Original article

Bone morphogenic protein-4 induces endothelial cell apoptosis through oxidative

stress-dependent p38MAPK and JNK pathway
Xiao Yu Tian
a, 1
, Lai Hang Yung
a, 1
, Wing Tak Wong
a
, Jian Liu
a
, Fung Ping Leung
a
, Limei Liu
a
,
Yangchao Chen
a
, Siu Kai Kong
b
, Kin Ming Kwan
b
, Siu Man Ng
c
, Paul B.S. Lai
c
, Lai Ming Yung
a
,
Xiaoqiang Yao
a
, Yu Huang
a,

a
Institute of Vascular Medicine, Li Ka Shing Institute of Health Sciences, School of Biomedical Sciences, Chinese University of Hong Kong; Hong Kong SAR, China
b
School of Life Sciences, Chinese University of Hong Kong; Hong Kong SAR, China
c
Department of Surgery, Chinese University of Hong Kong; Hong Kong SAR, China
a b s t r a c t a r t i c l e i n f o
Article history:
Received 5 August 2011
Received in revised form 16 September 2011
Accepted 17 October 2011
Available online 25 October 2011
Keywords:
Bone morphogenic protein-4
Reactive oxygen species
Mitogen-activated protein kinase
Apoptosis
Endothelial cells
The expression of bone morphogenic protein 4 (BMP4), a new pro-inammatory marker, is increased by dis-
turbed ow in endothelial cells (ECs). BMP4 stimulates production of reactive oxygen species (ROS) and
causes endothelial cell dysfunction. The present study examined BMP4-induced apoptosis in ECs and isolated
arteries from rat, mouse, and human, and the signaling pathways mediating BMP4-induced apoptosis. Apo-
ptosis was assessed by ow cytometry to detect Annexin-V positive cells, and terminal deoxynucleotidyl
transferase dUTP nick end (TUNEL) labeling. The superoxide production was measured by dihydroethidium
uorescence. BMP4 induced EC apoptosis in human mesenteric arteries, mouse aortic endothelium, rat pri-
mary ECs, and human ECs. BMP4-induced EC apoptosis was mediated through ROS production by activation
of NADPH oxidase, which led to cleaved caspase-3 expression. BMP4 also induced sequential activation of
p38 MAPK and JNK which was upstream of caspase 3 activation. Knockdown of BMP receptor 1A by lentiviral
shRNA or NOX4 siRNA transfection inhibited BMP4-induced ROS production, p38 and JNK phosphorylation,
and caspase-3 activation in ECs. JNK siRNA inhibited BMP4-induced JNK phosphorylation and caspase-3 acti-
vation. The present study delineates that BMP4 causes EC apoptosis through activation of caspase-3 in a ROS/
p38MAPK/JNK-dependent signaling cascade.
2011 Elsevier Ltd. All rights reserved.
1. Introduction
Bone morphogenic protein-4 (BMP4), originally discovered as a
bone growth factor [1] exerts an inammatory effect in blood vessels.
BMP4 and other members of the BMP family of ligands are detectable
in atherosclerotic plaques [24]. BMP4 induces monocyte adhesion in
response to oscillatory shear stress through NAD(P)H oxidase-
dependent ROS [5] and impairs endothelial function through activa-
tion of NAD(P)H oxidase [6,7] although the underlying mechanisms
of BMP4-induced vascular inammation remain incompletely known.
Apoptosis is essential for maintaining normal development of
multi-cellular organisms through elimination of unwanted cells [8].
In the cardiovascular system, however, elevated vascular cell apopto-
sis is associated with increased incidence of failing heart [9], ad-
vanced and unstable atherosclerotic plaques [10], hypertension [11],
and diabetes [12]. Endothelial cell (EC) apoptosis contributes to the
development of atherosclerosis through an increased permeability
of endothelial monolayer and subsequent uptake of lipids in the vas-
cular wall [13]. Clinical studies suggest a contributory role of EC apo-
ptosis in plaque destabilization and thrombosis [14].
In atherosclerotic plaques, ECs undergoing apoptosis produce high
levels of apoptotic blebs containing active oxidized phospholipid that
stimulate adhesion of monocytes to ECs [1517]. Enhanced produc-
tion of NAD(P)H oxidases-derived ROS, a hallmark in hypertension
and atherosclerosis, is found in ECs as a result of disturbed ow [18]
and ROS released by pro-inammatory cytokines and oxidized lipo-
proteins promote EC apoptosis [19]. However, it is unclear whether
ROS production induced by BMP4 can initiate a cascade of cellular
events leading to EC apoptosis.
Apart from the classical intracellular pathway initiated by BMPs
which involves receptor-mediated activation of Smad, existing evi-
dence shows that BMPs activate MAPKs in various cell types [20]
such as lung broblasts [20] and pulmonary arterial myocytes [21].
MAPKs serve as the downstream targets in H
2
O
2
-induced apoptosis
of rat VSMCs [22]. However, it remains unclear whether MAPKs are
involved in the signaling cascade triggered by BMP4 in ECs.
BMP4 can induce apoptosis in both vascular smooth muscle cells
[23] and ECs [24,25]. However, BMP4 also possesses anti-apoptotic ef-
fects [2628]. In the present study, we investigated the pro-apoptotic
Journal of Molecular and Cellular Cardiology 52 (2012) 237244
Corresponding author at: School of Biomedical Sciences, Chinese University of
Hong Kong, Shatin, NT, Hong Kong, China. Tel.: +852 26096787; fax: +852 26035022.
E-mail address: yu-huang@cuhk.edu.hk (Y. Huang).
1
Contributed equally to this work.
0022-2828/$ see front matter 2011 Elsevier Ltd. All rights reserved.
doi:10.1016/j.yjmcc.2011.10.013
Contents lists available at SciVerse ScienceDirect
Journal of Molecular and Cellular Cardiology
j our nal homepage: www. el sevi er . com/ l ocat e/ yj mcc
effect of BMP4 in ECs and arteries from different species and demon-
strated that BMP4-induced caspase-3 activation is mediated through
the sequential activation of BMPR1A, NADPHoxidase, and downstream
p38 MAPK and JNK.
