SEARCH

SEARCH BY CITATION

Keywords:

  • meloxicam;
  • COX-2 inhibitor;
  • ciglitazone;
  • PPARγ ligand;
  • ovarian cancer

Abstract

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

BACKGROUND.

It was recently reported that high expression of peroxisome proliferator-activated receptor γ (PPARγ) and low expression of cyclooxygenase-2 (COX-2) might be involved in the inhibition of ovarian tumor progression and confirmed that PPARγ activation could suppress COX-2 expression via the nuclear factor-κB pathway in ovarian cancer cells.

METHODS.

The current study investigated whether meloxicam, a selective COX-2 inhibitor, and ciglitazone, a ligand for PPARγ, inhibit the growth of human ovarian cancer cell lines and aimed to elucidate the molecular mechanism of their antitumor effect. Tumor growth and survival were examined in female nu/nu mice xenografted with subcutaneous OVCAR-3 tumors or with intraperitoneal DISS tumors and treated with meloxicam (162 ppm in diet, every day) or ciglitazone (15 mg/kg intraperitoneally once a week).

RESULTS.

Both meloxicam and ciglitazone treatments significantly suppressed the growth of OVCAR-3 tumors xenotransplanted subcutaneously and significantly prolonged the survival of mice with malignant ascites derived from DISS cells as compared with controls. Meloxicam treatment decreased COX-2 expression in tumors by 2.5-fold compared with that observed in untreated tumors. Although ciglitazone treatment did not alter COX-2 expression in tumors, it reduced the expression of microsomal prostaglandin (PG) E synthase, which converts COX-derived PGH2 to PGE2. Both meloxicam and ciglitazone decreased PGE2 levels in serum as well as in ascites. Reduced microvessel density and induced apoptosis were found in solid OVCAR-3 tumors treated with either meloxicam or ciglitazone.

CONCLUSIONS.

These results indicate that both meloxicam and ciglitazone produce antitumor effects against ovarian cancer in conjunction with reduced angiogenesis and induction of apoptosis. Cancer 2007; 110:791–800. © 2007 American Cancer Society.

Ovarian cancer is an insidious disease that has few specific symptoms in the early stages and most women with this disease present with an advanced stage at the time of diagnosis. The current management of advanced epithelial ovarian cancer generally includes cytoreductive surgery followed by combination chemotherapy. The combination of paclitaxel with a platinum analog is the preferred chemotherapy regimen in the treatment of newly diagnosed patients with this disease.1 Although such management induces favorable results in the treatment of ovarian cancer, the long-term survival of ovarian cancer patients remains unsatisfactory2 and an estimated 130,000 deaths per year still occur from ovarian cancer worldwide.3 Acquisition of drug tolerance by ovarian cancer cells is regarded as 1 of the causes underlying the failure to prolong the survival period of ovarian cancer patients.4 Although it is important to elucidate the mechanisms of drug resistance to overcome it, new medications, especially medications distinct from the known chemotherapeutic agents, remain to be developed with a view to improving survival and cure rates in ovarian cancer.

Cyclooxygenase (COX) has 2 isoforms, the constitutive COX-1 and the inducible COX-2.5 COX-1 is expressed in most tissues, whereas COX-2 is largely absent but is primarily responsible for prostaglandins (PGs) produced in inflammatory sites, suggesting that COX-2 plays a critical role in inflammation.6 Epidemiologic studies have shown a 40% to 50% reduction in mortality from colorectal carcinoma in continuous users of nonsteroidal antiinflammatory drugs (NSAIDs) compared with that of noncontinuous users.7 The antitumor effect of NSAIDs is due to their inhibition of COX-2.8 Many lines of evidence suggest that COX-2 might be involved in various aspects of carcinogenesis, tumor progression, and recurrence.9, 10 Shigemasa et al.11 reported that COX-2 expression might play an important role in the development of ovarian cancer. Ferrandina et al.12 demonstrated that increased COX-2 expression was associated with chemotherapy resistance and outcome in ovarian cancer patients and then Denkert et al.13 reported that COX-2 expression was an independent prognostic factor in ovarian cancer. Pharmacologic studies suggest that COX-2 is a useful therapeutic target and COX-2 inhibitors reduce the formation, growth and metastasis of experimental tumors.14, 15 Conversely, more recently COX-1 has also been found to play an important role in the growth of some types of malignant tumors via the production of angiogenic growth factor.16, 17

