Immunohistochemical localization of insulin-like growth factor binding protein-3 in eutopic and ectopic endometrial tissues

Transcription

1 FERTILITY AND STERILITY VOL. 72, NO. 6, DECEMBER 1999 Copyright 1999 American Society for Reproductive Medicine Published by Elsevier Science Inc. Printed on acid-free paper in U.S.A. Immunohistochemical localization of insulin-like growth factor binding protein-3 in eutopic and ectopic endometrial tissues Ali Akoum, Ph.D.,* André Lemay, M.D., Ph.D.,* Yves Lajeunesse,* Michel Marois, M.D., and Michael Koutsilieris, M.D., Ph.D. Pavillon Saint-François d Assise, CHUQ; and Pavillon CHUL, CHUQ, Québec, Québec, Canada Received March 19, 1999; revised and accepted July 16, This work was supported by a grant from the Medical Research Council of Canada (MA-1024) to André Lemay, Michael Koutsilieris and Ali Akoum. Ali Akoum is a chercheurboursier senior of the Fonds de la recherche en santé du Québec. Reprint requests: André Lemay, M.D., Ph.D., Unité d Endocrinologie de la Reproduction, Pavillon Saint-François d Assise, CHUQ 10, rue de l Espinay, Québec, Québec G1L 3L5 Canada (FAX: ; * Unité d Endocrinologie de la Reproduction, Pavillon Saint-François d Assise, CHUQ. Laboratoire de Pathologie, Pavillon Saint- François d Assise, CHUQ. Laboratoire d Endocrinologie Moleculaire, Pavillon CHUL, CHUQ /99/$20.00 PII S (99) Objective: To evaluate the expression of insulin-like growth factor binding protein-3 (IGFBP-3) in the eutopic endometrium and in endometriotic lesions. Design: Retrospective immunohistochemical study. Patient(s): Twenty-five normal women and 39 women with endometriosis. Intervention(s): Endometrial and endometriotic tissue biopsies obtained at laparoscopy. Main Outcome Measure(s): Expression of IGFBP-3 assessed by immunohistochemistry. Result(s): In the endometrium, positive immunostaining of IGFBP-3 was observed both in the stroma and the epithelial glands. The intensity of staining in the glands during the secretory phase was significantly higher in women with endometriosis compared with controls (P.018). An increased expression of IGFBP-3 over controls was found in stages I and II of the disease (P.018), whereas in stages III and IV, the difference between controls and women with endometriosis was not significant (P.300). In endometriotic tissues, a much-marked immunostaining of IGFBP-3 was noted in 90% of the glands and 67% of the stroma without apparent differences related to cycle phase. Conclusion(s): These data show intense staining of IGFBP-3 in endometriosis lesions and increased expression of the protein in the endometrium of patients with endometriosis compared to controls. This marked expression of IGFBP-3 could be related to its previous finding in the peritoneal fluid and to its potential involvement in the pathophysiology of endometriosis. (Fertil Steril 1999;72: by American Society for Reproductive Medicine.) Key Words: Endometriosis, endometrium, biopsy, IGFBP-3, immunohistochemistry Endometriosis, the growth of endometriallike tissue on the peritoneum and the organs of the pelvic cavity, depends on the secretion of sex-steroid hormones. There is cumulative evidence that the action of steroid hormones on the proliferation and differentiation of endometrial cells is mediated locally at least in part by growth factors (GFs). The expression, localization and steroid dependence of several GFs, including insulin-like growth factors (IGFs), have been demonstrated in the eutopic endometrium (1 5) and in endometriotic lesions (6, 7). Recent data also indicate the presence of numerous GFs, including IGFs, in the peritoneal fluid (PF) (8 10). To better understand the pathophysiology of endometriosis, we previously evaluated the mitogenic activity of the PF on various cell lines (11), described a purification procedure (12) and reported the finding of N-terminal truncated forms of IGF-binding protein-3 (IGFBP-3) that mediate an apparent preferential mitogenic action on endometrial epithelial cells (13). This last observation coincided with the recent documentation of the presence of IGFs, IGFBPs, and IGFBP-3 protease activity in the human PF (8, 14, 15). The IGFBPs are multifunctional proteins that regulate not only the transport of IGFs but also their presentation to cell surface receptors (16, 17). Recent data also indicate that IGFBP-3 can mediate direct stimulatory or inhibitory effects on several cell lines by IGFindependent mechanisms and specific binding sites (18 22). Tissue-specific regulation of IG- 1085

