Silibinin synergizes with mitoxantrone to inhibit cell growth and induce apoptosis in human prostate cancer cells
Version of Record online: 17 JAN 2007
Copyright © 2006 Wiley-Liss, Inc.
International Journal of Cancer
Volume 120, Issue 9, pages 2028–2033, 1 May 2007
How to Cite
Flaig, T. W., Su, L.-J., Harrison, G., Agarwal, R. and Glodé, L. M. (2007), Silibinin synergizes with mitoxantrone to inhibit cell growth and induce apoptosis in human prostate cancer cells. Int. J. Cancer, 120: 2028–2033. doi: 10.1002/ijc.22465
- Issue online: 28 FEB 2007
- Version of Record online: 17 JAN 2007
- Manuscript Accepted: 11 OCT 2006
- Manuscript Received: 18 AUG 2006
- NIH. Grant Numbers: 5P30CA46934, NRSA T32 CA079446-06A1
- prostate cancer
The purpose of these experiments was to assess the synergistic activity of silibinin with chemotherapy agents in clinical use against prostate cancer. Silybin-phytosome, a commercially available formulation containing silibinin, has recently been studied in a phase I clinical trial. The silibinin doses used in the present study are clinically achievable based on the preliminary phase I data. DU145, PC-3 and LNCaP prostate cancer cells were seeded in 96-well plates in triplicate. Twenty-four hours later, silibinin (10, 20 and 40 μM) and either mitoxantrone or docetaxel were added to the designated wells. Seventy-two hours post-treatment, cell viability was determined with a tetrazolium-based assay. The combination index (CI) for determination of a synergistic effect was calculated, with values of <0.9 indicating synergy and values >1.1 antagonism. Apoptosis was also assessed using a luminescent assay after 72 hr of treatment with media alone, silibinin, mitoxantrone, or silibinin plus mitoxantrone. Silibinin showed a synergistic effect with mitoxantrone, as measured by reduction in cell viability. The CI values ranged from 0.413 to 2.650 for the combination of silibinin and mitoxantrone; in contrast, treatment with docetaxel and silibinin showed little or no synergy, with CI values of 0.898–4.469. In concordance with these findings, the addition of silibinin increased the level of apoptosis compared to mitoxantrone alone, particularly in the PC-3 cells. The combination of silibinin and mitoxantrone exhibits a pattern of synergy in reducing cell viability with increased apoptosis. These data are important in the planning of future clinical applications of silibinin. © 2007 Wiley-Liss, Inc.
Prostate cancer is the most common non-skin cancer diagnosed in men. In 2005, the American Cancer Society estimates that 232,000 men were diagnosed and ∼30,000 died from prostate cancer in the United States.1 Early stage or organ confined cases may be cured with surgical prostatectomy or radiation therapy, although, some low-risk patients may also choose watchful waiting.2, 3 Disease that has spread beyond the prostate is generally first treated with hormonal ablation. The addition of cytotoxic chemotherapy is used in selected hormone-refractory patients, with this approach demonstrating a small (∼2 month) survival advantage.4, 5
Milk thistle (Silybum marianum) has a long history of use in humans and is commonly used in the treatment of liver disease.6, 7 Silymarin, the name of the crude milk thistle extract, is composed of several stereoisomers including silibinin (or silybin), silychristin and silydianin.8 Silibinin and silymarin have been shown to inhibit in vitro cell growth in several cancer models including prostate,9, 10, 11 bladder,12, 13 colon,14, 15 breast,16, 17 cervical,18 skin19, 20 and lung cancer.21, 22 Additional in vivo experiments have demonstrated an anti-cancer effect of silibinin and silymarin in models of prostate,23 skin,24, 25 bladder26 and colon cancer.27
