Angiopoietin-2 Exocytosis Is Stimulated by
Sphingosine-1-Phosphate in Human Blood and Lymphatic
Endothelial Cells
Cholsoon Jang, Young Jun Koh, Nam Kyu Lim, Hyun Jung Kang, Duk Hoon Kim, Sung Kwang Park,
Gyun Min Lee, Choon Ju Jeon, Gou Young Koh
Downloaded from http://atvb.ahajournals.org/ by guest on June 17, 2017
Objective—Although diverse functions of angiopoietin-2 (Ang2) have been revealed, little is known about upstream
signaling molecules regulating Ang2 exocytosis. We therefore investigated the mechanism of Ang2 exocytosis in human
blood and lymphatic endothelial cells (BECs and LECs) by stimulation with sphingosine-1-phosphate (S1P).
Methods and Results—By immunostaining and ELISA analyses using our newly developed human Ang2-specific
antibodies, Ang2 exocytosis from human endothelial cells was examined. Both exogenous and endogenous S1P trigger
rapid Ang2 exocytosis in time- and dose-dependent manners. Intriguingly, S1P-induced Ang2 exocytosis is higher in
LECs than BECs. These effects of S1P are mainly mediated by the endothelial differentiation gene receptor 1, which
subsequently activates its downstream phospholipase C and intracellular calcium mobilization to trigger Ang2
exocytosis. Consistently, S1P also dramatically stimulates Ang2 exocytosis from the ECs of ex vivo–incubated blood
vessels.
Conclusion—These results imply that the rapid secretion of Ang2 by exocytosis from endothelial cells is another possible
mechanism underlying S1P-induced angiogenesis and inflammation. (Arterioscler Thromb Vasc Biol. 2009;29:401-407.)
Key Words: angiogenesis 䡲 angiopoietin 䡲 endothelial differentiation gene receptor 䡲 lymphangiogenesis
䡲 Weibel-Palade bodies
A
sions compared to blood endothelial cells (BECs).12 Lymphatic vessels function as a key route for immune cell
trafficking and interstitial fluid drainage, and directly participate in inflammation and tissue homeostasis.13,14 Interestingly, Ang2 knockout mice exhibit defective lymphatic vessel
development,6 implicating essential role of Ang2 in development and function of LECs. Although the expression of Ang2
in LECs has been documented,15,16 presence of Ang2 granules and regulation of Ang2 secretion are still unknown.
Sphingosine-1-phosphate (S1P) is a phospholipid signaling
molecule generated from the plasma-membrane sphingosine
by sphingosine kinases 1 and 2.17 S1P is mainly secreted from
platelets, monocytes, and mast cells, thereby governing various cellular processes including cell proliferation, survival,
and migration.18 Moreover, S1P has a dual mode of action as
an extracellular ligand for cell-surface receptors and an
intracellular second messenger.19 In fact, diverse functions of
S1P are mostly contributed by its G-protein coupled–receptors (GPCR), which belong to the endothelial differentiation
gene (EDG) family.17,18,20 Among several EDG members,
only EDG1 and EDG3 are expressed to mediate diverse
effects of S1P on BEC physiologies like adherens junction
ngiopoietin-1 (Ang1) and angiopoietin-2 (Ang2) were
identified as key regulators of pathophysiological angiogenesis1– 4 along with vascular endothelial growth factors
(VEGFs). Both Ang1 and Ang2 are secretory proteins and
have an amino-terminal coiled-coil domain for their oligomerization and a carboxy-terminal fibrinogen-like domain
(FLD) for binding to their receptor, Tie2.1,2 In contrast to the
consistent role of Ang1 in stabilizing newly-formed vessels,1,4 more versatile functions have been reported for Ang2
such as angiogenesis, lymphangiogenesis, vascular permeability, and inflammation.2,5–9 Intriguingly, Ang2 is synthesized in advance and stored in Weibel-Palade bodies (WPBs)
of endothelial cells (ECs), thereby being rapidly secreted to
the external microenvironment by exocytosis when stimulated by extracellular signals.10,11 This implies that Ang2
activity is temporally regulated to exert controlled actions
onto Tie2-expressing cells via autocrine and paracrine rather
than endocrine manner. However, despite the importance of
Ang2 exocytosis control, signaling pathways regulating Ang2
exocytosis are poorly understood.
Lymphatic endothelial cells (LECs) have their own characteristic morphologies, functions, and specific gene expres-
Received June 22, 2008; revision accepted December 1, 2008.
From the National Research Laboratory of Vascular Biology and Department of Biological Sciences (C.J., Y.J.K., G.M.L., G.Y.K.), KAIST, Daejeon;
the Biotherapeutic Division (C.J., N.K.L., H.J.K., C.J.J.), GenExel-Sein Inc, Daejeon; and the Department of Internal Medicine (D.H.K., S.K.P.), Chonbuk
National University Medical School, Jeonju, Republic of Korea.
Correspondence to Gou Young Koh, National Research Laboratory for Vascular Biology, Department of Biological Sciences, KAIST, 373-1,
Guseong-dong, Daejeon, 305-701, Republic of Korea. E-mail [email protected]
© 2009 American Heart Association, Inc.
Arterioscler Thromb Vasc Biol is available at http://atvb.ahajournals.org
401
DOI: 10.1161/ATVBAHA.108.172676
402
Arterioscler Thromb Vasc Biol
Ang2 DAPI
A
vWF Prox-1
Merged
March 2009
B
HLEC
Ang2 DAPI
HLEC
Ang2 DAPI
HUVEC
Vein
Downloaded from http://atvb.ahajournals.org/ by guest on June 17, 2017
assembly and cell motility.21–23 However, little is known
about the role of S1P and its downstream signaling in LECs
during pathophysiological conditions.
In this study, we provide the first direct evidence of Ang2
exocytosis in primary cultured human LECs and in the
endothelial cells of ex vivo-incubated blood vessels. We also
define downstream signaling pathways of S1P-induced Ang2
exocytosis both in BECs and LECs.
Methods
Cell Culture and Reagent Treatment
HUVECs were prepared from human umbilical cords by collagenase
digestion as described previously.24 HLECs were purchased from
Clonetics (Walkersville, Md). Cells (4 to 6 passages) grown in
EBM-2 with supplements were starved with serum-free EBM-2 for
12 hours before treatment. The following reagents were treated:
10 ␮mol/L DMS for 1 hour; 200 ng/mL PTX for 12 hours;
10 ␮mol/L BAPTA or BAPTA-AM in DMEM, or calcium-free
DMEM for 10 minutes. Appropriate dissolving agents were treated
in parallel as controls. The Ang2 secretion after treatment was
normalized against the Ang2 secretion before treatment to eliminate
possible variance among samples.
Ang2 ELISA Analysis
Ninety-six–well plates were coated with 100 ng/well 4A1A3 in
coating solution (7.5 mmol/L Na2CO3, 17.5 mmol/L NaHCO3, pH
9.6) for o/n at 4°C. After blocking with 1% BSA in PBS for 1 hour
at RT, 200 ␮L of serially diluted standard Ang2 (from 1500 to 46.9
pg/mL) and samples (culture supernatant without dilution) were
incubated for 1.5 hour at RT. After incubating with 100 ng/well
biotin-labeled 4B33A11 for 1.5 hour at RT, streptavidine-HRP
(1:2500, Pierce) was added for 30 minutes at RT. Tetramethylbenzidine (Sigma) was used to visualize the signal. The absorbance was
determined at 450 nm using a SOFTmaxPRO ELISA reader (Molecular Devices).
