A selective estrogen receptor-{beta} agonist causes lesion regression in an experimentally induced model of endometriosis

Heather A. Harris1,3, Kaylon L. Bruner-Tran2, Xiaochun Zhang1, Kevin G. Osteen2 and C. Richard Lyttle1

1 Women's Health Research Institute, Wyeth Research, Collegeville, PA 19426 and 2 Department of Obstetrics and Gynecology, Vanderbilt University School of Medicine, Nashville, TN 37232, USA

3 To whom correspondence should be addressed at: Women's Health Research Institute, Wyeth Research, RN 3256, 500 Arcola Road, Collegeville, PA 19426, USA. Email: harrish{at}wyeth.com


    Abstract
 Top
 Abstract
 Introduction
 Materials and methods
 Results
 Discussion
 Acknowledgements
 References
 
BACKGROUND: Endometriosis is a common gynaecological problem of uncertain aetiology. It affects primarily young, reproductive-aged women and can result in chronic pelvic pain and infertility. Current approved therapies have significant side-effects and hysterectomy is employed as a final solution. ERB-041 is a selective estrogen receptor-{beta} (ER{beta}) agonist that has anti-inflammatory activity in preclinical models of arthritis and inflammatory bowel disease, but is inactive in many preclinical models of classic estrogen activity. Because endometriosis is now thought to be, at least in part, an inflammatory disease, we evaluated ERB-041's activity in an experimentally induced model of endometriosis. METHODS: Athymic nude mice (ovariectomized or intact) were implanted with tissue fragments of normal human endometrium. After establishment of lesions for 11–14 days, mice were treated with ERB-041 for 15–17 days. Upon euthanasia, the number of lesions, their size and location were noted. Five lesions were recovered for RNA analysis. RESULTS: Across six studies, ERB-041 caused complete lesion regression in 40–75% of the mice studied. The compound appeared to be equally effective in gonad-intact as in ovariectomized mice, and analysed recovered lesions expressed ER{alpha} but not ER{beta} mRNA. CONCLUSIONS: ERB-041 and possibly other ER{beta} selective agonists may be a useful new approach to treating endometriosis.

Key words: endometriosis/ERB-041/estrogen receptor-{beta}/immune system/nude mouse


    Introduction
 Top
 Abstract
 Introduction
 Materials and methods
 Results
 Discussion
 Acknowledgements
 References
 
Endometriosis is a disease defined by the establishment and growth of endometrial tissue outside the uterus. It is common in reproductive-aged women, although many may be asymptomatic. For women with symptoms, the primary symptom is pain: chronic pelvic pain, painful menstruation, painful intercourse. Frequently endometriosis is diagnosed in infertile women seeking assisted reproduction [for review, see American College of Obstetricians and Gynecologists (ACOG), 2000Go].

The causes of this disease continue to be debated. A leading hypothesis, first advanced by Sampson (1927)Go, is that viable endometrial tissue escapes into the peritoneal cavity during menstruation, instead of being shed through the vagina. Other hypotheses include coelomic or lymphatic metaplasia, and induction of mesenchyma by uterine factors (Witz, 2002Go).

The phenomenon of retrograde menstruation likely occurs in most women, and yet symptomatic disease develops in a relatively small percentage. The resolution of this apparent paradox may reside in differences between the immune systems of normal women and those with endometriosis (reviewed in Lebovic et al., 2001Go; Nothnick, 2001Go). The immune system of normal women may have a greater ability to recognize errant endometrial tissue as foreign and to remove it. That this process does not happen efficiently in women with endometriosis is supported by observations in immune system imbalances. For example, decreased natural killer (NK) cell or macrophage activity has been reported in endometriosis patients (Oosterlynck et al., 1991Go).

