Introduction

Ovarian cancer (OC), primarily epithelial OC, remains the most deadly gynecologic malignancy in the Western world. The incidence of this disease rises after women reach menopause and may be related to declines in specific reproductive hormones such as progesterone (P4) (Cramer et al., 1983; Riman et al., 1998). Conversely, increased exposure to pregnancy levels of progesterone appears to offer protection, since OC risk is reduced in women with higher numbers of births or a history of twin pregnancies (which is associated with higher P4 levels) (Adami et al., 1994, 1995; Lambe et al., 1999). The protective effect of pregnancy may be attributable to removal of genetically damaged cells from the ovarian surface epithelium (OSE) via P4-induced apoptosis. Consistent with this theory, P4 or other progestins have been shown to inhibit cell growth and/or induce apoptosis in normal and malignant human ovarian surface epithelial (HOSE) cells in vitro and in vivo (Bu et al., 1997; Hu and Deng, 2000; Syed et al., 2001; Yu et al., 2001; Rodriguez et al., 2002).

Apoptosis is a well-orchestrated, autonomous mode of cell death that eliminates superfluous, damaged, mutated, or aged cells (Zimmermann et al., 2001). The process is mediated by the action of a class of enzymes known as caspases (cysteine aspartate-specific proteases) (Alnemri et al., 1996; Earnshaw et al., 1999; Zou et al., 1999) that may be classified broadly into apoptotic initiators (caspase-2, -8, -9, and -10) and executioners (e.g., caspase-3, -6, and -7). The initiators are responsible for activating the executioners, which ultimately dismantle cells by cleaving key cellular and nuclear proteins. Two major apoptotic pathways utilizing different initiator caspases have been identified. One involves perturbation of the mitochondria, release of mitochondrial cytochrome C to the cytosol, formation of the Apaf-1 and caspase-9 complex, and subsequent activation of the enzyme and its downstream executioner caspases (caspase-3, -6, and -7) (Zou et al., 1999). The other is set off by the binding of specific ligands to cell surface 'death receptors' of the tumor necrosis factor (TNF)-receptor family (Ashkenazi and Dixit, 1998). Such an interaction results in the conversion of initiator procaspases into active enzymes, primarily caspase-8 and -10, and subsequent activation of executioner caspases. Among the TNF-receptor family, Fas (APO-1/CD95) is recognized as an important death receptor that transmits extracellular apoptotic signals intracellularly (Nagata, 1994; Song et al., 2000). Activation of the receptor is accomplished by binding of the Fas ligand (FasL), but could also be achieved by interaction with an anti-Fas antibody (Yonehara et al., 1989; Suda et al., 1993). The Fas/FasL system is considered a prime mediator of therapeutic cell killing in a variety of cancer cells, including those from OC (Ghahremani et al., 1998). In this regard, activation of caspase-8 has been recently identified as an important link to Fas-mediated apoptosis in two OC cell lines (Hayakawa et al., 2002).

We previously observed marked inhibition of cell growth when primary cultures of HOSE cells, immortalized HOSE, or OC cell lines were exposed to pregnancy-equivalent doses (10^-7–10^-6 M) of P4 (Syed et al., 2001). However, no information is yet available on how P4 triggers apoptosis in normal or malignant HOSE cells. The goal of this study is to determine (1) which initiator caspase, and (2) whether the mitochondria- or the death receptor-mediated pathway is involved in the P4-induced apoptosis in HOSE or OVCA cell lines.

Results

Induction of apoptosis in HOSE and OVCA cell cultures by P4

Treatment of the immortalized normal (HOSE 642, HOSE 12-12) and malignant (OVCA 429, OVCA 432) HOSE cell lines with 10^-6 M P4 induced a time-dependent increase in cells undergoing early and late apoptosis (Figure 1). The percentages of cells in early apoptosis (annexin V-positive/propidium iodide (PI)-negative) reached 13–18% after 6 h, peaked at 28–46% at 18 h, and declined to 10–20% by 36 h post-treatment. The percentages of cells in late apoptosis (annexin V- and PI-positive) continued to rise with increasing duration of hormone exposure from 2 to 8% at 6 h, 20 to 40% at 18 h, and 60 to 75% at 36 h.

Figure 1.

Time course of progesterone-induced apoptosis of normal HOSE and OVCA cell lines. Two normal HOSE (HOSE 642 and HOSE 12-12) and two OVCA (OVCA 429 and OVCA 432) cell lines were treated with 10^-6 M progesterone for 6, 18, and 36 h. Cells were stained with annexin V (gray bars) and PI (black bars), and the percentage of cells staining positive was analysed by flow cytometry. Open bars represent control cells for each time point. Data are the means of two experiments. Error bars represent means.e.m.

