Successful ovulation in plasminogen-deficient mice treated with the broad-spectrum matrix metalloproteinase inhibitor galardin

Transcription

1 Developmental Biology 295 (2006) Successful ovulation in plasminogen-deficient mice treated with the broad-spectrum matrix metalloproteinase inhibitor galardin Kui Liu a,1, Patrik Wahlberg a,1, Göran Leonardsson a,1, Anna-Carin Hägglund a,1, Annelii Ny a,1, Ida Bodén a, Carin Wibom a, Leif R. Lund b, Tor Ny a, a Department of Medical Biochemistry and Biophysics, Umeå University, SE Umeå, Sweden b Finsen Laboratory, Rigshospitalet, Strandboulevarden 49, DK-2100 Copenhagen, Denmark Received for publication 31 May 2005; revised 6 March 2006; accepted 30 March 2006 Available online 7 April 2006 Abstract Many studies have suggested the hypothesis that the plasminogen activator (PA) system and the matrix metalloproteinase (MMP) system, either separately or in combination, may provide the proteolytic activity that is required for rupture of the follicular wall at the time of ovulation. Our recent studies on ovulation in plasminogen (plg)-deficient mice have, however, shown that plasmin is not required for normal ovulation, leading us to the hypothesis that MMPs may be a more important source of proteolysis for this process. To investigate the role of MMPs and also the possibility of a functional overlap or synergy between the MMP and PA systems during ovulation, we have studied ovulation efficiency in wild-type and plg-deficient mice treated with the broad-spectrum MMP inhibitor galardin. We found that in both wild-type mice and heterozygous plg-deficient (plg +/ ) mice that had been treated with galardin prior to ovulation, there was a mild (18 20%) reduction in ovulation efficiency. Surprisingly, galardin treatment of plg-deficient (plg / ) mice only caused an additional 14% reduction in ovulation efficiency as compared to vehicle-treated plg / mice. Our data therefore suggest that although MMPs may play a role in degradation of the follicular wall, they may not be obligatory for ovulation. In contrast to previous studies on tissue remodeling during wound healing and placental development, we have demonstrated that there is no obvious functional overlap or synergy between the PA and MMP systems, which has previously been thought to be essential for the ovulatory process Elsevier Inc. All rights reserved. Keywords: Ovulation; Plasminogen activators; Matrix metalloproteinases; Mice Introduction Ovulation, the liberation of mature ova, is a fundamental process in mammals. It is required for the propagation of all mammalian species (Espey and Lipner, 1994). The ovulatory process involves follicular rupture, which is triggered by a surge of luteinizing hormone (LH) released from the pituitary, and it recurs in every reproductive cycle in female mammals (Erickson, 1986; Espey and Lipner, 1994). For breakdown of the follicular wall to take place, an extensive proteolytic degradation of basement membranes and connective tissues making up the follicular wall is required. Over the past 3 decades, many Corresponding author. Fax: address: tor.ny@medchem.umu.se (T. Ny). 1 These authors contributed equally. studies in both rodents and primates have suggested that two protease families, the plasminogen activator (PA) system and the matrix metalloproteinase (MMP) system, may play an important role in the degradation of the ovarian follicular wall (for references and reviews, see Beers et al., 1975; Strickland and Beers, 1976; Tsafriri and Reich, 1999; Liu et al., 1998; Curry and Osteen, 2001; Ny et al., 2002). Although a substantial body of indirect evidence obtained from several species has suggested that plasmin has a role in ovulation, our studies in plasminogen (plg)-deficient (plg / ) mice have provided genetic evidence that the absence of plasmin has only a mild effect on ovulation, manifested as an insignificant (13%) reduction in ovulation efficiency (Ny et al., 1999). This marginal reduction in ovulation efficiency may also be caused by delayed maturation of the plg / mice, seen as reduced body and ovarian weight (Ny et al., 1999). Based on this /$ - see front matter 2006 Elsevier Inc. All rights reserved. doi: /j.ydbio

2 616 K. Liu et al. / Developmental Biology 295 (2006) genetic evidence, we concluded that although the PA system may participate in the ovulatory process in mice, the role it plays is not indispensable. MMPs are also widely expressed in the ovary during the periovulatory period and many reports suggest that they may play an important role in rupture of the follicular wall during ovulation (for reviews, see Woessner, 1991; Tsafriri and Reich, 1999; Curry and Osteen, 2001; Ny et al., 2002). In other biological or pathological processes requiring proteolysis, including angiogenesis, tumor metastasis, mammary gland involution, adipose tissue development, wound healing, and placental development, a coordinated function between the PA and the MMP systems has been shown to be required (Saksela and Rifkin, 1988; Mignatti and Rifkin, 1993; Khokha et al., 1995; Lund et al., 1999; Lijnen et al., 2002; Ny et al., 2002; Solberg et al., 2003). For example, when MMP activities were effectively suppressed in plg / mice by treatment with the synthetic MMP inhibitor galardin (GM6001), the processes of wound healing (Lund et al., 1999) and placental development (Solberg et al., 2003) were completely blocked. Plasminogen deficiency alone, or galardin treatment alone, had much less effect. The pronounced synergism indicates that one or more