| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
Regular Article |
-Induced Luteolysis in Sheep1
a Department of Animal & Nutritional Sciences, University of New Hampshire, Durham, New Hampshire 03824
b Department of Animal Science, University of Connecticut, Storrs, Connecticut 06269
 |
ABSTRACT |
|---|
 TOP ABSTRACT INTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
(PGF2) typically initiates a cascade of events that leads to the functional and structural demise of the corpus luteum. A sheep model was used in which a 1-h, systemic infusion of PGF2 (20 µg/min) is given at midcycle. Such an infusion mimics the onset of spontaneous luteolysis by causing a transient decrease in peripheral plasma progesterone, which reaches a nadir (
60% of controls) at 8 h but returns to control levels by 16–24 h. We investigated whether PGF2 also influenced the endogenous protein levels of tissue inhibitors of metalloproteinases, TIMP-1 and TIMP-2, and matrix metalloproteinases, MMP-2 and MMP-9, all of which have been implicated in remodeling of the extracellular matrix (ECM). Corpora lutea (Day 11) were collected at 0 h and at 1, 8, 16, and 24 h post-PGF2 infusion (n = 3 sheep at each time). Immunoblot analysis revealed an immediate and precipitous decline in TIMP-1 (30 kDa) and TIMP-2 (19 kDa) protein levels (60% and 90%, respectively; P < 0.05) at the 1-h time point and remained depressed at 8 h (P < 0.05). Gelatin zymography and other procedures identified three MMPs (85, 70, and 64 kDa), which were shown to be the latent form of MMP-9 and the active and latent forms of MMP-2, respectively. In contrast to the rapid decrease in TIMP-1 and -2 levels, an increase in MMP-2 activity (165% of controls, P < 0.05) occurred at 8 h, which corresponded to the nadir in plasma progesterone. These early changes in TIMPs and MMPs indicate that alterations in the structure of the ECM by PGF2 may play a hitherto unsuspected role in the subsequent process of functional luteolysis.
corpus luteum, corpus luteum function, ovary, progesterone
|
INTRODUCTION |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
(PGF2) was identified as the luteolytic agent in sheep [3], and good evidence exists for a similar role in other species, such as the cow, pig, guinea pig, and pseudopregnant rat [2]. Luteolysis has a functional aspect, which is observed as a decrease in P production, and a structural aspect, which is the physical involution of the gland. Structural regressive processes include degradation of the extracellular matrix (ECM) [4]. The ECM, besides providing structural support between cells, is also implicated in modulating certain cell processes, such as differentiation, migration, gene expression, and apoptosis [4, 5]. Degradation of the ECM is facilitated by the action of a family of calcium- and zinc-dependent endopeptidases known as matrix metalloproteinases (MMPs) [6, 7]. Two family members, MMP-2 and -9 (gelatinase A and B, respectively), can degrade type IV collagen, which is a primary component of basement membrane [6, 7] and a constituent of luteal tissue [8]. The MMP activity is regulated, in part, by the tissue inhibitors of metalloproteinases (TIMPs), which form tight, noncovalent bonds with the enzymes, thereby inhibiting them [9, 10]. It is believed that the balance between these two groups of proteins underlies the structural integrity of the cellular environment.
Both MMPs and TIMPs have been implicated in many tissue-remodeling processes, including growth and regression of the CL. In the rat, an associated increase in the active form of MMP-2 protein was observed during structural involution of the CL [11, 12]. In cattle, luteal TIMP-1 and -2 mRNA expression increases significantly 8 h post-PGF2 treatment [13]. In sheep CL, TIMP-1 and -2 transcripts, but not MMPs, are well characterized [14, 15]. Within 2 h of exogenous or pharmacological PGF2 treatment, the cytoskeleton of large luteal cells decomposes [16], which is followed by diminished luteal mass within 24 h [17]. Interestingly, TIMP-1 protein levels in ovine CL decrease 6 h after a 25-mg dose of PGF2 i.m. [18]. These findings prompted us 1) to confirm the presence of TIMP-1, TIMP-2, MMP-2, and MMP-9 proteins in luteal tissue and 2) to correlate the temporal expression patterns of TIMPs and MMPs in luteal tissue with the temporary decline in plasma P following a 1-h, systemic infusion of PGF2, which was calculated to mimic the effect of the first endogenous pulse of PGF2 during luteolysis in sheep.
