| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
Regular Article |
a Laboratory of Developmental Biology, Department of Biotechnology, Seoul Women's University, Seoul 139-774, Korea
b Mirae & Heemang Ob/Gyn Clinic, Seoul 135-120, Korea
c Department of Biology, College of Natural Sciences, Inje University, Kimhae 621-749, Korea
d Department of Ob/Gyn, College of Medicine, Younsei University, Seoul 135-270, Korea
 |
ABSTRACT |
|---|
 TOP ABSTRACT INTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
female reproductive tract, fertilization, follicle, oviduct, ovulation
|
INTRODUCTION |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
The matrix metalloproteinases (MMPs) are a group of proteolytic enzymes requiring both zinc and calcium ions for their enzymatic activity. They can degrade various components of the extracellular matrix and basement membrane [8]. More than 20 enzymes belong to four major classes: the collagenases, the gelatinases, the stromelysins, and the membrane-type metalloproteinases [9]. Of these, gelatinases have a wide spectrum of substrate specificity, such as type IV collagen, laminin, fibronectin, and gelatin [9]. Two forms of gelatinases have been identified: a 72-kDa gelatinase A and a 92-kDa gelatinase B, which are referred to as MMP-2 and MMP-9, respectively. Of these, the 72-kDa gelatinase A is the most widely distributed of all MMPs and has been identified in a variety of tissues, including mammalian ovaries.
Like other MMPs, MMP-2 is synthesized as an inactive proform and is thought to be activated proteolytically either during or after secretion into the extracellular milieu. In mammals, only a single species of 72- to 68-kDa polypeptide chain is known to be an inactive proform, which is proteolytically processed into 68–62 or 45 kDa on activation [10, 11]. Mast cell proteinases [12], serine proteinases such as plasmin and kallikreins [13, 14], or other MMPs, including the membrane-type MMPs [15], have been suggested to act on the proform to produce an active form. The inactive precursor can also be activated in vitro by organomercurials, metal ions, thiol reagents, or detergents [8]. In the extracellular space, the enzymatic activity of MMP-2 is regulated by tissue inhibitors that form noncovalent complexes with MMP-2 [16, 17]. However, the exact mechanism of the activation/inactivation as well as the nature of other plasma inhibitors is not understood.
Mammalian ovaries consistently undergo tissue remodeling during the reproductive period. Many studies have demonstrated that MMPs play important roles in folliculogenesis, including development, luteinization, and luteolysis. Injection of synthetic metalloproteinase inhibitors has blocked ovulation in hamsters and rats [18, 19]. Others have shown that MMP-1 protein and its enzymatic activity of ovine and rabbit ovaries were concentrated at the site of future rupture in the capillary lumina at the apex of the follicle [20, 21]. During proestrus in rats, maximal expression of collagenase-3 was localized to theca cells and interstitial tissue of follicles [22]. Bovine granulosa cells (bGC) and human granulosa cells secreted MMP-2 as they became luteinized cells during in vitro culture [23–25]. Administration of eCG resulted in the increase of both mRNA expression and metalloproteinase activities of collagenase-3 and of 72- and 92-kDa gelatinases of the rat ovarian tissues during follicular growth and expansion [26]. Rat follicles undergoing structural luteolysis by GnRH agonist have been shown to increase the secretion of MMP-2 [27]. The MMPs are also found in the FF of mammals. Human FF exhibited type V collagenolytic activity that increased toward ovulation [28]. Ovine and porcine FFs, based on gelatin substrate zymography, showed the presence of many gelatinases [29, 30]. These observations suggest a possible involvement of FF MMPs in folliculogenesis. However, the precise nature of MMPs present in mammalian FF is not yet known.
