SEARCH

SEARCH BY CITATION

Keywords:

  • chemosensitizer;
  • ovarian carcinoma;
  • apoptosis;
  • caspase-2;
  • XIAP

Abstract

  1. Top of page
  2. Abstract
  3. MATERIALS AND METHODS
  4. RESULTS
  5. DISCUSSION
  6. REFERENCES

BACKGROUND

Previously, it was demonstrated that phenoxodiol induces apoptosis in epithelial ovarian carcinoma (EOC) cells and that it is capable of sensitizing these cells to Fas-mediated apoptosis. The objectives of this study were to determine whether phenoxodiol can also act as chemosensitizer to chemotherapeutic agents and to characterize the molecular mechanism behind its sensitizing effect.

METHODS

Ten EOC cell lines were used in this study. The effect of phenoxodiol on the inhibitory concentration 50% (IC50) of carboplatin, paclitaxel, and gemcitabine was determined by the CellTiter 96 Assay. The in vivo effect of combination treatments with phenoxodiol and the above-mentioned agents was determined in animal xenograft models. Apoptosis was measured using the Caspase-Glo Assay and the apoptotic cascade was characterized by Western blot analyses.

RESULTS

The results showed that phenoxodiol is able to sensitize EOC cells to carboplatin, paclitaxel, and gemcitabine both in vitro and in vivo. In addition, it was demonstrated that phenoxodiol is capable of inducing apoptosis by: 1) the activation of the mitochondrial pathway through caspase-2 and Bid signaling, and 2) the proteasomal degradation of the anti-apoptotic protein XIAP.

CONCLUSION

Understanding the components of the apoptotic pathway activated by phenoxodiol, which allows it to sensitize EOC cells to chemotherapeutic agents, will provide valuable information on the characteristic mode of action of a chemosensitizer. This will help in the identification of novel drugs and in the design of better strategies for combination therapy in patients with recurrent ovarian carcinoma. Cancer 2006. © 2005 American Cancer Society.

Ovarian carcinoma is a leading cause of cancer deaths in gynecologic malignancies. The high mortality rate is due to difficulties with early detection and the development of chemoresistance.

Induction of apoptosis is one of the key mechanisms of antitumor therapies, including chemotherapy.1, 2 Therefore, understanding the mechanisms regulating the apoptotic cascade in epithelial ovarian carcinoma (EOC) cells may lead to the development of new therapeutic approach for the treatment of ovarian carcinoma.

Phenoxodiol, an isoflavone derivative, has been shown to inhibit proliferation in a wide range of human cancer cell lines.3, 4 In our previous studies, we have shown that phenoxodiol induces cell death in EOC cells in a caspase-dependent manner.3 Furthermore, we have shown that phenoxodiol restores sensitivity of EOC cells to Fas-mediated apoptosis.3, 5 Whereas it has a potent proapoptotic effect on cancer cells, phenoxodiol does not exhibit any toxicity to normal ovarian epithelial cells.3

In the present study, we demonstrate that phenoxodiol is able to sensitize EOC cells to commonly used chemotherapeutic agents. We describe for the first time the use of a nontoxic compound, capable of simultaneously activating proapoptotic proteins and down-regulating antiapoptotic proteins, and consequently restoring chemosensitivity in EOC cells.

MATERIALS AND METHODS

  1. Top of page
  2. Abstract
  3. MATERIALS AND METHODS
  4. RESULTS
  5. DISCUSSION
  6. REFERENCES

Cell Lines, Drugs, Inhibitors, and Plasmids

Human EOC cell lines A2780 and CP706 were propagated in RPMI plus 10% fetal bovine serum (FBS) (Gemini Bio-Products, Woodland, CA) at 37 °C in a 5% CO2 atmosphere. Primary EOC cells were isolated from malignant ovarian ascites and cultured as previously described.3 Phenoxodiol was obtained from Novogen (Australia). Docetaxel was obtained from Aventis (Paris, France). All other reagents were purchased from Sigma Chemical (St. Louis, MO). Z-VAD-FMK and Z-VDVAD-FMK were obtained from R&D Systems (Minneapolis, MN). FLAG-labeled P-XIAP plasmid was a kind gift from Dr. Jin Cheng7 and transfected using FuGENE 6 Reagent (Roche Applied Science, Indianapolis, IN) according to the manufacturer's instructions.

