Fibrin and Collagen Differentially Regulate Human Dermal Microvascular Endothelial Cell Integrins: Stabilization of αv/β3 mRNA by Fibrin1
1999; Elsevier BV; Volume: 113; Issue: 6 Linguagem: Inglês
10.1046/j.1523-1747.1999.00786.x
ISSN1523-1747
AutoresXiaodong Feng, Richard A.F. Clark, Dennis K. Galanakis, Marcia G. Tonnesen,
Tópico(s)Blood properties and coagulation
ResumoIntegrin αvβ3 is specifically but transiently expressed on the tips of capillary sprouts as they invade the fibrin clot during angiogenesis of cutaneous wound repair. Specific blocking of αvβ3 function inhibits granulation tissue formation in cutaneous wounds. The mechanisms of regulation of αvβ3 expression on human dermal microvascular endothelial cells, however, have not been fully delineated. As αvβ3 was highly expressed on capillary sprouts in 5 d wounds rich in fibrin, but was almost undetectable on blood vessels in 7 d wounds rich in collagen, we hypothesized that the extracellular matrix environment could regulate human dermal micro- vascular endothelial cell αvβ3 expression. To address this, human dermal microvascular endothelial cells were cultured on surfaces coated with collagen, fibronectin, and gelatin, and mRNA levels of integrin αv/β3 were determined. Compared with human dermal microvascular endothelial cells on collagen, mRNA levels of αv/β3 were higher in human dermal microvascular endothelial cells on fibronectin and on gelatin. To simulate the in vivo environment better, human dermal microvascular endothelial cells cultured on collagen were overlaid by fibrin or collagen gels prior to assessment of αv/β3 mRNA levels. αv/β3 mRNA levels were higher in human dermal microvascular endothelial cells surrounded by a three-dimensional fibrin gel compared with a collagen gel, whether angiogenic factors were present or absent. As modulation of mRNA stability is a potential regulatory mechanism for integrin expression, integrin subunit mRNA stability was assessed. β3 mRNA decayed much faster than αv, α2, and β1 mRNA. Three-dimensional fibrin gels enhanced αv/β3 mRNA stability compared with collagen gels. We propose that the provisional matrix molecules in the wound clot regulate angiogenesis associated with cutaneous wound repair through their modulation of integrin receptor expression. Integrin αvβ3 is specifically but transiently expressed on the tips of capillary sprouts as they invade the fibrin clot during angiogenesis of cutaneous wound repair. Specific blocking of αvβ3 function inhibits granulation tissue formation in cutaneous wounds. The mechanisms of regulation of αvβ3 expression on human dermal microvascular endothelial cells, however, have not been fully delineated. As αvβ3 was highly expressed on capillary sprouts in 5 d wounds rich in fibrin, but was almost undetectable on blood vessels in 7 d wounds rich in collagen, we hypothesized that the extracellular matrix environment could regulate human dermal micro- vascular endothelial cell αvβ3 expression. To address this, human dermal microvascular endothelial cells were cultured on surfaces coated with collagen, fibronectin, and gelatin, and mRNA levels of integrin αv/β3 were determined. Compared with human dermal microvascular endothelial cells on collagen, mRNA levels of αv/β3 were higher in human dermal microvascular endothelial cells on fibronectin and on gelatin. To simulate the in vivo environment better, human dermal microvascular endothelial cells cultured on collagen were overlaid by fibrin or collagen gels prior to assessment of αv/β3 mRNA levels. αv/β3 mRNA levels were higher in human dermal microvascular endothelial cells surrounded by a three-dimensional fibrin gel compared with a collagen gel, whether angiogenic factors were present or absent. As modulation of mRNA stability