Hormonal regulation of apoptosis and the Fas and Fas ligand system ...

Molecular Human Reproduction Vol.8, No.5 pp. 447–455, 2002

Hormonal regulation of apoptosis and the Fas and Fas ligand system in human endometrial cells Joon Song, Thomas Rutherford, Frederick Naftolin, Santiago Brown and Gil Mor1 Department of Obstetrics and Gynecology, Yale University, School of Medicine, 333 Cedar St FMB 202, New Haven, CT 06520, USA 1To

whom correspondence should be addressed. E-mail: [email protected]

The process of apoptosis is responsible for normal cellular turnover in numerous tissues throughout the body. The endometrial layer of the uterus shows steroid-dependent cyclic changes in structure and function. After a proliferative and secretory phase, steroid support is withdrawn and the uterine epithelium is shed. We hypothesize that the apoptosis observed in endometrial cells following hormonal withdrawal is mediated by the Fas/Fas ligand (FasL) system. Normal endometrial cells and endometrial cancer cells were cultured in the presence of estrogen and progesterone. In order to mimic physiological hormonal changes, estrogen and progesterone were removed from the media. Apoptosis was determined by 3-[4,5-dimethylthiazol-2-yl]-2,5-dephenyl tetrazolium bromide (MTT) assay and propidium iodide staining, while Fas and FasL expression were evaluated by Western blot analysis. The endometrial cells expressed Fas and low levels of FasL. Withdrawal of estrogen and/or progesterone from the culture induced apoptosis causing an ~50% decrease in cell viability. This coincided with increased Fas and FasL expression. Treatment of the cells with anti-FasL antibody prevented cell death following hormonal withdrawal. Estrogen and progesterone therefore represent survival factors which hamper cell death by impeding the expression of apoptotic factors. Our results indicate that Fas-mediated apoptosis is important for endometrial cycling and suggest that dysregulation of the Fas/FasL interactions may have an important role in the development of endometrial cancer. Key words: apoptosis/endometrium/estrogen/Fas/progesterone

Introduction Apoptosis plays a critical role in the maintenance of tissue homeostasis and represents a physiological mechanism to eliminate excess or dysfunctional cells. As a female reproductive organ, the uterus is characterized by continuous cycles of endometrial cell growth and apoptosis in response to hormonal changes. The endometrium shows prominent steroid-dependent cyclic changes in structure and function in preparation for pregnancy. Thus, in the absence of conception, steroid support is withdrawn and the endometrium is shed. While menstruation was previously thought to be entirely the result of ischaemic necrosis, earlier morphological studies have also demonstrated the presence of apoptosis in the human endometrium and this has been confirmed by recently performed exemplary assessment of fragmented DNA. While growth factors and hormones actively stimulate proliferation and suppress default death pathways by delivering survival signals, cell death is efficiently triggered by membrane death receptors. A well-characterized apoptotic pathway in reproductive tissues involves the signal transduction pathway of Fas (Song et al., 2000), a receptor which upon binding its © European Society of Human Reproduction and Embryology

ligand (FasL) induces apoptosis through an autocrine/paracrine signalling pathway (Nagata, 1994). When activated, the intracellular death domain (DD) in Fas binds to Fas-associatedDD-containing protein (FADD), which then recruits the FADDlike IL-1β-converting enzyme (FLICE) (Boldin et al., 1995; Muzio et al., 1996; Kuwana et al., 1998). The activation of FLICE leads to the formation of the death-inducing signalling complex (DISC) that initiates downstream activation of caspases 3, 6 and 7, or mitochondrial damage by recruiting caspase-8. We and others have shown that Fas and FasL are expressed in the human endometrium throughout the menstrual cycle (Garcia-Velasco et al., 1998; Yamashita et al., 1999). During the late proliferative phase, these proteins are primarily retained within the cell’s Golgi apparatus and cytoplasmic vesicles and are unable to interact and cause apoptosis. During the secretory phase, these proteins are extruded as part of the cellular membranes, where Fas can bind FasL and signal apoptosis. It has been demonstrated that FasL exhibits peak expression during the secretory phase (Peng et al., 1998). This appears to be controlled by the ovarian steroids estrogen and progesterone, which represent the two main balancing survival factors in 447

J.Song et al.

the human endometrium. Under the influence of estrogen, endometrial cells proliferate as the result of estrogen-induced activation of anti-apoptotic genes and inhibition of pro-apoptotic genes. On the other hand, progesterone acts as a differentiation factor, blocking mitosis and inducing terminal differentiation (Slayden et al., 2000). Withdrawal of progesterone induces the menstrual cascade beginning with shrinkage of the entire endometrium, which is characterized by the activation of numerous genes, including pro-apoptotic factors (Rudolph-Owen et al., 1998). In previous studies, we have demonstrated changes in Fas and FasL protein expression in the rat ovary during the menstrual cycle (Sapi et al., 2000) and in the mammary gland during pregnancy and lactation (Song et al., 2000). In this study, we evaluate the role of sex hormones as survival factors which potentially regulate the expression of Fas and FasL in the normal endometrium.

