| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
Articles |
a INSERM U460, CHU Xavier Bichat, 75870 Paris Cedex, France
b Laboratoire Cassenne-Hoechst Marion Roussel, Paris La Défense Cedex, France
 |
ABSTRACT |
|---|
 TOP ABSTRACT INTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
|
INTRODUCTION |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
MMPs have different specificities, although there is considerable overlap, and together they can degrade most components of the extracellular matrix (ECM). The family comprises interstitial collagenases (such as MMP-1, which degrades collagen types I, II, and III), gelatinases (MMP-2 and MMP-9, which degrade collagen types IV, V, and X), stromelysins (MMP-3, -7, and -10, degrading types IV and IX collagens, laminin, fibronectin, elastin, and proteoglycans), and membrane-type MMPs (MMP-14, -15, and -16). MMPs are secreted as latent proenzymes (pro-MMPs) that undergo activation to a catalytically active form and are inhibited either by specific tissue inhibitors of metalloproteinases (TIMPs), or less specifically, by 
2 macroglobulin [12]. TIMP-1 binds to and inactivates MMP-1, -2, -3, and -9, while TIMP-2 binds to active forms of the same enzymes but also to the latent form of MMP-2 [13, 14]. The genes for MMPs and TIMPs are regulated by a number of factors, including steroid hormones, growth factors, and cytokines, with variations between MMPs, tissues, and cell types [12–15]. In vitro, MMPs can be activated by a number of proteases, including plasmin, MMP-3 (stromelysin-1), and the membrane type MT-MMP, or by treatment with organomercurial compounds. Most of these enzymes are expressed at low levels in adult tissues, and many are up-regulated during normal and pathological remodeling processes such as embryonic development, bone resorption, tissue repair, inflammation, and tumor invasion [12, 16, 17]. In particular, they are involved in several key reproductive events such as ovulation, embryo implantation, menstruation, and postpartum uterine involution [6–11, 18].
MMPs have been found in human endometrium, endometrial explants, and stromal cell cultures. The endometrial expression of mRNA for the various MMPs varies in cycle-specific patterns [7, 11, 19–21]. Some are not expressed during the progesterone-dominated secretory phase of the cycle [7, 19]; in addition, progesterone withdrawal increases production of pro-MMP2 and pro-MMP3 by decidualized stromal cells [22, 23] and activates endometrial proteases MMP-1, -2, -3, and -9 [11, 24, 25]. MMP-2 expression is also regulated by tumor necrosis factor- and interleukin-1 [26], suggesting a multifactorial control of metalloproteinase activity at menstruation. In contrast, there is relatively little cyclical variability of the tissue inhibitors TIMP-1 and TIMP-2 [27]; their presence was recently observed in all cellular compartments of the endometrium, including vascular cells, suggesting their involvement in maintaining vascular integrity [27]. However, although vascular changes, such as angiogenesis and growth of spiral arteries, are dependent upon sex steroids [28], the production of both MMPs and TIMPs in the human endometrial vessels and their possible role in menstruation and vascular remodeling have not been investigated in detail.
The present study was therefore undertaken to compare the localization and modulation of various MMPs (MMP-1, -2, -3, and -9) and of their inhibitors, TIMP-1 and -2, in individual biopsies throughout the menstrual cycle using immunohistochemistry on adjacent sections; special attention was given to the vascular structures of the functionalis. We show that, in addition to glandular, stromal, and hematic cells, vascular structures express several MMPs in a stage-dependent fashion. In particular, MMP-2 is expressed in newly formed capillary strands and is probably associated with angiogenesis; MMP-9 is found in developing arterioles; and MMP-2, -3, and -9 are present in vessels during menstruation. The interplay of MMPs and TIMPs in various endometrial structures is also discussed.
|
MATERIALS AND METHODS |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
Reagents for cell culture were from Gibco (Life Technologies, Cergy-Pontoise, France). Estradiol and progesterone were from Sigma Chemical Co. (St. Louis, MO). ICI 182780 was provided by Zeneca (Zeneca Pharmaceuticals, Cheshire, England). RU-486 (mifepristone) was provided by Roussel-UCLAF (Romainville, France). The MMP-2 cDNA was provided by the American Type Culture Collection (Rockville, MD), and the TIMP-2 cDNA was a gift from Dr. Dylan Edwards (Department Pharmacology and Therapeutics, University of Calgary, AB, Canada).
