VEGF121, a Vascular Endothelial Growth Factor (VEGF) Isoform Lacking Heparin Binding Ability, Requires Cell-surface Heparan Sulfates for Efficient Binding to the VEGF Receptors of Human Melanoma Cells
1995; Elsevier BV; Volume: 270; Issue: 19 Linguagem: Inglês
10.1074/jbc.270.19.11322
ISSN1083-351X
AutoresTzafra Cohen, Hela Gitay-Goren, Rivka Sharon, Masabumi Shibuya, Ruth Halaban, Ben-Zion Levi, Gera Neufeld,
Tópico(s)Glycosylation and Glycoproteins Research
ResumoFour vascular endothelial growth factor (VEGF) splice variants containing 121, 165, 189, and 206 amino acids are produced from a single human gene as a result of alternative splicing. VEGF121 is not a heparin-binding protein, while the other VEGF species possess heparin binding ability. YU-ZAZ6 human melanoma cells expressed the mRNA encoding the VEGF receptor flt-1, but not the mRNA encoding the VEGF receptor KDR/flk-1. Both VEGF121 and VEGF165 bound to the VEGF receptors of these cells. Unexpectedly, heparin inhibited the binding of VEGF121 as well as the binding of VEGF165 to the VEGF receptors of the melanoma cells. Digestion of the cells with heparinase also inhibited the binding of both VEGF variants. The VEGF165 binding ability of heparinase-digested cells could be partially restored by the addition of exogenous heparin to the binding reaction. In contrast, the addition of heparin to heparinase-digested cells did not restore VEGF121 binding. These results suggest that cell-surface heparan sulfates may regulate the binding ability of the VEGF receptors of the melanoma cells. They also indicate that heparin is not able to fully substitute for cell surface-associated heparan sulfates since VEGF121 binding to the VEGF receptors of heparinase-treated cells is not restored by heparin. These data suggest that changes in the composition of cell-surface heparin-like molecules may differentially affect the interaction of various VEGF isoforms with VEGF receptors. Four vascular endothelial growth factor (VEGF) splice variants containing 121, 165, 189, and 206 amino acids are produced from a single human gene as a result of alternative splicing. VEGF121 is not a heparin-binding protein, while the other VEGF species possess heparin binding ability. YU-ZAZ6 human melanoma cells expressed the mRNA encoding the VEGF receptor flt-1, but not the mRNA encoding the VEGF receptor KDR/flk-1. Both VEGF121 and VEGF165 bound to the VEGF receptors of these cells. Unexpectedly, heparin inhibited the binding of VEGF121 as well as the binding of VEGF165 to the VEGF receptors of the melanoma cells. Digestion of the cells with heparinase also inhibited the binding of both VEGF variants. The VEGF165 binding ability of heparinase-digested cells could be partially restored by the addition of exogenous heparin to the binding reaction. In contrast, the addition of heparin to heparinase-digested cells did not restore VEGF121 binding. These results suggest that cell-surface heparan sulfates may regulate the binding ability of the VEGF receptors of the melanoma cells. They also indicate that heparin is not able to fully substitute for cell surface-associated heparan sulfates since VEGF121 binding to the VEGF receptors of heparinase-treated cells is not restored by heparin. These data suggest that changes in the composition of cell-surface heparin-like molecules may differentially affect the interaction of various VEGF isoforms with VEGF receptors. Vascular endothelial growth factors (VEGFs) 1The abbreviations used are: VEGFsvascular endothelial growth factorsVEGF165165-amino acid VEGF isoformVEGF121121-amino acid VEGF isoformFGFfibroblast growth factorHUEhuman umbilical vein-derived endothelialPBSDulbecco's phosphate-buffered saline. are specific mitogens for vascular endothelial cells in vitro and angiogenic factors in vivo (1Ferrara N. Houck K. Jakeman L. Leung D.W. Endocr. Rev. 1992; 13: 18-32Crossref PubMed Scopus (1559) Google Scholar, 2Leung D.W. Cachianes G. Kuang W.J. Goeddel D.V. Ferrara N. Science. 