Materials and methods Chemicals RPMI 1640 and fetal bovine serum were purchased from Life Technologies (Grand Island, NY, USA). All other chemicals, unless otherwise specified, were obtained from Sigma Chemical Co. (St Louis, MO, USA). Tissue collection and cell culture Endometrial cell preparation Endometrial tissue samples were obtained from women undergoing diagnostic laparoscopy or hysterectomy for benign disease. The samples used for primary culture preparation were obtained from patients in the late proliferative or early secretory phase of the menstrual cycle. Informed consent was obtained from each woman prior to surgery using protocols approved by the Human Investigation Committee of Yale University. Tissues were kept in room temperature Hank’s balanced salt solution (HBSS) and transported to the laboratory for culture within 30 min. Glandular and stromal cell preparation has been previously described (Garcia-Velasco et al., 1998). In brief, endometrial cells were dispersed by incubation of tissue minces in HBSS containing HEPES (25 mmol/l), penicillin (200 IU/ml), streptomycin (200 mg/ml), collagenase-B (2 mg/ml, 15 IU/mg) [BoehringerMannheim, Gaithersburg, MD, USA (#1088-831)] and DNase (0.2 mg/ ml, 1500 IU/mg) for ~20 min at 37°C with constant agitation. Disaggregated endometrial stromal cells were separated from glands by filtration through a wire sieve (73 µm diameter pore); endometrial glands remained largely intact and were retained by the sieve, while stromal cells passed through into the filtrate. This separation was confirmed by direct microscopical observation. Cells were plated in Dulbecco’s modified Eagle’s medium (DMEM): Ham F12 (DMEM/F12 1:1, v/v) containing pen-streptomycine and antimycotics (1% v/v) and fetal bovine serum (FBS; 10% v/v) in plastic flasks (75 cm2), maintained at 37°C in a humidified chamber (5% CO2 in air) and allowed to replicate until confluent in a monolayer. Thereafter, the cells were passed by standard methods of trypsinization and plated in 6-well dishes for protein extraction and again allowed to replicate to confluence. After cells reached confluence, cultures were treated with 5% charcoal-stripped fetal bovine serum, phenol red-free media for 24 h before treatment with hormones was initiated. All treatments were conducted in charcoal-stripped serum, phenol red-free media.

448

Cell lines Human endometrial Ishikawa cells were purchased from the American Type Culture Collection (ATCC, Rockville, MD, USA). Cells were plated in DMEM containing antibiotics–antimycotics (1% v/v) and fetal bovine serum (10% v/v) in plastic flasks (75 cm2), maintained at 37°C in a humidified chamber (5% CO2 in air) and allowed to replicate until confluent in a monolayer. After cells reached 70–80% confluence, cultures were treated with serum-free, phenol red-free media for 24 h before the treatment with hormones was initiated. Hormonal treatment The medium for the experiment was serum- and phenol red-free DMEM Ham’s F-12 for Ishikawa cells or 5% charcoal-stripped serum for the primary cultures. Estradiol (10–8 mol/l) and progesterone (10–6 mol/l) were added to the cultures for 48 h, and then removed for another 24 or 48 h. Preparation for protein and Western blot analysis Proteins were prepared from Ishikawa cells and primary cultures of endometrial cells using TRIzol reagent (Gibco BRL, Gaithersburg, MD, USA) according to the manufacturer’s instructions. Proteins were separated by sodium dodecyl sulphate–polyacrylamide gel electrophoresis using 10% polyacrylamide gels and transferred to nitrocellulose membranes. Membranes were stained with Ponceau Red prior to the antibody incubation and the protein bands were photographed using a Kodak digital camera. Similarly, the gel was stained with Coomassie Blue and then photographed. Following immunoblotting, the intensity of the bands was normalized to the concentration of proteins present in the membranes and in the gel. The 1D Image Analysis Software (Scientific Imaging Kodak Company) was used for this purpose. Immunoblotting was performed after blocking the membranes with 5% powder milk in water. The blots were incubated first with antibodies for FasL (monoclonal antibody, clone 33; Transduction Laboratory, Lexington, KY, USA at 1:1000 dilution) or Fas (monoclonal antibody, B-10; Santa Cruz Biotechnology, Santa Cruz, CA, USA at 1:500 dilution) for 4 h. The secondary antibody (peroxidase-labelled horse anti-mouse; Vector, Burlingame, CA, USA) was developed with the TMB Peroxidase Substrate Kit (Vector). Since the specificity of clone 33 for FasL has been questioned, we confirmed the results using clone G247-4 from PharMingen (San Diego, CA, USA). No differences were found between the two antibodies. All the experiments were repeated at least three times and the intensity of the signal was analysed using a digital imaging analysis system (1D Image Analysis Software; Scientific Imaging Kodak Company). Actin was used as an internal control, in addition to Ponceau Red, to validate the right amount of protein loaded in the gels. Cell viability and Fas-mediated apoptosis Cell viability was measured by the MTT assay as previously described (Mor et al., 2000; Song et al., 2000). Single cell suspensions were prepared by passing the cells through a 21G needle, preventing the formation of aggregates. In each experiment triplicates were used and the experiments were repeated at least three times. We did not find differences in the proliferation rate between each individual well in the same group. A 200 µl suspension of 5⫻103 to 5⫻104 cells in minimal essential medium supplemented with 10% FBS for Ishikawa cells or in DMEM Ham’s F-12 supplemented with 10% FBS for endometrial primary cells was distributed into each well of a 96-well flat-bottomed microplate (Becton-Dickinson, Lincoln Park, NJ, USA) and incubated for 48 h. The medium was then removed and the cells were incubated with experimental medium containing various amounts of sex steroids as described above.

