Gold Nanoparticles Increase Endothelial Paracellular Permeability by Altering Components of Endothelial Tight Junctions, and Increase Blood-Brain Barrier Permeability in Mice
Gold Nanoparticles Increase Endothelial Paracellular Permeability by Altering Components of Endothelial Tight Junctions, and Increase Blood-Brain Barrier Permeability in Mice
Ching-Hao Li 0 4
Ming-Kwang Shyu 1
Cheng Jhan 3
Yu-Wen Cheng 2
Chi-Hao Tsai 3
Chen-Wei Liu 3
jj Ruei-Ming Chen 0
Jaw-Jou Kang 3
0 Graduate Institute of Medical Sciences, College of Medicine, Taipei Medical University , Taipei , Taiwan
1 Department of Obstetrics and Gynecology, National Taiwan University Hospital , Taipei , Taiwan
2 School of Pharmacy, Taipei Medicine University , Taipei , Taiwan
3 Institute of Toxicology, College of Medicine, National Taiwan University , Taipei , Taiwan
4 Department of Physiology, School of Medicine, College of Medicine, Taipei Medical University , Taipei , Taiwan
Gold nanoparticles (Au-NPs) are being increasingly used as constituents in cosmetics, biosensors, bioimaging, photothermal therapy, and targeted drug delivery. This elevated exposure to Au-NPs poses systemic risks in humans, particularly risks associated with the biodistribution of Au-NPs and their potent interaction with biological barriers. We treated human umbilical vein endothelial cells with Au-NPs and comprehensively examined the expression levels of tight junction (TJ) proteins such as occludin, claudin-5, junctional adhesion molecules, and zonula occludens-1 (ZO-1), as well as endothelial paracellular permeability and the intracellular signaling required for TJ organization. Moreover, we validated the effects of Au-NPs on the integrity of TJs in mouse brain microvascular endothelial cells in vitro and obtained direct evidence of their influence on blood-brain barrier (BBB) permeability in vivo. Treatment with Au-NPs caused a pronounced reduction of PKCf-dependent threonine phosphorylation of occludin and ZO-1, which resulted in the instability of endothelial TJs and led to proteasome-mediated degradation of TJ components. This impairment in the assembly of TJs between endothelial cells increased the permeability of the transendothelial paracellular passage and the BBB. Au-NPs increased endothelial paracellular permeability in vitro and elevated BBB permeability in vivo. Future studies must investigate the direct and indirect toxicity caused by Au-NP-induced endothelial TJ opening and thereby address the double-edged-sword effect of Au-NPs.
endothelial barrier; paracellular permeability; tight junction; protein kinase C zeta (PKCf); blood-brain barrier
AJ, adherens junction;
Au-NPs, gold nanoparticles;
Au-MPs, gold microparticles;
BBB, blood-brain barrier;
BMECs, brain microvascular endothelial cells;
HUVECs, human umbilical vein endothelial cells;
JAMs, junctional adhesion molecules;
MMP, matrix metalloproteinase;
PKC, protein kinase C;
TEER, trans-endothelial electrical resistance;
TEM, transmission electron microscopy;
TJ, tight junction;
VE-cadherin, vascular endothelial cadherin;
ZO-1, zonula occludens-1.
Tight junctions (TJs) are junctions formed between 2 adjacent
cells that are responsible for regulating the permeation of polar
solutes through the intercellular cleft (the so-called paracellular
permeability or barrier function)
. TJs are present
in both the epithelium and the endothelium. Whereas TJs are
concentrated on the apical side of epithelial cells, in endothelial
cells, the localization of TJs is intermingled with that of other
junctions. Moreover, TJ organization or tightness varies
considerably along the vascular tree: it is well developed in
arteries/arterioles and loosely elaborated in capillary venules. An
exception is the cerebral capillary, where TJs contribute to the
blood-brain barrier (BBB): TJs of brain microvascular endothelial
cells (BMECs) are tighter than those of peripheral microvessels
(Bazzoni and Dejana, 2004)
. The brain in all vertebrates contains
a BBB composed mainly of endothelial cells, astrocytes (as well
as their basement membranes), and pericytes
(Abbott et al.,
2006; Engelhardt and Sorokin, 2009)
. These ?neurovascular
units? are considered to constitute the physical barrier of the
BBB, which restricts paracellular diffusion of water-soluble
substances from blood to brain and maintains brain homeostasis
(Abbott, 2013). Current studies on the BBB widely use in vitro BBB
models prepared using nonBBB endothelium-originated cells
[eg, human umbilical vein endothelial cells (HUVECs), which
retain TJ-expressing phenotypes and paracellular permeability
(Beese et al., 2010; Liu et al., 2009)
Structurally, TJs are formed by numerous transmembrane
proteins [eg, occludin, claudins, and junctional adhesion molecules
(JAMs)], which seal the paracellular cleft by locally dimerizing with
each other. The cytoplasmic domains of these transmembrane
proteins bind to adaptor proteins such as zonula occludens-1
(ZO-1), and these adaptors mediate the anchorage of the
membranous TJ components to actin microfilaments and thereby stabilize
the TJ structure
(Bazzoni and Dejana, 2004)
. Consequently, TJ
disassembly results not only from a reduction in the expression of
transmembrane TJ proteins, but also from a dislocation of ZO-1
from the cell border
(Chen et al., 2009; Elias et al., 2009)
and this can
be reflected by a decrease in the trans-endothelial electrical
resistance (TEER) or an increase in paracellular permeability
(McLaughlin et al., 2004). Besides TJs, the ubiquitously distributed
adherens junctions (AJs) also mediate cell adhesion and
paracellular permeability. The transmembrane protein vascular endothelial
cadherin (VE-cadherin) is the most important component of
endothelial AJs. VE-cadherin binds through its cytoplasmic tail to
b-catenin, which in turn links to a-catenin and anchors the AJ
complex to the cytoskeleton. VE-cadherin overexpression reduces
cell proliferation and paracellular permeability, but enhances cell
adhesion and migration
(Bazzoni and Dejana, 2004; Harris and
Nelson, 2010; Taddei et al., 2008)
The integrity of TJs is regulated by various signaling factors,
among which the protein kinase C (PKC) family, a group of
membrane-associated protein kinases, is considered to be the
most common regulator
(Rao, 2009; Stamatovic et al., 2006;
Willis et al., 2010)
. At least 12 PKC isozymes have been identified
to date, and these can be classified into 3 categories (classical,
novel, and atypical) based on their cofactor requirements
et al., 2010)
. In endothelial cells, inflammatory stimuli
frequently induce a large increase in PKCa activity, which results
in a loss of the ability of TJs to form a tight barrier
(He et al.,
. Conversely, PKCf is continuously active under basal
resting conditions at least in certain cell types, and PKCf is the only
PKC isoform located at the cell border
(Mayanglambam et al.,
2011; Newton, 2010)
. Furthermore, PKCf has been shown to be a
part of the ZO complex, where it is the main kinase that
phosphorylates occludin on threonine or tyrosine/serine residues,
which determines the assembly or disassembly process of TJs,
(Dodane and Kachar, 1996; Jain et al., 2011; Zyrek
et al., 2007)
. Thus, the effects of PKCf inhibition on TJ
organization remain debated.
