Structural investigation of the VEGF receptor interaction with a helical antagonist peptide

Article abstract of DOI:10.1002/psc.2480

Angiogenesis is mainly regulated by the vascular endothelial growth factor (VEGF), a mitogen specific for endothelial cells, which binds two tyrosine kinase receptors, VEGFR1 and VEGFR2, on the surface of endothelial cells. Molecules targeting VEGF receptors are attractive to pharmacologically treat diseases associated with angiogenesis or to be used as probes in angiogenesis imaging. Recently, we reported a designed peptide targeting VEGF receptors and able to inhibit the VEGF-angiogenic response in vitro and in vivo. In this study, we employed NMR and molecular modeling methodology to investigate the molecular determinants of the interaction peptide-receptor. In particular, the peptide binding site on VEGFR1 domain 2 and the residues involved in receptor recognition have been determined. These results provide significant information to develop a new class of molecules able to recognize the VEGF receptors overexpressed in pathological angiogenesis.

Full text of DOI:10.1002/psc.2480

Special Issue Article
Received: 18 October 2012
Revised: 15 December 2012
Accepted: 17 December 2012
Published online in Wiley Online Library: 19 February 2013
(wileyonlinelibrary.com) DOI 10.1002/psc.2480
Structural investigation of the VEGF receptor
interaction with a helical antagonist peptide{
Donatella Diana,a Rossella Di Stasi,a Lucia De Rosa,a Carla Isernia,b
Luca D. D’Andreaa and Roberto Fattorussob*
Angiogenesis is mainly regulated by the vascular endothelial growth factor (VEGF), a mitogen specific for endothelial cells, which
binds two tyrosine kinase receptors, VEGFR1 and VEGFR2, on the surface of endothelial cells. Molecules targeting VEGF receptors
are attractive to pharmacologically treat diseases associated with angiogenesis or to be used as probes in angiogenesis imaging.
Recently, we reported a designed peptide targeting VEGF receptors and able to inhibit the VEGF-angiogenic response in vitro and
in vivo. In this study, we employed NMR and molecular modeling methodology to investigate the molecular determinants of the
interaction peptide-receptor. In particular, the peptide binding site on VEGFR1 domain 2 and the residues involved in receptor
recognition have been determined. These results provide significant information to develop a new class of molecules able to
recognize the VEGF receptors overexpressed in pathological angiogenesis. Copyright © 2013 European Peptide Society and
John Wiley & Sons, Ltd.
Keywords: VEGF; angiogenesis; NMR; peptide; interactions
formation and tumor growth. Thus, MA represents a candidate
for the treatment of diseases associated with excessive VEGF-
Introduction
The vascular endothelial growth factor (VEGF) is a homodimeric
cytokine-regulating angiogenesis mainly acting on endothelial
cells (EC) [1]. VEGF interacts with two tyrosine kinase transmem-
brane receptors (VEGFR1/Flt-1 and VEGFR2/KDR) on the surface
of EC. VEGF receptors contain an extracellular region, composed
of seven Ig-like domains, a short transmembrane helix, and the
intracellular kinase domain. The high-affinity binding of VEGF to
the extracellular region induces receptor dimerization and activa-
tion-stimulating transphosphorylation of the intracellular kinase
domain. The phosphorylated tyrosines act as docking sites for
the recruitment of intracellular proteins, which in turn activate
signaling pathways, involving key mediators, such as Akt and
ERK1/2, ending up in EC proliferation, migration, and survival
[2]. Structural and biochemical studies showed that VEGF binding
to VEGFR1 mainly involved domain 2 (VEGR1D2) [3], whereas
binding to VEGFR2 required receptor domains 2 and 3[4,5].
Many studies have been devoted to find novel molecules able
to modulate VEGF signaling targeting VEGF–VEGF receptor inter-
action [6–8]. Such molecules have potential applications in the
treatment of cancer and other pathologies depending on exces-
sive angiogenesis [9–11] to stimulate angiogenesis [11,12] and
for the imaging of VEGF receptors [13,14].
dependent angiogenesis.