2 TABLE 1 Clinical characteristics of 25 normal women and 39 women with endometriosis. Study group Control Endometriosis Characteristic Endometrium biopsy (n 25) Endometrium biopsy (n 22) Endometriosis biopsy (n 17) Mean age ( SD) (y) Infertility (%) Endometriosis (no. of patients) Stages I II 13 8 Stages III IV 9 9 Cycle phase (no. of patients) Proliferative Secretory FBP-3 proteolysis may also provide a mechanism to increase bioavailability of IGFs, depending on local needs or activity (23). Our data also indicate a possible direct action of IG- FBP-3 fragments purified from PF on epithelial cells derived from the endometrium (13). More recently, we demonstrated an association between mitogenic activity and proteolysis of IGFBP-3 related to urokinase in the peritoneal fluid (10). The above literature suggests that IGFBP-3, as part of an IGF/IGFBP-3/protease system, could be involved in the development of endometriosis. According to the retrograde menstruation theory (24), IGFBP-3 in the PF could come at least in part from endometrial cells regurgitated through the fallopian tubes at the time of the menses or directly from endometriotic implants. The messenger RNA (mrna) expression has been already reported in the endometrium of normal women showing a 3.9-fold increase in the amount of the IGFBP-3 transcript in the secretory endometrium compared to the proliferative endometrium (25) and more focally concentrated in endometrial capillaries during the secretory phase (26). The enhancement in the amount of IGFBP-3 in the conditioned media of endometrial stromal cells in the presence of estradiol and progesterone also indicated sex steroid-dependence of IGFBP-3 protein synthesis (25). However, the presence of IGFBP-3 in endometriosis lesions and in the endometrium of patients with endometriosis has not been documented. This study was based on the concept that endometriosis involves tissue remodeling and on recent evidence indicating the presence of IGFs, IGFBP-3, and protease activity in the PF. The purpose of this study was to examine IGFBP-3 expression in endometriotic implants and in the endometrium of patients having or not having evidence of endometriosis at laparoscopy. The immunohistochemical data presented in this report provide evidence for marked expression of IGFBP-3 in endometriotic lesions and in endometrial glands of women with initial endometriosis stages I II during the secretory phase of the menstrual cycle. MATERIALS AND METHODS Patients The present study was approved by the ethics committee of the St-François d Assise Hospital. Patients included in this study had a laparoscopy for investigation of presumed endometriosis because of pelvic pain and/or infertility, for second-look laparoscopy of previously known endometriosis or for tubal ligation. Patients who were taking hormones or anti-inflammatory drugs or had a coexisting pelvic inflammatory condition were excluded. An informed consent form was signed at a visit 2 to 5 days before laparoscopy, where a blood sample was drawn for measurement of progesterone and elimination of possible pregnancy by ( -hcg). At the time of laparoscopy, a biopsy specimen was taken from the endometrial cavity or from an endometriotic implant whenever appropriate. Simultaneous biopsies from the endometrium and an endometriotic lesion could be achieved only occasionally. As indicated in Table 1, biopsy specimens were obtained from the endometrium of 25 women without visual evidence of endometriosis at laparoscopy (endometrium from the control group). Twenty-two of these patients (88%) had one to four pregnancies before. Among the 3 nulliparous women, only 1 desired a child; the 2 others consulted for pelvic pain and tubal ligation. Twenty women (80%) in this group were requesting a tubal ligation. Biopsy specimens were also obtained from the endometrium of 22 patients with endometriosis (endometrium from the endometriosis group). Twelve of these patients (55%) had one to five previous pregnancies, whereas 10 (45%) had not conceived at the time of the study. Ten of the 12 patients consulting for infertility had had no 1086 Akoum et al. IGFBP-3 staining in endometrial tissues Vol. 72, No. 6, December 1999

