| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
Department of Molecular Medicine, Osaka Medical Center and Research Institute for Maternal and Child Health, Izumi, Osaka 594-1101, Japan
 |
ABSTRACT |
|---|
 TOP ABSTRACT INTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
endometrium, fertilization, implantation, matrix metalloproteinase, neutrophil, pregnancy, seminal vesicles, uterus
|
INTRODUCTION |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
For normal implantation, the activities of MMPs must be finely regulated, as demonstrated by the fact that a chemical inhibitor of MMPs reduced the length and overall size of decidua [3, 7]. In vivo, MMP activity is modulated by tissue inhibitors of metalloproteinases (TIMPs). TIMPs bind to the highly conserved zinc-binding site of active MMPs at molar equivalence and to other domains as well in MMP2 and MMP9. TIMPs 1 and 2 bind proforms of MMP9 (proMMP9) and MMP2 (proMMP2), respectively. TIMP3 is present in an ECM-bound form [8] and inhibits MMP9 more effectively than TIMP1 or TIMP2 [9]. The expression of TIMPs is regulated in the uterus during implantation and decidualization. The mRNAs of Timp1, Timp2, and Timp3 were highly expressed by the primary decidualized cells surrounding the implanting embryo on Day 6, and in both primary and secondary decidualized cells at the implantation site on Day 7 in rats [5]. Transgenic mice overexpressing Timp1 exhibited a phenotype similar to that of animals treated with MMP inhibitors (i.e., retardation of decidual remodeling and growth), and had irregularly oriented embryos in the uterus [3].
Taken together, these studies indicate important roles for MMP activity and its regulation in the processes of embryonal invasion and decidualization. However, MMP during the earlier preimplantation period has not been studied in sufficient detail regarding its occurrence, responsible molecules, and cellular origin. In the present study, we found that intense MMP9 activity is present just after mating, and that infiltrating neutrophils are the source of this activity.
|
MATERIALS AND METHODS |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
Animal protocols were approved by the institutional animal care and use committee at Osaka Medical Center and Research Institute for Maternal and Child Health. Adult ICR female and BDF1 male mice were obtained from SLC (Shizuoka, Japan) and were housed under temperature- and light-controlled conditions (lights-on from 0800 to 2000 h) with free access to food and water. The estrus cycle was classified into four stages (proestrus, estrus, metestrus, and diestrus) by vaginal smear and external view of the vagina on daily examination for two consecutive weeks. ICR female mice (6–9 wk) in the proestrus stage were mated with intact or vasectomized males. The day when a vaginal copulatory plug was first observed was designated Day 1 of pregnancy or pseudopregnancy. In mating with seminal vesicle (SV)-deficient males, vaginal smears were sampled to check for the presence of sperm, because no copulatory plugs formed. Uterine horns from the mice between 1300 and 1400 h on Days 1–8 of pregnancy were dissected. The implantation site in the uterus on Day 5 of pregnancy was visualized by i.v. injection of 0.1 ml of Evans blue dye solution (2% w/v in 0.9% w/v NaCl) 5 min before killing the mice. Vasectomized and SV-deficient males were prepared according to procedures described by Tremellen et al. [10].
To deplete neutrophils, female mice were injected i.p. with 100 µg of rat antigranulocyte RB6-8C5 monoclonal immunoglobulin G (IgG) (PharMingen, San Diego, CA), which suppresses neutrophils in peripheral blood for 8 days with a single administration [11–13].
Uterine luminal fluid was obtained 14 h after mating from a cut made near the cervical orifice by squeezing the horns, the ends of which were pinched with tweezers. Debris was sedimented and supernatants were collected for gelatin zymography.
