| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
Articles |
a Clinical and Experimental Endocrinology, Department of Obstetrics and Gynecology, University of Göttingen, 37075 Göttingen, Germany
 |
ABSTRACT |
|---|
 TOP ABSTRACT INTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
(PGF2) and tumor necrosis factor (TNF), we studied their effects on LLC under culture conditions. Treatment of cells with PGF2 or TNF (10-7 M or 3 x 10-9 M, respectively) induced a significantly higher release of MMPs, and gene expression was also significantly stimulated in comparison to that in untreated LLC. The gene expression of TIMPs remained unaffected by either treatment. It is concluded that at the beginning of luteolysis, MMPs are expressed and released in high amounts and that this is essential for the structural regression of the CL.
|
INTRODUCTION |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
Little is known about the MMP and TIMP gene expression during the course of the luteal phase in the pig. Therefore, we studied expression of their genes in CL obtained during the early, mid, and late luteal phases.
It is well known that steroidogenic cells of the CL consist of two different populations: the so-called small (SLC) and large luteal cells (LLC) [19, 20]. The LLC are luteinized granulosa cells, and they can be easily separated from other cell types by gravity methods [21, 22].
Luteolysis in the pig requires an intact uterus, and it is established that endometrial prostaglandin F2 (PGF2) initiates luteolysis [23]. Furthermore, it was shown that PGF2-induced luteolysis is accompanied by invasion of macrophages into the CL [24]. Macrophages are known to produce tumor necrosis factor (TNF) [25], and we showed recently that TNF makes the CL extremely susceptive to the luteolytic action of PGF2 [26]. Therefore, we also studied the effects of TNF and PGF2 on MMP and TIMP gene expression in LLC. In addition, the secretion of the three MMPs into the culture media was determined.
In summary, this study was initiated to test whether structural luteolysis of CL may involve MMPs and their TIMPs and whether regulatory effects of luteolysins PGF2 and TNF on gene expression and release of MMPs can be demonstrated.
|
MATERIALS AND METHODS |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
(10-7 M final dilution) or with TNF (3 x 10-9 M final dilution). After an additional incubation period of 48 h, the media of LLC cultures were collected and stored at -20°C for measuring MMP and progesterone secretion. Cultured LLC were subsequently harvested and used for RNA extraction. Reverse Transcription-Polymerase Chain Reaction(RT-PCR) and Cloning and Sequencing of PorcineMMP-1, MMP-2, MMP-9, TIMP-1, and TIMP-2 of LutealTissue and LLC
The RNA of intact luteal tissue and of cultured LLC (of at least three independent experiments with 6 wells per treatment) was isolated by the guanidinium thiocyanate-phenol-chloroform extraction method [28]. When cell cultures were finished, the media were removed and cells were lysed in 0.6 ml lysis buffer (4 M guanidinium thiocyanate, 25 mM sodium tricitrate, pH 7.0; 0.5% sodium-laurylsarcosine; 0.1 ml mercaptoethanol) and sonicated for 15 sec. This solution was transferred into a 1.5-ml reaction tube; and 60 µl 2 M sodium acetate, 0.6 ml phenol solution (pH 4.0), and 100 µl chloroform-isoamyl alcohol (24:1) were added. After mixing and incubation on ice for 15 min, the tubes were centrifuged to separate the different phases. The RNA-containing upper phase was transferred into a new reaction tube that contained 600 µl isopropanol for RNA precipitation at -20°C overnight. After centrifugation at 10 000 x g for 20 min, the RNA pellet was washed twice with 70% ethanol, dried, and resuspended in 20 µl RNase-free distilled water. The concentration and purity of the RNA were determined photometrically. An aliquot of total RNA (40 ng) was reverse transcribed by 200 U reverse transcriptase using Superscript Preamplification System (Gibco BRL, Karlsruhe, Germany). Twenty microliters of the reaction mixture contained single-strength RT buffer (50 mM Tris-HCl, pH 8.3; 75 mM KCl; 3 mM MgCl2; 0.01% gelatin) and 1 mM dNTP. The reaction was carried out at 22°C for 10 min, followed by a 50-min period at 42°C and a 10-min period at 95°C. One microliter of the resulting cDNA was subsequently amplified with 2.5 U Taq DNA polymerase (Gibco BRL) in 100 µl master mix containing double-strength PCR buffer (10 mM Tris-HCl, pH 7.5; 50 mM KCl; 1.5 mM MgCl2; 0.01% gelatin), 0.25 mM dNTP, and the appropriate primers (50 pmol each). PCR was carried out for 19–34 cycles in an automated thermocycler (Bio-metra, Göttingen, Germany) with the following profile: denaturation at 94°C for 1 min, primer annealing at 50–54°C for 1 min, primer extension at 72°C for 2 min. Ten microliters of resulting samples was brought onto a 1.5% agarose gel; and after electrophoresis in Tris-borate-ETDA buffer, the DNA fragment was stained with ethidium bromide and photographed under UV illumination. The negative film was used for densitometric scanning of the DNA signal.
