[CANCER RESEARCH 61, 5611–5618, July 15, 2001] Androgen Blocks Apoptosis of Hormone-dependent Prostate Cancer Cells 1 Kotohiko Kimura, Mark Markowski, Cai Bowen, and Edward P. Gelmann 2 Departments of Medicine and Human Oncology, Lombardi Cancer Center, Georgetown University, Washington, DC 20007-2197 ABSTRACT Androgen plays a critical role in the promotion and growth of prostate cancer. Androgen ablation has an expanding role in prostate cancer treatment and is now used to improve the efficacy of radiation therapy in addition to its role in treatment of metastatic disease. Here we show that androgen interferes with induction of prostate cancer cell death induced by a variety of stimuli. The effect of androgen on cell death occurs predominantly by interference with caspase activation and the inhibition of caspase cleavage in both the extrinsic and intrinsic cell death pathways. Androgen inhibited apoptosis induced by both tumor necrosis factor  (TNF-) and by Fas activation with or without concomitant irradiation. An antiapoptotic effect was seen in the presence of R1881, dihydrotestos- terone, and also 17-estradiol within 24 h of death induction. Sustained inhibition of apoptosis at 72 h was seen only with R1881, dihydrotestos- terone, cyproterone acetate, and hydroxyflutamide. Androgen treatment inhibited activation of caspases-8, -7, and -9 by TNF- / irradiation. Androgen attenuated BAX expression and blocked appearance of the proapoptotic p18 fragment of BAX. Androgen also abrogated BID cleav- age induced by TNF-  irradiation that contributed to a decrease in cytochrome c egress from mitochondria induced by TNF- / irradia- tion. There was also decreased mitochondrial depolarization in response to TNF-  irradiation. Production of the proapoptotic lipid metabolite ceramide was not affected by androgen, but androgen acted downstream from ceramide generation because R1881 blocked cell-death induction by bacterial sphingomyelinase. Inhibition of phosphoinositol-3-kinase activ- ity by wortmannin induced apoptosis that was also blocked by androgen, but there was no effect on protein levels or phosphorylation of AKT, indicating that R1881 did not interact with survival signaling of phosphoi- nositol-3-kinase. Lastly, androgen inhibited activation of nuclear factor-B during death induction, but the effect of androgen on cell death was not mediated by interference with the nuclear factor-B pathway. The data suggest that androgen induced blockade of caspase activation in both intrinsic and extrinsic cell death pathways and thereby was able to protect prostate cancer cells from apoptosis induced by diverse stimuli. INTRODUCTION Since the discovery by Huggins that androgen ablation benefited patients with advanced prostate cancer, we have come to understand that androgens play a critical role in the development, progression, and treatment of prostate cancer (1, 2). Currently androgen ablation is the only proven treatment that confers unequivocal, but temporary, benefit on patients with metastatic prostate cancer (3, 4). More re- cently use of androgen ablation as adjunctive treatment for localized prostate cancer has been undertaken. Studies during the last decade have shown that, when combined with radiation therapy, androgen ablation improved survival of patients with locally advanced prostate cancer (5). There is also increasing evidence that early androgen ablation, despite its long-term morbidity, improves cause-specific survival compared with delayed hormonal therapy in early metastatic prostate cancer (6 – 8). Androgens are also believed to be important prostate cancer promoters. Interference with the synthesis of DHT 3 , the androgen on which prostatic epithelial cells are most dependent, is under investigation for prostate cancer prevention in the nationwide randomized Prostate Cancer Prevention Trial (9, 10). We reported that androgens protected androgen-responsive LNCaP human prostate cancer cells from programmed cell death induced by etoposide (11). LNCaP cells undergo growth arrest but not apoptosis, in response to androgen deprivation, making these cells a good model for hormone-responsive prostate cancer. Our data suggested that cy- totoxic treatments for prostate cancer would have greater efficacy if delivered concurrently with androgen ablation. We have also shown that LNCaP cells are highly resistant to radiation-induced cell death but can be sensitized to irradiation by death ligands such as TNF- and agonistic Fas antibodies (12, 13). Androgen can inhibit cell death by acting as a survival factor for normal prostatic epithelial cells. Prostatic epithelium undergoes ap- optosis shortly after androgen deprivation. The rat prostate gland involutes within 3 weeks of castration (14, 15). The transplantable human PC-82 prostate cancer xenograft also undergoes apoptosis after castration (16). Human prostatic epithelium and the majority of pros- tate cancer cells respond quite rapidly to androgen ablation and undergo apoptosis (17, 18). The synthetic androgen R1881 has been shown to support LNCaP cell survival in the presence of a PI3K inhibitor that blocked AKT phosphorylation and thereby AKT activity (19). These experiments showed that androgen acted as a survival factor through pathway(s) independent of PI3K-PTEN-AKT. We now show that androgen blocks LNCaP cell death in large part by interfering with caspase activation in both intrinsic and extrinsic cell death pathways. These experiments provide further support for the notion that cytotoxic therapies may have greater effect on prostate cancer cells in the androgen-deprived milieu. MATERIALS AND METHODS Cell Culture and Apoptosis Assays. Cell culture conditions and death induction have been described previously (12, 13). Modified IMEM (Life Technology Inc., Gaithersburg, MD) containing 5% FCS was usually used for culturing LNCaP cells. For the experiments, the medium was replaced by the modified IMEM containing 5% charcoal-treated calf serum without phenol red 24 h before treatment with okadaic acid or death ligands. Synthetic androgen R1881 (10 -9 M, unless stated otherwise) was added to the culture when the culture medium was replaced by the medium with charcoal-treated calf serum. Other steroid hormones were used at the concentrations indicated. Cyproterone acetate was obtained from Sigma Chemical Co. Hydroxyflutamide was ob- tained from the Schering Corporation (20). Cell death was measured by in situ end labeling that we had previously shown correlates well with appearance of the morphological hallmarks of cell death (21). We routinely used 8 Gy for experiments in which irradiation was combined with TNF- (12). For some experiments with other agents, 20 Gy were used. Bacterial sphingomyelinase was used at a concentration of 300 mU/ml. Western Blotting. Western blotting was carried out as described previ- ously (12). Ceramide. Ceramide assays were done as described previously (12). Received 1/17/01; accepted 5/16/01. The costs of publication of this article were defrayed in part by the payment of page charges. This article must therefore be hereby marked advertisement in accordance with 18 U.S.C. Section 1734 solely to indicate this fact. 1 Supported by NIH Grant CA79912 (to E. P. G.). 2 To whom requests for reprints should be addressed, at Departments of Medicine and Human Oncology, Lombardi Cancer Center, Georgetown University, 3800 Reservoir Road, NW, Washington, DC 20007-2197. Phone: (202) 687-2207; Fax: (202) 784-1229; E-mail: Gelmanne@georgetown.edu. 3 The abbreviations used are: DHT, dihydrotestosterone; PARP, poly(ADP-ribose) polymerase; TNF, tumor necrosis factor; PI3K, phosphatidylinositol trisphosphate kinase; TNFR1, TNF receptor 1; IMEM, improved minimal essential medium; NF, nuclear factor; FACS, fluorescence-activated cell sorter/sorting. 5611 on May 28, 2016. © 2001 American Association for Cancer Research. cancerres.aacrjournals.org Downloaded from