Xenografts

Mouse xenograft models using CP70 were established as previously described.3 Phenoxodiol was formulated as a suspension in 1% carboxymethyl cellulose (CMC). Carboplatin and gemcitabine were formulated in phosphate-buffered saline (PBS), whereas paclitaxel was formulated in cremaphor/ethanol/PBS (1:1:98, v:v:v). Therapy began 8 days postinoculation and tumor measurements began on Days 12–14, depending on the experiment. Control groups received 1% CMC vehicle. Tumors were measured every third day as previously described.8 The protocol for combination therapy is described independently for each drug in the Results section. Each experiment was performed twice.

Cell Viability Assay

Cell viability was evaluated using the CellTiter 96 Aqueous One Solution Cell Proliferation Assay (Promega, Madison, WI) according to the manufacturer's instructions. The values from the treated cells were compared with the values generated from the untreated cells and reported as percent viability. Each experiment was performed in triplicate.

Protein Preparation

After drug treatment, protein was extracted and measured as previously described.3 For separation of the cytoplasmic and mitochondrial fractions, cell pellets were processed using the ApoAlert Cell Fractionation Kit (BD Biosciences, San Jose, CA) according to the manufacturer's instructions.

SDS-PAGE and Western Blots

Twenty μg protein was denatured in sample buffer (2.5% sodium dodecyl sulfate [SDS], 10% glycerol, 5% β-mercapto-ethanol, 0.15 M Tris, pH 6.8, and 0.01% bromophenol blue) and subjected to 12% SDS polyacrylamide gel electrophoresis (SDS-PAGE) as previously described.3 The following antibodies and concentrations were used: mouse anti-caspase-2 (BD Biosciences, 1:1,000), rabbit anti-Bid (Cell Signaling, Beverly, MA, 1:5,000), mouse anti-XIAP (BD Biosciences, 1:1,000), rabbit anti-phosphorylated Akt (Cell Signaling, 1:1,000), rabbit anti-Akt (Cell Signaling, 1:1,000), mouse anti-Bax (BD Biosciences, 1:500), rabbit anti-actin (Sigma, 1:10,000), rabbit anti-cytochrome c (BD Biosciences, 1:1,000), rabbit anti-HtrA2/Omi (R&D Systems, 1:5,000), mouse anti-Smac/DIABLO (Rockland, Gilbertsville, PA, 1:10,00), mouse anti-Cox-4 (BD Biosciences, 1:500), and mouse anti-FLAG (Sigma, 1:500). These dilutions have been found in our laboratory to be the optimum dilution for each respective antibody. Proteins were observed using enhanced chemiluminescence (Pierce, Rockford, IL).

Caspase-3/7, -8, and -9 Activity Assay

Ten μg of protein in 50 μL total volume was mixed with 50 μL of equilibrated Caspase-Glo 3/7, 8, or 9 reagents (Promega). After incubating at room temperature for 1 hour, luminescence was measured using TD 20/20 Luminometer (Turner Designs, Sunnyvale, CA). Blank values were subtracted and fold-increase in activity was calculated based on activity measured from untreated cells. Each sample was measured in triplicate.

Statistical Analysis

Data are expressed as mean ± standard deviation (SD). Statistical significance (P < 0.05) was determined using both one- and two-way analysis of variance (ANOVA) with Bonferonni correction.

RESULTS

  1. Top of page
  2. Abstract
  3. MATERIALS AND METHODS
  4. RESULTS
  5. DISCUSSION
  6. REFERENCES

Response of Primary EOC Cells to Standard Chemotherapy Agents

Our first objective was to characterize the in vitro response of EOC cells to standard chemotherapy agents. Eight primary EOC cell lines isolated from ascites and two established EOC cell lines, CP70 and A2780 were treated with increasing concentrations of carboplatin (1–100 μg/mL), paclitaxel (0.02–2 μM), and gemcitabine (0.01–0.3 μM) for 24 hours and cell viability was determined by the CellTiter 96 Assay. Table 1 summarizes the IC50 measured for each chemotherapeutic agent. The IC50 for most of the EOC cell lines was ≥ 100 μg/mL, ≥ 2 μM, and ≥ 0.3 μM for carboplatin, paclitaxel, and gemcitabine, respectively, which are equal to or greater than the highest dose used. Thus, in a majority of the cell lines a significant amount of cells remained viable after treatment with the highest dose.

