The G /T-15 Transgenic Mouse Model of Androgen-independent Prostate Cancer: Target Cells of Carcinogenesis and the Effect of the Vitamin D Analogue

Transgenic mouse models of prostate cancer provide unique opportunities to understand the molecular events in prostate carcinogenesis and for the preclinical testing of new therapies. We studied the G T-15 transgenic mouse line, which contains the human fetal globin promoter linked to SV40 T antigen (Tag) and which develops androgen-independent prostate cancer. Using the immunohistochemistry of normal mouse prostates before tumor formation, we showed that the target cells of carcinogenesis in G T-15 mice are located in the basal epithelial layer. We tested the efficacy of the 1,25(OH)2D3 analogue, EB 1089, to chemoprevent prostate cancer in these transgenic mice. Compared with treatment with placebo, treatment with EB 1089 at three different time points before the onset of prostate tumors in mice did not prevent or delay tumor onset. However, EB 1089 significantly inhibited prostate tumor growth. At the highest dose, EB1089 inhibited prostate tumor growth by 60% (P 0.0003) and the growth in the number of metastases, although this dose also caused significant hypercalcemia and weight loss. We conducted several in vitro experiments to explore why EB 1089 did not prevent the occurrence of the primary tumors. EB1089 significantly inhibited the growth of a Tag-expressing human prostate epithelial cell line, BPH-1, and an androgen-insensitive subline of LNCaP cells [which was not inhibited by 1,25(OH)2D3]. Thus, neither Tag expression nor androgen insensitivity explain the absence of chemopreventive effect. Conversely, neither 1,25(OH)2D3 nor EB 1089 inhibited the growth of the normal rat prostate basal epithelial cell line NRP-152. It is likely that EB 1089 was not effective in delaying the growth of the primary tumor in G T-15 transgenic mice because the target cells of carcinogenesis in these mice are located in the basal epithelial layer. We conclude that G T-15 transgenic mice are a useful model for testing vitamin D-based therapies in androgen-insensitive prostate cancer but are not suitable for studies of vitamin D-based chemoprevention. The superiority of EB 1089 over 1,25(OH)2D3 in the growth suppression of androgeninsensitive prostate cancer cells supports the use of EB 1089 in androgen-insensitive prostate cancer. Introduction AIPC is the second leading cause of cancer death in American men (1). Men with cancer that has spread beyond the prostate typically undergo androgen deprivation for palliation. However, the average duration of response to androgen deprivation is only 2 years, and there are no effective therapies for androgen-independent disease (2). Thus, effective treatments for AIPC are urgently needed. In addition to their responsiveness to androgens, prostate cancer cells respond to another member of the steroid hormone superfamily, 1,25(OH)2D3 (calcitriol). The therapeutic use of vitamin D metabolites is supported by epidemiological studies that first suggested that 1,25(OH)2D3 maintains the differentiated phenotype of prostate cells and that reduced serum levels of 1,25(OH)2D3 or its precursor, 25-hydroxyvitamin D3, permits the progression of preclinical prostate cancer to clinical disease (3–6). VDR is expressed in most prostate cancer cell lines, and high levels are necessary to mediate the antiproliferative effects in vitro (7). However, factors other than VDR density are also important in mediating the antiproliferative effect. For example, the androgen-dependent cell line LNCaP is more sensitive to 1,25(OH)2D3 compared with the androgenindependent PC-3 and DU 145 prostate cancer cell lines, and these differences are not solely attributable to differences in VDR levels (8). A Phase II clinical trial in AIPC showed that 1,25(OH)2D3 Received 10/12/01; revised 2/22/02; accepted 3/30/02. 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 Veterans Affairs (VA) Merit Review Entry Program Grant 9869-01 (to C. M. P-S.), by Grant R01 CA 68565 (to G. G. S.), by Department of Defense Grant DAMD-17-98-1-8525 (to B. A. R.), and by the VA Medical Research Service. G. A. H. has a Senior Research Career Scientist award from the Department of Veterans Affairs. 2 To whom requests for reprints should be addressed, at Veterans Affairs Medical Center, GRECC (11-GRC), 1201 NW 16 Street, Miami, FL 33125. Phone: (305) 324-4455, extension 4391; Fax: (305) 324-3365; E-mail: [email protected]. 