Characterization of Vascular Endothelial Cell Growth Factor Interactions with the Kinase Insert Domain-containing Receptor Tyrosine Kinase
1999; Elsevier BV; Volume: 274; Issue: 26 Linguagem: Inglês
10.1074/jbc.274.26.18421
ISSN1083-351X
AutoresSonia Cunningham, Tuan M. Tran, Maria P. Arrate, Tommy A. Brock,
Tópico(s)Coagulation, Bradykinin, Polyphosphates, and Angioedema
ResumoThe kinase insert domain-containing receptor (KDR) tyrosine kinase mediates calcium mobilization in endothelial cells and plays a key role during physiological and pathological angiogenesis. To provide a detailed understanding of how KDR is activated, we analyzed the kinetics of ligand-receptor interaction using BIAcore. Both predimerized (KDR-Fc) and monomeric (KDR-cbu) receptors were examined with vascular endothelial cell growth factor (VEGF) homodimers and VEGF/placental growth factor (PlGF) heterodimers. VEGF binds to KDR-Fc with ka = 3.6 ± 0.07e6, kd = 1.34 ± 0.19e−4, and KD = 37.1 ± 4.9 pm. These values are similar to those displayed by monomeric KDR where ka = 5.23 ± 1.4e6, kd = 2.74 ± 0.76e−4, and KD = 51.7 ± 5.8 pm were apparent. In contrast, VEGF/PlGF bound to KDR-Fc with ka = 7.3 ± 1.6e4,kd = 4.4 ± 1.2e−4, andKD = 6 ± 1.2 nm. Thus, the heterodimer displays a 160-fold reduced KD for binding to predimerized KDR, which is mainly a consequence of a 50-fold reduction in ka. We were unable to detect association between VEGF/PlGF and monomeric KDR. However, nanomolar concentrations of VEGF/PlGF were able to elicit weak calcium mobilization in endothelial cells. This latter observation may indicate partial predimerization of KDR on the cell surface or facilitation of binding due to accessory receptors. The kinase insert domain-containing receptor (KDR) tyrosine kinase mediates calcium mobilization in endothelial cells and plays a key role during physiological and pathological angiogenesis. To provide a detailed understanding of how KDR is activated, we analyzed the kinetics of ligand-receptor interaction using BIAcore. Both predimerized (KDR-Fc) and monomeric (KDR-cbu) receptors were examined with vascular endothelial cell growth factor (VEGF) homodimers and VEGF/placental growth factor (PlGF) heterodimers. VEGF binds to KDR-Fc with ka = 3.6 ± 0.07e6, kd = 1.34 ± 0.19e−4, and KD = 37.1 ± 4.9 pm. These values are similar to those displayed by monomeric KDR where ka = 5.23 ± 1.4e6, kd = 2.74 ± 0.76e−4, and KD = 51.7 ± 5.8 pm were apparent. In contrast, VEGF/PlGF bound to KDR-Fc with ka = 7.3 ± 1.6e4,kd = 4.4 ± 1.2e−4, andKD = 6 ± 1.2 nm. Thus, the heterodimer displays a 160-fold reduced KD for binding to predimerized KDR, which is mainly a consequence of a 50-fold reduction in ka. We were unable to detect association between VEGF/PlGF and monomeric KDR. However, nanomolar concentrations of VEGF/PlGF were able to elicit weak calcium mobilization in endothelial cells. This latter observation may indicate partial predimerization of KDR on the cell surface or facilitation of binding due to accessory receptors. Vascular endothelial cell growth factor (VEGF) 1The abbreviations used are: VEGF, vascular endothelial cell growth factor; PlGF, placental growth factor; KDR, kinase insert domain-containing receptor; RU, response unit(s); HUVEC, human umbilical vein endothelial cell.exerts its effects on the vasculature by stimulating mitogenesis of endothelial cells and altering vessel permeability (1Thomas K.A. J. Biol. Chem. 1996; 271: 603-606Abstract Full Text Full Text PDF PubMed Scopus (569) Google Scholar, 2Ferrara N. Houck K. Jakeman L. Leung D.W. Endocr. Rev. 1992; 13: 18-32Crossref PubMed Scopus (1559) Google Scholar). It belongs to a superfamily of growth factors that possess a cystine knot motif (3Sun P.D. Davis D.R. Annu. Rev. Biophys. Biomol. Struct. 