ANTICANCER RESEARCH 29: 1089-1094 (2009)

Indirect Antitumor Effects of Bisphosphonates on Prostatic LNCaP Cells Co-cultured with Bone Cells

TOMOAKI TANAKA, HIDENORI KAWASHIMA, KEIKO OHNISHI, KENTARO MATSUMURA, RIKIO YOSHIMURA, MASAHIDE MATSUYAMA, KATSUYUKI KURATSUKURI and TATSUYA NAKATANI

Department of Urology, Osaka City University Graduate School of Medicine, Osaka 545-8585, Japan

Abstract. Bisphosphonates are strong inhibitors of effectively inhibit bone absorption are available for osteoclastic bone resorption in both benign and malignant reducing the risk of skeletal complications in bone bone diseases. The nitrogen-containing bisphosphonates metastases of cancer as well as those of other solid (N-BPs) have strong cytotoxicity via inhibition of protein tumors characterized by osteoclastic lesions. prenylation in the mevalonate pathway, and also demonstrate Bisphosphonates are potent inhibitors of osteoclast activity direct cytostatic and proapoptotic effects on and survival, thereby reducing osteoclast-mediated bone cells. We confirmed the usefulness of a co-culture system absorption (1, 4-6). The chemical structure of bisphosphonates comprised of prostatic LNCaP cells, ST2 cells (mouse- contributes to their mechanism of action: Non-nitrogen- derived ) and MLC-6 cells (mouse-derived containing compounds (e.g. etidronate, clodronate) are osteoclasts) in vitro. N-BPs (pamidronate and zoledronic metabolized to cytotoxic analogues of ATP (7), while newer acid) inhibited both receptor transactivation and nitrogen-containing bisphosphonates (N-BPs; pamidronate, tumor cell proliferation by suppressing the activities of both ibandronate, zoledronic acid) have strong cytotoxic effects on osteoclasts and osteoblasts with low-dose exposure. This cells via inhibition of protein prenylation in the mevalonate indirect inhibition of prostate cancer cells via bone cells pathway (8, 9). The more potent N-BPs have also been could be beneficial in treating prostate cancer patients with demonstrated to exert direct cytostatic and proapoptotic effects bone metastases. on prostate cancer cell lines in vitro (10), the prevention of cancer cell adhesion and invasion via reduction of matrix Approximately 70% of patients with advanced prostate metalloproteinase (MMP) expressions (11-13), and the cancer have bone metastases (1-3). These bone metastases inhibition of testosterone-induced angiogenesis in a castrated are associated with considerable skeletal morbidity, animal model (14). Thus, treatment with N-BPs could exert a including severe bone pain requiring narcotics or palliative strong inhibitory effect on osteoclastic bone resorption and radiation therapy, pathological fractures, spinal cord also have antitumor potential against bone metastases, directly, compression and hypercalcemia of malignancy (HCM), or indirectly, via alterations in the bone microenvironment (5, which significantly decrease quality-of-life. These 8). In this study, we investigated the biological effects of bone complications result from excessive bone turnover, mainly cells on LNCaP cells and the antitumor effects of low doses bone formation which represents malignant bone lesions. of two N-BPs (pamidronate and zoledronic acid) on LNCaP While the principal characteristic of bone metastases in cells via both osteoblasts and osteoclasts in vitro. prostate cancer is osteoblastic lesions, bone absorption is initially needed to create the space for metastatic sites and Materials and Methods to enhance bone formation. Therefore, therapies that Materials. Pamidronate and zoledronic acid were provided by Novartis Pharma AG (Basel, Switzerland) and 5α-dihydrotesto- sterone (DHT) was purchased from Sigma-Aldrich (St. Louis, MO, Correspondence to: Dr. Tomoaki Tanaka, Department of Urology, USA). The dual-luciferase reporter assay was purchased from Osaka City University, Graduate School of Medicine, 1-4-3 Promega (Madison, WI, USA) and the WST-1 cell proliferation Asahimachi, Abeno-ku, Osaka 545-8585, Japan. Tel:+81 6 6645 assay was obtained from Takara Biochemicals (Tokyo, Japan). The 3857, Fax: +81 6 6647 4426, e-mail: [email protected] prostate cancer cell line, LNCaP, was purchased from the American cu.ac.jp Type Culture Collection (Rockville, MD, USA) and the mouse- derived cell line, ST2, and the mouse-derived osteoclast Key Words: Bisphosphonate, pamidronate, zoledronic acid, LNCaP, cell line, MLC-6, were obtained from Riken cell bank (Ibaragi, osteoblast, osteoclast. Japan). The reporter gene plasmid, a mouse mammary tumor virus

