Ovarian carcinoma is the fourth leading cause of cancer death in women and the most lethal gynecological malignancy. Approximately 70% of patients diagnosed with ovarian carcinoma will present at late stage of the disease (1). Modern surgery and cytotoxic chemotherapy will produce a transient clinical response, but the majority of patients will experience relapse and the response to the treatments is often short-lived (2). In many tumors, the actively dividing cells account for only a small proportion of the total cancer cells, with the remainder of the cells being in a nonproliferating state, which consists of both G0 (quiescent cells) and G1 cells (3, 4). Indeed, tumor cell populations in G0 range from 70–95%, among different types of cancer including epithelial ovarian cancer (OC) (5, 6). In a study looking at 100 OC tumors, the average expression of Ki-67, a proliferation marker present in all stages of the cell cycle except cells arrested at G0 and early G1, was 30% (5, 7). The nonproliferating cells within the tumor are viable but in a reversible state of growth arrest. Nevertheless, the nonproliferating cell population of tumors frequently poses a barrier to successful cancer therapies. Most chemotherapeutic drugs target proliferating cells, but the growth fraction of many tumors is low (3). Tumor cells in the nonproliferative stage are resistant to therapy and maintain the capacity to replenish a tumor after therapy, resulting in relapse (8). However, counter-intuitively, in cancers that respond poorly to therapy, it is the proliferating cells that are targeted. Therefore, it is tempting to explain all or most therapeutic failures by the persistence of nonproliferating tumor cells. Despite initial enthusiasm to reduce the nonproliferating cell population, in order to switch tumor cells into a division cycle (9), there have been no major therapeutic breakthroughs.
Numerous studies demonstrate that nonsteroidal anti-inflammatory drugs (NSAIDs) hold significant promise as anticancer therapeutics (10–12). Recent data on regular NSAID use indicated a reduction in the incidence of OC and inhibition of carcinogenesis (12–15). The basic mechanism of action of NSAIDs is to a large part attributed to their inhibition of cycloxygenase 2 (COX-2) (16). As COX-2 inhibition could differentially contribute to the antineoplastic, proapoptotic pathway targeted by NSAIDs in cancer cells, it is essential to unravel the molecular processes involved in apoptosis induction by NSAIDs. In this regard, we have recently described the mechanisms by which structurally different NSAIDs induce programmed cell death in various types of cancer cells including OC (17). Sulindac Sulfide and Diclofenac are the most potent NSAIDs that induce apoptosis in proliferating OC cells. Furthermore, we have demonstrated that NSAIDs lead to apoptotic induction through a pathway mediated by melanoma differentiation associated gene-7/Interleukin-24 (MDA-7/IL24) (18).
To obtain a comprehensive view of NSAID-mediated apoptosis in cancer cells we surveyed and compared a broad spectrum of NSAIDs for their capacities to induce apoptosis in nonproliferating OC cells. We also reasoned that by combining NSAIDs that target nonproliferating cells with NSAIDs that have effects on the proliferating cells, a dual-pathway targeting (DPT) approach, the effects of NSAIDs against OC might be enhanced. Here, we reveal that Flufenamic Acid, Flurbiprofen, Finasteride, Celocoxib, and Ibuprofen are the most potent NSAIDs that induce apoptosis in nonproliferating OC cells. Furthermore, we demonstrate that combinatorial therapy, the DPT approach, with Flurbiprofen and Sulindac Sulfide drastically reduces tumor growth in vivo when compared with a monotherapy.