Hormonal regulation of apoptosis and Fas/FasL in endometrium To determine the sensitivity of our system to Fas-mediated apoptosis, variable concentrations of mouse monoclonal IgG anti-Fas Ab (R&D Systems, Minneapolis, MN, USA) were added to 200 µl of medium for 24 h under each experimental condition. Thereafter, 20 µl MTT [3-(4,5-dimethylthiazol-2-yl)-2,5-diphenyltetrazolium bromide; Sigma) dye was added to each well 4 h before the end of the incubation. The unreacted dye and medium in the wells were poured off, and 100 µl of acidified isopropyl alcohol was added to solubilize the reactive crystals. The absorbance at 540 nm was measured by an automatic microplate reader (Model 550; Bio-Rad, Hercules, CA, USA). All assays were performed in triplicate. The percentage of anti-proliferative effect was calculated using the

formula: % of anti-proliferative effect ⫽ 100 (1–absorbance of experimental well/absorbance of positive control well). Detection of apoptotic cells The presence of apoptotic cells was determined by propidium iodide staining and TUNEL assay. Endometrial cells (1⫻105) were incubated in chamber slides and treated as described above. A total of 1 µg/ml propidium iodide was applied to the chambers after each hormonal treatment and fixed with 4% paraformaldehyde. The presence of positively stained endometrial cells was evaluated using a Zeiss

Figure 1. Effects of estrogen and progesterone on the proliferation of endometrial Ishikawa cells. Ishikawa cells were treated with estrogen and/or progesterone for 24 h. The MTT assay was performed to evaluate cell viability. Estradiol increased cell proliferation and progesterone opposed this estrogenic effect. Each experiment was repeated at least three times. *, **P ⬍ 0.05; *estrogen versus control; **estrogen versus estrogen plus progesterone.

Figure 2. Effects of hormonal withdrawal on endometrial cell viability. (A) Ishikawa cells were treated with estrogen and progesterone followed by the removal of either estrogen or progesterone or both for another 24 h. Removal of both hormones induced a statistically significant increase in the percentage of dead cells in the withdrawal group as compared with the control. The MTT assay was performed at the end of each experiment. *, **, ***, ****P ⬍ 0.05; *estrogen and progesterone versus control; **estrogen and progesterone versus progesterone withdrawal; ***estrogen and progesterone versus estrogen withdrawal; and ****estrogen and progesterone versus withdrawal of both estrogen, progesterone or both. No significant difference was found between groups 3 and 4. (B) Endometrial glandular epithelial cells and stromal primary cultures were treated with estrogen and progesterone followed by the removal of either estrogen or progesterone or both for another 24 h. The MTT assay was performed at the end of each experiment. The graph shows the response of a representative experiment using glandular epithelial cells. *, **, ***, ****, *****P ⬍ 0.05; *estrogen versus control; **estrogen and progesterone versus control; ***estrogen and progesterone versus progesterone withdrawal; ****estrogen and progesterone versus estrogen withdrawal; and *****estrogen and progesterone versus withdrawal of both estrogen and progesterone.

449

J.Song et al. fluorescent microscope and Image Analysis software Openlab 2 (Improvision, Lexiton, MA, USA). TUNEL assay To further confirm the presence of apoptotic cells following hormonal withdrawal, we carried out the TUNEL assay according to a previously described method (Gavrieli et al., 1992) with few modifications.

450

Briefly, cells were incubated with permeabilization solution (0.1% Triton X-100 in 0.1% sodium citrate) for 2 min on ice. Following three washes with phosphate-buffered saline (PBS) and one rinse with distilled deionized water, the sections were covered with terminal deoxynucleotidyl transferase (TdT) buffer (30 mmol/l Trizma base, pH 7.2, 140 mmol/l sodium cacodylate and 1 mmol/l cobalt chloride). Then, TdT buffer containing 0.2 IU/µl TdT (Boehringer-Mannheim)

Hormonal regulation of apoptosis and Fas/FasL in endometrium and 10 µmol/l fluorescein 16-dUTP (Boehringer-Mannheim) was added to the cells, and the slides were incubated in a humidified chamber at 37°C for 60 min. The reaction was terminated by transferring the slides to 50 mmol/l Tris–HCl, pH 7, for 15 min at room temperature. The slides were then washed with PBS. Endogenous peroxidase was inactivated with 0.3% H2O2 in methanol for 15 min at room temperature followed by 3⫻10 min washings with PBS. To block the non-specific binding of the antibodies, sections were incubated with 5% normal goat serum in PBS–5% bovine serum albumin for 30 min at room temperature and washed for 3⫻5 min with PBS. The sections were then incubated for 2 h at room temperature with an anti-fluorescein antibody, Fab fragment from sheep, conjugated with alkaline phosphatase (AP; BoehringerMannheim) and the AP substrate Blue Kit (Vector; blue colour). Blocking of apoptosis with an anti-FasL monoclonal antibody Endometrial Ishikawa cells were treated with different concentrations (20, 100 and 500 ng/ml) of either anti-FasL monoclonal antibody (NOK-mAb; PharMingen) or mouse IgG1 isotype-matched control monoclonal antibody (PharMingen) for 24 h prior to hormonal treatment. Cell viability was determined using the MTT assay. Each assay was performed at least three times in triplicate. Statistical analysis Data were analysed for statistical significance by using Student’s t-, Mann–Whitney or Kruskal–Wallis tests. All experiments were repeated two or three times with indistinguishable results.

