A Potent Indole-3-Carbinol–Derived Antitumor Agent with Pleiotropic Effects on Multiple Signaling Pathways in Prostate Cancer Cells Free

Indole-3-carbinol has emerged as a promising chemopreventive agent due to its in vivo efficacy in various animal models. However, indole-3-carbinol exhibits weak antiproliferative potency and is unstable in acidic milieu. Thus, this study was aimed at exploiting indole-3-carbinol to develop potent antitumor agents with improved chemical stability. This effort culminated in OSU-A9 {[1-(4-chloro-3-nitrobenzenesulfonyl)-1H-indol-3-yl]-methanol}, which is resistant to acid-catalyzed condensation, and exhibits 100-fold higher apoptosis-inducing activity than the parent compound. Relative to indole-3-carbinol, OSU-A9 displays a striking qualitative similarity in its effects on the phosphorylation or expression of multiple signaling targets, including Akt, mitogen-activated protein kinases, Bcl-2 family members, survivin, nuclear factor-κB, cyclin D1, p21, and p27. The ability of OSU-A9 to concurrently modulate this broad range of signaling targets underscores its in vitro and in vivo efficacy in prostate cancer cells. Nevertheless, despite this complex mode of mechanism, normal prostate epithelial cells were less susceptible to the antiproliferative effect of OSU-A9 than PC-3 and LNCaP prostate cancer cells. Treatment of athymic nude mice bearing established s.c. PC-3 xenograft tumors with OSU-A9 at 10 and 25 mg/kg i.p. for 42 days resulted in a 65% and 85%, respectively, suppression of tumor growth. Western blot analysis of representative biomarkers in tumor lysates revealed significant reductions in the intratumoral levels of phosphorylated (p-) Akt, Bcl-xL, and RelA, accompanied by robust increases in p-p38 levels. In conclusion, the ability of OSU-A9 to target multiple aspects of cancer cell survival with high potency suggests its clinical value in prostate cancer therapy. [Cancer Res 2007;67(16):7815–24]

The chemopreventive potential of indole-3-carbinol, a naturally occurring phytochemical in cruciferous vegetables (1), has received much attention in light of its shown in vivo efficacy to protect against chemically induced carcinogenesis in animals (2–7). Moreover, the clinical benefits of indole-3-carbionol have also been shown in human trials for cervical dysplasia (8), breast cancer (9, 10), and vulvar intraepithelial neoplasia (11). Despite these advances in translational research, the mechanism by which indole-3-carbinol inhibits tumorigenesis remains inconclusive, which, in part, might be attributable to its metabolic instability and complicated pharmacologic properties. In an acidic milieu, indole-3-carbinol undergoes spontaneous dehydration and condensation to form diindoylmethane and a series of oligomeric products (12, 13), all of which exhibit in vitro antiproliferative activities against cancer cells (14–18). Furthermore, mechanistic evidence indicates that indole-3-carbinol facilitates growth arrest and apoptosis by targeting a broad range of signaling pathways pertinent to cell cycle regulation and survival, including those mediated by Akt, nuclear factor-κB (NF-κB), Bcl-2, mitogen-activated protein (MAP) kinases, the cyclin-dependent kinase (CDK) inhibitors p21 and p27, and cyclin D1 (reviews: refs. 19–23). However, as these signaling targets often operate in a cell-specific fashion, it remains in dispute whether any of them could solely account for the effect of indole-3-carbinol on growth arrest and apoptosis in tumor cells (24).

From a mechanistic perspective, the ability of indole-3-carbinol and its metabolites to target a broad spectrum of signaling pathways underlies their antiproliferative effects. However, these agents suffer from metabolic instability, unpredictable pharmacokinetic properties, and low in vitro antiproliferative potency, which render therapeutic concentrations difficult to achieve in the body. Consequently, recent years have witnessed the use of diindoylmethane as a scaffold to carry out structural modifications, which has led to three distinct antitumor agents with higher potency reported in the literature: (p-substituted phenyl)-diindoylmethanes [peroxisome proliferator-activated receptor γ (PPARγ) agonists; refs. 18, 25–28], SR13668 (an Akt inhibitor; refs. 29, 30), and an indole-3-carbinol tetrameric derivative (a CDK6 inhibitor; ref. 31; Fig. 1A). These novel agents exhibit μmol/L potency in inducing apoptosis or cell cycle arrest, however, through signaling pathways distinct from that affected by diindoylmethane.

