Role In Ovarian Carcinogenesis Research Articles

Angiogenesis-related gene profile to predict outcome to multimodal therapy in advanced ovarian carcinoma.

5078 Background: Management of advanced ovarian carcinoma (AOC) involves surgery in order to achieve surgical cytoreduction followed by platinum (C) – paclitaxel (P) chemotherapy. A complete clinical remission is achieved in 40% to 50% of patients. Parameters such as age, extent of residual disease after surgery, and the histopathological subtypes are imperfect predictors of response. Angiogenesis plays a major role in ovarian carcinogenesis and progression. The aim of this study is to build a profile able to predict clinical response to multimodal first-line therapy. Methods: 61 patients with III/IV FIGO stage ovarian cancer who underwent surgical cytoreduction and received a C plus P regimen were included. RNAs were collected from formalin-fixed paraffin-embedded AOC samples. Expression levels of 82 angiogenesis related genes were measured using quantitative real time polymerase chain reaction. Clinical response was evaluated using CT after the completion of multimodal therapy. Statistical analysis was performed using a regression method to generate multiple models based on the significant genes. The accuracy of the models was evaluated using Receiver Operating Characteristic (ROC) curves. The Akaike Information Criterion based selection was used to find the most accurate one. Results: The median age at diagnosis was 53 years (range, 21 to 82 years). All patients had advanced disease (FIGO stages III/IV). Most of them had FIGO stage III (51, 83.6%), grade 3 tumors (35, 57.4%), and serous histology (42, 68.9%). 52 patients (85.2%) achieved an initial response (complete response or partial response by RECIST criteria) to this therapy. We found an independent model able to predict response to therapy comprising 8 genes with an Area Under the Curve (AUC) of 0.955 (p < 0.001). Leave-one-out cross validation was applied to avoid overfitting of the model, obtaining a corrected AUC of 0.880, 95%IC: 0,776-0,985. Conclusions: It is possible to generate a predictive model of clinical response for ovarian cancer based on angiogenesis related genes using formalin-fixed paraffin-embedded samples. Although these results are promising, they should be validated prospectively in larger series of ovarian cancer patients. No significant financial relationships to disclose.

Differential expression and phosphorylation of Pak1 and Pak2 in ovarian cancer: effects on prognosis and cell invasion

Ovarian cancer is a gynecological malignancy with high mortality. Therefore, the identification of novel prognostic and therapeutic targets is important. p21-activated kinases (Paks) are involved in cytoskeleton reorganization. This study investigated the clinical significance of total and phosphorylated (p) Pak1 and Pak2 as well as their functional roles in ovarian cancer. Expressions of Pak1, p-Pak1 Thr(212), Pak2 and p-Pak2 Ser(20) in ovarian normal and cancerous cell lines as well as in clinical samples of ovarian tumors were evaluated. The effects of Pak1 and Pak2 on ovarian cancer cell functions were determined. Pak1, p-Pak1 and p-Pak2 were overexpressed in ovarian cancer cell lines, and clinical samples of ovarian cancers were compared with benign ovarian lesions/inclusion cysts. Similar Pak2 expression levels were observed among normal and cancerous cell lines and clinical samples. After multiple testing correction, high Pak1 and nuclear p-Pak1 expression in ovarian cancers was significantly associated with histological type and tumor grade, respectively. Pak1 and p-Pak1 expression was associated with poor overall and disease-free survival. Pak1 was an independent prognostic factor. Knockdown of Pak1 and Pak2 in ovarian cancer cell lines reduced cell migration and invasion but did not affect cell proliferation and apoptosis. Knockdown of Pak1 also reduced p38 activation and downregulated vascular endothelial growth factor. Conversely, ectopic Pak1 overexpression enhanced ovarian cancer cell migration and invasion in a kinase-dependent manner, along with increased p38 activation. Our findings suggest that Pak1, p-Pak1 and p-Pak2 play important roles in ovarian carcinogenesis. Pak1 and p-Pak1 may be potential prognostic markers and therapeutic molecular targets in ovarian cancer.

