Molecular Human Reproduction, Vol. 6, No. 5, 479-485,
May 2000
© 2000 European Society of Human Reproduction and Embryology
Pregnancy |
Matrix metalloproteinases -2 and -9 and their endogenous tissue inhibitors in fetal membrane repair following fetoscopy in a rabbit model
1 Centre for Surgical Technologies, Faculty of Medicine, Catholic University of Leuven, Minderbroedersstraat 17, B-3000, Leuven, 2 Department of Obstetrics and Gynaecology and 3 Laboratory of Experimental Gynaecology, University Hospital `Gasthuisberg' Herestraat 49, B-3000 Leuven, Belgium and 4 Department of Obstetrics and Gynaecology, Centre for Reproductive Biology, University of Edinburgh, 37 Chalmers Street, Edinburgh EH3 9ET, UK
Abstract
The cellular mechanisms underlying fetal membrane repair are poorly understood. Matrix metalloproteinases (MMP) and the endogenous tissue inhibitors of metalloproteinases (TIMP) play a key role in the control of turnover of extracellular matrix in fetal membranes at normal parturition and preterm prelabour rupture of the fetal membranes (PPROM). The time course of secretion of MMP-2 (72 kDa, gelatinase A) and MMP-9 (92 kDa, gelatinase B) and TIMP into extra-embryonic coelomic, allantoic and amniotic fluids in a rabbit model was examined. Furthermore, to evaluate their role in fetal membrane repair, the changes induced by fetoscopy at mid-gestation (23 days; gestation length is 32 days) were investigated. Zymography showed predominantly secretion of latent MMP-2 at 18, 23 and 30 days of gestation in all gestational compartments. Reverse zymography detected a broad range of TIMP activity with molecular weights of 27–30 kDa (TIMP-1, glycosylated TIMP-3 and TIMP-4), 24 kDa (unglycosylated TIMP-3) and 21 kDa (TIMP-2). Following fetoscopy, both MMP-2 and TIMP increased significantly in amniotic fluid and extra-embryonic coelomic fluid, but not in allantoic fluid, as demonstrated by densitometric analyses. These findings indicate a modulating role for MMP and TIMP in the repair processes following a surgically induced fetal membrane defect.
fetal membranes/fetoscopy/matrix metalloproteinases/rabbit/TIMP
Introduction
Invasive techniques with direct access to the fetus and placenta are increasingly being used in fetal medicine. By definition, this involves the creation of a lesion in the myometrium, decidua and fetal membranes. In a minority of cases, a temporary or even persistent fluid leak may occur, which was earlier denominated `iatrogenic' preterm prelabour rupture of the fetal membranes (iPPROM) (Deprest and Gratacos, 1999). The risk for iPPROM has been estimated ~1% following amniocentesis at 15–16 weeks of gestation (Tabor et al., 1986
; Reece, 1997). For more invasive procedures, the risk increases significantly to become clinically important. For instance, fetoscopic Nd:Yag laser coagulation of chorionic plate vessels for twin-to-twin transfusion syndrome is complicated by iPPROM in up to 10% of the cases (Ville et al, 1998). For more complex procedures, such as fetoscopic umbilical cord ligation, the risk may be >30% (Deprest et al., 1998). If operative fetoscopy is to play a role in modern fetal medicine, the problem of iPPROM needs to be addressed.
For this purpose, a rabbit model was recently developed to study the effects of iatrogenic fetal membrane defects and to compare different surgical closure techniques to seal these lesions (Papadopoulos et al., 1998; Deprest et al., 1999; Gratacós et al., 1999a,b). As an additional step, it was hoped to obtain information regarding the cellular and enzymatic mechanisms underlying fetal membrane tissue repair, which remain largely unknown.