The present study aimed to identify gelatinases present in FF and to examine if any change occurred in the biochemical properties of FF gelatinases under the influence of oviductal fluid in bovine species.
|
MATERIALS AND METHODS |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
Acrylamide, bisacrylamide, N,N,N',N'-tetramethylethylenediamine, and bicinconinic acid protein assay kit were purchased from Bio-Rad (Hercules, CA). Ficoll, gold-labeled goat anti-mouse IgG antibody, and IntenSE BL kit were purchased from Amersham International (Buckinghamshire, U.K.). Mouse monoclonal antibodies against human MMP-2 and MMP-9 were purchased from Calbiochem (San Diego, CA). Other chemicals were purchased from the Sigma Chemical Company (St. Louis, MO). The EDTA was prepared as a stock solution of 50 mM in distilled water. The PMSF, phenanthroline, and aminophenylmercuric acetate (APMA) were dissolved into dimethyl sulfoxide as separate 10-mM stock solutions. In preliminary experiments, addition of 10% dimethyl sulfoxide to the body fluids during incubation did not affect the gelatinolytic activity. Soybean trypsin inhibitor (SBTI) was dissolved into distilled water to give a stock solution with a concentration of 10 mg/ml.
Bovine Follicular and Oviductal Fluids
Bovine oviducts and ovaries were collected from a local abattoir (Incheon, Korea). Reproductive histories of the cows were not available. Ovaries were transported to the laboratory on ice in sterile saline. After rinsing several times, FF of ovarian follicles was aspirated using a 22-gauge needle. Pooled FF was centrifuged at 4°C and 2000 x g for 30 min. The supernatant was taken as bovine (b) FF and kept frozen at -20°C. Before use, it was centrifuged again to remove any precipitate. The resulting pellet was used for the preparation of bGC homogenate after washing with PBS. The bovine oviductal fluid (bOF) was prepared as follows: Oviducts were rinsed several times with PBS and then transferred onto the filter paper and freed from the adherent lipid using scissors. They were then squeezed out between forceps to liberate inner fluid into the watch glass. The fluid was centrifuged at 4°C and 12 000 x g for 60 min. The supernatant was taken as the oviductal fluid and kept at -20°C until use. Fetal bovine serum (FBS) was purchased from Sigma. Bovine serum (bS) was collected from cows that donated follicles and oviducts. All body fluids were filtered using a 0.45-µm mesh membrane (Millex-HV; Millipore, Bedford, MA) immediately before incubation at 37°C. When necessary, EDTA, PMSF, phenanthroline, or SBTI was added to the bFF to give a final concentration of 5 mM or 1 mg/ml before incubation. The APMA was used at a 1 mM concentration.
bGC Homogenate
The pellet of bGC was overlaid on Ficoll and centrifuged at 400 x g for 30 min at 4°C. The cells were collected from the interface, followed by washing with PBS twice by centrifugation at 300 x g for 10 min at 4°C. After homogenization on ice in 10 mM sodium phosphate buffer (pH 7.4) containing 5 mM EDTA, 1 mM PMSF, and 1 mg/ml of SBTI, they were centrifuged again at 4°C and 1500 x g for 10 min. The supernatant was taken and assayed for the protein content by using bicinconinic acid protein assay kit. The homogenate was then mixed with SDS-PAGE sample buffer without mercaptoethanol.
Gelatin Zymography
The SDS-PAGE was used with the addition of 1 mg/ml of bovine skin gelatin (type B) to an 8% resolving gel as described previously [31]. Briefly, samples were dissolved into the SDS sample buffer in the absence of reducing agent without boiling. Unless indicated otherwise, each 0.75 µl of body fluid or 100 µg of protein of bGC homogenate were loaded onto a single well. After electrophoresis using a Hoefer mini gel kit (Hoefer Scientific Instruments, San Francisco, CA), gels were washed with 2.5% Triton X-100 in Tris-HCl buffer (pH 8.0) and gel incubation buffer (5 mM CaCl2, 0.02% NaN3, 50 mM Tris-HCl, pH 8.0). The washed gels were incubated with fresh incubation buffer for 48 h at 37°C. The reacted gels were stained with 0.5% Coomassie brilliant blue G-250 dye in 5% methanol and 7% glacial acetic acid and destained with distilled water. The clear bands on blue background were regarded as gelatinase bands, because gelatinases degrade gelatin in the acrylamide gel. Wide-range molecular markers from Sigma (M-3788) were used as a standard for SDS-PAGE gel.
