Reactive Oxygen Species-Caspase-3 Relationship inMediating Blood–Brain Barrier Endothelial CellHyperpermeability Following Oxygen–GlucoseDeprivation and Reoxygenation
HIMAKARNIKA ALLURI, HAYDEN W. STAGG, RICKESHA L. WILSON, ROBERT P. CLAYTON, DEVENDRA A.
SAWANT, MADHAVI KONERU, MADHAVA R. BEERAM, MATTHEW L. DAVIS, AND BINU THARAKAN
Departments of Surgery and Pediatrics, Texas A&M University Health Science Center College of Medicine and Scott & White Healthcare, Temple,
Texas, USA
Address for correspondence: Binu Tharakan, Ph.D., F.A.H.A, Department of Surgery, Scott and White Healthcare & Texas A&M University Health
Science Center College of Medicine, 702 S.W. H.K. Dodgen Loop, Temple, TX 76504, USA. E-mail: [email protected]
Received 30 August 2013; accepted 21 December 2013.
ABSTRACT
Objective: Microvascular hyperpermeability that occurs due to
breakdown of the BBB is a major contributor of brain vasogenic
edema, following IR injury. In microvascular endothelial cells,
increased ROS formation leads to caspase-3 activation following IR
injury. The specific mechanisms, by which ROS mediates micro-
vascular hyperpermeability following IR, are not clearly known. We
utilized an OGD-R in vitro model of IR injury to study this.
Methods: RBMEC were subjected to OGD-R in presence of a
caspase-3 inhibitor Z-DEVD, caspase-3 siRNA or an ROS inhibitor
L-AA. Cytochrome c levels were measured by ELISA and caspase-3
activity was measured fluorometrically. TJ integrity and cytoskeletal
assembly were studied using ZO-1 immunofluorescence and rho-
damine phalloidin staining for f-actin, respectively.
Results: OGD-R significantly increased monolayer permeability,
ROS formation, cytochrome c levels, and caspase-3 activity
(p < 0.05) and induced TJ disruption and actin stress fiber
formation. Z-DEVD, L-AA and caspase-3 siRNA significantly
attenuated OGD-R-induced hyperpermeability (p < 0.05) while
only L-AA decreased cytochrome c levels. Z-DEVD and L-AA
protected TJ integrity and actin cytoskeletal assembly.
Conclusions: These results suggest that OGD-R-induced hyper-
permeability is ROS and caspase-3 dependent and can be regulated
by their inhibitors.
KEY WORDS: vascular hyperpermeability, blood–brain barrier, endo-
thelial cells, ischemia reperfusion
Abbreviations used: AFC, 7-amino-4-trifluoromethylcoumarin;
BBB, blood–brain barrier; DCFDA, 2′,7′-dichlorofluorescein diac-
etate; EDTA, ethylenediaminetetraacetic acid; FITC, fluorescein
isothiocyanate; IR, ischemia reperfusion; L-AA, L-ascorbic acid;
MMP, matrix metalloproteinases; MPT, mitochondrial permeabil-
ity transition; OGD-R, oxygen–glucose deprivation and reoxygen-
ation; PBS, phosphate-buffered saline; RBMEC, rat brain
microvascular endothelial cells; ROS, reactive oxygen species;
SOD, superoxide dismutase; TJ, tight junction; TJP, tight junction
proteins; Z-DEVD, Z-D(OMe)-E(OMe)-V-D(OMe)-FMK; ZO-1,
zonula occludens-1.
Please cite this paper as: Alluri H, Stagg HW, Wilson RL, Clayton RP, Sawant DA, Koneru M, Beeram MR, Davis ML, Tharakan B. Reactive oxygen species-
caspase-3 relationship in mediating blood–brain barrier endothelial cell hyperpermeability following oxygen–glucose deprivation and reoxygenation.
Microcirculation 21: 187–195, 2014.