The matrix metalloproteinases (MMP) are important enzymes in tissue remodelling. They are capable of breakdown of extracellular matrix (ECM) with a broad range of substrate specificities. This matrix breakdown is one part of tissue remodelling, which also clearly involves matrix deposition, with this MMP activity reflecting local remodelling. MMP are involved in many aspects of reproductive function (Hulboy et al., 1997; Nagase and Woessner, 1999). They play a role in the development of the feto-maternal interface in early gestation and in the normal breakdown of the membranes during parturition at term (Riley et al., 1999a,b). The pathophysiology of `spontaneous' PPROM also indicates the importance of ECM remodelling in the inter-layer connective tissue of the fetal membranes by these enzymes (Parry and Strauss, 1998). The equilibrium between the collagenolytic MMP and their endogenous tissue inhibitors of metalloproteinases (TIMP) present in this layer can be unbalanced in the presence of infection and other conditions known to be risk factors for PPROM (Locksmith et al., 1999). Furthermore, it has been shown that MMP-2 (gelatinase-A) and MMP-9 (gelatinase-B) specifically break down collagen IV, a major component of basement membrane in fetal membranes (Malak et al., 1993). Although most work documenting the role of MMP and TIMP in fetal membrane remodelling has been done on human specimens from patients with spontaneous (i.e. not traumatic or iatrogenic) PPROM, it is plausible that these enzyme systems are also involved in repair mechanisms following a surgically induced fetal membrane defect.
In the present study, the secretion of MMP-2, MMP-9 and TIMP in the gestational cavities of the rabbit at different time points in gestation was measured. Secondly, their role was evaluated in the repair of the fetal membranes by measuring the secretion after causing a standardized surgical trauma to the fetal membranes at mid-gestation.
Materials and methods
Sample and tissue collection
Rabbits have a mean gestation of 32 days and mean litter size of eight. Each fetus lies within the amniotic cavity facing the placenta and its cord is attached antimesometrial to the choriovitelline circulation (Ramsey, 1982). In this species, the amniotic, extra-embryonic coelomic and allantoic cavities represent distinct compartments at mid-gestation (Figure 1), which can be identified and sampled using previously described techniques (Gratacos et al., 1999a). All animals were housed 2 days prior to surgery with ambient daylight and at room temperature. Animals undergoing fetoscopy were allowed only water during the last 12 h prior to this first intervention.
|
Normal pregnancy
To establish patterns of secretion during the second half of pregnancy, three different groups (18 days, 23 days and 30 days of gestation) of time-dated pregnant New Zealand rabbits were studied (n = 5 animals in each group). The rabbits were euthanized with a mixture of embutramide 200 mg mebezonium 50 mg tetracaïn hydrochloride 5 mg i.v. (T61®; Hoechst, Brussels, Belgium) and a midline laparotomy was performed to expose the pregnant uterus. Interventions on the uterus and membranes were done using an operating microscope (Carl Zeiss, Oberkochen, Germany; magnification x5–25) and micro-instruments. One gestational sac was selected in each animal, excluding the sacs at either the ovarian or cervical end of the uterine horns. A 2–3 mm myometrial incision was made on the antimesometrial side. Once the chorion was visualized, this myometrial incision was extended over two-thirds of the length of the gestational sac. The extra-embryonic coelomic fluid pockets were identified by microscopic visualization and a 22 G needle and syringe was used to aspirate a minimum volume of 0.3 ml of extra-embryonic coelomic fluid. The chorion was opened surgically, revealing the amnion, and a sample of amniotic fluid was retrieved in the same way. Finally, the amnion was further opened and a fetus exteriorized, so that the allantoic cavity could be accessed to sample allantoic fluid. Extra-embryonic coelomic fluid could not be obtained at 30 days when the coelomic space is obliterated due to close juxtaposition of amnion and chorion (Ramsey, 1982
).