0250-7005/2009 $2.00+.40 1089 ANTICANCER RESEARCH 29: 1089-1094 (2009)

Table I. Nucleotide sequence of primers used for reverse transcription- polymerase chain reaction.

Gene Nucleotide sequence (5’¡3’)

TRAP TTCCTCCAGGATGGATTCATGG CTGAAGATACTGCAGGTTGTGG COL1A1 CATGGGTCCTTCTGGTCCTCGT CTTGCCAGCTTCCCCATCATCT GAPDH ACCACAGTCCATGCCATCAC TCCACCACCCTGTTGCTGTA

TRAP, tartrate-resistant acid phosphatase; COL1A1, type Ι collagen; GAPDH, glyceraldehydes-3-phosphate dehydrogenase.

Figure 1. Schematic representation of in vitro co-culture system with (10 μM) or zoledronic acid (10 μM) was added, and the cells were LNCaP cells,osteoclasts (ML-6) and osteoblasts (ST2), followed by incubated for an additional 48 h. Premix WST-1 was added to the additional experiments. medium and the cells were incubated for 3 h. Absorbance was measured at 450 nm. All assays were carried out in triplicate.

RNA extraction and RT-PCR. Total RNA was extracted from the (MMTV)-luciferase reporter plasmid (pMMTV-luc) was kindly mixture of ST2 and MLC-6 with ISOGEN (NipponGene, Japan). provided by N.V. Organon (Oss, Netherland). Complementary DNA was synthesized from each total RNA (500 ng), subjected to PCR amplification with gene-specific PCR primers . The prostate cancer cell line, LNCaP, and mouse and Taq DNA polymerase using a Takara RNA PCR kit according to osteoblast cell line, ST2, were maintained in RPMI-1640 with 10% the manufacturer’s instructions. The primers designed for fetal calf serum (FCS). The mouse osteoclast cell line, MLC-6, was glyceraldehyde-3-phosphate dehydrogenase (GAPDH), tartrate- maintained in McCoy’s 5A medium with 20% FCS. resistant acid phosphatase (TRAP) and type 1 collagen (COL1A1) are listed in Table I. The PCR consisted of 27 cycles for RANK and Co-culture of LNCaP cells with bone cells and transient transfection. RANKL and 25 cycles for GAPDH. An in vitro bicompartment culture system was used (0.4 μm pore; Becton Dickinson Labware, Franklin Lakes, NJ, USA) (Figure 1). Statistical analysis. Data analysis was performed using Student’s t- In brief, LNCaP cells were seeded onto the six-well outer tissue test. P<0.05 was considered to denote a statistically significant 4 3 culture plate (1.0×10 cells/ well), and ST2 (5.0×10 cells/ well) and difference. MLC-6 (5.0×103 cells/ well) cells were plated on the inner culture insert. Co-culture was performed in RPMI-1640 with 10% FCS. Results After 120 h, LNCaP cells were transiently transfected using the FuGENE 6 transfection reagent (Roche) with 2 μg of a DNA Effects of bone cells on the transcriptional activity of AR in mixture containing the pMMTV-luc reporter plasmid (1 μg) and LNCaP cells. To examine whether a prostate cancer cell line, pRL-TK (40 ng), in accordance with the manufacturer’s instructions LNCaP, is affected by bone cells, we investigated the effects and grown for an additional 24 h in RPMI-1640 with 10% fetal bovine dialyzed serum (10,000 MW cut-off; Sigma). DHT (5 nM) of two bone cell lines (ST2 and MLC-6) on endogenous AR with or without either pamidronate (10 μM) or zoledronic acid (10 transcriptional activity in LNCaP cells. As shown in Figure μM) was then added and the cells were cultured for another 48 h. 2, the DHT-induced transcriptional activity of AR was not influenced by co-culture of LNCaP with either ST2 or MLC- Dual luciferase assay. Endogenous (AR) 6. In contrast, co-culture of both ST2 and MLC-6 in the transcriptional activities were monitored by a reporter gene assay same insert significantly enhanced DHT-induced AR activity using a pMMTV-luc. LNCaP cells were harvested, lysed in passive in the coexisting LNCaP cells. Accordingly, we examined the lysis buffer, subjected to two cycles of freeze/thaw and analyzed for luciferase activities using a dual luciferase reporter assay system effects of bisphosphonates on LNCaP cells via two bone cell (Promega) according to the methods recommended by the lines using this co-culture system with three different cell manufacturer. lines (LNCaP, ST2 and MLC-6).