Immunofluorescence Staining
ECs grown on 0.1% gelatin-coated coverslips or umbilical cord
slices were fixed with 3.7% paraformaldehyde in PBS for 30
minutes. After washing with 0.1% PBST, the samples were permeabilized with 0.5% PBST for 10 minutes. The cords were incubated
in 20% sucrose in PBS for o/n at 4°C, imbedded in OCT compound,
and sectioned using a Cryotome set to 30 ␮m thickness. The sections
were permeabilized with 0.1% PBST for 10 minutes. The samples
were incubated with blocking solution (5% BSA and 5% donkey
serum in 0.1% PBST) for 30 minutes and incubated with appropriate
primary antibodies in blocking solution for o/n at 4°C. The primary
antibodies were (1) mouse anti-human Ang2 antibody, 4B33A11
Artery
Figure 1. Development of Ang2 antibody
to visualize and quantitate Ang2 in ECs. A,
ECs were coimmunostained for Ang2
(green) and vWF (red, top and bottom) or
Prox-1 (red, middle). Scale bars, 5 ␮m. B,
Umbilical cord vessels were immunostained for Ang2 (green). Lower panels
show magnified views of the dotted
squares in upper panels. White bars,
100 ␮m; yellow bars, 20 ␮m.
(1:100); (2) rabbit anti-human von Willebrand factor (vWF; 1:200,
Sigma); (3) rabbit anti-human Prox-1 (1:200, Reliatech, Germany);
(4) rabbit anti-human EDG1 (1:100, Abcam); (5) goat anti-human
EDG3 (1:50, Santa Cruz Biotechnology); (6) FITC-conjugated
mouse anti-human smooth muscle actin (1:200, Sigma). After
washing with 0.1% PBST, the samples were incubated with appropriate secondary antibodies (1:1000) and DAPI (1:10000) in blocking solution for 3 hours. The samples were mounted with fluorescence mounting medium and observed using a Zeiss LSM 510
confocal microscope (Carl Zeiss).
The methods for generation of Ang2 antibodies, measurement of
S1P concentration, ex vivo incubation of umbilical cords, RT-PCR
analyses, siRNA transfection, Matrigel plug assay, and statistics are
described in supplemental materials (available online at
http://atvb.ahajournals.org).
Results
Visualization and Quantification of Human Ang2
in Endothelial Cells
Immunofluorescence staining using our newly developed
anti-human Ang2 antibody (clone No. 4B33A11) revealed
typical granular Ang2 localization in primary cultured
HUVECs (Figure 1A, top). These Ang2 granules were
colocalized with multimeric protein vWF, a major constituent
of WPB11 (Figure 1A, top), which is consistent with the
previous report.10 We also examined whether primary cultured HLECs contain Ang2 granules. HLECs were marked by
the LEC-specific transcriptional factor Prox-1 in the nuclei
(Figure 1A, middle).13 Intriguingly, Ang2 granules were
colocalized with vWF but unevenly localized to one side of
the perinuclear region of HLECs (Figure 1A, bottom). This
was different from the distribution of Ang2 granules in
HUVECs, in which they were widely dispersed in the
cytoplasm (Figure 1A, top). Using the same antibody, we also
detected Ang2 granules in the ECs of human umbilical cord
blood vessels (Figure 1B). Notably, immunopositive Ang2
signals in venous ECs were greater than those in arterial ECs
(Figure 1B). In contrast, no immunopositive Ang2 signals
were detected in nonendothelial cells of the umbilical cord
(supplemental Figure I). Immunoblotting analyses revealed
that clone 4B33A11 detected both nonreduced and reduced
forms of full-length and FLD of Ang2 protein, whereas clone
4A1A3 recognized only the nonreduced form of full-length
Ang2, indicating that they recognize different structural
epitopes on Ang2 protein (data not shown). By combining
Jang et al
HLEC
#
HUVEC
HLEC
400
*
200
0
C
0.1
1.0
#
* *
* * *
10
100 1000
S1P (nmol/L)
C
800
Ang2 (pg/ml)
Ang2 (pg/ml)
600
S1P 10 nmol/L S1P 1 mol/L
Control
S1P 1 mol/L
Ang2 vWF
HUVEC
B
S1P 10 nmol/L
403
Ang2 Prox-1
Control
A
Ang2 Exocytosis Stimulated by S1P
600
400
200
0
#
*
* *
*
* * *
#
10
HUVEC
#
HLEC
20 30
Time (min)
*
*
60
Downloaded from http://atvb.ahajournals.org/ by guest on June 17, 2017
Figure 2. Ang2 exocytosis is stimulated by exogenous S1P. A, ECs were coimmunostained for Ang2 (green) and vWF or Prox-1 (red)
after treatment with the indicated concentrations of S1P for 20 minutes. Scale bars, 10 ␮m. B, ECs were treated with control (C) or the
indicated concentrations of S1P for 20 minutes, and Ang2 secretion was measured by ELISA. C, ECs were treated with control
(squares) or 100 nmol/L S1P (circles) for the indicated times, and Ang2 secretion was measured by ELISA. *P⬍0.05 vs control (B) or
Time 0 (C); #P⬍0.05 vs HUVECs. n⫽4.
these antibodies, we developed a new human Ang2 ELISA
with high sensitivity (limit of quantification, ⬇50 pg/mL).
This ELISA method did not show any cross-reactivity with
other angiopoietins, including Ang3, Ang4, and COMPAng1, a potent and soluble variant of Ang125 (data not
shown), demonstrating high specificity of the Ang2 ELISA.
These sensitive and specific Ang2 antibodies allowed us to
investigate upstream signaling molecules regulating Ang2
exocytosis in human ECs.
Ang2 Exocytosis Is Stimulated by S1P in BECs
and LECs
Because both S1P and Ang2 modulate inflammation and
angiogenesis,5,17,18 and BECs abundantly express S1P receptors,21 we examined whether S1P stimulates Ang2 exocytosis
in vascular ECs. On various doses of S1P (0.1 to 1000
nmol/L) treatment, there were remarkable dose-dependent
depletions of Ang2 granules in primary cultured HUVECs
(Figure 2A, left). Consistently, S1P increased the amount of
Ang2 secretion in a dose-dependent manner (Figure 2B),
indicating that S1P actively stimulates Ang2 exocytosis from
BECs.10 Time-course experiments showed that S1P-induced
Ang2 exocytosis had already begun by 5 minutes after S1P
treatment (1 ␮mol/L), and it was much greater than the basal
level of Ang2 secretion (Figure 2C). Ang2 exocytosis occurred ⬇44% at 5 minutes and ⬇77% at 10 minutes during 1
hour of S1P stimulation (Figure 2C). Similarly, Ang2 exocytosis from HLECs also occurred in a S1P dose-dependent
manner, but the amounts of Ang2 secretion in HLECs were
greater (⬇1.4- to 2.1-fold) than those in HUVECs (Figure 2A
and 2B). Moreover, Ang2 secretion in HLECs was also rapid
and the increased amounts were again higher (⬇1.6- to
2.3-fold) than those in HUVECs (Figure 2C). On the S1P
stimulation, the greater Ang2 secretion from LECs than BECs
could be owing to higher intracellular Ang2 protein storage in
LECs than BECs (supplemental Figure II). However, the
treatment of S1P (1 ␮mol/L) did not affect Ang2 mRNA
synthesis up to 12 hours in both ECs (supplemental Figure
III), indicating that S1P stimulates Ang2 exocytosis without
immediate synthesis of Ang2.
