Research Article Increased Hsp27 after Androgen Ablation Facilitates Androgen- Independent Progression in Prostate Cancer via Signal Transducers and Activators of Transcription 3–Mediated Suppression of Apoptosis Palma Rocchi,1 Eliana Beraldi,1 Susan Ettinger,1 Ladan Fazli,1 Robert L. Vessella,3 Colleen Nelson,1 and Martin Gleave1,2 1The Prostate Centre, Vancouver General Hospital; 2Division of Urology, University of British Columbia, Vancouver, British Columbia, Canada; and 3Department of Urology, University of Washington, Seattle, Washington Abstract Unfortunately, androgen-independent (AI) progression is inevita- One strategy to improve therapies in prostate cancer involves ble, and the development of hormone-refractory disease and death targeting cytoprotective genes activated by androgen with- occurs within 2 to 3 years in most men. AI progression is a multi- drawal to delay the emergence of the androgen-independent factorial process by which cells acquire the ability to both survive (AI) phenotype. The objectives of this study were to define in the absence of androgens and proliferate using non-androgenic changes in Hsp27 levels after androgen ablation and to stimuli for mitogenesis, and involves variable combinations of evaluate the functional relevance of these changes in AI adaptive up-regulation of antiapoptotic genes, ligand-independent progression. Using a tissue microarray of 232 specimens of activation of the androgen receptor, and alternative signaling hormone-naı¨ve and post-hormone ablation–treated prostate pathways including Her2/neu, epidermal growth factor receptor, h cancer, we found that Hsp27 levels increase after androgen transforming growth factor- , and insulin-like growth factor-I ablation to become highly expressed (>4-fold, P V 0.01) in AI (1–3). Several antiapoptotic genes have been functionally linked to tumors. Hsp27 overexpression rendered LNCaP cells highly the development of hormone resistance, including Bcl-2, Bcl-xL, resistant to androgen withdrawal both in vitro and in vivo. and clusterin (4–6). The elucidation of the mechanisms that Tumor volume and serum prostate–specific antigen levels mediate AI progression, including key cytoprotective molecules, is increased 4.3- and 10-fold faster after castration when an important step towards identifying newly targeted therapeutic Hsp27 was overexpressed. Treatment of LNCaP tumor cells strategies. in vitro with Hsp27 antisense oligonucleotides (ASO) or short- Many components of survival and apoptotic pathways are interfering RNA suppressed Hsp27 levels in a dose-dependent regulated by molecular chaperones, such as heat shock proteins and sequence-specific manner increased the apoptotic sub– (Hsp). Using array analysis to compare gene expression profiles G0-G1 fraction and caspase-3 cleavage >2-fold, as well as before and after castration, we identified Hsp27 as one of the most decreased signal transducers and activators of transcription 3 highly expressed genes in AI prostate tumors (7). Hsp27 is a 27-kDa (Stat3) levels and its downstream genes, c-fos and sPLA-2. protein highly induced during the stress response to a wide variety The cytoprotection afforded by Hsp27 overexpression was of physiologic and environmental insults (8). Various roles have attenuated by Stat3 knockdown using specific Stat3 ASO. been proposed for Hsp27 to explain its cytoprotective effects during Coimmunoprecipitation and immunofluorescence confirmed cellular stress, including its role as a molecular chaperone, inhibitor that Hsp27 interacts with Stat3 and that Stat3 levels correlated of protein unfolding, direct interference with caspase activation, directly with Hsp27 levels. Hsp27 ASO treatment in athymic modulation of oxidative stress, and regulation of the cytoskeleton mice bearing LNCaP tumors significantly delayed LNCaP (9). Higher levels of Hsp27 are commonly detected in various tumor growth after castration, decreasing mean tumor cancers including breast (10), gastric (11), ovarian and endometrial volume and serum prostate–specific antigen levels by 57% (12, 13), osteosarcoma (14), glial tumors (15), and prostate (16). and 69%, respectively. These findings identify Hsp27 as a Increased Hsp27 levels in breast, endometrial, and gastric cancer modulator of Stat3-regulated apoptosis after androgen is associated with metastasis, poor prognosis, and resistance to ablation and as a potential therapeutic target in advanced chemotherapy or radiation (10, 17–19). prostate cancer. (Cancer Res 2005; 65(23): 11083-93) In prostate cancer, elevated Hsp27 expression has been linked to poor outcome, and is highly expressed in AI prostate cancer cells Introduction (7, 20–22). Bubendorf et al. (22) identified Hsp27 as an overex- Androgen withdrawal is the most effective