1989; 246: 1306-1309Crossref PubMed Scopus (4466) Google Scholar, 3Tischer E. Gospodarowicz D. Mitchell R. Silva M. Schilling J. Lau K. Crisp T. Fiddes J.C. Abraham J.A. Biochem. Biophys. Res. Commun. 1989; 165: 1198-1206Crossref PubMed Scopus (257) Google Scholar). They induce the permeabilization of blood vessels (VPFs) and are therefore also known as vascular permeability factors (VPFs) (4Keck P.J. Hauser S.D. Krivi G. Sanzo K. Warren T. Feder J. Connolly D.T. Science. 1989; 246: 1309-1312Crossref PubMed Scopus (1811) Google Scholar). Four VEGF species of 121, 165, 189, and 206 amino acids are produced from a single human gene containing eight exons as a result of alternative splicing (1Ferrara N. Houck K. Jakeman L. Leung D.W. Endocr. Rev. 1992; 13: 18-32Crossref PubMed Scopus (1559) Google Scholar, 5Houck K.A. Ferrara N. Winer J. Cachianes G. Li B. Leung D.W. Mol. Endocrinol. 1991; 5: 1806-1814Crossref PubMed Scopus (1241) Google Scholar). The active forms of these VEGF species are disulfide-linked homodimers (6Vaisman N. Gospodarowicz D. Neufeld G. J. Biol. Chem. 1990; 265: 19461-19466Abstract Full Text PDF PubMed Google Scholar, 7Gospodarowicz D. Abraham J.A. Schilling J. Proc. Natl. Acad. Sci. U. S. A. 1989; 86: 7311-7315Crossref PubMed Scopus (555) Google Scholar). VEGF121 differs from the larger VEGF isoforms in that it is the only VEGF type that does not bind to heparin (5Houck K.A. Ferrara N. Winer J. Cachianes G. Li B. Leung D.W. Mol. Endocrinol. 1991; 5: 1806-1814Crossref PubMed Scopus (1241) Google Scholar, 8Houck K.A. Leung D.W. Rowland A.M. Winer J. Ferrara N. J. Biol. Chem. 1992; 267: 26031-26037Abstract Full Text PDF PubMed Google Scholar). To date, the best characterized VEGF species is VEGF165. This VEGF isoform binds to three types of high affinity receptors on the cell surface of vascular endothelial cells (6Vaisman N. Gospodarowicz D. Neufeld G. J. Biol. Chem. 1990; 265: 19461-19466Abstract Full Text PDF PubMed Google Scholar, 9Plouet J. Moukadiri H. J. Biol. Chem. 1990; 265: 22071-22074Abstract Full Text PDF PubMed Google Scholar). Several types of non-endothelial cells such as human melanoma cells, monocytes, and HeLa cells also express VEGF receptors (10Gitay-Goren H. Soker S. Vlodavsky I. Neufeld G. J. Biol. Chem. 1992; 267: 6093-6098Abstract Full Text PDF PubMed Google Scholar, 11Gitay-Goren H. Halaban R. Neufeld G. Biochem. Biophys. Res. Commun. 1993; 190: 702-709Crossref PubMed Scopus (137) Google Scholar), but the binding of VEGF165 to these receptors does not seem to induce a mitogenic response. Activation of the VEGF receptors in several types of non-endothelial cells results in increased motility in the case of monocytes (12Clauss M. Gerlach M. Gerlach H. Brett J. Wang F. Familletti P.C. Pan Y.C. Olander J.V. Connolly D.T. Stern D. J. Exp. Med. 1990; 172: 1535-1545Crossref PubMed Scopus (768) Google Scholar, 13Shen H. Clauss M. Ryan J. Schmidt A.M. Tijburg P. Borden L. Connolly D. Stern D. Kao J. Blood. 1993; 81: 2767-2773Crossref PubMed Google Scholar), in increased production of insulin in beta cells (14Oberg C. Waltenberger J. Claessonwelsh L. Welsh M. Growth Factors. 1994; 10: 115-126Crossref PubMed Scopus (88) Google Scholar), and in the differentiation of osteoblasts (15Midy V. Plouet J. Biochem. Biophys. Res. Commun. 