Three-dimensional Type I Collagen Lattices Induce Coordinate Expression of Matrix Metalloproteinases MT1-MMP and MMP-2 in Microvascular Endothelial Cells
1998; Elsevier BV; Volume: 273; Issue: 6 Linguagem: Inglês
10.1074/jbc.273.6.3604
ISSN1083-351X
AutoresTara L. Haas, Sandra Davis, Joseph A. Madri,
Tópico(s)Cell Adhesion Molecules Research
ResumoMatrix metalloproteinases (MMPs) are hypothesized to play a key role in the processes of endothelial cell migration and matrix remodeling during angiogenesis. We utilized an in vitro model of microvascular endothelial cell angiogenesis, cells cultured within a collagen matrix, to investigate the MMP profile of endothelial cells undergoing angiogenesis. We demonstrated by gelatin zymography that monolayer cultures (two-dimensional) of endothelial cells constitutively expressed low levels of latent MMP-2, but that culture in a three-dimensional collagen matrix increased the total amount of MMP-2 mRNA and protein. Furthermore, 51% of total MMP-2 protein was activated in the three-dimensional culture lysates, compared with 3.5% in two-dimensional culture. The mRNA and protein of MT1-MMP, the putative activator of MMP-2, were up-regulated in endothelial cells cultured in three-dimensional as compared with two-dimensional culture. Treatment of cultures with MMP inhibitors blocked activation of MMP-2 and inhibited formation of endothelial cell networks within the collagen gel. Induction of MT1-MMP and MMP-2 appeared to be specific to collagen, inasmuch as culture of the endothelial cells on top of, or within, a Matrigel® matrix neither increased total MMP-2 nor increased activation of MMP-2. These results suggest that MT1-MMP activation of MMP-2 occurs in endothelial cells undergoing angiogenesis, that this activation has a functional role in endothelial cell organization, and that specific matrix interactions may be critical for the increased expression of MT1-MMP and MMP-2. Matrix metalloproteinases (MMPs) are hypothesized to play a key role in the processes of endothelial cell migration and matrix remodeling during angiogenesis. We utilized an in vitro model of microvascular endothelial cell angiogenesis, cells cultured within a collagen matrix, to investigate the MMP profile of endothelial cells undergoing angiogenesis. We demonstrated by gelatin zymography that monolayer cultures (two-dimensional) of endothelial cells constitutively expressed low levels of latent MMP-2, but that culture in a three-dimensional collagen matrix increased the total amount of MMP-2 mRNA and protein. Furthermore, 51% of total MMP-2 protein was activated in the three-dimensional culture lysates, compared with 3.5% in two-dimensional culture. The mRNA and protein of MT1-MMP, the putative activator of MMP-2, were up-regulated in endothelial cells cultured in three-dimensional as compared with two-dimensional culture. Treatment of cultures with MMP inhibitors blocked activation of MMP-2 and inhibited formation of endothelial cell networks within the collagen gel. Induction of MT1-MMP and MMP-2 appeared to be specific to collagen, inasmuch as culture of the endothelial cells on top of, or within, a Matrigel® matrix neither increased total MMP-2 nor increased activation of MMP-2. These results suggest that MT1-MMP activation of MMP-2 occurs in endothelial cells undergoing angiogenesis, that this activation has a functional role in endothelial cell organization, and that specific matrix interactions may be critical for the increased expression of MT1-MMP and MMP-2. Angiogenesis, which is the formation of new blood vessels from those pre-existing, occurs during development, wound healing, and tumor growth, and in response to stimuli such as exercise and hypoxia (1Madri J.A. Sankar S. Romanic A.M. Clark R.A.F. The Molecular and Cellular Biology of Wound Repair. Plenum Press, New York1996: 355-371Google Scholar, 2Pepper M.S. Arterioscler. Thromb. Vasc. Biol. 1997; 17: 605-619Crossref PubMed Scopus (263) Google Scholar, 3Folkman J. EXS. 1997; 79: 1-8PubMed Google Scholar). This process involves complex signaling events that cause the endothelial cells comprising capillaries to initiate proliferative and migratory phenotypes. The sprouting endothelial cells must break through their existing basement membrane and form contacts