|
MATERIALS AND METHODS |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
-Induced Luteolysis
The natural luteolysin of sheep, PGF2 is released from the uterus in a series of episodic pulses, each lasting approximately 1 h [19]. With this knowledge, a model was developed in the intact sheep in which a relatively high amount of PGF2 (0.22 µg/kg/min) is given systemically for 1 h [20]. This rate of infusion allows sufficient PGF2 to escape metabolism by the lungs and, thus, to reach the ovary via the systemic circulation. The temporary decline in P (40%) caused by such an infusion closely mimics the fall in P seen after the first endogenous pulse of PGF2 at the onset of luteolysis in sheep [21]. This model allowed us to harvest luteal tissue from PGF2-treated sheep at specific time points during a physiological induction of the very early stages of luteolysis.
Tissue Collection and Preparation
The estrous cycles of a group of mixed Suffolk and Dorset ewes, housed at the University of Connecticut in Storrs, were synchronized using two i.m. injections each of 5 mg of Lutalyse (UpJohn Co., Kalamazoo, MI) given 4 h apart. The ewes were observed twice daily for estrus using a vasectomized ram (Day 0 = estrus). On the 11th day postestrus, ewes were placed in metabolism cages, and cannulae (16-gauge) were inserted into both jugular veins under local anesthesia (2% [w/v] lidocaine). Each test animal was given a 1-h, systemic infusion of PGF2 (UpJohn) into the right jugular vein at a rate of 20 µg/min using a Harvard Infusion Pump (model no. 600-910/920; Harvard Apparatus Co., Holliston, MA) [20]. Control animals received no treatment. The CL were removed surgically via flank laparotomy under local anesthesia (2% [w/v] lidocaine), immediately placed on dry ice, and then stored at -80°C until analysis by zymography and immunoblotting. The CL were collected before PGF2 treatment at 0 h (controls) and at 1, 8, 16, and 24 h from the onset of the PGF2 infusion (n = 3 sheep at each time point, with the exception of 8 h, for which n = 2 sheep). To monitor peripheral plasma P, jugular blood samples (5 ml) were collected hourly via the left jugular vein cannula into heparinized tubes from time zero until the time of individual luteectomy. Animals that underwent luteectomy at 24 h had blood collected hourly until 16 h and then every 2 h until luteectomy. All animal procedures were approved by the University of Connecticut Animal Care and Use Committee.
Protein Extraction
Proteins were extracted in 2 M NaCl, 0.01 M Hepes, and 0.02% (w;clv) NaN3 (pH 7.6) as previously described [22, 23]. Briefly, tissues were thawed, weighed and kept on ice. Conditions were normalized using an 8:1 (v/w) ratio of extraction buffer to tissue (0.8 ml:0.1 mg). Samples were minced with a scalpel blade and homogenized before being placed on a Clay Adams Nutator Mixer (Beckton Dickinson, Sparks, MD) to extract for 24 h at 4°C. Extracts were then centrifuged at 2250 x g to pellet the homogenized tissue. Supernatants were subsequently centrifuged several times at 15 000 x g to further remove particulate matter. Finally, ultrafiltration through Centricon membranes (10 000 Mr cut-off; Amicon, Beverly, MA) at 5000 x g was used to dialyze (0.2 M NaCl, 1 mM CaCl2, 50 mM Tris, and 0.02% NaN3 [pH 7.6]) and to concentrate supernatants fourfold over the original volumes. Samples were stored at 4°C until analysis.
Radioimmunoassay for P
Jugular blood samples were spun in a centrifuge at 1400 x g at 4°C for 20 min, and the plasma was separated immediately and then stored at -20°C until P analysis. Plasma was extracted using petroleum ether, and P levels were determined using a previously described procedure [24]. The P-conjugate antibody was generated in sheep and used at a final dilution of 1:17 500 (v/v).
