Expression Profiling in Ovarian Clear Cell Carcinoma

Abstract

Of all of the epithelial ovarian cancers, clear cell carcinoma (CCC) of the ovary has the worst prognosis. We applied the oligonucleotide array technique to identify genes generally involved in CCC. Of the ∼12,600 genes that were analyzed, 28 were expressed significantly differently between four CCC and seven non-CCC cell lines. Among 16 up-regulated genes in CCC, we further investigated a transcription factor, hepatocyte nuclear factor-1β (HNF-1β). We validated up-regulation of HNF-1β in CCC in terms of both mRNA and protein level using real-time quantitative reverse transcriptase-polymerase chain reaction and immunoblotting. Immunohistochemical analysis of 83 surgically resected ovarian cancers showed that almost all CCC specimens (21 of 22 cases) had nuclear staining for HNF-1β, whereas most non-CCC specimens (60 of 61 cases) showed no immunostaining or only focal and faint staining in the nucleus. Furthermore, we investigated the significance of HNF-1β expression in CCC using RNA interference. The reduction of HNF-1β expression by RNA interference induced apoptotic cell death in ovarian CCC cells, which was confirmed by terminal dUTP nick-end labeling and fluorescence-activated cell-sorting analyses. Our results suggest that HNF-1β is not only an excellent CCC-specific molecular marker but also a molecular target for therapy of ovarian CCC. Of all of the epithelial ovarian cancers, clear cell carcinoma (CCC) of the ovary has the worst prognosis. We applied the oligonucleotide array technique to identify genes generally involved in CCC. Of the ∼12,600 genes that were analyzed, 28 were expressed significantly differently between four CCC and seven non-CCC cell lines. Among 16 up-regulated genes in CCC, we further investigated a transcription factor, hepatocyte nuclear factor-1β (HNF-1β). We validated up-regulation of HNF-1β in CCC in terms of both mRNA and protein level using real-time quantitative reverse transcriptase-polymerase chain reaction and immunoblotting. Immunohistochemical analysis of 83 surgically resected ovarian cancers showed that almost all CCC specimens (21 of 22 cases) had nuclear staining for HNF-1β, whereas most non-CCC specimens (60 of 61 cases) showed no immunostaining or only focal and faint staining in the nucleus. Furthermore, we investigated the significance of HNF-1β expression in CCC using RNA interference. The reduction of HNF-1β expression by RNA interference induced apoptotic cell death in ovarian CCC cells, which was confirmed by terminal dUTP nick-end labeling and fluorescence-activated cell-sorting analyses. Our results suggest that HNF-1β is not only an excellent CCC-specific molecular marker but also a molecular target for therapy of ovarian CCC. Epithelial ovarian cancer has the worst prognosis of all gynecological malignancies.1Scully RE Young RH Clement PB Rosai J Tumors of the Ovary, Maldeveloped Gonads, Fallopian Tube, and Broad Ligament. Armed Forces Institute of Pathology, Washington DC1999: 27-50Google Scholar Since the emergence of platinum-based chemotherapy, the survival rate of patients with epithelial ovarian cancer has improved dramatically.2Piver MS Ovarian carcinoma. A decade of progress.Cancer. 1984; 54: 2706-2715Crossref PubMed Scopus (90) Google Scholar Debulking surgery and adjuvant chemotherapy (such as a combination of paclitaxel and carboplatin) have now gained a position as the standard therapy for epithelial ovarian cancer.3Parmar MKB Adams M Balestrino M Bertelsen K Bonazzi C Calvert H Colombo N Delaloye JF Durando A Guthrie D Hagen B Harper P Mangioni C Perren T Poole C Qian W Rustin G Sandercock J Tumolo S Torri V Vecchione F Tinazzi A Uscinska B Collins S Flann M Buda A Taylor B Tannock I Souhami R Granzia-Valsecchi M Paclitaxel plus carboplatin versus standard chemotherapy with either single-agent carboplatin or cyclophosphamide, doxorubicin, and cisplatin in women with ovarian cancer: the ICON3 randomised trial.Lancet. 