Archives of Biochemistry and Biophysics 558 (2014) 14–27

Contents lists available at ScienceDirect

Archives of Biochemistry and Biophysics

journal homepage: www.elsevier.com/locate/yabbi

Endothelium-derived nitric oxide (NO) activates the NO-epidermal

growth factor receptor-mediated signaling pathway in bradykinin-
stimulated angiogenesis
Miriam S. Moraes c, Paulo E. Costa a, Wagner L. Batista b,g, Taysa Paschoalin b, Marli F. Curcio a,b,
Roberta E. Borges a, Murched O. Taha d, Fábio V. Fonseca h, Arnold Stern e,f, Hugo P. Monteiro a,⇑
a
Department of Biochemistry, Center for Cellular and Molecular Therapy – CTCMol, Universidade Federal de São Paulo, São Paulo, SP, Brazil
b
Department of Microbiology, Immunology, and Parasitology, Universidade Federal de São Paulo, São Paulo, SP, Brazil
c
Department of Biochemistry, Instituto de Química – Universidade de São Paulo, São Paulo, Brazil
d
Department of Surgery, Universidade Federal de São Paulo, São Paulo, SP, Brazil
e
Department of Pharmacology, New York University School of Medicine, New York, NY, USA
f
School of Medicine, Universidad Spíritu Santo – Ecuador, Guayaquil, Ecuador
g
Department of Biological Sciences, Universidade Federal de São Paulo/Campus Diadema, SP, Brazil
h
Department of Medicine, Institute for Transformative Molecular Medicine, Case Western University, Cleveland, OH, USA

a r t i c l e i n f o a b s t r a c t

Article history: Nitric oxide (NO) is involved in angiogenesis and stimulates the EGF-R signaling pathway. Stimulation of
Received 2 January 2014 different endothelial cell lines with bradykinin (BK) activates the endothelial NO synthase (eNOS) and
and in revised form 8 May 2014 promotes EGF-R tyrosine phosphorylation. Increase in NO production correlated with enhanced phos-
Available online 21 June 2014
phorylation of tyrosine residues and S-nitrosylation of the EGF-R. NO-mediated stimulatory effects on
tyrosine phosphorylation of the EGF-R, where cGMP independent. Inhibition of soluble guanylyl cyclase
Keywords: followed by BK stimulation of human umbilical vein endothelial cells (HUVECs) did not change tyrosine
Angiogenesis
phosphorylation levels of EGF-R. BK-stimulation of HUVEC promoted S-nitrosylation of the phosphatase
Bradykinin
Endothelial cells
SHP-1 and of p21Ras. Phosphorylation and activation of the ERK1/2 MAP kinases mediated by BK was
Epidermal growth factor receptor dependent on the activation of the B2 receptor, of the EGF-R, and of p21 Ras. Inhibition of BK-stimulated
Nitric oxide S-nitrosylation prevented the activation of the ERK1/2 MAP kinases. Furthermore, activated ERK1/2 MAP
S-nitrosylation kinases inhibited internalization of EGF-R by phosphorylating specific Thr residues of its cytoplasmic
domain. BK-induced proliferation of endothelial cells was partially inhibited by the NOS inhibitor
(L-NAME) and by the MEK inhibitor (PD98059). BK stimulated the expression of vascular endothelial growth
factor (VEGF). VEGF expression was dependent on the activation of the EGF-R, the B2 receptor, p21Ras,
and on NO generation. A MatrigelÒ-based in vitro assay for angiogenesis showed that BK induced the for-
mation of capillary-like structures in HUVEC, but not in those cells expressing a mutant of the EGF-R lack-
ing tyrosine kinase activity. Additionally, pre-treatment of BK-stimulated HUVEC with L-NAME, PD98059,
and with SU5416, a specific inhibitor of VEGFR resulted in inhibition of in vitro angiogenesis. Our findings
indicate that BK-mediated angiogenesis in endothelial cells involves the induction of the expression of
VEGF associated with the activation of the NO/EGF-R/p21Ras/ERK1/2 MAP kinases signaling pathway.
Ó 2014 Elsevier Inc. All rights reserved.

Introduction progression [1,2]. Binding of its endogenous ligand EGF promotes

receptor dimerization and trans-phosphorylation [1]. Hence, the
The epidermal growth factor receptor (EGF-R)1 is a key player in phosphorylation state of specific tyrosine residues at the EGF-R
the regulation of normal cell proliferation and initiation of tumor cytoplasmic domain determines the direct recruitment of

⇑ Corresponding author. Address: CTCMol, Universidade Federal de São Paulo, Rua Mirassol n° 207, CEP 04044-010 São Paulo, SP, Brazil.
E-mail address: hpmonte@uol.com.br (H.P. Monteiro).
1
Abbreviations used: NO, nitric oxide; BK, bradykinin; eNOS, endothelial NO synthase; HUVECs, human umbilical vein endothelial cells; VEGF, vascular endothelial growth
factor; EGF-R, epidermal growth factor receptor; PTP, protein tyrosine phosphatase; NOSs, NO synthases; iNOS, inducible NOS; nNOS, neuronal NOS; eNOS, endothelial NOS;
SNAP, S-nitroso-N-acetylpenicillamine; GPCR, G-protein coupled receptor; RAEC, rabbit aortic endothelial cells; FBS, fetal bovine serum; NEM, N-ethylmaleimide; BrdU, 5-bromo-
2-deoxyuridine; ANOVA, analysis of variance; KD, kinase-dead; WT, wild-type; NAC, N-acetylcysteine.

http://dx.doi.org/10.1016/j.abb.2014.06.011
0003-9861/Ó 2014 Elsevier Inc. All rights reserved.
 M.S. Moraes et al. / Archives of Biochemistry and Biophysics 558 (2014) 14–27 15

SH-2-containing signaling proteins [3]. It has been demonstrated Thus, we propose a new molecular mechanism by which NO
that SH-2 binds to SHP-1, an intracellular protein tyrosine phosphatase and S-nitrosylation and the NO/EGF-R/p21Ras/ERK1/2 MAP kinases
(PTP), and this complex is recruited to EGF-R [4]. The phosphorylated signaling pathway act in concert to regulate VEGF expression in
receptor is a docking site for adaptor proteins that assembles with endothelial cells stimulated with BK. Furthermore, signals derived
the small GTPase Ras leading to downstream activation of ERK1/2 from these pathways resulted in endothelial cell proliferation and
MAP kinases [1]. Furthermore, binding of EGF promotes EGF-R inter- angiogenesis.
nalization, down-regulation, degradation, and recycling [3,5].
Angiogenesis is a tightly regulated process of capillaries forma-
Materials and methods
tion from pre-existing blood vessels. Thus angiogenesis dysregula-
tion has been correlated with many human disorders such as:
Materials
chronic inflammation, diabetic retinopathy, and solid tumor forma-
tion [6,7]. Activation of EGF-R mediated signal transduction path-
The following reagents were purchased from Sigma–Aldrich (St.
ways has been associated with angiogenesis and vascular
Louis, MO, USA): BK, EGF, L-NAME (Nx-nitro-L-arginine methyl
endothelial growth factor (VEGF) expression [10]. VEGF is the prin-
ester hydrochloride), and the B2 receptor inhibitor HOE140. Anti-
cipal pro-angiogenic growth factor; it stimulates endothelial cell
ERK1/2, anti-phospho-ERK1/2 (Thr202/Tyr204), anti-eNOS, anti-
proliferation, migration, tube formation, and angiogenesis both
phospho-eNOS (Ser1177), anti-PTP101, and anti-P-Tyr1173 EGF-R
in vivo and in vitro [8]. Under physiologic conditions tissue vascula-
antibodies were from Cell Signaling Technologies (Beverly, MA,
ture can also be controlled by the modulation of VEGF expression [9].
USA). Anti EGF-R antibody was from Chemicon-Upstate Biotech-
VEGF and bradykinin (BK) are potent stimulators of macromo-
nology (Lake Placid, NY, USA). Phosphotyrosine (PY20) and SHP-1
lecular trans-endothelial flux. This dynamic process is character-
antibodies were from BD-Transduction Laboratories (Franklin
ized by cytoskeleton reorganization, intercellular openings
Lakes, NJ, USA). Anti-s-nitrosocysteine antibody was from AG
between adjacent endothelial cells, and by the release of nitric
Scientifics (San Diego, CA, USA). Appropriated secondary antibod-
oxide (NO) [11,12]. BK promotes angiogenesis by increasing vascu-
ies conjugated to horseradish peroxidase were from KPL (Gaithers-
lar permeability in the early phase of tumor development via B2
burg, MD, USA). The antibody anti-Pan-Ras and the inhibitors
receptor activation [13]. It was shown that NO can induce angio-
SU5416 (VEGF inhibitor), BIBX (EGFR inhibitor), and PD98059
genesis in vivo [14]. In mammalian cells, NO is generated by NO
(MEK inhibitor) were from Calbiochem (San Diego, CA, USA). Wes-
synthases (NOSs). NOSs enzymes produce NO through the catalytic
tern blotting developing reagents were obtained from Pierce
conversion L-arginine to L-citrulline, by the utilization of NADPH,
(Rockford, IL, USA).
O2 and BH4 as co-factors. Three NOSs isoforms are well character-
ized: inducible NOS (iNOS), and two constitutively expressed
enzymes, neuronal NOS (nNOS) and endothelial NOS (eNOS) [15]. Cell Cultures and transfections and generation of stable cell lines
In the endothelium eNOS release tightly controlled fluxes of NO Two endothelial cell lines were used in this study: Rabbit aortic
over short period of times to regulate vascular function [16]. NO endothelial cells (RAEC) and human umbilical vein endothelial
generated by eNOS in endothelium regulate vascular relaxation, cells (HUVECs). RAEC were kindly provided by Dr. Helena B. Nader
in part by the activation of guanylyl cyclase that raises cGMP levels (Department of Biochemistry at Universidade Federal de São Paulo,
[17]. In addition, S-nitrosylation of the thiol side chain of cysteine SP, Brazil). HUVEC were obtained from the American Type Culture
residues in proteins and peptides such as GSH by NO plays a major Collection. RAEC were cultivated in F12 medium supplemented
role in NO-mediated cell signaling events and protein–protein with 10% fetal bovine serum (FBS). HUVEC were cultivated in RPMI
interactions [18]. 1640 medium supplemented with 2.38 g/l Hepes and 10% FBS. All
We have previously demonstrated that physiological concentra- experiments were performed in serum-free medium at 37 °C and
tions of the nitrosothiols S-nitroso-N-acetylpenicillamine (SNAP) in a humidified atmosphere of 95% air and 5% CO2.
and S-nitrosoglutathione (GSNO) stimulate the EGF-R and the HUVEC express detectable levels of the EGF-R by western blot
p21Ras – ERK1/2 MAP kinases signaling pathways in several cell analysis. On the other hand, RAEC express low levels of the recep-
types. Activation of the p21Ras – ERK1/2 MAP kinases signaling tor. To obtain RAEC expressing detectable levels of a fully func-
pathway was correlated with: Ras S-nitrosylation; activation of tional EGF-R, cells were permanently transfected with the
the EGF-R stimulated by the ERK1/2 MAP kinases; promotion of prk5EGFR-WT plasmid vector. RAEC and HUVEC expressing a
cell cycle progression and stimulation of cell proliferation [19– non-functional (kinase-dead) EGF-R were obtained by permanent
23]. Conversely, others have shown an inhibitory action of NO over transfection with the prk5EGFR-KD plasmid vector [25]. Negative
EGF-R mediated tyrosine phosphorylation signaling events [24]. controls for all performed transfections were obtained with the
Therefore, the molecular mechanism by which NO modulates ectopic expression of the empty prk5 plasmid vector. In all cases,
EGF-R tyrosine phosphorylation and the physiological conse- cells were co-transfected with the pSV2-neo plasmid vector for
quences on cell signaling is still a focal point of controversies. selection purposes. Transfections were performed using the
In this work, we demonstrated that BK interacts with the G-pro- LipofectinÒ method (Invitrogen, Carlsbad, CA, USA). Cells stably
tein coupled receptor (GPCR) B2, stimulates endogenous NO pro- transfected were selected in 1 mg/ml GeneticinÒ (Invitrogen
duction, and promotes activation of EGF-R and its downstream Carlsbad, CA, USA) containing media. After isolating the colonies,
signaling pathway in endothelial cells. BK-mediated activation of the stable cell lines were maintained in 500 lg/ml GeneticinÒ
EGF-R occurred through the participation and/or interaction of containing media.
eNOS-derived NO, SHP-1, p21Ras, and the ERK1/2 MAP kinases.
BK/B2 promoted S-nitrosylation of EGF-R, SHP-1, and p21Ras. NO release in cell cultures supernatants
Downstream, activated ERK1/2 MAP kinases phosphorylated the Confluent HUVEC or RAEC cells monolayers were incubated in
cytoplasmic domain of EGF-R on specific threonine residues, starvation medium for 24 h, and then stimulated with BK for
thereby preventing receptor internalization and degradation. increasing period of time, in the presence or absence of inhibitors.
Moreover, endothelial cell proliferation and the expression levels Culture media were collected for NO measurements. A Nitric Oxide
of the pro-angiogenic factor VEGF were regulated by EGF-R/B2 Analyzer (NOA 280i, Sievers Instruments, Boulder, CO, USA) was
receptor/p21Ras and ERK1/2 MAP kinases, correlating positively utilized for the determination of NO metabolites nitrite/nitrate
with the endogenous production of NO in BK-stimulated cells. (NOx) in the samples. Collected media was injected into a
16 M.S. Moraes et al. / Archives of Biochemistry and Biophysics 558 (2014) 14–27