To identify the molecular details of the interaction between
VEGF receptor and MA, we report the structural characterization
of the interaction between MA and VEGFR1D2 by using NMR
spectroscopy and molecular modeling studies. In particular, MA
selectivity and binding to the VEGFR1D2 has been evaluated by
means of NMR chemical shift perturbation experiments, which
allow for the identification of the amino acid residues of VEGFR1D2
involved in MA recognition. Furthermore, the interaction between
MA and VEGFR1D2 was monitored by one-dimensional NMR
saturation transfer difference (STD) to define the binding epitope
of the ligand. Finally, a structural model of VEGFR1D2–MA complex
has been derived by molecular modeling on the basis of NMR-
derived constraints.
Materials and Methods
Chemicals
Fmoc-protected amino acids and coupling reagents were
purchased from Inbios (Naples, Italy). Rink amide MBHA resin
was obtained from Merck (Nottingham, UK). Peptide synthesis
Over the last several years, we reported the structural and
biological properties of several VEGFR binder peptides designed
on VEGF regions involved in VEGFR recognition such as the
N-terminal helix (residues 17–25) [15–23] and the b-hairpin
encompassing residues 79–92 [24].
Particularly, we have described the design, structure, and func-
tional characterization of a novel VEGF antagonist helical peptide
[21]. We showed that this peptide, MA, assumes in water a well-
defined helical conformation, binds to VEGF receptors, and
inhibits the VEGF biological activity on ECs. Furthermore, in vivo
experiments showed that MA inhibits VEGF-induced capillary
*
Correspondence to: Roberto Fattorusso, Dipartimento di Scienze Ambientali,
Seconda Università di Napoli, Via Vivaldi 43, Caserta, Italy. E-mail: roberto.
fattorusso@unina2.it
{
Special issue devoted to contributions presented at the 13th Naples Workshop
on Bioactive Peptides, June 7–10, 2012, Naples.
a Istituto di Biostrutture e Bioimmagini, CNR, Via Mezzocannone 16, Napoli, Italy
b Dipartimento di Scienze Ambientali, Seconda Università di Napoli, Via Vivaldi
43, Caserta, Italy
J. Pept. Sci. 2013; 19: 214–219
Copyright © 2013 European Peptide Society and John Wiley & Sons, Ltd.

STRUCTURAL STUDIES OF VEGF RECEPTOR BOUND TO AN ANTAGONIST PEPTIDE
solvents N,N,-DMF and DCM were obtained, respectively, from
Romil (Cambridge, UK) and Sigma-Aldrich (Milan, Italy) with the
lowest water content and used without further purification.
Piperidine was from Biosolve (Valkenswaard, Netherlands), and
N,N-DIPEA from Romil (Cambridge, UK), and acetic anhydride
triisopropylsilane, ethanedithiol, acetonitrile (CH3CN), and TFA
were from Sigma-Aldrich.
(Inalco, Milan, Italy). After 4–5 h, the cells were harvested by cen-
trifugation; the pellet was dissolved in 50-mM Tris-HCl, pH 8 con-
taining protease inhibitors cocktail (Roche, Basel, Swiss), to avoid
protein degradation, and the suspension was sonicated for 6min,
by using a Misonix Sonicator 3000 apparatus (Farmingdale, NY, USA)
with a micro tip probe and an impulse output of 1.5/2 (=9/12 W).
Bacterial lysate was then centrifuged (27200 g, 30 min, 4 ꢂC), and
the inclusion body pellet containing VEGFR1D2 was solubilized
in 50-mM Tris-HCl, 10-mM imidazole, and 8-M urea, pH 8. The sol-
ubilized VEGFR1D2 protein was loaded on a Ni2+–NTA resin
(45 min, 20 ꢂC) in the presence of 300-mM NaCl. The His-tagged
protein was refolded by equilibrating the resin in 50-mM Tris-HCl,
10-mM imidazole, and 300-mM NaCl, pH 8 with decreasing concen-
trations of urea, and then, it was eluted with increasing imidazole
concentration from 100 to 300 mM. The 15N-labeled recombinant
VEGFR1D2 was expressed according to Sambrook et al. (Sambrook,
J.; Fritch, E. F.; Maniatis, T. Molecular Cloning, 2nd ed.; Cold Spring
Harbor Laboratory: Cold Spring Harbor, 1989; pp 3, A3). The refold-
ing and protein purification were performed as described for the un-
labeled protein. Refolded VEGFR1D2 was dialyzed overnight in 50-
mM Tris-HCl, pH 7.0 and NaCl 250 mM at 4 ꢂC. Then, the cleavage
of the His-tagged protein was performed in the presence of glutathi-
one (3 mM reduced/0.3 mM oxidized) for 3 h at 30 ꢂC, adding a lab
prepared Tobacco Etch Virus protease to protein substrate in a
molar ratio (protease : substrate) of 1 : 35. Finally, VEGFR1D2
was purified to homogeneity by size exclusion chromatography
using a S75 column (GE Healthcare, Milan, Italy) equilibrated in
50-mM Tris-HCl and 250-mM NaCl, pH 7. Finally, it was concen-
trated until 0.9 mM by the Amicon Ultra system using a mem-
brane with a 3000 MW cut-off (Millipore, Milan, Italy).
