The paper by Zellweger et al. builds on the continuing story of clusterin (TRPM-2) in the development and progression of prostate cancer. This group have published a series of papers on this protein, showing that it correlates with progression to androgen-independence and resistance to apoptosis. One of their recent papers has shown that ‘knocking out’ clusterin increases radiation sensitivity in prostate cancer cells. The current paper reports that increasing the expression of clusterin in LNCaP cells increases the cell's resistance to radiation-induced apoptosis. Manipulating identified survival proteins has important implications in preventing androgen-independent progression. Clusterin is such a survival protein and represents an important drug target in the near future.
To evaluate the effect of clusterin overexpression on radiation-induced tumour growth rates and apoptosis in human prostate LNCaP cells, as prostate cancer cells are relatively resistant to radiation-induced apoptosis and local recurrences are common, but overexpression of the anti-apoptotic protein clusterin can accelerate progression to androgen-independence and to confer a chemoresistant phenotype in various prostate cancer models.
MATERIALS AND METHODS
Western blot analysis and immunohistochemistry were used to compare clusterin expression levels in parental (P) and clusterin-transfected (T) LNCaP cells in vitro and in vivo. The effects of radiation on clusterin-expression in both parental LNCaP/P and clusterin-transfected LNCaP/T tumours were analysed by Northern blot analysis. The cellular response to radiation was determined up to 3 weeks after irradiation using tetrazolium and re-growth assays, and cell-cycle analysis by flow cytometry.
Clusterin mRNA expression increased from undetectable to low levels in LNCaP/P tumours after radiation and more than three-fold in LNCaP/T tumours. Clusterin overexpression decreased the radiosensitivity in a time-dependent manner, reducing the extent of growth arrest and apoptosis by up to 54%. Re-growth assays showed that the improved survival rates of LNCaP/T cells after radiation did not change after 3 days, remaining constant over 3 weeks.
These results identify clusterin as a promoter of cell survival that may help mediate resistance to radiation-induced apoptosis. Furthermore, clusterin overexpression seems to provide an extended protection against radiation-induced cell cycle arrest and apoptosis.
Prostate cancer is the most frequent cancer in Western countries and the second leading cause of cancer-related deaths in men . Although radiation is capable of permanently eradicating localized prostate cancer, up to 30% of patients treated with potentially curative doses relapse at the sites of the irradiated tumours [2–4]. This indicates that there are variations in clonal sensitivity to the lethal effects of radiation within a given tumour.
The predominant mechanism by which radiation kills mammalian cells depends on mitosis. Double-stranded DNA breaks are considered to be the specific lesions resulting in lethal mutations or chromosomal aberrations that eventually lead to progeny cell death [5,6], which may be through an apoptotic process . Radiation-induced damage can also signal apoptosis directly, although it appears to be less prevalent than mitotic cell death . The pleiotropic nature of death pathways induced by radiation suggests that radiation resistance is likely to be regulated by a variety of mechanisms, each of which is associated with a specific death pathway. Whether radiation resistance of human prostate tumour clones is associated with a single mechanism, or a spectrum of mechanisms, is unknown, but the latter is more likely. Improved understanding of molecular changes in prostate cancer after radiation may yield potential for targeted modulation of radiation resistance.
Clusterin, also known as testosterone-repressed prostate message-2 or sulphated glycoprotein-2, was first isolated from ram rete testes fluid , and has been implicated in a wide variety of physiological and pathological processes, including tissue remodelling, reproduction, lipid transport, membrane protection, complement defence and apoptotic cell death . As clusterin expression is increased in various benign and malignant tissues undergoing apoptosis, it has been associated with cell death [11–14]. Recent observations show that clusterin acts in a chaperone-like manner, similar to that of small heat-shock proteins. Clusterin potently inhibits stress-induced protein precipitation in vitro and appears to be associated with cell survival, tumour progression and treatment resistance in vivo[15–17]. These results are supported by recent data from Bettuzzi et al. suggesting an anti-oncogenic role of clusterin in the prostate, thus controlling normal and transformed epithelial cells. In prostate cancer, experimental and clinical studies indicate that clusterin expression is associated with androgen-independent recurrences and has a protective role against apoptotic cell death [19,20]. Furthermore, inhibition of clusterin expression using sequence-specific antisense oligonucleotides enhances radiation sensitivity in human PC-3 tumours . Considering the strategic significance of clusterin in suppressing apoptosis, we examined whether clusterin overexpression could alter the radiosensitivity of LNCaP cells.