Current medical therapies are inadequate and include a progression from symptom management with oral contraceptives to more aggressive measures such as suppression of estrogen levels using GnRH agonists, surgery to reduce lesion burden and finally, hysterectomy (ACOG, 2000Go). Each of these treatments has significant efficacy or tolerability liabilities. For example, although oral contraceptives are often used as first-line therapy, little scientific evidence supports their efficacy (Moore et al., 2004Go). Treatment with GnRH agonists has a number of serious side-effects including bone loss, hot flushes, mood swings, and vaginal atrophy. In fact, its use is limited to 6 months. Clearly, an effective therapy with an improved side-effect profile is needed.

Two forms of the estrogen receptor have been described. The first, now called ER{alpha}, was cloned by Green et al. (1986)Go and is thought to be responsible for mediating many of estrogen's actions such as its trophic effects on the uterus and feedback of the hypothalamic–pituitary–ovarian axis (Couse and Korach, 1999Go). The second form (ER{beta}) was discovered more recently (Kuiper et al., 1996Go) and its function is not completely understood. Based on tissue distribution studies, it is not the dominant ER in uterus, although the human uterus does appear to express more ER{beta} than does the rodent (Couse et al., 1997Go; Kuiper et al., 1997Go; Matsuzaki et al., 1999Go; Pelletier and El-Alfy, 2000Go; Lecce et al., 2001Go).

We have developed a highly selective ER{beta} agonist that is >200-fold selective for ER{beta} based on a competitive radioligand binding assay. ERB-041 (Figure 1) has been extensively characterized in various models of estrogen action and found to be inactive on classic estrogenic targets such as the uterus, mammary gland and bone. However, it has potent anti-inflammatory activity in two in vivo models: the HLA-B27 transgenic rat and Lewis rat adjuvant-induced arthritis (Harris et al., 2003Go). We report here that this compound causes lesion regression in an experimentally induced model of endometriosis using human tissue and therefore may have utility in treating this disease.



View larger version (6K):
[in this window]
[in a new window]
 
Figure 1. The structure of ERB-041.

 

    Materials and methods
 Top
 Abstract
 Introduction
 Materials and methods
 Results
 Discussion
 Acknowledgements
 References
 
All reagents were obtained in the USA.

Acquisition of human tissues
Informed consent was obtained prior to endometrial biopsy collection. Although names were kept confidential, each patient's age, cycle stage and medication history were made available. Use of human tissues was approved by Vanderbilt University's Institutional Review Board and Committee for the Protection of Human Subjects. Normal endometrial tissues were acquired by biopsy during the proliferative phase (days 9–12) of the menstrual cycle from a donor population (aged 18–45 years) exhibiting normal menstrual cycles and no history of endometriosis. An endometrial thickness of >9 mm (confirmed by intravaginal ultrasound) and a serum progesterone level of <1.5 ng/ml were required for inclusion in this study. Individuals with a recent (≤3 months) history of hormone therapy (e.g. oral contraceptives) or other medications which might impact study results (e.g. anti-inflammatory agents) were excluded. Biopsies were obtained from the uterine fundus using a Pipelle® endometrial suction curette (Unimar, Inc., USA) and the tissue was washed in pre-warmed, phenol red-free Dulbecco's modified Eagle's medium/Ham's F-12 medium (DMEM/F-12) (Sigma, USA) to remove residual blood and mucus prior to culturing. A portion of each sample was formalin-fixed for histological confirmation of cycle stage.

Organ cultures of human tissues
Endometrial biopsies were dissected into small cubes (1 mm3) and 8–10 pieces per treatment group were suspended in tissue culture inserts (Millipore, USA) as previously described (Bruner et al., 1999Go). Each organ culture was maintained in 1 nmol/l 17{beta}-estradiol (Sigma) under serum-free conditions in DMEM/F-12 supplemented with 1% Insulin-Transferrin-Selenium (ITS+; Collaborative Biomedical, USA) and 0.1% Excyte (Miles Scientific, USA) and an antibiotic/antimycotic solution (Hybri-Max; Sigma, USA). Tissues were incubated overnight at 37°C in a humidified chamber with 5% CO2.