Oxidative stress-induced antioxidant enzyme expression is an early phenomenon in ovarian carcinogenesis

Oxidative stress and antioxidant enzymes have been widely investigated in various carcinomas. However, there is only some information about their role in ovarian carcinogenesis or in ovarian carcinomas in vivo. We studied immunohistochemical nuclear and/or cytoplasmic expression of oxidative stress markers 8-hydroxydeoxyguanosine (8-OHdG) and nitrotyrosine, as well as major antioxidative enzymes peroxiredoxins (PRDX) I–VI and thioredoxin (TXN) in ovarian tumours. The material consisted of 20 benign (10 serous, 10 mucinous) and 51 borderline (33 serous, 18 mucinous) epithelial ovarian tumours. The markers of oxidative stress, 8-OHdG and nitrotyrosine, were seen already in benign tumours (in 20% and 45% of the tumours, respectively) and their expression patterns were similar in benign and borderline tumours. The levels of PRDX II, III, IV, V and VI were significantly higher in borderline than in benign tumours ( p < 0.02 for all). Specifically for PRDX II (for both nuclear and cytoplasmic expression, p < 0.00005) and PRDX VI (for cytoplasmic expression, p = 0.0003 and for nuclear expression, p = 0.0005) the difference between benign and borderline tumours was remarkable. In general, serous benign and borderline tumours expressed higher antioxidant enzyme levels than mucinous ones. Nuclear TXN was expressed more strongly in benign than in borderline tumours ( p = 0.003). Oxidative stress occurs already in benign ovarian tumours and the levels are comparable to borderline tumours. However, some of the antioxidant enzymes, especially PRDX II and VI, are more profoundly induced in borderline ovarian tumours, reflecting their possible role as cancer preventers. This difference could also offer a potential tool for differential diagnosis between benign and borderline epithelial ovarian tumours.

Gene Mutations in Animal Models: Do Tumor Suppressor Genes, brca1 and brca2, Play a Role in Ovarian Carcinogenesis?

Ovarian cancer is the most lethal cause of death from gynecological malignancies in the Western world. Over 90% of human ovarian cancers arise in the ovarian surface epithelium (OSE). The OSE surrounding the ovary is simple mesothelium and squamous to flat-cubobidal mesothelial cells. This cell type of ovary has both epithelial and mesenchymal potential. Also OSE cells are regulated by many factors such as cytokines, growth factors, and multiple hormones. Nevertheless OSE function is poorly understood. In particular, ovarian cancers are closely related with hereditary predisposition. Hereditary ovarian tumors are commonly associated with mutations in tumor suppressor genes such as brca1 and brca2 genes. These genes play a role in maintenance of genome integrity, DNA repair, cell cycle control and apoptosis. Mutations in brca1 and/or brca2 may lead to carcinogenesis through distinct molecular pathways like estrogen-mediated proliferation, the presence of a p53 mutation, and the modulation of the activity of NF-kB. Especially the dysfunction of brca1 triggers the inactivation of p53 and a higher proportion of a p53 mutation is commonly linked to brca-linked ovarian tumorigenesis. The dysfunction of brca1 and/or brca2 can arise from multiple mechanisms in the regulation of both JNK and ERK1/2 signaling. For more effective diagnosis and therapy of ovarian cancer, the role of brca1 and/or brca2 in ovarian cancer has to be distinctively elucidated by the animal models in which the gene functions are deleted in mouse OSE cells and by the mechanisms by which these genes affect ovarian carcinogenesis.

Conditional knockout ofbrca1/2andp53in mouse ovarian surface epithelium: Do they play a role in ovarian carcinogenesis?

Alterations of genes are known to be critical for the induction of tumorigenesis, but the mechanism of ovarian carcinogenesis is little understood and remains to be elucidated. In this study, we investigated the roles of brca1, brca2 and p53 genes in the development of ovarian cancer using conditional knockout mice generated by a Cre-loxP recombinant system. Following the application of recombinant adenovirus expressing Cre in vitro, the proliferation of ovarian surface epithelium (OSE) was increased. For instance, a significant increase in cell growth was observed in OSE cells in vitro by conditional knockout isolated from the mice bearing concurrent floxed copies of brca1 and brca2/p53. However, the proliferative effect of the ovarian cells was not observed in concurrent brca1/brca2 or p53 knockout mice in vivo, indicating that we could not observe the direct evidence of the involvement of brca1, brca2, and p53 in ovarian carcinogenesis. Since morphological changes including tumor formation were not observed in mice bearing floxed copies of concurrent brca1/brca2 or p53, the inactivation of brca1/2 or p53 is not sufficient for the induction of tumor formation. Taken together, these results suggest that the deficiency of these genes may not be involved directly in the mechanism of ovarian carcinogenesis.