Full figure and legend (99K)

Progesterone-induced caspase activity in HOSE and OVCA cells

To identify the caspases involved in P4-induced apoptosis, we measured the enzymatic activity of caspase-3, -8, and -9 against synthetic caspase subtype-specific tetrapeptide substrates DEVD p-NA (for caspase-3), IETD p-NA (for caspase-8), and LEHD p-NA (for caspase-9), after 3 and 6 h of exposure of HOSE and OVCA cells to the hormone. The activities of caspase-8 and -3 in cultures of two HOSE cell lines were significantly higher (2–3.5-fold) than activities found in untreated cultures (Figure 2a). Similarly, exposure of OVCA cell lines to P4 resulted in a 1.5–3-fold increase in caspase-8 and -3 compared with their respective untreated controls. No significant changes in the activity of caspase-9 were detected in HOSE or OVCA cells following hormonal treatment (Figure 2a). We next investigated whether expression of the three caspases at the protein level was altered by P4 treatment. Western blot analyses revealed significant cleavage of caspase-8 and -3 in HOSE and OVCA cultures following P4 treatment, whereas the levels of caspase-9 in all four cell lines remained unchanged by the hormone (Figure 2b).

Figure 2.

Effect of progesterone on caspase-3, -8, and -9 in HOSE and OVCA cells. (a) Enzyme activities of cell lysates toward tetrapeptide caspase substrates in HOSE (HOSE 642 and HOSE 12-12) and OVCA (OVCA 429 and OVCA 432) after 3 and 6 h of exposure to 10^-6 M progesterone. The chromogenic substrates were DEVD-pNA (caspase-3), IETD-pNA (caspase-8), and LEHD-pNA (caspase-9). The caspase activity is expressed as fold relative to untreated controls and represented as the means of two experiments carried out in triplicate. Error bars represent means.e.m. Statistically significant increases in levels of caspases are shown by an asterisk (P<0.001–0.02) (b) Immunoblot analysis of caspase-3, -8, and -9 in HOSE and OVCA cells. Proteins were separated by electrophoresis, and immunoblots were probed with caspase-3, -8, and -9 antibodies. GAPDH was used as a loading control. A representative experiment of at least two others is presented, all with similar results

Full figure and legend (170K)

Effect of a caspase inhibitor on P4-induced apoptosis

We sought to assess the effect of a general caspase inhibitor, Z-VAD, on P4-induced apoptosis in HOSE and OVCA cell cultures. Addition of increasing concentrations of Z-VAD to HOSE or OVCA cell cultures exposed to 10^-6 M of P4 reversed the growth-inhibitory effects of the hormone in a concentration-dependent manner (Figure 3). This finding further supports the involvement of caspases in P4-induced apoptosis in normal and malignant HOSE cells.

Figure 3.

Effects of the general caspase inhibitor Z-VAD on progesterone-induced apoptosis. The HOSE and OVCA cells were pretreated with various concentrations of Z-VAD for 30 min and then treated with vehicle (open bars) or 10^-6 M progesterone (solid gray bars) for 4 days. Cell number was assessed by MTT assay as described in Materials and methods. Absorbance of wells not exposed to hormones was arbitrarily set as 1, and progesterone-treated cell growth was expressed as fold increase/decrease relative to control. Two experiments were performed in triplicate. Error bars represent means.e.m. Statistically significant changes are shown by asterisk (P<0.002–0.01)

Full figure and legend (111K)

Effects of P4 on FasL and Fas mRNA and protein expression in HOSE and OVCA cell lines

Since activation and enhanced expression of caspase-8 were associated with P4-induced apoptosis in HOSE and OVCA cells, next it is logical for us to determine whether P4-induced apoptosis is associated with alterations of Fas and/or FasL expression. Semiquantitative RT–PCR analyses of total cellular RNA prepared from immortalized normal HOSE cell lines (HOSE 642 and HOSE 12-12) and OVCA cell lines (OVCA 429, OVCA 432) revealed that Fas mRNA and protein were present in both normal and malignant HOSE cell lines at comparable levels, which remained unchanged following P4 treatment (Figure 4a and b). In contrast, expression of FasL, at both the transcript and protein levels, was altered by P4 in HOSE and OVCA cell lines, although in opposite ways (Figure 4a and b): P4 stimulated the expression of FasL in normal HOSE cell lines, whereas the hormone inhibited the expression of this death receptor ligand in OVCA cells. Differential regulation of FasL between HOSE and OVCA cells was consistently observed at the transcript and protein levels.

Figure 4.

Levels of Fas and FasL messenger RNA (mRNA) and protein in HOSE and OVCA cells. (a) Total RNA (1 g) was isolated from HOSE and OVCA cells treated with progesterone (10^-6 M) or with vehicle for 5 days, reverse-transcribed, and amplified by polymerase chain reaction (PCR). Products were separated by electrophoresis in a 2% agarose gel. A representative gel is shown in the upper panel (a). The reverse transcription (RT)–PCRs were performed from three total RNA preparations of each cell line. Relative mRNA levels are expressed as arbitrary units after normalization with GAPDH (b, lower panel). (b) Immunoblots were prepared as described in the legend of Figure 2 and probed with anti-Fas and anti-FasL antibodies. A representative blot is shown in the upper panel (b). Blots were run using three different cellular extracts of each cell line. Relative protein levels are expressed as arbitrary units after normalization with GAPDH. Open bars=untreated HOSE, solid bars=progesterone-treated HOSE, hatched bars=untreated OVCA, gray bars=progesterone-treated OVCA. Data are the means of three experiments. Error bars represent means.e.m. Statistically significant increases in levels of FasL mRNA and protein are shown by an asterisk (P<0.001–0.03). Statistically significant increases in levels of FasL mRNA and protein between untreated HOSE and OVCA are shown by dot (P<0.001–0.02)