TNF-␣–Induced Endogenous S1P Synthesis
Stimulates Ang2 Exocytosis
These observations led us to examine the effect of endogenous S1P on Ang2 exocytosis. Endogenous S1P is generated
by 2 sphingosine kinases and secreted via unknown mechanisms as an extracellular signaling molecule.17,18 Because
TNF-␣ is one of the most powerful stimulators of sphingosine
kinases,17 we treated HUVECs and HLECs with TNF-␣ and
assessed its effect on S1P secretion. Consequently, TNF-␣
(100 ng/mL) treatment induced S1P secretion from both
HUVECs and HLECs (Figure 3A). The secretion of S1P was
the highest at 10 minutes after TNF-␣ treatment, and its level
was decreased over time (Figure 3A). Consistently, various
doses of TNF-␣ (0.1 to 100 ng/mL) increased Ang2 exocytosis in a dose-dependent manner, and the secreted Ang2
levels were higher (⬇1.3- to 2.4-fold) in HLECs than
HUVECs (Figure 3B). To determine whether TNF-␣–induced Ang2 exocytosis is mediated by newly synthesized
S1P, we pretreated HUVECs and HLECs with a sphingosine
kinase inhibitor, dimethylsphingosine (DMS, 10 ␮mol/L),
before stimulating with S1P (1 ␮mol/L) or TNF-␣ (100
ng/mL). As a result, DMS substantially blocked Ang2 exocytosis by TNF-␣ but not by exogenous S1P (Figure 3C and
supplemental Figure IV), suggesting that newly synthesized
endogenous S1P stimulates Ang2 exocytosis in ECs.
PTX-Sensitive Receptor Mediates Extracellular
S1P-Induced Ang2 Exocytosis
Because S1P functions as an extracellular ligand or an
intracellular second messenger,19,20 we investigated which
form of S1P is responsible for inducing Ang2 exocytosis. We
used dehydro-S1P (dhS1P), a variant of S1P that has only
extracellular but not intracellular activity.26 Treatment of
various doses of S1P or dhS1P displayed similar efficacies in
stimulating Ang2 exocytosis in ECs (Figure 3D and supple-
80
60
40
20
0
C
*
*
*
*
5
*
*
**
400
200
*
*
- -TT
- + -+
HUVEC
HLEC
* *
* * *
* *
600
400
200
0
C
0.1 1 10 100
TNF-α (ng/ml)
HLEC
600
#
B
30
D
0
S or T - - S S
DMS - + - +
*
10
20
Time (min)
HLEC
600
Ang2 (pg/ml)
100ng/ml TNF-α
HUVEC
HLEC
Ang2(pg/ml)
S1P(nmol/L)
A
March 2009
S1P
dhS1P
400
200
E
HLEC
600
*
*
*#
*
S or T - - S S
PTX - + - +
- -TT
- +- +
Ang2 (pg/ml)
Arterioscler Thromb Vasc Biol
Ang2 (pg/ml)
404
0
400
200
0
C 1.0 10 100 1000
S1P (nM)
Downloaded from http://atvb.ahajournals.org/ by guest on June 17, 2017
mental Figure IV), suggesting that extracellular S1P stimulates Ang2 exocytosis. Extracellular S1P conveys its downstream signaling via plasma membrane GPCR.17,18 To
examine whether GPCR is involved in S1P-induced Ang2
exocytosis, HUVECs and HLECs were pretreated with a
specific GPCR blocker, PTX (200 ng/mL),21 before S1P
(1 ␮mol/L) or TNF-␣ (100 ng/mL) stimulation. Pretreatment
of PTX markedly blocked Ang2 exocytosis by S1P or TNF-␣
(Figure 3E and supplemental Figure IV), implicating that
GPCR is critical for mediating S1P-induced Ang2 exocytosis
in ECs.
EDG1 Mainly Mediates S1P-Induced
Ang2 Exocytosis
Because the EDG receptor family is the subset of GPCR that
mediates S1P-dependent signaling pathway,17,18 we assessed
the expression of various EDG receptors in HUVECs and
HLECs. RT-PCR analyses revealed that both ECs selectively
express EDG1 and EDG3 but not EDG5, EDG6, or EDG8
(Figure 4A), and quantitative RT-PCR analyses showed that
the expression levels of EDG1 were much higher (13.4-fold
Figure 3. Ang2 exocytosis is stimulated
by endogenous S1P via PTX-sensitive
receptors. A through E, Ang2 secretion
was measured by ELISA (n⫽4). A, ECs
pretreated with control (circles) or
10 ␮mol/L DMS (squares) for 1 hour were
treated with control (triangles) or TNF-␣. B
and D, ECs were treated with TNF-␣ (B),
S1P (D), or dhS1P (D) for 20 minutes. C
and E, HLECs pretreated with 10 ␮mol/L
DMS (C) or 200 ng/mL PTX (E) were
treated with S1P (S, 1 ␮mol/L) or TNF-␣
(T, 100 ng/mL) for 20 minutes. *P⬍0.05 vs
control. #P⬍0.05 vs S1P or TNF-␣.
in HUVECs and 18.7-fold in HLECs) than those of EDG3 in
ECs (supplemental Figure V), which is consistent with the
previous report.21 Moreover, immunostaining analyses
showed that EDG1 and EDG3 dramatically translocated from
plasma membrane into cytoplasm on S1P stimulation probably via receptor internalization27 (Figure 4B), suggesting that
EDG receptors on EC surface are actively responsive to S1P.
To further examine role of EDG1 and EDG3 in Ang2
exocytosis, we selectively reduced expression of EDG1 or
EDG3 using siRNAs targeting each receptor (si1 or si3,
respectively) and a scrambled random siRNA (sc) as a
negative control. RT-PCR and immunostaining analyses
showed that si1 or si3 but not sc selectively reduced mRNA
and protein levels of EDG1 or EDG3, respectively (Figure 4C
and 4D). Interestingly, S1P-induced Ang2 exocytosis was
blocked by si1 but not by si3 or sc in HUVECs and HLECs
(Figure 4E). Moreover, cotransfection of si1 and si3 did not
further decrease the si1-induced reduction of Ang2 exocytosis (Figure 4E). These data suggest that EDG1 is a major
subtype of EDG receptor mediating S1P-induced Ang2 exocytosis in both EC types.
Figure 4. EDG1 mainly mediates S1Pinduced Ang2 exocytosis. A, RT-PCR analyses of EDG receptors. GAPDH is used as
control. B and D, HUVECs were immunostained for EDG1 (red) or EDG3 (green) after
S1P (1 ␮mol/L) treatment for 20 minutes (B)
or after siRNA transfection for 72 hours (D).