Table 1. In Vitro Response of EOC Cells to Chemotherapy
Cell lineIC50 for carboplatin, μg/mLIC50 for paclitaxel, μMIC50 for gemcitabine, μM
  1. EOC: epithelial ovarian carcinoma.

CP70> 100< 0.20.01
A278050< 0.20.3
R12750< 0.2> 0.3
R179> 100< 2> 0.3
R182100> 2> 0.3
R385> 100> 2> 0.3
R389< 50> 20.1
R395> 100< 20.1
R396< 50> 20.2
R401> 10020.2

Phenoxodiol Chemosensitizes EOC Cells In Vitro

We have previously shown that phenoxodiol is able to induce cell death in EOC cells and that pretreatment with phenoxodiol sensitizes EOC cells to Fas-induced apoptosis.3 Our next objective was to determine whether phenoxodiol could also sensitize EOC cells to commonly used chemotherapeutic agents. Therefore, we tested whether pretreatment of EOC cells with phenoxodiol will affect their response to carboplatin, paclitaxel, and gemcitabine. EOC cells were pretreated with 10 μg/ml of phenoxodiol for 2 hours; afterward, phenoxodiol was removed from the media and the cells were treated with increasing concentrations of carboplatin, paclitaxel, or gemcitabine for an additional 24 hours. This dose of phenoxodiol has been previously determined to be IC50 for most of the tested EOC cells.3, 5, 9 This dose was used in this experiment for a shorter time (2 hr) to determine the chemosensitizing effect of pretreatment. Pretreatment with phenoxodiol sensitized the EOC cells as demonstrated by a significant shift in the IC50 of the three drugs tested. Figure 1 shows the results for A2780, CP70, R127, and R182. Pretreatment of EOC cells for 2 hours with phenoxodiol shifted the IC50 for carboplatin from 50–100 μg/mL to 1–10 μg/mL, whereas the IC50 for gemcitabine shifted from > 0.3 μM to < 0.01 μM (Fig. 1a–d). Likewise, the IC50 for paclitaxel shifted from > 2 μM to < 0.2 μM (Fig. 1e,f). Thus, pretreatment with phenoxodiol sensitizes EOC cells to chemotherapy by lowering the dose necessary to significantly reduce cell viability.

thumbnail image

Figure 1. In vitro sensitizing effect of phenoxodiol on EOC cells. EOC cells were treated with increasing concentrations of (a) carboplatin and (c) gemcitabine and (e) paclitaxel as monotherapy, and with (b) combination phenoxodiol/carboplatin (d) phenoxodiol/gemcitabine. The figure represents the results from 4 of 10 EOC cell lines tested. Each experiment was repeated at least three times.

Download figure to PowerPoint

Phenoxodiol Chemosensitizes EOC Cells In Vivo

Our in vitro results showed that pretreatment with phenoxodiol can sensitize EOC cells. To test this in vivo, we carried out combination therapy studies using a xenograft mouse model injected with EOC cells. The first combination treatment evaluated was phenoxodiol-carboplatin. For this we used the platinum-resistant CP70 cell line. Once the tumors were established, the animals were divided into four groups. Group one received phenoxodiol alone (25 mg/kg orally [p.o.] daily for 12 days), Group two received carboplatin alone (20 mg/kg intraperitoneally [i.p.] every 3 days for 12 days), Group three received the combination phenoxodiol/carboplatin, and Group four (control) received 1% CMC vehicle. Mice treated with phenoxodiol as monotherapy showed a 29% increase in tumor mass compared with the control, whereas mice treated with carboplatin alone showed a 6% decrease in tumor mass (Fig. 2a). A significant decrease in tumor mass (47%) was seen in mice given the phenoxodiol/carboplatin combination (P < 0.01 relative to carboplatin alone).