For each RT-PCR, 40 ng of total RNA was reverse transcribed and the linear part of the amplification curve determined. The resulting cycle numbers were 27 for MMP-1, 25 for MMP-2, 29 for MMP-9, 19 for TIMP-1, and 34 for TIMP-2. For each sample the ribosomal protein L7 was also amplified (24 cycles), and all values are expressed in relation to the housekeeping gene L7.
MMP-1 primers were selected from porcine MMP-1 sequence [29] resulting in a 424-base pair (bp) fragment. The upstream and downstream primers were 5'-TGA TGA AGA TGA AAG GTG-3' and 5'-ATC TCT ATC GGC AAT CTC-3'. MMP-2 primers were based on the mouse and human mRNA sequence [30, 31]. Upstream primer was 5'-ATG ATG GGG AGG CTG ACA-3' and downstream primer was 5'-GGA AGC GGA ACG GAA ACT-3' with a predicted size of 405 bp. Primers for MMP-9 were selected from mouse and bovine MMP-9 mRNA [32, 33] with a predicted size of 452 bp. The sequences of upstream and downstream primer were GGC ACC ACC ACA ACA TCA and GCG GTC GGC GTC GTA GTC. Primers for TIMP-1 were chosen from the human cDNA sequence [34]. The upstream primer was 5'-GCT TCT GGC ATC CTG TTG TTG-3' and downstream primer was 5'-GTC CGT CCA CAA GCA ATG AGT-3' with a predicted size of 492 bp. The upstream and downstream primers for TIMP-2 were synthesized based on the human and bovine cDNA sequence [35, 36] and were 5'-TTA TGG CAA CCC TAT CAA-3' and 5'-ACA GGA GCC GTC ACT TCT-3', respectively. The resulting DNA fragment had a predicted size of 421 bp. The housekeeping gene L7 [37] was determined by the following primers: 5'-AGA TGT ACA GAA CTG AAA TTC-3' and 5'-ATT TAC CAA GAG ATC GAC CAA-3' with a size of 353 bp.
The PCR fragments from MMPs and their inhibitors were cloned into pGEM-4Z vector (Promega, Madison, WI) and sequenced with the A.L.F.-sequencing system (Pharmacia, Freiburg, Germany) to confirm their identity.