1995; 24: 269-291Crossref PubMed Google Scholar). The expression levels of VEGF in the embryo are critical for development: even loss of a single VEGF allele is lethal in the mouse embryo and results in abnormal blood vessel development (4Ferrara N. Carver-Moore K. Chen H. Dowd M. Lu L. O'Shea K.S. Powell-Braxton L. Hillan K.J. Moore M.W. Nature. 1996; 380: 439-442Crossref PubMed Scopus (3061) Google Scholar, 5Carmeliet P. Ferreira V. Breier G. Pollefeyt S. Kieckens L. Gertsenstein M. Fahrig M. Vandenhoeck A. Harpal K. Eberhardt C. Declercq C. Pawling J. Moons L. Collen D. Risau W. Nagy A. Nature. 1996; 380: 435-439Crossref PubMed Scopus (3475) Google Scholar). VEGF binds with high affinity to the receptor tyrosine kinases Flt-1 and KDR/(Flk-1), which are localized predominantly, but not exclusively, to endothelial cells (1Thomas K.A. J. Biol. Chem. 1996; 271: 603-606Abstract Full Text Full Text PDF PubMed Scopus (569) Google Scholar). Currently, six isoforms of VEGF have been identified that are produced by alternative splicing, namely VEGF115, VEGF121, VEGF145, VEGF165, VEGF189, and VEGF206 (2Ferrara N. Houck K. Jakeman L. Leung D.W. Endocr. Rev. 1992; 13: 18-32Crossref PubMed Scopus (1559) Google Scholar, 6Poltorak 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 (455) Google Scholar, 7Sugihara T. Wadhwa R. Kaul S.C. Mitsui Y. J. Biol. Chem. 1998; 273: 3033-3038Abstract Full Text Full Text PDF PubMed Scopus (43) Google Scholar). The latter four species are heparin binding growth factors. The crystal structure of VEGF reveals a symmetrical antiparallel homodimer, which presents KDR/Flt-1 binding determinants at both poles of the dimer (8Muller Y.A. Li B. Christinger H.W. Wells J.A. Cunningham B.C. de Vos A.M. Proc. Natl. Acad. Sci. U. S. A. 1997; 94: 7192-7197Crossref PubMed Scopus (368) Google Scholar). Placental growth factor (PlGF), another member of the family, is portrayed by high affinity binding to Flt-1 but not KDR (9Kendall R.L. Wang G. DiSalvo J. Thomas K. Biochem. Biophys. Res. Commun. 1994; 201: 326-330Crossref PubMed Scopus (75) Google Scholar, 10Park 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, 11Sawano A. Takahashi T. Yamaguchi S. Aonuma M. Shibuya M. Cell Growth Diff. 1996; 7: 213-221PubMed Google Scholar). However, heterodimers formed between VEGF and PlGF allow interaction with both receptors (12Coa Y. Chen H. Zhou L. Chiang M.-K. Anand-Apte B. Weatherbee J.A. Wang Y. Fang F. Flanagan J.G. Tsang M.L.-S. J. Biol. Chem. 1996; 271: 3154-3162Crossref PubMed Scopus (259) Google Scholar). The KDR and Flt-1 receptor tyrosine kinases belong to the platelet-derived growth factor receptor family and are characterized by the presence of seven immunoglobulin-like folds in their extracellular domains (13De Vries C. Escobedo J.A. Ueno H. Houck K. Ferrara N. Williams T. Science. 1992; 255: 989-991Crossref PubMed Scopus (1895) Google Scholar, 14Terman B.I. Dougher-Vermazen M. Carrion M.E. Dimitrov D. Armellino D.C. Gospodarowicz D. Bohlen P. Biochem. Biophys. Res. Commun. 1992; 187: 1579-1586Crossref PubMed Scopus (1405) Google Scholar). Targeted disruption of each receptor results in a distinct lethal phenotype. While Flk-1 is essential for development of hematopoietic and endothelial cells, Flt-1 deficiency results in a disorganized vasculature (15Shalaby F. Rossant J. Yamaguchi T.P. Gertsenstein M. Wu X.-F. Breitman M.L. Schuh A.C. Nature. 1995; 376: 62-66Crossref PubMed Scopus (3370) Google Scholar, 16Fong G.-H. Rossant J. Gertsenstein M. Breitman M.L. Nature. 