Membrane-Type Matrix Metalloproteinases in Human Dermal Microvascular Endothelial Cells: Expression and Morphogenetic Correlation
1998; Elsevier BV; Volume: 111; Issue: 6 Linguagem: Inglês
10.1046/j.1523-1747.1998.00416.x
ISSN1523-1747
AutoresVincent T. Chan, Dan Ning Zhang, Usha Nagaravapu, Kevin Hultquist, Luz Romero, G. Scott Herron,
Tópico(s)Cell Adhesion Molecules Research
ResumoMembrane-type matrix metalloproteinases (MT-MMP) activate the zymogen form of MMP-2/Gelatinase A on cell surfaces and are expressed in invasive tumors. We sought to identify and characterize MT-MMP in a nonmalignant cell type that undergoes a physiologic and reversible invasive phenotype during angiogenesis. Human dermal microvascular endothelial cells (HDMEC) were isolated from neonatal tissue and purified by anti-CD31 (PECAM) affinity beads. MT-MMP-1 and -3 transcripts were amplified by reverse transcriptase-polymerase chain reaction and northern blots showed a single 4.5 kB mRNA for MT-MMP-1 that was modulated by angiogenic factors and phorbol ester. Immunoblotting of reduced cellular extracts with different MT-MMP-1 antibodies showed the presence of the 63–65 kDa and 57–60 kDa forms, as well as additional forms at lower molecular weights. HDMEC membranes extracted with Triton X114 were incubated with gelatin-sepharose purified MMP-2 and MMP-9 to show activation of proenzymes. Pre-incubation of HDMEC with anti-MT-MMP-1 antibodies decreased proMMP-2 conversion activity only. The movement of HDMEC and the formation of tubule-like structures in three-dimensional collagen gels was markedly delayed by preincubation with the same anti-MT-MMP-1 antibodies. These results demonstrate the presence of MT-MMP in cutaneous microvascular cellsin vitro. Modulation of these cell surface proteinases by angiogenic factors, demonstration of multiple processed forms, and specific attenuation of HDMEC morphogenetic patterns in three-dimensional collagen gels implicate their potential roles in the formation of new blood vessels in the skin. Membrane-type matrix metalloproteinases (MT-MMP) activate the zymogen form of MMP-2/Gelatinase A on cell surfaces and are expressed in invasive tumors. We sought to identify and characterize MT-MMP in a nonmalignant cell type that undergoes a physiologic and reversible invasive phenotype during angiogenesis. Human dermal microvascular endothelial cells (HDMEC) were isolated from neonatal tissue and purified by anti-CD31 (PECAM) affinity beads. MT-MMP-1 and -3 transcripts were amplified by reverse transcriptase-polymerase chain reaction and northern blots showed a single 4.5 kB mRNA for MT-MMP-1 that was modulated by angiogenic factors and phorbol ester. Immunoblotting of reduced cellular extracts with different MT-MMP-1 antibodies showed the presence of the 63–65 kDa and 57–60 kDa forms, as well as additional forms at lower molecular weights. HDMEC membranes extracted with Triton X114 were incubated with gelatin-sepharose purified MMP-2 and MMP-9 to show activation of proenzymes. Pre-incubation of HDMEC with anti-MT-MMP-1 antibodies decreased proMMP-2 conversion activity only. The movement of HDMEC and the formation of tubule-like structures in three-dimensional collagen gels was markedly delayed by preincubation with the same anti-MT-MMP-1 antibodies. These results demonstrate the presence of MT-MMP in cutaneous microvascular cellsin vitro. Modulation of these cell surface proteinases by angiogenic factors, demonstration of multiple processed forms, and specific attenuation of HDMEC morphogenetic patterns in three-dimensional collagen gels implicate their potential roles in the formation of new blood vessels in the skin. human dermal microvascular endothelial cell matrix metalloproteinase membrane-type matrix metalloproteinase