Scale bars, 5 ␮m. C, RT-PCR analyses of
EDG receptors in ECs transfected with
siRNA for 12 hours. E, ECs transfected with
siRNA for 72 hours were treated with S1P
(1 ␮mol/L) for 20 minutes, and Ang2 secretion was measured by ELISA (n⫽4). *P⬍0.05
vs nontransfected. #P⬍0.05 vs S1P⫹sc.
Jang et al
Ang2 Exocytosis Stimulated by S1P
405
mediates S1P-induced Ang2 exocytosis. We used Ca2⫹-free
media and membrane-permeable or -impermeable calcium
chelator (BAPTA-AM or BAPTA) to deplete intracellular or
extracellular calcium, respectively. Interestingly, Ang2 exocytosis was substantially suppressed by Ca2⫹-free media (⬇52% and
⬇60%) and BAPTA-AM pretreatment (⬇32% and ⬇37%) but
not by BAPTA pretreatment in HUVECs and HLECs (Figure
5C and supplemental Figure IV), suggesting that intracellular
calcium mediates Ang2 exocytosis in both ECs.
S1P Induces Ang2 Exocytosis From the ECs of Ex
Vivo–Incubated Umbilical Cord Vessels
Downloaded from http://atvb.ahajournals.org/ by guest on June 17, 2017
Figure 5. Phospholipase C and intracellular calcium mediate
S1P-induced Ang2 exocytosis. A through C, Ang2 secretion was
measured by ELISA (n⫽4). A, HUVECs pretreated with each
reagent (PD; 20 ␮mol/L PD98059, SB; 10 ␮mol/L SB203580,
LY; 5 ␮mol/L LY294002, U; 10 ␮mol/L U73122, Fk; 10 ␮mol/L
Forskolin) for 30 minutes were treated with S1P (1 ␮mol/L) for
20 minutes. B, HLECs pretreated with U73122 were treated with
S1P for 20 minutes. C, HLECs pretreated with DMEM containing BAPTA (B) or BAPTA-AM (A), or Ca2⫹-free DMEM (C) were
treated with S1P for 20 minutes. *P⬍0.05 vs control. #P⬍0.05
vs S1P.
Phospholipase C and Intracellular Calcium Are
Crucial Mediators of Ang2 Exocytosis by S1P
To further define downstream pathways mediating S1Pinduced Ang2 exocytosis, we examined the effect of all the
known mediators of EDG receptors, namely PI3K, ERK, p38,
phospholipase C (PLC), and adenylate cyclase on S1Pinduced Ang2 release, using specific inhibitors of each
pathway.17 As a result, only U73122, a PLC inhibitor,
significantly blocked S1P-induced Ang2 exocytosis in
HUVECs (Figure 5A). Moreover, pretreatment of U73122
(0.1 to 10 ␮mol/L) suppressed S1P-induced Ang2 exocytosis
in a dose-dependent manner in both ECs (Figure 5B and
supplemental Figure IV), indicating that PLC plays an important role in Ang2 exocytosis by S1P. Because PLCdependent calcium mobilization was known to promote
exocytosis,18 we examined whether calcium mobilization
Our findings in primary cultured ECs led us to further
investigate S1P-induced Ang2 exocytosis in ex vivoincubated umbilical cords. Cross-sectioned slices (⬇1 cm
length) of the cords were treated with control buffer, S1P, or
a powerful secretagogue, phorbol 12-myristate 13-acetate
(PMA) as a positive control. In control buffer–treated cords,
strong immunopositive Ang2 signals were specifically detected at the region of vWF-expressing ECs in the umbilical
vein (Figure 6A, left). In contrast, S1P (1 ␮mol/L)- or PMA
(50 ng/mL)-treated cords displayed considerably decreased
immunopositive Ang2 signals (Figure 6A, middle and data
not shown). Higher magnification images confirmed that
Ang2 signals were largely disappeared in the ECs of the
umbilical vein treated with S1P (Figure 6B). Consistently, the
amounts of Ang2 secretion were much higher in S1P- or
PMA-treated cords than control (Figure 6C). To exclude any
Ang2 secretion from non-ECs of cords, cross-sectioned slices
of the endothelium-denuded cord were treated with S1P
(Figure 6A, right). Basal amount of Ang2 secretion from the
endothelium-denuded cord was very low (⬇14% of intact
cord), and no substantial increase in S1P-induced Ang2
secretion was observed (Figure 6C), supporting that S1P
stimulates Ang2 exocytosis from the ECs of umbilical blood
vessels. Furthermore, to examine roles of S1P-induced Ang2
in angiogenesis, we performed an angiogenesis assay using a
Matrigel plug assay. Combination of S1P (0.5 nmol) and
Ang2 (40 ng)-containing Matrigel displayed markedly higher
blood vessel formation than either S1P (0.5 nmol)- or Ang2
(40 ng) alone-containing Matrigel (supplementary Figure VI),
suggesting that S1P-induced Ang2 release from BECs may
coordinately enhance angiogenesis.
Discussion
Using our newly developed antibodies against human Ang2,
we show that a vasoactive lipid signaling molecule, S1P,
potently activates endogenous Ang2 exocytosis from the
WPBs of BECs, LECs, and the endothelial cells of ex
vivo-incubated blood vessels. WPBs are EC-specific storage
granules containing at least 10 molecules, including vWF,
P-selectin, interleukin-8, CD63, and Ang2.11 Exocytosis of
these cargo molecules from WPBs is one of the earliest
responses of vascular stimulation, and plays pivotal roles in
thrombosis and inflammation.5,10,11 Consistently, Ang2 exocytosis from WPBs also rapidly occurs within a few minutes
after stimulation, indicating that Ang2 secretion is immediate
and dynamic.
Merged
vWF
Ang2
Control
Denuded
S1P 1 mol/L S1P 1 mol/L
B
Control
S1P 1 mol/L
C
600
Ang2 (pg/ml)
A
March 2009
vWF DAPI Ang2 DAPI
Arterioscler Thromb Vasc Biol
*
400
*
200
0
#
*
C S1P PMA S1P
Denuded
D
EDG1
Merged
406
Ang2
S1P
PLC
Ca2+
ER
WPB
Downloaded from http://atvb.ahajournals.org/ by guest on June 17, 2017
Figure 6. S1P induces Ang2 exocytosis from the ECs of umbilical cord vessels. A, Ex vivo-incubated intact (left and middle) or endothelium-denuded (right) umbilical cord sections were immunostained for Ang2 (green) and vWF (red) after treatment with control, S1P
(1 ␮mol/L), or PMA (50 ng/mL) for 1 hour. Scale bars, 200 ␮m. B, Magnified views of the dotted squares in A. Scale bars, 20 ␮m. C,
Ang2 secretion in A was measured by ELISA (n⫽4). *P⬍0.05 vs control (C). #P⬍0.05 vs S1P⫹intact cord. D, A schematic model of
S1P-induced Ang2 exocytosis. ER indicates endoplasmic reticulum.
Our data show that basal and S1P-induced Ang2 secretions
are higher in LECs than BECs. It is well known that LECs are
more prone to inflammatory and angiogenic stimuli than
BECs.28,29 Moreover, lymphatic capillaries have neither surrounding mural cells nor a basement membrane and exhibit
occasional gaps between cells, thereby being more permeable
for macromolecules than tightly associated blood vessels.12–14
In this sense, highly secreted Ang2 from LECs might be more
dynamically and freely diffused to Tie2-expressing cells like
LECs, BECs, and monocytes/macrophages and activate them
more strongly in the site of inflammation and angiogenesis.5,28 –30 Further studies using BEC- or LEC-specific Ang2
knockout mice could provide novel insights for understanding distinct roles of Ang2 in BECs and LECs in vivo.