3 previous pregnancy, whereas 2 were pregnant once before. Since it was quite difficult to obtain a tissue biopsy specimen from the endometrium and an endometriosis lesion in the same patient, a tissue sample was also obtained from the endometriosis lesions of 17 patients presenting with various stages of endometriosis (endometriosis tissue group) according to the revised classification of the American Fertility Society (1985). In this group, 9 patients (53%) consulted for infertility, whereas 8 (47%) had been pregnant previously. Biopsy specimens from endometriotic tissues were taken from peritoneal implants (n 10), from an ovarian implant (n 2) or from the wall of endometrioma (n 5). The mean age was comparable among the three groups. According to cycle history, serum progesterone level and appearance of endometrium at histology, it was easy to correctly define the cycle phase (proliferative or secretory) at the time of biopsy of the endometrium tissue in every patient. The correct determination of cycle phase was more difficult in biopsies of endometriosis lesions and could be ascertained in half of the patients. Material Endometrial and endometriotic tissue specimens obtained at laparoscopy were rapidly frozen on dry ice in optimal cutting temperature compound (Miles Inc., Elkhart, IN) and stored at 80 C until further use. A portion of the endometrial tissue was dated according to the criteria of Noyes et al. (27). The endometriotic nature of biopsy specimens from implants was confirmed histologically by the presence of stroma and endometrial glands. Blood samples were obtained by phlebotomy and were collected in Vacutainer tubes (Becton-Dickinson, Franklin Lakes, NJ) for serum separation. Blood human chorionic gonadotropin and progesterone levels were determined by enzyme-linked immunosorbant assay on an automatic station ES-300 (Boehringer Mannheim, Montréal, Canada) to eliminate a possible early pregnancy and to confirm the phase of the menstrual cycle. Immunohistochemistry Serial 4 5- m cryosections, prepared from endometrial and endometriotic tissue biopsy specimens, were placed on poly-l-lysine coated glass microscope slides and fixed first in 4% formaldehyde solution (Fisher Scientific, Montréal, Québec, Canada) for 20 minutes at room temperature and then in cold acetone/methanol (35/15, vol/vol) for 20 minutes at 20 C. Thereafter, sections were treated with 0.3% H 2 O 2 in absolute methanol for 20 minutes at room temperature to block endogenous peroxidase. Immunostaining was performed using an affinity purified polyclonal goat anti IGFBP-3 antibody (primary antibody) (Diagnostic Systems Laboratories, Webster, TX), a biotin-conjugated mouse antigoat antibody (second antibody) (Jackson Immunoresearch, West Grove, PA), streptavidine peroxidase (Jackson Immunoresearch) and diaminobenzidine (DAB) (Sigma Chemical Co., St. Louis, MO) as chromogen. Briefly, after blocking non-specific binding of primary antibody with 5% bovine serum albumin (BSA) in phosphate-buffered saline (PBS) for 30 minutes at room temperature, sections were rinsed in PBS, incubated for 90 minutes at room temperature with an appropriate dilution of primary antibody (2.5 g/ml in PBS containing 1% BSA), rinsed in PBS and incubated for 60 minutes at room temperature with the second antibody (1:500). Sections were then rinsed in PBS and the avidin-biotinylated horseradish peroxidase complex was applied for 30 minutes. After a PBS rinse followed by a 3-minute incubation with DAB:H 2 O 2 (0.5 mg/0.03% H 2 O 2 in PBS), sections were washed in tap water, counterstained with hematoxylin and mounted with Mowiol (Calbiochem-Novabiochem Corp., La Jolla, CA). Sections incubated without the primary antibody or with non-immune goat serum were included as negative controls in all experiments. As another negative control, the anti IG- FBP-3 antibody was incubated with an excess of IGFBP-3 (12.5 g/ml) overnight at 4 C prior to incubation with endometrial or endometriotic tissue sections. The immunostaining reactions were processed in batches that included several tissue biopsies from each group and the negative controls. Slides were viewed using a Leica microscope (Leica Mikroscopie und systeme GmbH, Postfach, Wetzlar, Germany), and photomicrographs were made by Kodak 100 ASA film (Kodak, Toronto, Ontario, Canada). IGFBP-3 immunostaining was evaluated in a blinded fashion by two independent observers without knowledge of laparoscopic findings. To take into account the distribution and