Full figure and legend (158K)

Interaction of P4 and Fas/FasL system in the induction of cell loss in HOSE and OVCA cultures

To determine whether the P4-induced cell kill in HOSE and OVCA cell cultures shares a common pathway with the Fas/FasL system, we treated cells singularly or jointly with P4 and an activating anti-Fas antibody. Treatment of cells separately with the anti-Fas antibody at 0.5 g/ml or with progesterone at 10^-6 M decreased HOSE and OVCA cell growth, but the effect of P4 was greater than that induced by the Fas-antibody. Treatment of cells jointly with the two inducers produced the maximal cell kill effect in HOSE and OVCA cells. Under all treatment conditions, the OVCA cells appeared to be less sensitive than the HOSE cells to P4 and the anti-Fas antibody, whether administered alone or conjointly (Figure 5a).

Figure 5.

Effects of an activating anti-Fas antibody and a blocking anti-FasL antibody on progesterone-induced apoptosis of HOSE and OVCA cell lines. The HOSE and OVCA cells were treated either with (a) 0.5 g/ml of a monoclonal activating anti-Fas (anti-Fas mAb) or (b) pretreated with 100 and 500 ng/ml of a blocking anti-FasL antibody for 30 min or with their isotype-matched control monoclonal antibody (control mAb) in the presence or absence of 10^-6 M progesterone. Cell number was assessed by MTT assay as described in Materials and methods. Absorbance of wells not exposed to hormones was arbitrarily set as 1, and progesterone-treated cell growth was expressed as fold increase/decrease relative to control. The experiment was performed in triplicate. Error bars represent means.e.m. Statistically significant changes are shown by asterisk (P<0.002–0.01)

Full figure and legend (227K)

Inhibition of P4-induced cell loss in HOSE and OVCA cultures by a blocking anti-FasL antibody

To confirm that the Fas/FasL system is involved in the P4-induced apoptosis of HOSE and OVCA cells, we incubated the cells with a blocking anti-FasL antibody 30 min before P4 treatment. Cells treated with P4 alone showed a significant decrease in cell numbers (Figure 5b); however, this hormonal effect was inhibited in a dose-dependent manner when cells were cotreated with a blocking anti-FasL antibody.

Discussion

Previous studies, including our own, have demonstrated the induction of growth inhibition or apoptosis by P4 in OC cell lines (Bu et al., 1997; Hu and Deng, 2000; Syed et al., 2001; Yu et al., 2001). We recently extended these observations to normal primary and immortalized HOSE cell cultures and demonstrated that high doses of P4 consistently inhibited cell growth in HOSE cells (Syed et al., 2001). In the present study, we definitively showed that induction of apoptosis is the mechanism by which P4 inhibits growth in HOSE and OVCA cell cultures. Early and late apoptotic cell numbers increased with time following P4 treatment. Most importantly, we found that induction of apoptosis by P4 in these cells utilized the caspase-8 initiator pathway rather than the mitochondrion-related caspase-9 pathway. These findings led us to uncover a previously unsuspected connection between P4, caspase-8 activation, and the cell-surface Fas/FasL signaling pathway. Finally, we were able to demonstrate direct regulation of FasL by P4 in HOSE and OVCA cell lines, although the direction of change was opposite.

During ovulation, OSE exfoliated from the dome of ovulatory follicles is repaired by generative stem cell replication, followed by migration of cells from the edges to the wound (Murdoch and McDonnel, 2002). The theory of incessant ovulation (Cramer and Welch, 1993) as a cause of ovarian carcinogenesis argues that successive bouts of ovulation-induced damage and repair-associated proliferation of the OSE creates genomic instability in some OSE cells (Cramer and Welch, 1993). Subsequent clonal expansion of OSE cells with unrepaired genomic errors increases the risk for OC. Yet, through the apoptotic action of P4, premalignant and early malignant cells could be eliminated from the OSE and thus reduce the risk of neoplastic transformation. The latter hypothesis is consistent with epidemiologic findings that higher parity, twin pregnancy, and pregnancy at advanced age lower OC risks (Adami et al., 1994, 1995; Lambe et al., 1999). Experimental studies, including the present report, provide additional support that P4, through induction of apoptosis in normal OSE and/or in OC cells, confer protection against OC (Bu et al., 1997; Hu and Deng, 2000; Syed et al., 2001; Yu et al., 2001; Rodriguez et al., 2002). An analogous scenario may also exist for the endometrium, since clinical and epidemiological findings suggest that exfoliation of premalignant and/or malignant cells from the endometrium via P4 action could reduce endometrial cancer risk (Akhmedkhanov et al., 2001).

