Isolation of rat chromosome-specific paint probes by bivariate flow sorting followed by degenerate oligonucleotide primed-PCR

Angiogenesis is an essential component of endometrial regeneration after menses in preparation for implantation. Vascular endothelial growth factor (VEGF) is a secreted angiogenic peptide with mitogenic activity specific for endothelial and trophoblast cells. VEGF-immunoreactivity was detected in glandular epithelium throughout the menstrual cycle by immunohistochemistry, but, showed cyclic variation in the stroma and the blood vessels. During the early proliferative phase, strong staining was seen in the glandular epithelial cells while staining in the stroma was confined to a subpopulation of stromal cells and endometrial blood vessels appeared negative. In contrast, very intense staining of the endometrial stromal cells was seen in the mid proliferative endometrium possibly due to increased synthesis of VEGF by oestrogen. In the late proliferative endometrium, staining was seen in the endothelial cells and the perivascular stromal cells around the endometrial blood vessels. The greatest degree of immunostaining of stromal cells was observed in the mid to late proliferative endometrium. Throughout the secretory phase no staining was seen around the endometrial blood vessels and staining of endometrial stromal cells was confined to early secretory endometrium. In the late secretory endometrium only the glands were positive to VEGF antibody. The observed increase in the immunostaining of stroma suggests increased production of VEGF from early to mid and late proliferative endometrium which parallels the increase in the oestradiol levels in the proliferative phase of the menstrual cycle. It is proposed that VEGF may serve as a paracrine mediator of the effects of ovarian steroids on endometrial vascular development.     

Studies on the mechanism of embryo implantation

Implantation is a complex process accomplished by synchronization and interactions between embryo and endometrium by local exchange of signals including a number of cytokines and growth factors and direct cell-cell and cell-matrix contact. However, the research in early events of human implantation is still in its infancy. This presentation comprises the results of our attempts to investigate the mechanisms of human implantation process in its early stage by cell-biological method, including establishment of experimental implantation model in vitro. 1. Human trophoblast of early stage of gestation showed active cell locomotion, active endocytosis, and invasion of endometrial cell monolayer in mixed cultures. Trophoblast invasion was later arrested by transformed endometrial cells similar to decidual cells in vivo. These results appeared to indicate the interactions between trophoblast and endometrial cells in implantation. 2. Coculture system of rabbit preimplantation blastocyst and endometrial epithelium reformed from isolated endometrial epithelial cells on basement membrane matrix (Matrigel) simulated the in vivo rabbit implantation processes. This coculture system may provide a useful experimental implantation model. 3. A human trophoblast cell line was established from chorionic tissues of normal early pregnancy. These cells were cytotrophoblast-like morphology and endocrine functions. They formed the villous structures similar to those in vivo in culture on Matrigel and invasion of Matrigel was observed. These indicated the extracellular matrix may affect the morphology and function of invading trophoblast in implantation site. 4. Human endometrial epithelial single cells were cultured on Matrigel. Reconstruction of gland followed by epithelium formation quite similar to in vivo structures by migration and proliferation of isolated cells was demonstrated. Height of gland was promoted by estrogen and initiation of epithelization was upregulated by platelet-derived growth factors. This system revealed the extracellular matrix regulated morphogenesis of endometrial epithelium in vivo and is an essential substrate in experimental implantation model of endometrial epithelium. 5. Parallel cultures of endometrial epithelial cells on Matrigel were carried out with the IVF. ET patients to evaluate the endometrial morphology at time of ET. Endometrial cultures were initiated in previous cycles on Matrigel and the sera of patients were added to her own cultures from 1st day of IVF treatment cycle. Evaluation of reformed epithelium revealed the apparently unsuitable morphology for implantation in group of patients who eventually failed in pregnancy. This system may provide a useful measures in evaluation of endometrial receptivity and modality of treatment for endometrial aberrations. 6. Cyclic changes of extracellular matrix components in endometrium were investigated. Collagen I, III, IV, V were immunohistochemically estimated. Relative levels of all types of collagen except for collagen V declined at early secretory phase. In rodents, not only collagen but also laminin and fibronectin levels declined at early secretory phase. These changes may facilitate trophoblast invasion of endometrium. Collagen V distributed in myometrial surface was found to consist of subunit (alpha 1)2 alpha 2 and trophoblast growth was inhibited on substrate of alpha 1 subunit. Collagen V in myometrial surface may have a role in blocking trophoblast invasion. 7. HGF (hepatocyte growth factor) mRNA was demonstrated to be expressed during menstruation and secretory phase in endometrium distinctly and its receptor in endometrial epithelial cells and decidual cells. Positive correlation between plasma HGF levels and ultrasonographic thickness of endometrium was observed at late secretory phase. Recombinant HGF promoted proliferation of endometrial epithelial cells and decidual cells and upregulated initiation of endometrial epithelization of Matrigel.     

