Mol. Hum. Reprod. Advance Access originally published online on August 25, 2006
Molecular Human Reproduction 2006 12(10):633-641; doi:10.1093/molehr/gal072
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
Identification of a calcium-dependent matrix metalloproteinase complex in rat chorioallantoid membranes during labour
1Direccion de Investigacion, 2Departamento de Bioquimica, Instituto Nacional de Perinatologia Isidro Espinosa de los Reyes, 3Departamento de Bioquimica, Facultad de Medicina, Universidad Nacional Autonoma de Mexico, and 4Departamento de Microscopia Electronica, Instituto Nacional de Perinatologia Isidro Espinosa de los Reyes, Mexico City, Mexico
5 To whom correspondence should be addressed at: Instituto Nacional de Perinatologia Isidro Espinosa de los Reyes, Montes Urales 800, Lomas de Virreyes, Mexico, D.F. 11000, Mexico. E-mail: fvadillo{at}servidor.inper.edu.mx
 |
Abstract |
|---|
 Top Abstract IntroductionMaterials and methodsResultsDiscussionReferences |
|---|
The induction of the expression of matrix metalloproteinases (MMPs) and their extracellular activation are key processes in connective tissue degradation in the chorioallantoid membrane during rat labour. However, the regulatory mechanisms remain largely unknown. Here, we report the identification of a calcium-dependent high molecular weight complex composed of MMP-9, MMP-3, MMP-2, tissue inhibitor of metalloproteinase 1 (TIMP-1) and TIMP-2, identified by zymography and western blotting. Molecular sieve chromatography confirmed the presence of a complex of MMPs and TIMPs with an exclusion volume >670 kDa. Differential scanning calorimetry of the complex confirmed the existence of a macromolecular complex that unfolds with a broad transition; it is denatured over a wide range of temperatures and has a Tm of 72°C in the presence of Ca2+. When denatured in the absence of Ca2+, there were at least eight transitions with Tms that corresponded to pro-MMP-9, MMP-9, pro-MMP-3, MMP-3, pro-MMP-2, MMP-2, TIMP-1 and TIMP-2. Co-localization of the same molecular components was demonstrated by confocal microscopy using cell-depleted chorioallantoid membranes. The assembly and disassembly of the complex can be reproduced at physiological concentrations of Ca2+. This complex provides a potential mechanism for the enzymatic regulation of MMPs, which may participate in connective tissue degradation leading to the rupture of the fetal membranes during labour.
Key words: chorioallantoid/extracellular matrix/matrix metalloproteinases/TIMP
|
Introduction |
|---|
|
TopAbstractIntroductionMaterials and methodsResultsDiscussionReferences |
|---|
The amnion and chorion are tissues with high contents of type I, III and IV collagens, which provide tensile strength to the fetal membranes (Bell and Malak, 1997
; Bryant-Greenwood, 1998). These undergo physical modifications at the end of pregnancy (McLaren et al., 1999) because of the loss of their tensile characteristics (Paavola et al., 1995). Normally, fetal membrane rupture shows structural alterations in the extracellular matrix, associated with the apoptosis of the amniotic epithelium (Lei et al., 1996). These mechanisms favour the process of birth immediately after fetal membrane disruption and are associated with the actions of extracellular matrix metalloproteinases (MMPs) (Lei et al., 1995; Xu et al., 2002). These comprise a family of 28 enzymes capable of degrading the structural macromolecules of connective tissues (Fata et al., 2000).
MMPs vary in molecular weight (MW), structural domain organization (Sang and Douglas, 1996), substrate specificity (Bode et al., 1999) and mechanisms of regulation (Nagase and Woessner, 1999). They show conserved protein domains: a signal peptide, a propeptide and a catalytic domain. All MMPs, except for MMP-7 and MMP-26, contain an additional carboxy terminal haemopexin-like domain (Li et al., 1995). MMP-2 and MMP-9 have a gelatin-binding fibronectin domain composed of three fibronectin repeats, before the active-site domain. Fibronectin repeats in gelatinases bind gelatin, laminin and type I and IV collagens (Allan et al., 1995). MMP-9 has a collagen type V-like domain before the haemopexin-like domain. MMPs are secreted as zymogens with a propeptide region that must be removed during activation (Springman et al., 1990; Bannikov et al., 2002). The biochemical mechanism for initiating the catalytic activity of mammalian MMPs is largely unknown but seems to occur via four different mechanisms: (i) the extracellular activation of MMPs by non-MMP proteins (Murphy et al., 1999); (ii) via activation by other MMPs (Imai et al., 1995); (iii) the intracellular activation of MT1-MMP (MMP-14) and MMP-11 by furin (Pei and Weiss, 1995); and (iv) MT-MMP-associated activation of MMP-2 (Cao et al., 1995). The inhibition of MMPs is carried out by a family of specific endogenous low MW MMP inhibitors termed tissue inhibitor of metalloproteinases (TIMPs) and by non-specific inhibitors such as 
2-macroglobulin (Brew et al., 2000; Baker et al., 2002). Others have shown that MMPs may be activated directly by reactive oxygen species (Maeda et al., 1998); however, there is no general agreement on this.