SPPS
MA peptide (acetyl-KLTWMELYQLAYKGI-amide) was synthesized
on Rink amide MBHA resin (0.5 mmol/g) using standard Fmoc/
tBu chemistry. Fmoc deprotection was performed by washing
the resin two times for 5 min with a solution of 30% piperidine
in DMF, and coupling reactions were performed with 10 equiv
of Fmoc-amino acid and 9.9 equiv of HBTU/HOBt, 20 equiv DIPEA
in DMF for 45 min at room temperature; capping step (1 ꢀ 5 min)
was performed with a solution of acetic anhydride (0.5 M)/HOBt
(0.015 M)/DIPEA (0.125 M) in DMF. Each step was followed by five
washing with DMF for 1 min. Peptide cleavage was performed
using
a mixture of TFA/H2O/ethanedithiol/triisopropylsilane
(94 : 2.5 : 2.5 : 1 v/v/v/v) for 3 h at room temperature. Resin was
separated by filtration, and the filtrate was precipitated with cold
diethyl. The precipitate was separated by centrifugation, solubi-
lized in water, and lyophilized.
Peptide Purification and Analysis
Peptide purifications were carried out by RP-HPLC on the Shimadzu
(Kyoto, Japan) LC-8A system, equipped with a UV–vis detector SPD-
10A using a semi-preparative Phenomenex (Torrance, CA, USA) Ju-
piter Proteo column (250ꢀ 10 mm, 90 Å) and a linear gradient from
20% to 80% of CH3CN (0.1% TFA) in H2O (0.1% TFA) in 30min at the
flow rate of 5 ml/min. Peptide identification and analysis were
performed on a Thermo LC-MS LCQ Deca XP MAX instrument
(Waltham, MA, USA) equipped with diode array detector combined
with an electrospray ion source and a quadrupole mass analyzer
using a Phenomenex Jupiter Proteo column (150 ꢀ 4.60 mm, 90 Å)
and a linear gradient from 20% to 80% of CH3CN (0.1% TFA) in
H2O (0.1% TFA) in 30 min. The synthesized peptide showed
purity above 95%, based on the chromatographic peak area
revealed at 210 nm.
NMR Spectroscopy
To map the binding site of MA on VEGFR1D2,
15N uniformly
labeled VEGFR1D2 (80 mM in 50-mM Tris-HCl and 250-mM NaCl,
pH 7) was titrated with increasing amounts of unlabeled peptide
(20, 40, 60, 80, 160, and 240 mM) [26]. Two-dimensional [1H, 15N]
heteronuclear single quantum coherence (HSQC) spectra were
acquired with 32 transients per t1 value. Presaturation of water
was employed during a recycle delay of 1.5 s. The 1 K complex
points were acquired in t2, with an acquisition time of 102.5 ms,
whereas 128 complex points were acquired in t1 with an acquisi-
tion time of 64 ms. Backbone 15N and HN resonance assignments
were obtained on the basis of previously reported assignments
[27]. To determine the per residue chemical shift perturbation
upon binding and account for differences in spectral widths
between 15N and 1H resonances, weighted average chemical shift
differences, ΔdΗΝ, were calculated for the amide 15N and 1H
resonances, using the equation:
MS m/z calculated: 1897.0 amu; found 949.4 [M + 2H]2+
Rt = 18.4 min.
.
Protein Expression and Purification
VEGFR1D2 protein was prepared according to Di Stasi et al. [25].