proteolytic events are essential for these processes, and that these critical proteolytic events can be the result of either plasmin activity or MMP activity (Lund et al., 1999; Solberg et al., 2003). Inspired by these studies, we wanted to test whether there is a functional overlap or synergy between plasmin and MMPs in follicular rupture during ovulation. We thus studied the effect of the broadspectrum MMP inhibitor galardin on ovulation in wild-type and plg / mice. Materials and methods Materials Pregnant mare serum gonadotropin (PMSG) and human chorionic gonadotropin (hcg) were purchased from Sigma (St. Louis, MO). McCoy's 5A medium (modified, without serum) was purchased from Invitrogen (Gaithersburg, MD). The gelatinolytic activity assay and purified human MMP-2 were purchased from Roche Molecular Biochemicals (Mannheim, Germany). S-2251, a chromogenic substrate for plasmin activity, was purchased from Chromogenix (Mölndal, Sweden). Ninety-six-well plates (Nunclon) were obtained from Nunc A/S (Roskilde, Denmark), and the microtiter plate reader Multiscan RC was from Labsystems (Stockholm, Sweden). Animals C57BL/6 wild-type mice were obtained from Bomholtgård Breeding and Research Center (Ry, Denmark). The plg-deficient mice used were as previously described (Ploplis et al., 1995). The plg / mice, and also their plasminogen heterozygous (plg +/ ) and plasminogen wild-type (plg +/+ ) F2 littermates, were obtained from plg +/ plg +/ breeding (backcrossed 10 generations to C57BL/ 6). The mice were genotyped as previously described (Ny et al., 1999). The mice were kept on a 12-h light/12-h dark cycle with the light cycle initiated at h, and were fed chow and water ad libitum. Experimental protocols were approved by the regional ethical committee of Umeå University. Gonadotropin-induced ovulation in mice To induce synchronized follicular growth and ovulation, immature 25-dayold female mice were injected i.p. with 1.5 IU of PMSG to stimulate follicular development, and with 5 IU hcg 48 h later to induce ovulation. Ovulation normally takes place h after hcg treatment (Rugh, 1990; Ny et al., 1999). Animals were sacrificed by cervical dislocation at selected time points after administration of hcg. To record the number of ova, the mice were sacrificed 20 h after hcg treatment and ova in the oviducts were counted as previously described (Leonardsson et al., 1995; Ny et al., 1999). Body weights and ovarian weights were also recorded. Treatment with MMP inhibitor The peptide hydroxamate MMP inhibitor galardin (Grobelny et al., 1992) was dissolved in 4% (w/v) carboxymethyl cellulose in 0.9% saline, which gave a final concentration of 20 mg/ml. At the time of hcg injection, galardin (100 mg/ kg body weight) was administrated i.p. This dosage of galardin has been proven to effectively inhibit MMP activities and to disrupt wound healing and placental development in plg-deficient mice (Lund et al., 1999; Solberg et al., 2003). The same volumes of vehicle (carboxymethyl cellulose) were injected as controls. In a separate experiment, 200 mg/kg body weight of galardin was also administered during follicular development at 3 time points: at the time of PMSG administration, 24 h after PMSG administration, and at the time of hcg administration. The same volume of vehicle (carboxymethyl cellulose) was used as control. Preparation of ovarian extract Ovaries were dissected free of adhering tissue, washed several times in cold McCoy's 5A medium, and kept at 80 C for further analysis. To ensure that the mice responded properly to hormone treatment, only mice with a body weight of 10 g or more and with ovaries with a total weight of 4 mg or more were used in the experiments. Ovaries (n = 8 16) were transferred to a pre-cooled homogenizer and extracts were prepared on ice, followed by centrifugation at 14,000 rpm for 10 min at 4 C. The supernatants were then collected and protein concentrations were measured using the bicinchoninic acid (BCA) protein assay. Gelatin zymography Ovarian extracts (15 μg of total protein) were analyzed by gelatin zymography in 7.5% SDS-polyacrylamide (SDS-PAGE) gels containing 1.8 mg/ml gelatin, as previously described (Masure et al., 1990). After electrophoresis, the gels were incubated in 2.5% (v/v) Triton X-100 for 2 20 min to remove SDS, and then incubated in a buffer containing 50 mm Tris, ph 7.5, 5 mm CaCl 2, 1% Triton X-100 and 0.02% NaN 3 for 50 h. Gelatin substrate assay Aliquots of 15 μg of ovarian extract were tested for gelatinolytic activity, using an MMP assay detecting degradation of biotinylated gelatin. The assay was performed according to the manufacturer's instructions (Roche). Briefly, MMPs in ovarian extracts were activated immediately prior to use by 2.5 mm aminophenylmercuric acetate (APMA) at ph Ovarian extracts containing activated MMPs were incubated at 37 C for 4 h in 50 mm Tris, 1 mm CaCl 2, 0.05% (v/v) Triton X-100, ph 7.5, with biotin-labeled gelatin. The sample was then transferred to a streptavidin-coated microtiter plate and incubated at room temperature for 30 min. After washing, the plate was incubated with a streptavidin peroxidase conjugate, which would bind to remaining free biotin residues on the biotin-labeled gelatin. Finally, the remaining biotin residues could be detected after incubation with a color substrate (ABTS ). Absorbance at 405 nm was measured for 30 min at 5 min intervals with a microtiter plate reader. In this assay, high gelatinase activity yields low absorbance, while low activity results in high absorbance. In situ gelatinase zymography In situ gelatinase zymography was performed on 10-μm cryosections as described by Leco et al. (2001). Briefly, the sections were covered with a layer of 0.5% low melting point agarose gel containing substrate buffer and 100 μg/ml fluorescein-conjugated DQ gelatine from Molecular Probes Inc. (Eugene, OR), overlayed with coverslips, and incubated in a humidity chamber at 37 C for 16 h. As a negative control, 10 mm EDTA, which chelates Zn 2+ and Ca 2+, was

