Sébastien Colette

Endometriomas as a possible cause of reduced ovarian reserve in women with endometriosis.

OBJECTIVE To evaluate the adverse effects of endometriomas on ovarian reserve. DESIGN Analysis of prospectively collected biopsy samples. SETTING Gynecology research unit in a university hospital. PATIENT(S) Women younger than age 35 years with endometriomas. INTERVENTION(S) Biopsy of normal cortex from ovaries affected by endometriomas (≤4 cm) and contralateral ovaries without cysts. MAIN OUTCOME MEASURE(S) Presence of cortex-specific stroma, observation of superficial endometriosis, follicular density, and presence of fibrosis. RESULT(S) Twenty samples of cortical tissue from ovaries with endometriomas and 11 from contralateral ovaries without cysts were analyzed. Follicular density was significantly lower in cortex from ovaries with endometriomas than in cortex from contralateral ovaries without cysts (mean ± SD = 6.3 ± 4.1/mm(3) vs 25.1 ± 15.0/mm(3)). Eleven (55%) cortical samples from ovaries with endometriomas showed fibrosis and concomitant loss of cortex-specific stroma, not observed in contralateral normal ovaries. Multivariate analysis revealed that the presence of endometrioma and fibrosis were significantly and independently associated with follicular density. CONCLUSION(S) Endometriotic cyst formation and associated structural tissue alterations in apparently normal ovarian cortex may be a cause of reduced ovarian reserve. Early diagnosis and intervention may be beneficial in women with endometriomas to protect their ovarian function.

Potential involvement of iron in the pathogenesis of peritoneal endometriosis

The aim of this study is to review the current literature associating endometriosis with iron and to discuss the potential causes and consequences of iron overload in the pelvic cavity. Indeed, iron is essential for all living organisms. However, excess iron can result in toxicity and is associated with pathological disorders. In endometriosis patients, iron overload has been demonstrated in the different components of the peritoneal cavity (peritoneal fluid, endometriotic lesions, peritoneum and macrophages). Animal models allow us to gather essential information on the origin, metabolism and effect of iron overload in endometriosis, which may originate from erythrocytes carried into the pelvic cavity mainly by retrograde menstruation. Peritoneal macrophages play an important role in the degradation of these erythrocytes and in subsequent peritoneal iron metabolism. Iron overload could affect a wide range of mechanisms involved in endometriosis development, such as oxidative stress or lesion proliferation. In conclusion, excess iron accumulation can result in toxicity and may be one of the factors contributing to the development of endometriosis. Treatment with an iron chelator could thus be beneficial in endometriosis patients to prevent iron overload in the pelvic cavity, thereby diminishing its deleterious effect.

Absence of aromatase protein and mRNA expression in endometriosis.

BACKGROUND Aromatase has been reported to be involved in estrogen biosynthesis and expressed in eutopic and ectopic endometrium of endometriosis patients. The objective of the present study was to investigate its expression and localization in three distinct types of endometriosis. METHODS Human peritoneal, ovarian and rectovaginal endometriotic lesions and matched eutopic endometrium were collected from patients during laparoscopy. Aromatase protein localization (immunohistochemistry, n = 63) and mRNA expression [quantitative polymerase chain reaction (Q-PCR), n = 64] were assessed. RESULTS No aromatase protein was detected by immunohistochemistry in either the glandular or stromal compartment of endometriotic lesions or eutopic endometrium, while it was strong in placental syncytiotrophoblasts, granulosa and internal theca cells from pre-ovulatory follicles, and luteal cells from corpus luteum. By Q-PCR, low but discernible levels of aromatase expression were found in endometriomas, probably due to follicular expression. Transcripts for aromatase were barely detectable in only a few peritoneal and rectovaginal endometriotic lesions, and a few eutopic endometrium samples, probably due to contaminating surrounding tissues (adipose tissue, intact peritoneum). CONCLUSIONS Unlike previous studies, we observed no aromatase protein in any of the endometriosis types, and barely detectable aromatase mRNA expression, suggesting that locally produced aromatase (within endometriotic lesions) may be less implicated in endometriosis development than previously postulated. Potential factors responsible for these discrepancies are discussed.

