- Email: [email protected]
Complement in human disease
REVIEW ARTICLE
Redox-Active Iron-Induced Oxidative Stress in the Pathogenesis of Clear Cell Carcinoma of the Ovary Yoshihiko Yamada, MD, PhD, Hiroshi Shigetomi, MD, Akira Onogi, MD, Shoji Haruta, MD, PhD, Ryuji Kawaguchi, MD, PhD, Shozo Yoshida, MD, PhD, Naoto Furukawa, MD, PhD, Akira Nagai, MD, Yasuhito Tanase, MD, PhD, Taihei Tsunemi, MD, Hidekazu Oi, MD, PhD, and Hiroshi Kobayashi, MD, PhD Objective: Epithelial ovarian cancer (EOC) is the most lethal pelvic gynecologic cancer. Clear cell carcinoma (CCC) and endometrioid adenocarcinoma (EAC) of the ovary have been associated with endometriosis, thus indicating that endometriosis has been believed to increase the risk of developing EOC. The aim of our review was to identify and synthesize the most current information on CCC with regard to molecular and pathophysiological distinctions. Method: This article reviews the English-language literature for molecular, pathogenetic, and pathophysiological studies on endometriosis and endometriosis-associated ovarian cancer (EAOC). In this review, we focus on the functions and roles of redox-active iron in CCC carcinogenesis. Results: The iron-induced reactive oxygen species signals can contribute to carcinogenesis via 3 major processes: step 1, by increasing oxidative stress, which promotes DNA mutagenesis, histone modification, chromatin remodeling, and gene products activation/ inactivation thus contributing to EAOC initiation; step 2, by activating detoxification and antiapoptotic pathways via the transcription factor hepatocyte nuclear factor 1A overexpression, thereby contributing to CCC promotion; and step 3, by creating an environment that supports sustained growth, angiogenesis, migration, and invasion of cancer cells via estrogen-dependent (EAC) or estrogen-independent (CCC) mechanisms, thus supporting tumor progression and metastasis. Conclusions: These aspects of reactive oxygen species biology will be discussed in the context of its relationship to EAOC carcinogenesis. Key Words: Ovarian cancer, Clear cell carcinoma, Oxidative stress, Iron, Redox Received October 19, 2010, and in revised form February 22, 2011. Accepted for publication May 4, 2011. (Int J Gynecol Cancer 2011;21: 1200Y1207)
Department of Obstetrics and Gynecology, Nara Medical University, Nara, Japan. Address correspondence and reprint requests to Hiroshi Kobayashi, MD, PhD, Department of Obstetrics and Gynecology, Nara Medical University, 840 Shijo-cho, Kashihara, Nara 634-8522, Japan. E-mail: [email protected]. This study was supported by KAKENHI (Japan Society for the Promotion of Science Grant-in-Aid). No potential conflicts of interest relevant to this article were reported. Copyright * 2011 by IGCS and ESGO ISSN: 1048-891X DOI: 10.1097/IGC.0b013e318222cfdd
1200
ovarian cancer (EOC) is a major health problem Eforpithelial women worldwide. Despite maximum debulking sur-
gery and platinum and taxane-based chemotherapy, EOC is the most lethal pelvic gynecologic cancer, with a 20% to 30% 5-year survival. The high fatality rate of cases results from the frequent diagnosis of EOC at an advanced stage and treatment in a similar fashion, despite evidence of considerable biologic heterogeneity. Epithelial ovarian cancer also displays great histologic heterogeneity depending on the cell type of origin, including serous cyst adenocarcinoma, mucinous cyst adenocarcinoma, endometrioid adenocarcinoma (EAC), and
International Journal of Gynecological Cancer
& Volume 21, Number 7, October 2011
Copyright © 2011 by IGCS and ESGO. Unauthorized reproduction of this article is prohibited.
International Journal of Gynecological Cancer
& Volume 21, Number 7, October 2011
clear cell carcinoma (CCC) of the ovary and others.1 Each EOC subtype is unique histologically, clinically, and molecularly. Serous and EAC tumors almost always presents in advanced stage. In contrast, CCC and mucinous tumors tend to present as tumors limited to the ovaries at presentation. Unlike serous carcinoma, CCC often presents as a large pelvic mass in early stages and thus is diagnosed early. Patients with CCC have a 2.5 times greater risk of venous thromboembolism than women with other histologic diagnoses of EOC. Clear cell carcinoma and EAC have been associated with endometriosis (also know as EAOC), thus indicating that endometriosis has been believed to increase the risk of developing EOC.2 There are notable differences among histologic types, with serous and endometrioid tumors being chemosensitive, whereas mucinous and CCC are more resistant to standard platinum/taxane chemotherapy. Many challenges to the treatment of EOC include the development of biomarkers that can rapidly predict or forecast disease outcome and the in vitro assays to detect chemosensitivity or resistance, the problems of late detection, peritoneal metastasis, drug resistance, and cancer recurrence even after initial response to treatment. The aim of this review was to identify and synthesize the most current information on CCC with regard to molecular, pathogenetic, and pathophysiological distinctions.
MATERIALS AND METHODS The present article reviews the English-language literature for biologic, pathogenetic, and pathophysiological studies on endometriosis and endometriosis-associated ovarian cancer (EAOC). We searched PubMed electronic databases for a 20-year period (1990Y2010), combining the keywords ‘‘genome-wide,’’ ‘‘microarray,’’ ‘‘proteomics,’’ ‘‘inflammation,’’ ‘‘oxidative stress,’’ ‘‘ROS,’’ ‘‘carcinogenesis,’’ ‘‘iron,’’ ‘‘HNF-1A,’’ ‘‘KRAS,’’ ‘‘PTEN,’’ ‘‘MAPK,’’ ‘‘ERK,’’ ‘‘PTP,’’ ‘‘mutation,’’ ‘‘chromatin remodeling,’’ and with ‘‘endometriosis,’’ ‘‘ovarian cancer,’’ ‘‘clear cell carcinoma (CCC),’’ or ‘‘endometriosis-associated ovarian cancer or EAOC.’’ Several recent studies are discussed in the context of pathogenesis of CCC. In addition, references in each article were searched to identify potentially missed studies for a 20-year period.
TUMORIGENESIS IN COMMON SUBTYPES OF EOC Previous studies have shown that EOC is not a single entity but is composed of a diverse group of tumors.3 Morphologic and molecular genetic features could more precisely stratify patients into different risk categories.4 Carcinogenesis in 4 common subtypes of EOC requires several somatic or inherited genetic alterations. Furthermore, epigenetic mechanisms also play an important role in the development and progression of ovarian cancer.5 The epigenetic alterations include DNA methylation, histone modifications and nucleosome remodeling.5 Most ovarian cancers arise because of the accumulation of genetic and epigenetic alterations, but the specific pathways for the development and progression of EOC remain poorly understood with regard to causality and biologic mechanisms.
Oxidative Stress, Ovarian Carcinogenesis
Recent molecular studies support a broad classification of EOC into 2 major types, designated as type 1 and type 2 (Table 1). Endometriosis serves as a precursor of ovarian cancer (EAOC), especially of the EAC and CCC subtypes.6 Chromosomal regions in endometriosis contain genes involved in ovarian tumorigenesis. The fundamental features of human neoplasms (genetic changes, loss of heterozygosity, specific gene mutations including KRAS, PTEN, and PIK3CA [phosphoinositide-3-kinase, catalytic, > polypeptide]) have been evaluated in endometriosis.7 More recent data implicate that ARID1A (AT-rich interactive domain-containing protein 1A) and PPP2R1A (protein phosphatase 2, regulatory subunit A, >) are frequently disrupted in CCC (see later paragraphs).8,9 ARID1A acts as a tumor-suppressor gene, and its mutation and loss of the gene product BAF250a can be seen in the preneoplastic lesions.8,9 It has been strongly speculated that the somatic mutation of ARID1A gene is an early event in the transformation of endometriosis into cancer.8,9 Sequence mutations of ARIA1A and PIK3CA occurred in 57% and 33% of CCC cases, respectively.8Y10 At present, the specific pathways such as ARIA1A, PIK3CA, PTEN, PPP2R1A, ERK (extracellular signalYregulated kinase), KRAS, BRAF, CTNNB1, and TP53 might be involved in the regulation of CCC carcinogenesis.8Y10 The animal model approach was recently used to demonstrate the role of KRAS and PTEN in the initiation of EAC.11 However, little is known about the development and progression of CCC. There have been no spontaneous CCC tumor mouse models available until now. In addition, genetic events that are involved in CCC pathogenesis might include not only point mutations but also the other types of genetic alterations such as gene amplification and deletion. The recent study provides the DNA copy number alterations in ovarian CCC.12 They identified for the first time discrete regions with either DNA copy number gain (Zinc finger protein 217, ZNF217) or loss (cyclin-dependent kinase inhibitor, CDKN2A/2B) in CCC tissues. Overexpression of ZNF217 conferred resistance to paclitaxel, stimulated
TABLE 1. Model of ovarian carcinogenesis Type 1 Histologic diagnosis
Mucinous Endometrioid Clear cell Characteristics Low grade and slow growing Preexisting lesions with adjacent benign- or borderline-like lesions Gene KRAS, BRAF, PTEN, mutations PIK3CA, ERK, CTNNB1, ARID1A, PPP2R1A
* 2011 IGCS and ESGO
Copyright © 2011 by IGCS and ESGO. Unauthorized reproduction of this article is prohibited.
Type 2 Serous
Rapidly progressing, high grade No evidence of precursor lesions P53, p16INK4a/RB
1201
Yamada et al
International Journal of Gynecological Cancer
cell proliferation, and antiapoptosis.13 Interestingly, homozygous deletion of CDKN2A/2B has been recognized as one of the major target genes involved in iron overload-induced carcinogenesis.14
PATHOPHYSIOLOGY OF CCC Oxidative Stress A Redox-Active Iron-Induced Oxidative Stress A considerable body of evidence provides support for a role of oxidative stress in ovarian carcinogenesis.6,15Y20 Markers of oxidative damage, such as strand breakage, DNA adducts, and lipid peroxidation products (eg, DNAmalondialdehyde adducts as an indication of lipid peroxidation; 8-oxo-7,8-dihydro-2¶-deoxyguanosine as an indication of DNA damage), can be detected in ovarian cancer tissue.21 We have previously described the hepatocyte nuclear factor (HNF) 1AYdependent pathophysiology of CCC and discussed its role in oxidative stressYinduced carcinogenesis.17,18,21 Eighty-seven percent of the specific genes upregulated in CCC were associated with the redox-related genes including oxidative and detoxification enzymes.20,21 Kajihara et al20 reported that these redox proteins may serve as an important molecular marker for the diagnosis and molecular understanding of CCC.
& Volume 21, Number 7, October 2011
The high frequency of genomic mutations found in EAOC may result from the presence of participation of mutagenic organics in the induction of carcinoma by endometriotic cyst fluids.15 Recent results also show that, in progressive stages of EOC, the oxidative stress strongly contributes to the uncontrolled tumor expansion. The CCC-specific gene signature contains typical markers of CCC, such as HNF-1A, versican (VCAN), and other genes that reflect oxidative stress.19,20 One of the most common mutagenic factors in CCC might be a redox-active iron19,20 (Fig. 1). A histologically normal ectopic endometrium bears genetic damages caused by iron-dependent oxidative stress. The persistent oxidative stress culminates in endometriotic cell damage and death due to the accumulated DNA aberrations. The specific cells that could avoid a cell death might become cancer.
