Fax: (510) 643-6791
Indole-3-carbinol induces a G1 cell cycle arrest and inhibits prostate-specific antigen production in human LNCaP prostate carcinoma cells
Article first published online: 29 OCT 2003
Copyright © 2003 American Cancer Society
Volume 98, Issue 11, pages 2511–2520, 1 December 2003
How to Cite
Zhang, J., Hsu B.A., J. C., Kinseth B.A., M. A., Bjeldanes, L. F. and Firestone, G. L. (2003), Indole-3-carbinol induces a G1 cell cycle arrest and inhibits prostate-specific antigen production in human LNCaP prostate carcinoma cells. Cancer, 98: 2511–2520. doi: 10.1002/cncr.11844
- Issue published online: 17 NOV 2003
- Article first published online: 29 OCT 2003
- Manuscript Accepted: 8 SEP 2003
- Manuscript Revised: 13 AUG 2003
- Manuscript Received: 16 JUN 2003
- NIH Public Health Service
- National Cancer Institute. Grant Number: CA69056
- California Department of Health Services. Grant Number: 99-00497V-10010
- prostate carcinoma cells;
- cell cycle control;
- growth inhibition;
- cyclin-dependent kinase;
- gene expression;
- protein kinase activity;
- prostate-specific antigen
Indole-3-carbinol (I3C), a naturally occurring component of Brassica vegetables, such as cabbage, broccoli, and Brussels sprouts, is a promising anticancer agent for certain reproductive tumor cells. The objective of the current study was to characterize the cell cycle effects of I3C in human prostate carcinoma cells.
The incorporation of [3H]thymidine and flow cytometry of propidium iodide–stained nuclei were used to monitor I3C-regulated changes in prostate carcinoma cell proliferation and cell cycle progression. Western blotting was used to document expression changes in cell cycle components and prostate-specific antigen (PSA) levels. The enzymatic activities of cyclin-dependent kinases (CDK) were tested by in vitro protein kinase assays using the retinoblastoma protein as a substrate.
I3C suppressed the growth of LNCaP prostate carcinoma cells in a dose-dependent manner by inducing a G1 block in cell cycle progression. I3C selectively inhibited the expression of CDK6 protein and transcripts and strongly stimulated the production of the p16 CDK inhibitor. In vitro protein kinase assays revealed the striking inhibition by I3C of immunoprecipitated CDK2 enzymatic activity and the relatively minor down-regulation of CDK4 enzymatic activity. In LNCaP prostate carcinoma cells, I3C treatment inhibited production of PSA, whereas combinations of I3C and the androgen antagonist flutamide more effectively inhibited DNA synthesis and PSA levels compared with either agent alone.
The results of the current study demonstrated that I3C has a potent antiproliferative effect in LNCaP and other human prostate carcinoma cells. These findings implicate this dietary indole as a potential chemotherapeutic agent for controlling the growth of human prostate carcinoma cells. Cancer 2003. © 2003 American Cancer Society.
Prostate carcinoma is the second leading cause of cancer-related death among men in the United States, after lung carcinoma, and accounts for approximately one-third of all male cancers.1 Although the discovery of prostate-specific antigen (PSA),2 a prostate-specific secretory protein, as an indicator of abnormal cell growth of the prostate gland2, 3 has improved the efficiency of early diagnosis of prostate carcinoma at the organ-confined stage,4, 5 the available treatments are limited. Because the growth of many prostate cancers, especially early-stage prostate carcinomas, is androgen dependent, several current therapies target the androgen supply to the tumor.1 These therapies, such as treatment with the antiandrogen flutamide and surgical removal of the testes, which are the predominant androgen-secretory glands, often lead to an initial inhibition of tumor development. However, in many cases, the tumors can overcome the endocrine blockage and resume proliferation independent of androgen.6 Therefore, an important goal is to identify new classes of potential therapeutic compounds for the prevention and treatment of prostate cancer.