Determination of MMP and Progesterone Secretion
MMP-1 release was determined by a "sandwich" ELISA (Amersham, Little Chalfont, England) using specific mouse anti-MMP-1, which did not cross-react with MMP-2 and MMP-9. This assay allows a more sensitive determination of MMP-1 than the zymographic assay. The method of zymography [3, 18] can be used to determine both gelatinases MMP-2 (72 kDa) and MMP-9 (92 kDa). Zymographic analysis was carried out by electrophoresis in 10% (w:v) SDS-polyacrylamide gels impregnated with 0.15% gelatin (w:v). One part of the culture medium was mixed with one part of nonreducing sample buffer (12.5 ml 1 M Tris-HCl, pH 6.8, 4.6 g SDS, 2 ml bromophenol blue, 20 ml glycerol mixed together with distilled water to 100 ml). Ten microliters of sample was applied to the gel, then run with Tris-glycine SDS-running buffer (30 g Tris, 144 g glycine, 10 g SDS to 10 L with distilled water, pH 8.6) at 125 volts until the marker dye reached the bottom of the gel. Thereafter the gel was incubated in renaturing buffer (2.5% Triton X-100 in 100 ml distilled water) for 30 min at room temperature. The renaturing buffer was decanted and replaced with 100 ml developing buffer (50 mM Tris-HCl, pH 7.6, 0.2 M NaCl, 5 mM CaCl2, 0.02% Brij). After equilibration for 30 min with gentle agitation, the buffer was replaced with fresh developing buffer followed by overnight incubation at 37°C. The gel was then stained with Coomassie Blue solution (6.25 g Coomassie Blue, 1.13 L methanol, 230 ml acetic acid with distilled water to 2.5 L) for 10 min followed by incubation (20 min) with destaining solution (1 L methanol, 1.5 L acetic acid with distilled water to 20 L). Quantification of MMP-2 and MMP-9 was achieved by computerized image analysis using two-dimensional scanning densitometry (Scan-Pak; Biometra). Molecular weights of collagenolytic enzymes were determined by comparison with SDS-PAGE molecular weight markers (Bio-Rad Laboratories, Hercules, CA) run in an adjacent lane.
Progesterone secretion by LLC was measured in culture media by a specific, commercial ELISA without extraction (Enzymun-Test, Progesterone; Boehringer-Mannheim, Mannheim, Germany). In brief, 3 µl of culture supernatant was diluted with 300 µl of human hormone-free serum and determined automatically in an ES 700 autoanalyzer (Boehringer-Mannheim). No influence of the culture media on the assay system was found under this condition.
Statistical Analysis
As there were variations from different tissue and cell preparations, the mean values of all tested parameters obtained from tissue or cells collected during the early luteal phase of each experiment were set as 100%, and the other time or treatment points were calculated in relation to this 100% value. This allows calculation of percentage related means and standard errors as well as statistical treatment of these data. To assess the statistical significance of treatment effects, one-way ANOVA was performed, followed by Dunnett's test for multiple comparison (according to the Graph Pad Prism, version 2; Graph Pad Software, San Diego, CA). P < 0.05 was considered significant.
|
RESULTS |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
Gene expression of the three MMPs and the two TIMPs in luteal tissue obtained at various stages of the estrous cycle is shown in Figure 1. Expression of all three MMP genes increased significantly (P < 0.05) with the age of the CL, while the expression of the TIMP genes showed an inverse pattern with lowest values in tissue of regressing CL (P < 0.05). Expression of the housekeeping gene L-7 did not change significantly.
|
MMP-1, MMP-2, and MMP-9 Secretion by Cultured LLC
The capacity of porcine LLC to secrete MMP-1 into the culture medium was measured by ELISA technique using specific mouse anti-MMP-1, while the simultaneous release of MMP-2 (72 kDa) and MMP-9 (92 kDa) was quantified by gelatin substrate zymography. The latter method allows visualization and quantification of collagenases by their gelatin-degrading activities in conditioned medium of porcine luteal cells (Fig. 2). When the LLC culture medium was serially diluted, the gelatinolytic bands became linearly weaker, indicating the validity of this method. Only a faint band of gelatinolytic activity was observed in unconditioned medium, which was probably due to the presence of 5% FCS.
|
Immunoreactive MMP-1 (Fig. 3, top) and bioactive MMP-2 and MMP-9 release (Fig. 3, bottom) from LLC increased as a function of age of the CL; i.e., significantly (P < 0.05) higher amounts of immunoreactive MMP-1 and bioactive MMP-2 and MMP-9 were released from cells prepared from aged (regressing) in comparison to young CL. The middle part of Figure 3 shows different zymographic activities in culture media from LLC harvested at different times of the luteal phase.