In Situ Zymography
Dissected uterine horns were placed in Cryomold (Miles Inc., Elkhart, IN), and then OCT compound (Sakura Finetechnical Co., Ltd., Tokyo, Japan) was used for tissue embedding. The cryomold was placed into liquid nitrogen and maintained for 30–40 sec to freeze the tissues. Frozen tissues were cut into 4-µm thickness, and immediately placed on the gelatin surface of MMP In Situ Zymo-Film (Wako Pure Chemicals Industries, Osaka, Japan) or on MMP-PT In Situ Zymo-Film (Wako) containing an MMP inhibitor, 1,10-phenanthroline. The film was incubated in a humid chamber for 3 or 5 h at 37°C according to the manufacturer's instructions. After incubation, the film was dried for 30 min at room temperature, and then stained with Biebrich Scarlet solution (Wako) for 4 min. The film was washed in running water for 7 min and air-dried for a few hours. MMP activities were evaluated by comparing the unstained areas from the digestion with total proteinases, including MMPs in the MMP film, to those from digestion with proteinases other than MMPs in the MMP-PT film.
Gelatin Zymography
Uterine horns were weighed, cut into small pieces, and homogenized in a RIPA buffer (150 mM NaCl, 1.0% NP-40, 0.5% deoxycholate, 0.1% SDS, and 50 mM Tris-HCl buffer, pH 8.0) at 1:4 (w/v) by a pellet mixer on ice. The lysate was centrifuged at 14 000 x g for 15 min at 4°C. The supernatant was collected and loaded on a 9% SDS polyacrylamide gel containing gelatin (1 g/ml) under nonreducing conditions. After electrophoresis, the gel was rinsed in 2.5% (w/v) Triton X-100 containing 50 mM Tris-HCl pH 7.5 and 100 mM NaCl for 90 min, and then incubated in 50 mM Tris-HCl pH 7.5 containing 50 mM CaCl2 for 16–18 h at 37°C. MMP activities were visualized as an unstained band after Coomassie brilliant blue staining. The gel image was acquired with a Printgraph (ATTO, Tokyo, Japan), and analyzed with Lane Analyzer software (ATTO).
Western Blot Analysis
Supernatants from tissue lysates were subjected to electrophoresis on a 9% SDS polyacrylamide gel under reducing conditions, and proteins were transferred to Immobilon-P membrane (Millipore, Bedford, MA). After blocking with 6% skim milk in PBS for 1 h at room temperature, blots were incubated overnight at 4°C with goat polyclonal antibody raised against the carboxy terminus of MMP9 (Santa Cruz Biotechnology, Santa Cruz, CA). After washing, the blots were incubated for 1 h at room temperature with donkey anti-goat immmunoglobulin G conjugated with horseradish peroxidase (Promega, Madison, WI), and then visualized with an Enhanced Chemiluminescence Plus kit (Amersham, Buckinghamshire, U.K.) followed by exposure on Kodak X-OMAT AR film (Kodak, Rochester, NY).
Reverse Transcription-Polymerase Chain Reaction
Total RNA from uterine horns was extracted with Isogen (Nippon Gene; Tokyo, Japan) according to the manufacturer's instructions. Total RNA was reverse transcribed and the resulting cDNA was then amplified by ReverTra Dash (Toyobo, Osaka, Japan) according to the manufacturer's protocol. The primer sequences for mouse Mmp9, Timp1, and beta actin cDNAs are presented in Table 1. The numbers of amplification cycles suitable for quantification were 32, 30, and 23 for Mmp9, Timp1, and beta actin, respectively, because all three were in the linearly amplifying phase. The amount of template cDNA in each polymerase chain reaction (PCR) was adjusted to give the same amplified beta actin amount in all samples. The products were electrophoresed on a 1.2% (w/v) agarose gel and visualized by ethidium bromide for quantification of specific transcripts. The gel image was processed using Lane Analyzer software as described above.