INTRODUCTION
Microvascular hyperpermeability, the excessive leakage of
fluid and proteins from small blood vessels into the
extravascular space is one of the major causes of vasogenic
brain edema and occurs in a variety of disease and clinical
conditions, the most important of which is IR injury [8]. IR
injury occurs as an early secondary injury leading to a
nonspecific, inflammatory response triggering mitochondrial
permeabilization and is associated with the production of
various free radicals also known as ROS [14,18,21,23,26]. IR
injury is one of the frequent causes of various debilitating
symptoms associated with stroke, cardiac diseases, peripheral
vascular disease, and traumatic brain and spinal cord injuries
[15,22,30,34,40]. A hallmark of IR injury is the rapid inflow
of blood following an acute/chronic occlusion of blood
vessels, triggering the formation of free radicals [9,13,29].
Compared to the peripheral blood vessels, the cerebral vessels
DOI:10.1111/micc.12110
Original Article
ª 2013 John Wiley & Sons Ltd 187
are more sensitive to blood flow changes, being composed of
high amount of unsaturated lipids and need high rate of
metabolism to occur, which renders them most vulnerable to
the reperfusion injury [25,33]. IR injury has been extensively
studied in various disease conditions but the intracellular
mechanisms that lead to IR-induced vascular hyperperme-
ability and cerebral edema are not clearly known. In this
study, using an established in vitro system of IR injury, that
created an environment of anoxia reperfusion, we studied
how oxygen and nutrient deprivation followed by reoxygen-
ation induces microvascular endothelial hyperpermeability
mediated by ROS and caspase-3.
Free radicals or ROS are highly reactive molecules with
short half-lives and the presence of the unpaired electrons
in their orbitals make them chemically unstable [6,31].
ROS includes superoxide, hydroxyl, nitric oxide, hydrogen
peroxide, and peroxynitrite radicals, formed mainly at the
level of the mitochondrial electron transport chain [8].
They trigger a variety of secondary injuries in the body,
such as lipid peroxidation of membranes (including the
mitochondrial membranes), inactivation of various pro-
teins, excitotoxicity, depolymerization of polysaccharides,
and nucleic acid disruption and inflammation [6,10,19]. In
the cell, mitochondria plays an important role in buffering
the free oxygen radicals by converting them into water in
the final step of the electron transport chain of oxidative
phosphorylation [12]. Pathogenic insult or IR trigger
increased ROS production, which causes redox imbalance
in the cell, outstripping the endogenous antioxidant
mechanisms like ascorbic acid, SOD, catalase, etc.
[16,21,24,34]. There is increasing evidence that ROS are
key mediators of endothelial barrier dysfunctions in the
body including in the BBB [7,27,31,43] but the mechanisms
by which they control microvascular permeability are not
clearly known.
BBB endothelial cell tight junctions are critical for main-
taining the homeostasis of brain and for regulating micro-
vascular permeability [2]. The tight junctions are composed
of transmembrane proteins such as claudins, occludins, and
junctional adhesion molecules and are connected to the actin
cytoskeleton through a scaffolding protein ZO-1 on the
cytoplasmic side [2–4,16,25]. The tight junctions are also one
of the primary targets for ROS and proteolytic enzymes such
as caspases and MMPs [5,9,15,16]. In this study, we
hypothesized that ROS formation following oxygen–glucosedeprivation/reoxygenation is a major inducer of mitochon-
drial cytochrome c release and subsequent caspase-3 mediated
BBB tight junction breakdown and hyperpermeability.
We mainly addressed the following questions in this
study: (i) Is there any involvement of mitochondrial
mechanisms that mediate caspase-3 induced breakdown of
the BBB resulting in microvascular hyperpermeability
following OGD-R? (ii) Is there ROS mediation in inducing
caspase-3 mediated breakdown of the BBB and associated
hyperpermeability? (iii) Is there an involvement of actin
cytoskeletal assembly in IR-induced barrier dysfunction and
permeability? (iv) Are endogenous antioxidants effective
against ROS mediated barrier dysfunction and hyperper-
meability?
