VEGF162, A New Heparin-binding Vascular Endothelial Growth Factor Splice Form That Is Expressed in Transformed Human Cells
2003; Elsevier BV; Volume: 278; Issue: 19 Linguagem: Inglês
10.1074/jbc.m212224200
ISSN1083-351X
AutoresTali Lange, Noga Guttmann‐Raviv, Limor Baruch, Marcelle Machluf, Gera Neufeld,
Tópico(s)Cancer, Hypoxia, and Metabolism
ResumoThe splice forms of vascular endothelial growth factor (VEGF) differ in biological properties such as the receptor types that they recognize and their interaction with heparan sulfate proteoglycans. We have identified a new VEGF mRNA splice form encoding a VEGF species containing 162 amino acids (VEGF162) in human A431 ovarian carcinoma cells. This novel mRNA contains the peptides encoded by exons 1–5, 6A, 6B, and 8 of the VEGF gene. Recombinant VEGF162is biologically active. It induces proliferation of endothelial cellsin vitro and angiogenesis in vivo as determined by the alginate bead assay. VEGF162 binds less efficiently than VEGF145 but more efficiently than VEGF165to a natural basement membrane produced by corneal endothelial cells. VEGF138, an artificial VEGF form that contains exon 6B but lacks exons 6A and 7, did not bind to this basement membrane at all, indicating that exon 6B probably interferes with the interaction of exon 6A with heparin and heparan sulfate proteoglycans. The splice forms of vascular endothelial growth factor (VEGF) differ in biological properties such as the receptor types that they recognize and their interaction with heparan sulfate proteoglycans. We have identified a new VEGF mRNA splice form encoding a VEGF species containing 162 amino acids (VEGF162) in human A431 ovarian carcinoma cells. This novel mRNA contains the peptides encoded by exons 1–5, 6A, 6B, and 8 of the VEGF gene. Recombinant VEGF162is biologically active. It induces proliferation of endothelial cellsin vitro and angiogenesis in vivo as determined by the alginate bead assay. VEGF162 binds less efficiently than VEGF145 but more efficiently than VEGF165to a natural basement membrane produced by corneal endothelial cells. VEGF138, an artificial VEGF form that contains exon 6B but lacks exons 6A and 7, did not bind to this basement membrane at all, indicating that exon 6B probably interferes with the interaction of exon 6A with heparin and heparan sulfate proteoglycans. vascular endothelial growth factor Chinese hamster ovary dihydrofolate reductase The various forms of vascular endothelial growth factor (VEGF)1 are generated by alternative splicing from a single gene (1Tischer E. Mitchell R. Hartman T. Silva M. Gospodarowicz D. Fiddes J.C. Abraham J.A. J. Biol. Chem. 1991; 266: 11947-11954Abstract Full Text PDF PubMed Google Scholar, 2Houck K.A. Ferrara N. Winer J. Cachianes G. Li B. Leung D.W. Mol. Endocrinol. 1991; 5: 1806-1814Crossref PubMed Scopus (1229) Google Scholar, 3Neufeld G. Cohen T. Gitay-Goren H. Poltorak Z. Tessler S. Gengrinovitch S. Levi B.-Z. Cancer Metastasis Rev. 1996; 15: 153-158Crossref PubMed Scopus (145) Google Scholar, 4Poltorak Z. Cohen T. Sivan R. Kandelis Y. Spira G. Vlodavsky I. Keshet E. Neufeld G. J. Biol. Chem. 1997; 272: 7151-7158Abstract Full Text Full Text PDF PubMed Scopus (446) Google Scholar). The domains encoded by exons 1–5 of the VEGF gene contain information required for the recognition of the tyrosine kinase VEGF receptors 1 (flt-1) and 2 (KDR/flk-1) (5Keyt B.A. Nguyen H.V. Berleau L.T. Duarte C.M. Park J. Chen H. Ferrara N. J. Biol. Chem. 1996; 271: 5638-5646Abstract Full Text Full Text PDF PubMed Scopus (421) Google Scholar) and are present in all the VEGF splice forms. Most VEGF splice forms are distinguished by the presence or absence of the peptides encoded by exons 6 and 7 of the VEGFgene that code for two independent heparin-binding domains. VEGF121 lacks both exons and does not bind to heparin. VEGF165 contains the exon 7-encoded peptide, VEGF145 contains the peptide encoded by exon 6A, and VEGF189 contains both exon 6A and exon 7 (3Neufeld G. Cohen T. Gitay-Goren H. Poltorak Z. Tessler S. Gengrinovitch S. Levi B.