DNA Adduct 8-Hydroxydeoxyguanosine, a Novel Putative Marker of Prognostic Significance in Ovarian Carcinoma

Previous studies have suggested the importance of reactive oxygen species in all the steps of carcinogenesis. Antioxidant enzymes are considered as the most specific and efficient way to protect cells from reactive oxygen species. The purpose of the current study was to identify the role of oxidative stress and major antioxidant enzymes in ovarian carcinomas. The material consisted of 68 invasive ovarian carcinomas which were studied by immunohistochemistry with antibodies to antioxidant enzymes peroxiredoxins (Prxs) I-VI and thioredoxin and oxidative stress markers nitrotyrosine and 8-hydroxydeoxyguanosine (8-OHdG). Both the intensity and the extent of the stainings were assessed, and the nuclear and cytoplasmic expressions were evaluated separately. The study revealed the hydroxyl radical-derived oxidative stress marker in DNA, 8-OHdG, to be a powerful prognostic factor in ovarian carcinoma (Kaplan-Meier survival log-rank-analysis P = 0.003; risk of death to ovarian carcinoma 2.69; 95% confidence interval 1.35-5.35. 8-OHdG was also associated with poor differentiation (P = 0.053), higher stage (P < 0.001), and non-optimal surgical outcome (P = 0.002). High cytoplasmic Prx IV immunostaining was associated with a better prognosis (P = 0.024), and elevated cytoplasmic expression rates of Prxs V (P = 0.043) and VI (P = 0.032) were associated with a higher stage. To conclude, it appears that hydroxyl radical-derived oxidative stress, but not nitric oxide radical-derived oxidative stress, plays a significant role in ovarian carcinogenesis. Immunohistochemical assessment of 8-OHdG could provide a useful prognostic marker in ovarian cancer.

Polymorphisms in the FGF2 Gene and Risk of Serous Ovarian Cancer: Results From the Ovarian Cancer Association Consortium

Fibroblast growth factor (FGF)-2 (basic) is a potent angiogenic molecule involved in tumor progression, and is one of several growth factors with a central role in ovarian carcinogenesis. We hypothesized that common single nucleotide polymorphisms (SNPs) in the FGF2 gene may alter angiogenic potential and thereby susceptibility to ovarian cancer. We analyzed 25 FGF2 tgSNPs using five independent study populations from the United States and Australia. Analysis was restricted to non-Hispanic White women with serous ovarian carcinoma (1269 cases and 2829 controls). There were no statistically significant associations between any FGF2 SNPs and ovarian cancer risk. There were two nominally statistically significant associations between heterozygosity for two FGF2 SNPs (rs308379 and rs308447; p < .05) and serous ovarian cancer risk in the combined dataset, but rare homozygous estimates did not achieve statistical significance, nor were they consistent with the log additive model of inheritance. Overall genetic variation in FGF2 does not appear to play a role in susceptibility to ovarian cancer.

Oncogenic transformation of human ovarian surface epithelial cells with defined cellular oncogenes

Ovarian surface epithelium (OSE) is considered to give rise to epithelial ovarian carcinomas (EOCs). To elucidate early processes contributing to the development of EOCs from the OSE, two batches of primary human OSE cells were transduced with non-viral human genes (mutant Cdk4, cyclinD1 and hTERT) so as to efficiently establish normal diploid OSE cells without chromosomal instability. Then defined genetic alterations frequently observed in EOCs were transduced into the OSE cells. A combination of p53 inactivation and oncogenic Kras transduction did not confer tumor-forming ability in immunodeficient mice, though additional transduction of Akt or combined transduction of c-myc with bcl-2 did result in tumor formation. In the latter case, tumors demonstrated phenotypes reminiscent of human EOCs, including cytokeratin expression, a highly aggressive phenotype, metastatic behavior and formation of ascites. These results indicate that inactivation of p53 and activation of the Ras pathway play critical roles in ovarian carcinogenesis in co-operation with the Akt or c-myc pathways. This first in vitro model system faithfully recapitulating the development of EOCs using normal human OSE cells should greatly facilitate further studies of EOCs.