Surgical trauma
To evaluate the effect of fetal membrane trauma and repair, time-dated New Zealand rabbits (n = 10) underwent fetoscopy at 23 days of gestation. A second operation for fluid sampling was then performed at 27 (n = 5) and 30 (n = 5) days of gestation. After premedication (ketamine 50 mg/kg i.m.; Ketalin®; Apharmo, Arnhem, Netherlands, and promazinum hydrochloridium 5 mg/kg i.m.; Prazine®; Libamedi, Brussels, Belgium), and antibiotic prophylaxis (Penicillin G 300 000 IU i.m.; Smith Kline Beecham, Brussels, Belgium), animals were anaesthetized (halothane 2–5% in O2; 1 l/min gas flow rate). Maternal heart rate and oxygen saturation were monitored using a pulse oximeter (Nellcor® N-20P; Nellcor Inc., Haasrode, Belgium). The animals were placed in the supine position and the abdomen was shaved under continuous vacuum aspiration, disinfected with povidone iodine (Iso-Betadine®; Asta Medica, Brussels, Belgium) and surgery was performed under sterile conditions. A midline abdominal incision was made to expose the pregnant uterus. Gestational sacs were counted and numbered and a maximum of one in three amniotic sacs, with exclusion of the gestational sacs at either the ovarian or cervical end of the uterine horn, were randomly assigned to the treatment (fetoscopy) or control group. The technique of `open access' fetoscopy in the rabbit has been described in detail previously (Gratacós et al., 1999a). Briefly, this involves a 2–3 mm myometrial incision to visualize and open the chorion. Gentle uterine pressure was given to make the amniotic membrane bulge through the chorionic incision and the amniotic sac was entered with the 2.0 mm needle under microscopic visualization. Fetoscopy was performed using a short 1.2 mm 10 000 pixel 0°-fibre endoscope (Karl Storz, Tüttlingen, Germany) and correct intra-amniotic position was confirmed. After withdrawal of the fetoscope and its needle, the access site was closed with a single myometrial suture of polypropylene monofil 6/0 (Prolene®; Ethicon, Dilbeek, Belgium), the uterus repositioned, and the abdomen closed in layers with polyglactine 3/0 (Vicryl®; Ethicon) for the fascia and intracutaneous nylon 2/0 suture (Ethilon®; Ethicon) for the skin. Postoperative uterine relaxation consisted in medroxyprogesterone acetate (4.5 mg i.m. Depo-Provera®; Pharmacia Upjohn, Puurs, Belgium). The animals were housed for the next 7 days in the same conditions as prior to surgery. The `second look' operations for fluid sampling were performed at 27 (n = 5) and 30 (n = 5) days of gestation, following an identical procedure as described above. From the 10 does that were operated upon in this series of experiments, one died of unknown reasons, and in two further rabbits all fetuses were dead and macerated at the second operation without obvious reason. These does were excluded from further analysis. Furthermore, in the treated group, only the sacs where membrane repair had occurred (fetus being alive, chorion and amnion sealed) were sampled, as well as the adjacent gestational sacs, as a control. Membrane sealing occurred in 13 of the total of 28 gestational sacs that underwent fetoscopy (seven at 27 days and six at 30 days of gestation) resulting in a repair rate of 46%. Fluid samples were stored at –20°C until processing.
All animals were treated in accordance to current guidelines on animal welfare and the experiments were approved by the Ethical Committee for animal experiments of the Faculty of Medicine of the Katholieke Universiteit Leuven, Belgium.
Measurement of MMP-2 and -9 by zymography
Gelatin zymography was used to detect gelatinase (MMP-2 and MMP-9) activities in amniotic, extra-embryonic coelomic and -allantoic fluids using methods described previously (Rawdanowicz et al., 1994) with minor modifications (Riley et al., 1999a). Samples (2.5 µl) were separated by sodium dodecyl sulphate (SDS)–polyacrylamide gel electrophoresis (PAGE; 7.5% gels; Minigel apparatus; Bio-Rad, Hemel Hempstead, Herts, UK) using gels containing gelatin (1 mg/ml) in non-reducing conditions. Gels were washed [2.5% (v/v) Triton X-100 in Tris-buffered saline (source of all chemicals from Sigma Chemical Co., St Louis MO, USA, unless specified otherwise) and incubated in digestion buffer (200 mmol/l NaCl, 50 m mol/l Tris, 5 mmol/l CaCl2, 1 µmol/l ZnCl2, 0.02% (v/v) Brij-35, pH 7.6] for 18 h at 37°C. Gels were then stained (0.5% Coomassie blue R250 in 30% methanol/10% glacial acetic acid in H2O; 3 h at 23°C), then destained (staining solution omitting Coomassie blue) to reveal discrete areas where degradation of gelatin by gelatinases was localized. Molecular weights were established by comparison with standard markers (Bio-rad). Human amniotic fluid collected at term during labour was used as a positive control, which clearly demonstrated the latent forms of MMP-2 and MMP-9 (Riley et al., 1999a), with a further band at 120 kDa which is probably a lipocalin–proMMP-9 complex (Kolkenbrock et al., 1996). Zymography permits detection of the latent as well as the active forms of MMP because exposure to SDS results in a conformational change associated with activation without the change in molecular weight associated with the endogenous activation by cleavage of the proportion of latent MMP.
