Apoptosis of Bovine Granulosa Cells After Serum Withdrawal Is ...

Download PDF
3 downloads 0 Views 986KB Size Report
Comment
ABSTRACT. Ovarian follicular atresia occurs by apoptosis of granulosa and theca cells. The Fas antigen (Fas), a cell surface receptor that triggers apoptosis ...
BIOLOGY OF REPRODUCTION 64, 518–526 (2001)

Apoptosis of Bovine Granulosa Cells After Serum Withdrawal Is Mediated by Fas Antigen (CD95) and Fas Ligand1 Che-Lin Hu, Robert G. Cowan, Rebecca M. Harman, Dale A. Porter, and Susan M. Quirk2 Department of Animal Science, Cornell University, Ithaca, New York 14853 ABSTRACT

events associated with apoptosis. The aim of the current studies was to determine whether the Fas antigen (Fas) plays a role in serum withdrawal-induced apoptosis of bovine granulosa cells. Fas is a cell surface receptor that, when engaged by Fas ligand (FasL) or specific agonistic antibodies, triggers apoptosis. Binding of an adaptor protein, FADD, to the death-effector domain of Fas activates a cascade of cysteine proteases known as caspases that cleave cellular substrates, resulting in cell death. Expression of Fas, FasL, or both has been demonstrated in ovarian follicles of rats, mice, humans, cows, and pigs [5, 9–17] and, under certain conditions, cultured follicular cells respond to treatment with FasL or agonistic anti-Fas antibodies by undergoing apoptosis [9, 13, 14, 16–18]. Thus, Fas/ FasL interaction is a potential signal for the initiation of apoptosis by follicle cells, resulting in atresia. The current study reports evidence that serum withdrawal-induced apoptosis of bovine granulosa cells is associated with increased expression of Fas and FasL. Furthermore, interaction between Fas and FasL is at least one of the mechanisms responsible for apoptosis. The fact that cells expressing FasL do not also stain with a marker for leukocytes provides evidence that FasL expression in the granulosa layer of follicles is not simply due to infiltration of immune cells that express FasL.

Ovarian follicular atresia occurs by apoptosis of granulosa and theca cells. The Fas antigen (Fas), a cell surface receptor that triggers apoptosis when activated by Fas ligand (FasL), may be involved in this process. A possible role of the Fas pathway in mediating serum withdrawal-induced apoptosis of granulosa cells was examined. Granulosa cells collected from 5- to 10-mm bovine follicles were cultured in DMEM-F12 containing serum for 3 days, deprived of serum, and live cells were counted at various times after serum withdrawal. Cell death increased significantly 6 h after serum withdrawal (21% 6 7%; P , 0.05 vs. 0 h) and continued to increase until 24 h (43% 6 6%). No further increases in cell death were observed through 72 h. Detection of the translocation of phosphatidylserine to the outer surface of the cell membrane by annexin V binding indicated that cells died by apoptosis. Quantitative reverse transcriptasepolymerase chain reaction assays showed no changes in Fas mRNA levels but a 4.7-fold increase in FasL mRNA 3 h after serum withdrawal (P , 0.05 vs. 0 h). FasL mRNA remained elevated through 24 h and returned to basal levels at 48 h. Immunohistochemical staining showed that both Fas and FasL protein increased on the cell surface within 3 h and remained elevated through 12 h (the last time point tested). Binding of FasL to Fas was blocked with two reagents that bind to the extracellular domain of FasL: an anti-FasL antibody and Fas:Fc, a chimeric protein consisting of the Fc portion of human immunoglobulin G and the extracellular domain of human Fas. Cell death 24 h after serum withdrawal was reduced 55% 6 10% and 34% 6 12% by anti-FasL antibody and Fas:Fc, respectively (P , 0.05 vs. no blocking protein). In conclusion, serum withdrawal-induced apoptosis of bovine granulosa cells is mediated at least partially by Fas/FasL interactions. These results are consistent with a potential role of Fas in an autocrine or paracrine pathway to trigger ovarian follicular atresia.

MATERIALS AND METHODS

Materials

All culture media and additives were obtained from Life Technologies, Inc. (Grand Island, NY). Sodium pyruvate, L-glutamine, and mouse immunoglobulin M (IgM) were from Sigma Chemical Company (St. Louis, MO). Tissue culture plates were obtained from Corning-Costar (Cambridge, MA) and Slide-Well chambers were from NuncIntermed (Naperville, IL). Polyclonal rabbit anti-rat FasL antibody (IgG; antibody C-178) was from Santa Cruz Biotechnology, Inc. (Santa Cruz, CA), monoclonal mouse antihuman Fas antibody (IgM; clone CH-11) and recombinant soluble human FasL were from Upstate Biotechnology (Lake Placid, NY), and mouse anti-bovine CD45 antibody was from Biosource (Camarillo, CA). Human Fas:immunoglobulin Fc was from Pharmingen (San Diego, CA). Rabbit IgG and phycoerythrin (PE)-conjugated goat antimouse IgM were obtained from Jackson ImmunoResearch Laboratories, Inc. (West Grove, PA). Alexa 488-conjugated goat anti-rabbit IgG and Alexa 488-conjugated annexin V (Vybrant apoptosis assay kit) were obtained from Molecular Probes (Eugene, OR). Avian myeloblastosis virus (AMV) reverse transcriptase (RT) was obtained from Promega (Madison, WI), random hexamer from Pharmacia (Piscataway, NJ), and Taq polymerase from Fisher (Pittsburgh, PA). Bovine interferon-g (IFN) was graciously provided by Dr. Dale Godson, Veterinary Infectious Disease Organization (Saskatoon, SK, Canada).