with and migrate along different extracellular matrix components, finally establishing a new, patent capillary (1Madri J.A. Sankar S. Romanic A.M. Clark R.A.F. The Molecular and Cellular Biology of Wound Repair. Plenum Press, New York1996: 355-371Google Scholar, 2Pepper M.S. Arterioscler. Thromb. Vasc. Biol. 1997; 17: 605-619Crossref PubMed Scopus (263) Google Scholar). In an effort to define the molecular mechanisms underlying these events, a number of in vitro cell systems have been established. These include growth of endothelial cells or blood vessel fragments in fibrin clots, on amnionic membranes, in collagen matrices, and on Matrigel® matrices (4Kubota Y. Kleinman H.K. Martin G.R. Lawley T.J. J. Cell Biol. 1988; 107: 1589-1598Crossref PubMed Scopus (986) Google Scholar, 5Madri J.A. Pratt B.M. J. Histochem. Cytochem. 1986; 34: 85-91Crossref PubMed Scopus (145) Google Scholar, 6Montesano R. Eur. J. Clin. Invest. 1992; 22: 504-515Crossref PubMed Scopus (85) Google Scholar). These models are characterized by re-organization that requires significant endothelial cell migration and/or invasion, and remodeling of the surrounding matrix molecules. It has been shown that microvascular endothelial cells undergo morphological changes that can include organization into tubelike structures when grown within a type I collagen matrix (7Montesano R. Orci L. Vassalli P. J. Cell Biol. 1983; 97: 1648-1652Crossref PubMed Scopus (513) Google Scholar, 8Madri J.A. Pratt B.M. Tucker A.M. J. Cell Biol. 1984; 106: 1375-1384Crossref Scopus (459) Google Scholar). The morphological changes are accompanied by changes in growth factor receptor profiles and extracellular matrix protein production (9Sankar S. Mahooti-Brooks N. Bensen L. McCarthy T.L. Centrella M. Madri J.A. J. Clin. Invest. 1996; 97: 1436-1446Crossref PubMed Scopus (153) Google Scholar,10Marx M. Perlmutter R.A. Madri J.A. J. Clin. Invest. 1994; 93: 131-139Crossref PubMed Scopus (125) Google Scholar). Matrix metalloproteinases (MMPs) 1The abbreviations used are: MMP, matrix metalloproteinase; MT-MMP, membrane-type MMP; TGFβ1, transforming growth factor β1; PMA, phorbol 12-myristate 13-acetate; MOPS, 4-morpholinepropanesulfonic acid. belong to a family of enzymes with diverse substrate specificity, ranging from multiple extracellular matrix components to growth factors, cytokines, and other proteinases (11Matrisian L. Trends Genet. 1990; 6: 121-125Abstract Full Text PDF PubMed Scopus (1532) Google Scholar, 12Woessner Jr., J.F. Ann. N. Y. Acad. Sci. 1994; 732: 11-21Crossref PubMed Scopus (436) Google Scholar). It was first recognized that matrix metalloproteinases play a role in angiogenesis, based on the observation that inhibition of MMP activity by endogenous tissue inhibitors of metalloproteinases or synthetic compounds could inhibitin vitro tube formation (13Montesano R. Orci L. Cell. 1985; 42: 469-477Abstract Full Text PDF PubMed Scopus (373) Google Scholar, 14Mignatti P. Rifkin D.B. Enzyme Protein. 1996; 49: 117-137Crossref PubMed Scopus (292) Google Scholar). Several studies have demonstrated that the gelatinases, MMP-2 and MMP-9, are involved in vascular cell migration and invasion assays (15Schnaper H.W. Grant D.S. Stetler-Stevenson W.G. Fridman R. D'Orazi G. Murphy A.N. Bird R.E. Hoythya M. Fuerst T.R. French D.L. Quigley J.P. Kleinman H.K. J. Cell Physiol. 1993; 156: 235-246Crossref PubMed Scopus (281) Google Scholar, 16Cornelius L.A. Nehring L.C. Roby J.D. Parks W.C. Welgus H.G. J. Invest. Dermatol. 1995; 105: 170-176Abstract Full Text PDF PubMed Scopus (128) Google Scholar). However, the regulation of MMP expression and activation during angiogenic events is not well understood. The recent cloning of several membrane-type MMPs (MT-MMPs) (17Sato H. Takino T. Okada Y. Cao J. Shinagawa A. Yamamoto E. Seiki M. Nature. 1994; 370: 61-65Crossref PubMed Scopus (2379) Google Scholar, 18Will H. Hinzmann B. Eur. J. Biochem. 