Endometrial cancer: experimental models useful for studies on

REVIEWEndocrine-Related Cancer (2003) 10 23–42

Endometrial cancer: experimental modelsuseful for studies on molecular aspects ofendometrial cancer and carcinogenesis

G VollmerMolecular Cell Physiology and Endocrinology, Institute of Zoology, Dresden University of Technology,Mommsenstr. 13, 01062 Dresden, Germany

(Requests for offprints should be addressed to G Vollmer; Email: [email protected])

Abstract

There is definitely a need for the development of new drugs for the treatment and cure of endometrialcancer. In addition there are various new drugs or phyto-remedies under development which areintended for use in the treatment and prevention of breast cancer, for the treatment of menopausalsymptoms and for hormone replacement therapy. The efficacy of novel drugs targeting steroidreceptors in endometrial cancers has to be evaluated and the safety of other endocrine measures onendometrial cancers or on endometrial carcinogenesis has to be assessed. For these experimentalpurposes five main classes of experimental models are available: spontaneous endometrialtumorigenesis models in inbred animals (Donryu rats, DA/Han rats, BDII/Han rats), inoculation tumorsfrom chunks of tumors (rat EnDA-tumor, human EnCa 101 tumor) or from inoculated tumor cell lines(rat RUCA-I cells, human Ishikawa and ECC-1 cells), developmental estrogenic exposure or chemicalcarcinogen exposure of CD-1 and ICR mice, transgenic approaches such as mice heterozygousregarding the tumor suppressor gene PTEN (pten + / − -mice) and endometrial tumor cell lines culturedunder conditions promoting in vivo-like morphology and functions e.g. cell culture on reconstitutedbasement membrane. Although the number of models is comparatively small, most aspects relatedto functions of estrogenic or gestagenic substances are assessable, particularly if variousexperimental models are combined. Whereas models based on human endometrial adenocarcinomacells are widely used, the properties and advantages of animal-derived models have mainly beenignored so far.

Endocrine-Related Cancer (2003) 10 23–42

Introduction

Endometrial cancer is the most frequently diagnosed malig-nancy of the female genital tract. Incidence rates of endo-metrial cancer are described as 10 to 25 women per100 000 with a clear geographic variation between e.g.European (Spain, UK, France) and North American (USAand Canada) countries, with a slightly higher incidence inNorth America (Parazzini et al. 1991). In 1998, an estimateof 36 000 new cases and 6300 deaths attributable to endo-metrial cancer were reported in the USA (Podratz et al.1998) resulting in an overall frequency which is onlyexceeded by breast, lung and colon cancer (Li et al. 1999).However, the mortality rate of endometrial cancer if com-pared with other cancers is low and the prognosis is excel-lent if it is detected in the early stages (Creasman 1997).The present knowledge of the role of endocrine factors inthe etiology of endometrial adenocarcinomas, their role in

Endocrine-Related Cancer (2003) 10 23–42 Online version via http://www.endocrinology.org1351-0088/03/010–023 2003 Society for Endocrinology Printed in Great Britain

the regulation of tumor growth, invasion and metastasis ofendometrial adenocarcinoma cells, as well as the potentialvalue of established and novel endocrine manipulations forprevention and treatment of endometrial adenocarcinomashave been exhaustively reviewed in a recent review article(Emons et al. 2000).

In the concluding remarks of the same paper it was statedthat neither progestin treatment, the major hormonal therapyof endometrial carcinoma, nor cytotoxic chemotherapyshowed substantial benefits in the adjuvant setting. There-fore, future research activities must evaluate new compoundsand new treatment strategies.

The endometrium is a classical hormone-dependenttissue and most of the endometrial adenocarcinomas arehormone-dependent tumors. A high percentage of thesetumors express the estrogen receptor(s) and/or progesteronereceptors. Investigators have, therefore, aimed at targetingsteroid hormone receptors by the development of novel sub-

Vollmer: Endometrial cancer

stances e.g. new antiestrogens, selective estrogen receptormodulators (SERMs) and aromatase inhibitors. For testingand characterization of these new substances appropriatemodels have to be available.

A further need for endometrial tumor models stems fromsafety considerations. Estrogens not only act as tumor pro-moters but also as carcinogens. This has consequences forthe approval of endocrine active substances intended for usein tumor therapy e.g. breast cancer, tumor prevention or hor-mone replacement therapy. The concerns are that substanceswhich are beneficial in one organ might harm another, e.g.tamoxifen, a common therapy in breast cancer, mightincrease the risk for endometrial cancer (Emons et al. 2000).Finally, plenty of phytoestrogens are recommended either assemipure substances or, as in the case of soy or red cloverextracts, for use as supportive measures in hormone replace-ment therapy as well as for use in lifestyle medicine. Noneof these substances has been exhaustively investigatedregarding its safety (Fugh-Berman & Kronenberg 2001, Balket al. 2002, Burton & Wells 2002).

These examples clearly demonstrate the need for exper-imental endometrial tumor models. The aim of this paperis to review the knowledge of endometrial tumor models.Particular emphasis is given to their applicability in vivo andto their potential responsiveness to natural as well as syn-thetic estrogens and progestins. The latter feature is a pre-requisite for the process of development and evaluation ofendocrine active substances.

Spontaneous endometrialadenocarcinoma of rats

Longevity studies revealed that female animals of four ratstrains die with an uncommonly high incidence rate for endo-metrial adenocarcinoma if kept to their natural life end. Dueto their clinical, pathophysiological, histological and bio-chemical features they all represent excellent experimentalmodels for endometrial carcinogenesis. Their potential asexperimental tumor models has so far not been appreciatedby the scientific community.

Han:Wistar rats

Already in 1981 Deerberg et al. published a report showingthat virgin Han:Wistar rats die with an incidence rate of 39%from tumors of the uterus if kept to their natural life end. Atotal of 35% of these tumors were endometrial adenocarcin-omas. This rat strain and the pathogenesis of its endometrialadenocarcinomas have not been investigated any further.

Donryu rats

Female Donryu rats have been presented as a hormone-dependent endometrial cancer model, an issue discussed in

24 www.endocrinology.org

detail below. This model is not well characterized regardinghistopathological properties and metastatic potential. Geneti-cally, in spontaneous endometrial cancers point mutations ofthe K-ras locus have been found recurrently (Tanoguchi etal. 1999). The latter mutation also has a rather frequentoccurrence in human endometrial adenocarcinoma (Boyd &Risinger 1991, Semczuk et al. 1998).

Female animals of the Donryu strain are characterizedby a similar mortality rate for endometrial adenocarcinomaas Han:Wistar rats and exhibit an incidence rate of 35.1%for endometrial adenocarcinoma and a total of proliferativelesions of approximately 60% (Nagaoka et al. 1990). Thepathogenesis of endometrial adenocarcinoma in this rat strainis characterized by some features which are discussed as riskfactors contributing to endometrial carcinogenesis in humans.Donryu rats are characterized by an imbalanced (increased)estradiol/progesterone rate which at the age of 12 months isincreased almost fivefold if compared, for example, withaged Fischer-344 rats (Nagaoka et al. 1990). In humans ano-vulatory cycles reflecting the estrogen/progesterone imbal-ance are regarded as a typical risk factor for endometrialadenocarcinoma (Barakat et al. 1997).

It has been proposed that endometrial carcinogenesis inDonryu rats is hormone dependent (Nagaoka et al. 1994).However, this conclusion has been drawn from experimentalobservations which should not necessarily be interpreted ashormone dependency. The first piece of evidence derivesfrom the above mentioned estrogen/progesterone imbalancein the serum of these animals. The other piece of evidencehas been deduced from comparative studies with Fischer-344rats which have a normal cyclicity and almost no advancedhistological changes such as hyperplastic or neoplasticlesions (Nagaoka et al. 1994) or alterations in the proliferat-ive capacity of the uterine epithelium (Ando-Lu et al. 1998).Further, reproductive experience reduced the incidence ofendometrial and mammary carcinoma. This finding is analo-gous to the situation discussed for humans (Terry et al.1999), although it only becomes apparent in Donryu rats afterthree periods of gestation. One- or twofold experiences wereineffective if compared with nulliparous animals (Nagaokaet al. 2000). However, the conclusive experiments provinghormone dependency, namely comparative studies in normalcyclic and ovariectomized animals are missing.

The overall yield of endometrial adenocarcinoma couldbe increased and the onset of endometrial adenocarcinomacould be considerably accelerated if animals were intraperi-toneally treated four times with a combination of estradioldiproprionate and N-methyl-N-nitrosourea (Nagaoka et al.1993) or if they received a single intra-uterine administrationof N-ethyl-N′-nitro-N-nitrosoguanidine via the vagina(Ando-Lu et al. 1994). High fat diets also led to an earlieronset of endometrial carcinogenensis which was paralleledby an earlier onset of imbalance in estrogen/progesteronelevels (Nagaoka et al. 1995). Transplacental administration


of diethylstilbestrol on days 17 and 19 at a dose of 0.1 mg/kg led to an increase in lesions of the female reproductivetract of Donryu female offspring, including an increase inendometrial adenocarcinoma (Kitamura et al. 1999).Although the overall process of carcinogenesis is acceleratedby these chemical carcinogens or the tumor promoting estro-gen, it is questionable whether results from studies applyingthis approach can be interpreted more accurately than studiesusing the spontaneous protocol. Treatment with chemicalcarcinogens induces mutations which unpredictably add togenetic alterations already present in Donryu rats, like thefrequent mutation of the Ki-ras locus (Tanoguchi et al.1999).

So far two therapeutic studies have been described usingthis model in the situation of spontaneous endometrial car-cinogenesis. Treatment with dietary indole-3-carbinolreduced the incidence rate of endometrial carcinogenesisconsiderably. The study showed a clear correlation betweenthe induction of estradiol 2-hydroxylation and a more nor-malized estrogen/progesterone level (Kojima et al. 1994). Ona mechanistic level this means chemoprevention is due toinduction of estradiol 2-hydroxylase. The other chemoprev-ention study was performed on N-ethyl-N′-nitro-N-nitrosoguanidine-treated Donryu rats. Tamoxifen treatmentsignificantly reduced the number of proliferative lesions inthe uteri which also was limited to a few hyperplastic lesions(Yoshida et al. 1998). The authors claimed that tamoxifenacted as an antiestrogen, because of decreased serum estra-diol levels. However, in none of the papers were any com-ments on estrogen receptor involvement made. Receptorinvolvement is crucial for any antiestrogenic function. Inaddition, the finding mentioned above does not reflect thesituation in humans. For the human situation it is wellaccepted that tamoxifen increases the relative risk for endo-metrial carcinogenesis (ACOG Committee Opinion 1996,Love et al. 1999), a calculation from epidemiological find-ings substantiated by the stimulation of proliferation in cul-tured endometrial adenocarcinoma cells by tamoxifen(Gusberg 1994, Creasman 1997).

DA/Han rats

DA/Han rats are characterized by an inbred background.Specific gene alterations are not yet documented. The over-whelming majority of tumors are moderately to well differen-tiated which makes the tumor phenotypically similar to mostof the human endometrial adenocarcinomas. This modelexhibits a highly metastatic phenotype (Fig. 1), but has notso far been used for intervention studies, because the tumorcan be transplanted into syngenic animals. There it gives riseto metastasizing endometrial adenocarcinomas which byhistopathological and biochemical means (expression ofestrogen receptor α (ER-α)) are undistinguishable from theparental tumor (Horn et al. 1993, 1994).

www.endocrinology.org 25

Female DA/Han animals die from endometrial adenocar-cinoma with an incidence rate of > 60% if kept to their natu-ral life end at 24–27 months of age. Ovariectomy prior toonset of cyclicity prevents endometrial carcinogenesis(Deerberg et al. 1985). This has to be interpreted as a clearindication of hormone dependency of the carcinogenic pro-cess. Sixty-three percent of all of the developing endometrialadenocarcinomas of the DA/Han rats metastasize into thelung, thereby using a lymphogenic pathway (Fig. 1; F Deer-berg, unpublished observations). Lymphatic invasion is acharacteristic of type II endometrial carcinoma (Emons et al.2000). Distant metastases are rare in cases of human endo-metrial cancer (Cook et al. 1999); however, among those fewcases the lung has been described as a frequent metastasessite (Bouros et al. 1996).

So far, these spontaneously occurring tumors have notbeen used for experimental purposes. However, from thesetumors a serially transplantable tumor called EnDA (Horn etal. 1993) and a cell line called RUCA-I (Schutze et al. 1992)which also gives rise to hormone-sensitive endometrial can-cers if inoculated into syngenic rats or athymic nude mice(see below) have been established (Fig. 2).

