Objective
There is growing evidence that progestins and nonsteroidal antiinflammatory drugs (NSAIDs) may prevent ovarian cancer. Because both induce apoptosis, we investigated the potential for synergistic impact of combined drug treatment on cell death.
Study Design
Using normal and malignant human ovarian epithelial cells and an NSAID-sensitive human colon cancer cell line, we evaluated the effects of progestins and NSAIDs alone and in combination on apoptosis.
Results
Both progestins and NSAIDs dose dependently inhibited cell growth ( P < .0001). Doses of NSAIDs or progestins that independently reduced cell viability by less than 30% synergistically reduced cell viability by 70-95% when combined. Similarly, the NSAID/progestin combination conferred 4- to 18-fold ( P < .05) increased apoptosis over either treatment alone.
Conclusion
Our results suggest it may be possible to combine progestins and NSAIDs to achieve ovarian cancer prevention at lower doses of each than are required for single administration, thereby lessening the risk of side effects posed by these agents.
There is significant potential to decrease ovarian cancer incidence and mortality through chemoprevention. Epidemiological evidence has shown that routine use of the combination estrogen-progestin oral contraceptive pill (OCP) confers a 30-50% reduction in the risk of developing epithelial ovarian cancer, suggesting that an effective pharmacological approach for the chemoprevention of ovarian cancer is possible. Investigations by our group have elucidated a mechanism that may be responsible for the ovarian cancer preventive effects of the OCP. Specifically, we have discovered that the progestin component of the OCP may be functioning as a chemopreventive agent in the ovarian surface epithelium by triggering apoptosis, a biological effect that is well known to be associated with cancer prevention. This raises the possibility that progestins enhance the clearance of genetically damaged ovarian epithelial cells, thereby significantly lowering ovarian cancer risk.
Several lines of human evidence support an ovarian cancer preventive role for progestins, including the following: (1) evidence of a 60% reduction in the risk of nonmucinous ovarian cancer in women who have used depomedroxyprogesterone acetate, a progestin-only contraceptive ; (2) evidence that twin gestation, which is associated with higher circulating levels of progesterone, is more protective against subsequent ovarian cancer risk than singleton pregnancy ; and (3) evidence that use of progestin-potent OCPs confers twice the ovarian cancer protective effect as use of OCPs containing weak progestins. Taken together, these data provide a strong rationale to evaluate progestins as ovarian cancer preventives and suggest that a pharmacological regimen that has enhanced chemopreventive biological potency in the ovarian epithelium will be more effective than lower-potency regimens.
The finding that progestins activate apoptosis in the ovarian epithelium suggests that it may be possible to develop other agents with similar biological effects as ovarian cancer preventives. In this regard, there is a growing body of epidemiological and laboratory evidence in support of nonsteroidal antiinflammatory drugs (NSAIDs) and other antiinflammatory agents as cancer-preventive agents for a variety of cancers, including ovarian cancer. Case-control comparisons suggest a reduction in ovarian cancer risk associated with use of NSAIDs as well as acetaminophen.
The molecular basis for the protective effect of these agents has not been well defined but may involve a direct chemopreventive biological effect on the ovarian epithelium similar to that induced by progestins. It has been shown, for example, that NSAIDs induce apoptosis in cells derived from human ovarian epithelium, similar to progestins.
Based on these data, we have hypothesized that progestins and NSAIDs target the early steps of carcinogenesis in the ovarian epithelium by activating apoptosis, thereby clearing dysplastic cells and resulting in effective cancer prevention. In this study, we sought to determine whether the effects of progestins and NSAIDs would be synergistic when these agents are combined. We show that in ovarian epithelial cells these 2 classes of drugs have a synergistic effect on cell death.
Materials and Methods
Materials
All culture media, sera, and reagents were purchased from Invitrogen (Carlsbad, CA), except MCDB105 and insulin, which were purchased from Sigma (St Louis, MO). All NSAIDs, acetaminophen, and progestins were purchased from Sigma, except celecoxib, which was purchased from LKT Laboratories (St Paul, MN).
Cell culture
The ovarian cancer cell lines OVCAR-3 (purchased from the American Type Culture Collection, Manassas, VA) and OVCAR-5 (kindly provided by Dr Tom Hamilton, Fox Chase Cancer Center, Philadelphia, PA) were grown in RPMI 1640 medium with glutamine supplemented with 10% heat-inactivated fetal bovine serum (FBS), 10 μg/mL insulin, and penicillin/streptomycin. HIO-118V, a transformed human ovarian surface epithelial cell line transfected with Simian virus 40 antigen to extend its life span (kindly provided by Dr Vimla Band, University of Nebraska Medical Center [Omaha, NE], with permission of Dr Andrew Godwin, Fox Chase Cancer Center) and normal ovarian epithelial cell cultures (NOE cells; harvested from a normal human ovary during diagnostic surgery under the approval of the institutional review board) were grown in a 1:1 mixture of Medium 199 and MCDB105 with 15% heat-inactivated FBS, glutamine, 10 ng/mL epidermal growth factor, and penicillin/streptomycin. The HT-29 colon cancer cell line (purchased from American Type Culture Collection) was grown in McCoy’s 5A medium with 10% heat-inactivated FBS and penicillin/streptomycin. All cell lines were maintained in 5% CO 2 at 37°C.