Materials
Endometrial biopsies were obtained from 33 consenting cycling women (aged 27–44 yr) [29]. Patients with endometrial pathology or those who had received steroid treatment were excluded, as were women receiving ovulation-induction treatment. Specimens of endometrium were obtained in the proliferative (n = 14: 7 mid, 7 late), secretory (n = 15: 7 early, 5 mid, 3 late), and menstrual (n = 4: 2 early, 2 late) phases of the cycle; the dates were confirmed by independent histological examination [30]. The biopsies were quickly frozen in isopentane precooled in liquid nitrogen and used for immunocytochemistry. Endometrial tissue obtained from premenopausal normally cycling women undergoing hysterectomy for benign nonendometrial pathology (i.e., fibroid uterus) was also used; it was either frozen for further RNA and protein analysis or immediately placed into Dulbecco's modified Eagle's medium (DMEM, containing 10% fetal calf serum [FCS], 100 U/ml penicillin, 0.1 mg/ml streptomycin, and 0.025 mg/ml fungizone) and processed for stromal cell preparation [28].
Immunocytochemistry
Cryostat sections (6-µm) were fixed in acetone at -20°C for 5 min. Immunocytochemical detection of MMPs was performed on frozen sections as previously described [28], using specific antibodies (Oncogene Sciences, Uniondale, NY). MMP-1 (clone 41-1E5), MMP-2 (clone 42-5D11), MMP-3 (clone 55-2A4), MMP-9 (clone 56-2A4), and inhibitors TIMP-1 (clone 7-6C1) and TIMP-2 (clone T2-101) have been fully characterized [31–36]. Western blot analysis of conditioned medium (20 µg of proteins) obtained from stromal cells incubated with both estradiol and progesterone confirmed the specificities of the antibodies used (presence of bands corresponding to proteins of Mr [x 10-3] 66–72, 28, and 21 related to MMP-2, TIMP-1, and TIMP-2, respectively). Antibodies anti-MMP-2 and anti-TIMP-1 from ANAWA (Zurich, Switzerland [37]) were used to corroborate some of the results. The immunocytochemical process included incubation overnight with affinity-purified mouse (5–20 µg/ml) or rabbit polyclonal antibodies (diluted 1:40) followed by incubation with biotinylated anti-mouse or anti-rabbit IgG and streptavidin-biotin peroxidase (Dakopatts, Glostrup, Denmark). Peroxidase reaction was performed using amino-ethylcarbazole substrate as previously described [28]. Some sections were lightly counterstained with Mayer's hematoxylin. Immunofluorescent detection of the same MMPs or TIMPs in stromal cells was performed using the same antibodies, as previously described [38].
Specificities of the immunocytochemical reactions were checked by incubating sections with irrelevant monoclonal or rabbit IgGs and by omission of the first antibody. Preabsorption of anti-TIMP-1 antibody for 12 h at 4°C with increasing amounts of purified recombinant TIMP-1 (5–20 µg/ml diluted antibody) before immunostaining was also performed, as an additional specificity control.
Adjacent sections were incubated with a polyclonal anti-Von Willebrand factor antibody (Dakopatts), which is a marker of vascular endothelial cells, or with an anti-smooth -actin antibody (Sigma) to identify vascular smooth muscle cells [28]. CD 34 monoclonal antibody was used both as a control of nonspecific labeling (for the monoclonal antibodies) and as an endothelial cell marker.
