Differential Actions of PAR2 and PAR1 in Stimulating
Human Endothelial Cell Exocytosis and Permeability
The Role of Rho-GTPases
Scott W. Klarenbach,* Adam Chipiuk,* Randy C. Nelson, Morley D. Hollenberg, Allan G. Murray
Downloaded from http://circres.ahajournals.org/ by guest on June 17, 2017
Abstract—Endothelial cell proteinase activated receptors (PARs) belong to a family of heterotrimeric G protein– coupled
receptors that are implicated in leukocyte accumulation and potentiation of reperfusion injury. We characterized the
effect and the signal transduction pathways recruited after stimulation of endothelial PAR2. We used von Willebrand
Factor (vWF) release and monolayer permeability to peroxidase to report Weibel-Palade body (WPB) exocytosis and
pore formation, respectively. Human umbilical vein endothelial cells (HUVECs) were stimulated with the selective
PAR2 agonist peptide SLIGRL-NH2 or PAR1 agonist peptide TFLLR-NH2. PAR2 stimulation resulted in WPB
exocytosis like PAR1 stimulation but, unlike PAR1, failed to increase monolayer permeability. BAPTA-AM inhibited
PAR2-induced exocytosis, indicating a PAR2 calcium– dependent signal in ECs. Moreover, PAR2-like PAR1-stimulated
exocytosis requires actin cytoskeleton remodeling, because vWF release is inhibited if the cells were pretreated with
Jasplakinolide. Rho-GTPase activity is required for PAR-stimulated exocytosis, because inactivation of this family of
actin-regulatory proteins with Clostridium difficile toxin B blocked exocytosis. Expression of dominant-negative mutant
Cdc4217N inhibited exocytosis whereas neither dominant-negative Rac17N expression nor C3 exotoxin treatment affected
vWF release. PAR2 stimulated RhoA-GTP weakly compared with the PAR1 agonist. We conclude that both PAR2 and
PAR1 elicit WP body exocytosis in a calcium and Cdc42 GTPase-dependent manner. In contrast, the differential effect
of PAR1 versus PAR2 activation to increase monolayer permeability correlates with weak RhoA activation by the PAR2
agonist. (Circ Res. 2003;92:272-278.)
Key Words: vascular endothelium 䡲 reperfusion injury 䡲 thrombin receptors 䡲 Rho-GTP-binding proteins
R
eperfusion of tissue after revascularization of an arterial
thrombosis or preservation of a solid-organ allograft is
accompanied by microvascular endothelial cell injury, local
activation of the coagulation cascade, and diminished endothelial anticoagulant properties.1–3 In addition, platelet and
leukocyte adhesion to the endothelium is thought to contribute to inhomogeneous perfusion and persistent ischemic
injury of the parenchymal cells of the tissue.4 – 6 Activation of
specific receptors on the microvascular endothelial cells by
the serine proteinases of the coagulation cascade appears to
be an important mechanistic link between tissue injury and
the early recruitment of leukocytes.2
The proteinase activated receptors (PARs) are a novel
family of G protein– coupled receptors that require proteolytic cleavage at the amino terminus to stimulate signaling
(see review7). Cleavage allows the newly exposed tethered
ligand to initiate an intramolecular interaction necessary for
signal transduction, which can be mimicked by peptide
analogues of the newly exposed N-terminus. The PAR family
members, PAR1– 4, are expressed constitutively on endothelial
cells, and PAR2 and PAR4 are further upregulated on inflamed
endothelium.8,9 PAR1, PAR3, and PAR4 are substrates for
thrombin, whereas PAR2 can be stimulated by the activity of
tryptase, a product of mast cell degranulation, and coagulation Factors VIIa and Xa.10,11
Stimulation of endothelial cells with thrombin or peptide
analogues of the PAR1-tethered ligand (eg, TRAP or
SFLLRN-NH2), results in reorganization of the cytoskeleton
and discharge of endothelial storage granules, or WeibelPalade bodies (WPBs). Exocytosis of WPBs plays a major
role in the recruitment of leukocytes to sites of inflammation.12 WP body exocytosis releases stored IL-8 and mobilizes the adhesion molecule P-selectin to the lumenal surface
of the endothelial cell.13,14 In addition, WPB exocytosis
promotes platelet adhesion through the release of von Willebrand’s factor (vWF).13 Thrombin stimulation elicits capillary leak in vivo and increased endothelial monolayer permeability in vitro.15,16 PAR stimulation, then, promotes
Original received February 14, 2002; resubmission received December 20, 2002; accepted January 10, 2003.
From the Department of Medicine (S.W.K., A.C., R.C.N., A.G.M.), University of Alberta, Edmonton, Alberta; and the Departments of Pharmacology
& Therapeutics and Medicine (M.D.H.), University of Calgary, Calgary, Alberta, Canada.
*Both authors contributed equally to this study.
Correspondence to Allan G. Murray, Dept of Medicine, Rm 260F HMRC, University of Alberta, Edmonton, AB, Canada T6G 2S2. E-mail
[email protected]
© 2003 American Heart Association, Inc.
Circulation Research is available at http://www.circresaha.org
DOI: 10.1161/01.RES.0000057386.15390.A3
272
Klarenbach et al
leukocyte adhesion to vascular endothelium and links inflammation and coagulation in a variety of pathological settings.
Stimulation of the PAR2 is also proinflammatory in vivo.
Injection of selective peptide analogues of the PAR2 receptor
tethered ligand is sufficient to initiate inflammation in vivo.17
Moreover, impaired inflammatory responses are observed in
the PAR2-deficient mouse.18 In vitro, trypsin, like thrombin or
TRAP, treatment elicits WPB exocytosis,19 but fails to induce
monolayer permeability.20 However, the proteinases may act
on surface proteins other than PARs,21 and the PAR1 agonist
peptides modeled on the PAR1 tethered ligand activate both
PAR1 and PAR2.22 Hence, characterization of the effects of
selective endothelial PAR2 stimulation, and the signaling
pathways recruited by PAR2 agonists to stimulate the endothelial cell remain to be explored.