Gold (Au) in its bulk form is inert, weakly toxic, and
biocompatible with various tissues, and gold compounds have long
been used for medical purposes. Thus, gold nanoparticles
(AuNPs) are also considered to be highly suitable for application in,
for example, cosmetics, biosensors, bioimaging, photothermal
therapy, and targeted drug delivery
(Khlebtsov and Dykmana,
2011; Panyala et al., 2009)
. However, the toxicity of several
nontoxic or weakly toxic bulk materials increases when the sizes of
these materials are reduced to the nanoscale. This increase in
toxicity might result from the unique toxicokinetic profiles of
nanoparticles, as well as from the ability of the particles to
overcome cell barriers. Given the continued increase in the
applications of Au-NPs, concerns regarding human safety have
received progressively more attention, but how Au-NPs affect
BBB permeability has remained poorly studied. Thus, in this
study, our aim was to elucidate the interactions of Au-NPs with
endothelial cells and barriers composed of endothelial cells. We
determined that Au-NP exposure reduced PKCf
phosphorylation and PKCf-dependent threonine phosphorylation on ZO-1
and occludin, and impaired ZO-1/occludin interaction, and that
this was followed by an enhancement of TJ protein degradation
and paracellular permeability.
MATERIALS AND METHODS
Chemicals. An uncoated Au-NP aqueous suspension (10 mg/ml)
was purchased from LiHo Chem Inc. (Taoyuan, Taiwan). Gold
microparticles (Au-MPs) (Sigma-Aldrich, St. Louis, MO; 326585)
were chosen as a reference control. All chemicals, unless
otherwise indicated, were obtained from Sigma-Aldrich. The
chemical stocks of MG-132 and Go6983 were prepared in dimethyl
sulfoxide (DMSO). For solvent control, the maximal final
concentration of DMSO in the medium was 2% (in vitro study), and
this did not adversely affect cell viability and paracellular
Au-NP characterization. The morphology, particle size, and
agglomeration of the gold particles were measured by using
transmission electron microscopy (TEM). A drop of the
nanoparticle suspension was allowed to air-dry onto a
Formvar-carboncoated 200 mesh copper grid, and then TEM images were
acquired on a JEM-2100 microscope operating at 120 kV (JEOL
Ltd., Tokyo, Japan). Approximately, 200 particles were counted
and measured for size distribution (AnalySIS software, Soft
Imaging Systems, Megaview III, Munster, Germany). The
particle size distribution of gold particles was measured in PBS
(137 mM NaCl, 2.7 mM KCl, 10 mM Na2HPO4, 1.8 mM KH2PO4, pH
7.4) and cell culture medium by dynamic light scattering (DLS)
method (Zetasizer 300, Malvern, UK). Furthermore, the zeta
potential of the gold particles was measured by laser Doppler
velocimetry at a constant temperature of 25 C in the same
device. The absorption spectra of gold particles were recorded
by using an ultraviolet-visible spectrophotometer (OPTIZEN;
Mecasys Co., Ltd., Korea) at the wavelength range 400?800 nm,
and sterile water was used as a standard for background
Primary endothelial cell (HUVEC and BMEC) isolation, culture, and
treatment. Primary vascular endothelial cells derived from
distinct donors were isolated from human umbilical cords by
means of 0.1% collagenase digestion (for detailed information,
please see the Supplementary Information). Primary cultures
were grown in a humidified 5%-CO2 atmosphere at 37 C in an
incubator, and passages between 3 and 6 were used in the
experiments. Culture medium was refreshed every 2 days.
Confluent monolayers of HUVECs were treated with gold
particles at indicated periods, and subjected to TJ protein (or
mRNA) analysis. For paracellular permeability and TEER assays,
postconfluent cells were incubated for 2?3 more days after cells
had reached confluence.
BMEC cultures were prepared by a modification of
et al.?s method (1983
) (for detailed information, please see the
Supplementary Information). Cell culture protocol for human
brain malignant glioma (GBM 8401) and human colonic
adenocarcinoma cell lines (Caco-2 and HCT-116) were also addressed
in the Supplementary Information.
Determination of cell viability: MTT assay. Cell viability was
evaluated based on the reduction of
3-(4,5-dimethyl-2-thiazolyl)2,5-diphenyl-2H-tetrazolium bromide (MTT).
Immunofluorescent staining of TJ markers. HUVECs were grown on
glass cover slips and treated as indicated for various
experiments. After treatment, the cells were fixed using 4%
paraformaldehyde, permeabilized with 0.1% Triton X-100 for 5 min at
room temperature, and incubated with blocking serum for
30 min. Next, the cells were incubated overnight (4 C) with a
primary ZO-1 antibody (1:100; Invitrogen, Camarillo, California;
Catalog #40-2200) and then with a fluorescein isothiocyanate
(FITC)-conjugated secondary antibody (1:1000 dilution) at room
temperature for 1 h in the dark. Lastly, nuclei were
counterstained with Hoechst 33342 (5 mM, Molecular Probes, Eugene,
Oregon) for 5 min and the samples were washed thrice and then
mounted on slides. Fluorescent images were captured by using
a Nikon MICROPHOT-FXA camera equipped with Media
Cybernetics Image-Pro Express software.
Analysis of endothelial paracellular permeability in vitro by using
FITC-dextran. The in vitro barrier permeability was determined
by measuring the transendothelial passage of water-soluble
tracers from the apical to the basolateral compartment
et al., 2010; Liu et al., 2009)
. To measure paracellular
permeability, 1 105 HUVECs were seeded on fibronectin-pre-coated
Transwell inserts (pore size, 0.4 lm) (Corning, Cambridge,
Massachusetts) and cultivated for 2?4 days until a postconfluent
cell monolayer was formed. We treated the cultures with
Au-NPs for 24 h and then added 200 lg/ml FITC-coupled
dextrans (molecular mass, 40 or 70 kDa) into the inner chamber;
30 min later, we measured the concentrations of FITC-dextran
in the outer chamber at 485/535 nm (Abs/Em) by using a
Paradigm Multi-Mode Plate Reader (Beckman Coulter Inc., Brea,
Measurement of TEER. HUVECs were seeded at 1 105 cells in
0.3 cm2 polyethylene terephtalate membrane inserts with 0.4 lm
pore size (Falcon, Catalog No. 353095) and cultivated for 2?4 days
as described previously. After treatment with gold particles, the
TEER was measured using an EVOM Epithelial Voltohmmeter
(World Precision Instruments; Sarasota, Florida). The normalized
TEER values were corrected for background resistance of the
culture insert and medium, and expressed as X cm2.