To perform semiquantitative evaluation, the relative intensity of the immunoreaction product was graded blindly with a light microscope (Zeiss, Oberkochen, Germany) at x100 and x200 magnification by 3 independent observers (M.P-A, S.F, and G.M). Semiquantitative evaluation categories, as previously described [28, 39], were as follows: -, negative; +, faint; ++, moderate; and +++, strong.
Isolation of Human Endometrial Cells
Isolation of stromal cells was performed as previously described [28]. Briefly, endometrial tissue was cut into small pieces and incubated for 45 min at 37°C in DMEM containing 0.1% collagenase (type 1) and 0.02% DNase I (Sigma). Nondissociated cells were discarded by filtration through nylon (100-µm and 35-µm pore size), and the cell suspension was centrifuged (400 x g). Cells were resuspended into DMEM/10% FCS and plated on Petri dishes at 37°C in a 95% air, 5% CO2 humidified atmosphere. After 30 min, unattached cells were discarded. Purity of the stromal cell preparation obtained at passage 3–4 was verified, as previously described [28], by staining with anti-vimentin antibody (clone V9; Dakopatts) and anti-leukocyte common antigen (Dakopatts).
Extraction of RNA from Cells and Northern Blot Analysis
Total RNA was isolated from the endometrium or from cultures of stromal cells using a modified guanidium isothiocyanate method (Trizol; Gibco BRL, Gaithersburg, MD) [40]. Messenger RNA for MMP-2, TIMP-1, and TIMP-2 were detected by Northern blot using specific human cDNA probes. Total RNA (20 µg) was size fractionated in formaldehyde-agarose (1%) gels and transferred to membranes (Hybond; Amersham Pharmacia Biotech, Piscataway, NJ). Prehybridization and hybridization were carried out in 5-strength SSC (single-strength SSC is 0.15 M sodium chloride and 0.015 M sodium citrate), 5-strength Denhardt's, 50% formamide, 0.1% SDS, and 100 ng/ml salmon sperm DNA. Radioactive labeling of the probes was performed using the random priming method. Glyceraldehyde phosphate dehydrogenase RNA was used to confirm equal RNA loading. Posthybridization washes were carried out (two times, 30 min each) with double-strength SSC + 0.1% SDS at 52°C and 0.1-strength SSC + 0.1% SDS at room temperature, followed by autoradiography.
|
RESULTS |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
The expressions of MMP-2 and -9 and TIMP-1 and -2 were investigated using Northern blotting, immunocytochemistry, and Western blotting on isolated endometrial stromal cell cultures and in conditioned medium. As shown in Figure 1A, Northern blot experiments indicated the expression of both MMP-2 (2.9 kilobases [kb]) and TIMP-2 (3.5 kb) transcripts in stromal and decidualized-like stromal cells under the influence of estradiol-17ß or of both estradiol and progesterone. Using immunofluorescence, MMP-2 (Fig. 1B) and, to a lesser extent, MMP-9 (not shown) were observed in isolated stromal cells. Western blot analysis of conditioned medium showed secretion of MMP-2, TIMP-1, and TIMP-2 (Fig. 1C) and confirmed the specificity of the antibodies.
|
Immunocytochemistry was carried out using 33 endometrial samples representing every stage of the menstrual cycle. MMP-1, MMP-2, MMP-3, MMP-9, TIMP-1, and TIMP-2 were studied on serial sections from each sample, using semiquantitative evaluation as described in Materials and Methods. Negative controls (irrelevant mouse IgG) were run in parallel to ensure specific labeling.