Using diverse signaling-specific blockers, we defined
downstream pathways of Ang2 exocytosis in ECs (Figure
6D). Our results suggest that EDG1 is a main signaling
pathway mediating Ang2 exocytosis because S1P-induced
Ang2 secretion is highly PTX-sensitive. Moreover, much
higher expressions of EDG1 than EDG3 in ECs further
support a major role of EDG1, rather than EDG3, in the
S1P-induced Ang2 exocytosis in ECs.21
While S1P possibly promotes inflammation by activating
various inflammatory molecules in WPBs,26 the angiogenic
function of S1P might be mainly associated with Ang2
release because Ang2 is required for dormant ECs to response
to angiogenic stimuli like Ang1 and VEGF.11,21 Ang2 was
known to potently destabilize quiescent ECs to exert angiogenesis and lymphangiogenesis with other growth factors.5,9,31–33 Indeed, we found that combined treatment of S1P
and Ang2 potently induced angiogenesis in the Matrigel plug
assay. Moreover, a recent study showed that S1P induces
lymphangiogenesis via EDG1/PLC/Ca2⫹-signaling pathway,34 which is a main pathway for S1P-induced Ang2
exocytosis in LECs as we have shown in this study. Thus, in
addition to direct roles of S1P in angiogenesis and lymphangiogenesis,21 S1P could have indirect roles in angiogenesis and lymphangiogenesis through Ang2 secretion. Con-
trary to other membrane-targeting molecules like vWF,
P-selectin, or CD63,11 Ang2 is a soluble factor acting as a
long-distance chemoattractant.5 Therefore, Ang2 could recruit circulating Tie2-expressing monocytes/macrophages to
the site of inflammation and tumor angiogenesis.30 Considering different subpopulations of WPBs (eg, mutual exclusive
localization of Ang2 and P-selectin even in the same ECs10)
and distinct mechanisms of WPB exocytosis (eg, selective
secretion of vWF by a WPB lingering kiss),35 future studies
regarding which subsets of WPBs selectively store Ang2 and
what kind of other molecules are secreted with Ang2 are
necessary.
To conclude, we propose that S1P derived from activated
immune cells and platelets leads to rapid exocytosis of Ang2
from ECs, which may exert various physiological functions
both in autocrine and paracrine manners.5,7,9 In these regards,
our present study would provide new insights into the
mechanism of cross-talk between S1P-secreting blood-born
cells and Ang2-secreting endothelial cells during dynamic
vascular homeostasis and diseases.
Acknowledgments
We thank Shinae Kang for critical reading of the manuscript.
Sources of Funding
This work was supported by the National Research Laboratory
Program (2004-02376, to G.Y.K.) of the Korean Ministry of Education, Science, and Technology.
Disclosures
None.
References
1. Suri C, Jones PF, Patan S, Bartunkova S, Maisonpierre PC, Davis S, Sato
TN, Yancopoulos GD. Requisite role of angiopoietin-1, a ligand for the
TIE2 receptor, during embryonic angiogenesis. Cell. 1996;87:1171–1180.
2. Maisonpierre PC, Suri C, Jones PF, Bartunkova S, Wiegand SJ, Radziejewski C, Compton D, McClain J, Aldrich TH, Papadopoulos N, Daly
TJ, Davis S, Sato TN, Yancopoulos GD. Angiopoietin-2, a natural an-
Jang et al
3.
4.
5.
6.
7.
8.
Downloaded from http://atvb.ahajournals.org/ by guest on June 17, 2017
9.
10.
11.
12.
13.
14.
15.
16.
17.
18.
19.
tagonist for Tie2 that disrupts in vivo angiogenesis. Science. 1997;277:
55– 60.
Yancopoulos GD, Davis S, Gale NW, Rudge JS, Wiegand SJ, Holash J.
Vascular-specific growth factors and blood vessel formation. Nature.
2000;407:242–248.
Brindle NP, Saharinen P, Alitalo K. Signaling and functions of
angiopoietin-1 in vascular protection. Circ Res. 2006;98:1014 –1023.
Fiedler U, Augustin HG. Angiopoietins: a link between angiogenesis and
inflammation. Trends Immunol. 2006;27:552–558.
Gale NW, Thurston G, Hackett SF, Renard R, Wang Q, McClain J,
Martin C, Witte C, Witte MH, Jackson D, Suri C, Campochiaro PA,
Wiegand SJ, Yancopoulos GD. Angiopoietin-2 is required for postnatal
angiogenesis and lymphatic patterning, and only the latter role is rescued
by Angiopoietin-1. Dev Cell. 2002;3:411– 423.
Fiedler U, Reiss Y, Scharpfenecker M, Grunow V, Koidl S, Thurston G,
Gale NW, Witzenrath M, Rosseau S, Suttorp N, Sobke A, Herrmann M,
Preissner KT, Vajkoczy P, Augustin HG. Angiopoietin-2 sensitizes endothelial cells to TNF-alpha and has a crucial role in the induction of
inflammation. Nat Med. 2006;12:235–239.
Bhandari V, Choo-Wing R, Lee CG, Zhu Z, Nedrelow JH, Chupp GL,
Zhang X, Matthay MA, Ware LB, Homer RJ, Lee PJ, Geick A, de
Fougerolles AR, Elias JA. Hyperoxia causes angiopoietin 2-mediated
acute lung injury and necrotic cell death. Nat Med. 2006;12:1286 –1293.
Kim KE, Cho CH, Kim HZ, Baluk P, McDonald DM, Koh GY. In vivo
actions of angiopoietins on quiescent and remodeling blood and lymphatic vessels in mouse airways and skin. Arterioscler Thromb Vasc Biol.
2007;27:564 –570.
Fiedler U, Scharpfenecker M, Koidl S, Hegen A, Grunow V, Schmidt JM,
Kriz W, Thurston G, Augustin HG. The Tie-2 ligand angiopoietin-2 is
stored in and rapidly released upon stimulation from endothelial cell
Weibel-Palade bodies. Blood. 2004;103:4150 – 4156.
Rondaij MG, Bierings R, Kragt A, van Mourik JA, Voorberg J. Dynamics
and plasticity of Weibel-Palade bodies in endothelial cells. Arterioscler
Thromb Vasc Biol. 2006;26:1002–1007.
Pepper MS, Skobe M. Lymphatic endothelium: morphological, molecular
and functional properties. J Cell Biol. 2003;163:209 –213.
Alitalo K, Tammela T, Petrova TV. Lymphangiogenesis in development
and human disease. Nature. 2005;438:946 –953.
Randolph GJ, Angeli V, Swartz MA. Dendritic-cell trafficking to lymph
nodes through lymphatic vessels. Nat Rev Immunol. 2005;5:617– 628.
Veikkola T, Lohela M, Ikenberg K, Mäkinen T, Korff T, Saaristo A,
Petrova T, Jeltsch M, Augustin HG, Alitalo K. Intrinsic versus microenvironmental regulation of lymphatic endothelial cell phenotype and
function. FASEB J. 2003;17:2006 –2013.