Analysis of TJ protein expression and PKCf phosphorylation: Western
blotting. Confluent cells grown on 60-mm dishes were treated
with gold particles for 12?48 h (for TJ protein analysis) or
10?60 min (for PKCf phosphorylation analysis). Subsequently,
cells were washed twice with ice-cold PBS and scraped using 200
ml of RIPA buffer
(Li et al., 2012)
. The samples were centrifuged at
14 000 g for 20 min at 4 C, and the supernatants were collected
and used as total cell lysates. Protein concentrations were
determined using Bradford reagent, and the lysates were then mixed
fully with a 4 sample buffer; next, proteins (60 mg/lane) were
resolved by means of SDS-PAGE and electrotransferred onto
PVDF membranes (Millipore, Bedford, Massachusetts). The
membranes were blocked using TBST containing 5% skim milk (and, if
necessary, the membranes were trimmed), and then incubated
overnight at 4 C with primary antibodies against b-catenin,
JAM1, claudin-5, PKCa, phospho-PKCa, PKCf, phospho-PKCf
(Epitomics, Burlingame, California), VE-cadherin (Genetex,
Irvine, California), and occludin (Proteintech, Chicago, Illinois).
The membranes were next washed in TBST and then incubated
for 2 h at room temperature with horseradish
peroxidaseconjugated secondary IgGs (1:5000). Finally, blots were developed
using enhanced chemiluminescence (Millipore), and images
were captured and densitometrically analyzed using the
BioSpectrum AC Imaging system (UVP, Upland, California). All
values were normalized relative to the control protein. Some of
the blots were stripped by using a commercial stripping buffer
(Pierce, Rockford, Illinois) and immunoprobed according to the
Characterizing protein-protein interaction in vitro:
co-immunoprecipitation. Whole-cell lysates were pre-cleared by incubating them
briefly with protein A magnetic beads (Millipore), and the
collected lysates (1 mg) were used in immunoprecipitations: We
added 1 mg/ml of the capture antibodies
anti-phospho-threonine (anti-p-Thr) (Santa Cruz, Dallas, Texas; sc-5267),
antioccludin (Proteintech; Catalog No. 13 409), or anti-ZO-1
(Epitomics, S1303) and incubated the samples overnight with
rotation at 4 C. Next, the samples were mixed with 50 ml of
protein A magnetic beads and incubated with agitation for another
2 h. The captured immunocomplexes were precipitated, washed
3 times with RIPA buffer, and mixed with 100 ml of the elution
buffer (125 mM Tris, 0.2% SDS, 0.1% Tween-20). Protein samples
were electrophoresed on 10% denaturing gels and analyzed as
described in the preceding section.
Analysis of mRNA-expression data of junction-associated proteins:
reverse transcription polymerase chain reaction. Total RNA was
isolated by using the TriPure reagent (Roach, Mannheim,
Germany) according to the manufacturer?s protocol. Total RNA
quality and quantity were assessed using a NanoDrop ND-1000
Spectrophotometer (Thermo, Wilmington, Delaware), and cDNA
was synthesized from the total RNA (3 lg) by using M-MLV
reverse transcriptase (Epicentre, Madison, Wisconsin) and Oligo
(dT16?18) primers. Next, the cDNA (2 ml) was amplified using a
Taq DNA polymerase mixture with appropriate primer sets
(Supplementary Table 1) and the following amplification
protocol: denaturation at 95 C for 5 min, followed by 30?35 cycles of
denaturation at 95 C for 30 s, annealing at 60 C for 30 s, and
extension at 72 C for 30 s, and then a final extension at 72 C for
10 min. The amplification was assessed by separating samples
on 1.5% agarose gels and staining with ethidium bromide
et al., 2012)
. The signal intensity was analyzed using a Gel Logic
100 Imaging system (Kodak, Rochester, New York) and mRNA
levels were normalized relative to that of the b-actin control.
Animal husbandry. All animals were acclimated and housed in
the Animal Center of National Taiwan University (NTUAC)
(Taipei, Taiwan) according to its standard operation protocol.
Animals were handled in all experiments in accordance with
the Guide for the Care and Use of Laboratory Animals (National
Academy of Sciences Press, 1996) and the procedures were
approved by the Institutional Animal Care and Use Committee
(approval number: LAC-2013-0038).
BBB permeability assay in vivo: Evans blue extravasation. We
purchased 6-week-old male ICR mice from NTUAC. Before testing,
the mice were grouped (5?6 mice/group) and acclimated for
1 week in the housing room. Mice in the test groups were
administered Au-NPs (12.5 or 25 mg/kg b.w.) by means of a
single intravenous injection through the tail vein; mice in the
negative control groups received the vehicle (saline). The mortality
and clinical behavior of the animals were observed daily. There
were no changes of behaviors or of the numbers of moribund or
dead animals that were related to treatment with the test
materials. After the treatment (24 or 48 h), BBB permeability was
determined based on Evans blue extravasation
(Wang et al.,
. In brief, Evans blue dye (2% in saline) was injected
intravenously at a dose of 2 ml/kg, and 2 h after the injection, the
animals were killed and perfused with saline to remove
intravascular Evans blue dye. Next, the brain was harvested,
weighed, and immersed in formamide (10 ml/g) at 55?60 C for
24 h, and then the supernatant was collected and its Evans blue
absorbance (at 610 nm) was measured using a microplate reader
(MRX-TC; Dynex Technology, Chantilly, Virginia).
Statistical analysis. All data are expressed as the
mean 6 standard deviation (SD) from at least 3 independent
experiments (N ^ 3). The significance of the difference between
the control and each experimental test condition was analyzed
by referring to Student?s t tests. The significance of the
difference between the control and various groups and among the
groups (eg, in dose-response assays) was determined using
One-Way Analysis of Variance (ANOVA), followed by Duncan?s
new multiple range method. P < .05 was considered statistically
Characteristics of Gold Particles
TEM images of gold particles are shown in Figures 1A and 1B.