There is specific induction of MMP-9, MMP-2 and MMP-1 in fetal membranes associated with labour in humans, rhesus monkeys and rats (Lei et al., 1995; Vadillo-Ortega et al., 1995, 2002; Maymon et al., 2000). There is also the expression of TIMP-2 and TIMP-1 in human and rat fetal membranes (Roswit et al., 1992; Rowe et al., 1997). The precise mechanisms of activation of these enzymes during labour still evade investigators. Meraz-Cruz et al. (2002) studied changes in MMP levels in rat amniotic fluid and fetal membrane extracts from days 15 to 21 of gestation and demonstrated the presence of MMP-9 solely at the end of gestation; by contrast, MMP-2 was present in low concentrations throughout, peaking at the end of gestation.
The rupture of the chorioallantoid membranes can be used as a natural model for the precise timing of the activation of different MMPs associated with normal labour. Here, we identified a complex of MMPs extracted from rat chorioallantoid membranes during labour; assembly and disassembly were dependent on extracellular Ca2+ concentrations.
|
Materials and methods |
|---|
|
TopAbstractIntroductionMaterials and methodsResultsDiscussionReferences |
|---|
Animals
The Internal Review Board of the Instituto Nacional de Perinatologia Isidro Espinosa de los Reyes, Mexico, authorized this protocol for experiments in animals (Approval Number 06071) in accordance with the guidelines of the National Research Council (1996)
.
Female Wistar rats obtained from our own colony, weighing 200–250 g, were used. We determined the phase of the ovarian cycle using the morphological changes of vaginal epithelium evaluated by vaginal smears. Mating was confirmed by the presence of sperm in vaginal smears the morning after females had been caged overnight with males, and this was considered the first day of gestation. The rats were synchronized with light/dark cycles of 12 h/12 h that produced very constant pregnancy lengths of 21 days. All animals were killed by cervical dislocation 12 h before the estimated time of delivery, and chorioallantoid membranes were separated manually for the extraction of membrane components (Meraz-Cruz et al., 2002).
Membrane extract preparation
Tissues were suspended in a buffer containing 50 mM Trizma Tris base (pH 7.4), 150 mM NaCl, 20 mM CaCl2, homogenized using a Polytron (Brinkman Instruments, Westbury, NY, USA) and centrifuged at 10 000 g for 15 min. The supernatants of this preparation are referred to as Ca2+-containing extracts. Pellets were washed extensively with the same solution, and once zero protein was present in the supernatant, they were treated with a buffer of 50 mM Trizma Tris base (pH 7.4), 150 mM NaCl, 40 mM EDTA. Eluted protein from the pellets was dialysed extensively against a buffer containing 50 mM Trizma Tris base (pH 7.4), 150 mM NaCl, to eliminate the EDTA. This preparation is referred to as the EDTA extract, and it was used as the source for the biochemical initial characterization of the MMPs present in chorioallantoid membranes (see next section).
Protein concentration was determined using an established method (Bradford, 1976), with bovine serum albumin as the standard. All procedures were performed at 4°C, unless stated otherwise. All reagents were purchased from Sigma-Aldrich (St Louis, MO, USA).
Electrophoresis and zymography
Sodium dodecyl sulphate–polyacrylamide gel electrophoresis (SDS–PAGE) was performed according to Laemmli (1970). Polyacrylamide gel (8%) containing 1% gelatin was used (Vadillo-Ortega et al., 1995) to identify the gelatinase associated with both membrane extracts. To determine metalloproteinase activity, we washed gels twice for 15 min in a solution containing 2.5% Triton X-100 and incubated in a buffer containing 50 mM Trizma Tris base (pH 7.4), 150 mM NaCl, 20 mM CaCl2. Controls with EDTA or o-phenanthroline to identify metalloproteinase activity were included in all experiments. Gelatinase activities were visualized by staining with 0.25% Coomassie Brilliant Blue R-250 in 45% methanol and 10% acetic acid and destained in a solution of 20% methanol and 10% acetic acid. Enzymatic activity standards for MMP-2 and MMP-9 were included in each experiment, using the supernatants of cultured U937 promyelocytes (Morodomi et al., 1992).
The activity of MMP bands was determined by image analysis using arbitrary measures (UVP, Cambridge, UK). MW was calculated against protein standards (Invitrogen, Life Technologies, Carlsbad, CA, USA), and linear regression was made for every lytic band.
Western blotting
Extracts (10.0 µg) were separated by 8% SDS electrophoresis and transferred to a nitrocellulose Immobilon-P membrane (Millipore, Medford, MA, USA), using the semi-dry system (Towbin et al., 1992). The identification of specific MMPs was determined using 1 µg/ml of monoclonal antibodies against MMP-9 (mouse anti-human, clone Ab-3), MMP-3 (mouse anti-human, clone Ab-1), MMP-2 (mouse anti-human, clone Ab-3), TIMP-1 (mouse anti-human, clone Ab-3) and TIMP-2 (mouse anti-human, clone Ab-2) (Calbiochem Biochemicals, La Jolla, CA, USA). All these antibodies cross-react with rat proteins. Primary antibodies were detected with VectaStain ABC reagents (Vector, Burlingame, CA, USA). Positive controls using purified recombinant proteins for MMP-9, MMP-3, MMP-2, TIMP-1 and TIMP-2 were included in each experiment (Calbiochem). The relative intensity of bands was determined using an image analyser (UVP, Southern CA, USA).