Detection of TIMP activity by reverse zymography
The activity of TIMP was detected by reverse zymography as described previously using a commercially available kit (University Technologies Inc., Calgary, Canada) with some minor adaptations (Hampton et al., 1994; Riley et al., 1999b). Samples of amniotic, extra-embryonic coelomic and allantoic fluids (2.5 µl) were separated according to molecular weight by PAGE (12% gels) containing gelatin (1 mg/ml) and a preparation of MMP-2 (conditioned medium from BHK-21 cells which constitutively express MMP-2; University Technologies Inc.) using a minigel apparatus. Gels were washed [wash buffer; 50 mmol/l Tris, 5 mmol/l CaCl2, 2.5% (v/v) Triton X-100; for 2.5 h at 23°C], incubated in reverse zymography digestion buffer (wash buffer excluding Triton X-100; for 17 h at 37°C), then stained [staining buffer; 0.5% Coomassie blue R250 (Bio-Rad) in 30% methanol/10% glacial acetic acid in H2O] and destained (staining buffer omitting the Coomassie blue). The inhibitory activity of TIMP with substrate degradation by MMP-2 appeared as dark bands against a lighter background. To confirm identity of these bands, TIMP were characterized by comparison with molecular weight markers (Bio-Rad), control standard solutions containing mouse TIMP-1, TIMP-2 and the glycosylated and unglycosylated forms of TIMP-3 (University Technologies Inc.), recombinant human TIMP-2 (Calbiochem, Nottingham, UK), and a sample of human amniotic fluid collected at term during labour that has been characterized previously (Riley et al., 1999b). Analysis of all samples by PAGE with the gelatin substrate omitted demonstrated no significant detectable underlying protein staining at the molecular weights at which TIMP were observed, demonstrating the specificity of the TIMP activity.
Data analysis
Fluid samples were examined by volume (not protein content), to reflect concentrations in vivo. The presence of TIMP in gels was analysed by transmission densitometry (G-700 densitometer, Bio-Rad) and relative intensities of equal areas were compared using integrated software (Quantity One®; Bio-Rad). All densitometric assessments of gels were performed within the optical density range of sensitivity (i.e. non-saturated pixels). Comparisons were only made between gels that had been run under identical conditions (i.e. same electrophoresis cassette and run, buffers, stains and incubation periods) with a control sample on each gel to permit normalization of data. Densitometric readings were analysed by analysis of variance (ANOVA) or Student's t-test, as appropriate. P < 0.05 was regarded as significant.
Results
MMP-2, MMP-9 and TIMP secretion into the extra-embryonic spaces
Gelatin zymography detected MMP-2 activity, predominantly as the latent form (72 kDa) in amniotic, allantoic and extra-embryonic coelomic fluid during the second half of pregnancy (Figure 2). The secretion of MMP-2 in the amniotic and allantoic fluid increased significantly (P < 0.05) with advancing gestation. There was no change in MMP-2 in extra-embryonic coelomic or allantoic fluid during this time period. MMP-9 activity, also in its latent form (92 kDa) was present in amniotic, allantoic and extra-embryonic coelomic fluid, but only in barely detectable amounts (Figure 2).
|
Reverse zymography detected a broad range of TIMP activity in amniotic, allantoic and extra-embryonic coelomic fluids (Table I
). These TIMP activities aligned with standards to TIMP-1 (30 kDa) and TIMP-2 (21 kDa), and TIMP-3 (glycosylated form of 30 kDa; unglycosylated form of 24 kDa). There were no differences in TIMP activities in amniotic, allantoic or extra-embryonic coelomic fluid between the different gestational ages. Only low amounts of TIMP-2 were detectable as a band at 21 kDa. In amniotic fluid, this band was observed at 18 days of gestation, but could not be distinguished from the background and hence was undetectable at 23 and 30 days of gestation.