MATERIALS AND METHODS
RBMECPrimary cultures of RBMECs derived from adult Sprague
Dawley rats were obtained from Cell Applications (San
Diego, CA, USA). RBMECs were initially grown on 0.05%
fibronectin-coated cell culture dishes, using rat brain endo-
thelial cell complete media in a cell culture incubator (95%
O2, 5% CO2 at 37°C). RBMECs were treated with 0.25%
trypsin-EDTA for cell detachment and were grown on
fibronectin-coated Transwell inserts, chamber slides or
regular dishes for experimental purposes. RBMEC passages
6–8 were chosen for all the experiments.
Chemicals and ReagentsRBMEC Culture Medium was purchased from Cell Appli-
cations, Inc. Dulbecco’s Modified Eagle Medium without
glucose was obtained from Life Technologies (Grand Island,
NY, USA). Transwell-24 well plates were obtained from
Corning, USA. 5% Fibronectin from bovine plasma, L-AA,
Nunc Lab Tek II-CC, 8-well glass chamber slides and FITC-
Dextran-10 were purchased from Sigma Aldrich (St. Louis,
MO, USA). Rabbit anti- ZO-1 and rhodamine phalloidin
were purchased from Life Technologies. QIA70 Caspase-3
activity fluorometric assay kit was bought from EMD
Millipore/Calbiochem (Billerica, MA, USA). DCFDA cellular
ROS detection assay kit was bought from Abcam, USA.
Caspase-3 siRNA and control siRNA were obtained from
Dharmacon/Thermoscientific (Pittsburgh, PA, USA). Pierce
BCA protein assay kit was bought from Thermo Scientific
(Waltham, MA, USA). Cytochrome c ELISA kit and Z-
DEVD-FMK (Z-D(OMe)-E(OMe)-V-D(OMe)-FMK) were
obtained from R&D Systems (Minneapolis, MN, USA).
OGD-R In Vitro ModelOGD-R was performed as described previously [40,42,43].
Briefly, RBMECs were grown in RBMEC complete growth
media as monolayers on Transwell inserts, chamber slides or
regular dishes in a cell culture incubator (95% O2, 5% CO2 at
37°C) as described above. Later, the cells were exposed to no
glucose DMEM (Life Technologies) and placed in a hypoxic/
anoxic chamber (95% N2 and 5% CO2 at 37°C) in a cell
culture incubator. This was followed by a reperfusion face by
exposure to normal incubation conditions (95% O2, 5% CO2
at 37°C) and regular RBMEC growth medium for one hour
and three hours, respectively.
H. Alluri et al.
188 ª 2013 John Wiley & Sons Ltd
Effect of OGD-R on Brain MicrovascularEndothelial Cell Monolayer PermeabilityRBMEC monolayers were grown on fibronectin-coated
Transwell inserts for 96 hours, followed by treatment with
Z-DEVD (10 lM) or L-AA (10 mM). Cells were later
exposed to oxygen–glucose deprivation (two hours) followed
by reoxygenation (one hour and three hours) as described
above. The following treatment groups were maintained:
Control (untreated group), OGD-R one hour, OGD-R three
hours, Z-DEVD + OGD-R one hour, Z-DEVD + OGD-R
three hours, L-AA + OGD-R one hour, L-AA + OGD-R
three hours, Z-DEVD, L-AA. FITC-dextran-10 (1 mg/mL)
probe was applied to the upper reservoir of the monolayer
plate for 30 minutes. Samples were collected from the lower
compartment and fluorescent intensity was measured (485-
and 520-nm, excitation and emission, respectively).The
change in permeability was calculated as a percentage of
the untreated control values. Each experimental group
consisted of five replicates.
Effect of Caspase-3 Knockdown and OGD-R onBrain Microvascular Endothelial Cell MonolayerPermeabilityRBMECs were grown on fibronectin-coated Transwell
membranes as monolayers for 96 hours. They were treated
with control siRNA or caspase-3 siRNA for 48 hours. This
was followed by exposure of caspase-3 siRNA group to OGD-
R (one hour and three hours) conditions. The following
treatment groups were maintained: Control (untreated),
OGD-R one hour, OGD-R three hours, Caspase-3 siR-
NA + OGD-R one hour, and Caspase-3 siRNA + OGD-R
three hours. FITC-dextran-10 (1 mg/mL) was applied in the
upper chamber for 30 minutes and the fluorescent intensity
was measured as described above. Each experimental group
consisted of five replicates.
