Survivin gene silencing sensitizes prostate cancer cells to selenium growth inhibition

MSA and MSC were obtained from PharmaSe (Lubbock, TX). The rabbit anti-human polyclonal antibodies against survivin and caspase 3 were from Santa Cruz Biotech (Santa Cruz, CA). The rabbit anti-human Ki-67 polyclonal antibody was obtained from Neomarker (Fremont, CA). Trizol, RPMI 1640, fetal bovine serum (FBS), and lipofectamine 2000 were purchased from Invitrogen (Carlsbad, CA). Acridine orange (AO) and ethidium bromide (EB) were from Sigma (St. Louis, MO).

Prostate cancer specimens were obtained from patients undergoing prostatectomy in the Second and the Third Affiliated Hospitals of Jilin University. Normal prostate tissues were obtained from patients undergoing surgery for benign prostatic hyperplasia (BPH) in these hospitals. This study was approved by the Research Ethics Committee of Norman Bethune College of Medicine, Jilin University and was in compliance with the Helsinki Declaration. The tissues were examined by the pathologists to confirm the diagnosis before IHC analyses. All specimens were fixed in 10% formalin, embedded in paraffin, and cut into 4 μm-thick slides. The slides were dewaxed, and the endogenous peroxidase activity was blocked by treatment with 3% hydrogen peroxide solution in methanol for 20 min. Epitope retrieval was performed by treating the slides with 10 mM sodium citrate buffer (pH 6.0) and heating in a microwave oven for two times at the high power for 6 min each. Non-specific binding was prevented by blocking with normal goat serum (1:10) for 10 min. Immunostaining of survivin was performed using a rabbit anti-human survivin polyclonal antibody. Goat anti-rabbit IgG conjugated with horseradish peroxidase was used as the secondary antibody. The staining procedure was carried out manually at room temperature, using an avidin-biotin-peroxidase complex method. The presence of survivin and Ki-67 was evaluated by staining with rabbit anti-survivin and anti-Ki-67 antibody, respectively. After incubation with the primary antibody for 60 min, the slides were incubated with the biotinylated goat anti-rabbit IgG (H+L) (DAKO, Carpinteria, CA) at 37°C for 30 min, followed by incubation with a 1:200 streptavidin-biotin-peroxidase complex (Sigma, St. Louis) for 30 min. Reactive products were visualized with 3,3'-diaminobenzidene (DAB) as the chromogen, and the slides were counterstained with hematoxylin and coverslipped. Sections previously known to express survivin and Ki-67 were included in each run, receiving either the primary antibody as the positive control, or a mouse IgG as the negative control. The stained slides were analyzed with a microscope at 200× magnification. The slides were examined by two independent pathologists who had no knowledge of the clinical outcomes and of the results obtained by the other. Differences in interpretation were reconciled by reviewing the slides jointly. Cellular brownish staining was scored as positive and the threshold was set at 10%. The proliferation index was defined as percent of tumor cells stained positively for Ki-67.

Oligonucleotides for anti-survivin shRNA expression were designed, synthesized, and cloned into the pGCsilencer-U6/Neo/GFP plasmid (GeneChem Co, Shanghai, China). The sequences are: sense, 5'-gatcccGCAGTTTGAAGAATTAACCttcaagaga GGTTAATTCTTCAAACTGCtttttggat-3'; and antisense, 5'-agctatccaaaaaGCAGTTTGAAGAATTAACCtctcttgaaGGTTAATTCTTCAAATGCgg-3'. Sequences in uppercase letters indicate nucleotides 394-412 of the survivin cDNA. BLAST searches of the human genome database were carried out to ensure that the sequences would not target any other transcripts. The recombinant plasmid was named pGCsh-survivin (referred to as sh-survivin hereafter). Recombinant pGCsh-scrambled plasmid (referred to as sh-scrambled hereafter), which used a scrambled sequence that has no significant homology to any mouse or human gene sequences, was constructed as the negative control.

The PC-3M, C4-2B, and 22Rv1 cells were cultured in RPMI 1640 medium supplemented with 10% FBS, 100 unit/ml of penicillin, 100 μg/ml of streptomycin, and 2 mM of L-glutamine and incubated at 37°C in 5% CO₂. For transfection, 5 × 10⁵ cells were plated onto 6-well plates 24 hours before transfection. Transfection was performed using Lipofectamine 2000 according to the manufacturer's instructions. Transfection media were removed 6 hours after transfection and replaced with fresh complete medium containing 10% FBS. Controls included lipofectamine 2000 treated cells and sh-scrambled transfected cells.