Results Effect of hormonal withdrawal on endometrial cell viability The apoptotic index in the human endometrium has been shown to correlate with changes in serum levels of estradiol and progesterone (Dahmoun et al., 1999). In order to further characterize the effects of changes in hormonal levels on endometrial cells, we developed an in-vitro system consisting of: (i) the human endometrial Ishikawa cell line; and (ii) primary cultures of glandular and stromal cells obtained from normal endometrium (late proliferative or early secretory phase). Using these two systems, we first evaluated the effect of hormonal treatment on cell viability as determined by the MTT assay. We found, as previously reported, that estradiol increased cell proliferation and that progesterone opposed this estrogenic effect (Figure 1). However, following incubation with estrogen and progesterone, removal of estrogen from the media for 24 h reduced the number of viable Ishikawa cells by 33% (P ⬍ 0.05; Figure 2A). In the case of progesterone withdrawal, cell viability decreased by 28%, while removal

of both estrogen and progesterone caused a 48% decrease (P ⬍ 0.05) in cell viability (Figure 2A). To further confirm these results in normal endometrial cells, we performed similar experiments using primary cultures of glandular and stromal endometrial cells. As seen with the Ishikawa cells, endometrial cells were sensitive to the removal of either estrogen or progesterone (Figure 2B). Thus, removal of estrogen or progesterone or a combination of both caused a statistically significant decrease in cell viability (Figure 2B). Presence of apoptotic cells following hormonal withdrawal In order to determine whether the decrease in the percentage of viable cells following hormonal withdrawal was due to apoptosis, we stained endometrial cells (both Ishikawa and primary culture cells) with propidium iodide (PI) and TUNEL following the hormonal treatment. As shown in Figure 3, no PI-positive cells were found in cultures treated with estrogen and progesterone. However, removal of estrogen and/or progesterone was associated with an increase in the number of PIand TUNEL-positive cells (Figure 3). Both assays, PI staining and TUNEL, gave similar results, confirming that the decrease in the number of cells was associated with apoptosis. Effect of hormonal withdrawal on Fas and FasL expression To study the mechanisms by which hormonal withdrawal may induce apoptosis of endometrial cells, we characterized the effect of estrogen and progesterone removal on Fas and FasL expression. Ishikawa cells and primary cultures of glandular and stromal cells were treated with either estrogen and progesterone or each hormone alone in the same concentration and conditions, as described above. Ishikawa cells in the presence of estrogen alone or estrogen and progesterone showed low levels of Fas protein expression (Figure 4A, columns 2 and 3). However, after progesterone removal Fas expression increased in a time-dependent manner. Thus, there were 15, 27, 43 and 200% increases after 3, 6, 24 and 48 h respectively (Figure 4A, columns 4–7). A similar effect on Fas expression was found following estrogen removal (Figure 5A, column 5) as well as withdrawal of both estrogen and progesterone (Figure 5A, columns 4–6). Signals from the Fas receptor are induced after binding with its ligand, FasL. Therefore, we examined the effect of hormonal changes on FasL expression. Contrary to Fas, FasL protein expression increased by 60% in cells treated with estrogen alone (Figure 4B, column 2) but did not increase with either progesterone or the combination of estrogen and progesterone

Figure 3. Detection of apoptosis in Ishikawa cells following hormonal withdrawal. (a) Propidium iodide staining. Ishikawa cells were treated as described in Figure 2. At 24 h after hormonal withdrawal, propidium iodide (1 µg/ml) was added to the cells and they were fixed with 4% paraformaldehyde. Presence of apoptotic cells was detected using a fluorescent microscope and image analysis software. (A) Endometrial cells in the presence of estrogen and progesterone (⫻250). (B) Endometrial cells following 24 h of estrogen and progesterone withdrawal (⫻250). Note the apoptotic cells with the background of live cells. (C) Same as B but at a higher magnification. Note the nuclear staining. The background shows the PI-negative cells (⫻400). (D) Anti-FasL treatment prevented apoptosis of endometrial cells following 24 h of hormonal withdrawal (⫻250). (b) TUNEL assay. Cells were treated as describe above and stained with TUNEL assay for the detection of apoptotic cells. The TUNEL assay showed similar results as those described for PI staining. (A) Endometrial cells in presence of estrogen and progesterone (⫻250). (B) Endometrial cells following 24 h of estrogen and progesterone withdrawal (⫻250). Note the apoptotic cells with the background of live cells. (C) Same as B but at a higher magnification. Note the nuclear staining. The background shows the PI-negative cells (⫻400). (D) Anti-FasL treatment prevented apoptosis of endometrial cells following 24 h of hormonal withdrawal (⫻250).

451

J.Song et al.