nitrogen-purge chamber containing vanadium (III) chloride in HCl Detection of S-nitrosylation using biotin derivatization coupled to
at 91 °C, which converts NOx to NO. For calibration, the area under Western blotting analysis
the curve was converted to micromolar NO using a sodium nitrite Detection of S-nitrosylated proteins was performed using the
standard curve as reference. Levels of NO detected in cell medium biotin switch method [27] with the following modifications. After
were corrected for background by subtracting the amount of NO treatments, harvested cells were rinsed two times with PBS con-
present in serum-free endothelial cell culture medium incubated taining 0.1 mM EDTA and 0.01 mM neocuproine and lysed in
under the same conditions in the absence of cells [26]. Experiments HEN buffer (25 mM Hepes, pH 7.7, 0.1 mM EDTA, 0.01 mM neo-
were performed at least in triplicate. cuproine) containing 0.5% 3-[(3-cholamidopropyl)dimethylammo-
nio]-1-propanesulfonate, 0.1% SDS, and 20 mM N-ethylmaleimide
Quantitative real-time reverse transcription polymerase chain (NEM) at 4 °C for 30 min to block free thiols. Lysates were cleared
reaction (qReal-time PCR) by centrifugation at 16,000g for 10 min at 4 °C, and excess NEM
Two to 5 lg of total RNA were treated with DNase (Invitrogen, was removed by protein precipitation by using cold acetone. Pro-
Carlsbad, CA, USA) according to the manufacturer’s instructions tein pellets were resuspended in HENS buffer (HEN + 1% SDS),
and an aliquot of 7.5 ll of the treated RNA was reverse-transcribed SNO bonds were decomposed by adding 20 mM sodium ascorbate,
to cDNA using the SuperScript First-Strand Synthesis System for and the resulting free thiols were reacted with 0.05 mM sulfhy-
RT-PCR (Invitrogen, Carlsbad, CA, USA). qReal-time PCR reactions dryl-specific biotinylating agent, N-(3-malemidylpropionyl) biocy-
were performed using SYBR Green PCR Master Mix (Applied tin (MPB, Molecular Probes) for 30 min at room temperature,
Biosystems, Foster City, CA, USA) in a GeneAmp 5700 Sequence resulting in biotinylation of S-nitrosothiols. After removal of excess
Detection System (Applied Biosystems, Foster City, CA, USA). Prim- MPB by another protein precipitation using acetone, SHP-1 and
ers were obtained from IDT (Coralville, IA, USA), and the sequences EGF-R were immunoprecipitated by using SHP-1 or EGF-R antibod-
optimized for real-time PCR were as follows: ies, respectively. Immunoprecipitates were washed three times
with HEN buffer and resuspended in Laemmli sample buffer, boiled
VEGF-A (SENSE) 50 CCT TCG TCC TCT CCT TAC C 30 at 95 °C for 5 min, loaded on 10% acrylamide gels, and transferred
VEGF-A (ANTI-SENSE) 50 ACA CAC AGC CAA GTC TCC 30 to nitrocellulose. Biotinylation of SHP-1 or EGF-R was detected on
FGF-B (SENSE) 50 GTC AAA CTA CAA CTC CAA GCA GAA G 30 the membrane by using horseradish peroxidase-linked streptavi-
FGF-B (ANTI-SENSE) 50 AGC CAG CAG CCG TCC ATC 30 din. To confirm equal amounts of the immunoprecipitated pro-
teins, biotinylated lysates were also subjected to Western
Western blotting analysis blotting for SHP-1, EGF-R, and p21Ras. All procedures until the bio-
Cells were starved for 24 h before pre-incubation with or with- tinylation step were performed in the dark. After incubation with
out inhibitors for 30 min and stimulated with BK for 5 min at 37 °C, appropriated HRP-conjugated secondary antibodies, blots were
in a humidified atmosphere of 95% air and 5% CO2. Cells were developed using the Super Signal system (Thermo-Pierce – Rock-
placed on ice, washed twice with ice-cold PBS. Lysates were ford, IL) and digitally registered on LAS-4000 (Fujifilm – Tokyo,
obtained by incubation with lysis buffer (50 mM Tris, pH 7.4, Japan).
150 mM NaCl, 0,25% sodium deoxycholate, 1% NP40, 1 mM NaF,
1 mM EDTA, 10 lg/m aprotinin, 10 lg/ml leupeptin, 1 mM PMSF, Detection of S-nitrosylation using the anti-s-nitrosocysteine antibody
1 mM) for 30 min on ice. Samples were then centrifuged at Western blot analysis using the anti-nitrosocysteine antibody
10,000g for 10 min and protein concentration on the supernatant was performed as previously described [22]. Briefly, at the indi-
was determined by Bradford. Protein content was adjusted, so that cated times of the treatment, cells were harvested and washed
each sample contained an equal amount of protein (50 lg/ml). three times in ice-cold PBS and lysed in ice-cold lysis buffer
Total cell lysates (50 lg/ml) were resolved in 10% SDS–PAGE gels (50 mM Tris–HCl pH 7.4, 150 mM NaCl, 1% NP-40, 0.25% deoxy-
and blotted onto Polyvinylidene fluoride (Immobilon-P – PVDF) cholic acid) 20 mM NEM). NEM was included to preserve the
Millipore (Billerica, MA) sheets. Blots were probed using specific protein thiol redox state present in intact cells by blocking all
antibodies against EGF-R, P-Tyr1173 EGF-R, ERK1/2, phospho- non-oxidized thiol groups, preventing further oxidation during
ERK1/2 (Thr202/Tyr204), eNOS, phospho-eNOS (Ser1177), subsequent processing. After incubation of the lysate on ice for
PTP101, Phosphotyrosine (PY20), SHP-1, p21Ras, and b-actin used 30 min, samples were centrifuged at 10,000g for 10 min and
as a protein loading control, when indicated. After incubation with the supernatant protein concentration was determined by Brad-
appropriated HRP-conjugated secondary antibodies, blots were ford. SHP-1 was immunoprecipitated using 1 lg of monoclonal
developed using the Super Signal system (Thermo-Pierce – Rock- SHP-1 antibody. After an overnight incubation at 4 °C, antigen anti-
ford, IL) and digitally registered on LAS-4000 (Fujifilm – Tokyo, body complexes were precipitated with 50 ll of protein G-agarose,
Japan). washed three times with lysis buffer, dissolved in 60 ll of electro-
phoresis non-reducing sample buffer, separated by electrophoresis
Immunoprecipitation on 10% SDS–PAGE gels, and transferred to PVDF membranes. Blots
For immunoprecipitation studies, lysates were incubated with were probed using specific antibodies against S-nitrosocysteine.
the immunoprecipitating antibody overnight at 4 °C followed by After incubation with appropriated HRP-conjugated secondary
additional 2 h incubation at 4 °C with protein G-Agarose. Lysates antibodies, blots were developed using the Super Signal system
were then centrifuged for 3 min at 10,000g and the supernatant (Thermo-Pierce – Rockford, IL) and digitally registered on LAS-
was discarded. Pellets were washed in phosphate-buffered saline 4000 (Fujifilm – Tokyo, Japan).
(3 times), resuspended in two times gel load buffer, and boiled
for 5 min prior to separation by SDS–polyacrylamide gel electro- 5-bromo-2-deoxyuridine (BrdU) incorporation assay
phoresis. Separated proteins were transferred onto PVDF mem- HUVEC cells were seeded (1.5  104 cells) on glass coverslips
branes. After transfer, the membrane was pre-incubated wit 5% and cultured for 24 h in RPMI supplemented with 10% FBS at
solution of blocking buffer for 1 h followed by overnight incubation 37 °C and 5% CO2. After 48 h cells were then starved overnight
with the indicated primary antibody. After incubation with appro- and then stimulated with BK (100 nM) for 12 h. Cells were labeled
priated HRP-conjugated secondary antibodies, blots were devel- with 100 lM 5-bromo-2-deoxyuridine (BrdU) for additional 12 h.
oped using the Super Signal system (Thermo-Pierce – Rockford, After labeling with BrdU, cells then harvested, washed and fixed.
IL) and digitally registered on LAS-4000 (Fujifilm – Tokyo, Japan). Fixed cells were incubated with an anti-BrdU antibody at room
 M.S. Moraes et al. / Archives of Biochemistry and Biophysics 558 (2014) 14–27 17

