Matrix Metalloproteinases in Endometrial Breakdown and Repair ...

4 downloads 0 Views 479KB Size Report
Jun 8, 2005 - bers of animal species: Old World primates and some bats. [13–15], both of ...... weak cytostatic effects against cancer cell lines in vitro, but.
BIOLOGY OF REPRODUCTION 73, 672–680 (2005) Published online before print 8 June 2005. DOI 10.1095/biolreprod.105.042473

Matrix Metalloproteinases in Endometrial Breakdown and Repair: Functional Significance in a Mouse Model1 Tu’uhevaha J. Kaitu’u,2,3,4 Jun Shen,3 Jin Zhang,3 Naomi B. Morison,3 and Lois A. Salamonsen3 Prince Henry’s Institute of Medical Research,3 Clayton, Victoria 3168, Australia Department of Obstetrics and Gynecology,4 Monash University, Clayton, Victoria 3168, Australia ABSTRACT

increased MMP activity [11] in human endometrium taken immediately before and during menstruation. Furthermore, studies with endometrial explants in culture [12], showed that inhibition of MMPs, but not of other classes of proteases, inhibited explant breakdown. Menstruation occurs naturally only in very small numbers of animal species: Old World primates and some bats [13–15], both of which are available for experimentation only at specialist facilities. MMPs have been shown to be associated with the follicular-luteal transition in rhesus monkeys [16], but to date no functional studies have been performed to establish whether or not MMPs are critical for menstruation. A mouse model of endometrial breakdown and repair, first developed in the 1980s [17], was recently established in our laboratory [18]; suitably steroidprimed ovariectomized mice are subjected to a decidualizing stimulus, and, once decidualization has progressed appropriately, progesterone (P) is withdrawn. Over the following 24 h, the integrity of the decidualized endometrium becomes progressively disturbed, and eventually the endometrium is entirely degraded and shed. Repair of the endometrium is equally rapid, with re-epithelialization preceding stromal restoration; repair is complete in most animals by 48 h after withdrawal of P. The present study examined in detail, the production and cellular location of MMPs during the processes of endometrial breakdown and repair in this mouse model and showed that these are remarkably similar to the patterns of MMP production during the perimenstrual and menstrual phases in the human endometrium. Furthermore, two inhibitors of MMPs—doxycycline, a relatively nonspecific MMP inhibitor, and batimastat, a broad-based, potent inhibitor of MMP activity—were separately administered to the mice during these processes to establish whether menstrual breakdown and repair proceed in the absence of MMP activity.

Considerable correlative evidence suggests an important role for matrix metalloproteinases (MMPs) in menstruation, a process which occurs naturally in very few species. In this study, MMP expression was examined in a mouse model of endometrial breakdown and repair and the functional importance of MMPs determined. In the model, progesterone support was withdrawn from mice in which endometrial decidualization had been induced; 24 h later, endometrial breakdown was complete, and the entire decidual zone had been shed. Re-epithelialization had occurred by 36 h, and the endometrium had undergone extensive restoration toward a predecidualized state by 48 h. Immunoreactive MMP9 and MMP7 colocalized with leukocyte subsets, particularly neutrophils, whereas MMP13 staining was always extracellular. MMP3 and MMP7 were abundant during re-epithelialization in close proximity to newly reforming epithelium. The functional importance of MMPs in these processes was examined using two MMP inhibitors, doxycycline and batimistat. Both inhibitors effectively reduced MMP activity, as assessed by in situ zymography, but did not have significant effects on endometrial breakdown or repair. This study demonstrates that although MMPs are present in abundance during endometrial breakdown and repair in this mouse model, they are not the key mediators of these processes.

female reproductive tract, decidua, menstrual cycle, progesterone, uterus

INTRODUCTION

Menstruation is the loss of most of the functionalis layer of the endometrium of women, accompanied by bleeding, that occurs following withdrawal of steroid hormone support at the end of each menstrual cycle. During the past decade, considerable evidence has been presented to support the contention that MMPs, proteases that have combined specificities for all components of extracellular matrix, are key players at menstruation [1, 2]. This evidence includes strong correlative data including immunolocalization of MMP proteins in perimenstrual phase endometrium [3–7], localization of MMP mRNA by Northern blot [8] and in situ hybridization [9, 10], and demonstration of

MATERIALS AND METHODS

Animals Female C57BL/6 mice of 8–12 wk of age were housed in standard conditions with food and water provided ad libitum and a constant light cycle of 12L:12D (lights-on from 0800 to 2000 h). Ethics approval for this project was granted by the Monash Medical Center Animal Ethics Committee.

Supported by the National Health and Medical Research Council of Australia (169003, 143798, 241000) and by an Australian Postgraduate Scholarship to T.J.K. 2 Correspondence: T.J. Kaitu’u, Prince Henry’s Institute of Medical Research, Level 4, Block E, Monash Medical Centre, 246 Clayton Road, Clayton, Victoria 3168, Australia. FAX: 61 3 9594 6125; e-mail: [email protected]

Induction of the Mouse Model of Endometrial Breakdown and Repair

Received: 30 March 2005. First decision: 29 April 2005. Accepted: 6 June 2005. Q 2005 by the Society for the Study of Reproduction, Inc. ISSN: 0006-3363. http://www.biolreprod.org

Induction of the mouse model of endometrial breakdown and repair was performed as previously described [18]. Surgery was performed under xylazine/ketamine-induced anesthesia. Mice were ovariectomized 7 days before the first of three daily s.c. injections of 100 ng 17b-estradiol (Sigma Chemical Co., St. Louis, MO) in arachis oil at approximately 0900 h. After the mice were rested for 3 days, P implants were inserted s.c. into

1

672

MMPS IN AN ENDOMETRIAL BREAKDOWN AND REPAIR MODEL the back of each mouse and 5 ng of 17b-estradiol in arachis oil was injected s.c. approximately 0900 h on that and the subsequent 2 days. The P implants were prepared by filling silastic tubes (0.062 inches internal diameter, Dow Corning, Midland, MI) with crystalline P (Sigma Chemical) and sealing the ends with polyethylene plugs, such that the functional length of each implant was 1 cm. The implants were incubated overnight at 378C in Tris-buffered saline pH 7.35 (TBS)/1% fetal calf serum (FCS; Trace Biosciences, Sydney, Australia). At approximately 1100 h on the day of the final 17b-estradiol injection, 20 ml of sesame oil was injected into the lumen of the right uterine horn of each mouse to induce decidualization. The left horn remained untreated as a control. The P implants were removed 49 h later. Mice were killed at the time of implant removal (0 h) and 12, 16, 20, 24, 36 and 48 h thereafter (at least three mice per time point) and the uteri were harvested for further analysis. Uteri were cleaned of fat and weighed. Any mouse in which the oil-treated horn had not decidualized (as evidenced by weight $ 400% of the untreated contralateral horn) was excluded from the study. Tissue was fixed in Carnoy fixative for 4 h or formalin fixative overnight and processed to wax. Some was snap frozen in OCT compound (Sakura Finetek, Torrance, CA) and stored at 2808C for no more than 2 wk before use. Other tissue was snap frozen in liquid nitrogen before storage at 2808C.

