Background
Prostate cancer is the most common malignancy in men in the United States [
1]. While the mortality of prostate cancer has been slightly reduced recently, it still contributes to 30,000 deaths annually with the majority from castration resistant prostate cancer (CRPC) [
2]. The androgen-signaling pathway, mediated mostly through the androgen receptor (AR), plays a critical role in the regulation of prostate cancer cell growth and survival [
3,
4]. Androgen deprivation is the standard therapy for advanced prostate cancer [
5]. However, within two to three years after initiating therapy, most patients relapse with a more aggressive form of prostate cancer, termed CRPC, for which there is no effective treatment.
The c-Met receptor tyrosine kinase (RTK) was originally discovered as an oncoprotein and has been implicated in the proliferation and progression of a wide variety of human malignancies, including prostate cancer [
6‐
9]. High c-Met expression is observed in late stages and metastases of prostate cancer [
8,
10]. Additionally, an inverse correlation between the expression of AR and c-Met has been observed in prostate epithelium and prostate cancer cell lines [
8,
10]. Recently, we demonstrated that AR suppressed c-Met transcription and that increased c-Met expression was induced by removal of androgens in prostate cancer cells [
11]. These data elucidated a biological role for AR in c-Met transcription that may directly contribute to the pathogenesis of CRPC. While the current androgen deprivation therapy represses the expression of growth promoting genes that are activated by the AR, it may also attenuate the suppressive role of AR on c-Met expression and contribute to tumor progression.
In this study, we directly assessed the inhibition of c-Met signalling pathway in prostate cancer cell growth and tumor formation and progression using two c-Met inhibitors, PHA-665752 and PF-2341066. These inhibitors have shown potency and specificity for inhibiting c-Met activation in a variety of human tumor cells [
12‐
16]. We first tested the anti-proliferative effect of these inhibitors on a variety of human prostate cancer cell lines. We then examined the effect of PF-2341066 in inhibiting the proliferation of LNCaP tumors in renal subcapsular mouse models. Finally, we assessed the effect of co-inhibition of c-Met and androgen signalling pathways in prostate cancer progression using PF-2341066 and castration in an orthotopic xenograft model. Through these
in vitro and
in vivo experimental approaches, we explored a future therapeutic strategy of combining c-Met inhibitors with standard androgen ablation therapy for advanced prostate cancer.
Methods
Cell culture
Human prostate cancer cell line LAPC4 was maintained in RPMI phenol-red free (Invitrogen, Carisbad, CA) supplemented with 5% fetal bovine serum (FBS, HyClone, Denver, CO) [
17‐
19]. Human prostate cancer cell line CWR22Rv1 was maintained in RPMI (Invitrogen) supplemented with 5% FBS and obtained from the American Tissue Culture Collection (ATCC) (CRL-2505). Human prostate cancer cell lines LNCaP, LNCaP C4-2, and LNCaP C4-2B were maintained in T medium (Invitrogen) supplemented with 5% FBS [
18]. Human prostate cancer cell lines DU-145 and PC-3 were maintained in DMEM supplemented with 5% FBS and obtained from ATCC (HTB-81, CRL-1435).
Cell proliferation and colony formation assays
Five hundred cells per well were seeded in triplicate in 96-well plates in suitable media mentioned above. Appropriate controls or designated concentrations of PHA-665752 or PF-2341066 dissolved in DMSO were added to each well after 4 hours, and cells were then incubated for up to 8 days. Cell proliferation assays were carried out using the MTS (3-(4,5-dimethylthiazol-2-yl)-5-(3-carboxymethoxyphenyl)-2-(4-sulfophenyl)-2H-tetrazolium) tetrazolium kit (Promega, Madison, WI) as suggested by the manufacturer. For colony formation assay, 50 cells per well were plated in quadruplicate in 6-well plates for 24 hr, and then 5 μM of PHA-665752 or PF-2341066 dissolved in DMSO was added to each well. Cells were then grown for 10 days then fixed and stained with crystal violet in methanol. Each visible colony (> 50 cells) was manually counted.
C-Met inhibitors
PHA-665752 [(2R)-1-[[5-[(Z)-[5-[[(2,6-Dichlorophenyl)methyl]sulfony l]-1,2-dihydro-2-oxo-3H-indol-3-ylidene]methyl]-2,4-dim ethyl-1H-pyrrol-3-yl]carbonyl]-2-(1-pyrrolidinylmethyl) pyrrolidine] was acquired from Tocris Bioscience, Missouri 12. PF-2341066 [(R)-3-[1-(2,6-dichloro-3-fluoro-phenyl)-ethoxy]-5-(1-piperidin-4-yl-1
H-pyrazol-4-yl)-pyridin-2-ylamine] was provided by Dr. James Christensen in Pfizer, La Jolla Laboratories [
14].
