JOURNAL OF CELLULAR PHYSIOLOGY 153:557-562 (1992)
Vascular Endothelial Growth Factor Induces Interstitial Collagenase Expression in Human Endothelial Cells ELAINE N. U N E M O R I , * NAPOLEONE FERRARA, EUGENE A. BAUER, AND E D W A R D P. A M E N T O Departments of immunology (€ N U., E P A.) and Cdrd/ovascu/ar Research (N F I , Cenentech, S >an Francisco, California 94080, and Department of Dermatology, Stanford University, Palo Alto, California (€ N U , t A R ) Y4305 Vascular endothelial growth factor (VECF) is a 45kDa secreted peptide that has potent mitogenic activity specific for endothelial cells in vitro and the ability to induce a strong angiogenic response in vivo. In the present study, 24 h treatment with VEGF resulted in a stimulation of expression of the metalloproteinase, interstitial collagenase, at the protein and mKNA levels 2.5-3.0-fold in human umbilical vein endothelial cells but not in human dermal fibroblasts. The dose response curve for collagenase induction was biphasic with the peak stimulatory response obtained by treatment of cells with 10-100 ng/ml (0.2-2 nM) VEGF. The dose response curve for collagenase induction overlapped with, but was not identical to, the response curve for proliferation, which showed VEGF mitogenic activity between < 0.1-50 ng/ml (< 0.002-1 nM). There was no induction seen in expression of other members of the matrix metalloproteinase family, including the 72kDa type IV collagenase, the 92kDa type V collagenase, or stromelysin. Expression of transcripts for the major metalloproteinase inhibitor, tissue inhibitor of metalloproteinases, was also unaltered by treatment with VECF (1-200 ngml). These studies demonstrate that in addition to stitnulating proliferation of endothelid cells, VEGF can d1so induce the expression of the only rnetalloproteinase that can initiate degradation of interstitial collagen types 1-111 under normal physiological conditions. Both responses are likely to contribute to the angiogenic potential Q 1992 WiIey-Liss, Inc. of this peptide.
The process of angiogenesis or neovascularization is required for tissue expansion during growth and tissue remodeling as it occurs during development and wound healing. The sprouting of new capillaries from existing vessels requires finite dismantling of endothelial cells from the constraints of a tubelike structure to proliferate, degradation of the extracellular matrix, both of the basement membrane and interstitial types, and reestablishment of cell-cell contacts (Moscatelli and Rifkin, 1988). Proteinases are arguably required for breaking cell-celljunctions but are clearly necessary for penetration and invasion of the extracellular matrix. Serine proteinases, such as plasmin (Mignatti et al., 1989) and metalloproteinases, such as collagenase (Herronetal., 1986; Mignatti et al., 1989), have been implicated in neovascularization by virtue of their substrate specificities and their induction by a number of putative angiogenic factors (Bauer et al., 1985; Chua et al., 1985; Dayer et al., 1985; Edwards et al., 1987). Although a number of polypeptide factors, including transforming growth factor-@,tumor necrosis factor-a, and angiogenin, can cause a neovascular response in vivo (Fett et al., 1986; Roberts et al., 1986; Leibovich et al., 19871, only acidic and basic fibroblast growth factor (bFGF) (Schweigerer et al., 1987), epidermal growth factor (Gospodarowicz et al., 1978), transforming growth facQ 1992 WILEY-LISS, INC
tor-ol (Schreiber et al., 1986),vascular permeability factor (Connolly et al., 1989), and vascular endothelial growth factor (VEGF) (Keck et al., 1989; Leung et al., 1989)are also directly mitogenic for endothelial cells in vitro. VEGF is secreted extracellularly, suggesting that unlike acidic and basic fibroblast growth factors, it is a diffusible angiogenic factor. VEGF is expressed by a number of tumor cells (Rosenthal et al., 19901, implicating its involvement in the neovascularization required during tumor growth, and is also distributed throughout many organs (Phillips et al., 1990; Jakeman et al., 1992), suggesting its importance in the development and/or maintenance of vascular beds. Recently, at least one of its receptors has been identified as the fms-like tyrosine kinase, Flt (DeVries et al., 1992). Because VEGF is known to directly stimulate the proliferative phase of anigogenesis, we tested its ability to influence another aspect of angiogenesis, extracellular matrix degradation. We have found that VEGF specifically induces expression of interstitial collagenase
Received April 9,1992; accepted June 25,1992. *To whom reprint requestskorrespondence should be addressed.