Materials and Methods
Downloaded from http://circres.ahajournals.org/ by guest on June 17, 2017
Reagents
PAR agonist peptides (AP) TFFLR-NH2, and SLIGRL-NH2 were
obtained from the peptide synthesis facility of the Health Sciences
Center of the University of Calgary or from the Alberta Peptide
Institute and were ⬎95% pure by HPLC. Jasplakinolide was obtained from Molecular Probes. ELISA reagents for vWF quantitation, were obtained from Cedarlane Laboratories, and a vWF
standard through Precision Biologic. Ionomycin and C3 exotoxin
were obtained from Calbiochem. The rho-kinase inhibitor Y-27632
was a gift of Welfide Corporation (Osaka, Japan). The Clostridium
difficile toxin B (TcdB) was obtained from List Biochemicals. The
rhotekin Rho binding domain (RBD)-GST fusion protein cDNA was
a generous gift of Dr M. Schwartz (Scripps Research Institute, San
Diego, Calif). Anti-RhoA mAb 26C4, and phospho-specific myosin
light chain kinase pAb, were obtained from Santa Cruz Biotechnology. Peroxidase-conjugated secondary pAb were from Jackson
Immunoresearch. M199, FBS, and Hanks balanced salt solution were
from Invitrogen. Cell dissociation solution, cytochalasin D, H-7, and
all other reagents were from Sigma.
Endothelial Cell Culture
Human umbilical vein endothelial cells (HUVECs) were isolated
from several umbilical cords, pooled, and cultured in complete media
(M199 with 20% FBS, penicillin, streptomycin, and glutamine).23
Cells were serially passed in complete media supplemented with
ECGS. HUVECs were cultured on a gelatin matrix and used in the
experiments at passage 2 to 5. The passaged endothelial cells were
harvested by washing twice with Hanks’, then incubated with a
nonenzymatic Cell Dissociation Solution (Sigma) until the ECs were
seen to lift off the plate. For use in an experiment, the harvested cells
were then replated onto fibronectin (Fn)-coated C6 wells and
incubated at 37°C in complete medium for 24 to 48 hours until
confluent.
Where indicated, Cytochalasin D (1 ␮mol/L), jasplakinolide (45
nmol/L), or DMSO vehicle were diluted in supplemented M199 and
added to the existing media. After 1 hour of incubation at 37°C, the
media was removed and discarded, and fresh complete media with
the agonist or vehicle diluted to the desired concentration was
applied to the HUVECs. To inhibit Rho-GTPases, HUVECs were
pretreated with 50 ␮g/mL C3 exotoxin for 24 hours and a mobility
shift of endothelial Rho on SDS-PAGE was confirmed as described
previously.24 To inhibit rho-kinase, endothelial cells were pretreated
with H-7 (0.5 ␮mol/L) or Y-27632 (15 ␮mol/L) for 120 minutes
before stimulation. To stimulate WP body exocytosis, the following
agonists were used: phorbal 12-myristate 13-acetate (PMA) at 100
nmol/L, PAR1 AP (TFLLR-NH2) at 30 ␮mol/L, PAR2 AP (SLIGRLNH2) at 30 ␮mol/L, and Ionomycin at 1 ␮mol/L, except where
otherwise specified.
Endothelial PAR2 Signals to Cdc42
273
Chemical inhibitors were used at concentrations that maintained
HUVEC viability ⬎85% of mock-treated controls as assessed by the
XTT assay of mitochondrial activity.25
ELISA Assay of vWF
Supernatant from mock- or agonist-stimulated HUVECs was removed into separate tubes and kept on ice until analysis. The vWF
concentration was determined by ELISA. Calibration was performed
with a vWF standard (Human reference plasma, Precision Biologics). Regulated vWF release was calculated by subtracting the vWF
concentration in unstimulated wells from the stimulated wells. The
effect of the inhibitors was determined by comparing the mean vWF
release to its control group. The mean difference of the treatment
among several experiments, as noted in the figure legends, was
calculated and tested for statistical significance (P⬍0.05) by
ANOVA using SPSS (SPSS).
Endothelial Monolayer Permeability
HUVECs at passage 3 were plated onto fibronectin-coated polycarbonate membranes (5-um pore size Transwell, Costar) at 6⫻104
cells/cm2 in M199/20% FBS. The monolayers were cultured for 5
days with media changes every 2 days and were not disturbed for 24
hours before the experiment. Horse radish peroxidase (HRP, MW
44kDa, Sigma) was added to the upper chamber at a final concentration of 1 ␮g/ml, then the specific peptide agonist was added at the
concentration indicated in the figure legends. Where indicated, the
monolayers were pretreated as described in the figure legend.
Medium was harvested from the lower chamber after 20 minutes,
and the HRP activity was determined colorimetrically by absorbance
at 490 nm of the o-phenyl diamine reaction product. The mean HRP
concentration from triplicate experimental wells was compared with
the HRP concentration in wells treated with ionomycin 10 ␮mol/L
according to the formula: (experimental⫺media control)/
(ionomyin⫺media control)⫻100⫽% maximum permeability.
The difference of the means between treatment and controls
among several experiments was calculated and tested for statistical
significance (P⬍0.05) by ANOVA using SPSS.