Peroxisome proliferator-activated receptor γ (PPARγ) is a member of a nuclear hormone receptor superfamily and provides a strong link between lipid metabolism and the regulation of gene transcription.18 Ligand-mediated activation of PPARγ has been linked to cellular differentiation, apoptosis, and antiinflammatory responses.19 Significant evidence from many experimental systems suggests that PPARγ plays an important role in carcinogenesis. PPARγ is up-regulated in malignant tissues, and ligands for PPARγ inhibit the growth of human colon, breast, gastric, and lung cancer cells.19–22 We recently reported that PPARγ activation could suppress COX-2 expression via the nuclear factor-κB pathway in ovarian cancer cells, suggesting the potential of PPARγ ligands as a new agent in the treatment of ovarian malignancies.23

Meloxicam is used worldwide as an NSAID and is categorized as a selective COX-2 inhibitor. It has been reported that meloxicam has an inhibitory effect on colorectal cancer cells24 and nonsmall-cell lung cancer cells.25 Conversely, ciglitazone is used clinically as an antihyperglycemic agent and is categorized as a PPARγ ligand. Ciglitazone decreases COX-2 expression and induced apoptosis in human colon cancer cells overexpressing COX-2.26 Although there has been significant interest in the role of COX-2 inhibitors and PPARγ ligands in colon, breast, and lung cancers,26–28 data on COX-2 inhibitors and PPARγ ligands in ovarian malignancies are quite scant. In this study we investigated the effect of meloxicam and ciglitazone on the growth of malignant ovarian tumors in vivo and aimed to elucidate the molecular mechanism of their antitumor effect. As a positive control for antitumor activity, we used cisplatin (cis-diaminedichloroplatinum [II]), which is a DNA-damaging agent used in the chemotherapy of human malignancy and is considered 1 of the key drugs for ovarian cancer treatment.

MATERIALS AND METHODS

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

Cell Lines and Cell Culture

OVCAR-3 was obtained from the American TypeCulture Collection (Rockville, MD) and DISS was kindly provided by Dr. Saga (Jichi Medical School, Tochigi, Japan). Both cell lines were derived from human epithelial ovarian adenocarcinoma, grown in RPMI-1640 medium supplemented with 10% fetal bovine serum (FBS), 100 U/mL penicillin, and 100 μg/mL streptomycin at 37°C in a water-saturated atmosphere with 5% carbon dioxide/95% air. Both cell lines grow in monolayer.

Animal Experimentation

Animal experiments were conducted in accordance with the Guidelines for Animal Experimentation, Hirosaki University. Eight-week-old female BALB/c nu/nu mice were used in this study. All mice were group-housed in plastic cages with stainless-steel grid tops in an air-conditioned and 12-hour light-dark cycle-maintained room at the Institute for Animal Experiments of Hirosaki University and consumed water and food ad libitum.

Cancer-Bearing Mouse Model

OVCAR-3 cells (5 × 106 cells) were inoculated subcutaneously in 500 μL of RPMI-1640 medium in the back region of the nude mice. All the mice were numbered, housed separately, and examined twice weekly for tumor development. One week after inoculation, small tumors were identified. Tumors were grown until the longer dimension reached 2 mm before starting treatment. The experimental mice were then divided into 4 groups containing 10 mice each (Day 0). Controls received only the basal diet. Each mouse in the meloxicam group was housed separately and given a solid diet supplemented with 162 ppm meloxicam (Boehringer Ingelheim, Germany) every day until the end of the study.29 Consumption was monitored every day. The ciglitazone group was administered ciglitazone (TOCRIS Bioscience, St. Louis, Mo) at a dose of 15 mg/kg intraperitoneally once a week until the end of the study.30 The cisplatin group was administered cisplatin (Nippon Kayaku, Tokyo, Japan) at a dose of 5 mg/kg intraperitoneally once on Day 0. The tumor dimensions were measured twice weekly using a vernier caliper and tumor volume was determined by external measurement according to the published method. Volume was determined by the equation V (mm3) = L × W2 × 0.5, in which L is length and W is width.31 Serum PGE2 concentration was determined on Day 7 and the mice were sacrificed on Day 21 to remove the tumor for pathologic and biochemical studies. Hematoxylin and eosin staining of the removed tumors showed human ovarian epithelial malignancy (data not shown).