|
Effect of fetoscopic membrane trauma and repair on secretion of MMP and TIMP
Zymography showed that the predominant gelatinase activity present in the amniotic fluid was due to MMP-2 (latent form; 72 kDa) (Figure 3a,b
). Some gelatinase activity was also detected at high molecular weights (>200 kDa) but the identity of these sources is not known. Densitometric analysis of this MMP-2 activity showed a significant (P < 0.05) increase following fetoscopy in amniotic fluid samples, when compared with matching controls at both day 27 and day 30 (Figure 3c). Furthermore, a significant (P < 0.05) increase for latent MMP-2 secretion was also found in the extra-embryonic coelomic fluid (Figure 4a,c). There was no change in MMP-2 secretion into the allantoic fluid (Figure 4b,c). The EECF was disrupted surgically and demonstrated an increase in MMP-2 concentrations. The allantoic cavities associated with the same fetuses were not disrupted during the surgical procedure and did not show an increase in MMP-2 activity.
|
|
Reverse zymography showed the presence of a wide range of TIMP activities in amniotic fluid collected on day 27 and day 30 of gestation. Densitometric analysis of these TIMP activities showed a significant (P < 0.05) increase at 24 kDa (corresponding to the unglycosylated form of TIMP-3) and 27–30 kDa (corresponding to TIMP-1, the glycosylated form of TIMP-3 and possibly TIMP-4) following fetal membrane trauma at both day 27 and 30 of gestation, when compared to matching control samples obtained from an adjacent fetus (Figure 5
). A similar significant (P < 0.05) increase following surgical trauma to the fetal membranes was observed in the extra-embryonic coelomic fluid (Figure 6a), but not in the allantoic fluid (Figure 6b,c). The chorion is disrupted surgically and demonstrates an increase in TIMP activity. The allantoic cavities associated with the same fetuses are not disrupted during the surgical procedure and do not show an increase in TIMP activity.
|
|
Discussion
These studies demonstrate the presence of MMP-2 and -9 and several isoforms of TIMP in the amniotic, extra-embryonic coelomic and allantoic fluid of the rabbit from day 18 of gestation to term. The predominant gelatinase present was the latent form of MMP-2, detected in increasing amounts as gestation proceeds. The fluid compartments showed TIMP activities corresponding to TIMP-1, glycosylated and unglycosylated TIMP-3 at all studied time points in gestation. TIMP-2 was almost undetectable in amniotic fluid.
In women, MMP and TIMP activity is present in the intrauterine compartments in amniotic and extra-embryonic coelomic fluids (Riley et al., 1999a,b). There are similarities in intrauterine development between the human at the end of the first trimester and the rabbit due to its relative immaturity in late gestation, as examined in these current studies. The distributions between the different gestational compartments, especially for MMP-2 and TIMP, suggest a comparable secretion and compartmentalization between rabbit and human.
Following surgical trauma and repair of the myometrial defect, earlier studies in the rabbit showed that about half of the traumatized sacs will reseal (Deprest et al., 1999). A similar figure (46%) was obtained in this study. Sufficient fluid samples could only be obtained from the different gestational compartments in these sacs where repair of the fetal membranes had occurred with re-accumulation of fluid. In these samples, consistent changes in secretion of MMP-2 and TIMP in amniotic and extra-embryonic coelomic fluid were observed both 4 and 7 days after surgical trauma, suggesting an important role for the gelatinases and their inhibitors in the fetal membrane repair processes following a standardized surgical trauma. The absence of any change observed in fluids sampled from the allantoic cavity, which was not entered surgically during fetoscopy, provided a most useful internal control. It demonstrated that secretion of MMP and TIMP with subsequent tissue repair following the surgical lesion was specific to the cavities that had been traumatized, and not adjacent compartments, despite being attached to the same conceptus.
MMP and TIMP play an important role in adult wound repair (Stricklin et al., 1993; Vaalamo et al., 1999), and their involvement in the breakdown of the ECM preceding rupture of the fetal membranes during parturition at term is now well documented (Parry and Strauss, 1998; Riley et al., 1999b). In both situations, turnover of the ECM is modulated via the MMP and TIMP equilibrium. These factors are controlled at the level of biosynthesis, secretion, activation and inhibition by cytokines released by resident cells, infiltrating neutrophils and other inflammatory cells (Flour, 1998; Thomson et al., 1999). The role of gelatinases and their regulation in fetal membrane repair has not been previously studied, but the present findings suggest a similar remodelling response to trauma.
