www.biochemj.org Biochem. J. (2008) 411, 515–522 (Printed in Great Britain)
515
doi:10.1042/BJ20080078
Catalytically inactive phospholipase A2 homologue binds to vascular endothelial growth factor receptor-2 via a C-terminal loop region Daisuke FUJISAWA*, Yasuo YAMAZAKI*, Bruno LOMONTE† and Takashi MORITA*1
VEGF (vascular endothelial growth factor) regulates neovascularization through binding to its receptor KDR (kinase insert domaincontaining receptor; VEGF receptor-2). We recently identified a catalytically inactive PLA2 (phospholipase A2 ) homologue (KDR-bp) in the venom of eastern cottonmouth (Agkistrodon piscivorus piscivorus) as a third KDR-binding protein, in addition to VEGF165 and tissue inhibitor of metalloproteinase-3. KDR-bp binds to the extracellular domain of KDR with a K d of 10−8 M, resulting in specific blockade of endothelial cell growth induced by VEGF165 . Inactive PLA2 homologues are widely distributed in the venoms of Viperidae snakes and are known to act as myotoxins. In the present study, we demonstrated that KDRbinding ability is a common characteristic for inactive PLA2 homologues in snake venom, but not for active PLA2 s such as neurotoxic and platelet aggregation-modulating PLA2 s. To
INTRODUCTION
VEGF (vascular endothelial growth factor)-A plays a central role in angiogenesis [1]. VEGF165 , the most abundant isoform of VEGF-A, binds two receptor tyrosine kinases, Flt-1 ( fmslike tyrosine kinase-1; VEGF receptor-1) and KDR (kinase insert domain-containing receptor; VEGF receptor-2), on the surface of vascular endothelial cells, and induces a range of several biological actions, such as proliferation and migration of endothelial cells and vascular permeability enhancement [1,2]. The VEGF family is currently growing, and five subtypes have been identified in mammals [1,3]: VEGF-A, VEGF-B, VEGF-C, VEGF-D and PlGF (placenta growth factor). VEGF family proteins have also been found in two exogenous sources: viral VEGF (also called VEGF-E) from the parapoxvirus and orf virus [4–6], and snake venom VEGF (also called VEGF-F) from several viper species [7,8]. VEGF-F, the most recently identified subtype in snake venom, selectively binds KDR and induces potent biological actions, such as hypotension, endothelial proliferation and vascular permeability, as compared with VEGF165 [7,9]. PLA2 (phospholipase A2 ) is known to be one of the major components of snake venom. It catalyses the hydrolysis of the sn-2 ester bond in glycerophospholipids in a calcium-dependent manner and releases non-esterified fatty acids and lysophospholipids [10]. Snake venoms contain several PLA2 s that exhibit unique biological activities, such as neurotoxicity, myotoxicity, cardiotoxicity, anticoagulant activity and platelet aggregationmodulating activity [11]. This diversification is thought to be the result of a unique evolutionary system, called accelerated
understand better the KDR and KDR-bp interaction, we resolved the binding region of KDR-bp using eight synthetic peptides designed based on the structure of KDR-bp. A synthetic peptide based on the structure of the C-terminal loop region of KDRbp showed high affinity for KDR, but other peptides did not, suggesting that the C-terminal loop region of KDR-bp is involved in the interaction with KDR. The results of the present study provide insight into the binding of inactive PLA2 homologues to KDR, and may also assist in the design of novel anti-KDR molecules for anti-angiogenic therapy. Key words: kinase insert domain-containing receptor (KDR), phospholipase A2 , synthetic peptide, snake venom, vascular endothelial growth factor (VEGF).
evolution, in the venom gland [12,13]. In this system, genes are efficiently substituted in exon regions rather than in introns, resulting in the acquisition of several toxins that display multiple biological functions. Some viper venoms contain catalytically inactive PLA2 homologues, in which Asp49 is substituted by lysine, serine, arginine or asparagine [14–18]. Asp49 is a calciumbinding residue and is essential for phospholipolytic activity [19]. These catalytically inactive PLA2 homologues are known to act as myotoxins [14]; however, the molecular mechanisms remain unknown. We recently identified a VEGF receptor-binding protein, designated KDR-bp (KDR-binding protein), from the venom of the Eastern Cottonmouth (Agkistrodon piscivorus piscivorus) [20]. KDR-bp is a catalytically inactive PLA2 homologue, Lys49 PLA2 , which possesses potent myotoxicity, and is the first exogenous molecule found to antagonize the VEGF receptor. KDR-bp binds to the extracellular domain of KDR with a K d of 10−8 M, but active PLA2 (Asp49 PLA2 ) from habu snake venom does not [20]. The binding to KDR is competitive with VEGF165 and it blocks the induction of endothelial cell proliferation by VEGF165 [20]. In the present study, we show that KDRbinding potential is a common characteristic for inactive PLA2 homologues. We further demonstrate that the C-terminal loop region of KDR-bp directly associates with the extracellular domain of KDR using synthetic peptides designed based on the sequence of KDR-bp. The results of the present study help to clarify the binding mode of inactive PLA2 homologues to KDR, and may be useful in the design of novel anti-KDR molecules for anti-angiogenic therapy.