thumbnail image

Figure 2. In vivo sensitizing effect of phenoxodiol on mouse xenograft model. (a) Effect of combination phenoxodiol/carboplatin on CP70 tumor growth. Tumor mass measured in animals treated with phenoxodiol (Pxd), carboplatin (Carbo), and phenoxodiol/carboplatin combination. *P < 0.01 relative to carboplatin alone. (b) Tumor mass measured in animals treated with phenoxodiol, paclitaxel, and phenoxodiol/paclitaxel combination. **P < 0.001 relative to paclitaxel alone. (c) Tumor mass measured in animals treated with phenoxodiol, gemcitabine, and phenoxodiol/gemcitabine combination. *P < 0.0004 relative to gemcitabine alone. Tumor volume was determined after 10 days of treatment. Red arrows indicate the location of the tumor. Data represent the mean ± SD of two independent experiments. Six to eight animals were tested in each group.

Download figure to PowerPoint

We next examined the effect of phenoxodiol/paclitaxel combination using the A2780 tumor model. For combination phenoxodiol/paclitaxel, mice were treated with phenoxodiol alone (10 mg/kg p.o. daily for 12 days), paclitaxel alone (2 mg/kg i.p. every 3 days for 12 days), or the combination phenoxodiol/paclitaxel. Control animals received vehicle alone. Treatment with phenoxodiol alone induced a 20% decrease in tumor mass compared with the control. Treatment with paclitaxel alone induced a 35% decrease in tumor mass, whereas the combination phenoxodiol/paclitaxel induced a 74% decrease in tumor mass (P < 0.001, relative to paclitaxel alone) (Fig. 2b).

The CP70 tumor model was also used to examine the effect of phenoxodiol/gemcitabine combination treatment. Mice were treated with either phenoxodiol alone, gemcitabine alone, or the combination phenoxodiol/gemcitabine. Mice treated with the combination therapy demonstrated significant tumor reduction compared with monotherapy (Fig. 2c).

Taken together, these results show that treatment with combination phenoxodiol/carboplatin, phenoxodiol/paclitaxel, and phenoxodiol/gemcitabine results in significant reduction of tumor size compared with monotherapy.

Phenoxodiol Induces Apoptosis in EOC Cells

It is well established that the induction of the apoptotic cascade is one of the main mechanisms of chemotherapy-induced cell death.1, 2 To determine whether the chemosensitizing effect of phenoxodiol demonstrated above is secondary to its ability to activate the apoptotic cascade, EOC cells were treated with monotherapy phenoxodiol (10 μg/ml), paclitaxel (2 μM), docetaxel (50 ng/ml), or carboplatin (50 μg/mL) for 24 hours. These doses are lower than the IC50 of each drug for the EOC cells tested (Table 1).9 A parallel experiment was performed where EOC cells were pretreated for 2 hours with 10 μg/mL of phenoxodiol before 24-hour treatment with the same doses of paclitaxel, docetaxel, or carboplatin. For the combination treatments, phenoxodiol was completely removed from the culture before treatment with the succeeding drug. The activation status of the main effector caspase, caspase-3, was evaluated using Western blot analysis and its activity was measured using the Caspase-Glo 3/7 Assay. Figure 3 shows the results for R182. Monotherapy with paclitaxel, docetaxel, or carboplatin did not result in the activation of caspase-3, as indicated by the absence of the p17 and p19 active forms of caspase-3 in the Western blots (Fig. 3a, lanes 6–8). In addition, low levels of caspase-3 activity were measured in the Caspase-Glo 3/7 Assay (Fig. 3b). However, in cells treated with phenoxodiol alone for 24 hours (Fig. 3a, lane 2) and in cells pretreated with phenoxodiol for 2 hours before treatment with paclitaxel, docetaxel, or carboplatin (Fig. 3a, lanes 3–5), the active forms of caspase-3 were detected and a significant increase in caspase-3 activity was observed (Fig. 3b).

thumbnail image

Figure 3. Effect of phenoxodiol pretreatment on caspase-3 activation and XIAP expression. EOC cells were treated as described in the Results section. Protein extracts were evaluated for (a) the expression of active caspase-3 and XIAP by Western blot analysis and (b) activity of caspase-3 by Caspase-Glo 3/7 assay. Data represent the mean ± SD of four independent experiments. *P < 0.000001 relative to each compound as monotherapy. The figure represents the results for R182 cells. Similar results were found with other EOC cell lines. Lanes 1–9: phenoxodiol, phenoxodiol/paclitaxel, phenoxodiol/carboplatin, phenoxodiol/docetaxel, carboplatin, paclitaxel, docetaxel, positive control. Phx, phenoxodiol; Pac, paclitaxel; Carbo, carboplatin; Doc, docetaxel.