When indicated, 5 mM EDTA, 5 mM PMSF, 5 mM phenanthroline, or 1 mg/ml of SBTI as a protease inhibitor was added to the gel incubation buffer after electrophoresis. Every zymographic result was confirmed by two or more repetitive experiments.
Immunoblotting
After nonreducing SDS gel electrophoresis, the gels were soaked in transfer buffer for 15 min. Transfer buffer was made of 25 mM Tris (pH 8.4), 192 mM glycine, and 10% methanol. The proteins were electrotransferred onto a polyvinylidene fluoride membrane (Immobilon-P; Millipore) for 1 h at 4°C and 200 mA. Before transfer, membrane was hydrated with absolute methanol for 15 sec, distilled water for 2 min, and then with transfer buffer for 5 min. After transfer, membrane was treated with methanol for 10 sec, dried on filter paper, and then treated again with methanol for 5 min. To saturate nonspecific binding sites, membrane was treated at 37°C for 1 h with PT buffer (10 mM sodium phosphate buffer [pH 7.4], 0.05% Tween 20, and 10 mM sodium azide) containing 5% BSA. It was then incubated for 1 h with PT buffer containing 1% normal goat serum and 1 µg/ml of antibody against human MMP-2 or MMP-9. After washing with PT buffer containing 0.1% BSA three times for 10 min each time, membrane was incubated for 1 h with PT buffer containing 1:100 diluted, gold-labeled goat anti-mouse IgG antibody. After washing the membrane three times, the signal was revealed by IntenSE BL kit according to the manufacturer's manual. Immunoblotting results were confirmed by three independent experiments.
|
RESULTS |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
The gelatinases of bFF obtained from eight cows are shown in Figure 1. Some variation was found in the intensity of each gelatinolytic protein band depending on the individual cows. However, most bFF typically exhibited gelatinases of 110 (GA110)-, 92-, 88–84-, and 62-kDa forms from the top of the gel. Of these, the 62-kDa gelatinase was the most prominent. Gelatinases of 92 and 58 kDa (not shown) were discernable in some samples but were not always apparent. In contrast, bOF obtained from six cows exhibited only two gelatinase bands of 97 and 62 kDa, of which the 62-kDa form was distinct in some bOF. However, gelatinases of most bOF showed very weak intensity (Fig. 1).
|
bOF Abolishes bFF GA110 Gelatinase
Surprisingly, when bFF was mixed with bOF and then incubated at 37°C, gelatinolytic activity of bFF GA110 dramatically diminished after 3 h (Fig. 2A). In particular, the disappearance of GA110 was distinct in bFF that was mixed with bOF in a 1:10 or a 1:1 ratio. Diminished activity of GA110 was even significant in bFF mixed with bOF in a 10:1 ratio. The reaction occurred so fast that the decreased gelatinolytic activity of GA110 was observed within 30 min after mixing at a 1:1 ratio (Fig. 2B). However, other bFF gelatinases of 92, 88–84, and 62 kDa were not affected by bOF when it was added to bFF in a 1:1 ratio (Fig. 2B).
|
To see if the disappearance of GA110 of bFF caused by bOF could be inhibited using some of the known protease inhibitors, 5 mM EDTA, 5 mM phenanthroline, 5 mM PMSF, or 1 mg/ml of SBTI was added to bFF and then mixed with bOF. After incubation of the mixture at 37°C for 3 h, the samples were analyzed for gelatinolytic activity (Fig. 2C). Interestingly, treatment of bFF with EDTA or phenanthroline increased the GA110 activity without affecting other gelatinase activities, and EDTA in particular greatly increased GA110 activity. Neither PMSF or SBTI affected the enzymatic activity of GA110 or of the other gelatinases. Nevertheless, addition of bOF to either of these pretreated bFFs consistently abolished the GA110 activity, regardless of whether the activity was enhanced by the inhibitors.