Liquid chromatography
EDTA extracts were subjected to fast protein liquid chromatography (FPLC) using a Superdex 200 column (Amersham Pharmacia Biotech, NJ, USA). EDTA extracts were applied to the column as prepared or after adding 20 mM CaCl2 (final concentration) and separated under appropriately equilibrated conditions in a buffer with or without 20 mM CaCl2. Five hundred micrograms of protein were injected into the column. Fractions of 1.0 ml were collected at a flow rate of 0.5 ml/min, and gelatinase activity was determined by zymography. Thyroglobulin (670 kDa), catalase (232 kDa), bovine albumin (68 kDa) and myoglobin (17 kDa) were run in a solution containing 50 mM Trizma Tris base (pH 7.4), 150 mM NaCl, as MW standards. The MWs of the EDTA-extract components were calculated from a linear regression curve.
Enzymatic activity of EDTA extracts
Enzymatic activity was evaluated in EDTA extracts against a gelatinase substrate, adding CaCl2 in a concentration range from 0.1 to 6 mM at 0.5 mM intervals. Specific activities present in these extracts were assayed using the MMP-2/MMP-9 substrate Ac-Pro-Leu-Gly-S-Leu-Leu-Gly-OC2H5 (Calbiochem). Assays were performed (Weingarten and Feder, 1985) with 10.0 µg protein of each sample in a solution containing 0.1 mM substrate, 50 mM HEPES (pH 7.4), 1.0 mM 4,4'-dithiodipyridine and incubated in triplicates for 6 h at 25°C. The 4,4'-dithiodipyridine reacts with the mercaptan hydrolysis fragment to form products, with absorbance at 324 nm.
Differential scanning calorimetry
Differential scanning calorimetry (DSC) was used to measure the transition temperature (Tm) of the proteins in the EDTA extracts in the presence and absence of Ca2+. Tm is the temperature at which the excess heat capacity is maximal. Extracts were dialysed in a buffer containing 0.15 mM NaCl and 20 mM Tris–malate (pH 7.4). Excess heat (Cp) was measured as a function of temperature; scans were obtained from about 1 mg/ml protein using a high-sensitivity differential scanning calorimeter VP-DSC (MicroCal, MA, USA). The samples and reference solutions were carefully degassed under vacuum for 5 min before loading the cells. When equilibrium was reached at 10°C, the temperature was increased to 100°C at 1°C/min. To assess for the reversibility of the protein denaturation, we rescanned the samples from 10 to 100°C after cooling down to 10°C. The baseline was corrected by subtracting the rescanned values from the scanned values. In all cases, reversible denaturation was subtracted to account for irreversible denaturation. DSC profiles were deconvoluted, and the best theoretical fit was calculated assuming irreversible denaturation (Biltonen and Freire, 1978).
Confocal microscopy
Cell-free chorioallantoid membranes were prepared by repeated freeze–thawing in 0.1% Triton X-100 0.01 M phosphate-buffered saline (PBS), 0.1 M NaCl, 0.003 M KCl (pH 7.4). After this treatment, membranes were washed extensively with PBS, fixed in neutral buffered formalin and embedded in paraffin wax for fluorescent immunostaining. The double staining of the membranes was performed using four pairs of primary antibodies; each included a combination of a fluorescein isothiocyanate (FITC)-labelled anti-MMP-9 with anti-MMP-2, anti-MMP-3, anti-TIMP-1 or anti-TIMP-2 (Calbiochem). All the primary antibodies (except anti-MMP-9) were identified using an Alexa Fluor 568-labelled secondary antibody (Molecular Probes, Eugene, OR, USA). A control lacking the primary antibody was included to verify that no signal was coming from the fluorescent secondary antibody. Slides were mounted with Vectashield (Vector Laboratories, UK), covered with glass coverslips and examined using confocal microscopy (Axiowert 100 M; Carl Zeiss, Germany).
|
Results |
|---|
|
TopAbstractIntroductionMaterials and methodsResultsDiscussionReferences |
|---|
Zymography gelatinase activity
Gel substrate assays (n = 30) of supernatants from chorioallantoid membrane homogenates prepared in a buffer containing 20 mM Ca2+ showed three lytic bands at estimated MWs of 92, 72 and 65 kDa (Figure 1, Panel A, lane 2). The gelatinolytic bands corresponded to pro-MMP-9, pro-MMP-2 and MMP-2, according to the relative migration of the U937-cell protein standard. The incubation of the zymograms in the presence of EDTA or o-phenanthroline inhibited all gelatinases, confirming the identity of the enzymes as metalloproteinases (data not shown).
|
Effects of EDTA on gelatinase activity
As controls, Ca2+-containing extracts were incubated in the presence of 40 mM EDTA and then applied to the gelatin substrate gel. Several new gelatinases appeared, and those previously identified in the non-treated samples (92, 72 and 65 kDa) were more visible. This treatment always resulted in the same pattern of bands, as depicted in Figure 1 (Panel A). To test the effects of Ca2+ on this phenomenon and its reversibility, we added an excess of CaCl2 to the EDTA extracts, and gelatinase activation was fully reversed (Figure 1, Panel A, lane 4). Control samples were kept at room temperature with no changes in gelatinolytic activity (Figure 1, Panel A, lane 5). To test the specificity of the metal-dependent assembly and disassembly of the gelatinases, we added to EDTA extracts equimolar concentrations of Mg2+, Mn2+ and Zn2+, but these metals could not reverse the appearance of additional gelatinolytic activities (data not shown).
