor that inhibits calcium-dependent chaperones, therefore increasing protein misfolding. Both chemicals are known to cause ER stress in several cell lines (31). Cell lysates were analyzed for p-eIF2α (S51), while β-actin was used as an internal control. The results showed that the level of p-eIF2α was very high in all mock-infected cells treated either with DTT or TG, but significantly decreased in PPRV-infected cells treated with DTT or TG. Likewise, neither DTT nor TG could alter the levels of GADD34 in the presence or absence of PPRV (Figures 3A–D). The level of p-eIF2α was significantly more repressed in PPRV-infected groups treated with DTT (Figure 3A, line 4) than in mock-infected treated with DTT (Figure 3A, line 2). In contrast, GADD34 was more upregulated in all PPRV-infected cells treated or untreated with DTT (Figure 3A, lines 3 and 4) compared to mock-infected cells (Figure 3A, lines 1 and 2). In PPRV-infected and TG-treated groups, the level of p-eIF2α was more repressed (Figure 3C, lines 8–10) compared to mock-infected groups treated in either the same condition (Figure 3C, lines 1–3) or with DMSO alone (Figure 3C, line 4). GADD34 was more upregulated in PPRV-infected groups treated either with or without TG (Figure 3C, lines 4–10) compared to mock-infected groups (Figure 3C, lines 1–3). These results support the previous observations that PPRV infection represses eIF2α phosphorylation and potentially induces GADD34. Additionally, the results demonstrate the ability of PPRV, to an extent, to evenly repress the chemically triggered ER stress-mediated eIF2α phosphorylation. Interestingly, cells treated with TG exhibited a reduced expression of PPRV N protein, therefore suggesting a potential ability of TG to inhibit PPRV replication. However, the reduced expression of PPRV N protein following the treatment with TG did not completely suppress its ability to repress eIF2α phosphorylation (Figure 3C, lines 8–10). These results provide evidence that PPRV can dephosphorylate eIF2α, even in the presence of some ER stressors such as DTT and TG.

To further investigate the role of eIF2α dephosphorylation in PPRV replication, two chemicals with known agonist/antagonist effects on the eIF2α phosphorylation were used to analyze PPRV replication in vitro: (1) Salubrinal is a selective inhibitor of eIF2α dephosphorylation (32, 33), and (2) integrated stress inhibitor (ISRIB) is known to reverse eIF2α phosphorylation and the restoration of cellular translation (34). Vero cells were infected with PPRV or mock-infected. After 2-h adsorption, infection media was changed to fresh media treated with different concentrations (5, 10, and 25 μM) of Salubrinal, followed by incubation up to 60 h.p.i. as described in the previous study (35). On the other hand, PPRV-infected or mock-infected cells were treated with 200 nM ISRIB, as described in the previous study (34), followed by incubation up to 48 h.p.i. Both treated and untreated samples were analyzed by RT-qPCR, as described in section Materials and Methods, or loaded on SDS-PAGE followed by probing with a monoclonal antibody against PPRV N protein to examine the expression of PPRV N protein after treatment. As shown in Figure 4A, there was a non-significant difference in expression of PPRV N protein in untreated cells (Figure 4A, line 1) and cells treated with DMSO alone (Figure 4A, line 2). In contrast, the relative expression of PPRV N protein decreased in a dose-dependent manner in cells treated with different concentrations of Salubrinal (Figure 4A, lines 3–5). Interestingly, following the treatment with Salubrinal, the level of PPRV N protein expression was reduced in correlation with the inhibition of GADD34 (Figure 4A, lines 3–5). Conversely, the expression of PPRV N protein was significantly reduced in cells treated with DMSO alone (Figure 4C, lines 5 and 6) when compared to cells treated with ISRIB (Figure 4C, lines 1 and 2) at 48 h.p.i. Moreover, in cells treated with ISRIB, the level of p-eIF2α was significantly decreased while the level of GADD34 significantly increased (Figure 4C, lines 1–4). At the mRNA level, similar results were observed for Salubrinal or ISRIB-treated cells analyzed by RT-qPCR, as shown in Figures 4E,F, respectively. The effect of ISRIB was more significant at the mRNA level compared to the protein expression level, as can be seen in Figures 4D,F. These data indicate that the dephosphorylation of eIF2α is beneficial to PPRV replication.