Experimental model of endometriosis
The model of endometriosis was performed as previously described (Bruner-Tran et al., 2002Go) with minor modifications. Prior to any invasive procedure or death, mice were anesthetized with Metofane (Pittman–Moore, USA). All studies using animals were performed using procedures approved by Vanderbilt University's Institutional Animal Care and Use Committee. Five week old athymic (ncr/nude), intact or ovariectomized mice (Harlan Sprague Dawley) were injected with a phosphate-buffered saline suspension of 8–10 human endometrial tissue fragments/mouse. Each study used tissue collected from a different donor. A typically sized endometrial biopsy yielded sufficient tissue for establishing disease in a maximum of 16 mice. In contrast to previous studies, these mice were not implanted with slow-releasing estradiol pellets. Tissue fragments had been incubated as described above for the preceding 24 h. Table I summarizes the experimental parameters for the six studies reported here. For all but one study, subcutaneous tissue injections were given on the ventral midline just below the umbilicus. Lesions were allowed to establish for 11–14 days and mice with visible subcutaneous disease were randomly allocated to treatment groups. Daily oral treatments (10 mg/kg or 1 mg/kg) were given in a volume of 0.1 ml by gavage. A vehicle of 2% Tween-80/0.5% methylcellulose was used to solubilize ERB-041. ERB-041 was obtained from the Wyeth Research compound library (Malamas et al., 2004Go). At the end of the experimental protocol, mice were euthanized by cervical dislocation and the subcutaneous area as well as the peritoneal cavity was examined for endometriosis. Lesions were counted, photographed and measured. Some recovered lesions were frozen in liquid nitrogen for RNA preparation.


View this table:
[in this window]
[in a new window]
 
Table I. Summary of experimental parameters

 
In one study (Study 6), the tissue injections were made intraperitoneally on the ventral midline just below the umbilicus. Since most of the lesions were expected to establish inside the peritoneal cavity, disease was unable to be assessed prior to randomization for treatment. Based on previous experience, we expected to establish disease in 90–95% of the mice. ERB-041 or vehicle was administered as described above.

Analysis of ER mRNA expression in recovered lesions
Total RNA was prepared from each individual recovered lesion. Each sample was homogenized in 1 ml of Trizol (Invitrogen, USA) at 4°C using a Polytron homogenizer PT1200C (BrinkMann, USA). After 0.2 ml chloroform extraction and centrifugation, ~0.5 ml aqueous phase was collected. The RNA from the aqueous phase was then purified by Qiagen RNeasy kits (Qiagen, USA). The genomic DNA in RNA sample was removed by on-column RNase-Free DNase set during RNA purification.

The RNA concentration was adjusted to 0.05 mg/ml for assay. mRNA expression was analysed using real-time quantitative PCR on an ABI PRISM 7700 Sequence Detection System according to the manufacturer's protocol (Applied Biosystems Inc., USA). The sequences of primers and labelled probes used for mRNA detection were designed in-house using Primer Express 2.0 (Applied Biosystems) and were as follows: human ER{alpha} forward primer, 5'-ATGATCAAACGCTCTAAGAAGAACAG-3'; reverse primer, 5'-ACTGAAGGGTCTGGTAGGATCATACT-3'; probe, 5'FAM-TCCCTGACGGCCGACCAGATG-TAMRA3'; human ER{beta} forward primer, 5'-GGTCCATCGCCAGTTATCACA-3'; reverse primer, 5'-GTGTTCTAGCGATCTTGCTTCACA-3'; probe, 5'VIC-CTGTATGCGGAACCTCAAAAGAGTCCCTG-TAMRA3'.