Affinity Precipitation of Rho-GTP
The RBD-GST fusion protein was prepared and isolated using
glutathione beads (Amersham, Baie d’Urfe, PQ), then used to
affinity precipitate GTP-Rho as described previously.26 Rho was
detected with anti-Rho mAb 26C4 then the blots were developed
using chemiluminescence (Supersignal, Pierce). Images were acquired from film using a CCD camera and the density of the bands
were quantitated using the Fluor-S Max software package (BioRad).
Transient Transfection of Endothelial Cells
Dominant-negative myc-tagged mutant Rac117N and Cdc4217N cDNA
were generously provided by Dr Gary Bokoch (Scripps) then
subcloned into pCSD7 (a gift of John Elliott, University of Alberta,
Edmonton, Canada) for transfection of HUVECs. The cDNAs were
sequenced to confirm the presence of the mutation before use.
Briefly, HUVECs were plated at confluence on Fn matrix in C-6
plates in growth medium 1 day before transfection. The cells were
washed with OptiMEM (Life Technologies), then 0.2 ␮g DNA/well
was introduced into the cells using the Effectene kit (Qiagen)
according to the instructions of the manufacturer. After 3 to 4 hours,
the wells were washed and the media was replaced with M199/10%
FBS and ECGS. Pilot experiments indicated that expression of the
myc-tagged mutant GTPase was optimal at approximately 24 hours
after transfection; therefore, the cells were used in the experiments as
indicated at this time. Endothelial cell expression of the myc-tagged
protein was confirmed in each experiment.
Results
Weibel-Palade Body Exocytosis Is Stimulated by
Activating Peptides Specific for PAR2
First, we sought to examine the effect of selective PAR2
activation compared with PAR1 activation on endothelial cell
274
Circulation Research
February 21, 2003
Downloaded from http://circres.ahajournals.org/ by guest on June 17, 2017
Figure 2. Protease-activated receptor-stimulated monolayer
permeability. Permeability of confluent endothelial monolayers to
HRP after stimulation with PAR1 (TFLLR-NH2, 50 ␮mol/L) or
PAR2 (SLIGRL-NH2, 50 ␮mol/L) was determined as described in
Materials and Methods. Agonist-stimulated monolayer permeability is shown relative to monolayers stimulated with ionomycin (10 ␮mol/L). Data represent the mean⫾SEM of 3 independent experiments.
Figure 1. Selective protease-activated receptor-2 agonist peptide stimulates Weibel-Palade body exocytosis. Confluent
monolayers of HUVECs were stimulated with the selective (A)
PAR1 (TFLLR-NH2) or (B) PAR2 agonist peptide (SLIGRL-NH2) or
PMA (100 nmol/L). Supernatants were collected after 60 minutes and assayed for vWF by ELISA. Regulated vWF release
was calculated as described in Materials and Methods and
expressed as a fraction of vWF release stimulated by PMA.
Mean⫾SEM of at least 4 independent experiments is shown.
exocytosis. Cells were stimulated with PAR1 (TFLLR-NH2),
or PAR2 (SLIGRL-NH2) agonist peptide, then the media was
assayed for vWF release. The amount of agonist peptidestimulated vWF release was compared with vWF release in
parallel wells maximally stimulated with PMA. The response
to PMA stimulation among different HUVEC cell lines was
variable. Hence, to facilitate comparison among independent
experiments, the effect of PAR agonist peptide was normalized to PMA-induced vWF release. Concentration-effect
curves were generated for both PAR1 and PAR2 agonist
peptides (Figure 1). The maximum release of vWF occurred
at a concentration range of 40 to 60 ␮mol/L for either peptide.
In this concentration range (ie, 50 ␮mol/L), TFLLR-NH2
selectively activates PAR1 and SLIGRL-NH2 selectively activates PAR2.22 No further increase of WP body exocytosis
occurred with higher concentrations (up to 150 ␮mol/L, data
not shown). After adjusting for constitutive release, PAR1 or
PAR2 agonist peptide stimulation at 50 ␮mol/L led to vWF
release of 1.1⫾0.1 (mean⫾SEM) U/mL and 1.1⫾0.1 U/mL
of vWF, respectively. These values corresponded to
21%⫾2% (mean⫾SEM) of PMA-induced vWF release. Further experiments used the agonist peptide at a concentration
(30 ␮mol/L) that approximates the EC50 for the response
elicited by the peptides. These experiments confirm that
either PAR2 or PAR1 can independently stimulate endothelial
exocytosis.
PAR2 Activation, Unlike PAR1, Fails to Increase
Monolayer Permeability
We next examined if selective PAR2 stimulation altered
endothelial monolayer integrity. Like the PAR1 agonist, the
PAR2 agonist peptide induced actin stress fiber formation in
confluent HUVECs (data not shown). Notwithstanding, unlike PAR1, PAR2 stimulation failed to increase monolayer
permeability (Figure 2). Ionomycin stimulation elicited a
marked increase in transfer of HRP across the monolayer
(18⫾6 versus 68⫾7 ng/cm2 per hour; mean⫾SEM) and
served as a comparator to which the effects of the PAR
agonists were normalized. These data indicate that although
PAR2, like PAR1 stimulation, is able to induce bundled f-actin
stress fiber formation, unlike PAR1, PAR2 fails to induce pore
formation.
Effect of Cytochalasin D and Jasplakinolide on
PAR2 Agonist Peptide-Stimulated Exocytosis
Both endothelial cell exocytosis and monolayer permeability
changes are thought to require rearrangement of the f-actin
cytoskeleton, hence we sought to determine if PAR2 stimulated f-actin reorganization. Cytochalasin D and jasplakinolide have been previously shown to depolymerize or stabilize
the f-actin cytoskeleton, respectively.27,28 To determine if the
endothelial cytoskeleton was reorganized in PAR2-stimulated
regulated exocytosis, HUVECs were pretreated with either
agent, then treated or not with peptide agonist. Neither agent
changed constitutive vWF release. However, cytochalasin D
pretreatment led to a modest augmentation of both PAR2- and
PAR1-regulated WP body exocytosis (Figure 3). These data
indicate that an intact cytoskeleton is not required for PARstimulated WP body exocytosis.