-Z. Cancer Metastasis Rev. 1996; 15: 153-158Crossref PubMed Scopus (145) Google Scholar, 6Houck 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, 7Park J.E. Keller G.A. Ferrara N. Mol. Biol. Cell. 1993; 4: 1317-1326Crossref PubMed Scopus (956) Google Scholar, 8Neufeld G. Cohen T. Gengrinovitch S. Poltorak Z. FASEB J. 1999; 13: 9-22Crossref PubMed Scopus (3128) Google Scholar). VEGF206 has a structure similar to that of VEGF189, except that it contains both exons 6A and 6B of the VEGF gene. However, the contribution of exon 6B to the biological properties of VEGF206 has not been studied. The amino acids encoded by exon 8 are present in most of the VEGF splice forms. However, it was recently found that in the novel VEGF form VEGF165b, the 6 amino acids encoded by exon 8 are replaced by 6 amino acids derived from a putative ninth exon to yield a VEGF form that probably inhibits angiogenesis (9Bates D.O. Cui T.G. Doughty J.M. Winkler M. Sugiono M. Shields J.D. Peat D. Gillatt D. Harper S.J. Cancer Res. 2002; 62: 4123-4131PubMed Google Scholar). In another recently described VEGF splice form, exon 8 is completely truncated to generate a 148-amino acid form (10Whittle C. Gillespie K. Harrison R. Mathieson P.W. Harper S.J. Clin. Sci. (Lond.). 1999; 97: 303-312Crossref PubMed Scopus (85) Google Scholar). In addition, it was recently found that VEGF forms possessing an extended N termini of unknown function also exist (11Meiron M. Anunu R. Scheinman E.J. Hashmueli S. Levi B.Z. Biochem. Biophys. Res. Commun. 2001; 282: 1053-1060Crossref PubMed Scopus (48) Google Scholar).Most VEGF isoforms induce proliferation of vascular endothelial cells, induce angiogenesis, and cause permeabilization of blood vessels (8Neufeld G. Cohen T. Gengrinovitch S. Poltorak Z. FASEB J. 1999; 13: 9-22Crossref PubMed Scopus (3128) Google Scholar). Recently, certain differences between some common VEGF splice forms have been reported. The neuropilin-1 and neuropilin-2 receptors function as receptors for axon guidance factors belonging to the semaphorin family (12Neufeld G. Cohen T. Shraga N. Lange T. Kessler O. Herzog Y. Trends Cardiovasc. Med. 2002; 12: 13-19Crossref PubMed Scopus (295) Google Scholar, 13Liu B.P. Strittmatter S.M. Curr. Opin. Cell Biol. 2001; 13: 619-626Crossref PubMed Scopus (152) Google Scholar). It was found that both neuropilins also function as VEGF receptors that differentiate between various forms of VEGF (14Gitay-Goren H. Cohen T. Tessler S. Soker S. Gengrinovitch S. Rockwell P. Klagsbrun M. Levi B.-Z. Neufeld G. J. Biol. Chem. 1996; 271: 5519-5523Abstract Full Text Full Text PDF PubMed Scopus (171) Google Scholar, 15Soker S. Takashima S. Miao H.Q. Neufeld G. Klagsbrun M. Cell. 1998; 92: 735-745Abstract Full Text Full Text PDF PubMed Scopus (2057) Google Scholar). Thus, VEGF121 cannot bind to either of the neuropilins, whereas VEGF165 binds efficiently to both receptors, and VEGF145 binds well to neuropilin-2 but not to neuropilin-1 (16Gluzman-Poltorak Z. Cohen T. Herzog Y. Neufeld G. J. Biol. Chem. 2000; 275: 18040-18045Abstract Full Text Full Text PDF PubMed Scopus (306) Google Scholar). Such differences are also reflected in the functional properties of the VEGF splice forms. It was found that mice expressing only VEGF121 do not develop properly (17Stalmans I. Ng Y.S. Rohan R. Fruttiger M. Bouche A. Yuce A. Fujisawa H. Hermans B. Shani M. Jansen S. Hicklin D. Anderson D.J. Gardiner T. Hammes H.P. Moons L. Dewerchin M. Collen D. Carmeliet P. D'Amore P.A. J. Clin. Invest. 2002; 109: 327-336Crossref PubMed Scopus (421) Google Scholar, 18Carmeliet P. Ng Y.S. Nuyens D. Theilmeier G. Brusselmans K. Cornelissen I. Ehler E. Kakkar V.V. Stalmans I. Mattot V. Perriard J.C. Dewerchin M. Flameng W. Nagy A. Lupu F. Moons L. Collen D. D'Amore P.A. Shima D.T. Nat. Med. 1999; 5: 495-502Crossref PubMed Scopus (557) Google Scholar) because the heparan sulfate-binding VEGF forms are required for correct branching of blood vessels during development (19Ruhrberg C. Gerhardt H. Golding M. Watson R. Ioannidou S. Fujisawa H. Betsholtz C. Shima D.T. Genes Dev. 2002; 16: 2684-2698Crossref PubMed Scopus (705) Google Scholar). Likewise, mice expressing only VEGF189 display impaired arterial development (17Stalmans I. Ng Y.S. Rohan R. Fruttiger M. Bouche A. Yuce A. Fujisawa H. Hermans B. Shani M. Jansen S. Hicklin D. Anderson D.J. Gardiner T. Hammes H.P. Moons L. Dewerchin M. Collen D. Carmeliet P. D'Amore P.A. J. Clin. Invest. 2002; 109: 327-336Crossref PubMed Scopus (421) Google Scholar), suggesting that each VEGF form possesses specific characteristics and that the various VEGF forms complement each other to achieve a balanced angiogenic response.Exon 6B was first identified in VEGF206 (Fig.1C), but the effect of exon 6B on the biological properties of VEGF206 had not been studied because VEGF206, like VEGF189, is not secreted into the medium of cells that produce these VEGF forms and is thus difficult to isolate and characterize. We report here the identification of a new exon 6B-containing form of VEGF expressed by A431 ovarian carcinoma cells. This VEGF form is 162 amino acids long (VEGF162). In this work, we have characterized the properties of VEGF162 and compared them with those of the closely related VEGF145 and those of VEGF138, an artificial exon 6B-containing VEGF form.DISCUSSIONVEGF206 was identified as a nonsecreted VEGF form in which alternative splicing in the region of exon 6 of the VEGF mRNA resulted in a 41-amino acid insertion relative to the more widely distributed VEGF165. The insertion included the 24 amino acids found in VEGF189 (2Houck K.A. Ferrara N. Winer J. Cachianes G. Li B. Leung D.W. Mol. Endocrinol. 