the intensity of staining, three different areas were randomly selected and scored at least three times using an arbitrary scale (0 absent, 1 light, 2 moderate, and 3 intense) (Fig. 1). Glands and stroma were evaluated separately. Statistical Analysis The IGFBP-3 staining scores are discrete data following an ordinal scale. Statistical analysis was performed using Fischer s exact test. Individual groups were compared using the same test (28). The Bonferroni procedure was applied when more than two groups were compared. Comparison of patient s age was performed using one-way analysis of variance. All analyses were performed using the Statistical Analyses System (SAS Institute, Inc., Cary, NC). Differences were considered as statistically significant for P.05. RESULTS Representative examples of IGFBP-3 immunostaining in the endometrium of women with or without endometriosis are shown in Figure 1. The purpose of this figure is to illustrate the scale of immunostaining; this set of micrographs show the different intensities of staining and the localization of IGFBP-3 throughout the cycle in a normal woman. Positive immunohistochemical staining of IGFBP-3 FERTILITY & STERILITY 1087

4 FIGURE 1 Immunohistochemical detection of IGFBP-3 in endometrial biopsies during the menstrual cycle ( 434). Endometrial sections were incubated with a goat polyclonal anti-igfbp-3 antibody (primary antibody), then with a biotin-conjugated mouse anti-goat antibody (second antibody), streptavidin peroxidase and DAB, respectively. Brown immunostaining was observed both in endometrial glands (g) and stroma (s). (A), Intensity 0, proliferative phase, day 7 of a normal woman. (B), Intensity 1, secretory phase, day 18 of a normal woman. (C), Intensity 2, secretory phase, day 21 of a normal woman. (D), Intensity 3, secretory phase, day 30 of a normal woman. is observed both in the stroma and epithelial glands of the endometrium. High intensity was mostly encountered during the secretory phase, showing tortuous glands and edematous stroma (D), whereas low intensity was generally more frequent during the follicular phase, containing small glands in a rather cellular stroma (B). This immunostaining was generally homogeneous, although there could be areas presenting different levels of staining. There was also positive staining of vessels occasionally seen in the overall observation of tissue sections, mainly in the secretory phase. As shown in Figure 2, the positive immunoreaction seen in the presence of anti-igfbp-3 antibody (A) was virtually absent when the primary antibody was replaced by an equal concentration of nonimmune goat immunoglobulins (B) or when it was preabsorbed with an excess of IGFBP-3 before incubation with endometrial section (data not shown). No immu Akoum et al. IGFBP-3 staining in endometrial tissues Vol. 72, No. 6, December 1999

5 FIGURE 2 Specificity of the immunoreaction observed with the goat anti-igfbp-3 polyclonal antibody in endometrial tissue ( 434) of a control woman. The intense immunoreaction shown by the anti-igfbp-3 antibody in endometrial gland (g) and stroma (s) (A) was virtually absent when this antibody was omitted (B). noreactivity was observed when the primary antibody was omitted (data not shown). Table 2 reports the distribution of patients with endometriosis and controls according to IGFBP-3 immunostaining score in the endometrium. During the proliferative phase, immunostaining was mostly negative or of low intensity compared to the more frequent and greater intensity immunostaining in the secretory phase. Among biopsy specimens obtained at different times during the secretory phase, an increase in intensity of immunostaining was found when progressing from early secretory phase specimens to late TABLE 2 Insulin-like growth factor binding protein-3 immunostaining score in the endometrium. Proliferative phase Secretory phase Variable No. of patients Signal intensity Signal intensity P value No. of patients P value Glands Controls Endometriosis patients * Stage I II Stage III IV Stroma Controls Endometriosis patients Stage I II Stage III IV Note: All P values were determined by the Fisher exact test. * P.05 (vs. controls). AP value of.025 is needed for statistical significance, following the Bonferroni procedure for comparison of two groups with one control. P.025 (vs. controls). FERTILITY & STERILITY 1089