temperature. After incubation, cells were then washed and incu- Bradykinin stimulates NO production in various endothelial cell lines
bated with IgG-FITC and with DAPI (5 mg/ml) for additional
20 min. Cells were washed and finally analyzed by fluorescence Next, we performed experiments to determine if the expression
microscopy. levels of (B1) and (B2) BK receptors would change upon BK treat-
ment and/or stable EGF-R over-expression in RAECWT and in
RAECKD cell lines by RT-PCR. As a result, a significant, however,
Cell proliferation assay
modest increase in B1 receptor expression was detected in RAECKD
HUVEC cells were seeded on 35 mm dishes and cultured in
cells. Conversely, B1 expression was slightly decreased in RAECWT
RPMI supplemented with 10% FBS. After 48 h, cells were treated
(Supplementary Information, S1). B2 expression levels exhibited
with or without with 100 nM BK in RPMI at 37 °C and 5% CO2 for
no significant differences on both cell clones. Therefore, we can
24 h. After 24 h, cells were harvested using trypsin, collected, and
conclude that, there was no correlation between EGF-R stable
counted. Cell number was determined by using a hemacytometer.
over-expression and the B2 receptor expression level.
It has been previously described that BK associates with the B2
In vitro angiogenesis assay on MatrigelÒ receptor and activated eNOS [30]. Based on these findings, we then
The in vitro angiogenesis assay on MatrigelÒ was performed ask if our endothelial cell lines (RAEC, RAECWT, RAECKD, and
essentially as described by Paschoalin et al. [28]. Briefly, MatrigelÒ HUVEC), could raise NO levels when stimulated with 100 nM BK
(Becton Dickinson, Franklin Lakes, NJ, USA) was allowed to poly- in a time course manner. All endothelial cell lines generated and
merize in a 96-well plate for 1 h a 37 °C, producing a thin coat. maintained NO production throughout time, peaking at 1 min
HUVECs were seeded at a density of 1  104 per coated well in (Fig. 2A). These data match previous NO measurements performed
RPMI supplemented with or without 0.1% FBS; and with or without in bovine aortic endothelial cells [31]. Next, we demonstrated that
100 nM BK stimulation; and in the presence or absence of different acutely BK treatment (1 min) lead to phosphorylation of eNOS-
inhibitors. Cells were allowed to grow for 24 h in a humidified Ser1179, same time point correspondent to maximal eNOS-NO
atmosphere of 95% air and 5% CO2. Tube formation was observed production upon BK stimulation (Fig. 2B). Collectively these data
using bright field microscopy. Captured images were analyzed suggest that under our experimental conditions BK is signaling
and processed with NIH Image 1.62 software that calculates the through the B2 receptor.
total number of pixels. Pixels were counted in at least three ran-
dom areas, and the average value was determined for each sample. BK stimulates tyrosine phosphorylation and cysteine S-nitrosylation of
The untreated endothelial cells (control) were defined as 100% tube the EGF-R, in endothelial cells
formation. Stimulation BK was determined as the percent increase
above control (untreated). In a special condition where we used Previously published data by our group, demonstrated that
HUVEC transfected with the kinase-defective mutant EGF-R, the exogenous NO can promote ligand independent tyrosine phos-
stimulated sample was obtained after incubation with 10% FBS. phorylation of EGF-R in vitro on endothelial cells [19,20]. Indeed,
phosphorylation of EGF-R by tyrosine kinases in the absence of
Statistical analysis ligand binding, is triggered by oxidants and elicit activation of spe-
Data are expressed as mean ± S.D for a series of at least three cific signaling cascades, as well as stress-related responses within
independent experiments. Student’s t tests or analysis of variance cells [32]. To determine whether endogenous NO could promote
(ANOVA) with the Student–Newman–Keuls post-test used to com- EGF-R tyrosine phosphorylation, we carried out EGF-R tyrosine
pare mean values as appropriate. p Values <0.05 were considered phosphorylation experiments after BK treatment. Here we demon-
to represent significant differences. strated that BK treatment alone lead to EGF-R tyrosine phosphory-
lation and peak within 1 and 5 min, returning to baseline levels
after 15 min (Fig. 3A and B). However, RAECsWT1/A3 and HUVECs
Results when stimulated with BK were able to maintain elevated levels
of tyrosine phosphorylated EGF-R after 30 min (Fig. 3A and B). Fur-
Characterization of endothelial cell lines over-expressing functional or thermore, the BK–EGF-R tyrosine phosphorylation time courses
non-functional EGF-R described here, overlapped with time course experiments of NO
release depicted on (Figs. 2A and 3D). To further demonstrate the
Endothelial cells express specific genes according to their tissue involvement of NO in the BK–EGF-R tyrosine phosphorylation
of origin, artery or vein [29]. Human umbilical vein cells (HUVEC) process; HUVEC cells were pre-incubated with the NOS inhibitor
express detectable levels of EGF-R and phosphorylated EGF-R L-NAME and stimulated with BK for 1 min. Inhibition of NO
(Fig. 1A). Western blot analysis showed that transfection of HUVEC production by eNOS with L-NAME significantly decreased EGF-R
with the kinase-defective mutant EGF-R (HUVECKD) resulted in tyrosine phosphorylation levels up to 60% and BK-eNOS-NO pro-
almost complete inhibition of EGF-R tyrosine phosphorylation lev- duction by 80%; confirming the role of NO (Fig. 3C and D). As posi-
els upon stimulation either with 20 ng/ml EGF or 100 nM BK tive control, complete inhibition of BK-stimulated EGF-R tyrosine
(Fig. 1A). Rabbit aortic endothelial cells (RAEC) express low levels phosphorylation was achieved with 1 lM of BIBX (a potent inhib-
of EGF-R (data not shown). However, after transfection and antibi- itor of the EGF-R tyrosine kinase activity) (Fig. 3E). Furthermore, in
otic selection, stable clones of RAEC with varying levels of EGF-R order to exclude the involvement of cGMP on this process, BK-
expression were selected. KD4/C2 (clone 3) and WT1/A3 (clone induced-EGF-R phosphorylation experiments were performed in
6), respectively, expressing the kinase-dead (KD) form of the the presence of 1H-[1,2,4] Oxadiazolo [4,3-a] quinoxalin-1-one – ODQ
EGF-R and the wild-type (WT) form of the EGF-R were selected (a soluble and potent guanylyl cyclase inhibitor) (Fig. S1). These
for these studies. We compared the effects of EGF on phosphoryla- data indicate that BK/NO-mediated stimulatory effects on EGF-R
tion of WT EGF-R and KD EGF-R RAEC stable cell line (Fig. 1B). tyrosine phosphorylation are cGMP independent. We also evalu-
Unlike the parental RAEC, the WT stable RAEC cell line (RAECWT) ated the role of an individual phosphorylation site in EGF-R using
exhibited a greater level of EGF-R phosphorylation. In the KD sta- a specific antibody against Tyr1173 residue within the cytoplasmic
ble-RAEC (RAECKD) there was no significant increase on tyrosine domain of the EGF-R. BK treatment increased phosphorylation of
phosphorylation of the EGF-R upon EGF stimulation, as expected Tyr1173 over basal levels. BK-induced Tyr1173-EGF-R-phosphory-
(Fig. 1B). lation was further confirmed by the attenuation of this effect with
18 M.S. Moraes et al. / Archives of Biochemistry and Biophysics 558 (2014) 14–27

A HUVEC HUVECKD

p-EGF-R

EGF-R

- + - - + - EGF (20 ng/mL)

- - + - - + BK (100 nM)

B RAECKD RAEC RAECWT

IP: EGF-R
WB: p-Tyr
β-actin

- + - + - + - + - + EGF (20 ng/mL)

KD1/F3 KD4/C2 WT2/C3 WT1/A3
Fig. 1. Relative expression of wild-type (WT) and non-functional kinase dead (KD) EGF-R in transfected RAEC cells. (A) RAEC cells were mock transfected or transfected with
the indicated variant of the EGF-R. Total cell extracts from RAEC cells expressing WT and KD EGF-R were resolved by SDS–PAGE and analyzed by immunoblotting with the
anti-EGF-R monoclonal antibody. Western blot analysis of KD (Lanes 1–3) and WT EGF-R transfected cells (Lanes 4–6). (B) RAEC cells transfected with WT EGF-R or KD EGF-R,
as indicated. Cells were serum deprived (24 h) and stimulated with EGF (20 ng/ml) for 5 min. Cell extracts were immunoprecipitated with the anti-EGF-R antibody followed
by immunoblotting with an anti-p-Tyr antibody. Blots are representative of three independent experiments.