Apnea following spinal anaesthesia in two former pre-term infants

Telomerase activity is associated with the proliferative activity of cells. In the endometrium, telomerase activity is higher in the proliferative phase than in the secretory phase of the menstrual cycle, suggesting that telomerase activity may occur primarily in the glandular epithelial cells. To test this, a dissociated cell culture of the endometrium was performed, and the telomerase activity in each cell fraction was analysed. Telomerase activity was found in all 10 endometrial tissues of the proliferative phase of the menstrual cycle. Both the fragments of epithelial glands and single cells, which were prepared by enzymatic dissociation, showed telomerase activity. In the 7 day cell culture, it was found in nine out of 10 epithelial cell enriched fractions, but in none of the stromal cell enriched fractions. Flow cytometric analysis showed that the epithelial enriched fraction was contaminated with a predominant number of stromal cells, while the stromal cell enriched fraction was comprised mostly of stromal cells with apparent proliferative activity. Our results suggest that telomerase activity of the endometrium occurs primarily in the epithelial cells in the endometrium and that the stromal cells do not express telomerase activity regardless of their potent proliferative activity.     

Immunolocalization of inhibin and activin subunits in human endometrium across the menstrual cycle

Inhibin/activin alphaC/alphaN and betaA subunits were localized immunohistochemically in the human endometrium throughout the menstrual cycle using an affinity-purified sheep polyclonal antibody raised against the alphaC/alphaN subunit and an affinity-purified rabbit polyclonal antibody raised against the betaA subunit. The betaB subunit was below the level of detection in all human endometrial samples tested. Immunoreactive inhibin alphaC/alphaN subunit was localized in the luminal epithelium, glandular epithelium, stromal tissues and vascular endothelium with no significant variation across the normal menstrual cycle. Immunoreactive betaA subunit, common to inhibin A and activins AA and AB was localized in the luminal and glandular epithelium and in migratory cells while the endometrial stromal cells, decidua, vascular smooth muscle and endothelium were devoid of immunoreactivity. A significant variation of immunoreactive betaA subunit was observed in glandular and luminal epithelium across the normal menstrual cycle. In proliferative endometrium, only a very low level of betaA immunostaining was seen in luminal and glandular epithelium, while the luminal epithelial staining increased significantly in the early secretory phase and remained relatively constant over the rest of the menstrual cycle. A progressive increase in betaA immunoreactivity was observed also in the glandular epithelium during the secretory phase reaching a maximum in the late secretory phases, and decreasing at menstruation. Co-localization studies on serial sections suggested that the migratory cells expressing strong betaA immunoreactivity were macrophages and neutrophils but not eosinophils or mast cells. Thus, cells within the human endometrium are capable of expressing inhibin/activin molecules in vivo. The variation in the pattern of secretion of the betaA subunit across the menstrual cycle suggests that activin peptides may have a physiological role in endometrial function.     