3 K. Liu et al. / Developmental Biology 295 (2006) included in the substrate buffer. Fluorescence images were taken with a Leica DC300F digital camera attached to a Leica DM LB microscope (Leica, Wetzlar, Germany). Results In vitro inhibition of ovarian MMP activities by the broad-spectrum MMP inhibitor galardin To test whether MMP activities detected in ovarian extracts can be efficiently inhibited by galardin in vitro, we induced follicular growth and ovulation in immature female mice and collected pre-ovulatory ovaries for preparation of ovarian extracts. Aliquots of 15 μg of ovarian extracts were analyzed by gelatin zymography to estimate the activities of MMP-2 and MMP-9 (gelatinase A and B, respectively) in the ovaries. As shown in Fig. 1A (lane 1), ovarian extracts prepared from ovaries just prior to ovulation contained high levels of MMP-2, and MMP-9 to a lesser extent. Both gelatinases appeared to be mostly in their latent pro-forms, as the molecular weights of their activities were reduced by approximately 10 kda following treatment with the MMP-activating agent aminophenylmercuric acetate (APMA) (data not shown) (Hägglund et al., 1999). As shown in Fig. 1A, lanes 2 5, addition of increasing amounts of galardin (0.016 μm to 2 μm) to the ovarian extracts efficiently inhibited the activities of both MMP-2 and MMP-9. The detectable gelatinolytic activity was completely abolished by 0.4 μm galardin, as measured by gelatin zymography (Fig. 1A, lane 4). As a control, full inhibition of the gelatinase activity was also observed when 5 mm EDTA was included as a chelating reagent to block MMP activity by neutralizing Ca 2+ and Zn 2+ in the assay buffer (data not shown). In vivo inhibition of ovarian MMPs by galardin treatment Previous reports have showed that i.p. injection of 100 mg galardin per kg of body weight effectively suppresses MMP activity in mice in vivo (Gijbels 1994; Alexander et al., 1996; Lund et al., 1999). This is a much higher dose as compared to the concentration of galardin we used in our in vitro experiment (Fig. 1). To test whether this dose of galardin is sufficient to inhibit MMP activities in the ovary, we injected galardin into wild-type C57BL/6 mice and measured the gelatinase activities in their ovarian extracts after galardin treatment. Mice were induced to ovulate with PMSG and hcg, and at the same time as hcg treatment, galardin or vehicle was injected. Six hours after galardin or vehicle injection, ovarian extracts were prepared and treated with APMA to convert pro- MMPs to their active forms before the gelatinolytic activity was determined (Hägglund et al., 1999). As shown in Fig. 1B, ovarian extracts from galardin-treated mice contained less than 5% of the gelatinolytic activity found in extracts from vehicletreated mice. When ovarian extracts from vehicle-treated mice and galardin-treated mice were mixed in a 1:1 ratio, still less than 5% of the gelatinolytic activity was measured in the mixture (Fig. 1B). This indicates that the amount of galardin in the ovarian tissue from inhibitor-treated mice was in excess, and Fig. 1. Galardin inactivates MMPs in ovarian extracts. (A) Ovarian gelatinolytic activity was inhibited by galardin in vitro. Twenty-five-day-old wild-type (C57BL/6) mice, previously treated with PMSG and hcg, were sacrificed 4 h after hcg treatment and ovarian extract was prepared. Fifteen-μg aliquots of the ovarian extract were separated by SDS-PAGE and the gel strips were then analyzed for gelatinolytic activity by gelatin zymography in the absence or presence of increasing amounts of galardin. The migration patterns of the proforms of MMP-2 and MMP-9 are indicated with arrows. (B) Ovarian gelatinolytic activity was inhibited by galardin in vivo. Twenty-five-day-old wild-type (C57BL/6) mice were similarly injected with PMSG and hcg as in panel A. At the time of hcg injection, 100 mg galardin per kg of body weight or the same volume of vehicle was injected into the mice. Six hours after hcg treatment, the mice were sacrificed and ovarian extracts were prepared and analyzed by gelatinolytic substrate assay. The sample containing a mixture of extracts from galardin- and vehicle-treated mice in a 1:1 ratio was also analyzed in this assay. The height of the bars represents mean ± SEM of relative gelatinolytic activity, where the activity level from extract of vehicle-treated control mice has been set to 100%. inhibited all additional gelatinolytic activity present in the untreated half of the sample. To test further whether galardin can inhibit MMPs in follicles in vivo, in situ gelatinase zymography was performed on ovary sections from mice 6 h after hcg/galardin treatment. Gelatinase activity was detected in the theca tissue of the follicles from vehicle-treated mice (Fig. 2A, large arrow). In ovaries from galardin-treated mice, only faint gelatinolytic activity was detected (Fig. 2B, small arrow). As a negative control, it was also possible to inhibit the gelatinase activity by the addition of 10 mm EDTA to the zymography gel (Fig. 2C, arrowhead).