BDII/Han rats

In terms of an endometrial cancer model the female rats ofthis strain are unique worldwide. Since being first describedin 1987 (Deerberg & Kaspareit 1987) it took almost a decadebefore they were used for cancer treatment experiments, andan additional five more years before investigators started tounravel their genetic peculiarities which contribute to thehigh incidence of endometrial cancer. This tumor model isgenetically well characterized (discussed in detail below),although some important genetic information is still missinge.g. mutation frequency of some tumor suppressor genes (e.g.APC, PTEN). Most histopathological properties known forhuman endometrial adenocarcinoma, e.g. tumor grading, arealso found in this spontaneous animal model for endometrialadenocarcinoma (Deerberg & Kaspareit 1987). Furthermore,this model has been investigated for the occurrence of pre-cancerous lesions and their immunohistochemical properties.In rat endometrial tissue the same precancerous lesions havebeen found which are also detectable in human specimens.In addition, as in hyperplastic and displastic human endo-metrium, tenascin-C, a stromal marker for proliferative epi-thelial disease, is strongly up-regulated in similar lesions ofrat endometrial tissue (Vollmer et al. 1990, 1991, Sasano etal. 1993). These findings illustrate the large similarities incell biological aspects in the pathogenesis of endometrialcancers in humans and rats. In experimental setups for theunderstanding of the molecular mechanisms of endometrialcarcinogenesis this model is of high value and should be con-sidered as one of the most important in vivo models. Thereare several experimental pieces of evidence that endometrial


Figure 1 Endometrial carcinogenesis and lung metastasis of DA/Han rats. In the upper panel the overall incidence ofendometrial carcinogenesis of DA/Han rats, dependent on age, is given. The lower panel shows the percentage of those tumorsdeveloping lung metastases. ECA, endometrial carcinoma.

carcinogenesis of female BDII/Han animals is hormonedependent and therefore represents an endometrial tumormodel for spontaneous hormonal carcinogenesis.

If female animals of this strain are kept to their naturallife end (around 27 months of age) they die from endometrialcarcinoma or metastases with an incidence rate of > 90%.The overwhelming majority (87%–97%) of all carcinomas inthe various experimental groups were endometrial adenocar-cinomas. The small remaining group consisted of anaplasticcarcinomas, adenosquamous carcinomas and squamous cellcarcinomas (Deerberg & Kaspareit 1987). This powerfulstudy was set up with 50 animals per experimental group


initially. In parallel experimental groups it was demonstratedthat neither maintenance of animals in a germ-free environ-ment nor feeding of a purified diet reduced the incidence rateof tumors. The group of retired breeders also exhibited anincidence rate of > 90% but was characterized by a signifi-cantly longer life span. This finding is a clear indication ofhormonal involvement in the carcinogenic process. Numbersof pregnancies are regarded as a life style factor inverselycorrelated to endometrial carcinogenesis in humans (Emonset al. 2000 and references therein). But even more impor-tantly, if BDII/Han rats are ovariectomized prior to estrouscyclicity, they die from causes other than endometrial


Figure 2 The EnDA/RUCA model. This figure shows the experimental possibilities of the syngenic EnDA/RUCA modelconsisting of spontaneous endometrial carcinogenesis of DA/Han rats, of the EnDA inoculation tumor and of cultured and/orinoculated RUCA-I cells.



carcinoma, the incidence rate for this tumor decreasing to0%. This is the strongest piece of evidence in favor of hor-mone-dependent endometrial carcinogenesis in BDII/Hanrats (Deerberg & Kaspareit 1987). This latter finding wasfurther substantiated in a more recent study in which life longtreatment with the progestin, melengestrol acetate, in threedifferent doses (0.1, 0.2 and 0.4 mg/kg/day) completely sup-pressed endometrial carcinogenesis (Deerberg et al. 1995).At the highest dose glucocorticoid-like side effects wereobserved.

Only recently has the hormonal carcinogenesis of inbredBDII/Han rats been recognized as a model system to studygenes involved in endometrial carcinogenesis. Using them asmodels for the understanding of genetic elements involvedin the ontogeny of endometrial adenocarcinoma in general,genetic changes occurring in the course of endometrial car-cinogenesis could be demonstrated in crossbreeding experi-ments applying comparative genomic hybridization (Helou etal. 2001). The most common aberration was amplification ofthe proximal region of rat chromosome 4. The genes Cdk6(cyclin dependent kinase 6) and Met (hepatocyte growthfactor receptor) were found to be located in the core of eachamplified region and amplified most recurrently and at thehighest level among the genes tested (Walentinsson et al.2001). These data suggested that up-regulation of Cdk6 and/or Met contributes to the development of endometrial cancersin BDII/Han rats. The human homolog of Cdk6 has beenpostulated to be an important player in cell cycle control.Evidence exists that Cdk6 provides the link between growthfactor stimulation and onset of cell cycle progression(Meyerson & Harlow 1994). However, the only documen-tation of Cdk6 overexpression in association with Cdk6amplification comes from human gliomas (Costello et al.1997). The Met protoocogene encodes a transmembranegrowth factor which binds the cytokine hepatocyte growthfactor/scatter factor (HGF/SF). The highest levels of thistyrosine kinase receptor are found in epithelial tissues. Theligand, in contrast, is predominantly expressed in mesenchy-mal tissues suggesting a paracrine mode of action(Gherardi & Stoker 1991). The major effect of HGF/SF func-tion mediated by Met is thought to be epithelial cell prolifer-ation and motility (Sugawara et al. 1997), including endo-metrial epithelium as well (Wagatsuma et al. 1998). Further,it has been shown that HGF/SF can stimulate invasion ofendometrial adenocarcinoma cell lines if they express MET(Bae-Jump et al. 1999).

Extra copies of chromosomal regions were found forchromosome 6 and eight cancer-related genes were predictedto be located in this chromosomal region of BDII/Han rats.These genes comprised the N-myc protooncogene, apolipo-protein B, the DEAD box gene, ornithine decarboxylase, pro-opiomelanocortin, ribonucleotide reductase, M2 polypeptideand syndecan. Amplification of N-myc was by far the highestsuggesting that Mycn amplification and overexpression con-


tributes to the development of this hormone-dependent tumor(Karlsson et al. 2001). In addition, three major chromosomalregions representing multiple susceptibility genes involved inthe development of endometrial adenocarcinoma in rat couldbe identified (Roshani et al. 2001). Loss of heterozygosityanalysis revealed three major losses of heterozygosityregions on chromosome 10 (Behboudi et al. 2001). By radi-ation hybrid mapping and single and dual color fluorescencein situ hybridization techniques a detailed chromosomal mapof the proximal part of chromosome 10 could be established.With this approach the regional localization of 14 genes,most of them cancer related, and of 5 microsatellite markerscould be determined (Behboudi et al. 2002).

Finally, in crossbreeding experiments of female BDII/Han rats with males of two strains with a low incidence ofendometrial adenocarcinoma, three chromosomal regions,which give rise to susceptibility for endometrial adenocarcin-oma, could be identified (Roshani et al. 2001). This is a clearindication that endometrial carcinogenesis of BDII/Han ratsexhibits heritable features. Interestingly, the genes affectingsusceptibility to endometrial adenocarcinoma were differentin the two crosses, suggesting that genes behind the suscepti-bility in BDII/Han animals may interact with various genesin different genetic backgrounds. Data in the literature sug-gest that this feature of heritability of endometrial carcino-genesis in BDII/Han rats is shared by at least two forms ofhuman endometrial cancer (Sandles et al. 1992, Gruber &Thompson 1996, Lynch et al. 1996). Whether or not thisheritability of endometrial carcinogenesis resembles the heri-table phenotype in human endometrial adenocarcinoma ofpatients with hereditary non-polyposis colorectal cancer(HNPCC) remains open. The latter phenotype would, at least,require microsatellite instability, mutation or alteration ofpromotor methylation of mismatch repair genes andmutations of the pten locus (Kuismanen et al. 2002, Whelanet al. 2002, Zhou et al. 2002a,b). A corresponding geneticanalysis of endometrial adenocarcinoma of BDII/Han rats isstill missing.

Mouse models

Mouse models historically have proved to be particularlyuseful in studying the effects of developmental exposure tohormones, particularly estrogens. This topic will be discussedin more detail below. In addition, the ICR mouse modelwhich responds to chemical carcinogens by induction ofendometrial cancer is described.

Transgenic mouse models are emerging tools for car-cinogenesis in general. In the context of endometrial carcino-genesis pten + / − transgenic mice have to be discussed. Theseanimals exhibit a phenotype strongly resembling humanCowden syndrome, a syndrome associated with high inci-dence of breast and endometrial neoplasia (Stambolic et al.


2000). This particular mouse strain, therefore, has to be dis-cussed as a potential endometrial cancer model.

Developmental hormone (estrogen) exposure

In various mouse and rat strains estrogens increase the inci-dence of tumorigenesis in several organs (Liehr 2001 andreferences therein), including the uterus, particularly duringdevelopmental exposure. In humans, epidemiologists haverecognized prolonged estrogen action or estrogen use unop-posed by gestagens as a risk factor for endometrial carcino-genesis (Key & Pike 1988, Weiderpass et al. 1999). From allthese data the question has been raised whether estrogens actas carcinogens (Boyd 1996). From mechanistic and molecu-lar studies more and more pieces of evidence have accumula-ted to strengthen this hypothesis (reviewed by Liehr 2000,2001).

The neonatal mouse has been proposed as a model forhormonal carcinogenesis of the endometrium (Newbold et al.1990). If outbred mice are treated neonatally (days 1–5) theywill develop endometrial adenocarcinoma in a dose-dependent manner and dependent on the strength of the estro-gen. In the initial paper, Newbold et al. (1990) showed thatdiethylstilbestrol and derivatives thereof are more potent thanestradiol. This model has been further characterized in termsof cellular differentiation and gene expression in the epi-thelium (Yoshida et al. 2000). In the meantime, this modelhas been successfully applied to test for the developmentaltoxicity of tamoxifen (Newbold et al. 1997), catecholestrog-ens (Newbold & Liehr 2000) and genistein (Newbold et al.2001). All these substances induce endometrial adenocarcin-oma if injected in neonatal mice during postnatal days 1–5. Whether this model can be replaced by transgenic miceoverexpressing ER and which exhibit an accelerated onset ofendometrial carcinogenesis (Couse et al. 1997) remains to beelucidated, because these tumors are characterized by a moreaggressive phenotype and may therefore be less representa-tive of the human situation than endometrial carcinogenesisin wild-type animals.

In summary, CD-1 mice appear to be an excellent modelto test for developmental toxicity of estrogens, thereby evalu-ating the capacity of test substances to induce hormone-dependent endometrial cancers.

Chemical carcinogen-inducedendometrial cancers in ICR mice

ICR mice respond to treatment with N-methyl-N-nitrosourea(NMU) or estradiol with the development of endometrialcancers within 30 weeks. This process is significantly acceler-ated if ICR mice are treated with both agents simultaneously(Niwa et al. 1991). The enhancing effects on endometrialcarcinogenesis initiated by NMU are not only detectable forestradiol but also for estrone and estriol (Niwa et al. 1993).


In comparison to human endometrial carcinogenesis, his-topathological examinations revealed that these tumorsdevelop from various preneoplastic, hyperplastic lesions,resembling the human situation (Niwa et al. 1991). Interest-ingly, the carcinogen and estradiol induce different preneo-plastic lesions. Estradiol induces glandular hyperplasia, whileadenomatous hyperplasia is induced by NMU. For the devel-opment of atypical hyperplasia the cooperative action ofestradiol and NMU is favorable (Niwa et al. 1996).

Whereas the histopathology of NMU-induced endomet-rial tumors in ICR mice resembles the situation describedfor endometrial carcinogenesis, the limited genetic analysisdescribed for the mouse model is different to human endo-metrial carcinogenesis. In specimens of human endometrialcancer the mutation of various Ras loci (Boyd & Risinger1991, Ignar-Trowbridge et al. 1992) and the p53 gene(Risinger et al. 1992) are found quite often. These kinds ofmutations are very rare in NMU-initiated endometrial tumorsin ICR mice (Murase et al. 1995).

With regard to steroid receptor expression and proteinlevels this model is poorly characterized. In a short-time pro-tocol in ovariectomized ICR mice, animals respond to estro-gen treatment by up-regulation of steady state mRNA levelsof the proto-oncogenes fos and jun (Niwa et al. 1998).

The NMU-induced and estradiol-enhanced endometrialcarcinogenesis proved to be particularly useful for treatmentstudies. Several hormonal and traditional treatment pro-cedures were tested leading with one exception to inhibitionor prevention of all tumors in the ICR model. Carcinogenesisinduced by a combination of estradiol and NMU wasinhibited by gestagen treatment (medroxyprogesterone acet-ate; Niwa et al. 1995) or by treatment with the antiestrogen,toremifen (Niwa et al. 2002). In contrast, treatment with theselective estrogen receptor modulator, tamoxifen, stimulatedendometrial carcinogenesis in NMU-treated animals with orwithout the combined treatment with estradiol (Niwa et al.1998).

