Investigation of the Role of Protein Kinase D in Human Rhinovirus Replication

Abstract

Picornavirus replication is known to cause extensive remodeling of Golgi and endoplasmic reticulum membranes, and a number of the host proteins involved in the viral replication complex have been identified, including oxysterol binding protein (OSBP) and phosphatidylinositol 4-kinase III beta (PI4KB). Since both OSBP and PI4KB are substrates for protein kinase D (PKD) and PKD is known to be involved in the control of Golgi membrane vesicular and lipid transport, we hypothesized that PKD played a role in viral replication. We present multiple lines of evidence in support of this hypothesis. First, infection of HeLa cells with human rhinovirus (HRV) induced the phosphorylation of PKD. Second, PKD inhibitors reduced HRV genome replication, protein expression, and titers in a concentration-dependent fashion and also blocked the replication of poliovirus (PV) and foot-and-mouth disease virus (FMDV) in a variety of cells. Third, HRV replication was significantly reduced in HeLa cells overexpressing wild-type and mutant forms of PKD1. Fourth, HRV genome replication was reduced in HAP1 cells in which the PKD1 gene was knocked out by clustered regularly interspaced short palindromic repeats (CRISPR)-Cas9. Although we have not identified the molecular mechanism through which PKD regulates viral replication, our data suggest that this is not due to enhanced interferon signaling or an inhibition of clathrin-mediated endocytosis, and PKD inhibitors do not need to be present during viral uptake. Our data show for the first time that targeting PKD with small molecules can inhibit the replication of HRV, PV, and FMDV, and therefore, PKD may represent a novel antiviral target for drug discovery.IMPORTANCE Picornaviruses remain an important family of human and animal pathogens for which we have a very limited arsenal of antiviral agents. HRV is the causative agent of the common cold, which in itself is a relatively trivial infection; however, in asthma and chronic obstructive pulmonary disease (COPD) patients, this virus is a major cause of exacerbations resulting in an increased use of medication, worsening symptoms, and, frequently, hospital admission. Thus, HRV represents a substantial health care and economic burden for which there are no approved therapies. We sought to identify a novel host target as a potential anti-HRV therapy. HRV infection induces the phosphorylation of PKD, and inhibitors of this kinase effectively block HRV replication at an early stage of the viral life cycle. Moreover, PKD inhibitors also block PV and FMDV replication. This is the first description that PKD may represent a target for antiviral drug discovery.

Keywords: Golgi membrane; antiviral; picornavirus; protein kinase D; rhinovirus; viral replication.

FIG 1

PKD phosphorylation during HRV infection. (A) The viral RNA level was measured by qRT-PCR and normalized to the 18S RNA level during the time course of infection. The graph shows the means (±SEM) of data from three independent experiments. (B) HeLa cells were infected with HRV16 at an MOI of 20 for 1 h, followed by a 7-h replication time course. Cell extracts were prepared every hour at 2 to 7 hpi (lanes 3 to 8) and analyzed by Western blotting. Uninfected cells are shown in lane 1, and uninfected cells treated with PDBu are shown in lane 2. Cells infected for 7 h with UV-inactivated virus are shown in lane 9. The blots from each experiment were immunostained with antibodies to phospho-PKD1/2 S744/S748 (pActivation loop), phospho-PKD1 S916 (pPKD1 S916), PKD1, HRV 2C, and lamin B1 (LB1). Arrowheads reveal the specific phosphorylated band. (C) A similar analysis was performed in an independent experiment with HeLa cells infected with HRV16 and immunostained with an antibody specific to phospho-PKD2 S876 (pPKD2 S876) and PKD2. (D) A similar experiment was performed in HBECs infected with HRV1B at an MOI of 20. Results of all experiments are representative of data from three independent repeats.