Changes in the expression of Fas and its ligand have been demonstrated in various types of cancer to be a mechanism for escape by tumor cells from proapoptotic signals and a possible mechanism by which they gain resistance to chemotherapeutic agents (O'Connell et al., 1996; Pinkoski and Green, 2000; Wajant 2002; ). Das et al. (2000) reported a significant decrease in the expression of Fas in ovarian, cervical, and endometrial carcinoma as compared with their normal counterparts. Decreased Fas expression was also detected in breast, esophageal, hepatocellular, colon, and lung carcinomas (Moller et al., 1994; Keane et al., 1996; Strand et al., 1996; Gratas et al., 1998). Thus, it appears that loss of Fas expression occurs in a wide variety of cancers. However, in the current study, we found that levels of Fas mRNA and protein did not differ in HOSE and OVCA cell lines and were unaffected by P4. This discrepancy might be due to the differences between Fas gene regulation in cell cultures and regulation in vivo. Nevertheless, it is important to note that HOSE and OVCA cell lines expressing significant amounts of Fas are able to respond to an activating anti-Fas antibody to undergo cell death. Conversely, treatment of cells with a blocking anti-FasL antibody prevented about 80% of P4-induced cell loss. Collectively, these findings provide strong evidence in support of activation of the Fas/FasL pathway as a crucial player in P4-induced apoptosis.

Few previous studies have focused on a potential connection between steroid hormones and Fas/FasL regulation, but evidence of such a connection has recently emerged. In normal endometrial glandular cells, estradiol, and P4 were shown to stimulate FasL expression (Selam et al., 2001). Ironically, withdrawal of estrogen and/or P4 from endometrial cells in culture also elevates Fas and FasL expression and induces apoptosis, which could be blocked by treatment of cells with an anti-FasL antibody (Song et al., 2002). Similar to normal endometrial cells, withdrawal of P4 from an endometrial cancer cell line (Ishikawa cells) stimulates Fas and FasL expression that were attended by increased apoptotic activity (Wang et al., 2003). Taken together, these studies suggest that both normal and malignant endometrial cells are dependent on estradiol/P4 for survival and withdrawal from sex steroid-support causes apoptosis via Fas/FasL signaling. To our knowledge, regulation of Fas and FasL by progesterone in normal and malignant ovarian surface epithelial cells has not been studied. The current report is the first to study steady-state Fas and FasL mRNA and protein levels, in relation to apoptosis, in HOSE and OVCA cell lines. We found that exposure of HOSE and OVCA cells to P4 does not alter levels of Fas mRNA and protein, but the hormone enhances levels of FasL mRNA and protein in HOSE cells and lowers their levels in OVCA cells. The mechanisms underlying opposite directional regulation of FasL transcripts/protein by P4 between HOSE and OVCA cells is currently unknown, but could be caused by a 'switch' in expression of transcriptional factors and/or their coregulators during neoplastic transformation. Most importantly, exposure of HOSE or OVCA cells to an activating anti-Fas antibody induced cell loss, whereas treatment of cells with a blocking anti-FasL antibody reduced the P4-induced cell loss. Collectively, these data suggest that P4 is an apoptosis-inducing agent for both HOSE and OVCA cells and its action is likely mediated via induction in a change in Fas to FasL ratio in these cell lines. It has been shown that the ratio of FasL to Fas plays a very important role in tumor progression (Pinkoski and Green 2000; Reimer et al., 2000; Wajant, 2002). Other recent reports pertinent to our findings include a recent demonstration of upregulation of FasL expression in normal ovarian epithelium by estrogen (Sapi et al., 2002) and in vivo increased FasL and Fas expression in malignant and borderline ovarian tumors (van Haaften-Day et al., 2003).

It is now believed that overexpression of FasL in cancer cells confers survival advantages (O'Connell et al., 2001), specifically by bestowing immune privilege or the so-called tumor immune escape. FasL contributes to immune privilege in tissues such as the eye and the testis by inducing apoptosis of activated lymphocytes (Griffith and Ferguson, 1997). In neoplastic cells, enhanced expression represents a potential mechanism for peripheral deletion of tumor-reactive T-cell clones. Additionally, FasL also suppresses T-cell activation and immunoglobulin production in B cells (Lepple-Wienhues et al., 1999; Sampalo et al., 2000). It is therefore not surprising to find enhanced FasL expression in many human carcinomas, including those of the colon, lung, esophagus, pancreas, and skin (Moller et al., 1994; O'Connell et al., 1996, 2001; Griffith and Ferguson, 1997; Ungefroren et al., 1998; Bennett et al., 1998, 2001). In accordance, levels of FasL mRNA and protein are also elevated in OVCA cells compared with that of normal HOSE cells. We postulate that elevated FasL levels help OVCA cells evade immune surveillance (O'Connell et al., 1996; Pinkoski and Green, 2000; Wajant 2002). In vivo, an increased frequency of FasL expression in OC specimens correlates with increasing malignant potential and poor prognosis (Das et al., 2000; Munakata et al., 2000). The fact that P4 was found to inhibit constitutively elevated FasL in OVCA cells raises the possibility that the hormone, in addition to promoting OC death, contributes to immune evasion by tumor cells.