Increased expression of Nlp, a potential oncogene in ovarian cancer, and its implication in carcinogenesis

Objective Nlp (Ninein-like protein), a novel centrosome protein involved in microtubule nucleation, has been studied extensively in our laboratory, and its overexpression has been found in some human tumors. To understand the role of Nlp in human ovarian cancer development, we studied the correlation of Nlp expression with clinicopathological parameters and survival in epithelial ovarian cancer, and the impact of Nlp overexpression on ovarian cancer cells. Methods Nlp expression in normal, borderline, benign and malignant epithelial ovarian tissues was examined by immunohistochemistry. The correlation between Nlp expression and tumor grade, FIGO stage and histological type was also evaluated. Survival was calculated using Kaplan–Meier estimates. Cell proliferation and apoptosis were assayed after stable transfection of pEGFP-C3-Nlp or empty vector in human ovarian cancer cell line SKOV3. Results Nlp was positive in 1 of 10 (10%) normal ovarian tissues, 5 of 34 (14.7%) benign tumors, 9 of 26 (34.6%) borderline tumors and 73 of 131 (56.0%) ovarian tumors. Nlp immunoreactivity intensity significantly correlated with tumor grade, but not with FIGO stage or histological type. Kaplan–Meier curves showed that Nlp overexpression was marginally associated with decreased overall survival. Overexpression of Nlp enhanced proliferation and inhibited apoptosis induced by paclitaxel in the SKOV3 cell line. Conclusions Overexpression of Nlp in ovarian tumors raises the possibility that Nlp may play a role in ovarian carcinogenesis.

Expression of estrogen receptor (ER) beta isoforms and Its correlation with DNA methylation in human epithelial ovarian cancer

11063 Background: Ovarian cancer is one of the leading causes of death from gynecological tumors in women. Several lines of evidence suggest that estrogens may play an important role in ovarian carcinogenesis, through binding with their receptors, ERα and ERβ. Two novel estrogen receptor β (ERβ) isoforms that diverge in their 5’-untranslated regions, ERβ mRNA (0K-1) and ERβ mRNA (0N-1), have recently been identified. This finding indicates that transcription of the human ERβ gene occurs from at least two different promoters, i.e., promoter 0K and promoter 0N. The purpose of this study was to study the expression of ERβ isoforms in primary cultures of normal human ovarian surface epithelial cells (OSE), a panel of ovarian cancer cell lines and in normal human ovarian tissues and ovarian cancer tissues; and to examine whether methylation of the two ERβ promoters is involved in regulation of ERβ gene expression or not. Methods: The expression of ERβ isoforms and the methylation status was evaluated in 8 OSE cells, 15 ovarian cancer cell lines, 6 normal, and 57 ovarian cancer tissues. Using quantitative real-time PCR techniques, mRNA expression levels of ERβ1, ERβ2 (ERβcx), ERβ3, ERβ4 and ERβ5 were evaluated. The methylation status of two ERβ promoters were confirmed by direct bisulfate sequencing. Results: ERβ1, ERβ2 and ERβ4 were significantly lower in ovarian cancer cell lines than OSE cells (p<0.01). In contrast, ERβ5 were significantly higher in ovarian cancer cell lines than OSE cells (p<0.01). Promoter 0N was unmethylated in OSE, but was extensively methylated in ovarian cancer cell lines. In contrast, promoter 0K was unmethylated in both normal and malignant ovarian epithelial cells. Correlation was detected between promoter 0N hypermethylation and loss of ERβ mRNA expression in ovarian cancer cell lines. These findings were also detected in ovarian cancer tissues. Conclusions: Decreased level of ERβ mRNA is considered to be associated with ovarian tumorigenesis, and that DNA methylation is an important mechanism for ERβ gene silencing in ovarian cancer. No significant financial relationships to disclose.

Comprehensive Evaluation of ESR2 Variation and Ovarian Cancer Risk

Studies indicate that estrogen receptor beta, encoded by the ESR2 gene on chromosome 14q, may play a role in ovarian carcinogenesis. Using the genetic structure data generated by the Breast and Prostate Cohort Consortium (BPC3), we have comprehensively characterized the role of haplotype diversity in ESR2 and risk of ovarian cancer. Five haplotypes with a frequency of > or =5% were observed in White subjects and five haplotype tagging SNPs (htSNP) were selected to capture the locus diversity with a minimum R(h)(2) of 0.81. The htSNPs were genotyped in 574 White controls, 417 White invasive ovarian cancer cases, and 123 White borderline ovarian cancer cases from case-control studies carried out in Los Angeles County from 1994 through 2004. No statistically significant association was observed between the five htSNPs and related haplotypes and risk of ovarian cancer overall. Haplotype D was associated with a nonstatistically significant increased risk of invasive ovarian cancer overall (odds ratio, 1.38; 95% confidence interval, 0.93-2.02; P = 0.11) relative to the most common haplotype and a statistically significant increased risk of invasive clear cell ovarian cancer (odds ratio, 3.88; 95% confidence interval, 1.28-11.73; P = 0.016). Haplotype D was also reported by the BPC3 to be associated with increased risk of breast cancer. This haplotype warrants further investigation to rule out any effect with invasive ovarian cancer risk.