Reverse Zymography
Reverse zymography was used to determine the presence of TIMPs. This procedure [25] involved several modifications to the one described below for gelatin zymography. Luteal samples were fractionated in 12% (w/v) polyacrylamide gels (to increase resolution of lower molecular weight TIMPs) that were impregnated with 0.05% (w/v) gelatin. After Triton X-100 washes, gels were incubated for 4 h at 37°C in conditioned medium of HT-1080 cells that were treated with dexamethasone (4 µM) and activated with aminophenylmercuric acetate (2 mM). HT-1080 (catalog no. CCL-121; American Type Culture Collection, Manassas, VA) is a human fibrosarcoma cell line known to produce several MMPs (e.g., MMP-2 and -9) and TIMPs (e.g., TIMP-1). Whereas MMPs in this conditioned medium digest the gelatin impregnated in the gel, the TIMPs in the luteal samples (i.e., in the gel) inhibit MMP action. Gels were then rinsed once with distilled water, placed in substrate buffer, and incubated overnight (15 h). After staining with Coomassie blue solution, gels were destained for 1 h using a solution of methanol, acetic acid, and water (50:10:40 [v/v]), followed by additional destaining with water overnight. Dark blue bands corresponding to the reported molecular weights for TIMPs appear on the gels.
Immunoblot Analysis
Immunoblotting was performed with modifications to the procedure described by Goldberg et al. [23]. Briefly, standard SDS-PAGE was conducted using 10% and 12% polyacrylamide gels, respectively, for MMPs and TIMPs to fractionate the proteins in each sample under nonreducing (MMP-2 and TIMP-1) and reducing (TIMP-2) conditions. Equal volumes of samples for MMP-2 and TIMP-1 immunoblots were mixed 1:1 (v/v) with sample buffer, whereas samples for TIMP-2 were boiled for 3 min in sample buffer (1:1 [v/v]) containing 200 µM dithiothreitol before loading in gels. After fractionation, proteins were electrophoretically transferred onto a nitrocellulose membrane (0.45 µm; Schleicher and Schuell, Keene, NH) at 200 mA for 3 h at approximately 4°C. All subsequent steps were performed at room temperature. Nonspecific binding sites were blocked with 5% milk in TBST buffer (0.15 M NaCl, 0.01 M Tris, and 0.05% [v/v] Tween-20; pH 8.0) for 2 h. Membranes were then incubated overnight (14 h) with primary mouse monoclonal antibodies that were specific for each assay: MMP-2 (2 µg/ml; catalog no. IM33L; Oncogene Research Products, Cambridge, MA), TIMP-1 (1 µg/ml; catalog no. IM32L; Oncogene Research Products), or TIMP-2 (5 µg/ml; catalog no. IM56; Oncogene Research Products). The following day, four 20-min washes in TBST buffer were done before incubation with a secondary antibody (goat anti-mouse) conjugated with horse radish peroxidase (HRP; 1:15 000 [v/v]; Bio-Rad Laboratories, Melville, NY) for 1 h. Afterward, four additional washes were followed by a 5-min incubation of the membranes with Super Signal West Pico Chemiluminescent Substrate (for detection of HRP; Pierce, Rockford, IL).
The relative molecular mass (Mr) of each immunoreactive protein was identified by comparison with broad-range molecular weight markers (Kaleidoscope; Bio-Rad) run in an adjacent lane. In addition, HT-1080-conditioned medium, containing MMPs and TIMPs, was used as a positive control throughout. Furthermore, a recombinant bovine TIMP-1 or human TIMP-2 protein standard (Oncogene Research Products) was included when appropriate. For the negative control, blots were not incubated with respective primary antibodies.
Documentation of Immunoblots
Visual records of immunoreactive bands were produced by the exposure of membranes to X-OMAT scientific imaging film (Eastman Kodak Co., Rochester, NY) under safelight conditions. Films of immunoblots were developed in a Konica (Wayne, NJ) automatic developer.