Jasplakinolide was used to test the requirements of each
agonist for remodeling of the intact cytoskeleton to elicit
vWF release (Figure 3). HUVECs pretreated with jasplakinolide then stimulated with PAR1 agonist peptide, TFLLR-NH2,
released 54⫾15% of vWF compared with mock-pretreated
cells (P⫽0.03, n⫽6 experiments). Similarly, jasplakinolide
inhibited vWF release on stimulation with PAR2 agonist
peptide, SLIGRL-NH2, to 43⫾16% of mock-pretreatment
release (P⫽0.05, n⫽5 experiments). These data indicate that
PAR2-mediated exocytosis requires active remodeling of the
endothelial f-actin cytoskeleton.
Klarenbach et al
Downloaded from http://circres.ahajournals.org/ by guest on June 17, 2017
Figure 3. Cytoskeletal remodeling is involved in Weibel-Palade
body exocytosis. Confluent monolayers of HUVECs were pretreated with either cytochalasin D (1 ␮mol/L; filled bars) or jasplakinolide (45 nmol/L; lined bars) or carrier for 1 hour, then
stimulated with PAR1 (TFLLR-NH2, 30 ␮mol/L), or PAR2 (SLIGRLNH2, 30 ␮mol/L). Supernatants were assayed for vWF by ELISA
as in Figure 1. Each pretreatment group was compared with the
carrier-pretreatment control. Mean⫾SEM of at least 4 independent experiments is shown. *P⬍0.05 by ANOVA.
PAR2 Exocytosis Requires Intracellular Calcium
The initiation of a calcium flux has previously been identified
to play a critical role in thrombin-stimulated WP body
exocytosis and monolayer permeability.29,30 To determine if
PAR2-stimulated exocytosis was similarly dependent on
calcium-dependent signaling, HUVEC monolayers were pretreated with BAPTA-AM, then stimulated with the PAR
agonist peptides. As noted in Figure 4A, PAR2-stimulated
vWF release was markedly inhibited under these conditions.
This indicates that WP body exocytosis, stimulated by either
PAR1 or PAR2, is dependent on elevated intracellular calcium
concentration.
Our data suggest that both cytoskeletal reorganization and
elevated intracellular calcium are required for efficient PARstimulated exocytosis. However, high concentrations of jasplakinolide may inhibit sustained intracellular calcium levels
after stimulation via G protein– coupled receptors.31 Because
a calcium flux is sufficient to induce WP body exocytosis,29
we sought to determine if calcium ionophore-induced vWF
release was affected by jasplakinolide. First, HUVECs were
stimulated with ionomycin to define a concentration-effect
curve (data not shown). Ionomycin (1 ␮mol/L) induced vWF
release comparable to that of the PAR-activating peptides at
30 to 50 ␮mol/L. Jasplakinolide pretreatment diminished
ionomycin-induced vWF release to 52⫾13% (Figure 4B;
P⫽0.02, n⫽5 experiments) of mock-pretreated controls. This
observation indicates that the effect of jasplakinolide lies
downstream of the generation of a calcium flux and suggests
that PAR2 stimulation results in a calcium-dependent reorganization of the endothelial cytoskeleton.
Involvement of Rho-GTPase Signaling After
PAR2 Stimulation
The Rho family of small GTP-binding proteins regulate the
f-actin cytoskeleton and have been implicated in thrombinstimulated increased endothelial monolayer permeability,16,32
and in exocytosis in a variety of other cell types.33,34 Rho
Endothelial PAR2 Signals to Cdc42
275
Figure 4. PAR2-stimulated exocytosis is dependent on intracellular calcium concentration. A, Confluent monolayers of
HUVECs were pretreated with BAPTA-AM (20 ␮mol/L for 30
min) where indicated, then stimulated with PAR2 agonist
(SLIGRL-NH2, 30 ␮mol/L). Supernatants were harvested and
vWF concentrations were determined by ELISA. vWF release
stimulated by the PAR2 agonist is shown relative to that elicited
by PMA (100 nmol/L). Mean⫾SEM of 3 independent experiments is shown. B, HUVECs were pretreated with cytochalasin
D (1 ␮mol/L), jasplakinolide (45 nmol/L), or carrier as in Figure 2,
then stimulated with ionomycin 1 ␮mol/L for 1 hour. Supernatants were assayed for vWF concentration by ELISA. Each pretreatment group was compared with the carrier-pretreatment
control. Mean⫾SEM of 4 independent experiments is shown.
*P⬍0.05 by ANOVA.
kinase, a Rho-GTP– dependent serine/threonine kinase, acts
to phosphorylate and inactivate myosin light chain phosphatase. Myosin light chain phosphorylation is also implicated in
both thrombin-stimulated permeability and exocytosis.16,29
We therefore sought to determine if a Rho-GTPasedependent signaling pathway was recruited after PAR2
stimulation.
We pretreated confluent HUVEC monolayers with TcdB
toxin to selectively inhibit the small GTPases, Rho, Rac, and
Cdc42. PAR2- and PAR1-mediated exocytosis were markedly
inhibited (Figure 5A). This strongly suggests that RhoGTPases serve a critically important function to effect PARstimulated exocytosis in endothelial cells.