Expression of transforming growth factor-beta 1 messenger ribonucleic acid and the modulation of deoxyribonucleic acid synthesis by transforming growth factor-beta 1 in human endometrial cells

OBJECTIVE: The purpose of this study was (1) to evaluate the potential sites of transforming growth factor-beta 1 synthesis in human endometrium by analyzing separated endometrial glands and stromal cells for transforming growth factor-beta 1 messenger ribonucleic acid by Northern analysis of total ribonucleic acid and (2) to investigate the effects of transforming growth factor-beta 1 on deoxyribonucleic acid synthesis in endometrial epithelial and stromal cells in culture. STUDY DESIGN: Endometrial glands and stroma from proliferative and secretory endometrium were isolated after collagenase treatment of endometrial tissue minces and were analyzed for transforming growth factor-beta 1 messenger ribonucleic acid by Northern analysis. We studied the effects of estradiol-17 beta and transforming growth factor-beta 1 on deoxyribonucleic acid synthesis in endometrial epithelium and transforming growth factor-beta 1 on stromal cells in culture by evaluating tritiated thymidine incorporation into trichloroacetic acid-precipitable material. RESULTS: Transforming growth factor-beta 1 messenger ribonucleic acid was detected for Northern analysis in separated endometrial stromal cells in levels that were greatest during the secretory phase and in greater levels than in epithelial cells from that same tissue. Transforming growth factor-beta 1 messenger ribonucleic acid in glandular epithelium in culture was not increased to detectable levels by treatment with transforming growth factor-beta 1. Deoxyribonucleic acid synthesis in endometrial glandular epithelium was inhibited by transforming growth factor-beta 1, but transforming growth factor-beta 1 stimulated deoxyribonucleic acid synthesis in endometrial stromal cells in culture. After treatment for 5 days with estradiol-17 beta (10(-8) mol/L), deoxyribonucleic acid synthesis in endometrial glands in culture was decreased by 40%. Transforming growth factor-beta 1 (1 ng/ml) did not alter this effect of estradiol-17 beta on deoxyribonucleic acid synthesis. CONCLUSIONS: Transforming growth factor-beta 1 acts to decrease deoxyribonucleic acid synthesis in epithelial cells and to increase it in stromal cells isolated from human endometrium and maintained in monolayer culture. Transforming growth factor-beta 1, potentially of stromal cell origin, could participate in the regulation of endometrial cell proliferation and differentiation in vivo.     

Immunohistochemical characterization of proliferation, oestrogen receptor and progesterone receptor expression in endometriosis: comparison of eutopic and ectopic endometrium with normal cycling endometrium

Recent studies examining oestrogen and progesterone receptor status and the proliferative activity of endometriotic lesions have produced conflicting reports. This study aimed to clarify the receptor status and proliferative activity of eutopic and ectopic endometrium from women with endometriosis and endometrium from normal women. Progesterone and oestrogen receptor expression and proliferative activity were studied in eutopic and ectopic endometrium from 30 women with endometriosis and in endometrium from 30 normal cycling women using microwave-pretreated paraffin-embedded sections stained with an avidin-biotin peroxidase technique. Progesterone and oestrogen receptor expression in the control endometrium did not differ from that of eutopic endometrium from women with endometriosis. Oestrogen receptor expression in ectopic endometrium increased from the proliferative to the late secretory phase. Epithelial progesterone receptor expression decreased during the cycle. Oestrogen receptor expression in both epithelium and stroma of ectopic endometrium was significantly higher than in eutopic endometrium throughout the cycle. In contrast, stromal progesterone receptor expression tended to be reduced in ectopic endometrium compared with eutopic tissue. Epithelial progesterone receptor expression was increased in ectopic endometrium but only in the late secretory phase. Although proliferative activity in the epithelium of control and eutopic endometrium was reduced from the proliferative to the late secretory phase, stromal activity did not vary. The proliferative activity in ectopic endometrium remained low and constant throughout the cycle. In the proliferative and early secretory phases, the proliferative activity of eutopic endometrium was increased compared with ectopic endometrium, but in the late secretory phase, levels were comparable. These findings challenge previous reports which have suggested that oestrogen receptors are reduced in ectopic tissue. This may have clinical implications for the development of novel treatments for endometriosis.     