2002; 360: 505-515Abstract Full Text Full Text PDF PubMed Scopus (512) Google Scholar However, we still face many problems in the therapy of epithelial ovarian cancer. One of the most difficult is the treatment of clear cell carcinoma (CCC). The incidence of CCC among epithelial ovarian cancers is not high (∼10%), but patients with CCC have a significantly worse prognosis than patients with serous carcinoma.4Tammela J Geisler JP Eskew Jr, PN Geisler HE Clear cell carcinoma of the ovary: poor prognosis compared to serous carcinoma.Eur J Gynaecol Oncol. 1998; 19: 438-440PubMed Google Scholar, 5O'Brien ME Schofield JB Tan S Fryatt I Fisher C Wiltshaw E Clear cell epithelial ovarian cancer (mesonephroid): bad prognosis only in early stages.Gynecol Oncol. 1993; 49: 250-254Abstract Full Text PDF PubMed Scopus (98) Google Scholar One of the reasons why CCC has such a poor prognosis is its low response to standard platinum-based chemotherapy.6Goff BA Sainz de la Cuesta R Muntz HG Fleischhacker D Ek M Rice LW Nikrui N Tamimi HK Cain JM Greer BE Fuller Jr, AF Clear cell carcinoma of the ovary: a distinct histologic type with poor prognosis and resistance to platinum-based chemotherapy in stage III disease.Gynecol Oncol. 1996; 60: 412-417Abstract Full Text PDF PubMed Scopus (317) Google Scholar From a pathogenetic viewpoint, CCC has a number of features distinguishing it from other epithelial ovarian cancers. The percentage of patients with stage I disease is significantly higher in patients with CCC (48.5%) than in those with serous adenocarcinoma (16.6%),7Sugiyama T Kamura T Kigawa J Terakawa N Kikuchi Y Kita T Suzuki M Sato I Taguchi K Clinical characteristics of clear cell carcinoma of the ovary: a distinct histologic type with poor prognosis and resistance to platinum-based chemotherapy.Cancer. 2000; 88: 2584-2589Crossref PubMed Scopus (698) Google Scholar which means the properties of invasion differ between CCC and non-CCCs. The incidence of p53 mutation differs between CCC (0%) and endometrioid adenocarcinoma (63%).8Okuda T Otsuka J Sekizawa A Saito H Makino R Kushima M Farina A Kuwano Y Okai T p53 mutations and overexpression affect prognosis of ovarian endometrioid cancer but not clear cell cancer.Gynecol Oncol. 2003; 88: 318-325Abstract Full Text Full Text PDF PubMed Scopus (85) Google Scholar The loss of heterozygosity pattern differs between CCC and non-CCCs.9Okada S Tsuda H Takarabe T Yoshikawa H Taketani Y Hirohashi S Allelotype analysis of common epithelial ovarian cancers with special reference to comparison between clear cell adenocarcinoma with other histological types.Jpn J Cancer Res. 2002; 93: 798-806Crossref PubMed Scopus (27) Google Scholar Immunohistochemical analysis has revealed that CCC shows trends such as weak expression of both p53 and cyclin A and markedly increased expression of both p21 and cyclin E compared with the other histological subtypes.10Shimizu M Nikaido T Toki T Shiozawa T Fujii S Clear cell carcinoma has an expression pattern of cell cycle regulatory molecules that is unique among ovarian adenocarcinomas.Cancer. 1999; 85: 669-677Crossref PubMed Scopus (84) Google Scholar Glutathione peroxidase 3 (GPX3) is overexpressed in CCC, which may explain the cancer's chemoresistance.11Hough CD Cho KR Zonderman AB Schwartz DR Morin PJ Coordinately up-regulated genes in ovarian cancer.Cancer Res. 2001; 61: 3869-3876PubMed Google Scholar Considering these facts, it is evident that CCC is not just another type of epithelial ovarian cancer but a distinct entity, and there is a need to determine its molecular pathogenesis if we are to improve its prognosis. Until the establishment of the genome-wide expression-analyzing technique such as serial analysis of gene expression12Velculescu VE Zhang L Vogelstein B Kinzler KW Serial analysis of gene expression.Science. 