    Results
 Top
 Abstract
 Introduction
 Materials and methods
 Results
 Discussion
 Acknowledgements
 References
 
Lesion regression in ovariectomized mice
Eutopic human endometrial tissue will establish lesions when implanted into nude mice. Three independent studies (Studies 1–3) were performed each using a different donor's tissue. These replicate studies were performed to assess the reproducibility of the response across studies and with different donor samples. Ovariectomized mice were used for initial ‘proof-of-concept’ studies to avoid the confounding variables of endogenous ovarian hormones. At the start of treatment all mice had established lesions, and in each study ERB-041 caused complete lesion regression in some mice. Across these studies, lesions completely regressed in 40–75% of the mice (depending on the study) and the average number of lesions per mouse decreased from 2.2 to 0.4 (Figure 2).



View larger version (10K):
[in this window]
[in a new window]
 
Figure 2. Average number of lesions per ovariectomized mouse. Data from Studies 1–3 using three different human donors are combined to show the lesions per mouse in vehicle and ERB-041 (10 mg/kg)-treated mice. A total of 10 mice were treated with vehicle and 15 mice treated with ERB-041. *Significantly less than vehicle, P<0.05 using Dunnett's test.

 
Lesion regression in intact mice
Since human endometriosis patients are typically young women with reproductive cycles, we next examined the activity of ERB-041 in gonad-intact mice. Indeed, this nude mouse model of endometriosis was developed using ovariectomized mice implanted with a slow-release pellet of 17{beta}-estradiol and the presence of this pellet increases the efficiency of lesion engraftment (Bruner et al., 1997Go). Therefore, non-selective estrogens do not cause lesion regression and it is possible that endogenous estrogens would block the effect of an ER{beta} selective ligand. However, in two independent studies using intact mice (Studies 4 and 5), ERB-041 still caused lesion regression. Two doses were studied. At 10 mg/kg, ERB-041 caused lesion regression in 40–60% of the mice. At 1 mg/kg, the results were more variable: in one study, lesions regressed in 80% of the mice and in the other study efficacy was 40%. These data are summarized in Table II.


View this table:
[in this window]
[in a new window]
 
Table II. Effect of ERB-041 on lesions in intact mice

 
Estrogen receptor mRNA expression
Some of the lesions remaining at the end of Study 1 and Study 3 were recovered, and RNA was prepared to assess ER expression. Four were from vehicle-treated mice and one was from an ERB-041-treated mouse. Expression of human ER{alpha} and ER{beta} was evaluated by quantitative, real-time RT–PCR. As shown in Figure 3, all these lesions expressed only ER{alpha}. Thymus RNA was included as a positive control for reagent function. The amount of ER per sample is highly variable (note log scale on y-axis) and is likely due to varying amounts of mouse tissue co-excised with the human tissue implants as ER expression is expressed as copies of human ER mRNA/ng total (human and mouse) RNA.



View larger version (18K):
[in this window]
[in a new window]
 
Figure 3. Estrogen receptor (ER) expression in recovered lesions. Total RNA was prepared from three lesions recovered from Study 1 as well as two lesions recovered from Study 3. The treatment the mice received is noted on the x-axis. Human ER{alpha} and ER{beta} expression was measured by quantitative real-time RT–PCR. Human thymus RNA was used as a control as it contains mRNA coding for both ER.

 
Lesion size and distribution
Lesion size was measured in all six studies and found to be highly variable. Combining data for Studies 1–6, no statistically significant differences were seen in the size of remaining lesions. Thus, either ERB-041 has an ‘all-or-none’ effect on lesion regression, or the numbers studied here were insufficient to detect a difference.

Although the human tissue implants were placed subcutaneously in Studies 1–5, some lesions were found inside the peritoneal cavity, presumably due to inadvertent penetration of the peritoneal wall during injection. In these studies, we noticed that no internal lesions were found in mice treated with ERB-041, whereas 27% of the lesions were internal in mice treated with vehicle (Table III). This observation suggested that ERB-041 was particularly effective at clearing lesions when they occurred internally. In order to more directly test this hypothesis, a separate study was performed (Study 6) where tissue fragments were intentionally placed in the peritoneal cavity. As shown in Table IV, 67% of lesions in vehicle-treated mice were inside the peritoneal cavity, whereas again no internal lesions were seen in mice treated with ERB-041. In this study, ERB-041 caused complete lesion regression in 80% of the mice.