In other paramyxoviruses, the P protein performs multiple functions. For example, in RPV, the P protein was shown to be required for replication/transcription (36), whereas in measles virus (MV), the interaction of P-N is required for biological processes such as transcription and regulation of protein translation, and controlling the cell cycle (9). In PPRV, the role of the P protein is not fully understood. To investigate the role of PPRV P protein in PPRV-mediated PERK/eIF2α repressions, HEK 293T cells were transfected with a His-tagged PPRV P protein-expressing plasmid or mock-transfected with an empty vector. The levels of p-PERK and p-eIF2α expression were then analyzed in comparison with mock-transfected cells. The results showed that the PPRV P protein alone was able to repress PERK/eIF2α phosphorylation in HEK293T cells (Figures 5A–C). To further investigate whether PPRV P protein, but not PPRV N protein could induce p-eIF2α downregulation, a Flag-tagged PPRV N protein expressing plasmid was used to transfect HEK293T cells and analyzed for the expression of p-eIF2α at indicated times. Results show that the level of p-eIF2α in cells transfected with PPRV N protein did not exhibit any significant p-eIF2α downregulation (Figures 5D,E). These results indicate that the PPRV P protein, but not the PPRV N protein, is at least one factor involved in PPRV-mediated eIF2α dephosphorylation.

It has been suggested that protein synthesis inhibition is an important protective mechanism against various stresses, including viral infection and ER stress. This immune response is mainly based on the phosphorylation state of eIF2α (37, 38). We have demonstrated the ability of PPRV to dephosphorylate eIF2α and a potential reciprocal interaction between PPRV replication and GADD34 upregulation during infection. Furthermore, we described the potential of PPRV P protein to inhibit PERK/eIF2α phosphorylation. It is known that GADD34 plays a role in pro-apoptotic or pro-survival downstream of the PERK/eIF2α pathway. To preliminarily investigate the probable molecular mechanism by which PPRV P protein blocks eIF2α phosphorylation, the status of the transcriptional related proteins downstream of the PERK/eIF2α pathway was investigated. HEK 293T cells were transfected with His-tagged P protein expressing plasmid or mock-transfected with an empty vector at indicated times. Whole-cell lysates were separated by SDS-PAGE and analyzed by Western blot (Figure 6A) or IFA (Figure 6C) for the protein expression state of transcription factor-4 (ATF4), GADD34, and pro-apoptotic transcription factor CCAT/enhancer-binding protein (CHOP). Results showed that in both Western blot and IFA, GADD34 and ATF4 were highly upregulated in the presence of PPRV P protein compared to vector-transfected cells, whereas the expression of CHOP showed a slight attenuation (Figures 6A,C). Surprisingly, the active form of ATF4 was markedly expressed in PPRV P protein-transfected cells but migrated abnormally on SDS-PAGE in a seemingly spliced manner, resulting in a higher molecular weight (approximately 70 kDa) than expected (50–58 kDa) (Figure 6A). To further investigate whether upregulation of GADD34 and ATF4 proteins under the effect of PPRV P protein was not common to other PPRV structural proteins, another set of HEK 293T cells were transfected (in the same condition) with Flag-tagged N protein or mock-transfected with an empty vector. Cell lysates were analyzed by Western blot probed with specific antibodies to detect GADD34, ATF4, and CHOP. As shown in Figure 6B, it was clear that, unlike PPRV P protein, the PPRV N protein could not activate either GADD34 or ATF4. In contrast, transfection of the individual PPRV N protein slightly decreased the level of GADD34 compared to mock-transfected cells from 24 h.p.t., whereas the ATF4 remained inactive. It is interesting to note that CHOP seems to be upregulated at 48 h.p.t. in PPRV N protein-transfected cells (Figure 6B). Taken together with the conclusion in section Viral P protein is involved in dephosphorylation of eIF2α during PPRV infection, it is reasonable to exclude PPRV N protein from the process of PPRV-mediated GADD34 upregulation and its subsequent eIF2α dephosphorylation. Consider that, in Figures 2C,D, we observed that GADD34 was more concentrated in the nucleus of Vero cells infected with PPRV and that PPRV P protein alone could upregulate GADD34 expression. Hence, we hypothesized that the P protein was involved in the cellular location of GADD34 in cells under PPRV infection. To investigate this, we co-transfected HEK293T cells with plasmids expressing His-tagged P and Flag-tagged GADD34 proteins for 36 h to visualize the location of the two proteins. Cells were fixed and subjected to IFA, as described in Materials and Methods, followed by double staining with anti-His and anti-Flag. The results showed that PPRV P protein and GADD34 were transiently co-expressed. The distribution of GADD34 in the nucleus was enhanced and consistent in co-transfected cells compared to cells transfected with P protein alone, as seen in Figures 6C,D. Interestingly, observed Vero cells transfected with GADD34 prior to PPRV infection showed a similar enhancement of GADD34 distribution in the nucleus and a more pronounced PPRV foci (see Figure 2D), suggesting a reciprocal effect on each other. Altogether, the results in this section indicate that PPRV P protein alone can induce upregulation of host cellular GADD34, thus supporting its involvement in the PPRV-mediated eIF2α dephosphorylation.