In this study, we report the use of indole-3-carbinol as a lead compound to develop antitumor agents with improved chemical stability and apoptosis-inducing potency. This structural optimization effort culminated in the generation of OSU-A9, a novel, acid-stable analogue with two orders of magnitude higher apoptosis-inducing potency than the parental indole-3-carbinol. Moreover, OSU-A9 retained the pleiotropic effects of indole-3-carbinol on multiple signaling targets associated with growth arrest and apoptosis. Equally important, despite this broad spectrum of pharmacologic activities, normal prostate epithelial cells were less sensitive to the antiproliferative effect of OSU-A9 relative to prostate cancer cell lines.

Reagents. 1H-indole-3-carbaldehyde (Sigma-Aldrich) was used as the starting material to synthesize 1H-indole-3-carbinol and series I to III agents (Fig. 1A), of which the synthesis will be published elsewhere. The identity and purity (≥99%) of these synthetic derivatives were verified by proton nuclear magnetic resonance, high-resolution mass spectrometry, and elemental analysis. For in vitro experiments, these agents at various concentrations were dissolved in DMSO, and were added to cells in medium with a final DMSO concentration of 0.1%. Rabbit polyclonal antibodies against various biomarkers were obtained from the following sources: phosphorylated (p-)⁴⁷³Ser Akt, p-Bad, Bad, phosphorylated extracellular signal-regulated kinases (p-ERK), phosphorylated c-Jun-NH₂-kinase (p-JNK), JNK, p-p38, p38, cyclin D1, and NF-κB, Cell Signaling Technologies; Akt, ERKs, p27, p21, Bax, Bcl-2, Bcl-xL, and AR, Santa Cruz Biotechnology; survivin, R&D Systems; and β-actin, Sigma-Aldrich. Mouse monoclonal anti-PARP antibody was purchased from PharMingen. The enhanced chemiluminescence system for detection of immunoblotted proteins was from GE Healthcare Bioscience. Other chemicals and biochemistry reagents were obtained from Sigma-Aldrich unless otherwise mentioned.

Cell culture. LNCaP androgen-responsive (p53^+/+) and PC-3 androgen-nonresponsive (p53^−/−) human prostate cancer cells were purchased from the American Type Tissue Collection and cultured in RPMI 1640 (Life Technologies) supplemented with 10% fetal bovine serum (FBS; Life Technologies). Normal human prostate epithelial cells were obtained from Cambrex Bioscience-Walkersville, and maintained in the vendor's recommended defined prostate epithelial growth medium. All cell types were cultured at 37°C in a humidified incubator containing 5% CO₂.

Cell viability analysis. The effect of test agents on cell viability was assessed by using the 3-(4,5-dimethylthiazol-2-yl)-2,5-diphenyl-2H-tetrazolium bromide (MTT) assay in 6 to 12 replicates. Cancer cells and prostate epithelial cells were grown in 5% FBS–supplemented RPMI 1640 or 5% FBS–supplemented prostate epithelial growth medium, respectively, in 96-well, flat-bottomed plates for 24 h, and then exposed to various concentrations of test agents in the same medium for the indicated time intervals. Controls received DMSO vehicle at a concentration equal to that in drug-treated cells. At the end of the treatment, the medium was removed, replaced by 200 μL of 0.5 mg/mL of MTT in the same medium, and cells were incubated in the CO₂ incubator at 37°C for 2 h. Supernatants were removed from the wells, and the reduced MTT dye was solubilized in 200 μL/well DMSO. Absorbance at 570 nm was determined on a plate reader.

Cell proliferation. PC-3 and LNCaP cells were seeded into six-well plates at ∼200,000 per well in 5% FBS–containing RPMI 1640. Following a 24 h attachment period, cells were treated in triplicate with the indicated concentrations of test agent or DMSO vehicle in 5% FBS–containing RPMI 1640. At different time intervals, cells were harvested by trypsinization and counted using a Coulter counter (Model Z1 D/T, Beckman Coulter).