LNCaP is an androgen-dependent human prostate cancer cell line and is considered to be highly radioresistant compared with other human cancer cell lines . Parental (P) LNCaP cells express very low levels of clusterin, thereby offering an ideal model to study the functional role of clusterin overexpression in radiation response using a stably transfected (T) LNCaP cell line .
MATERIALS AND METHODS
LNCaP cells were kindly provided by Dr L.W.K. Chung (University of Virginia, Charlottesville, VA) and maintained in RPMI 1640 (Life Technologies, Inc., Gaithersburg, MD) supplemented with 5% heat-inactivated fetal calf serum. Dr Martin Tenniswood (W. Alton Jones Cell Science Center, Lake Placid, NY) kindly provided PRC-CMV expression vector containing the 1.6-kb cDNA fragment encoding human clusterin. The expression vector was transfected into LNCaP/P cells by the liposome-mediated gene transfer method as previously described .
WESTERN BLOT ANALYSIS
Samples containing equal amounts of protein (15 µg) from lysates of cultured and irradiated LNCaP cells were subjected to SDS-PAGE and transferred to nitrocellulose filters. The filters were blocked overnight in PBS containing 5% nonfat milk powder at 4 °C and then incubated for 1 h with a 1 : 400-diluted antihuman clusterin goat polyclonal antibody (Santa Cruz Biotechnology Inc., CA), 1: 200-diluted antihuman cyclin D1 rabbit polyclonal antibody (Santa Cruz), or 1: 300-diluted antihuman beta-actin mouse monoclonal antibody (Sigma Chemical Co., St Louis, MO). The filters were then incubated for 30 min with 1: 10000-diluted horseradish peroxidase-conjugated antigoat, rabbit or mouse monoclonal antibody (Santa Cruz), and specific proteins detected using an enhanced chemiluminescence Western blotting analysis system (Amersham Life Science, Arlington Heights, IL).
NORTHERN BLOT ANALYSIS
Total RNA was isolated from irradiated LNCaP tumour tissues using the acid-guanidium thiocyanate-phenol-chloroform method. Electrophoresis, hybridization and washing conditions were carried out as previously reported . Human clusterin and β-actin cDNA probes were generated by RT-PCR from total RNA of human kidney using primers 5′–AAGGAAATTCAAAATGCTGTCAA-3′ (sense) and 5′-ACAGACAAGATCTCCCGGCACTT-3′ (antisense) for clusterin, and 5′–GGACGTGAC TGACTACCTCATGAA-3′ (sense) and 5′–TGATCCACATCTGCTGGAAGGTGG-3′ (antisense) for β-actin. The density of bands for clusterin was normalized against that of β-actin by densitometric analysis.
All irradiation was carried out using an X-ray machine at 250 kVp using a 0.5-mm copper filter. LNCaP cells were irradiated at room temperature in 100 mm culture dishes when cultures had reached ≈ 75% confluence, and received either a mock treatment for the control or 4 or 8 Gy at 2.17 Gy/min. After irradiation cultures were returned to their 5% CO2/37 °C incubation until harvesting at the times required.
The in vitro growth inhibitory effects of radiation on LNCaP/P and LNCaP/T cells were assessed using the tetrazolium-based assay (MTT) as previously described . After radiation with 4 Gy, 1 × 104 cells were seeded in each well of 96-well microtitre plates and allowed to attach overnight. Every 24 h over 5 days, growth was assessed by adding 20 µL of 5 mg/mL MTT (Sigma) in PBS to each well, followed by incubation for 1 h at 37 °C. The resulting formazan crystals were dissolved in DMSO. The optical density was determined with a microculture plate reader (Becton Dickinson Labware, Lincoln Park, NJ) at 540 nm. Absorbance values were normalised to the values obtained for the vehicle-treated cells to determine the percentage of survival. Each assay was performed in triplicate.
After irradiation with 8 Gy, plates were trypsinized and the number of cells per plate determined by Coulter and haemocytometer counting, including unattached cells in the overlaying medium. Individual plates were trypsinized every 4–7 days and cells counted using the same method. Relative growth was defined as the ratio of cell number on a given day divided by the cell number immediately after irradiation.