Briefly, VEGFR1D2 gene (DNA fragment encoding amino acids
129–229 of the human VEGFR1) was amplified by PCR from cDNA
of the full receptor. Then, it was purified and digested with NcoI
and XhoI (New England Biolabs, Ipswich, MA, USA) restriction
enzymes. The NcoI/XhoI gene fragment was ligated into the
pETM11 expression vector (EMBL, Hamburg, Germany) down-
stream to the His-tag and TEV recognition site sequences. The
identity of the insert in the resulting recombinant plasmid was
confirmed by DNA sequencing (PRIMM, Naples, Italy). The
plasmid construct was transformed into Escherichia coli BL21
Codon Plus (DE3) RIL cell strain (Stratagene, La Jolla, CA, USA) car-
rying an inducible T7 RNA polymerase gene. Bacterial culture of
1 l of prewarmed Luria-Bertani medium containing 50 mg mlꢁ1
kanamycin and 33 mg mlꢁ1 chloramphenicol was grown in shak-
ing flasks at 37 ꢂC until reaching 0.7/0.8 OD600. Then, they were
induced with 0.7-mM IPTG, Isopropyl b-D-1-thiogalactopyranoside
h
i
1=2
ꢀ
ꢁ
2
2
ΔdHN
¼
ðΔHNÞ þ 0:17 ꢀ Δ15N
where ΔH and ΔN are the differences between the free and bound
chemical shifts. The weighted average chemical shift differences
were mapped to the VEGFR1D2 NMR structure (1QSV.pdb) [27]
using MOLMOL graphics program [28]. CARA software [29] was used
to generate overlays of the two-dimensional spectra.
The dissociation constants (Kd) were estimated from the
1
changes in both H and 15N chemical shifts as observed in the
two-dimensional [1H, 15N] HSQC spectra of VEGFR1D2 upon
binding to MA peptide. The chemical shift changes, measured
J. Pept. Sci. 2013; 19: 214–219 Copyright © 2013 European Peptide Society and John Wiley & Sons, Ltd. wileyonlinelibrary.com/journal/jpepsci

DIANA ET AL.
at various protein–peptide ratios, were plotted versus ligand
concentration, and the results were fitted by nonlinear regression
according to the following equation [30] with GraphPad Prism,
version 4.00 (GraphPad Software, San Diego, CA, USA):
Upon progressive additions of unlabeled MA to a sample
containing 15N-labeled VEGFR1D2, we observed continuous
changes in 1H and 15N chemical shifts for several signals of
VEGFR1D2 in the two-dimensional 1H–15N HSQC spectra (Figure 1).
The VEGFR1D2 residues showing the most significant changes in
1H and 15N chemical shifts upon formation of the complex were
assessed by analyzing the normalized chemical shift deviations
(ΔdΗΝ) and were mapped onto the NMR solution structure of
the VEGFR-1D2 (Figure 2(A) and (B)). The most affected residues
are located in the N-terminal ba strand region encompassing
residues from 139 to145, in the loop separating bc and bc0
strands (residues 169–172), in the bf strand (residues 204–207),
and in the bg strand (residues 216–221). Thus, NMR titration
allows us to approximate the region of the VEGFR1D2 interacting
with the peptide. Notably, this region was involved in the
molecular recognition of the VEGF helical region that has been
(residues 17–25) exploited to design MA peptide. These chemical
shift changes indicated the formation of the MA/VEGFR1D2
complex in fast exchange on the NMR time scale. The Kd of the
MA/VEGFR1D2 complex was estimated to be in the micromolar
range (ꢄ46 ꢅ 13 mM) after an analysis of the NMR titration curves
of selected VEGFR1D2 residues (Figure 2(C)).
Δd ¼ Δdmax Kd þ ½Pꢃ þ ½Lꢃ
f
hꢂ
ꢃ
i
1=2
2
ꢁ
Kd þ ½Pꢃ þ ½Lꢃ
ꢁ 4½Pꢃ½Lꢃ
=2½Pꢃ
g
where Δd is the chemical shift change at various VEGFR1D2
/
ligand ratios, Δdmax is the chemical shift change at saturation,
Kd is the dissociation constants, and [P] and [L] are the protein
and ligand concentrations, respectively. All curve fitting showed
an R2 value ≥0.98.
