Secretion of monocyte chemotactic protein-1 by cytokine-stimulated endometrial cells of women with endometriosis*

FERTILITY AND STERILITY Copyright

©

Vol. 63, No.2, February 1995

1995 American Society for Reproductive Medicine

Printed on acid-free paper in U. S. A.

Secretion of monocyte chemotactic protein-l by cytokinestimulated endometrial cells of women with endometriosis*

Ali Akoum, Ph.D. t+ Andre Lemay, M.D.:!: Chantal Brunet, Ph.D.§

Jacques Hebert, M.D.§ Le groupe d' investigation en gynecologie II

Universite Laval; Le Centre Hospitalier de l'Universite Laval, Sainte-Fay; and Hopital St-Fran90is d'Assise, Quebec, Quebec, Canada

Objective: To evaluate in vitro the production of monocyte chemotactic protein-l (MCP-l) by endometrial cells of patients with and without endometriosis. Design: Primary cultures of stromal and epithelial cells isolated from human endometrium were exposed during 24 hours to different cytokines. Monocyte chemotactic protein-l secretion was analyzed in the culture medium. Setting: Gynecology clinic and laboratories of endocrinology of reproduction and immunology. Patients: Women presenting for infertility or pelvic pain in which endometriosis was diagnosed at laparoscopy (n = 6) and women presenting for tubal ligation without laparoscopic evidence of the disease (n = 6). Interventions: None. Main Outcome Measures: De novo secretion of MCP-l in the culture supernatant by immunoprecipitation and sodium dodecyl sulfate-polyacrylamide gel electrophoresis after metabolic labeling with 35S-cysteine. Results: The incubation of endometrial epithelial cells of endometriosis women with either interleukin -1{3 or tumor necrosis factor-a resulted in the appearance of at least two and sometimes three bands having approximately 15, 13, and 9 kd molecular weights. These bands were identified as three distinct species of MCP-l as their immunoprecipitation was prevented effectively in presence of an excess of cold MCP-l. In contrast, the endometrial epithelial cells of only one of six normal women produce significant levels of MCP-l under the same stimulation conditions. The stromal cells of both groups of subjects do not secrete appreciable amounts of MCP -1 or only small quantities in two cases of endometriosis. Conclusions: Monocyte chemotactic protein-l secretion is upregulated in cytokine-stimulated endometrial epithelial cells of women having endometriosis as compared with normal women without evidence of the disease. Such a difference at the level of eutopic endometrial cell may have a significance in the physiopathology of endometriosis. Fertil Steril 1995;63:322-28 Key Words: Endometriosis, endometrial cells, chemotactic factors, cytokines

Received March 10, 1994; revised and accepted August 26, 1994. * Presented at the 76th Annual Meeting of the Endocrine Society' Anaheim, California, June 15 to 18, 1994. t Reprint requests: Ali Akoum, Centre de recherche, Hopital St-Fran~ois d'Assise, 10 rue de l'Espinay, Quebec, Quebec, GIL 3L5, Canada (FAX: 418-525-4481). :j: Laboratoire d'endocrinologie de la reproduction, Department d'Obstetrique/Gynecologie, Universite Laval. § Le centre de Recherche en Inflammation et ImmunologieRhumatologie, Le Centre Hospitalier de l'Universite Laval.

322

Akoum et al.

MCP-l secretion by endometrial cells

Endometriosis is a gynecologic disorder characterized by proliferation of endometrial-like tissue outside the uterus mainly on the peritoneum and in the organs ofthe pelvic cavity. Several studies have reported an increase in number, activation level,