Microvascular endothelial cells actively remodel the extracellular matrix of basement membranes, provisional substrates, and mature interstitial tissues. These changes occur during both cutaneous wound repair and capillary sprouting in response to tumors and various pathologic skin conditions (Folkman and Shing, 1992Folkman J. Shing Y. Angiogenesis.J Biol Chem. 1992; 267: 10931-10934Abstract Full Text PDF PubMed Google Scholar,Birkedal-Hansen et al., 1993Birkedal-Hansen H. Moore Wgi Bodden M.K. et al.Matrix metalloproteinases: a review.Crit Rev Oral Biol Med. 1993; 4: 197-250Crossref PubMed Scopus (2565) Google Scholar;Clark, 1993Clark RaF Basics of cutaneous wound repair.J Dermatol Surg Oncol. 1993; 19: 693-706Crossref PubMed Scopus (177) Google Scholar). Expression of matrix degrading proteinases by microvascular endothelial cells is an early and crucial step in these processes. Both the plasminogen activator (PA)-plasmin system (Pepper et al., 1996Pepper M. Montesano R. Mandriota S. et al.Angiogenesis: a paradigm for balanced extracellular proteolysis during cell migration and morphogenesis.Enzyme Protein. 1996; 49: 138-162Crossref PubMed Scopus (197) Google Scholar) and matrix metalloproteinases (MMP) play key roles during angiogenesis (Liotta et al., 1991Liotta L.A. Steeg P.S. Stetler-Stevenson W.G. et al.Cancer metastasis and angiogenesis: an imbalance of positive and negative regulation.Cell. 1991; 64: 327-336Abstract Full Text PDF PubMed Scopus (2598) Google Scholar;Mignasti and Rifkin, 1996Mignasti P. Rifkin D.B. Plasminogen activators and matrix metalloproteinase in angiogenesis.Enzyme Protein. 1996; 49: 117-137Google Scholar). Like invasive tumor cells, microvascular endothelial cells migrate, proliferate, and may even circulate (Asahara et al., 1997Asahara T. Murohara T. Sullivan A. et al.Isolation of putative progenitor endothelial cells for angiogenesis.Sci. 1997; 275: 964-967Crossref PubMed Scopus (7356) Google Scholar) under specific conditions. The study of related similarities and differences between these two cell types serves as the foundation upon which much of our current understanding of tumor cell metastasis is built (Folkman, 1995Folkman J. Tumor Angiogenesis. W.B. Saunders, Philadelphia1995Google Scholar). Tight control of proteolytic events at specific locations and at different times during angiogenesis and tumor cell invasion is accomplished in several ways. Local basement membrane degradation at the tips of sprouting capillaries (Ausprunk and Folkman, 1977Ausprunk D.H. Folkman J. Migration and proliferation of endothelial cells in preformed and newly formed blood vessels during tumor angiogenesis.Microvascular Res. 1977; 14: 53-65Crossref PubMed Scopus (1008) Google Scholar) may depend on "invadopodial" structures that concentrate active proteinases at cell membrane projections (Chen, 1996Chen W.T. Proteases associated with invadopodia, and their role in degradation of extracellular matrix.Enzyme Protein. 1996; 49: 59-71Crossref PubMed Scopus (104) Google Scholar). These strucutures harness proteinase receptor systems that convert aqeuous phase enzymatic activity into solid state reactions, enormously increasing catalytic efficiency while compartmentalizing degradative events (Moscatelli and Rifkin, 1988Moscatelli D. Rifkin D.B. Membrane and matrix localization of proteinases: a common theme in tumor cell invasion and angiogenesis.Biochim Biophys Acta. 1988; 948: 67-85PubMed Google Scholar;Trelstad, 1988Trelstad R.L. The extracellular matrix is a soluble and solid-phase agonist and receptor.Arch Dermatol. 