Gelatin Zymography
Zymography was used to detect gelatinase (MMP-2 and -9) activity using previously described methods [23, 26] with minor modifications. Equal volumes of luteal extracts were mixed 1:1 (v/v) with sample buffer (10% SDS, 4% [w/v] sucrose, 0.1% [w/v] bromophenol blue, and 0.25 M Tris [pH 6.8]) and electrophoresed under nonreducing conditions at 200 V using the Mini-Protean II system (Bio-Rad) in 10% (w/v) polyacrylamide gels containing 0.05% (w/v) gelatin. Afterward, gels were washed twice (15 min each wash) in 2.5% (w/v) Triton X-100 to remove SDS, rinsed with distilled water, and incubated for 17–18 h at 37°C in substrate buffer (5 mM CaCl2 and 50 mM Tris [pH 8.0]). The next day, gels were stained with Coomassie blue R250 solution (0.5% [w/v] in a 1:3:6 [v/v] ratio of acetic acid, isopropanol, and distilled water) for 30 min and destained with distilled water for 2 days. The MMP activity was observed as zones (bands) of clearance against the blue-stained background of the gel. Adjacent lanes contained molecular weight markers (Novagen, Madison, WI) and the positive control, HT-1080-conditioned medium.
To verify that bands of clearance were the result of metal-dependent proteinases (i.e., MMPs), gels were incubated in substrate buffer containing 1,10-phenanthroline (10 mM; Sigma, St. Louis, MO). This chelator interferes with the zinc-containing active site of bonafide MMPs, preventing zones of clearance from forming. Furthermore, latent and active forms of MMP-2 and -9 were distinguished by incubating samples with 2 mM p-aminophenylmercuric acetate (APMA; Sigma) for 2 h before electrophoresis. It is purported that association with APMA causes cleavage of a peptide segment from the N-terminal end of latent MMPs, thus converting them to their active, lower molecular weight forms [27]. This causes the banding density of the latent form to decrease and that of the active form to increase.
Documentation of Zymograms
All gels produced by zymography and reverse zymography were photographed using a Polaroid Photo Documentation Camera FB-PDC-34 (Fisher Scientific, Pittsburgh, PA) and 665PN Polaroid black-and-white film (Polaroid Corporation, Cambridge, MA).
Quantification and Statistical Analysis
Films of the immunoblots and photographs of the zymograms were scanned using HP DeskScan II (Hewlett-Packard Co., Palo Alto, CA), and bands were digitized using Un-Scan-It (Gel Version 5.1; Silk Scientific, Inc., Orem, UT) to enable quantitative comparisons of band intensities by counting total pixels. The average numbers of total pixels from all samples at each of the five time points for each protein species was compared statistically using one-way ANOVA. This was followed by the Tukey test for pairwise comparisons to determine the level of significance between time points. These analyses were performed using SYSTAT software (SYSTAT, Inc., Evanston, IL).
|
RESULTS |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
Plasma P levels before PGF2 treatment ranged between 3.0 and 6.0 ng/ml. Therefore, subsequent changes in P for each animal were expressed as a percentage of controls. Following the infusion of PGF2, plasma P declined steadily and, by approximately 8 h postinfusion, reached 60% of controls (P < 0.05) before recovering again by 16 and 24 h (Fig. 1).
|
Identification of TIMPs by Reverse Zymography
To initially determine if TIMP-1 and -2 were present in our luteal samples, we used reverse zymography (Fig. 2). Visual observations revealed a band migrating at approximately 30 kDa, which is consistent with the Mr reported for TIMP-1. A second band migrating at approximately 19 kDa corresponded to the Mr reported for TIMP-2.
|
Identification of TIMPs by Immunoblotting
Immunoblotting was used to verify the presence of TIMP-1 and -2 in all sheep luteal extracts. Preliminary experiments demonstrated that immunoblot signals varied proportionately with dilutions of a representative sample (data not shown). Representative immunoblots for both TIMPs are shown in Figure 3. The TIMP-1 immunoblot (Fig. 3A) was produced using a mouse monoclonal antibody generated against bovine TIMP-1. Immunoreactive bands migrated at approximately 30 kDa, which is consistent with the Mr reported for the TIMP-1 protein. These bands also comigrated with two positive controls: a recombinant bovine TIMP-1 protein, and the conditioned media of HT-1080 cells. Visual observations revealed a strong signal for this protein at the preinfusion time point of 0 h, but immediately following PGF2 treatment, TIMP-1 levels decreased sharply and did not return to baseline values within 24 h.
|
To determine if another TIMP might compensate for the decreases seen in TIMP-1, we also investigated the expression of TIMP-2, which preferentially binds MMP-2. Immunoblots were produced using a mouse monoclonal antibody generated against a peptide corresponding to a sequence in human TIMP-2 (Fig. 3B). Immunoreactive bands comigrated with a recombinant human TIMP-2 protein at approximately 19 kDa, which is consistent with the Mr reported for TIMP-2. The protein expression pattern of TIMP-2 was similar to that of TIMP-1; TIMP-2 levels were highest at the preinfusion time point, fell markedly by 1 h post-PGF2 treatment, and did not recover by 24 h.