Prevention of endometrial carcinogenesis in the abovedescribed mouse model including a reduction of hyperplasticfoci was detected following treatment of animals with dana-zol (Niwa et al. 2000) or following treatment with traditionalJapanese or Chinese remedies such as extracts from Gly-cyrrhiza radix (Niwa et al. 1999) and Juzen-taiho-to (Niwaet al. 2001). Shimotsu-to was identified as the active prin-ciple in Juzen-taiho-to (Lian et al. 2002). Finally, preventionstudies with isoflavones revealed that both genistein anddaidzein exhibited an inhibitory effect on endometrial car-cinogensis in the estradiol/NMU ICR mouse model, particu-larly on the indices of atypical endometrial hyperplasia (Lianet al. 2001).

In summary, NMU-induced endometrial adenocarcin-omas in ICR mice are particularly sensitive to various hor-monal treatment procedures. The limitation of this model isits poor characterization of hormone responsiveness at a


molecular level. Neither the levels nor the qualities ofexpressed steroid hormone receptors are known nor is thereany information on hormonally triggered signal transductioncascades or responsive genes except for fos and jun.

Heterozygous mouse models with pten +/−

-mice as an example

PTEN/MMAC1/TEP1 has been detected as one of the mostcommonly mutated tumor suppressor genes in human cancer(Li et al. 1997, Steck et al. 1997). A significant rate of PTENmutations has also been reported for human endometrial aden-ocarcinoma (Risinger et al. 1997, 1998, Tashiro et al. 1997).This mutation has been found in up to 80% of cases of humanendometrial adenocarcinoma, and is already detectable in 33–55% of the precancerous lesions (Latta & Chapman 2002,Konopka et al. 2002). In addition to a mutation promoter,hypermethylation is an alternative way to inactivate tumorsuppressor genes. There is one report in the literature showingthat the frequency of promoter hypermethylation of the PTENtumor suppressor gene in endometrial adenocarcinoma isaround 19% (Salvesen et al. 2001).

Although the importance of a functional PTEN proteinhas been known for several years, the association of PTENexpression with a prognosis is not yet clear. It appears as ifloss of PTEN expression is associated with metastatic disease(Salvesen et al. 2002) and may serve as an independent prog-nostic marker for patients who undergo postoperative chemo-therapy (Kanamori et al. 2002).

In order to understand the biological role of this domi-nant tumor suppressor gene, transgenic PTEN knock-outmice have been created. A PTEN − / − mutation is lethal pre-sumably due to defective chorio-allantoic development(Suzuki et al. 1998).

Surprisingly, the mutation of one allele is sufficient tocause neoplasia in multiple organ systems in these pten + / −

-mice (Podsypanina et al. 1999). A particularly high inci-dence has also been detected in breast as well as in endomet-rial neoplasia and preneoplasia, strongly resembling humanCowden syndrome (Stambolic et al. 2000). Indeed, in thehuman situation and in addition to the occurrence of PTENmutations in sporadic tumor, germ-line mutations arebelieved to cause related autosomal dominant hamartomasyndromes, among them Cowden syndrome (Liaw et al.1997, Eng 1998). Although the phenotype detectable inpten + / − -mice in terms of the development of Cowden syn-drome and the association with the high incidence of devel-opment of precancerous and cancerous lesions of breast andendometrium is strikingly similar to the human situation,some differences do exist. Unlike the human situation, inpten + / − -mice neoplasia of the skin and brain were notablyabsent, whereas the observed changes in the endometriumwere very consistent (Podsypanina et al. 1999).


Endometrial carcinoma, except colon carcinoma itself, isthe most common malignancy in patients with hereditarynon-polyposis colon cancer (HNPCC). This disease is char-acterized by germ-line mutations in mismatch repair genesand by microsatellite instability (Fishel et al. 1993, Leach etal. 1993, Peltomaki et al. 1993, Bronner et al. 1994, Nico-laides et al. 1994, Papadopoulos et al. 1994). Surprisingly,patterns of microsatellite instability differ between endomet-rial and colorectal tumors from patients with HNPCC(Kuismanen et al. 2002).

To learn more about the involvement of these genes inthe pathogenesis of endometrial carcinogenesis, tumor speci-mens from patients with HNPCC and an established mutationrecord of either the MLH1 or MSH2 missmatch repair genelocus were subjected to analysis of the PTEN locus. Mutationof the mismatch repair loci and PTEN gene were verycommon in HNPCC, with PTEN mutation presumably pre-ceding mutations of the mismatch repair loci (Zhou et al.2002b). Interestingly, crosses of pten + / − - with mlh1 − /

− -mice were characterized by an accelerated endometrialtumorigenesis (Wang et al. 2002).

In summary, endometrial cancers from pten + / − -micewith or without mlh1 − / − may serve as an excellent modelfor hereditary endometrial cancer, because of the strongresemblance with human Cowden disease and humanHNPCC. In addition, Stambolic et al. (2000) proposedpten + / − -mice as a model for endometrial adenocarcinomaswhich develop in women with unopposed estrogen stimula-tion. These patients rather commonly suffer from loss of het-erozygosity at the PTEN locus and/or mutations in all stagesof endometrial hyperplasia (Risinger et al. 1997, 1998,Tashiro et al. 1997).

Whether pten + / − -mice can serve as a hormone-dependent cancer model or if they are suitable as models forhormonal treatment protocols remains open, because no dataare available on these issues. However, despite the lack ofthe latter information it has to be stated that pten + / − -micerepresent a promising new endometrial cancer model,because of the similarity to hereditary human endometrialcancer.

Inoculation tumors

These tumors are derived by inoculating chunks of tumors ordefined numbers of cultured cells e.g. subcutaneously or intothe fat pad of syngenic animals, athymic nude mice or rats.It is the scope of this paper to focus primarily on steroidhormone dependent or at least steroid hormone sensitive invivo models. For this reason in the following sections evi-dence for hormone responsiveness from in vitro data isbriefly summarized prior to the review of in vivo applicationdata of either model.


The RUCA-I/EnDA model

From a spontaneous endometrial adenocarcinoma a trans-plantable tumor called EnDA-tumor (Horn et al. 1993) anda cell line called RUCA-I (Schutze et al. 1992) have beenderived. These models have primarily been used to test novelantiestrogens, potential agonistic activities of phytoestrogensand o,p′-DDT as an example of an endocrine-disruptingchemical.

To test for antiestrogenic activities of antiestrogenschunks of EnDA-tumors or RUCA-I cells have been inocu-lated into syngenic DA/Han rats or athymic nude rats (Fig.2; Horn et al. 1993, Vollmer & Schneider 1996). Tumorgrowth from chunks of EnDA-tumors at the ectopic site aswell as formation of lymphogenic and pulmonary metastasisare estrogen-sensitive processes, because the tumor growthrate at the ectopic site, as well as the weight of ipsilateralaxillary lymph nodes and the number of lung metastases isreduced in ovariectomized animals if compared with intactanimals. These effects most likely are mediated by the ERα,which unlike the progesterone receptor is expressed in thesetumors. This experimental model has been successfullyapplied to study, for example, the antiestrogenic activity andfunction of ZK 119010 in an endometrial-derived experimen-tal tumor model (Horn et al. 1993, 1994). As indicated abovean endometrial adenocarcinoma cell line, the RUCA-I cellline, has been established from this transplantable EnDA-tumor. This cell line is characterized by the expression of theERα and by its estrogen responsiveness in vitro and in vivo(Fig. 2). If RUCA-I cells are cultured on reconstituted base-ment membrane (matrigel) they respond to treatment withestradiol by stimulation of proliferation and alteration of geneexpression e.g. by the upregulation of complement C3expression (Vollmer et al. 1995a) which is the major estro-gen responsive gene in the rat uterus (Sundstrom et al. 1989),as well as by down-regulation of fibronectin expression(Vollmer et al. 1995b) and up-regulation of clusterin geneexpression (Wunsche et al. 1998). In vivo, if inoculated sub-cutaneously into the hind limb of syngenic DA/Han ratsRUCA-I cells form estrogen-sensitive metastasizing adeno-carcinomas which by histological and biochemical means areindistinguishable from tumors arising from chunks of EnDA-tumors. Tumor growth and metastasis in both experimentalsetups are very fast. After approximately 30–35 days animalsare clinically ill from the metastatic disease if 0.5–1 × 106

cells are inoculated at the ectopic site (Horn et al. 1994,Vollmer & Schneider 1996).

In this transplantation approachRUCA-I cells at the inocu-lation site as well as metastases of these cells at lymphogenicsites and in the lung react very sensitively to antiestrogen treat-ment. The weight of the primary tumor at the ectopic site, theweight of the axillary ipsilateral lymph nodes as well as thenumber of lung metastases were found to be reduced compared


with control animals in response to antiestrogen treatment ofinoculated animals (Vollmer & Schneider 1996). Inectopically grown tumors, in lymph node metastases and in thenormal uterus (the latter organ was analyzed for controlpurposes) expression of estrogen-dependent genes was down-regulated significantly following treatment of animals with thepure antiestrogen ICI 182,780 (Wunsche et al. 1998), amongstthese genes were matrix metalloproteinases (Tushaus L,Hopert AC & Vollmer G, unpublished observations).

Inoculation of RUCA-I cells into ovariectomized DA/Han rats for the purpose of screening for estrogenic proper-ties of phytoestrogens and endocrine-disrupting chemicals inan endometrial-derived tumor model led to unexpectedresults. Although tumor weight increased following oralapplication of ethinyl estradiol in a 28-day treatment proto-col, the weight of the tumor tissue remained almost unaffec-ted following treatment of animals with a selected phytoes-trogen, genistein (Diel et al. 2001) or a representativeindustrial chemical o,p′-DDT (Diel et al., in preparation).Unlike ethinyl estradiol-treated animals, there was no sig-nificant alteration of gene expression detectable in tumor tis-sues of animals treated with genistein or o,p′-DDT. Controlinvestigations in the normal rat uterus showed that this organresponded very sensitively to treatment with ethinyl estra-diol, genistein and o,p′-DDT, both with an increase in tissueweight as well as up-regulation of estrogen-dependent geneexpression (Diel et al. 2001).

In conclusion, RUCA-I cells if inoculated into syngenicDA/Han rats represent a very sensitive estrogen-responsivetumor model for the purpose of antiestrogen screening andtesting in an endometrial-derived tumor model. The RUCA-Icell-derived model is of limited value for testing of agonisticestrogenic properties of e.g. suspected phytoestrogens orendocrine-disrupting chemicals. Experimentally, the EnDA/RUCA-I model shows a unique feature: it is possible to workin syngenic animals which means it is not necessary to useathymic nude mice or rats having an imperfect immunesystem which have to be kept in germ-free surroundings.

Ishikawa cells

Hormonal responsiveness of Ishikawa cellsin vitro

Since their first description (Nishida et al. 1985) Ishikawacells, a human endometrial adenocarcinoma cell line express-ing the estrogen and the progesterone receptors, are the mostwidespread human endometrial-derived cell culture model.Later on, applying the limited dilution method eighteen dif-ferent clones of Ishikawa cells were established (Nishida etal. 1996), fifteen of them being estrogen receptor positive.However, in terms of population doubling time, plating effi-ciency or saturation density there was no significant


difference among the clones (Nishida et al. 1996). In additionto the expression of functional estrogen receptors (Holinkaet al. 1986a, b) and progesterone receptors (Lessey et al.1996), Ishikawa cells express the androgen receptor (Lovelyet al. 2000) and the receptor for aryl hydrocarbons (Wormkeet al. 2000). There exists a large body of evidence for thehormonal responsiveness of Ishikawa cells in vitro. In sum-mary, the evidence for hormonal responsiveness is basedmainly on hormonal modification of proliferation, cellularfunctions or gene expression (for summary see Table 1 andreferences therein).

In vitro the cells are in use for the elucidation ofmolecular mechanisms of hormone action e.g. in drugdevelopment, in the drug discovery process, for testing ofpotential agonistic functions of antiestrogens or selectiveestrogen receptor modulators in an endometrial-derivedmodel (Labrie et al. 2001), in studies of ligand independentactivation of the estrogen receptor, in anchorage indepen-dent tumor growth (Holinka et al. 1989), in studies onfactors controlling hormonal receptivity (Appa Rao et al.2001), in environmental toxicology studies on the functionof phytoestrogens (Markiewicz et al. 1993, Liu et al. 2001,Frigo et al. 2002) and endocrine-disrupting chemicals(Bergeron et al. 1999) in an endometrial model, in parac-rine cell/cell-interaction studies (Yang et al. 2001, Arnoldet al. 2002), in studies on signaling cross-talk(Ignar-Trowbridge et al. 1995, Wormke et al. 2000), andothers (Kanishi et al. 2000).

