Download TNF-induced endothelial barrier disruption: beyond actin and Rho

1088
Review Article
TNF-induced endothelial barrier disruption: beyond actin and Rho
Beatriz Marcos-Ramiro*; Diego García-Weber*; Jaime Millán
Centro de Biología Molecular Severo Ochoa, CSIC-UAM, Madrid, Spain
Summary
The decrease of endothelial barrier function is central to the long-term
inflammatory response. A pathological alteration of the ability of endothelial cells to modulate the passage of cells and solutes across the
vessel underlies the development of inflammatory diseases such as
atherosclerosis and multiple sclerosis. The inflammatory cytokine tumour necrosis factor (TNF) mediates changes in the barrier properties
of the endothelium. TNF activates different Rho GTPases, increases
filamentous actin and remodels endothelial cell morphology. However, inhibition of actin-mediated remodelling is insufficient to prevent
endothelial barrier disruption in response to TNF, suggesting that additional molecular mechanisms are involved. Here we discuss, first, the
pivotal role of Rac-mediated generation of reactive oxygen species
(ROS) to regulate the integrity of endothelial cell-cell junctions and,
Correspondence to:
Jaime Millán
Centro de Biología Molecular Severo Ochoa
C/ Nicolás Cabrera 1, Universidad Autónoma de Madrid
Cantoblanco, 28049 Madrid, Spain
Tel.: +34 911964713, Fax: +34 911964420
E-mail: [email protected]
second, the ability of endothelial adhesion receptors such as ICAM-1,
VCAM-1 and PECAM-1, involved in leukocyte transendothelial migration, to control endothelial permeability to small molecules, often
through ROS generation. These adhesion receptors regulate endothelial barrier function in ways both dependent on and independent of
their engagement by immune cells, and orchestrate the crosstalk between leukocyte transendothelial migration and endothelial permeability during inflammation.
Keywords
Endothelial barrier function, permeability, adhesion, transendothelial
migration, Rho, Rac, Cdc42, ROCK, reactive oxygen species, Src, Fyn,
actin, VE-cadherin, ICAM-1, VCAM-1, JAM, PECAM-1, SHP phosphatases, metalloproteases
Received: April 1, 2014
Accepted after minor revision: June 16, 2014
Epub ahead of print: July 31, 2014
http://dx.doi.org/10.1160/TH14-04-0299
Thromb Haemost 2014; 112: 1088–1102
* Equal contribution.
Introduction
The inflammatory response is a tightly regulated process involved
in pathogen clearance and the repair of damaged tissue. It constitutes an alteration of tissue homeostasis in reaction to stress, which
is also detrimental to tissue function. An accurate spatiotemporal
restriction of the inflammatory response is therefore crucial for
preventing uncontrolled, systemic reactions that would be more
harmful than the initial pathological challenges (1).
Blood and lymphatic vessels enable tissue compartmentalisation during the initial stages of the innate and adaptive immune
responses and provide spatial constraints required during inflammatory reactions. The basal extravasation of immune cells that
survey the body is locally increased in the blood vessels surrounding an inflammatory focus. This increase occurs through stimulation of the endothelium by inflammatory factors secreted by activated, tissue-resident immune cells. Proinflammatory stimulation increases both endothelial permeability to small molecules,
such as chemokines and cytokines, and the expression of receptors
that mediate the endothelial adherence to immune cells. As a result, the leukocyte transendothelial migration towards the in-
flamed areas transiently increases. Excessive inflammatory activation of endothelial cells leads to aberrant endothelial barrier disruption, inducing local swelling and edema formation. Chronic,
unresolved endothelial inflammation that prolongs hyper permeability underlies the origin of diseases such as atherosclerosis and
multiple sclerosis, which adds further interest to the study of the
mechanisms regulating endothelial barriers.
Among the plethora of inflammatory mediators that contribute
to endothelial barrier disruption, the tumour necrosis factor
(TNF) is central to the initiation and termination of long-term inflammatory responses (2). TNF signals to the actin cytoskeleton by
promoting a notable morphological remodelling of endothelial
cells, which has been extensively investigated but whose inhibition
is not enough to prevent the progressive loss of endothelial barrier
properties in response to this cytokine. Here we review additional
molecular pathways, emerging as critical mediators of the transient disruption of endothelial barriers in response to TNF. They
include the disruption of cell-cell junctions by reactive oxygen
species (ROS) and the signalling by adhesion receptors involved in
leukocyte-endothelial interactions. This signalling occurs partly
through ROS generation and reveals a remarkable molecular
© Schattauer 2014
Thrombosis and Haemostasis 112.6/2014
Downloaded from www.thrombosis-online.com on 2017-06-17 | IP: 88.99.165.207
For personal or educational use only. No other uses without permission. All rights reserved.
1089
Marcos-Ramiro, García-Weber et al. Adhesion receptors and ROS in TNF-induced endothelial permeability
crosstalk between the increase of permeability to small molecules
and that of leukocyte transendothelial migration (TEM). This
should be taken into account in the design of therapeutic strategies
for controlling endothelial barrier disruption in pathological inflammation.
The cytokine TNF in the inflammatory
response
TNF, also known as TNFα, is a cytokine with pleiotropic effects in
many cell types that is expressed as a type II transmembrane protein mainly by macrophages, dendritic cells and T-cells (3, 4).
Upon arrival at the plasma membrane, some of the TNF molecules
are shed by the metalloproteinase TNFα-converting enzyme
(TACE), also known as a disintegrin and metalloproteinase-17
(ADAM-17), and released as a soluble homotrimer of 17 KD (5, 6).
TNF binds to two type I transmembrane receptors, TNFR1 and
TNFR2, which are homologous in their extracellular domain but
quite divergent with respect to the cytoplasmic segments that recruit different signalling machinery. TNFR1 is ubiquitously expressed whereas TNFR2 is expressed only in immune cells and endothelium (7). Both receptors lack intrinsic catalytic activity and
require assembly of supramolecular complexes of cytosolic proteins to transduce signals. A detailed description of the signalosome recruited by both receptors upon TNF interaction and the
subsequently elicited signalling pathways, has recently been reviewed elsewhere (7-10). Briefly, the activation of inhibitor of
kappa-B (IκB) kinases (IKKs) and their association with TNFR
signalling complexes are amongst the most important events of the
TNF signalling cascade. IKKs are the kinases that phosphorylate
the IκBα, leading to its ubiquitination and degradation. The absence of IκBα exposes the nuclear translocation sequence of the
p65 subunit of the nuclear factor (NF)-κB transcription factor,
which enters the nucleus and regulates a plethora of genes associated with long-term inflammatory responses (11). TNFR signalling complexes also activate the stress activated protein (SAP) kinases, p38 MAP kinase and the Jun N-terminal kinase (JNK),
which phosphorylate c-Jun. Phosphorylated c-Jun is a component
of the heterotrimeric transcription factor AP-1, which becomes activated and regulates another set of genes (12). In addition, TNFR1
engagement can induce apoptosis by activating Caspase8, and necrosis by assembling a multiprotein complex called necrosome
(13). The variation in composition of the TNFR signalosomes determines the relative induction of each pathway and the final outcome of TNF-mediated signalling. As a consequence, depending
on the cell type and the physiological context, TNF activates different subsets of target genes that promote diverse cellular effects
such as survival, apoptosis, necrosis, migration, proliferation, barrier disruption and cell–to-cell adhesion.
TNF signals to the actin cytoskeleton and
remodels cell-cell junctions through Rho
GTPases in endothelial cells
In addition to the stimulation of these canonical pathways that
lead to gene expression changes, TNF signals to the actin cytoskeleton by activating Rho GTPases. Rho GTPases cycle between a
GTP-loaded active form and a GDP-bound inactive conformation
(14). Rho GTPases are activated by guanine nucleotide exchange
factors (GEFs) and inactivated by GTPase-activating proteins
(GAPs) and GDP dissociation inhibitors (GDIs). GDIs form an inactive complex with the Rho GTPase in the cytosol. The local
combination of these regulators spatially confines and confers specificity on Rho-mediated signalling. Due to the central role that
this family plays in regulating cellular barriers, the TNF-mediated
signalling to Rho GTPases has been of great interest in the study of
the endothelial inflammatory response. So far, it has been established that TNF activates RhoA, Rac and Cdc42, the three best
known of the 22 members of the Rho family and the master regulators of filamentous actin (15, 16, 17-22). The molecular connection between TNFR engagement and Rho GTPase activation is
still poorly understood. Various GEFs have been reported to be activated downstream of TNF, mostly in non-endothelial cell types,
although the underlying molecular interactions still need to be explored. In addition, taking into account the similarities and functional cooperation found between Rho family proteins in different
physiological scenarios, the involvement of other members of the
family is quite probable and should be investigated in detail in the
future.
TNF induces actin stress fibres through RhoA and
ROCK
One of the main effects of TNF in endothelial cells is their morphological remodelling through the reorganisation of actin cytoskeleton, which has been observed in vivo (23) and in vitro (20, 24,
25). In response to this cytokine, the endothelium turns from a
pavement-like monolayer, containing a peripheral actin belt close
to stable cell-cell junctions, to a monolayer formed by elongated
cells enriched in actin stress fibres (20, 24, 25). The formation of
stress fibres is a paradigmatic outcome of RhoA-mediated signalling to Rho kinase (ROCK) (26, 27). ROCK phosphorylates and
inhibits myosin light chain phosphatase, which promotes myosin
light chain (MLC) phosphorylation, which causes the assembly of
these filaments and actomyosin contractility (28). ROCK can directly phosphorylate MLC, at least in in vitro assays (29). Since actomyosin contraction is a major mechanism of cell barrier disruption, it was hypothesised early on that this signalling pathway
was pivotal to TNF-mediated endothelial morphological reorganisation and long-term barrier alteration. However, experimental
evidence has confirmed this only for the case of endothelial morphological remodelling. As we shall see below, the central role of
Rho-ROCK in long-term barrier function decrease in response to
TNF is unclear.
Thrombosis and Haemostasis 112.6/2014
© Schattauer 2014
Downloaded from www.thrombosis-online.com on 2017-06-17 | IP: 88.99.165.207
For personal or educational use only. No other uses without permission. All rights reserved.
Marcos-Ramiro, García-Weber et al. Adhesion receptors and ROS in TNF-induced endothelial permeability
Stress fibres are connected to adherens junctions in a
ROCK-dependent manner in TNF-stimulated
endothelial cells
TNF transiently activates RhoA and its main effector ROCK in
different cell types, including macrovascular and microvascular
endothelial cells (17, 22, 30-32). Following gene silencing with
siRNA, Mong and Wang (33) demonstrated in microvascular endothelial cells that ROCKI and ROCKII isoforms have an additive
role in increasing MLC phosphorylation in response to TNF. However, ROCKII is the major Rho kinase activating actomyosin during long-term cytokine stimulation. Short-term pharmacological
inhibition of ROCK or Rho has confirmed the critical role of
ROCK in incrementing stress fibres in response to TNF (17, 30,
34).
Endothelial adherens junctions (AJs) are multiprotein junctional complexes formed by the vascular endothelial (VE)-cadherin. VE-cadherin complexes establish cell-cell homotypic interactions that are fundamental for maintaining endothelial barrier integrity (35). The VE-cadherin cytoplasmic tail nucleates a protein
complex formed by p120-catenin, β-catenin and α-catenin, the
latter linking the complex to actin filaments. Actin binding is
required to stabilise cell-cell junctions, but also regulates AJ dynamics in response to soluble mediators or transendothelial cellular migration. The most detailed analysis of AJ dynamics comes
from in vitro studies. In resting microvascular and macrovascular
endothelial cells, AJs exhibit a heterogeneous morphology that includes areas apparently devoid of F-actin (22). In contrast stress
fibres induced by TNF remodel AJs, which become discontinuous
and aligned with these actin microfilaments (▶ Figure 1). As expected, the formation of discontinuous AJs requires ROCK activity. Moreover, discontinuous AJs often connect stress fibre ends
coming from adjacent endothelial cells (34). Intriguingly, the stress
fibre increase occurring in response to TNF is not accompanied by
a comparable rise in the number of focal adhesions (FAs), the adhesive protein complexes that stabilise and attach these filaments
to the plasma membrane and the extracellular matrix. Indeed, abrogating AJs through VE-cadherin knockdown does not significantly diminish the number of stress fibres, but is sufficient to increase the number of FAs and alter cell morphology in confluent,
TNF-stimulated endothelial cells (34). These data suggest that discontinuous AJ stabilise tension-bearing, force-generating stress
fibres, independently of FAs, in TNF-stimulated endothelial
monolayers (34, 36). This connection participates in the actin-mediated morphological remodelling of endothelial cells upon TNF
exposure.