Peritoneal Carcinomatosis Mouse Model

DISS cells (0.5 × 107 cells) were inoculated into the peritoneal cavity of nude mice in 500 μL of sterile phosphate-buffered saline (PBS). It has been reported that the average survival of DISS cell-transplanted mice is about 30 days.32 The experimental mice were divided into 4 groups of 10 mice each. After confirming ascites on Day 7 the mice were treated in the same way as in the cancer-bearing mouse model. Ascites was aspirated on Day 21 for the determination of PGE2 and VEGF concentrations, then the survival time for each group was evaluated.

Western Blot Analysis

Removed tumor tissues were cut into small pieces and homogenized in PIPA buffer in ice. The homogenate was incubated overnight at 4°C followed by centrifuging at 15,000g for 15 minutes at 4°C. The supernatant fluids (50 μg protein) were electrophoresed through a 10% sodium dodecyl sulfate (SDS)-polyacrylamide gel and blotted as described previously.23 The protein concentration was determined using the Bradford method. The blots were probed with the following primary antibodies: COX-2 (IBL, Gunma, Japan) at 2 μg/mL, PPAR-γ (Cayman Chemical, Ann Arbor, Mich) at 1 μg/mL, microsomal PGE synthase (mPGES) (Cayman Chemical) at 1:1000, or β-actin (Sigma-Aldrich, St Louis, Mo) at 1:2000. The membranes probed by COX-2, PPAR-γ, and mPGES were incubated for 2 hours with antirabbit immunoglobulin (Ig) G conjugated to horseradish peroxidase and then immunoblots were visualized using diaminobenzidine (DAB) (Sigma-Aldrich) as a substrate of peroxidase. β-Actin was used as a loading control. The membranes probed by β-actin were incubated for 1 hour with biotinylated antimouse immunoglobin, transferred to avidin-biotin-peroxidase complex reagent (Vector Laboratories, Burlingame, Calif), and incubated in this solution for 30 minutes. DAB (Sigma-Aldrich) was used as a substrate. Quantification of the results was performed by scanning the membrane with Photoshop software (version 5.5; Adobe Systems, San Jose, Calif) followed by densitometry with the public domain software NIH Image (version 1.62; National Institutes of Health, Bethesda, Md).

Measurement of PGE2 in Serum and Ascites and Measurement of VEGF in Ascites

PGE2 concentrations were determined with PGE2 EIA system (R&D Systems, Minneapolis, Minn) according to the manufacturer's instructions. VEGF concentrations were determined using an enzyme-linked immunoadsorbent assay (ELISA) kit (R&D Systems) as described by Gu et al.33

Immunohistochemical Analysis and Microvessel Density

Six-μm sections of formalin-fixed and paraffin-embedded tissue specimens were stained by the established method previously described.34 Sections were incubated with antibodies specific for mPGES or VEGF (R&D Systems) for 1 hour and CD31 (R&D Systems) overnight. Slides were incubated with biotinylated species-specific appropriate secondary antibodies for 30 minutes and then exposed to avidin-biotin-peroxidase complex (Vector Laboratories). Sections were treated with 0.02% DAB as a chromogen and counterstained with hematoxylin. VEGF expression was evaluated according to a scoring method that uses the positive cell percentage and the staining intensity as reported previously.29 Microvessel density was determined as follows. The highly vascularized areas of the tumor stained with an anti-CD31 antibody were identified and CD31-positive microvessels were counted under high power in a field of 0.75 mm2 area (×200 magnification). Single endothelial cells or clusters of endothelial cells, with or without a lumen, were considered individual vessels. Microvessel density was expressed as the vessel number/high-power field in sections. Three fields were counted per animal and the average was taken as the microvessel density of each tumor.

Apoptosis

Apoptosis was measured in tissue sections by the terminal deoxyribonucleotidyl transferase-mediated dUTP-biotin nick end labeling (TUNEL) assay as described by Gavrieli et al.35 with some modifications. Briefly, 6-μm sections were stripped from proteins by incubation with 10 mg/mL proteinase-K for 15 minutes and immersed in 0.3% H2O2 in methanol for 15 minutes to block the endogenous peroxidase. The sections were then incubated in TdT mixture buffer (200 mM potassium cacodylate, 25 mM Tris-HCl [pH 6.5], 0.25 mg/mL BSA, 1 mM CoCl2, 0.01 mM biotin-dUTP, and 520 U/mL TdT) at 37°C for 1 hour. After rinsing in PBS the sections were exposed to avidin-biotin-peroxidase complex at 37°C for 30 minutes. Cells undergoing apoptosis were visualized with DAB. The number of stained tumor cells was counted in the 3 fields with highest activity under a high-power field of 0.75 mm2 area (×200 magnification) and the results were averaged.