1994; 199: 380-386Crossref PubMed Scopus (270) Google Scholar). In vascular endothelial cells as well as in melanoma cells, the presence of heparin-like molecules on the cell surface is required for the efficient binding of VEGF165 to the VEGF receptors (10Gitay-Goren H. Soker S. Vlodavsky I. Neufeld G. J. Biol. Chem. 1992; 267: 6093-6098Abstract Full Text PDF PubMed Google Scholar, 11Gitay-Goren H. Halaban R. Neufeld G. Biochem. Biophys. Res. Commun. 1993; 190: 702-709Crossref PubMed Scopus (137) Google Scholar). Two tyrosine kinase receptors for VEGF165 were identified. These receptors are the products of the KDR/flk-1 and flt-1 genes (16Devries C. Escobedo J.A. Ueno H. Houck K. Ferrara N. Williams L.T. Science. 1992; 255: 989-991Crossref PubMed Scopus (1896) Google Scholar, 17Millauer B. Wizigmann-Voos S. Schnurch H. Martinez R. Moller N.P.H. Risau W. Ullrich A. Cell. 1993; 72: 835-846Abstract Full Text PDF PubMed Scopus (1764) Google Scholar). The KDR/flk-1 receptor seems to mediate VEGF-induced mitogenic effects, while the flt-1 receptor by itself may not be able to induce cell proliferation (18Seetharam L. Gotoh N. Maru Y. Neufeld G. Yamaguchi S. Shibuya M. Oncogene. 1995; 10: 135-147PubMed Google Scholar, 19Waltenberger J. Claesson-welsh L. Siegbahn A. Shibuya M. Heldin C.H. J. Biol. Chem. 1994; 269: 26988-26995Abstract Full Text PDF PubMed Google Scholar, 20Park J.E. Chen H.H. Winer J. Houck K.A. Ferrara N. J. Biol. Chem. 1994; 269: 25646-25654Abstract Full Text PDF PubMed Google Scholar, 21Millauer B. Shawver L.K. Plate K.H. Risau W. Ullrich A. Nature. 1994; 367: 576-579Crossref PubMed Scopus (1162) Google Scholar). vascular endothelial growth factors 165-amino acid VEGF isoform 121-amino acid VEGF isoform fibroblast growth factor human umbilical vein-derived endothelial Dulbecco's phosphate-buffered saline. In the absence of heparin-like molecules, VEGF165 does not bind efficiently to the VEGF receptors of various cell types (10Gitay-Goren H. Soker S. Vlodavsky I. Neufeld G. J. Biol. Chem. 1992; 267: 6093-6098Abstract Full Text PDF PubMed Google Scholar, 11Gitay-Goren H. Halaban R. Neufeld G. Biochem. Biophys. Res. Commun. 1993; 190: 702-709Crossref PubMed Scopus (137) Google Scholar). It is possible that an interaction with heparin-like molecules induces a conformational change that is required for the efficient binding of VEGF165 to flk-1 (22Tessier S. Rockwell P. Hicklin D. Cohen T. Levi B.-Z. Witte L. Lemischka I.R. Neufeld G. J. Biol. Chem. 1994; 269: 12456-12461Abstract Full Text PDF PubMed Google Scholar). VEGF121 induces in vascular endothelial cells biological responses that seem to be similar to those induced by VEGF165, but does not bind to heparin. Therefore, it is possible that in contrast to VEGF165, VEGF121 may be permanently folded in a conformation that allows heparin-independent binding to VEGF receptors. However, cell-surface heparin-like molecules could also regulate the binding ability of VEGF receptors. Such a mechanism has been recently suggested with regard to the interaction of acidic fibroblast growth factor with FGF receptors (23Kan M.K. Wang F. Xu J.M. Crabb J.W. Hou J.Z. McKeehan W.L. Science. 1993; 259: 1918-1921Crossref PubMed Scopus (476) Google Scholar). We have studied the interaction of VEGF121 and VEGF165 with the VEGF receptors of fit-1-expressing YU-ZAZ6 melanoma cells to distinguish between the two possibilities described above. We find that VEGF121 and VEGF165 bind to common cell-surface VEGF receptors on these cells. Unexpectedly, although VEGF121 is unable to bind to heparin-like molecules, digestion of cell-surface heparan sulfates with heparinase inhibited VEGF121 binding to the VEGF receptors of the cells. Materials—VEGF166 and VEGF121 were purified from the conditioned medium of Sf9 insect cells infected with a baculovirus-based expression vector essentially as described (22Tessier S. Rockwell P. Hicklin D. Cohen T. Levi B.