4 618 K. Liu et al. / Developmental Biology 295 (2006) Fig. 2. In situ gelatinase activity is suppressed in ovaries from galardin-treated wild-type C57BL/6 mice. Twenty-five-day-old female wild-type (C57BL/6) mice were injected with PMSG and hcg to induce ovulation. At the time of hcg injection, 100 mg galardin per kg of body weight or same volume of vehicle was injected. The mice were sacrificed 6 h after hcg injection and in situ gelatinase zymography was performed on 10-μm frozen sections. Gelatinolytic activity appears as bright green fluorescence. The pictures show representative results from vehicle-treated mice (A), galardin-treated mice (B), and also galardin-treated mice with 10 mm EDTA added to the substrate buffer for in situ zymography (C). These results confirm the effectiveness of galardin as a broadspectrum MMP inhibitor. In addition, they indicate that the inhibitor reaches the follicles when used according to our protocol. Ovulation efficiency in galardin-treated mice We next examined the ovulation efficiency in both wild-type and plg-deficient mice in the presence and absence of galardin. Ovulation was induced by gonadotropins as described previously, and galardin or vehicle was injected at the same time as the hcg injection. Ovulation efficiency was then determined by counting the number of ova in the oviduct 20 h after the hcg injection, by which time ovulation is complete (Leonardsson et al., 1995; Ny et al., 1999). Wild-type C57BL/6 mice were first used to determine the optimal dose of inhibitor and the time point for galardin injection. Galardin (100 mg per kg of body weight) was injected at the same time as hcg, and 6 h before and 6 h after hcg, and the number of ova in the oviduct was recorded. Our data revealed that injection of 100 mg galardin per kg of body weight at the same time as hcg injection or 6 h before hcg gave equal reductions in ovulation efficiency. In contrast, injection of the inhibitor 6 h after hcg had no effect on ovulation efficiency (data not shown). A double-dose of galardin (200 mg of galardin per kg body weight), injected at the same time as hcg, did not suppress the ovulation efficiency any further, while injection of half the dose (50 mg of galardin per kg of body weight) caused approximately half of the degree of suppression (data not shown). This suggests that ovulation is suppressed by galardin in a dose-dependent manner, and that a sufficient amount of galardin is distributed to the ovary to obtain an optimal inhibition when 100 mg galardin per kg of body weight is injected at the same time as the hcg. Having established the protocol for in vivo inhibition of MMPs, we first studied the effect of galardin on ovulation in wild-type C57BL/6 mice. As shown in Table 1, ovulation efficiency was suppressed by 20% (P = 0.007) with galardin treatment relative to vehicle-treated C57BL/6 mice. This moderate inhibition suggests that MMPs may play a role in ovulation, but that they are not indispensable for this process. The effect of galardin on ovulation in wild-type mice is comparable to the 13% reduction in ovulation efficiency caused by plasminogen deficiency (Ny et al., 1999). We then proceeded to test whether follicular rupture involves a functional overlap between MMPs and plasmin, as seen in previous studies in other tissues (Lund et al., 1999; Solberg et al., 2003). Ovulation was induced in galardin-treated plg / mice as well as in plg +/ and plg +/+ littermates. As shown in Table 2, galardin treatment of plg +/+ and plg +/ mice caused a 19% (P = 0.056) and 18% (P = 0.022) reduction in ovulation efficiency, respectively, as compared to vehicle-treated plg +/+ and plg +/ mice. Although these reductions were similar to that in galardin-treated C57BL/6 mice (20% reduction in ovulation efficiency, Table 1), we noticed that the 19% reduction in ovulation efficiency in galardin-treated plg +/+ mice was not statistically significant (P = 0.056), which is different from the significant reduction in ovulation efficiency induced by galardin in commercial C57BL/6 mice (P = 0.007). This may have been caused by the higher body weight variations in the plg +/ mice that were bred in our own animal facility, which may have led to larger standard deviations in the oocyte numbers. Surprisingly, only a further reduction in ovulation efficiency of 14% relative to Table 1 Gonadotropin-induced ovulation in galardin-treated wild-type C57BL/6 mice Genotype Inhibitor treatment No. of mice No. of ova per mouse Reduced ovulation efficiency P value C57BL/6 Vehicle ± 1.8 C57BL/6 Galardin ± % Immature 25-day-old female wild-type (C57BL/6) mice were injected intraperitoneally with 1.5 IU of PMSG to stimulate follicular development, and 48 h later with 5 IU hcg to induce ovulation. At the same time that hcg was injected, 100 mg galardin per kg body weight or the same volume of vehicle was also injected into the mice. Animals were sacrificed 20 h after hcg treatment, and the number of ova in the oviduct was recorded. Numbers of ova per mouse represent mean ± SD and the reduction in ovulation efficiency represents galardin-treated vs. vehicle-treated mice.