Statistical Analysis

Differences in tumor volume between the groups were analyzed using the nonparametric Mann-Whitney U test. Survival curves were calculated by the Kaplan-Meier method, and the statistical significance of differences in the cumulative survival curves between the groups was evaluated by log-rank test. Other statistical analyses were performed using the Student t-test, chi-square test, or Fisher exact probability test. A result was deemed significant at P < .05.

RESULTS

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

Antitumor Effect of Meloxicam and Ciglitazone in Cancer-Bearing Mouse Model and Peritoneal Carcinomatosis Mouse Model

In the cancer-bearing mouse model, at the end of the experiment, the tumor volumes were 5.29 ± 1.25, 3.32 ± 0.89, 3.64 ± 0.74, 2.99 ± 0.77 cm3 (expressed as mean ± SD) in the control, cisplatin, meloxicam, and ciglitazone groups, respectively (Fig. 1A). The Mann-Whitney U test showed that tumor volume curves at Days 8–21 in the cisplatin, meloxicam, and ciglitazone groups were significantly lower (P < .0001) than in the control group (Fig. 1A). In the comparison of final tumor weights, tumor inhibition was 34% in the cisplatin group, 27% in the meloxicam group, and 47% in the ciglitazone group, tumors being significantly smaller in every treatment group than in the control group (Fig. 1B). In the peritoneal carcinomatosis model, survival times were significantly prolonged in the cisplatin, meloxicam, and ciglitazone groups, compared with the control group (Fig. 1C). No significant difference in survival time was observed between the cisplatin and meloxicam or ciglitazone groups. Treatment with meloxicam or ciglitazone did not result in weight reduction or gain and did not cause any adverse effects.

thumbnail image

Figure 1. Antitumor effect of meloxicam and ciglitazone in a cancer-bearing mouse model (OVCAR-3) and peritoneal carcinomatosa mouse model (DISS). (A) Comparison of tumor growth in cancer-bearing mice. A Mann-Whitney U test showed that tumor volume curves at Days 8 to 21 in the cisplatin (CDDP), meloxicam, and ciglitazone groups were significantly lower than in the control group (P < .0001). (B) Comparison of final tumor weights in cancer-bearing mice. The final tumor weight was significantly smaller in every group than in the control group. Results represent means ± standard deviation. *P < .01 and **P < .001 versus the control group, respectively. (C) Comparison of survival period in peritoneal carcinomatosa mice. The statistical significance of differences in the 100%-cumulative survival curves between the groups was evaluated by the log-rank test. Survival times were significantly prolonged in the cisplatin, meloxicam, and ciglitazone groups compared with the control group (P < .0005, P < .0001, P < .02, respectively).

Download figure to PowerPoint

Altered Expression of COX-2 and PPARγ in Tumors by Meloxicam or Ciglitazone Administration

To examine whether the expression of COX-2 and PPARγ in tumors is altered after meloxicam and ciglitazone treatment, COX-2 and PPARγ expression was evaluated by Western blotting. This analysis revealed that after ciglitazone treatment, PPARγ expression increased 2.8-fold, whereas COX-2 expression did not change, and that after meloxicam treatment COX-2 expression decreased 2.5-fold and PPARγ expression increased 2.2-fold (Fig. 2). There was no significant effect of cisplatin on COX-2 and PPARγ expression (Fig. 2).

thumbnail image

Figure 2. Altered expression of cyclooxygenase (COX)-2 and PPARγ in tumors treated with meloxicam or ciglitazone. COX-2 expression decreased 2.5-fold and PPARγ expression simultaneously increased approximately 2-fold in tumors treated with meloxicam. PPARγ expression increased 2.8-fold but COX-2 expression did not alter in tumors treated with ciglitazone. Densitometric values represent the actin-corrected means of 6 blots. Blots shown are from 1 representative experiment of 6. CDDP indicates cisplatin; kDa, kilodaltons.