3 The abbreviations used are: AIPC, androgen-independent prostate cancer; 1,25(OH)2D3, 1,25-dihydroxyvitamin D3; VDR, vitamin D receptor; PSA, prostate-specific antigen; Tag, (SV40) T antigen; RPA, RNase protection assay; AR, androgen receptor; FBS, fetal bovine serum; PIN, prostate intraepithelial neoplasia; RT-PCR, reverse transcription-PCR; BPH, benign prostatic hyperplasia; TRAMP, transgenic adenocarcinoma mouse prostate; GAPDH, glyceraldehyde 3-phosphate dehydrogenase. 555 Vol. 11, 555–563, June 2002 Cancer Epidemiology, Biomarkers & Prevention Research. on September 20, 2017. © 2002 American Association for Cancer cebp.aacrjournals.org Downloaded from could lower serum PSA levels in some men (9). However, treatment with 1,25(OH)2D3 causes significant hypercalcemia and/or hypercalciurea (10). Synthetic analogues of 1,25(OH)2D3, such as EB 1089 (Seocalcitol; Ref. 11), with potent antiproliferative effects but reduced calcemic effects, can inhibit androgenindependent PC-3 prostate cancer cells in vitro (12, 13). EB 1089, acting through VDR, has a cell growth-inhibition mechanism similar to that of 1,25(OH)2D3, but with longer lasting and stronger effects. In LNCaP cells, EB 1089 inhibits the growth of prostate cancer cells by inducing G1 cell cycle block in vitro and has been shown to inhibit tumor growth in vivo without producing hypercalcemia (14, 15). In addition, a recent comparison of EB 1089 and 1,25(OH)2D3 treatment in the MATLyLu model of advanced AIPC showed that EB 1089 was as effective as 1,25(OH)2D3 in inhibiting metastases but was significantly less calcemic (16). These results suggest that EB 1089 may offer a therapeutic option in AIPC. In addition to its therapeutic uses, vitamin D and its analogues may be candidates for use in prostate cancer chemoprevention (17). The rationale for the use of vitamin D metabolites in chemoprevention is that vitamin D may maintain the differentiated phenotype of prostatic cells and delay or reverse the carcinogenic process before invasion and metastasis occur (5, 18). Testing these hypotheses in animals ideally requires models in which the cancer originates from normal prostate epithelial cells in their natural microenvironment and progresses through multiple stages, similar to human prostate cancer (19, 20). One study using the Lobound-Wistar rat model of urogenital cancer showed that treatment with the less calcemic 1,25(OH)2D3 analogue Ro24-5531 resulted in a limited chemoprevention effect (21). Recently, the advent of several transgenic mouse models of prostate cancer that target the expression of SV40 Tag to specific prostate epithelial cells have provided more suitable systems to test the chemoprevention and therapeutic potential of drugs (20, 22). The present study used G /T-15 transgenic mice (23, 24) to test the efficacy of EB 1089 in the prevention of tumor onset and in the growth inhibition of AIPC. Unlike transgenic models using prostate-specific promoters to target Tag to the prostate epithelial cells, the G /T-15 transgenic mice use the fetal G -globin gene promoter (25). The progression of prostate cancer in 75% of the transgenic males has important similarities to the progression of prostate cancer in men, e.g., originating from high-grade PIN s and progressing to advanced metastatic carcinomas. These tumors are clearly androgenindependent because castration of transgenic males before prostate tumor formation still results in the development of prostate tumors (24). Although Tag is not the cause of prostate cancer in men, the G /T-15 transgenic mice serve as a model of an aggressive, highly metastatic form of AIPC, which is the cause of virtually all deaths from prostate cancer in men. Materials and Methods Reagents. EB 1089 was synthesized by Leo Pharmaceuticals (Ballerup, Denmark). EB 1089 was used as a stock solution of 100 mg/ml in Solutol H15 (10 mg/ml), 7.7 mg/ml sodium phosphate dibasic anhydrous, 1 mg/ml sodium phosphate monobasic anhydrous, 2.8 mg/ml sodium chloride, and 10 mg/ml sodium ascorbate and was stored in the dark at 4°C. 1,25(OH)2D3, obtained from Biomol (Plymouth Meeting, PA), was dissolved in ethanol and stored in the dark at 20°C. Immunohistochemistry. Our previous results showed that the onset of prostate tumors in G /T-15 transgenic mice occurs between 16 and 32 weeks of age (24). To identify the target cells of carcinogenesis, prostates from G /T-15 transgenic mice (14–32 weeks old; n 10) without palpable or visible tumors were removed at necropsy, fixed in 10% buffered formalin for 6 h, dehydrated, embedded in paraffin, and sectioned at 5 m. Immunostaining for Tag was