1988; 124: 706-708Crossref Scopus (3) Google Scholar;Plow and Miles, 1990Plow E.F. Miles L.A. Plasminogen receptors in the mediation of pericellular proteolysis.Cell Differ Dev. 1990; 32: 293-298Crossref PubMed Scopus (45) Google Scholar;Sedo et al., 1996Sedo A. Mandys V. Krepela S. et al.Cell membrane-bound proteases: not "only" proteolysis.Physiol Res. 1996; 45: 169-176Google Scholar). On endothelial cell membranes, these receptor systems include both the uPA receptor (Barnathan et al., 1990Barnathan E.S. Kup A. Kariko K. et al.Characterization of human urokinase-type plasminogen activator receptor protein and messenger RNA.Blood. 1990; 76: 1795-1806Crossref PubMed Google Scholar) and the αvβ3 integrin (Brooks et al., 1996Brooks P.C. Stromblad S. Sanders L.C. et al.Localization of matrix metalloproteinase MMP-2 to the surface of invasive cells by interaction with Integrin alpha-v Beta 3.Cell. 1996; 85: 683-693Abstract Full Text Full Text PDF PubMed Scopus (1395) Google Scholar). For invasive tumor cells and stromal cells, membrane-type matrix metalloproteinases (MT-MMP) appear to link active proteinases to cell surfaces (Sato et al., 1994Sato H. Takino T. Okada Y. et al.A matrix metalloproteinase expressed on the surface of invasive tumor cells.Nature. 1994; 370: 61-65Crossref PubMed Scopus (2315) Google Scholar;Strongin et al., 1995Strongin A.Y. Collier I. Bannikov G. et al.Mechanism of cell surface activation of 72 kDa type IV collagenase: isolation of the activated form of the membrane metalloprotease.J Biol Chem. 1995; 270: 5331-5338Crossref PubMed Scopus (1399) Google Scholar). MT-MMP form a subfamily of at least four members (MMP-14, -15, -16, and -17), the first three of which share sequence homology, intracellular activation by the furin protease system, a COOH-terminal transmembrane domain, and the ability to activate pro-MMP-2 via formation of a ternary complex with TIMP-2 (Sato et al., 1997Sato H. Okada Y. Seiki M. et al.Membrane-type matrix metalloproteinases (MT-MMP) in cell invasion.Thrombosis Haemostasis. 1997; 78: 497-500PubMed Google Scholar). In fact, MT-MMP may serve dual roles as both a cell surface "docking" system for activating matrix-bound proMMP-2 at invadapodial sites (Nakahara et al., 1997Nakahara H. Howard L. Thompson et al.Transmembrane/cytoplasmic domain-mediated membrane type 1-matrix metalloprotease docking to invadopodia is required for cell invasion.Proc Natl Acad Sci USA. 1997; 94: 7959-7964Crossref PubMed Scopus (351) Google Scholar), and directly as a matrix-degrading proteinase with broad collagenolytic, glycoproteolytic, and gelatinolytic activities (Ohuchi et al., 1997Ohuchi E. Imai K. Fujii Y. et al.Membrane type 1 matrix metalloproteinase digests interstitial collagens and other extracellular matrix macromolecules.J Biol Chem. 1997; 272: 2446-2451Crossref PubMed Scopus (804) Google Scholar). Stromal cell expression of MT-MMP during cutaneous wound repair and tumor formation strongly suggests involvement of cell surface proteinases in physiologic processes of the skin (Okada et al., 1997Okada A. Tomasetto C. Lutz Y. et al.Expression of matrix metalloproteinases during rat skin wound healing: evidence that membrane type-1 matrix metalloproteinase is a stromal activator of pro-gelatinase A.J Cell Biol. 1997; 137: 67-77Crossref PubMed Scopus (191) Google Scholar). We report the characterization of MT-MMP-1 expression by human dermal microvascular endothelial cells (HDEMC)in vitro at the transcript and protein levels. The same antibodies that recognize MT-MMP-1 species by immunoblotting can also modulate proMMP-2 activation activity and attenuate the morphogenetic pattern of HDMEC tubule formation in three-dimensional collagen gels. Dispase was supplied by Collaborative Research (Bedford, MA). Iscove's medium, penicillin, and streptomycin were purchased from Life Technologies GIBCO BRL (Grand Island, NY). Newborn bovine serum was obtained from Irving Scientific (Santa Ana, CA), and human