II Jacques Bergeron, M.D., Rodolphe Maheux, M.D., Georges Quesnel, M.D., Marc Villeneuve, M.D., Hopital St-Fran~ois d'Assise. Fertility and Sterility

and secretory products of peritoneal macrophages in association with the disease (1-5). A higher chemotactic activity for monocytes (6) and factors that may contribute to macrophage recruitment and activation (7, 8) have been found in the peritoneal fluid of patients. In a recent study (unpublished data), we have shown that endometriotic cells secrete a novel monocyte chemotactic and activating factor (MCP-1) in vitro upon stimulation with interleukin-1fj (IL-1m and tumor necrosis factor-a (TNFa). Elevated levels of these cytokines, which are produced mainly by macrophages, were found in the peritoneal fluid of women with endometriosis (4,5, 9). According to the transplantation theory, which implies ectopic implantation of uterine endometrial cells having migrated into the peritoneal cavity (10), it would be pertinent to examine if eutopic endometrial cells could also produce MCP-l. Such an endometrial cell function could be an important initial factor in the immunoinflammatory process taking place in the disease. Thus, we have investigated MCP-1 production by primary cultures of endometrial cells isolated from the endometrium of women with and without laparoscopic evidence of endometriosis at the time of laparoscopy. MATERIALS AND METHODS Source and Handling of Tissues

Tissue specimens used in this report were obtained from patients having given informed consent before laparoscopy. They originated from women having requested tubal ligation and without evidence of endometriosis at laparoscopy (n = 6) and from women having diagnosis of endometriosis at the time of laparoscopy performed for pain andj or infertility (n = 6). These women, aged between 19 and 37 years, otherwise had no other pelvic pathology and were not taking other hormonal or anti-inflammatory agents. A sterile Novak canula was used to obtain an adequate endometrial specimen. Endometrial biopsies were initially placed at 4°C in sterile Hanks' balanced salt solution (HBSS) containing 100 UjmL penicillin, 100 p,gj mL streptomycin, and 0.25 p,gjmL amphotericin B (GIBCO BRL, Burlington, Ontario, Canada) and transported to the laboratory. Tissue Dissociation and Cell Culture

To isolate epithelial and stromal cells, endometrial biopsies were dissociated as described by Smith and Kelly (11) with some modifications. Vol. 63, No.2, February 1995

Briefly, endometrial tissue was minced into small pieces and incubated for 1 hour at 37°C in a shaking water bath in HBSS supplemented with 1 mgjmL collagenase IA (Sigma, St. Louis, MO), 100 UjmL penicillin, 100 p,gjmL streptomycin, and 0.25 p,gj mL amphotericin B. The suspension produced by collagenase digestion consisted of single stromal cells and fragments of epithelial glands. These two populations of cells then were separated by a series of differential sedimentations and adhesions. Cells were pelleted by centrifugation in a 15-mL polyethylene centrifuge tube, resuspended in 10 mL Dulbecco's modified essential medium-F12 medium (DMEM-F12; GIBCO BRL, Burlington, Ontario, Canada) containing 10% fetal bovine serum (FBS) (ICN Biomedicals, St. Laurent, Quebec, Canada) and allowed to stand for 1 hour at room temperature. The top 8 mL of the medium (stromalenriched fraction) was transferred to a 75 cm2 tissue culture flask and incubated for 30 minutes at 37°C, 5% CO 2 , The adhering stromal cells were kept in culture in DMEM-F12 medium containing 10% FBS, whereas nonadhering cells were discarded. The bottom 2 mL of the medium (epithelial-enriched fraction) was submitted to a second sedimentation procedure after which residual contaminating stromal cells were eliminated by two successive adhesion procedures as described above. Purified epithelial glands then were harvested by centrifugation and cultured in 24-well culture plates in DMEM-F12 medium supplemented with 10% FBS and 10 ngjmL epidermal growth factor (EGF) (GIBCO BRL). Evaluation of Cell Homogeneity

The purity of cell populations was assessed morphologically by light microscopy and immunocytochemically using specific monoclonal antibodies to cytokeratins, vimentin, and a-actin. Monoclonal mouse anti-a smooth muscle actin (clone 1A4) was purchased from Sigma. Monoclonal antiepithelial keratin-AE1:AE3 Mix (lot No. lOH1) was obtained from ICN Biomedical. Stromal and epithelial cell cultures also were examined by flow cytometry for the presence of leukocytes using phycoerythrin-labeled anti-CD45 mouse monoclonal antibody (Becton Dickinson, Mississauga, Ontario, Canada) and for the presence of endothelial cells using a sheep antihuman factor VIII antiserum and a fluoresceinlabeled rabbit polyclonal antibody directed against sheep immunoglobulins (Atlantic Antibodies, Scarborough, ME). For immunocytochemical studAkoum et aI. MCP-l secretion by endometrial cells