apoptosis, granulosa cells, ovary

INTRODUCTION

More than 99% of ovarian follicles fail to ovulate and undergo atresia at various stages of development. Follicular atresia occurs by apoptosis, or programmed cell death, of granulosa and theca cells [1]. Gonadotropins and locally produced intraovarian growth factors regulate folliculogenesis and atresia [2, 3]. Removal of serum from culture media induces apoptosis of granulosa cells [4–6] and other types of cells [7, 8] and has been used as a model to study This work was supported by grant 98-35203-6220 from the U.S. Department of Agriculture and grant HD 32535 from the National Institutes of Health. 2 Correspondence: Susan M. Quirk, 258 Morrison Hall, Cornell University, Ithaca, NY 14853. FAX: 607 255 9829; e-mail: [email protected] 1

Received: 19 May 2000. First decision: 12 June 2000. Accepted: 14 September 2000. Q 2001 by the Society for the Study of Reproduction, Inc. ISSN: 0006-3363. http://www.biolreprod.org

518

FAS-MEDIATED APOPTOSIS OF GRANULOSA CELLS

Cell Culture and Treatment

Freshly excised cow ovaries from an abattoir were transported in saline at room temperature (approximately 1.5 h) and processed immediately thereafter. Granulosa cells were obtained by aspiration of 5- to 10-mm follicles with a 16gauge needle, followed by flushing follicles once with Dulbeccos modified Eagles medium (DMEM)-Hams F12 medium. Cells were collected by centrifugation, washed, counted, and cultured in DMEM-F12 (supplemented with 1 mM pyruvate, 2 mM glutamine, 100 U/ml penicillin, 100 mg/ml streptomycin, and 0.25 mg/ml fungizone) containing 10% FBS for 2 days. Cells were plated on Day 0 at 5 3 104 cells/well in 96-well plates for cell viability assays and at 5 3 105 cells/well in 6-well plates for analysis of RNA. Plating densities were determined experimentally so that cultures reached confluence between Days 2 and 3. On Day 2 of culture, media was changed to DMEM-F12 containing 5% FBS. On Day 3, viable cell number was determined to provide a baseline for subsequent cell counts. Media was immediately changed to DMEM-F12 without FBS, and viability was determined 0, 3, 6, 12, 24, 48, and 72 h after serum withdrawal. Cells were removed from the well by trypsin and live cells were counted in a hemacytometer using trypan blue-exclusion. Percent killing was calculated by subtracting the number of viable cells counted at each time after serum withdrawal from the number of viable cells counted just prior to serum withdrawal and dividing by the number of viable cells prior to serum withdrawal. Cultures were frozen at the same time points except for 72 h for subsequent preparation of total cellular RNA. Cultures at 72 h appeared insufficiently healthy for preparation of RNA. Four replicate wells were counted at each time point and experiments were repeated using three separate granulosa cell preparations. Assessment of Apoptosis

An early and specific marker for apoptosis is the translocation of phosphatidylserine from the inside to the outside of the cell membrane [19]. This process can be detected by the binding of fluorescent-labeled annexin V to phosphatidylserine in conjunction with the vital dye, propidium iodide. Healthy live cells do not stain with either annexin V or PI, cells in early stages of apoptosis stain with annexin V but not PI, whereas dead and dying cells stain with both annexin V and PI. Staining was performed using the Vybrant apoptosis assay kit according to the manufacturer’s instructions as described previously [16]. Cells were cultured as described above and stained 12 h after serum withdrawal. Counts of viable cells by trypan blue-exclusion indicated that the number of viable cells progressively decreased between 6 and 24 h, suggesting that a population of cells in the early stages of apoptosis should be detectable at 12 h. Cultures at later times would have predominantly dead cells. The assay was repeated on three separate granulosa cell preparations. Isolation of a Bovine FasL cDNA Clone

A 342-base pair (bp) fragment of bovine FasL cDNA spanning the positions of several introns and corresponding to position 387 to 729 of human FasL (GenBank accession number U11821) was cloned by reverse transcription-polymerase chain reaction (RT-PCR) of bovine spleen RNA. The fragment was amplified using a human sense primer (position 364 to 386, in exon 1 [20]) and a bovine antisense

519

primer (corresponding to human position 698 to 729, in exon 4). The bovine primer was designed using a 168-bp cDNA sequence for bovine FasL [21]. Amplification consisted of a preincubation at 948C for 5 min before adding Taq polymerase and then 40 cycles at 948C for 30 sec, 558C for 30 sec, and 728C for 30 sec. DNA sequencing of four independent clones indicated 88% and 82% fragment sequence homology with human and mouse FasL cDNA, respectively. Analysis of Fas and FasL mRNA Expression by Competitive RT-PCR