Vart RJ, Nikitenko LL, Lagos D, Trotter MW, Cannon M, Bourboulia D,
Gratrix F, Takeuchi Y, Boshoff C. Kaposi’s sarcoma-associated
herpesvirus-encoded interleukin-6 and G-protein-coupled receptor
regulate angiopoietin-2 expression in lymphatic endothelial cells. Cancer
Res. 2007;67:4042– 4051.
Spiegel S, Milstien S. Sphingosine-1-phosphate: an enigmatic signalling
lipid. Nat Rev Mol Cell Biol. 2003;4:397– 407.
Hannun YA, Obeid LM. Principles of bioactive lipid signalling: lessons
from sphingolipids. Nat Rev Mol Cell Biol. 2008;9:139 –150.
Van Brocklyn JR, Lee MJ, Menzeleev R, Olivera A, Edsall L, Cuvillier
O, Thomas DM, Coopman PJ, Thangada S, Liu CH, Hla T, Spiegel S.
20.
21.
22.
23.
24.
25.
26.
27.
28.
29.
30.
31.
32.
33.
34.
35.
Ang2 Exocytosis Stimulated by S1P
407
Dual actions of sphingosine-1-phosphate: extracellular through the
Gi-coupled receptor Edg-1 and intracellular to regulate proliferation and
survival. J Cell Biol. 1998;142:229 –240.
Lee MJ, Van Brocklyn JR, Thangada S, Liu CH, Hand AR, Menzeleev R,
Spiegel S, Hla T. Sphingosine-1-phosphate as a ligand for the G proteincoupled receptor EDG-1. Science. 1998;279:1552–1555.
Lee MJ, Thangada S, Claffey KP, Ancellin N, Liu CH, Kluk M, Volpi M,
Sha’afi RI, Hla T. Vascular endothelial cell adherens junction assembly
and morphogenesis induced by sphingosine-1-phosphate. Cell. 1999;99:
301–312.
Hobson JP, Rosenfeldt HM, Barak LS, Olivera A, Poulton S, Caron MG,
Milstien S, Spiegel S. Role of the sphingosine-1-phosphate receptor
EDG-1 in PDGF-induced cell motility. Science. 2001;291:1800 –1803.
Hla T, Lee MJ, Ancellin N, Paik JH, Kluk MJ. Lysophospholipids–receptor revelations. Science. 2001;294:1875–1878.
Kim I, Kim HG, So JN, Kim JH, Kwak HJ, Koh GY. Angiopoietin-1
regulates endothelial cell survival through the phosphatidylinositol 3⬘-Kinase/Akt signal transduction pathway. Circ Res. 2000;86:24 –29.
Cho CH, Kammerer RA, Lee HJ, Steinmetz MO, Ryu YS, Lee SH,
Yasunaga K, Kim KT, Kim I, Choi HH, Kim W, Kim SH, Park SK, Lee
GM, Koh GY. COMP-Ang1: a designed angiopoietin-1 variant with
nonleaky angiogenic activity. Proc Natl Acad Sci. 2004;101:5547–5552.
Matsushita K, Morrell CN, Lowenstein CJ. Sphingosine 1-phosphate
activates Weibel-Palade body exocytosis. Proc Natl Acad Sci. 2004;101:
11483–11487.
Liu CH, Thangada S, Lee MJ, Van Brocklyn JR, Spiegel S, Hla T.
Ligand-induced trafficking of the sphingosine-1-phosphate receptor
EDG-1. Mol Biol Cell. 1999;10:1179 –1190.
Baluk P, Tammela T, Ator E, Lyubynska N, Achen MG, Hicklin DJ,
Jeltsch M, Petrova TV, Pytowski B, Stacker SA, Ylä-Herttuala S, Jackson
DG, Alitalo K, McDonald DM. Pathogenesis of persistent lymphatic
vessel hyperplasia in chronic airway inflammation. J Clin Invest. 2005;
115:247–257.
Johnson LA, Clasper S, Holt AP, Lalor PF, Baban D, Jackson DG. An
inflammation-induced mechanism for leukocyte transmigration across
lymphatic vessel endothelium. J Exp Med. 2006;203:2763–2777.
Lewis CE, De Palma M, Naldini L. Tie2-expressing monocytes and tumor
angiogenesis: regulation by hypoxia and angiopoietin-2. Cancer Res.
2007;67:8429 – 8432.
Holash J, Maisonpierre PC, Compton D, Boland P, Alexander CR,
Zagzag D, Yancopoulos GD, Wiegand SJ. Vessel cooption, regression,
and growth in tumors mediated by angiopoietins and VEGF. Science.
1999;284:1994 –1998.
Lobov IB, Brooks PC, Lang RA. Angiopoietin-2 displays VEGFdependent modulation of capillary structure and endothelial cell survival
in vivo. Proc Natl Acad Sci U S A. 2002;99:11205–11210.
Scharpfenecker M, Fiedler U, Reiss Y, Augustin HG. The Tie-2 ligand
angiopoietin-2 destabilizes quiescent endothelium through an internal
autocrine loop mechanism. J Cell Sci. 2005;118:771–780.
Yoon CM, Hong BS, Moon HG, Lim S, Suh PG, Kim YK, Chae CB, Gho
YS. Sphingosine-1-phosphate promotes lymphangiogenesis by stimulating S1P1/Gi/PLC/Ca2⫹ signaling pathways. Blood. 2008;112:
1129 –1138.
Babich V, Meli A, Knipe L, Dempster JE, Skehel P, Hannah MJ, Carter
T. Selective release of molecules from Weibel Palade bodies during a
lingering kiss. Blood. 2008;111:5282–5290.
Downloaded from http://atvb.ahajournals.org/ by guest on June 17, 2017
Angiopoietin-2 Exocytosis Is Stimulated by Sphingosine-1-Phosphate in Human Blood and
Lymphatic Endothelial Cells
Cholsoon Jang, Young Jun Koh, Nam Kyu Lim, Hyun Jung Kang, Duk Hoon Kim, Sung
Kwang Park, Gyun Min Lee, Choon Ju Jeon and Gou Young Koh
Arterioscler Thromb Vasc Biol. 2009;29:401-407; originally published online December 26,
2008;
doi: 10.1161/ATVBAHA.108.172676
Arteriosclerosis, Thrombosis, and Vascular Biology is published by the American Heart Association, 7272
Greenville Avenue, Dallas, TX 75231
Copyright © 2008 American Heart Association, Inc. All rights reserved.
Print ISSN: 1079-5642. Online ISSN: 1524-4636
The online version of this article, along with updated information and services, is located on the
World Wide Web at:
http://atvb.ahajournals.org/content/29/3/401
Data Supplement (unedited) at:
http://atvb.ahajournals.org/content/suppl/2009/02/20/ATVBAHA.108.172676.DC1
Permissions: Requests for permissions to reproduce figures, tables, or portions of articles originally published
in Arteriosclerosis, Thrombosis, and Vascular Biology can be obtained via RightsLink, a service of the
Copyright Clearance Center, not the Editorial Office. Once the online version of the published article for
which permission is being requested is located, click Request Permissions in the middle column of the Web
page under Services. Further information about this process is available in the Permissions and Rights
Question and Answer document.