Download figure to PowerPoint

We also looked at the activation status of the antiapoptotic protein XIAP. Western blot analysis showed no change in the status of XIAP after monotherapy with paclitaxel, docetaxel, or carboplatin (Fig. 3a, lanes 6–8). However, a decrease in the p45 XIAP was seen in cells that were treated with phenoxodiol alone and in cells that received combination therapy (Fig. 3a, lanes 3–5). In addition, we observed the appearance of a p30 XIAP fragment in these cells, which corresponded with the activation of caspase-3.

Phenoxodiol-Induced Apoptosis Involves the Mitochondrial Pathway

Our next objective was to determine the molecular mechanisms involved in phenoxodiol-induced apoptosis. Thus, EOC cells were treated with phenoxodiol for different timepoints and the status of the apoptotic cascade was evaluated. The activity of the initiator caspases-8 and -9 and the effector caspase-3 was measured using the Caspase-Glo 8, 9, and 3/7 Assays, respectively, whereas the activation of caspase-2 was detected using Western blot analysis. The status of the proapoptotic proteins Bid, Bax, cytochrome c, Smac/DIABLO, and Omi/HtrA2 and the antiapoptotic proteins XIAP and Akt was likewise determined by Western blot analysis. Figure 4 shows the results for CP70 cells. Phenoxodiol is able to engage the apoptotic pathway in EOC cells as early as 4 hours posttreatment. At this time, we observed the activation of caspase-2 and Bid, the translocation of cytochrome c to the cytoplasm, and a decrease in p45 XIAP (Fig. 4a,c). At 8 hours posttreatment, we observed the translocation of Smac/DIABLO and Omi/HtrA2 and a decrease in the phosphorylated form (p-Akt) and total form of Akt (t-Akt). At 16 hours posttreatment, we saw the cleavage of Bax, significant increase in the activity of caspases-8, -9, and -3, and the appearance of p30 XIAP (Fig. 4a,b).

thumbnail image

Figure 4. Effect of phenoxodiol treatment on the apoptotic cascade. CP70 cells were treated with phenoxodiol (10 μg/mL) for different periods of time. Lysates were analyzed for (a) the expression of both pro- and antiapoptotic proteins by Western blot analyses and (b) activity of caspases-8, -9, and -3 using the Caspase-Glo 8, 9, 3/7 assays, respectively. (c) A parallel experiment was performed where the cytoplasmic and mitochondrial fractions were separated before Western blot analyses. Cox-4 was included to show the integrity of the cytoplasmic and mitochondrial preparations. The figures are representative results from CP70 cells. Similar results were observed with other EOC cells tested.

Download figure to PowerPoint

Phenoxodiol-Induced Apoptosis Involves the Simultaneous Activation of Caspase-2 and Down-regulation of XIAP

To dissect the early steps in phenoxodiol-induced apoptosis, we then focused our attention on the four proteins that were affected earliest: caspase-2, Bid, cytochrome c, and XIAP. We hypothesize that phenoxodiol first induces the activation of caspase-2, which then activates Bid. Bid then activates the mitochondrial pathway and induces the release of cytochrome c. We also hypothesize that the decrease in p45 XIAP is coming from a signal, independent from the one activating the mitochondrial pathway. Thus, EOC cells were treated with phenoxodiol for 4 hours in the presence or absence of the pan-caspase inhibitor, ZVAD-FMK, and the activation status of Bid and XIAP were determined by Western blot analysis. Figure 5 shows the results for CP70 cells. In the presence of the pan-caspase inhibitor, Bid's activation is inhibited, whereas the decrease in p45 XIAP was not affected (Fig. 5). This result showed preliminary evidence that indeed the signal that leads to the activation of the mitochondria is different from the signal that induces the decrease in p45 XIAP.

thumbnail image

Figure 5. Effect of the pan-caspase inhibitor ZVAD-FMK on phenoxodiol-induced XIAP down-regulation and Bid activation. EOC cells were treated with phenoxodiol (10 μg/mL × 4 hrs), in the presence or absence of the pan-caspase inhibitor Z-VAD-FMK (20 μM). Expression of XIAP and Bid were determined by Western blot analyses. The figure represents the results from CP70 cells. Pxd, phenoxodiol.