maternal serum from routine blood samples taken from healthy prepartum women at Stanford Medical Center. Gentamicin, amphotericin B, dibutyryl 3′5′-cyclic adenosine monophosphate (cAMP), isobutyl-methylxanthine (IBMX), gelatin, bovine serum albumin, and hypoxanthine were all products supplied by Sigma (St. Louis, MO). Medium was sterile prepared, aliquoted, and frozen at –20°C to prevent degradation of cAMP. Reagents used for purification of endothelial cells wereUlex europeasus 1-lectin (UEA-1) from Sigma, Dynabeads M-450 tosylactivated from Dynal (Oslo, Norway), and anti-human PECAM-1 antibody from Endogen (Woburn, MA). Mouse monoclonal antibodies [clone 114–1F2 (MTAb-1), clone 114–6G6 (MTAb-3)] and rabbit polyclonal antibody (MTAb-2) all directed against the same synthetic peptide spanning residue #160–173 within the catalytic domain of human MT-MMP-1, were from OncoGene Research Products (Calbiochem, La Jolla, CA). Rabbit polyclonal antibody directed against the "hinge" region of human MT-MMP-1 (MTAb-4) was from Chemicon (Temescula, CA). Mouse IgG (Sigma, St Louis, MO), rabbit IgG (Endogen), and BP180 monoclonal antibody (LAD-1–123 antigen) was a generous gift from Dr Peter Marinkovich (Stanford University, Stanford, CA). Anti-human Von Willebrand Factor VIII (F-VIII) was from DAKO. Cultures of human dermal microvascular endothelial cells were established by modifications of a previously described method (Karasek, 1989Karasek M.A. Microvascular endothelial cell culture.J Invest Dermatol. 1989; 93: 33S-38SAbstract Full Text PDF PubMed Scopus (63) Google Scholar;Normand and Karasek, 1995Normand J. Karasek M.A. A method for the isolation and serial propagation of keratinocytes, endothelial cells and fibroblasts from a single punch biopsy of human skin.In Vitro. 1995; 31: 447-455Google Scholar;Romero et al., 1997Romero L.I. Zhang D.-N. Herron G.S. et al.Interleukin-1 induces major phenotypic changes in human skin microvascular endothelial cells.J Cell Physiol. 1997; 173: 84-92Crossref PubMed Scopus (64) Google Scholar). Foreskin tissue obtained following routine circumcision of healthy newborns was collected in Hanks' balanced salt solution, supplemented with 2% newborn bovine serum, 200 U penicillin per ml, 100 μg streptomycin per ml, 20 μg gentamicin per ml, and 10 μg amphotericin B per ml. Tissue was cut into 5 mm square fragments and incubated overnight with 2500 U dispase per ml at 4°C. Epidermis was gently separated from the dermis and outward pressure was applied to the remaining dermis to extrude microvascular-associated cells. After 5 min centrifugation of the cell suspension at 1000 r.p.m., cells were resuspended in complete growth medium, consisting in Iscove's media supplemented with 8% newborn bovine serum, 2% human prepartum maternal serum, 2 × 10–4 M cAMP, 3.5 × 10–5 M IBMX, 1 × 10–4 M Hypoxanthine, 1.7 × 10–5 M thymidine, 200 U penicillin per ml, 100 μg streptomycin per ml, 10 μg gentamicin per ml, and 250 μg amphotericin B per ml. Cells were then plated onto 1% gelatin-coated 35 mm tissue culture plastic dishes (Becton Dickinson Labware, Lincoln Park, NJ) and incubated in a humidified atmosphere of 5% CO2 in air at 37°C. HDMEC from a pool of at least 10 foreskins were isolated by a double purification method using paramagnetic beads coated withUlex europeous lectin and anti-PECAM monoclonal antibodies (Romero et al., 1997Romero L.I. Zhang D.-N. Herron G.S. et al.Interleukin-1 induces major phenotypic changes in human skin microvascular endothelial cells.J Cell Physiol. 