ERK Regulation by Oxidative Stress There are several mechanisms by which the ERK pathway is activated. An activated ERK1/2 pathway might be critical to tumor growth, differentiation, and survival of EOC with KRAS or BRAF mutations. The ERK activity can also be controlled by not only KRAS or BRAF mutations but also specific protein tyrosine phosphatase (PTP) levels. Mitogenactivated protein kinase phosphatase 3 (MKP3) is a negative regulator of ERK1/222 (Fig. 1). Chan et al22 reported that
FIGURE 1. Redox-active iron-induced ROS signals contributes to carcinogenesis. The sources of the increased inflammation and oxidative stress may derive from the increased burden of repeated bleeding and from the increased amounts of ROS generated by redox-active heme and iron. These iron-induced ROS signals can contribute to carcinogenesis via 3 major processes: step 1, increasing oxidative stress promotes DNA mutagenesis, histone modification, chromatin remodeling, and gene products inactivation (initiation); step 2, activating detoxification and antiapoptotic pathways via the transcription factor HNF-1A overexpression (promotion); and step 3, E2-dependent (EAC) or E2-independent (CCC) mechanisms support further tumor progression. Over time, persistent excess iron exposure may downregulate ER> expression, which may lead to CCC carcinogenesis. Thunder symbol indicates increased susceptibility to ROS.
1202
* 2011 IGCS and ESGO
Copyright © 2011 by IGCS and ESGO. Unauthorized reproduction of this article is prohibited.
International Journal of Gynecological Cancer
& Volume 21, Number 7, October 2011
MKP3 often loses its expression in the protein level in ovarian cancer. The loss of MKP3 protein is associated with ubiquitination/proteosome degradation mediated by reactive oxygen species (ROS). A direct increase of the phospho-ERK level is observed on hydrogen peroxide treatment. These results suggest that the accumulation of ROS may cause the direct phosphorylation of ERK and the degradation of MKP3, which in turn leads to aberrant ERK1/2 activation and contributes to carcinogenesis.22 Because extramitochondrial source of ROS generator, NADPH oxidase (NOX1), is predominantly associated with the ROS generation, the NOX system may be critical for the malignant phenotype and carcinogenesis of some cancer cells. Thus, ROS can act as secondary messengers and control various signaling cascades.23 Oxidative stressYsensitive gene mutation and inactivation of the gene products greatly modulate the protein levels and their activities. Thus, the ERK protein function might be repressed by the mutation of the specific genes such as KRAS and inactivation of MKP3.
KRAS Regulation by Oxidative Stress The RAS-BRAF-MEK (mitogen-activated protein kinase kinase)YERK pathway plays a pivotal role in various cellular responses, including cellular growth, differentiation, and survival. Some of the targeted pathway inhibitors or antagonists of BRAF and its downstream effectors are in stages of clinical development for tailored combination strategies. Interestingly, identified RAS mutations are limited to a small number of sites. These mutations were enriched in amino acids 12, 13, 59, and 61, demonstrating that this reveals nonrandom patterns.24 For example, mutation in the RAS codon 12 results in a gain of function, subsequently activates RAF-MAPK, phosphoinositide 3¶-kinase (PI3K), NF-JB, and RAL-GTPases, thus resulting in accumulation of p16INK4a, p53, and p21 CYP1.24,25 Because human cancers with mutant KRAS coexpress the NF-JB target gene signatures, NF-JB may be another important KRAS downstream mediator. Even in the absence of obvious genetic somatic mutations, the RAS pathway is activated in some cancers.18 First, the RAS oncogene activation is involved in cellular detoxification as well as in oxidative stress. RAS induces constitutive expression of NOX1 through the MAPK pathway.26 Generation of ROS by RAS-induced NOX1 is required for anchorage-independent cell growth, anoikis resistance, augmented angiogenesis, and tumorigenesis.26 NOX1 signaling also causes oxidative inactivation of protein tyrosine phosphatases.26 Protein tyrosine phosphatases decrease growth factorYinduced signaling pathways such as DNA synthesis by inhibiting the tyrosine phosphorylation level of the growth factor receptors, suggesting that aberrant activation of the NOX1 activity benefits transformation phenotypes of a subset of cancer cells. Therefore, the RAS-induced, NOX1-mediated inactivation of protein tyrosine phosphatases serves as a regulatory switch for growth factor receptor activation (Fig. 1). Second, ROS were proposed to be involved in tumor metastasis, which is multifactorial steps, including the epithelial-mesenchymal transition (EMT), invasion of the tumor cells and angiogenesis. Reactive oxygen species activate epidermal growth factor receptor family and other tyrosine
Oxidative Stress, Ovarian Carcinogenesis
kinases, leading to activation of RAS proteins and multiple downstream signaling pathways. The epidermal growth factor also upregulates Snail and downregulates E-cadherin expression through production of H2O2 in human ovarian cancer cells. Furthermore, ROS potentiates cell invasion by upregulating gene expression of hepatocyte growth factor and Wnt/Acatenin as well as by modulating expression of urokinase-type plasminogen activator (Fig. 1). Third, ROS can induce activation of ERK1/2 and protein kinase B (AKT).18 The RAS/RAF/MEK/ERK and RAS/ PI3K/PTEN/AKT pathways interact with each other to regulate cell proliferation and tumorigenesis.18 Finally, ROS can cause cytosine deamination. This DNA modification may be relevant to glyoxal-induced mutations at GC pairs. Uracil in DNA generates promutagenic UG mispairs. The amino group of the G residue on the 3¶ side of the mispaired pyrimidine shows hindered rotation, leading to the specific gene mutation. Therefore, oxidative stress directly and indirectly activates the KRAS signaling cascades.
PTEN Regulation by Oxidative Stress Inactivation of PTEN by loss-of-function mutations and subsequent activation of PI3K-AKT signaling pathways have been detected in many human cancers, including brain, breast, prostate, and ovarian cancers.27 PTEN activity could also be regulated through phosphorylation-dependent modulation of protein stability and/or through reversible oxidation. First, a substantial proportion (approximately 10%) of PTEN proteins was oxidatively inactivated by H2O2 produced in various cell types including macrophages. The oxidation of the active site cysteine residue of this phosphatase abrogates its nucleophilic properties, thus rendering PTEN inactive. Therefore, oxidative stress can directly stimulate PI3Kdependent signaling, not by suppression of PTEN protein expression, but by inhibition of PTEN activity. Second, it has been suggested that DJ-1 (also known as PARK7 [Parkinson disease autosomal recessive early onset 7]) negatively regulates the tumor suppressor PTEN.28 Cysteine106 of DJ-1 is an essential amino acid for DJ-1 to exert its function. Oxidation of cysteine106 of DJ-1 occurred in cells treated with H2O2, accompanied by increased binding of oxidized DJ-1 to PTEN, decreased PTEN activity, and increased phosphorylation of AKT, leading to cell proliferation and transformation (Fig. 1). It functions as a redox-sensitive chaperone, as a sensor for oxidative stress (antioxidative stress reaction), and it apparently protects cancer cells and neurons against oxidative stress and cell death. Third, ROS generated by NOX1 may decrease PTEN activity by activating intermediate redox-sensitive protein kinases, leading to activation of Wnt/A-catenin pathway,29 which has important implications in carcinogenesis in a variety of human cancer (Fig. 1). Dysregulation of the Wnt signaling pathway, caused by loss of the adenomatous polyposis coli (APC) gene or mutational activation of A-catenin is sufficient to give rise to a tumor. Finally, PTEN activity is involved in redox regulation. For example, the loss of PTEN activity shows cellular susceptibility to oxidative stress through antioxidant-responsive
* 2011 IGCS and ESGO
Copyright © 2011 by IGCS and ESGO. Unauthorized reproduction of this article is prohibited.
1203
International Journal of Gynecological Cancer
Yamada et al
element (ARE)Ymediated transcription of detoxification genes. It also determines chromatin modifications leading to ARE activation. These data allow us to consider that PTEN activity is often regulated through oxidation stimulation.
Iron-Induced Carcinogenesis A growing body of evidence has accumulated that redox-active iron-induced oxidative stress is associated with carcinogenesis. First, oxidative stress, due to smoking or various other causes, triggers chronic inflammatory processes and changes in the immune system that is central to the pathogenesis of lung disease, including lung cancer, chronic obstructive lung disease, and atherosclerosis. Cigarette smoking is associated with a high risk for lung cancer. Heme iron intake also increased the risk of lung cancer. Second, patients with inflammatory bowel disease including ulcerative colitis are at increased risk for colorectal cancer, which involves a morphologic progression from inflamed and hyperplastic epithelia to dysplasia and adenocarcinoma. It has been shown that, with dietary iron supplementation, there is increased cancer development in dextran sulfate sodiumYinduced chronic ulcerative colitis. This suggests that inflammation-induced, iron-enhanced oxidative stress is a major player for colon carcinogenesis. Third, metal-induced mutagenicity and carcinogenicity depend on a multifactorial process. Carbon nanotubes might cause an asbestos-like pathologic lesion in the lung. Animal experiments showed that inhalation of carbon nanotubes (with an iron content of 17.7% wt) causes inflammatory response, oxidative stress, and fibrosis as well as mutations of KRAS gene locus in the lung of mice. Finally, iron is rich in the cyst fluids of endometrioma as well as EAOC.19 Iron is a possible cause of CCC carcinogenesis through iron-induced persistent oxidative stress.19 The major toxicity mechanisms include induction of inflammatory response and oxidative stress exacerbated by transition metals including iron. It could generate ROS and donate electrons for the generation of the superoxide radical and participate in the generation of hydroxyl radicals via the Fenton reaction (Fe2+ + H2O2jYFe3+ + OHj + OH). Inflammation provides a redox environment in which transition metals can exhibit their prooxidant potential. A combination of inflammatory response with catalytically competent metal would synergistically enhance damage to cells and DNA. There is an accumulating evidence that ROS can directly damage vital cellular components, including DNA, lipids, and proteins, leading to profound cellular toxicity (Fig. 1). For example, the hydroxyl radical may activate oncogenes or inactivate tumor suppressor genes through point mutations, activate chemical carcinogens, and prevent DNA repair. Among various classes of oxidative DNA damage, the most frequent, abundant, and investigated mutagenic base lesion caused by ROS is 7,8-dihydro-8-oxoguanine (8-oxoG) that can result in adenine mismatches during DNA replication.29 Point mutations generated via oxidative DNA damage might be involved in carcinogenesis and cancer progression. Another example is that acrolein, a major lipid peroxidation product, is chemically reactive and mutagenic.