bOF Specifically Acts on bFF GA110
Because higher molecular weight gelatinases are not likely to be synthesized per se, the possibility that some bFF gelatinases might be isoforms of the known gelatinase was examined using APMA, which is an activator of MMP proform. The bFF was treated with 1 mM APMA for 30 min, 3 h, or 24 h at 37°C and then analyzed by zymography. As seen in Figure 3A, most gelatinase bands, except for the 62-kDa band, disappeared within 30 min after treatment, and the concomitant increment of the 62-kDa band intensity was observed as treatment continued up to 24 h. These results showed that all gelatinases appearing in bFF were isoforms of the 62-kDa gelatinase.
|
Whether the gelatinolytic proteins appearing in bFF and bOF were indeed gelatinases was examined using protease inhibitors (Fig. 3B). When bFF pretreated with EDTA was electrophoresed and the gel incubated in the buffer containing 5 mM EDTA, which is a known inhibitor of MMP and Ca2+-dependent protease, all gelatinolytic activities, except that of the 88- to 84-kDa gelatinase, were abolished, regardless of whether bFF was pretreated. Gelatinases of bOF were also abolished completely by incubation with EDTA. Phenanthroline, which is a specific inhibitor of MMP, also gave a similar result as EDTA. However, PMSF, which is a serine/threonine protease inhibitor, failed to inhibit the enzymatic activity of gelatinases, including GA110 and the 62-kDa form. These results demonstrate that gelatinolytic proteins of bFF and bOF, except for the 88- to 84-kDa protease, of bFF were indeed gelatinases.
Identification of GA110 Gelatinase
To address the origin of GA110 appearing in bFF, its possible existence in bS and ovarian bGC was examined (Fig. 4). Without treatment with EDTA, bS and FBS exhibited a barely detectable level of GA110 activity. However, those treated with EDTA before zymography showed a greatly enhanced enzymatic activity of GA110, which was abolished by bOF as in bFF. In contrast, bGC did not show any gelatinolytic activity corresponding to GA110 even though pretreated with EDTA.
|
Finally, using monoclonal antibodies against human MMP-2 and MMP-9 that could recognize both inactive and active gelatinase isoforms, immunoblotting was carried out to identify the nature of the GA110 and 62-kDa gelatinases of bFF and bS (Fig. 5). Untreated bFF showed only the 62-kDa gelatinase, which was immunologically active against anti-MMP-2 antibody among many enzymatically active ones. However, bFF treated with EDTA showed an additional immunoreactive band of GA110 that was abolished by bOF. No change occurred in the immunoreactivity of the bFF 62-kDa form despite these treatments. Untreated bS did not show any immunoreactive band. Only after EDTA treatment did bS give an immunoreactive GA110, and bOF treatment of this bS resulted in the disappearance of GA110, accompanied by the appearance of the 62-kDa form. An additional immunoreactive 300-kDa protein, which was enzymatically inactive as seen in previous zymograms, was observed in both bFF and bS treated with EDTA or not. The anti-MMP-9 antibody failed to detect any discernable immunoreactive protein in bFF.
|
|
DISCUSSION |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
Among many gelatinases of bFF revealed by gelatin zymograms, GA110 and 62-kDa gelatinases were consistently found in bFF. However, treatment of bFF with APMA resulted in the disappearance of GA110, and this was accompanied by the concomitant increase in the intensity of 62-kDa gelatinase. Because APMA interacts with the precursors of MMPs, resulting in autolytic cleavage of the proenzyme itself to generate an active form [8], both the disappearance of GA110 and the increment of 62-kDa gelatinase by APMA indicate that GA110 might be an isoform of 62-kDa gelatinase. The immunoblot experiments showing that both GA110 and 62-kDa gelatinases were immunoreactive against the anti-human MMP-2 antibody support the idea that GA110 gelatinase is an isoform of MMP-2 and that the 62-kDa gelatinase is an active form of MMP-2.