is a potential regulatory mechanism for integrin expression, integrin subunit mRNA stability was assessed. β3 mRNA decayed much faster than αv, α2, and β1 mRNA. Three-dimensional fibrin gels enhanced αv/β3 mRNA stability compared with collagen gels. We propose that the provisional matrix molecules in the wound clot regulate angiogenesis associated with cutaneous wound repair through their modulation of integrin receptor expression. endothelial basal medium human dermal microvascular endothelial cells vascular endothelial cell growth factor Angiogenesis, an essential component of wound healing, tumor growth, and morphogenesis, includes the cellular processes of adhesion, invasion, migration, proliferation, and capillary tube formation of vascular endothelial cells (Arnold and West, 1991Arnold F. West D.C. Angiogenesis in wound healing.Pharmacol Ther. 1991; 52: 407-422Crossref PubMed Scopus (302) Google Scholar;Folkman et al., 1992Folkman J. Shing T. Angiogenesis J Biol Chem. 1992; 267: 10931-10934Abstract Full Text PDF PubMed Google Scholar;Folkman, 1994Folkman J. Angiogenesis and breast cancer.J Clin Oncol. 1994; 12: 441-443Crossref PubMed Scopus (221) Google Scholar;Bischoff, 1997Bischoff J. Cell adhesion and angiogenesis.J Clin Invest. 1997; 99: 373-375Crossref PubMed Scopus (104) Google Scholar). This highly regulated event involves complex, dynamic interactions between microvascular endothelial cells and extracellular matrix (ECM) proteins. Microvascular endothelial cell surface receptors specific for fibrin-rich provisional ECM are believed to be important in mediating the angiogenesis that occurs during wound repair and tumor growth (Madri et al., 1996Madri J.A. Sankar S. Romanic A.M. Angiogenesis.in: Clark R.A.F. The Molecular and Cellular Biology of Wound Repair. Plenum Press, New York1996: 355-371Google Scholar). Integrins, members of a superfamily of cell surface receptors, are transmembrane, noncovalently linked heterodimers consisting of one α-chain and one β-chain. Integrins recognize one or more specific ECM proteins (Hynes, 1992Hynes R.O. Integrins: Versatility, modulation, and signaling in cell adhesion.Cell. 1992; 69: 11-25Abstract Full Text PDF PubMed Scopus (9002) Google Scholar). Of the many endothelial integrin receptors, only the αvβ3 receptor is capable of recognizing all the provisional matrix proteins including fibrin and fibronectin (Cheresh et al., 1989Cheresh D.A. Berliner S.A. Vicente V. Ruggeri Z.M. Recognition of distinct adhesive sites on fibrinogen by related integrins on platelets and endothelial cells.Cell. 1989; 58: 945-953Abstract Full Text PDF PubMed Scopus (274) Google Scholar;Charo et al., 1990Charo I.F. Nannizzi L. Smith J.W. Cheresh D.A. The vitronectin receptor αvβ3 binds fibronectin and acts in concert with α5β1 in promoting cellular attachment and spreading on fibronectin.J Cell Biol. 1990; 111: 2795-2800Crossref PubMed Scopus (231) Google Scholar). Recently, αvβ3 has been identified as a marker of angiogenesis (Brooks et al., 1994Brooks P.C. Clark R.A.F. Cheresh D.A. Requirement of vascular integrin αvβ3 for angiogenesis.Science. 1994; 264: 569-571Crossref PubMed Scopus (2732) Google Scholar). It is minimally, if at all, expressed on resting or normal blood vessels, but is significantly upregulated on vascular cells within human tumors and in other biologic processes involving angiogenesis. Monoclonal antibody to αvβ3 can block angiogenesis induced by basic fibroblast growth factor (bFGF), tumor necrosis factor-α, or human melanoma fragments in the chick chorioallantoic membrane assay (Brooks et al., 1994Brooks P.C. Clark R.A.F. Cheresh D.A. Requirement of vascular integrin αvβ3 for angiogenesis.Science. 