For NMR STD sample [31], the protein solution was diluted in
NMR buffer (50-mM Tris, 100-mM NaCl, and 10% D2O). The final
sample contained 10 mM of unlabeled VEGFR-1D2 and 1-mM
peptide, leading to a protein : ligand ratio of 1 : 100. STD spectra
of the unlabeled VEGFR-1D2/peptide complex were recorded with
2048 scans and selective saturation of the protein resonances at
12 and ꢁ3 ppm. Time dependence of the saturation transfer
was investigated using an optimized saturation times ranging
from 0.2 to 4.0 s. STD NMR spectra were acquired using a series
of equally spaced 50-ms Gaussian-shaped pulses for selective
saturation, with 1-ms delay between the pulses. Subtraction of
the protein Free induction decay resonance was performed by
phase cycling.
Binding Epitope of the MA Peptide
To provide information on MA binding residues, one-dimensional
STD NMR experiments of MA in the presence of unlabeled
VEGFR1D2were performed. As a control experiment, STD NMR spec-
trum of the peptide without VEGFR1D2 protein was recorded. The
Molecular modeling studies
The HADDOCK web server [32] was used to generate a model of
the MA/VEGFR-1D2 complex. The NMR structures with the lowest
target functions of both MA and VEGFR-1D2 (1QSV.pdb) were
implemented for these studies. Chemical shift perturbation and
STD data were exploited to produce unambiguous interaction
restraints. In all of the docking runs, passive residues were set
automatically by the HADDOCK web server and the solvated
docking mode was turned on. In the first iteration (i.e. the rigid
body energy minimization), 1000 structures were calculated; in
the second iteration, the best 200 solutions were subjected to
semi-flexible simulated annealing, and a final refinement in water
was also performed The final 200 HADDOCK models were all
visually inspected, and solutions not compatible with our experi-
mental data and/or containing highly unusual protein orienta-
tions were soon removed. In particular, four clusters were
obtained, and the best structures of each cluster (according to
the HADDOCK score) were compared. Because the resulting
structures are very similar to each other, we decided to use the
one with the lowest (i.e. best) HADDOCK score as representative
of the MA/VEGFR1D2 complex.
Results and Discussion
NMR Mapping Studies of the MA Peptide Binding Site on
VEGFR1D2
In vitro binding studies showed that MA is able to interact specif-
ically and compete with VEGF for binding site on ECs, therefore
indicating that peptide binds to VEGF receptors [21]. To define
MA binding site on the receptor and to estimate the complex
dissociation constant (Kd), we performed an NMR titration using
the second domain of VEGFR1.
Figure 1. Mapping the VEGFR1D2 binding interface for MA. Overlay of
[1H, 15N] HSQC spectra of 15N-labeled VEGFR1D2 (80 mM) in absence (red)
and presence (blue) of unlabeled peptide MA (80 mM).
wileyonlinelibrary.com/journal/jpepsci Copyright © 2013 European Peptide Society and John Wiley & Sons, Ltd. J. Pept. Sci. 2013; 19: 214–219

STRUCTURAL STUDIES OF VEGF RECEPTOR BOUND TO AN ANTAGONIST PEPTIDE
Figure 2. (A) Graph reporting the normalized chemical shift deviations (Δd= [(ΔHN)2 + (0.17 ꢀ Δ15N)2]1/2) versus the residue. Residues localized at the
N-terminal and strand ba (Y139, S140, E141, I142, E144, I145) at the loop between strands bc and bc0 (L169, K170, K171, and F172), at the strand bf (L204,
L205, T206, and C207), and at the strand bg (Y216, K217, T218, N219, Y220, and L221) show normalized deviations with values higher that 0.1 ppm. (B)
Amino acids with normalized chemical shifts deviations (ΔdHN values) greater than 0.1 ppm are colored in blue on the three-dimensional solution
structure of VEGFR1D2 (conformer number 1, pdb code: 1QSV) in its ribbon and surface representations. (C) Chemical shift changes (ΔdHN) for the
H
15N resonances of the protein residues showing large variations in chemical shift upon formation of the MA/VEGFR1D2 complex as a function of the
ligand concentration. The curves represent the best fit solution of the quadratic equation that describes 1 : 1 complex formation.
