Figure 1.
Interaction between P and PLK1.
(A). Amino acid residue differences among N-termini of wild type P, Pcpi− and Pcpi+. (B). Interaction between P and endogenous PLK1 in infected cells. HeLa cells were infected with mock, PIV5 or rPIV5-CPI+ and immunoprecipitated (IP) with anti-P antibody. The immunoprecipitated proteins were resolved in SDS-PAGE and subjected to immunoblotting (IB) with antibody against PLK1. (C). Interaction between P and PLK1 in transfected cells. BSR T7 cells were transfected with plasmids encoding P or Pcpi+ along with Flag-PLK1. Vector (pCAGGS) was used to keep the total amount of transfected plasmids constant. At 18–20 hours post transfection, the transfected cells were metabolically labeled with 35S-Cys/Met for 3 h at 37°C. The cells were lysed by RIPA buffer and immunoprecipitated with Pk antibody or Flag antibody. The precipitated proteins were resolved by SDS-PAGE gel and visualized using a PhosphorImager.
Figure 2.
Effects of PLK1 inhibitor and over-expression of PLK1 on viral gene expression.
(A). Effect of PLK1 inhibitor BI 2536 on recombinant PIV5 expressing Renilla luciferase (rPIV5-RL). HeLa or BSR T7 cells in 24-well plates were infected with rPIV5-RL at MOI of 1 and incubated with BI 2536 at concentration of 0.01, 0.05, 0.1, 0.2, 0.5 and 1 µM. Renilla luciferase activity was measured at 18–20 hours post infection and average luciferase activity +/− standard derivation of mean (SEM) is shown. (B). Effect of PLK1 inhibitor BI 2536 on viral gene expression. HeLa cells were mock-infected or infected with PIV5 or rPIV5-CPI+ and incubated in 1 µM BI 2536. The cells were metabolically labeled with 35S-Met/Cys at 18–20 hours after infection and immunoprecipitated with Pk antibody, which recognizes V and P and co-precipitates NP and L. The right panel is labeled cell lysates without immunoprecipitation. (C). Quantification of Figure 2B. Six independent experiments were performed and the relative amount of the P protein was measured. The level of PIV5 P in infected cells with DMSO treatment was set at 100 and the rest was normalized to the level of PIV5 P. The average of relative expression level of P +/− SEM is shown. (D). Effects of PLK1 inhibitor on the mini-genome system. A mini-genome plasmid (pSMG-RL) that contains a Renilla luciferase (RL) reporter gene and from which a negative-sense mini-genome is generated from T7 RNA polymerase transcription in BSR T7 cells was described previously [21]. In the presence of NP and L and P or P from rPIV5-CPI+ (Pcpi+), this negative-sense RNA template is replicated and transcribed to give rise to the RL mRNA, resulting in luciferase activity. The pT7-FF-Luc plasmid, which contains an FF-Luc reporter gene as a transfection efficiency control, was transfected along with the plasmids. Firefly and Renilla luciferase activities were detected in cell lysates at 18 to 20 h post-transfection, as described in Experimental Procedures. BSR T7 cells were transfected with necessary plasmids and incubated with 0.25 µM BI 2536. At 18–20 hours post transfection, the cells were lysed and dual luciferase assay was performed. The relative luciferase activities are calculated at ratios of Renilla luciferase activity (indicative of mini-genome replication) versus FF-Luc activity (indicative of transfection efficiency). The average of relative luciferase activity +/− SEM is shown. (E). Effects of PLK1 overexpression on the mini-genome system. Plasmids encoding pCAGGS-Flag-PLK1 at various concentrations were transfected along with the mini-genome system as above. Dual luciferase assay was carried as described above. An aliquot of cell lysate was used for immunoblotting to detect expression levels of Flag-PLK1 (the bottom panel).
Figure 3.
Effect of kinase-deficient PLK1 on PIV5 protein expression.