Treatment with MMP Inhibitors Two inhibitors of MMPs (doxycycline and batimastat) were used in these studies. Doxycycline hydrochloride (Sigma) is a broad-based but nonspecific inhibitor of MMPs [19]. Batimastat is a hydroxamate MMP inhibitor that is highly specific for MMPs and the closely related ADAM (a disintegrin and metalloproteinase) enzymes, and is highly effective in inhibiting MMP activity in mice [20, 21]. Mice were randomly assigned into vehicle and treatment groups (n $ 4 mice per group). Doxycycline was prepared at 5 mg/ml and 0.1 ml was administered orally on the back of the tongue twice a day from the time of P withdrawal for 48 h. Batimastat (a gift from Vernalis Ltd, Wokingham, England) was prepared at 3 mg/ml, and 35 mg/kg of body weight was administered intraperitoneally 3 h before and 21 h after P withdrawal. Mice were killed at either 24 or 48 h after P implant removal and the uteri were harvested and processed as above.

Extraction of MMP from Tissues MMPs were extracted from samples of mouse uterus as described previously [22]. Weighed tissue was homogenized on ice in 0.01 mol CaCl2 L21 with 0.25% Triton-X100 (1:20, w/v) and centrifuged at 9000 3 g for 30 min. The pellet was resuspended in 10 volumes of high-calcium Tris buffer (0.1 mol CaCl2 L21, 0.05mol Tris L21, 0.15 mol NaCl L21, pH 7.5), heated to 608C for 6 min, and recentrifuged, and the supernatant (heat extract) was frozen. This method dissociates any metalloproteinases bound to substrate. The extract was subjected to gelatin zymography for analysis of MMP content.

Gelatin Zymography Protein activity in heat extracts was analyzed by zymography on 10% (w/v) SDS-polyacrylamide gels containing 1 mg/ml gelatin (all reagents from Bio-Rad, North Ryde) under nonreducing conditions and with standard MMPs as previously described [22]. Loading of samples was normalized such that a volume representing the same wet weight of tissue (1 mg) was applied to each lane. Gelatinase activity was visualized by negative staining and bands were identified by comparison with known human MMPs present in culture medium from baby hamster kidney cells transfected with MMP9. MMP identity of all bands was verified by incubation of parallel gels in the presence of EDTA (5 nM).

Immunohistochemical Analysis of MMP3, MMP7, MMP9, and MMP13 Immunohistochemistry was conducted to detect MMP3, MMP7, MMP9 and MMP13 in cross sections of uterus at different times following P withdrawal. Rabbit anti-rat MMP7(a) was a gift from Prof. J.F. Woessner Jr. (Univeristy of Florida, Gainesville, FL) [23], rat anti-human MMP7(b) monoclonal antibody (#338) was a gift from Prof. L.M. Matrisian (Vanderbilt University, Nashville, TN), rabbit anti-mouse MMP3 was a gift from Dr. L. Moons (Flanders Institute for Biotechnology, Leuven, Belgium), and mouse anti-rat MMP13 [24] was a gift from Dr. J.G. Lyons (University of Sydney, Sydney, Australia). Goat anti-mouse MMP9 was purchased from Santa Cruz Biotechnology (Santa Cruz, CA). Paraffin sec-

673

tions (5 mm) of Carnoy-fixed tissues (except when 2mm serial sections of formalin-fixed tissue were used in parallel with neutrophil detection) were dewaxed in histosol and rehydrated through descending grades of ethanol. They were then immersed in 0.1 M citrate buffer and heated for 5 min in a 700-W microwave oven set to medium. Once the slides had returned to room temperature (RT), they were rinsed in distilled (d) H2O and immersed in 3% H2O2 in methanol for 30 min at RT. Sections were then incubated with blocking solution containing 10% normal rabbit serum (MMP7b), 10% normal goat serum (MMP3 and MMP7a), 20% normal horse serum (MMP9), or 20% fetal calf serum (MMP13) in Tris-buffered saline pH 7.6 (TBS) for 20 min (10 min for MMP13) at room temperature. Sections were incubated for 1 h with primary antibodies diluted in 10% fetal calf serum (FCS)/TBS to 1mg/ml (MMP3), 1/500 (MMP7a), 1/300 (MMP7b) or 1mg/ml (MMP9), then washed sequentially in TBS, 0.6% (v/ v) Tween-20 in TBS, and three times in TBS. Biotinylated rabbit anti-rat IgG for MMP7a, biotinylated goat anti-rabbit IgG for MMP3 and MMP7, and biotinylated horse anti-goat IgG (DAKO, Glostrup, Denmark) for MMP9, diluted at 1/200 in 10% FCS/TBS, were applied for 45 min at RT and the slides washed as described above. For MMP13, an ARK kit (DAKO) was used according to the manufacturer’s instructions. In brief, the primary antibody was prepared to a final concentration of 1.5mg/ml in the biotinylated secondary antibody, incubated for 15 min at RT, and then blocking reagent (normal mouse IgG) was added and incubated for 5 min before application to sections for 30 min at RT. For all staining protocols the Strept ABC horseradish peroxidase (HRP) kit and diaminobenzidine solution (DAKO) were used in accordance with the manufacturer’s specifications to reveal the MMP3, MMP7a, MMP7b, MMP9, and MMP13 staining. Sections were lightly counterstained with Harris hematoxylin (Accustain; Sigma Diagnostics, Castle Hill, NSW, Australia), dehydrated, and mounted using DPX mounting medium. Negative controls were included for each tissue section by substitution of the primary antibody with a matching concentration of normal rat IgG for MMP7a, normal rabbit IgG for MMP3 and MMP7b, normal goat IgG (Sigma Diagnostics) for MMP9, and normal mouse IgG for MMP13.

Analysis of MMP9 Immunohistochemistry A microcomputer imaging device from Imaging Research (Brock University, St. Catherine, Ontario, Canada) was used to determine the average number of immunoreactive MMP9 positive cells per mm2. Cell numbers were analyzed separately in basal and decidualized areas of tissues. Cell counting was performed by the same observer and sections were scored blind [25].