Identification and characterization of cytosolic sulfotransferases in normal human endometrium

Understanding the factors which alter estrogen metabolism and activity in endometrial tissue is important because unopposed estrogen stimulation is an important risk factor in the development of endometrial carcinoma. The cyclic progression of the endometrium through proliferative and secretory phases is normally under the control of the ovarian hormones beta-estradiol (E2) and progesterone. One mechanism by which progesterone inhibits the activity of E2 in secretory endometrium is by elevating the degree of E2 sulfation, thereby reducing its ability to bind to the estrogen receptor and elicit a cellular response. Our laboratories have investigated the cytosolic sulfotransferases (STs) found in biopsies of both proliferative and secretory endometrium obtained from five normal pre-menopausal women who were not taking any drugs or steroids. Two of the human cytosolic STs were detected in human endometrial tissues. The phenol-sulfating form of phenol ST (P-PST) was found at varying levels in cytosol from both proliferative and secretory endometrium in all of the women studied but with no consistent correlation to the phase of the menstrual cycle. In contrast, estrogen ST (EST) was not detected in the proliferative endometrial cytosol of any of the women studied but was consistently found in all of the secretory endometrial cytosols. The presence and levels of these STs was confirmed by ST activity studies, immunoblot analysis and Northern blot analysis. These results indicate that the expression of EST in human endometrial tissues varies with the phase of the menstrual cycle and is most likely regulated by progesterone secreted from the ovaries.     

Tc-99m DTPA renal scintigraphy using deconvolution analysis with six functional images of the mean time to evaluate acute pyelonephritis

The objective of the present study was to evaluate the cyclic changes and regional localization of immunoreactive c-fos and prolactin (PRL) in the human endometrium, using immunohistochemistry. Immunoreactive PRL was found in the epithelium of 9.1% of the proliferative specimens and in 55.6% of the secretory specimens (p < 0.05, Fisher's exact test). In the endometrial stroma, immunoreactive PRL was present in 9.1 and 66.7% of the proliferative and secretory samples, respectively (p < 0.01). Immunoreactive c-fos predominated in the stroma and was identified in 54.5% of the specimens in the proliferative phase, but in only 7.1% of those in the secretory phase (p < 0.05). The progesterone/estradiol ratio was lower in the patients expressing immunoreactive c-fos (median = 13.1 ng/ml) compared to those who did not (median = 84.5 ng/ml, p < 0.05). We conclude that immunoreactive c-fos is found mostly in stromal cells during the proliferative phase of the menstrual cycle, and is sharply reduced during the secretory phase, when the endometrium is under progesterone stimulation - attested by PRL production.     

Sex-role reversal and the absence of extra-pair fertilization in Wilson''s phalaropes

Human endometrial leukocytes undergo regular cyclical changes during the menstrual cycle, with a striking increase in the phenotypically unusual population of CD56+ CD16- endometrial granulated lymphocytes (eGLs) in the late secretory phase and early pregnancy. The factors that regulate this increase in eGL numbers are unclear; their unusual morphology, however, has led to the suggestion that they undergo apoptosis at the end of the menstrual cycle. Apoptosis, bcl-2 expression, and proliferative activity were examined in the stroma of normal cycling, progesterone-treated, and early-pregnancy endometrium. The expression of bcl-2 and the Ki67 proliferation marker by highly purified (> 98% CD56+) eGLs from endometrium during the menstrual cycle and from first-trimester decidua was also studied. Apoptotic cells were rarely observed in the endometrial stroma of any of the samples examined. Stromal bcl-2 expression, however, increased from the proliferative to the premenstrual phase, and double immunohistochemical labeling demonstrated large numbers of bcl-2+ CD56+ eGLs. In contrast, Ki67 expression was high in the endometrial stroma during the proliferative phase, fell during the secretory phase, and rose again premenstrually, because of expression by eGLs. Isolated CD56+ eGLs also showed high bcl-2 and Ki67 expression at the end of the menstrual cycle. Unlike premenstrual endometrium, progesterone-treated endometrium and first-trimester decidua contained few proliferating cells, expressed high levels of bcl-2, and showed no evidence of apoptosis. Thus, eGLs do not undergo apoptosis in premenstrual endometrium, and their regulatory mechanisms remain to be clarified.     