The determination of absolute values for MMP and TIMP activities is difficult. Only relative intensities can be determined using the densitometric analyses. It is also impossible to identify precisely which TIMP isoform is measured. This would only be possible when antibodies become available that have not been raised in rabbits. For instance, it is currently not possible to establish the presence of TIMP-4 by reverse zymography as it has similar molecular weight to TIMP-1 and glycosylated TIMP-3. While the contribution of various isoforms needs further study, the overall relative increases in MMP and TIMP with trauma compared to controls are consistent.
The rabbit model has some limitations. It has a different type of placentation (haemodichorial) and possesses a slightly different relationship between amnion and chorion, with the extra-embryonic coelomic space persisting during the second half of pregnancy (Ramsey, 1982). This reflects the short gestation and relative immaturity of the rabbit neonate. However, this model has been shown to be valuable in the study of fetal membrane dynamics. Indeed, fetoscopy performed in this model reproduces persistence of fluid leak, chronic oligohydramnios and pulmonary hypoplasia, the main complications of iPPROM in humans (Gratacos et al., 1999a). Strategies have also been developed for closure of the fetoscopic access in this model (Deprest et al., 1999; Gratacos et al., 1999b). The current data support the theory that the biochemical basis of this sealing process is likely to be one of tissue remodelling involving MMP.
In summary, these findings demonstrate that MMP-2 and TIMP secretion is increased in the gestational compartments following trauma. This suggests that control of MMP activity, which reflects an increase in tissue remodelling of the extracellular matrix and an important role in the repair process in the fetal membranes. Further insight in the regulation of the MMP and TIMP secretion will offer the prospect of developing surgical and/or pharmacological means to modulate fetal membrane repair in order to limit the clinical consequences of PPROM.
Acknowledgments
This work is supported by the Biomed Programme of the European Commission (EUROFOETUS, BMH4-CT-97-2383), `Co-financiering van de Vlaamse Gemeenschap (COF 98/012)' en Fonds voor Wetenschappellijk Onderzoek Vlaanderen (FWO-G-0153-00). R.D. is recipient of a predoctoral research fellowship from the Faculty of Medicine, Katholieke Universiteit Leuven. E.G. is a recipient of a postdoctoral research fellowship from the European Commission. Work in S.C.R.'s laboratory is funded partly by the Scottish Hospital Endowments Research Trust (project no. 1389). The other members from the Eurofoetus group (Y.Ville, M.Dommergues, K.Hecher, T-H.Bui, U.Nicolini) are thanked for their efforts in helping setting up the Eurofoetus project.
Notes
5 To whom correspondence should be addressed at: Department of Obstetrics and Gynaecology, U.Z. Leuven Herestraat 49, B-3000 Leuven, Belgium 
References
Deprest J.A. and Gratacós E. (1999) Obstetrical endoscopy. Curr. Opin. Obstet. Gynecol., 11, 195–203.[Medline]
Deprest, J.A., Van Schoubroeck, D., Van Ballaer, P.P. et al. (1998) Fetoscopic cord ligation. Eur. J. Obstet. Gynecol. Reprod. Biol., 81, 157–164.[Web of Science][Medline]
Deprest, J.A., Papadopoulos, N.A., Decaluwé, H. et al. (1999) Closure techniques for fetoscopic access sites in the rabbit at midgestation. Hum. Reprod., 14, 1730–1734.
Flour, M. (1998) Wound re-epithelialisation. J. Surg. Pathol., 3, 9–18.
Gratacós, E., Yamamoto, H., Papadopoulos, N.A. et al. (1999) The midgestational rabbit as a model for the creation of membrane defects following needle-fetoscopy. Am. J. Obstet. Gynecol., 180, 1263–1267.[Web of Science][Medline]
Gratacós E., Wu, J., Yesildaglar, N. et al. (2000) Successful sealing of fetoscopic access sites with collagen plugs in the rabbit model. Am. J. Obstet. Gynecol., 182, 142–146.[Web of Science][Medline]
Hampton, A.L., Butt, A.R., Riley, S.C. et al. (1995) Tissue inhibitors of metalloproteinases in endometrium of ovariectomised, steroid-treated ewes and during the estrous cycle and early pregnancy. Biol. Reprod., 53, 302–311.[Abstract]
Hulboy, D.L., Rudolph, L.A. and Matrisian, L.M. (1997) Matrix metalloproteinases as mediators of reproductive function. Mol. Hum. Reprod., 3, 27–45.