Abbreviations used: KDR, kinase insert domain-containing receptor; KDR-bp, KDR-binding protein; MALDI–TOF, matrix-assisted laser-desorption ionization–time-of-flight; PLA2 , phospholipase A2 ; SPR, surface plasmon resonance; VEGF, vascular endothelial growth factor. 1 To whom correspondence should be addressed (email
[email protected]). The numbering of PLA2 s and their homologues corresponds to the numbering system proposed by Renetseder et al. [30]. Residue numbers for KDR-bp are given in parentheses. c The Authors Journal compilation c 2008 Biochemical Society
Biochemical Journal
*Department of Biochemistry, Meiji Pharmaceutical University, 2-522-1 Kiyose, Tokyo 204-8588, Japan, and †Instituto Clodomiro Picado, Facultad de Microbiolog´ıa, Universidad de Costa Rica, San Jos´e, Costa Rica
516
D. Fujisawa and others
EXPERIMENTAL
Molecular modelling
Materials
Tertiary model structures of basic protein I, basic protein II, ammodytin L, Cax-K49, ammodytoxin A, ammodytoxin B, ammodytoxin C, Tf -Asp49 PLA2 and ammodytin I2 were generated using homology modelling software Swiss-Pdb Viewer (http://www.expasy.ch/spdbv/). The crystal structures of KDR-bp (PDB code 1PPA) and myotoxin II (PDB code 1CLP) were used as templates for the generation of tertiary model structures of basic protein I, basic protein II, ammodytin L and Cax-K49. The crystal structures of Asp49 PLA2 of C. atrox (PDB code 1PP2) and Da-acidic Asp49 PLA2 (PDB code 1IJL) were used as templates for the generation of tertiary model structures of ammodytoxin A, ammodytoxin B, ammodytoxin C, Tf-Asp49 PLA2 and ammodytin I2 .
Freeze-dried venom from Agkistrodon piscivorus piscivorus, Crotalus atrox and Vipera ammodytes ammodytes was purchased from the Kentucky Reptile Zoo, Sigma–Aldrich and Latoxan respectively. Freeze-dried venom from Deinagkistrodon acutus and Trimeresurus flavoviridis was purchased from the Japan Snake Institute. The extracellular domain of KDR (immunoglobulin-like domain 1–7) was purchased from Calbiochem. Toxin purification
Snake venom PLA2 s and their homologues were isolated from the venom of A. p. piscivorus, C. atrox, D. acutus, T. flavoviridis and V. a. ammodytes by successive chromatography: gel-filtration, anion-exchange, cation-exchange or reverse-phase HPLC. Fractions containing snake venom PLA2 s and their homologues were detected by ELISA using an anti-KDR-bp antibody. First, fractions diluted with Tris-buffered saline were immobilized on a microtitre plate. After blocking with BSA, diluted anti-KDR-bp antiserum was added and incubated for 1 h at 37 ◦C. After washing with Tris-buffered saline containing 0.1 % Tween 20, peroxidase-conjugated anti-rabbit IgG antibody was added and incubated for 1 h at 37 ◦C. After washing, enzyme substrate mixture [1 mg/ml o-phenylenediamine and 0.06 % hydrogen peroxide in 0.1 M sodium citrate (pH 5.5)] was added, followed by the addition of 8 M sulfuric acid to stop the reaction. Colour development was measured at 492 nm. Identification of isolated proteins
Isolated snake venom PLA2 s and their homologues were identified by N-terminal amino acid sequence analysis after automated Edman degradation. Ammodytoxins A, B and C isolated from the venom of V. a. ammodytes were identified by N-terminal amino acid sequence analysis and MALDI–TOF (matrix-assisted laserdesorption ionization–time-of-flight) MS. The purity of isolated proteins was confirmed by SDS/PAGE. SPR (surface plasmon resonance) analysis
SPR analysis was performed using a Biacore machine. First, the carboxyl groups of carboxymethylated dextran on a CM4 sensor chip were activated by NHS (N-hydroxysuccinimide) and EDC [1-ethyl-3-(3-dimethylaminopropyl) carbodiimide hydrochloride]. Subsequently, the extracellular immunoglobulin-like domains 1–7 of KDR diluted with 10 mM sodium acetate (pH 4.5) were injected into the system at a flow rate of 10 µl/min, and were immobilized on a CM4 sensor chip. Remaining activated carboxyl groups were blocked with ethanolamine hydrochloride. Isolated proteins or peptides were diluted with 10 mM Hepes-buffered saline containing 3 mM EDTA and 0.005 % Surfactant P-20 and were injected into the system at a flow rate of 20 µl/min. Binding dissociation constants (K d ) were evaluated using BIAevaluation software. Peptide synthesis and purification