1991; 5: 1806-1814Crossref PubMed Scopus (1229) Google Scholar) and 17 additional amino acids. These two different portions of exon 6 have become known as exons 6A and 6B. VEGF206 is a relatively rare form of VEGF that is not released into the conditioned medium of producing cells and is therefore difficult to study. The presence of exon 6B in VEGF206 indicated that exon 6B may be present in additional forms and that the properties of such forms may perhaps be studied more readily. We therefore set out to look for such forms.We have found that A431 squamous carcinoma cells contain mRNA encoding an additional exon 6B-containing VEGF form. This form is identical to VEGF145 (4Poltorak Z. Cohen T. Sivan R. Kandelis Y. Spira G. Vlodavsky I. Keshet E. Neufeld G. J. Biol. Chem. 1997; 272: 7151-7158Abstract Full Text Full Text PDF PubMed Scopus (446) Google Scholar), except for the added amino acids encoded by exon 6B that lead to the production of a 162-amino acid-long VEGF form. This form may have been wrongly identified as VEGF165 because the mass of VEGF162 and the size of the mRNA encoding VEGF162 closely resemble VEGF165. Recombinant VEGF162 was secreted from CHO DHFR− cells and was biologically active as determined by endothelial cell proliferation assays and by its ability to induce angiogenesis in vivo. In these properties, it does not appear to be significantly different from VEGF145. VEGF162 bound to a native basement membrane produced by corneal endothelial cells. It bound to the basement membrane less efficiently than VEGF145 but better than VEGF165, which binds to such matrices relatively inefficiently, despite the substantial affinity that it displays toward heparin (4Poltorak Z. Cohen T. Sivan R. Kandelis Y. Spira G. Vlodavsky I. Keshet E. Neufeld G. J. Biol. Chem. 1997; 272: 7151-7158Abstract Full Text Full Text PDF PubMed Scopus (446) Google Scholar, 7Park J.E. Keller G.A. Ferrara N. Mol. Biol. Cell. 1993; 4: 1317-1326Crossref PubMed Scopus (956) Google Scholar). The synthetic VEGF form VEGF138, which contains exon 6B but lacks exon 6A, was not able to bind to this basement membrane, behaving very similarly in this respect to VEGF121 (4Poltorak Z. Cohen T. Sivan R. Kandelis Y. Spira G. Vlodavsky I. Keshet E. Neufeld G. J. Biol. Chem. 1997; 272: 7151-7158Abstract Full Text Full Text PDF PubMed Scopus (446) Google Scholar). It can therefore be concluded that exon 6B does not contribute to VEGF binding to the extracellular matrix. Rather, exon 6B seems to interfere with the interaction of exon 6A with basement membrane components, leading to a decreased extracellular matrix binding ability as compared with VEGF145. However, the interaction of VEGF162 with the basement membrane is still somewhat stronger than that of VEGF165.The evidence gathered in the past decade indicates that apparently insignificant differences in the properties of the VEGF splice forms have turned out to be biologically meaningful. Mice expressing only VEGF120 or VEGF188 develop abnormally, even if these VEGF forms are expressed at levels comparable with the expression levels of all VEGF forms put together (17Stalmans I. Ng Y.S. Rohan R. Fruttiger M. Bouche A. Yuce A. Fujisawa H. Hermans B. Shani M. Jansen S. Hicklin D. Anderson D.J. Gardiner T. Hammes H.P. Moons L. Dewerchin M. Collen D. Carmeliet P. D'Amore P.A. J. Clin. Invest. 2002; 109: 327-336Crossref PubMed Scopus (421) Google Scholar, 18Carmeliet P. Ng Y.S. Nuyens D. Theilmeier G. Brusselmans K. Cornelissen I. Ehler E. Kakkar V.V. Stalmans I. Mattot V. Perriard J.C. Dewerchin M. Flameng W. Nagy A. Lupu F. Moons L. Collen D. D'Amore P.A. Shima D.T. Nat. Med. 1999; 5: 495-502Crossref PubMed Scopus (557) Google Scholar, 19Ruhrberg C. Gerhardt H. Golding M. Watson R. Ioannidou S. Fujisawa H. Betsholtz C. Shima D.T. Genes Dev. 2002; 16: 2684-2698Crossref PubMed Scopus (705) Google Scholar, 30Mattot V. Moons L. Lupu F. Chernavvsky D. Gomez R.A. Collen D. Carmeliet P. J. Am. Soc. Nephrol. 