A RAEC RAECWT
10 20
Total NO release (µM)

Total NO release (µM)

8
15
6
10
4
5
2

0 0
0 1 5 15 30 min 0 1 5 15 30 min

RAECKD HUVEC
40 10
Total NO release (µM)

Total NO release (µM)

8
30
6
20
4
10
2

0 0
0 1 5 15 30 min 0 1 5 15 30 min

B HUVEC
p-eNOS

eNOS

0 1 5 15 30 min
100 nM BK

Fig. 2. Time-course for BK-stimulated activation of eNOS and NO release in endothelial cells. (A) HUVEC, RAEC, and RAEC-transfected cells either with WT (WT1/A3) or KD
(KD4/C2) variants of the EGF-R were serum-starved (24 h) and stimulated with BK (100 nM) for increasing periods (0–30 min). Released NO in the cell culture supernatants
was determined by a chemiluminescence based assay as described in Materials and methods. Values are significantly different from control (⁄p < 0.05; ⁄⁄p < 0.02; ⁄⁄⁄p < 0.01).
(B) HUVEC were serum-starved (24 h) and treated with BK (100 nM) for increasing periods (0–30 min). Cells were lysed, and total cell lysates were resolved on SDS–PAGE and
blotted with anti-phospho eNOS-Ser1177 antibody. Membranes were stripped of and probed against a specific anti-eNOS antibody. The immunoblot shown is representative
of three independent experiments.
 M.S. Moraes et al. / Archives of Biochemistry and Biophysics 558 (2014) 14–27 19

RAEC RAECWT1/A3
A IP: EGF-R
WB: p-Tyr
EGF-R

p-EGFR, Fold increase

1.6 * 3 *
* *
1.2
2
0.8
0.4 1

0 0
0 1 5 15 30 min 0 1 5 15 30 min
100 nM BK 100 nM BK

RAECKD4/C2 HUVEC
B IP: EGF-R
WB: p-Tyr

p-EGFR, Fold increase

EGF-R
* * 4 *
*
1.2
3
0.8 2
0.4 1
0 0
0 1 5 15 30 min 0 1 5 15 30 min
100 nM BK 100 nM BK

HUVEC
C IP: EGF-R D
WB: p-Tyr 250
*
EGFR *
% NO production 200
p-EGFR, Fold increase

3 * 150

2 100 #
*#
1 0.5

0 0
- + + BK (100 nM) - + + BK (100 nM)
- - + L-NAME (2 mM) - - + L-NAME (2 mM)

E HUVEC F
IP: EGF-R IP: EGF-R
WB: p-Tyr1173
WB: p-Tyr

EGF-R EGF-R
- + + BK (100 nM)
- 0.01 0.1 1.0 10.0 - BIBX (µM) - - + HOE140 (10 μM )
+ + + + + - BK (100 nM)

HUVEC
G
SNO-EGF-R

EGF-R

2.5
*

2
Fold Increase

1.5
1

0.5
0
- + + BK (100 nM)
- - + HOE 140 (10 µM)

Fig. 3. NO generation modulates tyrosine phosphorylation and cysteine S-nitrosylation of the EGF-R in BK-stimulated endothelial cell lines. (A) RAEC, and RAEC-transfected
cells with the WT variant of the EGF-R (WT1/A3) were serum-starved (24 h) and stimulated with BK (100 nM) for increasing periods (0–30 min), at 37 °C. Cell extracts were
immunoprecipitated with the anti-EGF-R antibody followed by immunoblotting with an anti-p-Tyr antibody. Relative densitometric values for phosphorylated bands are
shown in columns (⁄p < 0.05). Blots are representative of three independent experiments. (B) HUVEC, and RAEC-transfected cells with the KD variant of the EGF-R (KD4/C2)
were serum-starved (24 h) and stimulated with BK (100 nM) for increasing periods (0–30 min), at 37 °C. Cell extracts were immunoprecipitated with the anti-EGF-R antibody
followed by immunoblotting with an anti-p-Tyr antibody. Relative densitometric values for phosphorylated bands are shown in columns (⁄p < 0.05). Blots are representative
of three independent experiments. (C) HUVEC were serum-starved (24 h) and stimulated with BK (100 nM) during 1 min at 37 °C. Cells were incubated for 30 min with 2 mM
L-NAME before incubation with BK, where indicated. Cell extracts were immunoprecipitated with the anti-EGF-R antibody followed by immunoblotting with an anti-p-Tyr
antibody. Relative densitometric values for phosphorylated bands are shown in columns (⁄p < 0.05). Blots are representative of three independent experiments. (D) Released
NO in the cell culture supernatants was determined by a chemiluminescence based assay as described in Materials and methods. Data represent means ± SD from at least
three independent experiments. Values are significantly different from control (⁄p < 0.05). Values are significantly different from BK-treated cells (#p < 0.05). (E) HUVEC were
serum-starved (24 h) and stimulated with BK (100 nM) during 1 min at 37 °C. Cells were incubated for 30 min with increasing concentrations of BIBX (EGF-R inhibitor) before
incubation with BK, where indicated. Cell extracts were immunoprecipitated with the anti-EGF-R antibody followed by immunoblotting with an anti-p-Tyr antibody. Blots
are representative of three independent experiments. (F) HUVEC were serum-starved (24 h) and stimulated with BK (100 nM) during 1 min at 37 °C. Cells were incubated for
30 min with 10 lM HOE140 (B2 receptor antagonist) before incubation with BK, where indicated. Cell extracts were immunoprecipitated with the anti-EGF-R antibody
followed by immunoblotting with an anti-p-Tyr antibody. Blots are representative of three independent experiments. (G) HUVEC were serum-starved (24 h) and stimulated
with BK (100 nM) during 1 min at 37 °C. Cells were incubated for 30 min with 10 lM HOE140 before incubation with BK, where indicated. EGF-R cysteine S-nitrosylation was
detected in cell extracts using the biotin derivatization assay. Relative densitometric values for nitrosylated bands are shown in columns (⁄p < 0.05). Blots are representative
of three independent experiments.
20 M.S. Moraes et al. / Archives of Biochemistry and Biophysics 558 (2014) 14–27

experiments using the potent and selective BK B2 receptor antag- receptor activity by the selective antagonist HOE140 resulted in
onist HOE140 (Fig. 3F). inhibition of SHP-1 S-nitrosylation (Fig. 4A). In another set of
Tyrosine phosphorylation and cysteine S-nitrosylation have experiments, pre-incubation with L-NAME inhibited BK-induced
been associated with the activation of EGF-R signaling pathway SHP-1 S-nitrosylation, indicating the dependence on eNOS activity
in vitro on human basal-like breast cancer cells [33]. Based on these (Fig. 4B). Collectively, our findings demonstrate that in addition to
findings, we next examined whether EGF-R was S-nitrosylated in EGF-R, SHP-1 is another important target for BK-induced
BK-stimulated HUVECs and HUVECsKD cells. Treatment of HUVECs S-nitrosylation.
with BK resulted in endogenous EGF-R S-nitrosylation. EGF-R
S-nitrosylation levels were decreased with the B2 antagonist
HOE140 (Fig. 3G). Thus, we have demonstrated that BK binding
BK promotes cysteine S-nitrosylation of p21Ras and stimulates the
to B2 receptor lead to: an increase in NO production by eNOS, to
phosphorylation of ERK1/2 MAP kinases
ligand-independent EGF-R Tyr1173 phosphorylation, and EGF-R
S-nitrosylation.
Recently, our group demonstrated that BK stimulated p21Ras
activity in HUVEC cells in vitro [23]. Following up with this discov-
Cysteine S-nitrosylation of protein tyrosine phosphatase SHP-1 is ery, in this work, we demonstrated that NO derived from BK stim-
associated with BK-mediated activation of the EGF-R ulation promotes S-nitrosylation of p21Ras. Additionally,
inhibition of B2 receptor activity with the antagonist HOE140
Both SH-2 domains of the tyrosine phosphatase SHP-1, associ- resulted in inhibition of p21Ras S-nitrosylation. Pre-incubation
ates with phospho-Tyr1173 at the cytoplasmic domain of the with L-NAME inhibited BK-mediated p21Ras S-nitrosylation, again
EGF-R, this interaction negatively regulates receptor activity indicating the dependence on NO produced by eNOS activity
through dephosphorylation [4,34]. In addition, mild nitrosative (Fig. 5A). Downstream p21Ras, BK stimulates phosphorylation of
stress conditions can inhibit EGF-R dephosphorylation through the ERK1/2 MAP kinases. Thus, inhibition of B2 receptor activity
S-nitrosylation of a single conserved Cys residue at the catalytic with the antagonist HOE140 resulted in inhibition of ERK1/2
domain of SHP-1 [35,36]. For that reason, we investigated whether MAP kinases phosphorylation. Pre-incubation either with the
endogenous NO-derived from BK stimulation would promote reducing agent N-acetylcysteine (NAC), or with farnesyltransferase
S-nitrosylation of SHP-1 in HUVECs. Exposure of HUVECs to BK inhibitor of p21Ras (FPTII), resulted in inhibition of phosphoryla-
resulted in S-nitrosylation of SHP-1. In addition, inhibition of B2 tion of ERK1/2 MAP kinases (Fig. 5B).