View this table:
[in this window]
[in a new window]
 
Table III. Distribution of lesions at necropsy after intentional subcutaneous placement of tissue in Studies 1–5

 

View this table:
[in this window]
[in a new window]
 
Table IV. Distribution of lesions at necropsy after intentional intraperitoneal placement of tissue in Study 6

 

    Discussion
 Top
 Abstract
 Introduction
 Materials and methods
 Results
 Discussion
 Acknowledgements
 References
 
Endometriosis, for many women, is a debilitating disease not adequately addressed by currently available medical therapies. Current approved therapies are associated with serious side-effects and hysterectomy is used as a final solution. GnRH agonists are a currently approved non-surgical therapy for this disease but these agents induce a post-menopausal-like state with serious consequences such as bone loss. Thus their use is limited to 6 months (ACOG, 2000Go). Some of the drawbacks of GnRH therapy can be mitigated by replacement therapy with an estrogen and/or a progestin (ACOG, 2000Go). A number of non-hormone suppressing agents have been considered as potential therapeutics for endometriosis. These include tumour necrosis factor {alpha} inhibitors (reviewed in Nothnick and D'Hooghe, 2003Go), matrix metalloprotease inhibitors (Osteen et al., 2003Go) and anti-angiogenic agents (Hull et al., 2003Go).

Development of improved treatment options is hampered by the disease's complexity and a lack of tractable natural animal models. Although naturally occurring only in primates, several rodent models of this disease have been developed. One of the most widely used and well-established models involves transplantation of human endometrial tissue into nude mice, waiting until visible lesions develop and then treating with test compound. This model has been shown to share features with the native human disease, such as the morphological and histological appearance of lesions (Bruner et al., 1997Go; Grummer et al., 2001Go). We show here for the first time that an ER{beta} selective agonist, ERB-041, can cause lesion regression in an experimental model of endometriosis.

ERB-041 has been evaluated in this nude mouse xenograft model in six independent studies using six different human tissue samples (Table I). The size of the human tissue biopsy limited the number of mice able to be evaluated in each experiment to ≤16. All studies used a therapeutic regimen; that is, the compound treatment started after lesion establishment. Subcutaneous lesions (the only lesions visible before euthanasia) did not spontaneously regress in any of the vehicle-treated mice. Three studies used ovariectomized mice and tested ERB-041 at 10 mg/kg. In each study, ERB-041 caused complete lesion regression in some mice. Efficacy varied from 40 to 75%, depending on the study. Two additional studies used intact mice and evaluated two doses of ERB-041 (10 and 1 mg/kg). ERB-041 was similarly effective in intact as in ovariectomized mice, suggesting that endogenous levels of 17{beta}-estradiol (a non-receptor subtype selective estrogen) do not block the activity of an exogenous selective ER{beta} agonist.

Five lesions were examined at the study's end (four from vehicle-treated ovariectomized mice and one from an ovariectomized mouse treated with ERB-041) and all expressed only ER{alpha} as determined by real-time quantitative RT–PCR. This finding was somewhat unexpected as normal human endometrium does express low levels of ER{beta} (Matsuzaki et al., 1999Go; Taylor and Al-Azzawi, 2000Go; Lecce et al., 2001Go). Given our sample size limitations we did not examine expression of ER before transplantation into the mice. As most of the lesions examined for ER expression were from vehicle-treated mice, our observation of exclusive ER{alpha} expression suggests that ER{beta} was absent from the outset, and not influenced by compound treatment. To confirm this finding, as mentioned below, future studies will examine lesion dynamics during regression and ER{alpha}/{beta} expression will be one of the endpoints measured.