(A). Effects of overexpression of PLK1 kinase-dead mutant K82M on the mini-genome system. BSR T7 cells in 24-well plates were transfected with various concentrations of pCAGGS-Flag-PLK1 or pCAGGS-Flag-PLK1-K82M together with the plasmids necessary for the PIV5 mini-genome system. Dual luciferase assay was carried out after 18–20 h post transfection as before. NC: negative control, no plasmid encoding P; PC: positive control, no plasmid encoding PLK1 or PLK1 K82M. The relative activity unit of positive control was set as 100. The average of relative luciferase activity +/− SEM is shown. An aliquot of cell lysates was used for immunoblotting to detect the expression level of Flag-PLK1 or Flag-PLK1-K82M. (B). Interaction between P and PLK1-K82M. A plasmid encoding P was transfected into cells with a plasmid encoding Flag-PLK1 or Flag-PLK1-K82M. The cells were metabolically labeled and immunoprecipitated with anti-P or anti-Flag.
Figure 4.
Effects of PLK1 inhibitor on phosphorylation of P.
(A). Effects of BI 2536 on phosphorylation of P. Cells were mock-infected or infected with PIV5 or rPIV5-CPI+. At 18–20 hours post infection, the cells were metabolically labeled with 35S-Cys/Met or 33P-orthophosphate in the presence of 1 µM PLK1 inhibitor BI 2536 or DMSO as described in Experimental Procedures. The cells were lysed and immunoprecipitated with anti-P Pk antibody. (B). Quantification of effects of BI 2536 on phosphorylation of P. Three individual experiments as described in (A) were quantified. The average ratio of phosphorylated P from the 33P-labeling experiment to total amount of P as 35S-labeled P in PIV5 infected cells treated with DMSO was set to 100 and the others are normalized to the ratio. The average of relative level of phosphorylation of P +/− SEM is shown.
Figure 5.
PLK1 phosphorylates the P protein at S308.
(A). Prediction of PLK1 phosphorylation site within P protein. The center bold S/T residue is the PLK1 phosphorylation site. X, any amino acid residue; Ψ, amino acid residue with a hydrophobic side chain. (B). Effect of mutating S308 into A on the mini-genome activity. A mini-genome system as described before was used. To ensure a validity comparison of P vs. P-S308A, a range of concentrations of the P proteins in the experiments were used. Cell lysate aliquots of the transfected cells were subjected to immunoblotting using anti-NP or anti-P antibody. The average of relative luciferase activity +/− SEM is shown. (C). Effects of PLK1 inhibitor on the mini-genome system using P-S308A. To study the effect of PLK1 inhibitor on mini-genome activity with mutant P-S308A, BI 2536 was added to the mini-genome system using P-S308A in a similar fashion as in Fig. 2D. (D). Effects of PLK1 overexpression on the mini-genome system using P-S308A. Plasmids encoding pCAGGS-Flag-PLK1 at various concentrations were transfected along with the mini-genome system as above. Dual luciferase assay was carried as described above. An aliquot of cell lysate was used for immunoblotting to detect expression levels of Flag-PLK1. (E). Interaction between P-S308A and PLK1. A plasmid encoding Flag-PLK1 was transfected into cells with a plasmid encoding P or P-S308A. The cells were metabolically labeled and immunoprecipitated with anti-P or anti-Flag. (F). Phosphorylation of P by PLK1 in vitro. HeLa cells were transfected with plasmids pCAGGS-P or pCAGGS-P-S308A. P and P-S308A were purified from transfected cells using anti-P as described in Experimental Procedures. Kinase assays were carried out in a total volume of 30 µl containing 100 ng PLK1 for 60 minutes at room temperature. The bottom panel is the input of P and P-S308A with PLK1 on a SDS-PAGE with Coomassie blue staining.
Figure 6.
Effect of mutating S to A at position 308 in recombinant virus.