Ishikawa cells as an in vivo tumor model

Because of their hormone responsiveness in vitro Ishikawaendometrial adenocarcinoma cells have also been developedas an estrogen sensitive in vivo tumor model (Nishida et al.1986). As mentioned above, eighteen subclones were iso-lated, all of these subclones were found being transplantableinto athymic nude mice (Nishida et al. 1996). This model hasmainly been used for two purposes: (1) to elucidate generalmechanisms in tumor biology, particularly towards theunderstanding of estrogen and progesterone function in endo-metrial cancer and (2) as an endometrial model for hormonaltreatment of endometrial cancer.

The Ishikawa endometrial adenocarcinoma model provedto be useful for studies on the regulation of growth controlby steroid hormones in endometrial adenocarcinoma in vivo(Gong et al. 1994). The most important result of this studywas that 4-hydroxytamoxifen, like estradiol, stimulates endo-metrial tumor growth in vivo. This effect most likelycoincides with alteration of levels of expression of trans-forming growth factor (TGF)-α and TGF-β. In a similarapproach following inoculation of Ishikawa cells into the fatpad of nude mice it could be demonstrated that raloxifenexhibits a growth stimulatory potential in this assay(Barsalou et al. 2002).


Involvement of a switch in the angiogenic pathwaysduring tumorigenesis has become a recent focus of interest.The participation of the ERα in this process was studied inIshikawa cells following overexpression of the ERα protein.Overexpression of ERα not only significantly inhibited thegrowth of xenografted cells, but also down-regulated vascu-lar endothelial growth factor expression in tumor xenografts,resulting in a decreased vascularization of the tumors and theinhibition of the angiogenic agent integrin αvβ3 (Ali et al.2000). These experimental findings provide evidence in favorof the assumption that high levels of ERα may be beneficialin the control of endometrial cancer due to its inhibitoryeffects on angiogenic pathways.

Finally, endometrial carcinoma tumorgenicity could besuppressed following introduction of chromosome 18 intoIshikawa cells (Yamada et al. 1995). This chromosome isknown to harbor the presumptive tumor suppressor geneDCC and therefore experimental evidence is provided for thehypothesis of DCC being a candidate for an endometrial car-cinoma tumor suppressor gene.

In summary, Ishikawa cells represent a combined invitro/in vivo human endometrial tumor model which is par-ticularly suitable for the study of hormonal growth control.In addition, it may be useful in characterizing organ speci-ficity of natural and synthetic estrogens, like phytoestrogensor endocrine-disrupting chemicals.

Enca101 tumor and ECC-1 cells

ECC-1 cells as an in vitro model

ECC-1 endometrial adenocarcinoma cells (Satyaswaroop &Tabibzadeh 1991) have been derived from a transplantableendometrial adenocarcinoma called EnCa101 which wasestablished immediately after the technique of tumor growthin athymic nude mice became available (Satyaswaroop et al.1981). Most importantly, ECC-1 cells form tumors withglandular structures if inoculated into athymic nude mice(Satyaswaroop & Tabibzadeh 1991).

The EnCA101 tumor and the ECC-1 cell line respondto estrogen treatment (Satyaswaroop et al. 1983) and areparticularly rich in progesterone receptors (Clarke &Satyaswaroop 1985). In addition, tumors respond totamoxifen treatment by stimulation of growth (Jordan et al.1991, Tonetti et al. 1998), a feature which from epidemio-logical studies is expected to increase the risk for sporadichuman endometrial cancer during breast cancer therapy withtamoxifen (Emons et al. 2000 and references therein). Thereare many pieces of experimental evidence proving theresponsiveness of this model to treatment with sex steroidhormones (for summary see Table 2). For these reasons theEnCA101/ECC-1 tumor is the most widely used in vivohuman endometrial adenocarcinoma model.


Table 1 Hormonal responsiveness of Ishikawa cells

Receptor Ligand(s) Functional parameter Regulation Reference

ER E2 Proliferation Stimulation Holinka et al. (1986a,b)ER E2 Proliferation, colony formation Stimulation Holinka et al. (1989)ER 4-OHT Proliferation Stimulation Anzai et al. (1989)ER 4-OHT Proliferation Stimulation Croxtall et al. (1990)ER E2 Alkaline phosphatase Stimulation Holinka et al. (1986c)ER EM652, EM800 Alkaline phosphatase No regulation Labrie et al. (2001)ER E2 Migration through basement Stimulation Fujimoto et al. (1996b)

membraneER Resveratrol E2 induced alkaline phosphatase Down-regulation Bhat & Pezzuto (2001)

ERα Down-regulationERβ No regulation

ER E2, Tam Proliferation, plasminogen Up-regulation Hochner-Celnikier et al.activator (1997)

ER E2 Glycogen metabolism, gelatinase Up-regulationER Tam Glycogen metabolism, gelatinase Down-regulationER E2 Growth, cyclin D1, invasiveness, Up-regulation Mizumoto et al. (2002)

MMP-1, -7, -9ER E2 AF2 Stimulation Sakamoto et al. (2002)AR DHT, T Alkaline phosphatase Up-regulation Markiewicz & Gurpide (1997)PR P Growth Stimulation Holinka & Gurpide (1992)PR MPA Growth Suppression Shiozawa et al. (2001)

P27Kip Up-regulationPR P; E2 + P PAI-1 Up-regulation Fujimoto et al. (1996a)PR P Estrogen sulfotransferase Up-regulation Falany & Falany (1996)PR P α1ß1 integrin Up-regulation Lessey et al. (1996),

Castelbaum et al. (1997)PR MPA Sex hormone binding globulin Upregulation Misao et al. (1998)PR MPA 8 products from differential Up-/down-regulation Sakata et al. (1998)

display RT-PCR (gene dependent)PR P VEGFs Down-regulation of E2 Fujimoto et al. (1999)

induced inductionPR P Osteopontin Up-regulation Appa Rao et al. (2001)

Abbreviations: ER, estrogen receptor; E2, estradiol; 4-OHT, 4-hydroxytamoxifen; Tam, tamoxifen; DHT, dihydrotestosterone; T,testosterone; P, progesterone; MPA, medroxyprogesterone acetate; MMP, matrix metalloproteinase; PAI, plasminogen activatorinhibitor; VEGF, vascular endothelial growth factor.

EnCa101 and ECC-1 cells-derived tumors

In terms of hormonal regulation of tumor growth, inoculationof EnCa101 tumors or ECC-1 endometrial adenocarcinomacells into nude mice appears to be the most complete model.It is responsive to estrogens and progestins and it grows fol-lowing tamoxifen stimulation. In addition, the EnCA101tumors exhibit another feature which is highly advantageousin experimental cancer research: it can be used as a multi-sitetransplantation model thereby saving animals (Heitjan et al.2002). In these tumors the most interesting features of hor-monal influences on endometrial tumor growth can bestudied, e.g. testing of novel estrogen receptor antagonists,treatment of tamoxifen-stimulated tumor growth – a majorconcern as indicated above (ACOG Committee Opinion1996, Love et al. 1999) – and characterizing the effects ofprogestins, these being the adjuvant therapy for endometrialcarcinoma for decades (Emons et al. 2000 and referencestherein).


Once the estrogen dependency of tumor growth ofEnCa101 tumors (Satyaswaroop et al. 1983, Jordan et al.1991) and their estrogen-responsiveness as measured by geneexpression, particularly of progesterone receptor (Clarke etal. 1987) and of c-fos, (Sakakibara et al. 1992) had beenestablished, in a first series of larger experiments the effectsof antiestrogens and progestins on the growth of these tumorswere investigated. In fact, for progestins schedules for proge-stin administration were designed in this model (Mortel et al.1990).

For antiestrogens it could be shown that almost all anti-estrogens tested stimulated tumor growth although to a lesserdegree than tamoxifen (Gottardis et al. 1990) with the excep-tion of the pure antagonist ICI 164,384 which was not onlyvoid of tumor growth stimulatory activity but also blockedtamoxifen-induced tumor growth.

During combination treatments with tamoxifen and pro-gestins, resistance of EnCA101 tumors to treatment becameapparent. This resistance was attributed to a desensitization


Table 2 Hormonal responsiveness of EnCa101 tumors and ECC-1 cells

Cell line/ Receptor Ligand Functional parameter Regulation Referencetumor

EnCa101 ER E2 Growth in vivo Increase Satyaswaroop et al. (1993)EnCa101 ER Tam Growth in vivo Increase Jordan et al. (1991)EnCa101 ER E2/Tam Growth in vivo Increase Jordan et al. (1989)ECC-1 ER E2 Growth in vivo Increase Dardes et al. (2002a)

GW5638 Growth in vivo No effectECC-1 ER E2 Growth in vivo Stimulation Dardes et al. (2002b)

Tam, Ral Growth in vivo No effectECC-1 ER E2 Proliferation in vitro Stimulation Bergeron et al. (1999), Castro-Rivera

et al. (1999), Dardes et al. (2002b)EnCa101 ER E2 Progesterone receptor Stimulation Clarke et al. (1987)ECC-1 ER E2, BPA Progesterone receptor Stimulation Bergeron et al. (1999)EnCa101 ER E2, Tam c-fos Induction Sakakibara et al. (1992)ECC-1 ER E2 Cathepsin D Induction Castro-Rivera et al. 1999ECC-1 ER E2 GREB1 Induction Ghosh et al. (2000)ECC-1 ER E2, Ral ER-α Down-regulation Dardes et al. (2002b)

ICI182,780 ER-α DegradationE2 pS2, VEGF InductionE2 HER2/neu Down-regulation

Abbreviations: ER, estrogen receptor; E2, estradiol; Ral, raloxifen; Tam, tamoxifen; BPA, bisphenol A; VEGF, vascularendothelial growth factor; pS2, presenelin-2.

following down-regulation of the progesterone receptor(Satyaswaroop et al. 1992). Based on the results of anotherstudy in which progesterone levels, progesterone receptorlevels and rates of tumor growth were assessed, it was pre-dicted that intermittent progestin administration may resultin better control of endometrial cancer growth in the nudemouse model (Mortel et al. 1990).

The EnCa101 tumor model is characterized by anotherfeature – the estrogen-inducible expression of aromatase.Aromatase expression has been reported to be up-regulatedin endometrial adenocarcinoma, particularly in the stromaltissue compartment (Watanabe et al. 1995, Sasano et al.1996). Analyses of aromatase (Cyp19) gene polymorphismrevealed that these polymorphisms might represent one gen-etic risk factor for endometrial cancer development (Bersteinet al. 2001). Despite the clear up-regulation of aromataseactivity in endometrial cancer, the use of aromatase inhibitorsin a clinical setting has not been exhaustively investigatedand the clinical outcome is questionable (Rose et al. 2000,Berstein et al. 2002).

Estrogen-inducible aromatase activity, however, isinteresting for general tumor biology because the role ofandrogens in the growth of endometrial carcinoma is poorlyunderstood. The question of androgen sensitive endometrialtumor growth was addressed in this particular tumor model.There was little detectable growth promoting activity ofandrogens which was lower by far than that of estrogens.Surprisingly, there was no difference in the effects ofandrogens on the growth of EnCa101 tumors if the effects ofaromatizable and non-aromatizable androgens were com-pared (Legro et al. 2001).


In summary, the EnCa101/ECC-1 tumor model is themost complete endometrial tumor model because the effectsof estrogens, progestins and androgens can be assessed dueto the stable and, in part, estrogen inducible expression of theestrogen receptor, the progesterone receptor and aromatase.These concluding remarks have recently been substantiatedby studies on the characterization of some novel antiestro-gens: the tamoxifen analog GW5638 and the piperidinderivative ERA-923. These substances, like tamoxifen, blockgrowth of breast cancer cells and, in addition, more effec-tively block EnCa101 endometrial tumor growth(Greenberger et al. 2001, Dardes et al. 2002a).

Differentiation of endometrialadenocarcinoma cells in vitro

Endometrial adenocarcinomas differ in their ability to formglandular structures; however most of the sporadic occurringendometrial adenocarcinomas are well to moderately welldifferentiated. In conventional cell culture on plastic in thepresence of charcoal stripped fetal calf serum or in the pres-ence of a serum-free defined medium, endometrial adenocar-cinoma cells acquire a polygonal cell shape with a lowdegree of differentiation, and in a more narrow view cannotbe regarded as correlated to the relatively high differentiatedstatus of endometrial adenocarcinoma cells in vivo.