1995; 231: 602-608Crossref PubMed Scopus (317) Google Scholar, 19Takino T. Sato H. Shinagawa A. Seiki M. J. Biol. Chem. 1995; 270: 23013-23020Abstract Full Text Full Text PDF PubMed Scopus (448) Google Scholar) that each contain a putative transmembrane domain and appear to have substrate specificity for the pro-MMP-2 has led to considerable speculation concerning the role of the cell surface in regulating proteolytic activity, and the extent to which MT-MMPs may be involved in controlling proteolytic cascades involving MMP-2 (20Will H. Atkinson S.J. Butler G.S. Smith B. Murphy G. J. Biol. Chem. 1996; 271: 17119-17123Abstract Full Text Full Text PDF PubMed Scopus (505) Google Scholar, 21Sato H. Kinoshita T. Takino T. Nakayama K. Seiki M. FEBS Lett. 1996; 393: 101-104Crossref PubMed Scopus (302) Google Scholar, 22Cao J. Rehemtulla A. Bahou W. Zucker S. J. Biol. Chem. 1996; 271: 30174-30180Abstract Full Text Full Text PDF PubMed Scopus (102) Google Scholar, 23Strongin A.Y. Collier I. Bannikov G. Marmer B.L. Grant G.A. Goldberg G.I. J. Biol. Chem. 1995; 270: 5331-5338Abstract Full Text Full Text PDF PubMed Scopus (1438) Google Scholar, 24Corcoran M.L. Hewitt R.E. Kleiner Jr D.E. Stetler-Stevenson W.G. Enzyme Protein. 1996; 49: 7-19Crossref PubMed Scopus (186) Google Scholar). We hypothesized that the controlled activation of metalloproteinases would be necessary for the in vitro remodeling of endothelial cells cultured in a collagen matrix, and that this activation would require the transcriptional up-regulation of specific MMPs, including MT-MMPs. Utilizing primary cultures of rat microvascular endothelial cells, we demonstrated a coordinate increase in expression of MMP-2 and MT1-MMP (MMP-14) following culture within a type I collagen matrix, and that activation of MMP-2 correlated temporally with MT1-MMP induction. MMP inhibitors blocked the activation of MMP-2, and prevented the establishment of multicell networks in both primary cell cultures and in microvessel explant cultures. Furthermore, the three-dimensional type I collagen matrix provided unique signals that could not be duplicated by culturing cells either within a Matrigel matrix, or on top of thin coatings of type I collagen, suggesting that a malleable type I collagen matrix is important in signal transduction of these events. Rat capillary endothelial cells were harvested from the epididymal fat pads of Sprague-Dawley rats and cultured as described by Madri and Williams (25Madri J.M. Williams S.K. J. Cell Biol. 1983; 97: 153-165Crossref PubMed Scopus (453) Google Scholar). Twelve rats were used per preparation, and experiments were performed on two separate isolations of cells. Briefly, cells were grown on gelatin-coated tissue culture plates (1.5% gelatin in phosphate-buffered saline) and maintained in Dulbecco's modified Eagle's medium (Life Technologies, Inc.) containing 25% sterile-filtered conditioned bovine aortic endothelial cell medium and 10% fetal bovine serum. For all experiments, cells were cultured in a monolayer on type I collagen-coated plates (12 μg/ml) or in three-dimensional type I collagen gels (2.5 mg/ml acid-soluble type I collagen, buffered with Earle's salt and neutralized with sterile NaOH) at a density of 1.0 × 106 cells/ml collagen as described previously (8Madri J.A. Pratt B.M. Tucker A.M. J. Cell Biol. 1984; 106: 1375-1384Crossref Scopus (459) Google Scholar). Aliquots of the cells in collagen suspension were placed in bacteriological Petri dishes and incubated for 10 min at 37 °C, allowing gel polymerization prior to addition of medium. Gels remained attached to the plastic dish for the duration of the experiment. In variant experiments, cells were plated on top of or within a Matrigel matrix per manufacturer's specifications (Collaborative Biomedical Products, Inc.), or on top of a thin layer of polymerized collagen type I gel. In all experiments, cells were cultured in Dulbecco's modified Eagle's medium containing 10% sera previously passed over a gelatin-Sepharose column, in the absence of conditioned medium, to minimize contamination with MMP-2 or MMP-9. Drug treatments included the addition of transforming growth factor β1 (TGFβ1) at a final concentration of 0.5 ng/ml, or phorbol 12-myristate 13-acetate (PMA) (10 nm) immediately after plating, and again on the third day of culture. The MMP inhibitors marimastat and batimastat (BB-2516 and BB-94; a generous gift from British Biotech, Inc.) were prepared as 6 mmstock solutions in Me2SO, and used at a final concentration of 0.6 μm. Microvessel fragments were isolated from rat epididymal fat pads (20 rats/isolate) as described for endothelial cell culture (25Madri