MMP-2 (Gelatinase A, 72 kDa)
MMP-2 was predominant in endometrial stromal cells throughout the menstrual cycle and was detected in a more stage-dependent fashion in the vessels (Fig. 2 and Table 1). Extracellular MMP-2 was identified in the endometrial stromal space during the midproliferative phase (Fig. 2A); the immunostaining became localized to the stromal cells in the late proliferative phase (Fig. 2B) and persisted throughout the secretory phase (Fig. 2C), with a maximum production at Days 25–28 and 1–2 of the cycle (Fig. 2, D–E. Epithelial cells were mostly negative, and a faint labeling was detected only in the late proliferative phase.
|
|
Vascular structures MMP-2 finely decorated capillaries and arterial endothelium during the proliferative and secretory phases (Fig. 3, E and H, and Table 1); in the secretory phase, arteriolar smooth muscle cells were occasionally stained. Vascular immunolabeling was particularly intense at the end of the luteal phase, especially on capillaries and smooth muscle cells of the vascular walls (Fig. 3B).
|
MMP-9 (Gelatinase B, 92 kDa)
MMP-9 was expressed in a stage-dependent fashion in blood cells, vascular structures, and glandular epithelial cells (Fig. 4 and Table 1).
|
At the end of the menstrual period, MMP-9 was mostly found in a few round cells with ovoid nuclei (possibly macrophages) and in rare polymorphonuclear cells present in the interstitial tissue (Fig. 4A). During the proliferative phase, the interstitial staining persisted, mainly in leukocytes. During the secretory phase (Days 19–20), MMP-9 was observed in the glandular epithelium (Fig. 4B); the staining was often more intense at the apical portion of the cells, suggesting luminal secretion. No epithelial staining was observed after Day 20 (Fig. 4D). During the secretory phase, strong immunolabeling for MMP-9 was observed in round cells within the stroma, often arranged in clusters and located close to the glandular epithelium (see Fig. 4C); most of these cells expressed the macrophage-specific antigen CD 68 (not shown). The greatest number of immunostained macrophages was found during menstruation, especially in the necrotic regions. During this period, immunolabeling of some elongated stromal cells was also observed (Fig. 4D).
Vascular structures During the menstrual cycle, MMP-9 was found mainly in arterioles (Fig. 3, C, F, and I; Table 1); immunostaining was clearly detectable in the midfollicular and secretory phase (Days 19–20) (Fig. 3, F and I), but it was stronger in capillaries and in other vascular structures during menstruation (Figs. 3C and 4D).
MMP-3 (Stromelysin-1)
MMP-3 expression was mainly observed in perimenstrual and menstrual specimens (Fig. 5 and Table 1). Just after menstruation, MMP-3 immunostaining was diffuse in the interstitial space, while during the early proliferative phase it was present only in small round cells within the stroma (Fig. 5A). It was barely detectable in the interstitium and in the epithelium during the midproliferative (Days 6–9) (not shown) to the midsecretory (Days 19–24) period (Fig. 5B, Day 13; Fig. 5C, Day 18). Just before menstruation (Days 26–28), foci of interstitial cells intensely expressing MMP-3 were irregularly distributed in the stroma (Fig. 5D). These cellular foci expanded, especially around the glandular basement membranes at the onset of menses (Days 1–2) (Fig. 5E).
|
Vascular structures During menstruation, MMP-3 was expressed in vessels (Fig. 3, A, D, and G; Table 1); in the areas of endometrial breakdown, the smooth muscle cells of arterioles and the capillaries were strongly labeled (Fig. 3A). In contrast, diffuse light staining for MMP-3 was observed in and around the capillaries on Days 8–10 and 20–22 of the cycle (Fig. 3, D and G).
