Estradiol-Mediated Suppression of Apoptosis in the Rabbit Corpus Luteum Is Associated with a Shift in Expression of bcl-2 Family Members Favoring Cellular Survival¹

^a Department of On/Gyn, Johns Hopkins University School of Medicine, Baltimore, Maryland 21201 ^b Division of Reproductive Biology, Department of Population Dynamics, Johns Hopkins University, Baltimore, Maryland 21201 ^c Vincent Center for Reproductive Biology, Department of Ob/Gyn, Massachusetts General Hospital/ Harvard Medical School, Boston, Massachusetts 02114 ^d Department of Anatomy and Human Biology, The University of Western Australia, Nedlands, Perth, Western Australia, Australia

	ABSTRACT

TOP ABSTRACT INTRODUCTION MATERIALS AND METHODS RESULTS DISCUSSION REFERENCES

 
In the rabbit, estradiol is the primary luteotropic hormone. Estradiol withdrawal results in a rapid decline in serum progesterone and eventually in corpus luteum (CL) regression. The objective of this study was to determine whether estradiol modulates luteal cell apoptosis. In the first experiment, rabbits were randomly assigned to one of five experimental groups. An empty capsule (control) or estradiol-filled Silastic capsule was inserted s.c. on Day 0 of pseudopregnancy (day of hCG administration). On Day 11 of pseudopregnancy, some of the group I (control) and group II (estradiol capsule) rabbits were subjected to laparotomy, and one ovary from each rabbit was perfused in vitro to determine progesterone secretion rates. The CL from the contralateral ovary were dissected, snap-frozen, and stored at -70°C until analyzed for internucleosomal DNA cleavage (apoptosis). Estradiol-containing capsules were removed from some of the remaining rabbits on Days 8, 9, and 10 to initiate estradiol deprivation. Rabbits were then subjected to laparotomy 24, 48, or 72 h after capsule removal (groups III, IV, and V, respectively), and ovaries or CL were processed as described above. Deprivation of estradiol for 24 (group III), 48 (group IV), or 72 (group V) h in vivo reduced in vitro progesterone secretion rates by more than 90% as compared to that in ovaries collected from estradiol capsule-intact animals. After in vivo endogenous estradiol suppression, withdrawal of exogenous estradiol resulted in luteal cell apoptosis, which increased in a time-dependent manner. Northern blot analysis revealed an increase in bax mRNA levels and a decrease in bcl-x mRNA levels coincident with luteal cell apoptosis induced by estradiol withdrawal. These data demonstrate that changes in progesterone production caused by estradiol exposure and deprivation are in part related to luteal cell apoptosis, and alterations in the expression of bcl-2 gene family members may be one of the mechanisms by which estradiol exerts its luteotropic effect in the rabbit CL.

	INTRODUCTION

TOP ABSTRACT INTRODUCTION MATERIALS AND METHODS RESULTS DISCUSSION REFERENCES

 
Through its secretion of progesterone, the corpus luteum (CL) has a critical function in mammalian reproduction. Production of progesterone by the CL after ovulation is necessary for implantation of the early embryo and maintenance of early gestation. Inadequate CL function has been associated with infertility and recurrent pregnancy loss [1]. Regression of the CL in the absence of pregnancy is necessary to allow for recruitment of new follicles and in order for ovulation to occur in a subsequent cycle [1].

Estradiol is the primary luteotropic hormone in the rabbit. Withdrawal of estradiol results in a rapid decrease in serum progesterone in both intact and hypophysectomized pseudopregnant rabbits [2, 3]. Readministration of estradiol leads to recovery of serum progesterone concentrations to control levels [4]. The mechanism for the rapid decline in progesterone following withdrawal of estradiol support is not well understood, but there is increasing evidence from studies in rats and rabbits that withdrawal of estradiol increases accumulation of toxic metabolites of oxygen, elevates cytokine levels, and decreases blood flow to the CL [5–7]. Although all of these factors are likely to play a role in CL regression following estradiol withdrawal, the molecular mechanisms by which these factors trigger the onset of luteolysis remain to be elucidated.

Apoptosis is a widespread physiologic process that has steadily gained recognition due to its now clearly established roles in development, tissue homeostasis, and disease. Since the initial report describing the presence of internucleosomal DNA activity in rat granulosa and luteal cells [8], a role for apoptosis in follicular atresia [9–11] and luteal regression [12–18] has been demonstrated. Of direct relevance to the present studies, recent molecular investigations of functional and regressing CL collected from pseudopregnant and pregnant rabbit ovaries have provided evidence that internucleosomal (apoptotic) DNA cleavage is present in luteal cells of regressing but not functional CL [15, 19].

Apoptosis is recognized to be a gene-directed process [20]. Since its discovery over 10 years ago as an oncogenic protein involved in human tumors, Bcl-2 has defied attempts to determine the biochemical basis for its potent anti-apoptotic action. Identification of a number of Bcl-2 homologues, some of which bind to Bcl-2, suggests that Bcl-2 functions, at least in part, through protein-protein interactions. The first of these homologues, bax, is a death susceptibility gene. The Bax protein was originally identified via its ability to noncovalently interact with Bcl-2 in cells. This interaction is thought to blunt Bcl-2 bioactivity and thus may serve as one of its mechanisms of action as a death-inducing factor. In addition to bcl-2 and bax, other members of the bcl-2 gene family have been cloned, and several of these factors are expressed in the ovary [11]. One of these genes, bcl-x, is unique among the other members of this family as it is alternatively processed to yield both positive ("short" isoform, death inducer; bcl-x_Short) and negative ("long" isoform, death suppressor; bcl-x_Long) regulators of the cell death pathway. The members of Bcl-2 family of proteins are thought to be central regulators of cell death decision-making processes in most, if not all, cases, and thus these gene products serve at a critical intracellular checkpoint in apoptosis regulation [20].