2002; 13: 1548-1560Crossref PubMed Scopus (88) Google Scholar). These changes are the result of differential affinities to heparan sulfate proteoglycans and extracellular matrix components, differential recognition of VEGF receptors, and differential susceptibility to reactive oxygen species (8Neufeld G. Cohen T. Gengrinovitch S. Poltorak Z. FASEB J. 1999; 13: 9-22Crossref PubMed Scopus (3128) Google Scholar, 14Gitay-Goren H. Cohen T. Tessler S. Soker S. Gengrinovitch S. Rockwell P. Klagsbrun M. Levi B.-Z. Neufeld G. J. Biol. Chem. 1996; 271: 5519-5523Abstract Full Text Full Text PDF PubMed Scopus (171) Google Scholar, 15Soker S. Takashima S. Miao H.Q. Neufeld G. Klagsbrun M. Cell. 1998; 92: 735-745Abstract Full Text Full Text PDF PubMed Scopus (2057) Google Scholar, 16Gluzman-Poltorak Z. Cohen T. Herzog Y. Neufeld G. J. Biol. Chem. 2000; 275: 18040-18045Abstract Full Text Full Text PDF PubMed Scopus (306) Google Scholar, 31Gengrinovitch S. Berman B. David G. Witte L. Neufeld G. Ron D. J. Biol. Chem. 1999; 274: 10816-10822Abstract Full Text Full Text PDF PubMed Scopus (154) Google Scholar). These differences imply that certain forms of VEGF may act more efficiently than other forms in specific microenvironments and suggest that the differential splicing of VEGF may be more tightly regulated than is currently appreciated. Some evidence supporting such tight regulation is already available. For example, it was found that progesterone selectively up-regulates the expression of VEGF189 in decidual cells (32Ancelin M. Buteau-Lozano H. Meduri G. Osborne-Pellegrin M. Sordello S. Plouet J. Perrot-Applanat M. Proc. Natl. Acad. Sci. U. S. A. 2002; 99: 6023-6028Crossref PubMed Scopus (109) Google Scholar).To conclude, we have characterized a new, secreted, biologically active VEGF splice form. Whether this VEGF splice form has a biological role distinct from that of other VEGF forms is unclear at the moment, and it is not known whether there exist specific mechanisms that regulate the synthesis of the VEGF162 mRNA. Tools that allow easy discrimination between the expression patterns of VEGF162and the other splice forms will have to be developed to study thein vivo expression patterns of VEGF162. The elucidation of these questions, as well as the design of experiments aimed at the identification of the specific biological roles of VEGF162, will most likely be the focus for the continuation of the research in the near future. The various forms of vascular endothelial growth factor (VEGF)1 are generated by alternative splicing from a single gene (1Tischer E. Mitchell R. Hartman T. Silva M. Gospodarowicz D. Fiddes J.C. Abraham J.A. J. Biol. Chem. 1991; 266: 11947-11954Abstract Full Text PDF PubMed Google Scholar, 2Houck K.A. Ferrara N. Winer J. Cachianes G. Li B. Leung D.W. Mol. Endocrinol. 1991; 5: 1806-1814Crossref PubMed Scopus (1229) Google Scholar, 3Neufeld G. Cohen T. Gitay-Goren H. Poltorak Z. Tessler S. Gengrinovitch S. Levi B.-Z. Cancer Metastasis Rev. 1996; 15: 153-158Crossref PubMed Scopus (145) Google Scholar, 4Poltorak Z. Cohen T. Sivan R. Kandelis Y. Spira G. Vlodavsky I. Keshet E. Neufeld G. J. Biol. Chem. 1997; 272: 7151-7158Abstract Full Text Full Text PDF PubMed Scopus (446) Google Scholar). The domains encoded by exons 1–5 of the VEGF gene contain information required for the recognition of the tyrosine kinase VEGF receptors 1 (flt-1) and 2 (KDR/flk-1) (5Keyt B.A. Nguyen H.V. Berleau L.T. Duarte C.M. Park J. Chen H. Ferrara N. J. Biol. Chem. 1996; 271: 5638-5646Abstract Full Text Full Text PDF PubMed Scopus (421) Google Scholar) and are present in all the VEGF splice forms. Most VEGF splice forms are distinguished by the presence or absence of the peptides encoded by exons 6 and 7 of the VEGFgene that code for two independent heparin-binding domains. VEGF121 lacks both exons and does not bind to heparin. VEGF165 contains the exon 7-encoded peptide, VEGF145 contains the peptide encoded by exon 6A, and VEGF189 contains both exon 6A and exon 7 (3Neufeld G. Cohen T. Gitay-Goren H. Poltorak Z. Tessler S. Gengrinovitch S. Levi B.