The effects of survivin shRNA and MSA, alone or in combination, on cellular expression of survivin were measured by semi-quantitative RT-PCR and Western blot analysis. PC-3M cells were divided into 5 groups: (a) control (mock-transfected); (b) scrambled (transfected with sh-scrambled); (c) sh-survivin (transfected with sh-survivin); (d) MSA (treated with 5 μM MSA); (e) combination (transfected with sh-survivin and treated with 5 μM MSA), and cells were harvested after 66 hours following transfection and treatment. Total RNA was isolated using the Trizol reagent according to the manufacturer's instructions. Survivin gene expression was determined by semi-quantitative RT-PCR, using the primer pairs 5'-GAATTCATGGGTGCCCCGACGTTGCC-3' and 5'-AGATCTTTCTTCTTATTGTTGGTTTCC-3'. The housekeeping gene β-actin was amplified as an internal standard.

For Western blot analysis, cells were lysed with HEPES lysis buffer (30 mM HEPES, 1% Triton X-100, 10% glycerol, 5 mM MgCl₂, 25 mM NaF, 1 mM EDTA and 10 mM NaCl). Fifty micrograms of protein were electrophoresed on a 10% SDS-PAGE gel, transferred onto a PVDF membrane (Millipore, Bedford, MA), and probed with the anti-survivin or the anti-caspase-3 antibody. A goat anti-rabbit antibody labeled with horseradish peroxidase (Amersham) was added as the secondary antibody. Immunoblots were developed using a chemiluminescence detection system (Amersham Pharmacia Biotech).

Cell numbers were determined at 66 hours after treatment using the 3-(4,5-dimethylthiazol-2-yl)-2, 5-diphenyltetrazolium bromide (MTT) assay and quantified by using a microplate reader (Bio-Rad). For apoptosis, 10⁶ cells were fixed in 70% (v/v) ethanol, stained with PBS containing 50 μg/mL propidium iodide, 10 mg/mL RNase, and 0.1% (v/v) Tween-20 for 30 minutes at room temperature, and analyzed for cell apoptosis with a FACScan flow cytometer (Becton Dickinson). For AP/EB staining, PC-3M cells were collected by trypsinization, washed with PBS (pH 7.4), and fixed in 4% paraformaldehyde. Fixed cells were washed twice with PBS and resuspended in PBS containing 16 μg/ml AO and 16 μg/ml EB, then mounted on a glass slide and observed under a fluorescent microscope.

Five-week-old pathogen-free athymic nude mice were purchased from the Institute of Laboratory Animal Science, Chinese Academy of Medical Sciences (Beijing, China). PC-3M cells in exponential growth phase were harvested and single-cell suspensions (2 × 10⁶ cells in 100 μl PBS) were injected subcutaneously (s.c.) into the right dorsal flank of nude mice. Tumor size was measured every 2 to 3 days, and tumor volume calculated as 0.5236 × width ² × length. After palpable tumors had developed, the mice were divided randomly into five groups (six mice per group): (a) control (mock-transfected); (b) scrambled (transfected with sh-scrambled); (c) sh-survivin (transfected with sh-survivin); (d) MSC (given MSC at 100 μg/mouse/day); (e) combination (transfected with sh-survivin and given MSC). Transfection of the siRNAs was performed by an electroporation method. Twenty micrograms of plasmid suspended in 50 μl PBS was injected percutaneously into the tumors by using a syringe with a 27-gauge needle. Immediately after injection, tumors were pulsed with an electroporation generator (ECM830, BTX, Holliston, MA). Pulses were delivered at a frequency of 1 pulse/sec, 150 V/cm, for 50 milliseconds. This process was performed twice, first on day 15 and then on day 25. For selenium treatment, MSC was given by daily intragastric administration starting at day 15. Mice were sacrificed on day 35 when tumor burden in the control group approaching 10% of body weight. Tumors were excised, fixed in 4% buffered formalin, and embedded in paraffin. The animal study was conducted following internationally recognized guidelines and was approved by the Animal Research Committee of Norman Bethune College of Medicine, Jilin University.