performed as described previously (26) using a 1:100 dilution of rabbit polyclonal antibody to Tag. The secondary antibody was a biotinylated goat antirabbit IgG. Specific color was developed with the Vector ABC kit (Zymed Laboratories Inc., South San Francisco, CA) and enhanced with 3,3 -diaminobenzidine (DAB)-nickel chloride; the sections were counterstained with nuclear Fast Red. Treatment of G /T-15 Transgenic Males with EB 1089. We used the G /T-15 transgenic mouse model of AIPC (23, 24) to assess the in vivo antitumor effect of EB 1089. These mice began to develop prostate tumors by 16 weeks of age (24). At 11 weeks, three of four transgenic males expressed Tag mRNA in the prostate as determined by RPA (data not shown). For this reason, we chose three different treatment starting time points (14, 11, and 9 weeks) before tumor onset to test the ability of EB 1089 to chemoprevent prostate tumors. Transgenic mice (CBA C57) were identified by DNA slot blot analysis as described previously (24). Transgenic male mice (14, 11, and 9 weeks of age) without palpable tumors were randomly divided into experimental and control groups and were given injections i.p. three times a week with 0.1 ml of freshly prepared EB 1089 at doses of 0.5, 2, 3, 4, 5, and 10 g/kg body weight (BW) diluted in Solutol H15 or placebo control (Solutol H15). Mice were kept in a 12-hour day/night cycle and fed a normal rodent diet containing 0.95% calcium and 4.5 IU/g vitamin D3 (Laboratory Rodent Diet 5001; PMI Nutrition International, Purina Mills, Inc., Richmond, IN). Starting at 16 weeks, mice were palpated in the urogenital area 3 days a week to detect prostate tumor mass. End points were 21 days after palpable prostate tumor mass was first detected or at 24 weeks of age. The percentage of mice that developed prostate tumors by 24 weeks of age and the average age when tumors were first detected by palpation (age of onset) were determined for each EB 1089 dose group at 14, 11, and 9 weeks and compared with placebo controls. Mice without a palpable prostate tumor at the 24-week end point but with a visible tumor nodule on dissection of the prostate (usually weighing 25–100 mg) were considered positive for prostate tumor formation. All of the animal studies were carried out with the approval of the Institutional Animal Care and Use Committee at the Miami Veterans Affairs Medical Center (American Association for Accreditation of Laboratory Animal Care-accredited) and conducted in accordance with the NIH Guidelines for the Care and Use of Laboratory Animals. Prostate Tumor Weight, Serum Calcium, and Body Weight. Twenty-one days after prostate tumors were first detected by palpation, EB 1089and placebo-treated mice were anesthetized, their blood was collected by cardiac puncture and centrifuged, and the sera were frozen and stored at 80°C. Primary prostate tumors and visible metastases to lymph nodes, adrenal glands, or kidney were removed, and their wet weights determined. Serum calcium was measured by dry slide technology on an automated Kodak Ektachem 700XR clinical chemistry analyzer (Rochester, NY). The body weight for EB 1089treated and placebo-treated mice was determined at the end of the study. Statistical differences in the wet weights of primary tumor and total metastases, serum calcium, and final body weight between EB 1089and placebo-treated mice were determined using the two-tailed Student’s t test. VDR mRNA Expression by RT-PCR and RPA. RNA from mouse prostate, kidney, and G /T-15 prostate tumor tissue was 556 EB 1089 Treatment of Transgenic Mice with Prostate Cancer Research. on September 20, 2017. © 2002 American Association for Cancer cebp.aacrjournals.org Downloaded from isolated by the LiCl-urea method (27) and treated with RNasefree DNase. The following DNA oligonucleotides synthesized by Operon Technologies (Alameda, CA) were used for RTPCR to detect VDR mRNA in mouse prostate and G /T-15 prostate tumor RNA: forward, 5 -GAGTTCTTTTGGTTGGACA-3 ; reverse, 5 -CAGCCTTCACAGGTCATA-3 (28). Conditions for RT-PCR were: 2 min at 94°C for 1 cycle; 1 min at 94°C, 1 min at 55°C, and 2 min at 72°C for 35 cycles; and 7 min at 72°C for 1 cycle. The expected 209-bp fragment from mouse prostate was cloned into the TA vector pCRII (Invitrogen, Carlsbad, CA), and its identity was confirmed by DNA sequencing. VDR mRNA in mouse kidney, prostate, and G / T-15 prostate tumor was measured by RPA with P-labeled Sp6 polymerase-synthesized antisense RNA probe from EcoRV-digested mouse VDR/PCRII DNA, using conditions described previously (23). VDR