1997; 173: 84-92Crossref PubMed Scopus (64) Google Scholar). Unpurified, dermal microvascular cells prepared as described above represent a variable mixture of dermal fibroblasts (10%–20%), α-actin (+) pericytes/smooth muscle cells (10%–30%), endothelial cells (50%–80%), and small numbers of unidentified spindle and dendritic cells. Lectin-coated beads were used first followed by a second round of purification using anti-PECAM beads. Briefly, after primary cultures reached 80%–100% confluencey, cells were trypsinized and resuspended in Hanks' balanced salt solution containing 5% newborn bovine serum, mixed with lectin- or anti-PECAM coated paramagnetic beads at a concentration of 3–5 beads per cell and incubated for 30 min at 20°C. Endothelial cells bound to affinity beads were recovered with the magnetic particle concentrator, and were referred asUlex(+) or PECAM(+) cells. The remaining unbound cells were referred asUlex(–) or PECAM(–) cells and were used for some immunoblotting experiments described below. The selected population was then grown in complete media for at least one passage and evaluated for the expression of endothelial cell markers (Factor VIIIa and PECAM) by immunostaining and by fluorescence-activated cell sorter analysis. Primary HDMEC cells grown to passage 3–5 were routinely used for experiments described below and represented at least 95%–98% positive Factor VIIIa by fluorescence-activated cell sorter analysis. Dermal fibroblasts were prepared from explants of neonatal tissue in Dulbecco's modified Eagle's medium with 10% fetal calf serum as previously described (Normand and Karasek, 1995Normand J. Karasek M.A. A method for the isolation and serial propagation of keratinocytes, endothelial cells and fibroblasts from a single punch biopsy of human skin.In Vitro. 1995; 31: 447-455Google Scholar). HT1080 (ATCC#CCL-121) were thawed and grown to confluence in Eagle's minimal essential medium containing 10% heat inactivated fetal calf serum without antibiotics. Human recombinant IL-1β was obtained from Beckton Dickinson (Bedford, MA). Human recombinant bFGF was obtained from Life Technologies GIBCO BRL. One unit (U) of cytokine was established as the ED 50 of specific bioactivity and was 0.35 ng per ml for IL-1β and 10 ng per ml for bFGF. All cytokines were dissolved in sterile distilled nonpyrogenic water for irrigation (Abbot Laboratories, Chicago, IL) containing 1% endotoxin free bovine serum albumin. Dilution of the stock solutions were prepared at one time, aliquoted, and stored at –70°C until use. A new aliquote of cytokine was used for each experiment. Reverse transcriptase-polymerase chain reaction (RT-PCR) Total RNA was isolated from HDMEC cultures with RNA Stat-60 (Tel Test) Ten nanograms of total RNA was used as template for cDNA synthesis in a volume of 50 μl according to manufacturer's recommendations (Invitrogen, Carlsbad, CA), using Oligo dT as a primer for reverse transcription For PCR amplification, 3 μl of cDNA was used as template and amplification conditions were 95°C for 5 min followed by 95°C for 45 s, 55°C for 1 min, and 72°C for 1 min for 40 cycles in a Perkin Elmer Cetus 9600 thermal cycler Amplification was performed in a total volume of 50 μl containing 15 mM MgCl2, 01 mM of each nucleotide, 5 pmol of each primer, and 25 U of Taq Polymerase (Perkin Elmer, Norwalk, CN). All primer sequences were analyzed by the Blast homology search program (NCBI, Basic Blast, Entrez) using Genebank sequences (human MT-MMP-1 D26512 and human MT-3 MMP D50477) Low homology scores were found versus other human MMP sequences for the following primers To amplify the 328 bp MT-MMP-1 gene product the 5′ upstream sense primer was cgaagcctggctacagcaatatg and the 3′ downstream anti-sense primer was gagtatgccacatacgaggccattc; to amplify the 318 bp MT-3 MMP gene product the 5′ upstream sense primer was ggatggatacccaatgcaaattac and the 3′ downstream anti-sense primer was ctatcccaagccaatcacagtctgg Aliquots of 10 μl of the PCR products were analyzed by 15% agarose (FMC BioProducts, Natick, MA) gel electrophoresis The 328 bp MT-MMP-1 PCR product was subcloned into the pCRII vector (Invitrogen) and sequenced on both strands using the MT-MMP-1 5′ sense primer and the T7 primer There were no differences in sequences between the GeneBank sequence and the PCR amplified product Total RNA was isolated from purified HDMEC and HT1080 cells after treatment with phorbol myristate acetate (PMA), IL-1β, TNFα, VEGF(121), or bFGF for 16 h in serum-free media using the guanidine thiocyanate method (QIAGEN, Valencia, CA). Five micrograms of total RNA and RNA standards (Gibco) were electrophoresed on 1% formaldehyde/agarose denaturing gels (Ambion, Austin, TX), transferred overnight to nylon membranes (Boehringer, Indianapolis, IN), and cross-linked with UV light (Stratalinker 2400). Membranes were hybridized with DIG-labeled MT-MMP-1 and glyceraldehyde 3 dehydrogenase probes (Boehringer) overnight at 55°C, followed by washing according to the manufacturer's instructions. Hybridization signals were visualized by the luminescence detection kit (Boehringer) using Kodak film and bands were quantitated using the Digital Gel Imaging system by Kodak according to the manufacturer's instructions. Confluent and subconfluent cultures of HDMEC either untreated or treated overnight with 100 ng PMA per ml under serum-free conditions, confluentUlex (–) dermal cells, and HT1080 cell cultures were rinsed with phosphate-buffered saline, trypsinized, pelleted, and resuspended in phosphate-buffered saline. Cells were immediately lyzed by adding an equal volume of 2×Laemli sample buffer containing 10% β-mercaptoethanol and heating for 5 min at 95°C. All samples were normalized to equal protein loadings by OD 280 before subjecting to 10% sodium dodecyl sulfate/polyacrylamide gel electrophoresis. Proteins were transferred to nitrocellulose at 0.3 mA per cm2 overnight at 4°C. Membranes were blocked with 5% dry milk for 30 min, followed by incubation with primary antibodies for 6–12 h at 4°C as follows: 1:100 (MTAb-1, MTAb-3), 1:1000 (MTAb-2), 1:5000 (MTAb-4). After washing, bound antibodies were visualized by incubation with 1:2000 peroxidase-conjugated secondary antibody for 1 h at 20°C followed by visualization with the enhanced chemoluminescence western blotting detection system (Amersham, Arlington Heights, IL). Two different prestained protein standards were used: Rainbow Marker system (BioRad, Hercules, CA) and Benchmark prestained MW Protein Standards (Gibco). Acylamide was from Amresco (Solon, OH) and Tris/Glycine/sodium dodecyl sulfate was from BioRad. HDMEC and HT1080 were cultured to confluency and rinsed three times with plain Iscove's media. Anti-MT-MMP-1 monoclonal antibody 114–1F2 (MTAb-1) was added to cells at a final concentration of 5 μg per ml and incubated at room temperature for 1 h. Antibody was removed and the cells were washed three times with plain Iscove's media. Fresh 0.75% paraformaldehyde in phosphate-buffered saline was added and incubated for 5 min at room temperature. Paraformaldehyde was removed and the culture was rinsed three times with Tris-buffered saline (TBS; 50 mM Tris-HCl pH 7.4, 150 mM NaCl) and membrane extractions were performed. Triton X114 extraction of integral membrane proteins was performed as described byLewalle et al., 1995Lewalle J.M. Munaut C. Pichot B. et al.Plasma membrane-dependent activation of gelatinase A in human vascular endothelial cells.J Cellular Physiol. 