Follicular Fluid as a Favourable Environment for Endometrial and Endometriotic Cell Growth in Vitro

Follicular fluid from women with endometriosis has been shown to induce a higher endometrial cell proliferation than that derived from women without the disease. To elucidate this issue further, the aims of the present study were to compare the ability of follicular fluid and peritoneal fluid to stimulate both endometrial and endometriotic cell proliferation and to verify whether the mitogenic effect was merely sex steroid-dependent. Endometrial and endometriotic cells were cultured in follicular fluid or peritoneal fluid diluted in serum-free media; the growth induced in these conditions was compared with that obtained by culturing these cells in medium supplemented with charcoal stripped calf serum and a correspondent content of 17-β-oestradiol and progesterone. Follicular fluid was able to induce significantly higher cell proliferation than peritoneal fluid from controls, patients with endometriosis stage I–II and women with endometriosis stage III–IV (P < 0.05). Moreover, the growth in control media containing a corresponding amount of steroid hormones was significantly lower than that obtained with follicular or peritoneal fluids. This finding indicates that the stimulating effect is not simply related to the concentrations of 17-β-oestradiol and progesterone present in these fluids. Finally, based on these results and on other previous observations, the hypothesis that follicular fluid may be involved in the development of endometriotic ovarian cysts is discussed. Introduction this theory has been further supported by Nezhat et al. who suggested that the secondary involvement of functional ovarian The pathogenesis of ovarian endometriotic cysts is still contro-cysts by endometriosis is crucial to the development of large versial. At least three different explanations have been offered ovarian endometriomas (Nezhat et al., 1992). to explain the development of ovarian cystic endometriosis. Notwithstanding the aetiopathogenetic model considered, First, endometriomas may be formed by invagination of the there are only few and limited observations regarding the role ovarian cortex (Hughesdon, 1957; Brosens et al., 1994). played by the ovulatory ovarian activity in the development According to this hypothesis, the process would originate from of endometriotic cysts (Bahtiyar et al., 1998). In this study, a free superficial implant that is in contact with the ovarian we postulate that ovarian follicles may be involved in the surface and is sealed off by adhesions; the menstrual shedding pathogenesis of endometrioma, at least in promoting their and bleeding of this small implant would result in progressive growth. To investigate this hypothesis, we evaluated the ability invagination of the ovarian cortex and formation …


Introduction
this theory has been further supported by Nezhat et al. who suggested that the secondary involvement of functional ovarian The pathogenesis of ovarian endometriotic cysts is still controcysts by endometriosis is crucial to the development of large versial.At least three different explanations have been offered ovarian endometriomas (Nezhat et al., 1992).to explain the development of ovarian cystic endometriosis.
Notwithstanding the aetiopathogenetic model considered, First, endometriomas may be formed by invagination of the there are only few and limited observations regarding the role ovarian cortex (Hughesdon, 1957;Brosens et al., 1994).
played by the ovulatory ovarian activity in the development According to this hypothesis, the process would originate from of endometriotic cysts (Bahtiyar et al., 1998).In this study, a free superficial implant that is in contact with the ovarian we postulate that ovarian follicles may be involved in the surface and is sealed off by adhesions; the menstrual shedding pathogenesis of endometrioma, at least in promoting their and bleeding of this small implant would result in progressive growth.To investigate this hypothesis, we evaluated the ability invagination of the ovarian cortex and formation of the of follicular fluid to stimulate endometrial and endometriotic endometrioma.Alternatively, endometriotic cystic lesions cell proliferation in vitro.could originate by metaplasia of the coelomic epithelium that lines the small cystic inclusions frequently found in ovaries (Nisolle and Donnez, 1997).This hypothesis, recently Materials and methods emphasized by Nisolle and Donnez (Nisolle and Donnez, Subjects 1997), is essentially based on the reported considerable meta-All subjects enrolled in this study were Ͻ38 years and had regular plastic potential of the ovarian mesothelium (Serov et al., 1973).menstrual cycles.Women with previous autoimmune or neoplastic Finally, in the original article by Sampson, endometriomas were disorders were excluded from the study.The extent of endometriosis considered to arise as a result of invasion of functional cysts was staged according to the Revised American Fertility Society Classification (American Society for Reproductive Medicine, 1997).
by surface implants (Sampson, 1921).In more recent years, Approval for this study was granted by the local Human Institutional the endometria.Endometriotic cysts were excised at operation, biopsied and endometriotic tissue was dissected from underlying Investigation Committee.Written consent was obtained from each woman before the procedures.