Expression profile of histone deacetylases 1, 2 and 3 in ovarian cancer tissues

To investigate the expression levels of histone deacetylase (HDAC) 1, 2, and 3 in ovarian cancer tissues and normal ovarian tissues. Randomly assigned each of six patients with serous, mucinous and endometrioid ovarian cancer were included. Another six patients with normal ovarian tissue were included for comparison. RT-PCR was performed to quantify the levels of HDACs1-3 mRNA in the cancer and normal tissues. Western blot analysis was performed to measure the expression levels of HDACs1-3 protein. The HDACs1-3 expression pattern was also topologically examined by immunohistochemistry. Increased mRNA expressions of HDCA1, HDAC 2 and HDAC 3 were detected in 83%, 67% and 83% of 18 cancer tissue samples, compared to normal tissue samples. The relative densities of HDAC1 mRNA and HDAC3 mRNA in the serous, mucinous and endometrioid cancer tissues, and HDAC2 mRNA in serous cancer tissues were significantly higher than those of the normal tissues, respectively (p<0.05). Overexpression of HDAC1, HDAC2 and HDAC3 proteins were detected in 94%, 72% and 83% of 18 cancer samples, respectively. The relative densities of HDAC1 protein and HDAC3 protein in serous, mucinous and endometrioid cancer, and HDAC2 protein in serous and mucinous cancer tissues were significantly higher than those of normal tissues, respectively (p<0.05). Most cancer tissues expressed moderate to strong staining of HDACs1, 2 and 3 in immunohistochemistry. Staining of HDAC2 was weak in only one endometrioid cancer tissue. HDACs1-3 are over expressed in ovarian cancer tissues and probably play a significant role in ovarian carcinogenesis.

MEIS and PBX homeobox proteins in ovarian cancer

Three amino-acid loop extension (TALE) homeobox proteins MEIS and PBX are cofactors for HOX-class homeobox proteins, which control growth and differentiation during embryogenesis and homeostasis. We showed that MEIS and PBX expression are related to cisplatin resistance in ovarian cancer cell lines. Therefore, MEIS1, MEIS2 and PBX expression were investigated immunohistochemically in a tissue microarray ( N = 232) of ovarian cancers and ovarian surface epithelium ( N = 15). Results were related to clinicopathologic characteristics and survival. All cancers expressed MEIS1, MEIS2 and PBX in nucleus and cytoplasm. MEIS1 and 2 only stained nuclear in surface epithelium. Nuclear MEIS2 was negatively related to stage, grade and overall survival in univariate analyses. Additionally, MEIS and PBX RNA expression in ovarian surface epithelium and other normal tissues and ovarian cancer versus other tumour types using public array data sets were studied. In ovarian cancer, MEIS1 is highly expressed compared to other cancer types. In conclusion, MEIS and PBX are extensively expressed in ovarian carcinomas and may play a role in ovarian carcinogenesis.

Hormonal therapy in ovarian cancer

Ovarian carcinoma continues to be the leading cause of death due to gynecological malignancy. Epidemiologic studies indicate that steroid hormones play roles in ovarian carcinogenesis. Gonadotropins, estrogen, and androgen may be causative factors, while gonadotropin-releasing hormone and progesterone may be protective factors in ovarian cancer pathogenesis. Experimental studies have shown that hormonal receptors are expressed in ovarian cancer cells and mediate the growth-stimulatory or growth-inhibitory effects of the hormones on these cells. Hormonal therapeutic agents have been evaluated in several clinical trials. Most of these trials were conducted in patients with recurrent or refractory ovarian cancer, with modest efficacy and few side effects. Better understanding of the mechanisms through which hormones affect cell growth may improve the efficacy of hormonal therapy. Molecular markers that can reliably predict major clinical outcomes should be investigated further in well-designed trials.