1204
& Volume 21, Number 7, October 2011
In contrast, to counteract such oxidative damage in nucleic acids, cells are equipped with several defense mechanisms.30 Iron loading also causes up-regulation of antioxidant detoxification genes such as ferritin. Ferritin gene belongs to the ARE-regulated gene family. Ferritin is upregulated at the transcriptional level under oxidative conditions. Over time, the excessive ROS production can exceed the antioxidant defenses. Oxidative stress, caused by the imbalance between the generation and detoxification of ROS, plays an important role in carcinogenesis. Taken together, persistent excess iron overload might be associated with cell toxicity and induce carcinogenesis in humans, and oxidative DNA, protein, and lipid damage has been considered a key factor in carcinogenesis. However, direct causative evidence is lacking, and the molecular mechanisms remain obscure.
Mitochondrial Dysfunction The transcription factor HNF-1A may play a critical role in a cytoprotective effect against forthcoming oxidative stress. We analyzed HNF-1A siRNA-induced gene expression using a microarray approach.20 Several highly up- and downexpressed genes were confirmed by real-time polymerase chain reaction. Expression of OGG1 (8-oxoguanine DNA glycosylase), a repair enzyme, was decreased by treating with HNF-1A siRNA compared with control (our unpublished data, 2010). Removal of oxidized bases is initiated by a DNA glycosylase. A human DNA glycosylase encoded by the OGG1 gene has an activity to remove 8-oxoG from DNA and suppresses the mutagenic effect of 8-oxoG. Although there is no evidence whether OGG1 is a downstream target of HNF-1A, HNF-1A may upregulate OGG1 gene expression to counteract ROSmediated mitochondrial dysfunction. Mammalian cells are equipped with several defense mechanisms to prevent the accumulation of oxidative damage.30 Cells are armed with a vast repertoire of antioxidant defense mechanisms to counteract oxidative damage even in the genome. Thus far, at least 4 distinct repair enzymes (OGG1, MTH1 [7,8-dihydro-8-oxoguanine triphosphatase], MUTYH [mutY homolog], and APEX2 [apurinic/apyrimidinic endonuclease 2]) have been identified to function to repair or prevent oxidized bases, thereby suppressing carcinogenesis and cell death.30 Knockout of these repair enzymes in mice has been shown to produce a phenotype of higher cancer incidence. OGG1 knockout mice exhibit elevated levels of 8-oxoG in nuclear and mitochondrial DNA. The increased incidence of lung cancer, UV-induced skin cancer, or colon cancer has been reported in OGG1 knockout mice. Codon 12 of KRAS seems to be an important downstream target of oxidative DNA damage, resulting from OGG1 deficiency. The OGG1 Cysteine326 variant also seems to play a role in conferring increased risk of ovarian cancer. MTH1 protein hydrolyzes oxidized purine nucleoside triphosphates, such as 8-oxo-2¶-deoxyguanosine triphosphate and 2-hydroxy-2¶-deoxyadenosine triphosphate (2-OHdATP), to the corresponding nonmutagenic monophosphates.29 Increased susceptibility to carcinogenesis is observed in * 2011 IGCS and ESGO
Copyright © 2011 by IGCS and ESGO. Unauthorized reproduction of this article is prohibited.
International Journal of Gynecological Cancer
& Volume 21, Number 7, October 2011
Oxidative Stress, Ovarian Carcinogenesis
MTH1-null mice.29 Tumors were formed in the lungs, livers, and stomachs of MTH1-deficient mice. Germ-line mutations in MUTYH have been associated with recessive inheritance of multiple colorectal tumors. MUTYH mutation-associated polyposis (MAP) is characterized by a lifetime risk of colorectal cancer of up to 100%. A significant increase of MUTYH mutations has been observed in the incidence of ovarian cancers. It is unlikely, however, that MTH1 and APEX mutations are directly associated with ovarian cancer. There is no study whether these genes are modulated under the oxidative stress conditions in CCC.
HNF-1AYdependent pathway provides new insights into the regulation of glycogen synthesis and accumulation, possibly through up-regulation of glucose transporter type 2, glucose6-phosphatase, and dipeptidyl-peptidase 4.18 Although there is no relation between CCC and glycogen storage disease (GSD), the high risk for hepatocellular carcinoma in GSD I, III, VI, and IX has become evident, suggesting that excess glycogen accumulation is at risk for developing cancer. Excess glycogen accumulation may be associated with CCC carcinogenesis.
Estrogen Receptor
This review finally describes the evidence for the role of oxidative stress in the pathophysiology of CCC and discusses the oxidative stressYmediated chromatin remodeling. The pathophysiology of endometriosis and EAOC is characterized by persistent/chronic inflammation and oxidant/ antioxidant imbalance as described. The sources of the increased inflammation and oxidative stress may derive from the increased burden of repeated bleeding and from the increased amounts of ROS generated by redox-active heme and iron. Upregulation of iron-induced ROS expression is reflected by increased oxidative stress marker, 8-oxoG. The elevation in the levels of 8-oxoG is also associated with the disruption of chromatin remodeling via modulation of histone deacetylation, allowing access for transcription factor DNA binding33 (Fig. 1). These results clearly highlight the importance of oxidative stress in the induction of DNA damage and disruption of chromatin remodeling. In general, oxidative stressYdependent increased DNA damage results in global hypomethylation due to interference with the ability of DNA to function as a substrate for the DNA methyl transferase.34 DNA methylation also plays an important role in the regulation of gene expression, chromatin architecture, chromatin-associated proteins, and histone modification. The cytosine is the preferred base for CpG island methylation, whereas the guanine is the site for oxidative damage by causing profound DNA adducts. Current understanding of the roles of ROS in cellular functions as upregulation of proinflammatory responses, signaling pathways, activation of transcription factors, specific gene methylation or mutation, chromatin remodeling, and gene product expression will provide important information regarding pathologic processes contributing to CCC. Recent studies showed that the frequently mutated genes in CCC tumors are ARID1A, PIK3CA, PPP2R1A, and KRAS.8,9 The 2 mutated PPP2R1A and ARID1A genes were novel.8,9 PPP2R1A encodes a regulatory subunit of serine/ threonine phosphatase.9 The encoded protein is implicated in the negative control of cell growth and division. PPP2R1A has been shown to function as a tumor suppressor gene in breast cancer.35 In addition, ARID1A encodes AT-rich interactive domain-containing protein 1A, which participates in chromatin remodeling.8,9 The ARID1 chromatin-remodeling complex facilitates DNA access by the transcription machinery.36 Li et al36 recently reported that ARID1 is associated with an E3 ubiquitin ligase. The typical E3 ubiquitin ligases include not only ARID1 but also von HippelYLindau
Clear cell carcinomas and EACs are subtypes of EOC associated to endometriosis. In contrast to EAC, CCCs exhibit a significant decrease of cell-associated expression of estrogen receptor (ER) and progesterone receptor (PR). Endometrioid adenocarcinoma showed 100% and 92% immunostaining for ER> and PR, respectively.31 Conversely, 0% and 10% of CCC showed positive nuclear staining for ER> and PR, respectively.31 In addition to Fe2+-dependent free radical damage, 17A-estradiol (E2) also induces mitochondrial ROS production and subsequent free radical damage to DNA. Toxicological concentrations of E2 producing high concentrations of ROS may be directly toxic to the cells by activating apoptosis. Reactive oxygen species can also induce activation of ERK1/2 and AKT. Both kinases have been implicated in the phosphorylation of serine 118 and serine 167 on ER>, respectively. These phosphorylations lead to down-regulation of ER>. Over time, persistent ROS exposure to endometriosis downregulates ER> expression, which may lead to CCC carcinogenesis (Fig. 1). These data allow us to speculate that the more excess of free radicals in CCC rather than EAC will attack cellular macromolecules including DNA leading to carcinogenesis. The CCC and EAC subtypes exhibit a unique molecular profiles: HNF-1A+/detox+/ERj for CCC and HNF1Aj/detoxj/ER+ for EAC, respectively. These results suggest that the HNF-1A and ER> have different effects on pathogenesis of CCC and EAC. Epigenetic events, such as deacetylation and methylation of histones, might be involved in the regulation of promoter transcription. In a fraction of breast cancers, the loss of ER> protein expression is a result of the hypermethylation of the ER> promoter. Unfortunately, the exact interplay of these factors in the oxidative stressYdependent transcriptional repression activity is not yet well understood in EAOC.
Glycogen Accumulation Tumor-specific alterations in glucose and glycogen metabolism have been investigated for decades. Hypoxic microenvironments in tumors alter cancer cell metabolism. Cancer cells rely on anaerobic glycolysis rather than on respiration for ATP generation, a phenomenon known as the Warburg effect.32 Microscopic examination revealed abundant cytoplasmic glycogen content in CCC as demonstrated by PAS staining. Previous microarray studies have sought differences in glucose and glycogen metabolism in CCC and other histotypes.18 The
Chromatin Remodeling
* 2011 IGCS and ESGO
Copyright © 2011 by IGCS and ESGO. Unauthorized reproduction of this article is prohibited.