6 FIGURE 3 Immunohistochemical detection of IGFBP-3 in endometriotic lesions ( 434). The immunoreaction was revealed as described in the legend of Figure 1. Note the brown intense immunostaining observed in the presence of the primary antibody in endometriotic glands (g) and stroma (s) and the weak staining in fibroblastic tissue (f) surrounding the endometriosis implant (A). No immunoreactivity was detected in endometriotic tissue sections incubated with normal goat immunoglobulins instead of the primary antibody (B). secretory phase samples (data not shown). Statistical analysis of the results did not show any significant difference in the intensity of staining found in the glands or in the stroma during the proliferative phase of the menstrual cycle between women with and those without endometriosis. In the secretory phase, however, IGFBP-3 immunostaining found in the glands was significantly greater in endometriosis patients than in controls (P.018). Moreover, such an increased IGFBP-3 expression was limited to stage I and II endometriosis (P.018), whereas in the more advanced stages (III and IV), no significant difference was noted (P.300). There was no significant difference in IGFBP-3 immunostaining of the endometrium among fertile and infertile patients having endometriosis (P.329). Expression of IGFBP-3 was also examined by immunohistochemistry in endometriotic lesions. As shown in Figure 3, a strong immunoreaction was observed, which was readily evident in lesions showing small glands in cellular stroma (A). Note the lower intensity of staining in fibroblastic tissue surrounding the implant. Among the 17 biopsy specimens evaluated, a similarly intense staining was found in 90% of the glands and 67% of the stroma, regardless of whether the specimen came from a peritoneal or ovarian implant or from the wall of an endometrioma. No apparent difference related to the menstrual cycle or to the fertility status was noted. Insulin-like growth factor binding protein-3 immunostaining was also present in encountered vessels. In these lesions, epithelial cells of the glands strongly stain for cytokeratins, whereas stromal cells do not stain for this epithelial marker (data not shown). No immunostaining was observed in endometriotic sections incubated with normal goat immunoglobulins (B), with the primary antibody following preabsorption with an excess of IGFBP-3 or when the primary antibody was omitted (data not shown). DISCUSSION The results presented in this study show that the stroma and the glands of the endometrium expressed variable amounts of IGFBP-3 during the menstrual cycle. The immunostaining is generally of higher intensity in biopsy specimens taken during the secretory phase than in those obtained in the proliferative phase. These results are in agreement with studies reporting differential mrna expression of IGFBP-3 in secretory compared to proliferative phase endometrium in normal women (25). Another study found that IGFBP-3 mrna is primarily concentrated in endometrial capillaries and is increased in the secretory phase, largely because of the intense vascularization of endometrial tissue during this phase (26). In our study, the difference in IGFBP-3 immunostaining of the endometrium between the two phases was not due to staining of the vessels but to more intense coloration of the stroma and the glands Akoum et al. IGFBP-3 staining in endometrial tissues Vol. 72, No. 6, December 1999

7 It is interesting to consider that the results presented in this study reveal an increased expression of IGFBP-3 in the secretory phase of the eutopic endometrium in patients having endometriosis. This observation is reminiscent of our recent findings of increased expression of monocyte chemotactic protein-1 in situ in the endometrial glands of women with endometriosis and in vitro in endometrial epithelial cell cultures, suggesting that this disease is not only a local disorder restricted to the peritoneal cavity but could be associated with functional changes at the level of the eutopic endometrium (29, 30). It is also interesting to note that this increased expression is significant in women with stages I and II disease compared to patients with stage III and IV disease. This is also reminiscent of our recent findings of increased expression of monocyte chemotactic protein-1 in early stages of the disease (29, 30). Previous data indicated increased production capacities of prostaglandin