Figure 4. Fas and FasL protein expression in Ishikawa cells following progesterone withdrawal. A Western blot analysis for Fas and FasL was performed for endometrial cells 48 h after the withdrawal of progesterone. (A) Fas protein was present in low levels in Ishikawa cells, but increased in a time-dependent manner after progesterone was excluded from the culture. Peak expression was found after 48 h. (B) FasL protein increased following progesterone withdrawal. Lane 1: control; lane 2: cells treated with estrogen; lane 3: cells treated with estrogen and progesterone; lane 4: cells 3 h after withdrawal of progesterone; lane 5: cells 6 h after progesterone withdrawal; lane 6: cells 24 h after progesterone withdrawal; lane 7: cells 48 h after progesterone withdrawal.

Figure 5. Fas and FasL protein expression in Ishikawa cells following hormone withdrawal. A Western blot analysis for Fas and FasL was performed for endometrial Ishikawa cells after 48 h following withdrawal of estrogen and/or progesterone. (A) Fas protein expression increased after estrogen, progesterone, or estrogen and progesterone withdrawal (lanes 4, 5 and 6 respectively). (B) FasL expression increased in a similar manner to Fas expression. Note that the exclusion of both hormones had a more potent effect on FasL expression than the exclusion of progesterone or estradiol alone. Lane 1: control; lane 2: cells treated with estrogen and progesterone; lane 3: cells treated with progesterone; lane 4: cells 48 h after withdrawal of progesterone; lane 5: cells 48 h after withdrawal of estrogen; lane 6: cells 48 h after withdrawal of estrogen and progesterone.

452

Figure 6. Fas and FasL expression in endometrial primary culture cells. A Western blot analysis for Fas and FasL was performed with proteins obtained from primary cultures of endometrial cells prepared from normal cyclic endometrium. Endometrial glandular epithelial cells (EGC) or endometrial stromal cells (ESC) were treated with estrogen and progesterone for 48 h followed by the removal for 48 h of either progesterone or estrogen or both. Fas protein expression increased in EGC and ESC after the removal of either progesterone (lane 2), estrogen (lane 3) or the combination of both (lane 4). FasL expression showed a similar response to hormonal withdrawal. Lane 1: control cells treated in the presence of estrogen and progesterone; lane 2: cells 48 h after progesterone withdrawal; lane 3: cells 48 h after estrogen withdrawal; lane 4: cells 48 h after estrogen and progesterone withdrawal.

(Figure 4B, column 3 and data not shown). However, when endometrial cells were deprived of progesterone, FasL expression further increased in a time-dependent manner, very similar to the pattern noted with Fas expression, reaching its highest level of expression after 24 h (with a 316% increase in relation to the control; Figure 4B, columns 4–7). When we analysed the individual effects of estrogen or progesterone withdrawal on FasL expression, we found that the removal of each hormone had stimulatory effects on FasL expression. Quantification of the signal showed a 173% increase on Fas expression after 48 h of progesterone withdrawal (Figure 5B, column 4), a 260% increase in FasL expression after 48 h of estrogen withdrawal (Figure 5B, column 5), and a 300% increase after withdrawal of both estrogen and progesterone (Figure 5B, column 6). Similar experiments were performed with primary cultures of endometrial cells. As shown in Figure 6, Fas and FasL expression were up-regulated following hormonal withdrawal in both endometrial glandular cells and endometrial stromal cells.

Hormonal regulation of apoptosis and Fas/FasL in endometrium

Figure 8. Protective effect of anti-FasL antibody following hormonal withdrawal. A blocking anti-FasL antibody was applied to Ishikawa cells and primary culture of glandular and stromal cells prior to and during hormonal withdrawal. The MTT assay was performed to evaluate cell viability. After the removal of estrogen and progesterone, treatment with an anti-FasL antibody prevented cell death in a dose-dependent manner. No effect was found with a control IgG. * versus ** and * versus ***: P ⬍ 0.05. The data is from a representative study with Ishikawa cells. Similar results were found using endometrial primary cultures.

apoptosis in Fas-sensitive cells (Ogasawara et al., 1993). Cells treated in the presence of estrogen and progesterone were resistant to Fas-mediated apoptosis (Figure 7A, P ⬎ 0.05). However, following hormone withdrawal, the application of anti-Fas antibody further decreased the percentage of viable cells (Figure 7A, P ⬎ 0.05). Studies on Ishikawa cells showed that removal of estrogen and progesterone increased the sensitivity to anti-Fas antibody in a dose-dependent manner (Figure 7B, P ⬍ 0.05). Figure 7. Effect of hormonal withdrawal on Fas-mediated apoptosis. The sensitivity of Ishikawa cells and primary cultures of glandular and stromal cells to Fas-mediated apoptosis was evaluated by treating the cells with anti-Fas antibody following the hormonal treatments. The MTT assay was performed at the end of the antibody treatment. (A) Primary endometrial cells showed an increase in sensitivity to Fas-mediated apoptosis following hormonal withdrawal. *, **not significant; ***P ⬍ 0.05. (B) Following the removal of estrogen and progesterone, Ishikawa cells showed an increased sensitivity to the anti-Fas antibody in a dose-dependent manner, *P ⬍ 0.05. No effect was found with control IgG. Similar effects were found between Ishikawa cells and endometrial primary cultures.

