Selective COX-2 inhibitor (celecoxib) decreases cellular growth in prostate cancer cell lines independent of p53.

Celecoxib is a clinically available COX-2 inhibitor that has been reported to have antineoplastic activity. It has been proposed as a preventative agent for several types of early neoplastic lesions. Earlier studies have shown that sensitivity of prostatic carcinoma (PCa) to celecoxib is associated with apoptosis; however, these studies have not demonstrated adequately whether this effect is dependent on p53 status. We studied the relation between sensitivity to celecoxib and the phenotypic p53 status of PCa cells lines, LNCaP (wild type p53), PC3 (null p53) and DU145 (mutated p53). Cellular growth was assessed at 24, 48, 72 and 96 h after celecoxib treatment at concentrations of 0, 10, 30, 50, 70 and 100 µM using an MTT assay. Cellular proliferation (Ki-67 expression) was determined by immunocytochemistry. Phenotypic expression of p53 was analyzed by western blotting. The effects of celecoxib on cellular growth and its association with p53 were assessed after down-regulation of p53 using synthetic interfering RNAs (siRNA) in LNCaP cells. Expression of p53 and COX-2 at mRNA levels was assessed by quantitative real time polymerase reaction (qRT-PCR). We found that celecoxib inhibited cellular growth and proliferation in a dose-dependent manner in all three cell lines; LNCaP cells with a native p53 were the most sensitive to celecoxib. We observed a down- regulation effect on p53 in LNCaP cells exposed to ≥ 30 µM celecoxib for 72 h, but found no significant changes in the p53 levels of DU145 cells, which have a mutated p53. Reduced COX-2 expression was found with decreased p53 in LNCaP and PC-3 cells that were exposed to ≥ 20 µM of celecoxib for 72 h, but COX-2 expression was increased in DU145 cells. All three cell lines demonstrated pan-cytotoxicity when exposed to 100 µM celecoxib. When p53 expression was inhibited using siRNA in LNCaP cells, the inhibitory effects on cellular growth usually exerted by celecoxib were not changed significantly. Celecoxib reduces the growth of prostate cancer cell lines in part by decreasing proliferation, which suggests that the inhibition of growth of LNCaP cells by celecoxib is independent of normal levels of native p53.

Cell density influences the effect of celecoxib in two carcinoma cell lines.

It has been reported that when ovarian carcinoma cell lines are exposed to various concentrations of celecoxib, a COX-2 inhibitor, cell growth is decreased in a dose dependant manner. To examine further the effect of celecoxib, different cell densities of two carcinoma cell lines were exposed to various concentrations of celecoxib. LNCAP prostate and CAOV3 ovarian carcinoma cells were obtained from the American Type Culture Collection and maintained in Rosewell Park Memorial Institute 1640 and Dulbeceo's modified Eagle's medium, respectively. Each cell line was supplemented with 10% fetal bovine serum, 2 mM L-glutamine, and antibiotic-antimycotic solution, and placed in a humidified atmosphere containing 5% CO2 at 37 degrees C. After each cell line reached a confluency of 70-80%, 1,000, 2,000, 3,000, 5,000, 7,000 and 10,000 cells/well were seeded in 96 well plates in 100 microl medium/per well for 24 h. Each cell line was exposed to the same concentrations of celecoxib (10-100 microM) at each cell density for 72 h. Cell growth was assessed using a tetrazolium conversion assay. A significant decrease compared to controls was observed in cell growth at each cell density of LNCAP and CAOV3 cells plated with >or=30 microM and >or=50 microM celecoxib, respectively. When the cell growth curves were compared for each cell density at the same concentration of celecoxib, a significant decrease in cell growth was observed when LNCAP cells were plated at 10,000 cells/well and exposed to 10-100 microM celecoxib. At a cell density >or= 5,000 LNCAP cells/well, the inhibitory effect of celecoxib was less. Similarly, a significant decrease in cell growth was observed in CAOV3 cells plated at 1,000 cells/well compared to other cell numbers plated at the same drug concentrations. At a cell density of > 5,000 CAOV3 cells/well, the inhibitory effect of celecoxib was significantly less compared to other cell densities at the same concentration. We observed a more sensitive decrease in cell growth in both carcinoma lines studied at a cell density of 1,000 cells/well with exposure to 10-100 microM celecoxib. Both carcinoma cell lines were less sensitive at a cell density of 5,000 cells/well. Our results suggest that the inhibitory effect of celecoxib may be affected by cell density. Therefore, careful attention must be paid to determining the appropriate cell density for cytotoxicity studies.