Download figure to PowerPoint

Decreased PGE2 Concentration in Serum and Ascites With Meloxicam or Ciglitazone Administration

The mean concentration of PGE2 in serum was 695 ± 56 pg/mL for the control group, whereas it was 104 ± 25 pg/mL and 153 ± 28 pg/mL for the meloxicam and ciglitazone groups, respectively (Fig. 3, black bars). The concentration of PGE2 in serum decreased significantly in the meloxicam and ciglitazone groups compared with the control and cisplatin groups. The mean concentration of PGE2 in ascites was 708 ± 256 pg/mL and 423 ± 105 pg/mL for the control and cisplatin groups, respectively, whereas it was 152 ± 85 pg/mL and 179 ± 82 pg/mL for the meloxicam and ciglitazone groups, respectively (Fig. 3, white bars). The concentration of PGE2 in ascites was significantly lower in the meloxicam and ciglitazone groups than in the control and cisplatin groups.

thumbnail image

Figure 3. Determination of PGE2 concentrations in serum (black bars) and ascites (white bars). The concentration of PGE2 in both serum and ascites was significantly lower in the meloxicam and ciglitazone groups than in the control and cisplatin (CDDP) groups. Results represent means ± standard deviation. *P < .0005 and **P < .02 versus the control group, #P < .005 and ##P < .05 versus the cisplatin group, respectively.

Download figure to PowerPoint

Altered Expression of mPGES in Tumors Treated With Ciglitazone

To determine whether the expression of mPGES (a synthase that converts PGH2 to PGE2) alters in tumors treated with meloxicam or ciglitazone, the distribution of mPGES in tumors was immunohistochemically evaluated. In the control and meloxicam groups, mPGES expression was evenly distributed in the periphery and center of the tumor (Fig. 4A and 4C), whereas its expression was strong in tumor cells close to the stroma in the cisplatin group (Fig. 4B). In the ciglitazone group, overall mPGES expression was decreased and it was faintly expressed only in tumor cells close to the stroma (Fig. 4D). Western blot analysis demonstrated that the amount of intratumoral mPGES was much lower in the ciglitazone group compared with other groups (Fig. 4E).

thumbnail image

Figure 4. Immunohistochemical staining and Western blot analysis of OVCAR-3 tumors with anti-mPGES antibody. (A) OVCAR-3 tumor without treatment. (B) Tumor treated with cisplatin (CDDP). (C) Tumor treated with meloxicam. (D) Tumor treated with ciglitazone. Staining pictures shown are a single representative of each group of 10 tumors. Scale bars = 25 μm. (E) Intratumoral mPGE. Blots shown are 1 representative experiment of 3. kDa indicates kilodaltons.

Download figure to PowerPoint

Induction of Apoptosis and Reduced Microvessel Density in Tumors by Meloxicam or Ciglitazone

To evaluate the extent of apoptosis in tumor tissue in a cancer-bearing mouse model, apoptotic cells were stained by the TUNEL method and the number of TUNEL-positive cells per 0.75 mm2 was counted in a high-power field. The incidence of TUNEL-positive cells (number/mm2) was 12.8 ± 1.3 for the control group, 11.0 ± 1.8 for the cisplatin group, 28.3 ± 8.1 for the meloxicam group, and 29.7 ± 10.5 for the ciglitazone group (Fig. 5A). The incidence of apoptotic cells was significantly higher in the meloxicam and ciglitazone groups than in the control and cisplatin groups. We then examined the number of microvessels identified with CD31 in tumor tissues using immunostaining. Microvessel density (MVD) (number/mm2) was 23.3 ± 4.1 for the control group, 22.7 ± 8.0 for the cisplatin group, 8.9 ± 3.2 for the meloxicam group, and 16.0 ± 4.7 for the ciglitazone group, showing a significant decrease in the meloxicam and ciglitazone groups as compared with the control and cisplatin groups (Fig. 5B).