In the ovary, increased expression of bax, a pro-apoptotic bcl-2 gene family member, has been correlated with the occurrence of apoptosis in both granulosa cells during atresia [21, 22] and luteal cells during CL regression [14]. In agreement with the hypothesis that Bax acts in opposition to Bcl-2 or Bcl-x_Long to induce apoptosis [20], expression of bcl-x has also been characterized in the bovine CL [14] and in human granulosa-luteal cells [22]. Based on these observations and others regarding an important function for Bcl-2 family members in modulating apoptosis in various ovarian cell lineages [23–25], it has been proposed that cell death in the ovary is regulated by a conserved pathway composed of these central apoptosis-related genes [26].

The objectives of this study were to 1) determine whether estradiol modulates luteal cell apoptosis in the rabbit ovary, 2) correlate serum estradiol and progesterone levels with luteal cell DNA integrity after various in vivo manipulations, and 3) determine whether altered expression of bax and/or bcl-x correlates with estradiol deprivation-induced luteal regression in the rabbit ovary. In vivo and in vitro models were utilized to examine the effect of estradiol withdrawal on CL function, DNA integrity, and expression of bax and bcl-x.

	MATERIALS AND METHODS

TOP ABSTRACT INTRODUCTION MATERIALS AND METHODS RESULTS DISCUSSION REFERENCES

 
Animals

Sexually mature New Zealand White rabbits weighing an average of 3.5 kg were housed individually under controlled temperature and light and given free access to food (Purina rabbit chow; Ralston Purina Co., St. Louis, MO) and water. Rabbits received 100 IU of hCG (Organon, West Orange, NJ) via the marginal ear vein to induce pseudopregnancy. The day of injection was defined as Day 0 of pseudopregnancy. On the appropriate day of pseudopregnancy, rabbits were anesthetized with i.v. sodium pentobarbital (32 mg/kg), anticoagulated with heparin sulfate (120 U/kg), and then subjected to laparotomy. In the rabbit, pseudopregnancy lasts for 21 days, with progesterone secretion increasing from Days 1 to 11 and declining to baseline levels by Day 18 [15, 27, 28]. All protocols were reviewed and approved by the Johns Hopkins University Animal Care and Use Committee and the University of Western Australia Animal Ethics Committee.

In Vivo Study

Pseudopregnancy was induced as described above. On Day 0 of pseudopregnancy, a single polydimethyl-siloxane capsule (3 cm long, 3.35-mm internal diameter; Dow-Corning, Midland, MI) was implanted s.c. at the base of the neck under local anesthesia with 1% (w:v) lidocaine. Capsules either were empty or contained 2 cm (filled length) of estradiol (Sigma Chemical Co., St Louis, MO). Blood was drawn via the marginal ear vein for baseline serum steroid analyses.

Forty rabbits were randomly allocated to 5 groups (8 rabbits per group) as follows: group I (control), empty capsules implanted; group II (intact), estradiol capsules remaining in place for the duration of the experiment; group III (minus 24 h), estradiol capsules removed 24 h (Day 10) before removal of ovaries for perfusion/apoptosis studies; group IV (minus 48 h), estradiol capsules removed 48 h (Day 9) before laparotomy; and group V (minus 72 h), estradiol capsules removed 72 h (Day 8) before the start of experiments. Peripheral blood from the marginal ear vein was collected for estradiol and progesterone studies on the day of capsule removal from animals in groups III, IV, and V and from all animals on the day of laparotomy. On Day 11 of pseudopregnancy, all animals underwent laparotomy with removal of the ovaries for perfusion, DNA, and RNA studies.

In Vitro Ovarian Perfusion

The cannulation procedure and perfusion apparatus have been described in detail previously [28]. On Day 11 of pseudopregnancy, rabbits underwent laparotomy. After major anastomotic vessels were ligated, the unilateral ovarian artery and vein were cannulated in situ. Ovarian venous blood was sampled for steroid levels. The ovary with its cannulated vessels was removed and placed into the perfusion chamber. The perfusion apparatus consists of a chamber containing the ovary, an oxygenator, a reservoir, and a pump that maintains perfusate flow at 1.5 ml/min, approximating the normal rate of blood flow to the rabbit ovary [29]. One ovary from six rabbits in each group was perfused at 37°C with 150 ml of medium 199 (Gibco-BRL Life Technologies, Grand Island, NY), pH 7.4, supplemented with 3% BSA (Intergen, Purchase, NY), heparin sulfate (200 U/L), streptomycin sulfate (50 mg/L), and penicillin G (75 mg/L).

One-milliliter samples were obtained from the arterial and venous cannulae every 0.5 h for the first 2 h and thereafter every hour for a total of 6 h. Samples were replaced with fresh medium to maintain perfusate volume. After collection, samples were stored at 20°C until assayed for progesterone by RIA. Ovarian progesterone secretion was calculated by dividing the mean of the differences in concentration between perfusate venous and arterial samples by the perfusion time [28].

CL from the perfused ovaries were isolated and snap-frozen at the termination of the 6-h perfusion period. One ovary from 2 rabbits in each group was immediately perfusion fixed with 4% paraformaldehyde for in situ DNA-labeling studies. CL from the contralateral ovary of all rabbits were removed at the time of laparotomy, snap-frozen, and processed for DNA integrity or Northern blot analysis for bax and bcl-x gene expression.