Apoptosis analysis. Three apoptosis biomarkers were used to assess drug-induced apoptosis by Western blot analysis: cytochrome c release, caspase-3 activation, and PARP cleavage. Cytosolic-specific, mitochondria-free lysates for cytochrome c analysis was prepared according to an established procedure (32). After drug treatments, both the incubation medium and adherent cells in T-75 flasks were collected and centrifuged at 600 × g for 5 min. The pellet fraction was recovered, placed on ice, and triturated with 300 μL of a chilled hypotonic lysis solution [50 mmol/L PIPES-KOH (pH 7.4), containing 220 mmol/L mannitol, 68 mmol/L sucrose, 50 mmol/L KCl, 5 mmol/L EDTA, 2 mmol/L MgCl₂, 1 mmol/L DTT, and a mixture of protease inhibitors including 100 μmol/L AEBSF, 80 nmol/L aprotinin, 5 μmol/L bestatin, 1.5 μmol/L E-64 protease inhibitor, 2 μmol/L leupeptin, and 1 μmol/L pepstatin A]. After a 45-min incubation on ice, the mixture was centrifuged at 200 × g for 10 min. The supernatant was collected in a microcentrifuge tube and centrifuged at 14,000 rpm for 30 min. Equivalent amount of protein (50 μg) from each supernatant was subject to immunoblotting with anti–cytochrome c antibodies, as described below.

With regard to the other two apoptosis biomarkers, collected drug-treated cells were washed with ice-cold PBS and resuspended in lysis buffer containing 20 mmol/L Tris-HCl (pH 8), 137 mmol/L NaCl, 1 mmol/L CaCl₂, 10% glycerol, 1% NP40, 0.5% deoxycholate, 0.1% SDS, 100 μmol/L 4-(2-aminoethyl)benzenesulfonyl fluoride, leupeptin at 10 μg/mL, and aprotinin at 10 μg/mL. Soluble cell lysates were collected after centrifugation at 10,000 × g for 5 min. Equivalent amounts of proteins (60–100 μg) from each lysate were subject to immunoblotting with antibodies against caspase-3 or PARP as described below.

Immunoblotting. Biomarkers of apoptosis and signaling components associated with cell survival and growth arrest were assessed by Western immunoblotting as follows. Treated cells were washed in PBS, resuspended in SDS sample buffer, sonicated for 5 s, and then boiled for 5 min. After brief centrifugation, equivalent amounts of proteins from the soluble fractions of cell lysates were resolved in 10% SDS-polyacrylamide gels on a Minigel apparatus, and transferred to a nitrocellulose membrane using a semidry transfer cell. The transblotted membrane was washed thrice with TBS containing 0.05% Tween 20 (TBST). After blocking with TBST containing 5% nonfat milk for 60 min, the membrane was incubated with an appropriate primary antibody at 1:500 dilution (with the exception of anti–β-actin antibody, 1:2,000) in TBST–5% low fat milk at 4°C for 12 h, and was then washed thrice with TBST. The membrane was probed with goat anti-rabbit or anti-mouse IgG-horseradish peroxidase conjugates (1:2,500) for 90 min at room temperature, and washed thrice with TBST. The immunoblots were visualized by enhanced chemiluminescence.

Molecular modeling. Molecular structures of A7, OSU-A9, and A12 were subjected to energy minimization using the Merck Molecular Force Field program available as part of the Macromodel 7.0 software package (Schrodinger⁴

). The minimum conformations were then fully optimized at a density functional theory level of B3LYP/6-31G* basis set using Gaussian 03 (Gaussian, Inc.). All of the fully optimized structures were confirmed by normal mode analysis; no negative frequencies were found. Computations for electron density and electrostatic potential were then carried out for each of the fully optimized structures with a grid of 216,000 points using Gaussian 03. Molecular electrostatic potential maps for each compound were generated with the electrostatic potential mapped onto the electron density. The electron density isosurface value was 0.002 (electron/Å) with a range of −0.03 to 0.03 for the electrostatic potential. All molecular modeling calculations and manipulations were done on Silicon Graphics O2 (Silicon Graphics, Inc.).