558
UNEMORI ET AL
(Matrix Metalloproteinase-1) but none of the other major members of the matrix metalloproteinase family or of their inhibitor, tissue inhibitor of metalloproteinases (TIMP).
MATERIALS AND METHODS Cell culture Human umbilical endothelial (HUVE) cells were purchased from Clonetics (San Diego, CA) and used between passages 1-4. Cells were seeded a t the recommended density (4 x lo3 cellsicm') in EGM-UV Medium (Clonetics) consisting of MCDB 131 supplemented with hydrocortisone (1 ng1ml) and epidermal growth factor (10 ngiml). Fetal bovine serum (lo%), 2 mM L-glutamine, and a 1 2 5 0 dilution of Bovine Brain Extract (containing heparin), as provided by Clonetics, were also added to this medium. Cells were passaged every 4-5 days with gentle trypsinization and without centrifugation. For experiments, cells were cultured a t 4 X 10" cells/cm2 in the same medium. The following day, cells were washed and treated in EBM-UV (MCDB 131), supplemented with 2 mM L-glutamine. This medium did not contain serum, bovine brain extract, hydrocortisone, or epidermal growth factor. Cytokines Recombinant human VEGF was cloned, expressed, and purified at Genentech (Leung et al., 1989) as previously described. bFGF was purchased from R & D Systems (Minneapolis, MN).
5%type 111, Collagen Corp.) using '*C-acetic anhydride (Amersham), essentially as previously described (Cawston and Barrett, 1979).
Immunoblot analysis Procollagenase was detected by Western immunoblotting following 0.2% SDS-O.02Mquinine sulfate precipitation of proteins and electrophoresis on 10% acrylamide gels. Proteins were transferred electrophoretically to nitrocellulose by the method of Towbin et al. (1979). Filters were stained with a monoclonal antibody to collagenase, H18G8 (generous gift of Z. Werb, University of California, San Francisco). Following extensive washing, a biotinylated sheep antimouse second antibody and streptavidin-linked alkaline phosphatase were sequentially applied a t dilutions recommended by the manufacturer (Boehringer Mannheim). Blots were developed using nitroblue tetrazolium and bromochloroindoyl phosphate. Proliferation assay Cell number was measured using a n acid phosphatase assay, essentially a s described for endothelial cells (Connolly et al., 1986). Briefly, cells were washed in PBS, then lysed in a buffer composed of 0.1 M sodium acetate, pH 5.5,0.1% Triton X-100, and 10 mM p-nitrophenyl phosphate (Sigma 104 substrate). Cell lysates were incubated for 1-2 h , followed by termination of the reaction with 1 N NaOH. Color development was measured a t 405 nM on a 340 ATC plate reader (SLT Laboratories). A standard curve was generated by plating serial dilutions of a known number of HUVE cells, and incubation times, described above, had been optimized to fall within the linear part of the curve (data not shown).
Northern blot analysis Total cytoplasmic RNA was isolated from endothelial cells using RNazol (Cinna Biotecx, Friendswood, TX), according to manufacturer's instructions. RNA was separated on 1%agaroseiformaldehyde gels and transRESULTS ferred to nylon (Genescreen) in 25 mM sodium phosVEGF induces collagenase expression phate, pH6.5 (Greenberg, 1987). Northern blot analysis Human umbilical vein endothelial (HUVE) cells was performed using a 380 base pair ncoI cDNA fragment of the 72kDa type IV collagenase (Collier et al., were either untreated or treated with VEGF at a dose 1988), a 2.3 base pair Xba I fragment of the 92kDa type ranging from 1.0 to 200 ngiml in serum-free EBM for 24 V collagenase (Wilhelm et al., 19891, or a 760 base pair h. Total cytoplasmic RNA was harvested and screened EcoRl fragment of TIMP (kindly provided by S. Clark, by Northern blot analysis for the induction of collageGenetics Institute) (Gasson et al., 1985). Two 60-base nase mRNA. The results demonstrated that collagepair partially overlapping oligonucleotide probes were nase mRNA was induced in a dose-responsive manner designed according to each of the published sequences in the presence of 10, 50, and 100 ngiml VEGF (Fig. of collagenase (Goldberg et. al., 1986) and stromelysin 1A). The magnitude of stimulation of collagenase ex(Saus et al., 1988). All probes were 32P-labeled by ran- pression was similar to that seen following