Effect of Estradiol on Luteal Cell Apoptosis: In Vitro Perfusion Model

Rabbits received hCG and estradiol-containing capsules as described above. Control rabbits received empty capsules. On Day 11 of pseudopregnancy, ovaries from estradiol-treated and control rabbits were removed and perfused in vitro with and without estradiol (5 ng/ml) for 6 h. At the end of 6 h, CL were dissected, snap-frozen, and analyzed for DNA integrity. The dose of estradiol was selected based on previous studies [3].

Extraction and Analysis of DNA for Internucleosomal Cleavage

Genomic DNA was prepared from CL as originally described by Gross-Bellard et al. [30] and modified by Tilly and Hsueh [31], and Dharmarajan et al. [15]. After extraction, purification, and quantitation of DNA, samples were labeled on 3' ends with ³²P-dideoxy-ATP, 3000 Ci/mmol; Amersham Australia, Sydney, Australia) using the terminal transferase (Boehringer-Mannheim, Indianapolis, IN) reaction as described previously [15, 31]. Radiolabeled DNA samples were resolved by electrophoresis through 2% agarose gels at 50 volts (6.5 volts/cm) for 3–3.5 h. The gels were dried for 2 h without heat in a slab gel drier, sealed in plastic wrap, and exposed to X-Omat (Eastman Kodak, Rochester, NY) films at -70°C for autoradiographic analysis. After autoradiography, low molecular weight DNA fractions (< 15 kilobases [kb]) were excised from the gels, mixed with 3 ml of scintillation fluid, and counted in a beta counter to provide a quantitative estimate of the degree of internucleosomal DNA cleavage among samples, as described previously [15, 31].

In Situ Localization of DNA Fragmentation

The extent of DNA breakdown in fixed ovarian tissue sections was assessed using a nonisotopic, streptavidin-biotin-based in situ terminal transferase reaction as originally described [32] and modified [15, 33]. Briefly, tissues were perfusion fixed with 4% paraformaldehyde for 12 h (w:v in 75 mM PBS), embedded in paraffin, sectioned at 5-µm thickness, and transferred to glass slides. Slides containing the tissues to be analyzed were incubated at 60°C for 30 min. The tissue sections were deparaffinized by two 5-min washes in xylene and subsequently rehydrated through a graded ethanol series. After rehydration, sections were digested with 10 µg/ml proteinase-K (Boehringer-Mannheim) for 30 min at 37°C and then labeled by incubation with 250 U/ml of terminal transferase enzyme (Boehringer-Mannheim) and 50 µM biotin-14-deoxy-ATP (Gibco-BRL) as the labeling nucleotide for 15 min at 37°C. The tissue sections were washed to remove the reaction components, blocked with 3% BSA for 30 min at 21°C, and subsequently subjected to reaction with 30 µg/ml streptavidin conjugated to alkaline phosphatase (Sigma). The double-labeled sections were washed and then subjected to a colorimetric reaction with 400 µg/ml nitroblue tetrazolium salt (Gibco-BRL) and 350 µg/ml 5-bromo-4-chloro-3-indolyphosphate-toluidinium salt (Gibco-BRL) at 21°C for 15 min (the length of the colorimetric reaction for CL analysis was based on the time at which extensive labeling of granulosa cells in adjacent atretic follicles was observed [15]). The colorimetric reaction was terminated by placing the slides in 1 mM EDTA/10 mM Tris-HCl (pH 8.0), and the sections were then dehydrated, mounted under glass coverslips with Permount (Sigma), and analyzed for in situ DNA labeling (darkly stained cells) by light microscopy. As controls, the colorimetric reaction did not occur in cells of sections processed without either the terminal transferase enzyme or the labeling nucleotide (data not shown).

Isolation of cDNA Probes for Rabbit bcl-x and bax

The reverse transcription-polymerase chain reaction (RT-PCR) was utilized to obtain rabbit partial bcl-x and bax cDNA probes, as described previously [14, 21]. Total RNA was isolated from CL of Day 11 pseudopregnant rabbits and reverse transcribed into first-strand cDNA using random hexamer primers and avian myeloblastosis virus reverse transcriptase. Oligonucleotide primers were synthesized (DNA International, Lake Oswego, OR) based on mouse bcl-x and human bax cDNA sequences [21], and included appended 5'-end restriction enzyme sequences for subcloning. The first-strand cDNA was subjected to 35 cycles of PCR amplification using one of the primer sets (1-min denaturation at 94°C, 1-min annealing at 50°C, and 2-min extension at 72°C); and the amplified products were resolved through 1.5% agarose gels, isolated, purified, and subcloned into the pCRII plasmid (Invitrogen, San Diego, CA). Sequence analysis confirmed that the rabbit bcl-x and bax partial cDNAs were more than 90% homologous to reported mouse (bases 97–671) and human (bases 85–567) sequences (data not shown).

Preparation of cRNA and cDNA Probes

Antisense RNA probes complementary to rabbit bax and bcl-x mRNA sequences were synthesized by in vitro transcription from linearized plasmid templates using RNA polymerase, [-³²P]CTP (3000 Ci/mmol; Amersham), and the Gemini II Riboprobe Core System (Promega, Madison, WI) as previously described [21]. The cDNA probe for 18S rRNA was radiolabeled with [-³²P]>dCTP (3000 Ci/mmol; Amersham) using the random priming method [34] and then purified from unincorporated radionucleotides by column chromatography (Nick Columns, Pharmacia Biotech, Uppsala, Sweden).