FIG 2

Effect of CRT0066101 on HRV 2C and viral RNA expression following infection. (A) HeLa cells were pretreated for 1 h with increasing concentrations of CRT0066101, followed by infection with HRV16 at an MOI of 20 for 1 h. Following a 6-h replication period, RNA was extracted from cell lysates, and the viral RNA level was quantified by qRT-PCR and normalized to the 18S RNA level. The results show the means (±SEM) from three independent experiments, each performed in duplicate. The “input” level (dotted line) reflects the viral RNA that was cell bound at the start of the replication cycle. (B) HeLa cells were pretreated for 1 h with increasing concentrations of CRT0066101, followed by infection with HRV16 at an MOI of 20 for 1 h. Cell extracts were prepared following a 6-h replication period and analyzed by Western blotting with antibodies to autophosphorylation residue S916 of PKD1, PKD1, HRV 2C, and LB1. Controls are as follows: uninfected cells (lane 1), PDBu-treated cells (lane 2), and vehicle control-treated cells (lane 3). Cells treated with CRT0066101 at concentrations from 0.1 to 20 μM are shown in lanes 4 to 13. (C) HBECs cells were pretreated for 1 h with increasing concentrations of CRT0066101, followed by infection with HRV1B at an MOI of 20 for 1 h. Cell extracts were prepared following a 6-h replication period and analyzed by Western blotting with antibodies against HRV 2C and LB1. Uninfected cells are shown in lane 1, and vehicle control-treated cells are shown in lane 2. (D) The effect of CRT0066051 and XX-50 on HRV 2C protein expression was analyzed by using the same protocol as the one described above for panel B. Results from each experiment shown in panels B to D are representative of data from three independent repeats. (E) In order to confirm the pharmacodynamic effect of the inhibitors on cells, HeLa cells were treated for 8 h with CRT0066101 and XX-050 at 5 μM and CRT0066051 at 10 μM, followed by analysis by confocal microscopy. The Golgi apparatus was revealed by staining with an anti-GM130 antibody, followed by staining with an anti-rabbit antibody coupled to Alexa Fluor 546 (bar = 10 μm).

FIG 3

Effect of PKD inhibitors on picornavirus replication. (A) HeLa cells were infected with HRV16 at an MOI of 20, and replication was allowed to proceed for 6 h in the presence of increasing concentrations of CRT0066101, CRT0066051, or XX-050. Viral replication was determined as the endpoint titer (TCID₅₀). (B) HeLa cells and HBECs were infected with HRV1B at MOIs of 1 and 5, respectively, and replication was allowed to proceed for 6 h in the presence of increasing concentrations of CRT0066101. Viral replication was determined as the endpoint titer (TCID₅₀). (C and D) HeLa cells were infected with PV (C) and BHK21 cells were infected with FMDV (D) at an MOI of 20, replication was allowed to proceed for 6.5 h in the presence of increasing concentrations of CRT0066101, and viral replication was determined as the endpoint titer (TCID₅₀). All the virus titer graphs show the means (±SEM) of data from three independent experiments. Differences between infected DMSO-treated cells and drug-treated cells were estimated by using one-way ANOVA with Dunnett's post hoc test. *, P < 0.05; ***, P < 0.001; ****, P < 0.0001. (E) HBECs and BHK21 and HeLa cells were incubated with increasing concentrations of CRT0066101 for 10, 8.5, and 8 h, respectively, and cell viability was determined as described above. (F) TEER was measured on HBECs grown in ALI cultures and treated with CRT0066101 at increasing concentrations for 48 and 72 h. Results in panels E and F show the means (±SEM) of data from three independent experiments.

