Therapeutic potential of CDK inhibitor NU2058 in androgen-independent prostate cancer
AC Rigas, CN Robson and NJ Curtin
Northern Institute for Cancer Research, Newcastle University, Framlington Place, Newcastle upon Tyne, UK
Antiandrogens are initially effective in controlling pros- tate cancer (CaP), the second most common cancer in men, but resistance, associated with the loss of androgen- regulatedcell cycle control, is a major problem. At present there is no effective treatment for androgen-independent prostate cancer (AIPC). Cellular proliferation is driven by cyclin-dependent kinases (CDKs) with kinase inhibitors (for example, p27) applying the breaks. We present the first investigation of the therapeutic potential of CDK inhibitors, using the guanine-based CDK inhibitor NU2058 (CDK2 IC50 ¼ 17 lM, CDK1 IC50 ¼ 26 lM), in comparison with the antiandrogen bicalutamide (Casodex) in AIPC cells. A panel of AIPC cells was found to be resistant to Casodex-induced growth inhibition, but with the exception of PC3 (GI50 ¼ 38 lM) and CWR22Rv1 (GI50 ¼ 46 lM) showed similar sensitivity to NU2058 (GI50 ¼ 10–17 lM) compared to androgen-sensitive LNCaP cells (GI50 ¼ 15 lM). In LNCaP cells and their Casodex-resistant derivative, LNCaP-cdxR, growth in- hibition by NU2058 was accompanied by a concentration- dependent increase in p27 levels, reduced CDK2 activity and pRb phosphorylation, a decrease in early gene expression and G1 cell cycle phase arrest in both cell lines. In response to Casodex, there were similar observations in LNCaP cells (GI50 ¼ 673 lM Casodex) but not in LNCaP-cdxR cells (GI50 ¼ 2475 lM Casodex).
Keywords: androgen-independent prostate cancer; bicalut- amide; cyclin-dependent kinase inhibitor; LNCaP
Introduction
Prostate cancer (CaP) is the second most prevalent type of cancer diagnosed in men in the western world and the leading cause of cancer mortality in older men (Edwards et al., 2005). Because early-stage CaPs depend on androgens for growth and survival, androgen deprivation by surgical or chemical castration has been used as the main therapeutic intervention in the treatment of the disease (Feldman and Feldman, 2001). Despite initial responsive- ness to androgen ablation therapy, tumor cells invariably relapse to an androgen-independent (AI) and antiandrogen- resistant state that ultimately leads to mortality. At present there is no effective treatment for androgen-independent prostate cancer (AIPC), highlighting the pressing need to develop new alternative therapeutic strategies.
Cellular proliferation is driven by the periodic association of cyclin-dependent kinases (CDKs) and their partner cyclins and controlled by kinase inhibitors, for example p27, and dysregulation of this finely orchestrated process is a characteristic of cancer. The progression from a quiescent state G0/G1 into S phase is driven by cyclin D1/CDK4/6 and cyclin E/CDK2 phosphorylation of pRB, release of E2F1 and activation of immediate early genes (for example thymidylate synthase, TS) (Harbour et al., 1999). Reduced levels of p27, an inhibitor of CDK2, and increased levels of CDK2 and cyclin E are indicators of poor prognosis and associated with androgen independence (Guo et al., 1997; Macri and Loda, 1998; Gregory et al., 2001; Karan et al., 2002). Androgen withdrawal caused G1 arrest in androgen-dependent LNCaP cells and the reintroduction of androgens resulted in increased expression of cyclins A and B1, and CDK1 and CDK2 with resultant increase in CDK2 activity and pRb phosphorylation, but no change in cyclins D1 or E or CDK4 or -6 or p16, p21 or p27 (Taneja et al., 2001). Antiandrogens block the androgen–androgen receptor (AR) binding that promotes cell cycle progression and inhibits cell cycle arrest (Masiello et al., 2002), therefore one potential tool for the treatment of AIPC is to inhibit cell cycle progression by inhibiting CDK2 directly.
Encouraging preclinical data were obtained with the pan-CDK inhibitor flavopiridol, which induced growth inhibition in CaP cells (Drees et al., 1997; Li et al., 2000). However, in a phase II clinical trial in patients with advanced hormone refractory CaP, flavopiridol demon- strated poor response as a single agent. The more selective CDK1/2 inhibitor, olomoucine, was synergistic in combi- nation with antiandrogen, in CaP cell lines (Knillova et al., 2004), but these specific CDK inhibitors have not been evaluated as single agents for the treatment of AIPC.