Rac regulates TNF-mediated actin and junction
remodelling
Rac is involved in endothelial barrier stabilisation and mediates
the barrier-protective effects of various soluble mediators. Rac
GTPase controls F-actin and often plays a role antagonistic to that
of RhoA. RhoA and Rac activities are inversely correlated
throughout TNF stimulation. Rac activity increases during the
first minutes of TNF activation (21, 37) but decreases during longterm stimulation (32, 38). Moreover, TNF-mediated Rac activation probably depends on the stimulatory conditions and the
vascular cells analysed, since other authors have convincingly reported sustained activation of this GTPase up to 20 hours poststimulation (39). The F-actin increase caused by TNF and the subsequent redistribution of cell-cell junctions is altered by the expression of a constitutive active mutant of Rac1, but also by the expression of a dominant-negative form (20). In addition, Rac1 gene
silencing significantly diminishes TNF-mediated cell elongation,
one of the main morphological traits of the endothelial response to
this cytokine (40). Collectively, these results suggest a role for this
GTPase as a mediator of endothelial cell remodelling upon TNF
beyond that of being a mere antagonist of RhoA function.
Cdc42 activity contributes to F-actin induction upon
TNF stimulation
The contribution of Cdc42 to TNF-mediated signalling has been
less extensively investigated than that of RhoA and Rac. Cdc42 is
activated a few minutes after TNF exposure (21, 39) and mediates
the prosurvival signalling (41) and the cytoskeletal remodelling induced by this cytokine (19, 20). Cdc42 is considered to be a stabilising mediator of endothelial barriers and to play an important
role in maintaining endothelial AJs (42), although its role in cellcell junction remodelling in the context of TNF stimulation remains to be established.
TNF-mediated increase of endothelial
permeability: beyond actin and Rho
Experiments performed in macrovascular endothelial cells with
specific siRNA or inhibitors reveal a cooperative role of ROCKI
and ROCKII in the homeostatic maintenance of the endothelial
barrier, which is disrupted upon double ROCK knockdown in the
absence of inflammation (43). There is also evidence from the use
of inhibitors, expression of dominant-negative mutants and siRNA
that demonstrates the essential role of the Rho-ROCK pathway
during acute contraction induced by mediators such as thrombin
and histamine (43-45). However, the function of Rho and ROCKmediated actin remodelling in TNF-induced barrier disruption is
more controversial. As mentioned above, the endothelial cytoskeletal and junctional reorganisation induced by TNF is mediated by
Rho, ROCK and Rac. However, inhibiting such reorganisation is
not always sufficient to prevent endothelial barrier disruption.
ROCK inhibition reduces the permeability increase within the first
minutes of TNF-stimulation in microvascular endothelial cells
(33). Several reports have clearly shown that inhibition of actin
polymerisation through Rho-ROCK and of myosin is not enough
to impair microvascular and macrovascular barrier disruption
after several hours of TNF exposure (17, 22, 32, 33, 46). This is in
agreement with the fact that stress fibres generated by long-term
TNF stimulation have little myosin activity, suggesting that these
filaments became stabilised but generate little tension in the ab-
© Schattauer 2014
Thrombosis and Haemostasis 112.6/2014
Downloaded from www.thrombosis-online.com on 2017-06-17 | IP: 88.99.165.207
For personal or educational use only. No other uses without permission. All rights reserved.
1090
1091
Marcos-Ramiro, García-Weber et al. Adhesion receptors and ROS in TNF-induced endothelial permeability
sence of further stimulation (17). Similarly, single Rac1 knockdown promotes a clear reduction of endothelial cell elongation in
response to TNF, but has a minor effect on permeability increase
and leukocyte transendothelial migration (40). Hence, the lack of
correlation between actin-mediated remodelling and alteration of
barrier function clearly suggests that, in addition to actomyosinmediated contractility, there must be mechanisms that promote
long-term barrier disruption in response to this cytokine. The research focus on endothelial permeability has been directed towards the actin-independent roles of Rac and the expression or relocalisation of adhesion receptors involved in leukocyte transmi-
gration. The latter emerge as central players in the molecular
mechanisms mediating endothelial permeability increase upon
TNF stimulation.
RAC and ROS in TNF-induced endothelial
permeability
In contrast to the minor effect that Rac1 knockdown has on endothelial permeability, expression of the Rac dominant negative
mutant Rac-T17N or incubation with the Rac inhibitor
Figure 1: TNF induces morphological and
junction remodelling in endothelial cells.
Unstimulated endothelial cells display a heterogeneous distribution of cell-cell junctions. Top
left images (1, 2) represent AJs distributed in reticular structures at overlapping areas between
cells, which contain a PECAM-1-positive Lateral
Border Recycling Compartment (LBRC) depleted
of F-actin. Discontinuous white line marks the
border of overlapping cells. (3) TNF induces actin
stress fibres that connect with adherens junctions and disperses the LBRC from the cell
border by unknown mechanisms. (4) TNF increases the expression of adhesion receptors
such as ICAM-1 and VCAM-1, which align with
stress fibres upon engagement. These preformed
stress fibres may work as platforms for signalling to cell-cell junctions. In addition, clustered
receptors contribute to increase actomyosin-mediated tension and barrier disruption as well as
for guiding crawling leukocytes towards sites of
diapedesis. Bottom confocal and super resolution images show representative examples of
the structures represented in the top drawings
(1–4). (1, 2) Reticular adherens junctions, with
low F-actin content (arrows), containing
PECAM-1. STED, stimulated emission depletion
super resolution microscopy. (3) Discontinuous
adherens junctions connecting adjacent stress
fibres. (4) Clustering of ICAM-1 with specific
antibodies induces its alignment with actin
stress fibres, which can signal in the proximity of
cell-cell contacts. Bars, 10 μm.
Thrombosis and Haemostasis 112.6/2014
© Schattauer 2014
Downloaded from www.thrombosis-online.com on 2017-06-17 | IP: 88.99.165.207
For personal or educational use only. No other uses without permission. All rights reserved.
Marcos-Ramiro, García-Weber et al. Adhesion receptors and ROS in TNF-induced endothelial permeability
NSC-23766 completely abrogates the TNF-mediated decrease of
TEER in human lung microvascular endothelial cells (47, 48, 49).
Mutants and inhibitors, both alter the function of at least Rac1,
Rac2 and Rac3, the closely related and members of the Rac
GTPase subfamily. This suggests that whereas Rac1 significantly
contributes to TNF-induced endothelial actin and morphological
remodelling (40), it overlaps in regulating barrier function
through additional mechanisms with the other two members of
the subfamily, which are also expressed in endothelial cells. Upstream of Rac, the Rac GEFs Vav, Tiam-1, Trio and P-Rex, are activated in response to TNF (37, 39, 40). Vav activation regulates canonical signalling pathways mediated by this cytokine and has
only been reported in non-endothelial cell types (21). T cell lymphoma invasion and metastasis (Tiam)-1 associates with VE-cadherin upon TNF stimulation (40). In other stimulatory contexts,
Tiam-1 is essential, together with Vav-2, for Rac-mediated regulation of endothelial barrier function (50-53). Nonetheless, the
roles of Tiam-1 and Vav-2 in TNF-induced barrier modulation
still need to be addressed. Trio expression and association with active Rac are increased in response to TNF. Unlike Tiam-1, Trio
does not specifically regulate endothelial barrier function but instead has a role in activating Ets-2, a transcription factor that induces VCAM-1 and ICAM-1 expression (39). As detailed below,
VCAM-1 and ICAM-1 regulate endothelial permeability, so it is
likely that Trio regulates endothelial barrier properties via these receptors. Finally, phosphatidylinositol 3,4,5-trisphosphate-dependent Rac exchanger (P-Rex)-1 also plays a general role in TNF
stimulation regulating not only barrier function but also NFκB signalling and adhesion receptor expression. However, to date, there
has been no comparative study of the relative contribution of each
of these GEFs to TNF-induced endothelial barrier disturbance (37,
40).
Rac GTPases play an important role as a component of the active NADPH oxidase (NOX) complex, which is the main source of
ROS in endothelial cells. The endothelium expresses various NOX
complexes, namely those formed by NOX1, NOX2, NOX4 and
NOX5 (54). In HUVECs, the most widely used human endothelial
cell system, NOX2 and NOX4 are the most abundant complexes
(55). The phagocyte NOX2 complex is the best characterised endothelial NOX and is formed by the catalytic gp91phox (NOX2)
subunit associated with the regulatory p22phox. The cytosolic subunits p40phox, p47phox and p67phox, and Rac form the active complex
that transfers electrons from NADPH to oxygen to form superoxide (02-) (54, 56, 57). Rac also binds to and regulates the NOX1
complex, which contains different cytosolic subunits, but does not
regulate either NOX4, which is constitutively active (58), or NOX5,
which is regulated by calcium (59). The TNFR1 signalosome recruits the NOX1 complex, including active Rac, upon TNF interaction, which mediates sustained Jnk activation and necrosis in
non-endothelial cells (60). Alternatively, ROS can be directly activated by the TNF signalosome component TRAF4, which binds to
p47phox and is required for the full TNF-mediated ROS induction
in murine microvascular endothelial cells (61). In addition, TNF
increases the membrane localisation of NOX2 and Rac1 in HUVECs (55). In human colon adenocarcinoma cells, the mixed-lin-
eage kinase domain-like (MLKL) (62) is associated with the necrosome complex and is also required for ROS induction and late Jnk
activation in response to TNF. However, neither the role of MLKL
in ROS production, nor the possible contribution of Rac to
TRAF4- or MLKL-mediated ROS formation, have yet been addressed in endothelial cells.
In human endothelial cells, P-Rex knockdown, expression of
dominant-negative Rac and specific Rac inhibitors abrogate ROS
production in response to TNF (37). The expression of a constitutively active mutant of Rac is also sufficient to increase ROS production. One of the main effects of elevated ROS production is the
oxidation of several residues from phosphatases and kinases,
thereby controlling the main system of cellular signalling (63).
ROS oxidise various amino acids such as those with aromatic residues, large and hydrophobic, which often causes conformational
changes in the proteins. ROS also induce irreversible carbonylation and/or racemisation of various amino acids that direct proteins for degradation (63). However, the sulfur-containing amino
acids cysteine and methionine are the most prone to oxidation.
Importantly, their oxidation is reversible, which makes this modification important in signal transduction. In Src-family kinase
members (64, 65), elevated ROS induces two cysteine residues,
Cys245 and Cys487 in the case of Src kinase (66), to form intramolecular disulfide bridges upon oxidation, which lock the kinase
in an active conformation taken it to a super activation state. The
nucleophilic chain of cysteines can be oxidised in three states sulfenic (-SOH), sulfinic (-SO2H) and sulfonic (-SO3H). The sulfenic
acid state is able to form a disulfide bond with other cysteine residue in the vicinity. The cellular antioxidants glutathione or thioredoxin can reverse this state of cysteine oxidation, but further oxidative stress can lead to the formation of sulfinic or sulfonic states
that become irreversible (63). Hyperoxidation of Src kinases cause
enzyme inactivation and results in their final degradation (63). Indeed, oxidation of Src Cys277, which causes the formation of
homodimers and results in Src inhibition, has also been described
(67). ROS can also indirectly control Src kinases by oxidising
phosphatases such as SH2-containing protein tyrosine phosphatase-1 and - 2 (SHP-1 and SHP-2), upstream regulators of Src (64).