Colocalization Studies Dual immunohistochemistry was performed on Carnoy-fixed tissues for macrophages and MMP7 or MMP9. Because of incompatible staining requirements for MMP9/MMP7 and neutrophils, serial 2 mm sections were cut from formalin-fixed tissues and adjacent sections subjected separately to the different protocols. MMP9 and MMP7 staining was performed as described above. Immunolocalization of neutrophils used the rat anti-mouse monoclonal MCA771GA (Serotec) diluted to 20 mg/ml in 10% FCS/TBS. After heating in citrate buffer for 5 min at high level and 3 min at medium-low level in a 700-W microwave oven, sections were immersed in 3% H2O2 in methanol for 5 min and blocked with 20% normal rabbit serum (NRS) for 5 min at RT. Primary antibody incubation was for 30 min at RT. Incubation with biotinylated rabbit anti-rat IgG (1:200 v/v in 10% FCS/TBS for 30 min) (DAKO) followed by the Strept ABC-alkaline phosphatase (AP) reagent preceded color development with fuchsin (DAKO). Immunolocalization of macrophages was performed on Carnoy-fixed tissue using the rat anti-murine F4/80 pan macrophage antibody (BMA Biomedicals, Rheinstrasse, Switzerland). Immunostaining consisted of microwave antigen retrieval (10 min at high level in a 700-W microwave oven), applications of peroxidase blocking agent for 10 min, primary antibody (0.81 mg/ml) diluted in 10% FCS/TBS) at 378C for 30 min, biotinylated rabbit anti-rat IgG (1:200 diluted in 10% FCS/TBS) for 30 min at RT, Strept-ABC AP reaction for 30 min and color development using fuchsin (DAKO) for 5 min. Dual immunohistochemistry/histochemistry was used on selected Carnoy-fixed specimens to identify whether MMP7 or MMP9 were present in natural killer (NK) cells [26]. MMP9 and MMP7 immunohistochemical staining was visualized as above, and NK cells were subsequently detected using biotinylated lectin-Dolichos biflorus agglutinin (Sigma Diagnostics) diluted at 1:200 in 10% FCS/TBS and applied to the tissue for 40 min at

674

KAITU’U ET AL.

FIG. 1. Gelatin zymography, demonstrating the presence of both latent (pro) and active forms of MMP9 and MMP2 in uterine tissue at selected times (in h) after P withdrawal. Two gels run separately are shown, with 20 h included as a reference point. The band at 65 kDa is active MMP2; at 72 kDa, proMMP2; at 84 kDa, active MMP9; at 92 kDa, monomeric MMP9; and at 180 kDa, dimeric MMP9. std: control containing known enzymes as marked.

378C, followed by Strept ABC/HRP (DAKO) for 30 min at RT and the fuchsin substrate-chromogen system (DAKO) for 5 min at RT.

In Situ Zymography Tissue sections were examined for the presence of active forms of gelatinases and collagenases by gelatin and collagen in situ zymography respectively, as previously described [11]. Substrates used were DQ gelatin from pig skin, fluorescein conjugate, and DQ collagen type 1 from bovine skin, fluorescein conjugate (Molecular Probes, Inc., Eugene, OR). Sevenmicron sections were cut from the center of each frozen tissue block, placed on poly-L-lysine-coated slides, fixed in 10% buffered formalin for 5 min at 48C, and washed three times with cold TBS. Where nuclear counterstaining was required, propidium iodide (Molecular Probes) diluted 1:50 (w/v) in TBS was applied for 8 min at RT and the tissue was then thoroughly washed with cold TBS. The slide was held in a darkened TBS bath at 48C until use. The desired substrate (DQ gelatin or DQ collagen) was dissolved to a final concentration of 25 mg/ml in a mixture of 2% gelatin and 2% sucrose in TBS with 0.02% sodium azide and 100 ml was layered over the tissue section, covered with a coverslip and incubated in a darkened humid chamber at 378C for 16 h. For control sections, the broad-spectrum MMP inhibitor phenanthroline (Sigma) was added to the section at a final concentration of 10 mM and incubated at 378C for 1 h before counterstaining. Each section was viewed using an Olympus Corp. (Birkeroed, Denmark) fluorescent microscope with a fluorescein isothiocyanate filter.

priately primed mice, the stromal cell population expanded and differentiated into decidual cells within 49 h (0 h after P withdrawal). Endometrial tissue breakdown was initiated 12–16 h after P withdrawal, and by 24 h, the entire decidual zone had been shed in most animals. Re-epithelialization was often nearly complete by 36 h and the endometrium was generally fully restored by 48 h. Variability was apparent along the length of each stimulated horn, some areas having progressed further in the process than others. By contrast to the stimulated horn, neither tissue breakdown nor repair was observed in the unstimulated horns, which did not undergo decidualization. Gelatin Zymography

Extracts from uteri taken from 0 to 16 h after P withdrawal were analyzed by gelatin zymography, which identifies both latent and active forms of gelatinase. Five major bands of gelatinolytic activity corresponding to proMMP2 (72 kDa), active MMP2 (;65 kDa), dimeric and monomeric proMMP9 (;180 and 92 kDa) and active MMP9 (84 kDa) were detected (Fig. 1). At 0 h, proMMP2 and a faint band of active MMP2 were detectable. By 12 h, additional bands of active MMP2, proMMP9, and dimeric proMMP9 were present. Although the bands representing the two forms of MMP2 remained until 20 h, the active form of MMP2 was barely detected at 24 or 48 h. The most striking change at the later time points was a substantial increase in both monomeric and dimeric forms of proMMP9 and the appearance of some active MMP9, although this was to a lesser extent at 24 h, perhaps because of the compromised state of the tissue. Immunostaining of MMP9

All statistical analysis was performed using one-way ANOVA followed by a Tukey-Kramer multiple comparison test. Significance was taken as P , 0.05.

MMP9-immunopositive cells were present throughout the endometrium in both decidualized and nondecidualized tissue (Fig. 2, A–K). In the nondecidualized horn, their abundance was low and did not alter throughout the experiment (not shown). In the decidualized horn at 0 h, MMP9 positive cells were mainly located in the basal zone of the decidualized endometrium and were in low abundance (Fig. 2A), but their numbers increased dramatically in the basal zone following P withdrawal (Fig. 2, B–E). Once breakdown of the decidual zone was in progression (20 and 24 h), very large numbers of MMP9 positive cells were also identified within areas of tissue breakdown (Fig. 2, D and E). As the tissue repaired, MMP9 cells were reduced in number at sites where re-epithelialization was complete, but were retained in substantial numbers where tissue debris remained (Fig. 2F). Quantitative analysis of the number of immunopositive cells in each of the basal and decidual zones revealed an increase, initially in the basal zone, which reached significance (P , 0.01) by 16 h. The number of immunopositive cells in the decidual zone also increased significantly by 16 h (P , 0.01), although there were always fewer MMP9 stained cells here than in the basal zone (Fig. 3). Double immunostaining or staining on serial sections was used to confirm the identity of the MMP9-immunopositive cells as predominantly neutrophils (Fig. 2, G and H), but some MMP9-positive macrophages were also found (Fig. 2, I and J). There were no MMP9positive NK cells (Fig. 2K).