To date, only two MMP-2 isoforms are known in mammals. One is an inactive proenzyme of 72 kDa, and the other is an active form of 62 kDa [32]. In addition to these two molecules, our results demonstrate that other MMP-2 isoforms are present in bFF and bS. One of them, GA110, has a higher molecular weight than the 72-kDa isoform, as revealed by gelatin gel zymography and immunoblot analyses. Because MMP-2 isoforms are not likely to be synthesized from other than a single known mRNA species [33], GA110 per se cannot be synthesized by cells. Rather, its appearance could be a result of the interaction of a 72- or 62-kDa isoform with other polypeptides. Whichever molecule either isoform is bound to, the interaction should be a covalent bonding, because it was not dissociated in SDS-gelatin gel. Some proteins are known to covalently bind MMPs. Human 
2-macroglobulin, pregnancy zone protein, and neutrophil gelatinase-associated lipocalin make a complex through covalent bonds with MMPs [34–37]. When MMPs interact with these molecules, they cleave the bait region of proteins, followed by formation of a covalent bond to the fragment that is produced. Possibly, GA110 is similarly produced from the covalent binding of MMP-2 to an as-yet-unidentified protein via bait region cleavage.
We have consistently observed that EDTA treatment of bFF and bS greatly increased the activity of GA110. Because EDTA inhibits the enzymatic activity of MMPs by chelating the divalent ions needed, the increased band intensity of GA110 by EDTA is due not to an increased enzymatic activity but, rather, to an increased number of GA110 molecules. This explanation is supported by the observation that the immunoreactivity of GA110 was barely discernable in untreated bFF, whereas it became distinct in bFF after EDTA treatment. How EDTA leads to an increase in GA110 molecules in bFF or bS is not currently known, but our immunoblot experiments showed that both bFF and serum exhibited the presence of a 300-kDa protein that strongly reacts with the anti-MMP-2 antibody. It remains to be elucidated whether this molecule might be related to the appearance of GA110 by EDTA.
Surprisingly, when bFF was mixed with bOF, we observed that bOF induced proteolytic degradation of the MMP-2 isoform, GA110, of bFF. Disappearance of GA110 by bOF was also observed in bS and FBS treated with EDTA to increase the enzymatic activity of GA110. Even within 30 min after its addition to the bFF, bOF could abolish the activity of bFF GA110. Although a similar disappearance of bFF GA110 was also observed with APMA treatment instead of bOF treatment, the action mechanism of APMA, however, was not the same as that of bOF. Additionally, bOF did not affect other gelatinase activities of bFF, such as the 92-kDa form, even after incubation for 24 h, whereas APMA treatment resulted in the disappearance of most bFF gelatinases, except for the 62-kDa form. Furthermore, bOF treatment was not accompanied by increased activity of the 62-kDa form, whereas APMA treatment resulted in the increase of 62-kDa form activity, with the concomitant disappearance of GA110. Therefore, the proteolytic effect of bOF on GA110 appears to be specific and might be a degrading rather than an activating one, unlike that of APMA on GA110.
Little information is available regarding the proteolytic enzymes in bOF. Recent experiments have shown that both oviductal secretion in and oviductal tissue extract from superovulated hamster females exhibited proteolytic activity, some of which was ascribed to a plasminogen activator [38]. However, this type of oviductal enzyme, the activity of which was inhibited by SBTI or PMSF, is not relevant to the bOF molecules responsible for the degradation of GA110, because the latter reaction was not inhibited by either SBTI or PMSF.
Compared to a number of gelatinases in bFF and bS, bOF exhibited only two gelatinases (97 and 62 kDa). The 62-kDa form, as deduced by its molecular weight, is supposed to be MMP-2. However, the enzymatic activities of both gelatinases were very weak compared to those of bFF and bS. Throughout the reproductive cycle, most mammalian ovaries and uteri undergo extensive remodeling of tissues, which require active participation of various MMPs [39]. Mammalian oviduct also exhibits cyclical changes of structures, such as the height of ciliated epithelium the and secretory activity of nonciliated epithelium [1]. However, drastic remodeling, as seen in the ovary and uterus, has not been reported in the oviductal tissues. Under these circumstances, oviductal tissues may not need to synthesize and secrete active gelatinases to remodel themselves. Thus, weak gelatinase activity of bOF due to a regulated synthesis might relate to the absence of drastic remodeling of the oviductal tissues.