2. Materials and methods
2.1. Isolation and primary culture of rat aortic endothelial cells (RAECs)
The experimental protocols were approved by the institutional
animal care and use committee and were consistent with the Guide
for the Care and Use of Laboratory Animals published by the National
Institutes of Health. RAECs were isolated from the thoracic aorta of
male SpragueDawley rats (260280 g) using an enzymatic digestion
method [29]. The aorta was incubated in phosphate-buffered-saline
(PBS) containing 0.2% collagenase with shaking for 15 min at 37 C,
then centrifuged for 5 min at 800 g. The cells were suspended in
RPMI medium 1640 (Gibco, Grand Island, NY, USA) containing 10%
fetal bovine serum (Gibco) and 1% penicillin/streptomycin and set-
tled for 1 h. Culture medium was changed afterwards. The identity
of the RAECs was conrmed by a positive staining of PECAM-1
(Santa Cruz, CA, USA), and used within the rst two passages. For
Fig. 1. BMP4 induced EC apoptosis in human arteries. (A) BMP4 caused DNA fragmenta-
tion as assessed by TUNEL staining. (B) BMP4 induced cleaved caspase-3 expression as
assessed by immunouorescence imaging in ECs of human mesenteric arteries. Represen-
tative images shown are from 4 independent experiments from different subjects. Bars
represent 100 m.
Fig. 2. BMP4 induced EC apoptosis in mouse and rat. BMP4 (10 ng/mL, 24 h)-induced cleaved caspase-3 expression as assayed by (A) immunouorescence of en face endothelium of
mouse aortae and (B) Western blotting of aortic tissue. Native endothelial cells were labeled with PECAM-1 (red) and cleaved caspase-3 (green), which was localized in the nucleus.
Data are meanSEM (n=6).

Pb0.05 compared with control;
#
Pb0.05 compared with BMP4. Photos are representative for samples from 4 different mice. BMP4 induces RAEC
apoptosis through ROS-dependent caspase-3 activation. (C) BMP4 treatment for 24 h caused a concentration-dependent apoptotic changes in RAECs as assessed by TUNEL staining.
(D) BMP4 induced cleaved caspase-3 expression in RAECs as compared with H
2
O
2
.

Pb0.05 compared with control from different rats.
238 X.Y. Tian et al. / Journal of Molecular and Cellular Cardiology 52 (2012) 237244
transfection experiment, RAECs were transfected with NOX4 siRNA
pool (SMARTpools, Thermo Scientic, Lafayette, CO, USA) or nontar-
geting siRNA as control by electroporation using Nucleofector II ma-
chine (Amaxa/Lonza, Walkersville, MD, USA) according to the
manufacturer's instruction.
2.2. Human endothelial cell culture and transfection
Human umbilical vein endothelial cells (HUVECs, Lonza, Basel,
Switzerland, No.CC-2517) were grown in EGM (Clonetics, San Diego,
USA) supplemented with bovine brain extract (BBE, Clonetics), penicil-
lin (100 g/mL) and streptomycin (100 g/mL) in gelatin-coated asks
and maintained at 37 C in a 95% O
2
plus 5% CO
2
condition. The cells
at passage 46 were used when at ~8090% conuency. For the knock-
down experiment, 80% conuent cells were transfected with prede-
signed JNK siRNA (Invitrogen, Carlsbad, CA, USA), with Lipofectamine
RNAiMAX (Invitrogen) in Opti-MEM for 24 h. After transfection, Opti-
MEM medium was replaced with EBM without phenol red for further
treatment.
2.3. Human mesenteric artery specimens
The present study was approved by the Joint Chinese University of
Hong KongNew Territories East Cluster Clinical Research Ethics Com-
mittee. Human small mesenteric arteries were harvestedduring surger-
y from 4 colon cancer patients, after obtaining their informed consent.
Arteries were dissected in PBS, and incubated with BMP4 (30 ng/mL,
24 h) in DMEM.
2.4. Chemical treatment
The ECs were treated with recombinant BMP4 (10, 30 and 100 ng/
mL, dissolved in 4 mM HCl with 0.1% BSA, R&D Systems, Minneapolis,
MN, USA) for 24 h and collected for apoptosis and protein assay. For
other drug treatments, ECs were treated with one of the following: nog-
gin(BMP4 antagonist, 100 ng/mL, R&DSystems), apocynin (NADPHox-
idase inhibitor, 100 mol/L, Sigma, St. Louis, MO, USA), tiron (1 mmol/L,
Sigma) plus DETCA (100 mol/L, Sigma) (T+D, ROS scavengers),
SP600125 (JNK inhibitor, 10 mol/L, Tocris, Bristol, UK), PD98059
(ERK inhibitor, 20 mol/L, Tocris), and SB202190 (p38 MAPK inhibitor,
10 mol/L, Tocris) for 30 min prior to BMP4 exposure.
2.5. Western blotting
After treatment, cells were trypsinized and homogenized in lysis
buffer containing complete protease and phosphatase inhibitor cocktail
(Roche Diagnostics, Indianapolis, IN, USA), followedby centrifugation at
20,000 g for 20 minto collect supernatants. The protein samples (20 g)
were separated with 12.5% SDS-polyacrylamide gel and transferred to
an Immobilon-P polyvinylidene diuoride membrane. Membranes
were blocked with 1% bovine serum albumin. Primary antibodies
against phospho-p38 MAPK, p38 MAPK, phospho-SAPK/JNK, SAPK/
Fig. 3. BMP4 induces RAEC apoptosis through ROS-dependent caspase-3 activation. (A) Noggin (100 ng/mL), apocynin (100 mol/L) or tiron (1 mmol/L) plus DETCA (T+D)
(100 mol/L) suppressed the increased apoptotic rate in RAECs in response to BMP4 (10 ng/mL) as assayed by TUNEL staining. (B) BMP4 increased intra-cellular generation of
O
2

in RAECs after 30-min treatment, which was blocked by noggin, apocynin, or T+D. (C) Western blot analysis showed that BMP4-induced increase of cleaved caspase-3 in
RAECs was also inhibited by noggin, apocynin, or T+D pre-treatment. Data are meanSEM (n=5).