-Z. Witte L. Lemischka I.R. Neufeld G. J. Biol. Chem. 1994; 269: 12456-12461Abstract Full Text PDF PubMed Google Scholar, 24Cohen T. Gitay-Goren H. Neufeld G. Levi B.-Z. Growth Factors. 1992; 7: 131-138Crossref PubMed Scopus (51) Google Scholar). The initial yield of VEGF121 was ~20 mg/liter of conditioned medium as determined by an enzyme-linked immunosorbent assay. VEGF121 was purified from serum-free conditioned medium by two purification steps including hydrophobic chromatography on phenyl-Sepharose and anion-exchange chromatography using a Bio-Rad Econo-pac Q cartridge. The procedure yielded ~1 mg of highly purified VEGFl21 as assessed by silver staining of SDS-polyacrylamide gels and by an enzyme-linked immunosorbent assay. Recombinant human basic FGF was prepared from bacteria as described previously (10Gitay-Goren H. Soker S. Vlodavsky I. Neufeld G. J. Biol. Chem. 1992; 267: 6093-6098Abstract Full Text PDF PubMed Google Scholar). Tissue culture medium and reagents were from Biological Industries (Beth Haemek, Israel), and tissue culture plastic ware was from Nunc. Disuccinimidyl suberate was from Pierce, Na125I was from DuPont NEN, and heparin-Sepharose was from Pharmacia Biotech Inc. Recombinant bacterial heparinase-1 was kindly provided by Dr. Zimermann (Ibex Technologies, Montreal). High molecular mass protein markers were obtained from Bio-Rad. Porcine mucosa-derived heparin (H-7005) was purchased from Sigma, as were all other chemicals. Cell Culture—The human melanoma cell line YU-ZAZ6 (25Halaban R. Kwon B.S. Ghosh S. Delli Bovi P. Baird A. Oncogene Res. 1988; 3: 177-186PubMed Google Scholar) was cultured as described (11Gitay-Goren H. Halaban R. Neufeld G. Biochem. Biophys. Res. Commun. 1993; 190: 702-709Crossref PubMed Scopus (137) Google Scholar). Human umbilical vein-derived endothelial (HUE) cells were isolated and cultured on gelatin-coated dishes in M-199 medium containing 20% fetal calf serum, 4 mM glutamine, antibiotics, and 2 ng/ml basic FGF, which was added every other day to the cells (26Neufeld G. Gospodarowicz D. J. Cell. Physiol. 1988; 136: 537-542Crossref PubMed Scopus (64) Google Scholar). Iodination of VEGF and Binding and Cross-linking Experiments—Iodination of 5 µg of human recombinant VEGF165 or VEGF121 was done using either the chloramine-T method or the IODO-GEN method with identical results (6Vaisman N. Gospodarowicz D. Neufeld G. J. Biol. Chem. 1990; 265: 19461-19466Abstract Full Text PDF PubMed Google Scholar, 10Gitay-Goren H. Soker S. Vlodavsky I. Neufeld G. J. Biol. Chem. 1992; 267: 6093-6098Abstract Full Text PDF PubMed Google Scholar, 11Gitay-Goren H. Halaban R. Neufeld G. Biochem. Biophys. Res. Commun. 1993; 190: 702-709Crossref PubMed Scopus (137) Google Scholar, 27Neufeld G. Gospodarowicz D. J. Biol. Chem. 1985; 260: 13860-13868Abstract Full Text PDF PubMed Google Scholar). Free iodine was separated from 125I-VEGF121 using gel filtration on a Sephadex G-25 column, whereas separation of free iodine from 125I-VEGF165 was accomplished using heparin-Sepharose affinity chromatography. The specific activities of 125I-VEGF121 and 125I-VEGF165 were 0.5–1.5 × 105 cpm/ng. YU-ZAZ6 cells were grown to confluence in 6- or 10-cm dishes (for cross-linking experiments) or in 24-well plates (for binding experiments) coated with gelatin. The cells were washed with cold PBS, and the binding of 125I-VEGF121 or 125I-VEGF165 was carried out in binding buffer containing Dulbecco's modified Eagle's medium, 25 mM HEPES, pH 7.4, and 0.1% gelatin (6Vaisman N. Gospodarowicz D. Neufeld G. J. Biol. Chem. 1990; 265: 19461-19466Abstract Full Text PDF PubMed Google Scholar) for 2 h at 4 °C. In binding experiments, the level of nonspecific binding was determined in the presence of 1–2 µg/ml of the