MATERIALS AND METHODS
The OC cell lines SKOV-3, CAOV-3, 36M2, and SW626 and the HEK 293 cells were obtained from the American Type Culture Collection (Rockville, MD). The MS-1 endothelial cell line and the F-12 foreskin fibroblast cell line were kindly provided by Peter Oettgen and Dr. Steven Goldring, respectively, Beth Israel Deaconess Medical. The OC cells were grown in RPMI medium from Gibco (Carlsbad, California). The MS-1, F-12, and HEK 293 cells were grown in DMEM medium from Gibco (Carlsbad, California). The medium was supplemented with 10% fetal bovine serum, 50 U of penicillin per mL, and 50 μg of streptomycin per mL (Gibco, California). The cells were maintained in a 5% CO2-humidified incubator at 37°C. For the starvation method, the cells were cultivated in medium without fetal bovine serum for 24–48 h.
Sulindac Sulfide, Sulindac Sulfone, Ibuprofen, Aspirin, Acetaminophen and Naproxen were obtained from Sigma-Aldrich (St. Louis, MO, USA). Meloxican, Celocoxib, Diclofenac, Finasteride, and Flufenamic acid were obtained from LKT laboratories (St. Paul, MN, USA). NS-398, Ebselen, and Flurbiprofen were purchased from Calbiochem (San Diego, CA, USA). The drugs were dissolved in dimethyl sulfoxide (DMSO) or ethanol. Cancer cells were treated in their specified media, without FBS, for 24 h, following 24 h serum starvation. The final concentration for each compound was: 50 μM Sulindac Sulfide, 5 mM Aspirin, 200 μM Ibuprofen, 50 μM Sulindac Sulfone, 1 mM Acetaminophen, 200 μM Naproxen, 200 μM NS-398, 50 μM Celecoxib, 40 μM Diclofenac, 50 μM Finasteride, 200 μM Flufenamic acid, 10 μM Meloxican, 50 μM Ebselen, and 20 nM Flurbiprofen. For controls, cells were treated with an equal amount of DMSO or ethanol, which was less than 0.1% final concentration.
Apoptosis assays have been described before (18) and cells were assayed by using the Apoptotic Cell Death Detection ELISA (Roche, Basel, Switzerland) and/or the Cell Death Detection (Nuclear Matrix Protein) ELISA (Oncogene, La Jolla, CA, USA) according to the manufacturer's instructions. Significant statistical difference of control samples against samples treated with NSAIDs was determined by the Student's t-test.
Cell cycle Analysis
OC cells were starved and cultured in serum free media. After 24–48 h the cells were trypsinized, washed twice with cold PBS containing 2% fetal bovine serum and fixed in 70% ethanol for 60 min at 4°C. The cells were then washed twice with PBS and stained with 200 μL of propidium iodide stock solution (50 μg/mL propidium iodide, 3.8 mM Sodium trisphosphate in PBS) supplemented with 50 μL RNase A (10 μg/mL) for 3 h at 4°C and then analysed with a FACScan cell sorter (Becton Dickinson, Franklin Lakes, NJ, USA). Ten thousand cells were collected and the cell cycle profiles were calculated by using Cellquest Software (Becton Dickinson, NJ, USA).
OC cells were starved and cultured in serum free media. After 24–48 h the proliferation assay was performed using Cell Proliferation Kit I (MTT; Roche) according to the manufacturer's protocol.
Eight-week-old severe combined immunodeficient (SCID)-beige mice were purchased from Taconic (Germantown, NY) and housed in a pathogen-free environment. The mice were randomly divided into four groups (n = 8 per group). Eight-week-old female SCID-beige mice were fed with one of the experimental diets supplemented with 200 ppm sulindac sulfide, 200 ppm Flurbiprofen, a combination of either agents or control solvents for 2 weeks prior to implantation. Immediately before implantation, SKOV-3 cells were trypsinized and resuspended in RPMI medium. Cell viability was determined by trypan blue dye exclusion and a single cell suspension with >90% viability was used for implantation. SKOV-3 cells (2 × 106 cells in 50 μL) were injected subcutaneously via a 30-gauge needle as described previously (19), and the animals were maintained on the experimental diets. The experiment was terminated when the average tumor weight in the control animals reached 2–5% of the body weight. The diets were prepared by Research Diets, Inc. (New Brunswick, NJ, USA). Body weight and food intake were measured weekly. Tumor size (volume) was measured every 5 days, starting at day 35 after implantation. Animals were sacrificed and tumors were dissected and measured at the end of the 8-week experiment. All procedures with animals were reviewed and approved by the Institutional Animal Care and Use Committee at the Beth Israel Deaconess Medical Centre according to NIH guidelines.