|
Immunohistochemistry
Frozen sections 4 µm in thickness were prepared using a cryostat and fixed at 4°C for 10 min with 4% paraformaldehyde in PBS (for MMP9) or an ethanol/acetone (1:1, v/v) solution (for macrophages and neutrophils). After washing in PBS, endogenous peroxidase activity was quenched by incubation with 0.3% hydrogen peroxide in methanol for 30 min at room temperature. Sections were washed with PBS and then incubated with 2% (w/v) BSA in PBS for 30 min. The sections were then incubated with goat polyclonal antibody to MMP9 (1:100), rat monoclonal antibody to macrophage F4/80 antigen (Serotec, Oxford, U.K.) (1:200), or RB6-8C5 antibody (1:50) diluted in PBS at 4°C for 16 h. The sections incubated with the goat antibody were incubated with rabbit biotinylated anti-goat IgG (Vector Laboratories; Burlingame, CA) (1:100) diluted in PBS containing 1.0% BSA for 45 min, and then Vectastain avidin-biotin complex reagent (Vector) for 30 min. The sections incubated with rat monoclonal antibodies were incubated with Histofine Simple stain mouse MAX-PO (Nichirei, Tokyo, Japan) for 30 min. Peroxidase substrate, 3-amino-9-methyl-carbazole (Nichirei), or 3-3'-diaminobenzidine (Dojindo, Kumamoto, Japan) was used for color development. Normal goat IgG or rat IgG was used as a negative control. Sections were counterstained with Mayer hematoxylin. Double-staining for MMP9 and Ly6G was carried out using fluorescein avidin D (1:500) (Vector) and rhodamine-conjugated anti-rat IgG Fab (1:100) (Chemicon, Temecula, CA), respectively, with Fluorescent Mounting Medium (DAKO, Carpinteria, CA).
Statistical Analyses
The significance of differences between groups was evaluated with analysis of variance and the Duncan multiple range test.
|
RESULTS |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
In situ zymography allows MMP activity to be discriminated from other proteolytic activities in tissue sections by comparing the gelatinolysis in two gelatin-coated films, with versus without the MMP inhibitor 1,10-phenanthroline. As shown in Figure 1, marked gelatinolytic activity was identified in the whole endometrium on Days 1 and 2 of pregnancy by 3-h incubation, while weak gelatinolysis occurred at the glands and accompanying blood vessels in the MMP-PT film. This showed MMP to be present in the endometrium on Days 1 and 2 of pregnancy. Gelatinolytic activity declined starting on Day 3 of pregnancy and was barely detectable in the endometrium on Day 5, in either film, with 3 h of incubation (Fig. 1, E–J). With longer incubation (5 h), gelatinolysis was observed in both the MMP and the MMP-PT films of Day 5 and Day 6 tissues (Fig. 2). MMP activity was found in the endometrium, and was prominent at the implantation site compared with that of the interimplantation site, on Day 5 (Fig. 2, A–D). Subsequently, the activity was localized in a thin layer of stroma adjacent to the remaining luminal epithelium (Fig. 2E), decidual crypts, and the periphery of stroma adjacent to the myometrium (Fig. 2G) on Day 6, but disappeared thereafter (data not shown).
|
|
Gelatin Zymography and Western Blotting
To determine the types of MMPs contributing the gelatinolysis, whole uteri were analyzed by conventional gelatin zymography. The extracts from uteri at proestrus and on Gestational Days 1–5 formed three major bands at 105, 67, and 62 kDa, which were assigned as proMMP9, proMMP2, and MMP2, respectively, based on size (Fig. 3A). Gelatinolysis quantification clearly showed proMMP9 activity to increase dramatically after mating and then decline after Day 3 (Fig. 3B). During the estrus cycle, proMMP9 increased significantly in the metestrus phase (Fig. 3, C and D). The increased proMMP9 protein levels on Day 1 were demonstrated by Western blotting analysis (Fig. 4). On the other hand, proMMP2 and MMP showed no significant change during the estrus cycle. ProMMP2 increased only slightly from Day 1 to Day 5, while MMP2 expression increased on Day 2. These results indicate the marked increase in the gelatinolytic activity in the endometrium on Days 1 and 2 to be the result of increased proMMP9 levels.
|
|
Mmp9 Transcripts in Uterus
We next performed semiquantitative reverse transcription (RT)-PCR to examine whether the proMMP9 enhancement after mating was caused by up-regulation of Mmp9 transcription (Fig. 5). Unexpectedly, levels of Mmp9 mRNA were unchanged. On the other hand, Timp1 mRNA increased as pregnancy progressed, an observation that was consistent with those of previous studies [3, 14].