323

ies, coverslip cultures of purified stromal and epithelial cells were fixed in 3.5% formaldehyde in phosphate-buffered saline (PBS) for 15 minutes at room temperature, rinsed three times in PBS, incubated for 90 minutes at room temperature with an appropriate dilution in PBS of the monoclonal antibody (primary antibody) determined by titration, rinsed four times with PBS, incubated for 1 hour at room temperature with 1/50 dilution in PBS of fluorescein isothiocyanate-conjugated antimouse immunoglobulin (second antibody) (ICN Biomedicals), rinsed four times with PBS, covered with mounting medium, and sealed. Cultures incubated without the primary antibody were included as controls in all experiments. Slides were observed under a microscope equipped for fluorescence and photographed. Culture Stimulation

Stromal cell cultures grown to confluence were dissociated with 0.25% trypsin and 1 mM ethylenediaminetetraacetic acid in Hank's solution, harvested by centrifugation, replated at 100,000 cells per well in 24-well tissue culture plates (Costar, Cambridge, MA) and allowed to grow to confluence in DMEM-F12 medium containing 10 JLg/mL insulin and 5 JLg/mL transferrin and supplemented with 10% FBS. Epithelial glands were distributed directly in 24-well culture plates and allowed to grow to confluence in the same medium described above supplemented with 10 ng/mL EGF. Cultures then were washed twice with 500 JLL RPMI-1640 medium (GIBCO BRL), incubated for 24 hours at 37°C, 5% CO 2 with the same medium, washed twice, and incubated with 500 JLL of cysteine-free RPMI-1640 medium (GIBCO BRL) with or without different concentrations of cytokines and 80 JLCi of 35S-cysteine (Amersham, Oakville, Ontario, Canada). The inhibition of protein synthesis was carried out by incubating the cultures with 50 JLg/ mL of cycloheximide (Sigma) together with the stimulating cytokine. After incubation for various periods of time, the culture supernatants were collected and MCP-1 production was analyzed by immunoprecipitation, sodium dodecyl sulfate-polyacrylamide gel electrophoresis (SDS-PAGE) and autoradiography. Immunoprecipitation and SDS-P AGE Analysis

Labeled culture supernatants were precleared with protein G-agarose (GIBCO BRL) and normal rabbit serum. Subsequent immunoprecipitation 324

Akoum et at.

MCP-l secretion by endometrial cells

was performed using protein G-agarose and rabbit anti-MCP-1 antibody. This antibody did not crossreact with several cytokines that are related closely to MCP -1, including IL-8 or macrophage inflammatory proteins 1a and 1{j (12). Beads were washed three times in a solution containing 25 mM tris(hydroxymethyl)aminomethane, pH 8.0, 150 mM NaCI, 0.05% NaN3, and 0.1% Nonidet P40 and immunoprecipitated proteins were boiled for 5 minutes in 40 JLL of Kornberg sample buffer. All samples were electrophoresed through 18% discontinuous SDS-polyacrylamide gels according to the method of Thomas and Kornberg (13). Proteins were visualized by fluorography by using X-OMat AR film (Kodak, Rochester, NY) for 7 days at -80°C. To compare the levels of MCP-1 secretion by endometrial cells issued from the different patients' a densitometric analysis of each MCP-1 band on the autoradiograms was performed using a Bioimage 110 S, software WBA (Millipore, Bedford, MA). The integrated optical density (IOD) varied between 0 and 1,611. The results in Table 1 are expressed in terms of total MCP-1 secretion in response to a given concentration of cytokine. An arbitrary unit was defined as equivalent to 250 IOD and (-) represents IOD values equivalent to background (0 to 55). RESULTS Characterization of Cell Cultures