Given that the recovered human lesions expressed ER{alpha} and not ER{beta}, we believe that ERB-041 exerted its effects on the host, rather than on the implanted tissue (e.g. by directly inducing apoptosis). Our current working hypothesis is that ERB-041 affects the host's immune system. Although lacking T cells, athymic nude mice do have fully functional NK cells (Hasui et al., 1989Go) and macrophages (Cheers and Waller, 1975Go). As NK cells and macrophages both express ER{beta} (Curran et al., 2001Go; Stygar et al., 2001Go; Vegeto et al., 2001Go; Henderson et al., 2003Go), both cells are possible targets for ERB-041.

During the course of our studies on ERB-041, we recorded lesion location. In studies 1–5, vehicle-treated mice typically had some number of internal lesions but no internal lesions were seen in any of the ERB-041-treated mice. Thus, it appeared that ERB-041 was more efficient at causing regression when the lesions were found inside the peritoneal cavity. These observations were confirmed in a study (study 6) where the tissue fragments were intentionally placed internally. The reason for this differential effect is unclear, but it is possible that ERB-041 stimulates the activity of peritoneal immune surveillance cells.

The primary goal of the studies reported here was to assess the effect of an ER{beta} agonist, ERB-041, on lesion regression. Having documented the compound's effectiveness in both gonadectomized and gonad-intact mice, future studies will concentrate on elucidating mechanism of action. In most studies, histological analysis was judged to be fruitless as mice were treated for >2 weeks with compound. Lesions remaining at the end of the study were refractory to treatment. In future studies, histological analysis, coupled with immunostaining for key cellular markers (e.g. NK cells, macrophages, new blood vessel growth) will be performed within a week of beginning compound treatment in an effort to view the process of regression.

ERB-041, if effective at reducing disease burden in human endometriosis patients, would have potential advantages over the currently approved GnRH therapies, which are thought to act primarily via suppression of the hypothalamic–pituitary–gonadal axis. Since ER{alpha} appears to be the receptor responsible for mediating the negative feedback to the hypothalamus and pituitary (Couse and Korach, 1999Go), one would expect that an ER{beta} selective ligand would not share the same side-effect profile as a GnRH agonist. In addition, ERB-041 lacks classic estrogenic activity in several in vivo model systems (Harris et al., 2003Go), suggesting it will lack the liabilities associated with non-receptor subtype selective estrogens. The unique side-effect profile of an ER{beta} selective agonist remains to be determined, and is under investigation. However, as ER{beta} regulates few if any genes in normal mice (Jelinsky et al., 2003Go), it is expected that ERB-041 will have a favourable side-effect profile.

In summary, we show for the first time that a selective ER{beta} agonist causes regression of human endometrial xenografts in a mouse model of endometriosis. These data suggest that further evaluation of ERB-041 may be warranted, for example, in primates with naturally occurring disease.


    Acknowledgements
 Top
 Abstract
 Introduction
 Materials and methods
 Results
 Discussion
 Acknowledgements
 References
 
The authors thank Greg Kopf for helpful discussions and Rick Winneker for valuable criticism of the manuscript. We are also grateful to Derek Janszen for assisting with the statistical analysis of the data and to Mike Malamas for providing ERB-041.


    References
 Top
 Abstract
 Introduction
 Materials and methods
 Results
 Discussion
 Acknowledgements
 References
 
American College of Obstetricians and Gynecologists (2000) Medical management of endometriosis. Int J Gynecol Obstet 71, 185–195.