FLOW CYTOMETRIC ANALYSIS
LNCaP cells were plated on 150 mm dishes; when the cells were 70% confluent, they were either irradiated with 8 Gy as described above or received a mock treatment for the control. At 3 h later all cells were trypsinized for passage in 150 mm dishes. After reaching confluency 3 days later, cells were trypsinized into a single cell suspension, centrifuged at 200 g for 10 min, and fixed with 70% ethanol and re-centrifuged. Cells were then incubated with RNAse (0.5 mg/mL) at 37 °C for 30 min, re-centrifuged and stained with 1 mL of 50 µg/mL propidium iodide (Sigma). The DNA profile was then analysed using a dual-laser flow cytometer (Beckman Coulter Epics Elite, Beckman Inc., Miami, FL).
Sections from formaldehyde-fixed, paraffin-embedded LNCaP tumours were deparaffinized by passing slides through 5 min washes of xylene (× 2) and graded ethanol successively. Endogenous peroxidase was blocked with 3% hydrogen peroxide in absolute methanol. Slides were autoclaved for 15 min in target-retrieval solution (Dako Co., Carpintiera, CA), and blocked with 1% BSA for 1 h at room temperature. After an overnight incubation at 4 °C with 1 : 50 antihuman clusterin goat polyclonal IgG (Santa Cruz), sections were covered with 1 : 200 horseradish peroxidase-conjugated antigoat IgG (Santa Cruz). Peroxidase activity on sections was detected by immersion in freshly mixed 3–3′ diaminobenzidine tetrachloride (0.6 mg/mL; Sigma) and 0.003% hydrogen peroxide for 5 min. The sections were then counter-stained with haematoxylin, dehydrated and cleared, and placed under a coverslip.
In vivo treatments
About 1 × 106 of either LNCaP/P or /T cells were inoculated subcutaneously, with 0.1 mL of Matrigel (Becton Dickinson) into the flank of 6–8-week-old male athymic mice under halothane anaesthesia (5% induction and 1.5% maintenance). When the tumours grew to 1.5 cm, usually 6–8 weeks after injection, the mice were irradiated with an X-ray machine at 250 kVp using a 0.5-mm copper filter. The device was fitted with a jig to provide shielding for the animal's body and support for the special animal restraint, which comprised a lead shell with an opening to allow for exposure of the tumour protruding from the animal's flank. Tumours received a total irradiation of 15 Gy in five equal doses delivered on five consecutive days. For immunohistochemical and Northern blot analysis, two tumours of each experimental group were harvested 5 days after the last irradiation. All animal procedures were performed according to the guidelines of the Canadian Council on Animal Care and with appropriate institutional certification.
Numerical data are expressed as the mean (sem), with the statistical significance of differences between values from the different experimental treatments determined using Student's t-test, and P < 0.05 (two-sided) taken to indicate statistically significant differences.
Western blot analysis and immunohistochemistry were used to compare clusterin expression levels in LNCaP/P and /T cells in vitro and in vivo. As shown in Fig. 1A, abundant levels of both unprocessed (60 kDa) and mature (40 kDa) forms of clusterin protein were detected in LNCaP/T, while LNCaP/P did not express detectable clusterin protein levels. Immunohistochemistry was used to confirm clusterin overexpression of LNCaP/T tumours in vivo. Whereas LNCaP/P tumours expressed no detectable clusterin (Fig. 1B), there was strong cytoplasmic clusterin staining (dark reaction product) in LNCaP/T tumours (Fig. 1C).
Northern blot analysis was used to determine the effects of radiation (15 Gy) on clusterin mRNA expression in both LNCaP/P and /T tumours. As shown in Fig. 2, clusterin mRNA expression increased after radiation from undetectable to low levels in LNCaP/P tumours and more than three-fold in LNCaP/T tumours.
The MTT and re-growth assays were used to compare changes in growth rates of LNCaP/P and /T cells up to 3 weeks after radiation with 4 Gy. As shown in Fig. 3A (MTT assay), clusterin overexpression significantly increased cell viability after radiation (P < 0.05, mean values after three experiments). To investigate the duration of this effect, re-growth assays were used to observe cells over several weeks. As shown in Fig. 3B, increased radioresistance of LNCaP/T cells lasted > 3 weeks (P < 0.05, mean values after three experiments).