For cells derived from a variety of organs, particularlyfor normal and malignant mammary cells, a huge body ofevidence has accumulated (for review see Hansen & Bissell2000) that culturing of glandular cells on reconstituted


Tüshaus L, Holpert A C & Vollmer G, in preparationProteases (MMP2, MMP13)

Strunck et al. (2001)Expression of phosphatases (L3-PSP)

Treeck et al. (1998)icb-1 expression (differentiation marker)

Strunck E & Vollmer G, unpublished observationsMitochondrial gene expression

Pastor U, Strunck E & Vollmer G, unpublished observationsExpression of cytokeratins

Strunck and Vollmer (1996)Expression of integrins

Strunck et al. (1996)Gene expression in general

Vollmer et al. (1995a,b), Wünsche et al. (1998)Hormone responsiveness

Tan et al. (1999)Proliferation

Satyaswaroop & Tabibzadeh. (1991), Hopfer et al. (1996), Behrens et al.(1996), Pinelli et al. (1998)

Cellular morphology and ultrastructure

ReferenceProcesses regulated by basement membrane

Adenocarcinoma cell(dedifferentiated)

Alteration ofgene expression

basementmembrane

Adenocarcinoma cell onbasement membrane

(differentiated)

Stromalmatrix

TN-C rich

Figure 3 Responses of endometrial adenocarcinoma cells to contact with reconstituted basement membrane. The upper part isa schematic drawing of the situation of an endometrial adenocarcinoma cell grown in conventional cell culture on plastic(dedifferentiated) and on reconstituted basement membrane (differentiated). This figure also indicates the presence of thestromal matrix underneath the basement membrane, which in cases of endometrial adenocarcinoma is particularly rich intenascin-C (TN-C). The lower part shows the processes regulated by the basement membrane. icb-1, induced by contact tobasement membrane; L3-PSP, L3-phosphoserine phosphatase; MMP, matrix metalloproteinase.

basement membrane induces differentiation processes,thereby increasing physiological responsiveness considerablyif compared with cells kept under conventional cell cultureconditions. The evidence for morphological and functionaldifferentiation of human endometrial adenocarcinoma cellsof Ishikawa (Pinelli et al. 1998), of the ECC-1 cell line(Satyaswaroop et al. 1991) and of the RUCA-I rat endomet-rial adenocarcinoma cells (Vollmer et al. 1995a) is summar-ized in Fig. 3. The most important observation is that themorphological differentiation strongly correlates with func-tional differentiation, the latter being most importantly evi-denced by the induction of hormone responsiveness.

With the culturing of endometrial adenocarcinoma cellson reconstituted basement membrane a more pronounced


physiological phenotype of these cells is achieved, allowingthe assessment of physiological in vivo functions in an invitro model. This assay can be modified into an in vitroinvasion assay for the assessment of hormonal influences ontumor cell invasion through basement membranes (Fujimotoet al. 1996b). Finally, a modification of this assay has beendescribed by reconstituting the natural situation even furtherand adding conditioned medium of endometrial stromal cellsor cultured endometrial stromal cells to the system. In thisway two important parameters in endometrial tumor growthcan be assessed: the contribution of paracrine cellular inter-actions to tumor growth and the hormonal regulation of cellsof the two tissue compartments in a single cell culture dish(Arnold et al. 2002).


Conclusions

There is definitely a need for the development of substancesfor the treatment and cure of endometrial cancer. In additionthere are various new drugs or phyto-remedies under devel-opment which are intended to be used in the treatment andprevention of breast cancer, for the treatment of menopausalsymptoms and for hormone replacement therapy. The effi-cacy of novel drugs targeting steroid receptors in endometrialcancers has to be evaluated and the safety of other endocrinemeasures on endometrial cancers or on endometrial carcino-genesis has to be assessed. For these purposes a relativelysmall number of animal models or combined in vivo/in vitrotumor models, based on human endometrial adenocarcinomacells, are available. However, these models alone or in com-bination cover all needs for drug development or testing ofdrug safety in relation to endometrial cancer. While modelsconsisting of human endometrial adenocarcinoma cells trans-planted to athymic nude mice are commonly used, the poten-tial of animal models consisting of spontaneous endometrialcarcinogenesis or transplantation of rat endometrial aden-ocarcinoma cells to syngenic animals is largely ignored. Firstof all, with these models treatment protocols for chronictreatment of both primary tumors and metastases can beestablished. And even if the inoculation approach is usedthere is still a considerable advantage because syngenic ani-mals can be used. Thus the use of the cost intensive athymicnude animal, a more artificial model which has to be keptin germ-free surroundings because of its imperfect immunesystem, can be avoided.

In the future, the number of studies using transgenic ani-mals and therapy approaches in these animals will definitelyincrease. The advantage of these tools is that the geneticmodification leading to the carcinogenic process is more pre-cisely defined. In this way single genes or correspondingsignal transduction pathways can be targeted in quite a selec-tive way.

Acknowledgements

The author is grateful to Dr Deerberg for giving authorizationfor the use of unpublished data on lung metastasis of endo-metrial adenocarcinoma in DA/Han rats.

The author wishes to thank Drs Susanne Starcke, Jan-nette Wober and Oliver Zierau for critical reading of themanuscript and for their helpful suggestions. In addition,special thanks are due to Dr Oliver Zierau for his assistancewith the art work.

This paper was supported by the Deutsche Forschungs-gemeinschaft Vo410/5-4, Vo410/6-1 and Vo410/6-3.

Dedication

This paper is dedicated to honour the work of the recentlydeceased Dr Friedrich Deerberg, formerly of the German


Institute for Laboratory Animals. Dr Friedrich Deerberg sys-tematically investigated spontaneous carcinogenesis in inbredlaboratory animals and developed experimental animalmodels useful for studying cancer treatment and prevention.Amongst these models are two spontaneous hormone-dependent endometrial cancer models which are exhaustivelydiscussed in this paper. Personally, the author is indebted toDr Deerberg for supplying him with the DA/Han and BD/Han rat endometrial cancer models and the cell lines devel-oped thereof.

References

ACOG Committee Opinion 1996 Tamoxifen and endometrialcancer. International Journal of Gynecology and Obstetrics 53197–199.

Ali SH, O’Donnell AL, Balu D, Pohl MB, Seyler MJ, Mohamed S,Mousa S & Dandona P (2000) Estrogen receptor-α in theinhibition of cancer growth and angiogenesis. Cancer Research60 7094–7098.

Ando-Lu J, Takahashi M, Imai S, Ishihara R, Kitamura T, IijimaT, Takano S, Nishiyama K, Suzuki K & Maekawa A 1994High-yield induction of uterine endometrial adenocarcinomas inDonryu rats by a single intra-uterine administration ofN-ethyl-N′-nitro-N-nitrosoguanidine via the vagina. JapaneseJournal of Cancer Research 85 789–793.

Ando-Lu J, Sasahara K, Nishiyama K, Takano S, Takahashi M,Yoshida M & Maekawa A 1998 Strain-differences inproliferative activity of uterine endometrial cells in Donryu andFischer 344 rats. Experimental and Toxicologic Pathology 50185–190.

Anzai Y, Holinka CF, Kuramoto H & Gurpide E 1989 Stimulatoryeffects of 4-hydroxytamoxifen on proliferation of humanendometrial adenocarcinoma cells (Ishikawa line). CancerResearch 49 2362–2365.

Appa Rao KB, Murray MJ, Fritz MA, Meyer WR, Chambers AF,Truong PR & Lessey BA 2001 Osteopontin and its receptorανβ3 integrin are coexpressed in the human endometrium duringmenstrual cycle but regulated differentially. Journal of ClinicalEndocrinology and Metabolism 86 4991–5000.

Arnold JT, Lessey BA, Seppala M & Kaufman DG 2002 Effect ofnormal endometrial stroma on growth and differentiation inIshikawa endometrial adenocarcinoma cells. Cancer Research 6279–88.

Bae-Jump V, Segreti EM & Vandermolen D 1999 Hepatocytegrowth factor (HGF) induces invasion of endometrial carcinomacell lines in vitro. Gynecological Oncology 73 265–272.

Balk JL, Whiteside DA, Naus G, DeFerrari E & Roberts JM 2002A pilot study of the effects of phytoestrogen supplementation onpostmenopausal endometrium. Journal of the Society forGynecological Investigation 9 238–242.

Barakat RR, Park RC, Grigsby PW, Muss HD & Norris HJ 1997Corpus: epithelial tumors. In Principles and Practice ofGynecologic Oncology, edn 2, pp 859–896. Eds WJ Hoskins,CA Perez & RC Young. Philadelphia: Lippincott-Raven.

Barsalou A, Dayan G, Anghel SI, Alaoui-Jamali M, Van de VeldeP & Mader S 2002 Growth-stimulatory and transcriptionactivation properties of raloxifene in human endometrialIshikawa cell. Molecular and Cellular Endocrinology 19065–73.


Behboudi A, Levan G, Hedrich HJ & Klinga-Levan K 2001 Highdensity marker loss of heterozygosity analysis of ratchromosome 10 in endometrial adenocarcinoma. GenesChromosomes Cancer 32 330–341.

Behboudi A, Roshani L, Kost-Alimova M, Sjostrand E,Montelius-Alatalo K, Rohme D, Klinga-Levan K & Stahl K2002 Detailed chromosomal and radiation hybrid mapping in theproximal part of rat chromosome 10 and gene order comparisonwith mouse and human. Mammalian Genome 13 302–309.

Behrens P, Meissner C, Hopfer H, Schumann J, Tan MI,Ellerbrake N, Strunck E & Vollmer G 1996 Laminin mediatesbasement membrane induced differentiation of HEC1Bendometrial adenocarcinoma cells. Biochemical and CellularBiology 74 875–886.

Bergeron RM, Thompson TB, Leonard LS, Pluta L & Gaido KW1999 Estrogenicity of bisphenol A in human endometrialcarcinoma cell line. Molecular and Cellular Endocrinology 150179–187.

Berstein LM, Imyanitov EN, Suspitsin EN, Grigoriev MY, SokolovEP, Togo A, Hanson KP, Poroshina TE, Vasiljev DA,Kovalevskij AY & Gamajunova VB 2001 CYP19 genepolymorphism in endometrial cancer patients. Journal of CancerResearch and Clinical Oncology 127 135–138.

Berstein L, Maximov S, Gershfeld E, Meshkova I, Gamajunova V,Tsyrlina E, Larionov A, Kovalevskij A & Vasilyev D 2002Neoadjuvant therapy of endometrial cancer with aromataseinhibitor letrozole: endocrine and clinical effects. EuropeanJournal of Gynecology and Reproductive Biology 105 161–165.

Bhat KPL & Pezzuto JM 2001 Resveratrol exhibits cytostatic andantiestrogenic properties with human endometrialadenocarcinoma (Ishikawa) cells. Cancer Research 61 6137–6144.

Bouros D, Papadakis K, Siafakas N & Fuller AF Jr 1996 Patternsof pulmonary metastasis from uterine cancer. Oncology 53 360–363.

Boyd J 1996 Estrogen as a carcinogen: the genetics and molecularbiology of human endometrial carcinoma. Progress in Clinicaland Biological Research 394 151–173.

Boyd J & Risinger JI 1991 Analysis of oncogene alterations inhuman endometrial carcinoma: prevalence of ras mutations.Molecular Carcinogenesis 4 189–195.

Bronner CE, Baker SM, Morrison PT, Warren G, Smith LG,Lescoe MK, Kane M, Earabino C, Lipford J & Lindblom A1994 Mutation in the DANN mismatch repair gene homologuehMLH1 is associated with hereditary non-polyposis coloncancer. Nature 368 258–261.

Burton JL & Wells M 2002 The effect of phytoestrogens on thefemale genital tract. Journal of Clinical Pathology 55 401–407.

Castelbaum AJ, Ying L, Somkuti SG, Sun J, Ilesami OA & LesseyBA 1997 Characterization of integrin expression in a welldifferentiated endometrial adenocarcinoma cell line (Ishikawa).Journal of Clinical Endocrinology and Metabolism 82 136–142.

Castro-Rivera E, Wormke M & Safe S 1999 Estrogen and arylhydrocarbon responsiveness of ECC1 endometrial cancer cells.Molecular and Cellular Endocrinology 150 11–21.

Clarke CL & Satyaswaroop PG 1985 Photoaffinity labeling of theprogesterone receptor from human endometrial carcinoma.Cancer Research 45 5417–5420.

Clarke CL, Feil PD & Satyaswaroop PG 1987 Progesteronereceptor regulation by 17beta-estradiol in human endometrialcarcinoma grown in nude mice. Endocrinology 121 1642–1648.

Cook AM, Lodge N & Blake P 1999 Stage IV endometrial


carcinoma: a 10 year review of patients. British Journal ofRadiology 72 485–488.

Costello JF, Plass C, Arap W, Chapman VM, Held WA, BergerMS, Su Huang HJ & Cavenee WK 1997 Cyclin-dependentkinase 6 (CDK6) amplification in human gliomas identifiedusing two dimensional separation of genomic DNA. CancerResearch 57 1250–1254.