O6-cyclohexylmethylguanine, NU2058, is a competi- tive inhibitor of CDK2 with respect to ATP (Ki value CDK2, 1273 mM) that binds in the ATP binding pocket in a different orientation from other purine-based inhibitors, including olomoucine and roscovotine (Arris et al., 2000). NU2058, which is a slightly less potent inhibitor of CDK1 (CDK2 IC50 ¼ 17 mM and CDK1 IC50 ¼ 26 mM), is the lead compound in a structure-based drug discovery program to develop more potent and selective CDK inhibitors (Hard- castle et al., 2004). Cellular studies using NU2058 demon- strated inhibition of both MCF-7 breast cancer cell proliferation and target protein phosphorylation consistent with CDK1 and CDK2 inhibition.
We report here an investigation of NU2058 in comparison with the antiandrogen bicalutamide (Caso- dex) in parental androgen-dependent LNCaP cells and a panel of AIPC cells, including the LNCaP sublines LNCaP-cdxR and LNCaP-AI. All AIPC cells were resistant to Casodex-induced growth inhibition but showed more broadly similar sensitivity to NU2058 compared with LNCaP cells. The effects of NU2058 and Casodex on cell cycle events and cell growth were determined in the matched LNCaP cells and LNCaP- cdxR cells, selected for growth in the presence of Casodex to reflect the development of AIPC in patients. Growth inhibition by NU2058 was accompanied by an arrest in the G1 phase of the cell cycle, a reduction in pRb phosphorylation, an increase in p27 levels and a decrease in early gene expression in both cell lines. In response to Casodex, there were similar observations in LNCaP cells but not in LNCaP-cdxR cells. These investigations provide the first evidence that CDK2 inhibition represents a suitable therapeutic strategy for the control of CaP that fails to respond to antiandrogens.
Results
NU2058-induced growth inhibition
The sensitivity to Casodex- and NU2058-induced growth inhibition was determined in CaP cell lines (Figure 1) and displayed a four-fold resistance to Casodex compared to parental cells. The effect of Casodex and NU2058 on cell growth was also investigated in an additional A1 LNCaP cell subline, LNCaP-AI. LNCaP-AI cells along with the AI and AR negative DU145 and PC3, and AI CWR22Rv1, which express a truncated AR lacking part of its ligand-binding domain (Tepper et al., 2002), were all 4 five-fold resistant to Casodex compared to parental cells (GI50430mM). In contrast, LNCaP-cdxR, LNCaP-AI and DU145 cells displayed similar sensitivity to NU2058, whereas PC3 cells and CWR22Rv1 cells were two- and four-fold less sensitive to NU2058 inhibition, respectively, possibly reflecting the different origin of these cells. Further investigations of the mechanisms responsible for growth inhibition by Casodex and NU2058 were con- ducted in hormone-sensitive and -resistant LNCaP cells using 6 and 24 mM Casodex (GI50 in LNCaP and LNCaP- cdxR, respectively) and 15 and 45 mM NU2058 (approx- imate GI50 and GI80 in both cell lines).
Figure 1 Growth inhibition by Casodex and NU2058 in CaP cell lines. Cell growth, estimated by staining with sulforhodamine B, after exposure for 96 h to increasing concentrations of (a) Casodex and (b) NU2058. Data normalized by comparison with drug-free controls (0.3% DMSO for Casodex; 0.1% DMSO for NU2058). Data are mean7s.d. values from three independent experiments. CaP, prostate cancer; DMSO, dimethylsulfoxide.
NU2058 induces cell cycle reduction in Sphase
We investigated the effects of Casodex and NU2058 on cell cycle distribution in both LNCaP and LNCaP-cdxR cells lines. Determination of the normal cell cycle distribution by flow cytometry of propidium iodide- stained cells revealed a large population in G1-phase (X60%) and a small population of cells in S phase (p20%) making changes in G1–S-phase transition difficult to detect. Nevertheless, a concentration-depen- dent increase in G1 and decrease in S-phase fraction was observed in both parental and LNCaP-cdxR cells treated with NU2058 for 24 or 48 h (Table 2a). Casodex caused only a modest decrease in S phase and increase in G1, which was more marked in the parental cells at 24 h (Table 2b). To confirm these data, we investigated the incorporation of the thymidine analog bromodeoxyur- idine (BrdUrd) into the DNA of proliferating cells to obtain an estimate for the fraction of cells in S phase. Cells were treated with Casodex or NU2058 and harvested at 24 and 48 h for analysis of BrdUrd incorporation by flow cytometry (Figure 2). Casodex inhibited S-phase entry to a greater extent in parental cells, such that after 48 h treatment with 24 mM Casodex resulted in a 36–40% reduction of BrdUrd incorpora- tion in LNCaP cells and 22–25% reduction in LNCaP- cdxR cells. NU2058 caused a much more profound suppression of S phase in both cell lines. After 48 h treatment with 45 mM NU2058, BrdUrd incorporation was inhibited by 84–86% and 95–99% in LNCaP and LNCaP-cdxR cells, respectively.