Specific NSAIDS Are Potent Inducers of Apoptosis in Nonproliferating OC Cells
A broad panel of NSAIDs (18) was tested for their ability to induce apoptosis in proliferating OC cells. We used the starvation method to synchronize cells in a G0/G1 state and confirmed the nonproliferative state of the OC cells by FACS analysis and proliferation assays (Figs. 1a and 1b). The concentrations for all NSAIDs used in this study were selected carefully and are comparable to the achievable plasma concentrations (18). However, some drug concentrations exceeded the physiologically achievable doses (18). Apoptosis was measured 24 h after treatment of SW626, CAOV-3, SKOV-3, and 36M2 OC cells with this set of NSAIDs, revealing that a variety of, but not all NSAIDs-induced apoptosis in nonproliferating cells. Strong inducers of apoptosis across all the cell lines tested included Flufenamic Acid, Flurbiprofen, Finasteride, Celocoxib, and Ibuprofen when compared with the solvent control (P < 0.0001). Treatment with the remaining drugs resulted in only marginal apoptotic induction (Fig. 1c). Interestingly, a similar treatment of proliferating OC cells with this set of NSAIDs demonstrated significantly different sensitivity to apoptosis induction. Sulindac Sulfide treatment which we had previously reported to strongly induce apoptosis in proliferating OC cells had no effect on nonproliferating OC cells (17).
We extended our studies to determine the lowest dose of the stronger inducers that would still induce programmed cell death of nonproliferating OC cells. The concentrations of selected NSAIDs were chosen 2, 5, and 10 times lower than the physiologically achievable doses used in Fig. 1c. Apoptosis was measured in nonproliferating OC cells 24 h after treatment. Our results show that Flufenamic acid and Celocoxib concentrations applied at 10 times lower doses still effectively induced apoptosis, whereas Flurbiprofen and Celocoxib can be reduced fivefold and Finasteride can be reduced twofold (Fig. 1d).
Effects of NSAID Combinations on Nonproliferating OC Cells
Using the lowest dose of each NSAID, determined in Fig. 1d, we systematically analyzed the levels of apoptosis induction by combinations of NSAIDs in nonproliferative OC cells. After synchronizing the cells at the G0 stage (Figs. 1a and 1b), a panel of NSAIDs including Flufenamic acid, Flurbiprofen, Celocoxib, Finasteride, and Ibuprofen were tested for their abilities to induce apoptosis alone and in combination. SW626, CAOV3, and SKOV3 OC cells were treated with 5 μM Flufenamic acid, 4 nM Flurbiprofen, 40 μM Celocoxib, 25 μM Finasteride, or 200 μM Ibuprofen, and combinations of each of them. Apoptosis was measured 24 h after treatment revealing that all combinations of NSAIDs tested induced apoptosis significantly more than either of them alone in OC cells (Fig. 2). However, some NSAID combinations were more effective in apoptosis induction than others. Certain combinations such as Flufenamic acid and Flurbiprofen, Flufenamic acid and Celocoxib, Flufenamic acid and Finasteride, Flubiprofen and Celocoxib, and Celocoxib and Finasteride had the strongest effects on apoptosis induction. These drug combinations, on average, increased apoptotic cell death up to threefold when compared with the control (Fig. 2). Taken together, our results establish an unambiguous role for NSAID-mediated induction of apoptosis in nonproliferative OC cells with clear implications for cancer treatment.