|
Male Factor in ProMMP9 Expression
Increased levels of proMMP9 protein without mRNA expression in the uterus might be of male origin. We analyzed uterine luminal fluids containing sperm and seminal plasma collected from the uterus on Day 1 after mating by gelatin zymography. Five or 15 µl of uterine luminal fluid (corresponding to approximately 5% or 15%, respectively, of total fluid recovered from a single mouse) contained fewer MMPs than 0.1 µl of uterine tissue extracts prepared after luminal flushing (Fig. 6). This result indicated uterine luminal fluid to not be the source of proMMP9 in uterine tissue on Day 1 of pregnancy.
|
Next, we focused on the involvement of seminal vesicle secretions. To this end, males without SV were generated and mated with normal females. These resulting pregnancies will be referred to as SV pregnancies hereafter. For comparison, pseudopregnancy was induced by mating with vasectomized males. As shown in Figure 7, MMP expressions in uterine tissue extracts were similar in normal pregnancy and pseudopregnancy. In contrast, no substantial increase in proMMP9 was observed in SV pregnancy.
|
Immunohistochemistry of MMP9
The temporal and spatial distributions of uterine MMP9 protein were investigated by immunohistochemistry (Figs. 8 and 9). MMP9 was detected in the glandular epithelium throughout the estrus cycle (Fig. 8, A–D) and in the luminal epithelium during estrus (Fig. 8B). MMP9 was detected in the stroma during metestrus (Fig. 8C). On Day 1 of pregnancy, intense MMP9 staining was observed in the serosa, in addition to weak staining at the luminal and glandular epithelia (Fig. 9A). Subsequently, the localization of MMP9 progressed inward and was concentrated in the stroma on Day 2 (Fig. 9G). Staining gradually diminished starting on Day 3, and was undetectable by Day 5 (Fig. 9, J–L). On the other hand, with SV pregnancy, only sparse staining was observed in the stroma on Day 1, while localization was more evident in luminal and glandular epithelia (Fig. 9D).
|
|
Contribution of Neutrophils to MMP9 Expression in Uterus
Seminal vesicle secretions reportedly initiate the postmating inflammatory cascade by up-regulating the synthesis and release of colony-stimulating factor 2 (CSF2), and the resulting accumulation of CSF2 in uterine fluid was accompanied by markedly increased recruitment and activation of macrophages, neutrophils, and eosinophils [16, 17]. Neutrophils were identified as being responsible for rapid secretion of proMMP9 because these cells release the preformed enzyme stored in granules [18]. Therefore, we investigated the colocalization of neutrophils or macrophages with MMP9 in serial sections. As shown in Figure 9, the distribution of neutrophils, but not that of macrophages, was identical to that of MMP9 on Days 1 and 2 of normal pregnancy and Day 1 of SV pregnancy. These MMP9-bearing cells were confirmed to be neutrophils by double-staining with anti-MMP9 and RB6-8C5 on Day 2 (Fig. 10).
|
Finally, the contribution of neutrophils to MMP9 activity in the Day 1 uterus was verified by in vivo depletion of neutrophils. Neutropenic mice were generated by RB6-8C5 administration 4 or 6 days before mating. As expected, proMMP9 expression was reduced in the treated mice (Fig. 11).
|
|
DISCUSSION |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
Whether Mmp9 mRNA is expressed in the preimplantation uterus is controversial, possibly due to differences in methods used for detection. In mice, Mmp9 mRNA was not detected by Northern blotting or by in situ hybridization in the endometrium on Days 1–4 [2] or on Days 4–8 [4], whereas it was detected on Days 1–5 [6] and on Day 8 [4] by RT-PCR. In the present study, Mmp9 mRNA was detected at low levels on Days 1–5 by RT-PCR, but was stable during this period. This result contradicts a previous report describing the expression of Mmp9 mRNA concomitant with MMP9 activity, which peaked on Days 1–2 of pregnancy [6]. Presumably, the MMP9 was, at least in part, of extrauterine origin, because zymographic MMP9 activity on Days 6–10 did not correlate with the transcript on Northern blotting [3].