A IP: SHP-1
WB: SNO-Cys

SHP-1

- + + BK (100 nM)
- - + HOE140 (10 μM)

HUVEC
B
IP: SHP-1
WB: SNO-Cys
SHP-1
3.5
*
SNO SHP-1, Fold increase

3
*
2.5 *
2
1.5
1
0.5
0
0 1 5 30 1 5 30 min (BK 100 nM)
- - - - + + + L-NAME (2 mM)
Fig. 4. BK stimulated cysteine S-nitrosylation of SHP-1. (A) HUVEC were serum-starved (24 h) and stimulated with BK (100 nM) during 1 min at 37 °C. Cells were incubated
for 30 min with 10 lM HOE140 before incubation with BK, where indicated. After immunoprecipitation with an antibody against SHP-1, S-nitrosylation of SHP-1 was
detected using an anti-nitrosocysteine (anti-SNO-Cys) antibody. Blots are representative of three independent experiments. (B) HUVECs were exposed to 100 nM BK in the
indicated period (0–30 min). Cells were incubated for 30 min with 2 mM L-NAME before incubation with BK, where indicated. After immunoprecipitation with an antibody
against SHP-1, S-nitrosylation of SHP-1 was detected using the anti-SNO-Cys antibody. The immunoblot shown is representative of three independent experiments. Averaged
data quantified by laser-densitometry of immunoblots, expressed as relative intensity of the bands (means ± SD for three independent experiments. ⁄p < 0.05 vs. control cells).
 M.S. Moraes et al. / Archives of Biochemistry and Biophysics 558 (2014) 14–27 21

A A RAEC RAECWT2/B4
SNO-Ras p-ERK1/2
ERK1/2

p-ERK, Fold increase

3 * 6
Ras *
2 4
- + + + BK (100 nM) 1 2
- - + - HOE140 (10 µM) 0
0 1 5 15 30 min 0
- - - + L-NAME (2 mM) 0 1 5 15 30 min
100 nM BK 100 nM BK

B HUVEC RAECKD4/C2
p-ERK p-ERK1/2
ERK1/2

p-ERK, Fold Increase

p-ERK, Fold Increase

6
* *
ERK 2 4

1 2
0.5 * 0
p-ERK, Fold increase

0
0 1 5 15 30 min 0 1 5 15 30 min
0.4 100 nM BK 100 nM BK

0.3 RAECKD4/C2
B
p-ERK1/2
0.2
ERK1/2
0.1

p-ERK, Fold Increase

5
*
0 4
- + + + + BK (100 nM) 3
- - + - - HOE140 (10 µM) 2
- - - + - NAC (10 mM) 1
- - - - + FPT III (25 µM) 0
0 1 5 30 1 5 30 min
Fig. 5. S-nitrosylation of p21Ras and activation of the ERK1/2 MAP kinases in BK EGF
response to BK stimulation of HUVEC. (A) HUVEC were serum-starved (24 h) and
stimulated with BK (100 nM) for increasing 5 min. Cells were incubated for 30 min C HUVEC
with 2 mM L-NAME, or with 10 lM HOE140 before incubation with BK, where
indicated. p21Ras cysteine S-nitrosylation was detected in cell extracts using the p-ERK1/2
biotin derivatization assay. The immunoblot shown is representative of three
independent experiments. (B) HUVEC were serum-starved (24 h) and stimulated ERK1/2
with BK (100 nM) for 30 min. Cells were incubated for 30 min with 10 lM HOE140,
or 10 mM NAC, or 25 lM FPT II before incubation with BK, where indicated. Cells - + + + BK (100 nM)
were lysed, and total cell lysates were resolved on SDS–PAGE and blotted with anti- - - + - L-NAME (2 mM)
phospho ERK1/2 MAP kinases antibody. Membranes were stripped of and probed - - - + PD98059 (20 µM)
against a specific anti-ERK1/2 MAP kinases antibody. Relative densitometric values
for phosphorylated bands are shown in columns (⁄p < 0.05). The immunoblot shown Fig. 6. ERK1/2 MAP kinases activation in response to BK stimulation of endothelial
is representative of three independent experiments. cell lines. (A) RAEC, RAEC-transfected cells either with WT (WT2/B4) or KD (KD4/
C2) variants of the EGF-R, and HUVEC were serum-starved (24 h) and stimulated
with BK (100 nM) for increasing periods (0–30 min). Cells were lysed, and total cell
lysates were resolved on SDS–PAGE and blotted with anti-phospho ERK1/2 MAP
Differential phosphorylation profiles of ERK1/2 MAP kinases kinases antibody. Membranes were stripped of and probed against a specific anti-
stimulated by BK in functional versus non-functional EGF-R expressing ERK1/2 MAP kinases antibody. Relative densitometric values for phosphorylated
bands are shown in columns (⁄p < 0.05). The immunoblots shown are representative
endothelial cells
of three independent experiments. (B) RAEC-transfected cells with a KD variant of
the EGF-R (KD4/C2) were serum-starved (24 h) and stimulated either with BK
Stimulation of a variety of cell lines with BK promotes phos- (100 nM) or with EGF (20 ng/ml) for increasing periods (0–30 min). Cells were
phorylation and activation of the ERK1/2 MAP kinases [37]. Here, lysed, and total cell lysates were resolved on SDS–PAGE and blotted with anti-
we predicted that stimulation of endothelial cells expressing a fully phospho ERK1/2 MAP kinases antibody. Membranes were stripped of and probed
against a specific anti-ERK1/2 MAP kinases antibody. Relative densitometric values
functional (RAEC, RAECWT2/B4, and HUVEC) or a non-functional for phosphorylated bands are shown in columns (⁄p < 0.05). The immunoblot shown
EGF-R (RAECKD4/C2) with BK, could result in differential phosphor- is representative of three independent experiments. (C) HUVEC were serum-starved
ylation profiles of the ERK1/2 MAP kinases. Indeed, stimulation of (24 h) and stimulated with BK (100 nM) during 5 min at 37 °C. Before incubation
endothelial cells expressing a fully functional EGF-R with BK dis- with BK, cells were incubated for 30 min either with 2 mM L-NAME or with 20 lM
PD98059, where indicated. Cell lysates were resolved on SDS–PAGE and blotted
played a sustained phosphorylation profile of the ERK1/2 MAP
with anti-phospho ERK1/2 MAP kinases antibody. Membranes were stripped of and
kinases (Fig. 6A). Conversely, stimulation of RAEC KD4/C2 (non- probed against a specific anti-ERK1/2 MAP kinases antibody. The immunoblot
functional EGF-R) with BK showed a rapid ERK1/2 MAP kinase shown is representative of three independent experiments.
phosphorylation profile (1 min), followed by fast dephosphoryla-
tion (Fig. 6 A and B). In addition, stimulation of RAECKD4/C2 with
EGF did not result in stimulation of phosphorylation of the ERK1/ kinase MEK caused a marked reduction on phosphorylation levels
2 (Fig. 6B). Similar findings were obtained when we stimulated of ERK1/2 MAP kinases in HUVEC stimulated by BK, whereas, the
HUVECKD (not shown). In addition, inhibition of the upstream NOS inhibitor L-NAME did not show significant effects (Fig. 6C).
22 M.S. Moraes et al. / Archives of Biochemistry and Biophysics 558 (2014) 14–27

BK promotes ERK-dependent phosphorylation of EGF-R at a specific [20, and Fig. 6C]. These observations led us to examine the possibil-
site ity of a putative EGF-R phosphorylation pattern that could be med-
iated by the ERK1/2 MAP kinases, when endothelial cells are
Next, we show that BK similarly to EGF, stimulated tyrosine stimulated with BK. We used a specific mouse monoclonal
phosphorylation of the EGF-R and phosphorylation of downstream antibody that is immunoreactive with a consensus sequence con-
ERK1/2 MAP kinases (Fig. 7A). Previous published data from our taining a phospho-Thr residue followed by a Pro residue [38].
group, demonstrated that incubation of RAEC cell with physiolog- P-Thr/Pro was strongly evident in HUVEC and RAEC cell lines stim-
ical concentrations of NO donors resulted in stimulation of the ulated with BK, conversely EGF treated cells displayed weaker
EGF-R tyrosine phosphorylation. In addition, PD98059 (a MEK immunoreactivity. P-Thr/Pro enhanced immunoreactivity paral-
inhibitor) blocked NO-stimulated EGF-R tyrosine phosphorylation leled with the pattern of activation of ERK1/2 MAP kinases

HUVEC RAECWT1/A3
A IP: EGF-R
WB: p-Tyr

p-ERK1/2

EGF-R
- + - - + - BK (100 nM)
- - + - - + EGF (20 ng/mL)

HUVEC RAECWT1/A3
B IP: EGF-R
WB: PTP101
IP: EGF-R
WB: pTyr

EGF-R

p-ERK1/2

ERK1/2
- + - - + - BK (100nM)
- - + - - + EGF (20 ng/mL)

HUVEC RAEC WT1/A3

C IP: EGF-R
WB: PTP101
EGF-R
- + - - + - - + - - + - BK (100nM)
- - + - - + - - + - - + EGF (20 ng/mL)
- - - + + + - - - + + + PD98059 (20 µM)