The origin of GA110 in bFF does not seem to be ovarian bCG. Rather, it appears in bFF by a selective transudation from bS. Both adult bS and FBS showed the presence of GA110, the enzymatic activity of which was greatly increased by EDTA treatment of bFF. The GA110 of these body fluids was also immunologically active against the anti-MMP-2 antibody. In contrast, bCG homogenate showed a lack of enzymatically active GA110, even though it was prepared in the presence of EDTA. Earlier findings similarly showed that freshly isolated human granulosa cells exhibited little MMP-2 activity [24, 25]. Therefore, the GA110 of bFF obviously originated from bS. However, circulating bS isoforms do not appear to simply be accumulated within ovarian follicles. Our results showed many differences between the gelatinase profiles of bFF and bS. For example, bFF showed 62-kDa gelatinase as the predominant form, whereas bS exhibited that 62-kDa gelatinase was not as intense as in bFF, as in pig FF [30]. These differences imply that GA110 might appear in bFF via selective bS transudation by follicle cells, as has been suggested previously [40].
Ovine and porcine FFs also exhibit many gelatinases, including those presumed to be MMP-2 and -9 [29, 30]. Similar to bFF gelatinases, no detectable change was found in the FF gelatinase levels of these animals during follicular growth (i.e., change depending on the size of the follicles). However, the MMP-2 isoform corresponding to GA110 of bFF and bS was not detected in the FF of these animals. Gelatin gels of these studies were incubated for 24 h in the gel incubation buffer after SDS-PAGE, whereas gels of the present study were incubated for 48 h in the same buffer to increase the sensitivity. In addition, a larger sample amount (0.75 µl) was examined in this study than in the previous ones (0.5 µl). These differences in incubation time and sample amount might account for the absence of GA110 in porcine and ovine FFs in the previous studies, although the possibility of a species difference cannot be eliminated.
Throughout this study, we have made some interesting findings. First, bFF and bS share a unique isoform of MMP-2, GA110, which might be produced from the covalent bonding of MMP-2 to an unknown molecule. Second, bOF possesses little gelatinase activity compared to bFF and bS. Third, and most important, GA110 of bFF and bS can be specifically degraded by the oviductal fluid component. Based on these observations, we conclude that bFF and oviductal components interact with each other, resulting in biochemical modification of the former components. However, the physiological meaning of this interaction remains to be elucidated. Further studies of the nature of GA110 and bOF components might give insight regarding the role of mutual interaction in the events of fertilization and/or embryonic development in mammals.
|
FOOTNOTES |
|---|
1 Supported by grant 2000-1-20700-002-3 from the Basic Research Program of the Korea Science and Engineering Foundation. 
2 Correspondence: Haekwon Kim, Laboratory of Developmental Biology, Department of Biotechnology, Seoul Women's University, Kongnung-dong, Nowon-ku, Seoul 139-774, Korea. FAX: 082 2 970 5669; hwkim{at}swu.ac.kr
Accepted: July 24, 2001.
Received: March 12, 2001.
|
REFERENCES |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
2-macroglobulin. Arch Biochem Biophys 1997; 347:62-68[CrossRef][Medline]2-macroglobulin in the acute and chronic wound environments. J Invest Dermatol 1998; 110:771-776[CrossRef][Medline]This article has been cited by other articles:
|
|
 |
|
  F. Gaytan, E. Tarradas, C. Bellido, C. Morales, and J. E. Sanchez-Criado Prostaglandin E1 Inhibits Abnormal Follicle Rupture and Restores Ovulation in Indomethacin-Treated Rats Biol Reprod, October 1, 2002; 67(4): 1140 - 1147. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