The purity of endometrial stromal and epithelial cell cultures first was examined morphologically by phase-contrast microscopy. Stromal cells could be distinguished by their predominant fibroblast-like appearance, whereas epithelial cells formed colonies that spread out gradually, exhibiting a large and polygonal morphology (Fig. 1). The cultures were characterized further by immunocytochemistry using specific monoclonal antibodies to cytokeratins, vimentin, and a-actin. Epithelial cells were positive for cytokeratins (Fig. 2A) and a-actin (Fig. 2B), whereas stromal cells were stained positively only for vimentin (Fig. 2C); there were no cytokeratin-positive cells detectable in stromal cell cultures, whereas epithelial cell cultures were contaminated by a small number of stromal cells (approximately 1 %). Furthermore, flow cytometric analysis have shown that either epithelial or stromal cell cultures contained <0.5% contaminating CD45-positive cells (leukocytes) and factor VIIIpositive cells (endothelial cells). The absence of Fertility and Sterility

Table 1 Cytokine-Induced MCP-l Production by Endometrial Stromal and Epithelial Cells of Normal Women and of Patients with Endometriosis TNFa

IL-l~

Patients

Stromal

Epithelial

Stromal

1 ng/mL

No endometriosis (n = 6) 1 to 5 6 Pelvic endometriosis (n = 6) 1, stage I 2, stage I 3, stage II 4, stage lIt 5, stage II 6, stage IV

Epithelial

Stromal

No stimulation

Epithelial

Stromal

Epithelial

100 ng/mL

10 ng/mL

1 1

10 1 2 3 2 1

1

5 3 2 3 2 4

3 3 4 1

* Densitometric analysis of each MCP-l band on the autoradiograms was performed and an arbitrary unit was defined as equivalent to 250 IOD. The symbol "-" represents IOD values equivalent to background (0 to 55).

t The stage of the disease (I, II, III, or IV) was determined at laparoscopy according to the revised American Fertility Society classification. This takes into account number, size, and location of endometrial implants, plaques, endometriomas, and/or adhesions (16).

such a contamination is important, as these cells have the ability to produce MCP-1 (14, 15).

endometrial cells of patients with endometriosis. The incubation of epithelial cells either with IL-1{3 or with TNFa resulted in the appearance of at least two and sometimes three bands having molecular weights corresponding to approximately 15, 13, and 9 kd. The band having the lowest molecular weight (approximately 9 kd) has not always been detectable. This phenomenon was cell line dependent and was not due to differences in gel resolution. These bands were identified as three distinct species of MCP-1, because their immunoprecipitation was effectively prevented in presence of an excess of cold MCP-1 (1 JLg/mL), thereby confirming the specificity of the rabbit anti-MCP-1 antibody for MCP-l. This antibody previously has been shown not to cross-react with several cytokines that are related closely to MCP-1, including IL-8 and macrophage inflammatory proteins 1a and 1{3 (12). The higher molecular weight bands (> 29 kd) were not inhibited by cold ligand competition, indicating that they were nonspecifically binding to the protein G-agarose during immunoprecipitation. The incubation of endometrial epithelial cells of women with endometriosis with different concentrations of IL-1{3 and TNFa (0.1 to 100 ng/mL) resulted in a dose-dependent increase ofMCP-1 secretion (data not shown). Monocyte chemotactic protein-1 secretion was detected initially after approximately 9 hours of incubation with the cytokines and increased over 48 hours of stimulation (data not shown). Significant levels ofMCP-1 were produced after 24 hours of incubation with IL-1{3 (1

MCP-l De Novo Synthesis

Using 35S-cysteine labeling of protein synthesis and a rabbit polyclonal antibody specific to MCP-1, MCP-1 was identified in the culture media after immunoprecipitation and protein separation on SDS-PAGE. Figure 3 shows an autoradiogram from a representative experiment illustrating the cytokine-induced de novo secretion of MCP-1 by

Figure 1 Morphological characterization of endometrial cell cultures. (A), Phase-contrast microscopy of a confluent culture of epithelial cells; large and polygonal cells with obvious epithelial morphology can be seen. (B), Phase-contrast microscopy of a confluent culture of stromal cells. Cell morphology is comparable to that of fibroblast (spindle-shape) and more rounded cells can also be seen (Magnification was XI25). Vol. 63, No.2, February 1995

Akoum et al.