Inhibitor TIMP-1
The staining for TIMP-1 was intense in vessels and lighter in cells within the stroma (Fig. 6 and Table 2). Absorption of the antibody with purified TIMP-1 abolished the labeling (not shown). During the early and mid follicular phase (Fig. 6, A–B), TIMP-1 was strongly expressed in the arterioles and in capillaries, to decrease in the late follicular phase (not shown); it was lightly detectable in the stromal cells. In the secretory phase (Days 15–24) (Fig. 6, C–D), capillaries and spiral arteries (especially the smooth muscle cells) were also labeled. In the midluteal phase, newly formed capillary strands were finely decorated with anti-TIMP1 antibodies, while a diffuse staining was observed in the interstitial space, and a few positive cells with round nuclei were found within the stroma (not shown). Labeling of vascular structures persisted before menstruation (Fig. 6E). In contrast, during menstruation (Days 28 and 1–2), a heterogeneous expression for TIMP-1 was observed in arteriolar smooth muscle cells and in the stroma: staining was present in non-necrotic tissue and was especially abundant in several groups of arterioles separating necrotic from non-necrotic areas (Fig. 6F).
|
|
Inhibitor TIMP-2
TIMP-2 expression was predominant in vessels, especially arterioles, and was lighter in glands and stromal space throughout the menstrual cycle (Fig. 7 and Table 2). Strong immunostaining was observed in arterioles and capillaries in the proliferative endometrium (Fig. 7, A–C); at the beginning of the proliferative phase (Fig. 7A), TIMP-2 was also found in the interstitial space and in the basal portion of endometrial glands. During the secretory phase (Fig. 7, D and G), arterioles and, to a lesser extent, capillaries and the basement membrane of most glands were labeled. Stromal cells were focally stained. Newly formed capillary strands were also labeled during the mid and late luteal phase (Fig. 7, F–G). During menstruation, TIMP-2 labeling in the endometrial tissue increased, reaching a heterogeneous distribution at Days 1–2, similar to that of TIMP-1: the localization was strong in the undisrupted areas, and TIMP-2 was very abundant in several groups of arterioles delimiting necrotic from non-necrotic areas (Fig. 7H).
|
Cyclic Variations of the Expression of MMPs and Their Inhibitors in the Endometrial Vessels
Our new findings of the localization of several MMPs and their inhibitors TIMP-1 and TIMP-2 in endometrial vessels prompted us to further analyze modifications of their expression according to the endometrial phase and relatively to each other (Figs. 3, 6, 7). MMP-1 (interstitial collagenase 1) was also analyzed in relation to blood vessels.
Premenstrual period and menstruation MMP expression in the vessels was especially striking during the menstrual phase, when the majority of tissue breakdown occurs. MMP-2 and MMP-3 were observed in capillaries and arterioles at the end of the luteal phase, while the expression of MMP-1, MMP-3, and MMP-9 was strongly associated with areas of endometrial destruction during menstruation (Fig. 3, A–C; Fig. 8, A–B); MMP-3 was intensely expressed in smooth muscle cells and in the basal portion of capillaries and arterioles (Fig. 3A). In contrast to MMPs, TIMP-1 and TIMP-2 were strongly expressed in all endometrial vessels throughout the menstrual cycle; during menstruation (Days 28–2), as shown in Figures 6–7, a more pronounced labeling for TIMP-1 and TIMP-2 was evident on groups of arterioles delimitating necrotic from non-necrotic areas. The labeling was especially evident in the smooth muscle cells of arterioles.
|
Proliferative and secretory phases MMP-1, TIMP-1, and TIMP-2 were present in the vascular structures during the proliferative phase, while MMP-3 was not; and expression of MMP-9 was occasionally observed (Fig. 3, D–F). Throughout the secretory phase, MMP-1 and MMP-2 but not MMP-3 were detected in the vessels (Fig. 3, G–I). Diffuse staining for MMP-3 was occasionally present in the stromal tissue surrounding capillaries on Days 8–10 and 20–22, corresponding to periods of edema and angiogenesis (Fig. 3, G–H). MMP-2 finely decorated the basement membrane of the capillaries and occasionally the smooth muscle cells of arterioles (Fig. 2; Fig. 3, E and H); MMP-9 was evident in spiral arteries during the secretory (Days 19–20) and menstrual phases (Fig. 3I). The vascular structures also strongly expressed TIMP-1 and TIMP-2 in the secretory phase (Days 15–24), with a greater expression of TIMP-1 in arterioles; capillaries and newly formed capillary strands were strongly labeled for TIMP-2 during the midluteal phase and before the onset of menstruation (Days 26–28) (Fig. 3, D–F; Fig. 7, D and G).