thumbnail image

Figure 5. Determination of apoptosis and microvessel density (MVD) in tumors and determination of vascular endothelial growth factor (VEGF) in tumors and ascites. (A) Comparison of the incidence of apoptotic cells in tumors. The incidence of apoptotic cells was significantly higher in the meloxicam and ciglitazone groups than in the control and cisplatin (CDDP) groups. Results represent means ± standard deviation (SD). *P < .001 and **P < .05 versus the control group, and #P < .001 and ##P < .05 versus the cisplatin group, respectively. (B) MVD identified with anti-CD31 antibody in tumors. MVD was decreased significantly in the meloxicam and ciglitazone groups compared with the control and cisplatin groups. Results represent means ± SD. *P < .001 and **P < .05 versus the control group, and #P < .001 and ##P < .05 versus the cisplatin group. (C) Semiquantification of VEGF expression in tumors. VEGF decreased significantly in the meloxicam and ciglitazone groups compared with the control and cisplatin groups. Results represent means ± SD. §P < .05 versus the control and cisplatin group. (D) Determination of VEGF concentration in ascites. VEGF decreased significantly in the meloxicam and ciglitazone groups compared with the control and cisplatin groups. Results represent means ± SD. §P < .05 and §§P < .005 versus the control and cisplatin groups, respectively. TUNEL indicates terminal deoxynucleotidyl transferase mediated dUTP nick end labeling assay.

Download figure to PowerPoint

Decreased Expression of VEGF in Tumors and Ascites With Meloxicam or Ciglitazone

The amount of VEGF in tumor tissues was semiquantified using immunohistochemical staining. The amount of VEGF was 5.4 ± 0.5 for the control group, 4.3 ± 0.7 for the cisplatin group, 2.8 ± 0.9 for the meloxicam group, and 2.4 ± 1.3 for the ciglitazone group (Fig. 5C). The amount of VEGF was significantly lower in the meloxicam and ciglitazone groups compared with the control and cisplatin groups. The amount of VEGF in ascites measured by ELISA was 582.3 ± 112.0 pg/mL for the control group, 391.3 ± 44.8 pg/mL for the cisplatin group, 254.3 ± 41.1 pg/mL for the meloxicam group, and 109.0 ± 28.1 pg/mL for the ciglitazone group (Fig. 5D). The values were significantly lower in the meloxicam and ciglitazone groups than in the control and cisplatin groups.

DISCUSSION

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

In this study it emerged that both meloxicam and ciglitazone suppressed the growth of solid malignant tumors and peritoneal carcinomatosa derived from human ovarian cancer cells in conjunction with reduction of angiogenesis and significant induction of apoptosis. Their antitumor effect was comparable to that of cisplatin (Fig. 1). This is the first report describing the inhibitory effect of a COX-2 inhibitor and a PPARγ ligand on the growth of ovarian cancer.

As shown in Figure 3, meloxicam and ciglitazone reduced PGE2 production in the serum of subcutaneously xenografted mice and in the malignant ascites. Western blot analysis revealed that meloxicam not only down-regulated expression of COX-2 but also up-regulated that of PPARγ, whereas ciglitazone up-regulated that of PPARγ without modulating COX-2 (Fig. 2). Thus, it is suggested that meloxicam and ciglitazone led to inhibition of PGE2 by COX-2-dependent and independent routes, respectively. It has been shown that PGE2 enhances angiogenesis through the induction of VEGF36 and represses apoptosis by maintaining Bc1-2 expression.37 Munkarah et al.38 also reported that in vitro PGE2 treatment stimulated proliferation of ovarian cancer cells and reduced apoptosis. PGE2 production is involved in the ability of cancer cells to invade, metastasize, and grow.39 Moreover, mPGES, which converts COX-derived PGH2 to PGE2, plays an important role in releasing PGE2 from cancer cells40 and its enhanced expression is important in tumorigenesis. Cells overexpressing both COX-2 and mPGES produced more PGE2, grew faster, and exhibited more aberrant morphology than those expressing either COX-2 or mPGES,41 indicating the pivotal role of PGE2 in tumorigenesis. We found in this study that, whereas meloxicam did not alter expression of mPGES in OVCAR-3 tumors, ciglitazone treatment reduced its expression in the tumors (Fig. 4), resulting in a decrease in the level of PGE2. These results suggest that meloxicam and ciglitazone elicit their antitumor effects by suppressing PGE2 production via different pathways. The PGE2 measured in serum and ascites might come from circulating monocytes and peritoneal macrophages as well as from cancer cells.