1995; 165: 475-483Crossref PubMed Scopus (62) Google Scholar. One milliliter of 1.5% Triton X114 in TBS was added to each 60 mm culture plate of HDMEC and incubated at 4°C for 30 min. The samples were scraped and homogenized with a pasture pipette on ice followed by centrifuging at 12,000 ×g for 5 min at 4°C to remove extracellular matrix. Detergent and aqueous phases were partitioned by placing them in a 37°C water bath for 5 min. The samples were centrifuged at 12,000 ×g for 2 min and the aqueous (upper) phase was discarded. The detergent (lower) phase was washed three times by addition of ice cold TBS with subsequent incubation at 37°C and centrifugation as described above. The final detergent phase containing solubilized membrane proteins was used for gelatinase activation assays. HDMEC were cultured to confluency in a T175 flask and treated overnight with PMA. Supernatant was collected and incubated with 100 μl packed gelatin sepharose beads (Sigma) at 4°C. Beads were centrifuged and washed three times with ice cold TBS. The gelatinases were eluted with 1 mg gelatin per ml in TBS solution in one-fourth the volume of the original supernatant. Zymography was performed to detect activity. Purified unactivated gelatinases were added to the membrane extracts and incubated for 72 h at 37°C with gentle rotation. As a control, 1.5% Triton X114 alone was incubated with purified gelatinases alone under identical conditions. Gelatin zymography was then performed to determine activation of the gelatinases by the membrane extracts according to published methods (Herron et al., 1986Herron G.S. Banda M.J. Clark E.J. et al.Secretion of metalloproteinase by stimulated capillary endothelial cells. II. Expression of collagenase and stromelysin activities is regulated by endogenous inhibitors.J Biol Chem. 1986; 261: 2814-2818Abstract Full Text PDF PubMed Google Scholar). Confluent monolayer HDMEC at the second or third passage were incubated overnight in MTAb-1 antibody (final concentration 5 μg per ml) or BP180 monoclonal antibody (LAD-1) at the same concentration. Media was removed and the cultures were overlaid with a 1:1 mixture of Vitrogen 100 (Collagen Biomaterials, Montvale, NJ) in 2×Iscove's media, pH 7.4. Solidification of the collagen gel occurred within 30 min. Cells were then photographed at different times using a Nikon Diaphot inverted microscope. Confluent monolayers of primary HDMEC expressed both MT-1 and MT-3 MMP transcripts constitutively as detected by RT-PCR using primers specific for unique sequences in the two MT-MMP genes (data not shown). A single RT-PCR product was obtained for each species and the 328 bp MT-MMP-1 product was excised from the gel, subcloned, and sequenced to confirm its identity. Because all MT-MMP share sequence homology and because multiple MT-MMP species can be coexpressed in the same tissue, we sought to characterize MT-MMP-1 transcript expression by stimulated HDMEC using northern blot hybridization to ensure that a single mRNA transcript could be detected and followed under different conditions.Figure 1 shows one 4.5 Kb hybridization signal when total RNA was reacted with the (sequenced) MT-MMP-1 PCR product under all conditions tested. There was no evidence of different sized splice variants or hybridization to other MT-MMP gene products and MT-MMP-1 transcript levels were specifically increased by a variety of different angiogenic agents (Figure 1a. The amount of MT-MMP-1 transcript expressed by phorbol-stimulated HDMEC represents a greater than 2-fold increase in hybridization signal relative to unstimulated HDMEC and is comparable with unstimulated HT1080 fibrosarcoma cells (Figure 1c). Comparatively less but moderate increases in HDMEC MT-MMP-1 mRNA levels were also observed after 16 h stimulation with IL-1β, TNF-α, VEGF, and bFGF (Figure 1a. Dermal fibroblasts showed levels of MT-MMP-1 mRNA comparable with HDMEC (data not shown). To verify our transcript studies and investigate the molecular forms of MT-MMP-1 protein expressed by different dermal cell types we compared the MT-MMP-1 protein expressed