Extraction of RNA and Northern Blot Analysis

Total RNA was isolated from frozen CL using the single-step guanidinium thiocyanate method [35], resuspended in diethylpyrocarbonate-treated water, and quantitated by reading the absorbance at 260 nm using a spectrophotometer. Total RNA samples (10 µg/sample) were fractionated by electrophoresis through denaturing (formaldehyde) agarose gels, transferred by capillary action to pure nitrocellulose membranes, and UV-crosslinked (UV Crosslinker RPN 2500; Amersham International, Buckinghamshire, UK). Membranes were hybridized under highly stringent conditions at 65°C as described previously [21]. After cRNA probe hybridization analysis, the radioactivity on the blots was allowed to decay, and the blots were then hybridized with the radiolabeled 18S rRNA probe at 42°C under conditions previously described [36, 37]. All data were then normalized relative to 18S rRNA levels in each sample by exposing the Northern blot to a Fuji imaging plate (Bas-IIs; Fuji Photo Film Co., Tokyo, Japan) and scanning the resultant images using a Fuji Bioimager.

Progesterone and Estradiol RIA

Progesterone and estradiol concentrations were measured with commercial RIA kits (Diagnostic Products Corporation, Los Angeles, CA). The sensitivity was 8 pg/ml for the estradiol assay and 0.05 ng/ml for the progesterone assay. All samples and standards were assayed in duplicate. The inter- and intraassay coefficients of variation at the concentrations obtained in these experiments were 7.5% and 6.6% for progesterone and 4.4% and 6.8% for estradiol, respectively.

Statistical Analyses

The Generalized Linear Interactive Modeling statistical program (1985, Royal Statistical Society, London) was used for all statistical analyses. A histogram of the standardized residuals was used to test for normality of the data. All data were found to follow a normal distribution, thereby allowing for the use of parametric statistical calculations. ANOVA was used to evaluate differences in means among the groups. When differences were found versus the experimental control, these differences were further evaluated by Student-Newman-Keuls test. A p value less than 0.05 was considered statistically significant. All results, with the exception of representative autoradiograms or histological photomicrographs, are presented as the mean ± SEM of combined data from the replicate experiments.

	RESULTS

TOP ABSTRACT INTRODUCTION MATERIALS AND METHODS RESULTS DISCUSSION REFERENCES

 
Steroid Hormones: In Vivo Studies

The administration of estradiol via Silastic capsules to rabbits from Day 0 to Day 8 of pseudopregnancy caused serum levels of estradiol to rise significantly (p < 0.01) from control levels (empty capsule) at the time of capsule removal (Fig. 1A). Despite the increase in serum estradiol levels, serum progesterone levels did not change relative to levels in the control group (empty capsule; Fig. 1B). Figure 1 shows that 24 h after estradiol capsule removal (group III), peripheral serum estradiol concentrations were significantly reduced compared to those in the estradiol-intact group (group II) but did not decrease below the control level (group I) (Fig. 1A). A significant decrease in progesterone was observed 24 h after capsule removal (group III) (Fig. 1B). No significant further decrease in either estradiol or progesterone levels was observed 48 h (group IV) and 72 h (group V) after capsule removal. Figure 2A shows estradiol levels in the ovarian vein in control (group I) and estradiol-intact (group II) rabbits and in rabbits 24 h (group III), 48 h (group IV), and 72 h (group V) after capsule removal. In contrast to peripheral levels, estradiol levels in ovarian venous blood were lower (p < 0.01) in all four treated groups than in controls (Fig. 2A), reflecting suppression of ovarian estradiol production induced by exogenous estradiol administration. Progesterone production in the estradiol capsule-intact group (group II) was not different than in controls (group I) but was significantly reduced in group III (24 h), group IV (48 h), and group V (72 h) (p < 0.01 for all versus group II) (Fig. 2B).

View larger version (58K): [in this window] [in a new window]   FIG. 1. Peripheral estradiol (A) and progesterone (B) levels in Day 11 pseudopregnant rabbits implanted with empty (group I) or estradiol-containing capsules (group II), or in rabbits 24 (group III), 48 (group IV), and 72 h (group V) after estradiol capsule removal. Estradiol levels were significantly decreased in groups III, IV, and V (p < 0.05) compared to group II. Progesterone levels were significantly reduced by removal of estradiol capsules (p < 0.01, groups III, IV, and V). Values are the mean ± SEM (n = 6 per group), and those without common letter superscripts differ significantly (p < 0.05, ANOVA).

View larger version (54K): [in this window] [in a new window]   FIG. 2. Ovarian venous estradiol (A) and progesterone (B) levels. Estradiol implants significantly (p < 0.01) reduced endogenous estradiol production in groups II–V compared to control (group I). Progesterone levels were significantly (p < 0.01) reduced at 24 (group III), 48 (group IV), and 72 h (group V) after capsule removal. Values are the mean ± SEM (n = 6 per group), and those without common letter superscripts differ significantly (p < 0.05, ANOVA).

Steroid Hormones: In Vitro Studies

Figure 3 depicts ovarian progesterone secretion rates by in vitro-perfused rabbit ovaries collected from rabbits in the five experimental groups. Ovarian progesterone secretion rates were calculated by obtaining samples every 60 min and are presented as the mean for the 6-h perfusion period. The rate of progesterone secretion (6 g/h per ovary) did not change over the 6-h perfusion period within any group examined (data not shown). Despite the removal of the estradiol capsule at different times during pseudopregnancy, the CL responded to the removal in a similar fashion in each group. Removal of estradiol capsules (groups III, IV, and V) significantly (p < 0.001) reduced the progesterone secretion rates, mirroring the changes in progesterone secretion rates seen in vivo in each of the five groups.