1995; 376: 66-70Crossref PubMed Scopus (2224) Google Scholar). Several reports have demonstrated the importance of the second immunoglobulin domain of Flt-1 for VEGF interactions (17Davis-Smyth T. Chen H. Park J. Presta L.G. Ferrara N. EMBO J. 1996; 15: 4919-4927Crossref PubMed Scopus (181) Google Scholar, 18Cunningham S.A. Stephan C.C. Arrate M.P. Ayer K.G. Brock T.A. Biochem. Biophys. Res. Commun. 1997; 231: 596-599Crossref PubMed Scopus (30) Google Scholar, 19Barleon B. Totzke F. Herzog C. Blanke S. Kremmer E. Siemeister G. Marme D. Martiny-Baron G. J. Biol. Chem. 1997; 272: 10382-10388Abstract Full Text Full Text PDF PubMed Scopus (187) Google Scholar, 20Tanaka K. Yamaguchi S. Sawano A. Shibuya M. Jpn. J. Cancer Res. 1997; 88: 867-876Crossref PubMed Scopus (58) Google Scholar). Further, the co-crystal structure of VEGF with domain 2 of Flt-1 has revealed the receptor amino acids that directly contact VEGF (21Wiesmann C. Fuh G. Christinger H.W. Eigenbrot C. Wells J.A. de Vos A.M. Cell. 1997; 91: 695-704Abstract Full Text Full Text PDF PubMed Scopus (417) Google Scholar). For KDR, crucial VEGF binding determinants extend through immunoglobulin-like domain 3 (22Fuh G. Li B. Crowley C. Cunningham B. Wells J.A. J. Biol. Chem. 1998; 273: 11197-11204Abstract Full Text Full Text PDF PubMed Scopus (238) Google Scholar, 23Kaplan J.B. Sridharan L. Zaccardi J.A. Dougher-Vermazen M. Terman B.I. Growth Factors. 1997; 14: 243-256Crossref PubMed Scopus (28) Google Scholar). Although KDR binds VEGF with lower affinity than Flt-1, it appears to be the main receptor through which VEGF mediates mitogenic signaling (14Terman B.I. Dougher-Vermazen M. Carrion M.E. Dimitrov D. Armellino D.C. Gospodarowicz D. Bohlen P. Biochem. Biophys. Res. Commun. 1992; 187: 1579-1586Crossref PubMed Scopus (1405) Google Scholar, 24Waltenberger J. Claesson-Welsh L. Siegbahn A. Shibuya M. Heldin C.-H. J. Biol. Chem. 1994; 269: 26988-26995Abstract Full Text PDF PubMed Google Scholar). This is further demonstrated in vivo where endothelial expression of a dominant-negative Flk-1 mutant inhibits tumor angiogenesis (25Millauer B. Shawver L.K. Plate K.H. Risau W. Ullrich A. Nature. 1994; 367: 576-579Crossref PubMed Scopus (1162) Google Scholar). Angiogenesis is the process whereby new blood vessels develop from the existing vasculature (26Folkman J. Shing Y. J. Biol. Chem. 1992; 267: 10931-10934Abstract Full Text PDF PubMed Google Scholar). In the adult, physiological angiogenesis is limited but occurs for example during corpus luteum development and wound healing. However, pathological angiogenesis is a component of many diseases such as solid tumor growth, retinopathy, and rheumatoid arthritis (27Folkman J. Nat. Med. 1995; 1: 27-31Crossref PubMed Scopus (7235) Google Scholar). In these situations VEGF expression is up-regulated. Thus, in recent years, much effort has been placed on identifying inhibitors of the VEGF/receptor interaction. A compound that inhibits the association of VEGF with its receptor would conceivably be more effective than one that enhances dissociation. In order to gain a more detailed profile of VEGF binding to its receptors, we have initiated a real time kinetic study using the BIAcore. We chose to concentrate on the KDR receptor due to the key role that it plays in endothelial cell mitogenesis and performed a focused study with VEGF and VEGF/PlGF. Our work reports a detailed kinetic analysis for both monomeric and dimeric forms of KDR with these ligands using surface plasmon resonance. Ligand/receptor interactions were performed on a BIAcore 2000 (BIAcore, Piscataway, NJ). Protein A or calmodulin was cross-linked to the dextran surface of a CM5 sensor chip. Protein A was immobilized using amine coupling with 1-ethyl-3-[3-dimethylaminopropyl]carbodiimide hydrochloride andN-hydrosuccinimide to a density of 1000–2000 response units (RU). Calmodulin was immobilized using thiol coupling chemistry to a similar density. The hetero-bifunctional cross-linker Sulph-m-Maleimidobenzoyl-N-hydoxysuccinimide ester was attached to the dextran surface through ethylenediamine following 1-ethyl-3-[3-dimethylaminopropyl]carbodiimide hydrochloride andN-hydrosuccinimide activation. A mutant calmodulin with a free cysteine residue at position 18 (a gift from Dr. J. Putkey) was subsequently cross-linked to the surface. Recombinant receptor extracellular domains, fused to either the constant region of the mouse heavy chain IgG2a (KDR-Fc) or possessing a calmodulin binding tag (KDR-cbu), were captured to approximately 100 RU. For KDR-Fc, interaction of VEGF homodimer and VEGF/PlGF heterodimers was achieved at concentrations between 0.625 and 70 nm in HEPES-buffered saline consisting of: 10 mm HEPES, pH 7.4, 150 mm NaCl, 3.4 mm EDTA, 0.005% surfactant P20. For association, ligand was applied using the kinject program at 20–60 μl/min for 3 min. Dissociation was effected with HEPES-buffered saline applied over the surface at 60 μl/min for 10 min. When KDR-cbu was utilized, it was necessary to perform the experiments in a calcium-containing buffer to maintain capture of the receptor to the calmodulin surface. Experiments were conducted in 20 mm Tris, pH 7.5, 150 mmNaCl, 1.5 mm CaCl2, 0.005% Tween 20. The receptor/ligand complex was stripped from the protein A surface using 10 mm glycine, 200 mm NaCl, pH 1.8, and from the calmodulin surface using 20 mm Tris, pH 7.5, 150 mm NaCl, 1 mm EGTA, 0.005% Tween 20. For both protocols, it was necessary to follow with a brief pulse of 0.025% SDS to obtain full regeneration. The chip was allowed to stabilize for at least 10 min prior to subsequent use. Sensorgrams were subjected to global analysis using BIAevaluation software 3.0. Global fitting analyzes association and dissociation data for all VEGF concentrations simultaneously, and by doing so allows a more accurate description ofka/kd. A simple 1:1 Langmuir model was employed to fit the data, where A +B [dharrow] AB; A = analyte (VEGF), B = ligand (KDR), AB = complex;ka = association rate constant (m−1 s−1),kd = dissociation rate constant (s−1). Data were also analyzed with a 1:1 model, where mass transfer (kt = rate constant for mass transfer, s−1) or drift was taken into consideration. The fitting model was chosen based on the closeness of the calculated fit to the actual data (sensorgram overlay) and an examination of residual plots (data not shown). Table I averages data from different receptor preparations and CM5 chips.Table IParameter estimates for VEGF and VEGF/PlGF interactions with KDRReceptorLigandk ak dK Dn1/Ms1/sKDR(1–7)-FcVEGF3.6 ± 0.07e61.34 ± 0.19e−437.1 ± 4.9 pm3KDR(1–7)-cbuVEGF5.23 ± 1.4e62.74 ± 0.76e−451.7 ± 5.8 pm4KDR(1–7)-FcVEGF/PlGF7.3 ± 1.6e64.4 ± 1.15e−46.0 ± 1.2 nm4KDR(1–7)-cbuVEGF/PlGF0003KDR(1–3)-FcVEGF4.72 ± 1.0e60.67 ± 0.11e−414.5 ± 1.0 pm3Data were globally fit using BIA evaluation 3.0 software. -Fc, predimerized KDR; -cbu, monomeric KDR; (1Thomas K.A. J. Biol. Chem. 1996; 271: 603-606Abstract Full Text Full Text PDF PubMed Scopus (569) Google Scholar, 2Ferrara N. Houck K. Jakeman L. Leung D.W. Endocr. Rev. 1992; 13: 18-32Crossref PubMed Scopus (1559) Google Scholar, 3Sun P.D. Davis D.R. Annu. Rev. Biophys. Biomol. Struct. 1995; 24: 269-291Crossref PubMed Google Scholar, 4Ferrara N. Carver-Moore K. Chen H. Dowd M. Lu L. O'Shea K.S. Powell-Braxton L. Hillan K.J. Moore M.W. Nature. 1996; 380: 439-442Crossref PubMed Scopus (3061) Google Scholar, 5Carmeliet P. Ferreira V. Breier G. Pollefeyt S. Kieckens L. Gertsenstein M. Fahrig M. Vandenhoeck A. Harpal K. Eberhardt C. Declercq C. Pawling J. Moons L. Collen D. Risau W. Nagy A. Nature. 