Quantitative measurements, shown as histograms, revealed
that the average particle sizes of Au-NPs and Au-MPs were
40 6 1 and 637 6 9 nm, respectively. Au-NPs were observed as
dispersed, unaggregated primary particles, and they showed a
peak absorbance (the so-called localized surface plasmon
resonance) between the wavelengths of 500 and 550 nm
(Supplementary Fig. 1A). By contrast, the Au-MPs
morphologically appeared as compacted crystals, and in most fields,
AuMPs formed aggregates, which resulted in an
absorbancespectrum shift to the far-red region (Supplementary Fig. 1B).
The particle-size distributions measured using the DLS method
are shown in Figures 1C and 1D. The average zeta potentials of
Au-NP and Au-MP suspensions were 12.2 6 0.7 and 21.0 6
3.7 mV, respectively (Supplementary Fig. 2).
Effects of Au-NPs on Cell Viability
HUVEC and GBM 8401 cultures were incubated with Au-NPs (or
Au-MPs) for 24 h and then their viability was measured. In the
Au-NP-treatment groups, at concentrations up to 100 lg/ml, the
viability levels (% relative to control) of HUVECs and GBM 8401
cells were 90.1 6 0.4 and 89.7 6 1.3, respectively, whereas cell
viability remained unaffected in cultures treated with Au-MPs
(Supplementary Table 2). The Au-NPs and Au-MPs used in this
study were neither genotoxic (in vivo micronucleus assay) nor
mutagenic (Ames test) (Supplementary Table 3).
Au-NPs Reduced TJ Protein Expression, Decreased Endothelial
TJ Integrity, and Increased Paracellular Permeability in Treated HUVECs
The paracellular permeability of HUVECs treated with Au-NPs
(or Au-MPs) was measured using FITC-labeled dextrans. As
compared with the tracer flux of the control group, no marked
changes in the flux were observed in Au-MP-treated HUVECs; by
contrast, in cells treated with Au-NPs (50 mg/ml), we measured
an increase in the flux of 40- and 70-kDa FITC-dextrans (Fig. 2A).
The normalized TEER of the HUVEC monolayer was reduced
from 34.3 6 1.5 to 23.7 6 1.8 X cm2 after Au-NP treatment (50 mg/
ml; n ? 5), whereas there was no significant reduction in the
TEER of the HUVEC monolayer exposed to Au-MPs (37.1 6 1.4
X cm2; n ? 4).
Next, to examine TJ integrity, we performed
immunofluorescence staining for ZO-1. Unlike in control cells, in cells treated
with H2O2 (500 mM), discontinuous or intermittent ZO-1-positive
strands displaying intercellular gaps were observed, which
suggested that the TJ structure was damaged. These intercellular
gaps were also clearly detected in HUVECs treated with Au-NPs
(50 mg/ml), but the continuity of the ZO-1-positive strands was
retained in HUVECs treated with Au-MPs (Fig. 2B).
We next performed Western blotting to analyze changes in
the levels of AJ and TJ proteins. In Au-MP-treated HUVECs, no
changes were detected in the levels of either AJ proteins
(VEcadherin and b-catenin) or TJ proteins (occludin, JAM-1, and
claudin-5). However, treatment with Au-NPs (10, 50, and 100 mg/
ml) markedly diminished the expression of occludin, JAM-1,
and claudin-5 in HUVECs, but not that of VE-cadherin and
bcatenin (Fig. 3A and Supplementary Fig. 3). The expression of
occludin, JAM-1, and claudin-5 began to decrease in a
statistically significant manner at 12 h after treatment, and this
reduction in expression persisted for at least 48 h (Fig. 3B). We
speculated that the Au-NP-induced increase in paracellular
permeability and impairment of TJ tightness might have resulted
from the reduction in TJ protein expression.
In contrast to Au-NP treatment, which caused a
downregulation of TJ proteins, treatment with aggregates of Au-NPs exerted
no effects on these proteins; this suggested that the
downregulation of TJ proteins was caused by nano-sized Au
primary particles (Supplementary Fig. 4).
Au-NP-Mediated TJ Protein Downregulation Was
Specific to Endothelial Cells
In addition to the permeability of the endothelial barrier, the
permeability of the epithelial barrier also depends on TJ
organization. However, we observed that Au-NPs induced no changes
in occludin and JAM-1 protein levels in human intestinal
epithelial cells (HCT-116 and Caco-2) or human brain malignant
glioma GBM 8401 cells (Supplementary Fig. 5). These data
revealed that Au-NP treatment downregulated TJ proteins
specifically in endothelial cells.
Au-NP-Mediated Degradation of TJ Proteins
Was Regulated Posttranscriptionally
The mRNAs levels of VE-cadherin, b-catenin, ZO-1, occludin,
JAM-1, and claudin-5 were not distinct in control HUVECs and
HUVECs treated with either Au-NPs or Au-MPs (Fig. 4A and
Supplementary Fig. 6). By contrast, in cells co-incubated with
the 26 S proteasome inhibitor MG-132 (10 mM) and Au-NPs for
4 h, Au-NP-mediated TJ protein degradation was completely
rescued (Fig. 4B), which suggested that Au-NP-induced degradation
of TJ proteins was regulated by the 26 S proteasome pathway.
Under certain pathological conditions, matrix
metalloproteinases (MMPs), especially MMP-9, have been shown to degrade
occludin and claudin-5 and play a critical role in BBB failure
(Chen et al., 2009)
. However, in this study, we detected no
marked changes in the mRNA levels of MMPs (MMP-1, -2, -9, -10,
-11, and -14) after treatment with Au-NPs (or Au-MPs).
Moreover, no clear changes were detected in the levels of either
pro-MMP-9 or cleaved MMP-9 (Supplementary Fig. 7). These
data excluded the involvement of MMPs in Au-NP-mediated TJ
Au-NPs Inhibited PKCf Kinase Maturation and
Accelerated TJ Protein Degradation
Members of the PKC family are widely recognized to regulate
the TJ barrier function. However, distinct PKC isoforms function
in different tissues and play diverse roles in regulating the
maintenance of cell junctions. All PKC isoforms undergo a
kinase maturation process requiring phosphorylation in their
catalytic domain, such as phosphorylation of Thr-497 in PKCa and
Thr-560 in PKCf
(Hirai and Chida, 2003)
. We found that in
AuNP-treated cells, the levels of the phosphorylated form of PKCf
(Thr-560) were decreased in both time- and
concentrationdependent manners (Fig. 5A and Supplementary Fig. 8A).