Western blotting
To prepare the whole-cell lysate, cells were washed with cold PBS and resuspended in radioimmunoprecipitation assay (RIPA) buffer, 1% NP40, 0.1% SDS, 50 mM NaF, 0.2 mM Na3VO4, 0.5 mM DTT, 150 mM NaCl, 2 mM EDTA, 10 mM sodium phosphate buffer (pH 7.2). Protein samples boiled in SDS-sample buffer were resolved on a 6% SDS-PAGE, transferred onto a nitrocellulose membrane, and then blotted with a polyclonal anti-c-Met antibody (Santa Cruz Biotechnology, Santa Cruz, CA), monoclonal anti-phospho-c-Met antibody (Cell Signaling Technology, Danvers, MA), or monoclonal anti-tubulin antibody (Thermo Scientific, Fremont, CA). Blots were detected using the enhanced chemiluminescence kit (Amersham, Arlington Heights, IL).
Animal studies
All animal studies performed in this study conformed to the Helsinki Declaration and to local legislation. The research was approved by the ethics committee of the Administrative Panel on Laboratory Animal Care at Stanford University. Male nonobese diabetic-severe combined immunodeficient (NOD-SCID) mice (6 weeks old) were obtained from Jackson Laboratory (Bar Harbor, ME), and maintained and cared in accordance with the Guide for the Care and Use of Laboratory Animals.
LNCaP C4-2 cells were suspended in unpolymerized rat tail collagen to make collagen grafts (5 × 10
5 cells per graft). Two grafts were implanted under the renal capsule of each male NOD-SCID mouse that was then subcutaneously implanted with a testosterone pellet (10 mg/mouse) [
20,
21]. At 1 week post-grafting, one group of randomly chosen mice was started on daily treatment with PF-2341066 at 50 mg/kg given in water by oral gavage for 3 weeks. At 4 weeks post-grafting, these animals were sacrificed with the control group on water only. The grafts from the mice on water (n = 8) or PF-2341066 (n = 10) were harvested, weighed, and fixed for pathological analyses. Significant differences in graft weight between the treated versus the control groups were determined using Student's
t test.
For the orthotopic xenograft mouse model, LNCaP cells (1 × 10
6) were suspended in 50 μl of unpolymerized rat tail collagen and injected into the dorsal prostate of male NOD-SCID mice, supplemented with testosterone pellets (10 mg/mouse) [
22‐
25]. At five weeks post-implantation, a subset of mice (n = 4) was sacrificed to collect prostates for histopathologic analyses as a pre-castration group. The remaining mice were castrated and the testosterone pellets were removed. At one week post-castration, animals were either given daily treatment with PF-2341066 at 50 mg/kg in water (n = 7) or water only as controls (n = 6) by oral gavage for 3 weeks. Then, the mice were sacrificed and the prostate and pelvic lymph nodes were harvested and fixed for pathological analyses.
Immunohistochemistry
Tissues were fixed in 10% neutral buffered formalin, processed, cut at 4 μm intervals, dewaxed, and hydrated as described previously [
11]. Endogenous peroxidase activity was blocked with 0.5% hydrogen peroxide and then incubated with the blocking solution (ImmunoVision Technology, Springdale, AR). The adjacent sections were incubated with primary antibody or nonimmune mouse or rabbit IgG overnight at 4°C (Zymed Corp, So. San Francisco, CA). The mouse anti-human AR (Santa Cruz Biotechnology), rabbit anti-human PSA (DAKO, Carpinteria, CA), or rabbit anti-human Ki67 (Leica Microsystems, Bannockburn, IL) was used at dilutions of 1:300, 1:3000, or 1:3000, respectively, in PBS with 1% goat serum. Following incubation, sections were washed, incubated with a biotinylated goat anti-rabbit or anti-mouse antibody (Vector Laboratories, Burlingame, CA) at a dilution of 1:500, and then incubated with avidin-biotin complex (Vector Laboratories) at a dilution of 1:500. All sections used for immunohistochemistry were lightly counterstained with 5% (w/v) Harris hematoxylin. Stained sections were analyzed using an Olympus BX51 light microscope and DP70 camera with a 20× objective lens. Quantitative Ki67 index was determined by counting the number of Ki-67-positive cells relative to the number of total cells in more than five fields of two representative sections from each sample via light microscopy.
Discussion
Since the pioneering work in 1941 by Charles Huggins and Clarence Hodges that castration could be effective for metastatic prostate cancer, androgen deprivation therapy has been widely used to treat advanced prostate cancer [
29]. Although different medications have been developed and applied to patients for the purpose of reducing the level of androgens or competitively repressing AR function, the fundamental premise of this therapy has remained almost unchanged. Within 2-3 years of the therapies, most patients develop more aggressive disease, CRPC. Therefore, it is urgent to develop more effective therapeutic strategies for advanced prostate cancer to prevent the development of CRPC.