Time Course of TIMP Expression Following PGF2 Infusion
Immunoblots representing all luteal samples for TIMP-1 and -2 were subjected to densitometric analysis (Fig. 3). The TIMP-1 (Fig. 3A) protein levels decreased significantly (P < 0.05) at the 1-h (arbitrary pixel units, mean ± SD, 17.71 ± 1.6) and 8-h (18.11 ± 4.78) time points versus control values (45.1 ± 13.9). This initial depression began to recover at the 16- and 24-h time points, but it did not fully return to preinfusion values. Similarly, TIMP-2 (Fig. 3B) also decreased significantly (P < 0.05) from controls (17.68 ± 5.53) at the 1-h (2.08 ± 0.87) and 8-h (1.56 ± 0.01) time points. However, TIMP-2 expression remained low and was still significantly (P < 0.05) less than controls at 24 h (4.2 ± 3.18).
Identification of MMPs by Zymography and Immunoblotting
Identification of MMPs was determined by zymography. A representative zymogram of luteal samples from the 0-, 1-, 8-, 16-, and 24-h time points is shown in Figure 4. Visual observations revealed the presence of three zones of clearance in all samples. A strong band of enzyme activity had a Mr of approximately 70 kDa, whereas a weaker band of enzyme activity had a Mr of approximately 64 kDa. These two enzyme species have molecular weights consistent with those reported for MMP-2 family members. A third band of enzyme activity had a Mr of approximately 85 kDa, which is consistent with reported MMP-9 family members.
|
To further characterize these MMP enzymes, gels were loaded with three sample types previously shown to possess MMP activity (data not shown) and then incubated with the MMP-inhibitor 1,10-phenanthroline (Fig. 5A). This resulted in complete ablation of all clearance zones in all sheep luteal samples as well as the two positive controls, a cow luteal sample, and conditioned media of HT-1080 cells, indicating that these were truly divalent cation-dependent enzymes. Furthermore, pretreatment of samples before electrophoresis with APMA diminished the intensity of the 70-kDa band, whereas the 64-kDa band remained unaltered (Fig. 5B), indicating that the 70- and 64-kDa enzyme species likely were the latent and active forms, respectively, of MMP-2. Also, APMA treatment markedly diminished the intensity of the 85-kDa band, indicating that it was probably the latent form of MMP-9.
|
Finally, immunoblotting was used to verify the identity of MMP-2 (Fig. 5C). A single immunoreactive band with a Mr of approximately 70 kDa was present in a representative set of samples and comigrated with the HT-1080 positive control. Immunodetection of MMP-9 was unsuccessful despite use of three different antibody preparations.
Time Course of MMP Expression Following PGF2 Infusion
All zymograms were subjected to densitometric analysis, which revealed that the latent forms of MMP-2 and -9 remained unchanged (P > 0.05) over the course of the experiment (data not shown). However, a statistically significant increase (P < 0.05) was observed in the active form of MMP-2 at 8 h following PGF2 infusion (control, 4.8 ± 0.84; 8 h, 7.94 ± 0.27) (Fig. 6).
|
|
DISCUSSION |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
is administered to initiate rapid and irreversible regression of the CL. Whereas this approach is useful for investigating many aspects of luteolysis, it may not be physiologically relevant. To our knowledge, the present study is the first to examine the temporal changes in MMPs and TIMPs under simulated physiological conditions at the onset of luteolysis.