-Z. Cancer Metastasis Rev. 1996; 15: 153-158Crossref PubMed Scopus (145) Google Scholar, 6Houck 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, 7Park J.E. Keller G.A. Ferrara N. Mol. Biol. Cell. 1993; 4: 1317-1326Crossref PubMed Scopus (956) Google Scholar, 8Neufeld G. Cohen T. Gengrinovitch S. Poltorak Z. FASEB J. 1999; 13: 9-22Crossref PubMed Scopus (3128) Google Scholar). VEGF206 has a structure similar to that of VEGF189, except that it contains both exons 6A and 6B of the VEGF gene. However, the contribution of exon 6B to the biological properties of VEGF206 has not been studied. The amino acids encoded by exon 8 are present in most of the VEGF splice forms. However, it was recently found that in the novel VEGF form VEGF165b, the 6 amino acids encoded by exon 8 are replaced by 6 amino acids derived from a putative ninth exon to yield a VEGF form that probably inhibits angiogenesis (9Bates D.O. Cui T.G. Doughty J.M. Winkler M. Sugiono M. Shields J.D. Peat D. Gillatt D. Harper S.J. Cancer Res. 2002; 62: 4123-4131PubMed Google Scholar). In another recently described VEGF splice form, exon 8 is completely truncated to generate a 148-amino acid form (10Whittle C. Gillespie K. Harrison R. Mathieson P.W. Harper S.J. Clin. Sci. (Lond.). 1999; 97: 303-312Crossref PubMed Scopus (85) Google Scholar). In addition, it was recently found that VEGF forms possessing an extended N termini of unknown function also exist (11Meiron M. Anunu R. Scheinman E.J. Hashmueli S. Levi B.Z. Biochem. Biophys. Res. Commun. 2001; 282: 1053-1060Crossref PubMed Scopus (48) Google Scholar). Most VEGF isoforms induce proliferation of vascular endothelial cells, induce angiogenesis, and cause permeabilization of blood vessels (8Neufeld G. Cohen T. Gengrinovitch S. Poltorak Z. FASEB J. 1999; 13: 9-22Crossref PubMed Scopus (3128) Google Scholar). Recently, certain differences between some common VEGF splice forms have been reported. The neuropilin-1 and neuropilin-2 receptors function as receptors for axon guidance factors belonging to the semaphorin family (12Neufeld G. Cohen T. Shraga N. Lange T. Kessler O. Herzog Y. Trends Cardiovasc. Med. 2002; 12: 13-19Crossref PubMed Scopus (295) Google Scholar, 13Liu B.P. Strittmatter S.M. Curr. Opin. Cell Biol. 2001; 13: 619-626Crossref PubMed Scopus (152) Google Scholar). It was found that both neuropilins also function as VEGF receptors that differentiate between various forms of VEGF (14Gitay-Goren H. Cohen T. Tessler S. Soker S. Gengrinovitch S. Rockwell P. Klagsbrun M. Levi B.-Z. Neufeld G. J. Biol. Chem. 1996; 271: 5519-5523Abstract Full Text Full Text PDF PubMed Scopus (171) Google Scholar, 15Soker S. Takashima S. Miao H.Q. Neufeld G. Klagsbrun M. Cell. 1998; 92: 735-745Abstract Full Text Full Text PDF PubMed Scopus (2057) Google Scholar). Thus, VEGF121 cannot bind to either of the neuropilins, whereas VEGF165 binds efficiently to both receptors, and VEGF145 binds well to neuropilin-2 but not to neuropilin-1 (16Gluzman-Poltorak Z. Cohen T. Herzog Y. Neufeld G. J. Biol. Chem. 2000; 275: 18040-18045Abstract Full Text Full Text PDF PubMed Scopus (306) Google Scholar). Such differences are also reflected in the functional properties of the VEGF splice forms. It was found that mice expressing only VEGF121 do not develop properly (17Stalmans I. Ng Y.S. Rohan R. Fruttiger M. Bouche A. Yuce A. Fujisawa H. Hermans B. Shani M. Jansen S. Hicklin D. Anderson D.J. Gardiner T. Hammes H.P. Moons L. Dewerchin M. Collen D. Carmeliet P. D'Amore P.A. J. Clin. Invest. 2002; 109: 327-336Crossref PubMed Scopus (421) Google Scholar, 18Carmeliet P. Ng Y.S. Nuyens D. Theilmeier G. Brusselmans K. Cornelissen I. Ehler E. Kakkar V.V. Stalmans I. Mattot V. Perriard J.C. Dewerchin M. Flameng W. Nagy A. Lupu F. Moons L. Collen D. D'Amore P.A. Shima D.T. Nat. Med. 1999; 5: 495-502Crossref PubMed Scopus (557) Google Scholar) because the heparan sulfate-binding VEGF forms are required for correct branching of blood vessels during development (19Ruhrberg C. Gerhardt H. Golding M. Watson R. Ioannidou S. Fujisawa H. Betsholtz C. Shima D.T. Genes Dev. 2002; 16: 2684-2698Crossref PubMed Scopus (705) Google Scholar). Likewise, mice expressing only VEGF189 display impaired arterial development (17Stalmans I. Ng Y.S. Rohan R. Fruttiger M. Bouche A. Yuce A. Fujisawa H. Hermans B. Shani M. Jansen S. Hicklin D. Anderson D.J. Gardiner T. Hammes H.P. Moons L. Dewerchin M. Collen D. Carmeliet P. D'Amore P.A. J. Clin. Invest. 2002; 109: 327-336Crossref PubMed Scopus (421) Google Scholar), suggesting that each VEGF form possesses specific characteristics and that the various VEGF forms complement each other to achieve a balanced angiogenic response. Exon 6B was first identified in VEGF206 (Fig.1C), but the effect of exon 6B on the biological properties of VEGF206 had not been studied because VEGF206, like VEGF189, is not secreted into the medium of cells that produce these VEGF forms and is thus difficult to isolate and characterize. We report here the identification of a new exon 6B-containing form of VEGF expressed by A431 ovarian carcinoma cells. This VEGF form is 162 amino acids long (VEGF162). In this work, we have characterized the properties of VEGF162 and compared them with those of the closely related VEGF145 and those of VEGF138, an artificial exon 6B-containing VEGF form. DISCUSSIONVEGF206 was identified as a nonsecreted VEGF form in which alternative splicing in the region of exon 6 of the VEGF mRNA resulted in a 41-amino acid insertion relative to the more widely distributed VEGF165. The insertion included the 24 amino acids found in VEGF189 (2Houck K.A. Ferrara N. Winer J. Cachianes G. Li B. Leung D.W. Mol. Endocrinol. 