Involvement of cyclin-dependent kinase inhibitor p27Kip1 in growth inhibition of endometrium in the secretory phase and of hyperplastic endometrium treated with progesterone

A cyclin-dependent kinase (cdk) inhibitor, p27Kip1 (p27), binds to the cyclin E-cdk2 complex and functions as a suppressor of cell cycle promotion. Here, the involvement of p27 in the growth of normal human endometrium was immunohistochemically studied, and the findings were compared with those of Ki-67, cyclin E and cdk2. In addition, to elucidate the effect of progesterone on the expression of p27, tissues from patients with endometrial hyperplasia were examined before and after the administration of medroxyprogesterone acetate (MPA) for the treatment of this disease. In the glandular cells of the normal endometrium, p27 was negligible during the proliferative phase, whereas it was markedly increased in the secretory phase. The staining pattern of Ki-67 was the reverse. Cyclin E/cdk2-positive cells were observed throughout the menstrual cycle. In the secretory phase, the cyclin E/cdk2-positive cells were also positive for p27, suggesting an interaction between these molecules. Stromal cells, especially in the basalis, showed a consistent expression of p27 throughout the menstrual cycle. The expression of p27 in hyperplastic epithelia before the MPA treatment was negligible, whereas it was greatly increased after the treatment. The Ki-67 positivity decreased after the treatment. These findings suggest that p27 is involved in the progesterone-induced growth suppression of normal and hyperplastic endometria.     

Iron deposition at mineralization fronts and osteoid formation following chronic cadmium exposure in ovariectomized rats

Integrins are heterodimeric glycoproteins that have been found to undergo dynamic temporal and spatial changes in distribution in the endometrium during the menstrual cycle in women. Likewise the extracellular matrix (ECM) ligands for these receptors are likely to play a role in the establishment of a receptive endometrium. To develop primate models to study the role of these molecules in the cascade of molecular events leading to implantation, integrin expression and associated changes in ECM were investigated during the menstrual cycle and in early pregnancy in the baboon. Antibodies specific for the integrins (alpha(1-6) and alpha(v); beta1, beta3, and beta4) and ECM (laminin, collagen IV, fibronectin) were utilized. In addition, cytokeratin and alpha-smooth muscle actin were used as epithelial, stromal, and smooth muscle cell markers, respectively. Endometrium was obtained in duplicate or triplicate during the menstrual cycle and early pregnancy. Changes observed during the natural menstrual cycle were confirmed using ovariectomized, steroid-treated animals. Constitutively expressed integrins on the endometrial epithelium included the collagen/laminin receptors: alpha2, alpha3, alpha6, and beta4. The pattern of expression correlated well with the distribution of ECM in this tissue. Collagen IV was confined to the basement membrane of glandular epithelium and blood vessels. Laminin immunostaining was found in the basement membrane, mostly in the stroma of the basal region, in the glandular endometrium and vasculature. Fibronectin was present throughout the stroma but not in the basement membrane. The collagen receptor alpha1 beta1 and fibronectin receptor alpha4 beta1 appeared in the glandular epithelium in the luteal phase. As in the human, alpha1 and alpha4 disappeared from the glandular epithelium with the establishment of pregnancy. In contrast, the alpha4 beta3 vitronectin receptor appeared in the glandular epithelium only in pregnancy or following long-term steroid treatment with estrogen and progesterone but not during the time of uterine receptivity associated with the initial period of embryo attachment. Osteopontin, an ECM ligand for alpha(v) beta3, was coexpressed with this integrin in invading cytotrophoblasts, glandular epithelium, and decidualizing stromal cells. Decidualization in the baboon was associated with changes in integrin expression similar to those found in humans: there was an increase in alpha1, alpha3, alpha6, beta1, and alpha(v) beta3 in the decidualized stromal cells. Laminin and collagen IV expression also increased at the implantation site and throughout the endometrium. In contrast, fibronectin expression was most evident at the implantation site and corresponded to alpha5 expression on the invading cytotrophoblasts. In summary, marked similarities were found in the expression of ECM and the integrin receptors between the baboon and the human endometrium throughout the menstrual cycle and in pregnancy. Cycle-specific integrins, alpha1, and alpha4, were present on epithelial cells during the secretory phase. Delayed expression of alpha(v) beta3 in baboon endometrial glands correlated closely with the time of enhanced glandular secretory activity in this primate. The baboon appears to be an excellent model for the investigation of the role of integrins and ECM leading to successful implantation.     