Inhibition of hormonal withdrawal-induced apoptosis in endometrial cells by an anti-FasL antibody To confirm that the Fas/FasL system is involved in the hormonal withdrawal-induced apoptosis of endometrial cells, we incubated the cells with a blocking anti-FasL antibody prior to and during hormonal removal. As shown in Figures 2 and 8, cell viability decreased by 48% following the removal of estrogen and progesterone; however, this effect was inhibited when cells were treated with blocking anti-FasL antibody in a dose-dependent manner (Figure 8, P ⬍ 0.05). This was also confirmed using PI and TUNEL staining; no apoptotic cells were found in the cultures treated with anti-FasL monoclonal antibody prior to the hormonal removal (Figure 3D).

Sensitivity to Fas-mediated apoptosis We next sought to confirm that the decrease in viable cells found after hormonal withdrawal was indeed related to the expression of Fas and FasL. We treated endometrial cells with an anti-Fas monoclonal antibody that is known to induce

Discussion The glandular and stromal components of human endometrium are under strict hormonal control. During the menstrual cycle, proliferation, stimulated by estradiol, reaches its maximum at 453

J.Song et al.

the time of ovulation (Ferenzy and Guralnic, 1983). Progesterone inhibits estradiol-induced cell proliferation of endometrial epithelium after ovulation and induces differentiation into secretory cells (Clarke and Sutherland, 1990). In the absence of implantation, a decrease in the levels of sex hormones induces a cascade of events resulting in the shedding of the endometrium, known as menstruation (Noyes et al., 1975). The control of cell proliferation and tissue remodelling involves a delicate balance between hormones acting as survival factors and the activation of default death pathways. The present study illustrates the relationship between sex hormones (estrogen and progesterone) and the Fas/FasL system in the regulation of endometrial cell growth and apoptosis. The growth, development and maintenance of the female sex organs involves a balance between the actions of the two major female sex steroids estradiol and progesterone (Clarke and Sutherland, 1990). Estrogen acts as a survival factor by inhibiting default apoptotic pathways, either directly or through growth factors and cytokines. Progesterone prepares the uterine lining for implantation of the embryo, making nutrients available for its development (Nayak et al., 2000; Slayden et al., 2000). This differential effect of progesterone has two biological consequences: one is to inhibit further proliferation and the second is to induce a commitment to cell death or apoptosis at the end of the cell’s biological function. Originally, menstruation was regarded as being associated with ischaemic necrosis of the functional layer of the endometrium. More recently, electron microscopic studies have revealed the presence of apoptotic bodies in human endometrial epithelial cells during the late secretory phase (Kokawa et al., 1996). The capacity of hormonal withdrawal to give rise to apoptosis has been well documented in the endometrium (Rotello et al., 1992; Nayak et al., 2000). Using electromicroscopy and immunocytochemistry techniques, Yamashita and co-workers described the presence of Fas and FasL in the normal human endometrium (Yamashita et al., 1999). Furthermore, the authors demonstrated that Fas and FasL proteins are localized on the Golgi apparatuses and vesicles of endometrial cells during the late proliferative phase. During the secretory phase, Fas and FasL are incorporated into cell membranes and are coexpressed on the cell membrane of the endometrial glands throughout this last phase of the menstrual cycle (Yamashita et al., 1999). This pattern of expression suggested that apoptosis of the glandular epithelium might be mediated by the Fas/ FasL system, either in an autocrine or paracrine way through soluble FasL (French et al., 1997; Suda et al., 1997). According to our results, that may be the case. Thus, we have found that in the presence of estrogen and progesterone, Fas-mediated apoptosis in the glandular epithelium is inhibited. Following hormonal withdrawal (menstrual phase) Fas and FasL are coexpressed, and this expression is followed by apoptosis. The regulation of FasL expression by estrogen follows a slightly different pattern to that of Fas protein. Its expression is increased under estrogen treatment. Although estrogen has been proven to be a hormone that promotes cell survival, stimulation of endometrial cells through the estrogen receptor leads to increased cell surface expression of FasL, an apoptotic protein. This estrogenic effect demonstrates that during cell 454

growth and differentiation there is a tight connection between the signalling pathways that support cell survival and those that culminate in apoptosis. This result is in accordance with our previous study showing that estrogen up-regulates FasL expression in breast cancer cells (Mor et al., 2000). In order to explain this apparently contradictory effect, we proposed that FasL expression in proliferative cells represents a mechanism by which estrogen contributes to maintaining the balance between cell proliferation and cell death. According to our model, apoptosis of proliferative cells is prevented through the inhibition of Fas. Changes in the endometrial microenvironment induced by hormonal withdrawal will deprive the cells of their support/ survival stimuli, thereby allowing the activation of the apoptotic machinery through the expression and activation of Fas receptor. The co-expression of Fas and FasL on the surface of the cells is followed by apoptosis. Tumour suppresser genes, such as p53 and c-myc (Okazawa et al., 1996; Fuchs et al., 1997; Roberts et al., 1997), induce Fas expression and/or its translocation to the cell surface, where it meets its ligand (FasL) and induces apoptosis. On the other hand, tumour survival factor bcl-2 may block Fas/ FasL-mediated apoptosis in human endometrium during the proliferative phase (Otsuki et al., 1994; Yamashita et al., 1999). Apoptosis allows both normal tissue remodelling and the removal of mutated cells. However, alterations in Fas expression/activation may lead to the proliferation of transformed cells expressing FasL but not Fas on their surface (Niehans et al., 1997). FasL expression on neoplastic cells is further enhanced by cytokines and hormones produced by the immune infiltrate present in the vicinity of the tumour (Mor et al., 1998a,b). Consequently, tumours may elude immune surveillance by inducing, via the Fas/FasL system, the apoptosis of activated lymphocytes (Gutierrez et al., 1999). In this study, the exclusion of hormones from the medium induced Fas protein expression as well as FasL. Cells incubated in hormone-containing medium were resistant to Fas-mediated apoptosis, but became sensitive following hormone withdrawal. Furthermore, treatment of the cells with a blocking anti-FasL antibody during hormonal removal prevented apoptosis. This not only suggests that FasL is biologically active, but also confirms that the up-regulation of Fas is a crucial event for cell death following hormonal withdrawal (Song et al., 2000). In summary, the late follicular phase of the menstrual cycle, characterized by a shift of the sex hormonal environment, induces apoptosis of endometrial cells, representing the initial step of endometrial remodelling. Progesterone, the main hormone during the secretory phase, has a major role in converting cells from a proliferative to a functional stage, thus preparing the tissue for implantation. Without pregnancy, a rapid decrease in hormonal levels induces apoptosis in a proportion of endometrial cells, which eventually spreads to the entire endometrial lining. In our in-vitro model, we have shown that apoptosis is one of the crucial mechanisms in cell death following hormonal withdrawal and that the Fas/FasL system may represent one of the main pathways mediating tissue remodelling. Furthermore, our results suggest that it is the