5 K. Liu et al. / Developmental Biology 295 (2006) Table 2 Gonadotropin-induced ovulation in galardin-treated plg / mice Genotype Inhibitor treatment No. of mice No. of ova per mouse Reduced ovulation efficiency (galardin vs. vehicle) P value Plg +/+ Vehicle ± 2.2 Plg +/+ Galardin ± % Plg +/ Vehicle ± 1.8 Plg +/ Galardin ± % Plg / Vehicle ± 1.6 Plg / Galardin ± % Ovulation was induced in immature 25-day-old female wild-type (plg +/+ ), heterozygous (plg +/ ), and plg-deficient (plg / ) mice as described in the legend to Table 1. One hundred mg galardin per kg body weight, or the same volume of vehicle, was injected into the mice at the time of hcg injection. Animals were sacrificed 20 h after hcg treatment, and the number of ova in the oviduct was recorded. Numbers of ova per mouse represent mean ± SD and the reduction in ovulation efficiency represents galardin-treated vs. vehicle-treated mice. The reduction in ovulation efficiency between untreated plg / mice and untreated plg +/+ mice was 17% (P = 0.039). The reduction in ovulation efficiency between inhibitor-treated plg / mice and untreated plg +/+ mice was 28% (P = 0.002). vehicle-treated plg / mice was observed when plg / mice were treated with galardin. This mild reduction is comparable to the reduction seen in galardin-treated plg +/+ or plg +/ mice (19% and 18%, respectively) (Table 2). When the ovulation efficiencies of vehicle-treated plg +/+ mice (8.8 ova per mouse) and galardin-treated plg / mice (6.3 ova per mouse) were compared, a 28% reduction (P = 0.002) in ovulation efficiency was seen in the galardin-treated plg / mice. This reduction represents the effect of combined plasminogen deficiency and MMP inhibition on ovulation, and thus suggests a very mild additive effect of the PA and the MMP systems on ovulation. Effects of galardin and plasminogen deficiency on the onset of follicular wall rupture In order to study whether the suppression of MMP activities by galardin alone, or a combined MMP suppression and plasminogen deficiency may delay the onset of follicular wall rupture, we first treated C57BL/6 wild-type mice with galardin or vehicle at the time of hcg injection. Twelve hours after treatment with hcg, the mice were sacrificed and the presence of ova in the bursa or oviduct was investigated. As previously reported for this model, approximately 50% of C57BL/6 mice have started to ovulate at this time point (Ny et al., 1999). Similarly, as shown in Table 3, 5 of 10 vehicle-treated mice, and 5 of 11 galardin-treated mice had started the ovulation process at this time point, indicating that galardin treatment does not cause any apparent delay in the gonadotropin-stimulated onset of ovulation. Previous studies have shown that the onset of follicular wall rupture is not delayed in plg-deficient mice (Ny et al., 1999). Thus, plasminogen deficiency alone, or galardin treatment alone, does not appear to cause any apparent delay in onset of follicular wall rupture. To study whether the combined inhibition of these two enzyme systems would cause a delay in the onset of follicular wall rupture, ovulation was induced in plasminogen wild-type (plg +/+ ), heterozygous (plg +/ ) and homozygous (plg / ) mice as described above. As shown in Table 3, similar numbers of galardin-treated plg / mice and vehicle-treated plg +/+ mice had started to ovulate 12 h after hcg treatment. A similar percentage of mice that had started to ovulate at this time point was also seen in the galardin- and vehicle-treated plg +/ and plg +/+ groups of mice. Discussion Many previous studies in both rodents and primates have suggested that the PA system and the MMP system may play an important role in the degradation of the ovarian follicular wall (Beers et al., 1975; Strickland and Beers, 1976; Liu et al., 1998; Tsafriri and Reich, 1999; Curry and Osteen, 2001; Ny et al., 2002). However, recent studies of ovulation in PA- and plgdeficient mice have revealed that ovulation can occur rather normally in the absence of a functional PA system (Leonardsson et al., 1995; Ny et al., 1999). We and others have found that several MMPs, including MMP-2, MMP-9, stromelysin-3, MT1-MMP and MMP-19, are present in the mouse and rat ovary at the time of ovulation. Some MMPs are also upregulated in pre-ovulatory follicles prior to ovulation, suggesting that they may be involved in the tissue degradation that leads to follicle rupture (Hägglund et al., 1999; Liu et al., 1998; Curry and Osteen, 2001; Jo and Curry, 2004; Jo et al., 2004). These findings have therefore led to the hypothesis that it is the MMP system that may play a more prominent role in ovulation, or that the PA and MMP systems may cooperate in this process. In this report, we have studied the functional role of the MMP system during gonadotropin-induced follicular development and ovulation, and investigated whether the PA and MMP systems cooperate functionally in these processes. To address these questions, we administered the broad-spectrum hydroxamate MMP inhibitor galardin to wild-type or plg-deficient mice. Our data show that although galardin efficiently suppresses ovarian MMP activity in vivo, the ovulation efficiency was only modestly reduced (by 19%) in galardin-treated wild-type mice. Table 3 Onset of follicular wall rupture Genotypes Treatment No. of mice used No. of mice ovulated Percentage of ovulated mice C57BL/6 Vehicle C57BL/6 Galardin Plg +/+ Vehicle Plg +/+ Galardin Plg +/ Vehicle Plg +/ Galardin Plg / Vehicle Plg / Galardin Twenty-five-day-old female wild-type (C57BL/6), plg +/+, plg +/, and plg / mice were injected with PMSG and hcg to induce ovulation, as described in the legend to Table 1. At the time of hcg injection, 100 mg galardin per kg of body weight or the same volume of vehicle was injected. The mice were sacrificed 12 h after hcg injection, and ovaries were removed and examined for the presence of ova in the bursa or in the oviduct, for indications of onset of follicular rupture.