All peptides were synthesized with a peptide synthesizer using an Fmoc (fluoren-9-ylmethoxycarbonyl) strategy. After deblocking and cleavage from resin, synthesized peptides were dissolved in ultrapure water or DMSO, and purified by reverse-phase HPLC. The quality of synthetic peptides was confirmed by amino acid sequencing, MALDI–TOF MS and amino acid composition analysis. c The Authors Journal compilation c 2008 Biochemical Society
RESULTS Isolation of snake venom PLA2 s and their homologues from viper venoms
In order to examine the correlation between KDR-binding potential and myotoxicity, we isolated 15 PLA2 homologues from several snake venoms (listed in Table 1; for isolation procedures see the Experimental section): seven catalytically inactive PLA2 homologues with myotoxicity (myotoxin II, acutohaemolysin, KDR-bp, basic protein I, basic protein II, ammodytin L and Cax-K49), two catalytically active PLA2 s with myotoxicity (AppAsp49 PLA2 and Tf -Asp49 PLA2 ), a non-toxic PLA2 (ammodytin I2 ), three neurotoxic PLA2 s (ammodytoxin A, ammodytoxin B and ammodytoxin C), a heterodimeric PLA2 with neurotoxicity (vipoxin) and a platelet aggregation-inhibiting PLA2 (Da-acidic Asp49 PLA2 ). Isolated proteins were identified by N-terminal amino acid sequence analysis (see Supplementary Table 1 at http:// www.BiochemJ.org/bj/411/bj4110515add.htm), and homogeneity was judged by SDS/PAGE (Supplementary Figure 1 at http:// www.BiochemJ.org/bj/411/bj4110515add.htm). Because ammodytoxins are reported to be highly homologous proteins among isoforms, differing by only a few residues near the C-terminus, we identified them by MALDI–TOF MS with < 1.0 % experimental error (Supplementary Figure 2 and Supplementary Table 2 at http://www.BiochemJ.org/bj/411/bj4110515add.htm). Binding potential of snake venom PLA2 s and their homologues to the extracellular domain of KDR
We next investigated the KDR-binding potential of isolated snake venom PLA2 s and their homologues by SPR analysis using Biacore. All myotoxic PLA2 homologues (KDR-bp, myotoxin II, Cax-K49, acutohaemolysin, basic protein I, basic protein II and ammodytin L) bound to the immobilized extracellular domain of KDR with an essentially equal affinity (K d was 10−8 M), but PLA2 s exhibiting biological activities such as neurotoxicity, anticoagulant activity or platelet aggregation-modulating activity did not (Figure 1 and Table 1). In addition, we noted that App-Asp49 PLA2 , catalytically active PLA2 with myotoxicity, also bound to KDR, although its binding affinity was 10-fold weaker than inactive PLA2 homologues (Table 1). Identification of KDR-binding site on KDR-bp
For better understanding of the interactions between inactive PLA2 homologues and KDR, we next tried to identify the KDRbinding site in KDR-bp. Sequence alignment of snake venom PLA2 s and their homologues revealed several conserved amino
Novel high-affinity peptide bound to KDR Table 1
517
Binding properties of snake venom PLA2 homologues
Association (k ass ) and dissociation (k diss ) rate constants, and dissociation constants (K d ) of snake venom PLA2 homologues bound to immobilized KDRD1−7 . Snake venom PLA2 homologues were subjected to SPR assay in several independent experiments (n = 2–6). k ass and k diss were calculated using two-state reaction model fitting. K d was calculated from the ratio k diss /k ass . Nsb, no specific binding. *Note that myotoxicity of Tf -Asp49 PLA2 is dependent on enzymatic activity, indicating that the mechanism of myotoxicity is distinct from that of other myotoxic PLA2 homologues [31]. PLA2 homologues
k ass (M−1 · s−1 ) (× 104 )
k diss (s−1 ) (× 10−4 )
K d (M) (× 10−8 )
PLA2 activity
Bioactivity
Myotoxin II Acutohaemolysin KDR-bp Basic protein II Basic protein I Ammodytin L Cax-K49 App -Asp49PLA2 Tf -Asp49PLA2 Ammodytin I2 Ammodytoxin A Ammodytoxin B Ammodytoxin C Vipoxin Da -acidic Asp49PLA2
1.3 + − 0.0 0.9 + − 0.5 1.1 + − 0.9 2.5 + − 0.2 2.5 + − 0.5 1.2 + − 0.4 1.8 + − 1.4 1.7 + − 1.0 Nsb Nsb Nsb Nsb Nsb Nsb Nsb