It is of note that the effect of TNF on ROS-mediated oxidation of
Src at molecular level has not been studied yet. Rac-induced ROS
regulate endothelial permeability by elevating the tyrosine phosphorylation of VE-cadherin (68), which prevents the binding of
β-catenin and p120-catenin and results in inhibition of endothelial
barrier function (37, 69, 70). Exposure to tyrosine kinase inhibitors blocks TNF-induced endothelial permeability, illustrating
the importance of tyrosine phosphorylation in this process (70,
71).
Src-family kinases can phosphorylate not only VE-cadherin,
but also β-catenin, p120-catenin and other junctional molecules
such as PECAM-1 (40, 71, 72, 37, 54, 63). Amongst the Src kinase
members, some authors propose that Fyn, but not Src or Yes, are
fundamental to TNF-mediated barrier disruption and phosphorylation of VE-cadherin, γ-catenin and p120-catenin, at least in
human lung microvascular endothelial cells (71). Alternatively,
ROS-regulated tyrosine kinases other than Src, such as Pyk2 (73),
© Schattauer 2014
Thrombosis and Haemostasis 112.6/2014
Downloaded from www.thrombosis-online.com on 2017-06-17 | IP: 88.99.165.207
For personal or educational use only. No other uses without permission. All rights reserved.
1092
1093
Marcos-Ramiro, García-Weber et al. Adhesion receptors and ROS in TNF-induced endothelial permeability
Figure 2: TNF induces endothelial barrier
disruption through Rac activation and ROS
generation. TNF activates Rac and generates
reactive oxygen species (ROS) through NOX
complexes. Rac and ROS activates different tyrosine kinases and phosphatases that contribute
to phosphorylate and destabilise adherens junctions. Rac also mediates the increase of adhesion receptor expression via Ets-2 and NF-κB.
also associate with and regulate the phosphorylation of VE-cadherin and AJ disruption in response to TNF (40). The closest
relative of Pyk2, FAK, is also activated by TNF in non-endothelial
cell types (74-76). FAK regulates endothelial barriers in response
to other soluble mediators such as thrombin, S1P or VEGF (77, 78,
79) making plausible its involvement in TNF-mediated increase in
permeability. In addition, FAK and Pyk2 autophosphorylation recruit Src kinases (80, 81), so both families of kinases may have a
cooperative effect on cell-cell junctions. Together, these data imply
that whereas an increase in Rho-ROCK activity and F-actin promotes morphological changes in the endothelium upon TNF exposure, the Rac-ROS-Src/Pyk2 kinase axis is responsible for the
loss of barrier function through the phosphorylation and destabilisation of cell-cell junction complexes (▶ Figure 2).
Control of endothelial permeability by
TNF-regulated adhesion receptors
The molecular mechanisms that regulate the passage of immune
cells through the endothelium have been thoroughly investigated
in the last 25 years, although the differences found between heterogeneous vascular beds and leukocyte subsets, as well as between different inflammatory challenges, are still not fully understood (82-84). In the absence of inflammation, leukocyte TEM is
tightly controlled to the extent that only a small subset of immune
cells are able to extravasate, acting as “immune guards” and helping to transmit information to the adaptive immune system. In
contrast, in inflammatory foci, TNF secretion activates a transcriptional program in the neighbouring endothelium that includes a dramatic increase of chemokine release and the expression of adhesion receptors. These receptors include the lectin
E-selectin (85) and the immunoglobulin superfamily members
ICAM-1 and VCAM-1, which are the ligands of β2 and β1 integrins, respectively (86-88). These expression changes mediate the
local increase in leukocyte adhesion to endothelial cells and the
subsequent leukocyte infiltration of the inflamed tissues. It is of
note that TNF not only increases the expression of some receptors,
but also changes the localisation of some others. For example, TNF
favours initial endothelial cell–leukocyte interactions, called tethering and rolling, not only by increasing E-selectin, but also by
inducing the translocation of P-selectin from the Weibel–Palade
bodies to the plasma membrane (85, 89). These initial contacts
and the interaction with surface chemokines mature leukocyte adhesions through changes in the activation state of the mentioned
integrins, which adhere firmly to VCAM-1 and ICAM-1. Arrested
immune cells then crawl on the endothelium and transmigrate between two endothelial cells (paracellular diapedesis) or through
transcellular channels away from cell-cell junctions (transcellular
diapedesis) (90). In the paracellular diapedesis, leukocytes encounter another subset of adhesion receptors, such as PECAM-1,
CD99 and junctional adhesion molecules (JAMs) that guide cells
across the endothelium. Again, TNF does not modulate the expression of these junctional receptors, but determines their localisation, at least in the cases of PECAM-1 (91, 22) and JAM-A (92).
In addition to their role in paracellular diapedesis, the adhesion receptors ICAM-1 and PECAM-1 help guide cells that follow transcellular diapedesis (93-95).
Thus, TNF changes expression and localisation of receptors
that participate in the different steps of the cascade of interactions
between immune cells and the endothelium. The rise in endothelial permeability to small molecules parallels leukocyte TEM during the inflammatory course. In recent years, evidence has accumulated that points to an additional role for adhesion receptors
in controlling endothelial permeability, a role that, it is important
Thrombosis and Haemostasis 112.6/2014
© Schattauer 2014
Downloaded from www.thrombosis-online.com on 2017-06-17 | IP: 88.99.165.207
For personal or educational use only. No other uses without permission. All rights reserved.
Marcos-Ramiro, García-Weber et al. Adhesion receptors and ROS in TNF-induced endothelial permeability
to note, is both dependent on and independent of their interactions with leukocyte counterreceptors.
Intercellular adhesion molecule-1 (ICAM-1) and
endothelial barrier regulation in response to TNF
TNF increases ICAM-1 expression more than 15-fold in human
umbilical vein endothelial cells (HUVECs) in vitro (96). In endothelial cells, TNF-1 induces ICAM-1 expression mainly via
TNFR1 (97, 98). The DNA region upstream the ICAM-1 translation start site, involved in the translational increase of ICAM-1
mediated by TNF and other cytokines, has been well mapped and
consist on variant NF-κB binding site flanked by other regulatory
sequences (99). NF-κB/Rel complexes can be composed of different subunits that in turn can bind other regulatory factors. The
transcriptional regulation of ICAM-1 in different cell types
through NF-κB have been extensively reviewed elsewhere (99,
100). In HUVECs, the RelA homodimer and NF-κB1/RelA heterodimer bind to the NF-κB binding site of ICAM-1 (101). Over
expression of RelA activated ICAM-1 promoter, whereas NF-κB1
overexpression reduces such transactivation (101). In human lung
microvascular endothelial cells, TNF also regulates ICAM-1 expression post-transcriptionally by controlling mRNA stability
through the p38 mitogen-activated protein kinase (MAPK)-activated protein kinase 2 (MK2) pathway (102, 103).
ICAM-1 is the counter receptor of leukocyte β2 integrins.
ICAM-1 is a type I transmembrane receptor of the immunoglobulin (Ig) superfamily that contains five extracellular Ig domains. In
TNF-stimulated endothelium, ICAM-1 localises in plasma membrane microvilli and cell-cell junctions (104-106). The ICAM-1 cytoplasmic tail interacts with proteins that connect the receptor to
subcortical filamentous actin. Artificial overexpression by retroviral transduction of ICAM-1 at levels comparable to those induced
by TNF is sufficient to increase endothelial permeability significantly. Interestingly, these levels of expression do not affect the
ratio of filamentous to globular actin and exert a moderate effect
on tight junctions, suggesting an actin-independent effect of the
receptor on permeability (107). Higher levels of ICAM-1 completely disrupt the endothelial barrier, increase F-actin staining
and fragment the linear staining of adherens and tight junctions.
In accordance, ICAM-1 silencing partially reduces the reorganisation of cell-cell junctions occurring in response to TNF (107). It
is of note that overexpression of a truncated form of ICAM-1 lacking the cytoplasmic segment is also sufficient to induce endothelial
permeability (107). This, together with the uncoupled effects of
ICAM-1 overexpression on F-actin remodelling and permeability,
suggests that the transmembrane and extracellular domains of the
ICAM-1 contribute significantly to endothelial permeability in the
absence of any receptor engagement (107).
ERM proteins are connectors of subcortical F-actin with cytoplasmic domains of membrane-associated proteins, including endothelial ICAM-1 (90, 108). The association of ERM proteins with
ICAM-1 determines its localisation on surface F-actin-rich microvilli (109). Interestingly, TNF-induced endothelial barrier disruption is mediated by the activation of ERM proteins and cell-cell
junctions (110). The contribution of ICAM-1 to this effect has not
been analysed to date. However, in the absence of TNF stimulation, knockdown of a raft-protein of unreported function,
known as myeloid-associated differentiation marker (MYADM),
induces a proinflammatory phenotype that features hallmarks of
TNF stimulation such as raised levels of endothelial permeability,
activation of ERM proteins and ICAM-1 expression (96). Barrier
disruption caused by MYADM depletion requires ICAM-1 expression, which in turn depends on ERM activation (96). Collectively, these results strongly suggest a role for MYADM as a
negative regulator of the inflammatory response and permeability
through the control of ICAM-1 expression. However, the possible
participation of MYADM and ICAM-1 in TNF-induced barrier
disruption has not been tested yet. TNF exposure is not sufficient
to reduce MYADM (96), and further studies are required to place
this novel protein within any known inflammatory signalosome.
The molecular bases underlying a ligand-independent role of
ICAM-1 on endothelial permeability have not been well characterised to date. In the different cellular context of the blood-testis
barrier, ICAM-1 coimmunoprecipitates and colocalises with
ZO-1, claudin, N-cadherin and β-catenin at cell-cell contacts
(111). In contrast to what was observed in dermal endothelial cells
(107), overexpression of full ICAM-1 reduced epithelial permeability, whereas a truncated form containing ICAM-1 extracellular
domain increased permeability and reduced the expression of proteins involved in the formation of cell-cell junctions. Although the
components of cell-cell junctions certainly differ between bloodtestis and endothelial barriers, a role for ICAM-1 as a constituent
stabiliser/destabiliser of junctional complexes in the endothelium
is a possibility worth investigating in the future.
ICAM-1-mediated endothelial barrier disruption upon
β2-integrin engagement
Based on their findings with the ICAM-KO mice, the laboratory of
Sarelius has proposed a thought-provoking alternative, in which
the increase in vascular permeability to solutes is mediated by
ICAM-1-dependent signalling upon leukocyte interaction (112,
113). Endothelial cell remodelling in response to TNF would thus
be secondary for barrier control. This increase in permeability
would depend almost exclusively on the transcriptional regulation
of adhesion receptors that bind leukocyte ligands and signal to increase overall vascular leakiness. Sumagin el al. (112, 113) showed
that antibody-mediated engagement of ICAM-1 is sufficient to increase vascular permeability in vivo, whereas blocking antibodies
against β2 integrins dramatically decreased permeability in venules and arterioles. Consistent with this, vascular permeability in
ICAM-1 KO and CD18 KO mice was also reduced with respect to
control mice (113). Preventing leukocyte rolling by blocking P-selectin, an early step in the cascade of leukocyte-endothelial interactions, also strongly diminished in vivo vascular permeability in
unstimulated and TNF-stimulated vessels (112). Nonetheless, the
authors also pointed out that in unstimulated arterioles and venules, leukocyte-endothelial cell interactions are negligible while
permeability was largely reduced in ICAM-1 KO mice. Moreover,
© Schattauer 2014
Thrombosis and Haemostasis 112.6/2014
Downloaded from www.thrombosis-online.com on 2017-06-17 | IP: 88.99.165.207
For personal or educational use only. No other uses without permission. All rights reserved.