*Significantly different from LNCaP (Po0.05). aGI50 is the concen- tration of drug required to inhibit growth by 50% after 4 days of continuous exposure. bData are mean7s.d. of GI50 values from at least three independent experiments. cCWR22Rv1 cells have a doubling time of 3 days, so GI50 values were calculated after 9 days of continuous exposure.
NU2058 inhibits phosphorylation of pRb
We examined the effect of NU2058 on the phosphor- ylation status of Rb protein by western blotting using an antibody specific to pRb phosphorylated at T821 (Figure 3). T821 is phosphorylated in vitro preferentially by cyclin E/cyclin A–Cdk2 complexes, and phosphor- ylation at this site inhibits E2F binding (Knudsen and Wang, 1997). Treatment of LNCaP and LNCaP-cdxR cells with NU2058 for 24 and 48 h caused a reduction in Rb phosphorylation (Figure 3a). Densitometric analysis of the western blots revealed a 52% reduction in phosphorylation at T821 in LNCaP cells and 76% reduction in LNCaP-cdxR cells after exposure to 15 mM of NU2058 for 24 h. A similar loss was observed at 48 h. After treatment with the higher concentration of NU2058 (45 mM), phosphorylation at T821 was reduced by 82% in LNCaP cells and 78% in LNCaP-cdxR cells at 24 h. Similar effects were observed at 48 h. These results are consistent with NU2058 acting as a CDK2 inhibitor and the profound inhibition of BrdU incor- poration. Surprisingly, treatment with Casodex for 24 and 48 h increased phosphorylation of pRb at T821 in both cell lines (Figure 3b), suggesting that growth inhibition was not mediated by CDK2 inhibition.
In addition to the T821 site, NU2058 reduced phosphorylation of pRb at S807/811, a target for phosphorylation by CDK4/6 in both LNCaP and LNCaP-cdxR cells (Figure 3c). However, Casodex reduced pRb phosphorylation at S807/811 in a time- and concentration-dependent manner in LNCaP cells but not in LNCaP-cdxR cells (Figure 3d). These results suggest a possible role for the AR signaling pathway in CDK4/6 activity.
Neither NU2058 nor Casodex treatment of LNCaP and LNCaP-cdxR cells caused any significant difference in total pRb levels or a significant increase in hypopho- sphorylated Rb levels (data not shown).
NU2058 inhibits TS expression
Phosphorylation of pRb has been shown to be required for the induction and maintenance of the expression of ‘immediate early’ dNTP metabolic enzymes, including dihydrofolate reductase, ribonucleotide reductase sub- units and TS, in proliferating cells (Angus et al., 2002). We, therefore, determined the effect of NU2058 and Casodex on the expression levels of TS by immunoblot- ting cell extracts with an anti-TS antibody (Figure 4). NU2058 caused a marked concentration-dependent decrease in TS levels in both LNCaP and LNCaP-cdxR cells (Figure 4a) such that following exposure to 45 mM
Figure 3 NU2058 inhibits CDK2-specific T821 and CDK4/6-specific S807/811-pRb phosphorylation. Densitometric analysis of Western analysis for phospho T821- (a and b) and S807/811-pRb (c and d) in cell lysates of LNCaP and LNCaP-cdxR cells exposed to NU2058 (a and c) and Casodex (b and d) for 24 and 48 h expressed relative to vehicle controls. Data are mean7s.d. values from at least three independent experiments. Significant difference from control is given by *Pp0.05.
NU2058 TS expression was reduced by 94% in LNCaP cells and by 82% in LNCaP-cdxR cells. In comparison to NU2058, Casodex caused a less marked, but nevertheless concentration- and time-dependent decrease in TS levels (Figure 4b) consistent with its more modest effect on S phase (Figure 2). The suppression of TS expression by Casodex in LNCaP-cdxR cells, while similar to that in the parental cells at 24 h, was less pronounced at 48 h.