treatment dom priming (Boehringer Mannheim) ( c x ~ ~ P - ~ C Twith P , bFGF (10 ngiml), which was used a s a positive -dATP, - 3,000 Ciimmol, Amersham). Hybridizations control (Mignatti et al., 1989). Normal human dermal were carried out at 42°C. Filters were washed in fibroblasts did not respond to 50-100 ng/ml VEGF 0.2 x SSC + 1%SDS at 57°C for 1 h. To strip blots, (data not shown). As previously described, VEGF is mitogenic for endothelial cells (Keck et al., 1989; Leung nylon membranes were boiled in H,O for 3 min. et al., 1989). HUVE cells proliferated in response to Collagenase assay 0.1-50 ngiml VEGF a t 24 h with no effect seen at either Collagenase activity in conditioned media was mea- 100 or 200 ngiml VEGF (Fig. 1B). To assay for secretion of collagenase protein by these sured in the 14C-collagen fibril assay following activation of procollagenase in the medium with TPCK- cells, conditioned media from control cultures and those trypsin (10 pgiml) (Worthington) for 30 min a t 25"C, treated with VEGF were assayed by Western blot analfollowed by addition of soybean trypsin inhibitor (60 ysis using a monoclonal antibody to collagenase. Alpg/ml) (Sigma) (Werb and Burleigh, 1974). One unit of though untreated cultures secreted some procollagecollagenolytic activity hydrolyzed 1 pg collagen per nase, consistent with Northern blot results, the min at 37°C. The collagen substrate was obtained by procollagenase doublet at 52157kDa was clearly inacetylation of purified calf skin collagen (95% type I, duced following 24 h treatment with 50 ngiml VEGF, a s
559
VEGF INDUCES COLLAGENASE
A
A
kDa 97-
3 proCL
- CL
43 -
29-
28s
-
18s
C VEGF bFGF
B
a, 0
"0 \
x
LL
c
0
VEGF (ng/ml)
1.0-
'5 .- 0.8 c
0
2 0.6 .-
-> . I -
400, I
T
B
0.4 a, 0,
(d 5 0.2
300
0
=
0.0 0
200
10
50
100
200
VEGF (ng/mI) 100
-
0 0
0.1
1
10
50
100 200
VEGF (ng/mI) Fig. 1. VEGF induced collagenase mRNA expression (A) and proliferation (B) in HUVE cells. A. HUVE cells were untreated, treated with 1-200 ngiml VEGF, or bFGP (10 ng/ml) for 24 h. Total cytoplasmic RNA was extracted and electrophoresed as described in Materials and Methods. The blot was hybridized with two oligonucleotides specific for human collagenase (CL). Staining of the 28Si18S ribosomal bands by ethidiurn bromide confirmed equal loading of total RNA. €3. Cell number was assayed using a n assay for acid phoaphatase. Changes in cell number following VEGF treatment (0.1-200 ng/ml) was expressed a s fold increases over that seen in untreated cultures. Columns represent means + SE, with n = 4.
Fig. 2. VEGF stimulated collagenase protein expression (A) and collagenolytic activity (B) in HUVE cells. A. The procollagenase (proCL) doublet at 52157kDa was immunoprecipitated using a monoclonal antibody to collagenase from proteins secreted by untreated HUVE cells or those treated for 24 h with VEGF (50 ngiml) or bFGF (10 ngirnl). B. Collagenase activity was measured in conditioned media from untreated cultures or those treated with 10-200 ngiml VEGF in the 14C-collagen fibril assay. Columns represent means + SE of three samples in a representative dose-response experiment.
well as with bFGF (10 ng/ml) (Fig. 2A). To see whether collagenolytic activity could be detected in the media, conditioned media were tested in the 14C-collagenfibril assay (Fig. 2B). VEGF stimulated expression of trypsin-activatable collagenase in the same dose-responsive manner demonstrable by Northern blot analysis. The peak stirnulatory response occurred following
560
UNEMORI
Er AL
CL 72 kDa type IV CL
92 kDa type V CL
-
28s
-18s
Fig. 3. Northern analysis of metalloproteinase and inhibitor expression by HUVE cells following VEGF treatment. Total cytoplasmic RNA was extracted from endothelial cells following no treatment or treatment with 10 and 100 ng/ml VEGF for 24 h. Following transfer of RNA to nylon, the blot was prohed sequentially for collagenase t CIJ,
the 72kDa type IV collagenase, the 92kDa type V collagenase, and TIMP using probes described in Materials and Methods. The ribosomal bands stained with ethidium bromide confirmed equal loading of total RNA.
treatment of cells with 50 ngiml VEGF, with a 2.5-fold increase in collagenolytic activity detectable.
but its level remained unchanged in the presence of VEGF.