From these experiments, we knew that adding EDTA to the tissue extracts revealed the presence of an extra amount of gelatinases when analysed by zymography. This was why we used an EDTA buffer to treat the tissue pellets for membrane extract preparation. When the EDTA extracts of the membrane pellets were subjected to zymography, a pattern of gelatinases appeared that was indistinguishable from the EDTA-activated initial extract described above (Figure 1, Panel A, lane 3). Moreover, the assembly and disassembly of the complexes in the presence of Ca2+ were the same for the EDTA extract (data not shown). As expected, regular SDS–PAGE revealed more bands in the first extract than in the EDTA extract (Figure 1, Panels B and C), yielding 0.2 ± 0.05 mg/g of protein/g of chorioallantoid membranes (n = 20). From this point, all experiments were performed using the EDTA extracts.
Protein composition and western blot analysis
Western blotting of EDTA extracts revealed the presence of immunoreactive MMP-9, MMP-2, MMP-3, TIMP-1 and TIMP-2 (Figure 2; n = 5). Recombinant MMP-9, MMP-3, MMP-2, TIMP-1 and TIMP-2 proteins were used as positive controls. The antibody against MMP-9 showed not only the presence of 92 and 82 kDa forms at a ratio of 2:1 but also two bands of 115 and 215 kDa. Anti-MMP-2 showed immunoreactivity at MWs of 72 and 65 kDa at a ratio of 1:2. Anti-MMP-3 showed immunoreactivity at MWs of 60 and 45 kDa, at a ratio of 1:1. TIMP-1 and TIMP-2 appeared as single bands at an MW of 29 kDa.
|
Molecular sieve chromatography
Figure 3A shows a representative elution profile of gelatinases present in the EDTA extracts when analysed by zymography. Different MW gelatinases compatible with those observed by western blot were also identified with this technique (n = 5). Most of the gelatinolytic activity appeared at 92 kDa (
28% of the loaded protein; n = 5), followed by the fractions at 72 kDa (16% of the protein; n = 5) and 120 kDa (14%; n = 5). Figure 3B depicts a zymogram analysis of the EDTA extracts separated in the molecular sieve column after the addition of 20 mM CaCl2. This profile of gelatinase activity was completely different from that observed for the EDTA extracts alone, showing only one peak of gelatinase activity eluting near the exclusion volume of the column, corresponding to an MW 
670 kDa and making up 11.5% of the loaded protein. Eluted fractions corresponding to this high MW peak were then treated with EDTA and subjected to zymography. Gelatinase activity shown in the box inside Panel B in Figure 3 indicates that this peak eluting near the exclusion limit contained all the previously observed gelatinases.
|
Denaturation profile of the protein complex by DSC
DSC profile obtained from the EDTA extracts is shown in Figure 4; the extract denatured in at least eight transitions at temperatures (Tm) ranging from of 35.5°C to 92.6°C (n = 3). To identify the components of the protein complex observed in the thermograms, we used the denaturation profiles of recombinant proteins (Meraz-Cruz N et al., submitted for publication). This analysis gave the following estimated composition: pro-MMP-9, active MMP-9, pro-MMP-3, active MMP-3, pro-MMP-2, active MMP-2, TIMP-2 and TIMP-1. As the DSC profiles were performed under non-standarized conditions to calculate actual molar ratio contribution of each component, molecular weight of the complex was not calculated.
|
A broad peak (endothermic transition) corresponding to the denaturation of several proteins was obtained when 20 mM CaCl2 was added to the EDTA extract which was then analysed by DSC, revealing an associated complex (Figure 4, Panel B). The transition temperature (Tm) of this peak was 72°C with a calculated enthalpy (
H) of 870 mcal/kJ per mol, as determined by deconvolution (Biltonen and Freire, 1978).
According to these observations, the minimal MW of the complex was 480 kDa, calculated as the sum of each individual molecular mass without taking into account the real stoichiometry of the different components.
Calcium dependency of the enzymatic activity of the protein complex
Figure 5A shows the MMP-2/MMP-9 combined activity associated with the EDTA extract as a function of CaCl2 concentration (n = 5). There was no detectable gelatinolytic activity at Ca2+ concentrations <0.7 mM or >2.2 mM. Figure 5B shows the gelatinolytic activity of the EDTA extract in parallel zymogram gels, demonstrating that the activity observed in Figure 5A had a similar dependency on Ca2+ concentration.
|
Confocal microscopy
We used cell-depleted chorioallantoid membranes for immunostaining to reproduce the conditions from which the EDTA extracts were derived. All the materials used for the characterization of the complex were derived from the elution of the chorioallantoid PBS-washed pellets with EDTA. All the cell-free membranes were positive for double staining with any combination of the antibodies. MMP-2, MMP-3, TIMP-1 and TIMP-2 co-localized with MMP-9, revealing in some cases the presence of strong double-positive spots in the tissues (Figure 6). Controls, to which primary antibodies were not added, resulted in no staining as did membranes that were extensively washed with a buffer containing EDTA before any combination of the antibodies was added (data not shown).
|
|
Discussion |
|---|
|
TopAbstractIntroductionMaterials and methodsResultsDiscussionReferences |
|---|
The rat chorioallantoid membrane is a useful experimental model to evaluate the physiological activation of MMPs, because during labour there is a sudden increase in the expression and activation of these enzymes. Previously (Meraz et al., 2003
) and in this article, we were able to demonstrate increased MMP activities in rat fetal membrane extracts during labour, namely for MMP-2 and MMP-9. However, here and in other reports using different experimental models (Niu et al., 2000; Vadillo-Ortega et al., 2002; Xu et al., 2002), the active forms of MMP-2 and MMP-9 were not found by zymography. This finding raises a biological inconsistency because only the active forms of these enzymes can have a role in connective tissue degradation, leading to the rupture of the membranes during rat labour. Rather than discarding the participation of these molecules, any fault in the methodologies we used to demonstrate active MMP forms must be ruled out first.