1094
1095
Marcos-Ramiro, García-Weber et al. Adhesion receptors and ROS in TNF-induced endothelial permeability
the effect of the absence of ICAM-1 on permeability in unstimulated and TNF-stimulated animals was much stronger than that
caused by the absence or the inhibition of CD18. Together, these
results imply the existence of an engagement-independent contribution of ICAM-1 to the control of vascular permeability also in
vivo. Indeed, whereas ICAM-1 KO mice showed a marked reduction in permeability in all the vascular beds analysed, antibodymediated engagement of the receptor induced permeability only in
arterioles but not in venules. This suggests the co-existence, in
vivo, of ligand-dependent and ligand-independent roles for
ICAM-1 in regulating barrier function, whose relative importance
depends on the type of vessel analysed (113).
ICAM-1 signals to AJs upon engagement
What are the molecular mechanisms underlying the effect of
ICAM-1 engagement on barrier function? Data from in vitro experimental approaches highlight the importance of ICAM-1 signalling to F-actin upon clustering. From its C-terminal segment,
ICAM-1 interacts with proteins such as filamin, α-actinin, or ERM
proteins, which bind F-actin (90). Antibody-mediated clustering
induces ICAM-1 alignment with moesin and actin stress fibres
(93, 114), which are anchored by their ends to cell-cell junctions
(34). Clustered ICAM-1 not only aligns with these actin filaments
but also activates RhoA, generates more stress fibres and eventually moves to the periphery of the cell, close to cell–cell junctions,
where the receptor may generate tension and transduce signals to
junctional protein complexes (93, 115). Thus, TNF would assemble stress fibres that do not generate tension by themselves, but
that would work as platforms where ICAM-1 rapidly clusters and
induces transient contraction at cell borders. Indeed, as occurs
with TNF, ICAM-1 engagement is sufficient to induce VE-cadherin phosphorylation in tyrosine residues and increase paracellular
permeability (104, 115). Further, ICAM-1 crosslinking activates
Trio and Rac (116) and produces ROS (117, 118), although the signalling pathways that lead VE-cadherin phosphorylation have not
been fully elucidated for this adhesion receptor. For instance, they
may (104), or may not involve (115) the activation of Src and
Pyk2, in response to receptor clustering (117, 104, 119). Although
the effect of ICAM-1-induced tyrosine phosphorylation of VEcadherin in endothelial permeability has not yet been examined, it
is well known that inhibition of Src-regulated tyrosine phosphorylation of VE-cadherin greatly impairs the increase in endothelial
permeability in response to TNF (71), LPS (120) and VEGF (121).
In addition, the modulation that the different phosphorylated tyrosine residues exert on VE-cadherin and AJs also remains to be
investigated in detail. For instance, VE-cadherin phosphorylation
in response to ICAM-1 does not cause significant catenin dissociation (115), but VE-cadherin mutants that prevent tyrosine phosphorylation in two cytoplasmic tyrosine residues dampen permeability induction and p120-catenin dissociation in response to
VEGF (68). The picture is complicated by the fact that, depending
on the stimulus and vascular bed analysed, different kinases from
the Src family are activated and different VE-cadherin tyrosine
residues are phosphorylated (115, 71, 121, 68, 122). A clear
example of the role of different tyrosine residues of VE-cadherin
function comes from a recent report demonstrating the different
roles of Y685 and Y731 in murine VE-cadherin. TNF and other
inflammatory mediators increase Y685 phosphorylation to induce
permeability, whereas T-cells increase Y685 phosphorylation and
promote Y732 dephosphorylation, the latter being necessary for
neutrophil extravasation (123). Finally, it should be taken into account that the local increase in actomyosin-mediated tension provoked by ICAM-1 activation of RhoA may open intercellular gaps
solely due to the abrupt changes in cell morphology with no active,
signalling-mediated disruption of junctional complexes. This may
provide an alternative mechanism for barrier rupture in response
to receptor signalling (▶ Figure 3) (114, 124, 125).
A significant proportion of antibody-crosslinked ICAM-1 is
translocated to peripheral domains enriched in caveolae, where
actin filaments converge and where the receptor can undergo
transcytosis (93). In addition to forming part of the channels that
mediate leukocyte transcellular diapedesis (93, 95, 126), caveolae
mediate transcytosis of plasma solutes such as albumin (127).
Consistent with this, ICAM-1 signals to caveolae and mediates the
phosphorylation of caveolin-1 by Src, which produces endothelial
hyperpermeability that is dependent on the caveolar vesicular system (128). It is not clear, however, whether this hyperpermeability
involves the modulation of transcytosis only or of both transcytosis and paracellular permeability, as shown for other soluble mediators (▶ Figure 3) (129).
VCAM-1 signals to AJs upon engagement through
ROS generation
Another adhesion receptor dramatically increased in the endothelium by TNF is the vascular cell adhesion molecule-1 (VCAM-1).
TNF-mediated increase of VCAM-1 also occurs through TNFR1
activation of the NF-κB pathway (97, 98). Two NF-kB sites proximal to VCAM-1 promoter are necessary for its transcription (100,
130, 131). In HUVECs, the NF-κB1/RelA heterodimer bind to the
NF-κB binding site of VCAM-1 gene and controls its translation
(130). Like ICAM-1, VCAM-1 also associates with ERM (106),
aligns with and increases stress fibres upon crosslinking (93, 132).
However, parallel crosslinking of ICAM-1 and VCAM-1 produces
differential outcomes, mostly depending on the experimental system being examined. Using bone marrow endothelial cells, van
Buul et al. (133) demonstrated that VCAM-1 crosslinking increased permeability whereas ICAM-1 crosslinking did not. In
contrast, in vivo crosslinking of these two receptors revealed a role
for ICAM-1 in permeability induction in unstimulated venules
and arterioles, but not for VCAM-1 (112, 113). None of these reports addressed the relative expression of the two receptors, and it
is not possible to determine whether receptor expression or the
different experimental settings determines the prevalence of one
or other receptor in signal transduction. However, using antibodycoated beads as surrogate leukocytes that uniquely bind ICAM-1
or VCAM-1 has revealed that single receptor engagement can induce co-clustering of the other receptor (134, 135). This strongly
suggests that, in a physiological context, leukocytes interacting
Thrombosis and Haemostasis 112.6/2014
© Schattauer 2014
Downloaded from www.thrombosis-online.com on 2017-06-17 | IP: 88.99.165.207
For personal or educational use only. No other uses without permission. All rights reserved.
Marcos-Ramiro, García-Weber et al. Adhesion receptors and ROS in TNF-induced endothelial permeability
Figure 3: Endothelial adhesion receptors regulated by TNF induce
endothelial barrier disruption. TNF induces a long-term inflammatory response that implies generating stress fibres and expression of adhesion receptors, namely ICAM-1 and VCAM-1, involved in leukocyte adhesion and
transendothelial migration. Inhibition of TNF-induced actin remodelling
through RhoA or Rac is not sufficient to prevent permeability increase.
Hence, additional mechanisms are likely to contribute to barrier disruption.
Upregulated ICAM-1 and VCAM-1 associate to F-actin and promote actomyosin tension and activation of connectors between plasma membrane and
F-actin, such as ERM proteins, which induce cell contraction and barrier function decrease. Clustered ICAM-1 is internalized in caveolae and increases caveolin-1-mediated albumin permeability. TNF also causes dispersion from
junctions and from the parajunctional LBRC of proteins playing a dual function as adhesion receptors and cell-cell junction components such as JAMs
and PECAM-1. ICAM-1 and VCAM-1 activate Rac and generate reactive
oxygen species (ROS) that destabilise adherens junctions through various
signalling pathways.
with TNF-activated endothelial monolayers modulate vascular
permeability through simultaneous signalling from both adhesion
receptors.
Rac1 GTPase is activated downstream of crosslinked ICAM-1
and VCAM-1 (132, 136). However, the effect of endothelial Rac1
activation upon receptor clustering has been characterised only for
VCAM-1. Rac1 inhibition prevents intercellular gaps opening in
response to VCAM-1 crosslinking (137). Further, Rac activity is
necessary for the VCAM-1-mediated generation of ROS (137),
which are produced at similar levels by leukocyte adhesion to endothelium. Importantly, dampening ROS levels with ROS scavengers inhibits the effect of active Rac on endothelial permeability
(132). Like TNF, VCAM-1-induced ROS may activate Src family
kinases and trigger disruption of cell-cell junctions in a manner
similar to TNF. However, some additional players have been identified in the VCAM-1-mediated signalling pathway. First, an additional molecular mechanism for VCAM-1 and ROS regulation of
endothelial cell–cell junctions has been recently proposed (138).
Vockel et al. showed that VCAM-1-induced Rac activity and ROS
generation promotes dissociation between VE-cadherin and VEPTP, leading to the disruption of AJs. VE-PTP is a receptor-type
protein tyrosine phosphatase expressed in endothelium that associates with VE-cadherin and maintains AJs (139). Antibody engagement of VCAM-1 or lymphocyte adhesion clearly disrupts the
interaction between these two proteins, leading to VE-cadherin
hyperphosphorylation through the tyrosine phosphatase Pyk2
(140). These findings therefore suggest that immune cells transiently regulate endothelial barrier function during TEM through
VCAM-1-mediated transient disruption of the VE-PTP-VE-cadherin complex (▶ Figure 3).
Second, VCAM-1-induced ROS also stimulate matrix metalloproteinases (MMPs), a family of zinc-dependent endopeptidases
with structural homologies that regulate a plethora of events related to inflammation, cancer and tissue remodelling (141). In particular, VCAM-1 stimulates MMP2 and MMP9, as determined by
zymography (142). The real contribution of MMPs to endothelial
permeability during leukocyte transmigration has not yet been determined (143, 144, 145, 146). However, in a broader context, it
has been shown that MMP inhibition impairs TNF-induced VEcadherin shedding in HUVECs (147), suggesting a role for these
proteases in regulating cell-cell junctions.
TNF and junctional adhesion molecules (JAMs)
JAMs are a paradigm of proteins playing a double function in
regulating permeability and leukocyte transmigration. JAMs establish cell-cell junctional complexes that regulate permeability in
epithelium and endothelium (148, 149). JAM-family members
contain two extracellular immunoglobulin-like domains, a single
transmembrane segment and a short cytoplasmic tail with a PDZdomain-binding motif. They can establish homophilic and heterophilic interactions in cis and trans (148). JAM-A, JAM-B and
© Schattauer 2014
Thrombosis and Haemostasis 112.6/2014
Downloaded from www.thrombosis-online.com on 2017-06-17 | IP: 88.99.165.207
For personal or educational use only. No other uses without permission. All rights reserved.
1096
1097
Marcos-Ramiro, García-Weber et al. Adhesion receptors and ROS in TNF-induced endothelial permeability
JAM-C are expressed in endothelial cells, localised in tight junctions and associated with tight junction components, such as ZO-1
(150-152), and with polarity proteins such as PAR3 (152, 153). In
addition, JAM-A and JAM-C interact with β2 integrins LFA-1
(154) and Mac-1 (155), respectively, whereas JAM-B is a counterreceptor of the β1 integrin VLA-4 (156). Furthermore, JAM-B is a
counterreceptor of JAM-C (157), and JAM-C expression is
required for an efficient JAM-B-VLA-4 interaction (156). Although there is sound evidence that JAM proteins help regulate
epithelial barrier function (158-160), the research on endothelial
JAM proteins has mainly focused on their role in TEM of immune
cells and their function in maintaining endothelial permeability
has not been elucidated for all of them. Only the role of JAM-C as
a negative regulator of endothelial barrier function has been addressed in some detail (161, 162).