To determine if RhoA activity elicited by PAR2 stimulation
was involved in exocytosis, we tested the effect of specific
inhibitors of the RhoA signaling pathway. First, we inhibited
Rho activity with C3 exotoxin.24 C3 pretreatment did not
inhibit vWF release using the PAR2 agonist peptide to
stimulate endothelial cell exocytosis (Figure 5A). We similarly observed that inhibition of Rho-kinase with either
Y-27632 or H-7 (data not shown) failed to inhibit, and tended
to augment, PAR-stimulated vWF release. In contrast, PAR1
stimulation of monolayer permeability was markedly attenuated by inhibition of Rho kinase with Y-27632 (Figure 2).
These data indicate that RhoA-GTPase signaling is not
required for PAR2-stimulated exocytosis and suggest that
other Rho-GTPase family members account for the TcdBmediated inhibition.
To identify the Rho-GTPase family member involved in
PAR-stimulated exocytosis, we expressed dominant-
276
Circulation Research
February 21, 2003
Downloaded from http://circres.ahajournals.org/ by guest on June 17, 2017
Figure 5. Rho-family GTPases are involved in PAR-stimulated
exocytosis. A, Confluent monolayers of HUVECs were pretreated with C3 exoenzyme (50 ␮g/mL) for 24 hours, Y27632 (15
␮mol/L) for 2 hours, or TcdB (0.5 ␮g/mL for 4 hours) to inactivate RhoA, B, and C; Rho kinase; or Rho, Rac, and Cdc42,
respectively. Cells were stimulated with TFLLR-NH2 (30 ␮mol/L)
or SLIGRL-NH2 (30 ␮mol/L) agonist peptides, and supernatants
were analyzed for vWF as above. Data reflect the mean⫾SEM
effect of the inhibitor vs media alone control pretreatment in at
least 3 independent experiments. B, HUVECs were transiently
transfected with dominant-negative mutant Rac17N or Cdc4217N
as described in Materials and Methods, then stimulated with
PAR1 (TFLLR-NH2, 30 ␮mol/L) or PAR2 (SLIGRL-NH2, 30 ␮mol/L)
agonist peptides 24 hours after transfection. vWF concentration
in culture supernatants was quantitated by ELISA as in Figure 1.
Data reflect the mean⫾SEM of 3 independent experiments.
*P⬍0.05 by ANOVA.
Figure 6. Rho activation follows PAR2 stimulation. A, Confluent
monolayers of HUVECs were stimulated with PAR1 (TFLLR-NH2,
50 ␮mol/L) or PAR2 (SLIGRL-NH2, 50 ␮mol/L) agonist for the
indicated time. Rho-GTP was affinity precipitated with GSTRBD, then Rho was resolved by SDS-PAGE and immunoblotted
as described in Materials and Methods. B, Densitometric analysis of mean⫾SEM RhoA-GTP blots from 3 independent
experiments.
MLC phosphorylation by either agonist peptide was inhibited
by pretreatment of the HUVECs with Y27632 (data not
shown). These observations indicate that the failure of PAR2
stimulation to elicit monolayer pore formation correlates with
weak signaling through RhoA.
Discussion
The major observations of this article are as follows: first,
PAR2-stimulated exocytosis is dependent on both remodeling
negative mutant Rac117N or Cdc4217N in the endothelial
cells. Both PAR2- and PAR1-stimulated exocytosis was
inhibited in HUVEC-expressing Cdc4217N but not Rac117N
(Figure 5B). Taken together these observations indicate
that stimulation of either endothelial PAR receptor recruits
a signal that is dependent on Cdc42 activity to elicit
exocytosis.
PAR2 Stimulation Elicits Weak RhoA Activity
PAR1-stimulated monolayer permeability depends on Rho
kinase activity (Figure 2), hence we directly assessed RhoA
activation after PAR2 stimulation to determine if the failure of
PAR2 to elicit pore formation correlated with RhoA-GTP. As
shown in Figure 6, RhoA-GTP was robustly activated by
PAR1 stimulation, but only weakly by PAR2 stimulation at
receptor-selective concentrations of the agonist peptides. To
determine if the weak RhoA-GTP stimulation by PAR2 could
account for the failure of pore formation through failing to
increase cell contractility, we examined myosin light chain
phosphorylation. Both PAR2 and PAR1 agonist peptides
induced phosphorylation of myosin light chain, with a much
greater effect of PAR1 (Figure 7). Interestingly, stimulation of
Figure 7. PAR2 stimulates endothelial myosin light chain phosphorylation. Confluent HUVEC monolayers were stimulated with
PAR1 (TFLLR-NH2, 50 ␮mol/L) or PAR2 (SLIGRL-NH2, 50 ␮mol/L)
for the indicated time. Cell lysates were resolved by SDS-PAGE,
and immunoblotted for phosphorylated myosin light chain. Data
are representative of 2 independent experiments. B, Densitometric analysis of mean P-MLC/total MLC blot of each condition
from the experiment in A.
Klarenbach et al
Downloaded from http://circres.ahajournals.org/ by guest on June 17, 2017
the actin cytoskeleton and eliciting an intracellular calcium
flux. PAR2-stimulated exocytosis does not require either
myosin light chain phosphorylation, or RhoA-GTPasedependent Rho kinase activity. However, both PAR2 and
PAR1-stimulated exocytosis is sensitive to TcdB inhibition of
the Rho family of GTPases. Cdc42 signaling appears to be the
principal Rho-GTPase required for endothelial WeibelPalade body exocytosis. Interestingly, unlike PAR1, PAR2
failed to elicit increased endothelial monolayer permeability,
despite generating stress fibers and myosin light chain phosphorylation to enable cell contractility. Activity of RhoAGTP and myosin light chain phosphorylation are required for
PAR1-mediated monolayer permeability.