Angiogenesis plays a crucial role in tumor development and progression. PGE2 is a potent inducer of angiogenesis in vivo and induces expression of angiogenic regulatory proteins such as VEGF.42 We examined whether meloxicam or ciglitazone treatment in vivo reduced VEGF levels in the solid tumors and the malignant ascites as well as MVD in the solid tumors, and found significantly lower levels of VEGF in meloxicam- or ciglitazone-treated tumors and ascites, accompanied by a significant reduction in MVD compared with the control and cisplatin groups (Fig. 5). These results suggest that PGE2 plays a role in mediating angiogenesis in ovarian cancer. Additional mechanisms are undoubtedly involved in mediating the angiogenic effects of PGE2 via COX-2-dependent or -independent routes, but the present results suggest that PGE2 affects angiogenesis partly by enhancing VEGF secretion from tumor cells. Naruse et al.43 reported that meloxicam markedly reduced the expression of VEGF in a lung lesion metastatic from osteosarcoma compared with control tissues. Celecoxib, a COX-2 inhibitor, might suppress growth of lung and breast tumors due to the potent antiangiogenic activity it produces by reducing COX-2-derived PGE2.28, 44 Conversely, PPARγ ligands also play a potent role in modulating angiogenesis. They not only inhibit VEGF receptor-2 expression but also act as a blocker of VEGF-receptor signaling.45, 46 Keshamouni et al.45 reported that PPARγ ligands repress tumor growth by inhibition of angiogenesis via blocking the production of ELR+CXC chemokines mediated through antagonizing NF-κB activation. Sarayba et al.46 also reported that pioglitazone, a PPARγ ligand, significantly decreased the density of angiogenesis in a VEGF-induced neovascular model. Taking these results together with ours suggests that the suppressive effects of meloxicam and ciglitazone on tumor growth might be due to decreased angiogenesis through inhibited VEGF production in relation to PGE2 reduction. We recently reported that in ovarian cancer cells transfection of carbonyl reductase, which converts PGE2 to PGF, inhibited VEGF production, resulting in induction of apoptosis and inhibition of tumor growth.32

In this study both meloxicam and ciglitazone induced apoptosis in ovarian cancer cells. Reduced production of PGE2 after meloxicam or ciglitazone treatment may be the mechanism underlying enhanced tumor cell apoptosis. It has been shown that PGE2 reduces the basal apoptotic rate by increasing the level of antiapoptotic proteins such as Bcl-2.37 Many pathways that mediate the apoptotic effect of COX-2 inhibitors have been identified in colon cancer cells. Celecoxib may induce apoptosis in part via a COX-2-dependent pathway that causes activation of caspase-3 and -9 together with cytochrome c release, and in part a COX-2-independent pathway such as down-regulation of cyclin A and B1 and up-regulation of p21, Waf1, and Kip1.47 Another COX-2 inhibitor, NS-398, is reported to induce apoptosis through a MEK/ERK pathway.48 Thus, a clear picture of the exact pathways by which COX-2 inhibitors induce apoptosis has yet to emerge. Conversely, PPARγ also plays a crucial role in apoptosis in a variety of cells. PPARγ activation resulted in enhanced apoptosis of some types of malignant cells such as prostate cancer cells49 and gastric cancer cells.50 Martelli et al.51 reported that apoptosis was induced in thyroid carcinoma cells transfected with PPARγ. More recently, Yang and Frucht26 reported that ciglitazone induced colon cancer cells to undergo apoptosis through the PPARγ pathway mediated by a caspase-independent mechanism. Taken together, the present results suggest that the suppressive effects of meloxicam and ciglitazone on tumor growth might be related to the induction of apoptosis via PGE2 reduction. Anticancer drugs such as docetaxel induce apoptosis through a mild to moderate reduction in PGE2 activity.52

In this study, meloxicam suppressed COX-2 expression in OVCAR-3 tumor (Fig. 2). The mechanisms of COX-2 regulation by NSAIDs vary between NSAIDs and cell types.53 COX-2 regulation by NSAIDs is via their inhibition of COX activity, whose endproducts exert a feedback on COX-2 expression.53 In addition, regulation of COX activity is associated with PPAR activation.54 The possibility may be suggested that decreased expression of COX-2 in tumors treated with meloxicam is via PPARγ activation (Fig. 2).

We conclude that meloxicam and ciglitazone could be effective agents in human ovarian cancer and should be clinically tested alone and in combination with other molecular-targeted agents or cytotoxic drugs.

Acknowledgements

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

Supported in part by a Grant-in-Aid for Cancer Research (No. 16591632) from the Ministry of Education, Science and Culture of Japan and by the Karoji Memorial Fund of the Hirosaki University School of Medicine.

REFERENCES

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