In this article, we had a serendipitous finding when the samples of tissue extracts were first pre-incubated with EDTA and then applied to zymogram gels and later incubated under optimal conditions for demonstrating MMP activity. Under these conditions, several new gelatinase bands appeared and others were reinforced. This created a paradoxical phenomenon in that the removal of Ca2+ before electrophoretic separation revealed several unsuspected MMP activities. Moreover, this treatment showed the presence of the expected active forms of MMP-2, MMP-3 and MMP-9. We suspected that a Ca2+-dependent complex of MMP may have been present in the tissue extracts. To initiate the biochemical characterization of the hypothetical MMP complex, we decided to make an enriched extract using a different approach, taking into account the observed EDTA-mediated disaggregation of gelatinases. First, we observed that the treatment of the homogenate pellets by an EDTA-containing solution eluted a significant amount of the same hypothetical complex of gelatinases. This EDTA-eluted preparation was used for all further experiments. An additional property of this complex emerged immediately, suggesting that these molecular components were bound to the extracellular matrix, where they can be dissociated by the chelation of Ca2+ using EDTA. We first characterized this as a Ca2+-dependent process because only the presence or absence of this particular divalent cation reproduced the hypothetical assembly and disassembly of the detected MMP and TIMPs, respectively. To characterize the individual components present in the EDTA-treated ‘disassembled complex’, we used western blotting; this revealed the presence of pro-MMP-9, active MMP-9, pro-MMP-3, active MMP-3, pro-MMP-2, active MMP-2, TIMP-1 and TIMP-2. In addition to these MMPs, additional high MW MMP-9 immunoreactive proteins were observed at 115 and 215 kDa. They may correspond to the described association of MMP-9 with 2-microglobulin or lipocalin [neutrophil gelatinase-associated lipocalin (NGAL), a 25-kDa associated protein] to produce the 125 kDa form (Triebel et al., 1992a; Yan et al., 2001) or the dimeric form of 215 kDa (Triebel et al., 1992b). To further characterize the contribution of the different protein components, we studied their structural characteristics through their unfolding profile by DSC using recombinant proteins as standards. When the EDTA extract was heated, it showed a denaturation profile in which several transitions were clearly identified as pro-MMP-9, MMP-9, pro-MMP-2, MMP-2, pro-MMP-3, MMP-3, TIMP-1 and TIMP-2. The Ca2+-added extract protein complex unfolded as a single broad transition peak with a lack of exothermic transition suggested the formation of a complex rather than protein aggregation (Ortega and Lepock, 1995).
The EDTA extract was a semi-purified extract, as we eliminated virtually all the cellular contaminants by repeated washing of the fetal membrane homogenates. We then treated the biomatrix with an EDTA washout step, which eluted the same gelatinases we observed in the original extracts. Furthermore, we duplicated the assembly and disassembly of the Ca2+-dependent complex of gelatinases and TIMPs with these enriched extracts. This suggests that the identified complex exists in vivo, forming a stable complex with the extracellular matrix. We confirmed this assumption using confocal microscopy to co-localize the presence of all the molecular components in the chorioallantoid biomatrix, obtained through detergent treatment and freeze–thawing of the fetal membranes. The heterogeneous distribution of the complexes, as revealed by confocal microscopy, may suggest the existence of specific extracellular matrix components to which the complex may have affinity such as specific collagen types. Several other reports have revealed that the extracellular matrix may be a source of tightly bound MMPs. The demonstration of collagenase bound to connective tissue in vivo has been reported in normal rat tissues, in the experimental models of liver cirrhosis and in the rat uterus (Montfort and Perez Tamayo, 1975, 1978). Pardo et al. (1980) reported a collagenolytic enzyme, probably MMP-13, as a constant contaminant bound tightly to collagen fibres. Their suggestion of a probable physiological role of this substrate-bound enzyme is supported by our findings here.
The initial biochemical characterization of the EDTA extract involved molecular sieve chromatography, which corroborated the existence of the same gelatinases that had been identified by western blotting and DSC. When Ca2+ was added to this extract before loading into the column, these enzymes disappeared, and only one fraction of high MW was demonstrated. Furthermore, this peak eluting in the exclusion volume of the column could be dissociated into the same molecular components by the chelation of the available Ca2+. Surprisingly, only 15% of the loaded protein could be recovered from the molecular sieve column when Ca2+ was present in the solutions; these are conditions favouring the formation of the complex. This again suggests that this is a very high MW complex that does not enter Superdex 200 resin or the acrylamide mesh during SDS–PAGE or zymography. Indeed, the minimal estimated MW mass of this complex is 480 kDa, which may make this complex one of the biggest multicatalytic protease complexes described (Dutta and Berman, 2005). This proposed size of this complex can explain the confocal microscopy results, which revealed the presence of large clusters of immunoreactive co-localized proteins in cell-free chorioallantoid membranes. This is direct evidence that a complex of MMP forms part of the extracellular degradome, at least in rat chorioallantoid membranes.