Nuclear magnetic resonance analysis of acid stability. Indole-3-carbinol and OSU-A9, 20 mg each, were dissolved in 1 mL of CD₃OD. To each solution, 100 μL of deuterium-labeled HCl were added, and nuclear magnetic resonance (NMR) spectra were recorded in a 300 MHz NMR spectrometer at room temperature at different time intervals.

Luciferase assay for PPARγ activation. The PPAR-response element (PPRE)-x3-TK-Luc reporter vector contains three copies of the PPRE upstream of the thymidine kinase promoter-luciferase fusion gene and was kindly provided by Dr. Bruce Spiegelman (Harvard University, Cambridge, MA). The reporter gene assay was carried out as previously described (33). In brief, PC-3 cells were cultured in a 100-mm plate in phenol red–free RPMI 1640 containing 10% FBS until the achieved 50% to 70% confluency, after which they were transfected with 6 μg of the plasmid using Fugene 6 (Roche) in RPMI 1640. For each transfection, herpes simplex virus-thymidine kinase promoter–driven Renilla luciferase (phRL-TK) was used as an internal control for normalization. Following transfections, cells were treated as indicated in RPMI 1640 containing 10% charcoal-stripped FBS. Cells were then collected into Passive Lysis Buffer (Promega), and luciferase activities in the cell lysates were determined by luminometry. All transfection experiments were carried out in triplicate plates and repeated separately at least thrice.

In vivo studies. Intact male NCr athymic nude mice (5–7 weeks of age) were obtained from the National Cancer Institute. The mice were group housed under conditions of constant photoperiod (12 h light/12 h dark) with ad libitum access to sterilized food and water. All experimental procedures using these mice were done in accordance with protocols approved by the Institutional Laboratory Animal Care and Use Committee of The Ohio State University.

Each mouse was inoculated s.c. in the right flank with 5 × 10⁵ PC-3 cells in a total volume of 0.1 mL serum-free medium containing 50% Matrigel (BD Biosciences) under isoflurane anesthesia. As tumors became established (mean starting tumor volume, 109 ± 10 mm³), mice were randomized to three groups (n = 7) that received the following treatments: (a) OSU-A9 at 10 mg/kg body weight qd, (b) OSU-A9 at 25 mg/kg qd, and (c) DMSO vehicle. Mice received treatments by i.p. injection (50 μL/mouse) for the duration of the study. Tumors were measured weekly using calipers and their volumes calculated using a standard formula, as follows: width² × length × 0.52. Body weights were measured weekly. At terminal sacrifice, a complete necropsy was done on all mice and PC-3 tumors were harvested. A portion of each tumor was snap-frozen in liquid nitrogen and stored at −80°C until needed for Western blot analysis of relevant biomarkers, and the remainder was fixed in 10% formalin. All other tissues were fixed overnight in 10% formalin then transferred to 70% ethanol. Four-micrometer-thick, paraffin-embedded tissue sections were stained with H&E by standard procedures. A core list of tissues from three mice per group were evaluated microscopically animal by animal by a veterinary pathologist in accordance with Society of Toxicologic Pathology–proposed guidelines for repeat-dose toxicity studies (34), with the exception of spinal cord and thymus. Blood from each mouse was submitted to The Ohio State University Veterinary Clinical Laboratory Services for evaluation of serum chemistry and hematologic variables routinely evaluated in a clinical setting (35).

Statistical analysis. Differences in relative PPARγ activation in vitro and among group means of tumor volume in vivo were analyzed for statistical significance using one-way ANOVA followed by the Neuman-Keuls test for multiple comparisons. Differences were considered significant at P < 0.05. Statistical analyses were done using SPSS for Windows (SPSS, Inc.).

Structure-activity relationship. The intrinsic instability of indole-3-carbinol in acidic conditions arises from the vinyl hemiaminal moiety of the indole ring, which renders the molecule highly susceptible to acid-catalyzed dehydration and condensation to form oligomeric products (13). Mechanistically, this acid lability could be circumvented by blocking the vinyl hemiaminal function through substitution. Accordingly, to test the premise that indole-3-carbinol was amenable to structural optimization, we synthesized three series of N-substituted analogues: (1-aryloyl-1H-indol-3-yl)-methanols (I), (1-arylsulfonyl-1H-indol-3-yl)-carbaldehydes (II), and (1-arylsulfonyl-1H-indol-3-yl)-methanols (III; Fig. 1B).