However, the levels of the phosphorylated/non-phosphorylated
forms of PKCa Src, FAK, and Rac1 remained unaffected by Au-NP
treatment (Supplementary Fig. 8B). Go? 6983, a broad-spectrum
PKC inhibitor, not only inhibited PKCf Thr-560 phosphorylation
(30-min treatment), but also downregulated the expression of TJ
proteins (24-h treatment). Moreover, following
pretreatment/coincubation with G o?6983, Au-NP-mediated TJ protein degradation
was enhanced (Fig. 5B), and immunofluorescence labeling
confirmed that the TJ structure at the cell borders was disassembled
(Supplementary Fig. 9). These data suggested the requirement of
PKCf activation for maintaining TJ integrity.
The results of complementary experiments showed that
arachidonic acid (50 mM), an activator of PKCf, induced PKCf
Thr-560 phosphorylation (30-min treatment), but did not affect
TJ protein expression (24-h treatment). However, following
pretreatment/co-incubation with arachidonic acid,
Au-NPmediated TJ protein degradation was partially rescued (Fig. 5C),
as was the integrity of TJs at the cell border (Supplementary Fig.
9). These data indicated that Au-NPs might induce TJ
disassembly in a PKCf-dependent manner and lead to the degradation of
Au-NP Treatment Reduced the Threonine Phosphorylation of Occludin
and ZO-1. Hyper-phosphorylation on Ser and Thr residues of
ZO-1 and occludin is necessary for occludin/ZO-1 association
. The results of immunoprecipitation experiments
showed that threonine phosphorylation of captured ZO-1 and
occludin was markedly lower after treatment with Au-NPs than
after treatment with Au-MPs. Moreover, Au-NP treatment
reduced the occludin/ZO-1 association in
co-immunoprecipitation assays (Figs. 6A and 6B), and the amounts of ZO-1 and
occludin immunoprecipitated by the p-Thr capture antibody
were clearly decreased after Au-NP treatment, but were
unaffected in the control and Au-MP-treatment groups (Fig. 6C).
Au-NPs Decreased TJ Integrity of BMEC Monolayers and Enhanced
BBB Permeability In Vivo. We validated the Au-NP-induced
reduction in endothelial TJ integrity by using isolated mouse BMECs.
Immunofluorescence staining showed that ZO-1 and occludin
were expressed mainly on the cell border and that the staining
appeared as continuous strands in the control and
Au-MPtreatment groups. However, after treatment with Au-NPs or
H2O2, these strands appeared discontinuous and the
fluorescence became scattered (Fig. 7A). Lastly, we examined the effect
of Au-NPs on BBB permeability by means of Evans blue
extravasation in vivo. In control mice, at 2 h after intravenous injection
of Evans blue, the dye was not detected in the brain. By contrast,
in mice that were administered Au-NPs (12.5 and 25 mg/kg b.w.,
i.v.), blue staining of the brain was evident at 24 and 48 h after
FIG. 6. Au-NP treatment reduced threonine phosphorylation on ZO-1 and occludin. Proteins were immunoprecipitated using anti-ZO-1, anti-occludin, or
antiphospho-threonine (p-Thr) antibodies, as described in ?Materials and Methods? section. Representative images from 4 independent experiments are shown (N ? 4); IgG
was used as the loading control. A, In HUVECs treated with Au-NPs (50 mg/ml; 12 h), threonine phosphorylation of captured ZO-1 and the associated occludin was
clearly lower than the corresponding phosphorylation levels in the control sample. However, Au-MP treatment did not affect the phosphorylation of these proteins. B,
When co-immunoprecipitation assays were performed using the anti-occludin capture antibody, the threonine phosphorylation of captured occludin and the
associated ZO-1 was once again found to be decreased following Au-NP treatment; no changes were observed in the control and Au-MP-treatment groups. C, Lastly, in
immunoprecipitates obtained using the anti-p-Thr capture antibody, the levels of ZO-1 and occludin were diminished following Au-NP treatment, whereas they were
unaffected by Au-MP treatment.
dosing. Quantification of the data showed that the Evans blue
content was increased in a statistically significant manner at
24 h after administration of the 2 doses of Au-NPs (Fig. 7B).
Adjoining cells closely adhere to each other through 3
intercellular junctions, AJs, TJs, and gap junctions. Specifically, AJs and TJs
are responsible for regulating paracellular permeability, which is
tightly controlled in the endothelium. Molecules such as
vascular endothelial growth factor (VEGF)
(Gavard and Gutkind, 2006)
(Xu et al., 2012)
, and actin-depolymerizing agents
et al., 2010)
, which disturb the stability of the
VE-cadherin/bcatenin complex or AJs, typically damage the endothelial barrier
(Harris and Nelson, 2010)
. The presence of AJs is also required for
stabilizing TJ formation
(Taddei et al., 2008)
. In this study, we
found that Au-NP treatment increased endothelial barrier
permeability in vitro, but it did not affect the expression of
VEcadherin and b-catenin; this suggests that paracellular
permeability is more likely regulated by TJs than by AJs.
TJ complexes contain a wide spectrum of proteins, and
among these, occludin, claudins, and JAMs are integral
membrane proteins. One or more claudins, which share high
sequence identity but are differentially expressed in various
tissues, constitute the core of the TJ complexes. Ochratoxin A
(McLaughlin et al., 2004)
, human chorionic gonadotropin
(Rodewald et al., 2009)
(Argaw et al., 2009)
(Pinton et al., 2009)
, which cause the removal of specific
claudins from TJs, have been shown to allow claudin pore
formation and ion permeability. Claudin-5 is expressed specifically
by endothelial cells
(Gonc?alves et al., 2013)
, and an in vitro study
showed that overexpressed claudin-5 became concentrated at
endothelial cell borders, where it interacted with ZO-1 and
ensured tight contact between cells
(Umeda et al., 2006)
. In the
vascular tree, the levels of occludin expression were reported to
determine endothelial barrier permeability
(Hirase et al., 1997)
Like claudins, occludin binds directly to ZO-1 through its
Cterminal domain, and disruption of the occludin/ZO-1
interaction (especially using an occludin mutant containing a
C-terminal deletion) was shown to cause the removal of occludin from
(Balda et al., 1996)
. In the case of JAMs, humans express 3
isoforms, and the isoform JAM-1, which is universally expressed in
the epithelium and the endothelium, also functionally interacts
with ZO-1 and contributes to TJ (and BBB) tightness
et al., 2008)
. In this study, Au-NP treatment diminished the
integrity of TJs and the expression of TJ proteins, which
suggests that endothelial paracellular permeability is increased as
a result of Au-NP-mediated TJ protein degradation and TJ
TJ integrity can be regulated by various factors such as
inflammatory cytokines, growth factors, oxidative stress,
, and nanoparticles
(Brun et al., 2014;
Trickler et al., 2014)
. Treatment with Cu-NPs, Ag-NPs, and
TiO2NPs has been reported to increase the generation of free radicals
(Sarkar et al., 2014) and the release of inflammatory cytokines
(Marano et al., 2011)
and to increase paracellular permeability
(Brun et al., 2014; Trickler et al., 2014)
. Both free radicals and
inflammatory cytokines are widely recognized to induce TJ
damage, and thus the nanoparticles might impair TJ structure
and function by elevating these active factors. However,
because no available evidence suggests a correlation between
Au-NP exposure and the formation of free radicals or
inflammatory cytokines (Trickler et al., 2011), a novel mechanism might
be involved in Au-NP-mediated TJ regulation.
