Expression of adiponectin and its receptors is altered in epithelial ovarian tumors and ascites-derived ovarian cancer cell lines.

OBJECTIVES: Recent evidence suggests that higher body mass index is associated with a modest increase in ovarian cancer risk. Reduced serum levels of adiponectin are correlated with obesity and increased cancer risk. The objectives of the present study are to determine if expressions of adiponectin and its receptors, AdipoR1 and AdipoR2, are altered in epithelial ovarian tumors and ascites-derived ovarian cancer cell lines and to determine if plasma adiponectin levels are altered in the chicken model of ovarian cancer. METHODS: Adiponectin, AdipoR1, and AdipoR2 mRNA concentrations in ovaries and chicken ovarian cancer (COVCAR) cell lines were determined by quantitative real-time polymerase chain reaction analysis. Existence of adiponectin isoforms in the ovaries and COVCAR cells was identified by nondenaturing gel electrophoresis. Adiponectin, AdipoR1, and AdipoR2 protein amounts were determined by Western blot analysis. Plasma total adiponectin levels were determined by an enzyme immunoassay. RESULTS: Adiponectin, AdipoR1, and AdipoR2 mRNA concentrations were significantly lower in cancerous ovaries and COVCAR cell lines compared with normal ovaries and normal ovarian surface epithelial (NOSE) cells, respectively. Adiponectin in ovary and COVCAR cell lines appeared as a heavy-molecular-weight isoform that is greater than 720-kd mass. In addition, a lower-molecular-weight adiponectin isoform was found in COVCAR cells but not in NOSE cells. Adiponectin and AdipoR1 protein concentrations were not different in COVCAR cell lines compared with NOSE cells. However, AdipoR2 protein concentrations were significantly higher in cancerous ovaries but lower in COVCAR cell lines compared with normal ovaries and NOSE cells, respectively. Plasma adiponectin concentrations were not different in chickens that had ovarian carcinoma compared with control animals. CONCLUSIONS: Expression of adiponectin in ovarian tumors and in metastatic ovarian tumor cells is likely to affect cellular metabolism and proliferation through activating AdipoR1 and/or AdipoR2. Plasma adiponectin levels may not be predictive of advanced stages of ovarian tumor in the chicken model.

The restricted ovulator chicken: a unique animal model for investigating the etiology of ovarian cancer.

OBJECTIVE: The main goal of this study was to compare the incidence of ovarian cancer (OC) in 2 genetically different lines of hens--one that generally fails to lay eggs (the mutant "restricted ovulator" [RO] strain) and the other consisting of the wild-type (WT) siblings of the mutant RO hens. METHODS: Individual egg production data were obtained over a 972-day period for 31 RO hens and 33 WT hens. At 38 months of age, hens were killed, and their abdominal cavities were examined for any gross evidence of tumors. Samples of ovarian tissue were processed and assessed for histopathology and protein expression of ovalbumin. Plasma estradiol concentrations were also determined. RESULTS: Only 1 (3%) of the 31 RO hens was diagnosed with OC as compared with 9 (27%) of the 33 WT hens (P G 0.05). Wild-type hens laid more eggs than did RO hens during the 31-month collection period (average of 422 vs 28, respectively; P < 0.05). Although there was no difference in overall rate of ovulation between hens with and without OC, WT hens diagnosed with OC laid a greater percentage of their total number of eggs in the first year of production. Plasma estradiol concentrations were higher (P < 0.01) in RO versus WT hens. CONCLUSIONS: The results of this study strongly suggest that the number of ovulatory events is directly related to the incidence of OC in chickens. Clearly, other factors modify the risk of OC because there was no difference in ovulation rate between WT hens with and without OC. The mutant RO hen represents a valuable animal model for studying the etiology of OC.

Characterization of ovarian surface epithelial cells from the hen: a unique model for ovarian cancer.