J. Pept. Sci. 2013; 19: 214–219 Copyright © 2013 European Peptide Society and John Wiley & Sons, Ltd. wileyonlinelibrary.com/journal/jpepsci

DIANA ET AL.
sample containing unlabeled VEGFR1D2 protein showed saturation
transfer from the protein to the ligand in the STD spectra, thus
confirming peptide–protein interaction. In particular, the STD spec-
trum of MA in a 100-fold excess over the unlabeled VEGFR1D2 pro-
tein (Figure 3) clearly demonstrates the involvement of backbone
HN of residues Trp4 and Met5 and of aromatic protons of
residues Trp4, Tyr8, and Tyr12. Furthermore, the signal of the HNe
protons of the side chain of the Gln9 showed similar large STD
intensities. All MA residues involved in protein recognition occupy
the same face of the helix (Figure 4) and correspond to the inte-
racting residues hypothesized in the design step [24]. The nature
of the interacting residues suggests that hydrophobic interactions
drive the formation of MA/VEGFR1D2 complex.
complex has been generated with the HADDOCK web server on
the basis of chemical shift perturbation and STD data. Particularly,
residues Y139, S140, E141, I142, E144, I145, L169, K170, K171,
F172, L204, L205, T206, C207, Y216, K217, T218, N219, Y220, and
L221 of the VEGFR1D2 and residues W4, Y8, Q9, and Y12 of the
MA were considered active. In all of the docking runs, passive
residues were set automatically by the HADDOCK web server
and the solvated docking mode was turned on. The structure of
the complex derived from the molecular modeling studies
(Figure 5) shows that the orientation of MA bound to the
receptor binding site is consistent with what is observed for the
VEGF a1 helix (residues 17–25) interaction with VEGFR1D2 [27]
Moreover, docking results, in agreement with the NMR binding
studies, suggest that intermolecular interactions between MA
and VEGFR1D2 include a network of p–p interactions contacts.
Most of these stabilizing hydrophobic intermolecular interactions
involve the aromatic ring of the residues Trp4, Tyr8, and Tyr12 of
MA peptide and Y139, I145, F172, Y216, L204, Y220, and L221
Molecular Modeling of the MA/VEGR1D2 Complex
To gain structural information about the molecular interface of
the bound MA peptide, a structural model of MA/VEGFR1D2
Figure 3. Epitope mapping for MA using one-dimensional STD NMR. Expansions of (A) the reference one-dimensional 1H NMR spectrum of MA in 50-mM Tris
and 100-mM NaCl at pH 7.0 at 298 K and 600MHz and (B) the NMR STD spectrum of MA in the presence of VEGFR1D2 using the same condition as in (A).
Figure 5. Docking studies. The derived molecular modeling of the MA/
VEGFR1D2 complex. Residues that are most affected by binding in our
chemical shifts perturbation studies (ΔdHN > 0.1 ppm) have been colored
in orange on the surface representations of the VEGFR1D2. The side chains
of the residues on the MA surface that may contribute to important inter-
actions at the complex interface are shown as a neon representation.
Figure 4. MA binding epitope. The side chain of the residues interacting
with the receptor, as revealed by NMR STD experiments, is shown as a
neon representation.
wileyonlinelibrary.com/journal/jpepsci Copyright © 2013 European Peptide Society and John Wiley & Sons, Ltd. J. Pept. Sci. 2013; 19: 214–219

STRUCTURAL STUDIES OF VEGF RECEPTOR BOUND TO AN ANTAGONIST PEPTIDE
15 D’Andrea LD, Iaccarino G, Fattorusso R, Sorriento D, Carannante C,
residues on the VEGFR-1D2 surface. Potential hydrogen bonds can
also be identified involving amide – groups on the VEGFR1D2 sur-
face and the hydroxyl group of the tyrosine side chains of
the peptide.
Capasso D, Trimarco B, Pedone C. Targeting angiogenesis: structural
characterization and biological properties of a de novo engineered
VEGF mimicking peptide. Proc. Natl. Acad. Sci. U.S.A. 2005; 102:
14215–14220.
16 Diana D, Ziaco B, Colombo G, Scarabelli G, Romanelli A, Pedone C,
Fattorusso R, D’Andrea LD. Structural determinants of the unusual
helix stability of a de novo engineered vascular endothelial growth
factor (VEGF) mimicking peptide. Chem. Eur. J. 2008; 14: 4164–4166.