Peritoneal endometriosis is an inflammatory disease.

Peritoneal endometriosis is a chronic inflammatory disease characterized by increased numbers of peritoneal macrophages and their secreted products. Inflammation plays a major role in pain and infertility associated with endometriosis, but is also extensively involved in the molecular processes that lead to peritoneal lesion development. Peritoneal oxidative stress is currently thought to be a major constituent of the endometriosis-associated inflammatory response. Excessive production of reactive oxygen species, secondary to peritoneal influx of pro-oxidants such as heme and iron during retrograde menstruation, may induce cellular damage and increased proinflammatory gene expression through nuclear factor-kappa B activation. In particular, prostaglandin biosynthetic enzyme expression is regulated by this transcriptional factor, and increased peritoneal prostaglandin concentrations have been demonstrated in endometriosis. In the light of available data collected from patient biopsies, as well as in vitro and in vivo studies, the respective involvement and potential molecular interactions of iron, nuclear factor-kappa B and prostaglandins in the pathogenesis of endometriosis are explored and discussed. The key role of peritoneal macrophages is emphasized and potential therapeutic targets are examined.

Iron storage is significantly increased in peritoneal macrophages of endometriosis patients and correlates with iron overload in peritoneal fluid

OBJECTIVE To further investigate peritoneal iron disruption in endometriosis by studying iron storage in peritoneal macrophages of patients with endometriosis compared with controls. DESIGN Cross-sectional study. SETTING Academic gynecology research unit in a university hospital. PATIENT(S) Fifty patients undergoing laparoscopy. INTERVENTION(S) Collection of peritoneal fluid samples (N = 50) from patients with (n = 27) and without (n = 23) endometriosis undergoing laparoscopy. MAIN OUTCOME MEASURE(S) Quantification of peritoneal macrophage ferritin by immunocytochemical staining and immunodensitometry and measurement of peritoneal iron, transferrin, ferritin, and prohepcidin concentrations. RESULT(S) The optical density of peritoneal macrophage ferritin staining was statistically significantly higher in endometriosis patients than in controls. Higher iron concentrations, transferrin saturations, and ferritin concentrations were also detected in case of endometriosis. A statistically significant positive correlation was found between the optical density of macrophage ferritin staining and peritoneal iron concentrations in endometriosis and control patients. CONCLUSION(S) Iron storage is statistically significantly increased in peritoneal macrophages of patients with endometriosis and correlates with iron overload in peritoneal fluid. The potential implications of iron accumulation in peritoneal macrophages in case of endometriosis are discussed.

Expression of eicosanoid biosynthetic and catabolic enzymes in peritoneal endometriosis