Epidemiologic studies show that the consumption of phytochemical-rich plant foods, including whole grains, vegetables, and fruits, is highly associated with a reduced risk for certain cancers, including prostate carcinoma.7, 8 These findings suggest that dietary plants produce a largely untapped source of compounds with potent antiproliferative and/or anticarcinogenic properties.9 One such phytochemical is indole-3-carbinol (I3C), an autolysis product of glucobrassicin, a naturally occurring component of Brassica vegetables, such as cabbage, broccoli, and Brussels sprouts.10, 11 High doses of I3C in the diet of rodents greatly reduced the incidence of certain spontaneous and carcinogen-induced reproductive tissue tumors and other cancer types.12–16 This indole also tested positive as a chemopreventative agent in several short-term bioassays relevant to carcinogen-induced DNA damage, tumor initiation and promotion, and oxidative stress.17 The topical application of I3C can inhibit tumor incidence in a mouse skin carcinoma model,18 suggesting that direct exposure to I3C has potent chemotherapeutic properties in certain classes of tumors. We have established that the direct treatment of cultured human breast carcinoma cells with I3C potently inhibits the growth of either estrogen-responsive or estrogen-independent human breast carcinoma cells.19, 20 I3C mediates this antiproliferative effect by inducing a G1 cell cycle arrest through disruptions in the expression and activity of specific G1-acting cell cycle components.19–21 Although several studies have investigated the apoptotic effects of high concentrations of I3C in prostate carcinoma cells,22, 23 relatively little is known about the cell cycle effects that selectively control the antiproliferative and anticancer response of this indole.
Prostate carcinomas, as well as many other human tumors, display specific alterations in the expression and/or activity of individual cell cycle components.24–27 Extensive studies have shown the loss of expression or function of p21WAF1/CIP1 and p27KIP1 in primary prostate tumors as well as in cultured prostate carcinoma cells.28–31 Prostate carcinomas (primary or secondary, androgen sensitive or refractory) also can exhibit decreased or absent expression or mutation of the p53 tumor suppressor protein,1 which is a key transcriptional inducer of p21WAF1/CIP1.32 The increased expression of p16INK4 also was observed in tumor xenografts that are growth suppressed.33 Alterations in the expression of Rb, CDK2, and G1-acting cyclins also have been observed in many tumor types, including androgen-responsive and nonresponsive prostate carcinoma.25, 34–36 Therefore, chemotherapeutic agents that target cell cycle components could be an effective alternative or additive treatment to the androgen-targeting therapies for prostate carcinoma. In the current study, we characterized the effects of I3C, alone and in combination with an androgen antagonist, on the prostate carcinoma cell cycle and control of PSA production. The results implicate I3C as a potential chemotherapeutic agent for controlling the growth of human prostate carcinoma cells.
MATERIALS AND METHODS
Dulbecco modified Eagle medium (DMEM), fetal bovine serum (FBS), calcium-free and magnesium-free phosphate-buffered saline (PBS), trypsin-ethylenediaminetetraacetic acid, and the antibiotics penicillin and streptomycin were supplied by BioWhittaker (Walkersville, MD). [3H]thymidine (84 curies [Ci]/mmol) and [gamma-32P]adenosine triphosphate (ATP) (3000 Ci/mmol) were purchased from NEN Life Science Products (Boston, MA). I3C was obtained from Aldrich (Milwaukee, WI) and recrystallized in hot toluene before use. Flutamide was purchased from Sigma (St. Louis, MO). Antibodies for CDK2, CDK4, CDK6, p21WAF1/CIP1, p16, and cyclin D2 were purchased from Santa Cruz Biotechnology (Santa Cruz, CA). Antibodies for cyclin E, cyclin D1, and cyclin D3 were purchased from BD PharMingen (San Diego, CA), and the anti-PSA antibody was obtained from Dako (Carpinteria, CA). Mouse monoclonal anti-CDK2 was purchased from BABCO (Berkeley, CA). The sources of other reagents are either listed in the following sections or were of the highest purity available.