Calcium is required for MMP activity in two ways. First, Ca2+ ions are associated with the catalytic domains of MMPs, and they are essential for the folding of the molecule for substrate recognition and catalysis. On the contrary, a single Ca2+ ion is associated with the haemopexin domain of the MMP and is believed to participate in the interaction with other molecules, but it does not affect the catalytic function of these enzymes. There is evidence that the haemopexin domain has affinity for heparin and that this interaction depends on Ca2+ (Wallon and Overall, 1997). We could not purify the proposed complex using traditional methodologies, so we were unable to study the mechanism of interaction between the molecular components. However, in this report we provide evidence that MMP-2, MMP-3 and MMP-9 in both active and inactive forms, in addition to TIMP-1 and TIMP-2, are tightly bound in a complex that has affinity for the extracellular matrix, depending on Ca2+-associated non-described interactions.
We have not identified this complex in other tissue extracts, but, because the regular buffers used for MMP extraction contain millimolar concentrations of Ca2+, it is possible that the absence of active forms of MMP in membranes and other tissue extracts under zymography analysis can be explained by our findings.
Physiological role of the high MW complex in the control of MMP activity
The presence of a protein complex comprising active and inactive forms of different MMPs and TIMPs suggests a mechanism of latent enzyme activity that may participate in the tissue control of MMP activity. This assumption was evaluated over a range of physiological concentrations of Ca2+, and the phenomenon of Ca2+-mediated activation mechanism was found to be reproducible. According to these experiments, MMP activities can experience a transition in structure and function, forming a stable complex at high physiological concentrations of Ca2+ (>2.0 mM) that confers the latency to these enzymes. On the contrary, if Ca2+ falls <2.0 mM, we hypothesize that the complex dissociates and the active enzymes can act on their respective substrates. Such a change in the extracellular concentration of Ca2+ can occur in this microenvironment during labour, because myometrial contraction generates a flux of Ca2+ that is directed from the extracellular space and amniotic fluid towards the myometrium. Thus, it is feasible that Ca2+ concentrations up to 2.0 mM can be attained in fetal membranes (Sanborn, 2000). The formation of this complex of active and inactive MMPs may allow synchronization between myometrium contraction and membrane rupture. This complex may have an additional physiological advantage because it could be used as a store of preformed enzymes, which once activated could induce a massive degradation of extracellular matrix components. This appears to be the case for the degradation of fetal membranes, which are ruptured only during the last stages of labour, usually in a matter of minutes. This occurs even though the synthesis of MMP is induced several hours before labour in rats (Lei et al., 1995) and several days before in humans (Vadillo-Ortega et al., 1995) and non-human primates (Vadillo-Ortega et al., 2002). Therefore, it is unlikely that the rupture of the membranes can be ascribed to the active de-novo synthesis of MMP and simultaneous extracellular activation.
Here, we have identified a mechanism of enzymatic control with potential significance for the regulation of human parturition, producing a stepwise process leading to the rupture of the fetal membranes. We propose that the first step is the formation of an extracellular complex that may be generated simultaneously with MMP secretion to the extracellular space and may generate a latent enzyme system under physiological concentrations of Ca2+. This would result in the progressive accumulation of a massive amount of extracellular matrix-degrading enzymes during the first stages of labour. We also propose that, once effective myometrium contractions are induced, this results in a progressive decrease in extracellular Ca2+ making the complex unstable; the latency of the MMPS is thus lost, provoking the rupture of the membranes. This hypothesis takes into account that Ca2+ concentrations in this microenvironment depend on a wide range of physiological and pathological conditions, such as early uterine contractions. Thus, it is related potentially to the mechanisms of pre-labour rupture of the membranes. Further research on the molecular characteristics of this complex is needed to assess its physiological role during parturition in rodents and humans.
|
Acknowledgements |
|---|
|
TopAbstractIntroductionMaterials and methodsResultsDiscussionReferences |
|---|
This research was supported by CONACyT grant 26177-M and CONACyT Scholar Fellowship 93301 for N.M.C.
|
References |
|---|
|
TopAbstractIntroductionMaterials and methodsResultsDiscussionReferences |
|---|
Allan JA, Docherty AJ, Barker PJ, Huskisson NS, Reynolds JJ and Murphy G (1995) Binding of gelatinases A and B to type-I collagen and other matrix components. Biochem J 309,299–306.
Baker A, Edwards D and Murphy G (2002) Metalloproteinase inhibitors: biological actions and therapeutics opportunities. J Cell Sci 115,3719–3727.
Bannikov GA, Karelina TV, Collier IE, Marmer BL and Goldberg GI (2002) Substrate binding of gelatinase B induces its enzymatic activity in the presence of intact propeptide. J Biol Chem 277,16022–16027.
Bell SC and Malak TM (1997) Structural and cellular biology of the fetal membranes. In Elder MG, Romero R and Lamont RF (eds) Preterm Labor. Churchill Livingstone, New York, pp. 401–428.