parenchyma.In previous studies, we have successfully established and employed endometrial cell cultures from endometrial samples Follicular fluid and from endometriotic cysts biopsies (Vigano `et al., 1993(Vigano `et al., , 2000;;Somigliana et al., 1996;Di Blasio et al., 1997).Briefly, tissues were Follicular fluids were obtained from 12 women undergoing IVF and gently minced into small pieces (1-2 mm 3 ) and washed in fresh embryo transfer procedures at the Reproductive Centre, Reggio Emilia medium to remove mucus or debris.Thereafter, they were incubated Hospital for tubal factor or male infertility.In these women a for 2 h at 37°C in a shaking water bath in 10 ml Ham's F-10 preliminary laparoscopy ruled out the presence of endometriosis.A (BioWhittaker, PBI International, Milan, Italy) containing 0.2% collastandard IVF protocol was used as described elsewhere (Vigano `et al., genase (Boehringer-Mannheim Biochemicals, Milan, Italy).After 1998).Briefly, these women were pretreated with gonadotrophinseveral washings, the cell suspension was digested in a 0.05% trypsin releasing hormone analogue (buserelin acetate, Suprefact; Hoechst, solution for 3-5 min.Cells, which represent a mixture of both stromal L'Aquila, Italy) started in the mid-luteal phase of the preceding cycle and epithelial components, were washed twice in Ham's F-10 and continued for 14 days.Follicle-stimulating hormone (FSH, supplemented with 10% fetal calf serum (FCS; Flow Laboratories, Metrodin, Serono Laboratories, Inc., Rome, Italy) was started there-Opera, Milan, Italy) and antibiotics and counted using a Cell Counter after with dose variations depending on individual responses.Follic-(Seac, Milan, Italy).Specifically, ~50 000-100 000 cells/well were ular development was monitored by daily measurements of serum allowed to adhere selectively to 24-well tissue culture plates and oestradiol and by ultrasonographic measurements of follicle diameter.
cultured in 1 ml of Ham's F-10 with 10% FCS, 2 mmol/l L-glutamine, Human chorionic gonadotropin (HCG, 10 000 IU, Profasi; Serono antibiotics and 2.5 µg/ml fungizone in a humidified atmosphere of Laboratories, Inc.) was administered when the size of at least two 95% air and 5% CO 2 at 37°C.leading follicles reached 15 mm or serum oestradiol levels Ͼ1 nmol/ mature follicle.Transvaginal ultrasound-guided oocyte retrieval was Proliferation assay performed 36 h later and the contents of visible follicles were Duplicates of endometrial and endometriotic cells cultures established aspirated.Care was taken to select only follicular fluids that were at day -1 were allowed to proliferate in eight different medium clear, non-bloody and that did not contain flushing medium.Oocytes conditions from day 0: were identified and separated.The follicular aspirates obtained were centrifuged at 800 g for 10 min at room temperature to separate them Ham's F-10 with 10% pooled follicular fluids from cells and then stored at -20°C.At the time of experiment, samples Ham's F-10 with 10% pooled peritoneal fluids from controls were thawed and pooled.17-β-Oestradiol and progesterone concentra-Ham's F-10 with 10% pooled peritoneal fluids from endometriosis tions of the pools were measured.Overall, three different pools were stage I-II obtained using follicular fluids from at least three different women for Ham's F-10 with 10% pooled peritoneal fluids from endometriosis each pool.