Thus, the JAMs are all prime candidates for regulating the
crosstalk between immune cell adhesion and permeability in the
endothelium during inflammation, and they may also contribute
to the TNF-mediated changes of endothelial barrier features. The
expression of JAM-A and JAM-B is regulated by TNF in cultured
aortic or lymphatic endothelial cells (163-166). A significant increase in JAM-B expression has been found in arterioles within inflammatory foci in lung with chronic bronchopneumonia, suggesting that its expression may be regulated by some inflammatory
mediators (165, 167). As mentioned, TNF induces the cleavage
and shedding of JAM-A in HUVECs. However, other researchers
working on the same cellular model did not find a significant decrease in total or surface JAM-A levels (17, 92, 164, 166). Instead,
they detected significant dispersion of this molecule from junctions (92). The molecular events mediating JAM-A relocation or
shedding and the contribution of JAMs to TNF-mediated barrier
disruption in vascular endothelium have not been yet investigated.
TNF regulates endothelial barrier function through
platelet-endothelial cell adhesion molecule-1
(PECAM-1)
PECAM-1 is a transmembrane receptor of the immunoglobulin
superfamily that forms homophilic interactions between adjacent
endothelial cells or between endothelial cells and leukocytes. It is
important for efficient leukocyte diapedesis, being involved in its
paracellular and transcellular routes (94, 168). In the last decade it
has become clear that PECAM-1 is a regulator of endothelial permeability. PECAM-1 KO mice exhibit hyperpermeability in the
blood brain barrier, an early onset of experimental autoimmune
encephalomyelitis (EAE), the experimental animal model of
multiple sclerosis. The lack of this receptor also causes a prolonged
increase in permeability in response to acute contraction caused
by histamine (169, 170) and thrombin (171). The cytoplasmic segment of PECAM-1 is a scaffold for phosphatases and components
of cell-cell junctions. PECAM-1 contains an immunoreceptor tyrosine-based inhibitory motif (ITIM), with two tyrosine residues
(Y663 and Y686) that become phosphorylated by Src, Csk and Fer
kinases in response to different stimuli (172, 173, 174). This phosphorylated ITIM recruits SH2 domain-containing protein tyrosine
phosphatases (SHPs), of which SHP-2 is the best characterised in
endothelial cells (169, 175). On the other hand, PECAM-1 also
binds to β-catenin (176) and γ-catenin (177) in endothelial cells.
γ-catenin binds the region coded by exon 13, which overlaps with
the ITIM motif, whereas β-catenin binds a region closer to the
protein C-terminus coded by the exon 15. The association of these
two catenins with PECAM-1 depends on PECAM-1 tail phosphorylation status. Reciprocally, PECAM-1 negatively regulates
vascular β-catenin phosphorylation in tyrosine residues in response to acute contraction provoked by histamine. This phosphorylation is mediated by the recruitment of SHP-2 to
PECAM-1, which dephosphorylates β-catenin in the proximity of
the receptor. Accordingly, histamine induces association of β-catenin to PECAM-1 over a period consistent with the process of endothelial barrier reformation. In the absence of PECAM-1, SHP-2
is not available at junctions and β-catenin
is not dephosphorylated. This impairs β-catenin association to VE-cadherin during
junctional reannealing, thereby debilitating intercellular junctions
(169, 176, 177). In addition, SHP-2 not only interacts with and
phosphorylates β-catenin during barrier formation (178, 179), but
also may target VE-cadherin, since SHP-2 inhibition increases
both β-catenin and VE-cadherin tyrosine phosphorylation in pulmonary endothelial cells (180). Finally, similar to SHP-2, SHP-1 is
associated with PECAM-1 in non-endothelial cell types (172, 181),
which suggests that it may also play a role regulating endothelial
PECAM-1 and cell–cell junctions.
The detailed analysis of PECAM-1 localisation has provided
important information about the spatiotemporal relationship between PECAM-1 and AJs and its role in barrier function. It has
been reported that PECAM-1 is localised in a subjunctional compartment known as the lateral border recycling compartment
(LBRC), from where it constitutively recycles to the junctional surface (168) (▶ Figure 3). This localisation is necessary for
PECAM-1 function in leukocyte diapedesis, although the nature
and organisation of this compartment are still unclear. Importantly, PECAM-1 at the LBRC crosstalks with AJs, which form a
junctional reticulum in resting cells. Analysis by superresolution
confocal microscopy showed that the distributions of PECAM-1
and AJ overlap at these reticular borders, but form different compartments (22). Indeed, experiments expressing p120-catenin
tagged to two different fluorescent proteins in two adjacent endothelial cells demonstrated that reticular AJs are exclusively distributed in the areas of contact between two adjacent cells, which
suggests that they are surface, intercellular complexes, but not an
intracellular compartment. Reticular AJs, however, determine the
localisation of the internal LBRC (22).
TNF increases PECAM-1 phosphorylation, implying a stronger
association with SH2-containing proteins, and disperses
PECAM-1 from cell borders over the entire cell plasma membrane
(72, 91). Localisation studies suggest that perijunctional JAM-A is
also a component of the LBRC (182) and, as mentioned above, is
also dispersed in response to TNF, suggesting that this cytokine
may disrupt the whole LBRC rather than dispersing individual receptors. It is not yet clear how TNF disrupts the LBRC. The distribution of the LBRC is controlled by reticular AJs, which in turn
Thrombosis and Haemostasis 112.6/2014
© Schattauer 2014
Downloaded from www.thrombosis-online.com on 2017-06-17 | IP: 88.99.165.207
For personal or educational use only. No other uses without permission. All rights reserved.
Marcos-Ramiro, García-Weber et al. Adhesion receptors and ROS in TNF-induced endothelial permeability
are distributed at junctional regions with low F-actin content (22).
TNF increases stress fibres that become associated from their ends
to AJs (34), decreasing reticular organisation of cell-cell junctions
which may thereby lead to LBRC dispersion (22). However, reduction of F-actin increase by the ROCK inhibitor Y27632 does not
prevent the disappearance of reticular AJ, just as it likewise does
not prevent permeability increase. This implies that additional
mechanisms may regulate PECAM-1 dispersion from cell–cell
borders during long-term inflammation. SHP phosphatases are
candidates for regulating PECAM-1 localisation in response to
TNF because they are involved in TNF-mediated signalling:
SHP-2 is associated with IKK and mediates NF-κB activation in
response to TNF (183); SHP-1 and SHP-2 associate with the death
domain of TNFR1 in response to TNF, thereby modulating cytokine-mediated survival (184). TNF also induces SHP-2 activity in
HUVECs (185) as well a moderate increment in SHP-1 expression
and activity (186, 187). To date, however, the effect of TNF-mediated activation of SHPs on the localisation of PECAM-1 has not
been investigated.
Finally, PECAM-1 inhibition with blocking antibodies or by
gene silencing mimics the effect of TNF on cell-cell junctions and
endothelial barrier dysfunction, but no other TNF-related features
such as actin remodelling or ICAM-1 expression (22). Collectively,
these data suggest that PECAM-1 dispersion is, indeed, what disrupts low actin reticular junctions and not vice versa, strongly supporting PECAM-1 and, probably, LBRC dispersion as an additional, actin-independent mechanism that contributes to permeability increase during TNF stimulation (▶ Figure 1 and ▶ Figure
3).
like that generated by leukocyte interactions during TEM. Rather
than a cytoskeletal network responsible for the gradual and prolonged barrier function decline that TNF induces, stress fibres
could instead be a prerequisite for enabling immune cells to signal
efficiently to endothelial cell-cell junctions. Such signalling would
involve generating actomyosin-mediated tension and ROS to produce successful paracellular diapedesis (and subsequent hyperpermeability). On the other hand, at least for ICAM-1 and PECAM-1,
a role in modulating endothelial barrier function independent of
their encounter with immune cells is also emerging from recent reports. Nevertheless, many questions regarding the contribution of
these and other receptors to TNF-mediated barrier disruption remain unanswered, but it seems clear that such a contribution
largely depends on their relative expression levels, on the vascular
beds in which they are expressed and on their molecular crosstalk
to cell-cell junctional proteins. A deeper knowledge of human endothelial and vascular heterogeneity seems essential if we want to
fully elucidate the molecular mechanisms that contribute to vessel
leakiness during long-term physiological and pathological inflammation.
Acknowledgements
The expert technical advice of José I. Belio López from the graphic
design facility of the CBMSO, Madrid, is gratefully acknowledged.
This work was supported by grants SAF2011–22624 (to J.M.) from
the Ministerio de Ciencia e Innovación; grant S2010/BMD-2305
from Comunidad de Madrid; and grant Convenio Colaboración
Fundación Jiménez Díaz. B.M.R. and D.G.W. are recipients of an
FPI fellowship from the Ministerio de Ciencia e Innovación.
Conflicts of interest
Concluding remarks
None declared.
The fact that inhibiting actin remodelling through Rho GTPases
has less effect than expected on endothelial barrier disruption
upon TNF exposure indicates the existence of additional mechanisms involved in this process. In recent years, the Rac GTPase subfamily emerges as a main regulator of endothelial barrier function,
not only through the direct regulation of actin filaments, but also
as part of the protein complexes that elevate ROS production and
modulate cell-cell junctions in response to cytokines or leukocyte
adhesion. However, if actin stress fibres are not essential for increasing endothelial permeability in response to TNF in the absence of immune cells, they may play other roles in the endothelial
inflammatory response. In this review we have presented evidence
that leukocyte-endothelial cell interactions are of fundamental importance for barrier disruption during inflammation in experimental in vivo models. As mentioned above, stress fibres become
connected to engaged adhesion receptors such as ICAM-1 and
VCAM-1 (114, 93) and to discontinuous AJs from their ends (34).
This type of AJs contains vinculin, a protein involved in transmitting mechanical forces that is required for TNF-induced junction
remodelling (34, 36, 188). The identification of a mechanosensory
protein in the junctional structures that hold stress fibres, suggest
that these actin filaments are able to withstand mechanical stress,
References
1. Medzhitov R. Inflammation 2010: new adventures of an old flame. Cell 2010;
140: 771-776.
2. Madge LA, Pober JS. TNF signalling in vascular endothelial cells. Exp Mol Pathol 2001; 70: 317-325.
3. Ruddle N H. Tumor necrosis factor (TNF-alpha) and lymphotoxin (TNF-beta).
Curr Opin Immunol 1992; 4: 327-332.
4. Bradley JR. TNF-mediated inflammatory disease. J Pathol 2008; 214: 149-160.
5. Scheller J, et al. ADAM17: a molecular switch to control inflammation and tissue regeneration. Trends Immunol 2011; 32: 380-387.
6. Black RA, et al. A metalloproteinase disintegrin that releases tumour-necrosis
factor-alpha from cells. Nature 1997; 385: 729-733.
7. Cabal-Hierro L, Lazo PS. Signal transduction by tumor necrosis factor receptors. Cell Signal 2012; 24: 1297-1305.
8. Waters JP, et al. Tumour necrosis factor and cancer. J Pathol 2013; 230: 241-248.
9. Waters JP, et al. Tumour necrosis factor in infectious disease. J Pathol 2013; 230:
132-147.
10. Chu WM. Tumor necrosis factor. Cancer Lett 2013; 328: 222-225.
11. Vallabhapurapu S, Karin M. Regulation and function of NF-kappaB transcription factors in the immune system. Ann Rev Immunol 2009; 27: 693-733.
12. Brenner DA, et al. Prolonged activation of jun and collagenase genes by tumour
necrosis factor-alpha. Nature 1989; 337: 661-663.
13. Vandenabeele P, et al. The role of the kinases RIP1 and RIP3 in TNF-induced
necrosis. Sci Signal 2010; 3: re4.
© Schattauer 2014
Thrombosis and Haemostasis 112.6/2014
Downloaded from www.thrombosis-online.com on 2017-06-17 | IP: 88.99.165.207
For personal or educational use only. No other uses without permission. All rights reserved.