The RT-PCR assay for bovine Fas mRNA was described previously [16]and was modified to facilitate concomitant assay of Fas and FasL mRNA. Total RNA was prepared from cultures of granulosa cells at 0, 3, 6, 12, 24, and 48 h after serum withdrawal. RNA (0.8 mg) was reverse transcribed in the presence of various amounts of internal bovine Fas and FasL RNA standards (10–1300 attomoles/reaction for Fas and 2–205 attomoles/reaction for FasL) using AMV-RT and random hexamer primers. Fas and FasL RNA standards were prepared by in vitro transcription of fragments of mutated Fas antigen cDNA as described [16] and FasL cDNA. The bovine FasL cDNA fragment contained a 49-bp deletion internal to PCR primer binding sites (corresponding to position 652 to 700 of human FasL cDNA [GenBank accession number U11821]). Complementary DNA in RT reactions was amplified in the presence of 32PdCTP. For analysis of Fas mRNA, PCR primers were designed as described [16] to span the position of three introns and to generate a 206-bp fragment for the sample RNA and a 159-bp fragment for the internal RNA standard. For analysis of FasL mRNA, primers were designed to generate a 267-bp fragment for the sample RNA and a 218-bp fragment for the internal RNA standard (positions of 59 and 39 primers correspond to positions 462 to 482 in exon 2 and 729 to 701 in exon 4 of human FasL, respectively; numbering according to GenBank accession number U11821). Amplification consisted of a preincubation at 948C for 5 min before adding Taq polymerase and then 40 cycles (Fas) or 37 cycles (FasL) at 948C for 30 sec, 608C for 30 sec, and 728C for 30 sec. PCR products were fractionated on a 2% agarose gel. The gel was dried and radioactive signal was quantified on a Fuji BAS1000 PhosphorImager. The concentration of Fas or FasL mRNA in each sample was calculated by regression of the log (sample signal/standard signal) vs log standard concentration as previously described [16]. The sample concentration equals the standard concentration at the point where the sample signal equals the standard signal. The sample concentration is corrected for the 50-bp (Fas) or 49-bp (FasL) difference in the PCR product length between the sample and the standard. Samples from the same experiment were assayed together. Each assay included an RNA pool (prepared from bovine granulosa cells for Fas; bovine spleen cells for FasL) used to calculate the between assay coefficient of variation (coefficient of variation [CV] 5 9.2% for Fas assays and 16.0% for FasL assays; n 5 3 each). Immunohistochemistry of Fas, FasL, and CD45

Granulosa cells were plated at 7 3104 to 1 3 105 cells/ well on 8-well glass slides on Day 0 and cultured as described above in DMEM-F12 containing FBS for 72 h before serum withdrawal. At various times after serum withdrawal (0, 3, 6, and 12 h), cells were fixed in acetone for

520

HU ET AL.

FIG. 1. Serum withdrawal-induced cell death in granulosa cell cultures. Granulosa cells were cultured in DMEM-F12 containing FBS for 3 days and then cultured in media without FBS (time 5 0 h). Cells were counted at different times after serum withdrawal. Percent killed was calculated based on cell counts of 0-h cultures in DMEM-F12 1 FBS. Bars represent the mean 6 SEM of three experiments using separate granulosa cell preparations. Bars with different letters are significantly different (P , 0.05).

2 min at 2208C, and rinsed twice in 13 PBS, 5 min each. Later times were not tested due to the degrading quality of the cultures. Cells were incubated in PBS-2% NGS containing 2 mg/ml anti-Fas antibody or 10 mg/ml anti-FasL antibody at 378C for 1 h. Nonspecific binding was assessed with mouse IgM or rabbit IgG, respectively. Two mg/ml PE-conjugated goat anti-mouse IgM and 2 mg/ml Alexa 488-conjugated goat anti-rabbit IgG were used for the fluorescent detection of anti-Fas and -FasL, respectively. Immunohistochemistry for Fas and FasL was performed on three separate granulosa cell preparations. Granulosa cell preparations were incubated with 2.5 mg/ ml anti-CD45 antibody to determine the percentage of leukocytes present. Alexa 488-conjugated goat anti-mouse IgG was used for fluorescent detection. Cells were fixed and incubated as described above. Dual staining with 1 mg/ml PI was performed after CD45 staining to visualize the granulosa cell nuclei. Images were obtained of four randomly chosen fields using a Spot II Digital Camera (Diagnostic Instruments, Sterling Heights, MI) and the number of CD45-positive cells and total cells (PI-stained nuclei) were counted. Fields contained a minimum of 139 cells. The experiment was performed on three separate granulosa cell preparations. As a positive control, bovine lymphocytes were added to existing cultures of granulosa cells just prior to fixation and attached to the plate by centrifugation of the culture dish at 400 3 g for 5 min. Effect of Blocking Fas/FasL Binding on Serum Withdrawal-Induced Apoptosis

The requirement for interaction between Fas and FasL expressed on granulosa cells for serum withdrawal-induced apoptosis was studied using reagents that block Fas-FasL binding. Recombinant Fas:Fc protein, which consists of the extracellular portion of human Fas and the Fc portion of human IgG, binds to soluble or membrane-associated FasL and prevents interaction with Fas expressed on target cells, thus blocking FasL-induced killing [22–24]. A preliminary experiment was conducted to test whether an anti-rat FasL

antibody directed against the extracellular domain (Santa Cruz Biotechnology antibody C-178) was an effective antagonist of FasL binding to Fas. Granulosa cells were cultured as described above and culture media changed on Day 2 to DMEM-F12 supplemented with 100 ng/ml insulin, 5 mg/ml transferrin, 20 nM selenium, and 0.1% BSA. The change to defined media lacking serum was necessary because previous studies in our laboratory showed that FasLinduced killing was inhibited in the presence of FBS [18]. On Day 3 of culture, cells were treated with 0 or 100 ng/ ml recombinant soluble human FasL (maximal effective dose) in the presence or absence of 5 mg/ml anti-FasL antibody or 5 mg/ml rabbit IgG. Cells were counted 24 h later using trypan blue-exclusion as described above. Subsequently, the effect of anti-FasL antibody or Fas:Fc on serum withdrawal-induced apoptosis was tested. Cells were cultured as described above in DMEM-F12 containing FBS for 3 days. Media was then changed to serum-free DMEMF12 without treatment, with 20 mg/ml Fas:Fc, 5 mg/ml antiFasL antibody, or 5 mg/ml rabbit IgG. Cells were counted 24 h later using trypan blue-exclusion as described above. The experiment was repeated on three separate granulosa cell preparations. Statistical Analysis