|
The release of progesterone by the LLC showed an inverse pattern to MMP secretion (Fig. 4); i.e., progesterone was highest in culture media in which LLC prepared from young CL were cultivated and lowest (P < 0.05) in culture media in which LLC stemming from regressing CL were cultivated.
|
Figure 5 details the effects of PGF2 and TNF treatment on progesterone release (top) and on MMP-1, MMP-2, MMP-9, TIMP-1, and TIMP-2 gene expression (bottom) from LLC prepared from CL of the midluteal phase. While both compounds inhibited progesterone release significantly (P < 0.05), they stimulated the expression of the three MMP genes, an effect that was also significant (P < 0.05). In contrast, no up-regulation was observed on TIMP and on L7 gene expression.
|
Figure 6 details the effects of PGF2 and TNF on MMP-1 release (as determined by ELISA) and on MMP-2 and MMP-9 release (as determined by zymography). Both the prostaglandin and the cytokine stimulated the release of MMP significantly (P < 0.05).
|
To verify the identity of PCR products, the DNA fragments were cloned and sequenced. The amino acid sequence of the MMP-1 fragment was 100% identical to the predicted porcine MMP-1 protein, while the porcine MMP-2 and MMP-9 fragments were 96% and 92%, respectively, identical to the reported human MMP-2 and mouse MMP-9 amino acid sequences. The protein structure of the porcine TIMP-1 fragment had a 89% homology to the human TIMP-1 amino acid sequence, while the protein sequence of the TIMP-2 DNA fragment was 98% identical to the mouse TIMP-2.
|
DISCUSSION |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
The present results demonstrate for the first time the gene expression of MMP-1, MMP-2, and MMP-9 as well as of TIMP-1 and TIMP-2 in porcine luteal tissue. While gene expression of all three MMPs increased from the early to the late luteal phase, mRNA levels of TIMP-1 and TIMP-2 decreased continuously to be lowest during the time of luteal regression. These results were corroborated by our cell culture experiments in which an analogous release pattern for MMP-1, MMP-2, and MMP-9 was observed. In contrast to experiments from intact luteal tissue, which contains various cell types, the use of cell cultures allows the demonstration of MMP release from the steroidogenic LLC as well as the release of progesterone as an indicator of functional cell integrity. As shown by gel electrophoresis under nondenaturating conditions followed by zymography, gelatinolytic activity of MMP-2 and MMP-9 was present with the expected molecular masses of 72 kDa and 92 kDa. The MMP-1 release was determined by ELISA technique because the gelatinolytic activity of this MMP is too low to allow the development of a zymographic assay. In contrast to the secretion rates of MMPs, we demonstrated high LLC-progesterone release during the early and midluteal phase that decreased when LLC were prepared from CL of the late luteal phase. The release pattern of the three MMPs supports the hypothesis that increasing amounts of collagenolytic activities may play an important role in structural remodeling that takes place during the regression of the CL. MMP-1 and both gelatinases are also expressed in human CL, in which simultaneous determination of MMP-2 and MMP-9 release by zymography demonstrated increasing secretion rates of gelatinases from the midluteal to the late luteal phase [15]. Unfortunately, there is no sensitive bioassay to measure TIMP activities in the culture medium; reverse zymography [7, 40] proved to be too insensitive. The rate of expression of the TIMP-1 and TIMP-2 genes, however, may indicate that the synthesis rates and possibly also the release rates of both TIMPs are lowest at a time when the CL undergoes structural luteolysis. Although increasing or decreasing expression of genes does not necessarily reflect the translational capacity of cells to synthesize and to posttranslationally process the bioactive product, the parallelism between gene expression of all three MMPs with the release rates of the respective product may indirectly support the conclusion that the reduced expression of the two TIMP genes also indicates less production and release of these two inhibitors. As a net effect of increasing MMP and decreasing TIMP release, the proteolytic activities of the three MMPs would become dominant during luteolysis. In agreement with our findings, previous reports have suggested that TIMP-1 [17] and TIMP-2 [16] gene expression is up-regulated at the early stage of luteal development and that their mRNAs are present in LLC. The physiological function of TIMP-1 and TIMP-2 throughout the life span of CL is unclear at present. There has been discussion that maximal expression of TIMPs during the early luteal phase may be associated with the process of angiogenesis during CL development [16, 17].