D HUVEC RAECWT1/A3
IP: EGF-R
WB: PTP101
EGF-R
β-actin
0 1 2 4 1 2 4 0 1 2 4 1 2 4 h
EGF BK EGF BK

Fig. 7. BK stimulated tyrosine phosphorylation and the ERK consensus phosphorylation sites on the EGF-R in endothelial cell lines. (A) Serum starved (24 h) HUVEC and RAEC
transfected with a WT variant of the EGF-R (WT1/A3) were incubated with BK (100 nM) or EGF (20 ng/ml) for 15 min. Cell lysates were immunoprecipitated with anti-EGFR
antibody followed by immunoblotting with an anti-p-Tyr antibody. Cell lysates were resolved on SDS–PAGE and blotted with anti-phospho ERK1/2 MAP kinases antibody.
The immunoblots shown are representative of three independent experiments. (B) Serum starved (24 h) HUVEC and RAEC transfected with a WT variant of the EGF-R (WT1/
A3) were incubated with BK (100 nM) or EGF (20 ng/ml) for 15 min. Cell lysates were immunoprecipitated with anti-EGFR antibody followed by immunoblotting with the
anti-PTP101 antibody which recognizes ERK1/2 MAP kinases consensus phosphorylation sites (upper panel). Cell lysates were immunoprecipitated with anti-EGFR antibody
followed by immunoblotting with an anti-p-Tyr antibody (mid panel). Cell lysates were resolved on SDS–PAGE and blotted with anti-phospho ERK1/2 MAP kinases antibody.
Membranes were stripped of and probed against a specific anti-ERK1/2 MAP kinases antibody (lower panel). The immunoblots shown are representative of three independent
experiments. (C) Serum starved (24 h) HUVEC and RAEC transfected with a KD variant of the EGF-R (KD4/C2) were incubated with BK (100 nM) or EGF (20 ng/ml) for 15 min.
Before incubation with BK or EGF, where indicated cells were incubated for 30 min with 20 lM PD98059 a MEK inhibitor. Cell lysates were immunoprecipitated with anti-
EGFR antibody followed by immunoblotting with the anti-PTP101 antibody and with the EGF-R antibody. The immunoblots shown are representative of three independent
experiments. (D) serum starved HUVEC and RAEC transfected with a WT variant of the EGF-R (WT1/A3), were pre-incubated with 20 lM PD98059 for 30 min prior to
stimulation with 100 nM BK or 20 ng/ml EGF for 15 min. Cell lysates were immunoprecipitated with anti-EGF-R antibody followed by immunoblotting with the PTP101
antibody (upper panels). Immunoblotting was also performed with the anti-EGF-R antibody (mid panel). Immunoblotting with an anti-b-actin antibody is shown in the lower
panel. Results shown are representative of three independent experiments.
 M.S. Moraes et al. / Archives of Biochemistry and Biophysics 558 (2014) 14–27 23

(Fig. 7B). Pre-incubation of cells with MEK inhibitor (PD98059) fol- For this reason, we next, carried out experiments to examine
lowed by stimulation either with BK or EGF, resulted in strong inhi- how VEGF-A mRNA expression could be modulated by the inhibi-
bition of pro-directed EGF-R-Thr phosphorylation (Fig. 7C). tion of several signaling components: EGF-R phosphorylation
It has been described that BK promotes ERK1/2 MAP kinases (BIBX), bradykinin B2 receptor activity (HOE140), p21Ras (FPTII),
activation. Activated ERK1/2 directly phosphorylate EGF-R at Ser/ and endogenous NO production (L-NAME). Consequently, all these
Thr residues, this post-translational modification confers greater pharmacological compounds promoted down-regulation of VEGF-
receptor stability [39,40]. Based on these data, we evaluated EGF- A mRNA expression, to the same extent, after BK stimulation in
R stability on HUVEC and RAECWT1/A3 endothelial cell lines submit- HUVEC cells (Fig. S2).
ted to chronic treatments (up to 4 h) with either EGF or with BK. We also examined whether the EGF-R could influence on the
After 1-h stimulation, EGF promoted EGF-R internalization and production of fibroblast growth factor-B, another important
degradation. On the other hand, BK stimulation maintained EGF- angiogenic factor [43], in HUVEC stimulated with BK. BK did not
R expression levels up to 4 h of continuous stimulation (Fig. 7D). stimulate the expression of the mRNA levels of fibroblast growth
All together, these data illustrate the molecular mechanism by factor-B (not shown).
which BK treatment promoted a sustained signaling downstream
EGF-R associated with threonine phosphorylation of the receptor. NO, EGF, EGF-R, ERK1/2 MAP kinases, and VEGF mediated the effects of
BK on HUVEC in vitro angiogenesis
Participation of NO, EGF-R, and the ERK1/2 MAP kinases on BK-
stimulated proliferation of HUVEC We next examined the role of EGF-R, and the ability of BK to
stimulate formation and stabilization of capillary-like pro-
The angiogenic process is dependent on endothelial cell growth angiogenic structures on HUVECs cultured on MatrigelÒ (in vitro
stimulation [9]. Accordingly, we next determined the BK effects on angiogenesis). Capillary-like pro-angiogenic structures were signif-
HUVEC cell proliferation in two ways: by measuring BrdU incorpo- icantly inhibited in stable HUVEC cell lines over-expressing the
ration (DNA synthesis) and cell counting. BK treatment leads to an EGF-R kinase-defective mutant after BK or EGF stimulation
increase in BrdU incorporation (Fig. 8 left panel) that correlated (Fig. 9A). Moreover, inhibition of EGF-R with 1 lM BIBX also
with an increment of the total cell (Fig. 8 – right panel). To assess resulted in substantial inhibition of capillary-like pro-angiogenic
the role played by: NO, EGF-R activation and ERK1/2 MAP kinases structures formation (Fig. 9B). Accordingly, inhibition of ERK1/2
in BK-induced endothelial cell proliferation we performed the MAP kinases (key downstream EGF-R signaling pathway effectors)
same experiments, but in the presence of specific inhibitors. The with its selective inhibitor PD98059 resulted in inhibition of BK-
list of inhibitors employed comprises: L-NAME (NOS inhibitor), stimulated in vitro angiogenesis (Fig. 9B). We also assessed the
BIBX (a specific EGF-R inhibitor [41]), and PD98059 (MEK inhibi- direct participation of NO in BK-induced in vitro angiogenesis by
tor). Thus, sub-confluent HUVEC cultures were pre-treated with inhibiting eNOS activity with L-NAME. Formation of capillary-like
these inhibitors for 30 min prior to BK exposure. L-NAME pre- pro-angiogenic structures was abolished after eNOS inhibition
sented a slight but significant inhibition of DNA synthesis and cell (Fig. 9B). Finally we verified the implication of VEGF and its cog-
growth, whereas BIBX was capable of blocking both processes. nate receptor in this process. Pre-incubation of HUVEC cells with
PD98059 also suppressed BK-induced HUVEC proliferation (Fig. 8 the specific inhibitor of the VEGF receptor (SU5416) prevented
– right panel). BK-stimulated formation of capillary-like pro-angiogenic struc-
tures (Fig. 9C).
EGF-R, B2 receptor, p21Ras and NO regulate the expression levels of
VEGF in BK stimulated HUVEC Discussion

VEGF stimulates several cellular responses including: prolifera- BK-induced endogenous NO production in endothelial cells pro-
tion, migration, tube formation, and angiogenesis of endothelial motes cell proliferation and angiogenesis. Inasmuch, B2 receptor
cells, both in vivo or in vitro [8]. It has been previously shown that activation by BK stimulates EGF-R tyrosine phosphorylation. EGF-
inhibition of EGF-R signaling pathways inhibit EGF-R-dependent R is phosphorylated in response to GPCR agonists in a process
VEGF production in human micro-vascular endothelial cells [42]. defined as trans-activation [44,45]. Positive or negative

*
* 160
BrdU incorporation

120
Proliferation
(% Control)

#
(% Control)

# # 120 #
# #
80
80
40 40

0 0
- + + + + - + + + + BK (100nM)
- - + - - - - + - - L-NAME (2 mM)
- - - + - - - - + - PD98059 (20 µM)
- - - - + - - - - + BIBX (1 µM)

Fig. 8. Effects of L-NAME, PD98059 and the EGFR inhibitor BIBX on BK-induced proliferation of endothelial cells. (A) Serum-starved (24 h) HUVECs were pre-incubated for
30 min with increasing concentrations (0–10 lM) of the EGF-R inhibitor BIBX. Cell lysates were immunoprecipitated with the anti-EGF-R antibody followed by
immunoblotting with an anti-p-Tyr antibody. Immunoblotting was also performed with the anti-EGF-R antibody. Results shown are representative of at least three
independent experiments. (B – Left panel) DNA synthesis was determined by incorporation of BrDU in cells exposed to 100 nM BK for 24 h. Where indicated, L-NAME,
PD98059, and BIBX were added. Labeled nuclei were counted with the aid of an ocular grid at 100. Data are presented as the total labeled nuclei counted per slide
(means ± SD for three independent experiments. ⁄p < 0.05 vs. control cells). (B – Right panel) HUVEC were seeded on a 24-wells plate at 2  104 cells per well and cultured
overnight in complete medium until cell attachment. Cells were incubated for 24 h in the presence or absence of 100 nM BK. Where indicated, L-NAME, PD98059, and BIBX
were added. Cells were trypsinized, collected, and counted in a hemacytometer. Values are reported in the bar graphs and expressed as means ± SD (n = 3; ⁄p < 0.01 vs. control
cells; #p < 0.05 vs. control cells).
24 M.S. Moraes et al. / Archives of Biochemistry and Biophysics 558 (2014) 14–27

HUVECKD
Control BK EGF FBS
A

% Tube formation
400
*
300
200
100
0
- + - - BK (100 nM)
- - + + EGF (20 ng/mL)
- - - + FBS (10%)

HUVEC
B Control BK BK/BIBX BK/PD98059 BK/L-NAME

300
% Tube formation

*
200

100

0
- + + + + BK (100 nM)
- - + - - BIBX (1μM)
- - - + - PD98059 (20μM)
- - + - + L-NAME (2 mM)

C Control BK BK + SU5416
% Tube formation

300 *
HUVEC 200

100
0
- + + BK (100 nM)
- - + SU5416 (1μM)