The LNCaP, MDA-PCa-2b, PC-3, and DU-145 human prostate carcinoma cell lines were purchased from American Type Culture Collection (ATCC; Manassas, VA). The androgen-responsive LNCaP prostate carcinoma cells originally were derived from a Lymph Node Carcinoma of the Prostate,37 and MDA-PCa-2b prostate carcinoma cells were derived from a bone metastasis of prostate carcinoma.38 Two androgen-nonresponsive cell lines were used: PC3 cells were derived from a human prostatic adenocarcinoma metastatic to the bone,39 and DU-145 prostate carcinoma cells were derived from a human prostate adenocarcinoma metastatic to the brain.40 LNCaP, DU-145, and PC3 cells were grown in DMEM supplemented with 30 mM 4-(2-hydroxyethyl)-1-piperazineethanesulfonic acid (HEPES), 10% FBS, 60 units/mL penicillin, and 60 units/mL streptomycin. The MDA-PCa-2b cells were cultured in Kaighn-modified Ham F12 medium with 2 mM L-glutamine that was modified by the ATCC to contain 1.5 g/L sodium bicarbonate. The medium was supplemented with 25 ng/mL cholera toxin, 10 ng/mL epidermal growth factor, 0.005 mM phosphoethanolamine, 100 pg/mL hydrocortisone, 45 nM selenious acid, 0.005 mg/mL insulin, and 20% FBS. All cells were maintained at 37 °C in humidified air containing 5% CO2 at a medium-to-high level of confluence, depending on the time course of the experiment.
Treatments with I3C and Flutamide
I3C and flutamide were dissolved in 99.9% high-performance liquid chromatography grade dimethyl sulfoxide (DMSO; Aldrich) at concentrations 1000 times greater than the final medium concentration. In all experiments, 1 μL of the concentrated agent was added per milliliter of medium, and for the vehicle control, 1 μL DMSO was added per milliliter of medium. For the [3H]thymidine incorporation experiments, cells were plated onto 24-well tissue culture plates and treated with combinations of 100 μM I3C and/or 5 μM flutamide for 48 hours. The same incubation time and concentration of I3C and/or flutamide were used for the Western blot analysis.
Flow Cytometric Analyses of DNA Content and [3H]Thymidine Incorporation
LNCaP cells (4 × 104) were plated onto Corning six-well tissue culture dishes (Corning, NY) and were treated with the indicated concentrations of I3C. The medium was changed every 24 hours. Cells were incubated for 96 hours and hypotonically lysed in 1 mL of DNA staining solution (0.5 mg/mL propidium iodide (PI), 0.1% sodium citrate, 0.05% Triton X-100 [Sigma, St. Louis, MO]). Lysates were filtered using 60 μm Nitex flow mesh (Sefar America, Kansas City, MO) to remove cell membranes. PI-stained nuclei were detected using a PL-2 detector with a 575 nm band pass filter on a Beckman-Coulter (Fullerton, CA) fluorescence-activated cell sorter analyzer with laser output adjusted to deliver 15 megawatts at 488 nm. Ten thousand nuclei were analyzed from each sample at a rate of approximately 600 nuclei per second. The percentages of cells within the G1, S, and G2/M phases of the cell cycle were determined by analyzing the histographic output with the multicycle computer program MPLUS, provided by Phoenix Flow Systems (San Diego, CA), in the Cancer Research Laboratory Microchemical Facility at the University of California at Berkeley. For the [3H]thymidine incorporation experiments, the cells were pulse-labeled with 3 μCi [3H]thymidine per plate for 2 hours at the indicated times of indole treatment and washed 3 times with ice-cold 10% trichloroacetic acid to precipitate the DNA, followed by neutralization with 0.3 M NaOH. The radioactivity was quantified by scintillation counting, and the results from triplicate samples were averaged and expressed as counts per minute of [3H]thymidine incorporation per well.