Silibinin's anti-neoplastic actions appear to work through several different pathways.28 One of the most prominent effects seen in preclinical studies of silibinin is G1 arrest and apoptosis,29 with an increase of the cyclin dependent kinase inhibitors kip1/p27 and cip1/p21.17 Further study has shown that silibinin causes decreased phosphorylation of retinoblastoma protein leading to stability of the complex formed with E2F.9, 30 Silibinin's effect is accompanied by an inhibition of erbB131 with reduced ligand binding to this receptor demonstrated in prostate cancer cell lines.32 Using an in vivo model, silibinin increases insulin-like growth factor binding protein 3 levels after dietary feeding in a murine xenograft model of human prostate cancer.23 Transcription factor NF-κB, an important component in cell survival and chemotherapy resistance, is also inhibited by silibinin.11, 33, 34
Silibinin increases the efficacy of several chemotherapy agents both in vitro and in vivo. It acts synergistically with doxorubicin to inhibit growth via apoptosis in the human prostate cancer DU145 cells,35 and sensitizes these same cells to the anti-neoplastic effects of cisplatin and carboplatin.36 Silibinin also synergizes with doxorubicin to inhibit growth in an in vivo xenograft model of lung cancer.37
In addition to these preclinical studies with silibinin, we have conducted a Phase I human clinical trial of high-dose oral Silybin-Phytosome™. Preliminary results of this study have been reported in abstract form, indicating that at daily doses of 5 to 20 g, peak blood concentrations of >100 μM were achieved in some patients from 30 to 60 min after dosing.38 This high-dose regimen is in contrast to a recent pharmacokinetic study of standard dose silibinin (1.4 g daily) in colorectal cancer patients, which demonstrated that measurable tissue levels were achievable at that dosing level, with a terminal plasma half-life of 3.4 hr.39
Although, preclinical data would support the use of silibinin in combination with doxorubicin, cisplatin and carboplatin, these agents are not commonly used in the treatment of prostate cancer. Docetaxel, a microtubule inhibitor, is currently a first-line chemotherapy agent in advanced prostate cancer.4, 5 Mitoxantrone, an anthracenedione and topoisomerase II inhibitor, is also Food and Drug Administration approved for prostate cancer and in clinical use. Thus, both docetaxel and mitoxantrone were selected for further study based on their clinical relevance in prostate cancer. The experimental results presented here are essential in planning for future human clinical trials of silibinin in combination with cytotoxic chemotherapy in the treatment of prostate cancer, in addition to furthering our understanding of silibinin's mechanism of action.
Material and methods
Cell culture and reagents
Human prostate cancer cells DU145, LNCaP and PC-3 (American Type Culture Collection, Manassas, VA) were grown in OptiMEM (Gibco, Grand Island, NY). Ten percent Fetal Bovine Serum (Gemini, Woodland, CA) and 1% streptomycin-penicillin sulfate (Life Technologies, Grand Island, NY) were added to the media. All cell lines were incubated at 37°C with 5% CO2. Silibinin and docetaxel (both from Sigma, Saint Louis, MO) were dissolved in Dimethyl Sulphoxide (DMSO) and ethanol respectively; mitoxantrone (Serono, Rockland, MA) was obtained in solution.
Cell viability assay
Cell viability was assessed using a tetrazolium-based assay (CellTiter 96 AQueous One Solution-Promega Corporation, Madison, WI). One thousand cells in 50 μl of media per well were plated in 96-well plates in triplicate using the 3 cell lines. Twenty-four hours after plating, the combination treatments were added with 25 μl of 4× silibinin and 25 μl of 4× chemotherapy (either docetaxel or mitoxantrone) to give a total volume of 100 μl in each well. DMSO, in equal amounts to the treatment conditions, was added to the media in the control condition. Seventy-two hours after the treatment, 20 μl of the AQueous One Solution was added to each well. Colorimetric analysis using a 96-well plate reader (Vmax Kinetic Microplate Reader, Molecular Devices, Sunnyvale, CA) was performed between 1 and 4 hr (wavelength of 490 nm), depending on cell type and cell density. Cell viability assays were performed in triplicate. Statistical analysis was prepared with the SPSS software program (13.0 for Windows, 2004-Chicago, IL).