Data were analyzed by a randomized complete block ANOVA. Duncan’s new multiple range test was used for comparison of means when overall significance was observed. Messenger RNA concentration was log-transformed prior to analysis. RESULTS

Serum Withdrawal-Induced Killing of Bovine Granulosa Cells

Bovine granulosa cells cultured in DMEM-F12 containing serum for 3 days were deprived of serum, and counted at 0, 3, 6, 12, 24, 48, and 72 h after serum removal. Significant killing was observed 6 h after serum withdrawal (20.7%, P , 0.05 vs. cell counts at 0 h) and the percentage killed continued to increase until 24 h (42.6%; Fig. 1). There was no further increase in the percentage of cells killed by serum-withdrawal at 48 h (51.5%) and 72 h (51.3%). Cells in control wells (time 5 0 h, no serum withdrawal) contained 6.9 6 0.4 3 104 cells/well, 38% greater than the number of cells plated. Serum Withdrawal-Induced Killing Occurs by Apoptosis

During apoptosis phosphatidylserine is translocated from the cytoplasmic to the outer surface of the cell membrane. This can be detected by the binding of fluorescent-labeled annexin V to the external membrane of intact cells that exclude the nuclear dye, PI [16, 19]. Dead cells, both apoptotic and necrotic, and cells in late stages of apoptosis with compromised membranes, are detected by positive staining of both annexin V and PI. Cultures exposed to serum withdrawal for 12 h contained numerous cells which displayed bright staining with annexin V but not with PI, indicating that these cells were in the early stage of apoptosis (Fig. 2). Fewer cells stained positively with annexin V and PI, which was indicative of dead cells. Cells cultured in serum displayed only sporadic staining with annexin V, and staining with PI was absent, which was indicative of healthy cultures with few dead or dying cells.

FAS-MEDIATED APOPTOSIS OF GRANULOSA CELLS

521

FIG. 2. Externalization of phosphatidylserine in apoptotic granulosa cells following serum withdrawal. Granulosa cells were cultured in DMEM-F12 containing FBS for 3 days and then cultured in media without FBS for 12 h (bottom panels) or cultured in media containing FBS for 12 h (top panels). Cells were stained for phosphatidylserine with Alexa 488-conjugated annexin V and with PI as a vital stain and viewed under epifluorescent illumination and by phase contrast microscopy. Arrows indicate examples of cells in early stages of apoptosis that stained positively for annexin V but did not take up PI. Arrowheads indicate examples of dead cells that stained positively for annexin V and PI. Magnification 3250.

Fas and FasL Expression in Bovine Granulosa Cells

Fas and FasL mRNA levels were measured using previously reported [16] and newly validated competitive RTPCR assays, respectively. An example of an RT-PCR assay of FasL mRNA expression by granulosa cells cultured in serum-free media is shown in Figure 3. Fas mRNA was detectable in granulosa cells cultured for 3 days (time 5 0 h) in media containing serum and did not change during the 48 h of serum deprivation (Fig. 4a). FasL mRNA was also present in granulosa cells cultured in serum-containing media (time 5 0 h). By 3 h after serum withdrawal, FasL mRNA expression increased 4.7-fold, remained elevated up to 24 h, and returned to basal levels by 48 h (P , 0.05; Fig. 4b). Immunohistochemical Staining of Fas and FasL

The previously reported ability of anti-human Fas antibody (CH-11) to detect bovine Fas was confirmed by positive staining of a bovine kidney cell line, MDBK, which expresses Fas (Fig. 5f) [25]. The intensity of staining increased after incubation with 200 U/ml IFN (Fig. 5g), a response reported in other cell types (reviewed in [13]). Granulosa cells stained positively for Fas at 0 h (Fig. 5a) and intensity of staining was increased at 3, 6, and 12 h after serum withdrawal (Fig. 5, b–d). Staining was not done at time points after 12 h because the deteriorating health of

the cultures prevented adherence of cells during the staining procedure. An anti-rat FasL antibody, which is reported to cross-react with human and mouse FasL, was validated for use in bovine tissue. There was intense staining of Sertoli cells within bovine seminiferous tubules (Fig. 6f), cells previously demonstrated to express high levels of FasL in the rat [26]. Staining of granulosa cells for FasL was detectable at 0 h (Fig. 6a) and was increased in intensity at 3, 6, and 12 h after serum withdrawal (Fig. 6, b–d). Control mouse IgM and rabbit IgG produced negligible staining (Fig. 5e and Fig. 6e, respectively). CD45 Staining of Lymphocytes and Macrophages in Granulosa Cell Cultures Because lymphocytes and macrophages may express FasL, granulosa cell cultures were tested for contaminating leukocytes by staining for CD45, an antigen present on the surface of lymphoid cells. Less than 1% (0.7% 6 0.6%, n 5 3) of cells were stained by anti-CD45 antibody (Fig. 7, a and b). As a positive control bovine lymphocytes were added to granulosa cell cultures, resulting in distinct positive staining of small spherical cells by the CD45 antibody (Fig. 7, c and d). Serum Withdrawal-Induced Apoptosis Occurs by Fas/FasL Interactions To test whether serum withdrawal-induced apoptosis of bovine granulosa cells is mediated through the Fas path-

522

HU ET AL.