1991; 5: 1806-1814Crossref PubMed Scopus (1229) Google Scholar) and 17 additional amino acids. These two different portions of exon 6 have become known as exons 6A and 6B. VEGF206 is a relatively rare form of VEGF that is not released into the conditioned medium of producing cells and is therefore difficult to study. The presence of exon 6B in VEGF206 indicated that exon 6B may be present in additional forms and that the properties of such forms may perhaps be studied more readily. We therefore set out to look for such forms.We have found that A431 squamous carcinoma cells contain mRNA encoding an additional exon 6B-containing VEGF form. This form is identical to VEGF145 (4Poltorak Z. Cohen T. Sivan R. Kandelis Y. Spira G. Vlodavsky I. Keshet E. Neufeld G. J. Biol. Chem. 1997; 272: 7151-7158Abstract Full Text Full Text PDF PubMed Scopus (446) Google Scholar), except for the added amino acids encoded by exon 6B that lead to the production of a 162-amino acid-long VEGF form. This form may have been wrongly identified as VEGF165 because the mass of VEGF162 and the size of the mRNA encoding VEGF162 closely resemble VEGF165. Recombinant VEGF162 was secreted from CHO DHFR− cells and was biologically active as determined by endothelial cell proliferation assays and by its ability to induce angiogenesis in vivo. In these properties, it does not appear to be significantly different from VEGF145. VEGF162 bound to a native basement membrane produced by corneal endothelial cells. It bound to the basement membrane less efficiently than VEGF145 but better than VEGF165, which binds to such matrices relatively inefficiently, despite the substantial affinity that it displays toward heparin (4Poltorak Z. Cohen T. Sivan R. Kandelis Y. Spira G. Vlodavsky I. Keshet E. Neufeld G. J. Biol. Chem. 1997; 272: 7151-7158Abstract Full Text Full Text PDF PubMed Scopus (446) Google Scholar, 7Park J.E. Keller G.A. Ferrara N. Mol. Biol. Cell. 1993; 4: 1317-1326Crossref PubMed Scopus (956) Google Scholar). The synthetic VEGF form VEGF138, which contains exon 6B but lacks exon 6A, was not able to bind to this basement membrane, behaving very similarly in this respect to VEGF121 (4Poltorak Z. Cohen T. Sivan R. Kandelis Y. Spira G. Vlodavsky I. Keshet E. Neufeld G. J. Biol. Chem. 1997; 272: 7151-7158Abstract Full Text Full Text PDF PubMed Scopus (446) Google Scholar). It can therefore be concluded that exon 6B does not contribute to VEGF binding to the extracellular matrix. Rather, exon 6B seems to interfere with the interaction of exon 6A with basement membrane components, leading to a decreased extracellular matrix binding ability as compared with VEGF145. However, the interaction of VEGF162 with the basement membrane is still somewhat stronger than that of VEGF165.The evidence gathered in the past decade indicates that apparently insignificant differences in the properties of the VEGF splice forms have turned out to be biologically meaningful. Mice expressing only VEGF120 or VEGF188 develop abnormally, even if these VEGF forms are expressed at levels comparable with the expression levels of all VEGF forms put together (17Stalmans I. Ng Y.S. Rohan R. Fruttiger M. Bouche A. Yuce A. Fujisawa H. Hermans B. Shani M. Jansen S. Hicklin D. Anderson D.J. Gardiner T. Hammes H.P. Moons L. Dewerchin M. Collen D. Carmeliet P. D'Amore P.A. J. Clin. Invest. 2002; 109: 327-336Crossref PubMed Scopus (421) Google Scholar, 18Carmeliet P. Ng Y.S. Nuyens D. Theilmeier G. Brusselmans K. Cornelissen I. Ehler E. Kakkar V.V. Stalmans I. Mattot V. Perriard J.C. Dewerchin M. Flameng W. Nagy A. Lupu F. Moons L. Collen D. D'Amore P.A. Shima D.T. Nat. Med. 1999; 5: 495-502Crossref PubMed Scopus (557) Google Scholar, 19Ruhrberg C. Gerhardt H. Golding M. Watson R. Ioannidou S. Fujisawa H. Betsholtz C. Shima D.T. Genes Dev. 2002; 16: 2684-2698Crossref PubMed Scopus (705) Google Scholar, 30Mattot V. Moons L. Lupu F. Chernavvsky D. Gomez R.A. Collen D. Carmeliet P. J. Am. Soc. Nephrol. 