Immunohistochemical localization of extracellular matrix proteins in luteal phase endometrium of fertile and infertile patients

The lack of expression of certain components involved in cell adhesion and migration is believed to contribute to endometrial dysfunction and implantation failure. The purpose of this study was to investigate whether luteal phase endometrium in women with unexplained infertility differs, with respect to specific extracellular matrix (ECM) proteins, from endometrium of normal fertile women. A panel of monoclonal antibodies to collagen type IV, fibronectin and laminin was used to characterize the localization of ECM components in the different endometrial compartments. Precisely timed endometrial biopsies obtained at 4, 7, 10 and 13 days following the luteinizing hormone surge were obtained from 22 normal fertile women (group 1) and 24 women suffering from unexplained infertility (group 2). Paraffin-embedded sections were labelled using the streptavidin-biotin alkaline phosphatase technique. In group 1, collagen type IV, fibronectin and laminin were absent from the luminal epithelium but present in stromal cells and the basement membrane of glands and blood vessels. In group 2, these components were absent from all endometrial regions using equivalent titres of antibody to those used in group 1. This suggests that the endometrium of women with unexplained infertility demonstrates defects in the distribution of certain ECM glycoproteins. A possible consequence of this defect may be implantation failure.     

Expression of telomerase activity in human endometrium is localized to epithelial glandular cells and regulated in a menstrual phase-dependent manner correlated with cell proliferation

Telomerase activity is observed in most malignant tumors and germ cells, whereas normal somatic cells usually do not express it. Human endometrium is composed of glandular and stromal components and exhibits dramatic changes in proliferative activity during the menstrual cycle, which is exquisitely regulated by estrogen function. We previously reported that normal human endometrium expresses telomerase activity. However, it remains unclear which of the above components are the major sources of telomerase activity and how levels of telomerase activity are regulated over the menstrual cycle. Quantitative analysis of telomerase activity revealed that it changes dramatically over the course of the menstrual cycle and is strictly regulated in a menstrual-phase-dependent manner. Maximal activity equivalent to that in endometrial cancer was present in late proliferative phase, and minimal activity in late secretory phase. Postmenopausal endometrium and endometrium treated with anti-estrogen drugs exhibited decreased telomerase activity. Testing isolated epithelial glandular cells and stromal cells, we found that telomerase activity was localized to epithelial glandular cells. In situ RNA hybridization analysis also revealed epithelial-specific expression of human telomerase RNA. In vitro analysis of cultured epithelial cells demonstrated that telomerase activity is correlated with epithelial proliferation but not affected by estrogen treatment. These findings suggest that expression of telomerase activity is specific to epithelial cells and linked to cell proliferative status. The involvement of estrogen in telomerase regulation remains to be elucidated.     