FIG. 3. A) Phosphorimage of FasL RT-PCR products from granulosa cells at 0 and 6 h after serum withdrawal separated on a 2% agarose gel, showing competition of wild-type and standard FasL. B) Log plot of sample to standard intensity ratio vs. standard dose for the RT-PCR products shown in the top panel. The amount of sample FasL mRNA is the point at which the log intensity ratio 5 0 (i.e., sample band intensity 5 standard band intensity).

way, the effect of blocking Fas/FasL interaction using Fas: Fc and anti-FasL antibody was examined. In preliminary experiments the anti-FasL antibody inhibited killing by soluble human FasL 72% 6 11% (P , 0.05 vs. killing by soluble FasL 1 rabbit IgG, n 5 3), indicating that the antibody effectively blocked Fas/FasL interactions. Subsequent experiments tested the effect of blocking reagents on serum withdrawal-induced apoptosis of granulosa cells. Cell death 24 h after serum withdrawal was reduced 34% 6 12% and 55% 6 1% by Fas:Fc and anti-FasL antibody, respectively (P , 0.05 vs. no blocking protein; Fig. 8). Addition of 5 mg/ml rabbit IgG had no effect. DISCUSSION

Bovine granulosa cells subjected to serum withdrawal were used as a model system to study the potential role of the Fas pathway in mediating apoptosis within the granulosa cell layer of ovarian follicles. Cell death occurred by apoptosis, based on positive staining of phosphatidylserine on the outer cell membrane by annexin V [19]. Annexin V binding is useful for detecting early stages of apoptosis when phosphatidylserine is exposed on the outer cell membrane but the membrane remains intact and PI is excluded. At later stages of apoptosis, cells stain positively for both annexin V and PI and cannot be distinguished from necrotic cells. Furthermore, dead cells become detached from the plate and are not available for staining with annexin V. Therefore, annexin V binding was used to qualitatively assess the presence of apoptotic cells. The percentage of cells killed by serum withdrawal was calculated by comparing

FIG. 4. Expression of Fas and FasL mRNA in bovine granulosa cells following removal of serum from culture media. Cells were cultured in DMEM-F12 containing FBS for 3 days and then cultured in media without FBS and harvested at 0, 3, 6, 12, 24, and 48 h. The amount of Fas (a) and FasL (b) mRNA were quantified by competitive RT-PCR. Bars represent the mean 6 SEM of three experiments using separate granulosa cell preparations. Bars with different letters are significantly different (P , 0.05).

the number of live cells attached to dishes in serum-free cultures relative to the number of live cells in cultures containing serum at time 5 0 h. The results show that bovine granulosa cells express both Fas and FasL. Discrepancies exist in results of studies to localize the sites of FasL expression in the ovary [9, 10, 12, 27]. Despite this, a number of studies have shown that FasL expression occurs in the granulosa cell layer of the follicle and, because of its increased abundance in atretic follicles, is most likely expressed by granulosa cells rather than by nonovarian cell types such as infiltrating lymphocytes. However, T lymphocytes, macrophages, neutrophils, and mast cells infiltrate the theca and granulosa cell layers of the follicle at distinct times during follicle development [24, 28, 29]. These leukocytes may express FasL [30–32]. The fact that less than 1% of the cultured bovine granulosa cells stained positively for CD45, whereas essentially all of

FAS-MEDIATED APOPTOSIS OF GRANULOSA CELLS

523

FIG. 5. Immunohistochemistry of Fas expression by granulosa cells. Granulosa cells were cultured in DMEM-F12 containing FBS for 3 days and then cultured in media without FBS. Cells were processed for immunohistochemical staining for Fas at 0, 3, 6, and 12 h after serum withdrawal (a–d, respectively). Granulosa cells stained at 12 h after serum withdrawal with mouse IgM as control are shown in e. This procedure was repeated with three separate granulosa cell preparations. A positive control for staining by the anti-Fas antibody was MDBK cells treated with IFN (g) or without IFN (f). Magnification 3270.

the cells stained positively for FasL, provides strong evidence that granulosa cells express FasL. Expression of FasL mRNA and immunohistochemical staining for FasL protein increased after serum withdrawal. In contrast, Fas mRNA levels did not change after serum withdrawal but staining for Fas protein increased. These findings are supported by a previous study using pig granulosa cells in which Fas and FasL protein increased after serum withdrawal at the single 48-h time point examined

[5]. A possible explanation for the finding that Fas protein expression increased in the absence of an increase in Fas mRNA is that the distribution of Fas in the cell may change after serum withdrawal. Such a mechanism was demonstrated in human vascular smooth muscle cells: Following activation of p53, Fas protein sequestered in the Golgi complex was rapidly translocated to the cell membrane, where it bound FasL and initiated apoptosis [33]. Expression of p53 increases during apoptosis in granulosa cells [5, 34]

FIG. 6. Immunohistochemistry of FasL expression by granulosa cells. Granulosa cells were cultured in DMEM-F12 containing FBS for 3 days and then cultured in media without FBS. Cells were processed for immunohistochemical staining of FasL at 0, 3, 6, and 12 h after serum withdrawal (a–d, respectively). Granulosa cells stained at 12 h after serum withdrawal with rabbit IgG as control are shown in e. This procedure was repeated with three separate granulosa cell preparations. Magnification 3270. A bovine testis tissue section was processed as a positive control for staining by the anti-rat FasL antibody (f). Magnification 3135.