There is overwhelming evidence that PGF2 of uterine origin is necessary to induce luteolysis in the pig [23, 26]. As demonstrated earlier [22] and confirmed in this study, PGF2 indeed reduces progesterone secretion from porcine LLC. Recently, we demonstrated that the capability of PGF2 to induce luteolysis is largely enhanced by preexposure of the CL to TNF [26]. In the present experiment we demonstrate again that TNF per se reduces progesterone production. Hence, macrophages that produce large amounts of TNF [25] and that invade the CL at the time of luteolysis [24] appear to be also involved in functional luteolysis. In the present experiment we demonstrate that in cultured LLC, both the eicosanoid and the cytokine also stimulate the gene expression and the secretion of the three studied MMPs, which for the first time demonstrates that both compounds may also be involved in structural luteolysis. It is interesting, however, to note that gene expression of TIMP-1 and TIMP-2 was not significantly altered by PGF2 or TNF. However, in luteal tissue obtained during the time of luteolysis, TIMP-1 and TIMP-2 gene expression was significantly reduced, indicating that factors other than PGF2 and TNF may be involved in regulating TIMP-1 and TIMP-2 gene expression.
The present experiments were also performed to analyze which cell type might express the MMP and TIMP genes. Percoll sedimentation and separation techniques did not yield a 100% pure cell population, although the LLC preparation was only slightly contaminated (< 8%) with SLC. Stamouli et al. [39] reported suppressed MMP-2 and MMP-9 in human luteinized granulosa cells, which are the progenitor cells of LLC. Furthermore, Smith et al. found gene expression of both TIMP-1 [17] and TIMP-2 [16] in bovine and ovine LLC, respectively. Since LLC are also estrogen and progesterone receptive [37] and since changes in the gene expression of MMPs and TIMPs occur that vary with the stage of the luteal phase, it is possible that the observed changes in MMP secretion and in the gene expression of MMPs and TIMPs are regulated by the two luteal steroids progesterone and estradiol.
Whether or not steroidogenic SLC also produce MMPs or TIMPs could not be addressed in these experiments because attempts to separate SLC from other nonsteroidogenic cells yielded too high contamination with endothelial cells and lymphocytes.
Outside of the ovary, endothelial cells and cells of the white blood cell line, particularly macrophages, were also shown to produce MMPs and TIMPs [40–42]; hence, within the CL, not only the steroidogenic cells but also nonsteroidogenic cells could be a source of these enzymes and their inhibitors.
In summary, we showed increased MMP-1, MMP-2, and MMP-9 and decreased TIMP-1 and TIMP-2 gene expression in regressing porcine CL. LLC harvested from these regressing CL secreted significantly more MMP-1, MMP-2, and MMP-9 in comparison to LLC from young or middle-aged CL. We demonstrated further that the luteolytic substances PGF2 and TNF stimulated the gene expression and the release of all three MMPs from LLC kept under culture conditions while TIMP gene expression was unaffected by these treatments. The release pattern and gene expression of the three MMPs and their TIMPs allow one to hypothesize that structural remodeling processes that take place during luteolysis are at least in part due to activation of the MMPs and inhibition of the respective TIMPs.
|
FOOTNOTES |
|---|
1 This study was generously funded by the German Research Society (DFG Grant No. Wu 60/10-3). 
2 Correspondence: L. Pitzel, Division of Clinical and Experimental Endocrinology, Department of Obstetrics and Gynecology, University of Göttingen, Robert-Koch-Strasse 40, D-37075 Göttingen/Germany. FAX: 551 396518; ufkendo{at}med.uni-goettingen.de
Accepted: December 1, 1999.