#
Pb0.05 in compared with control;

Pb0.05 compared with BMP4.
239 X.Y. Tian et al. / Journal of Molecular and Cellular Cardiology 52 (2012) 237244
JNK (1:500, Cell signaling Technology, Beverly, MA, USA), cleaved
caspase-3 (1:2000 for Western blotting; 1:200 for immunouores-
cence, Cell signaling Technology), caspase-3 (1:2000, Calbiochem, San
Diego, CA, USA), NOX4 (Abcam, Cambridge, UK), BMPR1A (Santa Cruz,
CA, USA), housekeeping GAPDH (1:50000, Ambion, Austin, TX, USA),
and -actin (1:4000, Sigma, MO, USA) were used.
2.6. Terminal deoxynucleotidyl transferase-mediated dUTP nick end-
labeling (TUNEL) assay
The TUNEL assay was used to detect DNA fragmentation in situ
using the ApopTag apoptosis detection kit (Chemicon, Temecula, CA,
USA). Briey, cells or tissue sections were xed with 4% paraformal-
dehyde, washed and stained according to the manufacturer's instruc-
tions. The samples were then counterstained with hematoxylin. Photos
were taken under a Leica DMRBE microscope.
2.7. Flow cytometry
For analyses of cell apoptosis, the apoptotic rate was measured by PI
(propidium iodide) and annexin-V staining with ow cytometry
according to manufacturer's instructions (BD Biosciences, CA, USA).
Briey, after treatments, cells were trypsinized and collected by centri-
fugation. The cells were washed and re-suspended in binding buffer at
110
6
cells/mL. 110
5
cells were stained with PI and Annexin-V FITC
for 15 min before analysis by FACSort ow cytometer (BD Biosciences).
The instrument was set to collect 110
5
cells and the prole was ana-
lyzed using CellQuest software.
2.8. Dihydroethidium (DHE) uorescence imaging
After treatment of the RAECs with 10 ng/mL BMP4 and inhibitors,
cells were rinsed with normal physiological saline solution (NPSS), and
incubated with DHE (5 mol/L) at 37 C in dark for 15 min, then washed
twice with NPSS. Fluorescence was observed by confocal microscope
(515-nm excitation; 585-nm long pass lter; Olympus FV1000, Tokyo,
Japan). DHE uorescence intensity was analyzed by Fluoview (version
1.5). For each section, a square region with an area of 80 m80 m
was selected for analysis. The summarized data was expressed as com-
pared with control to indicate the fold change in uorescence intensity
among different treatments.
2.9. Immunouorescence staining
Expression of cleaved caspase-3 in human mesenteric artery was
detected. The human artery samples were then xed with 4% parafor-
maldehyde for overnight, cut into two segments, and kept in frozen
and parafn blocks. Both frozen and parafn sections were prepared
for immunostaining. Parafn sections were heated in citrate buffer for
antigen retrieval. Frozen sections were dried and rinsed with 0.01% Tri-
ton in PBS once for 30 s. Samples were blocked by 5% normal donkey
serum for 30 min. Then, sections were incubated with cleaved
caspase-3 primary antibodies at 4 C overnight. Afterwards, slides
were washed with PBS and incubated with the secondary antibodies
(AlexaFluor546, Molecular Probes, Eugene, OR. USA) for 1 h at room
temperature. After washing twice with PBS, uorescence was observed
by uorescence microscope.
2.10. Immunouorescence staining of enface endotheliumfrommouse aorta
Aortic rings (2 mm length segment from C57BL/6 J mouse aorta)
were treated with BMP4 (10 ng/mL) and noggin (100 ng/mL) in DMEM
for 24 h, and xedwith4%paraformaldehyde, thenwashed withPBS, fol-
lowed by the same procedure as frozen section. After being incubated
with secondary antibody (AlexaFluor546, AlexaFluor488), aortic rings
were cut open and the endothelium side was placed upside down on
the coverslip, with another coverslip placed on the top for mounting,
then observed under confocal microscope (Olympus FV1000, Tokyo,
Japan).
2.11. Constructs, lentivirus production and transduction
Two shRNAs (short hairpin RNAs) targeting mouse Bmpr1a were
designed: shRNA1: 5- GCT GTT AAA TTC AAC AGT GAC ACA AAT G -3;
shRNA2: 5- TCT CTC TAT GAC TTC CTG AAA TGT GCC -3, and one
scramble shRNA as a control [30]. DNA fragments containing shRNAs
sequence were synthesized and cloned into lentiviral RNAi (RNA inter-
ference) vector pLUNIG after annealing as described [31]. The VSV-
G-pseudotyped lentiviruses were produced by cotransfecting 293T
cells with the transfer vector and three packaging vectors: pMDLg/
pRRE, pRSV-REV, and pCMV-VSVG. Subsequent purication was
Fig. 4. ROS mediates p38 MAPK and JNK activation in BMP4-induced apoptosis in
RAECs. (A) The up-regulation of cleaved caspase-3 in response to BMP4 was reversed
by SB202190 or SP600125. (B) Increased level of phospho-p38 MAPK in response to
BMP4 (10 ng/mL) for 2 h was prevented by treatment with noggin (100 ng/mL), apoc-
ynin (100 mol/L), or tiron (1 mmol/L) plus DETCA (100 mol/L) (T+D) without af-
fecting total p38 MAPK. (C) The elevated phospho-JNK level stimulated by 4 h
exposure to BMP4 was abolished by noggin, apocynin and T+D without affecting
total JNK. (D) Western blot analysis showed that activation of JNK following 4-hour
treatment with BMP4 in RAECs was reduced by MAPK inhibitor SB202190 (10 mol/
L) and JNK inhibitor SP600125 (10 mol/L). (E and F) BMP4 (10 ng/mL, 24 h) induced
phosphorylations of p38 and JNK in mouse aortae, which were inhibited by noggin
(100 ng/mL, 24 h).Data are expressed as meanSEM (n=4) from different rats or
mice.