Determination of combination index
Cell viability data were analyzed with the CalcuSyn software program (Biosoft, Ferguson, MO-version 2-2005), which utilizes the statistical method described by Chou and Talalay to calculate the combination index (CI) values.40 Using a non-constant ratio setting, the CI values were calculated for each condition.
Using a 96-well plate system, 5,000 cells from each of the 3 cell lines were seeded in 50 μl of media, as above, in triplicate. After 24 hr, 50 μl of media, silibinin and mitoxantrone were added to create 4 treatment conditions: media alone, silibinin (20 μM), mitoxantrone (50 nM) and silibinin plus mitoxantrone (20 μM and 50 nM respectively). After 72 hr, caspase-3 and caspase-7 activity was assessed by adding 100 μl of Caspase-Glo 3/7 (Promega, Madison, WI) solution to each well. Approximately 45 min later, the plates were placed in a Victor 1420 multilevel counter (Wallac, Gaithersburg, MD) for analysis. The cell viability for each treatment condition was assessed concurrently as described, with the apoptosis results corrected for differences in viability and the corresponding loss of luminescence from the non-viable cells.
Cells from each of the 3 cell lines (5 × 104 in 500 ml of OptiMEM, with 10% Fetal Bovine Serum and 1% streptomycin-penicillin sulfate) were plated into 12-well plates. Twenty-four hours later, the cells were treated with silibinin (20 μM final concentration), mitoxantrone (50 nM final concentration), and DMSO in the control condition. After 72 hr, the cells were stained with propidium iodide and bisbanzimide (both from Sigma, St. Louis, MO) giving a final concentration of 0.1 μg/ml to visually assess apoptosis; a Nikon Eclipse, TE 2000-S microscope (Nikon, Melville, NY) was used for visualization at 200× power.
Short duration silibinin treatment
The effect of short duration silibinin treatments was also examined. DU145, LNCaP and PC-3 cells were seeded (3,000/well) in a 96-well plate in 50 μl of media as above. Twenty-four hours later, 50 μl of media containing mitoxantrone was added to each well, for a final mitoxantrone concentration of 50 nM. After 2 hr, the media was removed and quickly replaced with either the standard media or silibinin containing media (20 μM). After an additional 2 hr, the media was again removed from all wells and replaced with fresh media. After 72 hr these treatment started, cell viability was again assessed by adding 20 μl of AQueous One Solution and utilizing the plate reader, as above.
Synergy between silibinin and mitoxantrone in prostate cancer cell lines
The effect of silibinin and clinically relevant chemotherapy treatments on cell viability in prostate cancer cells was examined first. Cell viability for DU145, LNCaP and PC-3 cells treated with 50 nM mitoxantrone and varying doses of silibinin is shown in Figure 1. While both DU145 and LNCaP cells showed decreased viability with combination treatment, PC-3 cells exhibited minimal change in viability. In contrast to the effect observed with silibinin and mitoxantrone in DU145 and LNCaP cells, the combination of silibinin (10, 20 and 40 μM) and docetaxel (2.5–5 nM) showed little if any enhancement of docetaxel's efficacy in all 3 cell lines (data not shown).
Next, the synergistic and antagonistic effects of silibinin (doses of 10, 20 and 40 μM) in combination with docetaxel (doses of 0.62, 1.25, 2.5 and 5 nM) or mitoxantrone (doses of 25, 50, 100 and 200 nM) in DU145, LNCaP, and PC-3 were quantified. These dose ranges were selected to approximate the IC50 of docetaxel and mitoxantrone, with the silibinin dose based on the clinically achievable levels from preliminary phase I human data. Using AQueous One Solution in 96-well plates, cell viability was assessed 72 hr after the treatments. The CI values are on a continuum with respect to synergy, with values less than 0.9 consistent with synergy and values greater than 1.1 showing antagonism for a given combination. The mitoxantrone/silibinin treatment demonstrated significant synergy with CI values of 0.515–0.929, 0.521–0.967 and 0.413–2.65 for DU145, LNCaP and PC-3 cells respectively (Fig. 2). The only clear antagonism seen in the mitoxantrone combination was with low doses of mitoxantrone, 25 nM, a dose at which mitoxantrone's effect is small either with or without silibinin.