Cell viability/proliferation testing
Cells were seeded overnight in 96-well plates at 2500-5000 cells/well. The culture medium was replaced with fresh medium containing experimental agents (progestins and/or NSAIDs in varying concentrations to yield 20-30% decreased proliferation in each respective cell line). Cells were treated for 72 hours, at which time untreated controls were nearly confluent. All treatments were performed in quadruplicate. Viability was tested using the formazan dye-based 3-(4,5-dimethylthiazol-2-yl)-5-(3-carboxymethoxyphenyl)-2-(4-sulfophenyl)-2H-tetrazolium, inner salt (MTS) assay (Promega, Madison, WI) according to the manufacturer’s specifications. The plates were incubated at 37°C until the untreated wells exhibited an A 490 of 0.7-0.9. Wells containing medium alone were used as blanks. Viability was expressed as a percentage of untreated controls. All experiments were repeated at least 3 times.
Detection of apoptosis by activated caspase-3 and TUNEL assays
OVCAR-3, OVCAR-5, and HIO-118V cells were seeded overnight into 60 mm dishes at 2.8 × 10 5 , 1.6 × 10 5 and 1.6 × 10 5 cells/dish, respectively. The culture medium was then replaced with fresh medium containing progesterone, sulindac sulfide, or celecoxib or progesterone combined with sulindac sulfide or celecoxib and incubated for 48 hours. Vehicle concentrations did not exceed 0.125% for dimethylsulfoxide and 0.32% for ethanol.
All cells, floating and adherent, were collected, washed, fixed, and stained using either fluorescein isothiocyanate (FITC)-labeled antibody to active caspase-3 (APO Active 3) purchased from Cell Technology (Beverly, MA) or deoxynucleotidyltransferase deoxyuridine 5-triphosphate (dUTP) nick end labeling (TUNEL) using the APO-Direct kit purchased from BD Pharmingen (San Diego, CA). Both assays were performed according to the manufacturers’ protocols, with the following modification to the caspase-3 assay: 0.1% saponin was used in the rinse buffer. Ten thousand cells were measured by FACS Calibur (BD Biosciences, Franklin Lakes, NJ), and the data were analyzed using Cellquest 3.3 software (BD Biosciences).
FITC fluorescence was collected through a 488 bandpass filter and forward and side scatter were analyzed on a linear scale. For the caspase-3 assay, histograms, on a logarithmic scale, were used to assess the percentage of caspase-3–labeled cells for each treatment group. Gates were based on the untreated controls. For the TUNEL assay, each sample was gated to include only a singlet population using dual parameters of deoxyribonucleic acid (DNA) width (x-axis) and DNA area (y-axis). In assessing apoptosis, the resultant gates were used to generate dual parametric graphs of DNA area (x-axis) and FITC-dUTP (y-axis). Gates were based on increased dUTP-FITC labeling compared with the untreated controls. In both experiments, results are expressed as fold increase over controls. Both sets of experiments were repeated at least 3 times and/or run in triplicate.
Determination of synergy
Data were analyzed isobolographically using CalcuSyn software (Biosoft, Cambridge, UK) to characterize the inhibitory effect drug combinations had on the cell lines. Raw data for each drug or drug combination were entered singly to generate a median effect plot. From this plot, the combination index (CI) equation was applied to determine whether the drug effects were additive, synergistic, or antagonistic. CI values of less than 1, 1, or greater than 1 indicated synergy, additivity, or antagonism, respectively.
Fluorescent caspase assay
Fluorescent caspase-3 and -9 substrates and their inhibitors were obtained from Calbiochem (San Diego, CA). Cells were treated for 20-24 hours at drug concentrations that killed most cells as measured by MTS assay at 72 hours. Cell lysates were harvested and combined with a reaction buffer and the uncleaved, fluorescent substrate, according to the manufacturer’s instructions. Kinetic readings were made every 5 minutes for 1 hour on a SpectraMax Gemini XS fluorescent plate reader (Molecular Devices, Sunnyvale, CA). Results were normalized with blank samples and untreated cell lysates and are expressed as fold differences relative to untreated samples.
Results
Treatment with progestins and NSAIDs inhibits cell growth
The 2 ovarian cancer cell lines, OVCAR-3 and OVCAR-5, a primary normal ovarian epithelial cell culture, NOE, a transformed ovarian epithelial cell line, HIO-118V, and the colon cancer cell line, HT-29, were tested in dose-response experiments for the effects of progestins and NSAIDs.
All cell types were growth inhibited in a dose-dependent manner when exposed to progestins ( P < .0001; Figure 1 , A) . Relative to the normal and transformed ovarian epithelial cell cultures, the cancer cell lines OVCAR-3, OVCAR-5, and HT-29 were less sensitive to the progestins levonorgestrel and norethindrone and more sensitive to progesterone. The 50% lethal dose (LD 50 ) values from levonorgestrel treated cells ranged from 70 μM in the HIO-118V cells to 359 μM in the HT-29 cells, whereas norethindrone LD 50 values ranged from 141 μM in the HIO-118V cells to greater than 400 μM in the OVCAR-5 and HT-29 cells ( Table 1 ).