Biltonen RL and Freire E (1978) Thermodynamic characterization of conformational states of biological macromolecules using differential scanning calorimetry. CRC Crit Rev Biochem 5,85–124.[ISI][Medline]
Bode W, Fernández-Catalán C, Tschesche H, Grams F, Nagase H and Maskos K (1999) Structural properties of matrix metalloproteinases. Cell Mol Life Sci 55,639–652.[CrossRef][ISI][Medline]
Bradford M (1976) A rapid and sensitive method for the quantitation of microgram quantities of protein utilizing the principle of protein-dye binding. Anal Biochem 72,248–254.[CrossRef][ISI][Medline]
Brew K, Dinakarpandian D and Nagase H (2000) Tissue inhibitors of metalloproteinases: evolution, structure and function. Biochim Biophys Acta 1477,267–283.[CrossRef][Medline]
Bryant-Greenwood GD (1998) The extracellular matrix of the human fetal membranes: structure and function. Placenta 19,1–11.[ISI][Medline]
Cao J, Sato H, Takino T and Seiki M (1995) The C-terminal region of membrane type matrix metalloproteinase is a functional transmembrane domain required for pro-gelatinase A activation. J Biol Chem 270,801–805.
Dutta S and Berman H (2005) Large macromolecular complexes in the protein data bank: a status report. Structure 13,381–388.[Medline]
Fata JE, Ho AT-V, Leco KJ, Moorehead RA and Khokha R (2000) Cellular turnover and extracellular matrix remodeling in female reproductive tissues: functions of metalloproteinases and their inhibitors. Cell Mol Life Sci 57,77–95.[CrossRef][ISI][Medline]
Imai K, Yokohama Y, Nakanishi Y, Ohuchi E, Fuji Y, Nakai N and Okada Y (1995) Matrix metalloproteinase 7 (matrilysin) from human rectal carcinoma cells. Activation of the precursor, interaction with other matrix metalloproteinases and enzymic properties. J Biol Chem 270,6691–6697.
Laemmli UK (1970) Cleavage of structural proteins during the assembly of the head of bacteriophage T4. Nature 227,680–685.[CrossRef][Medline]
Lei H, Vadillo-Ortega F, Paavola LG and Strauss JF III (1995) 92-kDa gelatinase (matrix metalloproteinase-9) is induced in rat amnion immediately prior to parturition. Biol Reprod 53,339–344.[Abstract]
Lei H, Furth EE, Kalluri R, Chiou T, Tilly KI, Tilly JL, Elkon KB, Jeffrey JJ and Strauss JF III (1996) A program of cell death and extracellular matrix degradation is activated in the amnion before the onset of labor. J Clin Invest 98,1971–1978.[ISI][Medline]
Li J, Brick P, O’Hare MC, Skarzynski T, Lloyd LF, Curry VA, Clark IM, Bigg HF, Hazleman BL, Cawston TE et al. (1995) Structure of full-length porcine synovial collagenase reveals a C-terminal domain containing a calcium-linked, four-bladed ß-propeller. Structure 3,541–549.[Medline]
Maeda H, Okamoto T and Akaike T (1998) Human matrix metalloprotease activation by insults of bacterial infection involving proteases and free radicals. Biol Chem 379,193–200.[ISI][Medline]
Maymon E, Romero R, Pacora P, Gervasi MT, Gomez R, Edwin SS and Yoon BH (2000) Evidence for the participation of interstitial collagenase (matrix metalloproteinase 1) in preterm premature rupture of membranes. Am J Obstet Gynecol 183,914–920.[CrossRef][ISI][Medline]
McLaren J, Malak TM and Bell SC (1999) Structural characteristics of term human fetal membranes prior to labour: identification of an area of altered morphology overlying the cervix. Hum Reprod 14,237–241.
Meraz-Cruz N, Molina DG and Vadillo-Ortega F (2003) Cambios secuenciales de metaloproteinasas de matriz extracelular durante la gestación y el trabajo de parto en el corioalantoides de la rata. Rev Invest Clin 55,36–42.[ISI][Medline]
Montfort I and Perez Tamayo R (1975) The distribution of collagenase in normal rat tissues. J Histochem Cytochem 23,910–920.[Abstract]
Montfort I and Perez Tamayo R (1978) Collagenase in experimental carbon tetrachloride cirrhosis of the liver. Am J Pathol 92,411–420.[Abstract]
Morodomi T, Ogata Y, Sasaguri Y, Morimatsu M and Nagase H (1992) Purification and characterization of matrix metalloproteinase 9 from U937 monocytic leukaemia and HT1080 fibrosarcoma cells. Biochem J 285,603–611.
Murphy G, Stanton H, Cowell S, Butler G, Knauper V, Atkinson S and Gavrilovic J (1999) Mechanisms for pro matrix metalloproteinase activation. APMIS 107,38–44.[ISI][Medline]
Nagase H and Woessner JF Jr (1999) Metalloproteinases. J Biol Chem 274,21491–21494.
National Research Council (1996) Guide for the Care and Use of Laboratory Animals. Institute of Laboratory Animal Resources, Commission on Life Sciences. National Academy Press, Washington DC.