Western Blot Analysis
After the indicated treatments, cells were harvested in RIPA buffer (150 mM NaCl, 0.5% deoxycholate, 0.1% NoNidet-p40 (NP-40; Flulta Biochemitra, Switzerland), 0.1% sodium dodecyl sulfate [SDS], 50 mM Tris) containing protease and phosphatase inhibitors (50 μg/mL phenylmethyl sulfonyl fluoride [PMSF], 10 μg/mL aprotinin, 5 μg/mL leupeptin, 0.1 μg/mL NaF, 1 mM dithiothreitol [DTT], 0.1 mM sodium orthovanadate, and 0.1 mM beta-glycerol phosphate). Equal amounts of total cellular protein were mixed with loading buffer (25% glycerol, 0.075% SDS, 1.25 mL 14.4 M 2-mercaptoethanol, 10% bromophenol blue, 3.13% stacking gel buffer), fractionated by electrophoresis on 10% polyacrylamide/0.1% SDS resolving gels, electrophoretically transferred to nitrocellulose membranes, and prepared for Western blotting as described previously.19, 20 The working concentration for all antibodies was 1 μg/mL and immunoreactive proteins were detected after 1–3 hours of incubation at room temperature with the appropriate horseradish peroxidase–conjugated secondary antibodies. Blots were treated with ECL reagents (NEN Life Science Products), and all proteins were detected by autoradiography. Equal protein loading was ascertained by Ponceau-S staining of blotted membranes.
Cyclin-Dependent Kinase Immunoprecipitation and In Vitro Kinase Assay
LNCaP cells were harvested in PBS then stored as dry pellets at −70 °C. For the immunoprecipitation (IP), cells were lysed for 15 minutes in IP buffer (50 mM HEPES, pH 7.9; 150 mM NaCl; 0.1% NP-40) containing protease and phosphatase inhibitors (50 μg/mL PMSF, 10 μg/mL aprotinin, 5 μg/mL leupeptin, 0.1 μg/mL NaF, 1 mM DTT, 0.1 mM sodium orthovanadate, 0.1 mM beta-glycerol phosphate). Samples were diluted to 500–700 μg protein in 0.5 ml IP buffer and then precleared for 1 hour at 4 °C with 20 μL of a 1:1 slurry of protein-G Sepharose beads (Pharmacia Biotech, Uppsala, Sweden) in IP buffer and 1 μg rabbit immunoglobulin G. After centrifugation at 14,000 × g for 5 minutes at 4 °C to remove the beads, the precleared supernatant was added to 20 μL of a 1:1 slurry of protein-G Sepharose beads, 5 μg of specific antibody was added to each sample, and the sample was incubated on a rocking platform at 4 °C for 4 hours. The beads then were washed five times with IP buffer and twice with kinase buffer (50 mM HEPES, 10 mM MgCl2, 5 mM MnCl2, 0.1 μg/mL NaF, 10 μg/mL beta-glycerol phosphate, and 0.1 mM sodium orthovanadate). One-half of the immunoprecipitated sample was checked by Western blot analysis to confirm the IP and to compare the protein loading of each sample. The enzymatic activity of immunoprecipitated CDK2 and CDK4 was determined as previously described.19, 20
Antiproliferative Effects of I3C in Human Prostate Carcinoma Cell Lines
As an initial test to determine whether I3C can directly regulate the growth of human prostate carcinoma cells, LNCaP cells were cultured at subconfluence and treated with several concentrations of I3C for 48 hours. Cell proliferation was monitored by pulse labeling the cells with [3H]thymidine during the last 2 hours of indole treatment. Analysis of [3H]thymidine incorporation revealed strong dose-dependent inhibition of DNA synthesis with half-maximal response at approximately 60 μM I3C (Fig. 1A), compared with the DMSO vehicle–treated control cells. Quantitation of the resulting number of cells using a Coulter cell counter confirmed that I3C strongly reduced the rate of LNCaP cell growth (data not shown). Concentrations of I3C greater than 200 μM had a toxic effect on the prostate tumor cells, whereas 100 μM I3C maximally inhibited the growth of LNCaP cells without affecting viability. Therefore, 100 μM I3C was used routinely in subsequent experiments.