J.M. Williams S.K. J. Cell Biol. 1983; 97: 153-165Crossref PubMed Scopus (453) Google Scholar). These fragments included small arterioles, capillaries, and venules. At the final stage of isolation, fragments were resuspended in type I collagen gel and cultured in the same conditions as the endothelial cell three-dimensional cultures, using triplicate dishes per culture condition. Medium was replaced daily. Some cultures were treated daily with marimastat (0.6 μm). After 3 days or 5 days of culture, media were collected and utilized for zymography analysis and the cultures were processed for histological examination. Collagen gels were fixed with 4% paraformaldehyde and processed using routine paraffin-embedding procedures. Sections (6 or 20 μm thickness) were stained with hematoxylin and eosin prior to microscopic examination. Cells were lysed in 120 mm Tris-HCl buffer (pH 8.7), 0.1% Triton X-100, 0.01% sodium azide, and 5% glycerol. Collagen gels were homogenized in this buffer. Lysates were pelleted to remove cellular and collagen debris, and protein in the supernatant was quantitated using a bicinchonic acid assay (BCA; Pierce). Media from the cultures were collected for analysis on occasions that cultures were used for histological analysis. 10 μg of protein, or 2 μl of unconcentrated culture media, per sample were prepared in non-denaturing loading buffer and size fractionated in a 10% SDS-polyacrylamide gel impregnated with 0.02% gelatin. The gels were then washed in 2.5% Triton X-100 for 1 h to remove SDS, washed two times with water, then incubated for 24 h at 37 °C in a 50 mm Tris-HCl buffer, pH 8.0, containing either 5 mm calcium chloride or 10 mm EDTA (negative control for MMP activity). Gels were subsequently fixed with 50% methanol and 10% acetic acid containing 0.25% Coomassie Blue R250. Gelatinase activity appeared as clear bands within the stained gel. Gels were dried and then scanned (300 d.p.i.) using an Arcus II scanner, and band intensities were calculated with Biomax image analysis software using a Power Macintosh 7100/80 computer. Cells were lysed in a radioimmune precipitation buffer (0.1% SDS, 0.5% sodium deoxycholate, and 1% Nonidet P-40 in phosphate-buffered saline) containing protease inhibitors (“Complete” buffer; Boehringer Mannheim). Lysates were pelleted to remove cellular debris and collagen, and then protein in the supernatant was quantitated using BCA. 10 or 20 μg of protein/sample were prepared in denaturing conditions and size fractionated in a 12% SDS-polyacrylamide gel. Gels were blotted onto polyvinylidene fluoride membranes using semidry blotting conditions. Membranes were blocked for 1 h (25 °C) in Tris-buffered saline containing 0.5% Tween 20 and 5% milk. Primary antibodies were diluted in blocking solution and incubations were done overnight (4 °C). Affinity-purified polyclonal anti-MMP-2 (used 1:2000) and anti-TIMP-2 (1:1000) were purchased from Chemicon, Inc. (Temecula, CA). Two affinity-purified polyclonal anti-MT-MMP antibodies were used at 1:1000 (gifts from Immunex, Inc. (Seattle, WA) and S. Weiss (University of Michigan, Ann Arbor, MI)). An affinity-purified polyclonal anti-vimentin antibody (prepared by J. Madri) was used to verify equal protein loading per lane, as we had previously assessed vimentin levels to remain constant between monolayer and three-dimensional conditions. 2T. L. Haas, S. J. Davis, and J. A. Madri, unpublished observations. Secondary antibody (goat anti-rabbit, horseradish preoxidase-conjugated; Amersham Corp.) was used at 1:5000. Enhanced chemiluminescence (ECL) detection (Amersham Corp.) was performed per manufacturer's instructions. Total cellular RNA was extracted from endothelial cells in monolayer or three-dimensional culture after solubilizing the cells or collagen gels in Trizol (Life Technologies, Inc.), according to manufacturer's instructions. RNA concentration was determined by spectrophotometer. 5 or 10 μg of total RNA was denatured in sample buffer (20 mm MOPS, 6% formaldehyde, 50% formamide), electrophoresed through a 1% agarose-formaldehyde gel, and then transferred to a GeneScreen Plus membrane (NEN Life Science Products) by capillary transfer using 10 × SSC. Prehybridization and hybridization were carried out according to manufacturer's directions. 