| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
Regular Article |
a Department of Medical Biosciences, Medical Biochemistry, Umeå University, S-90187 Umeå, Sweden
b Institut de Genetique et de Biologie Moleculaire et Cellularie, CNRS/INSERM/ULP, BP 163, 67404 Illkirch Cedex, C.U. de Strasbourg, France
ABSTRACT
Apoptotic processes are often associated with an intense proteolytic remodeling of the extracellular matrix (ECM). Proteolytic degradation of the ECM can also be a signal that induces apoptosis. Here, we have investigated the expression pattern and functional role of the matrix metalloproteinase stromelysin-3 in follicular atresia. Twenty-four hours after the treatment of immature female mice with a low dose of eCG, both apoptosis and the stromelysin-3 mRNA expression were suppressed approximately threefold. However, the initial suppression of apoptosis and stromelysin-3 expression was followed by a time-dependent increase, and 96 h after eCG treatment, the levels were similar to those of untreated control mice. In 15- to 16-day-old juvenile mice, the ovary consisted of relatively undeveloped follicles, and almost no apoptosis and only low stromelysin-3 mRNA expression were observed. However, at the age of 21 days, when several antral follicles were present, a fivefold induction in both apoptosis and stromelysin-3 mRNA expression was detected. For both models, in situ analysis revealed that the expression of stromelysin-3 mRNA was localized to the granulosa cells of atretic follicles. To address the functional role of stromelysin-3 in follicular atresia, stromelysin-3-deficient mice were studied. However, no difference in the pattern of apoptotic DNA fragmentation and no apparent morphological differences were observed when ovaries from wild-type and stromelysin-3-deficient mice were compared. Taken together, our data indicate that stromelysin-3 is induced during follicular atresia, but that this protease is not obligatory for initiation or completion of the atretic process.
apoptosis, follicle, ovary
INTRODUCTION
Apoptosis plays an important role during embryonic development and postnatal tissue remodeling by eliminating discrete cell populations or cells that have developed improperly. The ovary is characterized by extensive tissue remodeling and massive cell death of both somatic and germ cells. In the human ovary, for example, as few as 0.1% of the follicles actually ovulate during reproductive life, whereas more than 99% are degenerated [1]. The degenerative process by which follicles are eliminated is termed atresia, and recent morphological and biochemical studies have demonstrated that follicle atresia is a hormonally controlled apoptotic process [2–4]. Studies of other physiological systems have shown that the nature of the extracellular matrix (ECM) can influence apoptotic programs in mammalian cells [5]. An example of this is involution of the mammary gland, which follows expression of the lactational phenotype and is characterized by degradation of ECM by matrix metalloproteinases (MMPs) [6]. This proteolytic degradation of the ECM seems to be a signal that induces the accompanying apoptotic process [7]. In a similar fashion, neuronal death in the hippocampus is promoted by degradation of the ECM protein laminin by plasmin [8, 9].
One protease that has been implicated in apoptotic processes is stromelysin-3 (i.e., MMP-11) [10–12]. Stromelysin-3 is a member of the MMP family, which is a family of extracellular proteases that currently consists of approximately 20 different members. The MMPs share a similar domain structure, including a Zn2+-binding site in the catalytic domain, and they are synthesized as proteolytically inactive proenzymes. Together, the MMPs have enzymatic activity against virtually all components of the ECM [13, 14]. Stromelysin-3 was first identified as a protein highly expressed in stromal cells surrounding invasive breast carcinomas [15]. Since then, abnormal expression of this MMP has been observed in various carcinomas, including those of the skin, ovary, and lung [16]. Stromelysin-3 is also expressed in some physiological conditions associated with intense tissue remodeling and, notably, in areas of extensive apoptotic cell death [10–12, 15].
In this study, we used two different mouse models of follicular atresia to examine the regulation and spatial expression pattern of stromelysin-3 mRNA during apoptosis in the ovary. In addition, mice deficient for stromelysin-3 were used to study the functional role of stromelysin-3 in follicular atresia. Our data suggest that stromelysin-3 may play a role in ECM remodeling or tissue degradation during follicular atresia, but that proteolysis mediated by stromelysin-3 is not a prerequisite for induction of the apoptotic process.