The use of dimethylsulfoxide as a vehicle in cell culture experiments using ovarian carcinoma cell lines.

Dimethylsulfoxide (DMSO) is a well-known solvent that is commonly used in the laboratory. We selected DMSO as the vehicle for an experiment designed to determine if several nonsteroidal anti-inflammatory agents inhibit the growth of Caov-3, OVCAR-3, and SK-OV-3 ovarian carcinoma cell lines. Using the tetrazolium conversion assay, however, we observed some variability in the number of cells present in each ovarian carcinoma cell line with varying concentrations of DMSO (10(-6)-10(-2) M) compared to medium alone. Similarly, when Caov-3, OVCAR-3, and SK-OV-3 cells were treated with 10(-4) M DMSO plus medium (Dulbecco's Modified Eagle Medium with 10% fetal bovine serum) and plated on coverslips, the total number of cells present in 60 random fields increased significantly (P < 0.0001) for each ovarian carcinoma cell line treated with DMSO compared to medium alone. Ethanol did not demonstrate such prominent effects on cellular growth. Our observations are important to consider when selecting an appropriate solvent, especially for growth inhibition studies using Caov-3, OVCAR-3, and SK-OV-3 cell lines.

Immunohistochemical expression of molecular markers in an avian model: a potential model for preclinical evaluation of agents for ovarian cancer chemoprevention.

OBJECTIVE: A significant obstacle confronting the evaluation of potential chemopreventive compounds in ovarian carcinoma is the absence of an animal model of spontaneous ovarian carcinogenesis. A potential model of adenocarcinoma has been described in the laying hen (Gallus domesticus). The purpose of this study was to evaluate the immunohistochemical expression of available antibodies that have been utilized in chemoprevention studies in this potential model of epithelial carcinoma. METHODS: Two hundred 2-year-old hens were sacrificed at Auburn University in accordance with IUACUC guidelines. Of these hens, 8 animals were thought grossly to have ovarian carcinoma and ascites. The tumors from these 8 hens were fixed in neutral-buffered formalin and processed to paraffin blocks. Hematoxylin and eosin stains were used to document the histologic presence of adenocarcinoma. Immunohistochemical evaluation for expression of antigen was performed using the following antibodies: CA125, CEA, cytokeratin, EGFR, erbB-2, Ki-67, Lewis Y, p27, PCNA, Tag 72, TGF-alpha, Muc 1, and Muc 2. RESULTS: Upon microscopic examination by a pathologist eight specimens were documented as adenocarcinomas. Several antibodies to antigens that are frequently expressed in human ovarian cancer were cross-reactive in the laying hen. Of these, cytokeratin AE1/AE3, pan cytokeratin, EGFR, Lewis Y, CEA, Tag 72, and erbB-2 stained the chicken carcinomas. EGFR and p185erbB-2 stained diffusely, and cytokeratin AE1/AE3, pan cytokeratin, Lewis Y, CEA, and Tag 72 were focally positive in the tumor. The aforementioned antibodies which have been useful as surrogate endpoints in chemoprevention trials and which also stained the chicken carcinomas included PCNA, p27, and TGF-alpha Antibodies that were not cross-reactive include CA 125, Ki-67, Muc 1, and Muc 2. CONCLUSION: The data presented in this pilot study support the potential utility of an avian model of spontaneously arising adenocarcinoma in which to study chemopreventive agents. More importantly, the influence of chemoprevention protocols on the expression of relevant antigens can be determined using available antibodies that are cross-reactive in this model. Thus, changes in the phenotypic expression of surrogate endpoint biomarkers as identified by cross-reactive antibodies can aid in the development of chemoprevention trials for human ovarian cancer.

Effects of acute and chronic body weight gain reductions in the evaluation of agents for efficacy in mammary cancer prevention.