corresponding VEGF isoform. At the end of the binding reaction, the cells were washed three times with PBS containing 0.1% bovine serum albumin. The cells were lysed with 0.2 N NaOH, and aliquots were counted in a γ-counter. In cross-linking experiments, the washed cell layer was incubated with disuccinimidyl suberate, followed by SDS-polyacrylamide gel electrophoresis chromatography of cell lysates and autoradiography as described previously for 125I-VEGF165 (10Gitay-Goren H. Soker S. Vlodavsky I. Neufeld G. J. Biol. Chem. 1992; 267: 6093-6098Abstract Full Text PDF PubMed Google Scholar, 11Gitay-Goren H. Halaban R. Neufeld G. Biochem. Biophys. Res. Commun. 1993; 190: 702-709Crossref PubMed Scopus (137) Google Scholar). All experiments were repeated at least three times with similar results. Error bars represent the deviation from the mean in duplicate or triplicate measurements. Expression of flt-1 and KDR/flk-1 mRNAs—Total RNA was prepared from YU-ZAZ6 and HUE cells using the Tri-reagent system (Molecular Research Center, Inc., Cincinnati, OH) according to the vendor's instructions. The RNA (20 µg) was applied to each lane and electrophoresed on 1.2% agarose containing 2.2 m formaldehyde. The RNA was blotted onto Nytran filters (Schleicher & Schuell) and hybridized to a 1.8-kilobase HindIII fragment derived from the 5′-end of the flt-1 cDNA (28Shibuya M. Yamaguchi S. Yamane A. Ikeda T. Tojo A. Matsushime H. Sato M. Oncogene. 1990; 5: 519-524PubMed Google Scholar) or to a 0.8-kilobase fragment (nucleotides 728–1519) of the KDR VEGF receptor (29Terman B.I. Carrion M.E. Kovacs E. Rasmussen B.A. Eddy R.L. Shows T.B. Oncogene. 1991; 6: 1677-1683PubMed Google Scholar, 30Terman B.L. Dougher-Vermazen M. Carrion M.E. Dimitrov D. Armellino D.C. Gospodarowicz D. Böhlen P. Biochem. Biophys. Res. Commun. 1992; 187: 1579-1586Crossref PubMed Scopus (1405) Google Scholar). Hybridization was carried out at 42 °C for 48 h in a mixture containing 50% formamide. Filters were washed at high stringency (0.1 × SSC plus 0.1% SDS at 60 °C). The filters were then autoradiographed. VEGF121 Stimulates the Proliferation of HUE Cells and Binds to the VEGF Receptors of YU-ZAZ6 Melanoma Cells—Cells from various melanoma cell lines express VEGF receptors (11Gitay-Goren H. Halaban R. Neufeld G. Biochem. Biophys. Res. Commun. 1993; 190: 702-709Crossref PubMed Scopus (137) Google Scholar), but do not respond to VEGF165 stimulation by proliferation. YU-ZAZ6 cells express the mRNA of the VEGF receptor flt-1, but do not contain detectable levels of the mRNA encoding the VEGF receptor KDR/flk-1, while HUE cells contain detectable levels of both transcripts (Fig. 1) (22Tessier S. Rockwell P. Hicklin D. Cohen T. Levi B.-Z. Witte L. Lemischka I.R. Neufeld G. J. Biol. Chem. 1994; 269: 12456-12461Abstract Full Text PDF PubMed Google Scholar, 30Terman B.L. Dougher-Vermazen M. Carrion M.E. Dimitrov D. Armellino D.C. Gospodarowicz D. Böhlen P. Biochem. Biophys. Res. Commun. 1992; 187: 1579-1586Crossref PubMed Scopus (1405) Google Scholar). Purified VEGF121 stimulated the proliferation of HUE cells at a half-maximal concentration of 1–2 ng/ml (Fig. 2). To compare the receptor binding characteristics of VEGF121 and VEGF165, both VEGF isoforms were iodinated and used in binding and cross-linking studies. 