17 Dudar GK, D’Andrea LD, Di Stasi R, Pedone C, Wallace JL. A vascular
endothelial growth factor mimetic accelerates gastric ulcer healing
in an iNOS-dependent manner. Am. J. Physiol. Gastrointest. Liver Phy-
siol. 2008; 295: G374–G381.
18 Santulli G, Ciccarelli M, Palumbo G, Campanile A, Galasso G, Ziaco B,
Altobelli GG, Cimini V, Piscione F, D’Andrea LD, Pedone C, Trimarco
B, Iaccarino G. In vivo properties of the proangiogenic peptide QK. J.
Transl. Med. 2009; 7: 41–50.
Conclusions
The molecular determinants of the binding surface between the
helical antagonist peptide MA and VEGFR1D2 were examined by
combining solution NMR studies and molecular modeling. The
peptide binding site on VEGFR1D2 has been characterized, and
it overlaps with the binding region of VEGF helix 17–25.
Furthermore, peptide residues involved in receptor recognition
have been determined.
19 Diana D, Ziaco B, Scarabelli G, Pedone C, Colombo G, D’Andrea LD,
Fattorusso R. Structural analysis of a helical peptide unfolding path-
way. Chem. Eur. J. 2010; 16: 5400–5407.
20 Gautier B, Goncalves V, Diana D, Di Stasi R, Teillet F, Lenoir C,
Huguenot F, Garbay C, Fattorusso R, D’Andrea LD, Vidal M, Inguimbert
N. Biochemical and structural analysis of the binding determinants of
a vascular endothelial growth factor receptor peptidic antagonist.
J. Med. Chem. 2012; 53: 4428–4440.
21 Basile A, Del Gatto A, Diana D, Di Stasi R, Falco A, Festa M, Rosati
A, Barbieri A, Franco R, Arra C, Pedone C, Fattorusso R, Turco MC,
D’Andrea LD. Characterization of a designed vascular endothelial
growth factor receptor antagonist helical peptide with antiangiogenic
activity in vivo. J. Med. Chem. 2011; 54: 1391–1400.
22 Ziaco B, Diana D, Capasso D, Palumbo R, Celentano V, Di Stasi R,
Fattorusso R, D’Andrea LD. C-terminal truncation of vascular
endothelial growth factor mimetic helical peptide preserves structural
and receptor binding properties. Biochem. Biophys. Res. Commun.
2012; 424: 290–294.
These data demonstrate that to develop peptides targeting
protein–protein interactions, an accurate secondary structure
motif stabilization keeping the three-dimensional arrangements
of the interacting residues may represent a successful structure-
based drug design approach. These results provide significant
information to develop a new class of molecules able to recognize
the VEGF receptors overexpressed in pathological angiogenesis.
Acknowledgements
This work was supported by MIUR (PON Ricerca e Competitività
2007–2013 #PON 01_2388, PRIN 2008-20084MMXNM_004). We
thank G. Filograsso, C. Chiarella, L. Zona, and L. De Luca for
administrative technical assistance.
23 Finetti F, Basile A, Capasso D, Di Gaetano S, Di Stasi R, Pascale M, Turco
CM, Ziche M, Morbidelli L, D’Andrea LD. Functional and pharmacolog-
ical characterization of a VEGF mimetic peptide on reparative
angiogenesis. Biochem. Pharmacol. 2012; 84: 303–311.
References
1 Ferrara N, Gerber HP, Le Couter J. The biology of VEGF and its receptors.
Nat. Med. 2003; 9: 669–676.
2 Olsson AK, Dimberg A, Kreuger J, Claesson-Welsh L. VEGF receptor
signalling – in control of vascular function. Nat. Rev. Mol. Cell Biol.
2006; 7: 359–371.
24 Diana D, Basile A, De Rosa L, Di Stasi R, Auriemma S, Arra C, Pedone C,
Turco MC, Fattorusso R, D’Andrea LD. b-hairpin peptide that targets
vascular endothelial growth factor (VEGF) receptors: design, NMR
characterization, and biological activity. J. Biol. Chem. 2011; 286:
41680–41691.