1098
1099
Marcos-Ramiro, García-Weber et al. Adhesion receptors and ROS in TNF-induced endothelial permeability
14. Ridley AJ. Rho family proteins: coordinating cell responses. Trends Cell Biol
2001; 11: 471-477.
15. Wennerberg K, Der CJ. Rho-family GTPases: it's not only Rac and Rho (and I
like it). J Cell Sci 2004; 117: 1301-1312.
16. Ridley AJ. Rho GTPases and actin dynamics in membrane protrusions and vesicle trafficking. Trends Cell Biol 2006; 16: 522-529.
17. McKenzie JA, Ridley AJ. Roles of Rho/ROCK and MLCK in TNF-alpha-induced
changes in endothelial morphology and permeability. J Cell Physiol 2007; 213:
221-228.
18. Hanna AN, et al. Tumor necrosis factor-alpha induces stress fibre formation
through ceramide production: role of sphingosine kinase. Mol Biol Cell 2001;
12: 3618-3630.
19. Puls A, et al. Activation of the small GTPase Cdc42 by the inflammatory cytokines TNF(alpha) and IL-1, and by the Epstein-Barr virus transforming protein
LMP1. J Cell Sci 1999; 112: 2983-2992.
20. Wojciak-Stothard B, et al. Regulation of TNF-alpha-induced reorganisation of
the actin cytoskeleton and cell-cell junctions by Rho, Rac, and Cdc42 in human
endothelial cells. J Cell Physiol 1998; 176: 150-165.
21. Kant, S. et al. TNF-stimulated MAP kinase activation mediated by a Rho family
GTPase signalling pathway. Genes Dev 2011; 25: 2069-2078.
22. Fernandez-Martin L, et al. Crosstalk between reticular adherens junctions and
platelet endothelial cell adhesion molecule-1 regulates endothelial barrier function. Arterioscl Thromb Vasc Biol 2012; 32: e90-e102.
23. Hocking DC, et al. Mechanisms of pulmonary edema induced by tumor necrosis factor-alpha. Circulation Res 1990; 67: 68-77.
24. Stolpen AH, et al. Recombinant tumor necrosis factor and immune interferon
act singly and in combination to reorganize human vascular endothelial cell
monolayers. Am J Pathol 1986; 123: 16-24.
25. Deli MA, et al. Exposure of tumor necrosis factor-alpha to luminal membrane of
bovine brain capillary endothelial cells cocultured with astrocytes induces a delayed increase of permeability and cytoplasmic stress fibre formation of actin. J
Neurosci Res 1995; 41: 717-726.
26. Kimura K, et al. Regulation of myosin phosphatase by Rho and Rho-associated
kinase (Rho-kinase). Science 1996; 273: 245-248.
27. Watanabe N, et al. Cooperation between mDia1 and ROCK in Rho-induced
actin reorganisation. Nature Cell Biol 1999; 1: 136-143.
28. Chrzanowska-Wodnicka M, Burridge K. Rho-stimulated contractility drives the
formation of stress fibres and focal adhesions. J Cell Biol 1996; 133: 1403-1415.
29. Amano M, et al. Phosphorylation and activation of myosin by Rho-associated
kinase (Rho-kinase). J Biol Chem 1996; 271: 20246-20249.
30. Mong PY, et al. Activation of Rho kinase by TNF-alpha is required for JNK activation in human pulmonary microvascular endothelial cells. J Immunol 2008;
180: 550-558.
31. Campos SB, et al. Cytokine-induced F-actin reorganisation in endothelial cells
involves RhoA activation. Am J Physiol Renal Physiol 2009; 296: F487-495.
32. Schlegel N, Waschke J. Impaired cAMP and Rac 1 signalling contribute to TNFalpha-induced endothelial barrier breakdown in microvascular endothelium.
Microcirculation 2009; 16: 521-533.
33. Mong PY, Wang Q. Activation of Rho kinase isoforms in lung endothelial cells
during inflammation. J Immunol 2009; 182: 2385-2394.
34. Millan J, et al. Adherens junctions connect stress fibres between adjacent endothelial cells. BMC Biol 2010; 8: 11.
35. Dejana E. Endothelial cell-cell junctions: happy together. Nat Rev Mol Cell Biol
2004; 5: 261-270.
36. Huveneers S, et al. Vinculin associates with endothelial VE-cadherin junctions
to control force-dependent remodelling. J Cell Biol 2012; 196: 641-652.
37. Naikawadi RP, et al. A critical role for phosphatidylinositol (3,4,5)-trisphosphate-dependent Rac exchanger 1 in endothelial junction disruption and vascular hyperpermeability. Circulation Res 2012; 11: 1517-1527.
38. Shao M, et al. Caveolin-1 regulates Rac1 activation and rat pulmonary microvascular endothelial hyperpermeability induced by TNF-alpha. PLoS ONE
2013; 8: e55213.
39. Van Rijssel J, et al. The Rho-GEF Trio regulates a novel pro-inflammatory pathway through the transcription factor Ets2. Biol Open 2013; 2: 569-579.
40. Cain RJ, et al. The PI3K p110alpha isoform regulates endothelial adherens junctions via Pyk2 and Rac1. J Cell Biol 2010; 188: 863-876.
41. Papakonstanti EA, Stournaras C. Tumor necrosis factor-alpha promotes survival
of opossum kidney cells via Cdc42-induced phospholipase C-gamma1 activation and actin filament redistribution. Mol Biol Cell 2004; 15: 1273-1286.
42. Vandenbroucke E, et al. Regulation of endothelial junctional permeability. Ann
NY Acad Sci 2008; 1123: 134-145.
43. van Nieuw Amerongen GP, et al. Involvement of Rho kinase in endothelial barrier maintenance. Arterioscl Thromb Vasc Biol 2007; 27: 2332-2339.
44. Wojciak-Stothard B, et al. Rho and Rac but not Cdc42 regulate endothelial cell
permeability. J Cell Sci 2001; 114: 1343-1355.
45. Wojciak-Stothard B, Ridley AJ. Rho GTPases and the regulation of endothelial
permeability. Vascul Pharmacol 2002; 39: 187-199.
46. Petrache I. et al. Differential effect of MLC kinase in TNF-alpha-induced endothelial cell apoptosis and barrier dysfunction. Am J Physiol Lung Cell Mol
Physiol 2001; 280: L1168-1178.
47. Reed SC, et al. Rickettsia parkeri invasion of diverse host cells involves an
Arp2/3 complex, WAVE complex and Rho-family GTPase-dependent pathway.
Cell Microbiol 2012; 14: 529-545.
48. Bandman O, et al. Complexity of inflammatory responses in endothelial cells
and vascular smooth muscle cells determined by microarray analysis. Ann NY
Acad Sci 2002; 975: 77-90.
49. Giusti B, et al. Desmoglein-2-integrin Beta-8 interaction regulates actin assembly in endothelial cells: deregulation in systemic sclerosis. PLoS ONE 2013;
8: e68117.
50. Birukova AA, et al. HGF attenuates thrombin-induced endothelial permeability
by Tiam1-mediated activation of the Rac pathway and by Tiam1/Rac-dependent inhibition of the Rho pathway. Faseb J 2007; 21: 2776-2786.
51. Di Lorenzo A, et al. eNOS derived nitric oxide regulates endothelial barrier
function via VE cadherin and Rho GTPases. J Cell Sci 2013; Epub ahead of
print.
52. Gavard J, Gutkind JS. VEGF controls endothelial-cell permeability by promoting the beta-arrestin-dependent endocytosis of VE-cadherin. Nat Cell Biol
2006; 8: 1223-1234.
53. Birukova AA, et al. Epac/Rap and PKA are novel mechanisms of ANP-induced
Rac-mediated pulmonary endothelial barrier protection. J Cell Physiol 2008;
215: 715-724.
54. Konior A, et al. NADPH Oxidases in Vascular Pathology. Antioxid Redox Signal
2013; Epub ahead of print.
55. Van Buul JD, et al. Expression and localisation of NOX2 and NOX4 in primary
human endothelial cells. Antioxid Redox Signal 2005; 7: 308-317.
56. Frey RS, et al. NADPH oxidase-dependent signalling in endothelial cells: role in
physiology and pathophysiology. Antioxid Redox Signal 2009; 11: 791-810.
57. Hordijk PL. Regulation of NADPH oxidases: the role of Rac proteins. Circulation Res 2006; 98: 453-462.
58. Martyn KD, et al. Functional analysis of Nox4 reveals unique characteristics
compared to other NADPH oxidases. Cell Signal 2006; 18: 69-82.
59. Bedard K, et al. NOX5: from basic biology to signalling and disease. Free Radical Biol Med 2012; 52: 725-734.
60. Kim YS, et al. TNF-induced activation of the Nox1 NADPH oxidase and its role
in the induction of necrotic cell death. Mol Cell 2007; 26: 675-687.
61. Li JM, et al. Acute tumor necrosis factor alpha signalling via NADPH oxidase in
microvascular endothelial cells: role of p47phox phosphorylation and binding
to TRAF4. Mol Cell Biol 2005; 25: 2320-2330.
62. Zhao J, et al. Mixed lineage kinase domain-like is a key receptor interacting protein 3 downstream component of TNF-induced necrosis. Proc Natl Acad Sci
USA 2012; 109: 5322-5327.
63. Corcoran A, Cotter TG. Redox regulation of protein kinases. Febs J 2013; 280:
1944-1965.
64. Giannoni E, Chiarugi P. Redox Circuitries Driving Src Regulation. Antiox
Redox Signal 2013; Epub ahead of print.
65. Saksena S, et al. Role of Fyn and PI3K in H2O2-induced inhibition of apical Cl-/
OH- exchange activity in human intestinal epithelial cells. Biochem J 2008; 416:
99-108.
66. Giannoni E, et al. Intracellular reactive oxygen species activate Src tyrosine kinase during cell adhesion and anchorage-dependent cell growth. Mol Cell Biol
2005; 25: 6391-6403.
67. Kemble DJ, Sun G. Direct and specific inactivation of protein tyrosine kinases in
the Src and FGFR families by reversible cysteine oxidation. Proc Natl Acad Sci
USA 2009; 106: 5070-5075.
68. Monaghan-Benson E, Burridge K. The regulation of vascular endothelial
growth factor-induced microvascular permeability requires Rac and reactive
oxygen species. J Biol Chem 2009; 284: 25602-25611.
Thrombosis and Haemostasis 112.6/2014
© Schattauer 2014
Downloaded from www.thrombosis-online.com on 2017-06-17 | IP: 88.99.165.207
For personal or educational use only. No other uses without permission. All rights reserved.
Marcos-Ramiro, García-Weber et al. Adhesion receptors and ROS in TNF-induced endothelial permeability
69. Potter MD, et al. Tyrosine phosphorylation of VE-cadherin prevents binding of
p120- and beta-catenin and maintains the cellular mesenchymal state. J Biol
Chem 2005; 280: 31906-31912.
70. Nwariaku FE, et al. Tyrosine phosphorylation of vascular endothelial cadherin
and the regulation of microvascular permeability. Surgery 2002; 132: 180-185.
71. Angelini DJ, et al. TNF-alpha increases tyrosine phosphorylation of vascular endothelial cadherin and opens the paracellular pathway through fyn activation in
human lung endothelia. Am J Physiol Lung Cell Mol Physiol 2006; 291:
L1232-1245.
72. Ferrero E, et al. Tumor necrosis factor alpha-induced vascular leakage involves
PECAM1 phosphorylation. Cancer Res 1996; 56: 3211-3215.
73. van Buul JD, et al. Proline-rich tyrosine kinase 2 (Pyk2) mediates vascular endothelial-cadherin-based cell-cell adhesion by regulating beta-catenin tyrosine
phosphorylation. J Biol Chem 2005; 280: 21129-21136.