125I-VEGF121 and 125I-VEGF165 were as active in endothelial cell proliferation assays as the corresponding unlabeled VEGF species (data not shown).Fig. 2Mitogenic activity of VEGF121 and VEGF165. HUE cells were seeded in 24-well plates (10,000 cells/well) and incubated in the presence of increasing concentrations of either VEGF121 (■) or VEGF165 (□), which were added to the cells every other day. The cells were trypsinized and counted in a Coulter counter after 4 days.View Large Image Figure ViewerDownload Hi-res image Download (PPT) 125I-VEGF121 was able to bind to the VEGF receptors of the melanoma cells. The binding of 125I-VEGF121 was inhibited by unlabeled VEGF165 (Fig. 3A), indicating that 125I-VEGF165 and 125I-VEGF121 bind to common receptors. The dissociation constants of 125I-VEGF121 and 125I-VEGF165 binding were determined according to the method of Scatchard (31Scatchard G. Ann. N. Y. Acad. Sci. 1949; 51: 660-672Crossref Scopus (17815) Google Scholar) using the LIGAND program (32Munson P.J. Rodbard D. Anal. Biochem. 1980; 107: 220-239Crossref PubMed Scopus (7772) Google Scholar). The dissociation constant of 125I-VEGF165 was 6 × 10−11 m, and that of 126I-VEGF121 was 6.5 × 10−11 m (data not shown). Cross-linking experiments also indicated that 125I-VEGF121 and 125I-VEGF165 bind to common VEGF receptors on the melanoma cells (Fig. 4). The apparent size of the 125I-VEGF121-receptor complexes appeared to be smaller by ~10 kDa compared with the apparent size of the 125I-VEGF165-receptor complexes (Fig. 4, compare lanes 1 and 5). This mass difference corresponds to the difference between the masses of VEGF121 and VEGF165 dimers.Fig. 4Cross-linking of 125I-VEGF165 and 125I-VEGF121 to the VEGF receptors of YU-ZAZ6 melanoma cells is inhibited by heparin and by an excess of unlabeled VEGF121 or VEGF165. YU-ZAZ6 cells were grown to confluence in 10-cm dishes (for 125I-VEGF121 binding) or in 6-cm dishes (for 125I-VEGF165 binding). 125I-VEGF165 (lanes 1–4) or 125I-VEGF121 (lanes 5–8) was bound and subsequently cross-linked to the cells. Some of the dishes received, in addition to the labeled growth factors, the following: heparin (1 µg/ml; lanes 2 and 6), VEGF121 (2 µg/ml; lanes 3 and 7), and VEGF165 (1 µg/ml; lanes 4 and 8). 125I-VEGF-receptor complexes were visualized by autoradiography of the 6% gel following SDS-polyacrylamide gel electrophoresis chromatography.View Large Image Figure ViewerDownload Hi-res image Download (PPT) Heparin Inhibits the Binding of 125I-VEGF121 and 125I-VEGF165 to the VEGF Receptors of YU-ZAZ6 Melanoma Cells—In contrast to VEGF165, VEGF121 is not a heparin-binding growth factor (8Houck K.A. Leung D.W. Rowland A.M. Winer J. Ferrara N. J. Biol. Chem. 1992; 267: 26031-26037Abstract Full Text PDF PubMed Google Scholar). When 125I-VEGF121 or 125I-VEGF165 was bound to the melanoma cells in the presence of increasing concentrations of heparin, the binding of both VEGF variants was inhibited (Fig. 3B). Inclusion of 1 µg/ml heparin in the binding reaction also completely inhibited the formation of 125I-VEGF121-receptor and 125I-VEGF165-receptor complexes in cross-linking experiments (Fig. 4, lanes 2 and 6, respectively). These results indicate that heparin has the capacity to regulate the VEGF binding ability of the VEGF receptors of the melanoma cells since VEGF121 lacks heparin binding ability. Digestion with Heparinase Abolishes 125I-VEGF121 and 125I-VEGF165 Binding to the VEGF Receptors of YU-ZAZ6 Melanoma Cells—To determine whether the binding of 125I-VEGF121 to the VEGF receptors of the cells is affected by cell-surface heparan sulfates, the cells were digested with heparinase. The VEGF receptors of heparinase-digested melanoma cells lost their 125I-VEGF121 and 125I-VEGF165 binding ability, indicating that cell-surface heparin-like molecules are required for the efficient binding of both VEGF species to the VEGF receptors (Fig. 5A, columns 3 and 7; and Fig. 5B, lanes 3 and 7). The loss of 125I-VEGF165 binding ability could be partially restored by the inclusion of 1 µg/ml heparin in the binding reaction (Fig. 5A, column 4; and Fig. 5B, lane 4). Surprisingly, the 125I-VEGF121 binding ability of the VEGF receptors of heparinase-digested cells could not be restored by the addition of exogenous heparin (Fig. 5A, column 8; and Fig. 5B, lane 8). This inability was not due to inappropriate concentrations of heparin or 125I-VEGF121. 