#
Pb0.05 compared with control;

Pb0.05 compared with BMP4.
240 X.Y. Tian et al. / Journal of Molecular and Cellular Cardiology 52 (2012) 237244
performed using ultracentrifugation. RAECs were cultured in 12-well
plates and transfected with lentivirus and 8 g/mL polybrene (Sigma).
2.12. Statistical analysis
Results represent meansSEM of n experiments and data ana-
lyzed by Student t test. Statistical signicance was determined by
two-tailed Student's t-test or one-way ANOVA followed by Bonferroni
post-hoc tests when more than two treatments were compared.
Pb0.05 was considered statistically signicant.
3. Results
3.1. BMP4 induced apoptosis in the endothelial cells from human, rat,
and mouse
First, we assessed the pro-apoptotic effect of BMP4 in isolated ar-
teries from human and mouse. BMP4 (30 ng/mL) induced DNA frag-
mentation as assessed by TUNEL staining (Fig. 1A) and increased the
level of cleaved caspase-3 as assessed by immunouorescence imag-
ing (Fig. 1B) in endothelium of human mesenteric arteries after 24 h
treatment with BMP4 (30 ng/mL). Expression of cleaved caspase-3
was also observed in isolated mouse aortae after BMP4 (10 ng/mL,
24 h) treatment, which was inhibited by co-incubation with noggin
(100 ng/mL), as assessed by immunouorescence staining of en face
endothelium from mouse aorta (Fig. 2A) and Western blot from aor-
tic tissue (Fig. 2B).
To further investigate the underlying mechanisms of BMP4-
induced EC apoptosis, we veried the pro-apoptotic effects of BMP4
in primary ECs from rat aorta (RAECs) and HUVECs. BMP4 caused ap-
optosis as measured by TUNEL staining in RAECs (Fig. 2C), and by
ow cytometry for detection of annexin-V positive cells in HUVECs
(Supplemental Fig. 1A and B) in a concentration-dependent manner.
BMP4 (24 h) also induced cleaved caspase-3 in RAECs as compared
to H
2
O
2
in RAECs (BMP4 10 ng/mL, Fig. 2D) and in HUVECs (BMP4
30 ng/mL, Supplemental Fig. 1B and E).
3.2. BMP4-induced EC apoptosis through ROS-dependent caspase-3
activation
After showing the pro-apoptotic effects of BMP4 in ECs from
different species, we studied whether such effects were mediated
by oxidative stress. Noggin (100 ng/mL), apocynin (100 mol/L), or
tiron (1 mmol/L) plus DETCA (100 mol/L) suppressed the increased
apoptotic rate in RAECs in response to BMP4 (10 ng/mL) (Fig. 3A).
BMP4 signicantly increased intra-cellular generation of O
2

in RAECs
after 30 min treatment, which was blocked by noggin, apocynin, or
tiron plus DETCA(Fig. 3B). Western blot analysis revealed that BMP4 in-
duced cleaved caspase-3 in RAECs was also inhibited by treatment with
noggin, apocynin, or tiron plus DETCA treatment (Fig. 3C). Similar pro-
Fig. 5. BMPR1A knockdown prevents BMP4-induced caspase-3 activation. (A) BMPR1A knockdown by two shRNA inhibited ROS production induced by BMP4 as measured by DHE
uorescence. Both shRNAs targeting BMPR1A inhibited p38 phosphorylation (B) and caspase-3 activation (C). Data are meanSEM (n=4).
#
Pb0.05 compared with control;

Pb0.05 compared with BMP4.

241 X.Y. Tian et al. / Journal of Molecular and Cellular Cardiology 52 (2012) 237244
apoptotic effect of BMP4 was also observed in HUVECs (Supplemental
Fig. 1C and E).
3.3. ROS mediated p38 MAPK and JNK activation in BMP4-induced EC
apoptosis
To further elucidate the signaling cascade leading to EC apoptosis
activated by BMP4-induced ROS, we studied whether p38 MAPK
and/or JNK were involved. Caspase-3 activation induced by BMP4
was inhibited by SB202190 (10 mol/L, p38 MAPK inhibitor) or
SP600125 (10 mol/L, JNK inhibitor) in RAECs (Fig. 4A). Similar results
were also found by TUNEL staining in RAECs (Supplemental Fig. 2A)
and also in HUVECs (Supplemental Fig. 1D and F). Phosphorylation
of p38 MAPK in response to BMP4 for 2 h was prevented by treatment
with noggin, apocynin, or tiron plus DETCA without affecting total p38
(Fig. 4B). Likewise, phosphorylation of JNK stimulated by 4-h expo-
sure to BMP4 was also abolished by noggin, apocynin, and tiron plus
DETCA without affecting total JNK (Fig. 4C).
Activation of p38 and JNK by BMP4 (10 ng/mL, 24 h) was also ob-
served in isolated mouse aortae, which were inhibited by co-treatment
with noggin (100 ng/mL) (Figs. 4E and F). To further elucidate the cross-
talk between p38 MAPK and JNK in BMP4-induced EC apoptosis,
we found that p38 inhibitor SB202190 (10 mol/L) inhibited JNK
phosphorylation induced by BMP4 (Fig. 4D), while JNK inhibitor
SP600125 (10 mol/L) did not inhibit p38 phosphorylation induced by
BMP4 (Supplemental Fig. 2B).
3.4. BMP4-induced EC apoptosis is mediated through BMPR1A
Since BMP4 have several BMP receptors including BMPR1A, BMPR1B,
and BMPR2, we studied whether BMPR1A is involved in BMP4-induced
EC apoptosis. BMPR1A knockdown by lentiviral shRNA transfection
inhibited BMP4-induced ROS production as measured by DHE uores-
cence (Fig. 5A). Moreover, BMPR1A knockdown reduced p38 phosphor-
ylation induced by BMP4 (Fig. 5B). In addition, BMP4-induced caspase-3
activation was also reduced (Fig. 5C). Efciency of transfection was ver-
ied by the reduced expression of BMPR1Aby both shRNAs compared to
scramble control (Supplemental Fig. 3A).