These agents (A1–A24) were evaluated for their ability to reduce cell viability of PC-3 (p53^−/−) androgen-nonresponsive and LNCaP (p53^+/+) androgen-responsive prostate cancer cells after 24-h exposure by MTT assay. Although the IC₅₀ values of all carboxamide derivatives (series I) were >50 μmol/L, some of the derivatives in the II and III series showed improved antitumor activities vis-à-vis indole-3-carbinol. Especially noteworthy is the compound A9 {[1-(4-chloro-3-nitrobenzenesulfonyl)-1H-indol-3-yl]-methanol; designated as OSU-A9}, which exhibited IC₅₀ values of 2 and 3.8 μmol/L for PC-3 and LNCaP cells, respectively (Fig. 1B). This antitumor potency was two orders of magnitude higher than that of indole-3-carbinol (IC₅₀, 512 and 267 μmol/L, respectively; Fig. 2A). Moreover, assessment of effects on nonmalignant cells revealed that prostate epithelial cells exhibited a 2.2- to 4.5-fold lower sensitivity to OSU-A9 (IC₅₀, 9 μmol/L) than the prostate cancer cells (Fig. 2A).

The in vitro efficacy of OSU-A9 vis-à-vis indole-3-carbinol in inhibiting the proliferation of PC-3 and LNCaP cells was also examined by direct counting of drug-treated cells (Fig. 2B). OSU-A9, at 2.5 μmol/L and above, completely suppressed cell proliferation and cause a net decrease in cell numbers in both cell lines, indicative of cytocidal activity. In line with the cell viability data, a concentration of at least 300 μmol/L was required for indole-3-carbinol to attain the same level of efficacy.

Molecular modeling analysis. The structure-activity analysis revealed a stringent structural requirement for the high potency of OSU-A9. Any change in the structure of the compound, such as substitution of the sulfonyl (-SO₂-) linker with a carbonyl (-CO-) function (i.e., A7) or replacement of the nitro (-NO₂) substituent with an amino (-NH₂) moiety (i.e., A12), resulted in a substantial loss of antitumor activity (Fig. 1B). To shed light onto the structural basis underlying this subtle structure-activity relationship, we analyzed the configuration and surface electrostatic potential of OSU-A9 versus those of A7 and A12 via modeling analysis (Fig. 3A).

As shown, OSU-A9 and A7 assumed vastly different configurations resulting from the structural differences between the tetrahedron-like sulfonyl (-SO₂-) and the triagonal-planar carbonyl (-CO-) linkers. Equally important, the electropositive nature of the nitrosubstituent on the appended phenyl ring was crucial to the ability of OSU-A9 to induce apoptosis. Replacement of the electron-withdrawing nitro group with an electron-donating amino function, as in A12, changed the regional surface potential, resulting in the loss of apoptosis-inducing activity.

OSU-A9 resists acid-catalyzed dimerization. We used a NMR technique to analyze the chemical stability of OSU-A9 versus indole-3-carbinol in 0.1 N HCl by monitoring changes in the proton signal associated with CH₂OH (Fig. 3B). Individual compounds (20 mg) were dissolved in 1 mL of deuterium-labeled methanol (CD₃OD). The NMR spectra revealed signals for the methylene protons (indicated by *) at 4.73 and 4.74 ppm for indole-3-carbinol and OSU-A9, respectively (top spectra, left and right). Addition of 100 μL of 0.1 N deuterium-labeled HCl to indole-3-carbinol resulted in an immediate shift of the CH₂ signal from 4.73 to 4.66 ppm (t = 5 min), indicating the chemical transformation of indole-3-carbinol to an acid reaction mixture consisting of diindoylmethane and other oligomeric products (12, 13). On the other hand, no appreciable change in the spectrum was noted after exposure of OSU-A9 to HCl for up to 8 h, indicating its significantly greater chemical stability.