6 620 K. Liu et al. / Developmental Biology 295 (2006) More surprisingly, only a minor additional suppression of ovulation efficiency (14%) was observed when galardin was administered to plg / mice. Thus, in contrast to similar studies with galardin in wound healing and placental development (Lund et al., 1999; Solberg et al., 2003), there appears to be no obvious functional overlap or synergy between the PA and the MMP systems in the ovulatory process. The use of young mice in this study offers the advantage that the ovulatory process can be studied before the onset of any pathological conditions observed later in life in plg / mice (Bugge et al., 1995; Ploplis et al., 1995). Similarly to our previous studies (Ny et al., 1999), the ovulation efficiency was slightly reduced (by 17%) in plg / mice as compared to plg +/+ mice. However, this reduction may be explained by the delay in maturation, manifested as a slightly lower body weight observed in plg / mice (Ny et al., 1999). For the MMP inhibition data presented here, the lower body weight of plg / mice does not influence the interpretation of our results, as these mice were compared to weight-matched untreated mice of similar genetic background. Galardin is a peptide-based zinc-chelating hydroxamate compound that has a general inhibitory mechanism against the action of all MMPs (Grobelny et al., 1992). MMP inhibitors similar or identical to galardin have been found to effectively suppress MMP activity in other physiological and pathological processes involving MMPs, including embryo implantation, experimental autoimmune encephalomyelitis, placental development, adipose tissue development and wound healing (Alexander et al., 1996; Gijbels et al., 1994; Lund et al., 1999; Graesser et al., 2000; Lijnen et al., 2002; Solberg et al., 2003). The inhibitory potency of galardin was characterized by in vitro and in vivo experiments. Excess amounts of inhibitory activity were detected in ovarian extracts from mice treated with galardin at 100 mg per kg of body weight. This amount of galardin caused more than 95% inhibition of the total APMA-activated gelatinolytic activity. As most MMP-2 and MMP-9 found in the ovary are in inactive zymogen forms (Hägglund et al., 1999), the amount of gelatinolytic activity inhibited is much higher than the normal in vivo level. As shown by in situ zymography, native gelatinolytic activity was efficiently suppressed by our galardin treatment. Taken together, these studies suggest that the amount of galardin used in this study (100 mg per kg of body weight) efficiently inhibits ovarian MMPs, and results in a mild suppression of ovulation efficiency. In wild-type C57BL/6, plg +/+ and also plg +/ mice, suppression of ovarian MMP activity by galardin caused a modest (18 20%) reduction in ovulation efficiency in gonadotropinprimed mice. This reduction is smaller than that found with similar synthetic MMP inhibitors in an in vitro perfused rat ovary model, where treatment with synthetic MMP inhibitors (SC and SC 44463) could suppress ovulation by up to 70 75% (Brännström et al., 1988; Butler et al., 1991). In addition, in the ewe, injection of an MMP-2 neutralizing antibody into the follicular antrum blocked ovulation (Gottsch et al., 2002). These differences may be explained by species differences between mice, rats and sheeps. At the same time that the ovulation blockage by in vitro perfused MMP inhibitors does reflect a role for MMPs in ovulation in rats (Brännström et al., 1988; Butler et al., 1991), there may also be physiological differences between the in vivo model used in the present study and the in vitro perfused ovary the in vitro perfusion model was disconnected from the circulatory system and therefore was not accessible to possible compensatory mechanisms. To test the hypothesis that MMPs and plasmin cooperate and/ or functionally complement each other during ovulation, we studied ovulation in plg / mice treated with galardin. Surprisingly, the ovulation efficiency was reduced by only an additional 14% when plg / mice were treated with galardin. Furthermore, neither wild-type mice nor plg / mice displayed any notable delay in the onset of ovulation when treated with galardin. In a separate experiment, plg +/+, plg +/, and plg / mice were treated with one daily dose of 200 mg/kg galardin starting at the time of PMSG treatment for 3 days. However, no additional reduction in ovulation efficiency was seen in these mice (data not shown). The lack of an obviously enhanced or synergistic suppression of ovulation efficiency when plg / mice were treated with galardin suggests that plasmin and MMPs may have a much less pronounced effect on ovulation than previously appreciated. Alternatively, the role of MMPs and plasmin may be effectively compensated for by other mechanisms in our protease-deficient mouse models. The unexpected finding in this study that neither plasmin nor MMPs seem to be obligatory for ovulation contrasts with the requirements on these two protease systems in wound healing and placental development (Rømer et al., 1996; Lund et al., 1999; Solberg et al., 2003). The observed discrepancy between ovulation and wound healing or placental development indicates that ovulation may be a more complex proteolytic process involving other perhaps still unidentified proteases. It also indicates that functional overlap between the PA and the MMP systems is tissue-specific. The involvement of additional proteases in ovulation could also explain the fact that many gene-deficient mouse strains lacking components of the PA or MMP systems are fertile, and can therefore ovulate normally (Carmeliet et al., 1993, 1994; Bugge et al., 1995; Ploplis et al., 1995; Dewerchin et al., 1996; Itoh et al., 1997; Shipley et al., 1996; Soloway et al., 1996, Wilson et al., 1997; Masson et al., 1998; Mudgett et al., 1998; Vu et al., 1998; Ny et al., 2002). Among many possible candidates from other protease families, it was reported that mice that are deficient in the disintegrin metalloprotease ADAMTS-1 showed markedly decreased ovulation efficiency, and increased percentage of anovulatory follicles (Shindo et al., 2000; Shozu et al., 2005; Mittaz et al., 2004). In addition, mice lacking progesterone receptor fail to ovulate and display a deficient expression of cathepsin L and ADAMTS-1 (Robker et al., 2000). This information indicates that ADAMTS-1 may be a crucial protease for ovulation in mice. In the current study, we have also wondered whether the slight reduction in ovulation efficiency in galardin-treated mice could be due to inhibition of ADAMTS-1. We have used western blot to measure the levels of cleaved versican, which is an indicator of ADAMTS-1 activity (Russell et al., 2003), in mouse ovaries after 12 h of hcg and galardin/ vehicle treatment (a time point that is just prior to ovulation).

