Fibroblast growth factor receptors (FGFRs) mediate the tumourigenic effects of FGFs in prostate cancer. These receptors are therefore potential therapeutic targets in the development of inhibitors to this pathway. To identify the most relevant targets, we simultaneously investigated FGFR1-4 expression using a prostate cancer tissue microarray (TMA) and in laser capture microdissected (LCM) prostate epithelial cells. In malignant prostates (n = 138) we observed significant FGFR1 and FGFR4 protein over-expression in comparison with benign prostates (n = 58; p < 0.0001). FGFR1 was expressed at high levels in the majority of tumours (69% of grade 3 or less, 74% of grade 4 and 70% of grade 5), while FGFR4 was strongly expressed in 83% of grade 5 cancers but in only 25% of grade 1-3 cancers (p < 0.0001). At the transcript level we observed a similar pattern, with FGFR1 and FGFR4 mRNA over-expressed in malignant epithelial cells compared to benign cells (p < 0.0005 and p < 0.05, respectively). While total FGFR2 was increased in some cancers, there was no association between expression and tumour grade or stage. Transcript analysis, however, revealed a switch in the predominant isoform expressed from FGFR2IIIb to FGFR2IIIc among malignant epithelial cells. In contrast, protein and transcript expression of FGFR3 was very similar between benign and cancer biopsies. The functional effect of targeting FGFR4 in prostate cancer cells has not previously been investigated. In in vitro experiments, suppression of FGFR4 by RNA interference effectively blocked prostate cancer cell proliferation (p < 0.0001) and invasion (p < 0.001) in response to exogenous stimulation. This effect was evident regardless of whether the cells expressed the FGFR4 Arg388 or Gly388 allele. In parallel experiments, FGFR3 suppression had no discernible effect on cancer cell behaviour. These results suggest evidence of selective over-expression of FGFR1 and FGFR4 in clinical prostate cancer and support the notion of targeted inhibition of these receptors to disrupt FGF signalling.

Transforming growth factor-1 (TGF 1) is inhibitory to most epithelia, but its role in the control of proliferation of prostatic epithelium is unclear. In some cells, TGF 1 inhibition is achieved by up-regulation of cyclindependent kinase (cdk) inhibitors including p15, p21 and p27.Our aims were to determine whether the effects of TGF 1 on human prostatic epithelial cell cycle kinetics were mediated by alterations in the levels of the cdk inhibitors p15, p16, p21 and p27 and hypophosphorylated retinoblastoma protein (Rb).Human prostatic epithelial cells in primary culture were grown in the presence of TGF 1 (0-10 ng/ml) for up to 4 days and proliferation assessed using a [ 3 H]thymidine uptake assay. Levels of p15, p16, p21 and p27 were measured at both mRNA and protein level by means of a reverse transcriptase PCR-based assay and Western analysis. Rb and cdk2 levels were measured.Exogenous TGF 1 (0-5 ng/ml) inhibited proliferation. This was associated with blocking of the cell cycle at G1, and up to 4-fold increases in p15, p21 and p27 mRNA levels, but no change was observed in p16 mRNA levels; these changes were not blocked by cycloheximide. Increased levels of p15, p21 and p27 protein were also accompanied by increased levels of hypo-phosphorylated Rb and decreased cdk2 kinase activity.TGF 1 has mainly inhibitory effects on benign human prostatic epithelium, which are caused by up-regulation of cdk inhibitors, hypo-phosphorylation of Rb and delaying of the cell cycle in G1.

No abstract

Gonadotropin-releasing hormone analogue (GnRHa) therapy is an established method of androgen withdrawal in the treatment of prostate cancer. The present study investigated if the expression of prostate GnRH receptors (GnRHRs) might influence the response to GnRHa. GnRHR protein expression was first studied in a panel of prostate cancer cell lines. In androgen-dependent cells, GnRHR expression was unchanged following acute or chronic androgen withdrawal. In these cells, GnRHa significantly inhibited androgen-induced cell proliferation (p = 0.01). In contrast, GnRHa was unable to further suppress basal levels of cell proliferation induced by androgen withdrawal. In androgen-independent prostate cancer cells, variable levels of GnRHR expression were observed. In these cells, GnRHa treatment blocked cell proliferation (p = 0.001) and invasion (up to 70%) induced by fibroblast growth factor stimulation. Crucially, this effect was only evident in cells that expressed high levels of the GnRHR. GnRHa treatment also significantly inhibited the ability of these cells to recover from a cytotoxic insult (50% inhibition). The clinical significance of prostate GnRHR was tested by immunohistochemistry in a preliminary cohort of patients treated with GnRHa or surgical castration. There was no association between GnRHR expression and pathological grade, clinical stage, time to PSA nadir (p = 0.82) (n = 35) or progression to hormone refractory disease (p = 0.22) (n = 21), irrespective of the treatment method. GnRHa therapy in the presence of high GnRHR expression however, was found to be associated with longer disease-specific survival (mean survival 85 months, p = 0.002). In contrast, high GnRHR expression was not associated with survival among surgically castrated patients (mean survival 50 months, p = 0.7). Taken together, these data support the notion of a functional interaction between GnRHa and the GnRHR, which results in an anti-tumourigenic effect on prostate cancer cells. Findings from this report have direct implications for the use of GnRHR as a novel therapeutic target in hormone refractory prostate cancer.