VEGF has no effect on expression of other metalloproteinases or TIMP Northern blot analysis of RNA extracted from untreated HUVE cells or those treated with 10 and 100 ngiml VEGF were used to see whether other metalloproteinases thought to contribute to angiogenesis were induced (Fig. 3). mRNA for the 72kDa type IV collagenase, which was expressed constitutively, was not induced reproducibly by VEGF treatment. consistent with previous reports, stromelysin was not expressed by resting umbilical vein endothelial cells and was also not induced following 24 h treatment with VEGF. The 92kDa type V collagenase was barely detectable a t the mRNA level in resting HUVE cells and was not inducible by VEGF. Conditioned media from these cells were also screened by gelatin zymography, which is a more sensitive method of detecting the 92kDa type V collagenase, but failed to show presence of or induction of this metalloproteinase following VEGF treatment (data not shown). mRNA for the major inhibitor of collagenase, TIMP, was readily detectable in untreated HUVE cells,
DISCUSSION Neovascularization in the adult occurs in a finite number of situations, including wound healing, and in certain pathologies such as tumor growth and diabetic retinopathies. The steps required for new vessel growth are biologically complex and probably require precise coordinated regulation of contributing components, including proliferation and migration of endothelial cells and matrix degradation. It is unclear whether a single agent that will stimulate all necessary components exists, although bFGF appears t o be the closest such agent (Presta et al., 1986; Schweigerer et al., 1987). However, this cytokine is not secreted in the classical sense and probably does not diffuse, so its role may be limited to stimulation of endothelial cell outgrowth in autocrine fashion (Sato and Rifkin, 1988; Tsuboi et al., 1990) or as a result of endothelial cell disruption during wounding (McNeil et al., 1989). VEGF is a secreted peptide that can stimulate angiogenesis in vivo, as well a s endothelial cell proliferation in vitro (Keek et al., 1989; Leung et al., 1989). This mitogenic activity, un-
561
VEGF INDUCES COLLAGENASE
like that of FGF and others, appears to be specific for endothelial cells, as i t has no such effect on epithelial cells, fibroblasts, lymphoid cells, or a variety of tumor cells (Rosenthal et al., 1990). Because of this specificity, it is possible that VEGF may play a unique role in regulating endothelial cell function. In the present study, we have evaluated the effect of this cytokine on the induction of a number of metalloproteinases that may play a role in the invasive processes required for neovasculation and have found that i t specifically induces expression of interstitial collagenase. This induction occurs at the mRNA and protein levels in the 1-2 nanomolar range. The increase in protein expression is directly correlated with a n increase in collagenolytic activity since TIMP levels remain unchanged in the face of increases in collagenase. Consistent with its mitogenic specificity, VEGF has no effect on collagenase expression in fibroblasts. The substrate specificity of collagenase includes collagen types 1-111(Gross and Nagai, 1965) and serine proteinase inhibitors (serpins) (Desrochers et al., 1991). The degradation of types I and I11 collagen by collagenase would be expected to facilitate their movement through the interstitium during the migration phase of angiogenesis. Recently, i t has been shown that bFGF-stimulated “angiogenesis” in vitro also requires the presence of collagenase, as invasion was inhibited by the addition of 1 , l O phenanthroline or TIMP to the system (Mignatti et al., 1989). It is interesting to note that VEGF stimulates collagenase expression at a dose range that overlaps with, but is not identical to, that required for mitogenicity, indicating that perhaps at least two intracellular signalling pathways are relevant to VEGF activity. The concentration of VEGF required to stimulate HUVE cell proliferation spans the range of B 0.1-50 ng/ml ( 60.002-1.0 nM), whereas 10-100 ng/ml (0.2-2 nM) are required for collagenase stimulation. Interestingly, a