We studied selective agonists of both the PAR2 and PAR1
G protein– coupled receptors and observed that each stimulated similar amounts of vWF release from the endothelial
cell storage pool. The use of a selective activator of PAR2
indicates that the PAR2 system can have as great an impact on
Weibel-Palade body exocytosis and hence lumenal display of
P-selectin as the thrombin-regulated PAR1 system. Interestingly, like PAR1, PAR2-mediated exocytosis required f-actin
reorganization, and involves a calcium-dependent rather than
a cAMP-dependent second messenger. A similar requirement
for f-actin reorganization was observed using ionomycin to
mimic the calcium flux initiated by the PAR receptors,
indicating that the requirement for f-actin remodeling likely
occurred downstream of the PAR-stimulated calcium signal.
These observations are in agreement with previous work that
inferred activation of the PAR2 receptor using trypsin to
stimulate Weibel-Palade body exocytosis19 and that detected
a calcium flux in endothelial cells after trypsin stimulation.35
Coupling of PAR2 to endothelial cytoskeletal-associated
proteins has not been examined previously. However, regulated exocytosis by mast cells or pancreatic acinar cells can be
facilitated by dissolution of the cortical actin barrier and is
thought to be a prerequisite for vesicle fusion to the plasma
membrane.36,37 The endothelial PAR2 stimulated calcium
signal is required for exocytosis and may regulate disassembly of cortical f-actin,38 for example by augmenting the
activity of actin-severing proteins.39 The sensitivity of PAR2
stimulated exocytosis to the actin-stabilizing effect of jasplakinolide is consistent with this model.
Regulation of f-actin reorganization occurs under the
influence of the Rho family of monomeric GTP-binding
proteins.40 For example, Rac1 is linked to PAR signaling
pathways that regulate stress fiber induction and cortical actin
remodeling.32 In addition, GTP-␥S–induced exocytosis in
permeabilized mast cells is inhibited by C3 exotoxin, which
specifically ADP-ribosylates and inactivates Rho.33 Rac and
Cdc42 activity are also implicated in regulated exocytosis in
basophilic leukemic cells and pancreatic ␤-cells.34,41,42 Our
observation that pretreatment of the HUVECs with TcdB, to
inactivate Rho, Rac, and Cdc42, markedly inhibits PARstimulated vWF release, clearly implicating Rho-GTPases in
both the PAR2 and PAR1 signaling pathways that direct
Weibel-Palade body exocytosis.
Rho-GTP rapidly increases after stimulation of HUVECs
with either the PAR2 or PAR1 agonist. To determine the
participation of Rho directly in PAR-stimulated WP body
Endothelial PAR2 Signals to Cdc42
277
exocytosis, we used the C3 exotoxin to selectively modify
and abrogate interaction of Rho with downstream effector
molecules. We found no consistent inhibitory effect of C3
exotoxin when used at concentrations known to markedly
ADP-ribosylate endothelial Rho proteins.24 Moreover, inhibition of Rho kinase with Y27632 also fails to block
PAR-stimulated exocytosis. This result indicates that neither
RhoA nor Rho kinase activity is required for WP body
exocytosis after stimulation of either PAR, and confirm
recent data examining the role of RhoA in thrombinstimulated exocytosis.43 In contrast, we observed that expression of the Cdc4217N mutant significantly decreased PARstimulated Weibel-Palade body exocytosis. Because the
transfection efficiency in these experiments was approximately 30% to 50%, diminished Cdc42 activity likely accounts for much of the inhibition observed with TcdB
treatment. Taken together these experiments implicate an
obligatory role for Cdc42 activity in the cascade of events
leading to endothelial exocytosis downstream of either proteinase activated receptor. However, we cannot exclude that
Rho-GTPase family members function partially redundantly
in this system.
Interestingly, despite the ability to mobilize intracellular
calcium and elicit f-actin remodeling and myosin light chain
phosphorylation, PAR2 signaling, in contrast with PAR1, fails
to increase endothelial monolayer permeability. Our observations confirm and extend those of Compton et al20 who
observed no endothelial monolayer permeability change after
trypsin stimulation. We observe that PAR1-mediated monolayer permeability required Rho-GTP and Rho kinase activity
for this event and demonstrate that PAR2 recruited Rho-GTP,
albeit only weakly. These results indicate that unlike PAR1,
PAR2 fails to sufficiently activate the RhoA, and possibly
other, signal transduction pathways necessary for monolayer
pore formation.
In summary, we describe a role for cytoskeleton remodeling and the Rho family of GTP-binding proteins in both PAR1
and PAR2 signaling to endothelial Weibel-Palade body exocytosis. The signal involves Cdc42 GTP-binding protein
acting independently of Rho kinase or Rac1. Our data are
consistent with a model of Cdc42 GTP-binding proteindependent cortical cytoskeletal remodeling. However, in
comparison with PAR1, PAR2 elicits less RhoA-GTP and
MLC phosphorylation and does not elicit increased monolayer permeability.
Acknowledgments
A.G.M. is the recipient of a Clinician Investigator award from the
Alberta Heritage Foundation for Medical Research. This work was
supported in part by an operating grant from the Kidney Foundation of
Canada and Aventis/Canadian Blood Services Research Fund to
A.G.M.
References
1. Golino P, Ragni M, Cirillo P, Avvedimento VE, Feliciello A, Esposito N,
Scognamiglio A, Trimarco B, Iaccarino G, Condorelli M, Chiariello M,
Ambrosio G. Effects of tissue factor induced by oxygen free radicals on
coronary flow during reperfusion. Nat Med. 1996;2:35– 40.
2. Erlich JH, Boyle EM, Labriola J, Kovacich JC, Santucci RA, Fearns C,
Morgan EN, Yun W, Luther T, Kojikawa O, Martin TR, Pohlman TH,
Verrier ED, Mackman N. Inhibition of the tissue factor-thrombin pathway
278
3.