Received: September 23, 1999.
|
REFERENCES |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
-macroglobulin interactions. J Biol Chem 1989; 264:393–401., oxytocin, and cytokines on steroidogenesis of porcine luteal cells. Endocrinology 1993; 132:751–756.[Abstract] as the luteolysin in swine: I. Effect of prostaglandin F2 in hysterectomized gilts. Prostaglandins 1976; 11:737–743.[CrossRef][Medline]-induced luteolysis. Biol Reprod 1994; 50:10–15.[Abstract] and tumor necrosis factor to induce luteolysis in the pig. Biol Reprod 1998; 58:1310–1315.This article has been cited by other articles:
|
|
 |
|
  M. Myers, M. C. Lamont, S. van den Driesche, N. Mary, K. J. Thong, S. G. Hillier, and W. C. Duncan Role of Luteal Glucocorticoid Metabolism during Maternal Recognition of Pregnancy in Women Endocrinology, December 1, 2007; 148(12): 5769 - 5779. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
M. Myers, E. Gay, A. S. McNeilly, H. M. Fraser, and W. C. Duncan In Vitro Evidence Suggests Activin-A May Promote Tissue Remodeling Associated with Human Luteolysis Endocrinology, August 1, 2007; 148(8): 3730 - 3739. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 K. A. Young and R. L. Stouffer Gonadotropin and Steroid Regulation of Matrix Metalloproteinases and Their Endogenous Tissue Inhibitors in the Developed Corpus Luteum of the Rhesus Monkey During the Menstrual Cycle Biol Reprod, January 1, 2004; 70(1): 244 - 252. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 T. E. Curry Jr. and K. G. Osteen The Matrix Metalloproteinase System: Changes, Regulation, and Impact throughout the Ovarian and Uterine Reproductive Cycle Endocr. Rev., August 1, 2003; 24(4): 428 - 465. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
T. E. Curry Jr and S. E. Wheeler Cellular Localization of Tissue Inhibitors of Metalloproteinases in the Rat Ovary Throughout Pseudopregnancy Biol Reprod, December 1, 2002; 67(6): 1943 - 1951. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 K.A. Young, J.D. Hennebold, and R.L. Stouffer Dynamic expression of mRNAs and proteins for matrix metalloproteinases and their tissue inhibitors in the primate corpus luteum during the menstrual cycle Mol. Hum. Reprod., September 1, 2002; 8(9): 833 - 840. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
B. Zhang, L. Yan, M. A. Moses, and P. C.W. Tsang Bovine Membrane-Type 1 Matrix Metalloproteinase: Molecular Cloning and Expression in the Corpus Luteum Biol Reprod, July 1, 2002; 67(1): 99 - 106. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
T. A. Towle, P. C.W. Tsang, R. A. Milvae, M. K. Newbury, and J. A. McCracken Dynamic In Vivo Changes in Tissue Inhibitors of Metalloproteinases 1 and 2, and Matrix Metalloproteinases 2 and 9, During Prostaglandin F2{alpha}-Induced Luteolysis in Sheep Biol Reprod, May 1, 2002; 66(5): 1515 - 1521. [Abstract] [Full Text]
|
|
|
|
|
|
J. A. Lenhart, P. L. Ryan, K. M. Ohleth, S. S. Palmer, and C. A. Bagnell Relaxin Increases Secretion of Tissue Inhibitor of Matrix Metalloproteinase-1 and -2 during Uterine and Cervical Growth and Remodeling in the Pig Endocrinology, January 1, 2002; 143(1): 91 - 98. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
E. Shalev, S. Goldman, and I. Ben-Shlomo The balance between MMP-9 and MMP-2 and their tissue inhibitor (TIMP)-1 in luteinized granulosa cells: comparison between women with PCOS and normal ovulatory women Mol. Hum. Reprod., April 1, 2001; 7(4): 325 - 331. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