F by endometriotic implants in the early stages of this condition (31). These data support the concept that endometriosis is more active in the early stages. The stronger expression in the glands of the endometrium of endometriosis patients is not related to the infertility status in this group. No comparison can be made for infertility in the non-endometriosis group since our control group included only one infertile patient. The relationship between infertility and endometrial growth factors is presently unknown. The main role usually ascribed to IGFBP-3 is regulation of the transport of IGFs and their presentation to cell surface receptors (16). Elevated levels of mrna-encoding IGFs have been reported to be associated with the proliferative phase for IGF-I and with the secretory phase for IGF-II (1, 4). However, the functional relationship between IGF and IGFBP-3 in the endometrium has not been assessed, and there is no data on the expression of IGFs in the endometrium of patients having endometriosis. There is evidence, however, that IGF stimulates the proliferation of stromal cells in culture (9). This study also revealed an intense immunostaining of IGFBP-3 in endometriotic lesions. This was observed without a real comparison with the endometrium, since few paired biopsy specimens from the endometrium and endometriotic lesions were available. It is also important to consider that the majority of lesions evaluated stained with high intensity independent of the cycle phase. This is in contrast to the findings in the eutopic endometrium, where IGFBP-3 immunostaining is variable, generally of low intensity in the proliferative phase and of greater intensity in the secretory phase. This observation supports the idea that the ectopic endometrium is less responsive to sex steroids. In this study, the immunostaining appeared similar in biopsy specimens taken from peritoneal or ovarian implants or from the wall of an endometrioma. No apparent difference could be noted in relationship to the fertility status. A more elaborate study will be necessary to evaluate potential differences according to the various types and locations of lesions. The relationship between high IGFBP-3 expression in endometriotic tissue and the process of endometriosis is unknown. However, it would be interesting to speculate that it might relate to recent findings of IGFBP-3 fragments and protease activity against IGFBP in the PF (8, 13 15). Tissuespecific regulation of IGFBP-3 proteolysis would provide a mechanism to increase bioavailability of IGFs and stimulate cellular proliferation (23). Recent data indicate that IGFBP-3 mediates direct effects on animal fibroblasts cell lines and human breast cancer cells by IGF-independent mechanisms (18 20). Moreover, our data indicate that N-terminal IGFBP-3 fragments purified from the PF have a direct mitogenic effect on endometrial epithelial cells in culture (13). Our recent findings showed an association between the mitogenic activity and fragmentation of IGFBP-3, suggesting the presence of an IGF/IGFBP-3 protease system in the peritoneal fluid (25). Indeed, such a system is involved in several physiological and pathological conditions involving tissue proliferation or remodeling, such as the seminal plasma (32); the fluid of embryonic cavity (33); the cerebrospinal fluid of children with leukemia, central nervous system tumor or meningitis (34); and the synovial fluid in arthritis (35). The involvement of this system is also supported by several recent reports of the expression of various proteases in the eutopic endometrium and ectopic endometriotic implants (36 40). Moreover, several matrix metalloproteases have been reported to be released by endometrial stromal cells in culture (41). These observations make plausible that IGFBP-3 is part of a tissue-remodeling process in endometriosis. Thus, the findings of increased expression of IGFBP-3 in the secretory endometrium in endometriosis according to the stages of the disease and its high expression in endometriotic lesions provides further evidence for the involvement of the IGF/IGFBP-3/protease system in the pathophysiology of endometriosis. According to the retrograde menstruation concept, IGFBP-3 in the PF could come at least in part from endometrial cells regurgitated through the fallopian tubes at the time of the menses or directly from endometriotic implants. Further studies will be required to better understand the mechanisms underlying the high level of IGFBP-3 in endometriosis and whether there is a link with the increased expression of this protein in the eutopic endometrium of endometriosis patients. Acknowledgments: The authors thank Monique Longpré and Johanne Pelletier for their technical assistance in patient evaluation and specimen collection, Lucile Turcot-Lemay for statistical analysis, and Marie-Claude Boivin for typing the manuscript. FERTILITY & STERILITY 1091