3.5. NADPH oxidase-derived ROS is required for BMP4-induced apoptosis
Previous studies indicated that NADPH oxidase is the major source
of BMP4-induced ROS [6,32]. We studied whether NOX4 is involved in
BMP4-induced ROS production and apoptosis. NOX4 siRNA reduced
the NOX4 expression in RAECs (Supplemental Fig. 3B). ROS production
induced by BMP4 also reduced after NOX4 siRNA transfection (Figs. 6A
and B). In addition, NOX4 inhibition reduced phosphorylations of p38
(Fig. 6C) and JNK (Fig. 6D) in response to BMP4. BMP4-induced cleaved
caspase-3 expression was also inhibited (Figs. 6E and F).
3.6. Knockdown of JNK inhibits BMP4-induced caspase-3 activation
Based on the results that BMP4 induced JNK phosphorylation and
JNK inhibitor inhibited BMP4-induced EC apoptosis, we studied wheth-
er knockdown of JNK by siRNA could inhibit BMP4-induced caspase-3
Fig. 7. Knockdown of JNK prevents caspase-3 activation. (A) JNK siRNA suppressed both
the BMP4- and ROS-induced EC apoptosis by ow cytometry. (B) Treatment with JNK
siRNA reduced the level of cleaved caspase-3 in response to BMP4 or ROS in HUVECs.
Data are expressed as meanSEM(n=4).
#
Pb0.05 compared with control (with lipofec-
tamine RNAiMAX only). N.S. refers to no statistical difference compared with the control
group. ROS was generated using HXXO (100 mol/L hypoxanthine+0.01 u/mL xanthine
oxidase or H
2
O
2
(50 mol/L)).
Fig. 6. NOX4 mediates BMP4-induced ROS and caspase-3 activation. (A and B) NOX4
inhibition also reduced BMP4-induced ROS production as measured by DHE uores-
cence. NOX4 siRNA inhibited BMP4-induced cleaved p38 phosphorylation (C), JNK
phosphorylation (D), and caspase 3 expression (E and F). Data are meanSEM
(n=4).

Pb0.05 compared with control;
#
Pb0.05 compared with BMP4.
242 X.Y. Tian et al. / Journal of Molecular and Cellular Cardiology 52 (2012) 237244
activation. Treatment with 20 nmol/L JNK siRNA for 24 h reduced the
JNK expression in HUVECs (Supplemental Fig. 3C). Under the same
condition, JNK siRNA suppressed both the BMP4 (30 ng/mL)- and
ROS [both H
2
O
2
(100 mol/L) and hypoxanthine (100 mol/L) plus
xanthine oxidase (0.01 u/mL) (HXXO)]-induced EC apoptosis as deter-
mined by ow cytometry (Fig. 7A). Treatment with JNK siRNA reduced
the elevated expression of cleaved caspase-3 in response to BMP4 or
ROS in HUVECs (Fig. 7B).
4. Discussion
The present study demonstrated BMP4-induced EC apoptosis in
human, rat, and mouse. We showed that BMP4 activates caspase-3
in ECs and this action is mediated through ROS-dependent p38 and
JNK activation. Interfering with the BMP4 signaling by knockdown
of BMPR1A, NADPH oxidase subtype NOX4, or JNK prevented BMP4-
induced EC apoptosis.
Oxidative stress plays an important role in cellular events including
apoptosis in ECs [33,34]. Oscillatory shear stress up-regulates the BMP4
production, leading to inammatory responses such as increased
monocyte adhesion, through NOX1-based ROS and NF-B activation
[35,36]. BMP4 plays a crucial role inhypertension[37] andatherosclero-
sis [3840]. BMP4 infusion in mice induces hypertension through acti-
vation of NADPH oxidase [37]. In ECs, the apoptosis level increases
with inammation, and the apoptotic blebs are able to stimulate the at-
tachment of monocytes towards ECs, thus exaggerating vascular in-
ammation [4143]. In the present study, ROS scavengers or NOX4
siRNA inhibited BMP4-induced EC apoptosis, suggesting that this pro-
cess is also mediated through NADPHoxidase-derived ROS. This nding
is in line with previous studies showing that NADPH oxidase subunits,
NOX1, NOX2, and NOX4 contribute to EC apoptosis [4447]. MAPKs
play an important role in the regulation of different cellular activities
and there are three major signaling components, e.g., p38 MAPK, JNK/
SAPK and ERK. p38 MAPK and JNK/SAPK contribute to the regulation
of cell apoptosis upon stress stimuli [4850]. Depending on stimuli,
p38 MAPK and JNK/SAPK also have a role in cell proliferation and sur-
vival [50]. Onthe other hand, ERK is involved incell proliferation, differ-
entiation, and survival. MAPKs are regulated by ROS in ECs to express
proinammatory phenotype [5153]. The present study shows that
BMP4-induced EC apoptosis is associated with p38 and JNK activation
based on the following observations: (1) BMP4 increases p38 and JNK
phosphorylation; (2) inhibition of p38 and JNK reduce BMP4-induced
apoptosis; and (3) knockout of JNK inhibits BMP4-induced EC apopto-
sis. JNK signaling contributes to EC apoptosis triggered by other stimuli,
such as oxidized LDL, TNF, or high glucose [12,5456]. The present re-
sults also suggest that BMP4-induced EC apoptosis is partly mediated
through JNK activation.