by HDMEC to that expressed by both HT1080 cells (a standard in previous studies) and "Ulex(–)" dermal perivascular cells using immunoblotting.Figure 2 shows western analysis of MT-MMP-1 in cellular lysates using four different MT-MMP-1 antibodies. Those antibodies directed against the same MT-MMP-1 epitope within the catalytic domain gave slightly different results on immunoblots of HDMEC, HT1080, and Ulex(–) cells (Figure 2a–c). Figure 2(a) shows four prinicipal MT-MMP-1 forms in Ulex(–) cells and HDMEC and three forms in HT1080 cells: (i) a prominent species at ≈63–65 kDa that is not well visualized in HT1080; (ii) a species at 57–60 kDa that, in HT1080 s, appears to be a doublet (Figure 2a,lane 4); (iii) a faint species at 50–53 kDa; and (iv) a species at 43–44 kDa. Ulex(–) and HT1080 cells reacted with MTAb-1 to yield a more intense band at 57–60 kDaversus HDMEC. PMA treatment of HDMEC resulted in the loss of the 63–65 kDa form (Figure 2a,lane 3). Figure 2b shows monoclonal MTAb-3 reacted with only a single species in HDMEC (57–60 kDa) at low protein concentrations and low antibody titers (Figure 2b,lane 1). Higher protein loadings resulted in the appearance of the same immunoreactive species in HDMEC lysates as HT1080 cells: (i) 63–65 kDa; (ii) 57–60 kDa, which in HT180 cells appears as a doublet; (iii) 50–53 kDa; and (iv) faint 43–44 kDa (Figure 2b,lanes 3 and 4). Several lower molecular weight immunoreactive species appear in lysates at higher protein concentrations. InFigure 2c polyclonal MTAb-2 reacted with two principle species in HT1080 cells at 57–60 kDa and 43–44 kDa, whereas HDMEC contained these species in addition to a prominent 63–65 kDa form. PMA treatment decreased the 63–65 kDa form (Figure 2c,lane 3) but had little or no effect on the 57–60 kDa and 43–44 kDa species, consistent with the data inFigure 2a (lanes 2 and 3). InFigure 2d, analysis using polyclonal MTAb-4 directed against the "hinge region" of MT-MMP-1 reveals two major immunoreactive species in HT1080 cells and HDMEC of 63–65 kDa and 57–60 kDa (Figure 2dlanes 1–5). Each of these species is clearly a doublet in both cell types. Ulex(–) cells show these same species in addition to the 43–44 kDa species (Figure 2d) (lane). We found that protein loading, titering of antibodies, and blocking procedures were critical in obtaining consistent results by immunoblotting. Higher antibody concentrations resulted in the appearance of multiple higher and lower sized immunoreactive species that varied between different cell culture preparations. Immunoreactive species that appeared to be most sensitive to variation of culture and blotting conditions were the 43–44 kDa and 50–53 kDa bands (data not shown). We speculate that it is therefore possible that the 43–44 kDa and 50–53 kDa species represent cross-reactive proteins unrelated to MT-MMP-1. Control studies using either mouse IgG or a monoclonal antibody specific for the BP180 antigen or rabbit IgG as primary antibodies showed no specific bands on immunoblots of HDMEC, Ulex(–), or HT1080 s (data not shown). Similarly, omission of primary antibody showed no specific protein species in the size range described above. We reasoned that since MTAb recognize several MT-MMP-1 species in cellular lysates by immunoblotting, the MTAb directed against the catalytic domain of MT-MMP-1 may actually block the activity of the active enzyme in HDMEC membranes. To test this hypothesis, affinity purified pro-gelatinases were mixed with Triton X114-extracted HDMEC for 12 h followed by gelatin zymography (seeMaterials and Methods). MTAb-1 was then used to block catalytic activity with or without paraformaldehyde cross-linkingin vitro. The latter was added to ensure that the MTAb-1:MT MMP complex would not dissociate
Referência(s)