1205
International Journal of Gynecological Cancer
Yamada et al
(VHL) complexes, which serve as scaffolds. Alteration of the VHL gene by mutation, loss of heterozygosity, and promoter methylation has been found to be important to renal clear cell carcinoma (RCC) pathogenesis. When the VHL gene is mutated or lost, hypoxia-inducible factor 1> accumulates, leading to the increased expression of hypoxia-related genes, which are potentially important in RCC carcinogenesis and progression. The protein complex containing ARID1 protein is assembled in a manner similar to that for the well-characterized VHL complex.36 VHL recognizes and subsequently ubiquitinates hypoxia-inducible factor 1>, whereas ARID1 targets histone H2B.36 Posttranslational modifications of histones, such as H2B ubiquitination, might show a positive correlation with transcription activation. It has also been reported that, although there are no precise data on iron, heavy metals such as nickel cleaved off the histone H2B possibly due to oxidative stress. Although the molecular mechanism by which ARID1Adependent carcinogenicity is exerted is not fully understood, this gene product may regulate transcription of certain genes by altering the chromatin structure around those genes. These data allow us to speculate that the ARID1A gene product is part of an ubiquitin ligase complex that targets H2B for ubiquitination and proteasomal degradation, linking HOXA9 response genes to CCC oncogenesis. Taken together, these results suggest that oxidative stressYinduced aberrant chromatin remodeling including ARID1A contributes to the pathogenesis of CCC.9
CONCLUSIONS The present review integrates basic research observations in an attempt to provide a comprehensive understanding of how oxidative stress processes contribute to ovarian cancer (EAOC) development in endometriosis patients. This review also sheds light on future research on oxidative stress cascades. Cancer is a complex progressive multistep disease that involves the accumulation of both genetic and epigenetic abnormalities. The most common genetic alterations include chromosomal structure changes such as loss of heterozygosity, deletions, gains, amplification, and gene mutations such as base substitutions. The important epigenetic alterations are DNA methylation, histone modification, and chromatin remodeling. One of the most common mutagenic factors in endometriosis might be a redox-active iron. Repeated hemolysis occurring during the development of endometriosis results in high levels of free heme and iron.16 In patients with endometriosis, persistent iron overload has been demonstrated in the components of the endometrioma (endometriotic cyst fluid) and the peritoneal cavity (peritoneal fluid, endometriotic lesions, peritoneum, and macrophages). The genome-wide expression analysis demonstrated that the genes upregulated in endometriosis are involved in oxidative stress.16 Several important endometriosis-specific genes overlap with those known to be regulated by iron.16 The increased iron-induced oxidative stress eventually culminate in endometriotic cell damage and finally death due to the accumulated DNA aberrations. Conversely, HNF-1A overexpressed in endometriotic
1206
& Volume 21, Number 7, October 2011
epithelial cells acts as a potent antioxidant and detoxification and antiapoptotic agent to continue to grow. Persistent iron exposure can result in toxicity and is also associated with carcinogenesis. Not only gene mutations but also microenvironmental changes play an important role in EAOC carcinogenesis. The pathways such as ARID1A, PPP2R1A, MAPK, KRAS, PI3K, and PTEN are involved in the regulation of EAOC carcinogenesis8,9 (Fig. 1). The distribution pattern of base substitutions in somatic mutations might be nonrandom distribution, which generates mutation hotspots. First, most KRAS mutations identified have been G-to-T transversions at codon 12, suggesting that mutagens including ROS cause specific DNA adduct formation and contribute to carcinogenesis. Second, oxidative stress might directly stimulate PI3Kdependent signaling by suppression of PTEN expression or inhibition of PTEN activity. Oxidative stressYsensitive gene mutation and inactivation of the gene products greatly modulate the protein levels and its activities. Third, PPP2R1A acts as a tumor suppressor gene. This somatic mutation confers a risk of developing CCC. Fourth, ARID1A participates in aberrant chromatin remodeling. This mutation may be more susceptible to ROS-induced CCC. Finally, these redox-active iron-induced ROS signals can contribute to carcinogenesis via 3 major processes: step 1, by increasing oxidative stress, which promotes DNA mutagenesis, histone modification, chromatin remodeling, and gene products inactivation, thus contributing to EAOC initiation; step 2, by activating detoxification and antiapoptotic pathways via the up-regulation of the transcription factor HNF-1A expression, thereby contributing to CCC promotion; and step 3, by creating an environment that supports sustained growth, angiogenesis, migration, and invasion of cancer cells via E2-dependent (EAC) or E2independent (CCC) mechanisms, thus supporting tumor progression and metastasis. Step 1 of these changes seems to be common in CCC and EAC. These mutational and environmental analyses provide a window into the genetic landscape of EAOC. The great effort has made to establish a molecularly based carcinogenesis mechanism. HNF-1A has been identified as one of the most highly overexpressed genes in CCC but not EAC. Another candidate factor, which is associated with an initial switch to CCC carcinogenesis, might be the redox-active iron. The following issues must still be clarified: What triggers the susceptibility of specific genes to ROS? What causes activation or silencing of these specific gene? What essentially differentiates CCC from EAC? These tumor-specific aberrations provide clues to the cellular processes underlying carcinogenesis and have proven useful for diagnostic and therapeutic purposes. These aspects of ROS biology will be discussed in the context of its relationship to CCC carcinogenesis and therapeutic strategies. Understanding of the mechanisms of redox signaling leads to the development of novel therapies based on the pharmacological manipulation of antioxidants in endometriosis to prevent EAOC development. Antioxidants, for example, thiols or molecules that have dual antioxidant and anti-inflammatory activity, could be potential therapeutic agents and help prevent carcinogenesis. The proof of concept will come from clinical studies on the effectiveness of antioxidant therapy. * 2011 IGCS and ESGO
Copyright © 2011 by IGCS and ESGO. Unauthorized reproduction of this article is prohibited.
International Journal of Gynecological Cancer
& Volume 21, Number 7, October 2011
ACKNOWLEDGMENTS The authors thank all the study participants for their time and efforts. The authors thank Mikiko Kita for editorial assistance.
REFERENCES 1. Gubbels JA, Claussen N, Kapur AK, et al. The detection, treatment, and biology of epithelial ovarian cancer. J Ovarian Res. 2010;3:8. 2. Kobayashi H, Sumimoto K, Moniwa N, et al. Risk of developing ovarian cancer among women with ovarian endometrioma: a cohort study in Shizuoka, Japan. Int J Gynecol Cancer. 2007;17:37Y43. 3. Kurman RJ, Shih IeM. The origin and pathogenesis of epithelial ovarian cancer: a proposed unifying theory. Am J Surg Pathol. 2010;34:433Y443. 4. Shih IeM, Kurman RJ. Ovarian tumorigenesis: a proposed model based on morphological and molecular genetic analysis. Am J Pathol. 2004;164:1511Y1518. 5. Maldonado L, Hoque MO. Epigenomics and ovarian carcinoma. Biomark Med. 2010;4:543Y570. 6. Mandai M, Yamaguchi K, Matsumura N, et al. Ovarian cancer in endometriosis: molecular biology, pathology, and clinical management. Int J Clin Oncol. 2009;14:383Y391. 7. Vigano´ P, Somigliana E, Chiodo I, et al. Molecular mechanisms and biological plausibility underlying the malignant transformation of endometriosis: a critical analysis. Hum Reprod Update. 2006;12:77Y89. 8. Wiegand KC, Shah SP, Al-Agha OM, et al. ARID1A mutations in endometriosis-associated ovarian carcinomas. N Engl J Med. 2010;363:1532Y1543. 9. Jones S, Wang TL, Shih IM, et al. Frequent mutations of chromatin remodeling gene ARID1A in ovarian clear cell carcinoma. Science. 2010;330:228Y231. 10. Kuo KT, Mao TL, Jones S, et al. Frequent activating mutations of PIK3CA in ovarian clear cell carcinoma. Am J Pathol. 2009;174:1597Y1601. 11. Dinulescu DM, Ince TA, Quade BJ, et al. Role of K-ras and Pten in the development of mouse models of endometriosis and endometrioid ovarian cancer. Nat Med. 2005;11:63Y70. 12. Kuo KT, Mao TL, Chen X, et al. DNA copy numbers profiles in affinity-purified ovarian clear cell carcinoma. Clin Cancer Res. 2010;16:1997Y2008. 13. Thollet A, Vendrell JA, Payen L, et al. ZNF217 confers resistance to the pro-apoptotic signals of paclitaxel and aberrant expression of Aurora-A in breast cancer cells. Mol Cancer. 2010;9:291. 14. Toyokuni S. Mysterious link between iron overload and CDKN2A/2B. J Clin Biochem Nutr. 2011;48:46Y49. 15. Yamaguchi K, Mandai M, Toyokuni S, et al. Contents of endometriotic cysts, especially the high concentration of free iron, are a possible cause of carcinogenesis in the cysts through the iron-induced persistent oxidative stress. Clin Cancer Res. 2008;14:32Y40. 16. Kobayashi H, Yamada Y, Kanayama S, et al. The role of iron in the pathogenesis of endometriosis. Gynecol Endocrinol. 2009;25:39Y52. 17. Kobayashi H, Kajiwara H, Kanayama S, et al. Molecular pathogenesis of endometriosis-associated clear cell carcinoma of the ovary. Oncol Rep. 2009;22:233Y240. 18. Kobayashi H, Yamada Y, Kanayama S, et al. The role of hepatocyte nuclear factor-1A in the pathogenesis of clear cell carcinoma of the ovary. Int J Gynecol Cancer. 2009;19:471Y479.
Oxidative Stress, Ovarian Carcinogenesis
19. Yamaguchi K, Mandai M, Oura T, et al. Identification of an ovarian clear cell carcinoma gene signature that reflects inherent disease biology and the carcinogenic processes. Oncogene. 2010;29:1741Y1752. 20. Kajihara H, Yamada Y, Kanayama S, et al. Clear cell carcinoma of the ovary: potential pathogenic mechanisms. Oncol Rep. 2010;23:1193Y1203. 21. Sanchez M, Torres JV, Tormos C, et al. Impairment of antioxidant enzymes, lipid peroxidation and 8-oxo-2¶-deoxyguanosine in advanced epithelial ovarian carcinoma of a Spanish community. Cancer Lett. 2006;233:28Y35. 22. Chan DW, Liu VW, Tsao GS, et al. Loss of MKP3 mediated by oxidative stress enhances tumorigenicity and chemoresistance of ovarian cancer cells. Carcinogenesis. 2008;29:1742Y1750. 23. Kumar B, Koul S, Khandrika L, et al. Oxidative stress is inherent in prostate cancer cells and is required for aggressive phenotype. Cancer Res. 2008;68:1777Y1785. 24. Adjei AA. Blocking oncogenic Ras signaling for cancer therapy. J Natl Cancer Inst. 2001;93:1062Y1074. 25. Serrano M, Lin AW, McCurrach ME, et al. Oncogenic ras provokes premature cell senescence associated with accumulation of p53 and p16INK4a. Cell. 1997;88:593Y602. 26. Shinohara M, Shang WH, Kubodera M, et al. Nox1 redox signaling mediates oncogenic Ras-induced disruption of stress fibers and focal adhesions by down-regulating Rho. J Biol Chem. 2007;282:17640Y17648. 27. Wu X, Senechal K, Neshat MS, et al. The PTEN/MMAC1 tumor suppressor phosphatase functions as a negative regulator of the phosphoinositide 3-kinase/Akt pathway. Proc Natl Acad Sci U S A. 1998;95:15587Y15591. 28. Nagakubo D, Taira T, Kitaura H, et al. DJ-1, a novel oncogene which transforms mouse NIH3T3 cells in cooperation with ras. Biochem Biophys Res Commun. 1997;231:509Y513. 29. Nakabeppu Y, Sakumi K, Sakamoto K, et al. Mutagenesis and carcinogenesis caused by the oxidation of nucleic acids. Biol Chem. 2006;387:373Y379. 30. Nakabeppu Y, Tsuchimoto D, Ichinoe A, et al. Biological significance of the defense mechanisms against oxidative damage in nucleic acids caused by reactive oxygen species: from mitochondria to nuclei. Ann N Y Acad Sci. 2004;1011:101Y111. 31. Fujimura M, Hidaka T, Kataoka K, et al. Absence of estrogen receptor-> expression in human ovarian clear cell adenocarcinoma compared with ovarian serous, endometrioid, and mucinous adenocarcinoma. Am J Surg Pathol. 2001;25:667Y672. 32. Jelluma N, Yang X, Stokoe D, et al. Glucose withdrawal induces oxidative stress followed by apoptosis in glioblastoma cells but not in normal human astrocytes. Mol Cancer Res. 2006;4:319Y330. 33. Rahman I, Gilmour PS, Jimenez LA, et al. Oxidative stress induces histone acetylation in alveolar epithelial cells (A549). Am J Respir Crit Care Med. 2001;163:A61. 34. Donkena KV, Young CY, Tindall DJ. Oxidative stress and DNA methylation in prostate cancer. Obstet Gynecol Int. 2010;2010:302051. 35. Dupont WD, Breyer JP, Bradley KM, et al. Protein phosphatase 2A subunit gene haplotypes and proliferative breast disease modify breast cancer risk. Cancer. 2010;109. 36. Li XS, Trojer P, Matsumura T, et al. Mammalian SWI/SNFVa subunit BAF250/ARID1 is an E3 ubiquitin ligase that targets histone H2B. Mol Cell Biol. 2010;30:1673Y1688.
* 2011 IGCS and ESGO
Copyright © 2011 by IGCS and ESGO. Unauthorized reproduction of this article is prohibited.