form of systemic pressed gene in androgen-resistant CWR22 prostate xenografts. therapy for men with advanced prostate cancer, producing The objectives of this study were to define the changes in Hsp27 symptomatic and/or objective responses in >80% of patients. levels after androgen ablation and during AI progression in human prostate cancer, as well as to evaluate the functional relevance of these changes in AI prostate cancer. We chose the LNCaP tumor model, which closely mimics AI progression in humans by pro- Requests for reprints: Martin Gleave, Division of Urology, Vancouver Hospital, University of British Columbia, 27333 Heather Street, Vancouver, British Columbia, ducing prostate-specific antigen (PSA)–secreting and androgen- Canada V5Z 3J5. Phone: 604-875-5003; Fax: 604-875-5604; E-mail: gleave@ dependent tumors when injected into male immunodeficient mice, interchange.ubc.ca. I2005 American Association for Cancer Research. and develop non–androgen-regulated PSA gene expression after doi:10.1158/0008-5472.CAN-05-1840 castration as a surrogate end point of progression to AI (23). www.aacrjournals.org 11083 Cancer Res 2005; 65: (23). December 1, 2005 Downloaded from cancerres.aacrjournals.org on September 30, 2021. © 2005 American Association for Cancer Research. Cancer Research As part of our ongoing investigations to identify key pathways Research Laboratory at the Vancouver General Hospital. Specimens were mediating AI progression, we use forced overexpression for gain-of- chosen to represent various treatment durations of androgen withdrawal function analyses and antisense oligonucleotides (ASO) or short- therapy prior to radical prostatectomy ranging from no treatment interfering RNA (siRNA) for loss-of-function analyses to evaluate (n = 35), 0 to 3 months (n = 58), 3 to 6 months (n = 52), and 6 months (n = 57). AI tumors were also identified (n = 30). Most tissues were from the functional role of castration-induced changes in gene radical prostatectomy specimens, whereas AI tissues were obtained from expression. In this study, we characterized changes in Hsp27 after transurethral resections or metastatic lesions from men with hormone- androgen ablation in LNCaP tumors and employed Hsp27 gain- refractory disease. Metastatic lesions were obtained from warm autopsy and loss-of-function approaches to determine the effects of Hsp27 specimens of 13 men who succumbed to prostate cancer. All radical on LNCaP cell survival and tumor progression after androgen prostatectomy specimens in the array were Gleason grade V4 and clinical withdrawal in LNCaP cells in vitro and in vivo. stage T1 or T2. One metastatic prostate lesion was obtained per patient from seven bone lesions, three lymph node lesions, two liver lesions, and one adrenal lesion. Materials and Methods Tissue microarray analysis. A tissue microarray analysis was Tumor cell lines. The human prostatic cancer cell line LNCaP was constructed using a Beecher microarrayer from the above paraffin- kindly provided by Dr. Leland W.K. Chung (Emery University, Atlanta, GA) embedded specimens with matching H&E slides used to identify the and was maintained in RPMI 1640 (Life Technologies, Inc., Gaithersburg, tumor and accurately obtain the core sample. Each case is represented MD) supplemented with 5% FCS. with three cores in the tissue microarray analysis. Sections were Microarray experiments. Total RNA from each tumor sample, obtained deparaffinized and rehydrated through xylene and ethanol, then before and 35 days post-castration, were compared on the same chip to transferred to the 0.02% triton for permeabilization. Slides in citrate tumor samples from intact controls. A dye-swap for each pair was done to buffer (pH = 6) were heated in the steamer for 30 minutes. After cooling account for dye bias. Three tumors from each post-castration time point for 30 minutes and washing thrice for 5 minutes in PBS, the slides were were analyzed and results were confirmed with Northern analysis. incubated in 3% BSA for a further 30 minutes. The slides were Microarrays of 13,791 (70-mer) human oligos (Operon, Huntsville, AL) successively transferred to 3% H2O2 for 10 minutes and then were printed in duplicate in 3Â SSC onto amine-coated slides (Ezray, Apogent, incubated overnight with anti-Hsp27 antibodies from Nova Castra Fisher Scientific, Ottawa, Ontario, Canada) were supplied by the Array (Newcastle, United Kingdom) at the concentration of 1:400 in 1% BSA. Facility of the Prostate Centre at Vancouver General Hospital. Slides were The next day, the primary antibody was washed extensively with PBS and printed with a BioRobotics Microgrid II (Harvard Biosciences and Genomic the LSAB+ kit (Dako, Carpinteria, CA) was used as the detection system. Solutions, Ann Arbor,