1996; 380: 435-439Crossref PubMed Scopus (3475) Google Scholar, 6Poltorak 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 (455) Google Scholar, 7Sugihara T. Wadhwa R. Kaul S.C. Mitsui Y. J. Biol. Chem. 1998; 273: 3033-3038Abstract Full Text Full Text PDF PubMed Scopus (43) Google Scholar), full extracellular domain of KDR; (1Thomas K.A. J. Biol. Chem. 1996; 271: 603-606Abstract Full Text Full Text PDF PubMed Scopus (569) Google Scholar, 2Ferrara N. Houck K. Jakeman L. Leung D.W. Endocr. Rev. 1992; 13: 18-32Crossref PubMed Scopus (1559) Google Scholar, 3Sun P.D. Davis D.R. Annu. Rev. Biophys. Biomol. Struct. 1995; 24: 269-291Crossref PubMed Google Scholar), first three immunoglobulin-like domains of KDR. Open table in a new tab Data were globally fit using BIA evaluation 3.0 software. -Fc, predimerized KDR; -cbu, monomeric KDR; (1Thomas K.A. J. Biol. Chem. 1996; 271: 603-606Abstract Full Text Full Text PDF PubMed Scopus (569) Google Scholar, 2Ferrara N. Houck K. Jakeman L. Leung D.W. Endocr. Rev. 1992; 13: 18-32Crossref PubMed Scopus (1559) Google Scholar, 3Sun P.D. Davis D.R. Annu. Rev. Biophys. Biomol. Struct. 1995; 24: 269-291Crossref PubMed Google Scholar, 4Ferrara N. Carver-Moore K. Chen H. Dowd M. Lu L. O'Shea K.S. Powell-Braxton L. Hillan K.J. Moore M.W. Nature. 1996; 380: 439-442Crossref PubMed Scopus (3061) Google Scholar, 5Carmeliet P. Ferreira V. Breier G. Pollefeyt S. Kieckens L. Gertsenstein M. Fahrig M. Vandenhoeck A. Harpal K. Eberhardt C. Declercq C. Pawling J. Moons L. Collen D. Risau W. Nagy A. Nature. 1996; 380: 435-439Crossref PubMed Scopus (3475) Google Scholar, 6Poltorak 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 (455) Google Scholar, 7Sugihara T. Wadhwa R. Kaul S.C. Mitsui Y. J. Biol. Chem. 1998; 273: 3033-3038Abstract Full Text Full Text PDF PubMed Scopus (43) Google Scholar), full extracellular domain of KDR; (1Thomas K.A. J. Biol. Chem. 1996; 271: 603-606Abstract Full Text Full Text PDF PubMed Scopus (569) Google Scholar, 2Ferrara N. Houck K. Jakeman L. Leung D.W. Endocr. Rev. 1992; 13: 18-32Crossref PubMed Scopus (1559) Google Scholar, 3Sun P.D. Davis D.R. Annu. Rev. Biophys. Biomol. Struct. 1995; 24: 269-291Crossref PubMed Google Scholar), first three immunoglobulin-like domains of KDR. For dimeric secreted receptors, the KDR extracellular domain was subcloned into the pBacPak9 vector, which contained the constant region of murine IgG2A as described previously (18Cunningham S.A. Stephan C.C. Arrate M.P. Ayer K.G. Brock T.A. Biochem. Biophys. Res. Commun. 1997; 231: 596-599Crossref PubMed Scopus (30) Google Scholar, 28Cunningham S.A. Tran T.M. Arrate M.P. Bjercke R. Brock T.A. Am. J. Physiol. 1999; 276: 176-181Crossref PubMed Google Scholar). Briefly, the KDR extracellular domain was modified by polymerase chain reaction to generate a KpnI site immediately prior to the transmembrane sequence. This was accomplished by amplifying a 181-base pair segment located at the C terminus of the extracellular domain using sense 5′- TGTAGAATTCTCAGGCATTGTA -3′ and antisense primers 5′ - GAATGGTACCTTCCAAGTTCGTCTTTTC-3′. The product was digested at an internal SphI site and at the introduced KpnI site, and ligated to the remaining extracellular domain (KDR was a previous gift from Dr. B. Terman). Recombinant dimeric fusion protein (KDR-Fc) was expressed in Sf21 cells and purified from the culture media using standard protein A affinity chromatography. For monomeric secreted receptors, a calmodulin-binding tag was subcloned into the KpnI site of pFastBac1 using the following annealed primers: sense, 5′- CAAGCGCCGCTGGAAGAAGAACTTTATCGCCGTGAGCGCCGCCAACCGCTTCAAGAAGATCAGCTCCAGCGGCGCCCTGCGTAC-3′; and antisense, 