View larger version (29K): [in this window] [in a new window]   FIG. 3. Progesterone secretion by the in vitro-perfused ovary. Progesterone secretion was significantly reduced by removal of estradiol capsules (p < 0.01; groups III, IV, and V). Values are the mean ± SEM (n = 6 per group), and those without common letter superscripts differ significantly (p < 0.05, ANOVA).

Estradiol Regulation of Apoptosis in the CL: In Vivo Studies

Figure 4A shows the analysis of DNA integrity in CL collected from rabbits in the five experimental groups. Evidence of a time-dependent effect of estradiol withdrawal on internucleosomal DNA cleavage in the CL was observed. Quantitative analysis of DNA integrity in CL, assessed by beta-counting of low molecular weight (< 15 kb) radiolabeled DNA fractions after electrophoresis, is shown in Figure 4B. A significant (p < 0.01) increase in DNA breakdown was observed at 48 h (group IV) and 72 h (group V) after estradiol capsule withdrawal versus the level in the estradiol-intact group (group II).

View larger version (53K): [in this window] [in a new window]   FIG. 4. A) Autoradiographic analysis of DNA integrity in rabbit CL collected from control (group I) and estradiol capsule-intact (group II) rabbits and from rabbits 24, 48, and 72 h after estradiol capsule removal (groups III, IV, and V, respectively). Evidence of a time-dependent effect of estradiol withdrawal on apoptotic DNA fragmentation was observed. Maximum DNA fragmentation was observed 72 h after estradiol withdrawal. B) Quantitative analysis of DNA integrity in CL as assessed by beta-counting of low molecular weight (MW; < 15 kb) radiolabeled DNA fractions after electrophoresis. Values are mean ± SEM of results of three independent experiments. Different letters indicate significant differences (p < 0.01).

In Situ DNA-Labeling Analysis

To determine which cell types were undergoing apoptosis in the CL, one ovary from the control group (group I) and one from the group deprived of estradiol for 72 h (group V) were perfusion fixed with 4% paraformaldehyde, embedded in paraffin, sectioned, transferred to glass slides, and subsequently analyzed by nonisotopic in situ DNA labeling for the localization of DNA strand breaks associated with apoptosis (Fig. 5). These experiments revealed that luteal cells present in CL of group V ovaries possessed extensive DNA cleavage as compared with the absence of labeling in cells of the control CL (group I).

View larger version (102K): [in this window] [in a new window]   FIG. 5. In situ DNA-labeling analysis of control CL (A) or CL collected 72 h after estradiol capsule withdrawal (B). Extensive DNA labeling (darkly stained cells) was detected in luteal cells of CL collected 72 h after capsule removal (group V) compared to control CL (group I). LC, Luteal cells. x400.

Effect of Estradiol on Apoptosis in the CL: In Vitro Perfusion Model

Figure 6 shows the autoradiographic and quantitative analyses of DNA integrity in samples extracted from individual CL collected from ovaries of group II rabbits on Day 11 of pseudopregnancy (endogenous estradiol is suppressed by the presence of the estradiol capsule) and perfused in vitro without and with estradiol for 6 h. After in vivo endogenous estradiol suppression, CL exhibited a significant (p < 0.01) increase in apoptosis over the 6-h perfusion period in medium alone. However, addition of estradiol significantly inhibited luteal cell apoptosis (Fig. 6).

View larger version (49K): [in this window] [in a new window]   FIG. 6. A) Autoradiographic analysis of DNA extracted from individual CL collected from ovaries of Day 11 estradiol-implanted pseudopregnant rabbits (group II; endogenous estradiol suppressed) perfused in vitro without (-E2) and with estradiol (+E2) for 6 h. B) Quantitative analysis of DNA integrity in individual CL collected from experiments as described in legend to A. The extent of low molecular weight DNA labeling was analyzed relative to those levels measured in DNA samples prepared from healthy CL snap-frozen immediately after isolation (Time 0). Values are mean ± SEM of results of three independent experiments. Different letters indicate significant differences (p < 0.01).

Northern Blot Analysis of bcl-x and bax Gene Expression After Estradiol Withdrawal

This final experiment was performed to determine whether there is an effect of estradiol on bcl-x and/or bax gene expression. Northern blot analysis of total RNA extracted from CL of ovaries from groups I–V indicated the presence of bcl-x and bax mRNA transcripts of 3.0 and 1.0 kb, respectively (Fig. 7). The significance was calculated based on the pooled values of samples I and II. Levels of bcl-x mRNA were significantly reduced 24 h after capsule removal (group III) (67 ± 8% of the levels in groups I and II; p < 0.05, n = 3) and remained decreased in groups IV and V (64 ± 6% and 76 ± 7%, respectively), whereas levels of bax mRNA were significantly increased 24 h (group III) after estradiol capsule removal to 186 ± 12% of the levels in groups I and II and remained increased in groups IV and V (p < 0.05, n = 3) (Fig. 7).

View larger version (108K): [in this window] [in a new window]   FIG. 7. Northern blot analysis of bcl-x (A), bax (B), and 18S (C) mRNA levels in the CL of ovaries from rabbits in groups I–V. The blots were prepared as described in Materials and Methods. Levels of bcl-x mRNA were significantly reduced 24 h after capsule removal (group III), whereas levels of bax mRNA were significantly increased 24 h after capsule removal (group III). See text for quantitative assessments across the replicate experiments.