A role for MMP9 in implantation has been suggested by these studies: first, a chemical inhibitor, 3-(N-hydroxycarbamoyl)-2(R)-isobutylpropionyl-L-tryptophan methylamide, which inhibits all types of MMPs, reduced decidual length and overall size in mice [3]; and second, doxycyclin produced a similar result in rats [7]. More specifically, mice deficient in the Mmp9 gene exhibited mild infertility, while neither the fetal nor the perinatal survival of the Mmp9-deficient mice was affected [19].
The present results indicate that the pro form of MMP9, or proMMP9, is most likely to be active in vivo. ProMMP9 is converted to MMP9 in vitro by various proteases such as MMP1, MMP2, MMP3, and MMP7, tissue kallikrein, and plasminogen activator [20], but MMP9 is not always found in tissues [21–23]. It has been suggested that proMMP9 acquires its activity by binding to ligands or to substrates without proteolytic activation [24]. In addition, binding to the plasma membrane of neutrophils enables proMMP9 to evade inhibition by TIMPs [25]. Therefore, proMMP9 in the endometrium would be active in its latent form even in the presence of abundant TIMPs in the uterus.
Neutrophils entered uterine tissues in response to SV secretions. This is consistent with an earlier report in which removing SVs from male mice reduced the number of postcoital neutrophils found in the endometrium [16]. As-yet-unknown factors in seminal plasma stimulate a response resembling the classical inflammatory cascade in rodents, pigs, rabbits, and humans [26]. The inflammatory response was characterized by extensive infiltration of activated neutrophils, eosinophils, and macrophages into the endometrial stroma in rodents [27–29]. Prostaglandin E is abundant in human seminal plasma and stimulates release of a proinflammatory cytokine, interleukin 8, from nonpregnant cervical explants, while it is undetectable in rodents [30]. Transforming growth factor beta 1 (TGFß1) is identified as the inflammation-inducing moiety in seminal plasma, and seminal TGFß1 initiates endometrial leukocyte infiltration by up-regulating epithelial cell expression of CSF2 [31]. However, intrauterine injection of TGFß1 resulted in the infiltration of macrophages and eosinophils but not of neutrophils as reported by Tremellen et al. [10], and consistently, TGFß1 did not enhance the MMP9 expression in endometrium by intrauterine injection (our unpublished observation). Other cytokines and chemokines are also implicated as mediators of macrophage and granulocyte recruitment and activation, including CSF1, regulated and normal T cell expressed and secreted (RANTES), macrophage inflammatory proteins 1 alpha and 1 beta, and monocyte chemotactic protein 1 [31–33]. These factors may synergize with TGFß1 in seminal plasma for neutrophil recruitment.
A series of studies have suggested that seminal plasma plays a role in pregnancy outcome: mating with an SV-deficient male reduced the litter size in rodents [34], pretreatment of the uterus with seminal plasma increased the litter size and the implantation rate in pigs [35], and insemination around the time of embryo transfer facilitated implantation and embryonic development in the setting of in vitro fertilization/assisted pregnancy in humans [36]. Neutrophils are attracted by seminal plasma, and are involved in tissue remodeling and immunoregulation. ProMMP9 binding to alphaMbeta2 integrin in neutrophils is reportedly necessary for their recruitment to inflammatory sites, probably via degradation of integrin-directed matrix bonds [37]. Taken together, these observations indicate that MMP9 from neutrophils in the preimplantation uterus are likely to play an important role in fertilization via remodeling of the uterine architecture, which enhances endometrial receptivity to the conceptus.
|
FOOTNOTES |
|---|
2 Correspondence: Yoshinao Wada, Department of Molecular Medicine, Osaka Medical Center and Research Institute for Maternal and Child Health, 840 Murodo-cho, Izumi, Osaka 594-1101, Japan. FAX: 81 725 56 1220; waday{at}mch.pref.osaka.jp
Received: 14 December 2004.
First decision: 16 January 2005.
Accepted: 15 March 2005.
|
REFERENCES |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
Related articles in Biol Reprod:
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