FIG 4

Effect of CRT0066101 on viral entry. (A) HeLa cells were infected with HRV16 (MOI of 20) for 6 h, and CRT0066101 (5 μM) was added at the following different time points: 1 h before infection (−1), during the 1-h virus infection period (0), and every hour after the time of virus adsorption (+1, +2, +3, +4, and +5). Cells extracts were prepared at the end of the 6-h replication period and analyzed by Western blotting with anti-2C and anti-LB1 antibodies. Uninfected cells and DMSO-treated cells infected for 6 h were used as controls. Data from a representative experiment from three independent repeats are shown. (B) 2C Western blots were scanned as described in Materials and Methods and quantified by using ImageJ. The mean 2C/LB1 ratio (±SEM) is shown as a percentage of the value for the DMSO control from three independent experiments. (C) In parallel, virus was extracted from the cell lysates, and viral replication was quantified by endpoint titer determination (TCID₅₀). Results are the means (±SEM) of data from three independent experiments, each done in triplicate. Differences between DMSO-treated cells (−) and the rest of the conditions in both panels B and C were estimated by using one-way ANOVA with Dunnett's post hoc test. *, P < 0.05; **, P < 0.01; ***, P < 0.001; ****, P < 0.0001. (D) HeLa cells were grown on coverslips and pretreated with DMSO or CRT0066101 at 5 μM for 1 h at 37°C. Human transferrin conjugated to Alexa Fluor 488 was added at a 75-μg/ml final concentration, and cells were incubated at 37°C for 1 h in the presence of DMSO or CRT0066101. Cells were stained with an anti-GM130 antibody followed by an anti-mouse antibody coupled to Alexa Fluor 546 and analyzed by confocal microscopy. Transferrin quantification is shown as the mean fluorescence intensity (MFI) from multiple low-power images. For each experiment, images from five different low-power fields with approximately 250 cells per field were quantified by using ImageJ. Data shown are the means (±SEM) from three independent experiments. No statistically significant difference between DMSO- and CRT0066101-treated cells was found by performing a two-tailed t test. (E) High-power images of HeLa cells showing internalized transferrin (green) and Golgi membrane (GM130) (red) staining from a representative experiment (bar = 10 μm).

FIG 5

Effect of PKD1 mutant overexpression on HRV16 replication. (A) HeLa cells were transfected with PKD1 plasmids for 18 h, and cells were harvested and processed for Western blotting with specific antibodies against the phosphorylation site in the activation loop (pActivation loop) and autophosphorylation residue S916 of PKD1 and with antibodies to PKD1 and LB1 as loading controls. An anti-Myc antibody was used to check tagged-protein expression. Untransfected cells (−) and empty Myc plasmid-transfected cells (Myc) were used as controls. (B) HeLa cells were transfected as described above for panel A, fixed, immunostained by using an anti-Myc antibody followed by an anti-mouse antibody coupled to Alexa Fluor 488 and with an anti-GM130 antibody followed by an anti-rabbit antibody coupled to Alexa Fluor 546, and analyzed by confocal microscopy (bar = 10 μm). (C) HeLa cells were either untransfected (−) or transfected with each PKD1 construct, incubated for 18 h, and then infected with HRV16 at an MOI of 1 for 1 h, and replication was allowed to proceed for 6 h. Cells were harvested, and cell lysates were processed for endpoint titer determination by a TCID₅₀ assay. The graph shows the means (±SEM) of data from 4 experiments for Myc, the wt, and KD and from 3 experiments for ΔCT, S916A, and ΔPH, and each independent experiment was performed in duplicate. Differences between empty Myc plasmid-transfected cells and PKD1 construct-transfected cells were estimated by one-way ANOVA with Dunnett's post hoc test. *, P < 0.05; **, P < 0.01.

FIG 6

Effect of PKD knockout on HRV replication. (A) Parental wt HAP1 cells were infected with HRV1B at an MOI of 5 for 1 h, and replication was allowed to proceed for up to a 6-h time course. Cells were harvested every hour from 2 to 6 hpi, RNA was extracted from cell lysates, and the viral RNA level was quantified by qRT-PCR and normalized to the 18S RNA level. The 0-h time point corresponds to viral RNA that was cell bound at the start of the replication cycle. Results are shown as means (±SEM) of data from three independent experiments, each performed in duplicate. (B) Parental wt HAP1 cells were infected with HRV1B for 1 h at an MOI of 5, and replication was allowed to proceed for various times up to 6 hpi. Cell extracts were prepared and analyzed by Western blotting with anti-HRV 2C and anti-LB1 antibodies. (C) All clones were infected with HRV1B at an MOI of 5 for 1 h, followed by a replication period of up to 6 h. Viral RNA was extracted and quantified by qRT-PCR, the level was normalized to the total cellular 18S RNA level, and the fold increase in the viral RNA level from 3 to 6 hpi was calculated for each clone (Δ HRV RNA). Results are the means (±SEM) from four independent experiments, each performed in duplicate. Differences between parental wt HAP1 and knockout cells were estimated by one-way ANOVA with a two-tailed t test post hoc analysis. *, P < 0.05. The fold increase was statistically different between wt cells and clone 001 (P = 0.0488). (D) Cell extracts were prepared from the parental wt HAP1 clone (wt) (lane 1), PKD1 knockout clones (clones 1 and 12) (lanes 2 and 3), PKD2 knockout clones (clones 8 and 23) (lanes 4 and 5), and a double-knockout clone (DKO) (lane 6) and analyzed by Western blotting with anti-PKD1- and anti-PKD2-specific antibodies and an anti-LB1 loading control. The PKD1-specific band is shown by an arrowhead.