125I-VEGF121 binding could not be detected in the presence of heparin concentrations ranging from 0.01 to 100 µg/ml, and binding could not be detected when 125I-VEGF121 concentrations as high as 150 ng/ml were used (data not shown). The Inhibitory Effect of Heparin Is Mediated by Cell-surface Heparin-binding Sites—To find out whether cell-surface heparin receptors mediate the inhibitory effects of exogenous heparin, melanoma cells were incubated with heparin (1 µg/ml) prior to 125I-VEGF121 or 125I-VEGF165 binding. The cells were washed extensively with PBS, and 125I-VEGF121 or 125I-VEGF165 was subsequently bound to the cells in the absence of heparin. The specific binding of both l25I-VEGF121 and 125I-VEGF165 was inhibited in cells that were preincubated with heparin (Fig. 6). These results indicate that heparin binds to heparin receptors on the cell surface and that the immobilized heparin or the heparin-activated heparin receptors are subsequently able to modulate the binding ability of the VEGF receptors of the melanoma cells. We have demonstrated here that VEGF121 and VEGF165 bind to common receptors on YU-ZAZ6 melanoma cells. Furthermore, we show that the interaction of both VEGF isoforms with the VEGF receptors of the cells is inhibited by exogenous heparin. Since VEGF121 is not a heparin-binding protein, this observation implies that heparin can regulate the binding ability of certain cell-surface VEGF receptors. This conclusion is further supported by experiments that demonstrate that the binding of both VEGF isoforms to the VEGF receptors of the melanoma cells is inhibited when cell-surface heparan sulfates are removed by heparinase digestion. A similar regulation of the binding ability of FGF receptor-1 by heparin-like molecules has also been described (23Kan M.K. Wang F. Xu J.M. Crabb J.W. Hou J.Z. McKeehan W.L. Science. 1993; 259: 1918-1921Crossref PubMed Scopus (476) Google Scholar). In that case, there is evidence indicating that heparin binds to FGF receptor-1. However, the mechanism by which heparin affects the VEGF binding ability of the VEGF receptors of YU-ZAZ6 cells may not require direct binding of heparin-like molecules to the VEGF receptors. Cell-surface heparin and heparan sulfate receptors may be activated as a result of heparin binding (33Cavari S. Vannucchi S. FEBS Lett. 1993; 323: 155-158Crossref PubMed Scopus (18) Google Scholar, 34Barzu T. Molho P. Tobelem G. Petitou M. Caen J. Biochim. Biophys. Acta. 1985; 845: 196-203Crossref PubMed Scopus (190) Google Scholar), and these could consequently modulate the VEGF binding ability of the receptors. The existence of such heparin-binding receptors on the cell surface of the melanoma cells is indicated by the persistence of the modulatory effects of heparin following extensive washing of heparin-treated cells. Our experiments suggest that heparin may regulate the VEGF binding ability of the VEGF receptor flt-1 since YU-ZAZ6 melanoma cells express the mRNA encoding this receptor, but not the mRNA encoding the other known VEGF receptor type, KDR/flk-1. This possibility is supported by data indicating that a soluble form of this receptor is indeed a heparin-binding protein (35Kendall R.L. Thomas K.A. Proc. Natl. Acad. Sci. U. S. A. 1993; 90: 10705-10709Crossref PubMed Scopus (1204) Google Scholar). However, the molecular mass of the 125I-VEGF-containing cross-linked products in YU-ZAZ6 cells is smaller than the expected mass of the flt-1-containing complexes (20Park J.E. Chen H.H. Winer J. Houck K.A. Ferrara