BMP ligands exert both pro- and anti-apoptotic effects in ECs in dif-
ferent conditions. BMP2 and BMP4 stimulate cell proliferation and an-
giogenesis in pulmonary artery ECs [57], human microvascular ECs
[58], and endothelial precursor cells [28]. However, BMP4 can also
cause apoptosis in human ECs including HUVECs [24,25], suggesting
that the effect of BMP4 may vary depending on EC types and culture
conditions. BMPR1A is involved in apoptosis during development in
mouse [5961]. In addition, Smad1/5 signaling mediates BMP4-
induced apoptosis while Smad6/7 protects ECs from apoptosis in dif-
ferent ECs [24,61]. In the present study, we found that BMPR1A-
mediated the pro-apoptotic effect of BMP4 in ECs is inhibited by
knockdown of BMPR1A using shRNA. More importantly, BMP4-
induced Smad activation was unaffected by ROS scavengers, p38 or
JNK inhibitor, or knockdown of NOX4 (Supplemental Figure S4B).
And ROS-generating agents such as H
2
O
2
or HXXO did not increase
Smad phosphorylation (Supplemental Figure S4B), suggesting that
BMP4 and BMPR1A-induced caspase-3 activation in ECs is most likely
to be mediated through oxidative stress rather than the Smad path-
way, which differs from the previous reported ndings.
In summary, BMP4 causes EC apoptosis in cultured ECs and arteries
fromhuman, mouse, and rat. The pro-apoptotic effect of BMP4 is medi-
ated through BMPR1A. BMP4-induced caspase-3 activation is mediated
through NADPH oxidase-derived ROS and downstream activation of
p38 and JNK. These ndings extend our understanding of the positive
role of BMP4 signaling in EC apoptosis and associated vascular dysfunc-
tion under pathological situations such as hypertension.
Disclosures
None.
Acknowledgments
This study was supported by Hong Kong General Research Fund
(CUHK 465308, 466110, and 465611), National Basic Research Program
of China (2012CB517805), and CUHK Focused Investment Scheme.
Appendix A. Supplementary data
Supplementary data to this article can be found online at doi:10.
1016/j.yjmcc.2011.10.013.
References
[1] Li RH, Wozney JM. Delivering on the promise of bone morphogenetic proteins.
Trends Biotechnol Jul 2001;19(7):25565.
[2] Bostrom K, Watson KE, Horn S, Wortham C, Herman IM, Demer LL. Bone morpho-
genetic protein expression in human atherosclerotic lesions. J Clin Invest Apr
1993;91(4):18009.
[3] Dhore CR, Cleutjens JP, Lutgens E, Cleutjens KB, Geusens PP, Kitslaar PJ, et al. Differen-
tial expression of bone matrix regulatory proteins in human atherosclerotic plaques.
Arterioscler Thromb Vasc Biol Dec 2001;21(12):19982003.
[4] Schluesener HJ, Meyermann R. Immunolocalization of BMP-6, a novel TGF-
beta-related cytokine, in normal and atherosclerotic smooth muscle cells. Athero-
sclerosis Mar 1995;113(2):1536.
[5] Sorescu GP, Sykes M, Weiss D, Platt MO, Saha A, Hwang J, et al. Bone morphogenic
protein 4 produced in endothelial cells by oscillatory shear stress stimulates an in-
ammatory response. J Biol Chem Aug 15, 2003;278(33):3112835.
[6] Miriyala S, Gongora Nieto MC, Mingone C, Smith D, Dikalov S, Harrison DG, et al.
Bone morphogenic protein-4 induces hypertension in mice: role of noggin, vascu-
lar NADPH oxidases, and impaired vasorelaxation. Circulation Jun 20, 2006;113
(24):281825.
[7] Wong WT, Tian XY, Chen Y, Leung FP, Liu L, Lee HK, et al. Bone morphogenic
protein-4 impairs endothelial function through oxidative stress-dependent
cyclooxygenase-2 upregulation: implications on hypertension. Circ Res Oct 15,
2010;107(8):98491.
[8] White E. Life, death, and the pursuit of apoptosis. Genes Dev Jan 1996;10(1):115.
[9] Hetts S. To die or not to die: an overview of apoptosis and its role in disease. JAMA
Jan 1998;279(4):3007.
[10] Bauriedel G, Hutter R, Welsch U, Bach R, Sievert H, Lderitz B. Role of smooth
muscle cell death in advanced coronary primary lesions: implications for plaque
instability. Cardiovasc Res Feb 1999;41(2):4808.
[11] Okura T, Watanabe S, Jiang Y, Nakamura M, Takata Y, Yang Z, et al. Soluble Fas li-
gand and atherosclerosis in hypertensive patients. J Hypertens May 2002;20(5):
8958.
[12] Ho FM, Liu SH, Liau CS, Huang PJ, Lin-Shiau SY. High glucose-induced apoptosis in
humanendothelial cells is mediatedby sequential activations of c-JunNH(2)-terminal
kinase and caspase-3. Circulation Jun 6, 2000;101(22):261824.
[13] Choy J, Granville D, Hunt D, McManus B. Endothelial cell apoptosis: biochemical
characteristics and potential implications for atherosclerosis. J Mol Cell Cardiol
Sep 2001;33(9):167390.
[14] Tricot O, Mallat Z, Heymes C, Belmin J, Lesche G, Tedgui A. Relation between en-
dothelial cell apoptosis and blood ow direction in human atherosclerotic pla-
ques. Circulation May 2000;101(21):24503.
[15] Coleman ML, Sahai EA, Yeo M, Bosch M, Dewar A, Olson MF. Membrane blebbing
during apoptosis results from caspase-mediated activation of ROCK I. Nat Cell Biol
Apr 2001;3(4):33945.
[16] Mallat Z, Tedgui A. Current perspective on the role of apoptosis in atherothrom-
botic disease. Circ Res May 25, 2001;88(10):9981003.
[17] Huber J, Vales A, Mitulovic G, Blumer M, Schmid R, Witztum JL, et al. Oxidized
membrane vesicles and blebs from apoptotic cells contain biologically active oxi-
dized phospholipids that induce monocyteendothelial interactions. Arterioscler
Thromb Vasc Biol Jan 2002;22(1):1017.