NSAID Combination Therapy Is More Effective in Reducing OC Tumor Growth In Vivo Than Monotherapy
We have previously shown that Sulindac Sulfide is a potent inducer of apoptosis and inhibits tumor growth in proliferating OC cells (16). We selected Flurbiprofen from the panel of NSAIDs tested here to carry out the DPT approach, as it is one of the most potent inducers of apoptosis as a monotherapy and when applied as a combinatorial therapy with other NSAIDs in nonproliferating cells, and the dosage could be scaled down significantly (Fig. 1d). In order to test the DPT approach whether the combination of a NSAID targeting nonproliferating cells and a NSAID with therapeutic effect in proliferating cells is more effective in inhibiting tumor formation in mice than a monotherapy, we first tested this combination on normal cells in vitro. As seen in Fig. 2b, this combination has no effect on normal cells in vitro. We then proceeded to test the combination in vivo by injecting SKOV-3 OC cells subcutaneously into SCID mice. The mice were randomly divided into four groups and fed one of three diets through the entire experiment: control diet, diet supplemented with 200 ppm Sulindac Sulfide, 100 ppm Flurbiprofen, or a combination of both. Two months later the animals were examined for tumor formation and tumor volume. In order to evaluate the toxicity of the drugs used, animal body weight was monitored in the mice during the course of the experiment. As seen in Fig. 3b, no significant weight loss was detected in the animals indicating that this dosage is safe for use. All mice developed tumors indicating that these particular doses of NSAIDs did not prevent tumor formation. As is evident in Fig. 3, on day 60, Sulindac Sulfide and Flurbiprofen treatment reduced the average tumor volume by 45 and 40%, respectively, when compared with the control diet (P < 0.0001), confirming its antitumor efficacy. However, tumor volume was further decreased by 57% (P < 0.0001) in mice treated with a combination of a NSAID, which targets nonproliferating cells and a NSAID with therapeutic effect in proliferating cells, demonstrating that combinatorial therapy (DPT approach) promotes a stronger inhibitory effect than monotherapy.
These results strongly support the hypothesis that dual NSAID drug treatment regiments with distinct target specificities can act synergistically and more effectively inhibit OC growth in vivo than a single NSAID treatment. Our experiments confirm that NSAIDs, which target specifically nonproliferative cancer cells in combination with NSAID therapy that focuses on the proliferative stage of the disease, the DPT approach, may provide opportunities for rationally designing new combination treatment modalities for OC.
The lack of effective therapies for OC reflects in part the lack of drug treatments that target nonproliferating tumor cells. There is a lack of knowledge about the mechanisms of action of drugs that may have the potential to target nonproliferating OC cells. In a recent study, the importance of quiescent (nonproliferating) cell populations in radiation therapy was highlighted in a lymphoma mouse model, showing that this population of cells is far less radio-sensitive than the dividing cell populations and presents a barrier to effective therapy (20). Here, we systematically and comprehensively evaluated the effects of a broad panel of NSAIDs with potential anticancer activities toward nonproliferating OC cells and identified possible combinations of NSAIDs for targeting both nonproliferating and actively dividing OC cells. Flufenamic Acid, Flurbiprofen, Finasteride, Celocoxib, and Ibuprofen emerged as the most effective NSAIDs that can induce programmed cell death in nonproliferating OC cell. Furthermore, a combination of these agents results in an enhanced apoptotic effect.
We have recently established NSAID therapy for proliferating OC cells (17). Interestingly, in this study, we demonstrate that NSAIDs that have potent effects in proliferating cells display only a marginal effect on apoptosis induction of nonproliferating cells. These data support a rational DPT approach for treating OC that embodies a therapy that targets specifically nonproliferating cancer cells in combination with a drug regimen that is operational on cells in the proliferative stage of the disease. Although many studies have looked at the role of NSAIDs in cancer therapy, the majority of these have focused on COX inhibition as a mechanism of action for these drugs (21–23). Decreased sensitivity to chemotherapy and poor prognostic outcome has been associated with ovarian carcinomas expressing COX-2 (24). However, it has been shown that COX-2 is only expressed in a subset of ovarian carcinomas (25).