The expression of angiopoietin-1, angiopoietin-2, Tie-2 and vascular endothelial growth factor mRNA in normal ovary, benign ovarian cyst, and epithelial ovarian cancer

Objective : Vascular endothelial growth factor (VEGF) increases angiogenesis. It is known that one of angiogenetic factors, angiopoietin-1 (Ang-1) stabilizes maturity of normal vessels and angiopoietin-2 (Ang-2) antagonizes the action of angiopoietin-1 by binding Tie-2 that is their receptor. Recently those factors are known to be involved in carcinogenesis. The aim was to investigate the mRNA expression of those factors in epithelial ovarian cancer including normal ovary and ovarian cyst and to evaluate their role in ovarian carcinogenesis. Methods : The tissue samples of 16 ovarian cyst, and 29 epithelial ovarian cancer patients who were diagnosed at Ewha Woman's University MokDong Hospital from 1997 to 2003, and 18 normal ovary subjects who had hysterectomy and salpingoophorectomy due to uterine myoma and confirmed as normal controls, were obtained. The mRNA expressions of Angiopoietin-1, Angiopoietin-2, Tie-2 and Vascular endothelial growth factor were measured by Quantitative Competitive polymerase chain reaction (PCR). Their expressions were analyzed with ANOVA test and Spearman correlation test. Results : The expressions of Ang-1 mRNA were higher in normal ovary than those of ovarian cyst and ovarian cancer (p

Ovarian Carcinomas: CCN Genes Are Aberrantly Expressed and CCN1 Promotes Proliferation of these Cells

The connective tissue growth factor/cysteine-rich 61/nephroblastoma overexpressed (CCN) family consists of six matricellular proteins that are involved in various cellular functions, such as proliferation, development, and angiogenesis. The purpose of this study was to explore the possibility that CCN genes are involved in ovarian cancers. We quantified CCN expression in a series of 59 ovarian cancers using quantitative real-time reverse transcription-PCR. CCN1 protein levels were further determined by immunohistochemistry and Western blot analysis. Overexpression and inhibition of CCN1 expression by small interfering RNA were used to examine its role in ovarian cancer cell proliferation in vitro and in vivo. We found dysregulation of levels of the various CCN mRNAs in ovarian cancers compared with their expression in normal whole ovaries. Expression of CCN1 protein was detected in normal ovarian epithelial cells and ovarian tumors as well as in ovarian cancer cell lines. Furthermore, estrogen increased CCN1 mRNA and protein levels in ovarian cancer cells. Ectopic expression of CCN1 enhanced the growth of ovarian cancer cells in liquid culture, whereas inhibition of its expression decreased proliferation and increased apoptosis in these cells. The observed changes in cell growth were accompanied with activation of Akt and extracellular signal-regulated kinase (ERK) signaling pathways. Stable expression of CCN1 in SKOV3 cells significantly increased tumorigenicity in nude mice. Finally, overexpression of CCN1 conferred resistant to carboplatin-induced apoptosis in SKOV3 cells. This is the first study to show abnormalities in CCN expression in ovarian carcinomas. Furthermore, our results suggest that CCN1 may play a role in ovarian carcinogenesis by stimulating survival and antiapoptotic signaling pathways.

Methylation and Messenger RNA Expression of p15INK4b but Not p16INK4a Are Independent Risk Factors for Ovarian Cancer

The purpose of this research was to compare methylation status and mRNA expression of p15INK4b and p16INK4a in serous epithelial ovarian cancer tissues and normal ovarian tissues. We analyzed the DNA methylation status and mRNA expression of p15INK4b and p16INK4a in 52 ovarian cancer specimens and 40 normal ovarian specimens by using methylation-specific PCR and real-time reverse transcription-PCR, respectively. Although the p15INK4b and p16INK4a mRNA expression levels were highly correlated with each other (P < 0.001), the methylation status did not seem to be linked with levels of mRNA expression, as no association between the two events was found for either gene. Promoter hypermethylation of p15(INK4b) was more common in ovarian cancer (30.8% for the 52 cases) than in normal ovaries (5% for the 40 controls without ovarian cancer; P = 0.005) but not methylation of p16INK4a (25% for cancer versus 37.5% for normal; P = 0.288). The relative mRNA expression levels of p15INK4b were significantly lower in ovarian cancer (12.9%) than in normal ovaries (41.7%; P = 0.008) but not those of p16INK4a (27% for cases versus 32.8% for controls; P = 0.754). Only high methylation rate and low mRNA expression of p15INK4b were independent risk factors for ovarian cancer (adjusted odds ratio, 5.67; 95% confidence interval, 0.85-37.9 for high methylation rate and odds ratio, 8.98; 95% confidence interval, 1.58-50.9 for low mRNA expression, respectively). Our results suggest that epigenetic alterations in p15INK4b but not p16INK4a have an important role in ovarian carcinogenesis and that mechanisms other than methylation may exist to reduce gene expression of p15INK4b in ovarian cancer.