Studies were performed to determine the effects of moderate decreases in body weight gain on mammary carcinogenesis. The levels of depressions in weight gain were those often observed in the evaluation of chemopreventive agents. In the first experiment, the effects of acute and chronic reductions of body weight gain when started after carcinogen treatment were examined in young rats (MNU at 50 days of age). Significant decreases (36%) in mammary cancers occurred in groups of rats that underwent a 12% acute reduction in body weight gain as compared with ad libitum controls. In contrast, chronic weight reductions of up to 12% had minimal effects on cancer multiplicities, while a 15% chronic reduction significantly decreased cancer numbers (26%). A second experiment evaluated the efficacy of toremifene (7.0 mg/kg diet), an estrogen/anti-estrogen, and the effect of toremifene-matched body weight gain reduction that occurred during the study. Toremifene caused a chronic reduction in body weight that resulted in a 10% decrease in final body weight at the end of the study. While toremifene-treated rats exhibited a 67% decrease in the number of mammary cancers, the rats which similarly exhibited a 10% decrease in final body weight showed only a 14% decrease in cancer number. Thus, the weight effects observed with toremifene, similar estrogens/anti-estrogens, and other classes of chemopreventive compounds (where chronic body weight reductions are 10% or less) imply that the body weight reduction has a limited effect on overall chemopreventive activity. A third study examined the effect of chronic body weight gain reduction on mammary cancers induced in older rats (MNU given at 100 days of age). This model more closely resembles the status of the breast tissue of mature women currently enrolled in clinical trials of chemopreventive agents. Under these conditions chronic reductions in body weight up to 15% had minimal effects on mammary carcinogenesis. These data further demonstrated that acute body weight reductions in young rats at the time of carcinogen treatment can be a concern in interpretation of the chemopreventive activity of an agent, but that moderate chronic depressions of body weight gain probably do not play a significant role.

Longitudinal analysis of mammary cancer multiplicity in chemoprevention studies.

BACKGROUND: The effect of chemopreventive agents on cancer multiplicity is of primary interest in animal studies. The nature of data collected from chemoprevention studies may be analyzed by a longitudinal analysis of repeatedly measured cancer multiplicity data. METHODS: We determined the number of mammary cancers over the entire follow-up period for varying doses of two chemopreventive agents. Longitudinal analyses were performed to model the number of cancers over different time intervals. RESULTS: There was a significant increase in the number of cancers between six to seven weeks post-carcinogen administration in the control group. Varying patterns of cancer development were observed at different doses of chemopreventive agents including a delay in onset of tumor growth compared to the control group. CONCLUSION: Longitudinal data analysis complements traditional analyses by providing detailed information regarding the effect of chemopreventive agents on the pattern of tumor development throughout the follow-up period. Importantly, some chemopreventive agents may delay time to appearance of mammary cancers without causing a significant difference in cancer multiplicity.

Effect of reduced body weight gain on the evaluation of chemopreventive agents in the methylnitrosourea-induced mammary cancer model.

These studies examined whether the small to moderate reductions in body weight gain (< or = 15%) affect mammary carcinogenesis. Beginning 1 week prior to methylnitrosourea (MNU) administration (experiment 1), rats received diets supplemented with 4-hydroxyphenylretinamide (4-HPR) (782 mg/kg of diet) and retinyl acetate (328 mg/kg of diet) or underwent food restrictions. Rats were administered an i.v. dose of MNU (50 mg/kg body wt) at 50 days of age. Although the final body weights were similarly depressed by 4-HPR (8%) and by retinyl acetate (11%) from rats fed ad libitum, the kinetics of inhibition were quite different. Whereas 4-HPR caused an acute decrease in body weight at the time it was administered, the effect of retinyl acetate was more chronic. At 110 days after the administration of MNU, the average number of mammary cancers per rat was 4.9 for rats fed ad libitum, 1.3 for rats fed 4-HPR, 3.1 when body weights were matched to 4-HPR-treated rats, 1.9 for retinyl acetate and 3.2 when body weights were matched to retinyl acetate. Experiment II was performed to determine the minimal degree of acute body weight gain reduction that would alter MNU-induced mammary carcinogenesis. Body weight gain depressions of 3, 6, 9, 12 and 15% were initiated at 43 days of age by dietary restrictions and MNU was administered at 50 days of age. At 120 days after MNU, the percentage decreases in mammary cancer multiplicity in the various groups were 14, 15, 41, 44 and 55%, respectively. These data demonstrate that moderate reductions (9-15%) in body weight gain, in particular when occurring during the initiation and early promotion stages can greatly affect cancer multiplicity.