Cell proliferation assay. To evaluate the effect of bisphosphonates Inhibition of bone cell-induced AR activity in LNCaP cells by against tumor cell proliferation, we conducted a WST-1 assay (Premix N-BPs (pamidronate, zoledronic acid). The transcriptional WST-1 Cell Proliferation System; Takara Bio, Japan). In brief, LNCaP activities of AR in LNCaP cells co-cultured with or without cells were seeded onto the six-well outer tissue culture plate (1.0×104 cells/ well), and ST2 (5.0×103 cells/ well) and MLC-6 (5.0×103 cells/ ST2 and MLC-6 cells were evaluated by a dual luciferase well) cells were plated onto the inner culture insert. Co-culture was reporter assay after incubation with pamidronate or performed in RPMI-1640 with 10% FCS. After 120 h, pamidronate zoledronic acid for 2 days. As shown in Figure 3A, the DHT-

1090 Tanaka et al: Effects of Bisphosphonates on Prostate Cancer Cells via Bone Cells

Figure 2. Effects of bone cells on AR activity of LNCaP cells. As described in the Materials and Methods section, at 5 days after incubation of LNCaP cells with or without ST2 cells and MLC-6 cells, LNCaP cells were transfected with 1 μg of the pMMTV-luc reporter plasmid and 40 ng of the pRL-TK vector (as the internal control) for 24 h. The following day, DHT was added at a 5 nM concentration. The relative luciferase activities were measured using the dual-luciferase reporter assay system after incubation for an additional 24 h. Values measured without DHT in LNCaP cells not co-cultured with bone cells were defined as l (control). These values represent the means±SD from three independent experiments. *P<0.01 as compared to the values measured with DHT in LNCaP cells alone.

induced transcriptional activity of AR in LNCaP cells was not significantly increased as compared with those of LNCaP influenced directly by exposure to N-BPs at 10 μM. On the cells alone. Furthermore, both bisphosphonates significantly other hand, the enhanced transcriptional activity of AR in inhibited LNCaP cell proliferation in the co-culture with LNCaP cells co-cultured with two bone cell lines was bone cells. These observations indicate that both N-BPs have significantly inhibited by treatment with 10 μM concentration indirect effects on LNCaP cell proliferation, and that the of bisphosphonates. Both ST2 and MLC-6 cells co-cultured newer bisphosphonate, zoledronic acid, also has a direct in the same inserts were harvested and total RNA was effect on LNCaP cell growth in the same setting as above. isolated. RT-PCR revealed that bisphosphonates reduced gene expressions of TRAP specific for osteoclasts (MLC-6) and Discussion COL1A1 specific for osteoblasts (ST2) (Figure 3B). These results indicate that bisphosphonates indirectly inhibited the In advanced prostate cancer, bone is the preferred site of transcriptional activity of AR in LNCaP cells through . The majority of patients with suppression of the activities of ST2 and MLC-6 cells. ultimately develop -independent lesions after initial androgen ablation therapy and suffer from the complications Effects of N-BPs on cell proliferation of LNCaP co-cultured associated with bone metastases, which result in significant with or without bone cells. The cell proliferation of LNCaP skeletal morbidity, such as pathological bone fractures and cells co-cultured with or without bone cells was evaluated spinal cord compression (1, 2). While the principal by WST-1 assay after incubation with bisphosphonates characteristic of bone metastases in prostate cancer is (Figure 4). The proliferation of LNCaP cells alone was not osteoblastic lesions, bone absorption is initially needed to markedly affected by 10 μM pamidronate, but was create the space for metastatic sites and enhance bone significantly inhibited by 10 μM zoledronic acid. In contrast, formation. Therefore, bisphosphonates that effectively inhibit the proliferation of LNCaP cells co-cultured with bone cells bone absorption can reduce the risk of skeletal complications