Apoptotic cells in tissues were determined by using the TUNEL assay (Roche Diagnostics, Indianapolis, IN) according to the manufacturer's instructions. Briefly, after deparaffinization and rehydration, sections were incubated with proteinase K (16.2 μg/mL in 10 mM Tris·HCL, pH 7.4) for 20 min at 37°C. Slides were rinsed with PBS and incubated with 3% H₂O₂ in methanol for 10 min at room temperature to block endogenous peroxidase activity, followed by PBS washing and incubation in 0.1% Triton X-100 in 0.1% sodium citrate for 2 min on ice. Sections were then incubated with a mixture of terminal deoxynucleotidyl transferase (TdT) solution and biotin-dUTP solution in a humidified chamber at 37°C for 60 min, followed by washing with PBS and incubation with peroxidase-conjugated streptavidin in a humidified chamber for 30 min at room temperature. After an additional wash with PBS, the slides were incubated in DAB and counterstained with hematoxylin. Paraffin-embedded sections of normal tonsils were used as positive control. Negative control was obtained by replacing the TdT solution with distilled water. The presence of clear nuclear staining was indicative of apoptotic cells. At least 1,000 tumor cell nuclei were randomly selected and examined. The number of TUNEL-positive tumor cell nuclei was counted and the apoptotic index was calculated as follows: apoptotic index = (number of apoptotic cells/total cell number) × 100%.

Chi-square test was used to analyze the difference in survivin expression between normal and cancerous prostate tissues. Pearson correlation analysis was performed to test the correlation of Gleason Score and survivin expression in the prostate cancer cases. Data were presented as mean ± standard deviation (SD) when appropriate. Two-tailed Student's t-test was used to evaluate the differences between treatment and control values, and P < 0.05 was considered statistically significant.

We first analyzed the expression of survivin by IHC staining in a collection of prostate cancer and normal tissues in a Chinese cohort. As shown in Fig. 1, the survivin protein was virtually undetectable in normal prostatic tissue but highly expressed in prostate cancer. The cellular localization of survivin is mostly cytoplasmic, but some weak staining can be found in the nucleus. Overall, survivin was expressed in 24 of 28 (85.7%) prostate cancer specimens, and no survivin expression could be detected in 22 normal prostate tissues (Table 1). The difference is statistically significant (P < 0.001). Pearson correlation analysis identified a strong correlation between Gleason score and survivin expression among the cases (R = 0.7446, P < 0.0001). Please see additional file-1 for the original data used to perform this analysis. These results confirmed previous findings that survivin is expressed in clinical prostate cancers [5], suggesting survivin is a promising target for prostate cancer therapy.

Bild vergrößern

Since MSA has been shown to induce apoptotic cell death in a number of prostate cancer cell lines [15, 16], we set out to examine the effect of MSA on apoptosis in PC-3M cells. PC-3M is a metastatic subline derived from PC-3. We chose PC-3M because it has the highest level of survivin expression among the cell lines we have tested (data not shown). Cells were treated with 5 μM MSA for 48 and 66 hr and the extent of apoptosis was quantified by flow cytometric analysis of cells labeled with propidium iodide (PI). This dose of selenium has been shown to be achievable in human subjects taking selenium supplementation [17]. As shown in Fig 2, only 0.9% of the control cells were positively labeled with PI. In cells treated with MSA, the proportion of apoptotic cells increased to 3.6% by 48 hr, and to 14.7% by 72 hr. These results showed that MSA was capable of inducing apoptosis in PC-3M cells, but the effect was modest and a long treatment (>48 hr) was needed to achieve a significant induction of apoptosis.

Bild vergrößern

As a member of the IAP family, the best known function of survivin is to protect the cells from undergoing apoptosis. It has been shown that selenium decreased survivin expression in several prostate cancer cell lines, including LNCaP, PC-3, and C4-2 [18]. We decided to examine survivin expression in PC-3M cells following MSA treatment. In contrast to what have been observed in other cell lines, both semi-quantitative RT-PCR and Western blotting analysis showed that treatment with 5 μM MSA for 66 hr did not influence survivin expression in PC-3M cells (Fig. 3A & 3B), indicating the induction of apoptosis by MSA in these cells is not mediated by decreasing survivin expression. In addition to PC-3M cells, we analyzed survivin expression in C4-2B and 22Rv1 prostate cancer cells following MSA treatment and the results were similar (Fig. 3B).

Bild vergrößern

Based on the data from the previous experiments, we hypothesized that survivin knockdown will sensitize prostate cancer cells to growth inhibition by MSA. To test this hypothesis, we designed a plasmid-based shRNA vector for survivin gene silencing and tested its efficacy. As shown in Fig. 3A, sh-survivin effectively reduced the level of the survivin transcript, whereas the scrambled shRNA had no effect. Western blotting analysis confirmed the down-regulation of survivin expression by sh-survivin in all 3 cell lines (Fig. 3B). Quantitation by volume densitometry showed the knockdown efficiency at the protein level is in the range of 60-70% in these cell lines.