similar dissociation between mitogenic and collagenase stimulatory activity has been observed in fibroblast responses to platelet-derived growth factor (Bauer et al., 1985), with which VEGF shares some homology. It is unclear what the basis for the biphasic nature of both responses to VEGF is. VEGF receptor down-regulation and transmodulation of some other receptor by VEGF may both be involved. Such responses to transforming growth factor-p, e.g., have been reported (Keski-Oja et al., 1988). The finding that VEGF triggers proliferation a t a lower concentration than collagenase stimulation is consistent with what is believed to be the temporal order in which these activities occur during the sprouting of new vessels from existing ones. It is conceivable that this escalating concentration of VEGF could be encountered by endothelial cells in vivo following release of VEGF from a distant source, such as a macrophage. That VEGF did not induce the expression of other metalloproteinases capable of degrading components of the basement membrane, such a s type IV collagen and laminin, was surprising and may indicate that other cytokines are required for completion of all of the steps necessary for angiogenesis. Although the expression of serine proteinases, such a s plasminogen activator, fol-
lowing VEGF treatment was not investigated here, i t has been shown that both tissue plasminogen activator and urokinase-type plasminogen activator are induced by VEGF (Pepper, Ferrara, Orci, and Montesano, submitted). Previous work has demonstrated that plasminogen activator has limited ability to degrade fibronectin (Quigley et al., 1987) and t h a t plasmin is capable of degrading laminin and fibronectin (Liotta et al., 1981). In addition, another important biological effect of the plasminogen activator-plasmin system in the context of matrix degradation lies in the ability of plasmin to activate collagenase (Werb e t al., 1977). The coinduction of plasminogen activator and procollagenase by VEGF, therefore, is suggestive of cooperativity of serine and metalloproteinases in effecting the matrix degradation necessary for vessel sprouting. That collagenase can also inactivate the serpin, plasminogen activator inhibitor-2 (Desrochers et al., 1991),would also favor a prodegradation environment. Since metalloproteinases capable of degrading other matrix proteins, such as the basement membrane component type IV collagen, are not induced by VEGF, it is likely that other cytokines must act in concert with VEGF to complete the degradative steps necessary for angiogenesis t o occur. It is also important to note that umbilical vein endothelial cells, which were used in this study, may not mimic completely the phenotype of microvascular endothelial cells. In summary, VEGF stimulates the production of interstitial collagenase in human endothelial cells. Its expression confers upon these cells a n enhanced ability to degrade interstitial collagens, which constitute a major barrier to endothelial cells during angiogenesis. This activity, as well as its previously characterized mitogenic potential, is consistent with VEGFs role as a n angiogenic peptide.
ACKNOWLEDGMENTS This work was supported in part by NIH grant AR19537. LITERATURE CITED Bauer, E.A., Cooper, T.W., Huang, J.S., Altman, J., and Deuel, T.F. (1985)Stimulation of in vitro human skin collagenase expression by platelet-derived growth factor. Proc. Natl. Acad. Sci. USA, 82.41324136. Cawston, T.E., and Barrett, A.J. (1979)A rapid and reproducible assay for collagenase using [l-’4C]acetylated collagen. Anal. Biochem., 99.340-345. Chua, C.C., Geiman, D.E., Keller, G.H., and Ladda, R.L. (1985) Induction of collagenase secretion in human fibroblast cultures by growth promoting factors. J. Biol. Chem., 260:5213-5216. Collier, I.E., Wilhelm, S.M., Eisen, A.Z., Marmer, B.L., Grant, G.A., Seltzer, J.L., Kronberger, A,, He, C., Bauer, E.A., and Goldberg, G.I. (1988) H-ras oncogene-transformed human bronchial epithelial cells (TBE-1) secrete a single metalloprotease capable of degrading basement membrane collagen. J. Biol. Chem., 263:6579-6587. Connolly, D.T., Knight, M.B., Harakas, N.K., Wittwer, A.J., and Feder, J. (1986) Determination of the number of endothelial cells in culture using a n acid phosphatase assay. Anal. Biochem., 152:136 140. Connolly, D.T., Heuvelman, D.M., Nelson, R., Olander, J.V., Eppley, B.L., Delfino, J.J., Siegel, N.R., Leimgruber, R.M., and Feder, J. (1989) Tumor vascular permeability factor stimulates endothelial cell growth and angiogenesis. J . Clin. Invest., 84.1470-1478,