524

HU ET AL.

FIG. 7. Detection of leukocytes in granulosa cell cultures by CD45 staining. Granulosa cultures (a and b) or granulosa cell cultures with exogenous bovine lymphocytes added (c and d) were stained with anti-bovine CD45 antibody. Cells were observed under epifluorescent illumination (a and c) or by phase contrast (b and d). This procedure was repeated with three separate granulosa cell preparations. Magnification 3160.

and is associated with increased responsiveness to Fas-mediated apoptosis [35]. Thus, redistribution of Fas protein to the cell membrane may have occurred in response to serum withdrawal. The rapid increase in Fas protein expression, observed 3 h after serum withdrawal, is consistent with this idea. Further experiments are necessary to determine whether redistribution of Fas protein occurs in granulosa cells. The fact that staining for Fas and FasL protein increased in bovine granulosa cells shortly after serum withdrawal suggests that apoptosis might be mediated by Fas/FasL interactions. Furthermore, the effectiveness of reagents that block Fas/FasL binding (Fas:Fc and anti-FasL antibody) to inhibit serum withdrawal-induced apoptosis suggests that Fas/FasL interactions were at least partially responsible. The failure of anti-FasL or Fas:Fc to completely block killing suggests that other triggers of apoptosis may also be operative. The dose of anti-FasL used was determined experimentally to be maximal, while Fas:Fc was used at the highest reasonable concentration based on the manufacturer’s formulation. However, the effectiveness of these two reagents may be hampered by two factors. First, FasL expressed on the surface of cells in confluence with other cells may not be accessible to the blocking reagents. Second, affinity of the reagents for bovine FasL may limit their effectiveness. Thus, the relative contribution of Fas/FasL to spontaneous apoptosis cannot be fully ascertained. Fas/ FasL interactions were required for c-myc-induced apoptosis of fibroblasts in the absence of serum [22], stress-induced apoptosis of Jurkat T cells [23], and for apoptosis of neuronal cells in response to withdrawal of survival factors [36]. In a fibroblast cell line, activation of the protein kinase Akt-1 by IGF-1, which is associated with cell survival, prevented the transcription factor, forkhead, from localizing to the nucleus and stimulating transcription of FasL [37]. Expression of Fas and responsiveness of cells to FasL-induced killing vary in healthy and atretic follicles during the bovine estrous cycle in a manner consistent with a role of the Fas pathway in follicular atresia [17]. In the rat, expression

FIG. 8. Effect of blocking FasL binding to Fas on serum withdrawalinduced apoptosis of bovine granulosa cells. Granulosa cells were cultured in DMEM-F12 containing FBS for 3 days and then cultured for 24 h in media without FBS and containing 20 mg/ml Fas:Fc, 5 mg/ml antiFasL antibody, or 5 mg/ml rabbit IgG. Cells were counted and percent killed calculated based on cell counts of 0-h cultures in DMEM-F12 1 FBS. Bars represent the mean 6 SEM of three experiments using separate granulosa cell preparations. Bars with different letters are significantly different (P , 0.05).

of Fas [9, 10, 35] and FasL protein [10, 35] were elevated in atretic relative to healthy follicles. Fas expression was elevated in atretic antral follicles in the human ovary [15]. The results of the current study provide evidence that there is a functional Fas pathway within the granulosa cell layer of the follicle and that this pathway is activated by apoptotic stimuli such as removal of survival factors. Equally important to changes in Fas and FasL expression for inducing apoptosis may be the inactivation of pathways that, in the presence of growth factors, prevent Fas-mediated killing. In our previous studies, induction of apoptosis in bovine granulosa cells by soluble human FasL was blocked by the presence of serum in the culture media or by addition of growth factors such as IGF-1 to defined media [16, 18]. In addition, responsiveness to Fas-mediated killing was not directly correlated with the level of Fas expression. Because FasL was added exogenously, it was not a limiting factor in responsiveness. The presence of labile protein inhibitors of the Fas pathway in mouse granulosa cells was demonstrated by experiments in which cycloheximide promoted Fas-mediated killing in serum-containing media [13]. Inhibitors of the Fas pathway have been identified, including FAP-1 [38] and FLIP [39], and expression of a general class of inhibitor of apoptosis proteins (IAPs) [40] has been demonstrated in the rat ovary [41]. Therefore, the loss of survival factors within ovarian follicles that inhibit the Fas pathway, together with increased expression of Fas and FasL, may trigger follicular atresia. A substantial number of cell types that express both Fas and FasL have been identified [42]. In T lymphocytes, for example, paracrine interactions between Fas- and FasL-expressing cells induce apoptosis that is critical for regulation of the immune response [30]. Autocrine interaction between Fas and FasL has also been demonstrated in lymphocytes undergoing activation-induced cell death [30]. The fact that both Fas and FasL are expressed in the granulosa cell layer of the follicle and increase during apoptosis/ atresia suggests that autocrine interactions, paracrine interactions, or both between Fas and FasL may be involved. FasL is known to be active both as a transmembrane protein