Reprints: Information about reprints can be found online at:
http://www.lww.com/reprints
Subscriptions: Information about subscribing to Arteriosclerosis, Thrombosis, and Vascular Biology is online
at:
http://atvb.ahajournals.org//subscriptions/
ATVB/2008/172676-R3
Supplement Material
Angiopoietin-2 Exocytosis Is Stimulated by Sphingosine-1-Phosphate
in Human Blood and Lymphatic Endothelial Cells
Cholsoon Jang1,2, Young Jun Koh1, Nam Kyu Lim2, Hyun Jung Kang2, Duk Hoon Kim3,
Sung Kwang Park3, Gyun Min Lee1, Choon Ju Jeon2, Gou Young Koh1
1
National Research Laboratory of Vascular Biology and Department of Biological
Sciences, KAIST. Daejeon, 305-701, Republic of Korea
2
Biotherapeutic Division, GenExel-Sein Inc., Daejeon, 305-701, Republic of Korea
3
Department of Internal Medicine, Chonbuk National University Medical School,
Jeonju, 561-712, Republic of Korea
Address Correspondence to:
Gou Young Koh
National Research Laboratory for Vascular Biology
Department of Biological Sciences, KAIST
373-1, Guseong-dong, Daejeon, 305-701
Republic of Korea
Phone: 82-42-350-2638
Fax: 82-42-350-8160
E-mail:[email protected]
ATVB/2008/172676-R3
Supplemental Materials and Methods
Materials
Flag-tagged recombinant full-length Ang2 and a deletion construct containing only
fibrinogen like domain (FLD) of Ang2 were generated as previously described.1
Recombinant human VEGF-A and TNF-α proteins were purchased from eBioscience
(San Diego, CA). Dulbecco’s Modified Eagle Medium (DMEM) and calcium-free
DMEM were purchased from GIBCO (Grand Island, NY). Endothelium basal medium2 (EBM-2) with supplements was purchased from Clonetics (Walkersville, MD). S1P,
dihydro-S1P (dhS1P), and D-erythro-N,N-dimethylsphingosine (DMS) were purchased
from Biomol (Plymouth Meeting, PA). S1P was dissolved in methanol, and
subsequently diluted into distilled water containing fatty acid-free bovine serum
albumin (BSA) according to the manufacturer’s instructions. Pertussis toxin (PTX),
U73122, 1,2-bis(2-aminophenoxy)ethane-N,N,N',N'-tetrapotassium salt (BAPTA), 1,2bis(2-aminophenoxy)ethane-N,N,N',N'-tetraacetic acid acetoxymethyl ester (BAPTAAM), phorbol 12-myristate 13-acetate (PMA), PD98059, SB203580, LY294002,
forskolin, anti-Flag M2 antibody, anti-tubulin antibody, and 4',6-diamidino-2phenylindole (DAPI) were purchased from Sigma-Aldrich (St. Louis, MO).
Generation of Ang2 antibodies
Generation of mouse monoclonal antibodies was performed as described previously.2
Briefly, purified recombinant human Ang2 protein was used as an antigen to immunize
BALB/c mice. After four subcutaneous and one intravenous injection of Ang2 every 2
weeks, splenocytes were collected and immortalized by fusion with mouse myeloma
cells (F0; ATCC CRL-1645). The cells were grown in 96-well plates in hypoxanthine
ATVB/2008/172676-R3
aminopterin thymidine medium, and the culture supernatant was screened by ELISA
using purified Ang2 protein as a coating antigen. Two mouse monoclonal antibodies
against human Ang2 (clone No. 4A1A3 and 4B33A11) were purified from the culture
supernatant of hybridomas grown in serum-free medium by affinity chromatography on
a protein G-Sepharose column (Amersham Pharmacia Biotech, Piscataway, NJ).
Measurement of S1P concentration
HUVEC and HLEC were grown to a full confluence in 60-mm dishes (~1 x 106
cells/dish) in EBM-2 containing serum, and then were subjected to serum starvation in
serum-free EBM-2 for 12 hr before being stimulated with TNF-α (100 ng/mL) in 0.6
mL of serum-free EBM-2 containing only 0.1% fatty acid-free BSA (as a carrier for
secreted S1P). Then the media was collected and subjected to S1P measurement.
To measure S1P in the media, we used the S1P competitive ELISA assay kit (Echelon
Biosciences, San Diego, CA) according to the manufacturer’s protocol. According to
our experience, the lowest detection limit of the kit is ~20 nmol/L in 300 µL of sample.
According to the data sheet, specifically, the antibody in the kit has no cross-reactivity
with 13 related-lipids, including ceramide, ceramide-1-phosphate, lysophosphatidic acid,
and dihydrosphingosine.
Ex vivo-incubation of human umbilical cords
After perfusion with PBS, the middle portions of umbilical cords were cut at 1 cm with
a sterile surgical knife. In some cases, venous and arterial endothelial cells were wiped
off with a sterile cotton swab. The cord slices were transferred to 6-well plates and
Ang2 secretion was measured by ELISA before and after treatment. The cord slices
ATVB/2008/172676-R3
were cryosectioned for immunofluorescence staining.
RT-PCR analyses
RNA was extracted using the Total RNA Isolation System (Promega, Madison, WI),
and each cDNA was made using the Reverse Transcription System (Promega).
Quantitative real-time PCR was performed with the SYBR Premix Ex TaqTM (Takara,
Japan) using the iCycler iQ5 Real-time PCR system (Bio-Rad, Hercules, CA). PCR
reactions were performed with appropriate primers (Supplemental Table I) and cycles.
SiRNA transfection
SiRNAs targeted to human EDG1 or EDG3 were constructed as previously described.3
The siRNA duplexes and a negative control random scrambled siRNA were synthesized
by Bioneer Inc. (Daejeon, Korea). The transfection of siRNA duplex into HUVEC and
HLEC was performed using an X-tremeGENE siRNA transfection agent according to
the manufacturer’s instructions (Roche, Herts, UK). Briefly, after incubation of 1 μg
siRNA duplex with 5 μL transfection agent in 100 μL serum-free EBM-2 media for 20
min, the mixture was directly added to the cells grown on 6-well plates with 50%
confluence (final volume ~2.2 mL of serum-free EBM-2 media). The media was
changed after 4 hr-incubation and the cells were further incubated for 8 hr, and RNA
was purified to perform RT-PCR analyses. The measurements of Ang2 secretion were
performed at 72 hr after transfection with siRNA duplex.
Matrigel plug assay
The growth factor-reduced Matrigel (100 µl; BD Biosciences, San Jose, CA) containing
ATVB/2008/172676-R3
S1P (0.5 nmol), Ang2 (40 ng), or both S1P and Ang2 was subcutaneously implanted
into adult male C57BL/6 mice, and then the extents of angiogenesis were compared at
14 days after the implantation. Blood vessels were immunostained with anti-PECAM-1
antibody (diluted 1:1000; Reliatech, Germany). The measurements of blood vessel
densities in the Matrigel were made from immunostained tissue sections by
photographic analysis using ImageJ software (http://rsb.info.nih.gov/ij) after converting
the images into 8-bit gray scale. The measurements were made at a screen magnification
of 100X, each 0.81 mm2 in area at 0.9 mm inside the boundaries of the Matrigel, and 3
mice were used per group. To exclude background fluorescence, only pixels over a
certain level (>50 intensity value) were taken.