7 K. Liu et al. / Developmental Biology 295 (2006) However, our results showed that versican cleavage is not significantly altered in ovaries from mice that were treated with galardin (data not shown), indicating that either ADAMTS-1 activity is not significantly suppressed by galardin, or that some other proteases can still cleave versican. In conclusion, the data presented in this study suggest that neither the PA system nor the MMP system either individually or in combination is essential for the degradation of the follicular wall at the time of ovulation. Our results thus challenge the hypothesis that degradation of the follicular wall is mediated by a proteolytic cascade generated from cooperation between the PA and MMP systems, and they suggest that other as yet unidentified proteases may play an important role in this process. Acknowledgments We wish to thank Drs Peter Carmeliet and Victoria Ploplis, who provided the gene-deficient mice. We also thank Pradeep Reddy for excellent technical assistance. This work was supported by the National Cancer Foundation (grant B03-04XAB), the Swedish Research Council (grants VR and VR ), the Swedish Natural Science Foundation (grant B-BU ), and the Norrland Cancer Research Fund/Lions' Cancer Research Fund (grants LP and LP ). References Alexander, C.M., Hansell, E.J., Behrendtsen, O., Flannery, M.L., Kishnani, N.S., Hawkes, S.P., Werb, Z., Expression and function of matrix metalloproteinases and their inhibitors at the maternal embryonic boundary during mouse embryo implantation. Development 122, Beers, W.H., Strickland, S., Reich, E., Ovarian plasminogen activator: relationship to ovulation and hormonal regulation. Cell 6, Brännström, M., Woessner Jr., J.F., Koos, R.D., Sear, C.H., LeMaire, W.J., Inhibitors of mammalian tissue collagenase and metalloproteinases suppress ovulation in the perfused rat ovary. Endocrinology 122, Bugge, T.H., Flick, M.J., Daugherty, C.C., Degen, J.L., Plasminogen deficiency causes severe thrombosis but is compatible with development and reproduction. Genes Dev. 9, Butler, T.A., Zhu, C., Mueller, R.A., Fuller, G.C., Lemaire, W.J., Woessner Jr., J.F., Inhibition of ovulation in the perfused rat ovary by the synthetic collagenase inhibitor SC Biol. Reprod. 44, Carmeliet, P., Kieckens, L., Schoonjans, L., Ream, B., van Nuffelen, A., Prendergast, G., Cole, M., Bronson, R., Collen, D., Mulligan, R.C., Plasminogen activator inhibitor-1 gene-deficient mice. I. Generation by homologous recombination and characterization. J. Clin. Invest. 92, Carmeliet, P., Schoonjans, L., Kieckens, L., Ream, B., Degen, J., Bronson, R., De Vos, R., van den Oord, J.J., Collen, D., Mulligan, R.C., Physiological consequences of loss of plasminogen activator gene function in mice. Nature 368, Curry Jr., T.E., Osteen, K.G., Cyclic changes in the matrix metalloproteinase system in the ovary and uterus. Biol. Reprod. 64, Dewerchin, M., Nuffelen, A.V., Wallays, G., Bouche, A., Moons, L., Carmeliet, P., Mulligan, R.C., Collen, D., Generation and characterization of urokinase receptor-deficient mice. J. Clin. Invest. 97, Erickson, G.F., An analysis of follicle development and ovum maturation. Semin. Reprod. Endocrinol. 4, Espey, L.L., Lipner, H., Ovulation. In: Knobil, E., Neill, J.D. (Eds.), Physiology of Reproduction, 2nd ed. Raven Press, NY, pp Gijbels, K., Galardy, R.E., Steinman, L., Reversal of experimental autoimmune encephalomyelitis with a hydroxamate inhibitor of matrix metalloproteases. J. Clin. Invest. 94, Gottsch, M.L., Van Kirk, E.A., Murdoch, W.J., Role of matrix metalloproteinase 2 in the ovulatory folliculo-luteal transition of ewes. Reproduction 124, Graesser, D., Mahooti, S., Madri, J.A., Distinct roles for matrix metalloproteinase-2 and alpha4 integrin in autoimmune T cell extravasation and residency in brain parenchyma during experimental autoimmune encephalomyelitis. J. Neuroimmunol. 109, Grobelny, D., Poncz, L., Galardy, R.E., Inhibition of human skin fibroblast collagenase, thermolysin, and Pseudomonas aeruginosa elastase by peptide hydroxamic acids. Biochemistry 31, Hägglund, A.C., Ny, A., Leonardsson, G., Ny, T., Regulation and localization of matrix metalloproteinases and inhibitors of metalloproteinases in the mouse ovary during gonadotropin induced ovulation. Endocrinology 140, Itoh, T., Ikeda, T., Gomi, H., Nakao, S., Suzuki, T., Itohara, S., Unaltered secretion of beta-amyloid precursor protein in gelatinase A (matrix metalloproteinase 2)-deficient mice. J. Biol. Chem. 272, Jo, M., Curry Jr., T.E., Regulation of matrix metalloproteinase-19 messenger RNA expression in the rat ovary. Biol. Reprod. 71, Jo, M., Thomas, L.E., Wheeler, S.E., Curry Jr., T.E., Membrane type 1- matrix metalloproteinase (MMP)-associated MMP-2 activation increases in the rat ovary in response to an ovulatory dose of human chorionic gonadotropin. Biol. Reprod. 70, Khokha, R., Martin, D.C., Fata, J.E., Utilization of transgenic mice in the study of matrix degrading proteinases and their inhibitors. Cancer Metastasis Rev. 14, Leco, K.J., Waterhouse, P., Sanchez, O.H., Gowing, K.L., Poole, A.R., Wakeham, A., Mak, T.W., Khokha, R., Spontaneous air space enlargement in the lungs of mice lacking tissue inhibitor of metalloproteinases-3 (TIMP-3). J. Clin. Invest. 108, Leonardsson, G., Peng, X.