2002; 13: 1548-1560Crossref PubMed Scopus (88) Google Scholar). These changes are the result of differential affinities to heparan sulfate proteoglycans and extracellular matrix components, differential recognition of VEGF receptors, and differential susceptibility to reactive oxygen species (8Neufeld G. Cohen T. Gengrinovitch S. Poltorak Z. FASEB J. 1999; 13: 9-22Crossref PubMed Scopus (3128) Google Scholar, 14Gitay-Goren H. Cohen T. Tessler S. Soker S. Gengrinovitch S. Rockwell P. Klagsbrun M. Levi B.-Z. Neufeld G. J. Biol. Chem. 1996; 271: 5519-5523Abstract Full Text Full Text PDF PubMed Scopus (171) Google Scholar, 15Soker S. Takashima S. Miao H.Q. Neufeld G. Klagsbrun M. Cell. 1998; 92: 735-745Abstract Full Text Full Text PDF PubMed Scopus (2057) Google Scholar, 16Gluzman-Poltorak Z. Cohen T. Herzog Y. Neufeld G. J. Biol. Chem. 2000; 275: 18040-18045Abstract Full Text Full Text PDF PubMed Scopus (306) Google Scholar, 31Gengrinovitch S. Berman B. David G. Witte L. Neufeld G. Ron D. J. Biol. Chem. 1999; 274: 10816-10822Abstract Full Text Full Text PDF PubMed Scopus (154) Google Scholar). These differences imply that certain forms of VEGF may act more efficiently than other forms in specific microenvironments and suggest that the differential splicing of VEGF may be more tightly regulated than is currently appreciated. Some evidence supporting such tight regulation is already available. For example, it was found that progesterone selectively up-regulates the expression of VEGF189 in decidual cells (32Ancelin M. Buteau-Lozano H. Meduri G. Osborne-Pellegrin M. Sordello S. Plouet J. Perrot-Applanat M. Proc. Natl. Acad. Sci. U. S. A. 2002; 99: 6023-6028Crossref PubMed Scopus (109) Google Scholar).To conclude, we have characterized a new, secreted, biologically active VEGF splice form. Whether this VEGF splice form has a biological role distinct from that of other VEGF forms is unclear at the moment, and it is not known whether there exist specific mechanisms that regulate the synthesis of the VEGF162 mRNA. Tools that allow easy discrimination between the expression patterns of VEGF162and the other splice forms will have to be developed to study thein vivo expression patterns of VEGF162. The elucidation of these questions, as well as the design of experiments aimed at the identification of the specific biological roles of VEGF162, will most likely be the focus for the continuation of the research in the near future. VEGF206 was identified as a nonsecreted VEGF form in which alternative splicing in the region of exon 6 of the VEGF mRNA resulted in a 41-amino acid insertion relative to the more widely distributed VEGF165. The insertion included the 24 amino acids found in VEGF189 (2Houck K.A. Ferrara N. Winer J. Cachianes G. Li B. Leung D.W. Mol. Endocrinol. 1991; 5: 1806-1814Crossref PubMed Scopus (1229) Google Scholar) and 17 additional amino acids. These two different portions of exon 6 have become known as exons 6A and 6B. VEGF206 is a relatively rare form of VEGF that is not released into the conditioned medium of producing cells and is therefore difficult to study. The presence of exon 6B in VEGF206 indicated that exon 6B may be present in additional forms and that the properties of such forms may perhaps be studied more readily. We therefore set out to look for such forms. We have found that A431 squamous carcinoma cells contain mRNA encoding an additional exon 6B-containing VEGF form. This form is identical to VEGF145 (4Poltorak Z. Cohen T. Sivan R. Kandelis Y. Spira G. Vlodavsky I. Keshet E. Neufeld G. J. Biol. Chem. 1997; 272: 7151-7158Abstract Full Text Full Text PDF PubMed Scopus (446) Google Scholar), except for the added amino acids encoded by exon 6B that lead to the production of a 162-amino acid-long VEGF form. This form may have been wrongly identified as VEGF165 because the mass of VEGF162 and the size of the mRNA encoding VEGF162 closely resemble VEGF165. Recombinant VEGF162 was secreted from CHO DHFR− cells and was biologically active as determined by endothelial cell proliferation assays and by its ability to induce angiogenesis in vivo. In these properties, it does not appear to be significantly different from VEGF145. VEGF162 bound to a native basement membrane produced by corneal endothelial cells. It bound to the basement membrane less efficiently than VEGF145 but better than VEGF165, which binds to such matrices relatively inefficiently, despite the substantial affinity that it displays toward heparin (4Poltorak Z. Cohen T. Sivan R. Kandelis Y. Spira G. Vlodavsky I. Keshet E. Neufeld G. J. Biol. Chem. 1997; 272: 7151-7158Abstract Full Text Full Text PDF PubMed Scopus (446) Google Scholar, 7Park J.E. Keller G.A. Ferrara N. Mol. Biol. Cell. 1993; 4: 1317-1326Crossref PubMed Scopus (956) Google Scholar). The