RESULTS

Immunostaining of MMP7

As previously described [18], following administration of the decidual stimulus into the uterine lumen of appro-

Different immunoreactive MMP7 cellular staining was detected by two well-verified antibodies in epithelial cells

Assessment of Extent of Tissue Breakdown Cross sections of formalin-fixed uterine tissue were deparaffinized and hydrated by processing sections through Histosol (Sigma Chemical) and a graded series of ethanol to dH2O. The hydrated sections were stained with hematoxylin-eosin using standard procedures. Stained sections were dehydrated and mounted under coverslips using DPX mounting medium (BDH Laboratory Supplies, Poole, England). Sections were then examined for extent of tissue breakdown and/or repair. To quantify the data, relative areas of tissue undergoing breakdown and total endometrial area were assessed on the cross sections, using an OC-M micrometer reticule as previously described [22]. The area of breakdown was expressed as a percentage of total cross sectional endometrial area. All measurements were made by a single observer, blinded to the treatment group. One section from each of n 5 10 (control) and n 5 7 (batimistat) animals were examined.

Statistical Analysis

MMPS IN AN ENDOMETRIAL BREAKDOWN AND REPAIR MODEL

675

FIG. 3. Numbers of MMP9-positive leukocytes (cells/mm2) in the basal (open bar) and decidual (black bar) zones of the uterus at 0–24 h after P withdrawal. MMP9-positive cells significantly increased as time after P withdrawal progressed, but their number was always higher in the basal zone compared to the decidual zone. Paired t-tests demonstrated significant differences (P # 0.05) between basal and decidual zone after 12 h. A single asterisk indicates significant (P # 0.05) differences from previous time point in basal zone. A double asterisk indicates significant (P # 0.05) differences from previous time point in decidual zone. Data are represented as mean (n 5 8–12) 6 SD.

when the extracellular matrix was seen to become disorganized and subsequently was widespread throughout the breaking-down tissue (Fig. 4A). In contrast to MMP7 and MMP9 immunoreactivity, MMP13 staining was always extracellular. FIG. 2. Immunostaining for MMP9 and MMP7 in cross sections of mouse uterus at given times after P withdrawal. MMP9 at 0 h (A), 12 h (B), 16 h (C), 20 h (D), 24 h (E), and 48 h (F). At 0 h, MMP9 was in low abundance, but it increased as time after P withdrawal progressed, specifically in the basal zone. Most MMP9-positive cells (G) also stained with a neutrophil marker (H), and some MMP9-positive cells also stained as macrophages (I, J). NK cells (K) were all MMP9 negative. L–Q) MMP7 staining in leukocytes in breakdown areas around sites of repair (L, M) and in newly forming epithelium (N, O). Most MMP7-positive cells also stained with a neutrophil marker (P, Q). Bars 5 200mm (A–F), 10mm (G– K), 50mm (L, M, O), 100mm (N), 20mm (P, Q). Asterisk indicates doublestained cells. Arrow indicates non-double-stained cells.

and in some leukocytes (Fig. 2, L–Q). While decidualization was in progress, there was no obvious change of MMP7 expression in epithelial cells (data not shown), but MMP7 expression in leukocytes within the basal and decidual area increased with time (Fig. 2L) and reached a peak at 24 h after P withdrawal (Fig. 2M). In the area of tissue breakdown, more MMP7-immunoreactive cells were obvious compared to the basal area. In uteri where re-epithelialization was occurring ($ 24 h), there was strong staining for MMP7 in the reforming epithelium (Fig. 2N) and also abundant MMP7 positive leukocytes around this epithelium (Fig. 2O). Identification of the subsets of leukocytes with MMP71 phenotype demonstrated that most MMP7-immunoreactive leukocytes were neutrophils (Fig. 2, P and Q). No MMP7-positive macrophages or NK cells were detected (not shown). Immunostaining of MMP13

Immunoreactive MMP13 (mouse collagenase) was not detected in the endometrium of the mice until tissue breakdown was evident (20 h, 24 h). It was first apparent at the interface between the basal zone and the decidual zone,

Immunostaining of MMP3

Positive MMP3 immunostaining was confined to focal areas of endometrial stromal/decidual cells and their adjacent extracellular matrix at 0 h and 12 h when tissue remodeling associated with decidualization was still in progress. Similar foci of staining were also apparent from 16 h to 24 h (Fig. 4C). Such patchy staining is very similar to that seen in normal human endometrium during menstruation [4]. Subsequently, when the tissue was undergoing repair (48 h; Fig. 4D) specific staining was apparent in close proximity to the newly reforming epithelium in both stromal cells and leukocytes. This staining was not apparent once tissue had undergone complete restoration. In Situ Zymography

Both collagenase activity and gelatinase activity can be detected by in situ zymography, depending upon the substrate used. Collagenase activity was detected in the decidualized horns before P withdrawal (0 h). Bright fluorescent (green) spots were apparent around the glands at the edge of the decidual area and less intense fluorescence was apparent within the decidua (Fig. 4E). The latter is similar in intensity to that seen in decidualized endometrium late in the normal menstrual cycle [11]. Gelatinase activity at time 0 was low, although some discrete spots of fluorescence were seen also around the edge of the decidual area (Fig. 4F). At 12 h more collagenase activity was apparent within the decidual area, but only on the mesometrial side. Very little gelatinase activity was seen at this time. At 16 h, there was very little activity of either class of enzyme. By 20 h, when tissue destruction was underway, both collagenase and gelatinase activities were apparent around the edge of the area of tissue breakdown, and at 24 h, the time of max-

676

KAITU’U ET AL.

mals, there was only a small quantity of debris remaining within the lumen, and the endometrium had undergone extensive restoration toward a predecidualized state. Doxycycline had no significant effect on endometrial restoration in treated mice compared to the control mice (Fig. 5E). Effects of Batimistat on Production and Activity of MMPs