[18] Hwang J, Saha A, Boo YC, Sorescu GP, McNally JS, Holland SM, et al. Oscillatory
shear stress stimulates endothelial production of O2- from p47phox-dependent
NAD(P)H oxidases, leading to monocyte adhesion. J Biol Chem Nov 21,
2003;278(47):472918.
243 X.Y. Tian et al. / Journal of Molecular and Cellular Cardiology 52 (2012) 237244
[19] Li AE, Ito H, Rovira II, KimKS, Takeda K, Yu ZY, et al. A role for reactive oxygen species
in endothelial cell anoikis. Circ Res Aug 20, 1999;85(4):30410.
[20] Jeffery T, Upton P, Trembath R, Morrell N. BMP4 inhibits proliferation and pro-
motes myocyte differentiation of lung broblasts via Smad1 and JNK pathways.
Am J Physiol Lung Cell Mol Physiol Feb 2005;288(2):L3708.
[21] Yang X, Lee P, Long L, Trembath R, Morrell N. BMP4 induces HO-1 via a Smad-
independent, p38MAPK-dependent pathway in pulmonary artery myocytes. Am
J Respir Cell Mol Biol Nov 2007;37(5):598605.
[22] Guyton K, Liu Y, Gorospe M, Xu Q, Holbrook N. Activation of mitogen-activated
protein kinase by H2O2. Role in cell survival following oxidant injury. J Biol
Chem Feb 1996;271(8):413842.
[23] Lagna G, Nguyen PH, Ni W, Hata A. BMP-dependent activation of caspase-9 and
caspase-8 mediates apoptosis inpulmonary artery smooth muscle cells. AmJ Physiol
Lung Cell Mol Physiol Nov 2006;291(5):L105967.
[24] Kiyono M, Shibuya M. Inhibitory Smad transcription factors protect arterial endo-
thelial cells from apoptosis induced by BMP4. Oncogene Nov 16, 2006;25(54):
71317.
[25] Kiyono M, Shibuya M. Bone morphogenetic protein 4 mediates apoptosis of capillary
endothelial cells during rat pupillary membrane regression. Mol Cell Biol Jul 2003;23
(13):462736.
[26] Zhou X, Sheng Y, Yang R, Kong X. Nicotine promotes cardiomyocyte apoptosis via
oxidative stress and altered apoptosis-related gene expression. Cardiology
2010;115(4):24350.
[27] Frank DB, Abtahi A, Yamaguchi DJ, Manning S, Shyr Y, Pozzi A, et al. Bone morpho-
genetic protein 4 promotes pulmonary vascular remodeling in hypoxic pulmo-
nary hypertension. Circ Res Sep 2, 2005;97(5):496504.
[28] Heinke J, Wehofsits L, Zhou Q, Zoeller C, Baar KM, Helbing T, et al. BMPER is an en-
dothelial cell regulator and controls bone morphogenetic protein-4-dependent
angiogenesis. Circ Res Oct 10, 2008;103(8):80412.
[29] Liu CQ, Leung FP, Wong SL, Wong WT, Lau CW, Lu L, et al. Thromboxane prostanoid
receptor activation impairs endothelial nitric oxide-dependent vasorelaxations: the
role of Rho kinase. Biochem Pharmacol Aug 15, 2009;78(4):37481.
[30] Chen Y, Stamatoyannopoulos G, Song CZ. Down-regulation of CXCR4 by inducible
small interfering RNA inhibits breast cancer cell invasion in vitro. Cancer Res Aug
15, 2003;63(16):48014.
[31] Chen Y, Lin MC, Yao H, Wang H, Zhang AQ, Yu J, et al. Lentivirus-mediated RNA inter-
ference targeting enhancer of zeste homolog 2 inhibits hepatocellular carcinoma
growth through down-regulation of stathmin. Hepatology Jul 2007;46(1):2008.
[32] Jo H, Song H, Mowbray A. Role of NADPH oxidases in disturbed ow- and BMP4-
induced inammation and atherosclerosis. Antioxid Redox Signal Sep-Oct 2006;8
(910):160919.
[33] Cuda G, Paterno R, Ceravolo R, Candigliota M, Perrotti N, Perticone F, et al. Protec-
tion of human endothelial cells from oxidative stress: role of Ras-ERK1/2 signal-
ing. Circulation Feb 26, 2002;105(8):96874.
[34] Park S, Kim JA, Choi S, Suh SH. Superoxide is a potential culprit of caspase-3 de-
pendent endothelial cell death induced by lysophosphatidylcholine. J Physiol
Pharmacol Aug 2010;61(4):37581.
[35] Sorescu G, Sykes M, Weiss D, Platt M, Saha A, Hwang J, et al. Bone morphogenic
protein 4 produced in endothelial cells by oscillatory shear stress stimulates an
inammatory response. J Biol Chem Aug 2003;278(33):3112835.
[36] Sorescu G, Song H, Tressel S, Hwang J, Dikalov S, Smith D, et al. Bone morphogenic
protein 4 produced in endothelial cells by oscillatory shear stress induces mono-
cyte adhesion by stimulating reactive oxygen species production from a nox1-
based NADPH oxidase. Circ Res Oct 2004;95(8):7739.
[37] Miriyala S, Gongora Nieto M, Mingone C, Smith D, Dikalov S, Harrison D, et al. Bone
morphogenic protein-4 induces hypertension in mice: role of noggin, vascular
NADPHoxidases, andimpairedvasorelaxation. CirculationJun2006;113(24):281825.
[38] Dhore C, Cleutjens J, Lutgens E, Cleutjens K, Geusens P, Kitslaar P, et al. Differential
expression of bone matrix regulatory proteins in human atherosclerotic plaques.
Arterioscler Thromb Vasc Biol Dec 2001;21(12):19982003.
[39] Bostrm K, Watson K, Horn S, Wortham C, Herman I, Demer L. Bone morphoge-
netic protein expression in human atherosclerotic lesions. J Clin Invest Apr
1993;91(4):18009.
[40] Mohler ER, Gannon F, Reynolds C, Zimmerman R, Keane M, Kaplan F. Bone forma-
tion and inammation in cardiac valves. Circulation Mar 2001;103(11):15228.