4.
5.
6.
7.
8.
9.
Downloaded from http://circres.ahajournals.org/ by guest on June 17, 2017
10.
11.
12.
13.
14.
15.
16.
17.
18.
19.
20.
21.
Circulation Research
February 21, 2003
limits infarct size after myocardial ischemia-reperfusion injury by
reducing inflammation. Am J Pathol. 2000;157:1849 –1862.
Ejiri S, Eguchi Y, Kishida A, Ishigami F, Kurumi Y, Tani T, Kodama M.
Cellular distribution of thrombomodulin as an early marker for warm
ischemic liver injury in porcine liver transplantation: protective effect of
prostaglandin I2 analogue and tauroursodeoxycholic acid. Transplantation. 2001;71:721–726.
Palazzo AJ, Jones SP, Girod WG, Anderson DC, Granger DN, Lefer DJ.
Myocardial ischemia-reperfusion injury in CD18- and ICAM-1– deficient
mice. Am J Physiol. 1998;275:H2300 –H2307.
Yadav SS, Howell DN, Steeber DA, Harland RC, Tedder TF, Clavien PA.
P-Selectin mediates reperfusion injury through neutrophil and platelet
sequestration in the warm ischemic mouse liver. Hepatology. 1999;29:
1494 –1502.
Singbartl K, Green SA, Ley K. Blocking P-selectin protects from ischemia/reperfusion-induced acute renal failure. FASEB J. 2000;14:48 –54.
MacFarlane SR, Seatter MJ, Kanke T, Hunter GD, Plevin R. Proteinaseactivated receptors. Pharmacol Rev. 2001;53:245–282.
Mirza H, Yatsula V, Bahou WF. The proteinase activated receptor-2
(PAR-2) mediates mitogenic responses in human vascular endothelial
cells. J Clin Invest. 1996;97:1705–1714.
Hamilton JR, Frauman AG, Cocks TM. Increased expression of proteaseactivated receptor-2 (PAR2) and PAR4 in human coronary artery by
inflammatory stimuli unveils endothelium-dependent relaxations to
PAR2 and PAR4 agonists. Circ Res. 2001;89:92–98.
Fox MT, Harriott P, Walker B, Stone SR. Identification of potential
activators of proteinase-activated receptor-2. FEBS Lett. 1997;417:
267–269.
Riewald M, Ruf W. Mechanistic coupling of protease signaling and
initiation of coagulation by tissue factor. Proc Natl Acad Sci U S A.
2001;98:7742–7747.
Robinson SD, Frenette PS, Rayburn H, Cummiskey M, Ullman-Cullere
MD, Wagner DD, Hynes RO. Multiple, targeted deficiencies in selectins
reveal a predominant role for P-selectin in leukocyte recruitment. Proc
Natl Acad Sci U S A. 1999;96:11452–11457.
Hattori R, Hamilton KK, Fugate RD, McEver RP, Sims PJ. Stimulated
secretion of endothelial von Willebrand factor is accompanied by rapid
redistribution to the cell surface of the intracellular granule membrane
protein GMP-140. J Biol Chem. 1989;264:7768 –7771.
Wolff B, Burns AR, Middleton J, Rot A. Endothelial cell “memory” of
inflammatory stimulation: human venular endothelial cells store interleukin 8 in Weibel-Palade bodies. J Exp Med. 1998;188:1757–1762.
Cirino G, Cicala C, Bucci MR, Sorrentino L, Maraganore JM, Stone SR.
Thrombin functions as an inflammatory mediator through activation of its
receptor. J Exp Med. 1996;183:821– 827.
van Nieuw Amerongen GP, van Delft S, Vermeer MA, Collard JG, van
Hinsbergh VW. Activation of RhoA by thrombin in endothelial hyperpermeability: role of Rho kinase and protein tyrosine kinases. Circ Res.
2000;87:335–340.
Vergnolle N, Hollenberg MD, Sharkey KA, Wallace JL. Characterization
of the inflammatory response to proteinase-activated receptor-2 (PAR2)activating peptides in the rat paw. Br J Pharmacol. 1999;127:1083–1090.
Lindner JR, Kahn ML, Coughlan SR, Sambrano GR, Schauble E,
Bernstein D, Foy D, Hafezi-Moghadam A, Ley K. Delayed onset of
inflammation in protease-activated receptor-2– deficient mice. J Immunol.
2000;165:6505– 6510.
Storck J, Kusters B, Vahland M, Morys-Wortmann C, Zimmermann ER.
Trypsin induced von Willebrand factor release from human endothelial
cells in mediated by PAR-2 activation. Thromb Res. 1996;84:463– 473.
Compton SJ, Cairns JA, Holgate ST, Walls AF. Human mast cell tryptase
stimulates the release of an IL-8-dependent neutrophil chemotactic
activity from human umbilical vein endothelial cells (HUVEC). Clin Exp
Immunol. 2000;121:31–36.
Lafleur MA, Hollenberg MD, Atkinson SJ, Knauper V, Murphy G,
Edwards DR. Activation of pro-(matrix metalloproteinase-2) (proMMP-2) by thrombin is membrane-type-MMP-dependent in human
umbilical vein endothelial cells and generates a distinct 63 kDa active
species. Biochem J. 2001;357:107–115.
22. Kawabata A, Saifeddine M, Al-Ani B, Leblond L, Hollenberg MD. Evaluation
of proteinase-activated receptor-1 (PAR1) agonists and antagonists using a
cultured cell receptor desensitization assay: activation of PAR2 by PAR1targeted ligands. J Pharmacol Exp Ther. 1999;288:358–370.
23. Gimbrone MA. Culture of vascular endothelium. Prog Hemost Thromb.
1976;3:1–28.