FAS-MEDIATED APOPTOSIS OF GRANULOSA CELLS

requiring cell-cell contact to activate Fas and in a soluble form produced when a metalloproteinase cleaves the extracellular portion of FasL from the cell membrane [30]. Information is not available to indicate whether the membrane-associated or soluble form of FasL is active under most examples of Fas-mediated apoptosis. Approximately 50% of the cells died following serum withdrawal, with apoptosis occurring during the first 24 h and little or no further death from 24 to 72 h. Immunohistochemistry suggested that nearly all cells were expressing Fas and FasL despite the fact that 50% of the cells were resistant to apoptosis. In hepatocytes subjected to serum withdrawal a similar pattern of cell death was observed, with 42% death within 72 h and most death occurring by 24 h [8]. A possibility requiring testing is that granulosa cells may be susceptible to apoptosis only during particular stages of the cell cycle. An interaction between the cell cycle and susceptibility to apoptosis has been demonstrated in other cell types. In a number of cell types, withdrawal into the G0 or resting stage of the cell cycle is associated with resistance to apoptosis [43]. In summary, granulosa cells may express both Fas and FasL and expression of both proteins is increased during serum withdrawal-induced apoptosis. Serum withdrawal-induced apoptosis is mediated at least partially by Fas/FasL interactions. These findings support a potential role for an endogenous Fas pathway in granulosa cells, activated in response to removal of survival factors, to induce ovarian follicular atresia.

14. 15.

16. 17.

18. 19.

20. 21.

22. 23.

REFERENCES 1. Hsu SY, Hsueh AJW. Hormonal regulation of apoptosis. An ovarian perspective. Trends Endocrinol Metab 1997; 8:207–213. 2. Richards JS. Hormonal control of gene expression in the ovary. Endocrine Rev 1994; 15:725–751. 3. Chun S-Y, Hsueh AJW. Paracrine mechanisms of ovarian follicle apoptosis. J Reprod Immunol 1998; 39:63–75. 4. Tilly JL, Billig H, Kowalski KI, Hsueh JW. Epidermal growth factor and basic fibroblast growth factor suppress the spontaneous onset of apoptosis in cultured rat ovarian granulosa cells and follicles by a tyrosine kinase-dependent mechanism. Mol Endocrinol 1992; 6:1942– 1950. 5. Peng X, Maruo T, Matsuo H, Takekida S, Deguchi J. 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. Endocrinol J 1998; 45: 247–253. 6. Guthrie HD, Garrett WM, Cooper BS. Follicle-stimulating hormone and insulin-like growth factor-I attenuate apoptosis in cultured porcine granulosa cells. Biol Reprod 1998; 58:390–396. 7. Cotter TG, Al-Rubeai M. Cell death (apoptosis) in cell culture systems. Trends Biotechnol 1995; 13:150–155. 8. Evans-Storms RB, Cidlowski JA. Delineation of an antiapoptotic action of glucocorticoids in hepatoma cells: the role of Nuclear FactorkB. Endocrinology 2000; 141:1854–1862. 9. Hakuno N, Koji T, Yano T, Kobayashi N, Tsutsumi O, Taketani Y, Nakane PK. Fas/APO-1/CD95 system as a mediator of granulosa cell apoptosis in ovarian follicle atresia. Endocrinology 1996; 137:1938– 1948. 10. Kim J-M, Boone DL, Auyeung A, Tsang BK. Granulosa cell apoptosis induced at the penultimate stage of follicular development is associated with increased levels of Fas and Fas ligand in the rat ovary. Biol Reprod 1998; 58:1170–1176. 11. Sakamaki K, Yoshida H, Nishimura Y, Nishikawa S-I, Manabe N, Yonehara Y. Involvement of Fas antigen in ovarian follicular atresia and luteolysis. Mol Reprod Dev 1997; 47:11–18. 12. Guo MW, Xu JP, Mori E, Sato E, Saito S, Mori T. Expression of Fas ligand in murine ovary. Am J Reprod Immunol 1997; 37:391–398. 13. Quirk SM, Porter DA, Huber SC, Cowan RG. Potentiation of Fasmediated killing of murine granulosa cells by interferon-g, tumor ne-

24.

25. 26. 27. 28.

29. 30. 31.

32. 33. 34.

35.