1995; 270: 484-487Crossref PubMed Scopus (3633) Google Scholar or cDNA microarray13Schena M Shalon D Davis RW Brown PO Quantitative monitoring of gene expression patterns with a complementary DNA microarray.Science. 1995; 270: 467-470Crossref PubMed Scopus (7711) Google Scholar, 14Lockhart DJ Dong H Byrne MC Follettie MT Gallo MV Chee MS Mittmann M Wang C Kobayashi M Horton H Brown EL Expression monitoring by hybridization to high-density oligonucleotide arrays.Nat Biotechnol. 1996; 14: 1675-1680Crossref PubMed Scopus (2831) Google Scholar there was only fragmentary knowledge about the molecular pathogenesis of epithelial ovarian cancer. Since then, there have been extensive studies and rapid progress in our understanding of the molecular pathogenesis of various tumors15Golub TR Slonim DK Tamayo P Huard C Gaasenbeek M Mesirov JP Coller H Loh ML Downing JR Caligiuri MA Bloomfield CD Lander ES Molecular classification of cancer: class discovery and class prediction by gene expression monitoring.Science. 1999; 286: 531-537Crossref PubMed Scopus (9355) Google Scholar, 16Higgins JP Shinghal R Gill H Reese JH Terris M Cohen RJ Fero M Pollack JR van de Rijn M Brooks JD Gene expression patterns in renal cell carcinoma assessed by complementary DNA microarray.Am J Pathol. 2003; 162: 925-932Abstract Full Text Full Text PDF PubMed Scopus (223) Google Scholar, 17Varambally S Dhanasekaran SM Zhou M Barrette TR Kumar-Sinha C Sanda MG Ghosh D Pienta KJ Sewalt RG Otte AP Rubin MA Chinnaiyan AM The polycomb group protein EZH2 is involved in progression of prostate cancer.Nature. 2002; 419: 624-629Crossref PubMed Scopus (2270) Google Scholar including epithelial ovarian cancer.18Wang K Gan L Jeffery E Gayle M Gown AM Skelly M Nelson PS Ng WV Schummer M Hood L Mulligan J Monitoring gene expression profile changes in ovarian carcinomas using cDNA microarray.Gene. 1999; 229: 101-108Crossref PubMed Scopus (286) Google Scholar, 19Hough CD Sherman-Baust CA Pizer ES Montz FJ Im DD Rosenshein NB Cho KR Riggins GJ Morin PJ Large-scale serial analysis of gene expression reveals genes differentially expressed in ovarian cancer.Cancer Res. 2000; 60: 6281-6287PubMed Google Scholar, 20Ono K Tanaka T Tsunoda T Kitahara O Kihara C Okamoto A Ochiai K Takagi T Nakamura Y Identification by cDNA microarray of genes involved in ovarian carcinogenesis.Cancer Res. 2000; 60: 5007-5011PubMed Google Scholar, 21Welsh JB Zarrinkar PP Sapinoso LM Kern SG Behling CA Monk BJ Lockhart DJ Burger RA Hampton GM Analysis of gene expression profiles in normal and neoplastic ovarian tissue samples identifies candidate molecular markers of epithelial ovarian cancer.Proc Natl Acad Sci USA. 2001; 98: 1176-1181Crossref PubMed Scopus Google Scholar, DR R G Taylor JM R Y H SM Cho KR Gene expression in ovarian cancer both and distinguishing clear cell from other ovarian Res. 2002; Google Scholar such studies the genome-wide expression analysis have many genes involved in ovarian the and significance of almost all of such genes still we that a between four CCC cell and seven non-CCC cell in terms of molecular We that hepatocyte nuclear factor-1β a transcription factor, is up-regulated both the mRNA and protein level in CCC, other epithelial ovarian cancers, using of ovarian cancer cell and of 83 surgically resected analysis revealed that HNF-1β expression was to CCC, we investigated the significance of HNF-1β expression in CCC using RNA interference a A S and interference by RNA in 1998; PubMed Scopus Google Scholar, JR RW of interference to that and in the 1998; Full Text Full Text PDF PubMed Scopus Google Scholar, SM J W A K T of RNA interference in 2001; PubMed Scopus Google Scholar Ovarian CCC were with interference RNA the HNF-1β dUTP nick-end labeling and fluorescence-activated cell-sorting revealed that the reduction of HNF-1β induced apoptotic cell death in ovarian CCC Our that HNF-1β not only an excellent CCC-specific molecular marker but also a molecular target for the therapy of ovarian CCC. and cell were by and were by was by