Download figure to PowerPoint

Phenoxodiol-Induced Apoptosis Depends on the Activation of Caspase-2

Because we hypothesize that caspase-2 acts upstream of Bid, we then treated EOC cells with phenoxodiol for 24 hours in the presence or absence the specific caspase-2 inhibitor, Z-VDVAD-FMK, and the status of caspase-2, Bid, and XIAP was determined by Western blot analysis. Our results showed that in the presence of the caspase-2 inhibitor, phenoxodiol-induced Bid activation is inhibited, whereas the decrease in p45 XIAP was not affected (Fig. 6a). In addition, in the presence of the caspase-2 inhibitor, phenoxodiol-induced activation of caspases-8, -9, and -3 were significantly inhibited (Fig. 6b). Taken together, these results show that phenoxodiol-induced apoptosis depends on the initial activation of caspase-2, and that indeed there are separate signals activating the mitochondrial pathway and inducing the decrease in p45 XIAP.

thumbnail image

Figure 6. Phenoxodiol-induced apoptosis is caspase-2 dependent. EOC cells were treated with phenxodiol (10 μg/mL × 24 hrs), in the presence or absence of the specific caspase-2 inhibitor Z-VDVAD-FMK (25 μM). (a) Activation status of caspase-2 and Bid and levels of XIAP were determined by Western blot analysis. (b) Activities of caspases-8, -9, and -3 were measured by the Caspase-Glo assay. The figure represents the results from R179 cells. Pxd, phenoxodiol.

Download figure to PowerPoint

Phenoxodiol-Induced Apoptosis Depends on the Early Proteasome Degradation of XIAP

It has been shown that XIAP is a target for proteasome degradation.7 Because our results showed that the decrease in p45 XIAP, seen 4 hours posttreatment, was caspase-independent (Fig. 5), we hypothesized that phenoxodiol is inducing the proteasomal degradation of XIAP at around the same time that caspase-2 is activated (Fig. 4a). XIAP has been shown to be a substrate of Akt and that phosphorylated XIAP cannot be targeted by the proteasome.7 Thus, EOC cells were transfected with a phospho-mimic form of XIAP (P-XIAP), which is resistant to proteasomal degradation, and then treated with 10 μg/ml phenoxodiol for 4 hours and the status of XIAP was determined by Western blot analysis. A parallel experiment was performed with transfected cells that were treated with phenoxodiol for 24 hours. Because caspase-3 activation was observed after 16 hours posttreatment with phenoxodiol (Fig. 4b), the 24-hour timepoint allowed us to evaluate the impact of inhibiting the proteasomal degradation of XIAP on the full activation of apoptosis by measuring caspase-3 activity. Wildtype EOC cells served as control. Figure 7 shows the results for R179. In wildtype R179 cells, there was 40% decrease in p45 XIAP after 4 hours of treatment with phenoxodiol. However, in cells transfected with P-XIAP there was only a 20% decrease in p45 XIAP (Fig. 7a,b). The level of activation of caspase-3 was also different between the wildtype and transfected EOC cells. In wildtype cells, a 10-fold increase in caspase-3 activity was observed after phenoxodiol treatment (Fig. 7c). In the transfected cells, however, caspase-3 activity increased only 3-fold. These results suggest that the phenoxodiol-induced decrease in p45 XIAP is proteasome-dependent and that the full induction of the apoptotic cascade depends not only on the activation of caspase-2 but also on the simultaneous proteasomal degradation of XIAP.

thumbnail image

Figure 7. Phenoxodiol-induced down-regulation of XIAP is proteasome-dependent. Wildtype EOC cells and EOC cells transfected with P-XIAP were treated with phenoxodiol (10 μg/mL) for 4 hours and 24 hours. (a) Levels of FLAG-labeled-P-XIAP and status of XIAP were determined by Western blot analysis. (b) Densitometer readings were measured to determine the percentage decrease in XIAP. (c) Activity of caspase-3 was measured using the Caspase-Glo 3/7 assay. Wt, wildtype, P-XIAP, phospho-mimic XIAP, Pxd, phenoxodiol.