32P-Labeled cDNA probes encoding mouse MMP-2 and human MT1-MMP (cDNA were kind gifts of K. Tryggvason (University of Oulu, Oulu, Finland) and M. Seiki (Kanazawa University, Kanazawa, Ishikawa, Japan), respectively) were prepared by random primer labeling (Stratagene, Inc.), separated from unincorporated nucleotides using NucTrap columns (Stratagene, Inc.) and added to fresh hybridization buffer. Blots were washed using two washes of 2 × SSC (25 °C), two washes of 2 × SSC, 2% SDS (65 °C), and two washes of 0.1 × SSC (25 °C), then exposed to HyperFilm (Amersham Corp.) for several hours or overnight. Blots were stripped and reprobed with a 28 S ribosomal probe (Ambion, Inc.) to normalize for RNA loading. Films were scanned (300 d.p.i.) using an Arcus II scanner, and band intensities were calculated using Biomax image analysis software. Intensities of experimental bands were normalized based on load, according to the measured 28 S densities. Statistical analyses were performed on the normalized values, using a Student's t test, with significance established asp < 0.05. To define MMP activities of endothelial cells grown as a monolayer on type I collagen (two-dimensional), or grown within a gel composed type 1 collagen (three-dimensional), we compared proteolytic activity between the two conditions using gelatin zymography. Gelatin zymography is a sensitive technique for the direct detection of any MMPs that can proteolyze gelatin (collagenases, gelatinases, stromelysins, and matrilysin). Lysates from monolayer cultures and three-dimensional cultures both contained gelatinase activity (Fig. 1 A). MMP-2 (72-kDa gelatinase) was seen almost entirely in the latent form in endothelial cell monolayers, with only 3.5 ± 1.9% (n = 8) in the active form (62 kDa). In contrast, 51 ± 4.5% of MMP-2 in the three-dimensional cultures was in the activated form, with the total amount of MMP-2 (latent + active) approximately 300 ± 75% that of two-dimensional cultures (n = 8). MMP-1 (interstitial collagenase; 52 kDa, latent size) was not evident in either culture condition, based on the complete absence of bands smaller than 62 kDa by gelatin zymography. Latent MMP-9 was not observed in monolayer cultures and was infrequently detected in three-dimensional cultures. Active MMP-9 was never detected. Interestingly, the gelatinase profile was not altered by treatment of cell cultures with angiogenic factors TGFβ1 or the phorbol ester PMA. Initially, we utilized lysates rather than collected media to ensure that all cell-associated MMP activity would be detected. Media collected from these cultures and assessed by zymography showed the same pattern of MMP-2 induction and activation with the exception that, after 1 day in culture, less MMP-2 protein (both latent and active) was detectable in the media than in the cell lysates; after 3 days of culture, levels of protein appeared similar in media and in lysates (data not shown). Northern (Fig. 1 B) and Western (Fig. 1 C) blotting were performed to confirm the low level, constitutive expression of MMP-2 in endothelial cell monolayer culture and the increase of both MMP-2 mRNA (Table I) and protein in three-dimensional culture. mRNA and protein levels were elevated after 1 day of three-dimensional culture, remained elevated after 3 days of three-dimensional culture, and were not affected by TGFβ1 or PMA treatment, consistent with zymography results.Table IComparison of changes in mRNA levels of MMP-2, MT1-MMP, and TIMP-2 in two- and three-dimensional cultureTwo-dimensionalThree-dimensionalUntreatedPMAUntreatedPMAMMP-211.1 ± 0.045.0 ± 1.81-aSignificantly different from the untreated samples (P < 0.05).6.6 ± 0.91-aSignificantly different from the untreated samples (P < 0.05).,1-bNo significant difference from the three-dimensional untreated samples (P > 0.05).MT1-MMP11.1 ± 0.013.2 ± 1.11-aSignificantly different from the untreated samples (P < 0.05).1.8 ± 0.41-aSignificantly different from the untreated samples (P < 0.05).,1-bNo significant difference from the three-dimensional untreated samples (P > 0.05).TIMP-2 4.0 kb10.86 ± 0.21.2 ± 0.62.1 ± 0.3 1.2 kb10.86 ± 0.20.9 ± 0.21.2 ± 0.21-a Significantly different from the untreated samples (P < 0.05).1-b No significant difference from the three-dimensional untreated samples (P > 0.05). Open table in a new tab Investigators using several cell systems have shown that activation of MMP-2 involves an MT-MMP (20Will H. Atkinson S.J. Butler G.S. Smith B. Murphy G. J. Biol. Chem. 1996; 271: 17119-17123Abstract Full Text Full Text PDF PubMed Scopus (505) Google Scholar, 21Sato H. Kinoshita T. Takino T. Nakayama K. Seiki M. FEBS Lett. 1996; 393: 101-104Crossref PubMed Scopus (302) Google Scholar, 22Cao J. Rehemtulla A. Bahou W. Zucker S. J. Biol. Chem. 1996; 271: 30174-30180Abstract Full Text Full Text PDF PubMed Scopus (102) Google Scholar, 23Strongin A.Y. Collier I. Bannikov G. Marmer B.L. Grant G.A. Goldberg G.I. J. Biol. Chem. 1995; 270: 5331-5338Abstract Full Text Full Text PDF PubMed Scopus (1438) Google Scholar, 24Corcoran M.L. Hewitt R.E. Kleiner Jr D.E. Stetler-Stevenson W.G. Enzyme Protein. 1996; 49: 7-19Crossref PubMed Scopus (186) Google Scholar). Although several MT-MMP genes have been cloned, MT1-MMP (or MMP-14) is the most thoroughly documented in its ability to activate MMP-2. Thus, we assessed the amount of MT1-MMP present in monolayer and three-dimensional cultures. Very little MT1-MMP mRNA or protein were detectable in monolayer cultures, but both mRNA and protein were induced in three-dimensional cultures (Table I; Fig. 2 A). By Western blotting, MT1-MMP was detected as a single band of approximately 63 kDa; smaller (processed) forms of the protein were not detectable. The time course of MT1-MMP induction (detectable within 1 day of three-dimensional culture) correlated with the detection of activated MMP-2 by zymography (see Fig. 1 A). MT1-MMP mRNA levels were insensitive to either TGFβ1 or PMA treatment. Tissue inhibitor of matrix metalloproteinases-2 (TIMP-2) is postulated, at low concentrations, to be a necessary component for the activation of MMP-2 (23Strongin A.Y. Collier I. Bannikov G. Marmer B.L. Grant G.A. Goldberg G.I. J. Biol. Chem. 1995; 270: 5331-5338Abstract Full Text Full Text PDF PubMed Scopus (1438) Google Scholar), and has been shown to function as a competitive inhibitor of MMPs at higher concentrations. Based on the large fraction of activated MMP-2 seen by gelatin zymography of three-dimensional cultures, we hypothesized that TIMP-2 levels were sufficient to allow activation of MMP-2, but low enough so as not to inhibit MT1-MMP or MMP-2 function, thus conferring an overall proteolytic phenotype to the culture. Northern and Western blotting established that TIMP-2 mRNA and protein levels in fact did not change in three-dimensional culture (Table I; Fig. 2 B). An increase in the amount of activated MMP-2 and no change in the level of TIMP-2 protein implied a shift to increased proteolytic activity of MMP-2. Again, TIMP-2 mRNA transcripts were not sensitive to PMA or TGFβ1 treatment. Coordinated expression of MT1-MMP with MMP-2 appeared consistent with MT1-MMP acting as the activator of pro-MMP-2. To strengthen this argument, we utilized the nonselective MMP inhibitors marimastat or batimastat (high affinity for multiple matrix metalloproteinases) (26Wojtowicz-Praga S.M. Dickson R.B. Hawkins M.J. Invest. New Drugs. 1997; 15: 61-75Crossref PubMed Scopus (419) Google Scholar), hypothesizing that if pro-MMP-2 were activated through an MMP-dependent pathway (i.e. MT1-MMP), then MMP inhibitors would block cleavage and activation of pro-MMP-2. One day treatment of three-dimensional cultures with marimastat (0.6 μm) blocked the cleavage of MMP-2 to the active 62-kDa form (Fig. 3). Inhibition of MMP-2 activation was also seen at 3 days after marimastat treatment. The total amount of MMP-2 protein in three-dimensional culture was not affected by marimastat treatment, as assessed by densitometry of zymography bands. Microvascular endothelial cells underwent morphological changes when cultured within a three-dimensional type I collagen matrix, as has been described previously (27Merwin J.R. Anderson J.M. Kocher O. Itallie C.M.V. Madri J.A. J. Cell Physiol. 