	DISCUSSION

TOP ABSTRACT INTRODUCTION MATERIALS AND METHODS RESULTS DISCUSSION REFERENCES

 
Estradiol has been established as the primary luteotropic hormone in rabbits, and withdrawal of estradiol leads to luteolysis [38]. We utilized in vivo and in vitro models to study the potential molecular mechanisms through which estradiol controls the CL life span. Suppression of endogenous ovarian estradiol production was achieved in vivo through implantation of an s.c. estradiol capsule [38]. Within 24 h after removal of the estradiol capsule, serum progesterone concentrations reached a nadir, indicating that CL function had become dependent upon exogenous estradiol. We then investigated the possibility that estradiol withdrawal could induce apoptosis in the CL, since we have previously demonstrated that apoptosis is associated with luteal regression in the pseudopregnant rabbit [15].

In the present study, the extent of internucleosomal DNA fragmentation was found to be regulated by estradiol. Following in vivo endogenous estradiol suppression, withdrawal of exogenous estradiol resulted in luteal cell apoptosis, which increased in a time-dependent manner. In situ detection of DNA fragmentation in paraffin sections showed that apoptotic DNA breakdown was present in luteal cells. Deprivation of estradiol for 24 h in vivo also reduced in vitro progesterone secretion rates by more than 90% as compared to that in ovaries collected from estradiol capsule-intact animals. However, this effect occurred initially in the absence of observed luteal cell apoptosis, since complete functional regression (loss of progesterone secretion) occurred as early as 24 h after estradiol capsule withdrawal while internucleosomal DNA cleavage was not observed until sometime between 24 and 48 h after estradiol withdrawal. Similar conclusions have been obtained in both cattle and sheep using prostaglandin F₂ injection as the luteolytic stimulus. In these species, serum progesterone decreased within 8 h of prostaglandin F₂ injection whereas apoptotic DNA was not detected until 12–24 h after injection [39, 40].

Previous data on CL blood flow analyses after estradiol withdrawal have demonstrated that blood flow to the CL did not decrease until 48 h after capsule removal, whereas progesterone secretion decreased as early as 24 h after estradiol withdrawal [7]. On the basis of this observation and previous observations by other investigators, it is clear that the functional and structural life spans of the CL are separate entities that may be regulated differently by endocrine signaling events. Interestingly, addition of estradiol to ovaries perfused in vitro inhibited luteal cell apoptosis. Thus, an early reversal of the inhibition of steroidogenesis may preclude the occurrence of apoptosis and structural regression in the CL.

Although it is not completely understood how estradiol exerts its potent luteotropic effects in the rabbit, it is known that rabbit luteal cells contain estrogen receptors [41]. Furthermore, several studies have demonstrated a correlation between the number of receptors and luteal progesterone production [42, 43], and withdrawal of estrogen support results in a decreased number of estrogen receptors in the rabbit CL. Steroid receptors are transcription factors [44]. When bound to specific ligands, ovarian receptors may directly regulate steroid-dependent transcription leading to changes in expression of genes directly involved in regulating apoptosis. In the present study, we observed that estradiol deprivation led to a significant increase in bax mRNA levels and a significant reduction in bcl-x message in the CL within 24 h, suggesting that alterations in the expression of these cell death genes may be one mechanism by which estradiol exerts its luteotropic effect in the rabbit CL. Moreover, the finding that bax and bcl-x mRNA levels changed at a point prior to the occurrence of significant levels of internucleosomal DNA cleavage supports the concept of a temporal series of events activated by estradiol withdrawal in the rabbit CL. A re-setting of the bax:bcl-x rheostat favoring higher bax levels appears to precede and therefore may be at least partly responsible for the onset of apoptosis during luteolysis.

In summary, these data demonstrate that estradiol withdrawal leads to diminished CL function that may initially be reversible but with time results in irreversible luteolysis, owing to activation of apoptosis. These studies have also provided the first evidence that two genes encoding members of the Bcl-2 family of cell death regulators, bax and bcl-x, are expressed and steroid-regulated in the rabbit CL. We propose from these findings that estrogen may maintain luteal cell integrity by directly altering the bax:bcl-x to favor reduced bax, thus allowing for a suppression of apoptosis.

	ACKNOWLEDGMENTS

 
The authors would like to acknowledge the technical assistance of Mrs. Bev Smith in the assay of progesterone and estradiol.

	FOOTNOTES

 
¹ This study was supported by the Rockefeller Foundation (A.M.D.), National Institute of Health Grants R01-HD19430 (A.M.D.), R01-HD34226 (J.L.T.), and R01-AG12279 (J.L.T.), National Health and Medical Research Council of Australia (A.M.D.), Australian Research Council (A.M.D.), Raine Foundation (A.M.D.), and a Johns Hopkins University Institutional Research Grant.

² Correspondence. FAX: 61–8-9380–1051; dharma{at}anhb.uwa.edu.au

³ Current address: 2804 Fountain Grove Terrace, Olney, MD 20832.

⁴ Current address: Department of Ob/Gyn, The University of Tokyo Faculty of Medicine, 7–3-1 Hongo, Bunkyo-ku, Tokyo 113, Japan.

⁵ Current address: 2001 Marcus Avenue, Suite N213, Lake Success, NY 11042.

⁶ Current address: Department of Gyn/Ob, Ramathibodi Hospital, Mahidol University School of Medicine, Bangkok, Thailand.

Accepted: May 18, 1998.

Received: October 31, 1997.