Couse JF, Davis VL, Hanson RB, Jefferson WN, McLachlan JA,Bullock BC, Newbold RR & Korach KS 1997 Accelerated onsetof uterine tumors in transgenic mice with aberrant expression ofthe estrogen receptor after neonatal exposure to diethylstilbestrol.Molecular Carcinogenesis 19 236–242.

Creasman WT 1997 Endometrial cancer: incidence, prognosticfactors, diagnosis and treatment. Seminars in Oncology 24 SI140–SI 150.

Croxtall JD, Elder MG & White JO 1990 Hormonal control ofproliferation in the Ishikawa endometrial adenocarcinoma cellline. Journal of Steroid Biochemistry 35 665–669.

Dardes RC, O’Reagan RM, Gajdos C, Robinson SP, Bentrem D,De Los Reyes A & Jordan VC 2002a Effects of a new clinicallyrelevant antiestrogen (GW5638) related to tamoxifen on breastand endometrial cancer growth in vivo. Clinical CancerResearch 8 1995–2001.

Dardes RC, Schafer, JM, Pearce ST, Osipo C, Chen B & JordanVC 2002b Regulation of estrogen target genes and growth byselective estrogen-receptor modulators in endometrial cancercells. Gynecological Oncology 85 498–506.

Deerberg F & Kaspareit J 1987 Endometrial carcinoma in BDII/Han rats: model of a spontaneous hormone-dependent tumor.Journal of the National Cancer Institute 78 1245–1251.

Deerberg F, Rehm S & Pittermann W 1981 Uncommon frequencyof adenocarcinoma of the uterus in virgin Han:Wistar rats.Veterinary Pathology 18 707–713.

Deerberg F, Rehm S & Rapp KG 1985 Survival data andage-associated spontaneous diseases in two strains of rats.In The Contribution of Laboratory Animal Science to theWelfare of Man and Animals, pp 171–179. Eds J Archibald,J Ditchfield & HC Rowsell. Stuttgart and New York:Fisher-Verlag.

Deerberg F, Pohlmeyer G, Lorcher K & Petrow V 1995 Totalsuppression of spontaneous endometrial carcinoma in BDII/Hanrats by melengestrol acetate. Oncology 52 319–325.

Diel P, Smolnikar K, Schulz T, Laudenbach-Leschowski U,Michna H & Vollmer G 2001 Phytoestrogens andcarcinogenesis – differential effects of genistein in experimentalmodels of normal and malignant rat endometeium. HumanReproduction 16 997–1006.

Emons G, Fleckenstein G, Hinney B, Huschmand A & Heyl W2000 Hormonal interactions in endometrial cancer.Endocrine-Related Cancer 7 227–242.

Eng C 1998 Genetics of Cowden syndrome: through the lookingglass of oncology. International Journal of Oncology 12 701–710.

Falany JL & Falany CN 1996 Regulation of estrogensulfotransferase in human endometrial adenocarcinoma cells byprogesterone. Endocrinology 137 1395–1405.

Fishel R, Lescoe MK, Rao MR, Copeland NG, Jenkins NA, GarberJ, Kane M & Kolodner R 1993 The human mutator genehomolog MSH2 and its association with hereditary nonpolyposiscolon cancer. Cell 75 1215–1225.

Frigo DE, Duong BN, Melnik LI, Schief LS, Collins-Burrow BM,Pace DK, McLachlan JA & Burrow ME 2002 Flavonoid


phytochemicals regulate activator protein-1 signal transductionpathways in endometrial and kidney stable cell lines. Journal ofNutrition 132 1848–1853.

Fugh-Berman A & Kronenberg F 2001 Red clover (Trifoliumpratense) for menopausal women: current state of knowledge.Menopause 8 333–337.

Fujimoto J, Hori M, Ichigo S & Tamaya 1996a Sex steroidsregulate the expression of plasminogen activator inhibitor-1(PAI-1) and its mRNA in uterine endometrial cancer cells.Journal of Steroid Biochemistry and Molecular Biology 59 1–8.

Fujimoto J, Hori M, Ichigo S, Morishita S & Tamaya T 1996bEstrogen activates migration potential of endometrial cancercells through basement membrane. Tumour Biology 17 48–57.

Fujimoto J, Sakaguchi H, Hirose R, Ichigo S & Tamaya 1999Progestins suppress estrogen-induced expression of vascularendothelia growth factor (VEGF) subtypes in uterine endometrialcancer cells. Cancer Letters 141 63–71.

Gherardi E & Stoker M 1991 Hepatocyte growth factor-scatterfactor: mitogen, motogen and met. Cancer Cells 3 227–232.

Ghosh MG, Thompson DA & Weigel RJ 2000 PDZK1 andGREB1 are estrogen regulated genes expressed inhormone-responsive breast cancer. Cancer Research 60 6367–6375.

Gong Y, Murphy LC & Murphy LJ 1994 Hormonal regulation ofproliferation and transforming growth factors gene expression inhuman endometrial adenocarcinoma xenografts. Journal ofSteroid Biochemistry and Molecular Biology 50 13–19.

Gottardis MM, Ricchio ME, Satyaswaroop PG & Jordan VC 1990Effect of steroidal and nonsteroidal antiestrogens on the growthof tamoxifen-stimulated human endometrial carcinoma(EnCa101) in athymic mice. Cancer Research 50 3189–3192.

Greenberger LM, Annable T, Collins KI, Komm BS, LyttleCR, Miller CP, Satyaswaroop PG, Zhang Y & Frost P 2001A new antiestrogen, 2-(4-hydroxy-phenyl)-3-methyl-1-[4-(2-piperidin-1-yl-ethoxy)-benzyl]-1H-indol-5-ol hydrochloride(ERA-923), inhibits the growth of tamoxifen-sensitive and-resistant tumors and is devoid of uterotropic effects in mice andrats. Clinical Cancer Research 7 3166–3177.

Gruber SB & Thompson WD 1996 A population-based study ofendometrial cancer and familial risk in younger women. Cancerand steroid hormone study group. Cancer, Epidemiology,Biomarkers & Prevention 5 411–417.

Gusberg SB 1994 Estrogen and endometrial cancer: an epilogue ala recherche du temps perdu. Gynecology and Oncology 52 3–9.

Hansen RK & Bissell MJ 2000 Tissue architecture and breastcancer: the role of extracellular matrix and steroid hormones.Endocrine-Related Cancer 7 95–113.

Heitjan DF, Derr JA & Satyaswaroop PG 1992 The multi-sitetumour transplantation model for human endometrial carcinoma:a statistical evaluation. Cell Proliferation 25 193–203.

Helou K, Walentinsson A, Beckmann B, Johansson A, Hedrich HJ,Szpirer C, Klinga-Levan K & Levan G 2001 Analysis of geneticchanges in rat endometrial carcinomas by means of comparativegenomic hybridization. Cancer Genetics and Cytogenetics 127118–127.

Hochner-Celnikier D, Greenfield C, Finci-Yeheskel, Milwidsky A,Gutman A, Goldman-Wohl D, Yagel S & Mayer M 1997Tamoxifen exerts oestrogen-agonistic effects on proliferation andplasminogen activation, but not on gelatinase activity, glycogenmetabolism and p53 protein expression, in cultures ofoestrogen-responsive human endometrial adenocarcinoma cells.Molecular Human Reproduction 3 1019–1027.


Holinka CF & Gurpide E 1992 Growth-promoting effects ofprogesterone in human endometrial cancer cell line(Ishikawa-Var I). Journal of Steroid Biochemistry and MolecularBiology 43 635–640.

Holinka CF, Hata H, Kuramoto H & Gurpide E 1986a Responsesto estradiol in a human endometrial adenocarcinoma cell line(Ishikawa). Journal of Steroid Biochemistry 24 85–89.

Holinka CF, Hata H, Gravanis A, Kuramoto H & Grupide E 1986bEffects of estradiol on proliferation of endometrialadenocarcinoma cells (Ishikawa cells). Journal of SteroidBiochemistry 25 781–786.

Holinka CF, Hata H, Kuramoto H & Gurpide E 1986c Effects ofsteroid hormones and antisteroids on alkaline phosphataseactivity in human endometrial cancer cells (Ishikawa line).Cancer Research 46 2771–2774.

Holinka CF, Anzai Y, Hata H, Kimmel N, Kuramoto H & GurpideE 1989 Proliferation and responsiveness to estrogen of humanendometrial cancer cells under serum-free culture conditions.Cancer Research 49 32078–33301.

Hopfer H, Rinehart CA Jr, Kaufman DG & Vollmer G 1996Basement membrane induced differentiation of HEC-1B(L)endometrial adenocarcinoma cells affects both morphology andgene expression. Biochemical Cell Biology 74 165–177.

Horn DW, Vollmer G, Deerberg F & Schneider MR 1993 TheEnDA endometrial adenocarcinoma: an oestrogen-sensitive,metastasizing in vivo tumour model of the rat. Journal of CancerResearch and Clinical Oncology 119 450–456.

Horn DW, Vollmer G, von Angerer E & Schneider MR 1994Effect of the nonsteroidal antiestrogen ZK 119,010 on growthand metastasis of the EnDA endometrial carcinoma.International Journal of Cancer 58 426–429.

Ignar-Trowbridge D, Risinger JI, Dent GA, Kohler M, Berchuk A,McLachlan JA & Boyd J 1992 Mutations of the Ki-ras oncogenein endometrial carcinoma. American Journal of Obstetrics andGynecology 167 227–232.

Ignar-Trowbridge DM, Pimentel M, Teng CT, Korach KS &McLachlan JA 1995 Cross talk between peptide growth factorand estrogen receptor signaling systems. Environmental HealthPerspectives 103 (Suppl. 7) 35–38.

Jordan VC, Gottardis MM, Robinson SP & Friedl A 1989Immune-deficient animals to study ‘hormone-dependent’ breastand endometrial cancer. Journal of Steroid Biochemistry 34 169–176.

Jordan VC, Gottardis MM & Satyaswaroop PG 1991Tamoxifen-stimulated growth of human endometrial carcinoma.Annals of the New York Acadamy of Sciences 622 439–446.

Kanamori Y, Kigawa J, Itamochi H, Sultana H, Suzuki M, OhwadaM, Kanamura T, Sugiyama T, Kikuchi Y, Kita T, Fujiwara K &Terakawa N 2002 Pten expression is associated with prognosisfor patients with advanced endometrial carcinoma undergoingpostoperative chemotherapy. International Journal of Cancer100 686–689.

Kanishi Y, Kobayashi Y, Noda S, Ishizuka B & Saito B 2000Differential growth inhibitory effect of melatonin on twoendometrial cancer cell lines. Journal of Pineal Research 28227–233.

Karlsson A, Helou K, Walentinsson A, Hedrich HJ, Szpirer C &Levan G 2001 Amplification of Mycn, Ddx1, Rrm2 and Odc1 inrat uterine endometrial carcinomas. Genes Chromosomes Cancer31 345–356.

Key TJA & Pike MC 1988 The dose–effect relationship between‘unopposed’ estrogens and endometrial mitotic rate: its central


role in explaining and predicting endometrial cancer risk. BritishJournal of Cancer 57 202–212.

Kitamura T, Nishimura S, Sasahara K, Yoshida M, Ando J,Takahashi M, Shirai T & Maekawa A 1999 Transplacentaladministration of diethylstilbestrol (DES) causes lesions infemale reproductive organs of Donryu rats, includingendometrial neoplasia. Cancer Letters 141 219–228.

Kojima T, Tanaka T & Mori H 1994 Chemoprevention ofspontaneous endometrial cancer in female Donryu rats by dietaryindole-3-carbinol. Cancer Research 54 1446–1449.

Konopka B, Paszko Z, Janiec-Jankowska A & Golulda M 2002Assessment of the quality and frequency of mutations occurencein PTEN gene in endometrial carcinomas and hyperplasias.Cancer Letters 178 43–51.

Kuismanen SA, Moisio AL, Schweizer P, Truninger K, SalovaaraR, Arola J, Butzow R, Jiricny J, Nystrom-Lahti M & PeltomakiP 2002 Endometrial and colorectal tumors from patients withhereditary nonpolyposis colon cancer display different patternsof microsatellite instability. American Journal of Pathology 1601953–1958.

Labrie F, Labrie C, Belanger A, Simard J, Giguere V, TremblayA & Tremblay G 2001 EM-652 (SCH57068), a pure SERMhaving complete antiestrogenic activity in the mammary glandand endometrium. Journal of Steroid Biochemistry andMolecular Biology 79 213–225.

Latta E & Chapman WB 2002 PTEN mutations and evolvingconcepts in endometrial neoplasia. Current Opinion in Obstetricsand Gynecology 14 59–65.