Analysis of DNA synthesis over a 144-hour time course revealed that 100 μM I3C reduced the rate of [3H]thymidine incorporation to nearly 50% of the level observed in the control group of LNCaP prostate carcinoma cells after 24 hours of treatment. Only 48 hours of exposure to this indole caused the maximal inhibition of [3H]thymidine incorporation, with the rate of DNA synthesis remaining low for the next 96 hours (Fig. 1B). After 72 hours of I3C treatment, one set of cells was cultured in the absence of indole. Fig. 1B also shows that the rate of [3H]thymidine incorporation increased after indole withdrawal, indicating that 100 μM I3C had no affect on cell viability. Trypan blue exclusion, the lack of cells with a sub G1 DNA content, and no changes in overall cellular morphology (data not shown) further confirmed that treatment with 100 μM I3C did not alter LNCaP cell viability or result in an apoptotic response.
LNCaP cells are androgen-responsive and highly tumorigenic prostate carcinoma cells.35, 41 To determine whether I3C can inhibit the growth of different classes of human prostate carcinoma cells, the effects of I3C on the rate of DNA synthesis were compared in both androgen-responsive (LNCaP and MDA-PCa-2b) and androgen-nonresponsive (PC-3 and DU-145) human prostate carcinoma cells. Both DU-145 and PC-3 cells exhibit relatively high rates of proliferation, whereas LNCaP and MDA-PCa-2b cells are slow-growing cell lines. The rate of [3H]thymidine incorporation was determined at the indicated times over a 72-hour time course of I3C-treated (I3C) or vehicle control (DMSO) cells. As shown in Figure 2, C inhibited the DNA synthesis of both androgen-responsive (AR+) prostate cell lines, whereas only one of the androgen receptor–negative cell lines (DU-145) experienced growth-inhibition by I3C. The growth of PC-3 remained unresponsive to indole treatment. These results suggest that indole treatment will be able to inhibit the growth of a wide range of human prostate carcinoma cell types and, potentially, many distinct classes of prostate tumors. The following experiments will focus on the androgen-responsive LNCaP prostate carcinoma cells because of their relatively rapid growth and wide use in the field as a model cell line for androgen-responsive prostate carcinoma.
I3C Induces a G1 Block in Cell Cycle Progression in Human LNCaP Prostate Tumor Cells
To assess the cell cycle effects of I3C, LNCaP cells were treated with several concentrations of I3C for 96 hours and then were hypotonically lysed in the presence of PI to stain the nuclear DNA. Flow cytometry profiles of nuclear DNA content revealed that I3C induced a G1 cell cycle arrest of these human prostate carcinoma cells. As shown in Figure 3, in the absence of I3C, the LNCaP cells grow as an asynchronous population in all phases of the cell cycle. Increasing concentrations of I3C caused a significant enhancement in the number of cells displaying a G1 phase level of DNA content, from 50–55% of the untreated cell population to 75–80% of the cell populations treated with 120 μM I3C. Consistent with the [3H]thymidine incorporation results, I3C caused a corresponding decrease in the number of S-phase cells, from 17–20% in the absence of indole to 7–10% cells treated with the higher concentrations of I3C (Fig. 3). The relative number of cells with a G2/M DNA content remained approximately the same at all concentrations of I3C. These results suggest that I3C inhibited the proliferation of LNCaP cells by inducing a G1 block in cell cycle progression.