1994; 264: 569-571Crossref PubMed Scopus (2732) Google Scholar). In a porcine cutaneous wound healing model, αvβ3 is focally and transiently expressed on the tips of capillary sprouts invading the wound clot, rich in fibrin and fibronectin. Addition of αvβ3 antagonists, including specific monoclonal antibodies or cyclic peptides, to healing wounds transiently inhibits granulation tissue formation (Clark et al., 1996Clark R.A.F. Tonnesen M.G. Gailit J. Cheresh D.A. Transient functional expression of αvβ3 on vascular cells during wound repair.Am J Pathol. 1996; 148: 1407-1421PubMed Google Scholar). Thus, modulation of αvβ3 expression on microvascular endothelial cells can regulate angiogenesis. The regulatory mechanism of αvβ3 expression, however, is not completely understood. It has been reported that angiogenic factors, such as bFGF and vascular endothelial cell growth factor (VEGF), induce increased levels of αvβ3 expression by cultured human microvascular endothelial cells (Enenstein et al., 1992Enenstein J. Waleh N.S. Kramer R.H. Basic FGF and TGF-β differentially modulate integrin expression of human microvascular endothelial cells.Exp Cell Res. 1992; 203: 499-503Crossref PubMed Scopus (151) Google Scholar;Sepp et al., 1994Sepp N.T. Li L. Lee K.H. Brown E.J. Caughman S.W. Lawley T.J. Swerlick R.A. Basic fibroblast growth factor increases expression of the αvβ3 integrin complex on human microvascular endothelial cells.J Invest Dermatol. 1994; 103: 295-299Abstract Full Text PDF PubMed Scopus (115) Google Scholar;Senger et al., 1996Senger D.R. Ledbetter S.R. Claffey K.P. Papadopoulos-Sergiou A. Perruzzi C.A. Detmar M. Stimulation of endothelial cell migration by vascular permeability factor/vascular endothelial growth factor through cooperative mechanisms involving the αvβ3 integrin, osteopontin, and thrombin.Am J Pathol. 1996; 149: 293-305PubMed Google Scholar). Nevertheless, it is difficult to explain how these diffusable factors produce the focal, transient expression of αvβ3 on the tips of angiogenic capillary sprouts invading the fibrin clot during wound repair. Recently it has been suggested that the ECM may participate in control of integrin expression. For example, alterations of the ECM environment can regulate the expression of integrins induced by platelet-derived growth factor (PDGF) on dermal fibroblasts (Xu and Clark, 1996Xu J. Clark R.A.F. Extracellular matrix alters PDGF regulation of fibroblast integrins.J Cell Biol. 1996; 132: 239-249Crossref PubMed Scopus (184) Google Scholar). α3β1 and α5β1, provisional matrix integrins, are induced by PDGF when cells are in a fibrin or fibronectin environment, but not in a collagen milieu. In contrast, α2β1, a collagen receptor, is super-induced by PDGF when the cells are in a collagen gel compared with a fibrin gel.Delcommenne and Streuli, 1995Delcommenne M. Streuli C.H. Control of integrin expression by extracellular matrix.J Biol Chem. 1995; 270: 26794-26801Crossref PubMed Scopus (141) Google Scholar reported that mouse mammary epithelial cells plated on tissue culture plastic dramatically upregulate steady-state levels of mRNA encoding the α1, α2, α3, α5, α6, α7, αv, and β1 integrin subunits, in contrast to cells cultured on a basement membrane matrix or to cells in vivo. Thus the integrin–ECM interaction is reciprocal: integrins mediate cell attachment to matrix, and the matrix, at least to a degree, controls the expression and activity of integrins (Ruoslahti and Engvall, 1997Ruoslahti E. Engvall E. Integrins and vascular extracellular matrix assembly.J Clin Invest. 