FIG 7

Effect of CRT0066101 on interferon signaling. (A) Effect of CRT0066101 on STAT1 phosphorylation at residue Y701 in HeLa cells infected with HRV16. Cells were either untreated (lane 1) or treated with 30 U/ml IFN-β for 15 min (lane 2), the DMSO vehicle (lane 3), or increasing concentrations of CRT0066101 for 1 h, followed by a 6-h replication period. Cell extracts were prepared and analyzed by Western blotting with antibodies to pSTAT1 Y701, STAT1, HRV 2C, and LB1. Data shown are representative of results from three independent experiments. (B and C) To determine the effect of CRT0066101 on ISG expression, RNA was extracted from HRV16-infected HeLa cells after 20 h of culture in the presence of the DMSO vehicle or 1 μM, 2 μM, or 3.5 μM CRT0066101. UV-inactivated virus was included as a control. Viral replication was confirmed by measuring the levels of HRV16 RNA (HRV) (B) and OAS mRNA (C) as a representative ISG. The results are the means (±SEM) of data from four independent experiments, each performed in duplicate. Differences between infected DMSO-treated cells and infected CRT0066101-treated cells were determined by one-way ANOVA with Dunnett's post hoc analysis. *, P < 0.05; ****, P < 0.0001. (D) HeLa cells were pretreated for 1 h with the DMSO vehicle or CRT0066101 at 5 μM, followed by stimulation with IFN-β (30 U/ml) for 4, 6, and 8 h in the presence of DMSO or CRT0066101. Cells were harvested, and RNA was extracted and processed for qRT-PCR. The OAS mRNA level was measured and normalized to the 18S RNA level. The results are the means (±SEM) of data from three independent experiments, each performed in duplicate. Differences between DMSO-treated and CRT0066101-treated cells at each time point were determined by two-way ANOVA with Sidak's post hoc test. *, P < 0.05; **, P < 0.01.

FIG 8

Effect of blocking of type I interferon receptor signaling on CRT0066101. In order to determine if a blockade of the type I interferon receptor (IFNAR2) influenced the ability of CRT0066101 to inhibit viral replication, HeLa cells were left untreated (lane 1) or pretreated with DMSO (lanes 2 to 4) or 5 μM CRT0066101 (lane 5 to 7) for 1 h. Cells were subsequently infected with HRV16 for 1 h (lanes 2 to 7), and replication was allowed to proceed for a further 4 h. During the viral infection and replication periods, cells were treated with a blocking antibody to IFNAR2 alone (lane 3) or with an isotype-matched control antibody alone (lane 4) or cotreated with CRT0066101 and a blocking antibody (lane 6) or CRT0066101 with an isotype control (lane 7). As additional controls, cells were treated with 30 U/ml of IFN-β for 4 h (lanes 8 to 10), with no antibody (lane 8), with anti-IFNAR2 (lane 9), or with an isotype control antibody (lane 10). Cell extracts were prepared and analyzed by Western blotting with antibodies to pSTAT1 Y701, STAT1, HRV 2C, and LB1. The pSTAT1 Y701 blot is shown at high and low exposures, and data shown are from a representative experiment from three independent repeats.