MCP-l secretion by endometrial cells

325

laparoscopic evidence of pelvic endometriosis. The endometrial epithelial cells of all women with endometriosis secrete various levels of MCP-l. However, the difference in the level of the de novosynthesized material could not be correlated with the stage of the disease as more patients will be required for such an investigation. In contrast, the endometrial epithelial cells of one of six patients with normal pelvises secrete significant amounts of MCP-1 under the same stimulation conditions. The stromal cells of both groups of patients do not produce appreciable levels of MCP-1 or only small quantities in two cases of endometriosis. DISCUSSION

According to the transplantation theory first put forward by Sampson (10), endometriosis is secondary to ectopic implantation and proliferation of en-

14.3

Figure 2 Immunocytochemical characterization of endometrial epithelial and stromal cells in culture. (A), Immunostaining of epithelial cells with anticytokeratin antibodies. (B), Immunostaining of epithelial cells with anti-a actin antibody. (C), Immunostaining of stromal cells with antivimentin antibody (Magnification was X260).

to 10 ng/mL) and TNFa (100 ng/mL). Thus, these conditions were maintained in subsequent stimulations. Table 1 summarizes the results of the study in six women with normal pelvises and six women with 326

Akoum et al.

MCP-l secretion by endometrial cells

1234567 Figure 3 Sodium dodecyl sulfate-PAGE analysis of cytokineinduced de novo MCP-1 secretion by endometrial epithelial cells of patients with endometriosis as described under Materials and Methods. Lane 1, molecular weight markers; lane 2, 1 ng/mL IL-1iJ; lane 3,10 ng/mL IL-1iJ; lane 4,100 ng/mL TNFa; lane 5, the culture was stimulated with 10 ng/mL IL-1iJ as described before, the supernatant was collected and an excess of unlabeled MCP-1 (1 ILg/mL) was added before immunoprecipitation to inhibit the immunoprecipitation of 35S-cysteine-Iabeled MCP-1 and to confirm, thereby, the specificity of MCP-1 antiserum for MCP-1; lane 6,10 ng/mL IL-1iJ and 50 ILg/mL cycloheximide; lane 7, medium alone (control).

Fertility and Sterility

dometrial cells having migrated by tubal spill from the uterus into the pelvic cavity of women during menstruation. Retrograde seeding is a common phenomenon in most menstruating women. Genetic predisposition has been evoked to explain why endometrial cells would implant ectopically into some particular patients and not into others (17). Immunologic defects described in endometriosis patients also may tolerate the implantation and development of endometriotic foci (18). On the other hand, a growing body of evidence indicates that an immunoinflammatory process involving a humoral response to endometrial and endometriotic antigens (19, 20) as well as an increased number of T lymphocytes and activated macrophages in the peritoneal fluid of patients (1-5, 7) is associated with the disease. Macrophages may have an important role in the development of endometriotic lesions by secreting growth factors for endometrial cells (21, 22) and also may have an adverse effect on the fertility of patients by secreting numerous embryotoxic cytokines (4, 5, 9). Interestingly, recent studies on the peritoneal fluid of endometriosis patients have shown an increased chemotactic activity for monocytes (6) and elevated levels of monocyte chemoattractant cytokines such as Regulated on Activation, Normal T Expressed and Secreted (RANTES) (7) and transforming growth factor-{1 (8). In previous work (Akoum A, Lemay A, Brunet C, Hebert J, unpublished data), we have demonstrated that endometriotic cells can produce MCP-l in response to IL-l{1 and TNFa, cytokines that are produced mainly by activated macrophages and found in elevated levels in the peritoneal fluid of patient. In this report we show that uterine endometrial cells also have the ability to produce MCP-l in response to the same cytokines. Furthermore, in endometriosis, the epithelial endometrial cells seem to have the intrinsic ability to produce elevated levels of MCP-l when compared with those of normal women without laparoscopic evidence of the disease. Two and sometimes three different species of MCP-l were detected. This phenomenon reported in other cell types (12) was cell line dependent. It has been reported that the core protein (approximately 9 kd) may be synthesized but not necessarily secreted by all cells that produce MCP-l. Differences in the processing of a-linked carbohydrates account for the heterogeneity of MCP-l produced by different cell types (23). The physiopathological significance of this finding may be interesting. On one hand it would sugVol. 63, No.2, February 1995