N. J. Biol. Chem. 1994; 269: 25646-25654Abstract Full Text PDF PubMed Google Scholar). It is therefore possible that the VEGF-receptor complexes seen in the melanoma cells contain shorter forms of the flt-1 receptor or alternatively another VEGF receptor type. Our findings do not rule out a role for heparin-induced regulatory effects that are mediated through an interaction between heparin and a heparin-binding VEGF isoform such as VEGF165. In the case of VEGF165, the overall modulatory effect of heparin probably represents a summation of effects resulting from the binding of heparin to the growth factor and from the effects of heparin at the VEGF receptor level. In contrast, VEGF121 does not bind to heparin. Therefore, all the effects of heparin or heparan sulfate on VEGF121 binding apparently occur at the VEGF receptor level. Heparin and heparan sulfates may therefore modulate in dissimilar ways the biological activities of VEGF121 and VEGF165. Indeed, when the effects of heparin on the binding of VEGF121 and VEGF165 to heparinase-treated melanoma cells are compared, such differential effects of heparin are observed. Heparin is able to partially restore the binding of VEGF165 to the VEGF receptors of the heparinase-treated cells, but its addition is not sufficient for the restoration of VEGF121 binding. A specific heparan sulfate may be required to restore VEGF121 binding to the VEGF receptors of heparinase-treated YU-ZAZ6 cells, and we are currently exploring this possibility. The effect of exogenous heparin on the binding of VEGF165 to the VEGF receptors of YU-ZAZ6 melanoma cells seems to be modulated by cell surface-located heparan sulfates. When VEGF165 is bound to untreated cells, heparin inhibits the binding of VEGF165. Upon removal of the endogenous cell-surface heparan sulfates, an identical concentration of heparin potentiates 125I-VEGF165 binding. These results are in agreement with published data showing that heparin stimulates VEGF165 binding to cell-surface VEGF receptors in a variety of cell types including HUE cells and several types of melanoma cells (10Gitay-Goren H. Soker S. Vlodavsky I. Neufeld G. J. Biol. Chem. 1992; 267: 6093-6098Abstract Full Text PDF PubMed Google Scholar, 11Gitay-Goren H. Halaban R. Neufeld G. Biochem. Biophys. Res. Commun. 1993; 190: 702-709Crossref PubMed Scopus (137) Google Scholar). Heparin potentiates the binding of VEGF165 to a soluble fusion protein containing the extracellular domain of the flk-1 receptor, but inhibits the binding of 125I-VEGF165 to NIH-3T3 cells expressing a flk-1/c-fms chimeric receptor (22Tessier S. Rockwell P. Hicklin D. Cohen T. Levi B.-Z. Witte L. Lemischka I.R. Neufeld G. J. Biol. Chem. 1994; 269: 12456-12461Abstract Full Text PDF PubMed Google Scholar). It therefore seems that different cell types express different populations of cell-surface heparan sulfates, and these heparan sulfates may differentially affect the interaction of VEGF165 with VEGF receptors. In conclusion, our results suggest that the function of some VEGF receptor species may be regulated, either directly or indirectly, by heparin-like molecules. The various VEGF isoforms differ with respect to their heparin binding capabilities, and complex effects may be generated by heparin-like molecules under diverse in vivo conditions. Changes in the levels of heparan sulfates in melanoma cells as a function of their metastatic potential have been reported (36Moczar M. Caux F. Bailly M. Berthier O. Dore J.F. Clin. Exp. Metastasis. 1993; 11: 462-471Crossref PubMed Scopus (20) Google Scholar, 37) and may perhaps be related to the regulation of the activity of growth factors such as VEGF and FGF by heparan sulfates. We thank Michal Migdal for the KDR probe.
Referência(s)