525

crosis factor-a and cycloheximide. Endocrinology 1998; 139:4860– 4869. Quirk SM, Cowan RG, Joshi SG, Henrikson KP. Fas antigen-mediated apoptosis in human granulosa/luteal cells. Biol Reprod 1995; 52:279– 287. Kondo H, Maruo T, Peng X, Mochizuki M. Immunological evidence for the expression of the Fas antigen in the infant and adult human ovary during follicular regression and atresia. J Clin Endocrinol Metab 1996; 81:2702–2710. Vickers SL, Cowan RG, Harman RM, Porter DA, Quirk SM. Expression and activity of the Fas antigen in bovine ovarian follicle cells. Biol Reprod 2000; 62:54–61. Porter DA, Vickers SL, Cowan RG, Huber SC, Quirk SM. Expression and function of Fas antigen vary in bovine granulosa and theca cells during ovarian follicular development and atresia. Biol Reprod 2000; 62:62–66. Quirk SM, Harman RM, Cowan RG. Regulation of Fas antigen (Fas, CD95)-mediated apoptosis of bovine granulosa cells by serum and growth factors. Biol Reprod 2000; 63:1278–1284. Martin SJ, Reutelingsperger CPM, McGahon AJ, Rader JA, van Schie RCAA, LaFace DM, Green DR. Early redistribution of plasma membrane phosphatidylserine is a general feature of apoptosis regardless of the initiating stimulus: inhibition by overexpression of Bcl-2 and Abl. J Exp Med 1995; 182:1545–1556. Takahashi T, Tanaka M, Inazawa J, Abe T, Suda T, Nagata S. Human Fas ligand: gene structure, chromosomal location and species specificity. Int Immunol 1994; 6:1567–1574. Hirano A, Brown WC, Trigona W, Tuo W, Estes DM. Kinetics of expression and subset distribution of the TNF superfamily members CD40 ligand and Fas ligand on T lymphocytes in cattle. Vet Immunol Immunopath 1998; 61:251–263. Hueber AO, Zornig M, Lyon D, Suda T, Nagata S, Evan GI. Requirement for the CD95 receptor-ligand pathway in c-Myc-induced apoptosis. Science 1997; 278:1305–1309. Faris M, Kokot N, Latinis K, Kasibhatla S, Green DR, Koretzky GA, Nel A. The c-jun n-terminal kinase cascade plays a role in stressinduced apoptosis in Jurkat cells by upregulating Fas ligand expression. J Immunol 1998; 160:134–144. Liles WC, Kiener PA, Ledbetter JA, Aruffo A, Klebanoff SJ. Differential expression of Fas (CD95) and Fas ligand on normal human phagocytes: implications for the regulation of apoptosis in neutrophils. J Exp Med 1996; 184:429–440. Yoo J, Stone RT, Beattie CW. Cloning and characterization of the bovine fas. DNA Cell Biol 1996; 15:227–234. Lee J, Richburg JH, Younkin SC, Boekelheide K. The fas system is a key regulator of germ cell apoptosis in the testis. Endocrinology 1997; 138:2081–2088. Xu JP, Li X, Mori E, Sato E, Saito S, Guo MW, Mori T. Expression of Fas-Fas ligand system associated with atresia in murine ovary. Zygote 1997; 5:321–327. Best CL, Pudney J, Welch WR, Burger N, Hill JA. Localization and characterization of white blood cell populations within the human ovary throughout the menstrual cycle and menopause. Hum Reprod 1996; 11:790–797. Brannstrom M, Mayrhoffer G, Robertson SA. Localization of leukocyte subsets in the rat ovary during the periovulatory period. Biol Reprod 1993; 48:277–286. Matiba B, Mariani SM, Krammer PH. The CD95 system and the death of a lymphocyte. Immunology 1997; 9:59–68. Kiener PA, Davis PM, Starling GC, Mehlin C, Klebanoff SJ, Ledbetter JA, Liles WC. Differential induction of apoptosis by Fas-Fas ligand interactions in human monocytes and macrophages. J Exp Med 1997; 185:1511–1516. Wagelie-Steffen AL, Hartmann K, Vliagoftis H, Metcalfe DD. Fas ligand (FasL, CD95L, APO-1L) expression in murine mast cells. Immunology 1998; 94:569–574. Bennett M, MacDonald K, Chan S-W, Luzio JP, Simari R, Weissberg P. Cell surface trafficking of Fas: a rapid mechanism of p53-mediated apoptosis. Science 1998; 282:290–293. Tilly KI, Banerjee S, Banerjee PP, Tilly JL. Expression of the p53 and Wilms’ tumor suppressor genes in the rat ovary: gonadotropin repression in vivo and immunohistochemical localization of nuclear p53 protein to apoptotic granulosa cells of atretic follicles. Endocrinology 1995; 136:1394–1402. Kim J-M, Yoon Y-D, Tsang BK. Involvement of the Fas/Fas ligand system in p53-mediated granulosa cell apoptosis during follicular development and atresia. Endocrinology 1999; 140:2307–2317.

526

HU ET AL.

36. Le-Niculescu H, Bonfoco E, Kasuya Y, Claret FX, Green DR, Karin M. Withdrawal of survival factors results in activation of the JNK pathway in neuronal cells leading to Fas ligand induction and cell death. Mol Cell Biol 1999; 19:751–763. 37. Brunet A, Bonni A, Zigmond MJ, Lin MZ, Juo P, Hu LS, Anderson MJ, Arden KC, Bienis J, Greenberg ME. Akt promotes cell survival by phosphorylating and inhibiting a forkhead transcription factor. Cell 1999; 96:857–868. 38. Sato T, Irie S, Kitada S, Reed JC. FAP-1: a protein tyrosine phosphatase that associates with Fas. Science 1995; 268:411–415. 39. Irmler M, Thome M, Hahne M, Schneider P, Hofmann K, Steiner V, Bodmer J-L, Schroter M, Burns K, Mattmann C, Rimoldi D, French LE, Tschopp J. Inhibition of death receptor signals by cellular FLIP. Nature 1997; 388:190–195.

40. Liston P, Roy N, Tamai K, Lefebvre C, Baird S, Cherton-Horvat G, Farahani R, McLean M, Ikeda J, MacKenzie A, Korneluk RG. Suppression of apoptosis in mammalian cells by NAIP and a related family of IAP genes. Nature 1996; 379:349–353. 41. Li J, Kim J-M, Liston P, Li M, Miyazaki T, Makenzie AE, Korneluk RG, Tsang BK. Expression of inhibitor of apoptosis proteins (IAPs) in rat granulosa cells during ovarian follicular development and atresia. Endocrinology 1998; 139:1321–1328. 42. French LE, Hahne M, Viard I, Radlgruber G, Zanone R, Becker K, Muller C, Tschopp J. Fas and Fas ligand in embryos and adult mice: ligand expression in several immune-privileged tissues and coexpression in adult tissues characterized by apoptotic cell turnover. J Cell Biol 1996; 133:335–343. 43. Meikrantz W, Schlegel R. Apoptosis and the cell cycle. J Cell Biochem 1995; 58:160–174.