[41] Coleman M, Sahai E, Yeo M, Bosch M, Dewar A, Olson M. Membrane blebbing dur-
ing apoptosis results from caspase-mediated activation of ROCK I. Nat Cell Biol
Apr 2001;3(4):33945.
[42] Mallat Z, Tedgui A. Current perspective on the role of apoptosis in atherothrom-
botic disease. Circ Res May 2001;88(10):9981003.
[43] Huber J, Vales A, Mitulovic G, Blumer M, Schmid R, Witztum J, et al. Oxidized
membrane vesicles and blebs from apoptotic cells contain biologically active oxi-
dized phospholipids that induce monocyteendothelial interactions. Arterioscler
Thromb Vasc Biol Jan 2002;22(1):1017.
[44] Quagliaro L, Piconi L, Assaloni R, Martinelli L, Motz E, Ceriello A. Intermittent high
glucose enhances apoptosis related to oxidative stress in human umbilical vein
endothelial cells: the role of protein kinase C and NAD(P)H-oxidase activation. Di-
abetes Nov 2003;52(11):2795804.
[45] Li JM, Fan LM, George VT, Brooks G. Nox2 regulates endothelial cell cycle arrest
and apoptosis via p21cip1 and p53. Free Radic Biol Med Sep 15, 2007;43(6):
97686.
[46] Basuroy S, Bhattacharya S, Lefer CW, Parfenova H. Nox4 NADPH oxidase medi-
ates oxidative stress and apoptosis caused by TNF-alpha in cerebral vascular en-
dothelial cells. Am J Physiol Cell Physiol Mar 2009;296(3):C42232.
[47] Teng RJ, Eis A, Bakhutashvili I, Arul N, Konduri GG. Increased superoxide produc-
tion contributes to the impaired angiogenesis of fetal pulmonary arteries with in
utero pulmonary hypertension. Am J Physiol Lung Cell Mol Physiol Jul 2009;297
(1):L18495.
[48] Ono K, Han J. The p38 signal transduction pathway: activation and function. Cell
Signal Jan 2000;12(1):113.
[49] Barr RK, Bogoyevitch MA. The c-Jun N-terminal protein kinase family of mitogen-
activated protein kinases (JNK MAPKs). Int J Biochem Cell Biol Nov 2001;33(11):
104763.
[50] Junttila MR, Li SP, Westermarck J. Phosphatase-mediated crosstalk between
MAPK signaling pathways in the regulation of cell survival. FASEB J Apr 2008;22
(4):95465.
[51] Griendling KK, Sorescu D, Lassegue B, Ushio-Fukai M. Modulation of protein ki-
nase activity and gene expression by reactive oxygen species and their role in vas-
cular physiology and pathophysiology. Arterioscler Thromb Vasc Biol Oct 2000;20
(10):217583.
[52] Csiszar A, Ahmad M, Smith KE, Labinskyy N, Gao Q, Kaley G, et al. Bone morpho-
genetic protein-2 induces proinammatory endothelial phenotype. Am J Pathol
Feb 2006;168(2):62938.
[53] Anilkumar N, Weber R, Zhang M, Brewer A, Shah AM. Nox4 and nox2 NADPH ox-
idases mediate distinct cellular redox signaling responses to agonist stimulation.
Arterioscler Thromb Vasc Biol Jul 2008;28(7):134754.
[54] Takabe W, Li R, Ai L, Yu F, Berliner JA, Hsiai TK. Oxidized low-density lipoprotein-
activated c-Jun NH2-terminal kinase regulates manganese superoxide dismutase
ubiquitination: implication for mitochondrial redox status and apoptosis. Arter-
ioscler Thromb Vasc Biol Mar 2010;30(3):43641.
[55] Garin G, Abe J, Mohan A, Lu W, Yan C, Newby AC, et al. Flow antagonizes TNF-
alpha signaling in endothelial cells by inhibiting caspase-dependent PKC zeta pro-
cessing. Circ Res Jul 6, 2007;101(1):97105.
[56] Ho FM, Lin WW, Chen BC, Chao CM, Yang CR, Lin LY, et al. High glucose-induced
apoptosis in human vascular endothelial cells is mediated through NF-kappaB
and c-Jun NH2-terminal kinase pathway and prevented by PI3K/Akt/eNOS path-
way. Cell Signal Mar 2006;18(3):3919.
[57] Teichert-Kuliszewska K, Kutryk MJ, Kuliszewski MA, Karoubi G, Courtman DW,
Zucco L, et al. Bone morphogenetic protein receptor-2 signaling promotes pulmo-
nary arterial endothelial cell survival: implications for loss-of-function mutations
in the pathogenesis of pulmonary hypertension. Circ Res Feb 3, 2006;98(2):
20917.
[58] Suzuki Y, Montagne K, Nishihara A, Watabe T, Miyazono K. BMPs promote prolif-
eration and migration of endothelial cells via stimulation of VEGF-A/VEGFR2 and
angiopoietin-1/Tie2 signalling. J Biochem Feb 2008;143(2):199206.
[59] El-Bizri N, Guignabert C, Wang L, Cheng A, Stankunas K, Chang CP, et al. SM22al-
pha-targeted deletion of bone morphogenetic protein receptor 1A in mice impairs
cardiac and vascular development, and inuences organogenesis. Development
Sep 2008;135(17):298191.
[60] Suzuki K, Bachiller D, Chen YP, Kamikawa M, Ogi H, Haraguchi R, et al. Regulation
of outgrowth and apoptosis for the terminal appendage: external genitalia devel-
opment by concerted actions of BMP signaling [corrected]. Development Dec
2003;130(25):620920.
[61] Kiyono M, Shibuya M. Bone morphogenetic protein 4 mediates apoptosis of cap-
illary endothelial cells during rat pupillary membrane regression. Mol Cell Biol
Jul 2003;23(13):462736.
244 X.Y. Tian et al. / Journal of Molecular and Cellular Cardiology 52 (2012) 237244