25 Di Stasi R, Diana D, Capasso D, Palumbo R, Romanelli A, Pedone C,
Fattorusso R, D’Andrea LD. VEGFR1(D2) In drug discovery: expression
and molecular characterization. Biopolymers 2010; 94: 800–809
26 Pellecchia M, Bertini I, Cowburn D, Dalvit C, Giralt E, Jahnke W,
James TL, Homans SW, Kessler H, Luchinat C, Meyer B, Oschkinat
H, Peng J, Schwalbe H, Siegal G. Perspectives on NMR in drug
discovery: a technique comes of age. Nat. Rev. Drug Discov. 2008;
7: 738–745.
3 Wiesmann C, Fuh G, Christinger HW, Eigenbrot C, Wells JA, de Vos AM.
Crystal structure at 1.7 Angstrom resolution of VEGF in complex with
domain 2 of the Flt-1 receptor. Cell 1997; 91: 695–704.
4 Fuh G, Li B, Crowley C, Cunningham B, James A. Requirements for
binding and signaling of the kinase domain receptor for vascular
endothelial growth factor. J. Biol. Chem. 1998; 273: 11197–11204.
5 Leppänen VM, Prota AE, Jeltsch M, Anisimov A, Kalkkinen N, Strandin
T, Lankinen H, Goldman A, Ballmer-Hofer K, Alitalo K. Structural deter-
minants of growth factor binding and specificity by VEGF receptor 2.
Proc. Natl. Acad. Sci. U.S.A. 2010; 107: 2425–2430.
6 D’Andrea LD, Del Gatto A, Pedone C, Benedetti E. Peptide-based
molecules in angiogenesis. Chem. Biol. Drug Des. 2006; 67: 115–126.
7 D’Andrea LD, Del Gatto A, De Rosa L, Romanelli A, Pedone C. Peptides
targeting angiogenesis related growth factor receptors. Curr. Pharm.
Des. 2009; 15: 2414–2429.
8 Grothey A, Galanis E. Targeting angiogenesis: progress with anti-VEGF
treatment with large molecules, Nat. Rev. Clin. Oncol. 2009; 6: 507–518.
9 Folkman J. Angiogenesis. Annu. Rev. Med. 2006; 57: 1–18.
10 Folkman J. Angiogenesis: an organizing principle for drug discovery?
Nat. Rev. Drug Discov. 2007; 6: 273–286.
11 Ferrara N, Kerbel RS. Angiogenesis as a therapeutic target, Nature
2005; 438: 967–974.
12 Jacobs J. Combating cardiovascular disease with angiogenic therapy,
Drug Discov. Today 2007; 12: 1040–1045.
13 Cai W, Chen X. Multimodality molecular imaging of tumor angiogen-
esis, J. Nucl. Med. 2008; 49: 113S–128S.
14 D’Andrea LD, Romanelli A, Di Stasi R, Pedone C. Bioinorganic aspects
of angiogenesis. Dalton Trans. 2010; 39: 7625–7636.
27 Starovasnik MA, Christinger HW, Wiesmann C, Champe MA, De Vos
AM, Skelton NJ. Solution structure of the VEGF-binding domain of
Flt-1: comparison of its free and bound states. J. Mol. Biol. 1999;
293: 531–544.
28 Koradi R, Billeter M, Wüthrich K. MOLMOL: a program for display
and analysis of macromolecular structures. J. Mol. Graph. 1996;
14: 29–32.
29 Masse JE, Keller R. AutoLink: automated sequential resonance
assignment of biopolymers from NMR data by relative-hypothesis-
prioritization-based simulated logic. J. Magn. Reson. 2005; 174: 133–151.
30 Morton CJ, Pugh DJ, Brown EL, Kahmann JD, Renzoni DAC, Campbell
ID. Solution structure and peptide binding of the SH3 domain from
human Fyn. Structure 1996; 4: 705–714.
31 Klages J, Coles M, Kessler H. NMR based screening: a powerful tool in
fragment-based drug discovery. Mol. Biosyst. 2006; 2: 318–23.
32 Dominguez C, Boelens R, Bonvin AM. HADDOCK: a protein-protein
docking approach based on biochemical or biophysical information.
J. Am. Chem. Soc. 2003; 125: 1731–1737.
J. Pept. Sci. 2013; 19: 214–219 Copyright © 2013 European Peptide Society and John Wiley & Sons, Ltd. wileyonlinelibrary.com/journal/jpepsci