0.6 + − 0.3 0.7 + − 0.7 1.1 + − 1.4 1.8 + − 1.3 3.1 + − 2.2 1.7 + − 0.9 2.7 + − 0.6 25.7 + − 24.5 Nsb Nsb Nsb Nsb Nsb Nsb Nsb
0.5 + − 0.0 0.6 + − 0.4 0.7 + − 0.5 0.7 + − 0.5 1.3 + − 1.0 1.3 + − 0.5 2.1 + − 1.2 13.3 + − 11.8 Nsb Nsb Nsb Nsb Nsb Nsb Nsb
− − − − − − − + + + + + + + +
Myotoxicity Myotoxicity Myotoxicity Myotoxicity Myotoxicity Myotoxicity Myotoxicity Myotoxicity Myotoxicity* Non-toxic Neurotoxicity Neurotoxicity Neurotoxicity Neurotoxicity Anti-platelet aggregation
Figure 1
Myotoxic PLA2 homologues bind to the extracellular domain of KDR, but other PLA2 s do not
The extracellular domain of KDR was immobilized on a CM4 sensor chip. Several concentrations (10, 20, 40, 60, 80 and 100 nM) of isolated PLA2 s and their homologues were then applied. Binding properties are summarized in Table 1.
acid residues: Lys7 and Glu13 (Lys7 and Glu12 in KDR-bp) in the N-terminal α-helix region (α1), Lys38 and Lys53 (Lys37 and Lys52 in KDR-bp) near the enzymatic active site and Lys78 and Lys80 (Lys69 and Lys71 in KDR-bp) in an anti-parallel β-sheet region
(β3 and β4) (the β-wing region; bold characters in Figure 2). In addition, the C-terminal loop region of KDR-binding PLA2 homologues is rich in basic amino acid residues (Figure 2). In the tertiary structure of KDR-bp, conserved residues in the c The Authors Journal compilation c 2008 Biochemical Society
518
Figure 2
D. Fujisawa and others
Comparison between amino acid sequences of KDR-binding PLA2 homologues and those of non-KDR-binding PLA2 s
Residue numbers are amino acid numbers of KDR-bp. Lower residue numbers correspond to those of bovine group IB PLA2 (bovine GIB). For the secondary structure of PLA2 and their homologues, α-helices and β-strands are represented as columns and arrows respectively. Cysteine residues and highly conserved residues among PLA2 s and their homologues are shown in black and grey respectively. Conserved residues among KDR-binding PLA2 homologues and basic residues in the C-terminal loop region are indicated in bold.
N-terminal α-helix region are located adjacent to the β-wing region, whereas conserved residues around the active site (Lys38 and Lys53 ) are near the C-terminal loop region (Figure 3). These conserved amino acid residues among KDR-binding PLA2 homologues are located on the surface of the KDR-bp molecule, and are predicted to form positively charged clusters (Figure 3). Therefore we speculated that the N-terminal α-helix, β-wing and C-terminal loop regions are possible interaction sites for KDR-binding by PLA2 homologues. In order to confirm this notion, we first designed six synthetic peptides based on the primary structure of the N-terminal αhelix region (peptides α-1 and α-2), the β-wing region (peptides β-1 and β-2) and the C-terminal loop region [peptides C-1 and C-2(C127S)] of KDR-bp (Figure 4). Cysteine residues in these peptides were substituted with serine residues in order to prevent synthetic peptides from dimerizing via disulfide bonds (highlighted serine residues in Figure 4B). The quality of synthetic peptides was confirmed by amino acid sequencing, amino acid composition analysis and MALDI–TOF MS. SPR analysis revealed that peptides α-1 and α-2, which mimic the Nterminal α-helix region, did not bind to the extracellular domain of KDR at all (Figures 5A and 5B). Although the C-terminal loop region peptide C-1 failed to interact with KDR, peptide C-2(C127S) bound to the extracellular domain of KDR with essentially equal affinity (K d = 10−8 M) as KDR-bp (Figures 5E and 5F and Table 2). We also tested the binding ability of another synthetic peptide in which cysteine residues were substituted c The Authors Journal compilation c 2008 Biochemical Society