1997; 99: 1149-1152Crossref PubMed Scopus (73) Google Scholar). The regulatory effect of ECM on integrin expression in microvascular endothelial cells, particularly αvβ3, however, has not been extensively examined. To address whether alterations in the ECM environment regulate vascular αvβ3 expression, we cultured normal human dermal microvascular endothelial cells (HDMEC) on surfaces coated with collagen, fibronectin, and gelatin, and examined mRNA levels of integrin αv/β3. In addition, fibrin or collagen three-dimensional gels were overlaid on HDMEC cultured on collagen to simulate the in vivo ECM environment and to determine whether fibrin and collagen could differentially regulate αv/β3 mRNA levels. Our data indicated that the ECM environment can control mRNA levels of αv/β3 in microvascular endothelial cells. Furthermore, in this study we demonstrate that integrin subunit β3 mRNA has a very low stability which can be increased by interaction of microvascular endothelial cells with fibrin. HDMEC were isolated from human neonatal foreskins as previously described (Kubota et al., 1988Kubota Y. Kleinman H.K. Martin G.R. Lawley T.J. Role of laminin and basement membrane in the morphological differentiation of human endothelial cells into capillary-like structures.J Cell Biol. 1988; 107: 1589-1598Crossref PubMed Scopus (975) Google Scholar), with minor modifications. Briefly, after initial harvest from minced trypsinized human foreskins, microvascular endothelial cells were further purified on a Percoll density gradient. HDMEC were suspended in endothelial growth medium consisting of endothelial basal medium (EBM) (Clonetics, San Diego, CA) supplemented with 10 ng epidermal growth factor per ml and 0.4% bovine brain extract (Clonetics); and 17.5 μg dibutyryl cyclic adenosine monophosphate per ml and 1 μg hydrocortisone per ml (Sigma, St Louis, MO) in the presence of 15% normal human serum (Biowhittaker, Walkersville, MD), and plated on gelatin-coated tissue culture flasks. Endothelial cell cultures were characterized and determined to be > 99% pure on the basis of the formation of typical cobblestone monolayers in culture, positive immunostaining staining for factor VIII-related antigen, and selective uptake of acetylated low-density lipoprotein. Experiments were done on HDMEC below passage 10. For experi- mentation, HDMEC at 80% confluence were harvested by trypsin and ethylenediamine tetraacetic acid, neutralized by soybean trypsin inhibitor, and resuspended in EBM with 1% normal human serum. To determine the effect of ECM proteins on αvβ3 mRNA expression, HDMEC were cultured on ECM protein-coated surfaces either alone, or overlaid with a protein gel, for varying periods of time up to 24 h prior to RNA isolation. Specifically, 6 × 105 cells were plated in 60 × 15 mm tissue culture dishes (Becton Dickinson Labware, Lincoln Park, NJ), wet-coated by either collagen type I (10 μg per ml) (Celtrix, Santa Clara, CA), fibronectin (10 μg per ml) (Life Technologies, Grand Island, NY), or gelatin (0.5%). For the gel overlay experiments (Chalupowicz et al., 1995Chalupowicz D.G. Chowdhury Z.A. Bach T.L. Barsigian C. Martinez J. Fibrin II induces endothelial cell capillary tube formation.J Cell Biol. 1995; 130: 207-215Crossref PubMed Scopus (128) Google Scholar), HDMEC were plated on collagen type I coated dishes. After 24 h of incubation at 37°C, medium was aspirated and 2 ml of ECM gel mixture was added to each cell monolayer. Fibrin gel mixture was prepared by adding human thrombin (0.3 U per ml) (New York Blood Center, New York, NY) to purified human fibrinogen (1 mg per ml), isolated as previously described (Mosesson and Sherry, 1966Mosesson M. Sherry S. The preparation and properties of human fibrinogen of relatively high solubility.Biochemistry. 