In the sheep model system used during the present study, PGF2 was given systemically for 1 h at 20 µg/min. Such a rate of infusion allows approximately 1% of the infused PGF2 to escape metabolism by the lungs [17, 28]. The amount of PGF2 reaching the ovary elicits a 40% decrease in plasma P [20]. This reduction in plasma P is essentially identical to that seen following the first endogenous uterine pulse of PGF2 during spontaneous luteolysis in normally cycling sheep [21]. It is important to note that this single exposure to PGF2 is not, by itself, luteolytic, because the CL fully recovers its functional capabilities within 16–24 h. During natural luteolysis, five to six additional pulses of PGF2 would occur approximately 8 h apart to ensure irreversible luteal regression [29].
This physiologically accurate model of early luteolysis was used to study the effects of PGF2 on MMP and TIMP protein expression in vivo. Both MMPs and TIMPs have been implicated in tissue remodeling processes that require the breakdown of ECM, resulting in the loss of cell adhesion and apoptosis [30, 31]. During luteal regression, these same cellular processes promote structural involution of the CL. It is believed that ECM turnover is regulated, in part, by the balance between the proteolytic action of MMPs on matrix components and the inhibitory action of endogenous TIMPs on these MMPs.
In human luteal tissue, TIMP-1 mRNA expression significantly decreased after PGF2 induced luteolysis [32], and a decrease in ovine TIMP-1 protein was observed within 6 h of a pharmacologic dose of PGF2 [18]. It has also been reported that TIMP-2 mRNA levels decrease significantly in sheep CL during the late luteal phase [15]. In the present study, however, the coordinate decline of both TIMP-1 and -2 proteins, followed by an increase in MMP-2 activity, resulted in a pronounced imbalance between these proteins that most likely would favor enzyme activity. Indeed, recent evidence suggests that low levels of TIMP-2 may serve to enhance MMP-2 activity, and TIMP-2 is often found in association with latent (pro) MMP-2, selectively bound at the noncatalytic ends (C-terminus) of each molecule [33]. The proMMP-2, complexed with TIMP-2 in this manner, can be activated by membrane type 1-MMP [34, 35]. Therefore, a low-level, stoichiometric amount of TIMP-2 may be necessary to form the molecular complexes that lead to cell surface MMP-2 activation [34], promoting proteolysis of luteal ECM.
Contrary to the findings that TIMP mRNA and protein decrease after the onset of luteolysis, Juengel et al. [13] reported that both TIMP-1 and -2 mRNA from bovine CL increased significantly from control levels 8 h after a pharmacological dose of PGF2 (25 mg of Lutalyse). The differences in the responses between studies may be the result of differences in the dosage levels of PGF2 or of species variation. Another possibility is that PGF2 may not immediately induce negative changes in the synthesis of TIMP proteins. Instead, physiologic conditions set in motion by PGF2 (e.g., formation of oxygen radicals) may degrade TIMP proteins already present in the cellular milieu.
In the present study, the significant drop in TIMP-1 and -2 proteins immediately following the infusion period leads us to speculate regarding the reason for such a sharp and rapid decline. A study by Frears et al. [36] reported that peroxynitrite (ONOO-) was capable of inactivating TIMP-1, thereby potentiating MMP-mediated ECM degradation. Peroxynitrite is formed by the reaction of nitric oxide (NO) with the superoxide anion (
). A gaseous radical, NO, is widely distributed throughout various tissues of the body and can be produced by macrophages as well [37]. Coincidentally, invasion of immune cells is reported to be an early event in the luteolytic process [38]. Also, NO is synthesized by the enzyme NO synthase (NOS), which has the ability to catalyze production. The subsequent generation of the potent radical ONOO- may catabolize TIMPs as well as other biological molecules, thereby damaging them [39].
Of additional interest, NO stimulates PGF2 production in cultured bovine luteal cells [40]. In turn, PGF2 up-regulates NOS activity in cultured rabbit luteal cells [41]. Estevez et al. [42] postulate that a positive feedback mechanism between PGF2 and NO exists to ensure luteal regression in the rat. In fact, intraluteal administration of an inhibitor of NOS (N-
-nitro-L-arginine methyl ester) prolongs the life span of the bovine CL [43]. Therefore, the effects of both NO and PGF2 could mediate the sharp decline in TIMPs seen during the present study. Further investigation of this system in the CL is needed to provide additional information.