Figure 4 NU2058 inhibits TS expression. Western blot for TS protein and densitometric analysis of expression relative to vehicle controls in LNCaP and LNCaP-cdxR cells exposed to NU2058 (a) and Casodex (b) for 24 and 48 h. Data are mean7s.d. values from at least three independent experiments. Significant difference from control is given by *Pp0.05. TS, thymidylate synthase.
NU2058 increases p27 protein expression
Because NU2058 induced a G1 arrest in the cell lines (Table 2), we investigated the expression of CDK inhibitor proteins, p21 and p27, which are known to elicit G1 arrest when overexpressed. Interaction of p27 with CDK2 and cyclin E prevents subsequent entry into the S phase of the cell cycle. Treatment with 45 mM NU2058 increased p27 levels by threefold by 48 h in LNCaP cells (Figure 5a) and by five- and sixfold at 24 and 48 h, respectively, in LNCaP-cdxR cells (Figure 5c). Casodex had no significant effect on p27 protein expression in either cell line (Figures 5b and d), consistent with its lack of effect on pRb phosphoryla- tion at T821. Furthermore, neither NU2058 nor Casodex showed any significant effect on p21 levels in either cell line.
CDK2 kinase activity is decreased in NU2058-treated cells
We measured CDK2 kinase activity in LNCaP and LNCaP-cdxR cells following treatment with NU2058 and Casodex (Figure 6). CDK2 was immunoprecipi- tated from cells and radioactive kinase assays were performed using histone H1 as substrate. Both LNCaP and LNCaP-cdxR cells exposed to NU2058 displayed a decrease in CDK2 activity (Figure 6a). In marked contrast, Casodex (24 mM) only inhibited CDK2 activity in LNCaP cells but had no effect on LNCaP-cdxR cells (Figure 6b).
Discussion
At present there is no effective therapy for AIPC and there is therefore a pressing need for new therapeutic approaches. AIPC is associated with the loss of the CDK2 inhibitor, p27 (Macri and Loda, 1998), and the upregulation of CDK2 and cyclin E suggesting that inhibiting CDK2 activity directly may be a useful approach to restore cell cycle arrest and thus inhibit proliferation. We investigated the therapeutic potential of the guanine-based CDK inhibitor NU2058 in AIPC. Initial studies in a panel of AIPC cells confirmed their resistance to Casodex-induced growth inhibition and demonstrated a range of sensitivities to NU2058, some of which may have been attributable to their different genetic origin. To compare the effects of NU2058 and Casodex on androgen-dependent and -independent CaP cells of the same genetic background, we created antiandrogen-resistant LNCaP cells. This allowed us to examine the effects of both Casodex and NU2058 in matched cell lines, reflecting the initial antiandrogen- responsive state of CaP and the development of AIPC at relapse, without the added complication of additional genetic differences that confound the interpretation of data obtained using cell lines established from different patients. The LNCaP-cdxR and LNCaP-AI cell lines represent the clinical pathogenesis of AIPC and are more likely to be related to the parental cell line, thus the effects observed are more probably due to loss of androgen dependence. Experimental development of resistance to Casodex has been shown to have a variety of causes, for example mutation of the AR (Hara et al., 2003) or increased AR expression (Kokontis et al., 2005), no doubt reflecting the multiple potential causes of clinical resistance. Interestingly, AR levels in our LNCaP-cdxR cell line are similar to the parental LNCaPs and sequencing of the AR did not reveal any additional mutations (data not shown). These Casodex-resistant cell lines are a useful tool for the development of therapies for AIPC.
Figure 5 NU2058 increases p27 expression. Western blot for p21 and p27 proteins and densitometric analysis of the expression of p27/p21 relative to controls in LNCaP (a and b) and LNCaP-cdxR (c and d) cells exposed to NU2058 (a and c) and Casodex (b and d) for 24 and 48 h. Data are mean7s.d. values from at least three independent experiments. Significant difference from control is given by *Pp0.05, **P ¼ 0.06.