1091 ANTICANCER RESEARCH 29: 1089-1094 (2009)

Figure 3. Inhibition of AR activity in LNCaP cells via the effects of N-BPs on bone cells. A, At 5 days after incubation of LNCaP cells with or without ST2 and MLC-6 cells, LNCaP cells were transfected with 1 μg of the pMMTV-luc reporter plasmid and 40 ng of the pRL-TK vector (as the internal control) for 24 h. The following day, DHT and N-BPs (pamidronate or zoledronic acid) were added at 5 nM and 10 μM, respectively. The relative luciferase activities were measured using the dual-luciferase reporter assay system after incubation for an additional 48 h. Values measured without DHT and N-BPs in LNCaP cells not co-cultured with bone cells were defined as l (control). Results (means±SD) are representative of three independent experiments. *P<0.05, as compared to the values measured with DHT in LNCaP cells co-cultured with ST2 and MLC-6 cells (Student’s t-test, n=3). B, To evaluate the effects of N-BPs against ST2 cells and MLC-6 cells, TRAP and COL1A1 were used as surrogate gene markers specific for osteoclasts (MLC-6) and osteoblasts (ST2), respectively. At 48 h after exposure to N-BPs, total RNA was extracted and semiquantitative RT-PCR was performed using primers against TRAP, COL1A1 and GAPDH.

in bone metastases of prostate cancer as well as those of The N-BPs such as pamidronate and zoledronic acid have other solid tumors characterized by osteoclastic lesions (4, been shown to be clinically effective for ameliorating some 5). These compounds preferentially accumulate in bone and bone-associated disorders (e.g. pathological fractures, spinal inhibit bone absorption by suppressing the activities of cord compression) derived from bone metastasis of certain functional osteoclasts. tumors, including prostate cancer (15-17). A vicious cycle

1092 Tanaka et al: Effects of Bisphosphonates on Prostate Cancer Cells via Bone Cells

Figure 4. Inhibition of proliferation of LNCaP cells via the effects of N-BPs on bone cells. As described in the Materials and Methods section, N- BPs (10 μM) were added to the LNCaP cells co-existing with or without bone cells at 5 days after starting the co-culture. The number of viable LNCaP cells among co-cultured cells treated with or without N-BPs for 48 h was determined by WST-1 assay. Values measured without N-BPs in LNCaP cells not co-cultured with bone cells were defined as l (control). Three individual experiments were performed. Results were expressed as means±SD. *P<0.01 as compared to the control (Student’s t-test, n=3). †P<0.01 as compared to the values measured in LNCaP cells co-cultured with bone cells (Student’s t-test, n=3).