Zymography results showed no change in proMMP-2 as a result of PGF2 infusion. This is consistent with information that most cell types responsible for producing MMP-2 do so in a largely constitutive manner [6]. However, a statistically significant increase (165% of controls) was observed in the active form of MMP-2, which is consistent with observations reported in the rat CL [11]. The MMP-2 has the ability to degrade basement membrane (collagen type IV), which is a well-known constituent of bovine and ovine CL [8]. Therefore, the increase in active MMP-2 may enhance the structurally regressive processes observed during luteolysis. Furthermore, in the present study, no significant change was observed in the expression of latent MMP-9 due to PGF2 treatment. Although MMP-9 has been characterized in the late-phase CL of the human [32], pig [44], and cow [23], it has not yet been assessed in these species after the specific induction of luteolysis by PGF2. Therefore, to our knowledge, the detection of MMP-9 by zymography in the regressing sheep CL during the present study is a novel finding. Its function and role, however, will require further study.
The relationship between PGF2, regulators of the ECM (MMPs and TIMPs), and P levels is complex. The ECM proteins and the cytoskeleton are connected via transmembrane proteins called integrins [45, 46]. When TIMP levels fall, active MMPs may disrupt ECM-integrin relationships, thus affecting cytoskeleton stability. Indeed, following PGF2 treatment, this may be why disintegration occurred in the ovine luteal cell cytoskeleton [16], the stability of which is believed to be critical for the intracellular transport of substrate molecules (e.g., cholesterol, pregnenolone) used for steroidogenesis [47]. In this connection, Puthalakath et al. [48] have shown that disruption of the ECM and its associated integrin signaling system could promote apoptosis in luteal cells by the newly described sensor Bmf, which appears to couple integrin-linked alterations in the cytoskeleton to the activation of apoptosis. Furthermore, an association exists between TIMP-1 and steroid biosynthesis. The steroid acute regulatory protein (StAR) and TIMP-1 have a 124-base pair homology in their nucleotide sequences [49]. In follicular fluid, increased P levels correlate to increased levels of TIMP-1 [50], which is consistent with the inhibitor's ability to stimulate steroid production [51]. In contrast, gelatinase (MMP-2 and -9) and collagenase activities are inhibited by P [52]. Thus, the early changes in the regulators of the ECM observed during the present study could play a critical role in mediating the process of functional luteolysis, such as by affecting membrane/receptor conformation or other pathways that lead to altered P production by luteal cells.
In summary, after a 1-h, systemic pulse of PGF2, luteal TIMP proteins drop acutely at 1 h after the start of infusion, followed by an increase in MMP-2 activity that coincided with the nadir in plasma P levels at 8 h. Thus, in sheep, it appears that structural luteolysis, as demonstrated by a marked change in known regulators of the ECM, may actually occur before the onset of functional luteolysis, as manifested by a later decline in plasma P.
|
FOOTNOTES |
|---|
1 Supported in part by Northeast Regional Project NE-161 to P.C.W.T. and USDA grant 98–35203-6635 to J.A.M.C. This is scientific contribution 2101 from the New Hampshire Agricultural Experiment Station. 
2 Correspondence: Paul C.W. Tsang, University of New Hampshire, Department of Animal & Nutritional Sciences, 129 Main Street, Kendall Hall, Durham, NH 03824. FAX: 603 862 3758; pct{at}cisunix.unh.edu
Accepted: December 12, 2001.
Received: July 10, 2001.
|
REFERENCES |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
identified as a luteolytic hormone in the sheep. Nat New Biol 1972; 238:129-134[CrossRef][Medline] from the ovine uterus. Endocrinology 1974; 95:1340-1349[Medline]. Biol Reprod 1997; 56:(suppl 1):197 (abstract) and the timing of events associated with luteolysis in ewes with estrous cycles of different lengths. J Reprod Fertil 1988; 83:517-526[Abstract] early in the bovine estrous cycle. J Reprod Fertil 1980; 59:393-396[Abstract] synthesis and release in the ewe during the initial establishment of pregnancy. J Reprod Fertil 1988; 83:527-536[Abstract] and progesterone during luteolysis in the rat. Medicina (B Aires) 1999; 59:463-465[Medline]This article has been cited by other articles:
|
|
 |
|
  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 |