Expression of katG in Mycobacterium tuberculosis is associated with its growth and persistence in mice and guinea pigs

The integrin alpha(v)beta3 functions in both cell-cell and cell-extracellular matrix adhesion, and has reported roles in platelet aggregation, immune function, tissue repair, tumour invasion, angiogenesis and uterine receptivity. The aim of this study was to use immunohistochemistry to describe the vascular and glandular expression of integrin alpha(v)beta3 in formalin fixed, paraffin embedded endometrium obtained from women with (n = 29) and without (n = 24) endometriosis. The results showed a significant increase in the percentage of vessels expressing alpha(v)beta3 in the endometrium of women with endometriosis compared with controls (P = 0.0001). This difference was more pronounced in the secretory phase (P = 0.001) than the proliferative phase (P = 0.016). There was no correlation between vascular alpha(v)beta3 expression and the endothelial cell proliferation index (P > 0.05). Vascular sprouts were not observed in any of the 53 endometrial tissues obtained from women with or without endometriosis throughout the menstrual cycle. Results from semi-quantitative scoring of gland immunostaining showed that neither controls (P = 0.3329) nor the endometriosis group (P = 0.2260) had any significant changes in terms of alpha(v)beta3 expression between the different stages of the menstrual cycle. There was also no difference in glandular alpha(v)beta3 expression between women with and without endometriosis (P = 0.4302). These results provide evidence for increased endometrial angiogenesis in women with endometriosis compared with controls, and suggest that glandular expression of alpha(v)beta3 is not related to uterine receptivity per se.     

Vascular endothelial growth factor (VEGF) in endometriosis

Angiogenesis is likely to be involved in the pathogenesis of endometriosis. According to the transplantation theory, when the exfoliated endometrium is attached to the peritoneal layer, the establishment of a new blood supply is essential for the survival of the endometrial implant and development of endometriosis. From the known angiogenic factors, vascular endothelial growth factor (VEGF) has emerged as a pivotally important regulator of normal angiogenesis and pathological neovascularization. The VEGF protein was evaluated immunohistochemically in the eutopic endometrium of 10 women without endometriosis (group I) at laparoscopy and the eutopic endometrium and peritoneal endometriotic lesions of 43 women with endometriosis (group II). VEGF histological scores were 9.7 +/- 4.3 and 4.0 +/- 2.6 respectively in the epithelium and stroma of the eutopic endometrium of group I women, and 10.3 +/- 2.3 and 3.6 +/- 2.3 respectively in women of group II. In red lesions, the VEGF scores were 11.1 +/- 3.0 in the epithelium and 5.1 +/- 3.0 in the stroma, and in black lesions were 8.6 +/- 2.7 and 1.6 +/- 1.6, respectively. Significantly lower values were observed in black lesions as compared with eutopic endometrium and red lesions, the values of which were similar. Scores were also evaluated according to the phase of the cycle. In eutopic as well as ectopic endometrium, no significant cyclic variations were observed throughout the cycle. However, VEGF content was found to be higher in the eutopic glandular epithelium of women with endometriosis during the late secretory phase, possibly suggesting a more likely tendency to implant. In contrast, significantly higher VEGF content was noted in red lesions as compared with black lesions. During all phases of the cycle, the VEGF content in stromal cells of red lesions was higher than in black lesions. Similarities in VEGF content were observed in the glandular epithelium of the eutopic endometrium of women with endometriosis and red lesions, suggesting that endometriosis probably arises from the peritoneal seeding of viable endometrial cells during retrograde menstruation and that red lesions can be considered as the first stage of implantation. After the attachment phase, the high VEGF levels could provoke an increase in the subperitoneal vascular network and facilitate implantation and viability in the retroperitoneal space. Lower VEGF levels in black lesions explain the decrease in both stromal vascularization, followed by fibrosis and inactivation of the implant.