Kolkenbrock, H., Hecker-Kia, A., Orgel, D. et al. (1996) Progelatinase B forms from human neutrophils: complex formation of monomer/lipocalin with TIMP-1. Biol. Chem., 377, 529–533.[Web of Science][Medline]
Locksmith, G.J., Clark, P., Duff, P. et al. (1999) Amniotic fluid metalloproteinase-9 levels in woman with preterm labour and suspected intra-amniotic infection. Obstet. Gynecol., 94, 1–6.[Web of Science][Medline]
Nagase, H. and Woessner, J.F. (1999) Matrix metalloproteinases. J. Biol. Chem., 274, 21491–21494.
Malak, T.M., Ockleford, C.D., Bell, S.C. et al. (1993) Confocal immunofluorescence localisation of collagen types I, III, IV, V and VI and their ultrastructural organisation in term human fetal membranes. Placenta, 14, 385–406[Web of Science][Medline]
Papadopoulos, N.A., Van Ballaer, P.P., Ordoñez, J.L. et al (1998) Fetal membrane closure techniques after hystero-amniotomy in the rabbit model. Am. J. Obstet. Gynecol., 178, 938–942.[Web of Science][Medline]
Parry, S. and Strauss, J.F. (1998) Premature rupture of the fetal membranes. N. Engl. J. Med., 338, 663–670.
Ramsey, E.M. (1982) The Placenta: Human and Animal. Praeger, New York, p. 107.
Rawdanowicz, T.J., Hampton, A.L., Nagase, H. et al. (1994) Matrix metalloproteinase production by cultured endometrial stromal cells: identification of interstitial collagenase, gelatinase-A, gelatinase-B, and stromelysin-1 and their differential regulation by interleukin-1
and tumor necrosis factor . J. Clin. Endocrin. Metab., 79, 530—536.[Abstract]
Reece, A.E. (1997) Early and midtrimester genetic amniocentesis. Safety and outcome. Fetal Diagn. Ther., 24, 71—81.
Riley, S.C., Leask, R., Chard, T. et al. (1999a) Secretion of matrix-metalloproteinase-2, matrix metalloproteinase-9 and tissue inhibitor of metalloproteinases into the intrauterine compartments during early pregnancy. Mol. Hum. Reprod., 5, 376–381.
Riley, S.C., Leask, R., Denison, F.C. et al. (1999b) Secretion of tissue inhibitors of metalloproteinases by human fetal membranes, decidua and placenta at parturition. J. Endocrinol., 162, 351–359.[Abstract]
Stricklin, G.P., Li, L., Jancic, V. et al. (1993) Localisation of mRNA's representing collagenases and TIMPs in sections of healing human burn wounds. Am. J. Pathol., 143, 1657–66.[Abstract]
Tabor, A., Madsen, M., Obe, E.B. et al. (1986) Randomised controlled trial of genetic amniocentesis in 4606 low risk women. Lancet, i, 287–293.
Thomson A.J., Telfer, J.F., Young, A. et al. (1999) Leukocytes infiltrate the myometrium during human parturition: further evidence that labour is an inflammatory process. Hum. Reprod., 14, 229–236.
Vaalamo, M., Leivo, T. and Saarialho-Kere, U. (1999) Differential expression of tissue inhibitors of metalloproteinases (TIMP-1, -2, -3, and-4) in normal and abnormal wound healing. Hum. Pathol., 30, 95–802.
Ville, Y., Hecher, K., Gagnon, A. et al. (1998). Endoscopic laser coagulation in the management of severe twin-to-twin-transfusion syndrome. Br. J. Obstet. Gynaecol., 105, 446–453.[Web of Science][Medline]
Submitted on December 10, 1999; accepted on February 28, 2000.
CiteULike 
Connotea 
Del.icio.us What's this?
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||