To further develop the hen as a model of ovarian adenocarcinoma, we have studied normal and neoplastic ovaries as well as cultured cells from the ovarian surface epithelium (OSE). We characterized the OSE layer of the hen for specific histologic markers and evaluated these markers on tumor tissue. We also isolated and characterized the epithelial cells that are the likely source of the ovarian tumors of the hen. The surface epithelium of normal ovaries demonstrated positive staining for cytokeratin, proliferating cell nuclear antigen (PCNA), progesterone receptor (PR), and negative staining for vimentin. Ovarian tumors demonstrated positive cytokeratin, PCNA, PR, and weak vimentin staining in the gland-like areas. Epithelial cell cultures were obtained by an explant method utilizing small and large yellow follicles. These cells were positive for cytokeratin and negative for vimentin on Days 1 and 3. By Day 10, cytokeratin protein expression was less for some cells, and vimentin expression was weakly present in some cells. Expression of PCNA was observed at Days 1 and 3, but was rarely seen in cells cultured for 10 days. Expression of PR was observed on Day 10 after 24-hr estrogen treatment. Epithelial cells grew slowly in culture, and were susceptible to trypsin or other dissociation treatments.

The hen as a model of ovarian cancer.

The domestic laying hen is the only non-human animal that spontaneously develops ovarian cancer with a high prevalence. Hens ovulate prolifically, and this has made the hen intuitively appealing as a model of this disease in light of epidemiological evidence that ovulation rate is highly correlated with the risk of human ovarian cancer. As in women, ovarian cancer in the hen is age-related and it is also grossly and histologically similar to that in humans. In both women and hens, the cancer metastasizes to similar tissues with an accumulation of ascites fluid. Some aggressive ovarian cancers in women arise from cells in the oviduct; this is intriguing because ovarian cancers in the hen express an oviductal protein that is normally absent in the ovary.

Gene expression profiling reveals differentially expressed genes in ovarian cancer of the hen: support for oviductal origin?

Ovarian cancer has a high mortality rate due, in part, to the lack of early detection and incomplete understanding of the origin of the disease. The hen is the only spontaneous model of ovarian cancer and can therefore aid in the identification and testing of early detection strategies and therapeutics. Our aim was to combine the use of the hen animal model and microarray technology to identify differentially expressed genes in ovarian tissue from normal hens compared with hens with ovarian cancer. We found that the transcripts up-regulated in chicken ovarian tumors were enriched for oviduct-related genes. Quantitative real-time PCR and immunohistochemistry confirmed expression of oviduct-related genes in normal oviduct and in ovaries from hens with early- and late-stage ovarian tumors, but not in normal ovarian surface epithelium. In addition, one of the oviduct-related genes identified in our analysis, paired box 2 has been implicated in human ovarian cancer and may serve as a marker of the disease. Furthermore, estrogen receptor 1 mRNA is over-expressed in early-stage tumors, suggesting that expression of the oviduct-related genes may be regulated by estrogen. We have also identified oviduct-related genes that encode secreted proteins that could represent putative serum biomarkers. The expression of oviduct-related genes in early-stage tumors is similar to what is seen in human ovarian cancer, with tumors resembling normal Müllerian epithelium. These data suggest that chicken ovarian tumors may arise from alternative sites, including the oviduct.

Ovarian tumor expression of an oviductal protein in the hen: a model for human serous ovarian adenocarcinoma.

OBJECTIVE: We hypothesized that ovarian tumors without oviductal involvement would not express the oviductal protein ovalbumin, the major protein found in the magnum of the hen's oviduct. METHODS: On the basis of gross visual examination, tissues samples were removed from hens determined to have ovarian tumors and were processed, embedded in paraffin, sectioned, mounted on glass slides, and stained with hematoxylin and eosin. Ovarian tumors and other peritoneal lesions were evaluated histologically. Paraffin sections of ovarian and oviductal tissue were deparaffinized and evaluated for the protein expression of ovalbumin, proliferating cell nuclear antigen (PCNA), and progesterone receptor (PR). RESULTS: Hens with ovarian adenocarcinoma without (n = 10) or with (n = 6) oviductal involvement were positive for ovalbumin in the ovary. Ovary sections from normal hens (n = 9) were negative, and oviductal sections from normal hens (n = 3) were positive for ovalbumin. Expression of PCNA protein was abundant in all ovarian tumors (16 of 16). Oviductal epithelial cells strongly expressed PCNA protein. Expression of PR was observed in 9 of 14 ovarian tumors. CONCLUSION: The presence of ovalbumin in ovarian tumors in the absence of any oviductal involvement suggests that ovarian tumors dedifferentiate during the disease process and thereby resemble serous-type ovarian tumors in women.