We observed that whereas LNCaP and LNCaP-cdxR cells display differential sensitivity to Casodex they are similarly sensitive to growth inhibition by NU2058, thus demonstrating sensitivity to CDK2-inhibiting drugs. We investigated the mechanism underlying the growth inhibitory effects of NU2058 further. NU2058 caused an accumulation of cells in G0/G1 and a profound decrease in the S-phase fraction. Consistent with these results, NU2058 also inhibited phosphorylation of pRb at CDK2 selective targets and inhibited the kinase activity of cellular CDK2 in both cells lines. Such G1 accumulation and reduced phosphorylation of pRB T821 has been observed with CDK2 RNA knockdown but not CDK1 knockdown (Cai et al., 2006), confirming that CDK2 is the primary target of NU2058 in LNCaP cells. The antiproliferative effect of NU2058 was also associated with the modulation of TS levels, which are also controlled by pRb and associated with S-phase entry. Interestingly, the NU2058-induced G1 arrest was accompanied by increased expression of p27 providing further insights into the potential mechanism of cell cycle arrest by NU2058. Levels of p27 protein are regulated by ubiquitin-dependent proteolysis (Kawama- ta et al., 1995; Ponce-Castaneda et al., 1995; Ferrando et al., 1996), which involves phosphorylation at Thr187 by CDK2 and recognition of Thr187-phospho-p27 by the SCFSkp2 ubiquitination system (Montagnoli et al., 1999). Thus, inhibition of CDK2 by NU2058 may increase p27 levels by inhibiting p27 degrada- tion, thereby contributing to further CDK2 inhibition with time.
Figure 6 NU2058 inhibits CDK2 activity. CDK2 activity was measured in LNCaP and LNCaP-cdxR cells exposed to (a) NU2058 or
(b) Casodex. Samples were assayed for H1 kinase activity in CDK2 immunoprecipitates and analysed by SDS–PAGE and autoradiography. Data are from a single experiment. SDS–PAGE, sodium dodecyl sulfate–polyacrylamide gel electrophoresis.
In contrast, Casodex caused a less marked accumula- tion in G1 and reduction in S phase in LNCaP cells, similar to that observed following androgen withdrawal in this cell line that was associated with a decrease in CDK4/6 but not CDK2 phosphorylation of pRb. This observation contrasts with previous reports of increased CDK2 and cyclin A but not CDK4/6 or cyclin D1 after androgen stimulation of androgen-starved LNCaP cells (Taneja et al., 2001) and suggests a possible role for the AR signaling pathway in CDK4/6 activity.
In addition to its established function in G1 and S phases, CDK2 has also been described to have a role in G2/M (Hu et al., 2001). NU2058 also inhibits the G2/M CDK, CDK1, but we did not observe G2/M accumulation indicating that at the concentrations used in the work reported here NU2058 was acting princi- pally by CDK2 inhibition. However, a role for CDK1- cyclin E in the regulation of the G1/S-phase transition has also been suggested (Aleem et al., 2005). Thus, the dual inhibition of CDK1/CDK2 could contribute to the observed G1 arrest. Alternatively, it is possible that direct or indirect effects on CDK4/6 activity, alone or in concert with CDK2 inhibition are causing arrest.
We report here the first demonstration of the potential utility of small molecule CDK inhibitors alone for the treatment of AIPC, coupled with an in-depth investigation of the underlying mechanism. Although suitability of CDK2 as a therapeutic target has been questioned (Tetsu and McCormick, 2003), our data show that NU2058 can inhibit the growth of CaP cells in a manner consistent with inhibition of CDK2. This is further supported by the observation that NU2058 and CDK2 knockdown with siRNA caused similar effects in MCF7 cells (Neil Johnson, manuscript submitted for publication) as we demonstrate here. These data provide preliminary evidence that inhibition of CDK2 is a valid therapeutic maneuver in advanced CaP.
Materials and methods
Materials
Casodex (ICI 176,334; 2(RS)-40-cyano-3-(fluorophenylsulp- honyl)-2-hydroxy-2-methyl-30-(trifluoromethyl)propionanilide;bicalutamide) was a kind gift from AstraZeneca Pharmaceu- ticals (Cheshire, UK). g[32P]ATP (specific activity ¼ 0.37 MBq/ ml) was from Amersham Biosciences (Buckinghamshire, UK). NU2058 was provided by Professor RJ Griffin (Newcastle University). All other chemicals were from Sigma (Poole, UK) unless stated otherwise.