Download figure to PowerPoint

DISCUSSION

  1. Top of page
  2. Abstract
  3. MATERIALS AND METHODS
  4. RESULTS
  5. DISCUSSION
  6. REFERENCES

Chemoresistance is a major hurdle in the treatment of ovarian carcinoma. Cancers that initially respond to standard chemotherapy often recur, with selective outgrowth of a tumor cell subpopulation that is resistant to further treatment. These tumors frequently display resistance, not only to the original agent, but also to a variety of functionally and structurally unrelated compounds. Thus, a drug that can resensitize these cancer cells to already commonly used agents will be beneficial, especially to patients with recurrent/refractory disease. The present study demonstrates how a drug like phenoxodiol, which can immediately and simultaneously activate proapoptotic proteins and down-regulate antiapoptotic proteins, can sensitize chemoresistant EOC cells. Phenoxodiol not only activated the caspases but also concomitantly promoted the down-regulation of XIAP and Akt (Fig. 4a-b), thereby potentiating the cytotoxic effects of chemotherapy. To our knowledge, this is the first report of a compound's capability to function as a chemosensitizer both in vitro (Fig. 1) and in vivo (Fig. 2).

The animal studies presented herein confirmed our in vitro results, which showed that phenoxodiol could chemosensitize EOC cells to commonly use chemotherapeutic agents. Pretreatment with phenoxodiol not only enhanced the cytotoxic effect of carboplatin, paclitaxel, and gemcitabine but also reduced the dose of the drug necessary to obtain an effective antitumoral effect (Fig. 2a–c). A relevant aspect in order for a compound to act as a chemosensitizer is to show lack of toxicity. We found in the animal studies that phenoxodiol is well tolerated when administered p.o., i.v., or i.p. This was confirmed in humans, where we found minimal treatment-related toxicity after weekly administration of phenoxodiol.10 A current clinical trial is ongoing to further validate the utility of phenoxodiol in the clinical setting.

Our results show that phenoxodiol is able to fully activate the apoptotic pathway, as evidenced by the activation of the main effector caspase, caspase-3, in the same cells in which carboplatin, paclitaxel, and docetaxel failed (Fig. 3). This suggests that the apoptotic pathway is functional in these EOC cells but is activated only in response to phenoxodiol. Recent studies from our group and others suggest that intracellular inhibitors of caspases are main players in drug-induced apoptosis.3, 9, 11–13 Thus, we sought to characterize the effects of phenoxodiol not only on caspase activation but also on the status of antiapoptotic proteins.

We showed so far that phenoxodiol-induced apoptosis begins with caspase-2-dependent Bid activation (4 hrs posttreatment; Fig. 4a), which engages the mitochondrial pathway, allowing the release of cytochrome c, Smac/DIABLO, and Omi/HtrA2 (Fig. 4c). Therefore, we propose that these initial events represent the main changes that will lead to the activation of caspase-9, followed by the activation of caspase-3. At a later time the cascade is amplified through Bax, which results in an even more significant increase on the activities of caspases-8, -9, and -3 (Fig. 4a,b).

We also looked at the status of the antiapoptotic proteins XIAP and Akt after phenoxodiol treatment. Previously, we showed that phenoxodiol-induced apoptosis in EOC cells involves XIAP down-regulation.3, 5 After our initial findings, we further characterized the molecular pathway mediating this effect. Interestingly, we observed two different processes in XIAP down-regulation. First, we saw an early effect (4 hrs posttreatment) characterized by a decrease in the level of the p45 XIAP (Fig. 4a). This effect is caspase-independent because the decrease was not blocked by the pan-caspase inhibitor Z-VAD-FMK (Fig. 5). A potential mechanism mediating this effect may be XIAP self-ubiquitination and proteasomal degradation, as previously described.7 To test this, we transfected EOC cells with P-XIAP, a form of XIAP which is resistant to proteasomal degradation.7 Our results confirmed our hypothesis and showed that the early effect of phenoxodiol on XIAP, characterized by the decrease in the p45 XIAP, is proteasome-dependent (Fig. 7a,b). In addition, we showed that phenoxodiol-induced apoptosis not only depends on caspase-2 activation (Fig. 6), but also on this early degradation of XIAP (Fig. 7).