exists involving tumor cells and bone cells in the However, higher doses of bisphosphonates were used in microenvironment of prostate cancer bone metastasis (5, 8). these experiments than are routinely used clinically for Specifically, the growth factors and cytokines, such as PTHrP patients with bone metastasis. On the contrary, the and IL6, secreted from metastatic prostate cancer cells induce concentration (10 μM) of N-BPs used in our experiments differentiation and proliferation of osteoclasts via the receptor corresponds to 10 times the maximum serum concentration activator of nuclear factor kappaB (RANK)/RANK ligand after clinical administration to patients (19). (RANKL) pathway. Activated osteoclasts accelerate bone In the current study, we established an in vitro experimental resorption and release growth factors, such as insulin-like model derived from prostate cancer cells, osteoblasts and growth factor- 1(IGF-1), basic fibroblast growth factor (bFGF) osteoclasts in order to investigate the actions of and transforming growth factor beta-1 (TGF-β). These growth bisphosphonates in the microenvironment of prostate cancer factors enhance tumor cell activity, thereby generating both bone metastasis (Figure 1). Interestingly, these bisphospho- osteoclastic and osteoblastic factors. nates, pamidronate and zoledronic acid, inhibited both AR Generally, it is difficult to determine whether the transactivation activity and tumor cell proliferation of antitumor effects of bisphosphonates are attributable to direct LNCaP cells indirectly by suppressing the activities of both effects on tumor cells or indirect effects, i.e. decreased osteoblasts and osteoclasts at exposures approaching clinical release of growth factors from the bone matrix. Several sets doses. As previously stated, bisphosphonates have a high of in vitro data have indicated that bisphosphonates have affinity for hydroxyapatite in bone minerals and the direct antitumor activity, for instance, tumor cell apoptosis concentration in bone is more than 100-fold that in serum and inhibition of cell proliferation (4, 10, 13). Furthermore, after bisphosphonate administration (8). It is possible that previous data on one of the N-BPs, minodronate, showed the accumulation of bisphosphonates in bone may achieve a inhibition of both tumor cell adhesion and invasion activity concentration adequate for exerting direct antitumor effects. through regulation of the SDF-1/CXCR-4 pathway (18). Fundamentally, our current data endorse the concept that