OSU-A9 facilitates apoptosis by targeting multiple signaling pathways identical to that of indole-3-carbinol. Evidence suggest that the antiproliferative effect of OSU-A9 was, at least in part, attributable to apoptosis, reminiscent of that of indole-3-carbinol. Western blot analysis showed a dose-dependent effect of both agents on cytochrome c release into the cytoplasm, caspase-3 activation, and PARP cleavage (Fig. 4A). The effects of indole-3-carbinol and OSU-A9 on these two apoptosis-related biomarkers were qualitatively similar, albeit with a 100-fold difference in potency.

From a translational perspective, the mechanism by which OSU-A9 mediated antiproliferative effects warranted investigation. The finding that many indole derivatives, including diindoylmethanes, exhibited PPARγ agonist activity (18, 25–28, 36, 37) raised the possibility that PPARγ activation might contribute to the apoptosis-inducing effect of OSU-A9. Accordingly, we used an established PPRE-luciferase reporter assay (33) to assess the ability of OSU-A9 to transactivate PPARγ in PC-3 cells. However, even at 20 μmol/L, OSU-A9 lacked appreciable activity in PPARγ transactivation, whereas the PPARγ agonist rosiglitazone induced significant activation of the reporter gene (Fig. 4B).

Indole-3-carbinol has been reported to target an array of signaling pathways to induce apoptosis and cell cycle arrest in cancer cells (reviews: refs. 19–23). Consequently, we examined the dose-dependent effect of OSU-A9 vis-à-vis indole-3-carbinol in PC-3 cells on the phosphorylation and/or expression status of a series of molecular targets reported for indole-3-carbinol in the literature. These targets comprised three categories of biomarkers: (a) phosphorylation of signaling kinases: Akt and its downstream effector GSK3β, ERKs, JNKs, and p38; (b) phosphorylation/expression of Bcl-2 family members: Bad, Bax, Bcl-2, Bcl-xL, and Mcl-1; (c) expression of other apoptosis and cell-cycle regulatory proteins: survivin, NF-κB/RelA, cyclin D1, p27, and p21 (Fig. 5).

Despite a 100-fold difference in antitumor potency, the pharmacologic profiles of OSU-A9 and indole-3-carbinol in interfering with these target proteins were virtually identical. As shown, both agents facilitated dose-dependent dephosphorylation of Akt and its substrates GSK3β and Bad, accompanied by increased phosphorylation of ERKs, JNK, and p38. Moreover, the dose-dependent effects of OSU-A9 on suppressing the expression of Bcl-2, Bcl-xL, Mcl-1, survivin, NF-κB, and cyclin D1, and on up-regulating the expression of Bax, p27, and p21 paralleled those of indole-3-carbinol.

Together, these data suggest that OSU-A9 retains the pleiotropic effects of indole-3-carbinol with regard to the activating apoptosis machinery by targeting multiple signaling pathways. Pursuant to these findings, the in vivo antitumor potential of OSU-A9 was further assessed in a PC-3 xenograft animal model.

OSU-A9 suppresses PC-3 tumor xenograft growth in vivo. The maximum tolerated dose of OSU-A9 in athymic nude mice was determined by i.p. injection at 5, 10, 25, and 50 mg/kg/d (n = 3) continuously for 14 days (data not shown). No more than 10% weight loss was noted at doses up to 25 mg/kg (Fig. 6A , inset).

To examine the in vivo antitumor efficacy of OSU-A9, athymic nude mice bearing established s.c. PC-3 tumor xenografts (mean tumor volume, 109 ± 10 mm³) were treated i.p. for 42 days with OSU-A9 at 10 or 25 mg/kg daily, or with vehicle. As shown in Fig. 6A, treatment of mice with 10 and 25 mg/kg of OSU-A9 significantly inhibited PC-3 tumor growth by 66% and 85%, respectively, relative to vehicle-treated controls (P < 0.01). Importantly, mice seemed to tolerate treatment with OSU-A9 without overt signs of toxicity, without significant loss of body weight compared with vehicle-treated controls (Fig. 6A , inset), and with normal hematologic and serum chemistry findings. The sole lesions observed at necropsy in all OSU-A9–treated mice were mild to moderate amounts of fibrous adhesions that were limited to the peritoneal surfaces of tissues, including the colon, seminal vesicles, and body wall, and did not seem to compromise function. Microscopically, there were no treatment-associated lesions in the array of tissues examined in the OSU-A9–treated mice, with the exception of the affected peritoneum, which was variably expanded by mixtures of granulation and fibrous tissue and activated macrophages. No lesions were observed in the vehicle-treated controls.