As mentioned earlier in this section, the interaction between
ZO-1 and occludin (and also JAM-1 and claudin-5) helps stabilize
the TJ structure. Increasing evidence indicates that the
phosphorylation of occludin and ZO-1 can regulate TJ assembly or
(Bazzoni and Dejana, 2004)
. This regulation
has been shown to depend mainly on the kinases/phosphatases
involved and the residues/positions phosphorylated
For example, occludin phosphorylation on Thr-424/438
(by PKCf) appears to be required for TJ assembly
(Jain et al.,
. However, phosphorylation on Tyr and Ser residues
typically attenuates occludin/ZO-1 complex formation
et al., 2009; Murakami et al., 2012; Raleigh et al., 2011; Yamamoto
et al., 2008)
. In this study, treatment of endothelial cells with
Au-NPs caused a reduction in the threonine phosphorylation of
ZO-1 and occludin and diminished the ZO-1/occludin
Most previous studies indicate that PKCs play a central role
in TJ regulation and exert both stimulatory and inhibitory
effects. In this study, PKCa did not respond to Au-NPs, but PKCf
was inactivated after Au-NP treatment. Stimuli such as TNF-a
(Aveleira et al., 2010)
(Titchenell et al., 2012)
(Omri et al., 2013)
, and hypoxia
(Willis et al., 2010)
induce the hyper-activation of PKCf, subsequently compromise
TJ complexes. Inhibition of PKCf hyper-activation restores not
only the integrity of TJs, but also their barrier functions.
Conversely, PKCf can also serve as a direct substrate for
phosphoinositide-dependent protein kinase 1 and PI-3 kinase
signaling. Moreover, PKCf is recruited to the membrane and is
constitutively active in the absence of agonist stimulation
. Membrane-recruited PKCf directly interacts
with occludin, induces the phosphorylation of occludin on
Thr424/438 residues and positively regulates TJ assembly
et al., 1993)
. In this study, co-incubation of cells with Au-NPs
and a PKCf inhibitor (Go? 6983) or activator (arachidonic acid)
exacerbated or partly alleviated Au-NP-mediated TJ protein
degradation, respectively. These data suggested that a basal level of
PKCf activation was crucial for endothelial TJ assembly, and
that PKCf inhibition caused by Au-NP treatment led to a
disruption of barrier function.
Three barriers exist between the blood and the brain: the
BBB, the blood-cerebrospinal fluid barrier (BCSFB), and the
arachnoid barrier. In the arachnoid barrier, a multi-layered
epithelium containing TJs is present, whereas in the BCSFB,
capillary endothelial cells are fenestrated, and TJs are formed
between epithelial cells at the apical surface (the CSF-facing
surface). Thus, the BCSFB and the arachnoid barrier contain
epithelial TJs, rather than endothelial TJs
(Abbott, 2013; Engelhardt
and Sorokin, 2009)
. Our data suggested that Au-NP-mediated
downregulation of TJ proteins occurred in endothelial cells, but
not in epithelial cells. Based on the results of both in vitro and
in vivo assays, we conclude that Au-NP-induced Evans dye
diffusion in brain might result from endothelial TJ damage and BBB
In this study, we first determined that Au-NPs increased
endothelial paracellular permeability in vitro and enhanced BBB
permeability in vivo. Our results also provided an enhanced
understanding of the mechanism of action of Au-NPs: the
nanoparticles inactivated PKCf, suppressed threonine
phosphorylation on occludin and ZO-1, and perturbed occludin/ZO-1
complex formation, and thus caused TJ disassembly and TJ
protein degradation. These results demonstrated that Au-NPs,
which hold great potential for use in biomedicine, can
substantially increase endothelial permeability and thereby enhance
the transport of other molecules, xenobiotics, or pathogens into
the brain. This double-edged-sword effect of Au-NPs must
continue to receive attention and must be investigated further in
Supplementary data are available online at http://toxsci.
This study was financed in part by the grants
TMU101-AE1B46 (funded by Taipei Medical University),
DOH102-FDA31103 (funded by the Food and Drug Administration,
Ministry of Health and Welfare, Taiwan), and MOST
1032320-B-038-033 (funded by the Ministry of Science and
The authors have no conflicts of interest to declare.
Abbott , N. J. ( 2013 ). Blood-brain barrier structure and function and the challenges for CNS drug delivery . J. Inherit. Metab. Dis . 36 , 437 - 449 .
Abbott , N. J. , Ro? nnba? ck, L., and Hansson , E. ( 2006 ). Astrocyteendothelial interactions at the blood-brain barrier . Nat. Rev. Neurosci. 7 , 41 - 53 .
Argaw , A. T. , Gurfein , B. T. , Zhang, Y. , Zameer , A. , and John , G. R. ( 2009 ). VEGF-mediated disruption of endothelial CLN-5 promotes blood-brain barrier breakdown . Proc. Natl Acad. Sci. U. S. A. 106 , 1977 - 1982 .
Aveleira , C. A. , Lin , C. M. , Abcouwer , S. F. , Ambro? sio, A. F. , and Antonetti , D. A. ( 2010 ). TNF-a signals through PKCf/NF-jB to alter the tight junction complex and increase retinal endothelial cell permeability . Diabetes 59 , 2872 - 2882 .
Balda , M. S. , Gonzalez-Mariscal , L. , Matter , K. , Cereijido , M. , and Anderson , J. M. ( 1993 ). Assembly of the tight junction: the role of diacylglycerol . J. Cell Biol . 123 , 293 - 302 .
Balda , M. S. , Whitney , J. A. , Flores , C. , Gonzalez , S. , Cereijido , M. , and Matter , K. ( 1996 ). Functional dissociation of paracellular permeability and transepithelial electrical resistance and disruption of the apical-basolateral intramembrane diffusion barrier by expression of a mutant tight junction membrane protein . J. Cell Biol . 134 , 1031 - 1049 .