MATERIALS AND METHODS
Materials
Paraformaldehyde and eCG were purchased from Sigma Chemical Company (St. Louis, MO). McCoys 5A medium (modified without serum) and Klenow fragment of DNA polymerase I were obtained from Gibco BRL, Life Technologies, Inc. (Gaithersburg, MD). Tissue-Tec OCT Compound was purchased from Miles, Inc. (Elkhart, IL). The riboprobe in vitro transcription system was from Promega (Madison, WI). Anti-digoxigenin antibodies, digoxigenin-labeled uridine triphosphate (UTP), terminal transferase, and biotin-16-deoxyuridine triphosphate (dUTP) were obtained from Roche Diagnostics Scandinavia AB (Bromma, Sweden). SuperFrost*/Plus microscope slides were purchased from Menzel-Glaser (Braunschweig, Germany). The Ultraspec RNA Isolation System was from Biotecx Laboratories (Houston, TX). The 
-35S-deoxyadenosine triphosphate (dATP) (1000 Ci/mmol) and -32P-UTP (800 Ci/mmol) were obtained from Amersham Pharmacia Biotech (Buckinghamshire, England).
Animals
Immature female mice (C57BL/6J) were obtained from Bomholt Gård Breeding and Research Centre Ltd.-Boommice (Ry, Denmark). The generation of stromelysin-3-deficient mice has been described previously [17]. The stromelysin-3-deficient mice had been back-crossed to C57BL/6J for 10 generations, and the genotype of the mice was confirmed by Southern blot analysis [17]. The mice were fed a regular chow-and-water diet. A 12L:12D cycle was maintained, with the light cycle initiated at 0600 h. Experimental protocols were approved by the regional ethical committee of Umeå University.
Two models were used to study follicular atresia. In the first model, follicular atresia was induced by treating 23-day-old immature female mice (body weight 
9–10 g) with a single injection of 1.5 IU of eCG [18]. The mice were then killed at 24, 48, 72, or 96 h after eCG treatment, and ovaries were collected. In the second model, ovaries from juvenile mice (15–21 days of age) were collected. The ovaries used for in situ hybridization were fixed overnight at 4°C in a freshly prepared solution of 4% w:v paraformaldehyde in PBS, cryoprotected by an overnight incubation at 4°C in a solution of 30% sucrose in PBS, and then embedded in Tissue-Tek OCT Compound.
Synthesis of RNA Probe
The mouse stromelysin-3 fragment (nucleotides 250–685) [10] was obtained by reverse transcription-polymerase chain reaction and ligated into the pT7 Blue T-vector (Novagen, Madison, WI). Before transcription, the plasmid was linearized such that antisense or sense RNA probes could be obtained. For Northern blot analysis, transcription was performed using -32P-UTP and an in vitro transcription system from Promega. The specific activity of the probe was 2–5 x 108 cpm/µg RNA. The riboprobe used for in situ hybridization was labeled with digoxigenin-labeled UTP and the appropriate RNA polymerase.
RNA Preparation and Analysis
Total RNA from mouse ovaries was isolated with the Ultraspec RNA Isolation System. For Northern blot analysis, total RNA (5–10 µg) was fractionated by agarose gel electrophoresis in the presence of formaldehyde, then transferred to Hybond-N filters (Amersham) according to the supplier's instructions. The prehybridization and hybridization were performed as previously described [19]. The relative abundance of specific mRNA was analyzed with a PhosphorImager (Molecular Dynamics, Inc., Sunnyvale, CA) and normalized to the relative abundance of glyceraldehyde phosphate dehydrogenase mRNA.
In Situ Hybridization
The in situ hybridization was performed as previously described [20, 21]. Photographs were taken with a Leica (Heerbrugg, Switzerland) camera attached to a Leica DMBL microscope.
Apoptotic DNA Fragmentation Analysis
Genomic DNA was extracted from ovaries as previously described [22]. The DNA, in 500-ng aliquots, was labeled at the 3' ends with -35S-dATP and Klenow enzyme [23]. After 2 h of incubation at 4°C, the labeled DNA was ethanol precipitated and then fractionated by agarose gel electrophoresis (2% agarose). The degree of apoptotic DNA fragmentation was analyzed with a PhosphorImager.
In Situ DNA Fragmentation Analysis
Cryostat sections of 10 µm were collected on SuperFrost*/Plus slides and fixed in 4% paraformaldehyde (in PBS) for 15 min. After washing, sections were treated with 0.6% H2O2 for 30 min to inactivate endogenous peroxidase. The 3' ends of the DNA fragments were labeled with biotin-16-dUTP by incubation with terminal transferase for 2 h at 37°C. The sections were further incubated for 1 h with biotinylated peroxidase-avidin complexes. The DNA fragmentation was visualized by staining with aminoethyl carbazole chromogenic substrate (Dako, Copenhagen, Denmark). When terminal transferase was omitted from the procedure, no color reaction was detected.