1990; 142: 117-128Crossref PubMed Scopus (152) Google Scholar). Within a day of culturing, the cells took on an elongated shape, often with multiple extended processes. Continuing the culture, multicell interactions resembling tube structures were formed (Fig. 4 A). Treatment with TGFβ1 augmented this phenomenon, with observable formation of multicellular structures exhibiting lumen structures (Fig. 4 C). PMA has been used to induce the in vitroformation of tubes by human endothelial cells (13Montesano R. Orci L. Cell. 1985; 42: 469-477Abstract Full Text PDF PubMed Scopus (373) Google Scholar); however, the rat microvascular endothelial cells did not increase tube formation in response to PMA (data not shown). Treatment of the endothelial cell three-dimensional cultures with the MMP inhibitor marimastat, either in the presence or absence of TGFβ1, resulted in cells that appeared either rounded or stellate, as if unable to progress beyond the initial stages of cell attachment and spreading. In these cultures, there was a marked failure of cells to elongate and to establish tubelike structures (Fig. 4, B and D). In the presence of TGFβ1, marimastat treatment appeared to inhibit the organization of cell aggregates into structures containing lumina. Treatment of endothelial cell monolayer cultures with the MMP inhibitor had no noticeable effects on cell morphology or behavior (data not shown). The effects of MMP inhibitor treatment on three-dimensional cultures supported the involvement of MMPs in establishing lumina-containing, multicell networks. Cells in monolayer culture conditions never formed networks or tubelike structures. Explant cultures were used to further define the roles of MMP-2 and MT1-MMP in a more complex in vitro system. Thus, we assessed the zymographic profile and the morphology of microvessel fragments cultured within a type I collagen matrix. The microvessel fragments contained a mixed population of endothelial cells, smooth muscle cells and pericytes. Despite this mixed population of cells, we hypothesized that the MMP-2 induction and activation would occur in like manner to the endothelial cell cultures. Gelatin zymography was performed on lysates of freshly isolated fragments, and on media collected from 3-day explant cultures (Fig. 5). Lysates from freshly isolated blood vessel fragments contained multiple matrix metalloproteinases, including latent MMP-2, and several non-MMP gelatinases (as evidenced by bands appearing in the EDTA control gel). Gelatinolytic proteins other than MMP-2 detected in the explant lysates but not in the cultured endothelial cell lysates are most likely derived from one or several of the multiple non-endothelial cell types present within the explant cultures. Media collected after 3 days of culture in type I collagen showed a significant increase in the total amount of MMP-2 and in the fraction of active MMP-2. At this time point, multiple endothelial cell sprouts from the pre-existing vessel fragments were observable (Fig. 6 A). Notably, marimastat treatment of these cultures greatly reduced the amount of active MMP-2 (Fig. 5), as well as the extent of endothelial cell sprouting (Fig. 6 B). These findings are consistent with results using the endothelial cell cultures (see Fig. 3), thus strengthening the hypothesis that MMP-2 and MT1-MMP play specific roles in angiogenesis.Figure 6MMP inhibitors block angiogenic sprouting of microvessels in explant cultures. After 3 days of culture, many microvessel fragments formed multiple endothelial cell sprouts that migrated into the collagen matrix (A). However, marimastat treatment (0.6 μm) greatly reduced the number and length of the endothelial sprouts (B). In both panels, original microvessel fragments (denoted by black dots), had a rough surface appearance, and individual cells are not detectable. On the other hand, the endothelial sprouts (denoted by arrows) were thin and had a smooth surface, and endothelial cells comprising the sprouts were oriented parallel to the long axis of the sprout. Scale bar represents 100 μm. These images are representative fields of view from triplicate cultures of one isolate of rat microvessels.View Large Image Figure ViewerDownload (PPT) Finally, we considered whether the induction of MMP-2 protein and its activation were responses general to other types extracellular matrices or whether these responses were specific to the type I collagen matrix. Neither the induction of MMP-2 protein nor the activation of pro-MMP-2 occurred when endothelial cells were cultured either on, or within, a Matrigel matrix (Fig. 7 A), suggesti
Referência(s)