synthetic VEGF form VEGF138, which contains exon 6B but lacks exon 6A, was not able to bind to this basement membrane, behaving very similarly in this respect to VEGF121 (4Poltorak Z. Cohen T. Sivan R. Kandelis Y. Spira G. Vlodavsky I. Keshet E. Neufeld G. J. Biol. Chem. 1997; 272: 7151-7158Abstract Full Text Full Text PDF PubMed Scopus (446) Google Scholar). It can therefore be concluded that exon 6B does not contribute to VEGF binding to the extracellular matrix. Rather, exon 6B seems to interfere with the interaction of exon 6A with basement membrane components, leading to a decreased extracellular matrix binding ability as compared with VEGF145. However, the interaction of VEGF162 with the basement membrane is still somewhat stronger than that of VEGF165. The evidence gathered in the past decade indicates that apparently insignificant differences in the properties of the VEGF splice forms have turned out to be biologically meaningful. Mice expressing only VEGF120 or VEGF188 develop abnormally, even if these VEGF forms are expressed at levels comparable with the expression levels of all VEGF forms put together (17Stalmans I. Ng Y.S. Rohan R. Fruttiger M. Bouche A. Yuce A. Fujisawa H. Hermans B. Shani M. Jansen S. Hicklin D. Anderson D.J. Gardiner T. Hammes H.P. Moons L. Dewerchin M. Collen D. Carmeliet P. D'Amore P.A. J. Clin. Invest. 2002; 109: 327-336Crossref PubMed Scopus (421) Google Scholar, 18Carmeliet P. Ng Y.S. Nuyens D. Theilmeier G. Brusselmans K. Cornelissen I. Ehler E. Kakkar V.V. Stalmans I. Mattot V. Perriard J.C. Dewerchin M. Flameng W. Nagy A. Lupu F. Moons L. Collen D. D'Amore P.A. Shima D.T. Nat. Med. 1999; 5: 495-502Crossref PubMed Scopus (557) Google Scholar, 19Ruhrberg C. Gerhardt H. Golding M. Watson R. Ioannidou S. Fujisawa H. Betsholtz C. Shima D.T. Genes Dev. 2002; 16: 2684-2698Crossref PubMed Scopus (705) Google Scholar, 30Mattot V. Moons L. Lupu F. Chernavvsky D. Gomez R.A. Collen D. Carmeliet P. J. Am. Soc. Nephrol. 2002; 13: 1548-1560Crossref PubMed Scopus (88) Google Scholar). These changes are the result of differential affinities to heparan sulfate proteoglycans and extracellular matrix components, differential recognition of VEGF receptors, and differential susceptibility to reactive oxygen species (8Neufeld G. Cohen T. Gengrinovitch S. Poltorak Z. FASEB J. 1999; 13: 9-22Crossref PubMed Scopus (3128) Google Scholar, 14Gitay-Goren H. Cohen T. Tessler S. Soker S. Gengrinovitch S. Rockwell P. Klagsbrun M. Levi B.-Z. Neufeld G. J. Biol. Chem. 1996; 271: 5519-5523Abstract Full Text Full Text PDF PubMed Scopus (171) Google Scholar, 15Soker S. Takashima S. Miao H.Q. Neufeld G. Klagsbrun M. Cell. 1998; 92: 735-745Abstract Full Text Full Text PDF PubMed Scopus (2057) Google Scholar, 16Gluzman-Poltorak Z. Cohen T. Herzog Y. Neufeld G. J. Biol. Chem. 2000; 275: 18040-18045Abstract Full Text Full Text PDF PubMed Scopus (306) Google Scholar, 31Gengrinovitch S. Berman B. David G. Witte L. Neufeld G. Ron D. J. Biol. Chem. 1999; 274: 10816-10822Abstract Full Text Full Text PDF PubMed Scopus (154) Google Scholar). These differences imply that certain forms of VEGF may act more efficiently than other forms in specific microenvironments and suggest that the differential splicing of VEGF may be more tightly regulated than is currently appreciated. Some evidence supporting such tight regulation is already available. For example, it was found that progesterone selectively up-regulates the expression of VEGF189 in decidual cells (32Ancelin M. Buteau-Lozano H. Meduri G. Osborne-Pellegrin M. Sordello S. Plouet J. Perrot-Applanat M. Proc. Natl. Acad. Sci. U. S. A. 2002; 99: 6023-6028Crossref PubMed Scopus (109) Google Scholar). To conclude, we have characterized a new, secreted, biologically active VEGF splice form. Whether this VEGF splice form has a biological role distinct from that of other VEGF forms is unclear at the moment, and it is not known whether there exist specific mechanisms that regulate the synthesis of the VEGF162 mRNA. Tools that allow easy discrimination between the expression patterns of VEGF162and the other splice forms will have to be developed to study thein vivo expression patterns of VEGF162. The elucidation of these questions, as well as the design of experiments aimed at the identification of the specific biological roles of VEGF162, will most likely be the focus for the continuation of the research in the near future. We thank Dr. Ofra Kessler for critical comments and excellent technical tips. We thank Dr. Israel Vlodavsky for invaluable help with basement membrane binding experiments.
Referência(s)