Batimistat was administered to mice 3 h before P withdrawal to determine whether specific inhibition of MMP activity would alter tissue breakdown or repair in the mouse model. Uteri were harvested and examined 24 and 48 h after P withdrawal. At 24 h after P withdrawal, both gelatinase and collagenase activity were barely detectable in uteri from batimistat treated animals (Fig. 4, I and L). However, no significant effect on tissue breakdown was noted, with the breaking-down area being 69% 6 13% and 72% 6 10% (mean 6 SD) of the total area for control and batimistat-treated uteri respectively (Figs. 5C and 6). Additionally, no difference in the degree of decidualization, examined by desmin immunostaining, a marker for decidualization (data not shown), was evident between treatment groups. Because the action of batimistat is specifically to inhibit MMP activity, the lack of effect observed on MMP immunostaining was as expected (data not shown). By 48 h after P withdrawal, tissue from batimistat-treated animals had undergone extensive restoration toward a predecidualized state, indicating no significant effect of batimistat on endometrial restoration compared to control mice (Fig. 5F). FIG. 4. MMP13 (A) and MMP3 (C, D) immunolocalization following P withdrawal, and collagenase and gelatinase activity following P withdrawal. For immunostaining, transverse sections of representative uterine horns, counterstained with hematoxylin, are shown (A). MMP13 is widespread in areas of tissue destruction at 24 h after P withdrawal. B) Negative control, same tissue as (A). C) MMP3 is present at foci within the extracellular matrix during tissue breakdown (16–24 h), and (D) specific staining is apparent around reforming epithelium during tissue repair (48 h). Both collagenase (E, J–L) and gelatinase (F, G–I) activity were detected by in situ zymography. At 0 h, collagenase activity (E) was detected as bright green fluorescent spots around the glands at the edge of the decidual tissue. By 24 h (J), collagenase activity was apparent in limited specific areas and was decreased under either doxycycline (K) or batimistat (L) treatment. At 0 h, gelatinase activity (F) was limited to discrete areas within the decidua, but it was apparent at 24 h (G) in specific areas of tissue breakdown. After treatment with either doxycycline (H) or batimistat (I), gelatinase activity was significantly reduced Red fluorescence represents propidium iodide nuclei staining. Bars 5 25mm (A–D), 10mm (E–L).

imal tissue breakdown, both activities were apparent only in limited specific areas (Fig. 4, G and J). Effects of Doxycycline on Production and Activity of MMPs

Doxycycline was administered to mice from the time of P withdrawal, and uteri were examined 24 and 48 h later to determine whether this treatment affected MMP activity and whether such inhibition of MMPs had any major effect on tissue breakdown or repair in this model. Both gelatinase and collagenase activities, as detected by in situ zymography, were inhibited by doxycycline: the number of fluorescence foci in the decidual zone decreased substantially in the doxycycline-treated group by 24 h after P withdrawal (Fig. 4H). Importantly, there was no detectable effect on tissue breakdown (Fig. 5B). By 48 h after P withdrawal in doxycycline-treated ani-

DISCUSSION

In this study, sequential expression and activity of endometrial MMP2, MMP3, MMP7, MMP9, and MMP13 were examined in a mouse model for endometrial breakdown and repair [18]. Both the cellular location and changes in expression of these MMPs, as tissue breakdown was initiated and completed and as endometrial repair proceeded, mimicked closely those seen in the human endometrium during the perimenstrual and menstrual phases of the cycle. Two inhibitors of MMPs were used in this study; both were proven to be effective in inhibiting MMP activity, but, surprisingly, they had little effect on either the extent of tissue breakdown or subsequent repair. In the human, strong correlative evidence exists supporting a role for MMPs in menstruation. Studies using in vitro culture systems of isolated stromal and epithelial cells and endometrial explants have demonstrated the presence (either protein or mRNA) of a number of MMPs in premenstrual and menstrual endometrium, including MMP1, MMP2, MMP3, MMP7, MMP9, MMP10, and MMP11 [1– 6, 11, 12, 27]. Although mRNA and protein localization studies lend support to a role for MMPs in menstruation, they are not indicative of MMP action or activity. Critical evidence for the importance of MMPs in menstruation came from an in vitro study using MMP inhibition compared with inhibition of other classes of proteases, which showed that only MMP inhibition stopped breakdown of endometrial explants [12]. Additionally, in situ zymography studies demonstrated elevated MMP activity before and during menstruation, at a level significantly higher than that during other phases of the cycle [11]. In this study, both MMP7 and MMP9 were colocalized with various leukocyte markers. In human endometrium, the number of leukocytes, particularly neutrophils, rises substantially immediately premenstrually, when they are

MMPS IN AN ENDOMETRIAL BREAKDOWN AND REPAIR MODEL

677

FIG. 5. Hematoxylin and eosin staining of morphological features of control tissue at 24 (A) and 48 h (D) after P withdrawal. After treatment with doxycycline (B, E) and batimistat (C, F), no effect on endometrial breakdown (B, C) or repair (E, F) was observed. Bars 5 100mm.

believed to play a role in tissue destruction through their production of various enzymes and proteases, including MMPs [1]. MMP9 has previously been colocalized to neutrophils in regions of tissue degradation [28], and thus it is expected that the role of neutrophil MMP9 in the mouse model is one of degradation. In particular, its basal localization indicates a likely role in the destruction of extracellular matrix surrounding decidual tissue, allowing the necrosis or degradation of this tissue by other factors. This supposition is supported by the observation that at the time of maximal tissue breakdown gelatinase activity is restricted to the basal zone in a similar pattern to that observed for MMP9. MMP13 is also expected to play a significant role in tissue destruction in this model, with immunoreactivity only detected once breakdown was evident. In the human mRNA, as well as activity, for the equivalent human homologue MMP1, is significantly upregulated only at the time of menstruation and not during other phases of the cycle [8, 29]. The widespread expression pattern of MMP13 protein in this mouse model suggests a role in the breakdown of decidual tissue that may occur subsequent to or in concert with the action of MMP9. MMP7-positive neutrophils were also detected maximally 24 h after P withdrawal. However, in contrast to basally-located MMP9-positive neutrophils, cells expressing MMP7 were more abundant in and around central areas of tissue destruction. The differential expression of MMP family members by neutrophils in distinct areas of tissue destruction may result from disparate regulatory signaling within each focal area. It is well recognized that within tissue, leukocytes, including neutrophils, are capable of producing a plethora of regulatory molecules, including cytokines, chemokines. and a range of enzymes that are important either directly in matrix degradation or indirectly by activation of other proteases [30]. In addition to their role in tissue destruction, MMPs also play a significant role in tissue repair [31] of many tissues including the skin [32], airways [33, 34], cornea [35] and intestine [36]. Functional studies in mice have revealed a significant impairment of wound healing capabilities in both MMP7 [31] and MMP3 [37] knock-out mice. In human endometrial epithelium, MMP7 is strongly ex-

pressed during the menstrual phase. MMP7 has previously been shown to be important for re-epithelialization following wounding in the trachea [38]. In our mouse model, MMP7 was detected in both epithelial cells and some leukocytes during breakdown, and was markedly upregulated during repair. The presence of MMP7 at sites of re-epithelialization, both in the reforming epithelium as well as in surrounding leukocytes, strongly suggests a role for MMP7 also in tissue repair in the endometrium. MMP3 was also identified in and around areas of re-epithelialization, but in contrast to MMP7, MMP3 protein was significantly upregulated only in stromal cells surrounding sites of re-epithelialization, and disappeared once the tissue had completely reverted to a predecidualized state. A similar pattern of MMP3 mRNA expression has been reported in corneal wound healing [39]. In this tissue, MMP3 is thought to play a role in basement membrane assembly, and stromal remodeling. The switching-off of MMP3 following the reestablishment of a mature epithelium is postulated to be

FIG. 6. Quantitative analysis of breaking-down tissue as a percentage of total endometrial area from control (black bar) or batimistat (open bar) treated animals. Data represents the mean 6 SD for n 5 10 (control) and n 5 7 (batimistat) animals. No significant difference between treatment groups was detected.