74. Schlaepfer DD, et al. Tumor necrosis factor-alpha stimulates focal adhesion kinase activity required for mitogen-activated kinase-associated interleukin 6 expression. J Biol Chem 2007; 282: 17450-17459.
75. Tseng WP., et al. FAK activation is required for TNF-alpha-induced IL-6 production in myoblasts. J Cell Physiol 2010; 223: 389-396.
76. Mon NN, et al. A role for focal adhesion kinase signalling in tumor necrosis factor-alpha-dependent matrix metalloproteinase-9 production in a cholangiocarcinoma cell line, CCKS1. Cancer Res 2006; 66: 6778-6784.
77. Knezevic N, et al. The G protein betagamma subunit mediates reannealing of
adherens junctions to reverse endothelial permeability increase by thrombin. J
Exp Med 2009; 206: 2761-2777.
78. Belvitch P, Dudek SM. Role of FAK in S1P-regulated endothelial permeability.
Microvasc Res 2012; 83: 22-30.
79. Chen XL, et al. VEGF-induced vascular permeability is mediated by FAK. Develop Cell 2012; 22: 146-157.
80. Dikic I, et al. A role for Pyk2 and Src in linking G-protein-coupled receptors
with MAP kinase activation. Nature 1996; 383: 547-550.
81. Matsui A, et al. Central role of calcium-dependent tyrosine kinase PYK2 in endothelial nitric oxide synthase-mediated angiogenic response and vascular
function. Circulation 2007; 116: 1041-1051.
82. Kolaczkowska E, Kubes P. Neutrophil recruitment and function in health and
inflammation. Nature Rev Immunol 2013; 13: 159-175.
83. Shi C, Pamer EG. Monocyte recruitment during infection and inflammation.
Nature Rev Immunol 2011; 11: 762-774.
84. Shechter R, et al. Orchestrated leukocyte recruitment to immune-privileged
sites: absolute barriers versus educational gates. Nature reviews. Immunol 2013;
13: 206-218.
85. Pober JS, et al. Two distinct monokines, interleukin 1 and tumor necrosis factor,
each independently induce biosynthesis and transient expression of the same
antigen on the surface of cultured human vascular endothelial cells. J Immunol
1986; 136: 1680-1687.
86. Osborn L, et al. Direct expression cloning of vascular cell adhesion molecule 1, a
cytokine-induced endothelial protein that binds to lymphocytes. Cell 1989; 59:
1203-1211.
87. Shimizu Y, et al. Four molecular pathways of T cell adhesion to endothelial cells:
roles of LFA-1, VCAM-1, and ELAM-1 and changes in pathway hierarchy under
different activation conditions. J Cell Biol 1991; 113: 1203-1212.
88. Pober JS, et al. Activation of cultured human endothelial cells by recombinant
lymphotoxin: comparison with tumor necrosis factor and interleukin 1 species.
J Immunol 1987; 138: 3319-3324.
89. Liu Z, et al. Differential regulation of human and murine P-selectin expression
and function in vivo. J Exp Med 2010; 207: 2975-2987.
90. Reglero-Real N, et al. Endothelial membrane reorganisation during leukocyte
extravasation. Cell Mol Life Sci 2012; 69: 3079-3099.
91. Romer LH, et al. IFN-gamma and TNF-alpha induce redistribution of
PECAM-1 (CD31) on human endothelial cells. J Immunol 1995; 154:
6582-6592.
92. Martinez-Estrada OM, et al. Opposite effects of tumor necrosis factor and soluble fibronectin on junctional adhesion molecule-A in endothelial cells. Am J
Physiol Lung Cell Mol Physiol 2005; 288: L1081-1088.
93. Millan J, et al. Lymphocyte transcellular migration occurs through recruitment
of endothelial ICAM-1 to caveola- and F-actin-rich domains. Nat Cell Biol
2006; 8: 113-123.
94. Carman CV, et al. Transcellular diapedesis is initiated by invasive podosomes.
Immunity 2007; 26: 784-797.
95. Keuschnigg J, et al. The prototype endothelial marker PAL-E is a leukocyte trafficking molecule. Blood 2009; 114: 478-484.
96. Aranda JF, et al. MYADM controls endothelial barrier function through ERMdependent regulation of ICAM-1 expression. Mol Biol Cell 2013; 24: 483-494.
97. Zhou Z, et al. TNFR1-induced NF-kappaB, but not ERK, p38MAPK or JNK activation, mediates TNF-induced ICAM-1 and VCAM-1 expression on endothelial cells. Cell Signal 2007; 19: 1238-1248.
98. Mackay F, et al. Tumor necrosis factor alpha (TNF-alpha)-induced cell adhesion
to human endothelial cells is under dominant control of one TNF receptor type,
TNF-R55. J Exp Med 1993; 177: 1277-1286.
99. Roebuck KA, Finnegan A. Regulation of intercellular adhesion molecule-1
(CD54) gene expression. J Leukocyte Biol 1999; 66: 876-888.
100. Collins T, et al. Transcriptional regulation of endothelial cell adhesion molecules: NF-kappa B and cytokine-inducible enhancers. FASEB J 1995; 9:
899-909.
101. Ledebur HC, Parks TP. Transcriptional regulation of the intercellular adhesion
molecule-1 gene by inflammatory cytokines in human endothelial cells. Essential roles of a variant NF-kappa B site and p65 homodimers. J Biol Chem 1995;
270: 933-943.
102. Shi JX, et al. MK2 posttranscriptionally regulates TNF-alpha-induced expression of ICAM-1 and IL-8 via tristetraprolin in human pulmonary microvascular endothelial cells. Am J Physiol Lung Cell Mol Physiol 2012; 302:
L793-799.
103. Su X, et al. Post-transcriptional regulation of TNF-induced expression of
ICAM-1 and IL-8 in human lung microvascular endothelial cells: an obligatory
role for the p38 MAPK-MK2 pathway dissociated with HSP27. Biochim Biophys Acta 2008; 1783: 1623-1631.
104. Allingham MJ, et al. ICAM-1-mediated, Src- and Pyk2-dependent vascular endothelial cadherin tyrosine phosphorylation is required for leukocyte transendothelial migration. J Immunol 2007; 179: 4053-4064.
105. Wojciak-Stothard B, et al. Monocyte adhesion and spreading on human endothelial cells is dependent on Rho-regulated receptor clustering. J Cell Biol
1999; 145: 1293-1307.
106. Barreiro O, et al. Dynamic interaction of VCAM-1 and ICAM-1 with moesin
and ezrin in a novel endothelial docking structure for adherent leukocytes. J
Cell Biol 2002; 157: 1233-1245.
107. Clark PR, et al. Increased ICAM-1 expression causes endothelial cell leakiness,
cytoskeletal reorganisation and junctional alterations. J Invest Dermatol 2007;
127: 762-774.
108. Fehon RG, et al. Organising the cell cortex: the role of ERM proteins. Nat Rev
Mol Cell Biol 2010; 11: 276-287.
109. Oh HM, et al. RKIKK motif in the intracellular domain is critical for spatial
and dynamic organisation of ICAM-1: functional implication for the leukocyte
adhesion and transmigration. Mol Biol Cell 2007; 18:, 2322-2335.
110. Koss M, et al. Ezrin/radixin/moesin proteins are phosphorylated by TNF-alpha
and modulate permeability increases in human pulmonary microvascular endothelial cells. J Immunol 2006; 176: 1218-1227.
111. Xiao X, et al. Intercellular adhesion molecule-1 is a regulator of blood-testis
barrier function. J Cell Sci 2012; 125: 5677-5689.
112. Sumagin R, et al. Leukocyte rolling and adhesion both contribute to regulation
of microvascular permeability to albumin via ligation of ICAM-1. Am J Physiol
Cell Physiol 2011; 301: C804-813.
113. Sumagin R, et al. Leukocyte-endothelial cell interactions are linked to vascular
permeability via ICAM-1-mediated signalling. Am J Physiol Heart Circ Physiol
2008; 295: H969-H977.
114. Thompson PW, et al. Intercellular adhesion molecule (ICAM)-1, but not
ICAM-2, activates RhoA and stimulates c-fos and rhoA transcription in endothelial cells. J Immunol 2002; 169: 1007-1013.
115. Turowski P, et al. Phosphorylation of vascular endothelial cadherin controls
lymphocyte emigration. J Cell Sci 2008; 121: 29-37.
116. van Rijssel J, et al. The Rho-guanine nucleotide exchange factor Trio controls
leukocyte transendothelial migration by promoting docking structure
formation. Mol Biol Cell 2012; 23: 2831-2844.
117. Durieu-Trautmann O, et al. Intercellular adhesion molecule 1 activation induces tyrosine phosphorylation of the cytoskeleton-associated protein cortactin in brain microvessel endothelial cells. J Biol Chem 1994; 269: 12536-12540.
118. Wolf SI, et al. Agonistic anti-ICAM-1 antibodies in scleroderma: activation of
endothelial pro-inflammatory cascades. Vasc Pharmacol 2013; 59: 19-26.
© Schattauer 2014
Thrombosis and Haemostasis 112.6/2014
Downloaded from www.thrombosis-online.com on 2017-06-17 | IP: 88.99.165.207
For personal or educational use only. No other uses without permission. All rights reserved.
1100
1101
Marcos-Ramiro, García-Weber et al. Adhesion receptors and ROS in TNF-induced endothelial permeability
119. Etienne-Manneville S, et al. ICAM-1-coupled cytoskeletal rearrangements and
transendothelial lymphocyte migration involve intracellular calcium signalling
in brain endothelial cell lines. J Immunol 2000; 165: 3375-3383.
120. Gong P, et al. TLR4 signalling is coupled to SRC family kinase activation, tyrosine phosphorylation of zonula adherens proteins, and opening of the paracellular pathway in human lung microvascular endothelia. J Biol Chem 2008; 283:
13437-13449.
121. Wallez Y. et al. Src kinase phosphorylates vascular endothelial-cadherin in response to vascular endothelial growth factor: identification of tyrosine 685 as
the unique target site. Oncogene 2007; 26: 1067-1077.
122. Tilghman RW, Hoover RL. The Src-cortactin pathway is required for clustering
of E-selectin and ICAM-1 in endothelial cells. Faseb J 2002; 16: 1257-1259.
123. Wessel F, et al. Leukocyte extravasation and vascular permeability are each
controlled in vivo by different tyrosine residues of VE-cadherin. Nature Immunol 2014; 15: 223-230.
124. Etienne S, et al. ICAM-1 signalling pathways associated with Rho activation in
microvascular brain endothelial cells. J Immunol 1998; 161: 5755-5761.
125. Adamson P, et al. Lymphocyte migration through brain endothelial cell monolayers involves signalling through endothelial ICAM-1 via a rho-dependent
pathway. J Immunol 1999; 162: 2964-2973.
126. Carman CV, Springer TA. A transmigratory cup in leukocyte diapedesis both
through individual vascular endothelial cells and between them. J Cell Biol
2004; 167: 377-388.
127. Minshall RD, et al. Endothelial cell-surface gp60 activates vesicle formation
and trafficking via G(i)-coupled Src kinase signalling pathway. J Cell Biol 2000;
150: 1057-1070.
128. Hu G, et al. Intercellular adhesion molecule-1-dependent neutrophil adhesion
to endothelial cells induces caveolae-mediated pulmonary vascular hyperpermeability. Circulation Res 2008; 102: e120-131.
129. Sun Y, et al. Phosphorylation of caveolin-1 regulates oxidant-induced pulmonary vascular permeability via paracellular and transcellular pathways. Circulation Res 2009; 105: 676-685.
130. Neish AS, et al. Functional analysis of the human vascular cell adhesion molecule 1 promoter. J Exp Med 1992; 176: 1583-1593.
131. Iademarco MF, et al. Characterisation of the promoter for vascular cell adhesion molecule-1 (VCAM-1). J Biol Chem 1992; 267: 16323-16329.