1207
Redox-Active Iron-Induced Oxidative Stress in the Pathogenesis of Clear Cell Carcinoma of the Ovary Yoshihiko Yamada, MD, PhD, Hiroshi Shigetomi, MD, Akira Onogi, MD, Shoji Haruta, MD, PhD, Ryuji Kawaguchi, MD, PhD, Shozo Yoshida, MD, PhD, Naoto Furukawa, MD, PhD, Akira Nagai, MD, Yasuhito Tanase, MD, PhD, Taihei Tsunemi, MD, Hidekazu Oi, MD, PhD, and Hiroshi Kobayashi, MD, PhD Objective: Epithelial ovarian cancer (EOC) is the most lethal pelvic gynecologic cancer. Clear cell carcinoma (CCC) and endometrioid adenocarcinoma (EAC) of the ovary have been associated with endometriosis, thus indicating that endometriosis has been believed to increase the risk of developing EOC. The aim of our review was to identify and synthesize the most current information on CCC with regard to molecular and pathophysiological distinctions. Method: This article reviews the English-language literature for molecular, pathogenetic, and pathophysiological studies on endometriosis and endometriosis-associated ovarian cancer (EAOC). In this review, we focus on the functions and roles of redox-active iron in CCC carcinogenesis. Results: The iron-induced reactive oxygen species signals can contribute to carcinogenesis via 3 major processes: step 1, by increasing oxidative stress, which promotes DNA mutagenesis, histone modification, chromatin remodeling, and gene products activation/ inactivation thus contributing to EAOC initiation; step 2, by activating detoxification and antiapoptotic pathways via the transcription factor hepatocyte nuclear factor 1A overexpression, thereby contributing to CCC promotion; and step 3, by creating an environment that supports sustained growth, angiogenesis, migration, and invasion of cancer cells via estrogen-dependent (EAC) or estrogen-independent (CCC) mechanisms, thus supporting tumor progression and metastasis. Conclusions: These aspects of reactive oxygen species biology will be discussed in the context of its relationship to EAOC carcinogenesis. Key Words: Ovarian cancer, Clear cell carcinoma, Oxidative stress, Iron, Redox Received October 19, 2010, and in revised form February 22, 2011. Accepted for publication May 4, 2011. (Int J Gynecol Cancer 2011;21: 1200Y1207)
Department of Obstetrics and Gynecology, Nara Medical University, Nara, Japan. Address correspondence and reprint requests to Hiroshi Kobayashi, MD, PhD, Department of Obstetrics and Gynecology, Nara Medical University, 840 Shijo-cho, Kashihara, Nara 634-8522, Japan. E-mail: [email protected]. This study was supported by KAKENHI (Japan Society for the Promotion of Science Grant-in-Aid). No potential conflicts of interest relevant to this article were reported. Copyright * 2011 by IGCS and ESGO ISSN: 1048-891X DOI: 10.1097/IGC.0b013e318222cfdd
1200
ovarian cancer (EOC) is a major health problem Eforpithelial women worldwide. Despite maximum debulking sur-
gery and platinum and taxane-based chemotherapy, EOC is the most lethal pelvic gynecologic cancer, with a 20% to 30% 5-year survival. The high fatality rate of cases results from the frequent diagnosis of EOC at an advanced stage and treatment in a similar fashion, despite evidence of considerable biologic heterogeneity. Epithelial ovarian cancer also displays great histologic heterogeneity depending on the cell type of origin, including serous cyst adenocarcinoma, mucinous cyst adenocarcinoma, endometrioid adenocarcinoma (EAC), and
International Journal of Gynecological Cancer
& Volume 21, Number 7, October 2011
Copyright © 2011 by IGCS and ESGO. Unauthorized reproduction of this article is prohibited.
International Journal of Gynecological Cancer
& Volume 21, Number 7, October 2011
clear cell carcinoma (CCC) of the ovary and others.1 Each EOC subtype is unique histologically, clinically, and molecularly. Serous and EAC tumors almost always presents in advanced stage. In contrast, CCC and mucinous tumors tend to present as tumors limited to the ovaries at presentation. Unlike serous carcinoma, CCC often presents as a large pelvic mass in early stages and thus is diagnosed early. Patients with CCC have a 2.5 times greater risk of venous thromboembolism than women with other histologic diagnoses of EOC. Clear cell carcinoma and EAC have been associated with endometriosis (also know as EAOC), thus indicating that endometriosis has been believed to increase the risk of developing EOC.2 There are notable differences among histologic types, with serous and endometrioid tumors being chemosensitive, whereas mucinous and CCC are more resistant to standard platinum/taxane chemotherapy. Many challenges to the treatment of EOC include the development of biomarkers that can rapidly predict or forecast disease outcome and the in vitro assays to detect chemosensitivity or resistance, the problems of late detection, peritoneal metastasis, drug resistance, and cancer recurrence even after initial response to treatment. The aim of this review was to identify and synthesize the most current information on CCC with regard to molecular, pathogenetic, and pathophysiological distinctions.
MATERIALS AND METHODS The present article reviews the English-language literature for biologic, pathogenetic, and pathophysiological studies on endometriosis and endometriosis-associated ovarian cancer (EAOC). We searched PubMed electronic databases for a 20-year period (1990Y2010), combining the keywords ‘‘genome-wide,’’ ‘‘microarray,’’ ‘‘proteomics,’’ ‘‘inflammation,’’ ‘‘oxidative stress,’’ ‘‘ROS,’’ ‘‘carcinogenesis,’’ ‘‘iron,’’ ‘‘HNF-1A,’’ ‘‘KRAS,’’ ‘‘PTEN,’’ ‘‘MAPK,’’ ‘‘ERK,’’ ‘‘PTP,’’ ‘‘mutation,’’ ‘‘chromatin remodeling,’’ and with ‘‘endometriosis,’’ ‘‘ovarian cancer,’’ ‘‘clear cell carcinoma (CCC),’’ or ‘‘endometriosis-associated ovarian cancer or EAOC.’’ Several recent studies are discussed in the context of pathogenesis of CCC. In addition, references in each article were searched to identify potentially missed studies for a 20-year period.
TUMORIGENESIS IN COMMON SUBTYPES OF EOC Previous studies have shown that EOC is not a single entity but is composed of a diverse group of tumors.3 Morphologic and molecular genetic features could more precisely stratify patients into different risk categories.4 Carcinogenesis in 4 common subtypes of EOC requires several somatic or inherited genetic alterations. Furthermore, epigenetic mechanisms also play an important role in the development and progression of ovarian cancer.5 The epigenetic alterations include DNA methylation, histone modifications and nucleosome remodeling.5 Most ovarian cancers arise because of the accumulation of genetic and epigenetic alterations, but the specific pathways for the development and progression of EOC remain poorly understood with regard to causality and biologic mechanisms.
Oxidative Stress, Ovarian Carcinogenesis
Recent molecular studies support a broad classification of EOC into 2 major types, designated as type 1 and type 2 (Table 1). Endometriosis serves as a precursor of ovarian cancer (EAOC), especially of the EAC and CCC subtypes.6 Chromosomal regions in endometriosis contain genes involved in ovarian tumorigenesis. The fundamental features of human neoplasms (genetic changes, loss of heterozygosity, specific gene mutations including KRAS, PTEN, and PIK3CA [phosphoinositide-3-kinase, catalytic, > polypeptide]) have been evaluated in endometriosis.7 More recent data implicate that ARID1A (AT-rich interactive domain-containing protein 1A) and PPP2R1A (protein phosphatase 2, regulatory subunit A, >) are frequently disrupted in CCC (see later paragraphs).8,9 ARID1A acts as a tumor-suppressor gene, and its mutation and loss of the gene product BAF250a can be seen in the preneoplastic lesions.8,9 It has been strongly speculated that the somatic mutation of ARID1A gene is an early event in the transformation of endometriosis into cancer.8,9 Sequence mutations of ARIA1A and PIK3CA occurred in 57% and 33% of CCC cases, respectively.8Y10 At present, the specific pathways such as ARIA1A, PIK3CA, PTEN, PPP2R1A, ERK (extracellular signalYregulated kinase), KRAS, BRAF, CTNNB1, and TP53 might be involved in the regulation of CCC carcinogenesis.8Y10 The animal model approach was recently used to demonstrate the role of KRAS and PTEN in the initiation of EAC.11 However, little is known about the development and progression of CCC. There have been no spontaneous CCC tumor mouse models available until now. In addition, genetic events that are involved in CCC pathogenesis might include not only point mutations but also the other types of genetic alterations such as gene amplification and deletion. The recent study provides the DNA copy number alterations in ovarian CCC.12 They identified for the first time discrete regions with either DNA copy number gain (Zinc finger protein 217, ZNF217) or loss (cyclin-dependent kinase inhibitor, CDKN2A/2B) in CCC tissues. Overexpression of ZNF217 conferred resistance to paclitaxel, stimulated
TABLE 1. Model of ovarian carcinogenesis Type 1 Histologic diagnosis
Mucinous Endometrioid Clear cell Characteristics Low grade and slow growing Preexisting lesions with adjacent benign- or borderline-like lesions Gene KRAS, BRAF, PTEN, mutations PIK3CA, ERK, CTNNB1, ARID1A, PPP2R1A
* 2011 IGCS and ESGO
Copyright © 2011 by IGCS and ESGO. Unauthorized reproduction of this article is prohibited.
Type 2 Serous
Rapidly progressing, high grade No evidence of precursor lesions P53, p16INK4a/RB
1201
Yamada et al
International Journal of Gynecological Cancer
cell proliferation, and antiapoptosis.13 Interestingly, homozygous deletion of CDKN2A/2B has been recognized as one of the major target genes involved in iron overload-induced carcinogenesis.14
PATHOPHYSIOLOGY OF CCC Oxidative Stress A Redox-Active Iron-Induced Oxidative Stress A considerable body of evidence provides support for a role of oxidative stress in ovarian carcinogenesis.6,15Y20 Markers of oxidative damage, such as strand breakage, DNA adducts, and lipid peroxidation products (eg, DNAmalondialdehyde adducts as an indication of lipid peroxidation; 8-oxo-7,8-dihydro-2¶-deoxyguanosine as an indication of DNA damage), can be detected in ovarian cancer tissue.21 We have previously described the hepatocyte nuclear factor (HNF) 1AYdependent pathophysiology of CCC and discussed its role in oxidative stressYinduced carcinogenesis.17,18,21 Eighty-seven percent of the specific genes upregulated in CCC were associated with the redox-related genes including oxidative and detoxification enzymes.20,21 Kajihara et al20 reported that these redox proteins may serve as an important molecular marker for the diagnosis and molecular understanding of CCC.
& Volume 21, Number 7, October 2011
The high frequency of genomic mutations found in EAOC may result from the presence of participation of mutagenic organics in the induction of carcinoma by endometriotic cyst fluids.15 Recent results also show that, in progressive stages of EOC, the oxidative stress strongly contributes to the uncontrolled tumor expansion. The CCC-specific gene signature contains typical markers of CCC, such as HNF-1A, versican (VCAN), and other genes that reflect oxidative stress.19,20 One of the most common mutagenic factors in CCC might be a redox-active iron19,20 (Fig. 1). A histologically normal ectopic endometrium bears genetic damages caused by iron-dependent oxidative stress. The persistent oxidative stress culminates in endometriotic cell damage and death due to the accumulated DNA aberrations. The specific cells that could avoid a cell death might become cancer.