Leach FS, Nicolaides NC, Wei YF, Liu B, Jen J, Parsons R,Peltomaki P, Sistonen P, Aaltonen LA & Nystrom-Lahti M 1993Mutation of a mutS homolog in hereditary nonpolyposis coloncancer. Cell 75 1215–1225.

Legro RS, Kunselman AR, Miller AS & Satyaswaroop PG 2001Role of androgens in the growth of endometrial carcinoma: an invivo animal model. American Journal of Obstetrics andGynecology 184 303–308.

Lessey BA, Ilesanmi AO, Castelbaum AJ, Somkuti SG, ChwaliszK & Satyaswaroop P 1996 Characterization of functionalprogesterone receptor in an endometrial adenocarcinoma cell line(Ishikawa): progesterone-induced expression of the α1 integrin.Journal of Steroid Biochemistry and Molecular Biology 59 31–39.

Li C, Samsiol G & Iosif C 1999 Quality of life in endometrialcancer survivors. Maturitas 31 227–236.

Li J, Yen C, Liaw D, Podsypanina K, Bose S, Wang SL, Puc J,Miliaresis C, Rodgers L, McCombie R, Bigner SH, GiovanellaBC, Ittmann M, Tycko B, Hibshoosh H, Wigler MH & ParsonsR 1997 PTEN, a putative protein tyrosine phosphatase genemutated in human brain, breast, and prostate cancer. Science 2751943–1947.

Lian Z, Niwa K, Tagami K, Hashimoto M, Gao J, Yokuyama Y,Mori H & Tamaya T 2001 Preventive effects of isoflavones,genistein and daidzein, on estradiol-17beta-related endometrialcarcinogenesis in mice. Japanese Journal of Cancer Research 92726–734.

Lian Z, Niwa K, Gao J, Tagami K, Hashimoto M, Yokoyama Y,Mori H & Tamaya T 2002 Shimotsu-to is the agent inJuzen-taiho-to responsible for the prevention of endometrialcarcinogenesis in mice. Cancer Letters 182 19–26.

Liaw D, Marsh DJ, Li J, Dahia PL, Wang SI, Zheng Z, Bose S,Call KM, Tsou HC, Paecocke M, Eng C & Parsons R 1997Germline mutations of the PTEN gene in Cowden disease, an


inherited breast and thyroid cancer syndrome. Nature Genetics16 64–67.

Liehr JG 2000 Is estradiol a genotoxic mutagenic carcinogen?Endocrine Reviews 21 40–54.

Liehr JG 2001 Genotoxicity of the steroidal estrogens oestrone andoestradiol: possible mechanisms of uterine and mammary cancerdevelopment. Human Reproduction Update 7 273–281.

Liu J, Burdette JE, Xu H, Gu C, Van Breemen RB, Bhat KPL,Booth N, Constantinou AI, Pezzuto JM, Fong HHS, FarnsworthNR & Bolton JL 2001 Evaluation of estrogenic activity of plantextracts for the potential treatment of menopausal symptoms.Journal of Agriculture and Food Chemistry 49 2472–2479.

Love CDB, Muir BB, Scrimgeou JB, Leonard RCF, Dillon P &Dixon JM 1999 Investigation of endometrial abnormalities inasymptomatic women treated with tamoxifen and an evaluationof the role of endometrial screening. Journal of ClinicalOncology 17 2050–2054.

Lovely LP, Appa Rao KB, Gui Y & Lessey BA 2000Characterization of androgen receptors in a well-differentiatedendometrial adenocarcinoma cell line (Ishikawa). Journal ofSteroid Biochemistry and Molecular Biology 74 235–241.

Lynch HT, Krush AJ, Larsen AL & Magnuson CW 1966Endometrial carcinoma: multiple primary malignancies,constitutional factors, and heredity. American Journal of MedicalScience 252 381–390.

Markiewicz & Gurpide 1997 Estrogenic and progestagenicacitivities of physiologic and synthetic androgens, as measuredby in vitro bioassays. Methods and Findings in Experimentaland Clinical Pharmacology 19 215–222.

Markiewicz L, Garey J, Adlercreutz H & Gurpide E 1993 In vitrobioassay of non-steroidal phytoestrogens Journal of SteroidBiochemistry and Molecular Biology 45 399–405.

Meyerson M & Harlow E 1994 Identification of G1 kinase activityfor cdk6, a novel cyclin D partner. Molecular and CellularBiology 14 2077–2086.

Misao R, Nakanishi Y, Fujimoto J & Tamaya T 1998 Effect ofmedroxyprogesterone acetate on sex hormone-binding globulinmRNA expression in the human endometrial cancer cell lineIshikawa. European Journal of Endocrinology 138 574–582.

Mizumoto H, Saito T, Ashihara K, Nishimura M, Tanaka R &Kudo R 2002 Acceleration of invasive activity via matrixmetalloproteinases by transfection of the estrogen receptor-alphagene in endometrial carcinoma cells. International Journal ofCancer 100 401–406.

Mortel R, Zaino RJ & Satyaswaroop PG 1990 Designing aschedule of progestin administration in the control ofendometrial carcinoma growth in the nude mouse model.American Journal of Obstetrics and Gynecology 162 928–934.

Murase T, Niwa K, Morishita S, Itoh N, Mori H, Tanaka T &Tamaya T 1995 Rare occurrence of p53 and ras gene mutationsin preneoplastic and neoplastic mouse endometrial lesioninduced by N-methyl-N-nitrosourea. Cancer Letters 92 223–227.

Nagaoka T, Onodera H, Matsushima A, Todate A, Shibutani M,Ogasawara H & Maekawa A 1990 Spontaneous uterineadenocarcinomas in aged rats and their relation to endocrineimbalance. Journal of Cancer Research and Clinical Oncology116 623–628.

Nagaoka T, Takeuchi M, Onodera H, Mitsumori K, Lu J &Maekawa A 1993 Experimental induction of uterineadenocarcinoma in rats by estrogen and N-methyl-N-nitrosourea.In Vivo 7 525–530.


Nagaoka T, Takeuchi, M, Onodera H, Matsushima Y, Ando-LuJ & Maekawa A 1994 Sequential observation of spontaneousendometrial adenocarcinoma development in Donryu rats.Toxicological Pathology 22 261–269.

Nagaoka T, Onodera H, Hayashi Y & Maekawa A 1995 Influenceof high-fat diets on the occurence of spontaneous uterineendometrial adenocarcinomas in rats. Teratogenesis,Carcinogenesis and Mutagenesis 15 167–177.

Nagaoka T, Takegawa K, Takeuchi M & Maekawa A 2000 Effectsof reproduction on spontaneous development of endometrialadenocarcinomas and mammary tumors in Donryu rats. JapaneseJournal of Cancer Research 91 375–382.

Newbold RR & Liehr JG 2000 Induction of uterineadenocarcinoma in CD-1 mice by catechol estrogens. CancerResearch 60 235–237.

Newbold RR, Bullock BC & McLachlan JA 1990 Uterineadenocarcinoma in mice following developmental treatment withestrogens: a model for hormonal carcinogenesis. CancerResearch 50 7677–7681.

Newbold RR, Jefferson WN, Padilla-Burgos E & Bullock BC 1997Uterine carcinoma in mice treated neonatally with tamoxifen.Carcinogenesis 18 2293–2298.

Newbold RR, Banks EP, Bullock B & Jefferson WN 2001 Uterineadenocarcinoma in mice treated neonatally with genistein.Cancer Research 61 4325–4328.

Nicolaides NC, Papadopoulos N, Liu B, Wie YF, Carter KC,Ruben SM, Rosen CA, Haseltine WA, Fleischmann RD &Fraser CM 1994 Mutations of two PMS homologues inhereditary nonpolyposis colon cancer. Nature 371 75–80.

Nishida M, Kasahara K, Kaneko M, Iwasaki H & Hayashi K 1985Establishment of a new human endometrial adenocarcinoma cellline, Ishikawa cells, containing estrogen and progesteronereceptors. Nippon Sanka Fujinka Gakkai Zasshi 37 1103–1111.

Nishida M, Kasahara K, Tsuji T, Kaneko M & Iwasaki H 1986Mechanisms of the antitumor action of gestagens on endometrialcancer. Nippon Sanka Fujinka Gakkai Zasshi 38 2214–2222.

Nishida M, Kasahara K, Oki A, Satoh T, Arai Y & Kubo T 1996Establishment of eighteen clones of Ishikawa cells. Human Cell9 109–116.

Niwa K, Tanaka T, Mori H, Yokoyama Y, Furui T, Mori H &Tamaya T 1991 Rapid induction of endometrial carcinoma inICR mice treated with N-methyl-N-nitrosourea and17beta-estradiol. Japanese Journal of Cancer Research 82 1391–1396.

Niwa K, Murase T, Furui T, Morishita S, Mori H, Tanaka T, MoriH & Tamaya T 1993 Enhancing effects of estrogens onendometrial carcinogenesis initiated by N-methyl-N-nitrosoeureain ICR mice. Japanese Journal of Cancer Research 84 951–955.

Niwa K, Morishita S, Murase T, Itoh, N, Tanaka T, Mori H &Tamaya T 1995 Inhibitory effects of medroxyprogesteroneacetate on mouse endometrial carcinogenesis. Japanese Journalof Cancer Research 86 724–729.

Niwa K, Morishita S, Murase T, Mudigdo A, Tanaka T, Mori H &Tamaya T 1996 Chronological observation of mouse endometrialcarcinogenesis induced by N-methyl-N-nitrosourea and17beta-estradiol. Cancer Letters 104 115–119.

Niwa K, Morishita S, Hashimoto M, Itoh T, Fujimoto J, Mori H &Tamaya T 1998 Effects of tamoxifen on endometrialcarcinogenesis in mice. Japanese Journal of Cancer Research 89502–509.

Niwa K, Hashimoto M, Morishita S, Ykoyama Y, Mori H &Tamaya T 1999 Preventive effects of Glycyrrhizae radix extract


on estrogen-related endometrial carcinogenesis in mice. JapaneseJournal of Cancer Research 90 726–732.

Niwa K, Hashimoto M, Morishita S, Yokoyama Y, Lian Z, TagamiK, Mori H & Tamaya T 2000 Preventive effects of danazol onendometrial carcinogenesis in mice. Cancer Letters 158 133–139.

Niwa K, Hashimoto M, Morishita S, Lian Z, Tagami K, Mori H &Tamaya T 2001 Preventive effects of Juzen-taiho-to onN-methyl-N-nitrosourea and estradiol-17beta-inducedendometrial carcinogenesis in mice. Carcinogenesis 22 587–591.

Niwa K, Hashimoto M, Lian Z, Gao J, Tagami K, Yokoyama Y,Mori H & Tamaya T 2002 Inhibitory effects of toremifene onN-methyl-N-nitrosourea and estradiol-17beta-inducedendometrial carcinogenesis in mice. Japanese Journal of CancerResearch 93 626–635.

Papadopoulos N, Nicolaides NC, Wei YF, Ruben SM, Carter KC,Rosen CA, Haseltine WA, Fleischmann RD, Fraser CM &Adams MD 1994 Mutation of mutL homolog in hereditary coloncancer. Science 263 1625–1629.

Parazzini F, La Cecchia C, Bocciolone L & Francheschi S 1991The epidemiology of endometrial cancer. Gynecology andOncology 41 1–16.

Peltomaki P, Lothe RA, Aaltonen LA, Pylkkanen L, Nystrom-LahtiM, Seruca M, David L, Holm R, Ryberg D & Haugen A 1993Microsatellite instability associated with tumors that characterizethe hereditary non-polyposis colorectal carcinoma syndrome.Cancer Research 53 5853–5855.

Pinelli DM, Drake J, Williams MC, Cavanagh D & Becker JL1998 Hormonal modulation of Ishikawa cells duringthree-dimensional growth in vitro. Journal of the Society forGynecological Investigation 5 217–223.

Podratz KC, Mariani A & Webb MJ 1998 Staging and therapeuticvalue of lymphadenectomy in endometrial cancer (Editorial).Gynecology and Oncology 70 163–164.

Podsypanina K, Ellenson LH, Nemes A, Gu J, Tamura M, YamadaK, Cordon-Cardo C, Catoretti G & Fisher PE 1999 Mutation ofPten/Mmac1 in mice causes neoplasia in multiple organ systems.PNAS 96 1563–1568.

Risinger JI, Dent GA, Ignar-Trowbridge D, McLachlan JA, TsaoMS, Senterman M & Boyd J 1992 p53 gene mutations in humanendometrial carcinoma. Molecular Carcinogenesis 5 250–253.

Risinger JI, Hayes AK, Berchuk A & Barrett JC 1997 PTEN/MMAC1 mutations in endometrial cancers. Cancer Research 574736–4738.

Risinger JI, Hayes K, Maxwell GL, Carney ME, Dodge RK,Barrett JC & Berchuk A 1998 PTEN mutation in endometrialcancers is associated with favorable clinical and pathologiccharacteristics. Clinical Cancer Research 4 3005–3010.