Cell culture
Tissue culture reagents were purchased from Sigma. LNCaP, DU145, PC3 and CWR22Rv1 cells were obtained from the American Type Culture Collection (Manassas, VA, USA). LNCaP, LNCaP-cxdR, DU145, PC3 and CWR22Rv1 cells were maintained in RPMI 1640 media supplemented with 10% fetal calf serum (FCS), L-glutamine, 1% penicillin and 1% streptomycin at 371C in 5% CO2 atmosphere. LNCaP-cdxR cells were generated by exposing LNCaP cells to increasing concentrations of Casodex in normal growth medium over a 4-month period. Individual clones displaying resistance to 2 mM Casodex were isolated and expanded in 2 mM Casodex- containing medium except for the duration of experiments. LNCaP-AI were generated by continuous culturing of LNCaP cells in RPMI 1640 media supplemented with 10% FCS that had been stripped of steroids by treatment with dextran-coated charcoal (Halkidou et al., 2003).
Growth inhibition assays
Cells were seeded into 96-well plates at a density shown previously to give exponential growth and approximately three cell doublings throughout the exposure period, that is 1.5 × 103 cells/well for CWR22Rv1 cells and 1 × 103 cells/ well for all other cell lines in 100 ml tissue culture medium. After overnight attachment, the cells were exposed to varying concentrations of Casodex and NU2058 for 9 days for CWR22Rv1 cells and 4 days for all other cell lines, then fixed and stained with sulforhodamine B as described previously (Skehan et al., 1990). The concentration required to inhibit cell growth by 50% (GI50) was calculated from point- to-point graphs using GraphPad Prism (San Diego, CA, USA) software.
Cell cycle analysis
Parental and derivative LNCaP cells seeded in 90 mm dishes were treated with Casodex or NU2058 for 24 and 48 h. Exactly 2 h before harvesting, cells were treated with 10 mM BrdUrd. Cells were harvested, pelleted and resuspended in 300 ml of ice-cold phosphate-buffered saline (PBS), fixed by gentle addition of 700 ml of ice-cold 100% ethanol and kept overnight at 41C. For BrdUrd staining of proliferating cells, cells were resuspended in 2.5 ml of 2 M HCl at room temperature for 20 min to isolate nuclei. Nuclei were then washed twice with 10 ml of ice-cold PBS and resuspended in 200 ml PBS, 0.5% Tween-20 and 1% bovine serum albumin (BSA) containing 4 ml of mouse anti-BrdUrd antibody (1:50, DAKO, Cambridgeshire, UK) for 1 h at room temperature. Nuclei were washed twice with 5 ml of ice-cold PBS and resuspended in 200 ml PBS, 0.5% Tween-20 and 1% BSA containing 8 ml of rabbit anti-mouse FITC conjugate antibody (1:25; DAKO) for 30 min in the dark at room temperature. Nuclei were washed twice more with 5 ml of ice-cold PBS and resuspended in 400 ml PBS, 50 ml 5% Triton X-100/50 mg/ml propidium iodide and 50 ml of 1 mg/ml RNase A and incubated in the dark for 30 min at room temperature. The fluorescence of stained cells was measured using an FACScan flow cytometer and the Cell Quest program (BD Biosciences, Oxford, UK).
Western blotting
Parental and derivative LNCaP cells seeded in 90 mm dishes were treated with Casodex or NU2058 for 24 and 48 h. At the end of the experiment, cells were washed with PBS and lysed with Laemmli lysis buffer (20% glycerol, 4% SDS and 100 mM Tris, pH 6.8). For immunoblot analysis, protein lysates were denatured in 4 × SDS–PAGE (sodium dodecyl sulfate–polyacrylamide gel electrophoresis) sample buffer and subjected to SDS–PAGE on 12% Tris-glycine gels (BioRad, Herfordshire, UK). The separated proteins were transferred onto nitrocellulose membrane followed by blocking with 5% nonfat milk powder in Tris-buffered saline (20 mM Tris, 500 mM NaCl) for 1 h at room temperature or overnight at 41C. Membranes were probed for the proteins levels of CDK2 phosphorylated pRb T821 (Biosource, Camarillo, CA, USA), CDK4/6 phosphorylated pRb S807/811 (Cell Signalling Technology, Herts, UK), TS (Abcam, Cambridge, UK), CDK2, p27 (Santa Cruz Biotechnologies, Heidelberg, Germany), p21 (Oncogene Research Products, Nottingham, UK) and a-tubulin using specific primary antibodies, followed by peroxidase-conjugated appropriate secondary antibody and visualization by the ECL detection system (Amersham Biosciences).