Fig. 9. Effects of EGF-R, ERK1/2 MAP kinases, NO, and VEGF on BK-stimulated formation of capillary-like pro-angiogenic structures (Tubes) derived from cultured endothelial
cells. HUVEC and HUVEC transiently transfected with a KD variant of the EGF-R (K745A) were plated on MatrigelÒ in medium supplemented with 0.2% FBS in the presence of
BK (100 nM). Phase-contrast pictures were taken for all experimental situations. A red arrow points to capillary-like pro-angiogenic structures. (A) Addition of 20 ng/ml EGF,
or 10% FBS, or 100 nM BK to HUVECKD cultured cells. Three independent experiments were performed and one representative picture of four different treatments is shown.
The formation and quantification of capillary-like pro-angiogenic structures was determined as described in Materials and methods. The number of pro-angiogenic structures
was counted after 24 h and bar graphs expressing mean values ± SD are shown. (B) Addition of 1 lM BIBX, 20 lM PD98059 or 2 mM L-NAME to HUVEC cultured cells
stimulated with 100 nM BK. Three independent experiments were performed and one representative picture of three different treatments is shown. The number of pro-
angiogenic structures was counted after 24 h and bar graphs expressing mean values ± SD are shown. (C) Addition of 1 lM SU5416 to HUVEC cultured cells stimulated with
100 nM BK. Three independent experiments were performed and one representative picture of three different treatments is shown. The number of pro-angiogenic structures
was counted after 24 h and bar graphs expressing mean values ± SD are shown. (For interpretation of the references to color in this figure legend, the reader is referred to the
web version of this article.)

modulation of receptor tyrosine kinase activities by GPCRs, such as, signaling pathways associated with eNOS Ser1177 phosphoryla-
the BK B2 receptor will depend on tissue and/or cell type [6,13]. tion and EGF-R Tyr1173 phosphorylation. Although independent,
BK stimulation of HUVEC and RAEC endothelial cells resulted in these signaling pathways might cross talk. Indeed we demon-
EGF-R tyrosine phosphorylation. BK-mediated EGF-R tyrosine strated that EGF-R phosphorylation was down-regulated in the
phosphorylation is dependent on a fully functional EGF-R. Con- presence of L-NAME (NOS inhibitor). In addition, NO mediated
versely, BK-stimulated NO production is independent of the integ- EGF-R activation was coupled to B2 receptor activation, occurring
rity of the EGF-R; suggesting the existence of two independent at different levels, and there is a possibility of interactions of two
 M.S. Moraes et al. / Archives of Biochemistry and Biophysics 558 (2014) 14–27 25

posttranslational modifications (S-nitrosylation and phosphoryla- be that the extent of S-nitrosylation and/or the percentage of the
tion). In that same context, mouse embryo fibroblasts when incu- receptor population that is S-nitrosylated might be different when
bated with the nitrosothiol, SNAP, results in S-nitrosylation and NO is generated endogenously. Experiments aiming the identifica-
Tyr416 Src kinase phosphorylation. In this case, both posttransla- tion of EGF-R-cysteine sites that undergo S-nitrosylation upon BK
tional modifications act in concert to activate Src kinase activity stimulation are currently being performed in our laboratory.
[23]. Our present findings on NO-stimulated EGF-R Tyr1173 phos- Here we show that SHP-1, a SH2 PTP domain that is recruited to
phorylation and S-nitrosylation are another evidence of positive EGF-R when phosphorylated at Tyr1173 [36], is S-nitrosylated when
cooperation between both posttranslational modifications in the endothelial cells are stimulated with BK. The unique microenviron-
activation of downstream EGF-R signaling pathways. ment surrounding the active site of PTPs confers the conserved
Previous published observations have demonstrated that Cys215 residue an unusually low pKa, thus being highly susceptible
S-nitrosylation of Cys166 and Cys305 on EGF-R at the ligand-bind- to oxidation or S-nitrosylation [32,46]. Moreover, it is known that
ing-pocket inhibits EGF-R tyrosine phosphorylation and activation. mild oxidative/nitrosative stress induces a transient S-nitrosylation
However one can say that these data were obtained in the presence of a single cysteine at the active site on the SHP-1 and SHP-2 pro-
of the NO donor DEA-NO (1, 1-diethyl-2-hydroxy-2-nitrosohydr- teins thereby inactivating the SH2 domain that down-regulate the
azine), therefore raising a possibility that EGF-R activity can be dif- PTP activity, [35]. In this work, incubation of endothelial cells with
ferentially modulated when NO is generated endogenously by L-NAME prevented endogenous SHP-1 S-nitrosylation and EGF-R
eNOS in endothelial cells [24]. Although the conclusions of both tyrosine phosphorylation, providing additional evidence that
studies are conflicting, the experimental conditions used on them S-nitrosylation is a positive regulator of the EGF-R signaling path-
are not comparable. Murillo-Carretero et al. [24] employed non- way in BK-stimulated endothelial cells. These findings are in agree-
physiological conditions by exposing human neuroblastoma ment with previous observations describing of EGF-R mediated
NB69 cells to millimolar concentrations of the NO donor DEA- tyrosine phosphorylation signaling in a number of cell types stimu-
NO. On the other hand, in this work, we demonstrate that endog- lated with exogenous NO sources [19,33].
enously NO produced by endothelial cells after BK stimulation In endothelial cells, the S-nitrosothiol SNAP promotes p21Ras
increases EGF-R Tyr1173 tyrosine phosphorylation promoting S-nitrosylation and activation of ERK1/2 MAP kinases [20,21].
receptor activation [3]. Furthermore, our findings strongly suggest Our results confirm, and extend these observations because,
that endogenous NO may S-nitrosylate a different set of cysteine endogenous NO derived from eNOS activation through phosphory-
residues on EGF-R that could be different than the residues charac- lation at Ser1177 promotes p21Ras S-nitrosylation, and trigger
terized by Murillo-Carretero et al. [24]. Another explanation could ERK1/2 MAP kinases activation through phosphorylation on our

Fig. 10. A schematic overview of the BK initiated, NO-epidermal growth factor receptor-mediated signaling pathway in angiogenesis. (1) BK stimulates the B2 receptor which
activates eNOS and generates NO. NO nitrosylates the EGF-R, the PTP SHP-1, and p21Ras. (2) S-nitrosylation, tyrosine phosphorylation and ERK1/2 dependent threonine
phosphorylation of the EGF-R. S-nitrosylation of p21Ras and activation of the ERK1/2 MAP kinases. A cross-talk between the two signaling pathways results in sustained
signaling that leads to endothelial cell proliferation and induction of the expression levels of VEGF. (3) Signals derived from BK/B2 receptor/NO/S-nitrosylation, EGF/EGF-R,
VEGF/VEGF-R result in induction of in vitro angiogenesis.
26 M.S. Moraes et al. / Archives of Biochemistry and Biophysics 558 (2014) 14–27