1966; 5: 2829-2835Crossref PubMed Scopus (174) Google Scholar) and dissolved in EBM. Collagen gel mixture was prepared by dissolving 1 mg per ml collagen type I in EBM. Human thrombin (0.3 U per ml) was also added to the collagen gel mixture to control for possible effects of thrombin on HDMEC. After the gels were allowed to polymerize for 30 min at 37°C, 3 ml of EBM with 1% normal human serum and 1% bovine serum albumin were added. In some experiments, bFGF (50 ng per ml) (Scios Nova, Mountainvale, CA) or VEGF (100 ng per ml) (Pepro Tech, Rocky Hill, NJ) was present in the medium. In experiments designed to measure mRNA stability, 60 μM 5,6-dichloro-1-b-D-ribofuranosyl-benzimidazole (DRB; Sigma), an inhibitor of transcription initiation (Zandomeni et al., 1983Zandomeni R. Buvick D. Ackerman S. Mittleman B. Weinmann R. Mechanism of action of DRB. III. Effect on specific in vitro initiation of transcription.J Mol Biol. 1983; 167: 561-574Crossref PubMed Scopus (85) Google Scholar), was added to the medium. Total RNA was isolated from HDMEC cultures on ECM-coated dishes and from gel overlay cultures using trizol reagent (Life Technologies) as previously described for fibroblasts (Xu and Clark, 1996Xu J. Clark R.A.F. Extracellular matrix alters PDGF regulation of fibroblast integrins.J Cell Biol. 1996; 132: 239-249Crossref PubMed Scopus (184) Google Scholar). One milliliter of trizol solution was used for each culture dish. For each overlay culture, the gel was first transferred to a centrifuge tube, and then RNA was extracted from the remaining cells on the dish. The transferred gel was spun in a microcentrifuge at 14 000 × g to remove H2O, dissolved in 1 ml trizol solution, and repeatedly passed through a 20½ gauge needle. Then RNA was isolated as above. For northern hybridization, 4–6 μg of total RNA was treated with glyoxal/dimethyl sulfoxide, separated by electrophoresis on 1% agarose gel in 10 mM phosphate buffer, pH 7.0, and transferred to Hyband + nylon membranes (Amersham, Arlington Heights, IL). Ethidium bromide (0.5 μg per ml) was included in the gel to monitor equal loading by the quantity of 18S and 28S ribosomal RNA present. cDNA probes of human integrin subunits αv, β3, α2, and β1 were labeled with [α-32P]dCTP by the random primer procedure (DuPont New England Nuclear, Boston, MA). Oligonucleotide probe complementary to 28S ribosomal RNA (Clontech, Palo Alto, CA) was end-labeled with [γ-32P]ATP (DuPont New England Nuclear) and polynucleotide kinase (Boehringer Mannheim). The filters were hybridized to the labeled probes in QuickHyb solution (Stratagene, La Jolla, CA) for 3 h at 68°C and washed according to the manufacturer's protocol. After autoradiography (Kodak X-Omat AR) at - 80°C for optimal exposure, signal intensity was determined by densitometry. Porcine cutaneous wounds and immunofluorescence staining Porcine cutaneous wounds were harvested at various times and then immunoprobed for the expression of integrin receptors as previously described (Clark et al., 1996Clark R.A.F. Tonnesen M.G. Gailit J. Cheresh D.A. Transient functional expression of αvβ3 on vascular cells during wound repair.Am J Pathol. 1996; 148: 1407-1421PubMed Google Scholar). Briefly, full-thickness wounds were made with an 8 mm punch on the backs of White Yorkshire pigs and harvested at the times indicated. Specimens were bisected; one-half was fixed in formalin and stained with Masson trichrome, the other half was frozen in liquid nitrogen for immunofluorescence studies. The following antibodies were used to identify integrins of interest: 7G2, a monoclonal antibody to the β3 subunit (Gresham et al., 1989Gresham H.D. Goodwin J.L. Allen P.M. Anderson D.C. Brown E.J. A novel member of the integrin receptor family mediates arg-gly-asp-stimulated neutrophil phagocytosis.J Cell Biol. 1989; 108: 1935-1943Crossref PubMed Scopus (148) Google Scholar); 23C6, a monoclonal antibody to the αvβ3 complex (Horton et al., 1985Horton M.A. Lewis D. McNulty K. Pringle J.A.S. Chambers T.J. Monoclonal antibodies to osteoclastomas (giant cell bone tumors): Definition of osteoclast-specific cellular antigens.Cancer Res. 1985; 45: 5663-5669PubMed Google Scholar); and 13C2, a monoclonal antibody to the αv subunit (Horton et al., 1985Horton M.A. Lewis D. McNulty K. Pringle J.A.S. Chambers T.J. Monoclonal antibodies to osteoclastomas (giant cell bone tumors): Definition of osteoclast-specific cellular antigens.Cancer Res. 1985; 45: 5663-5669PubMed Google Scholar). All antibodies were used at dilutions that gave maximal specific fluorescence and minimal background fluorescence on frozen tissue specimens. Bound antibody was detected by the avidin–biotin complex (ABC) technique. Stained specimens were observed and photographed using a Nikon microphot FXA epifluorescence microscope. To delineate further the relationship in vivo between microvascular endothelial cell expression of integrin αvβ3 and alteration of the ECM milieu during cutaneous wound repair, the expression of αv and β3 integrin subunits were compared in day 5 and 7 wounds. Previous investigations from our laboratory have shown that the fibrin clot persists in wounds up to 5 d after injury (McClain et al., 1996McClain S.A. Simon M. Jones E. et al.Mesenchymal cell activation is the rate-limiting step of granulation tissue induction.Am J Pathol. 1996; 149: 1257-1270PubMed Google Scholar), whereas 7 d wounds have a substantial organized collagen fiber network and little, if any, fibrin (Welch et al., 1990Welch M.P. Odland G.F. Clark R.A.F. Temporal relationships of F-actin bundle formation, collagen and fibronectin matrix assembly, and fibronectin receptor expression to wound contraction.J Cell Biol. 1990; 110: 133-145Crossref PubMed Scopus (283) Google Scholar;Clark et al., 1995Clark R.A.F. Nielsen L.D. Welch M.P. McPherson J.M. Collagen matrices attenuate the collagen-synthetic response of cultured fibroblasts to TGF-beta.J Cell Sci. 1995; 108: 1251-1261PubMed Google Scholar). Immunoprobing of day 5 and 7 porcine cutaneous wounds revealed that αv was strongly expressed on blood vessels in fibrin and fibronectin-rich day 5 wounds and in collagen-rich day 7 wounds (Figure 1a,b). In contrast, whereas capillary sprouts in day 5 wounds strongly expressed β3 (Figure 1c), blood vessels of mature granulation tissue in day 7, fibrin-deficient and collagen-rich wounds, were no longer positive for β3 (Figure 1d). As in vascular cells β3 is known to combine only with αv, the lack of β3 expression in day 7 wounds is consistent with transient expression of αvβ3 limited to the angiogenic capillary sprouts invading the fibrin clot at days 3–5, as previously shown (Clark et al., 1996Clark R.A.F. Tonnesen M.G. Gailit J. Cheresh D.A. Transient functional expression of αvβ3 on vascular cells during wound repair.Am J Pathol. 1996; 148: 1407-1421PubMed Google Scholar). The persistence of αv in day 7 wound blood vessels corresponds to the appearance of αvβ5 in the day 7 microvasculature (data not shown). The absence of β3, and αvβ3, expression in day 7 wound blood vessels is accompanied by accumulation of collagen in the wound space and clearance of the fibrin clot. These observations support the general concept that ECM can regulate integrin expression. To assess whether ECM proteins regulate integrin mRNA levels in microvascular endothelial cells, we measured mRNA steady-state levels of integrin subunits αv, β3, α2, and β1 in HDMEC grown on surfaces coated by collagen, fibronectin, and gelatin (Figure 2). Compared with cultures on collagen, mRNA levels of αv and β3 were higher in cultures on fibronectin and gelatin. HDMEC plated on fibronectin or gelatin demonstrated increased levels of αv mRNA for 24 h, whereas β3 mRNA levels were elevated for 8 h on fibronectin and 24 h on gelatin. There was little difference