132. van Wetering S, et al. VCAM-1-mediated Rac signalling controls endothelial
cell-cell contacts and leukocyte transmigration. Am J Physiol Cell Physiol
2003; 285: C343-352.
133. van Buul JD, et al. Migration of human hematopoietic progenitor cells across
bone marrow endothelium is regulated by vascular endothelial cadherin. J Immunol 2002; 168: 588-596.
134. van Buul JD, et al. ICAM-1 clustering on endothelial cells recruits VCAM-1. J
Biomed Biotechnol 2010; 2010: 120328.
135. Barreiro O, et al. Endothelial adhesion receptors are recruited to adherent leukocytes by inclusion in preformed tetraspanin nanoplatforms. J Cell Biol 2008;
183: 527-542.
136. van Buul JD, et al. RhoG regulates endothelial apical cup assembly downstream
from ICAM1 engagement and is involved in leukocyte trans-endothelial migration. J Cell Biol 2007; 178: 1279-1293.
137. van Wetering S, et al. Reactive oxygen species mediate Rac-induced loss of cellcell adhesion in primary human endothelial cells. J Cell Sci 2002; 115:
1837-1846.
138. Vockel M, Vestweber D. How T cells trigger the dissociation of the endothelial
receptor phosphatase VE-PTP from VE-cadherin. Blood 2013; 122: 2512-2522.
139. Nawroth R, et al. VE-PTP and VE-cadherin ectodomains interact to facilitate
regulation of phosphorylation and cell contacts. Embo J 2002; 21: 4885-4895.
140. Wolburg H, et al. Diapedesis of mononuclear cells across cerebral venules during experimental autoimmune encephalomyelitis leaves tight junctions intact.
Acta Neuropathol 2005; 109: 181-190.
141. Page-McCaw A, et al. Matrix metalloproteinases and the regulation of tissue
remodelling. Nature Rev Mol Cell Biol 2007; 8: 221-233.
142. Deem TL, Cook-Mills JM. Vascular cell adhesion molecule 1 (VCAM-1) activation of endothelial cell matrix metalloproteinases: role of reactive oxygen
species. Blood 2004; 104: 2385-2393.
143. Moll T, et al. In vitro degradation of endothelial catenins by a neutrophil protease. J Cell Biol 1998; 140: 403-407.
144. Dhawan S, et al. HIV-1 infection alters monocyte interactions with human
microvascular endothelial cells. J Immunol 1995; 154: 422-432.
145. Cook-Mills JM, et al. Vascular cell adhesion molecule-1 expression and signalling during disease: regulation by reactive oxygen species and antioxidants.
Antiox Redox Signal 2011; 15: 1607-1638.
146. Lessner SM, Galis ZS. Matrix metalloproteinases and vascular endotheliummononuclear cell close encounters. Trends Cardiovasc Med 2004; 14: 105-111.
147. Sidibe A, et al. Soluble VE-cadherin in rheumatoid arthritis patients correlates
with disease activity: evidence for tumor necrosis factor alpha-induced VEcadherin cleavage. Arthritis Reum 2012; 64: 77-87.
148. Weber C, et al. The role of junctional adhesion molecules in vascular inflammation. Nat Rev Immunol 2007; 7: 467-477.
149. Ebnet K, et al. Regulation of epithelial and endothelial junctions by PAR proteins. Front Biosci 2008; 13: 6520-6536.
150. Ebnet K, et al. Junctional adhesion molecule interacts with the PDZ domaincontaining proteins AF-6 and ZO-1. J Biol Chem 2000; 275: 27979-27988.
151. Bazzoni G, et al. Interaction of junctional adhesion molecule with the tight
junction components ZO-1, cingulin, and occludin. J Biol Chem 2000; 275:
20520-20526.
152. Ebnet K, et al. The junctional adhesion molecule (JAM) family members
JAM-2 and JAM-3 associate with the cell polarity protein PAR-3: a possible role
for JAMs in endothelial cell polarity. J Cell Sci 2003; 116: 3879-3891.
153. Itoh M, et al. Junctional adhesion molecule (JAM) binds to PAR-3: a possible
mechanism for the recruitment of PAR-3 to tight junctions. J Cell Biol 2001;
154: 491-497.
154. Ostermann G, et al. JAM-1 is a ligand of the beta(2) integrin LFA-1 involved in
transendothelial migration of leukocytes. Nat Immunol 2002; 3: 151-158.
155. Santoso S, et al. The junctional adhesion molecule 3 (JAM-3) on human platelets is a counterreceptor for the leukocyte integrin Mac-1. J Exp Med 2002;
196: 679-691.
156. Cunningham SA, et al. JAM2 interacts with alpha4beta1. Facilitation by JAM3.
J Biol Chem 2002; 277: 27589-27592.
157. Arrate MP, et al. Cloning of human junctional adhesion molecule 3 (JAM3)
and its identification as the JAM2 counter-receptor. J Biol Chem 2001; 276:
45826-45832.
158. Vetrano S, et al. Unique role of junctional adhesion molecule-a in maintaining
mucosal homeostasis in inflammatory bowel disease. Gastroenterology 2008;
135: 173-184.
159. Mandell KJ, et al. Involvement of the junctional adhesion molecule-1 (JAM1)
homodimer interface in regulation of epithelial barrier function. J Biol Chem
2004; 279: 16254-16262.
160. Liu Y, et al. Human junction adhesion molecule regulates tight junction resealing in epithelia. J Cell Sci 2000; 113: 2363-2374.
161. Li X, et al. JAM-C induces endothelial cell permeability through its association
and regulation of {beta}3 integrins. Arterioscl Thromb Vasc Biol 2009; 29:
1200-1206.
162. Orlova VV, et al. Junctional adhesion molecule-C regulates vascular endothelial permeability by modulating VE-cadherin-mediated cell-cell contacts. J Exp
Med 2006; 203: 2703-2714.
163. Sircar M, et al. Neutrophil transmigration under shear flow conditions in vitro
is junctional adhesion molecule-C independent. J Immunol 2007; 178:
5879-5887.
164. Ozaki H, et al. Cutting edge: combined treatment of TNF-alpha and IFNgamma causes redistribution of junctional adhesion molecule in human endothelial cells. J Immunol 1999; 163: 553-557.
165. Ueki T, et al. Expression of junctional adhesion molecules on the human lymphatic endothelium. Microvasc Res 2008; 75: 269-278.
166. Shaw SK, et al. Reduced expression of junctional adhesion molecule and platelet/endothelial cell adhesion molecule-1 (CD31) at human vascular endothelial junctions by cytokines tumor necrosis factor-alpha plus interferon-gamma
Does not reduce leukocyte transmigration under flow. Am J Pathol 2001; 159:
2281-2291.
167. Liang TW, et al. Vascular endothelial-junctional adhesion molecule (VE-JAM)/
JAM 2 interacts with T, NK, and dendritic cells through JAM 3. J Immunol
2002; 168: 1618-1626.
168. Mamdouh Z, et al. Targeted recycling of PECAM from endothelial surfaceconnected compartments during diapedesis. Nature 2003; 421: 748-753.
169. Biswas P, et al. PECAM-1 affects GSK-3beta-mediated beta-catenin phosphorylation and degradation. Am J Pathol 2006; 169: 314-324.
Thrombosis and Haemostasis 112.6/2014
© Schattauer 2014
Downloaded from www.thrombosis-online.com on 2017-06-17 | IP: 88.99.165.207
For personal or educational use only. No other uses without permission. All rights reserved.
Marcos-Ramiro, García-Weber et al. Adhesion receptors and ROS in TNF-induced endothelial permeability
170. Graesser D, et al. Altered vascular permeability and early onset of experimental
autoimmune encephalomyelitis in PECAM-1-deficient mice. J Clin Invest
2002; 109: 383-392.
171. Privratsky JR, et al. Relative contribution of PECAM-1 adhesion and signalling
to the maintenance of vascular integrity. J Cell Sci 2011; 124:, 1477-1485.
172. Cao MY, et al. Regulation of mouse PECAM-1 tyrosine phosphorylation by the
Src and Csk families of protein-tyrosine kinases. J Biol Chem 1998; 273:
15765-15772.
173. Lu TT, et al. Platelet endothelial cell adhesion molecule-1 is phosphorylatable
by c-Src, binds Src-Src homology 2 domain, and exhibits immunoreceptor tyrosine-based activation motif-like properties. J Biol Chem 1997; 272:
14442-14446.
174. Kogata N, et al. Identification of Fer tyrosine kinase localized on microtubules
as a platelet endothelial cell adhesion molecule-1 phosphorylating kinase in
vascular endothelial cells. Mol Biol Cell 2003; 14: 3553-3564.
175. Bird IN, et al. Homophilic PECAM-1(CD31) interactions prevent endothelial
cell apoptosis but do not support cell spreading or migration. J Cell Sci 1999;
112: 1989-1997.
176. Ilan N, et al. PECAM-1 (CD31) functions as a reservoir for and a modulator of
tyrosine-phosphorylated beta-catenin. J Cell Sci 1999; 112: 3005-3014.
177. Ilan N, et al. Platelet-endothelial cell adhesion molecule-1 (CD31), a scaffolding molecule for selected catenin family members whose binding is mediated
by different tyrosine and serine/threonine phosphorylation. J Biol Chem 2000;
275: 21435-21443.
178. Ukropec JA, et al. SHP2 association with VE-cadherin complexes in human endothelial cells is regulated by thrombin. J Biol Chem 2000; 275: 5983-5986.
179. Timmerman I, et al. The tyrosine phosphatase SHP2 regulates recovery of endothelial adherens junctions through control of beta-catenin phosphorylation.
Mol Biol Cell 2012; 23: 4212-4225.
180. Grinnell KL, et al. Role of protein tyrosine phosphatase SHP2 in barrier function of pulmonary endothelium. Am J Physiol. Lung Cell Mol Physiol 2010;
298: L361-370.
181. Henshall TL, et al. Src homology 2 domain-containing protein-tyrosine phosphatases, SHP-1 and SHP-2, are required for platelet endothelial cell adhesion
molecule-1/CD31-mediated inhibitory signalling. J Immunol 2001; 166:
3098-3106.
182. Mamdouh Z, et al. Transcellular migration of leukocytes is mediated by the endothelial lateral border recycling compartment. J Exp Med 2009; 206:
2795-2808.
183. You M, et al. Modulation of the nuclear factor kappa B pathway by Shp-2 tyrosine phosphatase in mediating the induction of interleukin (IL)-6 by IL-1 or
tumor necrosis factor. J Exp Med 2001; 101-110.
184. Daigle I, et al. Death receptors bind SHP-1 and block cytokine-induced antiapoptotic signalling in neutrophils. Nature Med 2002; 8: 61-67.
185. Lerner-Marmarosh N, et al. Inhibition of tumor necrosis factor-[alpha]-induced SHP-2 phosphatase activity by shear stress: a mechanism to reduce endothelial inflammation. Arterioscl Thromb Vasc Biol 2003; 23: 1775-1781.
186. Koch E, et al. The endothelial tyrosine phosphatase SHP-1 plays an important
role for vascular haemostasis in TNFalpha -induced inflammation in vivo.
Mediators Inflamm 2013; 2013: 279781.
187. Nakagami H, et al. Tumor necrosis factor-alpha inhibits growth factor-mediated cell proliferation through SHP-1 activation in endothelial cells. Arterioscl Thromb Vasc Biol 2002; 22: 238-242.
188. le Duc Q, et al. Vinculin potentiates E-cadherin mechanosensing and is recruited to actin-anchored sites within adherens junctions in a myosin II-dependent manner. J Cell Biol 2010; 189: 1107-1115.
© Schattauer 2014
Thrombosis and Haemostasis 112.6/2014
Downloaded from www.thrombosis-online.com on 2017-06-17 | IP: 88.99.165.207
For personal or educational use only. No other uses without permission. All rights reserved.
1102