BACKGROUND Increased peritoneal eicosanoid concentrations have been reported in endometriosis patients and might be important in disease-associated pain and inflammation. Here, we evaluated the expression of key biosynthetic and catabolic enzymes involved in this abnormal eicosanoid production in peritoneal macrophages and endometriotic lesions. METHODS Peritoneal macrophages, endometriotic lesions and matched eutopic endometrium were collected from endometriosis patients (n = 40). Peritoneal macrophages and eutopic endometrium samples were also collected from disease-free women (n = 25). Expression of type IIA secretory phospholipase A(2) (sPLA(2)-IIA), cyclooxygenase-2 (COX-2), microsomal prostaglandin E synthase-1 (mPGES-1), 15-hydroxyprostaglandin dehydrogenase (15-PGDH) and 5-lipoxygenase (5-LO) was quantified by real-time PCR, and these five key enzymes were localized by immunohistochemistry. RESULTS sPLA(2)-IIA, COX-2 and mPGES-1 mRNA was significantly increased in peritoneal macrophages of endometriosis patients compared with controls (P = 0.006, P = 0.016 and P = 0.025, respectively). In endometriosis patients, sPLA(2)-IIA, mPGES-1 and 15-PGDH mRNA was significantly enhanced in peritoneal lesions compared with matched eutopic endometrium (P < 0.001, P < 0.001 and P = 0.005, respectively). In eutopic endometrium, a significant decrease in 15-PGDH mRNA was found in the endometriosis group compared with controls (P = 0.023). Finally, sPLA(2)-IIA, COX-2, mPGES-1 and 15-PGDH immunostaining was found mainly in endometrial glands, whereas 5-LO was distributed throughout the glands and stroma. CONCLUSIONS Our study highlights an imbalance between eicosanoid biosynthesis and degradation in endometriosis patients. Both peritoneal macrophages and endometriotic lesions may be involved. Research into new molecules inhibiting biosynthetic enzymes (such as sPLA(2)-IIA and mPGES-1) and/or activating catabolic enzymes (such as 15-PGDH) may prove to be a major field of investigation in the development of targeted medical therapies.

Inhibition of steroid sulfatase decreases endometriosis in an in vivo murine model

BACKGROUND Steroid sulfatase (STS) is involved in estrogen biosynthesis and expressed in eutopic and ectopic endometrium of disease-free and endometriosis patients. The present study was designed to investigate its role in endometriosis development. METHODS Human endometrial explants were cultured on inserts for 24 h to assess the effectiveness of an STS inhibitor (STS-I), estradiol-3-O-sulfamate (E2MATE), on STS activity in endometrial tissue. Endometriosis was induced in mice, and E2MATE (or vehicle alone) was given orally for 21 days. Plasma estradiol levels were measured, and STS activity was assessed in murine organs (uterus, liver and leukocytes) and in lesions. Lesion number, weight and size (morphometry) were quantified. Lesion STS and progesterone receptor (PR) expression, proliferation and apoptosis rates were determined by immunohistochemistry. RESULTS In vitro, addition of 1 µM E2MATE to the culture medium resulted in decreased STS activity in endometrial explants (P < 0.001). Treatment of mice with E2MATE (1.0 and 0.5 mg/kg) did not modify plasma estradiol levels, but inhibited STS activity in murine uterus (P < 0.05), liver (P < 0.001) and leukocytes (P < 0.001), as well as in induced lesions (P < 0.05). E2MATE reduced lesion weight (P < 0.01) and size (P < 0.05), but had no impact on proliferation or apoptosis rates, nor STS protein expression. Stromal edema was observed in the uterus of animals treated with E2MATE, but not in the stroma of lesions. Increased PR expression was detected in endometriotic lesions (P < 0.001). CONCLUSIONS E2MATE was shown to effectively inhibit STS activity in endometrial tissue in vitro. In vivo, E2MATE decreased endometriosis development without affecting systemic estradiol levels. Use of STS-I could therefore be of potential interest in endometriosis treatment.

Differential expression of steroidogenic enzymes according to endometriosis type.