We next quantitated changes in cell viability after cells were transfected with the shRNAs and treated with various concentrations of MSA for 66 hr. The results are shown in Fig. 4. Cells transfected with sh-scrambled have similar viability as the mock-transfected control cells under all circumstances. Silencing of survivin reduced cell viability by 50-60% (P < 0.01). In the absence of sh-survivin, MSA reduced the cell viability in a dose-dependent manner in all three cell lines. However, cells transfected with sh-survivin are significantly more sensitive to MSA growth inhibition, manifested by the marked downward shift of the dose-response curves. In addition, cells transfected with sh-survivin and treated with MSA, the decrease in cell viability was significantly greater than sh-survivin alone (P < 0.01).

Bild vergrößern

Next, we analyzed the effect of apoptosis induction by sh-survivin and MSA in PC-3M cells by flow cytometry. Transfection and MSA treatment were carried out as described in the previous section. Very few PI-positive cells were found in untreated cells or in cells transfected with sh-scrambled (Fig. 5A). Silencing of survivin resulted in a robust increase of apoptotic cells, confirming the antiapoptotic role of survivin. The combination of survivin knockdown and MSA induced apoptosis more effectively than either alone (P < 0.01). In addition, AO/EB staining was performed to visualize the apoptotic cells and similar results were obtained (see additional file-2 for detail).

Bild vergrößern

Caspases are important mediators of apoptosis. Activation of caspase-3 by enzymatic cleavage is a hallmark event in caspase-dependent apoptosis. We investigated caspase-3 activation by Western blotting and the results are shown in Fig. 5B. Cleaved caspase-3 (19 KDa) was induced similarly by sh-survivin and MSA, whereas their combination led to a modestly but significantly stronger activation (P < 0.05). The changes in caspase-3 activation did not involve changes in gene expression, as the band intensity for procaspase-3 (32 KDa) was not changed by either sh-survivin, or MSA, or their combination. The modest increase in caspase-3 activation in the combined group suggest that the synergistic effect on apoptosis induction could also be mediated by caspase-independent mechanisms, since survivin has also been shown to inhibit both caspase-independent apoptosis [19]. In general, the apoptosis data are in agreement with the MTT results, showing that survivin gene knockdown greatly enhances the anticancer effect of selenium in prostate cancer cells.

A xenograft nude mouse tumor model with subcutaneously implanted PC-3M cells was used to investigate the effects of survivin knockdown and selenium in vivo. Fifteen days after injection, the tumors reached an average volume of 122.5 ± 15.36 mm³. The animals were then randomly divided into 5 groups receiving PBS, sh-scrambled, sh-survivin, MSC, or the combination of MSC and sh-survivin, respectively. MSC was administrated at a dose of 100 μg/mouse/day through an oral route and the shRNA plasmids were delivered by electroporation twice during the experiment, first at day 15 and then at day 25. As shown in Fig. 6A, tumor xenografts in the control and the scrambled groups grew at similar rates, whereas xenografts in all 3 treatment groups demonstrated reduced growth by day 30, when compared to their respective controls (P < 0.05). Consistent with the in vitro results, selenium by itself has a modest inhibitory effect on the growth of PC-3M xenografts, but the combination with survivin knockdown achieved a robust growth inhibition. In fact, tumors in the combination group maintained their original sizes after treatment, and were significantly smaller than those in the single treatment groups by day 35 (P < 0.05). Similar conclusions can be reached by examining the final tumor weight (Fig. 6B). These results demonstrate a strong inhibitory effect on tumor growth in vivo by the combination of survivin knockdown and selenium treatment.

Bild vergrößern

To explore the mechanisms of tumor inhibition in vivo, tumors were collected after the animals were sacrificed. The expression of survivin was first analyzed by IHC staining. As shown in Fig. 7A, the staining results showed that the electroporation-based delivery method was effective as survivin was down-regulated by sh-survivin. MSC has no effect on the expression or the cellular localization of survivin. Cell proliferation was quantitated by IHC staining of Ki-67 (Fig. 7B) and apoptotic cells were identified by the TUNEL assay (Fig. 7C). Tumors from the control or the sh-scrambled groups have proliferating cells (Ki-67 positive) and less apoptotic cells (TUNEL positive) than those in the sh-survivin or MSC. The lowest number of proliferating cells and highest number of apoptotic cells were found in tumors from the combination group. Quantitative analyses of the stained slides were performed and the results are summarized in Table 2. In tumors receiving sh-survivin or MSC, the proliferation index was significantly lower (P < 0.01), and the apoptotic index was significantly higher (P < 0.01), when compared with the control tumors. Once again, the combination group was more effective in blocking tumor growth and inducing cell death than either of the single treatment groups (P < 0.01). These results confirmed our in vitro observations and showed survivin knockdown greatly enhanced the efficacy of selenium in inhibiting the growth of prostate cancer cells in vivo.

Bild vergrößern