Hormonal regulation of apoptosis and Fas/FasL in endometrium

removal of hormones from the micro-environment that is the main factor controlling tissue turnover. Based on these results, we question the effectiveness of continuous versus cyclic progesterone treatment for protection against endometrial cancer in post-menopausal women receiving estrogen replacement therapy. Our results also indicate that the Fas/FasL system may be important for endometrial homeostasis and may play a central role in the removal of pre-cancer cells. Defects in this pathway may alter the normal maintenance of the tissue and increase the risk of endometrial cancer.

Acknowledgement This work was in part supported by grants from the National Institutes of Health RO1 HD37137-01A2 and RO1 CA92435-1.

References Boldin, M.P., Varfolomeev, E.E., Pancer, Z., Mett, I.L., Camonis, J.H. and Wallach, D. (1995) A novel protein that interacts with the death domain of Fas/APO1 contains a sequence motif related to the death domain. J. Biol. Chem., 270, 7795–7798. Clarke, C.L. and Sutherland, R.L. (1990) Progestin regulation of cellular proliferation. Endocr. Rev., 11, 266–301. Dahmoun, M., Boman, K., Cajander, S., Westin, P. and Backstrom, T. (1999) Apoptosis, proliferation and sex hormone receptors in superficial parts of human endometrium at the end of the secretory phase. J. Clin. Endocrinol. Metab., 84, 1737–1743. Ferenzy, A. and Guralnic, M. (1983) Endometrial microstructure: structure– function relationship throughout the menstrual cycle. Semin. Reprod. Endocrinol., 1, 205–219. French, L., Wilson, A., Hahne, M., Viard, I., Tschopp, J. and Macdonald, H. (1997) Fas ligand expression is restricted to nonlymphoid thymic components in situ. J. Immunol., 159, 2196–2202. Fuchs, E., McKenna, K. and Bedi, A. (1997) P53-dependent DNA damageinduced apoptosis requires Fas/Apo-1-independent activation of Cpp32beta. Cancer Res., 57, 2550–2554. Garcia-Velasco, J., Arici, A., Naftolin, F., Zreick, T. and Mor, G. (1998) Macrophage-derived growth factors regulate FasL expression in endometrial stromal cells. Mol. Hum. Reprod., 5, 642–650. Gavrieli, Y., Sherman, Y. and Ben-Sasson, S. (1992) Identification of programmed cell death in situ via specific labeling of nuclear DNA fragmentation. J. Cell Biol., 119, 493–501. Gutierrez, L., Eliza, M., Niven-Fairchild, T., Naftolin, F. and Mor, G. (1999) The Fas/Fas-Ligand system: a mechanism for immune evasion in human breast carcinomas. Breast Cancer Res. Treat., 54, 245–253. Kokawa, K., Shikone, T. and Nakano, R. (1996) Apoptosis in the human uterine endometrium during the menstrual cycle. J. Clin. Endocrinol. Metab., 81, 4144–4147. Kuwana, T., Smith, J.J., Muzio, M., Dixit, V., Newmeyer, D.D. and Kornbluth, S. (1998) Apoptosis induction by caspase-8 is amplified through the mitochondrial release of cytochrome c. J. Biol. Chem., 273, 16589–16594. Li, S., Han, H., Resnik, E., Carcangiu, M.L., Schwartz, P.E. and Yang-Feng, T.L. (1993) Advanced ovarian carcinoma: molecular evidence of unifocal origin. Gynecol. Oncol., 51, 21–25. Mor, G., Gutierrez, L., Eliza, M., Kahyaoglu, F. and Arici, A. (1998a) Fas– Fas ligand system induced apoptosis in human placenta and gestational trophoblastic disease. Am. J. Reprod. Immunol., 40, 89–95. Mor, G., Wei, Y., Santen, R., Gutierrez, L., Eliza, M., Lev, B., Harada, N.,