|
DISCUSSION |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
The antibodies used for this study have been extensively characterized [31–36]; analysis of the different patterns of immunostaining on serial sections from each individual biopsy, as well as Western blot analysis of conditioned medium from stromal cells, confirms their specificity and the absence of cross-reactions. Our findings of MMP-2 and TIMP-2 expression, both in vivo and in cultured stromal cells using Northern blot analysis and immunolocalization, also show that both mRNA(s) and protein(s) are coexpressed under hormonal stimulation. Although the present study does not indicate directly the cell type responsible for the production of MMPs and TIMPs, it is suggested that the proteins detected by immunostaining in stromal and vascular cells are secreted by them, as deduced from immunofluorescent studies (Fig. 1 and unpublished results).
Our findings of MMP-2 and -9 protein expression confirm previous data showing cyclical changes of metalloproteinase mRNAs in the endometrial epithelium and stroma during the menstrual cycle [7, 11, 19–21]. MMP-2 expression is predominant in stromal cells throughout the menstrual cycle, as previously described [20], but is detected in a more stage-dependent fashion in the vessels. The strong stromal expression found in late secretory and menstrual endometrium correlates with the increase of mRNA at this stage and the increase of both protein secretion and activation after progesterone withdrawal [11, 23, 24]. Expression of MMP-9 in polymorphonuclear cells and macrophages infiltrating the endometrium during the late secretory phase and menstrual endometrium is in agreement with previous findings [21] and correlates with the repression of the active form of MMP-9 by progesterone, as analyzed in human endometrial explants or in stromal cells in culture [22, 26]. MMP-9 in macrophages could be up-regulated by proinflammatory cytokines secreted during menstruation. In addition, the focal staining for MMP-9 observed in epithelial cells during Days 18–20, and that of MMP-3 at foci of edema during Days 20–24, suggest that those metalloproteinases might contribute to blastocyst implantation. The strongest staining for MMP-3 observed before the onset of menstruation (Days 25–28), associated with focal stromal disruption, and during menstruation has also been previously emphasized [21].
Expression of MMPs in Vessels
An important finding of this study is the expression of MMP-2, -3, and -9 in endometrial capillaries and/or arterioles, which has not been detailed before. MMP-2 and -9 are abundant in the secretory phase with a predominance of MMP-2 in capillaries, and of MMP-9 in spiral arteries at Days 19–20. MMP-2, -3, and -9 are strongly expressed during menstruation at foci of active tissue degradation. The weak production of MMP-3 in other phases of the cycle agrees with the negligible amounts of mRNA or protein found in endometrial specimens and with the suppression of its expression by progesterone [21, 41]. The vascular expression of MMP in a cyclic-dependent fashion contrasts with low MMP proteolytic activity in the arteries from the adult organism under physiological conditions [42]. Our results emphasize different functional roles of MMPs during the menstrual cycle and suggest their participation in vascular remodeling during angiogenesis and menstruation.
Endometrial angiogenesis during the midfollicular and periimplantation period (Days 20–24) is associated with the hormonally dependent production of vascular endothelial growth factor (VEGF), the main angiogenic and permeability factor of the endometrium [28, 43, 44]. During the proliferative and secretory periods, the expression of MMP-2 and VEGF [28] on newly formed endometrial capillaries suggests the interplay of metalloproteinases and angiogenic factors in the vascular growth of the endometrium, as evoked for tumor angiogenesis [17]. Possibly a shift in the balance of MMPs and their inhibitors within the capillaries or in their environment is required during endometrial angiogenesis. We have previously demonstrated the presence of VEGF, estrogen receptor, and progesterone receptor in the smooth muscle cells of spiral arteries [28, 39, 45] that develop after ovulation under both estradiol and progesterone influence [46]. The additional finding of MMP-9 in spiral arteries at Days 19–20 indicates its coparticipation in endometrial vascular development before implantation.