Rose PG, Brunetto VL, VanLe L, Bell J, Walker JL & Lee RB2000 A phase II trial of anastrozole in advanced recurrent orpersistent endometrial carcinoma: a Gynecologic OncologyGroup study. Gynecological Oncology 78 212–216.

Roshani L, Wedekind D, Szpirer J, Taib Z, Szpirer C, BeckmannB, Riviere M, Hedrich HJ & Klinga-Levan K 2001 Geneticidentification of multiple susceptibility genes involved in thedevelopment of endometrial carcinoma in a rat model.International Journal of Cancer 94 795–799.

Sakakibara K, Kan NC & Satyaswaroop PG 1992 Both17beta-estradiol and tamoxifen induce c-fos messengerribonucleic acid expression in human endometrial carcinomagrown in nude mice. American Journal of Obstetrics andGynecology 166 206–212.


Sakamoto T, Eguchi H, Omoto Y, Ayabe T, Mori H & Hayashi S2002 Estrogen receptor-mediated effects of tamoxifen on humanendometrial cancer cells. Molecular and Cellular Endocrinology192 93–104.

Sakata M, Kurachi H, Morishige K, Ogura K, Yamaguchi M,Nishio Y, Ikegami H, Miyake A & Murata Y 1998 MessengerRNA differential display reverse-transcriptase-polymerase-chain-reaction analysis of a progesterone-suppressive gene in ahuman endometrial-cancer cell line. International Journal ofCancer 78 125–129.

Salvesen HB, MacDonald N, Ryan A, Jacobs IJ, Lynch ED,Akslen LA & Das S 2001 PTEN methylation is associated withadvanced stage and microsatellite instability in endometrialcarcinomas. International Journal of Cancer 91 22–26.

Salvesen HB, Stefansson I, Kalvenes MB, Das S & Akslen LA2002 Loss of PTEN expression is associated with metastaticdisease in patients with endometrial carcinoma. Cancer 942185–2191.

Sandles LG, Shulman LP, Elias S, Photopulos GJ, Smiley LM,Posten WM & Simpson JL 1992 Endometrial carcinoma: geneticanalysis suggesting heritable site-specific uterine cancer.Gynecological Oncology 47 167–171.

Sasano H, Kaga K, Sato S, Yajima A, Nagura H & Harada N 1996Aromatase cytochrome P450 gene expression in endometrialcarcinoma. British Journal Cancer 74 1541–1544.

Sasano H, Nagura H, Watanabe K, Ito K, Tsuiki A, Sato S, YajimaA, Kusakabe M & Sakakura T 1993 Tenascin expression innormal and abnormal human endometrium. Modern Pathology 6323–326.

Satyaswaroop PG & Tabibzadeh SS 1991 Extracellular matrix andthe patterns of differentiation of human endometrial carcinomasin vitro and in vivo. Cancer Research 51 5661–5666.

Satyaswaroop PG, Zaino RJ & Mortel R 1983 Human endometrialadenocarcinoma transplanted into nude mice: growth regulationby estradiol. Science 219 58–60.

Satyaswaroop PG, Clarke CL, Zaino RJ & Mortel R 1992Apparent resistance in human endometrial carcinoma duringcombination treatment with tamoxifen and progestin may resultfrom desensitization following downregulation of tumorprogesterone receptor. Cancer Letters 62 107–114.

Schutze N, Kraft V, Deerberg F, Winking H, Knuppen R &Vollmer G 1992 Functions of estrogens and antiestrogens in therat endometrial adenocarcinoma cell lines RUCA-I andRUCA-II. International Journal of Cancer 52 941–949.

Semczuk A, Berbec H, Kostuch M, Cybulski M, Wojcierowski J &Baranowski W 1998 K-ras gene point mutations in humanendometrial carcinomas: correlation with clinicopathologicalfeatures and patients outcome. Journal of Cancer Research andClinical Oncology 124 695–700.

Shiozawa T, Horiuchi A, Kato K, Obinata M, Konishi I, Fujii S &Nikaido T 2001 Up-regulation of p27Kip1 by progestins isinvolved in the growth suppression of the normal and malignanthuman endometrial glandular cells. Endocrinology 142 4182–4188.

Stambolic V, Tsao M-S, Macpherson D, Suzuki A, ChapmanWB & Mak TW 2000 High incidence of breast and endometrialneoplasia resembling human Cowden syndrome in pten + / −

mice. Cancer Research 60 3605–3611.Steck P, Pershouse MA, Jasser SA, Yung WK, Lin H, Ligon AH,Langford LA, Baumgard ML, Hattier T, Davis T, Frye C, Hu R,Swedlund B, Teng DH & Tavtigian SV 1997 Identification of acandidate tumour suppresor gene, MMAC1, at chromosome


10q23.3 that is mutated in multiple advanced cancers. NatureGenetics 15 356–362.

Strunck E & Vollmer G 1996 Variants of integrin beta 4 subunit inhuman endometrial adenocarcinoma cells: mediators ofECM-induced differentiation? Biochemical and Cellular Biology74 867–873.

Strunck E, Hopert AC & Vollmer G 1996 Basement membraneregulates gene expression in HEC1B(L) endometrialadenocarcinoma cells. Biochemical and Biophysical ResearchCommunications 221 346–350.

Strunck E, Frank K, Tan MI & Vollmer G 2001 Expression ofL-3-phosphoserine phosphatase is regulated by reconstitutedbasement membrane. Biochemical and Biophysical ResearchCommunications 281 747–753.

Suguwara J, Fukaya T, Murakami T, Yoshida H & Yajima A 1997Hepatocyte growth factor stimulated proliferation, migration, andlumen formation of human endometrial epithelial cells in vitro.Biology of Reproduction 57 936–942.

Sundstrom SA, Komm BS, Ponce-de-Leon H, Yi Z, Teuscher C &Lyttle CR 1989 Estrogen regulation of tissue-specific expressionof complement C3. Journal of Biological Chemistry 264 16941–16947.

Suzuki A, de la Pompa JL, Stambolic V, Elia AJ, Sasaki T, delBarco Barrantes I, Ho A, Wakeham A, Itie A, Khoo W,Fukumoto M & Mak TW 1998 High cancer susceptibility andembryonic lethality associated with mutations of the PTENtumor suppresssor gene in mice. Current Biology 8 1169–1178.

Tan MI, Strunck E, Scholzen T, Gerdes J & Vollmer G 1999Extracellular matrix regulates steady-state mRNA levels of theproliferation associated protein Ki-67 in endometrial cancercells. Cancer Letters 140 145–152.

Tanoguchi K, Yaegashi N, Jiko K, Maekawa A, Sato S & YajimaA 1999 K-ras point mutations in spontaneously occurringendometrial adenocarcinomas in Donryu rats. Tohoku Journal ofExperimental Medicine 189 87–93.

Tashiro H, Blazes MS, Wu R, Cho KR, Bose S, Wang SI, Li J,Parsons R & Ellenson LH 1997 Mutations in PTEN are frequentin endometrial carcinoma but rare in other commongynecological malignancies. Cancer Research 57 3935–3940.

Terry P, Baron JA, Weiderpass E, Yuen J, Lichtenstein P & NyrenO 1999 Lifestyle and endometrial cancer risk: a cohort studyfrom the Swedish twin registry. International Journal of Cancer82 38–42.

Tonetti, DA, O’Reagan R, Tanjore S, England G & Jordan VC1998 Antiestrogen stimulated human endometrial cancer growth:laboratory and clinical considerations. Journal of SteroidBiochemistry and Molecular Biology 65 181–189.

Treeck O, Strunck E & Vollmer G 1998 A novel basementmembrane-induced gene identified in the human endometrialadenocarcinoma cell line (HEC1B) FEBS Letters 425 426–430.

Vollmer G & Schneider MR 1996 The rat endometrialadenocarcinoma cell line RUCA-I: a novel hormone-responsivein vivo/in vitro tumor model. Jourmal of Steroid Biochemistryand Molecular Biology 58 103–115.

Vollmer G, Siegal GP, Chiquet-Ehrismann R, Lightner VA,Arnholt H & Knuppen R 1990 Tenascin expression in thehuman endometrium and in endometrial adenocarcinomas.Laboratory Investigations 62 725–730.

Vollmer G, Deerberg F, Siegal GP & Knuppen R 1991 Alteredtenascin expression during spontaneous endometrialcarcinogenesis in the BDII/Han rat. Virchows Archive of B CellPathology Including Molecular Pathology 60 83–99.


Vollmer G, Ellerbrake N, Hopert A-C, Wunsche W & Knuppen R1995a Extracellular matrix induces hormone responsiveness inRUCA-I rat endometrial adenocarcinoma cells. Journal ofSteroid Biochemistry and Molecular Biology 52 259–269.

Vollmer G, Hopert AC, Ellerbrake N, Wunsche W & Knuppen R1995b Fibronectin is an estrogen repressed protein in RUCA-Irat endometrial adenocarcinoma cell. Journal of SteroidBiochemistry and Molecular Biology 54 141–139.

Wagatsuma S, Konno R, Sato S & Yajima A 1998 Tumorangiogenesis, hepatocyte growth factor, and c-Met expression inendometrial carcinoma. Cancer 82 520–530.

Walentinsson A, Helou K, Wallenius V, Hedrich HJ, Szpirer C &Levan G 2001 Independent amplification of two gene clusters onchromosome 4 in rat endometrial cancer: identification andmolecular characterization. Cancer Research 61 8263–8273.

Wang H, Douglas W, Lia M, Edelmann W, Kucherlapati R,Podsypanina K, Parsons R & Ellenson LH 2002 DNA mismatchrepair deficiency accelerates endometrial tumorigenesis in Ptenheterozygous mice. American Journal of Pathology 160 1481–1486.

Watanabe K, Sasano, H, Harada N, Ozaki M, Niikura H, Sato S &Yajima A 1995 Aromatase in human endometrial carcinoma andhyperplasia. Immunohistochemical, in situ hybridization, andbiochemical studies. American Journal of Pathology 146 491–500.

Weiderpass E, Adami H-O, Baron JA, Magnusson C, Bergstrom R,Lindgren A, Correia N & Persson I 1999 Risk of endometrialcancer following estrogen replacement with or withoutprogestins. Journal of the National Cancer Institute 91 1131–1137.

Whelan AJ, Babb S, Mutch DG, Rader J, Herzog TJ, Todd C,Ivanovich JL & Goodfellow PJ 2002 MSI in endometrialcarcinoma: absence of MLH1 promoter methylation is associatedwith increased familial risk for cancers. International Journal ofCancer 99 697–704.


Wormke M, Castro-Rivera E, Chen I & Safe S 2000 Estrogen andaryl hydrocarbon receptor expression and crosstalk in humanIshikawa endometrial cells. Journal of Steroid Biochemistry andMolecular Biology 72 197–207.

Wunsche W, Tenniswood MP, Hopert A-C, Schneider MR &Vollmer G 1998 Estrogenic regulation of clusterin mRNA innormal and malignant endometrial tissue. International Journalof Cancer 76 684–688.

Yamada H, Sasaki M, Honda T, Wake N, Boyd J, Oshimura M &Barrett JC 1995 Suppression of endometrial carcinoma celltumorigenicity by human chromosome 18. Genes ChromosomeCancer 13 18–24.

Yang S, Fang Z, Gurates B, Tamura M, Miller J, Ferrer K &Bulun SE 2001 Stromal PRs mediate induction of17beta-hydroxysteroid dehydrogenase type 2 expression inhuman endometrial epithelium: a paracrine mechanism forinactivation of E2. Molecular Endocrinology 15 2093–2105.

Yoshida A, Newbold RR & Dixon D 2000 Abnormaldifferentiation and p21 expression of endometrial epithelial cellsfollowing developmental exposure to diethylstilbestrol (DES).Toxicological Pathology 28 237–245.

Yoshida M, Kudoh K, Katsuda S, Takahashi M, Ando J &Maekawa A 1998 Inhibitory effect of uterine endometrialcarcinogenesis in Donryu rats by tamoxifen. Cancer Letters 1143–51.

Zhou CP, Loukola A, Salovaara R, Nystrom-Lahti M, Peltomaki P,de la Chapelle A, Aaltonen LA & Eng C 2002a PTENmutational spectra, expression levels, and subcellular localizationin microsatellite stable and unstable colorectal cancers. AmericanJournal of Pathology 161 439–447.

Zhou XP, Kuismanen S, Nystrom-Lahti M, Peltomaki P & Eng C2002b Distinct PTEN mutational spectra in hereditarynon-polyposis colon cancer syndrome-related endometrialcarcinomas compared to sporadic microsatellite unstable tumors.Human Molecular Genetics 11 445–450.

Documents

Endometrial cancer: experimental models useful for studies on