Protein Kinase D2 and D3 Promote Prostate Cancer Cell Bone Metastasis by Positively Regulating Runx2 in a MEK/ERK1/2-Dependent Manner

Abstract

Advanced-stage prostate tumors metastasize to the bone, often causing death. The protein kinase D (PKD) family has been implicated in prostate cancer development; however, its role in prostate cancer metastasis remains elusive. This study examined the contribution of PKD, particularly PKD2 and PKD3 (PKD2/3), to the metastatic potential of prostate cancer cells and the effect of PKD inhibition on prostate cancer bone metastasis in vivo. Depletion of PKD2/3 by siRNAs or inhibition by the PKD inhibitor CRT0066101 in AR-positive and AR-negative castration-resistant prostate cancer cells potently inhibited colony formation and cell migration. Depletion or inhibition of PKD2/3 significantly blocked tumor cell invasion and suppressed the expression of genes related to bone metastasis in the highly invasive PC3-ML cells. The reduced invasive activity resulting from PKD2/3 depletion was in part mediated by the transcription factor Runx2, as its silencing decreased PKD2/3-mediated metastatic gene expression through the mitogen-activated protein kinase kinase/extracellular signal-regulated kinase 1/2 signaling axis. Furthermore, inhibition of PKD by CRT0066101 potently decreased the frequency of bone micrometastases in a mouse model of bone metastasis based on intracardiac injection of PC3-ML cells. These results indicate that PKD2/3 plays an important role in the bone metastasis of prostate cancer cells, and its inhibition may be beneficial for the treatment of advanced prostate cancer.

Figure 1

Depletion of protein kinase D2 and D3 (PKD2/3) or inhibition of PKD suppresses the proliferation, migration, and invasion of PC3-ML prostate cancer cells. A: Expression of PKD1, PKD2, and PKD3 was analyzed in RWPE-1, LNCaP, and PC3-Ml cells by Western blot analysis. B and C: Depletion or inhibition of PKD blocked clonogenic ability of PC3-ML cells. PC3-ML (1000 cells per well) transfected with PKD2 or PKD3 siRNAs or exposed to CRT101 (5 μmol/L) (B) or transfected with an empty vector (EV) or a catalytically active PKD2 (PKD2-CA) plasmid (C). Left panel: Crystal violet stain. Right panel: Colony numbers from five random fields. D: Depletion or inhibition of PKD inhibited PC3-ML prostate tumor cell migration. Cells transfected with PKD2 or PKD3 siRNAs for 48 hours or treated with CRT0066101 (5 μmol/L) for 24 hours were stained with crystal violet (left panel), and wound closure was measured by using ImageJ software (right panel). E: Depletion or inhibition of PKD blocked PC3-ML prostate tumor cell invasion. PC3-ML cells were transfected with PKD2 or PKD3 siRNAs, and treated with the PKD inhibitor CRT0066101 (5 μmol/L). Right panel: Percent invasion was calculated as the percentage of the cells invaded through Matrigel inserts versus the total cells migrated through the control inserts. F: Matrigel invasion assay was performed on cells transfected with a control (EV) or a PKD2-CA plasmid. Cells invaded through Matrigel were imaged (left panel) and quantified (right panel). Data shown in the bar graphs are the average of three independent experiments with error bars representing SEM (B–F). ∗∗P < 0.01; ∗∗∗P < 0.001; ∗∗∗∗P < 0.0001. Original magnification, ×10 (D–F).

Figure 2

Depletion of protein kinase D2 and D3 (PKD2/3) or inhibition of PKD reduces the viability, migration, and clonogenicity of AR-positive castration-resistant prostate cancer cells. A: Expression of PKD1, PKD2, and PKD3 in the castration-resistant prostate cancer cell lines C4-2B and 22Rv1 was analyzed by Western blot analysis. B: Depletion or inhibition of PKD2/3 reduced the viability of C4-2B and 22Rv1 cells transfected with PKD2 or PKD3 siRNAs or treated with CRT101 (5 μmol/L). C: Effects of PKD1 knockdown on the viability of C4-2B cells. Cell viability at 48 and 72 hours after PKD1 siRNA transfection was measured by using the CCK-8 assay. D: Depletion or inhibition of PKD2/3 suppressed C4-2B cell migration. Cells were transfected with PKD2 or PKD3 siRNAs or treated with CRT0066101 (5 μmol/L). Representative data from one of three independent experiments with seven to nine determinations at each time point are shown. E: Knockdown or inhibition of PKD2/3 blocked clonogenicity of C4-2B and 22Rv1 cells. PKD2 and PKD3 siRNA-transfected or CRT101-treated C4-2B and 22Rv1 cells were stained by crystal violet and quantified. Representative data from one of two independent experiments with triplicate determinations are shown. Data are expressed as means ± SEM from three independent experiments (B and C). ∗P < 0.05; ∗∗P < 0.01; ∗∗∗P < 0.001; ∗∗∗∗P < 0.0001.