In agreement with the small or neutral effect noted in the initial viability studies, the docetaxel/silibinin treatments were largely devoid of synergy with CI values of 0.898–2.54, 0.921–2.32 and 0.895–4.47 for DU145, LNCaP and PC-3 cells respectively (Fig. 2). As evidenced by CI values of greater than 0.9, a neutral or antagonistic pattern was observed in all 3 cell lines with this combination.
Silibinin in combination with mitoxantrone induces apoptosis
Other investigations of silibinin have pointed to the induction of apoptosis as an important element of its anti-cancer efficacy. To further evaluate the mechanism of mitoxantrone/silibinin synergy, an assessment of the caspase-3 and caspase-7 levels were performed. The results of these apoptosis experiments are reported as fold-change in activity compared to control, after normalizing for differences in cell viability at the time of caspase activity assessment (Fig. 3). In all 3 cell lines, treatment with the relatively low dose of silibinin (20 μM) alone had no effect on the level of apoptosis over control. Consistent with previously published data, treatment with silibinin alone at higher doses, i.e. greater than 50 μM, did produce apoptosis (data not shown). Treatment with mitoxantrone alone did result in apoptosis in DU145 and LNCaP cells (3.84 and 1.79 times control, respectively), but not in the PC-3 cells (Fig. 3). The combination of silibinin and mitoxantrone increased the relative apoptosis in all cell lines, revealing a relative increase of 6.30, 2.18 and 2.23 fold over controls in the DU145, LNCaP and PC-3 cells, respectively (Fig. 3). To confirm silibinin's sensitization of PC-3 cells to mitoxantrone-induced apoptosis, the cells from these experiments were stained and visually observed for apoptosis. Using propidium iodide and bisbenzimide stains, the cells were assessed at 72 hr after treatment with silibinin (20 μM) and mitoxantrone (50 nM). The cell staining showed concordant visualization of the apoptosis measured by the caspase activity assay (Fig. 4).
Short duration treatments with silibinin show enhancement of mitoxantrone's activity
Since the half-life of oral silibinin in humans is short,39 the in vitro efficacy of short duration treatments of silibinin was studied. First, the cells were treated with mitoxantrone for 2 hr, after which time the media was replaced with either standard media or media containing silibinin (20 μM). After 2 additional hours, the media was again exchanged in all wells for standard media. Treatment with 2 hr of silibinin after mitoxantrone yielded a significant reduction in cell viability in all cell lines (Fig. 5). Silibinin's enhancement of mitoxantrone's activity with this treatment regimen was most prominent in the PC-3 cell line, where a 30% decrease in cell viability was observed (p < 0.001, Student's t test).
Silibinin synergistically decreased prostate cancer cell viability in combination with mitoxantrone; this pattern of synergy was seen in all 3 human prostate cancer cell lines examined. This was observed with an increase in apoptosis, as measured by active caspase-3 and caspase-7 levels. In contrast, silibinin did not act synergistically with docetaxel in these prostate cancer cells.
These data add to our understanding of silibinin's mechanism of action. Previous studies have shown that silibinin enhances the effects of several chemotherapy agents, all of which are DNA-damaging agents.35, 36 Of these, cisplatin and carboplatin act similarly to alkylating agents, binding and cross-linking DNA, and doxorubicin is a topoisomerase II inhibitor, contributing to DNA breakage. In the present study, it is notable that mitoxantrone inhibits topoisomerase II, in a similar manner to doxorubicin, whereas docetaxel binds to microtubules and inhibits cell division, without a known DNA-damaging effect. Recent data using mouse epidermal keratinocyte JB6 cells and ultraviolet light B have given insight into silibinin's activity with DNA-damaging agents.41 In these experiments, the addition of silibinin increased the percentage of apoptotic cells when compared to the ultraviolet light B treatments alone. Silibinin's enhancement of ultraviolet light B-induced apoptosis was seen with an upregulation of DNA protein kinase, an important element in sensing genotoxic stress, with this effect abrogated by the addition of a semi-selective inhibitor of DNA protein kinase. Considering the present data in this context, consistency is seen in the synergy of silibinin with the DNA-damaging agent mitoxantrone.