was from the Cancer and were from the for these cell and the cell are in cell were in with and in a of and Ovarian Cancer in a RNA was from cell using to the and with I The target was from of RNA from using a and the RNA to the of target to oligonucleotide to genes and expressed and of the were as by the analysis was with The hybridization were to for The were further and by From RNA cDNA was with an and quantitative was with a and a the cycle for for for and for of the of expression of in was The for of HNF-1β, and cDNA were as HNF-1β and and and Quantitative was including a as a were by the were with and The was and the was of were by and to the were by for with in of the were with for HNF-1β for and for as a in the The were and for with were using an We investigated a of 83 epithelial ovarian cancers surgically resected the Cancer between and The from to and of the patients had chemotherapy or other treatment features are in of ovarian and of normal ovarian from patients with other gynecological were also The were to the of the and the of and RE of ovarian of of cell stage in a of tissue were with with in in in a for and to for the were in normal in with for HNF-1β as the with and for with as the were for with using a and to the peroxidase reaction using as the and in by nuclear with was by and The results were the of the percentage of for HNF-1β no to to The and were to the significance of the between HNF-1β expression and the HNF-1β which was was from The HNF-1β was to the to to the of the to and SM J W A K T of RNA interference in 2001; PubMed Scopus Google Scholar were from were in a for real-time quantitative and the were with a of of using to the was by and The was using an to the the were with and with a The number of was and by the number of to the percentage of or with and were and with with the were for with of of A and in of of A of were in a and the was using identify genes expressed differently between CCC and we compared the gene of four ovarian CCC cell with those of seven cell of other using oligonucleotide We all with the expressed in the in of of than in of samples in between and with significance P to identify genes expressed differently between and non-CCCs. the we differently expressed genes from Furthermore, we genes that an than a in the between to identify genes with of expression in CCC or non-CCC genes were up-regulated and were in CCC compared with non-CCCs. Of the genes in we a transcription HNF-1β it is a transcription an in the D V L E R and are expressed PubMed Google Scholar but its up-regulation is among the tissue other than W Y T K H T H Expression of and in various histological of Pathol. 1998; PubMed Scopus Google Scholar and the significance of its expression in tissue is still HNF-1β was the most up-regulated transcription factor, we that it to the of CCC by its target we the expression level of HNF-1β mRNA by real-time quantitative The expression of HNF-1β mRNA with as by real-time quantitative were with those by microarray The expression of HNF-1β mRNA in were higher than those in by We the expression of HNF-1β protein using of HNF-1β in a of ovarian cancer cell is in The to HNF-1β protein was in all CCC cell lines. The in and was to to a protein of HNF-1β there are HNF-1β I M or a of the are by RNA 1993; PubMed Scopus Google Scholar was other than these The of the in cell was with the mRNA expression level from quantitative or no for HNF-1β was in of the non-CCC cell for determine HNF-1β is also up-regulated in surgically resected specimens of CCC the protein for analysis we the that we had for analysis all of the showed nuclear staining A to but most showed no immunostaining or only focal and faint staining in the D to of showed nuclear staining for HNF-1β no of normal ovarian which is to the of epithelial ovarian showed nuclear staining for HNF-1β The HNF-1β immunostaining for was significantly higher than that for by The HNF-1β expression patterns or low