in α2 mRNA levels among the three culture conditions. Interestingly, β1 mRNA levels were highest in cultures on collagen. Thus fibronectin and gelatin upregulated mRNA levels of αv and β3 compared with collagen. As in vivoβ3 was highly expressed on the microvasculature in day 5 wounds rich in fibrin, but not in day 7 wounds rich in collagen (Figure 1), we examined the ability of three-dimensional fibrin or collagen matrix to regulate mRNA levels of αv and β3 in microvascular endothelial cells. HDMEC cultured on collagen were overlaid by either fibrin gel or collagen gel to simulate the early and late wound ECM environment, respectively. Both αv and β3 mRNA levels were higher in cultures overlaid by fibrin gels than those in cultures overlaid by collagen gels (Figure 3a). The greatest difference in both αv and β3 mRNA levels between fibrin and collagen occurred at 24 h (Figure 3b). Thus, fibrin sustained αv and β3 steady-state mRNA levels compared with collagen. As the angiogenic factors, bFGF and VEGF, have been shown to increase αv/β3 mRNA levels in HDMEC (Enenstein et al., 1992Enenstein J. Waleh N.S. Kramer R.H. Basic FGF and TGF-β differentially modulate integrin expression of human microvascular endothelial cells.Exp Cell Res. 1992; 203: 499-503Crossref PubMed Scopus (151) Google Scholar;Sepp et al., 1994Sepp N.T. Li L. Lee K.H. Brown E.J. Caughman S.W. Lawley T.J. Swerlick R.A. Basic fibroblast growth factor increases expression of the αvβ3 integrin complex on human microvascular endothelial cells.J Invest Dermatol. 1994; 103: 295-299Abstract Full Text PDF PubMed Scopus (115) Google Scholar;Senger et al., 1996Senger D.R. Ledbetter S.R. Claffey K.P. Papadopoulos-Sergiou A. Perruzzi C.A. Detmar M. Stimulation of endothelial cell migration by vascular permeability factor/vascular endothelial growth factor through cooperative mechanisms involving the αvβ3 integrin, osteopontin, and thrombin.Am J Pathol. 1996; 149: 293-305PubMed Google Scholar), we wondered whether the ECM environment could modify the microvascular endothelial cell integrin response to angiogenic factors. To address this question, mRNA levels of αv/β3 in HDMEC overlaid by fibrin gel and stimulated with either bFGF (50 ng per ml) or VEGF (100 ng per ml) were compared with those of HDMEC overlaid by collagen gel. With or without stimulation by bFGF or VEGF, the mRNA levels of αv and β3 at 24 h were much higher in cells overlaid by fibrin gel than cells overlaid by collagen gel (Figure 4a,b). Furthermore, fibrin augmented both bFGF and VEGF induction of HDMEC αv/β3 mRNA compared with collagen (Figure 4a,b). These data confirm the importance of HDMEC interaction with the ECM, even in the presence of angiogenic factors, for the appropriate expression of integrin ECM receptors. As in cutaneous wounds in vivo, the expression of integrin subunit β3 in angiogenic capillary sprouts was transient (Figure 1), and as in vitro, β3 mRNA levels in HDMEC interacting with collagen were almost undetectable after 24 h (Figure 2,Figure 3,Figure 4) we wondered whether β3 mRNA might be relatively unstable. To investigate β3 mRNA stability, HDMEC cultured on immobilized type I collagen were treated with a RNA transcription initiation inhibitor, DRB. Quantitative northern analysis of αv, β3, α2, and β1 mRNA was performed as a function of time. β3 mRNA stability was strikingly low, with a rate of decay so rapid that the level was almost undetectable at 24 h (Figure 5). In contrast, the mRNA stabilities of the other integrin subunits, αv, α2, and β1, were much higher than that of β3. As αv/β3 mRNA levels were higher in HDMEC interacting with three-dimensional fibrin gels than in HDMEC interacting with three-dimensional collagen gels (Figure 3 and Figure 4), the effects of fibrin and collagen