Bazzoni , G. , and Dejana , E. ( 2004 ). Endothelial cell-to-cell junctions: molecular organization and role in vascular homeostasis . Physiol. Rev . 84 , 869 - 901 .
Beese , M. , Wyss , K. , Haubitz , M. , and Kirsch , T. ( 2010 ). Effect of cAMP derivates on assembly and maintenance of tight junctions in human umbilical vein endothelial cells . BMC Cell Biol . 11 , 68 .
Bowman , P. D. , Ennis , S. R. , Rarey , K. E. , Betz , A. L. , and Goldstein , G. W. ( 1983 ). Brain microvessel endothelial cells in tissue culture: a model for study of blood-brain barrier permeability . Ann. Neurol. 14 , 396 - 402 .
Brun , E. , Barreau , F. , Veronesi , G. , Fayard , B. , Sorieul , S. , Chane? ac, C. , Carapito , C. , Rabilloud , T. , Mabondzo , A. , Herlin-Boime , N. , et al. ( 2014 ). Titanium dioxide nanoparticle impact and translocation through ex vivo, in vivo and in vitro gut epithelia . Part. Fibre Toxicol . 11 , 13 .
Chen , F. , Ohashi , N. , Li , W. , Eckman , C. , and Nguyen , J. H. ( 2009 ). Disruptions of occludin and claudin-5 in brain endothelial cells in vitro and in brains of mice with acute liver failure . Hepatology 50 , 1914 - 1923 .
Dodane , V. , and Kachar , B. ( 1996 ). Identification of isoforms of G proteins and PKC that colocalize with tight junctions . J. Membr. Biol . 149 , 199 - 209 .
Elias , B. C. , Suzuki , T. , Seth , A. , Giorgianni , F. , Kale , G. , Shen , L. , Turner , J. R. , Naren , A. , Desiderio , D. M. , and Rao , R. ( 2009 ). Phosphorylation of Tyr-398 and Tyr-402 in occludin prevents its interaction with ZO-1 and destabilizes its assembly at the tight junctions . J. Biol. Chem . 284 , 1559 - 1569 .
Engelhardt , B. , and Sorokin , L. ( 2009 ). The blood-brain and the blood-cerebrospinal fluid barriers: function and dysfunction . Semin. Immunopathol . 31 , 497 - 511 .
Gavard , J. , and Gutkind , J. S. ( 2006 ). VEGF controls endothelialcell permeability by promoting the beta-arrestin-dependent endocytosis of VE-cadherin . Nat. Cell Biol . 8 , 1223 - 1234 .
Gonc?alves, A. , Ambro? sio, A. F. , and Fernandes , R. ( 2013 ). Regulation of claudins in blood-tissue barriers under physiological and pathological states . Tissue Barriers 1 , e24782 .
Harris , E. S. , and Nelson , W. J. ( 2010 ). VE-cadherin: at the front, center, and sides of endothelial cell organization and function . Curr. Opin. Cell. Biol . 22 , 651 - 658 .
He , F. , Yin , F. , Omran , A. , Yang , L. F. , Xiang , Q. L. , and Peng , J. ( 2012 ). PKC and RhoA signals cross-talk in Escherichia coli endotoxin induced alterations in brain endothelial permeability . Biochem. Biophys. Res. Commun . 425 , 182 - 188 .
Hirai , T. , and Chida , K. ( 2003 ). Protein kinase Czeta (PKCzeta): activation mechanisms and cellular functions . J. Biochem. 133 , 1 - 7 .
Hirase , T. , Staddon , J. M. , Saitou , M. , Ando-Akatsuka , Y. , Itoh , M. , Furuse , M. , Fujimoto , K. , Tsukita , S. , and Rubin , L. L. ( 1997 ). Occludin as a possible determinant of tight junction permeability in endothelial cells . J. Cell Sci . 110 , 1603 - 1613 .
Jain , S. , Suzuki , T. , Seth , A. , Samak , G. , and Rao , R. ( 2011 ). PKCf phosphorylates occludin and promotes assembly of epithelial tight junctions . Biochem. J . 437 , 289 - 299 .
Khlebtsov , N. , and Dykmana , L. ( 2011 ). Biodistribution and toxicity of engineered gold nanoparticles: a review of in vitro and in vivo studies . Chem. Soc. Rev . 40 , 1647 - 1671 .
Li , C. H. , Liao , P. L. , Shyu , M. K. , Liu , C. W. , Kao , C. C. , Huang , S. H., Cheng, Y. W., and Kang , J. J. ( 2012 ). Zinc oxide nanoparticlesinduced intercellular adhesion molecule 1 expression requires Rac1/Cdc42, mixed lineage kinase 3, and c-Jun Nterminal kinase activation in endothelial cells . Toxicol. Sci . 126 , 162 - 172 .
Liu , P. , Woda , M. , Ennis , F. A. , and Libraty , D. H. ( 2009 ). Dengue virus infection differentially regulates endothelial barrier function over time through type I interferon effects . J. Infect. Dis . 200 , 191 - 201 .
Marano , F. , Hussain , S. , Rodrigues-Lima , F. , Baeza-Squiban , A. , and Boland , S. ( 2011 ). Nanoparticles: molecular targets and cell signalling . Arch. Toxicol . 85 , 733 - 741 .
Mayanglambam , A. , Bhavanasi , D. , Vijayan , K. V. , and Kunapuli , S. P. ( 2011 ). Differential dephosphorylation of the protein kinase C-zeta (PKCf) in an integrin aIIbb3- dependent manner in platelets . Biochem. Pharmacol . 82 , 505 - 513 .
McLaughlin , J. , Padfield , P. J. , Burt , J. P. H. , and O'Neill , C. A. ( 2004 ). Ochratoxin A increases permeability through tight junctions by removal of specific claudin isoforms . Am. J. Physiol. Cell Physiol . 287 , C1412 - C1417 .
Millan , J. , Cain , R. J. , Reglero-Real , N. , Bigarella , C. , MarcosRamiro , B. , Fernandez-Martin , L. , Correas , I. , and Ridley , A. J. ( 2010 ). Adherens junctions connect stress fibres between adjacent endothelial cells . BMC Biol . 8 , 11 .
Murakami , T. , Frey , T. , Lin , C. , and Antonetti , D. A. ( 2012 ). Protein kinase Cb phosphorylates occludin regulating tight junction trafficking in vascular endothelial growth factor-induced permeability in vivo . Diabetes 61 , 1573 - 1583 .
Newton , A. C. ( 2010 ). Protein kinase C: poised to signal . Am. J. Physiol. Endocrinol. Metab . 298 , E395 - E402 .