24. Essler M, Amano M, Kruse HJ, Kaibuchi K, Weber PC, Aepfelbacher M.
Thrombin inactivates myosin light chain phosphatase via Rho and its
target Rho kinase in human endothelial cells. J Biol Chem. 1998;273:
21867–21874.
25. Murray AG, Khodadoust MM, Pober JS, Bothwell AL. Porcine aortic
endothelial cells activate human T cells: direct presentation of MHC
antigens and costimulation by ligands for human CD2 and CD28.
Immunity. 1994;1:57– 63.
26. Ren XD, Kiosses WB, Schwartz MA. Regulation of the small GTPbinding protein Rho by cell adhesion and the cytoskeleton. EMBO J.
1999;18:578 – 85.
27. Copper JA. Effects of cytochalasin and phalloidin on actin. J Cell Biol.
1987;105:1473–1478.
28. Bubb MR, Senderowicz AM, Sausville EA, Duncan KL, Korn ED.
Jasplakinolide, a cytotoxic natural product, induces actin polymerization
and competitively inhibits the binding of phalloidin to F-actin. J Biol
Chem. 1994;269:14869 –14871.
29. Birch KA, Pober JS, Zavoico GB, Means AR, Ewenstein BM. Calcium/
calmodulin transduces thrombin-stimulated secretion. J Cell Biol. 1992;
118:1501–1510.
30. Draijer R, Atsma DE, Van der Laarse A, van Hinsbergh VW. cGMP and
nitric oxide modulate thrombin-induced endothelial permeability. Circ
Res. 1995;76:199 –208.
31. Patterson RL, van Rossum DB, Gill DL. Store-operated Ca2⫹ entry:
evidence for a secretion-like coupling model. Cell. 1999;98:487– 499.
32. Wojciak-Stothard B, Potempa S, Eichholtz T, Ridley A. Rho and Rac but not
Cdc42 regulate endothelial cell permeability. J Cell Sci. 2001;114:1343–1355.
33. Norman JC, Price LS, Ridley AJ, Koffer A. The small GTP-binding
proteins, Rac and Rho, regulate cytoskeletal organization and exocytosis
in mast cells by parallel pathways. Mol Biol Cell. 1996;7:1429 –1442.
34. Kowluru A, Li G, Rabaglia ME, Segu VB, Hofmann F, Aktories K, Metz
SA. Evidence for differential roles of the Rho subfamily of GTP-binding
proteins in glucose- and calcium-induced insulin secretion from pancreatic beta cells. Biochem Pharmacol. 1997;54:1097–1108.
35. Vergnolle N, Macnaughton WK, Al-Ani B, Saifeddine M, Wallace JL,
Hollenberg MD. Proteinase-activated receptor 2 (PAR2)-activating peptides: identification of a receptor distinct from PAR2 that regulates
intestinal transport. Proc Natl Acad Sci U S A. 1998;95:7766 –7771.
36. Nakata T, Hirokawa N. Organization of cortical cytoskeleton of cultured
chromaffin cells and involvement in secretion as revealed by quickfreeze, deep-etching, and double-label immunoelectron microscopy.
J Neurosci. 1992;12:2186 –2197.
37. Muallem S, Kwiatkowska K, Xu X, Yin HL. Actin filament disassembly
is a sufficient final trigger for exocytosis in nonexcitable cells. J Cell Biol.
1995;128:589 –598.
38. Sullivan R, Price LS, Koffer A. Rho controls cortical F-actin disassembly
in addition to, but independently of, secretion in mast cells. J Biol Chem.
1999;274:38140 –38146.
39. Trifaro J, Rose SD, Lejen T, Elzagallaai A. Two pathways control
chromaffin cell cortical F-actin dynamics during exocytosis. Biochimie.
2000;82:339 –352.
40. Hall A. Rho GTPases and the actin cytoskeleton. Cell. 1998;279:
509 –514.
41. Prepens U, Just I, von Eichel-Streiber C, Aktories K. Inhibition of Fc
epsilon-RI-mediated activation of rat basophilic leukemia cells by Clostridium difficile toxin B (monoglucosyltransferase). J Biol Chem. 1996;
271:7324 –7329.
42. Djouder N, Prepens U, Aktories K, Cavalie A. Inhibition of calcium
release-activated calcium current by Rac/Cdc42-inactivating clostridial
cytotoxins in RBL cells. J Biol Chem. 2000;275:18732–18738.
43. Vischer UM, Barth H, Wollheim CB. Regulated von Willebrand factor
secretion is associated with agonist-specific patterns of cytoskeletal
remodeling in cultured endothelial cells. Arterioscler Thromb Vasc Biol.
2000;20:883– 891.
Downloaded from http://circres.ahajournals.org/ by guest on June 17, 2017
Differential Actions of PAR2 and PAR1 in Stimulating Human Endothelial Cell Exocytosis
and Permeability: The Role of Rho-GTPases
Scott W. Klarenbach, Adam Chipiuk, Randy C. Nelson, Morley D. Hollenberg and Allan G.
Murray
Circ Res. 2003;92:272-278; originally published online January 23, 2003;
doi: 10.1161/01.RES.0000057386.15390.A3
Circulation Research is published by the American Heart Association, 7272 Greenville Avenue, Dallas, TX 75231
Copyright © 2003 American Heart Association, Inc. All rights reserved.
Print ISSN: 0009-7330. Online ISSN: 1524-4571
The online version of this article, along with updated information and services, is located on the
World Wide Web at:
http://circres.ahajournals.org/content/92/3/272
Permissions: Requests for permissions to reproduce figures, tables, or portions of articles originally published
in Circulation Research 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 Circulation Research is online at:
http://circres.ahajournals.org//subscriptions/