CDK2 immunoprecipitation – kinase assay
Exponentially growing parental and derivative LNCaP cells (8–10 × 105) were exposed to Casodex or NU2058 for 1 h. Cells were washed in PBS, harvested and lysed in reaction lysis buffer (50 mM Tris, pH 7.5, 150 mM NaCl, 0.2 mM Na3VO4, 0.5% NP40, 1 mM PMSF, 1 mM dithiothreitol, 25 mg/ml leupeptin, 25 mg/ml aprotinin and 25 mg/ml pepstatin) on ice for 30 min. Lysates were centrifuged at 14 000 g for 3 min. Supernatants were precleared with 20 ml protein G sepharose (PGS) after three washes in reaction lysis buffer and rotated at 41C for 4 h. PGS and nonspecific-bound protein were removed by centrifugation at 14 000 g for 5 min. Immunoprecipitation was performed with 2 mg of polyclonal anti-CDK2 antibody (Santa Cruz Biotechnology) and samples were incubated overnight at 41C with rotation. After incubation, a further 20 ml of PGS were added to immunopre- cipitated samples and returned to 41C for 1 h with rotation. PGS with bound protein complexes was recovered by centrifugation at 14 000 g for 5 min and beads were washed once in wash buffer (PBS, 0.2% Triton X-100) and twice in PBS. Phosphorylation of histone H1 was measured by incubating the beads for 10 min at 301C with g[32P]ATP, 1 mM ATP, CDK buffer (50 mM Tris, pH 7.5, 5 mM MgCl2) and histone H1 (5 mg/ml) as substrate. The reaction was stopped by the addition of 50 ml Laemmli buffer. Samples were analysed by 12% SDS–PAGE and the gel was dried and subjected to autoradiography.
Statistical analysis
Statistically significant changes were determined by Student’s t-test (paired or unpaired as appropriate) using GraphPad Prism software. Pp0.05 was considered significant.
Acknowledgements
We thank AICR for financial support, AstraZeneca for Casodex and Professor RJ Griffin (Northern Institute for Cancer Research, School of Natural Sciences-Chemistry, Newcastle University) for NU2058.
References
Aleem E, Kiyokawa H, Kaldis P. (2005). Cdc2-cyclin E complexes regulate the G1/S phase transition. Nat Cell Biol 7: 831–836.
Angus SP, Wheeler LJ, Ranmal SA, Zhang X, Markey MP, Mathews CK et al. (2002). Retinoblastoma tumor suppres- sor targets dNTP metabolism to regulate DNA replication. J Biol Chem 277: 44376–44384.
Arris CE, Boyle FT, Calvert AH, Curtin NJ, Endicott JA, Garman EF et al. (2000). Identification of novel purine and pyrimidine cyclin-dependent kinase inhibitors with distinct molecular interactions and tumor cell growth inhibition profiles. J Med Chem 43: 2797–2804.
Cai D, Latham Jr VM, Zhang X, Shapiro GI. (2006). Combined depletion of cell cycle and transcriptional cyclin-dependent kinase activities induces apoptosis in cancer cells. Cancer Res 66: 9270–9280.
Drees M, Dengler WA, Roth T, Labonte H, Mayo J, Malspeis L et al. (1997). Flavopiridol (L86-8275): selective antitumor activity in vitro and activity in vivo for prostate carcinoma cells. Clin Cancer Res 3: 273–279.
Edwards BK, Brown ML, Wingo PA, Howe HL, Ward E, Ries LA et al. (2005). Annual report to the nation on the status of cancer, 1975–2002, featuring population-based trends in cancer treatment. J Natl Cancer Inst 97: 1407–1427. Feldman BJ, Feldman D. (2001). The development of androgen-independent prostate cancer. Nat Rev Cancer
1: 34–45.
Ferrando AA, Balbin M, Pendas AM, Vizoso F, Velasco G, Lopez-Otin C. (1996). Mutational analysis of the human cyclin-dependent kinase inhibitor p27kip1 in primary breast carcinomas. Hum Genet 97: 91–94.
Gregory CW, Johnson Jr RT, Presnell SC, Mohler JL, French FS. (2001). Androgen receptor regulation of G1 cyclin and cyclin-dependent kinase function in the CWR22 human prostate cancer xenograft. J Androl 22: 537–548.
Guo Y, Sklar GN, Borkowski A, Kyprianou N. (1997). Loss of the cyclin-dependent kinase inhibitor p27(Kip1) protein in human prostate cancer correlates with tumor grade. Clin Cancer Res 3: 2269–2274.