stage III-IV For each of these conditions, control medium was represented by

Peritoneal fluid
Ham's F-10 with 10% charcoal stripped (steroid depleted) calf Samples of peritoneal fluid were collected at the time of laparoscopy serum (Sigma) added with the corresponding concentration of 17-βfrom 38 women who, on the basis of the laparoscopic diagnosis, oestradiol and progesterone (Sigma) (Table I).Steroids were dissolved were divided into the following three groups: endometriosis stage in ethanol as concentrated stocks.The final ethanol concentration I-II (12 women), endometriosis stage III-IV (12 women) and controls was below 0.1% in all media.Content of 17-β-oestradiol in follicular (14 women).All the women were in early follicular or late luteal fluid pools ranged between 2.2ϫ10 6 and 3.5ϫ10 6 pg/ml, whereas phase of the menstrual cycle in order to avoid contamination of progesterone concentration was between 10ϫ10 6 and 14ϫ10 6 pg/ml.follicular fluid into peritoneal fluid.No woman had received hormones Levels of 17-β-oestradiol and progesterone in pools of peritoneal for at least 3 months.In the control group, laparoscopic examination fluids ranged between 300 and 640 pg/ml and between 18ϫ10 3 demonstrated normal pelvic organs in four cases, pelvic adhesions in and 31ϫ10 3 pg/ml respectively.All media were supplemented with two cases, benign ovarian pathology in six cases and benign uterine 2 mmol/l L-glutamine, antibiotics and 2.5 µg/ml fungizone and pathology in two cases.Peritoneal fluid samples were centrifuged at changed every other day.After 8 days of treatment, cells were 800 g for 10 min at room temperature to separate them from cells harvested by a 10 min incubation in 0.05% trypsin-0.02%EDTA and then stored at -20°C.At the time of experiment, samples were solution and counted using a Cell Counter (Seac).thawed and pooled.Each pool was established so that an equal number of early follicular and late luteal phase samples were used.
Statistical analysis 17-β-Oestradiol and progesterone concentrations of the pools were Data are expressed as median and range.Statistical analysis was measured.Overall, three different pools of peritoneal fluid for each performed using Statview SEϩ.Differences between groups were of the three groups of women were obtained.
determined by Wilcoxon signed-rank test or Friedman test as appropriate.P Ͻ 0.05 was considered as statistically significant.

Collection, isolation and culture of endometrium and endometriotic cyst
Samples of uterine endometrium were obtained at the time of there may be specific interactions between stroma and epithe-in media increased (Figure 1, upper panel).Furthermore, in preliminary experiments, proliferation of endometrial cells lium that could influence the reciprocal growth, we have specifically decided not to separate these two components.cultured in 10% follicular fluid was also tested for a timeresponse effect.As shown in Figure 1 (lower panel), an Results are reported in Tables I and II.Peritoneal fluids from controls, women with endometriosis stage I-II and increased proliferation with time was observed.endometriosis stage III-IV patients and follicular fluids displayed a significant increased ability to stimulate endometrial Discussion cell proliferation when compared to correspondent control media (Table I).Similar results were obtained for endometriotic

Results
In this study, we demonstrated that both peritoneal and follicular fluids are able to stimulate endometrial and endometriotic cell cells, with the exception of peritoneal fluid from women with endometriosis stage III-IV in which an increased but not growth in vitro.Moreover, we observed that this effect is much more evident using follicular fluids.statistically significant ability to stimulate these cells in culture was observed (Table II).Furthermore, follicular fluid induced a Peritoneal fluid contains many growth factors and cytokines which have been advocated to be involved in the implantation higher cellular proliferation than peritoneal fluid from controls, patients with endometriosis stage I-II and women with and growth of ectopic endometrial fragments (Koutsilieris et al., 1991).Interestingly, it has been reported that the endometriosis stage III-IV.This significant increase in cellular growth induced by follicular fluids could be observed for both peritoneal fluid of women with endometriosis would induce an increase in endometrial stromal cells proliferation (Surrey endometrial (Table I) and endometriotic cells (Table II).Conversely, stimulation of cellular proliferation induced by and Halme, 1990).However, results from other studies on this subject are conflicting (Meresman et al., 1997;Overton et al., peritoneal fluids obtained from the different groups of patients resulted similar (Tables I and II).
1997).In the present study, we failed to detect an increased ability to stimulate endometrial cells proliferation using peri-Endometrial cells were also treated with different concentrations of follicular fluid.At the concentrations tested, the rate toneal fluids from women with endometriosis.However, this topic was not the main objective of our experimental design of cell proliferation increased as the content of follicular fluid follicular fluid may represent an extremely favourable environment for endometrial and endometriotic cells growth.Indeed, a marked and statistically significant enhanced ability to stimulate proliferation of these cells was observed using follicular fluids obtained from partially luteinized follicles.