Vascular MMPs and Menstruation
Evidence for a role of MMPs (activated by declining progesterone concentrations) in starting the menstrual breakdown of tissue has been shown in vitro [8–11, 23, 25, 47]. Immunostaining for MMP-1, -2, -3, and -9 is found in vessels of the shedding endometrium, vascular MMP-2 and -3 being also present in the premenstrual phase. These findings have not been reported before. The vascular localization of MMP-3 during menstruation might be related to the role of MMP-3 in the activation of various metalloproteinase precursors, such as pro-MMP-1 and pro-MMP-9 [48–51], also detected in vascular structures during menstruation. Pro-MMP-3 can be activated by progesterone withdrawal, by the anti-progestin RU-486 [52], and by several enzymes (mast cell enzymes, plasmin, neutrophil elastase, and cathepsin G) [49, 50, 53, 54]. These new findings, together with the activation of MMPs upon progesterone deprivation, indicate their involvement in vascular degradation. Destabilization of endometrial microvasculature resulting from the degradation of the surrounding ECM is consistent with the heavy menstrual bleeding upon progesterone deprivation. In addition, the focal nature of menstrual MMP-2, -3, and -9 reactivity suggests that progesterone withdrawal is not the only factor responsible for in vivo induction of MMPs.
Expression of TIMPs in Vessels
Generation of full enzymatic activity of MMPs in vivo may depend on the availability of TIMP-1 or TIMP-2, whose distribution and role in the menstrual cycle have been less fully investigated than those of MMPs. TIMP-1 and TIMP-2 are detected in all cellular compartments of the endometrium, and their expression shows little cyclical variation (Table 2); those findings reflect previous data in the human [20, 28, 47] and sheep [55]. Our detailed findings also show strong immunolocalization of TIMP-1 and -2 in the arterioles and the capillaries throughout the cycle, with subtle variations from region to region depending on the stage; especially, vascular expression TIMP-1 and TIMP-2 becomes heterogeneous during the menstrual period. The stronger staining for TIMP-1 and TIMP-2 observed in the early-mid follicular phase could be related to down-regulation of these inhibitors by estradiol [55]; TIMP-1 and -2 could protect vessels during endometrial regeneration and maintain vascular integrity during the cycle. The stronger expression of TIMP-2 in the stroma and the glandular basement membranes at Days 15–24, when progesterone levels increase, could protect the endometrial tissue from proteases secreted by the trophoblast during implantation. The strong regional variations of both TIMP-1 and TIMP-2 expression during the perimenstrual period might explain the apparent discrepancies between the increased level of TIMP-1 mRNA reported at menstruation [47], the report of stable levels of this inhibitor under progesterone withdrawal [23], and the heterogeneous TIMP protein expression shown in our study and in the study from Zhang [27]. Heterogeneous expression was strongly observed for vascular TIMP-1 and -2, especially during menstruation. This dissimilar pattern of expression could reflect complex paracrine or autocrine mechanisms of regulation by individual cells within endometrial microenvironments. The modulation of TIMPs could be therefore quite complex, depending upon the species, the cell type, or the mesenchymal interactions [20, 27, 55, 56].
At menses, TIMPs are probably secreted in amounts insufficient to completely inhibit MMPs; in an in vitro model of menstruation [57], progesterone withdrawal alters the ratio of MMPs to TIMPs in favor of MMPs and, hence, of tissue degradation. The heterogeneous distribution of TIMP-1 and TIMP-2 during menstruation and their lower level in necrotic than in non-necrotic areas, associated with the increase of metalloproteases in the disrupted endometrium, is a new finding. Our observations imply that TIMPs limit the action of metalloproteinases in some areas, thus preserving the capacity for endometrial regeneration.