To correlate this in vivo tumor-suppressive response to mechanisms identified in vitro, the effects of OSU-A9 on six representative intratumoral biomarkers of drug activity were evaluated by immunoblotting of PC-3 tumor homogenates collected after 42 days of treatment. These biomarkers included the phosphorylation status of Akt and p38, and expression levels of Bcl-xL, and NF-κB/RelA. As shown in Fig. 6B and C, the effects of OSU-A9 on these biomarkers were qualitatively similar to those observed in vitro, and reflect the dose-dependent tumor suppression in vivo. Treatment with OSU-A9 i.p. at 10 and 25 mg/kg/d induced marked reductions in intratumoral levels of p-Akt (83 ± 8% and 94 ± 1% reductions, respectively, compared with vehicle-treated controls), Bcl-xL (34 ± 8% and 86 ± 4%, respectively), and RelA (60 ± 7% and 77 ± 7%, respectively), accompanied by increases in intratumoral levels of p-p38 (9.0 ± 3.5-fold and 10.1 ± 2.4-fold, respectively).

During the course of tumor progression, cancer cells constitutively up-regulate cell proliferation– and cell survival–regulatory signaling mechanisms, thereby overcoming genomic instability and/or acquiring a drug-resistant phenotype. From a clinical perspective, it is desirable to concomitantly target these molecular abnormalities by using a combination therapy or an agent with pleiotropic effects to optimize therapeutic outcomes (38). This rationale constitutes the molecular basis for structurally optimizing indole-3-carbinol to develop potent antitumor agents. This effort has culminated in the generation of OSU-A9, which provides considerable therapeutic advantages over indole-3-carbinol with respect to chemical stability and antitumor potency.

The introduction of a (3-chloro-2-nitrobenzene)sulfonyl substituent onto the indole nucleus endowed OSU-A9 with resistance to acid-catalyzed dehydration, and, equally important, a 100-fold higher apoptosis-inducing potency. The unique stereoelectronic property of this substituent allowed OSU-A9 to interact more effectively with target proteins compared with indole-3-carbinol. Meanwhile, this structural modification might also affect the kinetic behavior of drug absorption in different cell lines, as suggested by a crossover in the relative sensitivity to OSU-A9 versus indole-3-carbinol between PC-3 and LNCaP cells; that is, LNCaP cells were more sensitive to indole-3-carbinol–induced apoptosis than PC-3 cells, whereas a reversal in the relative susceptibility was noted with OSU-A9.

Relative to indole-3-carbinol, OSU-A9 displayed a striking similarity in the effects on modulating the phosphorylation or expression of a multitude of molecular targets, including Akt and its downstream effectors GSK3β and Bad; the MAP kinases ERKs, p38, and JNK; the Bcl-2 family members Bax, Bcl-2, Bcl-xL, and Mcl-1; the inhibitor of apoptosis protein survivin; NF-κB; cyclin D1; and the CDK inhibitors p21 and p27 (Fig. 6D). Among these targets, the ability of OSU-A9 and indole-3-carbinol to facilitate Akt dephosphorylation in conjunction with increased phosphorylation of ERKs, JNK, and p38 is mechanistically intriguing. This finding is reminiscent to that reported for a number of molecules, including thrombin (39), sphingosine 1-phosphate (40), and kainate (41), in different cell systems. Moreover, one study with thrombin in human endothelial cells suggests a mechanistic link between the activation of Rho/Rho-kinase signaling and drug-induced divergent effects on the phosphorylation status of Akt and MAP kinases (39), which is currently under investigation in OSU-A9–treated prostate cancer cells.