Our study also implies that the follicular fluid-mediated induction of endometrial cell proliferation is not merely due to steroid hormones.Indeed, the control media used in this study did actually contain a concentration of steroid hormones similar to the experimental conditions tested.Therefore, other factors present in follicular fluid may be responsible for this effect.In particular, growth factors such as vascular endothelial growth factor (Lee et al., 1997), insulin-like growth factors I and II (VanDessel et al., 1996), steroidogenesis-inducing protein (Khan et al., 1997) and tumour necrosis factor-α (Punnonen et al., 1992) and cytokines such as interleukin (IL)-1 (Chen et al., 1995), IL-6 (Buyalos et al., 1992), IL-8 (Arici et al., 1996), monocyte chemotactic protein-1 (Arici et al., 1997b), leukaemia inhibiting factor (Arici et al., 1997a), as well as prostaglandins have been detected in follicular fluid.Interestingly, direct evidences of interaction between growth factors and sex steroids in the regulation of endometrial stromal cell growth have been reported (Irwin et al., 1991).However, factors and molecular interactions that are responsible for the ability of follicular fluid to increase endometrial stromal cell proliferation markedly were not further investigated herein, and are beyond the scope of the present study.Exciting new insights might be drawn from future investigations in this field.
Our results are generally in keeping with those from Bahtiyar et al. in terms of reported ability of follicular fluid to stimulate endometrial cell growth (Bahtiyar et al., 1998).Moreover, in that study follicular fluid from women with endometriosis was shown to induce an increased cell proliferation than that obtained using follicular fluid from women without the disease.It should, however, be noted that these authors failed to evaluate the mitogenic ability of follicular fluid relative to other milieus and did not investigate whether this effect was merely steroid-dependent.Interestingly, they postulated that repetitive release of follicular content into the peritoneal fluid at the time of ovulation may play a role in the growth of peritoneal ectopic endometrial implants.Although this hypothesis might have some relevance, it is tempting to found that some large endometriomas also had histological characteristics of luteal or follicular ovarian cysts suggesting and we cannot exclude that increasing the number of peritoneal fluid pools tested would lead to some differences.Indeed, the that these lesions may develop as a result of secondary involvement of functional ovarian cysts (Nezhat et al., 1992).central purpose of this study was to evaluate the ability of follicular fluid to affect endometrial and endometriotic cell Moreover, very recently, Jain and Dalton showed by serial transvaginal tracking of ovarian follicles that a chocolate cyst proliferation.To this specific aim, comparison with peritoneal fluid was established, since this milieu is a well-known can develop from an ovarian follicle (Jain and Dalton, 1999).
In each of the 12 cases reported by these authors, the diagnosis mitogenic source for endometrial cells (Koutsilieris et al., 1991).In this contest, results from our study suggest that was successively confirmed laparoscopically.
mitogen for cell lines derived from ovarian surface epithelial carcinomas.
vitro fertilization. Fertil. Steril., 68, 305-311. Meresman, G.F., Baranao, R.I., Tenenbaum, A. et al. (1997) Effect of peritoneal fluid from patients with mild and severe endometriosis on endometrial stromal cell proliferation.Arch.Gynecol. Obstet.,259,[109][110][111][112][113][114][115] laparoscopy using an endometrial biopsy curette from 11 proliferative Endometrial and endometriotic cells were cultured in follicular phase women affected by benign ovarian pathology but without fluid and peritoneal fluid diluted in serum-free media; the evidence of endometriosis.Moreover, six samples of ovarian growth induced in these conditions was compared with that endometriotic cysts were obtained from six proliferative phase women obtained by culturing these cells in medium supplemented affected by endometriosis stage III-IV.Histopathological examination with charcoal stripped CS and an equal content of oestrogen confirmed the endometriotic character of the cysts and both the phase of the menstrual cycle and the absence of pathological conditions of and progesterone (control media).In the present study, since TableI.Endometrial cell proliferation in the presence of follicular fluid and peritoneal fluid from women with and without endometriosis.Experimental media consisted of Ham's F-10 plus 10% follicular fluid or 10% peritoneal fluid obtained from patients with endometriosis stage I-II, from patients with endometriosis stage III-IV and from women without the disease.Control media consisted of Ham's F-10 plus charcoalstripped call serum added with a content of 17-β-oestradiol and progesterone correspondent to concentrations present in the experimental 0.05 compared with the other three conditions.TableII.Endometriotic cell proliferation in the presence of follicular fluid and peritoneal fluid from women with and without endometriosis.Experimental media consisted of Ham's F-10 plus 10% follicular fluid or 10% peritoneal fluid obtained from patients with endometriosis stage I-Il, from patients with endometriosis stage III-IV and from women without the disease.Control media consisted of Ham's F-10 plus charcoalstripped calf serum added with a content of 17-β-oestradiol and progesterone correspondent to concentrations present in the experimental 0.05 compared with the other three conditions.

Figure 1 .
Figure 1.Effect of follicular fluid on endometrial cell proliferation