Little is known about which initiator caspase is involved in apoptosis of normal or malignant HOSE cells. Our finding that caspase-8 was involved in P4-induced apoptosis of both cell types may have important ramifications in designing better treatment protocols for OCs. Caspase-9 in OC cells has been shown to be more selectively activated by chemotherapeutic agents, whereas caspase-8 could be activated by a wide myriad of drugs in an FADD-independent manner (Milner et al., 2002). In this regard, we found additive effects of P4 and anti-Fas antibody on induction of apoptosis for OVCA cells. It is therefore perceivable that future OC treatment might entail a combined therapy utilizing P4 plus ligation of Fas or other death receptors.

In conclusion, to the best of our knowledge, this is the first report describing a P4-induced, caspase-8-initiated, Fas/FasL-dependent apoptosis pathway in HOSE and OVCA cell lines. This study also makes the novel observation that FasL expression in these cells is regulated by P4 in a manner consistent with the proposed role of the hormone in OC protection/treatment. Further investigations along this line should open exciting new avenues for therapeutic intervention for OC.

Materials and methods

Cell cultures and cell lines

The origin and culture conditions of HOSE and OVCA cell lines have been described previously (Syed et al., 2001, 2002). The HOSE cell lines HOSE 642 and HOSE 12-12 were derived from normal ovaries obtained from women with noncancer gynecologic indications, a 47-year-old normal woman, and a 39-year-old woman with ovarian stromal hyperplasia, respectively. Primary cultures were established from surface scrapings of these normal ovaries and immortalized with human papillomavirus E6 and E7 genes (Tsao et al., 1995). All primary cultures had an epithelial cell phenotype, that is, cobblestoned morphology, epithelial cytokeratin staining, low responsiveness to transforming growth factor-1, no detectable CA125 production, and lack of in vivo tumorigenicity (Tsao et al., 1995). OVCA cell lines OVCA 429 and OVCA 432, were established cell lines derived from patients with late-stage serous ovarian adenocarcinomas, as described by Bast et al. (1981). Well-characterized HOSE cell lines, HOSE 642 and HOSE 12-12, and OC cell lines, OVCA 429 and OVCA 432 were cultured at 37°C in a humidified atmosphere of 5% CO₂/95% air in a 1 : 1 mixture of medium 199/MCDB 105 (Sigma, St Louis, MO, USA) supplemented with 10% heat-inactivated fetal bovine serum (hi-FBS; Sigma), penicillin (100 U/ml; Sigma), and streptomycin (100 g/ml; Sigma) (the culture medium).

Treatment protocols for HOSE and OVCA cell lines

HOSE and OVCA cells were seeded at 2 10⁵ cells per T-25 flask (Falcon, Becton Dickinson Labware, Bedford, MA, USA) and maintained in the culture medium with charcoal-stripped hi-FBS for 48 h before exposure to 10^-6 M P4 (Sigma) dissolved in absolute ethanol or to the vehicle. Cell cultures were subsequently treated daily for 5 days. The final total concentration of ethanol in cell cultures never exceeded 0.1%. At the end of the treatment period, cells were harvested for total RNA and protein extraction. In a time-course study, HOSE and OVCA cell lines were treated with 10^-6 M of P4 for 6, 18, and 36 h.

To determine whether the P4-induced apoptosis in HOSE and OVCA cell lines was mediated by activation of intracellular caspases, cell cultures were pretreated with different concentrations (0.1–10 M) of the general caspase inhibitor Z-VAD for 30 min before exposure to 10^-6 M of P4 for 3 days.

To detect any synergistic/antagonistic action between P4 and Fas-activation on apoptosis of HOSE and OVCA cells, an 'activating' anti-human Fas monoclonal antibody (MAB142; R&D, Minneapolis, MN, USA) was used to activate cell-surface Fas receptors in the four cell lines in the absence and presence of P4. Mouse immunoglobulin G antibody (Santa Cruz Biotechnology, Santa Cruz, CA, USA) was used as a negative control antibody. Cell lines were plated at a density of 2000 cells per well in culture medium in 96-well plates. After 2 days, the medium was replaced with 200 l of fresh medium containing charcoal-stripped hi-FBS. The cells were then treated with the anti-Fas antibody or the control antibody at 500 ng/ml, alone or in the presence of 10^-6 of P4. Results were compared with those of cultures exposed to P4 only.

Further evidence that the Fas/FasL system is involved in P4-induced apoptosis in HOSE and OVCA cell lines was provided by setting up cell culture experiments as described in the previous paragraph. Instead of using an activating anti-Fas antibody, a 'blocking' anti-FasL antibody (G247-4, BD Biosciences Pharmingen, San Diego, CA, USA) at 100 and 500 ng/ml, alone or in the presence of 10^-6 of P4, was used.