ERK Regulation by Oxidative Stress There are several mechanisms by which the ERK pathway is activated. An activated ERK1/2 pathway might be critical to tumor growth, differentiation, and survival of EOC with KRAS or BRAF mutations. The ERK activity can also be controlled by not only KRAS or BRAF mutations but also specific protein tyrosine phosphatase (PTP) levels. Mitogenactivated protein kinase phosphatase 3 (MKP3) is a negative regulator of ERK1/222 (Fig. 1). Chan et al22 reported that
FIGURE 1. Redox-active iron-induced ROS signals contributes to carcinogenesis. The sources of the increased inflammation and oxidative stress may derive from the increased burden of repeated bleeding and from the increased amounts of ROS generated by redox-active heme and iron. These iron-induced ROS signals can contribute to carcinogenesis via 3 major processes: step 1, increasing oxidative stress promotes DNA mutagenesis, histone modification, chromatin remodeling, and gene products inactivation (initiation); step 2, activating detoxification and antiapoptotic pathways via the transcription factor HNF-1A overexpression (promotion); and step 3, E2-dependent (EAC) or E2-independent (CCC) mechanisms support further tumor progression. Over time, persistent excess iron exposure may downregulate ER> expression, which may lead to CCC carcinogenesis. Thunder symbol indicates increased susceptibility to ROS.
1202
* 2011 IGCS and ESGO
Copyright © 2011 by IGCS and ESGO. Unauthorized reproduction of this article is prohibited.
International Journal of Gynecological Cancer
& Volume 21, Number 7, October 2011
MKP3 often loses its expression in the protein level in ovarian cancer. The loss of MKP3 protein is associated with ubiquitination/proteosome degradation mediated by reactive oxygen species (ROS). A direct increase of the phospho-ERK level is observed on hydrogen peroxide treatment. These results suggest that the accumulation of ROS may cause the direct phosphorylation of ERK and the degradation of MKP3, which in turn leads to aberrant ERK1/2 activation and contributes to carcinogenesis.22 Because extramitochondrial source of ROS generator, NADPH oxidase (NOX1), is predominantly associated with the ROS generation, the NOX system may be critical for the malignant phenotype and carcinogenesis of some cancer cells. Thus, ROS can act as secondary messengers and control various signaling cascades.23 Oxidative stressYsensitive gene mutation and inactivation of the gene products greatly modulate the protein levels and their activities. Thus, the ERK protein function might be repressed by the mutation of the specific genes such as KRAS and inactivation of MKP3.
KRAS Regulation by Oxidative Stress The RAS-BRAF-MEK (mitogen-activated protein kinase kinase)YERK pathway plays a pivotal role in various cellular responses, including cellular growth, differentiation, and survival. Some of the targeted pathway inhibitors or antagonists of BRAF and its downstream effectors are in stages of clinical development for tailored combination strategies. Interestingly, identified RAS mutations are limited to a small number of sites. These mutations were enriched in amino acids 12, 13, 59, and 61, demonstrating that this reveals nonrandom patterns.24 For example, mutation in the RAS codon 12 results in a gain of function, subsequently activates RAF-MAPK, phosphoinositide 3¶-kinase (PI3K), NF-JB, and RAL-GTPases, thus resulting in accumulation of p16INK4a, p53, and p21 CYP1.24,25 Because human cancers with mutant KRAS coexpress the NF-JB target gene signatures, NF-JB may be another important KRAS downstream mediator. Even in the absence of obvious genetic somatic mutations, the RAS pathway is activated in some cancers.18 First, the RAS oncogene activation is involved in cellular detoxification as well as in oxidative stress. RAS induces constitutive expression of NOX1 through the MAPK pathway.26 Generation of ROS by RAS-induced NOX1 is required for anchorage-independent cell growth, anoikis resistance, augmented angiogenesis, and tumorigenesis.26 NOX1 signaling also causes oxidative inactivation of protein tyrosine phosphatases.26 Protein tyrosine phosphatases decrease growth factorYinduced signaling pathways such as DNA synthesis by inhibiting the tyrosine phosphorylation level of the growth factor receptors, suggesting that aberrant activation of the NOX1 activity benefits transformation phenotypes of a subset of cancer cells. Therefore, the RAS-induced, NOX1-mediated inactivation of protein tyrosine phosphatases serves as a regulatory switch for growth factor receptor activation (Fig. 1). Second, ROS were proposed to be involved in tumor metastasis, which is multifactorial steps, including the epithelial-mesenchymal transition (EMT), invasion of the tumor cells and angiogenesis. Reactive oxygen species activate epidermal growth factor receptor family and other tyrosine
Oxidative Stress, Ovarian Carcinogenesis
kinases, leading to activation of RAS proteins and multiple downstream signaling pathways. The epidermal growth factor also upregulates Snail and downregulates E-cadherin expression through production of H2O2 in human ovarian cancer cells. Furthermore, ROS potentiates cell invasion by upregulating gene expression of hepatocyte growth factor and Wnt/Acatenin as well as by modulating expression of urokinase-type plasminogen activator (Fig. 1). Third, ROS can induce activation of ERK1/2 and protein kinase B (AKT).18 The RAS/RAF/MEK/ERK and RAS/ PI3K/PTEN/AKT pathways interact with each other to regulate cell proliferation and tumorigenesis.18 Finally, ROS can cause cytosine deamination. This DNA modification may be relevant to glyoxal-induced mutations at GC pairs. Uracil in DNA generates promutagenic UG mispairs. The amino group of the G residue on the 3¶ side of the mispaired pyrimidine shows hindered rotation, leading to the specific gene mutation. Therefore, oxidative stress directly and indirectly activates the KRAS signaling cascades.
PTEN Regulation by Oxidative Stress Inactivation of PTEN by loss-of-function mutations and subsequent activation of PI3K-AKT signaling pathways have been detected in many human cancers, including brain, breast, prostate, and ovarian cancers.27 PTEN activity could also be regulated through phosphorylation-dependent modulation of protein stability and/or through reversible oxidation. First, a substantial proportion (approximately 10%) of PTEN proteins was oxidatively inactivated by H2O2 produced in various cell types including macrophages. The oxidation of the active site cysteine residue of this phosphatase abrogates its nucleophilic properties, thus rendering PTEN inactive. Therefore, oxidative stress can directly stimulate PI3Kdependent signaling, not by suppression of PTEN protein expression, but by inhibition of PTEN activity. Second, it has been suggested that DJ-1 (also known as PARK7 [Parkinson disease autosomal recessive early onset 7]) negatively regulates the tumor suppressor PTEN.28 Cysteine106 of DJ-1 is an essential amino acid for DJ-1 to exert its function. Oxidation of cysteine106 of DJ-1 occurred in cells treated with H2O2, accompanied by increased binding of oxidized DJ-1 to PTEN, decreased PTEN activity, and increased phosphorylation of AKT, leading to cell proliferation and transformation (Fig. 1). It functions as a redox-sensitive chaperone, as a sensor for oxidative stress (antioxidative stress reaction), and it apparently protects cancer cells and neurons against oxidative stress and cell death. Third, ROS generated by NOX1 may decrease PTEN activity by activating intermediate redox-sensitive protein kinases, leading to activation of Wnt/A-catenin pathway,29 which has important implications in carcinogenesis in a variety of human cancer (Fig. 1). Dysregulation of the Wnt signaling pathway, caused by loss of the adenomatous polyposis coli (APC) gene or mutational activation of A-catenin is sufficient to give rise to a tumor. Finally, PTEN activity is involved in redox regulation. For example, the loss of PTEN activity shows cellular susceptibility to oxidative stress through antioxidant-responsive
* 2011 IGCS and ESGO
Copyright © 2011 by IGCS and ESGO. Unauthorized reproduction of this article is prohibited.
1203
International Journal of Gynecological Cancer
Yamada et al
element (ARE)Ymediated transcription of detoxification genes. It also determines chromatin modifications leading to ARE activation. These data allow us to consider that PTEN activity is often regulated through oxidation stimulation.
Iron-Induced Carcinogenesis A growing body of evidence has accumulated that redox-active iron-induced oxidative stress is associated with carcinogenesis. First, oxidative stress, due to smoking or various other causes, triggers chronic inflammatory processes and changes in the immune system that is central to the pathogenesis of lung disease, including lung cancer, chronic obstructive lung disease, and atherosclerosis. Cigarette smoking is associated with a high risk for lung cancer. Heme iron intake also increased the risk of lung cancer. Second, patients with inflammatory bowel disease including ulcerative colitis are at increased risk for colorectal cancer, which involves a morphologic progression from inflamed and hyperplastic epithelia to dysplasia and adenocarcinoma. It has been shown that, with dietary iron supplementation, there is increased cancer development in dextran sulfate sodiumYinduced chronic ulcerative colitis. This suggests that inflammation-induced, iron-enhanced oxidative stress is a major player for colon carcinogenesis. Third, metal-induced mutagenicity and carcinogenicity depend on a multifactorial process. Carbon nanotubes might cause an asbestos-like pathologic lesion in the lung. Animal experiments showed that inhalation of carbon nanotubes (with an iron content of 17.7% wt) causes inflammatory response, oxidative stress, and fibrosis as well as mutations of KRAS gene locus in the lung of mice. Finally, iron is rich in the cyst fluids of endometrioma as well as EAOC.19 Iron is a possible cause of CCC carcinogenesis through iron-induced persistent oxidative stress.19 The major toxicity mechanisms include induction of inflammatory response and oxidative stress exacerbated by transition metals including iron. It could generate ROS and donate electrons for the generation of the superoxide radical and participate in the generation of hydroxyl radicals via the Fenton reaction (Fe2+ + H2O2jYFe3+ + OHj + OH). Inflammation provides a redox environment in which transition metals can exhibit their prooxidant potential. A combination of inflammatory response with catalytically competent metal would synergistically enhance damage to cells and DNA. There is an accumulating evidence that ROS can directly damage vital cellular components, including DNA, lipids, and proteins, leading to profound cellular toxicity (Fig. 1). For example, the hydroxyl radical may activate oncogenes or inactivate tumor suppressor genes through point mutations, activate chemical carcinogens, and prevent DNA repair. Among various classes of oxidative DNA damage, the most frequent, abundant, and investigated mutagenic base lesion caused by ROS is 7,8-dihydro-8-oxoguanine (8-oxoG) that can result in adenine mismatches during DNA replication.29 Point mutations generated via oxidative DNA damage might be involved in carcinogenesis and cancer progression. Another example is that acrolein, a major lipid peroxidation product, is chemically reactive and mutagenic.