gest a functional difference at the level of eutopic endometrial cell in endometriosis. On the other hand, endometrial cells that may be spilled into the peritoneal cavity of patients holds the potential to respond to local stimulating factors by producing MCP-l and to contribute to the local inflammatory process taking place in the disease. Monocyte chemotactic protein-l is known as a potent chemotactic and activating factor specific for monocytes (24) that express a significant number of MCP-l receptors (25). On the basis of these findings it is tempting to hypothesize that MCP-l is an attractive candidate as an effector cell mediator in the pathophysiology of endometriosis. Further studies will be necessary to ascertain whether MCP-l is present in the peritoneal fluid of patients and to compare the levels of MCP-l in women with and without the disease. It also will be necessary to examine MCP-l expression in endometrial tissue in situ and to determine its cellular localization. The reasons for which endometrial epithelial cells of women with endometriosis secrete elevated levels of MCP-l after in vitro stimulation by IL-l{1 and TNFa are undetermined. Antigenic differences have been found between the endometrium of women with and without endometriosis (19). A serum antibody recognizing a 34- kd endometrial antigen also has been identified in patients with endometriosis (20). It is not known if these immunologic changes are primary or secondary in the physiopathology of endometriosis, but it is tempting to postulate that they could be related to the functional change observed in our study. However, it remains to be shown that MCP-l is elaborated in response to antigenic stimulation. This study includes a small number of subjects (n = 12); it also is possible that low levels of MCP-l would not have been detectable by immunoprecipitation and SDS-PAGE and additional factors would be required in vitro for MCP-l secretion by endometrial stromal and epithelial cells. Nevertheless, MCP-l secretion seems to be upregulated in cytokine-stimulated endometrial epithelial cells of women who have laparoscopic evidence of endometriosis compared with normal women without evidence of the disease. Such a difference at the level of eutopic endometrial cell may play a significant role in the pathophysiology of endometriosis. Acknowledgments. Ali Akoum is recipient of the fellowship in Reproductive Endocrinology at Laval University, Quebec, Quebec, Canada, offered by Wyeth Ayerst Pharmaceuticals, Montreal, Quebec, Canada. The authors thank Shaun McColl, Akoum et al.

MCP-J secretion by endometrial cells

327

Ph.D., from the Centre de Recherche en Inflammation et Immunologie-Rhumatologie, Le Centre Hospitalier de l'Universite Laval, Sainte-Foy, Quebec, for providing the rabbit polyclonal anti-MCP-1 antibody and the synthetic MCP-1. They are grateful to Ms. Anette Bald for her technical assistance and to Ms. Josee Pardiac for typing the manuscript. Monoclonal mouse antivimentin antibody was a generous gift from Michel Vincent, Ph.D. (Centre Hospitalier de l'Universite Laval, Quebec, Canada).

REFERENCES 1. Haney AF, Muscato JJ, Weinberg JB. Peritoneal fluid cell populations in infertility patients. Fertil SteriI1981;35:6968. 2. Halme J, Becker S, Hammond MG, Raj MHG, Raj S. Increased activation of pelvic macrophages in infertile women with mild endometriosis. Am J Obstet Gynecol 1983; 145:333-7. 3. Zeller JM, Hening I, Radwanska E, Dmowski WP. Enhancement of human monocyte and peritoneal macrophage chemiluminescence activities in women with endometriosis. Am J Reprod Immunol 1987;13:78-83. 4. Eisermann J, Gast MJ, Pineda J, Odem RR, Collins JL. Tumor necrosis factor in peritoneal fluid of women undergoing laparascopic surgery. Fertil Steril 1988;50:573-9. 5. Taketani Y, Kuo TM, Miguno M. Comparison of cytokine levels and embryo toxicity in peritoneal fluid in infertile women with untreated and treated endometriosis. Am J Obstet GynecoI1992;167:765-70. 6. Leiva MC, Hasty LA, Pfeifer S, Mastroianni L, Lyttle CR. Increased chemotactic activity of peritoneal fluid in patients with endometriosis. Am J Obstet Gynecol 1991; 168:592-8. 7. Khorram 0, Taylor RN, Ryan IP, Schall TJ, Landers DV. Peritoneal fluid concentrations of the cytokine RANTES correlate with the severity of endometriosis. Am J Obstet Gynecol 1993;169:1545-9. 8. Oosterlynck DJ, Meuleman C, Waer M, Koninckx PRo Transforming growth factor-{3 activity is increased in peritoneal fluid from women with endometriosis. Obstet Gynecol 1994;83:287-92. 9. Fakih H, Baggett B, Holtz G, Tsang K-Y, Lee JC, Williamson HO. Interleukin-1: a possible role in the infertility associated with endometriosis. Fertil SteriI1987;47:213-7. 10. Sampson JA. Peritoneal endometriosis due to the menstrual dissemination of endometrial tissue into the peritoneal cavity. Am J Obstet Gynecol 1927;14:422-69. 11. Smith SK, Kelly RW. The effect of estradiol-17{3 and actin-