5′- GCAGGGCGCCGCTGGAGCTGATCTTCTTGAAGCGGTTGGCGGCGCTCACGGCGATGAAGTTCTTCTTCCAGCGGCGCTTGGTAC-3′. This resulted in deletion of the KpnI site at the 3′ end of the tag. The full KDR extracellular domain was subcloned from the pBacPak9 vector using BamHI and KpnI and ligated upstream of the tag. Recombinant monomeric fusion protein (KDR-cbu) was expressed in Sf21 cells and purified from the culture media using calmodulin affinity resin (Stratagene). The protein was eluted from the column in calcium-free buffer, using a NaCl gradient from 0 to 300 mm. KDR-cbu fusion eluted at approximately 150 mm NaCl. HUVECs were detached from monolayers and loaded with 2 μm FURA-2/AM for 30 min at 37 °C in phosphate-buffered saline. Cells were washed and resuspended at a final count of 1.0 × 106cells/ml. For inhibition with anti-KDR antibody, polyclonal serum at 1:100 was preincubated with the cells prior to stimulation (28Cunningham S.A. Tran T.M. Arrate M.P. Bjercke R. Brock T.A. Am. J. Physiol. 1999; 276: 176-181Crossref PubMed Google Scholar). Fluorescence was monitored at 340 and 380 nm and [Ca2+]i estimated as described previously (28Cunningham S.A. Tran T.M. Arrate M.P. Bjercke R. Brock T.A. Am. J. Physiol. 1999; 276: 176-181Crossref PubMed Google Scholar). For each experiment, a base-line resting fluorescence was measured for 100 s prior to stimulation with 0.05 nmVEGF165 or 18 nm VEGF/PlGF heterodimer. The kinetic experiments were performed on a BIAcore 2000, which has four flow cells allowing simultaneous measurements. This allows accurate monitoring of nonspecific binding to a control surface throughout the association/dissociation phases of ligand/receptor interaction. Amine coupling is the standard procedure used to achieve immobilization of ligand or receptors to the sensor chip surface. We specifically chose not to immobilize the VEGF ligand nor the KDR receptor directly to the surface in order to avoid modifying residues directly involved in the KDR/VEGF interaction. Our approach was to capture the KDR receptor in a manner that would allow VEGF full access to the binding interface. The experimental set-up for the series of studies conducted in this report is shown in Fig.1. The first sequence of experiments was performed on the KDR-Fc fusion protein, which presents a covalently dimerized receptor surface to the VEGF ligand. Dimerization occurs through a disulfide bond located in the hinge of the constant region of mouse IgG. Similar receptor constructs have been utilized extensively in the literature to assess VEGF interactions. Initial experiments were performed using a rabbit anti-mouse antibody as capture for the KDR-Fc fusion protein. This former was covalently cross-linked to the carboxymethylated dextran matrix of the CM5 sensor chips. However, an upward drift in the sensorgram following receptor interaction complicated the kinetic analysis. The use of protein A as capture eliminated this trend and was used for all subsequent experiments. Due to the very high affinity of VEGF for its receptor, KDR-Fc fusion was captured at low density, i.e.approximately 100 RU. This minimized mass transport effects due to VEGF165 rebinding following its dissociation from the receptor. To further reduce mass transport effects, VEGF165ligand was applied at a relatively fast flow rate of 60 μl/min. This ensured that binding was not limited due to inefficient delivery of VEGF to the receptor surface (29Fivash M. Towler E.M. Fisher R.J. Curr. Opin. Bio/Technol. 