	REFERENCES

TOP ABSTRACT INTRODUCTION MATERIALS AND METHODS RESULTS DISCUSSION REFERENCES

 

1. Niswender GD, Nett TM. Corpus luteum and its control in infraprimate species. In: Knobil E, Neill JD (eds.), Physiology of Reproduction. New York: Raven Press; 1994: 781–816.
2. Bill II CH, Keyes PL. 17ß-Estradiol maintains normal function of corpora lutea throughout pseudo-pregnancy in hypophysectomized rabbits. Biol Reprod 1983; 28:608–617.[Abstract]
3. Dharmarajan AM, Zanagnolo VL, Dasko LM, Zirkin BR, Ewing LL, Wallach EE. Estradiol regulation of the rabbit corpus luteum: in vivo and in vitro studies. Endocrinology 1991; 128:2678–2684.[Abstract/Free Full Text]
4. Bender EM, Miller JB, Possley RM, Keyes PL. Steroidogenic effect of 17ß-estradiol in the rabbit: stimulation of progesterone synthesis in prematurely regressing corpora lutea. Endocrinology 1978; 103:1937–1943.[Abstract/Free Full Text]
5. Hesla JS, Miyazaki T, Dasko LM, Wallach EE, Dharmarajan AM. Superoxide activity, lipid peroxide production and corpus luteum steroidogenesis during natural luteolysis and regression induced by oestradiol deprivation of the ovary in pseudopregnant rabbits. J Reprod Fertil 1992; 95:915–924.[Abstract/Free Full Text]
6. Bagavandos P, Wiggins RC, Kunkel SL, Remick DG, Keyes PL. Tumor necrosis factor production and accumulation of inflammatory cells in the corpus luteum of pseudopregnancy and pregnancy in rabbits. Biol Reprod 1990; 42:367–376.[Abstract]
7. Moutos D, Schryer B, Hurn P, Dharmarajan AM. Effect of exogenous oestrogen on blood flow and quantitative histology of the corpora lutea of the pseudopregnant rabbits. J Reprod Fertil 1995; 103:357–362.[Abstract/Free Full Text]
8. Zeleznik AJ, Ihrig LL, Bassett SG. Developmental expression of Ca⁺⁺/Mg⁺⁺ dependent endonuclease activity in rat granulosa and luteal cells. Endocrinology 1989; 125:2218–2220.[Abstract/Free Full Text]
9. Hughes FM Jr, Gorospe WC. Biochemical identification of apoptosis (programmed cell death) in granulosa cells: evidence for a potential mechanism underlying follicular atresia. Endocrinology 1991; 129:2415–2422.[Abstract/Free Full Text]
10. Tilly JL, Kowalski KI, Johnson AL, Hsueh AJW. Involvement of apoptosis in ovarian follicular atresia and postovulatory regression. Endocrinology 1991; 129:2799–2801.[Abstract/Free Full Text]
11. Tilly JL. Apoptosis and ovarian function. Rev Reprod 1996; 1:162–172.[Abstract]
12. Rueda BR, Wegner JA, Marion SL, Wahlen DD, Hoyer PB. Internucleosomal DNA fragmentation in ovine luteal tissue associated with luteolysis. In vivo and in vitro analysis. Biol Reprod 1995; 52:305–312.[Abstract]
13. Rueda BR, Tilly KI, Hansen TR, Hoyer PB, Tilly JL. Expression of superoxide dismutase, catalase and glutathione peroxidase in the bovine corpus luteum: evidence supporting a role for oxidative stress in luteolysis. Endocrine 1995: 3:227–232.
14. Rueda BR, Tilly K, Botros IW, Jolly PD, Hansen TR, Hoyer PB, Tilly JL. Increased bax and interleukin-1ß-converting enzyme (Ice) messenger ribonucleic acid levels coincide with apoptosis in the bovine corpus luteum during structural regression. Biol Reprod 1997; 56:186–193.[Abstract]
15. Dharmarajan AM, Goodman SB, Tilly KI, Tilly JL. Apoptosis during functional corpus luteum regression: evidence of a role for chorionic gonadotropin in promoting luteal cell survival. Endocrine 1994; 2:295–303.
16. Fraser HM, Lunn SF, Cowen GM, Illingworth PJ. Induced luteal regression in the primate: evidence for apoptosis and changes in c-myc protein. J Endocrinol 1995; 147:131–137.[Abstract/Free Full Text]
17. Shikone T, Yamoto M, Kokawa K, Yamashita K, Nishimori K, Nakano R. Apoptosis of human corpora lutea during cyclic luteal regression and early pregnancy. J Clin Endocrinol Metab 1996; 81:2376–2380.[Abstract]
18. Matsuyama S, Chang K-T, Kanuka H, Ohnishi M, Ikeda A, Nishihara M, Takahashi M. Occurrence of deoxyribonucleic acid fragmentation during prolactin induced structural luteolysis in cycling rats. Biol Reprod 1996; 54:1245–1251.[Abstract]
19. Dharmarajan AM, Goodman SB, Atiya N, Parkinson SP, Tilly JL. Role of apoptosis in functional luteolysis in the pregnant rabbit corpus luteum: evidence of a role for placental-derived factors in promoting luteal cell survival. In: Program of the 40th annual meeting of the Endocrine Society of Australia; 1997; Canberra, Australia. Abstract 35.
20. Yang E, Korsmeyer SJ. Molecular thanatopsis: a discourse on the BCL-2 family and cell death. Blood 1996; 88:386–401.[Free Full Text]