gel on αv/β3 mRNA stability were compared. After 15 min treatment with 60 μM DRB, HDMEC were overlaid by fibrin or collagen gel. Quantitative northern analysis of αv andβ3 mRNA was performed as a function of time. Both αv andβ3 mRNA decayed rapidly in HDMEC overlaid by three-dimensional collagen gel, with β3 mRNA becoming almost undetectable at 24 h (Figure 6), similar to the results observed when cells were plated on two-dimensional collagen fibers (Figure 5). Interestingly, αv mRNA decayed more rapidly in HDMEC overlaid with three-dimensional collagen gel (Figure 6) than in HDMEC plated on two-dimensional collagen fibers (Figure 5). In contrast, fibrin gel slowed the decay rate of both αv and β3 mRNA (Figure 6a,b). Thus, a three-dimensional fibrin matrix increased the mRNA stability of both αv and β3 compared with a three-dimensional collagen matrix. Cutaneous wound repair is a dynamic process which is tightly regulated both temporally and spatially. Early granulation tissue is formed by endothelial cells and fibroblasts which migrate from the periphery of the wound and invade the fibrin clot. As the granulation tissue matures, blood vessels are organized in a vertical array interwoven with a horizontal meshwork of fibroblasts. Alterations in the ECM environment in the wound space accompany the initiation and evolution of granulation tissue formation. After tissue injury, fibrinogen leaks from blood vessels at the wound periphery to form the fibrin clot which fills the wound space. The fibrin clot with its associated proteins, fibronectin and vitronectin, serves as a provisional matrix scaffold for cell migration (Yamada and Clark, 1996Yamada K.M. Clark R.A.F. Provisional matrix.in: Clark R.A.F. The Molecular and Cellular Biology of Wound Repair. Plenum Press, New York1996: 51-93Google Scholar). Collagen accumulation begins as the granulation tissue matures, and in small excisional wounds, collagen replaces fibrin in the wound space by 7 d (Welch et al., 1990Welch M.P. Odland G.F. Clark R.A.F. Temporal relationships of F-actin bundle formation, collagen and fibronectin matrix assembly, and fibronectin receptor expression to wound contraction.J Cell Biol. 1990; 110: 133-145Crossref PubMed Scopus (283) Google Scholar). As the ECM environment in the wound evolves, modulation of expression of specific endothelial cell surface ECM receptors is critical for the sequential events of wound angiogenesis and blood vessel maturation. Recently, we demonstrated that the endothelial cell integrin receptor αvβ3 is focally expressed in vivo on the tips of capillary sprouts as they invade the fibrin clot during early granulation tissue formation in a porcine model of cutaneous wound repair (Clark et al., 1996Clark R.A.F. Tonnesen M.G. Gailit J. Cheresh D.A. Transient functional expression of αvβ3 on vascular cells during wound repair.Am J Pathol. 1996; 148: 1407-1421PubMed Google Scholar). Addition of antagonists to αvβ3, either specific monoclonal antibodies or cyclic peptides, to healing wounds transiently inhibits granulation tissue formation (Clark et al., 1996Clark R.A.F. Tonnesen M.G. Gailit J. Cheresh D.A. Transient functional expression of αvβ3 on vascular cells during wound repair.Am J Pathol. 1996; 148: 1407-1421PubMed Google Scholar). This dependence of angiogenesis on αvβ3 has been shown in other model systems. Blocking αvβ3 with antibodies or cyclic peptides inhibits cytokine-induced or melanoma-induced angiogenic vessels in the chick chorioallantoic membrane assay (Brooks et al., 1994Brooks P.C. Clark R.A.F. Cheresh D.A. Requirement of vascular integrin αvβ3 for angiogenesis.Science. 1994; 264: 569-571Crossref PubMed Scopus (2732) Google Scholar), and angiogenesis and breast cancer growth in human skin grafted
Referência(s)