(A). Sequences of PIV5, rPIV5-V/P-S157A and rPIV5-P-S308A at regions around residue 157 and 308 of the P protein. The entire genome of the mutant viruses was sequenced and no mutation was observed except in the designated site (S157A or S308A). (B). Viral gene expression from PIV5, rPIV5-V/P-S157A and rPIV5-P-S308A-infected cells. HeLa cells were infected with PIV5, rPIV5-V/P-S157A, or rPIV5-P-S308A. At 16 hours post infection, the infected cells were processed for flow cytometry using anti-HN antibody as described in Experimental Procedures. Average mean fluorescence intensity of infected cells +/− SEM is graphed. (C). Effects of PLK1 inhibitor on phosphorylation of P of rPIV5-P-S308A. Cells were mock-infected or infected with PIV5 or rPIV5-P-S308A. At 18–20 hours post infection, the cells were metabolically labeled with 35S-Cys/Met or 33P-orthophosphate in the presence of 1 µM PLK1 inhibitor BI 2536 or DMSO as described in Fig. 4A. The cells were lysed and immunoprecipitated with ant-P Pk antibody. (D). Quantification of effects of BI 2536 on phosphorylation of P of rPIV5-P-S308A. Three individual experiments as described in (C) were quantified. The average ratio of phosphorylated P from the 33P-labeling experiment to total amount of P as 35S-labeled P in PIV5 infected cells treated with DMSO was set to 100 and the others are normalized to the ratio. The average of relative phosphorylation level of P +/− SEM is shown.
Figure 7.
Induction of apoptosis and cytokine expression by mutant viruses.
(A) Induction of cytopathic effect by virus infection. MDBK cells were infected with PIV5, rPIV5-V/P-S157A, or rPIV5-P-S308A. At 48 hours post infection, the infected cells were photographed using microscope (Nikon ECLIPSE TE300, Japan). (B) Propidium iodine (PI) staining. MDBK cells were infected and collected 2 days post infection as above. The cells were stained with PI and cellular DNA profiles were examined using flow cytometry. Percentages of sub G0-G1 population, which is considered apoptotic, are graphed. S157A, rPIV5-V/P-S157A; S308A, rPIV5-P-S308A. (C). Annexin V staining. The infected cells were stained with FITC-labeled annexin V and measured using a flow cytometer. Annexin V, which binds to phosphatidylserine (PS), is an indication of cells undergoing apoptosis when it is present on cell surface. Percentages of annexin V positive cells are graphed. (D) TUNEL assay. The infected cells were subjected to a TUNEL assay as described in Experimental Procedures. TUNEL positive cells were measured using a flow cytometer and percentages of TUNEL positive cells are graphed. (E). Growth of recombinant viruses. HeLa cells or Vero cells were infected with viruses at 0.01 MOI and media of infected cells were collected at indicated time points. Titers of viruses were determined using plaque assay. (F). Induction of IFN-β. HeLa cells were infected and levels of IFN-β were measured using ELISA at 2 days post infection. (G). Induction of IL-6. HeLa cells were infected and levels of IL-6 were measured using ELISA at 2 days post infection. Error bars are standard deviation of mean.
Figure 8.
A working model for regulation of PIV5 viral gene expression by PLK1.
For wild type PIV5, PLK1 binds to SSP motif (156–158) within the P protein, and then phosphorylates the P protein at S308, resulting in inhibition of PIV5 gene expression. The reduced viral gene expression supports efficient virus replication, but enables virus to avoid the induction of cell death and cytokine expression. For rPIV5-CPI+, SSP is mutated to SFP, preventing the binding of P to PLK1 and, thus, no phosphorylation of S308 by PLK1. As a result, viral gene expression is higher than that of PIV5, and cell death as well as cytokine production is induced, resulting in limiting replication and spread of virus. However, when PLK1 is over expressed, PLK1 can interact with S157F weakly through STP centered at T108 and reduces P-S157F mini-genome expression moderately. For rPIV5-P-S308A, while the SSP motif is intact and P-S308A still binds to PLK1; however, there is no phosphorylation of S308. This mutant virus, similar to rPIV5-CPI+, shows elevated viral gene expression and increased induction of cell death.