Detection of early and late apoptotic cells by fluorescence-assisted cell sorting (FACS) of cells stained with annexin V-FITC and/or PI

Annexin V-FITC/PI staining was performed with the apoptosis detection kit from Oncogene Research Products (San Diego, CA, USA) according to the manufacturer's instruction. One million cells (10⁶ cells/ml) were washed twice with cold PBS pH 7.2 and resuspended in 0.5 ml of suspension buffer. Cells were stained with annexin V-FITC (200 g) and subsequently with PI (30 g/ml in PBS). Flow cytometric analysis was performed with the following controls: unstained cells, cells stained with annexin V-FITC only, and cells stained with PI only.

Assays of caspase activity

After treatment with P4 for a defined period, the detached cells and the adherent cells were collected from each culture and suspended in 500 l of ice-cold lysis buffer provided with the Caspases Assay kit (MLB International, Watertown, MA, USA). After sonification, the cell lysate was centrifuged for 20 min at 14 000 g at 4°C. The resulting supernatants were analysed for protein concentrations by the Bradford dye-binding assay and stored at -20°C until use. Colorimetric enzymatic activity assays for caspases were performed according to the manufacturer's instructions.

RNA isolation and quantification of FAS, FASL, and glyceraldehyde 3-phosphate dehydrogenase (GADPH) mRNA by semiquantitative RT–PCR

Total RNA was isolated from P4-treated and untreated HOSE and OVCA cell lines with the Tri-reagent protocol (Sigma) according to the manufacturer's instructions. Total RNA (1–2 g) from each sample was reverse transcribed with the GeneAmp RNA PCR kit (Perkin-Elmer, Shelton, CT, USA).

The relative expression levels of FAS, FASL, and GAPDH mRNA were investigated by performing semiquantitive RT–PCRs in each cell culture. The oligonucleotide primers specific for human FAS, FASL, and GAPDH were published sequences (Das et al., 2000). The FAS primers were CAGAACTTGGAAGGCCTGCATC (forward) TCTGTTCTGCTGTGTCTTGGAC (reverse), which amplified a fragment of 682 bp. Specific FasL primers were ACACCTATGGAATTGTCCTGC (forward) and GACCAGAGAGAGCTCAGATACG (reverse), which amplified a fragment of 311 bp. GAPDH primers were CATCACCATCTTCCAGGAGC (forward) and GGATGATGTTCTGGAGAGCC (reverse), which amplified a fragment of 404 bp. Aliquots of 2-l first-strand cDNA were subjected to PCR amplification. Hot start PCR using AmpliTaq Gold DNA polymerase (Perkin-Elmer) was used in all amplification reactions. The annealing temperatures for the specific primers were 55°C for FasL and Fas and 60°C for GAPDH cDNA amplification. The number of amplification cycles was 34 for Fas and FasL and 25 for GAPDH. All other PCR conditions were the same, with initial denaturation for 6 min at 94°C, cycle denaturation for 1 min at 94°C, annealing for 1 min, and extension at 72°C for 1.5 min. The PCR products were fractionated on 2% agarose gel and quantified after ethidium bromide staining. Signal intensities of Fas and FasL were normalized to those of GAPDH to produce arbitrary units of relative message abundance. The reproducibility of the experiment was established by three independent experiments for cDNA synthesis and each cDNA was subjected to three separate PCRs. The means of the replicated measurements were calculated.

Immunoblot analysis

Cellular extracts were prepared as described earlier (Syed et al., 2002). Proteins from cell extracts (25 g) of normal and malignant cells were separated by 10% SDS–polyacrylamide gel electrophoresis and transferred to nitrocellulose membranes as described previously (Syed et al., 2002). After blocking, membranes were incubated with one of the following primary antibodies: an anti-Fas clone DX₂ monoclonal antibody (2 g/ml; R&D Systems, Minneapolis, MN, USA), M-20 polyclonal antibody (1 : 500 (v/v) dilution; Santa Cruz Biotechnology) for FasL, polyclonal anti-caspase-3 (1 : 500, Cell Signaling Technology), monoclonal anti-caspase-8, 1C12 (1 : 1000, Cell Signaling Technology), and polyclonal anti-caspase-9 (1 : 1000, Cell Signaling Technology) for caspase-3, -8, and -9, respectively, and GAPDH antibody (1 : 500, Biogenesis, Poole, UK). After membranes were washed, they were incubated with the secondary antibody at a 1 : 6000 dilution in 3% BSA or 3% nonfat dry milk in TBST for 1 h at room temperature. Blots were developed using enhanced chemiluminescence (ECL) protocol (Amersham Pharmacia Biotech, Little Chalfont, Buckinghamshire, UK).

MTT assay

The numbers of cells in the cultures were assayed by the MTT assay (Roche Diagnostics, Indianapolis, IN, USA) following the manufacturer's instructions and as described previously (Syed et al., 2001). Assays were performed in triplicate to generate mean values for controls and each treatment group. Cell numbers in untreated control cultures were arbitrarily assigned a value of 1. The relative cell growth was expressed as fold increases/decreases over control untreated cultures. Data points are group mean valuess.e.m. from three separate experiments.