1093 ANTICANCER RESEARCH 29: 1089-1094 (2009) indirect antitumor effects mediated by suppression of 12 Boissier S, Ferreras M, Peyruchaud O, Magnetto S, Ebetino FH, osteoblast and osteoclast activities contribute to significant Colombel M, Delmas P, Delaisse JM and Clezardin P: inhibition of prostate cancer cell proliferation, regardless of Bisphosphonates inhibit breast and prostate carcinoma cell invasion, an early event in the formation of bone metastases. direct antitumor effects. Cancer Res 60: 2949-2954, 2000. There are numerous reports suggesting the effectiveness 13 Matsunaga S, Iguchi K, Usui S and Hirano K: Incadronate of bisphosphonates in the treatment of bone metastases (3, induces cell detachment and apoptosis in prostatic PC-3 cells. 15-18, 20, 21). However, the mechanisms of action of Anticancer Res 27: 927-932, 2007. bisphosphonates in the bone metastatic microenvironment 14 Fournier P, Boissier S, Filleur S, Guglielmi J, Cabon F, Colombel remain unclear. M and Clezardin P: Bisphosphonates inhibit angiogenesis in vitro In conclusion, N-BPs exert indirect antitumor effects and testosterone-stimulated vascular regrowth in the ventral prostate in castrated rats. Cancer Res 62: 6538-6544, 2002. against prostate cancer cells, which are attributable to 15 Small EJ, Smith MR, Seaman JJ, Petrone S and Kowalski MO: suppression of bone cell activities at metastatic sites. Combined analysis of two multicenter, randomized, placebo- controlled studies of pamidronate disodium for the palliation of Acknowledgements bone pain in men with metastatic prostate cancer. J Clin Oncol 21: 4277-4284, 2003. We thank Novartis Pharma AG for providing us with the N-BPs 16 Saad F, Gleason DM, Murray R, Tchekmedyian S, Venner P, (pamidronate and zoledronic acid). This work was supported by Lacombe L, Chin JL, Vinholes JJ, Goas JA and Chen B: A grants from the Ministry of Education, Science, and Culture of randomized, placebo-controlled trial of zoledronic acid in Japan, and Osaka City University Medical Research Foundation. patients with hormone-refractory metastatic prostate carcinoma. J Natl Cancer Inst 94: 1458-1468, 2002. References 17 Saad F, Gleason DM, Murray R, Tchekmedyian S, Venner P, Lacombe L, Chin JL, Vinholes JJ, Goas JA and Zheng M: Long- term efficacy of zoledronic acid for the prevention of skeletal 1 Adami S: Bisphosphonates in prostate carcinoma. Cancer 80: complications in patients with metastatic hormone-refractory 1674-1679, 1997. prostate cancer. J Natl Cancer Inst 96: 879-882, 2004. 2 Coleman RE: Metastatic bone disease: clinical features, 18 Miwa S, Mizokami A, Keller ET, Taichman R, Zhang J and pathophysiology and treatment strategies. Cancer Treat Rev 27: Namiki M: The bisphosphonate YM529 inhibits osteolytic and 165-176, 2001. osteoblastic changes and CXCR-4-induced invasion in prostate 3 Coleman RE: Bisphosphonates: clinical experience. Oncologist cancer. Cancer Res 65: 8818-8825, 2005. 9(Suppl 4): 14-27, 2004. 19 Skerjanec A, Berenson J, Hsu C, Major P, Miller WH, Jr., Ravera 4 Dumon JC, Journe F, Kheddoumi N, Lagneaux L and Body JJ: C, Schran H, Seaman J and Waldmeier F: The pharmacokinetics Cytostatic and apoptotic effects of bisphosphonates on prostate and pharmacodynamics of zoledronic acid in cancer patients cancer cells. Eur Urol 45: 521-528; discussion 528-529, 2004. with varying degrees of renal function. J Clin Pharmacol 43: 5 Fleisch H: Bisphosphonates: mechanisms of action. Endocr Rev 154-162, 2003. 19: 80-100, 1998. 20 Ernst DS, Tannock IF, Winquist EW, Venner PM, Reyno L, Moore 6 Oades GM, Coxon J and Colston KW: The potential role of MJ, Chi K, Ding K, Elliott C and Parulekar W: Randomized, bisphosphonates in prostate cancer. Prostate Cancer Prostatic Dis double-blind, controlled trial of mitoxantrone/prednisone and 5: 264-272, 2002. clodronate versus mitoxantrone/prednisone and placebo in patients 7 Frith JC, Monkkonen J, Blackburn GM, Russell RG and Rogers with hormone-refractory prostate cancer and pain. J Clin Oncol MJ: Clodronate and liposome-encapsulated clodronate are 21: 3335-3342, 2003. metabolized to a toxic ATP analog, adenosine 5’-(beta, gamma- 21 Kim SJ, Uehara H, Yazici S, He J, Langley RR, Mathew P, Fan dichloromethylene) triphosphate, by mammalian cells in vitro. J D and Fidler IJ: Modulation of bone microenvironment with Bone Miner Res 12: 1358-1367, 1997. zoledronate enhances the therapeutic effects of STI571 and 8 Rogers MJ, Frith JC, Luckman SP, Coxon FP, Benford HL, paclitaxel against experimental bone metastasis of human Monkkonen J, Auriola S, Chilton KM and Russell RG: Molecular prostate cancer. Cancer Res 65: 3707-3715, 2005. mechanisms of action of bisphosphonates. Bone 24: 73S-79S, 1999. 9 Benford HL, Frith JC, Auriola S, Monkkonen J and Rogers MJ: Farnesol and geranylgeraniol prevent activation of caspases by aminobisphosphonates: biochemical evidence for two distinct pharmacological classes of bisphosphonate drugs. Mol Pharmacol 56: 131-140, 1999. 10 Oades GM, Senaratne SG, Clarke IA, Kirby RS and Colston KW: Nitrogen containing bisphosphonates induce apoptosis and inhibit the mevalonate pathway, impairing Ras membrane localization in prostate cancer cells. J Urol 170: 246-252, 2003. 11 Coxon JP, Oades GM, Kirby RS and Colston KW: Zoledronic acid induces apoptosis and inhibits adhesion to mineralized Received July 4, 2008 matrix in prostate cancer cells via inhibition of protein Revised November 28, 2008 prenylation. BJU Int 94: 164-170, 2004. Accepted January 19, 2009

1094