HNF-1β not with or histological The P for the in expression pattern of HNF-1β between the early stage and the stage was by but we that significance was to the stage of and the higher HNF-1β immunostaining for there was no between the HNF-1β for cases) and that for four cases) by no was between these immunostaining and HNF-1β Expression and in Epithelial Ovarian expression high group 3 to expression low group to Clear stage in a the up-regulated expression of HNF-1β in CCC has we analysis by reduction of HNF-1β expression with real-time quantitative we the of to the level of and HNF-1β mRNA in the and ovarian CCC cell significantly the expression of mRNA to or to of the level in with the but it not the expression of HNF-1β HNF-1β significantly the expression of HNF-1β mRNA to to or to of the level in with the but it not the expression of or HNF-1β not or HNF-1β mRNA in or cell not by using we induced in the and cell lines. using we these also the expression of or HNF-1β protein in the and cell the expression of HNF-1β the expression of HNF-1β protein and the of the HNF-1β protein almost the level in and The was or HNF-1β only the or HNF-1β and to for which or has no not the expression of was that and with HNF-1β from the and in the the a and of the other with as not or these the reduction of HNF-1β expression in and cells, we and analysis showed that or of the with HNF-1β the of The percentage of apoptotic was significantly higher than that in with in or in by shows the of HNF-1β the profile of and by were with or no in the percentage of in the of apoptotic cells, was the the other a in the was in with HNF-1β the of both in and The percentage of the in HNF-1β and or the that the HNF-1β we had for cells, we analysis of serous carcinoma and cells, which have weak or no HNF-1β expression not the protein of cells, that the both in and analysis showed no between the of HNF-1β and that of in both cell lines. that the HNF-1β we not have a and that of HNF-1β mRNA and reduction of HNF-1β protein apoptotic cell death in and with HNF-1β of the of the molecular pathogenesis of ovarian cancer have not been The establishment of genome-wide expression as as studies such K Gan L Jeffery E Gayle M Gown AM Skelly M Nelson PS Ng WV Schummer M Hood L Mulligan J Monitoring gene expression profile changes in ovarian carcinomas using cDNA microarray.Gene. 1999; 229: 101-108Crossref PubMed Scopus (286) Google Scholar, 19Hough CD Sherman-Baust CA Pizer ES Montz FJ Im DD Rosenshein NB Cho KR Riggins GJ Morin PJ Large-scale serial analysis of gene expression reveals genes differentially expressed in ovarian cancer.Cancer Res. 2000; 60: 6281-6287PubMed Google Scholar, 20Ono K Tanaka T Tsunoda T Kitahara O Kihara C Okamoto A Ochiai K Takagi T Nakamura Y Identification by cDNA microarray of genes involved in ovarian carcinogenesis.Cancer Res. 2000; 60: 5007-5011PubMed Google Scholar, 21Welsh JB Zarrinkar PP Sapinoso LM Kern SG Behling CA Monk BJ Lockhart DJ Burger RA Hampton GM Analysis of gene expression profiles in normal and neoplastic ovarian tissue samples identifies candidate molecular markers of epithelial ovarian cancer.Proc Natl Acad Sci USA. 2001; 98: 1176-1181Crossref PubMed Scopus Google Scholar, DR R G Taylor JM R Y H SM Cho KR Gene expression in ovarian cancer both and distinguishing clear cell from other ovarian Res. 2002; Google Scholar have the genes involved in ovarian cancer. Of the genes up-regulated in the CCC cell in HNF-1β and were also up-regulated in surgically resected CCC tissue to a microarray analysis of ovarian cancer tissue DR R G Taylor JM R Y H SM Cho KR Gene expression in ovarian cancer both and distinguishing clear cell from other ovarian Res. 2002; Google Scholar with such as is a in the microarray analysis of tissue whereas changes in cell including the expression the of a cell are in the analysis of cell lines. the between the results from surgically resected tissue and those from the cell that