Selective Effects of I3C on the Expression of G1-Acting Cell Cycle Components
To identify potential downstream targets of I3C, the expression of G1-acting cell cycle components was evaluated during a time course of I3C treatment. LNCaP cells were treated with or without 100 μM I3C for the indicated time periods (Fig. 4), and the levels of G1 CDKs, cyclins, and CDK inhibitors were evaluated by Western blot analysis. Among the three G1-acting CDKs (CDK2, CDK4, and CDK6), only CDK6 protein levels were strongly down-regulated in response to I3C treatment. The level of CDK6 is maximally reduced within 24 hours of indole treatment (Fig. 4); this reduction was accounted for by the down-regulation of CDK6 transcripts (data not shown). Importantly, no significant effect was observed on the expression of the two other G1-acting CDKs (CDK2 and CDK4). I3C did not consistently alter the level of any of the G1-acting cyclins (cyclin D1, cyclin D2, cyclin D3, or cyclin E), although in most experiments, a small increase in cyclin E and a minor decrease in cyclin D3 levels were observed. The Western blot analysis also revealed that I3C stimulated the production of the CDK2/CDK4 inhibitors, p21WAF1/CIP1 (Fig. 4) and p27KIP1 (data not shown), after only 72 hours of indole treatment in which the cells reached their maximal growth arrest. It is noteworthy that the production of p16, which inhibits either CDK4 or CDK6 activity, was strongly up-regulated by 48 hours of I3C treatment. Therefore, of all of the tested cell cycle components, only the down-regulation of CDK6 and the stimulation of p16 protein levels changed early enough to coincide with the inhibition of DNA synthesis (Fig. 1). These results suggest a functional relation between CDK activity and the I3C-mediated cell cycle arrest of prostate carcinoma cells.
Inhibition of the Immunoprecipated CDK-2 Enzymatic Activity in I3C-Treated LNCaP Cells
The control of G1 CDK enzymatic activity is critical for regulating cell cycle progression.26, 27 The activity of specific CDKs is regulated, in part, by the composition of the holoenzyme, which includes the appropriate cyclin and/or CDK inhibitory proteins. Therefore, even though the levels of CDK2 and CDK4 remain unaltered after I3C treatment, we evaluated the potential effects of I3C on CDK2 and CDK4-specific enzymatic activities. Because one of the key endogenous substrates for the G1 CDKs is the Rb protein, we determined the ability of CDK2 and CDK4 to phosphorylate Rb in vitro. LNCaP prostate carcinoma cells were treated with or without 100 μM I3C for the indicated times; and CDK2 or CDK4 then was isolated from each total cell extract using a low-stringency IP with the corresponding antibodies. For the kinase assays, one-half of each immunoprecipitated sample was incubated with the carboxy-terminal domain of Rb fused to glutathione S-transferase and [γ-32P]ATP. The amount of phosphorylated Rb was monitored by electrophoretic fractionation of the reaction products and the phosphorylated protein was visualized by autoradiography. The other half of the immunoprecipitated samples was analyzed by Western blot to confirm the efficiency and specificity of each IP. The kinase activity of the CDK2 protein complex immunoprecipitated from I3C-treated samples exhibited a distinct decrease compared with the untreated samples as early as after 18 hours of I3C treatment (Fig. 5). This relatively rapid change in CDK2 activity occurred in the absence of any noticeable changes in the levels of either the cyclins or CDK inhibitors that normally regulate or attenuate CDK2 enzymatic activity. In contrast to kinetics observed for CDK2, the enzymatic activity of CDK4 is reduced after 48 hours of I3C treatment, which is a time point at which an increase in the p16 CDK inhibitor protein is correlated with the overall growth arrest of the LNCaP cells.
I3C Inhibits Production of PSA from LNCaP Prostate Carcinoma Cells
PSA is a secreted serine/threonine protease. Its expression level is correlated with the development of prostate neoplasias and provides a marker for the effectiveness of anti–prostate carcinoma treatments.42 Therefore, a key biologic issue is whether I3C can alter PSA production in LNCaP cells concomitant with its potent antiproliferative effects. LNCaP cells were treated with or without I3C over a 72 hour time course, and Western blots of total cellular extracts were probed with anti–PSA-specific antibodies. I3C strongly inhibited cellular PSA production, which was clearly observable by 24 hours of indole treatment (Fig. 6). As a control, CDK2 production remained generally constant over the 72-hour time course.