Bruner KL, Matrisian LM, Rodgers WH, Gorstein F and Osteen KG (1997) Suppression of matrix metalloproteinases inhibits establishment of ectopic lesions by human endometrium in nude mice. J Clin Invest 99, 2851–2857.[Abstract/Free Full Text]

Bruner KL, Etsenberg E, Gorstein F and Osteen KG (1999) Progesterone and transforming growth factor-beta coordinately regulate suppression of endometrial matrix metalloproteinases in a model of experimental endometriosis. Steroids 64, 648–653.[CrossRef][ISI][Medline]

Bruner-Tran K, Eisenberg E, Yeaman G, Anderson T, McBean J and Osteen K (2002) Steroid and cytokine regulation of matrix metalloproteinase expression in endometriosis and the establishment of experimental endometriosis in nude mice. J Clin Endocrinol Metab 87, 4782–4791.[Abstract/Free Full Text]

Cheers C and Waller R (1975) Activated macrophages in congenitally athymic "nude mice" and in lethally irradiate mice. J Immunol 115, 844–847.[Abstract]

Couse JF and Korach KS (1999) Estrogen receptor null mice: what have we learned and where will they lead us? Endocr Rev 20, 358–417.[Abstract/Free Full Text]

Couse JF, Lindzey J, Grandien K, Gustafsson JA and Korach KS (1997) Tissue distribution and quantitative analysis of estrogen receptor-alpha (ERalpha) and estrogen receptor-beta (ERbeta) messenger ribonucleic acid in the wild-type and ERalpha-knockout mouse. Endocrinology 138, 4613–4621.[Abstract/Free Full Text]

Curran EM, Berghaus L, Vernetti N, Saporita A, Lubahn D and Estes D (2001) Natural killer cells express estrogen receptor-alpha and estrogen receptor-beta and can respond to estrogen via a non-estrogen receptor-alpha-mediated pathway. Cellular Immunol 214, 12–20.[CrossRef][ISI][Medline]

Green S, Walter P, Kumar V, Krust A, Bornert J, Argos P and Chambon P (1986) Human oestrogen receptor cDNA: sequence, expression and homology to v-erb-A. Nature 320, 134–139.[ISI][Medline]

Grummer R, Schwarzer F, Bainczyk K, Hess-Stumpp H, Regidor P, Schindler A and Winterhager E (2001) Peritoneal endometriosis: validation of an in vivo model. Hum Reprod 16, 1736–1743.[Abstract/Free Full Text]

Harris HA, Albert LM, Leathurby Y, Malamas MS, Mewshaw RE, Miller CP, Kharode YP, Marzolf J, Komm BS, Winneker RC et al. (2003) Evaluation of an estrogen receptor-{beta} agonist in animal models of human disease. Endocrinology 144, 4241–4249.[Abstract/Free Full Text]

Hasui M, Saikawa Y, Miura M, Takano N, Ueno Y, Yachie A, Miyawaki T and Taniguchi N (1989) Effector and precursor phenotypes of lymphokine-activated killer cells in mice with severe combined immunodeficiency (scid) and athymic (nude) mice. Cellular Immunol 120, 230–239.[CrossRef][ISI][Medline]

Henderson TA, Saunders PTK, Moffett-King A, Groome NP and Critchley HOD (2003) Steroid receptor expression in uterine natural killer cells. J Clin Endocrinol Metab 88, 440–449.[Abstract/Free Full Text]

Hull ML, Charnock-Jones DS, Chan CLK, Bruner-Tran KL, Osteen KG, Tom BDM, Fan TPD and Smith SK (2003) Antiangiogenic agents are effective inhibitors of endometriosis. J Clin Endocrinol Metab 88, 2889–2899.[Abstract/Free Full Text]

Jelinsky SA, Harris HA, Brown EL, Flanagan K, Zhang XC, Tunkey C, Lai KD, Lane MV, Simcoe DK and Evans MJ (2003) Global transcription profiling of estrogen activity: estrogen receptor a regulates gene expression in the kidney. Endocrinology 144, 701–710.[Abstract/Free Full Text]

Kuiper GGJM, Enmark E, Pelto-Huikko M, Nilsson S and Gustafsson J-A (1996) Cloning of a novel estrogen receptor expressed in rat prostate and ovary. Proc Natl Acad Sci USA 93, 5925–5930.[Abstract/Free Full Text]