Estrogen receptor beta, a possible tumor suppressor involved in ovarian carcinogenesis

Ovarian cancer is one of the leading causes of death from gynecological tumors in women. Several lines of evidence suggest that estrogens may play an important role in ovarian carcinogenesis, through their receptors, ERα and ERβ. Interestingly, malignant ovarian tumors originating from epithelial surface constitute about 90% of ovarian cancers and expressed low levels of ERβ, compared to normal tissues. In addition, restoration of ERβ in ovarian cancer cells, leads to strong inhibition of their proliferation and invasion, while apoptosis is enhanced. In this manuscript, recent data suggesting a possible tumor-suppressor role for ERβ in ovarian carcinogenesis are discussed.

Androgen Receptor Gene Methylation and Exon One CAG Repeat Length in Ovarian Cancer: Differences from Breast Cancer

More than one neoplastic founder clone can exist in benign epithelial tumours. Although theories of clonal selection make pluriclonality appear unlikely in carcinomas, published data do not exclude this possibility. This study looked for evidence of multiclonal X inactivation in ovarian carcinoma using AR methylation as a marker. Fifteen unifocal ovarian carcinomas and 14 multifocal carcinomas all in Scottish patients were studied. One representative formalin-fixed paraffin-embedded tumour block was chosen for each of the former and two for the latter. From each of these 43 tumour blocks three samples each of approximately 10(4) carcinoma cells were obtained by microdissection (129 in all). DNA released by proteinase K digestion was subjected to PCR amplification of the androgen receptor gene AR exon I CAG repeat polymorphism with and without prior digestion with methylation-sensitive restriction enzymes HpaII and HhaI. Complex amplification patterns were consistent with mosaic X inactivation in some ovarian carcinomas but acquired anomalies of AR methylation cannot be excluded. Parallel analysis of other X-linked polymorphic loci would strengthen the inference of clonality status from DNA methylation data in tumour X studies. Strikingly, the number of CAG repeats in the 29 ovarian tumour patients (median 16, range 11 - 20) was substantially fewer than in 34 previously studied breast cancer patients from the same scottish population (median 21, range 14 - 26; P < 0.0001), and women homozygous for the AR CAG repeat were over-represented in the ovarian cancer patients but not in the breast cancer series. These findings reinforce recent suggestions that AR may have a role in ovarian carcinogenesis.

Detection of human papillomavirus-16 in ovarian malignancy.

Human papillomavirus is the causal factor for cervical cancer. However, the role of HPV infection in ovarian cancer is unclear. This study aimed to determine the presence of human papillomavirus-16 (HPV-16) in ovarian cancer tissues. Archived human ovarian cancer tissues (N=54 cases, 50 are epithelial cancer, four are nonepithelial cancer) embedded in paraffin blocks were used. Controls are 30 nonmalignant ovarian tissue blocks. In situ hybridisation (ISH) and immunohistochemistry (IHC) were used to detect the presence of HPV-16 and p53 expression. In all, 52 or 36% of the epithelial ovarian tumours detected by ISH or IHC, respectively, were HPV-16 E6 positive. In contrast, only 6.7% of normal ovarian tissues were HPV-16 positive proved by ISH. Human papillomavirus-16 infection was significantly higher in cancer tissues compared to controls with an odds ratio of 16.7 (95% confidence interval [CI]=3.2–71.4, P<0.01). No significant correlation between HPV-16 infection and histological types of cancer was found (P>0.05). p53 gene expression was detected in 42% epithelial ovarian cancers. No correlation between p53 expression and HPV-16 infection was found. The results showed the presence of HPV-16 E6 in ovarian carcinoma, suggesting that HPV infection might play a role in ovarian carcinogenesis.