562
UNEMORI ET AL
Dayer, J.-M., Beutler, B., and Cerami, A. (1985) Cachectinitumor necrosis factor stimulates collagenase and prostaglandin E2 production by human synovial cells and dermal fibroblasts. J. Exp. Med., 162t2163. De Vries, C., Escobedo, J.A., Ueno, H.. Houck, K., Ferrara, N., and Williams, L.T. (1992) The fms-like tyrosine kinase, a receptor for vascular endothelial growth factor. Science,255t989-991. Desrochers, P.E., Jeffrey, J.J., and Weiss, S.J. (1991)Interstitial collagenase (matrix metalloproteinase-1)expresses serpinase activity. J. Clin. Invest., 87t2258-2265. Edwards, D.R., Murphy, G., Reynolds, J.J., Whitham, S.E., Docherty, A.J.P., Angel, P., and Heath, J.K. (1987) Transforming growth factor beta modulates the expression of collagenase and metalloproteinase inhibitor. EMBO J . , 6t1899-1904. Fett, J.W., Strydom, D.J., Lobb, K.R., Alderman, E.M., Bethune, J.L., Riordan, J. J.F., and Vallee, B.L. (1985) Isolation and characterization of angiogenin, an angiogenic protein from human carcinoma cells. Biochemistry, 24t5480-5486. Goldberg, G.I., Wilhelm, S.N., Kronberger, A,, Bauer, E.A., Grant, G.A., and Eisen, A.Z. (1986) Human fibroblast collagenase: Complete primary structure and homology to a n oncogene transformation-induced rat protein. J. Biol. Chem., 261t6600-6605. Gospodarowicz, D., Brown, K.D., Birdwell, C.R., and Zetter. B.R. (1978)Control of proliferation of human vascular endothelial cells I. Characterization of the response of human umbilical vein endothelial cells to fibroblast growth factor, and thrombin. J . Cell Biol., 77t774-788. Greenberg, M.E. (19871 Analysis of RNA by Northern hybridization. In: Current Protocols in Molecular Biology. F.M. Ausubel, R. Brent, R.E. Kingston, D.D. Moore, J.A. Smith, J.G. Seidman, eds. John Wiley & Sons, New York, sect. 4.9.1. Gross, J., and Nagai, Y. (1965) Specific degradation of the collagen molecule by tadpole collagenolytic enzyme. Proc. Natl. Acad. Sci. USA, 54t1197-1204. Herron, G.S., Werb, Z., Dwyer, K., and Bandd, M.J. (1986)Secretion of‘ metalloproteinases by stimulated capillary endothelial cells. I. Production of procollagenase and prostromelysin exceeds expression of proteolytic activity. J. Biol. Chem., 261r2810-2813. Jakeman, L.B., Winer, J., Bennett, G.L., Altar, C.A., and Ferrara, N. (1992) Binding sites for vascular anduthclial growth factor are Iocalized on endothelial cells in adult rat tissues. J . Clin. Invest., 893244253. Gasson, J.C., Golde, D.W., Kaufman, S.E., Westbrook, C.A., Hewick, R.M., Kaufman, R.J., Wong, G.G., Temple, P.A., Leary, A.C., Brown, E.L., Orr, E.C., and Clark, S.C. (1985) Molecular characterization and expression of the gene encoding human erythroid-potentiating activity. Nature, 316r768-771. Keck, P.J., Hauser, S.C., Krivi, G., Sanzo, K., Warren, T., Feder, J., and Connolly, D.T. (1989)Vascular permeability factor, a n endothelial cell mitogen related to PDGF. Science, 246r1309-1312. Keski-Oja, J., Raghow, R., Sawdey, M., Loskutoff, D.J., Postlethwaite, A.E., Kang, A.H., and Moses?H.L. (1988) Hegulation of mRNAs for type-1 plasminogen activator inhibitor, fibronectin, and type I procollagen by transforming growth factor+. J . Biol. Chem. 263:3111-3115. Leibovich, S.J., Polverini, P.J., Shepard, H.M., Wiseman, D.M., Shively, V., and Nuseir, N. (1987)Macrophaage-induced angiogenesis is mediated by tumour necrosis factor-a. Nature, 329t63CL632. Leung, D.W., Cachianes, G., Kuang, W.-J., Goeddel, D.V., and Ferrara, N. (1989) Vascular endothelial growth factor is a secreted angiogenic mitogen. Science, 246t1306-1309. Liotta, L.A., Goldfarb. R.H., Brundage, R., Siegal, G.P., Terranova, V., and Garbisa, S. (1981) Effect of plasminogen activator (urokinase), plasmin, and thrombin on glycoprotein and collagenous components of the basement membrane. Cancer Res., 41t4629-4636.