with an alanine residue instead of serine, designated peptide C2(C127A) (Figure 4). Peptide C-2(C127A) bound to KDR with essentially equal affinity as peptide C-2(C127S) (Supplementary Figure 3 at http://www.BiochemJ.org/bj/411/bj4110515add.htm), indicating that the substituted serine residue did not affect the KDR-binding activity of peptide C-2. Both β-wing region peptides, peptide β-1 and peptide β-2, bound to the extracellular domain of KDR, although the binding affinity was 1000-fold lower than that of peptide C-2 (K d = 10−4 M) (Figures 5C and 5D and Table 2). These results suggest that the β-wing region may participate in KDR-binding potential in addition to the C-terminal loop region. Next, we attempted to further localize the KDR-binding sites of KDR-bp, and designed two additional synthetic peptides based on the β-wing region and the C-terminal loop region, designated peptides β-3 and C-3 (Figure 4). Peptide β-3 bound to KDR with 100-fold higher affinity (K d of 10−6 M) than peptide β-1 (Figure 5G and Table 2). Peptide C-3 bound to KDR with 1000fold lower affinity (K d of 10−5 M) than peptide C-2 (Figure 5H and Table 2). These results indicate that residues 121–129 (110–117 in KDR-bp), corresponding to peptide C-3, are essential for KDRbinding activity, and that the overlap region between peptides C-1 and C-2, residues 115–120 (105–109 in KDR-bp), enhances KDR-binding activity. Thus KDR-bp directly interacts with KDR via its C-terminal loop region (residues 115–129 in PLA2 , residues 105–127 in KDR-bp) and the β-wing region (residues 52–80 in PLA2 , residues 51–71 in KDR-bp) supports this interaction.
Novel high-affinity peptide bound to KDR
Figure 3
519
Surface electrostatic potential model of KDR-bp
The N-terminal α-helix region is shown in a dotted circle. The β-wing region and C-terminal loop region are indicated by arrows. Negatively charged regions are shown in red, and positively charged regions are shown in blue. The orientation of (B) is rotated 180◦ from that of (A).
Figure 4
Design of synthetic peptides from KDR-bp
(A) Primary structure of KDR-bp. Secondary structures, α-helices and β-strands, are represented as columns and arrows respectively. Lines indicate the regions corresponding to synthetic peptides. (B) Amino acid sequences and designation of synthetic peptides. Cysteine residues that were substituted are shaded. c The Authors Journal compilation c 2008 Biochemical Society
520
Figure 5
D. Fujisawa and others
Interaction of N-terminal α-helix region, β-wing region and C-terminal loop region peptides with immobilized KDR
The extracellular domain of KDR was immobilized on a CM4 sensor chip. Synthetic peptides at 0.1, 0.5, 1, 5 and 10 µM for peptide α-1 (A) and peptide α-2 (B), at 1, 5, 10, 50 and 100 µM for peptide β-1 (C), peptide β-2 (D), peptide C-1 (E), peptide β-3 (G) and peptide C-3 (H) and at 20, 40, 60, 80 and 100 nM for peptide C-2 (F), were injected over the immobilized KDR at a flow rate of 20 µl/min. Binding properties are shown in Table 2. The corresponding region in KDR-bp of each synthetic peptide is highlighted in red.
DISCUSSION
KDR-bp, an enzymatically inactive PLA2 homologue found in the Eastern Cottonmouth, is a novel VEGF receptor-binding protein that is structurally distinct from already known proteins, such as VEGFs and TIMP-3 (tissue inhibitor of metalloproteinase-3) [21]. In the present study, we demonstrated that inactive PLA2 homologues in snake venoms are common ligands for KDR. Furthermore, we identified the KDR-binding sites of KDR-bp using synthetic peptides based on its primary structure. A synthetic peptide based on the C-terminal loop region of KDRbp, peptide C-2, bound to KDR with an affinity essentially equal to KDR-bp. Previous studies have shown that blockage of VEGF165 results in the inhibition of angiogenesis and tumour development, and is useful in the treatment of age-related macular c The Authors Journal compilation c 2008 Biochemical Society
degeneration [1,22]. Several groups have attempted to generate blocking peptides against the VEGF–VEGF receptor system. These were designed based on the structure of receptor-binding loops and heparin-binding region of the ligands [23–26], or generated by phage-display systems [27–29]. Our present results show that KDR-binding peptides designed based on snake venom proteins are a novel class of VEGF receptor-binding peptide with unique sequences and high affinity. Biological consequence of KDR-binding potential of inactive phospholipase A2 homologues