Statistical analysis

Data are expressed as the mean of two to four experiments, each in triplicate samples for individual treatments or dosage regimens. Statistical analysis was carried out with two-tailed Student's t-tests. Values are presented as the means.e.m. All statistical tests were considered to be statistically significant at P<0.05.

References

1. Adami HO, Hsieh CC, Lambe M, Trichopoulos D, Leon D, Persson I, Ekbom A and Janson PO. (1994). Lancet, 344, 1250–1254. | Article | PubMed | ChemPort |
2. Adami HO, Lambe M, Persson I, Ekbom A, Hsieh CC and Trichopoulos D. (1995). Lancet, 345, 789. | PubMed |
3. Akhmedkhanov A, Zeleniuch-Jacquotte A and Toniolo P. (2001). Ann. N.Y. Acad. Sci., 943, 296–315. | PubMed | ChemPort |
4. Alnemri ES, Livingston DJ, Nicholson DW, Salvesen G, Thornberry NA, Wong WW and Yuan J. (1996). Cell, 87, 171. | Article | PubMed | ISI | ChemPort |
5. Ashkenazi A and Dixit VM. (1998). Science, 281, 1305–1308. | Article | PubMed | ISI | ChemPort |
6. Bast Jr RC, Feeney M, Lazarus H, Nadler LM, Colvin RB and Knapp RC. (1981). J. Clin. Invest., 68, 1331–1337. | PubMed | ISI |
7. Bennett MW, O'Connell J, Houston A, Kelly J, O'Sullivan GC, Collins JK and Shanahan F. (2001). J. Clin. Pathol., 54, 598–604. | Article | PubMed | ISI | ChemPort |
8. Bennett MW, O'Connell J, O'Sullivan GC, Brady C, Roche D, Collins JK and Shanahan F. (1998). J. Immunol., 160, 5669–5675. | PubMed | ISI | ChemPort |
9. Bu SZ, Yin DL, Ren XH, Jiang LZ, Wu ZJ, Gao QR and Pei G. (1997). Cancer, 79, 1944–1950. | Article | PubMed | ChemPort |
10. Cramer DW, Hutchison B, Welch WR, Scully RE and Ryan KJ. (1983). J. Natl. Can. Inst., 71, 711–716.
11. Cramer DW and Welch WR. (1993). J. Natl. Cancer Inst., 71, 717–721.
12. Das H, Koizumi T, Sugimoto T, Chakraborty S, Ichimura T, Hasegawa K and Nishimura R. (2000). Br. J. Cancer, 82, 1682–1688. | Article | PubMed | ISI | ChemPort |
13. Earnshaw WC, Martins LM and Kaufmann SH. (1999). Annu. Rev. Biochem., 68, 383–424. | Article | PubMed | ISI | ChemPort |
14. Ghahremani M, Foghi A and Dorrington JH. (1998). Gynecol. Oncol., 70, 275–281. | PubMed |
15. Gratas C, Tohma Y, Barnas C, Taniere P, Hainaut P and Ohgaki H. (1998). Cancer Res., 58, 2057–2062. | PubMed | ISI | ChemPort |
16. Griffith TS and Ferguson TA. (1997). Immunol. Today, 18, 240–244. | Article | PubMed | ISI | ChemPort |
17. Hayakawa A, Wu J, Kawamoto Y, Zhou YW, Tanuma S, Nakashima I and Suzuki H. (2002). Apoptosis, 7, 107–113. | PubMed |
18. Hu Z and Deng X. (2000). Zhonghua Fu Chan Ke Za Zhi, 35, 423–426. | PubMed | ChemPort |
19. Keane MM, Ettenberg SA, Lowrey GA, Russell EK and Lipkowitz S. (1996). Cancer Res., 56, 4791–4798. | PubMed | ISI | ChemPort |
20. Lambe M, Wuu J, Rossing MA and Hsieh CC. (1999). Lancet, 353, 1941. | Article | PubMed | ISI | ChemPort |
21. Lepple-Wienhues A, Belka C, Laun T, Jekle A, Walter B, Wieland U, Welz M, Heil L, Kun J, Busch G, Weller M, Bamberg M, Gulbins E and Lang F. (1999). Proc. Natl. Acad. Sci. USA, 96, 13795–13800. | Article | PubMed | ChemPort |
22. Milner AE, Palmer DH, Hodgkin EA, Eliopoulos AG, Knox PG, Poole CJ, Kerr DJ and Young LS. (2002). Cell Death Differ., 9, 287–300. | Article |
23. Moller P, Koretz K, Leithauser F, Bruderlein S, Henne C, Quentmeier A and Krammer PH. (1994). Int. J. Cancer, 57, 371–377. | Article | PubMed | ISI | ChemPort |
24. Munakata S, Enomoto T, Tsujimoto M, Otsuki Y, Miwa H, Kanno H and Aozasa K. (2000). Br. J. Cancer, 84, 1446–1452.
25. Murdoch WJ and McDonnel AC. (2002). Reproduction, 123, 743–750. | Article | PubMed | ChemPort |
26. Nagata S. (1994). Philos. Trans. R. Soc. London B. Biol. Sci., 345, 281–287. | PubMed |