Effects of Combinations of I3C and Flutamide on LNCaP Cell DNA Synthesis and Expression of G1-Acting Cell Cycle Proteins, and Production of Prostate Specific Antigen PSA
LNCaP prostate carcinoma cells are highly androgen responsive.35, 41 When they are cultured in medium supplemented with 10% FBS, there are enough steroids and other proliferative factors to support the near-maximal proliferation of these cells in the absence of added exogenous androgens. Under these culture conditions, treatment with flutamide, an androgen antagonist that is used as a therapeutic agent for prostate carcinoma,43 inhibits the growth of LNCaP cells. In treated cells, flutamide is converted into its more active 2-hydroxylflutamide metabolite.44 Because LNCaP cells respond to I3C or flutamide, these cells were treated with the indicated combinations of 100 μM I3C and 5 μM flutamide for 48 hours. Treatment with either I3C or flutatmide alone strongly inhibited the incorporation of [3H]thymidine, whereas a combination of both reagents led to more effective growth inhibition, which virtually abolished DNA synthesis (Fig. 7). In parallel sets of cells, the expression levels of specific cell cycle proteins that are responsible for progression through G1 and/or transition into S phase were monitored by Western blot analysis of total cell extracts. I3C and flutamide had different effects on the expression levels of the tested cell cycle proteins (Fig. 7). I3C selectively decreased the level of CDK6 protein and increased the level of the p16 CDK inhibitor, whereas flutamide had no effect on either of these cell cycle proteins under our cell culture conditions. In addition, a combination of I3C and flutamide had the same effects as I3C alone on CDK6 and p16 production. The minor induction of p21WAF1/CIP1 that is observed after 48 hours of I3C treatment appears to be reproducibly disrupted in the presence of flutamide. The expression levels of the other cell cycle proteins tested, such as CDK2, CDK4, cyclin D1, and cyclin E, were not affected by either I3C or flutamide.
Western blot analysis of both cell extracts and the secreted fractions revealed that treatment with I3C or flutamide alone strongly down-regulated the level of both cellular and secreted PSA (Fig. 7B). Consistent with the effects on LNCaP cell proliferation, a combination of I3C and flutamide inhibited the production of secreted and cellular PSA to a greater extent (Fig. 7B). Our results demonstrated that although I3C and flutamide have different effects on the expression of cell cycle proteins, both reagents have striking effects on PSA production, suggesting a common regulatory step in the indole and antiandrogen pathways. These studies suggest that a combination of I3C and flutamide, which together have a potent antiproliferative effect on LNCaP cells, may represent an effective prostate carcinoma therapeutic.
Our results have established a direct link between the regulation of cell cycle control by the dietary indole I3C and the selective disruption in expression and activity of G1 cell cycle components. The I3C-mediated G1 cell cycle arrest of LNCaP cells is accompanied by a rapid reduction in CDK6 expression, an increase in the level of the p16 CDK inhibitor, and a strong inhibition in CDK2 enzymatic activity. Although a specific target protein that binds to either I3C or its cellular metabolite has not been identified, we propose that the direct binding of the indole to an existing cellular component initiates the signaling cascade that induces the G1 cell cycle arrest. We further speculate that the regulated changes in CDK6 expression, p16 production, and CDK2 activity that occur before the maximal growth arrest are likely to be direct consequences of the I3C signaling pathway in prostate carcinoma cells. We also observed a delayed and relatively modest inhibition of CDK4 enzymatic activity that is first observed at a time point (48 hours) of I3C treatment when the level of p16 is elevated and the levels of the p21WAF1/CIP1 and p27KIP1 CDK inhibitors begin to increase. This time frame was correlated with the overall reduction in the LNCaP cell proliferative state by I3C, suggesting that these changes may be more important to maintenance of the growth-arrested state than to mediation of the antiproliferative effect.