Halkidou K, Gnanapragasam VJ, Mehta PB, Logan IR, Brady ME, Cook S et al. (2003). Expression of Tip60, an androgen receptor coactivator, and its role in prostate cancer development. Oncogene 22: 2466–2477.
Hara T, Miyazaki J, Araki H, Yamaoka M, Kanzaki N, Kusaka M et al. (2003). Novel mutations of androgen receptor: a possible mechanism of bicalutamide withdrawal syndrome. Cancer Res 63: 149–153.
Harbour JW, Luo RX, Dei Santi A, Postigo AA, Dean DC. (1999). Cdk phosphorylation triggers sequential intramole- cular interactions that progressively block Rb functions as cells move through G1. Cell 98: 859–869.
Hardcastle IR, Arris CE, Bentley J, Boyle FT, Chen Y, Curtin NJ et al. (2004). N2-substituted O6-cyclohexylmethylgua- nine derivatives: potent inhibitors of cyclin-dependent kinases 1 and 2. J Med Chem 47: 3710–3722.
Hu B, Mitra J, van den Heuvel S, Enders GH. (2001). S and G2 phase roles for Cdk2 revealed by inducible expression of a dominant-negative mutant in human cells. Mol Cell Biol 21: 2755–2766.
Karan D, Kelly DL, Rizzino A, Lin MF, Batra SK. (2002). Expression profile of differentially-regulated genes during progression of androgen-independent growth in human prostate cancer cells. Carcinogenesis 23: 967–975.
Kawamata N, Morosetti R, Miller CW, Park D, Spirin KS, Nakamaki T et al. (1995). Molecular analysis of the cyclin- dependent kinase inhibitor gene p27/Kip1 in human malignancies. Cancer Res 55: 2266–2269.
Knillova J, Bouchal J, Hlobilkova A, Strnad M, Kolar Z. (2004). Synergic effects of the cyclin-dependent kinase (CDK) inhibitor olomoucine and androgen-antagonist bicalutamide on prostatic cancer cell lines. Neoplasma 51: 358–367.
Knudsen ES, Wang JY. (1997). Dual mechanisms for the inhibition of E2F binding to RB by cyclin-dependent kinase- mediated RB phosphorylation. Mol Cell Biol 17: 5771–5783. Kokontis JM, Hsu S, Chuu CP, Dang M, Fukuchi J, Hiipakka RA et al. (2005). Role of androgen receptor in the progression of human prostate tumor cells to androgen
independence and insensitivity. Prostate 65: 287–298.
Li Y, Chinni SR, Senderowicz AM, Sarkar FH. (2000). Induction of growth inhibition and apoptosis in prostate cancer cells by flavopiridol. Int J Oncol 17: 755–759.
Macri E, Loda M. (1998). Role of p27 in prostate carcinogen- esis. Cancer Metastasis Rev 17: 337–344.
Masiello D, Cheng S, Bubley GJ, Lu ML, Balk SP. (2002). Bicalutamide functions as an androgen receptor antagonist by assembly of a transcriptionally inactive receptor. J Biol Chem 277: 26321–26326.
Montagnoli A, Fiore F, Eytan E, Carrano AC, Draetta GF, Hershko A et al. (1999). Ubiquitination of p27 is regulated by Cdk-dependent phosphorylation and trimeric complex formation. Genes Dev 13: 1181–1189.
Ponce-Castaneda MV, Lee MH, Latres E, Polyak K, Lacombe L, Montgomery K et al. (1995). p27Kip1: chromosomal mapping to 12p12-12p13.1 and absence of mutations in human tumors. Cancer Res 55: 1211–1214.
Skehan P, Storeng R, Scudiero D, Monks A, McMahon J, Vistica D et al. (1990). New colorimetric cytotoxicity assay for anticancer-drug screening. J Natl Cancer Inst 82: 1107–1112.
Taneja SS, Ha S, Garabedian MJ. (2001). Androgen stimu- lated cellular proliferation in the human prostate cancer cell line LNCaP is associated with reduced retinoblastoma protein expression. J Cell Biochem 84: 188–199.
Tepper CG, Boucher DL, Ryan PE, Ma AH, Xia L, Lee LF et al. (2002). Characterization of a novel androgen receptor mutation in a relapsed CWR22 prostate cancer xenograft and cell line. SEL120-34A Cancer Res 62: 6606–6614.Tetsu O, McCormick F. (2003). Proliferation of cancer cells despite CDK2 inhibition. Cancer Cell 3: 233–245.