678

KAITU’U ET AL.

one of the signals for the cessation of wound healing, including MMP production [39], and this may also be true for the endometrium following menstruation. To examine the functional significance of MMPs in the endometrial breakdown and repair observed in the mouse model, two MMP inhibitors were employed. Tetracycline derivatives are capable of MMP inhibition via multiple mechanisms, including chelation of the zinc atom at the active site to interfere with MMP action, as well as interference of proteolytic activation of latent MMPs and reduction of enzyme expression [40–43]. Doxycycline is the most potent MMP inhibitor of the approved tetracyclines [44]. In vitro studies have demonstrated effects including inhibition of transcription or activity of MMP1, MMP2 (at least in part via action on MT1-MMP), MMP8, MMP9, and MMP13 [40, 45–49]. Tetracyclines also have effects on the interleukin-1 system, [45, 46], actions that would indirectly lead to a reduction in MMP action. They retard excessive tissue breakdown and bone resorption [19, 50] and affect chemotactic activity of leukocytes [51]. Administration of doxycycline to mice in very early pregnancy reduced the size of implantation sites, and this reflected the implantation-related phenotype of MMP9 null or TIMP1 transgenic mice [22, 52]. The second MMP inhibitor used, batimistat, is a substituted peptide analog of the peptide residues on one side of a principle cleavage site in type 1 collagen [53]. It is capable of producing specific potent reversible inhibition of MMPs via its hydroxamate group, which binds the zinc atom in an MMPs active site [54]. Batimistat has weak cytostatic effects against cancer cell lines in vitro, but is not cytotoxic. Batimistat also inhibits ADAM17 (also known as TNFa (tumor necrosis factor) converting enzyme or TACE), which interferes with the release of bioactive TNFa [55] and products of the actions of other ADAMs proteases. Clinical trials have demonstrated positive effects of batimistat in animal models of breast, ovarian, and colorectal cancer, including decreased growth of metastatic tumors [20, 53, 56–58]. Although the efficacy for inhibiting MMP activity in the endometrium was proven for batimistat and a clear reduction of activity was shown for doxycycline, neither had an effect on endometrial breakdown or repair in the mouse model. This was surprising, given the similarity in MMP expression patterns between perimenstrual endometrium in women and that seen in this model, and because MMPs are thought (although not proven) to be critical for menstruation in women. A major difference between the mouse and human endometrium is that in the human, decidualization of stromal cells occurs spontaneously and is initiated late in every cycle, whereas in the mouse, it is induced naturally only by the presence of a blastocyst. It has been proposed that the destruction and restoration of the endometrium is necessary because of the irreversibility of this differentiative process. Indeed, it has been noted that menstrual sloughing is more extensive in one species of fruit bat, the black mastiff bat (M. ater), in which endometrial differentiation progresses further in the nonpregnant cycle [15]. Correlation between decidual response and menstruation is also supported by differences between women and the rhesus macaque. In women, the decidual reaction begins around the spiral arterioles and spreads through the upper two thirds of the endometrium with significant enlargement of the periarteriolar stromal cells. In contrast, although there is some enlargement of stromal cells surrounding spiral arterioles, decidualization and bleeding in the rhesus monkey is minimal

[59]. It is of interest that in the mouse model used in the present studies, tissue breakdown and repair is observed only in the decidualized uterine horn, supporting the contention that initiation/progression of decidualization is closely linked with the onset of menstruation following withdrawal of P. However, an important difference between the mouse model and the human is that once decidualization is initiated in the mouse, it progresses much more extensively and at a more rapid pace than that observed in the human. A means to restrain decidualization of the mouse endometrium once it is underway has not been found. This extensive decidual response in the mouse model may be one reason for our inability to restrain the endometrial breakdown using MMP inhibition. It is also not clear why MMP inhibition had no observable effect on the tissue repair in the mouse model; menstrual/endometrial repair may require different mechanisms from wound healing [60]. Importantly, most wounds heal with scarring, whereas endometrial repair occurs without scarring, which otherwise occurs only during fetal development. However, in one study in the postpartum mouse [61] endometrial ‘nodules’ have been reported. In conclusion, the results of this study demonstrate that although MMPs are present during endometrial breakdown and repair in the mouse model, they are not the key contributors to these processes in this model. However, given that there are differences between the mouse model and the human, these studies do not confirm or refute the role of MMPs in menstruation or endometrial repair in women. Indeed, further investigation is required to define the limitations of this model and its applicability to events in the human. The function of infiltrating cells in these processes is currently under investigation to determine whether the production of cytokines or proteases other than MMPs are essential to endometrial breakdown or repair in this mouse model. ACKNOWLEDGMENTS We thank Professor JF Woessner Jr., Professor L Matrisian, Dr. Lieve Moons, Dr. JG Lyons, and Vernalis Ltd. for their generous gifts of reagents.

REFERENCES 1. Salamonsen LA, Woolley DE. Menstruation: induction by matrix metalloproteinases and inflammatory cells. J Reprod Immunol 1999; 44: 1–27. 2. Curry TE, Jr., Osteen KG. The matrix metalloproteinase system: changes, regulation, and impact throughout the ovarian and uterine reproductive cycle. Endocr Rev 2003; 24:428–465. 3. Zhang J, Hampton AL, Nie G, Salamonsen LA. Progesterone inhibits activation of latent matrix metalloproteinase (MMP)-2 by membranetype 1 MMP: enzymes coordinately expressed in human endometrium. Biol Reprod 2000; 62:85–94. 4. Jeziorska M, Nagase H, Salamonsen LA, Woolley DE. Immunolocalization of the matrix metalloproteinases gelatinase B and stromelysin 1 in human endometrium throughout the menstrual cycle. J Reprod Fertil 1996; 107:43–51. 5. Jeziorska M, Salamonsen LA, Woolley DE. Mast cell and eosinophil distribution and activation in human endometrium throughout the menstrual cycle. Biol Reprod 1995; 53:312–320. 6. Irwin JC, Kirk D, Gwatkin RB, Navre M, Cannon P, Giudice LC. Human endometrial matrix metalloproteinase-2, a putative menstrual proteinase. Hormonal regulation in cultured stromal cells and messenger RNA expression during the menstrual cycle. J Clin Invest 1996; 97:438–447. 7. Skinner JL, Riley SC, Gebbie AE, Glasier AF, Critchley HO. Regulation of matrix metalloproteinase-9 in endometrium during the men-

MMPS IN AN ENDOMETRIAL BREAKDOWN AND REPAIR MODEL

8. 9.