Western blot analysis was used to measure differences in cyclin-D1 expression in LNCaP/P and /T cells before and 3 days after irradiation with 4 Gy. Cyclin-D1 activates the kinase activity of G1 cyclin-dependent kinases in the cell cycle of proliferating eukaryotic cells . Whereas cyclin-D1 protein expression levels were equal in unirradiated LNCaP/P and /T cells, its expression was three times more after irradiation in LNCaP/T than in LNCaP/P cells, paralleling changes in radiation-induced growth rates (Fig. 4).
Flow cytometry was used to analyse the DNA profile of LNCaP/P and /T cells before and after irradiation. The cell-cycle distribution of both cell lines was not significantly different before radiation (Fig. 5A) . Three days after irradiation with 8 Gy, the proportion of cells undergoing apoptosis (sub G1–G0) was significantly lower in LNCaP/T (23%) than in LNCaP/P cells (50%; P < 0.05, after triplicate analysis using Student's t-test; Fig. 5B).
A better understanding of the molecular factors that determine the radiosensitivity of prostatic tumour cells will enable the optimization of radiation therapy for treating localized prostate cancer. The ability of radiation to induce apoptosis might be important in ultimately determining responsiveness. Kyprianou et al. showed that overexpression of a single-cell survival gene (bcl-2) significantly increased radiation resistance in human prostate cancer by delaying the onset of radiation-induced apoptosis, and clinical studies have confirmed bcl-2 as a predictive marker for the response to radiation [26–28]. Further research into factors regulating pro- and anti-apoptotic mechanisms in prostate cancer cells in response to ionizing radiation will enable strategies to be directed at influencing this critical balance.
One method to identify potential therapeutic targets associated with radiation resistance is to compare gene-expression in human prostate cancer cells before and after irradiation. The present in vivo results show the survival gene clusterin to be highly upregulated after irradiation in both LNCaP/P and /T prostate tumours. Considering the increasing evidence suggesting that greater clusterin expression in tumour cells is protective against apoptotic cell death [19,20], we hypothesized that clusterin overexpression may influence the apoptotic response of prostate tumour cells to radiation. The role of clusterin in apoptosis has been debated, with recent data suggesting a pro-apoptotic rather than the anti-apoptotic role after radiation in human MCF-7 breast cancer cells . However, the present results, and our previous studies [19,30–32], suggest that clusterin is protective against apoptosis induced by various stimuli. The seemingly paradoxical roles for clusterin in apoptosis are perhaps analogous to those ascribed to two forms of bcl-x that arise from alternative splicing [33–35]. Clusterin can exist in several molecular forms, although they apparently arise from post-translational modifications of a single mRNA transcript [36,37].
To clarify the functional role of clusterin in radiation, we compared the radiosensitivity of LNCaP/P and LNCaP/T cells. After radiation the viability of LNCaP/P cells decreased by more than half, as determined by the MTT assay. This result is consistent with flow cytometric analysis showing double the proportion of radiation-induced apoptotic cells in LNCaP/P cells. Furthermore, LNCaP/P cells expressed three-fold lower levels of cyclin D1-protein as a consequence of decreased proliferation activity. Finally, the relative numbers of LNCaP/P cells remained significantly lower over 3 weeks after radiation. These results indicate that overexpression of clusterin decreases radiation sensitivity by inhibiting apoptosis not only in a delayed but also an extended fashion.
In conclusion, collectively the present results show that overexpression of clusterin further increases the given resistance of LNCaP cells to radiation-induced apoptosis. This provides additional support for the hypothesis that upregulation of clusterin after various apoptotic signals represents an adaptive cell-survival mechanism, thereby offering a potential target for downregulation to improve the results of radiation therapy for patients with prostate cancer.
We thank Mary Bowden and Howard Tearle for their excellent technical assistance, and Dr Lawrence Meyer and Visia Dragowska (Department of Molecular Sciences, British Columbia Cancer Agency, Vancouver, Canada) for their support and technical assistance in flow cytometry. This work was supported by a grant from the Prostate Research Foundation of Canada. T.Z. was supported by the Swiss National Science Foundation, Krebsliga Basel, Lichtenstein-Stiftung and Freiwillige Akademische Gesellschaft of the University of Basel, Switzerland.
LNCaP/T, parental or transfected LNCaP cells.