and AR Western Blot Analysis. Nuclear extracts from mouse (C57) prostate and kidney and from G /T-15 prostate tumors were prepared according to the procedure of Dent and Latchman (29), and protein concentrations were determined with the Bio-Rad protein assay (Bio-Rad Laboratories, Hercules, CA). After the separation of 10 g of protein by SDSPAGE, proteins were transferred by electrophoresis to Immobilon-P membrane (Millipore Corp., Bedford, MA) and incubated in 5% nonfat dry milk, PBS, and 0.25% Tween 20 for 1 h. Rabbit polyclonal antibodies specific for VDR (1:1000 dilution, C-20; Santa Cruz Biotechnology, Santa Cruz, CA) and AR (1:1000 dilution., N-20; Santa Cruz Biotechnology) were diluted in 5% nonfat dry milk, PBS, and 0.25% Tween 20 and incubated overnight at 4°C. Membranes were washed in PBS and 0.25% Tween 20 (three times, 10 min each time) and incubated with horseradish peroxidase-conjugated secondary antibody (antirabbit 1:2000 dilution; Santa Cruz Biotechnology) for 1 h, washed in PBS and 0.25% Tween 20, and analyzed by exposure to X-ray film (X-Omat, Eastman Kodak Co., Rochester, NY) using enhanced chemiluminescence plus (ECL plus; Amersham Pharmacia Biotech, Arlington Heights, IL). Cell Culture and Treatment with 1,25(OH)2D3 and EB 1089. LNCaP-AI is an androgen-independent derivative of the human prostate cancer cell line LNCaP-FGC (Ref. 30; American Type Culture Collection, Manassas, VA), which was spontaneously derived in our laboratory. These cells express AR and PSA, similar to LNCaP-FGC (data not shown). LNCaP-AI cells were maintained in RPMI 1640 (Life Technologies, Inc.) with 5% FBS (Hyclone, Logan, UT), 100 units/ml penicillin, 100 g/ml streptomycin, and 0.25 g/ml amphotericin (Life Technologies, Inc.). Unlike androgen-dependent LNCaP-FGC, the LNCaP-AI cells are able to grow long-term in RPMI 1640 with 5% charcoal-stripped FBS (Hyclone) and are referred to as LNCaP-AI/CSS. The Tag-expressing human prostate epithelial cell line BPH-1, generously provided by S. Hayward, was maintained in RPMI 1640 with 5% FBS (31). The normal rat prostate basal epithelial cell line NRP-152, generously provided by D. Danielpour, was maintained in HEPES-free DMEM/F12 (1:1, v/v; Life Technologies, Inc.) with 5% FBS, antibiotic/ antimycotic, 20 ng/ml epidermal growth factor, 10 ng/ml cholera toxin, 5 g/ml insulin (Life Technologies, Inc.), and 0.1 M dexamethasone (Sigma, St. Louis, MO; Ref. 32). To determine and compare prostate cell growth inhibition by 1,25(OH)2D3 and EB 1089, 7.5 10 4 LNCaP-AI/CSS, 0.75 10 BPH-1, and 1.5 10 NRP-152 cells were seeded in 6-well plates in their corresponding medium containing 5% FBS and allowed to attach overnight. The next day, fresh media containing 1,25(OH)2D3 or EB 1089 (1 and 10 nM) or ethanol vehicle control (0.1% total volume) were added. After 3 days, the medium was changed and replenished. On the sixth day, cells were removed by Fig. 1. Prostatic target cells of carcinogenesis in G /T-15 transgenic mice. A, light micrograph (H&E, 400) of preneoplastic lesion similar to high-grade PIN (arrow) containing epithelial cells protruding into the lumen (L). B, light micrograph (nuclear Fast Red, 200) of Tag immunohistochemistry showing a normal-appearing prostate acini containing Tag-expressing epithelial cells located in the basal epithelial layer (arrows pointing to cells that contain black nucleus) next to an acini not containing Tag-expressing cells. C, light micrograph (nuclear Fast Red, 200) showing proliferating Tagexpressing cells (arrows) producing an elevation toward the surface layer into the lumen (L) of prostate gland. D, light micrograph (nuclear Fast Red, 200) showing prostate duct filled with Tag-expressing dysplastic cells (large arrow) next to a normalappearing prostate duct containing Tag-expressing epithelial cells in the basal layer (small arrows). 557 Cancer Epidemiology, Biomarkers & Prevention Research. on September 20, 2017. © 2002 American Association for Cancer cebp.aacrjournals.org Downloaded from trypsin-EDTA (Life Technologies, Inc.), and viable cells were counted with a Neubauer hemacytometer. In all of the experiments, the control-treated cells reached 80–90% confluency after 6 days of growth and were presumed to be in a log-growth phase. Cell numbers in each experiment were derived from the average value of quadruplicate wells repeated three independent times and calculated as percentage of vehicle control. Statistical differences between 1,25(OH)2D3and EB 1089-treated and control cells were determined by two-tailed Student’s t test.