Kuiper GGJM, Carlsson B, Grandian K, Enmark E, Haggblad J, Nilsson S and Gustafsson JA (1997) Comparison of the ligand binding specificity and transcript tissue distribution of estrogen receptor alpha and beta. Endocrinology 138, 863–870.[Abstract/Free Full Text]

Lebovic DI, Mueller MD and Taylor RN (2001) Immunobiology of endometriosis. Fertil Steril 75, 1–10.[CrossRef][ISI][Medline]

Lecce G, Meduri G, Ancelin M, Bergeron C and Perrot-Applanat M (2001) Presence of estrogen receptor beta in the human endometrium through the cycle: expression in glandular, stromal, and vascular cells. J Clin Endocrinol Metab 86, 1379–1386.[Abstract/Free Full Text]

Malamas MS, Manas ES, McDevitt RE, Gunawan I, Zhang XB, Collini MD, Miller CP, Dinh T, Henderson RA, Keith JC, Jr et al. (2004) Design and synthesis of aryl di-phenolic azoles as potent and selective estrogen receptor-{beta} ligands. J Med Chem 47, 5021–5040.[CrossRef][ISI][Medline]

Matsuzaki S, Fukaya T, Suzuki T, Murakami T, Sasano H and Yajima A (1999) Oestrogen receptor alpha and beta mRNA expression in human endometrium throughout the menstrual cycle. Mol Hum Reprod 5, 559–564.[Abstract/Free Full Text]

Moore J, Kennedy S and Prentice A (2004) Modern combined oral contraceptives for pain associated with endometriosis. Cochrane Database of Systematic Reviews 4.

Nothnick WB (2001) Treating endometriosis as an autoimmune disease. Fertil Steril 76, 223–231.[CrossRef][ISI][Medline]

Nothnick WB and D'Hooghe TM (2003) Medical management of endometriosis: novel targets and approaches towards the development of future treatment regimes. Gynecol Obstet Invest 55, 189–198.[ISI][Medline]

Oosterlynck D, Cornillie F, Waer M, Vandeputte M and Koninckx P (1991) Women with endometriosis show a defect in natural killer activity resulting in a decreased cytoxicity to autologous endometrium. Fertil Steril 56, 45–51.[ISI][Medline]

Osteen KG, Yeaman GR and Bruner-Tran KL (2003) Matrix metalloproteinases and endometriosis. Semin Reprod Med 21, 155–163.[CrossRef][ISI][Medline]

Pelletier G and El-Alfy M (2000) Immunocytochemical localization of estrogen receptors alpha and beta in the human reproductive organs. J Clin Endocrinol Metab 85, 4835–4840.[Abstract/Free Full Text]

Sampson JA (1927) Peritoneal endometriosis due to menstrual dissemination of endometrial tissue into the peritoneal cavity. Am J Obstet Gynecol 14, 422–469.

Stygar D, Wang H, Vladic YS, Ekman G, Eriksson H and Sahlin L (2001) Co-localization of oestrogen receptor beta and leukocyte markers in the human cervix. Mol Hum Reprod 7, 881–886.[Abstract/Free Full Text]

Taylor AH and Al-Azzawi F (2000) Immunolocalisation of oestrogen receptor beta in human tissues. J Mol Endocrinol 24, 145–155.[Abstract/Free Full Text]

Vegeto E, Bonincontro C, Pollio G, Sala A, Viappiani S, Nardi F, Brusadelli A, Viviani B, Ciana P and Maggi A (2001) Estrogen prevents the lipopolysaccharide-induced inflammatory response in microglia. J Neurosci 21, 1809–1818.[Abstract/Free Full Text]

Witz CA (2002) Pathogenesis of endometriosis. Gynecol Obstet Invest 53, 52–60.[CrossRef][ISI][Medline]

Submitted on October 21, 2004; resubmitted on November 30, 2004; accepted on December 3, 2004.