Our data provide strong evidence that multiple NSAIDs induce cancer cell death in nonproliferating OC cells, irrespective of their ability to block COX-2. For example, Finasteride is well known as a chemotherapeutic agent in prostate cancer, through its inhibition of 5α-reductase (26–28). Flurbiprofen, together with Ibuprofen have been implicated in the induction of p75NTR tumor suppressor-mediated apoptosis in prostate and bladder cancer cells (29, 30). Moreover, Celecoxib, a COX-2 inhibitor, has been shown to increase the cytotoxicity of docetaxel in lung cancer (31) and has been approved for use in familial adenomatous polyposis by the FDA (21). Phase II trials have also been conducted to test Celecoxib and carboplatin in patients with pretreated recurrent OC, and this approach is yielding promising results (32). In combination with oxaliplatin, Celecoxib has also been shown to inhibit proliferation and tumor growth in colon carcinoma and breast cancer cells (22, 33).
NSAIDs have been evaluated as therapies in different types of cancer (23, 34, 35) and have been shown to have a synergistic effect with known chemotherapeutic agents (31–33, 36–38). However, none of these studies have looked specifically at NSAIDs effects on nonproliferating cells. Here, we also show that Flurbiprofen, a strong inducer of apoptosis of nonproliferating OC cells, markedly reduced tumor volume by 40% when compared with the control group. Importantly, combinatorial therapy of the NSAID, Sulindac Sulfide, the most potent apoptotic inducer in proliferating OC cells (17), with Flurbiprofen, a strong inducer of apoptosis in nonproliferating cells, increased the therapeutic effect in vivo compared with the monotherapies. We have previously addressed the issue of toxicity and achievable plasma concentrations of Sulindac Sulfide (17, 18). We have previously shown that Sulindac sulfide inhibits proliferation in proliferating cancer cells (18) and suppresses growth of cancers in xenograft mouse models (17, 18). Sulindac sulfide reaches peak plasma concentrations of 30–50 μM, coming down to a steady state plasma concentration of 5–10 μM (18). We have previously demonstrated that the sulindac sulfide steady plasma concentration achievable in patients is still able to induce apoptosis in cancer cells (18).
The dose of Flurbiprofen used here is comparable to that used in previous cancer studies in mouse models (39). In this study 15–20 mg/kg/day was given to mice over an 18-week period, with no deaths as a result of toxicity (39). In a different mouse model study, a dose of 10 mg/kg given twice daily, over a period of 30 days, resulted in no deaths due to toxicity, with an average achievable plasma steady state concentration of 52.2 μg/mL, over a period of 30 min 4 h (40).
The use of NSAIDs or chemically modified NSAIDs that target nonproliferating cancer cells in combination with other therapies that focus on the proliferative stage of the disease should provide a way to develop more potent anticancer drug combinations. Accordingly, this study highlights a new combinatorial therapy for OC that employs a novel DPT approach to attack both nonproliferating and proliferating OC cells. This strategy also holds promise for enhancing current therapies for other cancers.
This work was supported with grants from the National Institutes of Health (NIH) grants 1R01 CA85467 (TAL); NIH grant R01 CA138540 (DS); NIH grants 1R01 CA097318, 1R01 CA127641, 1R01 CA134721, and P01 CA104177, and the Samuel Waxman Cancer Research Foundation and the National Foundation for Cancer Research (PBF); and Department of Defence grant OC0060439 and ICGEB (LFZ). We are thankful to Dr. Rodrigo E. Tamura for his guidance and insight. DS is the Harrison Endowed Scholar in Cancer Research and a Blick scholar. PBF holds the Thelma Newmeyer Corman Chair in Cancer Research and is a Samuel Waxman Cancer Research Foundation (SWCRF) Investigator and a National Foundation for Cancer Research (NFCR) Investigator.