1998; 9: 97-101Crossref PubMed Scopus (172) Google Scholar). Each concentration of VEGF165 was passed simultaneously over two flow cells, one in the presence and one in the absence of captured receptor. This allowed the subtraction of nonspecific binding effects and bulk refractive index. The sensorgrams shown reflect the data subtracted from the control flow cell and shows the closeness of the calculated fit. Fig. 2 shows a typical sensorgram overlay for varying VEGF concentrations. Since the off rate is extremely slow, dissociation was recorded over an extended period of 10 min to improve the accuracy of kd determination. Global analysis was performed on the data using BIAevaluation 3.0 software (30BIAcore AB BIAevaluation, Version 3. BIAcore AB, Piscataway, NJ1997Google Scholar). Gel filtration analysis of VEGF binding to KDR-Fc fusions shows that one dimeric ligand binds one dimeric receptor even in the presence of excess VEGF (22Fuh G. Li B. Crowley C. Cunningham B. Wells J.A. J. Biol. Chem. 1998; 273: 11197-11204Abstract Full Text Full Text PDF PubMed Scopus (238) Google Scholar). In keeping with this study, our data points fit well to a simple 1:1 (Langmuir) binding model. Further, the apparent binding stoichiometry = (RU VEGF/RU KDR-Fc) × (M r KDR-Fc/M r VEGF), where RU VEGF represents the level of VEGF bound to the sensor chip surface at saturation, and RU KDR, the level of immobilized receptor. The precise amount of KDR captured on the surface varied 20–30% between experiments. However, in generalizing for KDR(1–7)-Fc, stoichiometry 64 (20/100) × (220/42), i.e.approximately 1:1. Due to experimental optimization, it was not necessary to include mass transport into the calculation. Evaluation of the data with either a 1:1 or 1:1 + mass transfer model resulted in calculation of similar association and dissociation rate constants (data not shown). We evaluated the association rate constant,ka at 3.6 ± 0.07e6 (1/Ms) and the dissociation rate constant kd at 1.34 ± 0.19 e−4 (1/s) (Table I). The dissociation constant (KD) averages at 37.1 pm, which falls within the range of values published in numerous reports using radiolabeled VEGF165 binding to recombinant receptors in vitro. Having defined the conditions required to measure the kinetic parameters, we wished to address whether a monomeric receptor would bind VEGF with the same affinity as the dimerized species. For this purpose, we made recombinant KDR extracellular domain without the Fc tag. For affinity purification purposes, a calmodulin binding peptide tag was added to the C terminus to generate KDR-cbu. Fig.3 shows a representative experiment. Analysis of the data shows that the interactions of VEGF with the receptor monomer are very similar to those seen with a dimeric receptor. Over four experiments, ka averages at 5.23 ± 1.4e6 (1/Ms) and the kdat 2.74 ± 0.76e−4 (1/s); theKD = 51.7 pm (Table I). Since we are working at the lower limits of BIAcore sensitivity, we refrain from performing a statistical analysis on these numbers in order not to over interpret small differences. Our results are consistent with the model for VEGF/receptor interactions presented by the co-crystal structure (21Wiesmann C. Fuh G. Christinger H.W. Eigenbrot C. Wells J.A. de Vos A.M. Cell. 1997; 91: 695-704Abstract Full Text Full Text PDF PubMed Scopus (417) Google Scholar): the contact sites between each receptor monomer with either pole of the VEGF head to tail dimer, are identical. If one VEGF dimer bound to each KDR-cbu captured on the sensor chip surface, then some 50 RU of VEGF would be expected to bind at saturation (KDR-cbu mass = 92 kDa). We were unable to achieve this level. This may be explained by the ability of a VEGF dimer to bind to two receptor monomers and/or the presence of some inactive receptors. It has been reported that heterodimers of VEGF with PlGF can be cross-linked to KDR expressed as a soluble extracellular domain (12Coa Y. Chen H. Zhou L. Chiang M.