OBJECTIVE To evaluate, in peritoneal, ovarian, and rectovaginal endometriotic lesions, expression of steroidogenic enzymes involved in the activation and inactivation of estrogens: 17β-hydroxysteroid dehydrogenase type 1 (HSD17B1) and 2 (HSD17B2), estrone sulfotransferase (EST), and steroid sulfatase (STS). SETTING Academic gynecology research unit. DESIGN Retrospective study. PATIENT(S) Disease-free (n = 41) patients and patients with endometriosis (n = 79) were included for quantitative polymerase chain reaction (q-PCR) (15 disease-free, 33 endometriosis) and immunohistochemistry (26 disease-free, 46 endometriosis) studies. INTERVENTION(S) Q-PCR and immunohistochemistry. MAIN OUTCOME MEASURE(S) Evaluation of mRNA and protein expression. RESULT(S) Glandular HSD17B1, HSD17B2, and STS protein expression were demonstrated. HSD17B2 mRNA values were higher in the secretory phase of the menstrual cycle in the endometrium of disease-free women, but not in the eutopic endometrium of patients with endometriosis. HSD17B1 mRNA was equally expressed in the various tissues investigated, and EST mRNA was expressed at low levels in the different lesion types. HSD17B2 mRNA expression was decreased in ovarian and rectovaginal endometriosis compared with eutopic endometrium, while STS mRNA was increased in rectovaginal lesions compared with ovarian lesions. Ratios between pro- and antiestrogenic enzymes (STS/EST and HSD17B1/HSD17B2) were more in favor of estrogens in ovarian and rectovaginal endometriosis. CONCLUSION(S) In endometriosis development, local activation of estrogens appears to be important. STS and HSD17B1 inhibitors may therefore prove useful to treat the disease.

Induction of endometriotic nodules in an experimental baboon model mimicking human deep nodular lesions

OBJECTIVE To establish an experimental model for the study of deep nodular endometriosis. DESIGN Induction of nodular endometriosis in baboons by grafting different uterine specimens to the peritoneal cavity. SETTING Research and university facilities. ANIMAL(S) Ten baboons, to develop a model of induced deep nodular endometriosis. INTERVENTION(S) Biopsies of endometrium, and endometrium plus the junctional zone (JZ), full uterine thickness, and myometrium grafted to the peritoneum. MAIN OUTCOME MEASURE(S) Macroscopic descriptions recorded for observed induced lesions; staining with hematoxylin and eosin for histological evaluation and specific antibodies (CK22, CD10) for immunohistochemical studies; and analysis of surface area and volume of lesions, glandular density, and invasion of surrounding organs. RESULT(S) The incidence of induced nodular endometriosis was 100%, but the extent depended on the tissue grafted. Lesions induced after grafting specimens containing the JZ were statistically significantly larger than those not containing the JZ. Surrounding organ invasion was reported in more than 40% of lesions after grafting specimens containing the JZ. CONCLUSION(S) The first experimental model of nodular endometriosis allows investigation of deeper nodular lesions as well as invasion phenomena associated with nodular lesions.

Nerve fibers are absent in disease-free and eutopic endometrium, but present in endometriotic (especially deep) lesions

Objective Detection of nerve fibers in endometrial biopsies was recently proposed as a noninvasive diagnostic tool for endometriosis. However, their occurrence in the functional layer of endometrium still remains controversial. Nerve fibers were found to be present in endometriotic lesions themselves, which may account for some of the pain experienced by patients, but their origin is not clear. The objective of the present study was to reevaluate the presence of nerve fibers in endometrium and in different types of endometriotic lesions. Patients and Methods Nerve fiber density (PGP9.5 immunohistochemical analysis), unmyelinated nerve fiber presence (neurofilament immunohistochemical detection) and nerve growth factor expression were evaluated in endometrial (disease free: n = 20; endometriotic: n = 26) and endometriotic (peritoneal lesions: n = 11; ovarian lesions: n = 16; rectovaginal lesions: n = 27) samples. Results Endometrial biopsies were found to be mostly negative for nerve fibers. Nerve fiber density was higher in deep nodular lesions than in peritoneal (p<0.01) or ovarian (p<0.001) lesions. Around 30% of PGP9.5-positive nerve fibers were confirmed by neurofilament staining. Nerve growth factor expression was detected at higher levels in the stroma of deep-infiltrating lesions (p<0.05). Conclusions No nerve fibers were detected in endometrial biopsies (from healthy or endometriosis patients). However, nerve fibers were detected in endometriotic lesions. Most of them were found to be unmyelinated, suggesting they could be implicated in pain. Deep nodular lesions may be more neuroattractive through the action of nerve growth factor.