8 References 1. Boehm KD, Daimon M, Gorodeski IG, Sheean LA, Utian WH, Han J. Expression of the insulin-like and platelet-derived growth factor genes in human uterine tissues. Mol Reprod Dev 1990;27: Talavera F, Reynolds RK, Roberts JA, Menon KM. Insulin-like growth factor I receptors in normal and neoplastic human endometrium. Cancer Res 1990;50: Villeneuve MG, Plouffe L, Pantazis CG, Smith MT, Thompson WO, McDonough PG. Immunocytochemical study of insulin-like growth factor I and insulin-like growth factor I receptor in human endometrium during the normal menstrual cycle. Int J Gynecol Pathol 1993;12: Giudice LC, Dsupin BA, Jin IH, Vu TH, Hoffman AR. Differential expression of messenger ribonucleic acids encoding insulin-like growth factors and their receptors in human uterine endometrium and decidua. J Clin Endocrinol Metab 1993;76: Giudice LC. Growth factors and growth modulators in human uterine endometrium: their potential relevance to reproductive medicine. Fertil Steril 1994;61: Chang SY, Ho YS. Immunohistochemical analysis of insulin-like growth factor I, insulin-like growth factor I receptor and insulin-like growth factor II in endometriotic tissue and endometrium. Acta Obstet Gynecol Scand 1997;76: Sbracia M, Zupi E, Alo P, Manna C, Marconi D, Scarpellini F, et al. Differential expression of IGF-1 and ectopic endometria of women with endometriosis and in women without endometriosis. Am J Reprod Immunol 1997;37: Smith SK, Thomas EJ, Rock JA, eds. The endometrium and endometriosis. In: Thomas E, Roch J, eds. Modern approach to endometriosis. Boston: Kluwer Academic Publishers, 1991: Giudice LC, Dsupin BA, Gargosky SE, Rosenfeld RG, Irwin JC. The insulin-like growth factor system in human peritonal fluid: its effects on endometrial stromal cells and its potential relevance to endometriosis. J Clin Endocrinol Metab 1994;79: Lemay A, Koutsilieris M, Akoum A, Maheux R. Potential implications of peritoneal fluid mitogens related to proteolysis of insulin-like growth factor binding protein-3. In: Minaguchi H, Sugimoto V, eds. The Pathophysiology of Endometriosis. Endometriosis Today. Advances in Research and Practice. Proceeding of the VI World Congress on Endometriosis. Yokohama, Japan: Parthenon Publishing Group, 1997: Koutsilieris M, Allaire-Michaud L, Fortier M, Lemay A. Mitogen(s) for endometrial-like cells can be detected in human peritoneal fluid. Fertil Steril 1991;56: Koutsilieris M, Niklinski W, Frenette G, Lemay A. Heparin-sepharose binding growth factors in peritoneal fluid of women with endometriosis. Fertil Steril 1993;59: Koutsilieris M, Akoum A, Lazure C, Frenette G, Lemay A. N-terminal truncated forms of insulin-like growth factor binding protein-3 in the peritoneal fluid of women without laparoscopic evidence of endometriosis. Fertil Steril 1995;63: Taskin O, Giudice L, Mangal R, Dunn RC, Dsupin BA, Poindexter AN, et al. Insulin-like growth factor binding proteins in peritoneal fluid of women with minimal and mild endometriosis. Hum Reprod 1996;11: Koutsilieris M, Lavergne E, Lemay A. Association of protease activity against IGFBP-3 with peritoneal fluid mitogens: possible implication for the ectopic growth of endometrial cells in women with endometriosis. Anticancer Res 1997;17: Kelley KM, Oh Y, Gargosky SE, Gucev Z, Matsumoto T, Hwa V, et al. Insulin-like growth factor-binding proteins (IGFBPs) and their regulatory dynamics. Int J Biochem Cell Biol 1996;28: Mohan S, Baylink DJ, Pettis JL. Insulin-like growth factor (IGF)- binding proteins in serum do they have additional roles besides modulating the endocrine IGF actions? (Editorial) Clin Endocrinol Metab 1996;81: Delbé J, Blat C, Desauty G, Harel L. Presence of IDF45 (migfbp-3) binding sites on chick embryo fibroblasts. Biochem Biophys Res Commun 1991;179: McCusker RH, Busby WH, Dehoff MH, Camacho-Hubner C, Clemmons DR. Insulin-like growth factor (IGF) binding to cell monolayers