BK stimulated endothelial cells model. Although, pre-incubation of phosphorylation. Levels of EGF-R tyrosine phosphorylation can be
cells with L-NAME has no direct effect on ERK1/2 MAP kinases maintained through S-nitrosylation of SHP-1(a PTP-associated
phosphorylation levels, the inhibition of p21Ras activity and its EGF-R) inhibiting its phosphatase activity. Moreover, BK-induced
S-nitrosylation lead to the inhibition of ERK1/2 MAP kinases phos- p21Ras activation by S-nitrosylation results in stimulation of
phorylation that is usually triggered by BK cell treatment. downstream ERK1/2 MAP kinases signaling. Activated ERK1/2
BK treatment induces EGF-R Pro-directed Thr phosphorylation MAP kinases inhibit EGF-R internalization due phosphorylation of
and tyrosine phosphorylation. Pro-directed Thr phosphorylation specific Thr residues at the cytoplasmic receptor domain. Thus,
was inhibited by pretreatment with PD98059 (a MEK inhibitor), inhibition of EGF-R internalization results in constitutive long-
suggesting dependence on ERK1/2 mediated signaling pathways. term signaling of the receptor [49]. NO-derived from endothelial
Growth hormone up-regulates EGF-R signaling and increase recep- cell stimulation by BK, promotes direct and indirect posttransla-
tor half life at the cell surface in an ERK1/2 pathway-dependent tional modifications on EGF-R. Such modifications prime EGF-R
manner [47]. In this work, endothelial cell stimulation with BK pro- sustained signaling. We have demonstrated that EGF-R down-
moted EGF-R phosphorylation and/or activation. Moreover, BK stream signaling events, B2 receptor, p21Ras, and endogenous
treatment counteracts EGF agonist induced EGF-R desensitization NO production can modulate VEGF mRNA expression levels in
and internalization and consequent degradation. We have shown endothelial cells stimulated with BK.
that these effects were inhibited by PD98059, suggesting the par- Finally, BK-stimulated angiogenesis results from a combination
ticipation ERK1/2 MAP kinases signaling pathway during sustained of signals transduced by the BK/B2 receptor, NO/EGF-R/SHP-1,
EGF-R signaling that is a hallmark of BK-stimulated endothelial p21Ras/ERK1/2 MAP kinases signaling axis and by VEGF. This
cells. mechanism is described in Fig. 10.
It is well known that endothelial cell proliferation is associated
with neo-vascular expansion or angiogenesis [48]. In this work we
observed that, proliferation was stimulated by BK treatment in Acknowledgments
HUVEC, RAEC, and RAEC expressing wild type EGF-R; conversely
endothelial cell lines expressing EGF-R–kinase-dead BK treatment This work was supported by grants from FAPESP (Fundação de
can no longer stimulate proliferation. These findings suggest a Amparo a Pesquisa do Estado de São Paulo/Brazil, Proc. 2007/
major role for the EGF-R signaling pathway in BK induced endothe- 59617-6, 2010/19013-7, 2012/10470-1) and from Conselho Nac-
lial cell proliferation. ional de Desenvolvimento Científico e Tecnológico/Brazil (CNPq)
In our model, endothelial cells expressing a fully functional to H.P.M. M.S.M. received a doctorate fellowship from FAPESP.
EGF-R displayed a sustained activation of ERK1/2 MAP kinases
upon BK treatment. Sustained ERK1/2 MAP kinases activation
Appendix A. Supplementary data
and signaling is crucial for proper endothelial cell proliferation.
Indeed, sustained ERK1/2 MAP kinases activation and signaling
Supplementary data associated with this article can be found, in
was previously associated with cell cycle progression and prolifer-
the online version, at http://dx.doi.org/10.1016/j.abb.2014.06.011.
ation; when RAEC cells were exposed to low concentrations of
SNAP (a nitrosothiol) [20,21]. In this same context, experiments
performed here in the presence of BIBX suggested that pharmaco- References
logical inhibition of EGFR-tyrosine kinases in HUVECs expressing
WT-EGF-R or stimulation of HUVECs over-expressing a kinase- [1] M.A. Lemmon, J. Schlessinger, Cell 141 (2010) 1117–1134.
[2] G. Pines, W.J. Kostler, J. Yarden, FEBS Lett. 584 (2010) 2699–2706.
dead form of EGF-R, was enough to down-regulate BK-induced [3] R.N. Jorissen, F. Walker, N. Pouliot, T.P. Garrett, C.W. Ward, A.W. Burgess, Exp.
capillary-like pro-angiogenic structures formation on MatrigelÒ. Cell Res. 284 (2003) 31–53.
This negative modulation was also observed when BK-stimulated [4] H. Keilhack, T. Tenev, E. Nyakatura, J. Godovac-Zimmerman, L. Nielsen, K.
Seedorf, F.D. Bohmer, J. Biol. Chem. 273 (1998) 24839–24846.
HUVEC cells were pre-incubated with the NOS inhibitor and with [5] Y. Huang, S.O. Kim, J. Jiang, S.J. Frank, J. Biol. Chem. 278 (2003) 18902–18913.
MEK inhibitor. [6] M.J. Marinissen, J.S. Gutkind, Trends Pharmacol. Sci. 22 (2001) 368–376.
VEGF expression plays a major role in controlling tissue vascu- [7] J. Folkman, Nat. Med. 1 (1995) 27–31.
[8] P. Carmeliet, Nat. Med. 9 (2003) 653–660.
lature and angiogenesis under physiologic conditions [8]. Secretion
[9] I. Zachary, G. Gliki, Cardiovasc. Res. 49 (2001) 568–581.
of VEGF by tumor cells stimulates VEGF receptors in endothelial [10] J. Tabernero, Mol. Cancer Res. 5 (2007) 203–220.
cells of pre-existing blood vessels in the stroma promoting new [11] D. Thuringer, L. Maulon, C. Frelin, J. Biol. Chem. 277 (2002) 2028–2032.
blood vessel formation [10]. Here, we show that VEGF expression [12] F. Khur, J. Lowry, Y. Zhang, V. Brovkovych, R.A. Skidgel, Neuropeptides 44
(2010) 145–154.
can be down-regulated by many reasons, such as, blockade of [13] K. Ishihara, M. Kamata, I. Hayashi, S. Yamashima, M. Majima, Int.
EGF-R activity, B2 receptor inhibition, inhibition of p21Ras activity, Immunopharmacol. 2 (2002) 499–509.
and inhibition of endogenous NO production in HUVECs that was [14] M. Ziche, L. Morbidelli, E. Masini, S. Amerini, H.J. Granger, C.A. Maggi, P.
Geppetti, F. Ledda, J. Clin. Invest. 94 (1994) 2036–2044.
stimulated with BK. Moreover, the VEGF inhibitor SU5416 also [15] R.G. Knowles, S. Moncada, Biochem. J. 298 (1994) 249–258.
blocked BK-induced formation of capillary-like pro-angiogenic [16] W.C. Sessa, J. Thromb. Haemost. 7 (Suppl. 1) (2009) 35–37.
structures. Taken together our data suggested that in addition to [17] F. Murad, J. Clin. Invest. 78 (1986) 1–5.
[18] D.T. Hess, A. Matsumoto, S.-O. Kim, H.E. Marshall, J.S. Stamler, Nat. Rev. Mol.
signaling events triggered by BK/B2 receptor, NO/EGF-R/SHP-1, Cell Biol. 6 (2005) 150–166.
p21Ras/ERK1/2 MAP kinases, VEGF-mediated signaling contributes [19] T.M.S. Peranovich, A.M. da Silva, D.M. Fries, A. Stern, H.P. Monteiro, Biochem J.
to the formation of capillary-like pro-angiogenic structures. 305 (1995) 613–619.
[20] C.J. Rocha Oliveira, F. Schindler, A.M. Ventura, M.S. Moraes, R.J. Arai, V. Debbas,
A. Stern, H.P. Monteiro, Free Radic. Biol. Med. 35 (2003) 381–396.
[21] C.J. Rocha Oliveira, M.F. Curcio, M.S. Moraes, M. Tsujita, L.R. Travassos, A. Stern,
Conclusion H.P. Monteiro, Nitric Oxide 18 (2008) 241–255.
[22] M.F. Curcio, W.L. Batista, E. Linares, F.D. Nascimento, M.S. Moraes, R.E. Borges,
J. Sap, A. Stern, H.P. Monteiro, Antioxid. Redox Signal. 13 (2010) 109–125.
Endothelial cells when stimulated with BK up-regulate endoge- [23] W.L. Batista, F.T. Ogata, M.F. Curcio, R.B. Miguel, R.J. Arai, A.L. Matsuo, M.S.
nous NO production by eNOS promoting cell proliferation and Moraes, A. Stern, H.P. Monteiro, Antioxid. Redox Signal. 18 (2013) 221–238.
angiogenesis. NO, EGF-R, and ERK1/2 MAP kinases are involved in [24] M. Murillo-Carretero, A. Torroglosa, C. Castro, A. Villalobo, C. Estrada, Free
Radic. Biol. Med. 46 (2009) 471–479.
this process. NO activates EGF-R by a mechanism that includes [25] O. Kashles, Y. Yarden, R. Fischer, A. Ullrich, J. Schlessinger, Mol. Cell. Biol. 11
S-nitrosylation of multiple targets and EGF-R tyrosine (1991) 1459–1463.
 M.S. Moraes et al. / Archives of Biochemistry and Biophysics 558 (2014) 14–27 27

[26] V. Hampl, C.L. Walters, S.L. Archer, in: M.S. Feelisch, J.S. Stamler (Eds.), [36] D.M. Barrett, S.M. Black, H. Todor, R.K. Schmidt-Ullrich, K.S. Dawson, R.B.
Methods in Nitric Oxide Research, John Wiley & Sons Ltd., Chichester, West Mikkelsen, J. Biol. Chem. 280 (2005) 14453–14461.
Sussex, England, 1996, pp. 309–318. [37] C. Liebmann, Biol. Chem. 382 (2001) 49–55.
[27] M.T. Forrester, M.W. Foster, M. Benhar, J.S. Stamler, Free Radic. Biol. Med. 46 [38] X. Li, Y. Huang, J. Jing, S.J. Frank, Cell. Signal. 20 (2008) 2145–2155.
(2009) 119–126. [39] R.B. Pearson, B.E. Kemp, Methods Enzymol. 200 (1991) 62–81.
[28] T. Paschoalin, A.K. Carmona, E.G. Rodrigues, V. Oliveira, H.P. Monteiro, M.A. [40] P. Morrison, K. Takishima, M.R. Rosner, J. Biol. Chem. 268 (1993) 15536–15548.
Juliano, L. Juliano, L.R. Travassos, Mol. Cancer 6 (2007) 44. [41] M. Toulany, U. Kasten-Pisula, I. Brammer, S. Wang, J. Chen, K. Dittmann, M.
[29] J.T. Chi, H.Y. Chang, G. Haraldsen, F. Jahmsen, O.G. Troyanskaya, D.S. Chang, Z. Baumann, E. Dikomey, H.P. Rodemann, Clin. Cancer Res. 12 (2006) 4119–4126.
Wang, F.R. Stanley, M. van-de-Rijn, D. Botstein, P.O. Brown, Proc. Natl. Acad. [42] A. Hirata, S. Ogawa, T. Kometani, T. Kuwano, S. Naito, M. Kuwano, M. Ono,
Sci. U.S.A. 100 (2003) 10623–10628. Cancer Res. 62 (2002) 2554–2560.
[30] J.K. Liao, C.J. Homcy, J. Clin. Invest. 92 (1993) 2168–2172. [43] J. Folkman, M. Klagsbrun, Science 235 (1987) 442–447.
[31] S.W. Bae, H.S. Kim, Y.N. Cha, Y.S. Park, S.A. Jo, I. Jo, Biochem. Biophys. Res. [44] D.A. Linseman, C.W. Benjamin, D.A. Jones, J. Biol. Chem. 270 (1995) 12563–
Commun. 306 (2003) 981–987. 12568.
[32] H.P. Monteiro, R.J. Arai, L.R. Travassos, Antioxid. Redox Signal. 10 (2008) 843– [45] H. Daub, F.U. Weiss, C. Wallasch, A. Ullrich, Nature 379 (1996) 557–560.
889. [46] Y.-Y. Chen, H.-F. Huang, K.-H. Khoo, T.-C. Meng, Methods 42 (2007) 243–249.
[33] C.H. Switzer, S.A. Glynn, R.Y.-S. Cheng, L.A. Ridnour, J.E. Green, S. Ambs, D.A. [47] Y. Huang, Y. Chang, X. Wang, J. Jiang, S.J. Frank, Endocrinology 145 (2004)
Wink, Mol. Cancer Res. 10 (2012) 1203–1215. 3297–3306.
[34] S. Tomic, U. Greiser, R. Lammers, A. Kharitonenkov, E. Imyanitov, A. Ullrich, [48] D. Hanahan, J. Folkman, Cell 86 (1996) 353–364.
F.D. Bohmer, J. Biol. Chem. 270 (1995) 21277–21284. [49] H.S. Wiley, Exp. Cell Res. 284 (2003) 78–88.
[35] A. Caselli, G. Camici, G. Manao, G. Moneti, L. Pazzagli, G. Cappugi, G. Ramponi, J.
Biol. Chem. 269 (1994) 24878–24882.