The second effect of phenoxodiol on XIAP is its cleavage and the appearance of its p30 fragment, occurring 16 hours posttreatment (Fig. 4a). XIAP has been previously shown to be a substrate of caspases-8, -9, -3, -6, and -7 and that this reaction produces a p30 XIAP cleaved fragment corresponding to the BIR3 and RING domains of XIAP.14, 15 In addition, the p30 XIAP fragment has been shown to be a product of cleavage by Omi/HtrA2.16 Our previous studies demonstrated that the appearance of the p30 XIAP fragment after phenoxodiol treatment is blocked in the presence of the pan-caspase inhibitor ZVAD-FMK.3 In the current study, we show that 1) relocation of Omi/HtrA2 to the cytoplasm (8 hrs posttreatment) precedes the appearance of p30 XIAP (seen at 16 hrs posttreatment); 2) relocation of Omi/HtrA2 (8 hrs posttreatment) occurs after caspase-2 induced activation of the mitochondrial pathway (4 hrs posttreatment); and 3) the levels of p30 XIAP fragment after phenoxodiol treatment was reduced in the presence of caspase-2 inhibitor (Fig. 6a). Taken together, these results suggest that the p30 XIAP fragment may be the product of Omi/HtrA2 and that the Omi/HtrA2-induced XIAP cleavage depends on the activation of the caspase-2 and Bid-signaling pathway. This hypothesis is currently under investigation in our laboratory together with the possibility that caspase-2 may also directly induce XIAP cleavage.

Another interesting observation is that significant caspase-8, -9, and -3 activation correlates with the appearance of p30 XIAP (both occurring 16 hrs posttreatment). Johnson et al.17 previously suggested that this p30 XIAP cleaved fragment may act as a dominant-negative inhibitor of normal XIAP function, and therefore allow the full activation of the caspases. Whether the p30 XIAP cleaved fragment is the cause or effect of significant caspase activation remains to be determined.

We also observed that phenoxodiol treatment not only induces the down-regulation of antiapoptotic XIAP, but also induces the down-regulation of both p-Akt and t-Akt. Akt can be degraded by the proteasome,18 cleaved by caspases, or down-regulated as a consequence of a decrease in XIAP.19 Our results show that phenoxodiol-induced down-regulation of Akt (8 hrs posttreatment) occurs after the decrease in XIAP but before the full activation of caspases (Fig. 4a,b). The molecular mechanisms of phenoxodiol-induced Akt down-regulation and its impact on phenoxodiol-induced apoptosis remains to be determined.

Aguero et al.20 recently confirmed our previous findings that phenoxodiol is able to induce the apoptotic cascade, and in addition showed that phenoxodiol is able to arrest head and neck squamous cell carcinoma in G1 through the induction of p21 and inhibition of cdk2. They reported that phenoxodiol is able to exert these effects beginning 3 hours posttreatment. Combined with our recent findings, these results provide a detailed picture of the molecular mechanisms required for a chemosensitizer. Phenoxodiol, by 1) activating the caspase cascade, 2) inactivating the antiapoptotic proteins, and 3) inducing G1 arrest, all within 4 hours posttreatment, is able to create a pro-death environment, which significantly lowers the dose of cytotoxic drugs needed to significantly reduce cell viability in vitro and tumor size in vivo.

Although we have not yet defined the target molecule for phenoxodiol, the biological relevance of phenoxodiol action in ovarian carcinoma is its capacity to disrupt the intracellular protein circuit comprised of the survival and apoptotic factors. Phenoxodiol, when dosed orally at 50 mg/kg, reaches serum concentration of 250 μM after 30 minutes of administration (our unpublished data). Thus, the dose used in this study (10 μg/mL = 41.62 μM) is readily attainable in vivo.

The complexity of tumor biology requires a better approach for its treatment. The combination of factors that may act on the apoptotic pathway together with the pharmacologic stress/damage induced by the cytotoxic drugs could represent a better and more efficient approach to cancer treatment.

REFERENCES

  1. Top of page
  2. Abstract
  3. MATERIALS AND METHODS
  4. RESULTS
  5. DISCUSSION
  6. REFERENCES