10.

11.

12.

13. 14. 15. 16.

17.

18. 19.

20.

21. 22. 23. 24.

25.

26.

27. 28.

29.

strual cycle and following administration of intrauterine levonorgestrel. Hum Reprod 1999; 14:793–799. Hampton AL, Salamonsen LA. Expression of messenger ribonucleic acid encoding matrix metalloproteinases and their tissue inhibitors is related to menstruation. J Endocrinol 1994; 141:R1–3. Rodgers WH, Matrisian LM, Giudice LC, Dsupin B, Cannon P, Svitek C, Gorstein F, Osteen KG. Patterns of matrix metalloproteinase expression in cycling endometrium imply differential functions and regulation by steroid hormones. J Clin Invest 1994; 94:946–953. Rodgers WH, Osteen KG, Matrisian LM, Navre M, Giudice LC, Gorstein F. Expression and localization of matrilysin, a matrix metalloproteinase, in human endometrium during the reproductive cycle. Am J Obstet Gynecol 1993; 168:253–260. Zhang J, Salamonsen LA. In vivo evidence for active matrix metalloproteinases in human endometrium supports their role in tissue breakdown at menstruation. J Clin Endocrinol Metab 2002; 87:2346– 2351. Marbaix E, Kokorine I, Moulin P, Donnez J, Eeckhout Y, Courtoy PJ. Menstrual breakdown of human endometrium can be mimicked in vitro and is selectively and reversibly blocked by inhibitors of matrix metalloproteinases. Proc Natl Acad Sci U S A 1996; 93:9120–9125. Finn CA. Why do women and some other primates menstruate? Perspect Biol Med 1987; 30:566–574. Rasweiler JJt. Spontaneous decidual reactions and menstruation in the black mastiff bat, Molossus ater. Am J Anat 1991; 191:1–22. Rasweiler JJt, de Bonilla H. Menstruation in short-tailed fruit bats (Carollia spp.). J Reprod Fertil 1992; 95:231–248. Brenner RM, Rudolph L, Matrisian L, Slayden OD. Non-human primate models; artificial menstrual cycles, endometrial matrix metalloproteinases and s.c. endometrial grafts. Hum Reprod 1996; 11Suppl 2:150–164. Finn CA, Pope M. Vascular and cellular changes in the decidualized endometrium of the ovariectomized mouse following cessation of hormone treatment: a possible model for menstruation. J Endocrinol 1984; 100:295–300. Brasted M, White CA, Kennedy TG, Salamonsen LA. Mimicking the events of menstruation in the murine uterus. Biol Reprod 2003; 69: 1273–1280. Golub LM, Ramamurthy NS, McNamara TF, Greenwald RA, Rifkin BR. Tetracyclines inhibit connective tissue breakdown: new therapeutic implications for an old family of drugs. Crit Rev Oral Biol Med 1991; 2:297–321. Wang X, Fu X, Brown PD, Crimmin MJ, Hoffman RM. Matrix metalloproteinase inhibitor BB-94 (batimastat) inhibits human colon tumor growth and spread in a patient-like orthotopic model in nude mice. Cancer Res 1994; 54:4726–4728. Goss KJ, Brown PD, Matrisian LM. Differing effects of endogenous and synthetic inhibitors of metalloproteinases on intestinal tumorigenesis. Int J Cancer 1998; 78:629–635. Rechtman MP, Zhang J, Salamonsen LA. Effect of inhibition of matrix metalloproteinases on endometrial decidualization and implantation in mated rats. J Reprod Fertil 1999; 117:169–177. Yu WH, Woessner JF, Jr. Heparan sulfate proteoglycans as extracellular docking molecules for matrilysin (matrix metalloproteinase 7). J Biol Chem 2000; 275:4183–4191. Lyons JG, Birkedal-Hansen B, Pierson MC, Whitelock JM, BirkedalHansen H. Interleukin-1 beta and transforming growth factor-alpha/ epidermal growth factor induce expression of M(r) 95,000 type IV collagenase/gelatinase and interstitial fibroblast-type collagenase by rat mucosal keratinocytes. J Biol Chem 1993; 268:19143–19151. Iqbal J, Henry BA, Pompolo S, Rao A, Clarke IJ. Long-term alteration in bodyweight and food restriction does not affect the gene expression of either preproorexin or prodynorphin in the sheep. Neuroscience 2003; 118:217–226. King A, Hiby SE, Gardner L, Joseph S, Bowen JM, Verma S, Burrows TD, Loke YW. Recognition of trophoblast HLA class I molecules by decidual NK cell receptors–a review. Placenta 2000; 21(suppl A):S81– 85. Curry TE, Jr., Osteen KG. Cyclic changes in the matrix metalloproteinase system in the ovary and uterus. Biol Reprod 2001; 64:1285– 1296. Vincent AJ, Malakooti N, Zhang J, Rogers PA, Affandi B, Salamonsen LA. Endometrial breakdown in women using Norplant is associated with migratory cells expressing matrix metalloproteinase-9 (gelatinase B). Hum Reprod 1999; 14:807–815. Marbaix E, Kokorine I, Henriet P, Donnez J, Courtoy PJ, Eeckhout Y. The expression of interstitial collagenase in human endometrium is

30. 31. 32. 33. 34.

35. 36.

37. 38. 39.

40. 41.

42.

43.

44. 45. 46. 47. 48.

49.

50. 51.