Recent work including data from our lab has created a paradigm shift by showing a role of AR in transcriptional repression [
11,
30]. Our previous data demonstrated that AR suppressed c-Met transcription in LNCaP prostate cancer cells and castration of SCID mice bearing LNCaP xenografts induced c-Met expression in the tumors [
11]. Knockdown of c-Met using specific c-Met siRNA inhibited the induction of c-Met expression by androgen depletion and repressed prostate cancer cell growth [
31]. Decreasing
c-met expression using ribozyme technology resulted in reduced prostate cancer cell proliferation and orthotopic tumor formation using PC-3 cells [
26]. The c-Met receptor tyrosine kinase had been shown to be involved in tumor proliferation and progression and overexpression of c-Met had been associated with advanced prostate cancer [
6‐
10]. These data suggested that the therapeutic strategy of inhibiting the activation of both HGF/c-Met and AR signaling pathways should be considered in the treatment of advanced prostate cancer.
In this study, we used two recently developed small molecule inhibitors for c-Met receptor, PHA-665752 and PF-2341066, to test the inhibition of c-Met activation in the growth and progression of prostate cancer cells. First, we evaluated the efficacy of two small molecule inhibitors for c-Met, PHA-665752 and PF-2341066, in a variety of prostate cancer cell lines. We observed significant inhibitory effects of these inhibitors on all the cell lines used here. In addition, we also confirmed that these small molecule inhibitors block phosphorylation of c-Met to inhibit its activation and downstream signalling [
12,
14]. In this study, we compared the repressive effect of these small molecule inhibitors in androgen sensitive and insensitive cell lines. Interestingly, we observed a more pronounced effect of these inhibitors on androgen-insensitive cells than androgen-sensitive cells. This correlated with previous publications showing that androgen-insensitive cells possessed higher cellular levels of c-Met [
7,
8,
10,
32,
33]. As we have shown previously, although these androgen-insensitive cells remained AR positive, the repressive effect of AR on c-Met expression in these cells appeared diminished or reduced [
11]. Multiple lines of evidence have suggested that action and regulation mediated by AR may be different in androgen-sensitive and insensitive status [
34,
35]. Therefore, further investigation of the regulatory mechanisms of AR in these different stages would be extremely important to understand the molecular basis of the switch between androgen-sensitive and insensitive during prostate cancer progression.
To examine whether co-targeting androgen and c-Met signaling pathways could effectively impede the progression of prostate cancer, we initially demonstrated the feasibility and effectiveness of the c-Met inhibitor on repression of the proliferation of prostate cancer cells in a xenograft mouse model using an orally bioavailable c-Met small molecule inhibitor, PF-2341066 [
11,
20,
21]. We observed a significant 48% reduction in gross tumor weight and 45% reduction in proliferation when comparing PF-2341066 treated mice to control mice. Using an AR positive, androgen insensitive prostate cancer cell line, LNCaP C4-2, we showed for the first time that PF-2341066 inhibited the growth of AR positive prostate tumor xenografts
in vivo.
Next, we tested targeted therapy to c-Met with castration in an orthotopic mouse model to determine the effectiveness of combination therapy in inhibiting prostate cancer progression. Intraprostatic injection of LNCaP cells into SCID mice had been shown to lead to tumor formation within one month in this mouse model [
36‐
38]. Subsequent castration of these mice initially delayed tumor growth but LNCaP cells eventually regained growth ability and developed androgen-insensitive tumors [
22‐
25]. We chose a dosing schedule of three weeks based upon a prior study showing
in vivo effectiveness of PF-2341066 on prostate cancer cells [
14]. Intriguingly, we observed a significant 85% reduction in proliferation when comparing PF-2341066 treated castrated mice versus control castrated mice. This elucidated that targeting of c-Met signalling as an adjunct to standard androgen deprivation therapy may reduce the development of castration resistant tumors. In addition to PHA-665752 and PF-2341066 tested in our study, other small molecule c-Met receptor tyrosine kinase inhibitors have also been developed and showed an inhibitory effect in non-prostate and prostate cancer cell lines [
39,
40]. In particular, Dai et al. reported in PC-3 cells that their c-Met inhibitor had anti-motogenic and anti-proliferative effects depending upon the drug concentration. This was similar to the previously published data on PHA-665752 and PF-2341066 [
12,
14] Taken together, these most recent findings and the results from this study provided important information on the potential of selective molecular therapy for advanced prostate cancers.
Conclusions
Effective treatment for advanced prostate cancer has an urgent need in prostate cancer research. In this study, we demonstrated that two newly developed c-Met inhibitors, PHA-665752 and PF-2341066, repress the growth of a variety of prostate cancer cell lines. We observed an inhibitory effect of PF-2341066 on the proliferation of AR positive tumors in renal subcapsular mouse models. Moreover, using an orthotopic xenograft model, we showed that co-inhibition of c-Met and androgen signalling pathways using PF-2341066 and castration significantly reduced the proliferation of prostate cancer cells during progression after castration. Data from this study shed fresh light on possible future treatment by co-targeting c-Met and androgen signalling pathways in advanced prostate cancers.
Competing interests
JGC is an employee of Pfizer. The other authors declare that they have no competing interests.
Authors' contributions
ZS and JGC conceived of the study. WHT, CZ, and CC collected the data. WHT and ZS contributed to data analysis and wrote the manuscript. All authors read and approved the final manuscript.