In the present study we observed that the KDR-binding potential of snake venom PLA2 s and their homologues agreed with myotoxicity. However, given that the myotoxicity induced by
Novel high-affinity peptide bound to KDR Table 2
Binding properties of synthetic peptides
Dissociation constants (K d ) of synthetic peptides to immobilized KDRD1−7 . Synthetic peptides were subjected to SPR assay in several independent experiments (n = 2–13). k ass and k diss of KDR-bp, peptide C-2(C127S) and peptide C-2(C127A) were calculated using two-state reaction model fitting and K d was calculated from the ratio k diss /k ass . K d of other peptides was calculated using the steady-state affinity model. Nsb, no specific binding. Peptide
Residues
K d (M)
n
KDR-bp Peptide α-1 Peptide α-2 Peptide β-1 Peptide β-2 Peptide β-3 Peptide C-1 Peptide C-2(C127S) Peptide C-2(C127A) Peptide C-3
1–121 1–13 1–20 51–67 65–79 51–71 97–109 105–117 105–117 110–117
−9 6.9 + − 4.7 × 10 Nsb Nsb −4 2.2 + − 1.2 × 10−3 1.2 + 0.7 × 10 − −6 6.8 + − 1.4 × 10 Nsb −8 4.0 + − 1.1 × 10−8 2.5 + − 0.8 × 10−5 3.1 + − 0.3 × 10
6 6 7 7 7 3 7 13 2 3
521
C-terminal loop region peptide C-1 does not display KDR-binding potential, despite the similar electrostatic nature of C-terminal loop region peptides (calculated pI of peptides C-1, C-2 and C-3 were 9.8, 10.7 and 10.6 respectively). Thus the positively charged electrostatic nature around the C-terminal loop region of KDR-binding PLA2 homologues is partially responsible for its KDR-binding ability, but binding is not only dependent on the electrostatic nature of peptides. In conclusion, in the present study we demonstrate that all myotoxic PLA2 homologues exhibit KDR-binding potential, indicating that KDR-binding potential is a common property among myotoxic PLA2 homologues. Furthermore, using synthetic peptides, we confirmed that the C-terminal loop region of KDRbp interacts with KDR. This study assists in clarifying the binding mode of KDR and inactive PLA2 homologues, and may be advantageous for the design of novel VEGF receptor-binding peptides as an anti-angiogenic treatment. We would like to thank to Miss Yukiko Matsunaga for technical assistance. The work was supported by a Science Research Grants-in-Aid from the Ministry of Education, Science, and Culture of Japan (to T. M.).
KDR-binding PLA2 homologues is caused by binding, there remain some questions over why myotoxic PLA2 homologues selectively lyse skeletal muscle cells but not vascular endothelial cells [20], and why other KDR antagonists and blocking antibodies do not cause muscle cell death. VEGF165 is also known to be a potent vascular permeability factor [1]. We recently found that snake venom VEGF, VEGFF, exhibits more potent vascular permeability enhancement than VEGF165 . Moreover, KDR-bp synergistically enhances vascular permeability induced by VEGF165 and VEGF-F [9]. This phenomenon is thought to contribute to toxin penetration in prey animals due to the enhancement of vascular permeability induced by VEGF-F, and may explain the potent vascular permeability of Viperidae snake venoms. We found that a myotoxic PLA2 homologue from the venom of V. a. ammodytes, ammodytin L, also synergistically enhances vascular permeability induced by VEGFs [9]. Thus we believe that KDR-binding PLA2 homologues in snake venoms are important for enhancing vascular permeability in order to facilitate the penetration of toxins into tissues, and that KDR-binding ability may be responsible for this biological action rather than myotoxicity, although further investigation is needed. Binding mode of inactive phospholipase A2 homologues to VEGF receptor
We found that two peptides mimicking two distinct regions of KDR-bp (C-terminal loop region peptide and β-wing region) displayed KDR-binding potential (Figure 5). In the tertiary structure, the β-wing region is located on the opposite side of the Cterminal loop region (Figure 3). These data indicate that KDRbp interacts with KDR via two structurally distinct regions. The extracellular domain of KDR is composed of seven immunoglobulin-like domains, consisting of 745 amino acid residues, and the molecular mass of the extracellular domain of KDR is much larger than KDR-bp (121 amino acid residues). Therefore we believe that KDR-bp is able to bind to two distinct sites of the extracellular domain of KDR via the C-terminal loop and β-wing regions. PLA2 homologues that displayed KDR-binding ability formed positively charged clusters at the C-terminal loop region on the surface electrostatic potential model, whereas PLA2 s with no KDR-binding ability formed negatively charged clusters (dotted circles in Supplementary Figure 4 at http:// www.BiochemJ.org/bj/411/bj4110515add.htm). In addition, the