1204
& Volume 21, Number 7, October 2011
In contrast, to counteract such oxidative damage in nucleic acids, cells are equipped with several defense mechanisms.30 Iron loading also causes up-regulation of antioxidant detoxification genes such as ferritin. Ferritin gene belongs to the ARE-regulated gene family. Ferritin is upregulated at the transcriptional level under oxidative conditions. Over time, the excessive ROS production can exceed the antioxidant defenses. Oxidative stress, caused by the imbalance between the generation and detoxification of ROS, plays an important role in carcinogenesis. Taken together, persistent excess iron overload might be associated with cell toxicity and induce carcinogenesis in humans, and oxidative DNA, protein, and lipid damage has been considered a key factor in carcinogenesis. However, direct causative evidence is lacking, and the molecular mechanisms remain obscure.
Mitochondrial Dysfunction The transcription factor HNF-1A may play a critical role in a cytoprotective effect against forthcoming oxidative stress. We analyzed HNF-1A siRNA-induced gene expression using a microarray approach.20 Several highly up- and downexpressed genes were confirmed by real-time polymerase chain reaction. Expression of OGG1 (8-oxoguanine DNA glycosylase), a repair enzyme, was decreased by treating with HNF-1A siRNA compared with control (our unpublished data, 2010). Removal of oxidized bases is initiated by a DNA glycosylase. A human DNA glycosylase encoded by the OGG1 gene has an activity to remove 8-oxoG from DNA and suppresses the mutagenic effect of 8-oxoG. Although there is no evidence whether OGG1 is a downstream target of HNF-1A, HNF-1A may upregulate OGG1 gene expression to counteract ROSmediated mitochondrial dysfunction. Mammalian cells are equipped with several defense mechanisms to prevent the accumulation of oxidative damage.30 Cells are armed with a vast repertoire of antioxidant defense mechanisms to counteract oxidative damage even in the genome. Thus far, at least 4 distinct repair enzymes (OGG1, MTH1 [7,8-dihydro-8-oxoguanine triphosphatase], MUTYH [mutY homolog], and APEX2 [apurinic/apyrimidinic endonuclease 2]) have been identified to function to repair or prevent oxidized bases, thereby suppressing carcinogenesis and cell death.30 Knockout of these repair enzymes in mice has been shown to produce a phenotype of higher cancer incidence. OGG1 knockout mice exhibit elevated levels of 8-oxoG in nuclear and mitochondrial DNA. The increased incidence of lung cancer, UV-induced skin cancer, or colon cancer has been reported in OGG1 knockout mice. Codon 12 of KRAS seems to be an important downstream target of oxidative DNA damage, resulting from OGG1 deficiency. The OGG1 Cysteine326 variant also seems to play a role in conferring increased risk of ovarian cancer. MTH1 protein hydrolyzes oxidized purine nucleoside triphosphates, such as 8-oxo-2¶-deoxyguanosine triphosphate and 2-hydroxy-2¶-deoxyadenosine triphosphate (2-OHdATP), to the corresponding nonmutagenic monophosphates.29 Increased susceptibility to carcinogenesis is observed in * 2011 IGCS and ESGO
Copyright © 2011 by IGCS and ESGO. Unauthorized reproduction of this article is prohibited.
International Journal of Gynecological Cancer
& Volume 21, Number 7, October 2011
Oxidative Stress, Ovarian Carcinogenesis
MTH1-null mice.29 Tumors were formed in the lungs, livers, and stomachs of MTH1-deficient mice. Germ-line mutations in MUTYH have been associated with recessive inheritance of multiple colorectal tumors. MUTYH mutation-associated polyposis (MAP) is characterized by a lifetime risk of colorectal cancer of up to 100%. A significant increase of MUTYH mutations has been observed in the incidence of ovarian cancers. It is unlikely, however, that MTH1 and APEX mutations are directly associated with ovarian cancer. There is no study whether these genes are modulated under the oxidative stress conditions in CCC.
HNF-1AYdependent pathway provides new insights into the regulation of glycogen synthesis and accumulation, possibly through up-regulation of glucose transporter type 2, glucose6-phosphatase, and dipeptidyl-peptidase 4.18 Although there is no relation between CCC and glycogen storage disease (GSD), the high risk for hepatocellular carcinoma in GSD I, III, VI, and IX has become evident, suggesting that excess glycogen accumulation is at risk for developing cancer. Excess glycogen accumulation may be associated with CCC carcinogenesis.
Estrogen Receptor
This review finally describes the evidence for the role of oxidative stress in the pathophysiology of CCC and discusses the oxidative stressYmediated chromatin remodeling. The pathophysiology of endometriosis and EAOC is characterized by persistent/chronic inflammation and oxidant/ antioxidant imbalance as described. The sources of the increased inflammation and oxidative stress may derive from the increased burden of repeated bleeding and from the increased amounts of ROS generated by redox-active heme and iron. Upregulation of iron-induced ROS expression is reflected by increased oxidative stress marker, 8-oxoG. The elevation in the levels of 8-oxoG is also associated with the disruption of chromatin remodeling via modulation of histone deacetylation, allowing access for transcription factor DNA binding33 (Fig. 1). These results clearly highlight the importance of oxidative stress in the induction of DNA damage and disruption of chromatin remodeling. In general, oxidative stressYdependent increased DNA damage results in global hypomethylation due to interference with the ability of DNA to function as a substrate for the DNA methyl transferase.34 DNA methylation also plays an important role in the regulation of gene expression, chromatin architecture, chromatin-associated proteins, and histone modification. The cytosine is the preferred base for CpG island methylation, whereas the guanine is the site for oxidative damage by causing profound DNA adducts. Current understanding of the roles of ROS in cellular functions as upregulation of proinflammatory responses, signaling pathways, activation of transcription factors, specific gene methylation or mutation, chromatin remodeling, and gene product expression will provide important information regarding pathologic processes contributing to CCC. Recent studies showed that the frequently mutated genes in CCC tumors are ARID1A, PIK3CA, PPP2R1A, and KRAS.8,9 The 2 mutated PPP2R1A and ARID1A genes were novel.8,9 PPP2R1A encodes a regulatory subunit of serine/ threonine phosphatase.9 The encoded protein is implicated in the negative control of cell growth and division. PPP2R1A has been shown to function as a tumor suppressor gene in breast cancer.35 In addition, ARID1A encodes AT-rich interactive domain-containing protein 1A, which participates in chromatin remodeling.8,9 The ARID1 chromatin-remodeling complex facilitates DNA access by the transcription machinery.36 Li et al36 recently reported that ARID1 is associated with an E3 ubiquitin ligase. The typical E3 ubiquitin ligases include not only ARID1 but also von HippelYLindau
Clear cell carcinomas and EACs are subtypes of EOC associated to endometriosis. In contrast to EAC, CCCs exhibit a significant decrease of cell-associated expression of estrogen receptor (ER) and progesterone receptor (PR). Endometrioid adenocarcinoma showed 100% and 92% immunostaining for ER> and PR, respectively.31 Conversely, 0% and 10% of CCC showed positive nuclear staining for ER> and PR, respectively.31 In addition to Fe2+-dependent free radical damage, 17A-estradiol (E2) also induces mitochondrial ROS production and subsequent free radical damage to DNA. Toxicological concentrations of E2 producing high concentrations of ROS may be directly toxic to the cells by activating apoptosis. Reactive oxygen species can also induce activation of ERK1/2 and AKT. Both kinases have been implicated in the phosphorylation of serine 118 and serine 167 on ER>, respectively. These phosphorylations lead to down-regulation of ER>. Over time, persistent ROS exposure to endometriosis downregulates ER> expression, which may lead to CCC carcinogenesis (Fig. 1). These data allow us to speculate that the more excess of free radicals in CCC rather than EAC will attack cellular macromolecules including DNA leading to carcinogenesis. The CCC and EAC subtypes exhibit a unique molecular profiles: HNF-1A+/detox+/ERj for CCC and HNF1Aj/detoxj/ER+ for EAC, respectively. These results suggest that the HNF-1A and ER> have different effects on pathogenesis of CCC and EAC. Epigenetic events, such as deacetylation and methylation of histones, might be involved in the regulation of promoter transcription. In a fraction of breast cancers, the loss of ER> protein expression is a result of the hypermethylation of the ER> promoter. Unfortunately, the exact interplay of these factors in the oxidative stressYdependent transcriptional repression activity is not yet well understood in EAOC.
Glycogen Accumulation Tumor-specific alterations in glucose and glycogen metabolism have been investigated for decades. Hypoxic microenvironments in tumors alter cancer cell metabolism. Cancer cells rely on anaerobic glycolysis rather than on respiration for ATP generation, a phenomenon known as the Warburg effect.32 Microscopic examination revealed abundant cytoplasmic glycogen content in CCC as demonstrated by PAS staining. Previous microarray studies have sought differences in glucose and glycogen metabolism in CCC and other histotypes.18 The
Chromatin Remodeling
* 2011 IGCS and ESGO
Copyright © 2011 by IGCS and ESGO. Unauthorized reproduction of this article is prohibited.