Wang, J., Lysiak, J., Diano, S. and Naftolin, F. (1998b) Macrophages, estrogen and the microenvironment of breast cancer. J. Steroid Biochem. Mol. Biol., 67, 403–411. Mor, G., Kohen, F., Garcia-Velasco, J., Nilsen, J., Brown, W., Song, J. and Naftolin, F. (2000) Regulation of Fas ligand expression in breast cancer cells by estrogen: functional differences between estradiol and tamoxifen. J. Steroid Biochem. Mol. Biol., 73, 185–194. Muzio, M., Chinnaiyan, A.M., Kischkel, F.C., O’Rourke, K., Shevchenko, A., Ni, J., Scaffidi, C., Bretz, J.D., Zhang, M., Gentz, R. et al. (1996) FLICE, a novel FADD-homologous ICE/CED-3-like protease, is recruited to the CD95 (Fas/APO-1) death-inducing signaling complex. Cell, 85, 817–827. Nagata, S. (1994) Fas and Fas ligand: a death factor and its receptor. Adv. Immunol., 57, 129–135. Nayak, N.R., Critchley, H.O., Slayden, O.D., Menrad, A., Chwalisz, K., Baird, D.T. and Brenner, R.M. (2000) Progesterone withdrawal up-regulates vascular endothelial growth factor receptor type 2 in the superficial zone stroma of the human and macaque endometrium: potential relevance to menstruation. J. Clin. Endocrinol. Metab., 85, 3442–3452. Niehans, G., Brunner, T., Frizelle, S., Liston, J., Salerno, C., Knapp, D., Green, D. and Kratzke, R. (1997) Human lung carcinomas express Fas ligand. Cancer Res., 57, 1007–1012. Noyes, R.W., Hertig, A.T. and Rock, J. (1975) Dating the endometrial biopsy. Am. J. Obstet. Gynecol., 122, 262–263. Ogasawara, J., Watanabe-Fukunaga, R., Adachi, M., Matsuzawa, A., Kasugai, T., Kitamura, Y., Itoh, N., Suda, T. and Nagata, S. (1993) Lethal effect of the anti-Fas antibody in mice. Nature, 364, 806–809. Okazawa, H., Shimizu, J., Kamei, M., Imafuku, I., Hamada, H. and Kanazawa, I. (1996) Bcl-2 inhibits retinoic acid-induced apoptosis during the neural differentiation of embryonal stem cells. J. Cell Biol., 132, 955–968. Otsuki, Y., Misaki, O., Sugimoto, O., Ito, Y., Tsujimoto, Y. and Akao, Y. (1994) Cyclic bcl-2 gene expression in human uterine endometrium during menstrual cycle. Lancet, 344, 28–29. Peng, X., Maruo, T., Matsuo, H., Takekida, S. and Deguchi, J. (1998) Serum deprivation-induced apoptosis in cultured porcine granulosa cells is characterized by increased expression of p53 protein, Fas antigen and Fas ligand and by decreased expression of PCNA. Endocr. J., 45, 247–253. Roberts, R.A., Nebert, D.W., Hickman, J.A., Richburg, J.H. and Goldsworthy, T.L. (1997) Perturbation of the mitosis/apoptosis balance—a fundamental mechanism in toxicology. Fund. Appl. Toxicol., 38, 107–115. Rotello, R.J., Lieberman, R.C., Lepoff, R.B. and Gerschenson, L.E. (1992) Characterization of uterine epithelium apoptotic cell death kinetics and regulation by progesterone and RU 486 Am. J. Pathol., 140, 449–456. Rudolph-Owen, L.A., Slayden, O.D., Matrisian, L.M. and Brenner, R.M. (1998) Matrix metalloproteinase expression in Macaca mulatta endometrium: evidence for zone-specific regulatory tissue gradients. Biol. Reprod., 59, 1349–1359. Sapi, E., Brown, W., Aschkenazi, S., Tartaro, K., Lim, C., Munoz, A., Kacinski, B., Rutherford, T. and Mor, G. (2002) Regulation of Fas ligand expression by estrogen in normal ovary. J. Soc. Gynecol. Invest., in press. Slayden, O.D., Rubin, J.S., Lacey, D.L. and Brenner, R.M. (2000) Effects of keratinocyte growth factor in the endometrium of Rhesus macaques during the luteal–follicular transition. J. Clin. Endocrinol. Metab., 85, 275–285. Song, J., Sapi, E., Brown, W., Nilsen, J., Tartaro, K., Kacinski, B., Naftolin, F. and Mor, G. (2000) Expression of Fas and Fas ligand during pregnancy, lactation and involution and its potential role during mammary gland remodeling. J. Clin. Invest., 106, 1209–1224. Suda, T., Hashimoto, H., Tanaka, M., Ochi, T. and Nagata, S. (1997) Membrane Fas ligand kills human peripheral blood T lymphocytes and soluble Fas ligand blocks the killing. J. Exp. Med., 186, 2045–2050. Yamashita, H., Otsuki, Y., Matsumoto, K., Ueki, K. and Ueki, M. (1999) Fas ligand, Fas antigen and Bcl-2 expression in human endometrium during the menstrual cycle. Mol. Hum. Reprod., 5, 358–364. Submitted on May 14, 2001; resubmitted on October 11, 2001; accepted on February 8, 2002

455