REFERENCES 1 Ferrara, N. (2004) Vascular endothelial growth factor: basic science and clinical progress. Endocr. Rev. 25, 581–611 2 Olsson, A. K., Dimberg, A., Kreuger, J. and Claesson-Welsh, L. (2006) VEGF receptor signalling - in control of vascular function. Nat. Rev. Mol. Cell Biol. 7, 359–371 3 Yamazaki, Y. and Morita, T. (2006) Molecular and functional diversity of vascular endothelial growth factors. Mol. Diversity 10, 515–527 4 Ogawa, S., Oku, A., Sawano, A., Yamaguchi, S., Yazaki, Y. and Shibuya, M. (1998) A novel type of vascular endothelial growth factor, VEGF-E (NZ-7 VEGF), preferentially utilizes KDR/Flk-1 receptor and carries a potent mitotic activity without heparin-binding domain. J. Biol. Chem. 273, 31273–31282 5 Meyer, M., Clauss, M., Lepple-Wienhues, A., Waltenberger, J., Augustin, H. G., Ziche, M., Lanz, C., Buttner, M., Rziha, H. J. and Dehio, C. (1999) A novel vascular endothelial growth factor encoded by Orf virus, VEGF-E, mediates angiogenesis via signalling through VEGFR-2 (KDR) but not VEGFR-1 (Flt-1) receptor tyrosine kinases. EMBO J. 18, 363–374 6 Wise, L. M., Veikkola, T., Mercer, A. A., Savory, L. J., Fleming, S. B., Caesar, C., Vitali, A., Makinen, T., Alitalo, K. and Stacker, S. A. (1999) Vascular endothelial growth factor (VEGF)-like protein from orf virus NZ2 binds to VEGFR2 and neuropilin-1. Proc. Natl. Acad. Sci. U.S.A. 96, 3071–3076 7 Yamazaki, Y., Takani, K., Atoda, H. and Morita, T. (2003) Snake venom vascular endothelial growth factors (VEGFs) exhibit potent activity through their specific recognition of KDR (VEGF receptor 2). J. Biol. Chem. 278, 51985–51988 8 Takahashi, H., Hattori, S., Iwamatsu, A., Takizawa, H. and Shibuya, M. (2004) A novel snake venom vascular endothelial growth factor (VEGF) predominantly induces vascular permeability through preferential signaling via VEGF receptor-1. J. Biol. Chem. 279, 46304–46314 9 Yamazaki, Y., Nakano, Y., Imamura, T. and Morita, T. (2007) Augmentation of vascular permeability of VEGF is enhanced by KDR-binding proteins. Biochem. Biophys. Res. Commun. 355, 693–699 10 Six, D. A. and Dennis, E. A. (2000) The expanding superfamily of phospholipase A2 enzymes: classification and characterization. Biochim. Biophys. Acta 1488, 1–19 11 Kini, R. M. (2003) Excitement ahead: structure, function and mechanism of snake venom phospholipase A2 enzymes. Toxicon 42, 827–840 12 Nakashima, K., Nobuhisa, I., Deshimaru, M., Nakai, M., Ogawa, T., Shimohigashi, Y., Fukumaki, Y., Hattori, M., Sakaki, Y., Hattori, S. and Ohno, M. (1995) Accelerated evolution in the protein-coding regions is universal in crotalinae snake venom gland phospholipase A2 isozyme genes. Proc. Natl. Acad. Sci. U.S.A. 92, 5605–5609 13 Ohno, M., Chijiwa, T., Oda-Ueda, N., Ogawa, T. and Hattori, S. (2003) Molecular evolution of myotoxic phospholipases A2 from snake venom. Toxicon 42, 841–854 14 Lomonte, B., Angulo, Y. and Calderon, L. (2003) An overview of lysine-49 phospholipase A2 myotoxins from crotalid snake venoms and their structural determinants of myotoxic action. Toxicon 42, 885–901 15 Krizaj, I., Bieber, A. L., Ritonja, A. and Gubensek, F. (1991) The primary structure of ammodytin L, a myotoxic phospholipase A2 homologue from Vipera ammodytes venom. Eur. J. Biochem. 202, 1165–1168 c The Authors Journal compilation c 2008 Biochemical Society
522
D. Fujisawa and others