Data Analysis
All experiments for Northern blot analysis and 3'-end labeling were repeated at least three times. Each time, ovaries from three mice were used per time point. Data are expressed as mean ± SEM of three individual experiments. Statistical comparisons were made by one-way ANOVA followed by Newman-Keuls test. A value of P < 0.05 was considered to be significant. All experiments for in situ hybridization and in situ 3'-end labeling were repeated at least three times, and each time, ovaries from two different mice were used per time point.
RESULTS
Effects of Gonadotropin Treatment on Apoptosis and Stromelysin-3 mRNA Expression in the Ovary
To study the expression pattern of stromelysin-3 mRNA during apoptosis in the ovary, 23-day-old female mice were treated with 1.5 IU of eCG. At different time points after the hormone treatment, ovaries were collected and analyzed for the appearance of apoptotic cell death and stromelysin-3 mRNA expression. Apoptotic cell death was detected by the presence of DNA fragments in size multiples of 185–200 base pairs (Fig. 1A), and the degree of apoptosis was calculated by quantification of the low-molecular-weight DNA fragments (<15 kilobases [kb]) (Fig. 1B). In ovaries from untreated control mice, high levels of apoptosis were found. Northern blot analysis revealed that these ovaries also expressed high levels of stromelysin-3 mRNA (Fig. 1C). Treatment of the mice with eCG resulted in a suppression of both apoptosis and stromelysin-3 expression (Fig. 1, A–C). At 24 h after eCG treatment, a fourfold reduction in apoptotic DNA fragmentation (Fig. 1B) and a threefold reduction in stromelysin-3 mRNA expression (Fig. 1C) were observed as compared to the levels in untreated control mice. However, this initial decrease in apoptotic DNA fragmentation and stromelysin-3 mRNA expression was followed by a time-dependent increase until 96 h after the eCG treatment, at which time the level of apoptotic DNA fragmentation and stromelysin-3 mRNA expression was similar to that of ovaries from untreated control mice.
|
To identify atretic follicles and follicles expressing stromelysin-3 mRNA, in situ 3'-end labeling (Fig. 2A) and in situ hybridization (Fig. 2B) were performed on adjacent sections of an ovary collected 96 h after eCG treatment. Several of the follicles were atretic, as determined by in situ 3'-end labeling (Fig. 2A). In situ hybridization on the adjacent section (Fig. 2B) revealed that the stromelysin-3 mRNA expression was only localized to the granulosa cells of atretic follicles, whereas no expression was detected in the healthy follicles.
|
Age-Dependent Increase in Ovarian Apoptosis and Stromelysin-3 mRNA Expression in Juvenile Mice
Previous studies in the rat have shown that the level of apoptotic DNA fragmentation in the ovary is low until the age of 18 days, after which a major increase in ovarian cell apoptosis takes place [24]. To further examine if the stromelysin-3 expression is linked to apoptosis in the ovary, we adapted this model to the mouse. Ovaries from mice at different ages (15–21 days) were collected and analyzed for the appearance of apoptotic cell death and stromelysin-3 mRNA expression.
Low levels of apoptotic DNA fragmentation were detected on Days 15 and 16 (Fig. 3A). However, on Days 19 and 21, the level of DNA fragmentation had increased dramatically. Quantification revealed a fivefold increase in apoptosis on Day 21 as compared to the level on Day 15 (Fig. 3B). Analysis of the stromelysin-3 mRNA expression revealed a similar pattern, with low stromelysin-3 mRNA expression on Days 15 and 16 and increases of four- and fivefold on Days 19 and 21, respectively (Fig. 3C).
|
At the age of 15 days, the mouse ovary consisted of many secondary follicles; however, almost no staining by in situ 3'-end labeling was observed, which indicated that the level of apoptosis was very low at this age (Fig. 4A). Consistent with the Northern blot hybridization data (Fig. 3C), in situ hybridization on the adjacent slide revealed a negligible stromelysin-3 mRNA expression (Fig. 4B). At the age of 21 days, some ovarian follicles had developed to the antral stage. In situ analysis revealed that several of these antral follicles were atretic (Fig. 4C). In situ analysis also revealed that the stromelysin-3 mRNA expression was localized to the granulosa cells of the atretic follicles, whereas no expression of stromelysin-3 mRNA was detected in healthy follicles (Fig. 4D).