Together, the aforementioned signaling kinases/proteins regulate cell cycle and apoptosis at multiple levels, including transcriptional activation of gene expression, cell cycle checkpoint control, intracellular kinase signaling, mitochondrial integrity, and caspase activation, all of which are clinically relevant to the tumorigenesis and progression of prostate cancer. This broad range of antitumor activities underscores the in vitro and in vivo efficacy of OSU-A9 in prostate cancer cells. It is especially noteworthy that, despite this complicated mode of drug action, normal prostate epithelial cells were less susceptible to the antiproliferative effect of OSU-A9, reflecting the in vivo tolerance of this drug in tumor-bearing nude mice.

Assessment of the in vivo efficacy in tumor-bearing nude mice indicate that daily i.p. injection of OSU-A9 at 10 and 25 mg/kg/d for 42 days resulted in a 65% and 85%, respectively, suppression of established PC-3 xenograft tumor growth. Western blot analysis of the tumor lysates revealed that the extent of changes in the six representative biomarkers paralleled the dose-dependent tumor-suppressive activity of OSU-A9. The concerted action on these molecular targets underscores the therapeutic potential of OSU-A9 to be developed into a potent antitumor agent not only as a single agent, but also potentially in combination with other chemotherapeutic drugs. Previously, indole-3-carbinol was shown to sensitize prostate and breast cancer cells to cisplatin (19) and tamoxifen (42), respectively. Use of OSU-A9 as a chemosensitizing agent in prostate cancer cells is currently under way in this laboratory.

OSU-A9 was well tolerated by tumor-bearing mice after 6 weeks of repeat dosing. The absence of gross and microscopic lesions in major organs together with normal clinical and hematologic findings indicated that the intra-abdominal fibrous adhesions observed in the OSU-A9–treated animals were likely nonspecific reactions associated with chronic irritation induced by i.p. injection of the drug. Although concerns regarding potential toxicities arising from the use of drugs with pleiotropic actions warrant consideration, our findings suggest that, at least in the case of OSU-A9, such adverse effects are not an obligatory consequence.

Although indole-3-carbinol is considered as a chemopreventive compound, the studies described here were focused on the therapeutic efficacy of OSU-A9. Nevertheless, in light of the molecular heterogeneity of premalignant lesions and the multifactorial nature of carcinogenesis (43, 44), the pleiotropic effects of OSU-A9 on multiple signaling pathways, along with its enhanced chemical stability and lack of evident toxicity, suggest its potential in the context of chemoprevention. Using prostatic intraepithelial neoplasia (PIN) as a clinically relevant intervention point (45), we have evaluated, and continue to evaluate, the chemopreventive efficacy of other compounds developed in our laboratory in the transgenic adenocarcinoma of the mouse prostate model (46) in which the predictable occurrence of PIN lesions is a component of the multistage tumor progression of the model (47). Dysplastic lesions similar to PIN precede the occurrence of tumors in other models as well, such as the hormone-induced model of prostate carcinogenesis in the Noble rat, in which PIN was shown to be a suitable intermediate end point for chemoprevention studies (48). The chemopreventive efficacy of OSU-A9 clearly warrants investigation and studies to do so are planned.

In conclusion, our results show that the novel indole-3-carbinol–derived multitargeted agent, OSU-A9, is a potent antitumor agent that modulates multiple aspects of cancer cell cycle regulation and survival, including intracellular kinase signaling, cell cycle checkpoint control, mitochondrial integrity, and caspase activation. This broad spectrum of antitumor activities in conjunction with low toxicity underlies the translational potential of OSU-A9 and suggests its viability as part of a therapeutic strategy for prostate cancer.

Grant support: National Cancer Institute grant CA112250, grants from William R. Hearst Foundation and Prostate Cancer Foundation, and the Lucius A. Wing Endowed Chair Fund at the Ohio State University (C-S. Chen); National Science Council grants NSC-93018P and NSC-94-2314B-039-033 and financial support from China Medical University CMU94-180 (J-R. Weng) in Taiwan.

The costs of publication of this article were defrayed in part by the payment of page charges. This article must therefore be hereby marked advertisement in accordance with 18 U.S.C. Section 1734 solely to indicate this fact.