-K. Anand-Apte B. Weatherbee J.A. Wang Y. Fang F. Flanagan J.G. Tsang M.L.-S. J. Biol. Chem. 1996; 271: 3154-3162Crossref PubMed Scopus (259) Google Scholar). In order to study the kinetics of this interaction, we had to considerably reduce the flow rate and increase the heterodimer concentrations relative to the studies conducted with VEGF. Fig.4 depicts a representative sensorgram. A downward drift prior to VEGF addition is apparent. This decaying surface, due to receptor dissociation from the protein A surface, occurred sporadically throughout the study. It is likely related to the number of regenerations the surface has been exposed to. Nevertheless, the data can be easily fit with a 1:1 Langmuir model that incorporates drift into the analysis (provided in BIAevaluation 3.0 software). TableI shows that the ka is considerably reduced and the kd is somewhat enhanced. Global analysis of the data reveals a ka of 7.3 ± 1.6e4 (1/Ms) and kd of 4.4 ± 1.2e−4 (1/s); KD = 6 ±1.2 nm. Thus, the nanomolar KD is described by a 50-fold reduction in the rate of association with only an approximate 3-fold increase in the dissociation rate. Surprisingly, under similar conditions to those described above, we were unable to obtain an interaction of the heterodimer with the KDR monomer (Table I, sensorgram not shown). Neither, increasing the VEGF/PlGF concentration 4-fold, nor reducing the flow rate to 5 μl/min for 60 min, nor increasing receptor capture density 10-fold resulted in a detectable interaction (data not shown). This would suggest that the heterodimer would be unable to activate KDR receptors on the cell surface. In order to address this possibility, we performed calcium mobilization studies on HUVECs with VEGF homodimers and VEGF/PlGF heterodimers. Fig. 5(panel A) shows a typical response to VEGF; following a 1-min delay, a rapid and transient rise in intracellular calcium occurs. This is almost completely inhibited by a neutralizing KDR antibody (panel B) showing that KDR is the main receptor through which calcium signals are relayed (28Cunningham S.A. Tran T.M. Arrate M.P. Bjercke R. Brock T.A. Am. J. Physiol. 1999; 276: 176-181Crossref PubMed Google Scholar). Fig. 5 (panel C) shows the very small but perceptible calcium mobilization obtained with the VEGF/PlGF heterodimer. It was necessary to stimulate the cells with a 250-fold increased concentration compared with VEGF in order to obtain a measurable response. That this weak effect was achieved through VEGF/PlGF activating the KDR receptor is shown by a lack of mobilization following cell preincubation with neutralizing KDR antibody (panel D). Based on binding to the predimerized KDR-Fc, one may predict that an equivalent calcium mobilization would occur at VEGF/PlGF concentrations 160-fold greater than VEGF,i.e. at 8 nm. This is clearly not the case. Thus, this cellular assay and the inability of KDR-cbu to interact with VEGF/PlGF suggest that the heterodimer is inefficient at promoting receptor dimerization. Most recently, several reports have described that the first three immunoglobulin-like domains of KDR are sufficient for high affinity VEGF binding (22Fuh G. Li B. Crowley C. Cunningham B. Wells J.A. J. Biol. Chem. 1998; 273: 11197-11204Abstract Full Text Full Text PDF PubMed Scopus (238) Google Scholar, 23Kaplan J.B. Sridharan L. Zaccardi J.A. Dougher-Vermazen M. Terman B.I. Growth Factors. 1997; 14: 243-256Crossref PubMed Scopus (28) Google Scholar). Further, a neutralizing antibody directed against this region is sufficient to inhibit VEGF-mediated intracellular calcium mobilization. Nevertheless, although the overallKD may be similar, the individual rate constants may vary. In order to assess this, we constructed the first t
Referência(s)