|
Atresia in Stromelysin-3-Deficient Mice
The expression pattern of stromelysin-3 in the ovary suggests that this protease may be involved in ECM remodeling processes that are associated with the induction of apoptosis or the facilitation of apoptotic tissue remodeling and resorption. To test the functional role of stromelysin-3 for the apoptotic processes in the ovary, we analyzed mice that were deficient in stromelysin-3. Stromelysin-3-deficient, 23-day-old female mice that had been back-crossed to C57BL/6J for 10 generations were treated with 1.5 IU of eCG. At different time points after the hormone treatment, ovaries were collected and analyzed for the appearance of apoptotic cell death. The eCG treatment initially suppressed apoptosis by a factor of four as compared to the level in untreated mice (Fig. 5, A and B). After this, a time-dependent increase in apoptotic DNA fragmentation was detected until 96 h after the eCG treatment, when the level was similar to that of untreated control mice. The pattern of apoptotic DNA fragmentation in the stromelysin-3-deficient mice thus is similar to that observed in wild-type mice (compare Figs. 1B and 5B).
|
As shown above for wild-type mice, ovaries collected from stromelysin-3-deficient mice at 15–21 days of age also revealed low levels of apoptotic DNA fragmentation on Days 15 and 16, which was increased on Days 19 and 21 (Fig. 6A). Quantification by PhosphorImager showed that the level of apoptosis increased by four- and fivefold on Days 19 and 21, respectively (Fig. 6B). This magnitude of induction is similar to that seen in wild-type mice (compare Figs. 3B and 6B). Therefore, for both apoptosis models, the pattern of apoptotic DNA fragmentation was the same in both wild-type and stromelysin-3-deficient mice.
|
Morphological analyses of follicular atresia have identified stages of the degenerative process that characterize the atretic process. To investigate if the degenerative process was affected in stromelysin-3-deficient mice, we performed histological analyses of ovarian sections from young (15 days of age) as well as old (9 mo of age) female mice. However, these analyses revealed no apparent morphological differences between ovaries from wild-type and stromelysin-3-deficient mice (data not shown).
DISCUSSION
Previous studies have shown that degradation of the ECM by proteases, such as MMPs and plasmin, can be a signal that induces apoptosis. Because stromelysin-3 has been implicated in apoptotic processes, we have studied the expression pattern and functional role of stromelysin-3 in follicular atresia, a hormonally controlled apoptotic process [5, 10–12]. Two different mouse models were used to study the expression of stromelysin-3 during follicular atresia. Our study revealed a prominent induction of stromelysin-3 in atretic follicles undergoing apoptosis, which suggests that stromelysin-3 may play a role in ECM remodeling or tissue degradation associated with atresia. However, the apoptotic process and follicular atresia were apparently normal in stromelysin-3-deficient mice, which indicates that expression of stromelysin-3 is not required for the apoptotic process to proceed.
Two different apoptosis models were used in this study. In the first model, immature female mice were treated with eCG. This gonadotropin acts on the ovary both by recruiting small follicles into a more active growth phase and by "rescuing" follicles from atresia [25–27]. By using a DNA fragmentation assay, we could show that apoptosis was suppressed approximately fourfold at 24 h after eCG treatment as compared to the level in untreated control mice. However, because the eCG-treated mice did not receive any ovulation-inducing signal, they did not ovulate. Instead, the follicles entered the degenerative atretic pathway, and the initial decrease in apoptosis was followed by a time-dependent increase. At 96 h after eCG treatment, the apoptotic DNA fragmentation was at a level similar to that of untreated control mice (Fig. 1). In a second, alternative apoptosis model, ovaries from juvenile, 15- to 21-day-old female mice were studied during their first wave of follicle development. Ovaries from mice at 15–16 days of age contained many secondary follicles, and the level of apoptotic DNA fragmentation was low (Fig. 3). However, when the mice reached 19–21 days of age, an increase in apoptosis was observed that correlated with the appearance of more developed follicles in the ovary (Fig. 4). Consistent with a previous study in the rat [24], many of the antral follicles in the ovary of 21-day-old mice were atretic.
In an earlier study, in which the regulation of MMPs during ovulation was assessed, the expression pattern of stromelysin-3 was different from that of other MMPs [28]. During both follicular development and ovulation, the stromelysin-3 expression was only localized to granulosa cells of small- and middle-sized follicles. No expression could be detected in large, preovulatory follicles, which expressed other MMPs at the time of ovulation. This expression pattern, therefore, suggests that stromelysin-3 is not involved in follicular rupture but, rather, in other tissue remodeling events in the ovary. By using the two apoptosis models described above, we found that the stromelysin-3 mRNA expression was regulated in a manner similar to that of apoptosis in the ovary. Furthermore, in situ analysis localized the stromelysin-3 mRNA expression to the granulosa cells of atretic follicles (Figs. 2 and 4). Our data, therefore, indicate that induction of stromelysin-3 coincides with hormonally controlled apoptosis, which suggests that this protease may play a role in ECM remodeling or tissue destruction during apoptosis. One possibility is that the stromelysin-3 expression is a consequence of atresia, and that this protease, together with other proteases, is involved in downstream tissue-remodeling events. Alternatively, stromelysin-3 could have an effector role and be involved in the induction of atresia by ECM proteolysis, as has been discussed for other systems [5, 7].