is directly modulated by the addition of IGF-binding proteins. Endocrinology 1991;129: Cohen P, Lamson G, Okajima T, Rosenfeld RG. Transfection of the human insulin-like growth factor binding protein-3 gene into Balb/c fibroblasts inhibits cellular growth. Mol Endocrinol 1993;7: Oh Y, Muller HL, Lamson G, Rosenfeld RG. Insulin-like growth factor (IGF)-independent action of IGF-binding protein-3 in Hs578T human breast cancer cells. J Biol Chem 1993;268: Blat C, Villaudy J, Harel L. Density-dependent inhibition of mouse embryo fibroblast growth: involvement of IGFBP-3. Exp Cell Res 1994;215: Lamson G, Giudice LC, Cohen P, Liu F, Gargosky S, Muller HL, et al. Proteolysis of IGFBP-3 may be a common regulatory mechanism of IGF action in vivo. Growth Regul 1993;3: Sampson JA. Peritoneal endometriosis due to menstrual dissemination of endometrial tissue into the peritoneal cavity. Am J Obstet Gynecol 1927;14: Giudice LC, Milkowski DA, Lamson G, Rosenfeld RG, Irwin JC. Insulin-like growth factor binding proteins in human endometrium: steroid-dependent messenger ribonucleic acid expression and protein synthesis. J Clin Endocrinol Metab 1991;72: Zhou A, Dsupin BA, Giudice LC, Bondy CA. Insulin-like growth factor system gene expression in human endometrium during the menstrual cycle. J Clin Endocrinol Metab 1994;79: Noyes RW, Hertig AW, Rock J. Dating the endometrial biopsy. Fertil Steril 1950;1: Siegel S, Castellan NJ. Nonparametric statistics for the behavioral sciences. 2nd ed. New York: McGraw-Hill, Inc., Akoum A, Lemay A, McColl S, Turcot-Lemay L, Maheux R. Elevated concentration and biologic activity of monocyte chemotactic protein-1 in the peritoneal fluid of patients with endometriosis. Fertil Steril 1996;66: Jolicoeur C, Boutouil M, Drouin R, Paradis I, Lemay A, Akoum A. Increased expression of monocyte chemotactic protein-1 in the endometrium of women with endometriosis. Am J Pathol 1998;152: Vernon MW, Beard JS, Graves K, Wilson WA. Classification of endometriotic implants by morphologic appearance and capacity to synthesize prostaglandin F. Fertil Steril 1986;46: Cohen P, Graves HCP, Peehl DM, Kamarei M, Giudice LC, Rosenfeld RG. Prostate-specific antigen (PSA) is an insulin-like growth factor binding protein-3 protease found in seminal plasma. J Clin Endocrinol Metab 1992;75: Nonoshita LD, Wathen NC, Dsupin BA, Chard T, Giudice LC. Insulinlike growth factors (IGFs), IGF-binding proteins (IGFBPs), and proteolyzed IGFBP-3 in embryonic cavities in early human pregnancy: their potential relevance to maternal-embryonic and fetal interactions. J Clin Endocrinol Metab 1994;79: Muller HL, Oh Y, Gargosky SE, Lehrnbecher T, Hintz RL, Rosenfeld RG. Concentrations of insulin-like growth factor (IGF)-binding protein-3 (IGFBP-3), IGF, and IGFBP-3 protease activity in cerebrospinal fluid of children with leukemia, central nervous system tumor, or meningitis. J Clin Endocrinol Metab 1993;77: Matsumoto T, Gargosky SE, Iwasaki K, Rosenfeld RG. Identification and characterization of insulin-like growth factors (IGFs), IGF-binding proteins (IGFBPs), and IGFBP proteases in human synovial fluid. J Clin Endocrinol Metab 1996;81: Head JR, MacDonald PC, Casey ML. Cellular localization of membrane metalloendopeptidase (enkephalinae) in human endometrium during the ovarian cycle. J Clin Endocrinol Metab 1993;76: Rodgers WH, Osteen KG, Matrisian LM, Navre M, Giudice LC, Gorstein F. Expression and localization of matrilysin, a matrix metalloproteinase, in human endometrium during the reproductive cycle. Am J Obstet Gynecol 1993;168: Clements J, Mukhtar A. Glandular kallikreins and prostate-specific antigen are expressed in the human endometrium. J Clin Endocrinol Metab 1994;78: Fernandez-Shaw S, Marshall JM, Hicks B, Barlow DH, Starkey PM. Plasminogen activators in ectopic and uterine endometrium. Fertil Steril 1995;63: Bergqvist A, Ferno M, Mattson S. A comparison of cathepsin D levels in endometriotic tissue and in uterine endometrium. Fertil Steril 1996; 65: Rawdanowicz TJ, Hampton AL, Nagase H, Woolley DE, Salamonsen LA. Matrix metalloproteinase production by cultured human endometrial stromal cells: identification of interstitial collagenase, gelatinase-a, gelatine-b, and stromelysin-1 and their differential regulation by interleukin-1 and tumor necrosis factor. J Clin Endocrinol Metab 1994;79: Akoum et al. IGFBP-3 staining in endometrial tissues Vol. 72, No. 6, December 1999