328

Akoum et al.

MCP-J secretion by endometrial cells

omycin D on the release of PGF and PGE separated cells of human endometrium. Prostaglandins 1987;34:553-60. 12. Hachicha M, Rathanaswami P, Schall TJ, McColl SR. Production of monocyte chemotactic protein-1 in human type B synoviocytes. Arthritis Rheum 1993;36:26-34. 13. Thomas JO, Kornberg RD. An octamer of histones in chromatin and free solution. Proc Nat! Acad Sci USA 1975;72:2626-30. 14. Yoshimura T, Robinson EA, Tanaka S, Appela E, Leonard EJ. Purification and amino acid analysis of two human monocyte chemoattractants produced by phytohemagglutinin-stimulated human blood mononuclear leukocytes. J Immunol 1989;142:1956-62. 15. Sica A, Wang JM, Colotta F, Dejana E, Montovania A, Oppenheim JJ, et al. Monocyte chemotactic and activating factor gene expression induced in endothelial cells by IL-1 and tumor necrosis factor. J Immunol 1990;144:3034-8. 16. The American Fertility Society. Revised American Fertility Society classification of endometriosis: 1985. Fertil Steril 1985;43:351-2. 17. Malinak LR, Buttram VC Jr, Elias S, Simpson JL. Heritable aspects of endometriosis II. Clinical characteristics of familial endometriosis. Am J Obstet GynecoI1980;137:332. 18. Oosterlynck DJ, Comillie FJ, Waer M, Vandeputte M, Koninckx PR. Women with endometriosis show a defect in natural killer activity resulting in a decreased cytotoxicity to autologous endometrium. Fertil Steril 1991;56:45-51. 19. Garza D, Mathur S, Dowd MM, Smith LF, Williamson HO. Antigenic differences between the endometrium of women with and without endometriosis. J Reprod Med 1991;36: 177-82. 20. Rajkumark K, Malliah V, Simpson CWo Identifying the presence of antibodies against endometrial antigens. A preliminary study. J Reprod Med 1992;37:552-6. 21. Halme J, White C, Kauma S, Estes J, Haskill S. Peritoneal macrophages from patients with endometriosis release growth factor activity in vitro. J Clin Endocrinol Metab 1988;66: 1044-9. 22. Olive DL, Montoya I, Riehl RM, Schenken RS. Macrophage-conditioned media enhance endometrial stromal cell proliferation in vitro. Am J Obstet GynecoI1991;164:953-8. 23. Jiang Y, Valente AJ, Williamson MJ, Zhang L, Graves DT. Post-translational modification of monocyte-specific chemoattractant synthesized by glioma, osteosarcoma, and vascular smooth muscle cell. J Bioi Chern 1990;265:18318-21. 24. Leonard EJ, Yoshimura T. Human monocyte chemoattractant protein-1 (MCP-1). Immunol Today 1990;11:97-101. 25. Yoshimura T, Leonard EJ. Identification of high-affinity receptors for human monocyte chemoattractant protein-1 on human monocytes. J ImmunoI1990;145:292-7.

Fertility and Sterility