McNeil, P.L., Muthukrishnan, L.. Warder. E.. and D’Amore. P.A. (1989) Growth factors are released by mechanically wounded cells. J . Cell Biol., 109r811-822. Mignatti, P., Tsuboi, R., Robbins, E., and Rifkin. D.B. (1989) In vitro angiogenesis on the human amniotic membrane: Requirement for basic fibroblast growth factor-induced proteinases. J . Cell Biol., 108t671-682. Moscatelli, D., and Rifkin, D.B. (1988) Membrane and matrix localization of proteinases: A common theme in tumor cell invasion and angiogenesis. Biochim. Biophys. Acta, 948r67-85. Phillips, H.S., Hains, J., Leung, D.W., Ferrara, N. (1990) Vascular endothelial growth factor is expressed in rat corpus luteum. Endocrinology, 127t965-967. Presta, M., Moscatelli, D., Joseph-Silverstein, J., and Rifkin, D.B. (1986) Purification from a human hepatoma cell line of a basic fibroblast growth factor-like molecule that stimulates capillary endothelial cell plasminogen activator production, DNA synthesis and migration. Mol. Cell Biol., 6t4060-4066. Quigley, J.P., Gold, L.I., Schwimmer, R., and Sullivan, L.M. (1987) Limited cleavage of cellular fibronectin by plasminogen activator purified from transformed cells. Proc. Natl. Acad. Sci. USA, 84.2776-2780. Roberts, A.B., Sporn, M.B., Assoian, R.K., Smith, J.M., Roche, N.S., Wakefield, L.M., Heine, U.I., Liotta, L.A., Falanga, V., Kehrl, J.H., and Fauci, A S . (1986) Transforming growth factor type B: Rapid fibrosis and angiogenesis in vivo and stimulation of collagen formation in vitro. Proc. Natl. Acad. Sci. USA, 83t4167-4171. Rosenthal, H.A., Megyesi, J.F., Henzel, W.J., Ferrdra, N., and Folkman, J. (1990) Conditioned medium from mouse sarcoma 180 cells contains vascular endothelial growth factor. Growth Factors, 4t5359. Sato, Y., and Rifkin, D.R. (1988)Autocrine activities of basic fibroblast growth factor: Regulation of endothelial cell movement, plasminogen activator synthesis and DNA synthesis. J. Cell Biol., 107t1199-1205. Saus, J., Quinones, S., Otani, Y., Nagase, H., Harris, E.D. Jr., and Kurkinen, M. (1988) The complete primary structure of human matrix metalloproteinase-3. Identity with stromelysin. J. Biol. Chem., 263:67424745. Schreiber, A.B., Winkler, M.E., and Derynck, R. (1986) Transforming growth factor-u: A more potent angiogenic mediator than epidermal growth factor. Science, 232t1250-1253. Schweigerer, L., Neufeld, G., Freidman, J., Abraham, J.A., Fiddes, J.C., and Gospodarowicz, D. (1987) Capillary endothelial cells exDress basic fibroblast erowth factor. a mitonen that aromotes their bwn growth. Nature, 3kir257-259. Towbin, H., Staehlin, T., and Gordon, J. (1979)Electrophoretic transfer of proteins from polyacrylamide gels to nitrocellulose sheets: Procedure and some applications. Proc. Natl. Acad. Sci. USA, 76:4350-4354. Tsuboi, R., Sato, Y., and Rifkin, D.B. (1990) Correlation of cell migration, cell invasion, receptor number, proteinase production, and basic fibroblast growth factor levels in endothelial cells. J . Cell Biol., IIOt511-517. Werb, A., and Burleigh, M.C. (1974)A specific collagenase from rabbit fibroblasts in monolayer culture. Biochem. J., 137r373-385. Werb, Z., Mainardi, C., Vater, C.A., and Harris, E.D. (1977) Endogenous activation of latent collagenase by rheumatoid synovial cells: Evidence for a role of plasminogen activator. New Engl. J. Med., 2W10 17-1023. Wilhelm, S.M., Collier, I.E., Marmer, B.L., Eisen, A.Z., Grant, G.A., and Goldberg, G.I. (1989) SV-40-transformed human lung fibroblasts secrete a 92kDa type rV collagenase which is identical to that, secreted bv normal human macroahaees. J. Biol. Chem.. 264:17213-i7221. I~