1205
International Journal of Gynecological Cancer
Yamada et al
(VHL) complexes, which serve as scaffolds. Alteration of the VHL gene by mutation, loss of heterozygosity, and promoter methylation has been found to be important to renal clear cell carcinoma (RCC) pathogenesis. When the VHL gene is mutated or lost, hypoxia-inducible factor 1> accumulates, leading to the increased expression of hypoxia-related genes, which are potentially important in RCC carcinogenesis and progression. The protein complex containing ARID1 protein is assembled in a manner similar to that for the well-characterized VHL complex.36 VHL recognizes and subsequently ubiquitinates hypoxia-inducible factor 1>, whereas ARID1 targets histone H2B.36 Posttranslational modifications of histones, such as H2B ubiquitination, might show a positive correlation with transcription activation. It has also been reported that, although there are no precise data on iron, heavy metals such as nickel cleaved off the histone H2B possibly due to oxidative stress. Although the molecular mechanism by which ARID1Adependent carcinogenicity is exerted is not fully understood, this gene product may regulate transcription of certain genes by altering the chromatin structure around those genes. These data allow us to speculate that the ARID1A gene product is part of an ubiquitin ligase complex that targets H2B for ubiquitination and proteasomal degradation, linking HOXA9 response genes to CCC oncogenesis. Taken together, these results suggest that oxidative stressYinduced aberrant chromatin remodeling including ARID1A contributes to the pathogenesis of CCC.9
CONCLUSIONS The present review integrates basic research observations in an attempt to provide a comprehensive understanding of how oxidative stress processes contribute to ovarian cancer (EAOC) development in endometriosis patients. This review also sheds light on future research on oxidative stress cascades. Cancer is a complex progressive multistep disease that involves the accumulation of both genetic and epigenetic abnormalities. The most common genetic alterations include chromosomal structure changes such as loss of heterozygosity, deletions, gains, amplification, and gene mutations such as base substitutions. The important epigenetic alterations are DNA methylation, histone modification, and chromatin remodeling. One of the most common mutagenic factors in endometriosis might be a redox-active iron. Repeated hemolysis occurring during the development of endometriosis results in high levels of free heme and iron.16 In patients with endometriosis, persistent iron overload has been demonstrated in the components of the endometrioma (endometriotic cyst fluid) and the peritoneal cavity (peritoneal fluid, endometriotic lesions, peritoneum, and macrophages). The genome-wide expression analysis demonstrated that the genes upregulated in endometriosis are involved in oxidative stress.16 Several important endometriosis-specific genes overlap with those known to be regulated by iron.16 The increased iron-induced oxidative stress eventually culminate in endometriotic cell damage and finally death due to the accumulated DNA aberrations. Conversely, HNF-1A overexpressed in endometriotic
1206
& Volume 21, Number 7, October 2011
epithelial cells acts as a potent antioxidant and detoxification and antiapoptotic agent to continue to grow. Persistent iron exposure can result in toxicity and is also associated with carcinogenesis. Not only gene mutations but also microenvironmental changes play an important role in EAOC carcinogenesis. The pathways such as ARID1A, PPP2R1A, MAPK, KRAS, PI3K, and PTEN are involved in the regulation of EAOC carcinogenesis8,9 (Fig. 1). The distribution pattern of base substitutions in somatic mutations might be nonrandom distribution, which generates mutation hotspots. First, most KRAS mutations identified have been G-to-T transversions at codon 12, suggesting that mutagens including ROS cause specific DNA adduct formation and contribute to carcinogenesis. Second, oxidative stress might directly stimulate PI3Kdependent signaling by suppression of PTEN expression or inhibition of PTEN activity. Oxidative stressYsensitive gene mutation and inactivation of the gene products greatly modulate the protein levels and its activities. Third, PPP2R1A acts as a tumor suppressor gene. This somatic mutation confers a risk of developing CCC. Fourth, ARID1A participates in aberrant chromatin remodeling. This mutation may be more susceptible to ROS-induced CCC. Finally, these redox-active iron-induced ROS signals can contribute to carcinogenesis via 3 major processes: step 1, by increasing oxidative stress, which promotes DNA mutagenesis, histone modification, chromatin remodeling, and gene products inactivation, thus contributing to EAOC initiation; step 2, by activating detoxification and antiapoptotic pathways via the up-regulation of the transcription factor HNF-1A expression, thereby contributing to CCC promotion; and step 3, by creating an environment that supports sustained growth, angiogenesis, migration, and invasion of cancer cells via E2-dependent (EAC) or E2independent (CCC) mechanisms, thus supporting tumor progression and metastasis. Step 1 of these changes seems to be common in CCC and EAC. These mutational and environmental analyses provide a window into the genetic landscape of EAOC. The great effort has made to establish a molecularly based carcinogenesis mechanism. HNF-1A has been identified as one of the most highly overexpressed genes in CCC but not EAC. Another candidate factor, which is associated with an initial switch to CCC carcinogenesis, might be the redox-active iron. The following issues must still be clarified: What triggers the susceptibility of specific genes to ROS? What causes activation or silencing of these specific gene? What essentially differentiates CCC from EAC? These tumor-specific aberrations provide clues to the cellular processes underlying carcinogenesis and have proven useful for diagnostic and therapeutic purposes. These aspects of ROS biology will be discussed in the context of its relationship to CCC carcinogenesis and therapeutic strategies. Understanding of the mechanisms of redox signaling leads to the development of novel therapies based on the pharmacological manipulation of antioxidants in endometriosis to prevent EAOC development. Antioxidants, for example, thiols or molecules that have dual antioxidant and anti-inflammatory activity, could be potential therapeutic agents and help prevent carcinogenesis. The proof of concept will come from clinical studies on the effectiveness of antioxidant therapy. * 2011 IGCS and ESGO
Copyright © 2011 by IGCS and ESGO. Unauthorized reproduction of this article is prohibited.
International Journal of Gynecological Cancer
& Volume 21, Number 7, October 2011
ACKNOWLEDGMENTS The authors thank all the study participants for their time and efforts. The authors thank Mikiko Kita for editorial assistance.
REFERENCES 1. Gubbels JA, Claussen N, Kapur AK, et al. The detection, treatment, and biology of epithelial ovarian cancer. J Ovarian Res. 2010;3:8. 2. Kobayashi H, Sumimoto K, Moniwa N, et al. Risk of developing ovarian cancer among women with ovarian endometrioma: a cohort study in Shizuoka, Japan. Int J Gynecol Cancer. 2007;17:37Y43. 3. Kurman RJ, Shih IeM. The origin and pathogenesis of epithelial ovarian cancer: a proposed unifying theory. Am J Surg Pathol. 2010;34:433Y443. 4. Shih IeM, Kurman RJ. Ovarian tumorigenesis: a proposed model based on morphological and molecular genetic analysis. Am J Pathol. 2004;164:1511Y1518. 5. Maldonado L, Hoque MO. Epigenomics and ovarian carcinoma. Biomark Med. 2010;4:543Y570. 6. Mandai M, Yamaguchi K, Matsumura N, et al. Ovarian cancer in endometriosis: molecular biology, pathology, and clinical management. Int J Clin Oncol. 2009;14:383Y391. 7. Vigano´ P, Somigliana E, Chiodo I, et al. Molecular mechanisms and biological plausibility underlying the malignant transformation of endometriosis: a critical analysis. Hum Reprod Update. 2006;12:77Y89. 8. Wiegand KC, Shah SP, Al-Agha OM, et al. ARID1A mutations in endometriosis-associated ovarian carcinomas. N Engl J Med. 2010;363:1532Y1543. 9. Jones S, Wang TL, Shih IM, et al. Frequent mutations of chromatin remodeling gene ARID1A in ovarian clear cell carcinoma. Science. 2010;330:228Y231. 10. Kuo KT, Mao TL, Jones S, et al. Frequent activating mutations of PIK3CA in ovarian clear cell carcinoma. Am J Pathol. 2009;174:1597Y1601. 11. Dinulescu DM, Ince TA, Quade BJ, et al. Role of K-ras and Pten in the development of mouse models of endometriosis and endometrioid ovarian cancer. Nat Med. 2005;11:63Y70. 12. Kuo KT, Mao TL, Chen X, et al. DNA copy numbers profiles in affinity-purified ovarian clear cell carcinoma. Clin Cancer Res. 2010;16:1997Y2008. 13. Thollet A, Vendrell JA, Payen L, et al. ZNF217 confers resistance to the pro-apoptotic signals of paclitaxel and aberrant expression of Aurora-A in breast cancer cells. Mol Cancer. 2010;9:291. 14. Toyokuni S. Mysterious link between iron overload and CDKN2A/2B. J Clin Biochem Nutr. 2011;48:46Y49. 15. Yamaguchi K, Mandai M, Toyokuni S, et al. Contents of endometriotic cysts, especially the high concentration of free iron, are a possible cause of carcinogenesis in the cysts through the iron-induced persistent oxidative stress. Clin Cancer Res. 2008;14:32Y40. 16. Kobayashi H, Yamada Y, Kanayama S, et al. The role of iron in the pathogenesis of endometriosis. Gynecol Endocrinol. 2009;25:39Y52. 17. Kobayashi H, Kajiwara H, Kanayama S, et al. Molecular pathogenesis of endometriosis-associated clear cell carcinoma of the ovary. Oncol Rep. 2009;22:233Y240. 18. Kobayashi H, Yamada Y, Kanayama S, et al. The role of hepatocyte nuclear factor-1A in the pathogenesis of clear cell carcinoma of the ovary. Int J Gynecol Cancer. 2009;19:471Y479.
Oxidative Stress, Ovarian Carcinogenesis
19. Yamaguchi K, Mandai M, Oura T, et al. Identification of an ovarian clear cell carcinoma gene signature that reflects inherent disease biology and the carcinogenic processes. Oncogene. 2010;29:1741Y1752. 20. Kajihara H, Yamada Y, Kanayama S, et al. Clear cell carcinoma of the ovary: potential pathogenic mechanisms. Oncol Rep. 2010;23:1193Y1203. 21. Sanchez M, Torres JV, Tormos C, et al. Impairment of antioxidant enzymes, lipid peroxidation and 8-oxo-2¶-deoxyguanosine in advanced epithelial ovarian carcinoma of a Spanish community. Cancer Lett. 2006;233:28Y35. 22. Chan DW, Liu VW, Tsao GS, et al. Loss of MKP3 mediated by oxidative stress enhances tumorigenicity and chemoresistance of ovarian cancer cells. Carcinogenesis. 2008;29:1742Y1750. 23. Kumar B, Koul S, Khandrika L, et al. Oxidative stress is inherent in prostate cancer cells and is required for aggressive phenotype. Cancer Res. 2008;68:1777Y1785. 24. Adjei AA. Blocking oncogenic Ras signaling for cancer therapy. J Natl Cancer Inst. 2001;93:1062Y1074. 25. Serrano M, Lin AW, McCurrach ME, et al. Oncogenic ras provokes premature cell senescence associated with accumulation of p53 and p16INK4a. Cell. 1997;88:593Y602. 26. Shinohara M, Shang WH, Kubodera M, et al. Nox1 redox signaling mediates oncogenic Ras-induced disruption of stress fibers and focal adhesions by down-regulating Rho. J Biol Chem. 2007;282:17640Y17648. 27. Wu X, Senechal K, Neshat MS, et al. The PTEN/MMAC1 tumor suppressor phosphatase functions as a negative regulator of the phosphoinositide 3-kinase/Akt pathway. Proc Natl Acad Sci U S A. 1998;95:15587Y15591. 28. Nagakubo D, Taira T, Kitaura H, et al. DJ-1, a novel oncogene which transforms mouse NIH3T3 cells in cooperation with ras. Biochem Biophys Res Commun. 1997;231:509Y513. 29. Nakabeppu Y, Sakumi K, Sakamoto K, et al. Mutagenesis and carcinogenesis caused by the oxidation of nucleic acids. Biol Chem. 2006;387:373Y379. 30. Nakabeppu Y, Tsuchimoto D, Ichinoe A, et al. Biological significance of the defense mechanisms against oxidative damage in nucleic acids caused by reactive oxygen species: from mitochondria to nuclei. Ann N Y Acad Sci. 2004;1011:101Y111. 31. Fujimura M, Hidaka T, Kataoka K, et al. Absence of estrogen receptor-> expression in human ovarian clear cell adenocarcinoma compared with ovarian serous, endometrioid, and mucinous adenocarcinoma. Am J Surg Pathol. 2001;25:667Y672. 32. Jelluma N, Yang X, Stokoe D, et al. Glucose withdrawal induces oxidative stress followed by apoptosis in glioblastoma cells but not in normal human astrocytes. Mol Cancer Res. 2006;4:319Y330. 33. Rahman I, Gilmour PS, Jimenez LA, et al. Oxidative stress induces histone acetylation in alveolar epithelial cells (A549). Am J Respir Crit Care Med. 2001;163:A61. 34. Donkena KV, Young CY, Tindall DJ. Oxidative stress and DNA methylation in prostate cancer. Obstet Gynecol Int. 2010;2010:302051. 35. Dupont WD, Breyer JP, Bradley KM, et al. Protein phosphatase 2A subunit gene haplotypes and proliferative breast disease modify breast cancer risk. Cancer. 2010;109. 36. Li XS, Trojer P, Matsumura T, et al. Mammalian SWI/SNFVa subunit BAF250/ARID1 is an E3 ubiquitin ligase that targets histone H2B. Mol Cell Biol. 2010;30:1673Y1688.
* 2011 IGCS and ESGO
Copyright © 2011 by IGCS and ESGO. Unauthorized reproduction of this article is prohibited.
1207