27. O'Connell J, Houston A, Bennett MW, O'Sullivan GC and Shanahan F. (2001). Nat. Med., 7, 271–274. | Article | PubMed | ISI | ChemPort |
28. O'Connell J, O'Sullivan GC, Collins JK and Shanahan F. (1996). J. Exp. Med., 184, 1075–1082. | Article | PubMed | ChemPort |
29. Pinkoski MJ and Green DR. (2000). Curr. Opin. Genet. Dev., 10, 114–119. | Article | PubMed | ISI | ChemPort |
30. Reimer T, Herrnring C, Koczan D, Richter D, Gerber B, Kabelitz D, Friese K and Thiesen HJ. (2000). Cancer Res., 60, 822–828. | PubMed | ISI | ChemPort |
31. Riman T, Persson I and Nilsson S. (1998). Clin. Endocrinol. (Oxf.), 49, 695–707.
32. Rodriguez GC, Nagarsheth NP, Lee KL, Bentley RC, Walmer DK, Cline M, Whitaker RS, Isner P, Berchuck A, Dodge RK and Hughes CL. (2002). J. Natl. Cancer Inst., 94, 50–60. | PubMed | ChemPort |
33. Sampalo A, Navas G, Medina F, Segundo C, Camara C and Brieva JA. (2000). Blood, 96, 3168–3174. | PubMed | ISI | ChemPort |
34. Sapi E, Brown WD, Aschkenazi S, Lim C, Munoz A, Kacinski BM, Rutherford T and Mor G. (2002). J. Soc. Gynecol. Invest., 9, 243–250.
35. Selam B, Kayisli UA, Mulayim N and Arici A. (2001). Biol. Reprod., 65, 979–985. | PubMed |
36. Song K, Chen Y, Goke R, Wilmen A, Seidel C, Goke A and Hilliard B. (2000). J. Exp. Med., 191, 1095–1104. | Article | PubMed | ISI | ChemPort |
37. Song J, Rutherford T, Naftolin F, Brown S and Mor G. (2002). Mol. Hum. Reprod., 8, 447–455. | PubMed |
38. Strand S, Hofmann WJ, Hug H, Muller M, Otto G, Strand D, Mariani SM, Stremmel W, Krammer PH and Galle PR. (1996). Nat. Med., 2, 1361–1366. | Article | PubMed | ISI | ChemPort |
39. Suda T, Takahashi T, Golstein P and Nagata S. (1993). Cell, 75, 1169–1178. | Article | PubMed | ISI | ChemPort |
40. Syed V, Ulinski G, Mok SC, Yiu GK and Ho SM. (2001). Cancer Res., 61, 6768–6776. | PubMed | ISI | ChemPort |
41. Syed V, Ulinski G, Mok SC and Ho SM. (2002). J. Natl. Cancer Inst., 94, 617–629. | PubMed | ChemPort |
42. Tsao SW, Mok SC, Fey EG, Fletcher JA, Muto MG, Knapp RC and Berkowitz RS. (1995). Exp. Cell Res., 218, 499–507. | Article | PubMed | ISI | ChemPort |
43. Ungefroren H, Voss M, Jansen M, Roeder C, Henne-Bruns D, Kremer B and Kalthoff H. (1998). Cancer Res., 58, 1741–1749. | PubMed | ISI | ChemPort |
44. van Haaften-Day C, Russell P, Davies S, King NJ and Tattersall MH. (2003). Hum. Pathol., 34, 74–79. | Article | PubMed | ISI | ChemPort |
45. Wang S, Pudney J, Song J, Mor G, Schwartz PE and Zheng W. (2003). Gynecol. Oncol., 88, 108–117. | PubMed |
46. Wajant H. (2002). Science, 296, 1635–1636. | Article | PubMed | ISI | ChemPort |
47. Yonehara S, Ishii A and Yonehara M. (1989). J. Exp. Med., 169, 1747–1756. | Article | PubMed | ISI | ChemPort |
48. Yu S, Lee M, Shin S and Park J. (2001). J. Cell Biochem., 82, 445–451. | Article | PubMed | ChemPort |
49. Zimmermann KC, Bonzon C and Green DR. (2001). Pharmacol. Ther., 92, 57–70. | Article | PubMed | ISI | ChemPort |
50. Zou H, Li Y, Liu X and Wang X. (1999). J. Biol. Chem., 274, 11549–11556. | Article | PubMed | ISI | ChemPort |

Acknowledgements

This study was supported by an Army Ovarian Cancer Research Program Grant DAMD17-99-1-9563 (to S-M Ho) and NIH Grants CA091250 (to V Syed) and CA94221 (to S-M Ho). This publication was made possible by Grant Number 5 P30 DK32520 from the National Institute of Diabetes and Digestive and Kidney Diseases'. We thank Dr Samuel C Mok of Brigham and Women's Hospital, Boston, Massachusetts, for his generous gifts of HOSE and OVCA cell lines and editorial staff of Brigham and Women's Hospital, Boston, for editorial help.