Statistics
Values presented are means ± standard deviation (SD). Significant differences between
means were determined by Student’s t-test or analysis of variance followed by the
Student-Newman-Keuls test. Statistical significance was set at P<0.05.
ATVB/2008/172676-R3
Supplemental References
1. Kim KT, Choi HH, Steinmetz MO, Maco B, Kammerer RA, Ahn SY, Kim HZ, Lee
GM, Koh GY. Oligomerization and multimerization are critical for angiopoietin-1 to
bind and phosphorylate Tie2. J Biol Chem. 2005;280:20126-20131.
2. Lim NK, Kim JH, Oh MS, Lee S, Kim SY, Kim KS, Kang HJ, Hong HJ, Inn KS. An
anthrax lethal factor-neutralizing monoclonal antibody protects rats before and after
challenge with anthrax toxin. Infect Immun. 2005;73:6547-6551.
3. Li X, Stankovic M, Bonder CS, Hahn CN, Parsons M, Pitson SM, Xia P, Proia RL,
Vadas MA, Gamble JR. Basal and angiopoietin-1-mediated endothelial permeability
is regulated by sphingosine kinase-1. Blood. 2008;111:3489-3497.
ATVB/2008/172676-R3
Supplemental Figure Legends
Supplemental Figure I. Ang2 is specifically expressed in the vein endothelial cells
of human umbilical cords. The perivascular regions (upper) and the surrounding
muscle regions (lower) of the ex vivo-incubated human umbilical cord sections were coimmunostained for smooth muscle actin (SMA, green) and Ang2 (red), and
counterstained with DAPI. Scale bars, 10 µm.
Supplemental Figure II. HLEC stores more intracellular Ang2 than HUVEC. (A)
ECs were treated with control buffer (-) or 1 μmol/L S1P (+) for 20 min, and Ang2
secretion was measured by ELISA (n=4). *, P<0.05 vs control. #, P<0.05 vs HUVEC.
(B) ECs were treated with control buffer (-) or 1 μmol/L S1P (+) for 20 min, and the
amounts of intracellular Ang2 were measured by immunoblotting (IB). Tubulin is used
as control. Three independent experiments show similar results.
Supplemental Figure III. S1P does not affect Ang2 mRNA synthesis in ECs. RTPCR analyses of Ang2 in ECs after treatment with S1P (1 µmol/L) or VEGF-A (10
ng/mL) for the indicated times. GAPDH is used as control. Three independent
experiments show similar results.
Supplemental Figure V. S1P induces Ang2 exocytosis via PTX-sensitive receptors,
PLC and calcium signaling pathway. (A-E) Ang2 secretion was measured by ELISA
(n=4). (A and C), HUVEC pretreated with 10 μmol/L DMS (A) or 200 ng/mL PTX (C)
was treated with S1P (S, 1 μmol/L) or TNF-α (T, 100 ng/mL) for 20 min. *, P<0.05 vs
ATVB/2008/172676-R3
control. #, P<0.05 vs S1P or TNF-α. (B) HUVEC was treated with S1P or dhS1P for 20
min. (D) HUVEC pretreated with U73122 was treated with S1P for 20 min. (E)
HUVEC pretreated with DMEM containing BAPTA (B) or BAPTA-AM (A), or Ca2+free DMEM (C) was treated with S1P for 20 min. *, P<0.05 vs control. #, P<0.05 vs
S1P.
Supplemental Figure V. ECs express much higher levels of EDG1 than EDG3.
Quantitative real-time RT-PCR analyses of EDG1 and EDG3 in ECs. GAPDH was used
to normalize the expression levels of EDG1 and EDG3. Three independent experiments
show similar results. *, P<0.05 vs EDG3.
Supplemental Figure VI. Combined treatment of S1P and Ang2 potently induces
angiogenesis in the Matrigel plug assay. Matrigel (100 µl) containing control buffer
(C), Ang2 (A, 40 ng), S1P (S, 0.5 nmol), or both S1P and Ang2 (S+A) was
subcutaneously implanted into adult male C57BL/6 mice, and then the densities of
PECAM-1-positive blood vessels were measured at 14 days after the implantation. The
dotted lines indicate the boundaries of Matrigel. *, P<0.05 vs C. #, P<0.05 vs S1P only.
Scale bars, 200 µm.
ATVB/2008/172676-R3
Supplemental Table I
EDG1
EDG3
EDG5
EDG6
EDG8
Angiopoietin-2
GAPDH
Forward
5’-caaggcccaccgcagctc-3’
Reverse
5’-gagggagatgacccagcagg-3’
Forward
5’-ctccgagggcagcacgc-3’
Reverse
5’-ggttggccaccttacggctg-3’
Forward
5’-gtgctcattgcggtggccc-3’
Reverse
5’-ggagtggacgggacaggc-3’
Forward
5’-ggagaacttgctggtgctggc-3’
Reverse
5’-ccccgcaggtactcctgg-3’
Forward
5’-gagcgaggtcatcgtcctgc-3’
Reverse
5’-cagcgctggcaggagcc-3’
Forward
5’-gccagagatggacaactgccg-3’
Reverse
5’-gtctccatgagatcatgttgctgc-3’
Forward
5’-ggctgcttttaactctggtaaagtgg-3’
Reverse
5’-ccacagtttcccggagggg-3’
Supplemental Figure I
SMA
vein
muscle
Ang2
DAPI
Merged
Supplemental Figure II
HUVEC
#
*
6
*
4
2
+
(kDa) HUVEC
-
+
HLEC
70
50
S1P (1 µmol/L)
+
-
+
IB:
Ang2
IB:
Tubulin
intracellular
Ang2 (fold)
8
0
S1P (1 µmol/L)
B
HLEC
extracellular
A
Supplemental Figure III
HUVEC
S1P 1 µmol/L
0
1
3
VEGF-A 10 ng/mL
12
0
1
3
12 (hr)
Ang2
GAPDH
HLEC
S1P 1 µmol/L
0
Ang2
GAPDH
1
3
12
VEGF-A 10 ng/mL
0
1
3
12 (hr)
Supplemental Figure IV
#
*
200
100
0
S or T - - S S
- + - +
DMS
(10µmol/L)
D
Ang2 (pg/ml)
400
S1P
dhS1P
300
200
100
0
- - T T
- + - +
C
* *
300
200
10 100 1000
E
*
300
*
200
#
#
0
S or T - - S S
- + - +
PTX
(200ng/mL)
HUVEC
* *
0
**
300
#
*#
200
100
0
-
-
0.1
1
S1P
10
10
*#
*
100
400
100
U73122
(μmol/L)
1.0
S1P (nM)
HUVEC
400
HUVEC
Ang2 (pg/ml)
**
300
C
HUVEC
400
*
Ang2 (pg/ml)
400
Ang2 (pg/ml)
B
HUVEC
Ang2 (pg/ml)
A
- B A C
- B A C
Control
S1P
- - T T
- + - +
Fold to EDG3/GAPDH
Supplemental Figure V
HUVEC
25
20
*
HLEC
*
15
10
5
0
G
ED
1
G1 DG3
G3
D
D
E
E
E
Supplemental Figure VI
C
Ang2
PECAM-1
S1P
S1P+Ang2
Blood vessel density (%)
PECAM-1
#
*
50
40
*
30
20
*
10
0
C
S
A
S+A