21. Tilly JL, Tilly KI, Kenton ML, Johnson AL. Expression of members of the bcl-2 gene family in the immature rat ovary: equine chorionic gonadotropin-mediated inhibition of apoptosis is associated with decreased bax and constitutive bcl-2 and bcl-x_long messenger ribonucleic acid levels. Endocrinology 1995; 136:232–241.[Abstract]
22. Kugu K, Ratts VS, Piquette GN, Tilly KI, Tao X-J, Martimbeau S, Aberdeen GW, Krajewski S, Reed JC, Pepe GJ, Albrecht ED, Tilly JL. Analysis of apoptosis and expression of bcl-2 gene family members in the human and baboon ovary. Cell Death Differ 1998; 5:67–76.[CrossRef][Medline]
23. Ratts VS, Flaws JA, Kolp R, Sorenson CM, Tilly JL. Ablation of bcl-2 gene expression decreases the numbers of oocytes and primordial follicles established in the post-natal female mouse gonad. Endocrinology 1995; 136:3665–3668.[Abstract]
24. Knudson CM, Tung KSK, Tourtellotte WG, Brown GAJ, Korsmeyer SJ. Bax-deficient mice with lymphoid hyperplasia and male germ cell death. Science 1995; 270:96–99.[Abstract/Free Full Text]
25. Perez GI, Knudson CM, Leykin L, Korsmeyer SJ, Tilly JL. Apoptosis-associated signaling pathways are required for chemotherapy-mediated female germ cell destruction. Nat Med 1997; 3:1228–1232.[CrossRef][Medline]
26. Tilly JL, Tilly KI, Perez GI. The genes of cell death and cellular susceptibility to apoptosis in the ovary: a hypothesis. Cell Death Differ 1997; 4:180–187.[CrossRef][Medline]
27. Browning JY, Keyes PL, Wolfe RC. A comparison of serum progesterone 20-dihydroprogesterone, and estradiol-17ß in pregnant and pseudopregnant rabbits: evidence for post-implantation recognition of pregnancy. Biol Reprod 1980; 23:1014–1019.[Abstract]
28. Dharmarajan AM, Yoshimura Y, Sueoka K, Atlas SJ, Dubin NH, Zirkin BR, Ewing LL, Wallach EE. Progesterone secretion by the isolated perfused luteal ovary during pseudopregnancy in the rabbit. Biol Reprod 1988; 38:1137–1143.[Abstract]
29. Ahren K, Janson PO, Selstam G. Perfusion of ovaries in vitro and in vivo. Acat Endocrinol 1972; 158:285–309.
30. Gross-Bellard M, Oudet P, Chambon P. Isolation of high-molecular-weight DNA from mammalian cells. Eur J Biochem 1973; 36:32–38.[Medline]
31. Tilly JL, Hsueh AJW. Microscale autoradiographic method for the qualitative and quantitative analysis of apoptotic DNA fragmentation. J Cell Physiol 1993; 154:519–526.[CrossRef][Medline]
32. Gavrieli Y, Sherman Y, Ben-Sasson S. Identification of programmed cell death in situ via a specific labeling of nuclear DNA fragmentation. J Cell Biol 1992; 119:493–501.[Abstract/Free Full Text]
33. Tilly JL. Use of terminal transferase DNA labeling reaction for the biochemical and in situ analysis of apoptosis. In: Celis JE (ed.), Cell Biology: A Laboratory Handbook. San Diego: Academic Press; 1994: 330–337.
34. Feinberg AP, Vogelstein B. A technique for radiolabeling DNA restriction endonuclease fragments to high specific activity. Anal Biochem 1983; 132:6–13.[CrossRef][Medline]
35. Chomczynski P, Sacchi N. Single-step method of RNA isolation by acid guanidinium thiocyanate-phenol-chloroform extraction. Anal Biochem 1987; 162:156–159.[Medline]
36. Sridaran R, Hisheh S, Dharmarajan AM. Induction of apoptosis by a gonadotropin-releasing hormone agonist during early pregnancy in the rat. Apoptosis 1998; 3:51–57.
37. Tilly JL, Tilly KI. Inhibitors of oxidative stress mimic the ability of follicle-stimulating hormone to suppress apoptosis in cultured rat ovarian follicles. Endocrinology 1995; 136:242–252.[Abstract]
38. Wiltbank MC, Dysko RC, Gallagher KP, Keyes PL. Regulation of blood flow to the rabbit corpus luteum: effect of estradiol and human chorionic gonadotropin. Endocrinology 1989; 124:605–611.[Abstract/Free Full Text]
39. Juengel JL, Garverick HA, Johnson AL, Youngquist RS, Smith ME. Apoptosis during luteal regression in cattle. Endocrinology 1993; 132:249–254.[Abstract/Free Full Text]
40. Murdoch WJ. Temporal relationship between stress protein induction, progesterone withdrawal, and apoptosis in corpora lutea of ewes treated with prostaglandin F2 alpha. J Anim Sci 1995; 73:1789–1792.[Abstract]
41. Holt JA. Regulation of progesterone production in the rabbit corpus luteum. Biol Reprod 1989; 40:201–208.[Abstract]
42. Miller JB, Keyes PL. Transition of the rabbit corpus luteum to estrogen dependence during early luteal development. Endocrinology 1978; 102:31–38.[Abstract/Free Full Text]
43. Yuh K-CM, Keyes PL. Properties of nuclear and cytoplasmic estrogen receptor in the rabbit corpus luteum: evidence for translocation. Endocrinology 1979; 105:690–696.[Abstract/Free Full Text]
44. Bagchi MK, Tsai M-J, O'Malley BW, Tsai SY. Analysis of the mechanism of steroid hormone receptor-dependent gene activation in cell-free systems. Endocr Rev 1992; 13:525–535.[Abstract/Free Full Text]