Studies of other biological systems have shown that matrix-degrading proteases are not only expressed in areas associated with intense tissue remodeling and apoptotic cell death but can also be involved in the induction of apoptosis. Both in mammary gland involution and in neuronal cell death, apoptosis can be induced by proteolytic degradation of the ECM [7, 9]. In the mammary gland, ectopic expression of the MMP stromelysin-1 leads to unscheduled involution and mammary epithelial cell apoptosis, whereas inhibition of stromelysin-1 delays involution and mammary epithelial cell apoptosis [6, 29]. In a similar fashion, the serine protease plasmin has been shown to participate in excitotoxin-induced neuronal cell death in the hippocampus. In this system, excitotoxic injury leads to plasmin formation and a plasmin-catalyzed degradation of the ECM protein laminin, which induces neuronal cell death [8, 9].
Expression of stromelysin-3 has been detected in areas associated with intense tissue remodeling and apoptotic cell death, such as during mammary gland involution, tadpole metamorphosis, and limb-bud morphogenesis in the mouse embryo [10, 11, 30]. Several of these apoptotic processes are characterized by basement-membrane degradation, and a role for stromelysin-3 in this degradation process has been proposed. In ovarian follicles, alterations and subsequent degeneration of the follicular basement membrane is an early event in follicular atresia [31]. Our finding that stromelysin-3 is expressed in atretic follicles suggests that stromelysin-3, perhaps in cooperation with other proteases, could be involved in basement-membrane degradation during atresia.
Recently, stromelysin-3 gene-deficient mice were created. These mice were found to be fertile and did not exhibit obvious alterations in appearance or behavior [17]. We used these mice to test if the absence of stromelysin-3 would affect follicular atresia. By using the two apoptosis models described above, we could not detect any difference in the pattern of apoptotic DNA fragmentation between stromelysin-3-deficient and wild-type mice. Furthermore, a simple histologic evaluation revealed no apparent difference in ovarian morphology when ovarian sections from wild-type and stromelysin-3-deficient mice were compared. These data show that stromelysin-3 is not obligatory for initiation or completion of the atretic process. Alternatively, the role of stromelysin-3 is compensated for by other proteases in stromelysin-3-deficient mice, such as up-regulation of other MMPs, as has been shown during uterine involution in matrilysin- and stromelysin-1-deficient mice [32]. However, redundant mechanisms may exist, and the role of stromelysin-3 might be taken over by other proteases without any difference in expression pattern.
Taken together, our data show that the expression of stromelysin-3 mRNA in the ovary is regulated in a way similar to apoptosis. The stromelysin-3 expression is localized to atretic follicles, which indicates that stromelysin-3 may play a role in ECM remodeling or tissue degradation during apoptosis. However, no difference in apoptotic DNA fragmentation or ovarian morphology was observed when ovaries from wild-type and stromelysin-3-deficient mice were compared, suggesting that stromelysin-3 is neither obligatory nor has an effector role in inducing the apoptotic phenotype. Alternatively, redundant or compensatory mechanisms may be compensating for the lack of stromelysin-3 in stromelysin-3-deficient mice.
ACKNOWLEDGMENTS
We wish to thank James Snell for critically reading this manuscript.
FOOTNOTES
First decision: 4 August 2000.
1 Supported by the Swedish Medical Research Counsel (K97-13X-09709-07A), the Swedish Cancer Society (3912-B97-01XAB), and Cancerforskningsfonden in Umeå (LP1177/95). 
2 Correspondence. FAX: 46 90 136465;tor.ny{at}medchem.umu.se
Accepted: September 11, 2000.
Received: July 10, 2000.
REFERENCES
This article has been cited by other articles:
|
|
 |
|
  R. Wu, K. H. Van der Hoek, N. K. Ryan, R. J. Norman, and R. L. Robker Macrophage contributions to ovarian function Hum. Reprod. Update, March 1, 2004; 10(2): 119 - 133. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 T. E. Curry Jr. and K. G. Osteen The Matrix Metalloproteinase System: Changes, Regulation, and Impact throughout the Ovarian and Uterine Reproductive Cycle Endocr. Rev., August 1, 2003; 24(4): 428 - 465. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