679

controlled by progesterone and by oestradiol and is related to menstruation. Biochem J 1995; 305(Pt 3):1027–1030. Salamonsen LA, Zhang J, Brasted M. Leukocyte networks and human endometrial remodelling. J Reprod Immunol 2002; 57:95–108. Parks WC. Matrix metalloproteinases in repair. Wound Repair Regen 1999; 7:423–432. Pilcher BK, Dumin JA, Sudbeck BD, Krane SM, Welgus HG, Parks WC. The activity of collagenase-1 is required for keratinocyte migration on a type I collagen matrix. J Cell Biol 1997; 137:1445–1457. Dunsmore SE, Saarialho-Kere UK, Roby JD, Wilson CL, Matrisian LM, Welgus HG, Parks WC. Matrilysin expression and function in airway epithelium. J Clin Invest 1998; 102:1321–1331. Legrand C, Polette M, Tournier JM, de Bentzmann S, Huet E, Monteau M, Birembaut P. uPA/plasmin system-mediated MMP9 activation is implicated in bronchial epithelial cell migration. Exp Cell Res 2001; 264:326–336. Sivak JM, Fini ME. MMPs in the eye: emerging roles for matrix metalloproteinases in ocular physiology. Prog Retin Eye Res 2002; 21:1–14. Salmela MT, Pender SL, Karjalainen-Lindsberg ML, Puolakkainen P, Macdonald TT, Saarialho-Kere U. Collagenase-1 (MMP1), matrilysin-1 (MMP7), and stromelysin-2 (MMP10) are expressed by migrating enterocytes during intestinal wound healing. Scand J Gastroenterol 2004; 39:1095–1104. Bullard KM, Lund L, Mudgett JS, Mellin TN, Hunt TK, Murphy B, Ronan J, Werb Z, Banda MJ. Impaired wound contraction in stromelysin-1-deficient mice. Ann Surg 1999; 230:260–265. Parks W, Lopez-Boado Y, Wilson C. Matrilysin in epithelial repair and defense. Chest 2001; 120:36S–41S. Daniels JT, Geerling G, Alexander RA, Murphy G, Khaw PT, Saarialho-Kere U. Temporal and spatial expression of matrix metalloproteinases during wound healing of human corneal tissue. Exp Eye Res 2003; 77:653–664. Uitto VJ, Firth JD, Nip L, Golub LM. Doxycycline and chemically modified tetracyclines inhibit gelatinase A (MMP2) gene expression in human skin keratinocytes. Ann N Y Acad Sci 1994; 732:140–151. Sorsa T, Ramamurthy NS, Vernillo AT, Zhang X, Konttinen YT, Rifkin BR, Golub LM. Functional sites of chemically modified tetracyclines: inhibition of the oxidative activation of human neutrophil and chicken osteoclast pro-matrix metalloproteinases. J Rheumatol 1998; 25:975– 982. Smith GN, Jr., Mickler EA, Hasty KA, Brandt KD. Specificity of inhibition of matrix metalloproteinase activity by doxycycline: relationship to structure of the enzyme. Arthritis Rheum 1999; 42:1140– 1146. Lee HM, Golub LM, Cao J, Teronen O, Laitinen M, Salo T, Zucker S, Sorsa T. CMT-3, a non-antimicrobial tetracycline (TC), inhibits MT1-MMP activity: relevance to cancer. Curr Med Chem 2001; 8: 257–260. Burns FR, Stack MS, Gray RD, Paterson CA. Inhibition of purified collagenase from alkali-burned rabbit corneas. Invest Ophthalmol Vis Sci 1989; 30:1569–1575. Solomon A, Rosenblatt M, Li DQ, Liu Z, Monroy D, Ji Z, Lokeshwar BL, Pflugfelder SC. Doxycycline inhibition of interleukin-1 in the corneal epithelium. Invest Ophthalmol Vis Sci 2000; 41:2544–2557. Shlopov BV, Stuart JM, Gumanovskaya ML, Hasty KA. Regulation of cartilage collagenase by doxycycline. J Rheumatol 2001; 28:835– 842. Gu Y, Lee HM, Roemer EJ, Musacchia L, Golub LM, Simon SR. Inhibition of tumor cell invasiveness by chemically modified tetracyclines. Curr Med Chem 2001; 8:261–270. Lukkonen A, Sorsa T, Salo T, Tervahartiala T, Koivunen E, Golub L, Simon S, Stenman UH. Downregulation of trypsinogen-2 expression by chemically modified tetracyclines: association with reduced cancer cell migration. Int J Cancer 2000; 86:577–581. Curci JA, Mao D, Bohner DG, Allen BT, Rubin BG, Reilly JM, Sicard GA, Thompson RW. Preoperative treatment with doxycycline reduces aortic wall expression and activation of matrix metalloproteinases in patients with abdominal aortic aneurysms. J Vasc Surg 2000; 31:325– 342. Rifkin BR, Vernillo AT, Golub LM, Ramamurthy NS. Modulation of bone resorption by tetracyclines. Ann N Y Acad Sci 1994; 732:165– 180. Golub LM, Sorsa T, Lee HM, Ciancio S, Sorbi D, Ramamurthy NS, Gruber B, Salo T, Konttinen YT. Doxycycline inhibits neutrophil (PMN)-type matrix metalloproteinases in human adult periodontitis gingiva. J Clin Periodontol 1995; 22:100–109.

680

KAITU’U ET AL.

52. Alexander CM, Hansell EJ, Behrendtsen O, Flannery ML, Kishnani NS, Hawkes SP, Werb Z. Expression and function of matrix metalloproteinases and their inhibitors at the maternal-embryonic boundary during mouse embryo implantation. Development 1996; 122:1723– 1736. 53. Macaulay VM, O’Byrne KJ, Saunders MP, Braybrooke JP, Long L, Gleeson F, Mason CS, Harris AL, Brown P, Talbot DC. Phase I study of intrapleural batimastat (BB-94), a matrix metalloproteinase inhibitor, in the treatment of malignant pleural effusions. Clin Cancer Res 1999; 5:513–520. 54. Talbot DC, Brown PD. Experimental and clinical studies on the use of matrix metalloproteinase inhibitors for the treatment of cancer. Eur J Cancer 1996; 32A:2528–2533. 55. Gearing AJ, Beckett P, Christodoulou M, Churchill M, Clements J, Davidson AH, Drummond AH, Galloway WA, Gilbert R, Gordon JL, Leber TM, Mangan M, et al. Processing of tumour necrosis factoralpha precursor by metalloproteinases. Nature 1994; 370:555–557. 56. Wojtowicz-Praga S, Low J, Marshall J, Ness E, Dickson R, Barter J,

57.

58.

59. 60. 61.

Sale M, McCann P, Moore J, Cole A, Hawkins MJ. Phase I trial of a novel matrix metalloproteinase inhibitor batimastat (BB-94) in patients with advanced cancer. Invest New Drugs 1996; 14:193–202. Davies B, Brown PD, East N, Crimmin MJ, Balkwill FR. A synthetic matrix metalloproteinase inhibitor decreases tumor burden and prolongs survival of mice bearing human ovarian carcinoma xenografts. Cancer Res 1993; 53:2087–2091. Eccles SA, Box GM, Court WJ, Bone EA, Thomas W, Brown PD. Control of lymphatic and hematogenous metastasis of a rat mammary carcinoma by the matrix metalloproteinase inhibitor batimastat (BB94). Cancer Res 1996; 56:2815–2822. Brenner RM, Slayden OD. Cyclic changes in the primate oviduct and endometrium. In: Knobil E, Neill JD (eds.), The Physiology of Reproduction, vol. 2. New York: Raven Press; 1994:541–569. Salamonsen LA. Tissue injury and repair in the female human reproductive tract. Reproduction 2003; 125:301–311. Brandon JM. Decidualization in the post-partum uterus of the mouse. J Reprod Fertil 1990; 88:151–158.