Omri , S. , Behar-Cohen , F. , Rothschild , P. R. , Ge? lize? , E., Jonet , L. , Jeanny , J. C. , Omri , B. , and Crisantim P. ( 2013 ). PKCf mediates breakdown of outer blood-retinal barriers in diabetic retinopathy . PLoS One 8 , e81600 .
Panyala , N. R. , Pen? a -Me? ndez, E. M. , and Havel , J. ( 2009 ). Gold and nano-gold in medicine: overview, toxicology and perspectives . J. Appl. Biomed . 7 , 75 - 91 .
Pinton , P. , Nougayre` de, J. P. , Del Rio , J. C. , Moreno , C. , Marin , D. E. , Ferrier , L. , Bracarense , A. P. , Kolf-Clauw , M. , and Oswald , I. P. ( 2009 ). The food contaminant deoxynivalenol, decreases intestinal barrier permeability and reduces claudin expression . Toxicol. Appl. Pharmacol . 237 , 41 - 48 .
Raleigh , D. R. , Boe , D. M. , Yu , D. , Weber , C. R. , Marchiando , A. M. , Bradford , E. M. , Wang , Y. , Wu , L. , Schneeberger , E. E. , Shen , L. , et al. ( 2011 ). Occludin S408 phosphorylation regulates tight junction protein interactions and barrier function . J. Cell Biol . 193 , 565 - 582 .
Rao , R. ( 2008 ). Oxidative stress-induced disruption of epithelial and endothelial tight junctions . Front. Biosci . 13 , 7210 - 7226 .
Rao , R. ( 2009 ). Occludin phosphorylation in regulation of epithelial tight junctions . Ann. N. Y. Acad. Sci . 1165 , 62 - 68 .
Rodewald , M. , Herr , D. , Duncan , W. C. , Fraser , H. M. , Hack , G. , Konrad , R. , Gagsteiger , F. , Kreienberg , R. , and Wulff , C. ( 2009 ). Molecular mechanisms of ovarian hyperstimulation syndrome: paracrine reduction of endothelial claudin 5 by hCG in vitro is associated with increased endothelial permeability . Hum. Reprod . 24 , 1191 - 1199 .
Sarkar , A. , Ghosh , M. , and Sil , P. C. ( 2014 ). Nanotoxicity: oxidative stress mediated toxicity of metal and metal oxide nanoparticles . J. Nanosci. Nanotechnol . 14 , 730 - 743 .
Stamatovic , S. M. , Dimitrijevic , O. B. , Keep , R. F. , and Andjelkovic , A. V. ( 2006 ) . Protein kinase Calpha-RhoA cross-talk in CCL2- induced alterations in brain endothelial permeability . J. Biol. Chem . 281 , 8379 - 8388 .
Taddei , A. , Giampietro , C. , Conti , A. , Orsenigo , F. , Breviario , F. , Pirazzoli , V. , Potente , M. , Daly , C. , Dimmeler , S. , and Dejana , E. ( 2008 ). Endothelial adherens junctions control tight junctions by VE-cadherin-mediated upregulation of claudin-5 . Nat. Cell Biol . 10 , 923 - 934 .
Titchenell , P. M. , Lin , C. M. , Keil , J. M. , Sundstrom , J. M. , Smith , C. D. , and Antonetti , D. A. ( 2012 ). Novel atypical PKC inhibitors prevent vascular endothelial growth factor-induced bloodretinal barrier dysfunction . Biochem. J . 446 , 455 - 467 .
Trickler , W. J. , Lantz , S. M. , Murdock , R. C. , Schrand , A. M. , Robinson , B. L. , Newport , G. D. , Schlager , J. J. , Oldenburg , S. J. , Paule , M. G. , Slikker , W. , Jr. , et al. ( 2011 ). Brain microvessel endothelial cells responses to gold nanoparticles: In vitro proinflammatory mediators and permeability . Nanotoxicology 5 , 479 - 492 .
Trickler , W. J. , Lantz-McPeak , S. M. , Robinson , B. L. , Paule , M. G. , Slikker , W. Jr., Biris , A. S. , Schlager , J. J. , Hussain , S. M. , Kanungo , J. , et al. ( 2014 ). Porcine brain microvessel endothelial cells show pro-inflammatory response to the size and composition of metallic nanoparticles . Drug Metab. Rev . 46 , 224 - 231 .
Umeda , K. , Ikenouchi , J. , Katahira-Tayama , S. , Furuse , K. , Sasaki , H. , Nakayama , M. , Matsui , T. , Tsukita , S. , and Furuse , M. ( 2006 ). ZO-1 and ZO-2 independently determine where claudins are polymerized in tight-junction strand formation . Cell 126 , 741 - 754 .
Wang , P. , Liu , Y. , Shang , X. , and Xue , Y. ( 2011 ). CRM197-induced blood-brain barrier permeability increase is mediated by upregulation of caveolin-1 protein . J. Mol. Neurosci . 43 , 485 - 492 .
Willis , C. L. , Meske , D. S. , and Davis , T. P. ( 2010 ). Protein kinase C activation modulates reversible increase in cortical bloodbrain barrier permeability and tight junction protein expression during hypoxia and posthypoxic reoxygenation . J. Cereb. Blood Flow Metab . 30 , 1847 - 1859 .
Xu , M. , Chen , G. , Fu , W. , Liao , M. , Frank , J. A. , Bower , K. A. , Fang , S. , Zhang , Z. , Shi , X. , and Luo , J. ( 2012 ). Ethanol disrupts vascular endothelial barrier: implication in cancer metastasis . Toxicol. Sci . 127 , 42 - 53 .
Yamamoto , M. , Ramirez , S. H. , Sato , S. , Kiyota , T. , Cerny , R. L. , Kaibuchi , K. , Persidsky , Y. , and Ikezu , T. ( 2008 ). Phosphorylation of claudin-5 and occludin by rho kinase in brain endothelial cells . Am. J. Pathol . 172 , 521 - 533 .
Yeung , D. , Manias , J. L. , Stewart , D. J. , and Nag , S. ( 2008 ). Decreased junctional adhesion molecule-A expression during blood-brain barrier breakdown . Acta Neuropathol . 115 , 635 - 642 .
Zyrek , A. A. , Cichon , C. , Helms , S. , Enders , C. , Sonnenborn , U. , and Schmidt , M. A. ( 2007 ). Molecular mechanisms underlying the probiotic effects of Escherichia coli Nissle 1917 involve ZO2 and PKCzeta redistribution resulting in tight junction and epithelial barrier repair . Cell Microbiol . 9 , 804 - 816 .