Niu R, Okamoto T, Iwase K, Nomura S and Mizutani S (2000) Quantitative analysis of matrix metalloproteinases-2 and -9, and their tissue inhibitors-1 and -2 in human placenta throughout gestation. Life Sci 66,1127–1137.[CrossRef][ISI][Medline]
Ortega A and Lepock J (1995) Use of thermal analysis to distinguish magnesium and calcium stimulated ATPase activity in isolated transverse tubules from skeletal muscle. Biochim Biophys Acta 1233,7–13.[Medline]
Paavola LG, Furth EE, Delgado V, Boyd CO, Jacobs CC, Lei H and Strauss JF III (1995) Striking changes in the structure and organization of rat fetal membranes precede parturition. Biol Reprod 53,321–338.[Abstract]
Pardo A, Soto H, Montfort I and Perez Tamayo R (1980) Collagen-bound collagenase. Connect Tissue Res 7,253–261.[ISI][Medline]
Pei D and Weiss S (1995) Furin-dependent intracellular activation of the human stromelysin-3 zymogen. Nature 375,244–247.[CrossRef][Medline]
Roswit WT, McCourt DW, Partridge NC and Jeffrey JJ (1992) Purification and sequence analysis of two rat tissue inhibitors of metalloproteinases. Arch Biochem Biophys 292,402–410.[CrossRef][ISI][Medline]
Rowe TF, King LA, MacDonald PC and Casey ML (1997) Tissue inhibitor of metalloproteinase-1 and tissue inhibitor of metalloproteinase-2 expression in human amnion mesenchymal and epithelial cells. Am J Obstet Gynecol 176,915–921.[CrossRef][ISI][Medline]
Sanborn B (2000) Relationship of ion channel activity to control of myometrial calcium. J Soc Gynecol Investig 7,4–11.[ISI][Medline]
Sang QA and Douglas DA (1996) Computational sequence analysis of matrix metalloproteinases. J Protein Chem 15,137–160.[CrossRef][ISI][Medline]
Springman EB, Angleton EI, Birkedal-Hansen H and Van Wart HE (1990) Multiple modes of activation of latent human fibroblast collagenase: evidence for the role of a Cys73 active-site zinc complex in latency and a "cysteine switch" mechanism for activation. Proc Natl Acad Sci USA 87,364–368.
Towbin H, Staehelin T and Gordon J (1992) Electrophoretic transfer of proteins from polyacrylamide gels to nitrocellulose sheets: procedure and some applications. Biotechnology 24,145–149.[Medline]
Triebel S, Blaser J, Reinke H and Tschesche H (1992a) A 25 kDa 2-microglobulin-related protein is a component of the 125 kDa form of human gelatinase. FEBS Lett 314,386–388.[CrossRef][ISI][Medline]
Triebel S, Blaser J, Reinke H, Knauper V and Tschesche H (1992b) Mercurial activation of human PMN leucocyte type IV procollagenase (gelatinase). FEBS Lett 298,280–284.[CrossRef][ISI][Medline]
Vadillo-Ortega F, González-Avila G, Furth EE, Lei H, Muschel RJ, Stetler-Stevenson WG and Strauss JF III (1995) 92-kd type IV collagenase (matrix metalloproteinase-9) activity in human amniochorion increases with labor. Am J Pathol 146,148–156.[Abstract]
Vadillo-Ortega F, Sadowsky DW, Haluska GJ, Hernández-Guerrero C, Guevara-Silva R, Gravett MG and Novy MJ (2002) Identification of matrix metalloproteinase-9 in amniotic fluid and amniochorion in spontaneous labor and after experimental intrauterine infection or interleukin-1 beta infusion in pregnant rhesus monkeys. Am J Pathol 186,128–138.
Wallon UM and Overall CM (1997) The hemopexin-like domain (C domain) of human gelatinase A (matrix metalloproteinase-2) requires Ca2+ for fibronectin and heparin binding. Binding properties of recombinant gelatinase A C domain to extracellular matrix and basement membrane components. J Biol Chem 272,7473–7481.
Weingarten H and Feder J (1985) Spectrophotometric assay for vertebrate collagenase. Anal Biochem 147,437–440.[CrossRef][ISI][Medline]
Xu P, Alfaidy N and Challis JRG (2002) Expression of matrix metalloproteinase (MMP)-2 and MMP-9 in human placenta and fetal membranes in relation to preterm and term labor. J Clin Endocrinol 87,1353–1361.
Yan L, Borregaard N, Kjeldsen L and Moses MA (2001) The high molecular weight urinary matrix metalloproteinase (MMP) activity is a complex of gelatinase B/MMP-9 and neutrophil gelatinase-associated lipocalin (NGAL). Modulation of MMP-9 activity by NGAL. J Biol Chem 276,37258–37265.
Submitted on May 20, 2006; resubmitted on July 14, 2006; accepted on July 19, 2006.
CiteULike 
Connotea 
Del.icio.us What's this?
This article has been cited by other articles:
|
|
 |
|
  G. Garcia-Lopez, F. Vadillo-Ortega, H. Merchant-Larios, R. Maida-Claros, M. Osorio, D. Soriano-Becerril, H. Flores-Herrera, J. Beltran-Montoya, Y. Garfias-Becerra, and V. Zaga-Clavellina Evidence of in vitro differential secretion of 72 and 92 kDa type IV collagenases after selective exposure to lipopolysaccharide in human fetal membranes Mol. Hum. Reprod., June 1, 2007; 13(6): 409 - 418. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||