Figure 3

Depletion or inhibition of protein kinase D (PKD) blocks the expression of Runx2-regulated bone metastatic genes in prostate cancer cells. A: PC3-ML cells were transfected with PKD2 or PKD3 siRNA or treated with CRT101 (5 μmol/L). Transcript levels relative to the control, glyceraldehyde-3-phosphate dehydrogenase (GAPDH), were determined. B: PC3-ML cells were transfected with empty vector (EV), Flag–catalytically active PKD2 mutant (PKD2-CA), and Flag–dominant negative PKD2 mutant (ie, PKD2-DN). Transcript levels of alkaline phosphatase (ALP), focal adhesion kinase (FAK), osteocalcin (OCN), and osteopontin (OPN) were determined by using real-time RT-qPCR. Top right: Confirmation of the overexpressed PKD2-CA and PKD2-DN in PC3-ML by Western blot analysis. C: Runx expression in prostate cancer cells. Whole cell extracts of RWPE-1, LNCaP, and PC3-ML cells were subjected to Western blot analysis for Runx1-3 expression. D: PC3-ML were transfected with a control siRNA (si-NT) or a Runx2 siRNA (si-Runx2), followed by real time RT-qPCR for the indicated genes. Left: Western blot analysis confirmed the knockdown of Runx2. E: Matrigel invasion assay was performed on PC3-ML cells transfected with si-NT and si-Runx2. Representative images are shown (top). Percent invasion is shown in the bar graph (bottom). Each real-time quantitative PCR experiment was repeated with duplicates at least two to three times, and data represent the means ± SEM from all independent experiments. ∗P < 0.05; ∗∗P < 0.01; ∗∗∗P < 0.001; ∗∗∗∗P < 0.0001. Original magnification, ×10 (E).

Figure 4

Inactivation or knockdown of protein kinase D2 (PKD2) causes down-regulation of Runx2 protein in prostate cancer cells. A and B: Transcript and protein levels of Runx1/2 and PKD2/3 were analyzed by using quantitative real-time RT-PCR (A) or Western blot analysis (B) in PC3-ML cells transfected with PKD2 or PKD3 siRNAs or treated with CRT101 (5 μmol/L). C: Immunofluorescence staining in tumor cells transfected with PKD2 and PKD3 siRNAs or treated with CRT101. The nuclei are counterstained with DAPI. Representative images from three independent experiments are shown. D:Top: PC3-ML cells were treated with cycloheximide with or without CRT101 for indicated times, and cell lysates were immunoblotted for Runx2. Glyceraldehyde-3-phosphate dehydrogenase (GAPDH) served as loading control. Bottom: Half-life of Runx2 protein was measured from the blots above. Data are expressed as means ± SEM of three independent experiments (A and D). ∗P < 0.05; ∗∗P < 0.01; ∗∗∗∗P < 0.0001. Original magnification, ×20 (C). DMSO, dimethyl sulfoxide.

Figure 5

Protein kinase D (PKD) regulates bone metastatic gene expression through mitogen-activated protein kinase kinase/extracellular signal–regulated kinase (ERK). A: Catalytically active PKD2 mutant (PKD2-CA) overexpression increased ERK activity. PC3-ML cells were transfected with an empty vector (EV) or Flag–PKD2-CA, followed by Western blot analysis for the indicated genes. B: U0126 blocked ERK1/2 phosphorylation in PC3-ML cells. C and D: PC3-ML cells were treated with the mitogen-activated protein kinase kinase inhibitor U0126. The transcript levels of Runx2, alkaline phosphatase (ALP), focal adhesion kinase (FAK), osteocalcin (OCN), and osteopontin (OPN) were determined by using quantitative real-time RT-PCR. Data are expressed as means ± SEM of three independent experiments (C and D). ∗P < 0.05; ∗∗P < 0.01; ∗∗∗P < 0.001; ∗∗∗∗P < 0.0001. DMSO, dimethyl sulfoxide; GAPDH, glyceraldehyde-3-phosphate dehydrogenase.

Figure 6

Protein kinase D (PKD) inhibitor CRT0066101 impairs metastatic dissemination of PC3-ML cells to the athymic mouse bone. A: Schematic representation for the treatment of mice. Male athymic nude mice (n = 12) were inoculated by intracardiac injection with the ZsGreen PC3-ML cell lines. Animals were treated with CRT101 (80 mg/kg, gavage daily) or vehicle (5% dextrose). B: Body weight was recorded twice a week (all P > 0.05, nonparametric U-test). C: Representative micrographs of cultured cells from each treatment group are shown. D: After 28 days, skeletal micrometastasis was identified by confocal microscopy. Representative images showing ZsGreen PC3-ML cell micrometastases (top). Percent decrease in skeletal micrometastasis on CRT101-treated animals after 28 days of injection was quantified (bottom). ∗∗∗P < 0.001. Original magnification, ×10 (C and D). BF, bright field.

Supplemental Figure S1

The effect of protein kinase D2 and D3 (PKD2/3) knockdown or inactivation on Runx1 mRNA expression in prostate cancer cells. PC3-ML cells were transfected with PKD2 or PKD3 siRNA for 48 hours or treated with CRT101 (5 μmol/L) for 24 hours. The transcript levels of Runx1 (A), PKD2 (B), and PKD3 (C) were determined by quantitative real-time RT-PCR with glyceraldehyde-3-phosphate dehydrogenase as the internal control. Data are expressed as means ± SEM from two independent experiments with triplicate determinations. ∗∗∗∗P < 0.0001.