-R., Liu, K., Nordström, L., Carmeliet, P., Mulligan, R., Collen, D., Ny, T., Ovulation efficiency is reduced in mice that lack plasminogen activator gene function: functional redundancy among physiological plasminogen activators. Proc. Natl. Acad. Sci. U. S. A. 92, Lijnen, H.R., Maquoi, E., Hansen, L.B., Van Hoef, B., Frederix, L., Collen, D., Matrix metalloproteinase inhibition impairs adipose tissue development in mice. Arterioscler. Thromb. Vasc. Biol. 22, Liu, K., Wahlberg, P., Ny, T., Coordinated and cell-specific regulation of membrane type matrix metalloproteinase 1 (MT1-MMP) and its substrate matrix metalloproteinase 2 (MMP-2) by physiological signals during follicular development and ovulation. Endocrinology 139, Lund, L., Rømer, J., Bugge, T., Nielsen, B., Frandsen, T., Degen, J., Stephens, R., Danø, K., Functional overlap between two classes of matrixdegrading proteases. EMBO J. 18, Masson, R., Lefebvre, O., Noel, A., Fahime, M.E., Chenard, M.P., Wendling, C., Kebers, F., Lemeur, M., Dierich, A., Foidart, J.M., Basset, P., Rio, M.C., In vitro evidence that the stromelysin-3 metalloproteinase contributes in a paracrine manner to epithelial cell malignancy. J. Cell Biol. 140, Mignatti, P., Rifkin, D.B., Biology and biochemistry of proteinases in tumor invasion. Physiol. Rev. 73, Mittaz, L., Russell, D.L., Wilson, T., Brasted, M., Tkalcevic, J., Salamonsen, L.A., Hertzog, P.J., Pritchard, M.A., Adamts-1 is essential for the development and function of the urogenital system. Biol. Reprod. 70, Mudgett, J.S., Hutchinson, N.I., Chartrain, N.A., Forsyth, A.J., McDonnell, J., Singer, I.I., Bayne, E.K., Flanagan, J., Kawka, D., Shen, C.F., Stevens, K., Chen, H., Trumbauer, M., Visco, D.M., Susceptibility of stromelysin 1-deficient mice to collagen-induced arthritis and cartilage destruction. Arthritis Rheum. 41, Ny, A., Leonardsson, G., Hägglund, A.-C., Hägglöf, P., Ploplis, V.A., Carmeliet, P., Ny, T., Ovulation in plasminogen deficient mice. Endocrinology 140,

8 622 K. Liu et al. / Developmental Biology 295 (2006) Ny, T., Wahlberg, P., Brändström, I.J.M., Matrix remodeling in the ovary: regulation and functional role of the plasminogen activator and matrix metalloproteinase systems. Mol. Cell. Endocrinol. 187 (1 2), Ploplis, V.A., Carmeliet, P., Vazirzadeh, S., Van Vlaenderen, I., Moons, L., Plow, E.F., Collen, D., Effects of disruption of the plasminogen gene on thrombosis, growth, and health in mice. Circulation 92, Robker, R.L., Russell, D.L., Espey, L.L., Lydon, J.P., O'Malley, B.W., Richards, J.S., Progesterone-regulated genes in the ovulation process: ADAMTS-1 and cathepsin L proteases. Proc. Natl. Acad. Sci. U. S. A. 97, Rømer, J., Bugge, T.H., Pyke, C., Lund, L.R., Flick, M.J., Degen, J.L., Danø, K., Impaired wound healing in mice with a disrupted plasminogen gene. Nat. Med. 2, Rugh, R., The Mouse: Its Reproduction and Development. Oxford Univ. Press, New York, p Russell, D.L., Doyle, K.M.H., Ochsner, S.A., Sandy, J.D., Richards, J.S., Processing and localization of ADAMTS-1 and proteolytic cleavage of versican during cumulus matrix expansion and ovulation. J. Biol. Chem. 278, Saksela, O., Rifkin, D.B., Cell-associated plasminogen activation: regulation and physiological functions. Annu. Rev. Cell. Biol. 4, Shindo, T., Kurihara, H., Kuno, K., Yokoyama, H., Wada, T., Kurihara, Y., Imai, T., Wang, Y., Ogata, M., Nishimatsu, H., Moriyama, N., Oh-hashi, Y., Morita, H., Ishikawa, T., Nagai, R., Yazaki, Y., Matsushima, K., ADAMTS-1: a metalloproteinase-disintegrin essential for normal growth, fertility, and organ morphology and function. J. Clin. Invest. 105, Shipley, J.M., Wesselschmidt, R.L., Kobayashi, D.K., Ley, T.J., Shapiro, S.D., Metalloelastase is required for macrophage-mediated proteolysis and matrix invasion in mice. Proc. Natl. Acad. Sci. U. S. A. 93, Shozu, M., Minami, N., Yokoyama, H., Inoue, M., Kurihara, H., Matsushima, K., Kuno, K., ADAMTS-1 is involved in normal follicular development, ovulatory process and organization of the medullary vascular network in the ovary. J. Mol. Endocrinol. 35, Solberg, H., Rinkenberger, J., Danø, K., Werb, Z., Lund, L.R., A functional overlap of plasminogen and MMPs regulates vascularization during placental development. Development 130, Soloway, P.D., Alexander, C.M., Werb, Z., Jaenisch, R., Targeted mutagenesis of Timp-1 reveals that lung tumor invasion is influenced by Timp-1 genotype of the tumor but not by that of the host. Oncogene 13, Strickland, S., Beers, W.H., Studies on the role of plasminogen activator in ovulation. In vitro response of granulosa cells to gonadotropins, cyclic nucleotides, and prostaglandins. J. Biol. Chem. 251, Tsafriri, A., Reich, R., Molecular aspects of mammalian ovulation. Exp. Clin. Endocrinol. Diabetes 107 (1), Vu, T.H., Shipley, J.M., Bergers, G., Berger, J.E., Helms, A., Hanahan, D., Shapiro, S.D., Senior, R.M., Werb, Z., MMP-9/gelatinase B is a key regulator of growth plate angiogenesis and apoptosis of hypertrophic chondrocytes. Cell 93, Wilson, C.L., Heppner, K.J., Labosky, P.A., Hogan, B.L., Matrisian, L.M., Intestinal tumorigenesis is suppressed in mice lacking the metalloproteinase matrilysin. Proc. Natl. Acad. Sci. U. S. A. 94, Woessner Jr., J.F., Matrix metalloproteinases and their inhibitors in connective tissue remodeling. FASEB J. 5,