In conclusion, our observations implicate the participation of MMPs and TIMPs in vascular remodeling during angiogenesis and menstruation and suggest that endometrial cyclic changes involve vascular components. The occurrence of MMPs in both stromal and vascular compartments at the start of menstruation indicates that bleeding results from degradation of both stromal components and vascular walls. These observations will permit further investigations on vascular remodeling and also on irregular endometrial bleeding that may occur in pathologic conditions or during hormonal therapy, such as contraception and hormonal substitutive treatment.
|
ACKNOWLEDGMENTS |
|---|
|
FOOTNOTES |
|---|
2 Correspondence: M. Perrot-Applanat, INSERM U460, CHU Xavier Bichat, 16 Rue Henri Huchart, 75870 Paris Cedex, France. FAX: 33 1 44 85 61 56; applanat{at}inserm.bichat.fr
Accepted: June 14, 1999.
Received: February 3, 1999.
|
REFERENCES |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
nirch H, Frydman R, Chaouat G. In-vitro endometrial secretion of human interleukin for DA cells/leukaemia inhibitory factor by explant cultures from fertile and infertile women. Hum Reprod 1995; 10:2483–2488.This article has been cited by other articles:
|
|
 |
|
  J. V. Cockle, N. Gopichandran, J. J. Walker, M. I. Levene, and N. M. Orsi Matrix Metalloproteinases and Their Tissue Inhibitors in Preterm Perinatal Complications Reproductive Sciences, October 1, 2007; 14(7): 629 - 645. [Abstract] [PDF]
|
|
|
|
 |
|
 R.D. Catalano, H.O. Critchley, O. Heikinheimo, D.T. Baird, D. Hapangama, J.R.A. Sherwin, D.S. Charnock-Jones, S.K. Smith, and A.M. Sharkey Mifepristone induced progesterone withdrawal reveals novel regulatory pathways in human endometrium Mol. Hum. Reprod., September 1, 2007; 13(9): 641 - 654. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 T. Collette, R. Maheux, J. Mailloux, and A. Akoum Increased expression of matrix metalloproteinase-9 in the eutopic endometrial tissue of women with endometriosis Hum. Reprod., December 1, 2006; 21(12): 3059 - 3067. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
S. Malik, K. Day, I. Perrault, D.S. Charnock-Jones, and S. K. Smith Reduced levels of VEGF-A and MMP-2 and MMP-9 activity and increased TNF-{alpha} in menstrual endometrium and effluent in women with menorrhagia Hum. Reprod., August 1, 2006; 21(8): 2158 - 2166. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
C. Punyadeera, V.L. Thijssen, S. Tchaikovski, R. Kamps, B. Delvoux, G.A.J. Dunselman, A.F.P.M. de Goeij, A.W. Griffioen, and P.G. Groothuis Expression and regulation of vascular endothelial growth factor ligands and receptors during menstruation and post-menstrual repair of human endometrium Mol. Hum. Reprod., June 1, 2006; 12(6): 367 - 375. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 P. B. Cornet, C. Galant, Y. Eeckhout, P. J. Courtoy, E. Marbaix, and P. Henriet Regulation of Matrix Metalloproteinase-9/Gelatinase B Expression and Activation by Ovarian Steroids and LEFTY-A/Endometrial Bleeding-Associated Factor in the Human Endometrium J. Clin. Endocrinol. Metab., February 1, 2005; 90(2): 1001 - 1011. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
M. Plaisier, K. Kapiteijn, P. Koolwijk, C. Fijten, R. Hanemaaijer, J. M. Grimbergen, A. Mulder-Stapel, P. H. A. Quax, F. M. Helmerhorst, and V. W. M. van Hinsbergh Involvement of Membrane-Type Matrix Metalloproteinases (MT-MMPs) in Capillary Tube Formation by Human Endometrial Microvascular Endothelial Cells: Role of MT3-MMP |