these genes are features of ovarian CCC. the most up-regulated gene in is a overexpression is to poor prognosis and in S Terris DJ B Identification of as a marker for and cell Cancer Res. 2003; Google Scholar, M P R Y Y Wang B carcinomas using gene expression and 2003; PubMed Scopus Google Scholar is up-regulated in epithelial ovarian cancer CCC and compared with normal JH T as a for ovarian 2002; PubMed Scopus Google Scholar but its overexpression in CCC in was that overexpression of to the poor prognosis of CCC. the of and other and its overexpression is with H V R N A of and in response to in J 2002; PubMed Scopus Google Scholar of may for the of CCC which an in in response to DNA induced by such as JB GM S of nuclear and from by and PubMed Scopus Google Scholar of a transcription M H D R a to a the of a and transcription from PubMed Scopus Google Scholar was confirmed by real-time quantitative and analysis and involved in the of CCC protein has been to expressed a high level in A H S A B A P D a and expression in and PubMed Google Scholar and is to of cell and in P The of 1993; PubMed Scopus Google Scholar using surgically resected specimens with we that up-regulated HNF-1β expression the protein level was to ovarian CCC, of stage or histological than (21 of 22 cases) of showed cancer cells, whereas of 61 cases) of showed cancer was that no of or normal ovary showed HNF-1β immunostaining is with S T S Kobayashi H T K Kamura T H Ovarian with ovarian a and Oncol. 2000; Full Text PDF PubMed Scopus Google Scholar and is to of M of ovarian epithelial 2001; PubMed Scopus Google Scholar results of analysis that HNF-1β is a molecular marker for ovarian CCC. in HNF-1β immunostaining an excellent of distinguishing ovarian CCC from other which to now has been difficult with the standard HNF-1β or is a transcription that the or of genes that are expressed in a such as and G S hepatocyte nuclear with a of Natl Acad Sci USA. 85: PubMed Scopus Google Scholar, S M M A for transcription differs between and PubMed Scopus Google Scholar, J T M S is a that transcription and with PubMed Scopus Google Scholar HNF-1β and protein to J T M S is a that transcription and with PubMed Scopus Google are also to of M nuclear a transcription the of 2000; Google Scholar HNF-1β mutations Y N M H Y T K M O H T Y in hepatocyte nuclear gene with PubMed Scopus Google Scholar HNF-1β is overexpressed in W Y T K H T H Expression of and in various histological of Pathol. 1998; PubMed Scopus Google Scholar but its up-regulation has not been in other cancers the significance of HNF-1β expression in CCC, we analysis and to and ovarian CCC cell lines. these cell reduction of HNF-1β expression by apoptotic cell we not to and ovarian CCC cell using such as and We that the why we not gene to these cell was these had or the using to than of and whereas than of and not We not in and cell not the results to we that the reduction of HNF-1β by in and cell that HNF-1β expression was for the survival of CCC Our is by that of cell H H D Wang H C Byrne JH of results in cell and increased to but not to high cell 2002; PubMed Scopus Google Scholar However, it is not clear why reduction of in CCC or HNF-1β may or target genes to cell to the between of and changes by the reduction of HNF-1β are in our our results suggest that HNF-1β a molecular target for therapy of ovarian CCC. RNA as a in E Wang J N N J J P J RNA interference from 2003; PubMed Scopus Google Scholar or it may to C L J Zhang M MT ML L A to genes in cells, and by RNA 2003; PubMed Scopus Google Scholar or transcription H K M Y Y K K to carcinoma the expression of factor, and tissue and also cell and invasion Res. 2000; 60: Google Scholar we have a of genes that are expressed in ovarian CCC. we that the expression of HNF-1β was to CCC and for its The genes in HNF-1β, to that to the distinguishing features of CCC, including its and resistance to standard for ovarian cancer. We for and and for with