Previous work has highlighted the potential importance of p53 activation in silibinin-induced apoptosis.41, 42 Of the 3 cell types examined here, LNCaP cells are considered to have wild type p53, PC-3 cells have a deletion in codon 138 with negative immunohistochemical staining for p53, and DU145 cells have 2 mutations in codons 223 and 274 of p53, but stain positive for p53 expression.43 Notably, the PC-3 cells showed little apoptosis in response to either low-dose silibinin or mitoxantrone; however, the combination demonstrated a significant increase. Silibinin therefore appears to overcome the relative resistance of PC-3 cells to mitoxantrone-induced apoptosis. The role of p53 status in silibinin's synergy in prostate cancer cells is an area worthy of additional study.
Because of the short half-life of silibinin observed in phase I testing, the effect of short duration silibinin treatment was examined, as this is believed to be clinically relevant. The short duration silibinin treatment experiments presented here were designed to better simulate the short half-life of oral silibinin in humans40 and demonstrated the potential relevance of a 2 hr silibinin exposure. Unlike the initial experiments with 72 hr of concurrent treatment (Fig. 1), PC-3 cells had decreased viability with the short duration, sequential treatment with mitoxantrone followed by silibinin (Fig. 5). One explanation for this observed finding is that mitoxantrone treatment for 2 hr induces DNA damage, but little cell death. The addition of silibinin enhances the genotoxic stress response, yielding decreased cell viability, even with a 2 hr treatment.
Most prostate cancer patients currently requiring chemotherapy are started on docetaxel therapy, as this treatment has shown a survival advantage.4, 5 The best treatment for progressive disease after docetaxel treatment is not well defined, but mitoxantrone is approved for use in prostate cancer and employed in this setting. The preclinical data presented here show significant synergy for the combination of mitoxantrone and silibinin in these in vitro models of prostate cancer. With a favorable side effect profile and long history of human use, silibinin represents a good candidate for combination treatments. If validated in further in vivo experiments, this combination could be considered in a clinical trial for patients with progressive prostate cancer, despite docetaxel treatment.
In conclusion, silibinin synergizes with mitoxantrone in prostate cancer cell lines at doses that are clinically achievable in humans. This effect was seen with increased apoptosis, especially in PC-3 cells. Additionally, a pattern of benefit from combination treatment with mitoxantrone and silibinin was also seen with short duration treatments, mimicking the silibinin levels observed in preliminary human testing. As mitoxantrone is approved for use in prostate cancer, these data suggest that additional study of this combination is warranted in the treatment of prostate cancer.
- 7Milk thistle (Silybum marianum) for the therapy of liver disease. Am J Gastroenterol 1998; 93: 139–43., , , .Direct Link:
- 17Anticarcinogenic effect of a flavonoid antioxidant, silymarin, in human breast cancer cells MDA-MB 468: induction of G1 arrest through an increase in Cip1/p21 concomitant with a decrease in kinase activity of cyclin-dependent kinases and associated cyclins. Clin Cancer Res 1998; 4: 1055–64., , .
- 30Inhibition of retinoblastoma protein (Rb) phosphorylation at serine sites and an increase in Rb-E2F complex formation by silibinin in androgen-dependent human prostate carcinoma LNCaP cells: role in prostate cancer prevention. Mol Cancer Ther 2002; 1: 525–32., , .
- 38A phase I study of silibinin in hormone refractory prostate cancer. J Clin Oncol 2005; 23: 427S., , , , , .