16 Mebs, D., Kuch, U., Coronas, F. I., Batista, C. V., Gumprecht, A. and Possani, L. D. (2006) Biochemical and biological activities of the venom of the Chinese pitviper Zhaoermia mangshanensis , with the complete amino acid sequence and phylogenetic analysis of a novel Arg49 phospholipase A2 myotoxin. Toxicon 47, 797–811 17 Chijiwa, T., Tokunaga, E., Ikeda, R., Terada, K., Ogawa, T., Oda-Ueda, N., Hattori, S., Nozaki, M. and Ohno, M. (2006) Discovery of novel [Arg49]phospholipase A2 isozymes from Protobothrops elegans venom and regional evolution of Crotalinae snake venom phospholipase A2 isozymes in the southwestern islands of Japan and Taiwan. Toxicon 48, 672–682 18 Wei, J. F., Wei, X. L., Chen, Q. Y., Huang, T., Qiao, L. Y., Wang, W. Y., Xiong, Y. L. and He, S. H. (2006) N49 phospholipase A2 , a unique subgroup of snake venom group II phospholipase A2 . Biochim. Biophys. Acta 1760, 462–471 19 Bahnson, B. J. (2005) Structure, function and interfacial allosterism in phospholipase A2 : insight from the anion-assisted dimer. Arch. Biochem. Biophys. 433, 96–106 20 Yamazaki, Y., Matsunaga, Y., Nakano, Y. and Morita, T. (2005) Identification of vascular endothelial growth factor receptor-binding protein in the venom of eastern cottonmouth. A new role of snake venom myotoxic Lys49-phospholipase A2 . J. Biol. Chem. 280, 29989–29992 21 Qi, J. H., Ebrahem, Q., Moore, N., Murphy, G., Claesson-Welsh, L., Bond, M., Baker, A. and Anand-Apte, B. (2003) A novel function for tissue inhibitor of metalloproteinases-3 (TIMP3): inhibition of angiogenesis by blockage of VEGF binding to VEGF receptor-2. Nat. Med. 9, 407–415 22 Ng, E. W., Shima, D. T., Calias, P., Cunningham, Jr, E. T., Guyer, D. R. and Adamis, A. P. (2006) Pegaptanib, a targeted anti-VEGF aptamer for ocular vascular disease. Nat. Rev. Drug Discov. 5, 123–132 23 Soker, S., Gollamudi-Payne, S., Fidder, H., Charmahelli, H. and Klagsbrun, M. (1997) Inhibition of vascular endothelial growth factor (VEGF)-induced endothelial cell proliferation by a peptide corresponding to the exon 7-encoded domain of VEGF165. J. Biol. Chem. 272, 31582–31588 Received 8 January 2008/4 February 2008; accepted 6 February 2008 Published as BJ Immediate Publication 6 February 2008, doi:10.1042/BJ20080078
c The Authors Journal compilation c 2008 Biochemical Society
24 Jia, H., Jezequel, S., Lohr, M., Shaikh, S., Davis, D., Soker, S., Selwood, D. and Zachary, I. (2001) Peptides encoded by exon 6 of VEGF inhibit endothelial cell biological responses and angiogenesis induced by VEGF. Biochem. Biophys. Res. Commun. 283, 164–173 25 Zilberberg, L., Shinkaruk, S., Lequin, O., Rousseau, B., Hagedorn, M., Costa, F., Caronzolo, D., Balke, M., Canron, X., Convert, O. et al. (2003) Structure and inhibitory effects on angiogenesis and tumor development of a new vascular endothelial growth inhibitor. J. Biol. Chem. 278, 35564–35573 26 Yamazaki, Y., Tokunaga, Y., Takani, K. and Morita, T. (2005) Identification of the heparin-binding region of snake venom vascular endothelial growth factor (VEGF-F) and its blocking of VEGF-A165. Biochemistry 44, 8858–8864 27 Binetruy-Tournaire, R., Demangel, C., Malavaud, B., Vassy, R., Rouyre, S., Kraemer, M., Plouet, J., Derbin, C., Perret, G. and Mazie, J. C. (2000) Identification of a peptide blocking vascular endothelial growth factor (VEGF)-mediated angiogenesis. EMBO J. 19, 1525–1533 28 Hetian, L., Ping, A., Shumei, S., Xiaoying, L., Luowen, H., Jian, W., Lin, M., Meisheng, L., Junshan, Y. and Chengchao, S. (2002) A novel peptide isolated from a phage display library inhibits tumor growth and metastasis by blocking the binding of vascular endothelial growth factor to its kinase domain receptor. J. Biol. Chem. 277, 43137–43142 29 El-Mousawi, M., Tchistiakova, L., Yurchenko, L., Pietrzynski, G., Moreno, M., Stanimirovic, D., Ahmad, D. and Alakhov, V. (2003) A vascular endothelial growth factor high affinity receptor 1-specific peptide with antiangiogenic activity identified using a phage display peptide library. J. Biol. Chem. 278, 46681–46691 30 Renetseder, R., Brunie, S., Dijkstra, B. W., Drenth, J. and Sigler, P. B. (1985) A comparison of the crystal structures of phospholipase A2 from bovine pancreas and Crotalus atrox venom. J. Biol. Chem. 260, 11627–11634 31 Kihara, H., Uchikawa, R., Hattori, S. and Ohno, M. (1992) Myotoxicity and physiological effects of three Trimeresurus flavoviridis phospholipases A2 . Biochem. Int. 28, 895–903