In most instances, the enzymatic activity, but not the expression, of the G1-acting CDKs is regulated by antiproliferative factors.45 Only a limited number of studies have shown that an alteration in the level of CDK6 and its corresponding activity is correlated with the transformed or proliferative state of the cells. For example, tumor-specific amplification of CDK6 has been observed in human gliomas.46 It is noteworthy that previous reports have demonstrated that I3C inhibited CDK6 expression in both estrogen-responsive and nonresponsive human breast carcinoma cells.19, 21 Those observations, in combination with our studies involving LNCaP prostate carcinoma cells, suggest that the selective inhibition of CDK6 gene expression may be a hallmark of the cellular mechanism operating in reproductive tumor cells that allows these cell types to respond to I3C. In breast carcinoma cells, we have discovered that I3C treatment alters the function of Sp-1 transcription factors at the specific indole-regulated composite element in the CDK6 promoter,21 and we currently are attempting to determine whether I3C has a similar response in human prostate carcinoma cells. It is conceivable that the similarities in the developmental origins of mammary and prostate tissue,2 both arising from the mesoderm, may account for the existence of analogous indole signaling pathways, common I3C target proteins, and/or similar downstream transcription factors that control CDK6 gene expression.
A key distinction between the prostate and breast carcinoma cell responses to I3C is the stimulation in p16 protein production observed in the LNCaP cells. Because p16 associates with CDK6, it is possible that the decrease in CDK6 protein levels caused an increase in free p16 levels in the cell and, therefore, altered its turnover rate. The increase in p16 could contribute to the modest inhibition of CDK4 activity observed starting at 48 hours of I3C treatment. Therefore, in prostate carcinoma cells, several G1-acting cell cycle components are regulated simultaneously by I3C at different cellular levels of regulation. It is likely that this combination of regulatory mechanisms mediates the antiproliferative effects of this indole.
LNCaP prostate carcinoma cells, which initially were isolated from a lymph node carcinoma of the prostate, are highly androgen responsive.41 This steroid-responsive property allowed the direct comparison of the growth inhibitory effects of I3C with those of the androgen antagonist, flutamide. Our results established that the antiproliferative cascades activated by I3C and flutamide can cooperate to induce a more stringent growth arrest of the prostate carcinoma cells treated with a combination of both compounds compared with treatment with either agent alone. Therefore, it is tempting to consider that these agents could potentially be used in combination as a new therapy for androgen-responsive prostate carcinoma. The advantage of such a combination, especially using flutamide at suboptimal doses, is the potential to overcome certain physiologic limitations observed with high concentrations of the antiandrogen in traditional androgen ablation therapy.47
The cooperative effects of I3C and flutamide on prostate carcinoma cell growth suggest the existence of common gene targets of their respective signaling pathways. One such target gene is the PSA, a secreted protein that is used as a biomarker for the advancement and treatment of prostate carcinoma.48 Treatment with either I3C or flutamide down-regulated the levels of cellular and secreted PSA, whereas treatment with both agents virtually ablated the expression of this gene. Most prostate tumors are androgen responsive, and expression of PSA protein in prostate tissue has been shown to be correlated with fluctuating androgen levels during male development.49 In LNCaP cells, I3C strongly inhibited the transcription of the PSA gene (unpublished data). It is conceivable that the inhibitory effects of I3C on PSA levels may be either independent of or linked to the cell cycle arrest of LNCaP cells. Our results with cultured prostate carcinoma cells suggest that some caution should be taken in using only PSA as a biomarker for patients with prostate carcinoma who ingest significant amounts of I3C-containing vegetables concurrent with their antiandrogen treatment.
The current study has provided the necessary first experimental steps that are crucial for the eventual development of indoles, alone or in combination with known anti–prostate carcinoma treatments, in new in vivo chemotherapeutic strategies with reduced side effects for the control of prostate carcinoma. I3C represents an intriguing potential therapeutic agent for reproductive cancers. A key future direction will involve the elucidation of the precise I3C-activated signaling pathway that mediates the antiproliferative effects of dietary indoles.
The authors thank Erin Cram, Hahn Garcia, and other members of the Firestone and Bjeldanes Laboratories for their helpful and critical comments. The authors also thank Minnie Wu, Hal Light, Tim Labadie, and Cindy Huynh for their technical assistance.