Effect of OGD-R on BBB Tight Junction Integrityand Cytoskeletal AssemblyRBMECs were grown on chamber slides and pretreated with
Z-DEVD (10 lM; one hour) or L-AA (10 mM; one hour)
followed by exposure to OGD-R (one hour) as described
above. To study tight junction integrity, the cells were
processed for immunofluorescence localization of ZO-1. Cells
were fixed in 4% paraformaldehyde in PBS for 15 minutes and
permeabilized with 0.5% Triton X-100 in PBS for 15 minutes.
After blocking with 2% bovine serum albumin in PBS, cells
were incubated overnight with an anti-rabbit primary anti-
body against ZO-1(1:150), followed by incubation with an
FITC tagged anti-rabbit secondary antibody for one hour at
room temperature. The cells were washed and the slides
mounted using Vectashield antifade reagent containing DAPI
(Vector Laboratories, Burlingame, CA, USA). To study the
actin cytoskeleton, f-actin fibers were stained with rhodamine
phalloidin. For this purpose, cells were fixed and permeablized
as described above followed by exposure to rhodamine
Phalloidin (1:50) for 20 minutes, prior to mounting as above.
The cells were visualized and they were scanned at a single
optical plane with an Olympus Fluoview 300 confocal
microscope (Center Valley, PA, USA). Each experimental
group consisted of two replicates.
Effect of OGD-R on ROS FormationRBMECs were grown on 96 well plates for 96 hours and
pretreated with L-AA (10 mM; one hour) followed by
OGD-R (one hour and three hours) as described above. An
untreated control group and an L-AA treated group were
also maintained. ROS levels were determined using a
membrane-permeable dye DCFDA. DCFDA was applied to
the cells at 37°C for 30 minutes. Each experimental group
consisted of four replicates. ROS in the cells oxidize DCFDA,
yielding the fluorescent product -(-6)-chloromethyl-29,79-
dichlorodihydrofluorescein. The fluorescence intensity was
measured at an excitation of 485 nm and an emission of
535 nm.
Effect of OGD-R on Mitochondrial Release ofCytochrome cRBMECs were grown in culture dishes and treated with L-AA
(10 mM; one hour), followed by OGD-R (one hour and
three hours) as described above. An untreated group served
as the control group. Each experimental group consisted of
five replicates. Cell lysates were obtained by scraping the cells
in the presence of a lysis buffer (provided in the ELISA kit)
followed by centrifugation at 10,000 g for 60 minutes at 4°C.Supernatant was collected, protein concentration was esti-
mated using BCA method (Pierce, Rockford, IL, USA) and
were used to perform an ELISA (R&D Systems). Briefly, the
samples were added to a 96 well plate and treated with a
conjugate regent, transferred to a cytochrome c antibody-
coated microwell plate, and incubated at room temperature
for 60 minutes. The wells were washed and treated with a
substrate reagent and incubated for 30 minutes, followed by
addition of a stop solution. The optical density was read at
450 nm using a colorimetric plate reader. The concentration
of cytochrome c was calibrated from a standard curve.
Effect of OGD-R on Caspase-3 ActivityRBMECs were grown in culture dishes and treated with L-AA
(10 mM; one hour) alone or L-AA, followed by OGD-R (one
hour and three hours) as described above. An untreated
group served as the control group. Caspase-3 activity was
measured using a caspase-3 activity assay kit (EMD
Millipore/Calbiochem). The Z-DEVD substrate provided in
the kit was labeled with a fluorescent molecule, AFC. The cell
Blood-Brain Barrier and Hyperpermeability
ª 2013 John Wiley & Sons Ltd 189
lysates were obtained as described above using lysis buffer
provided in the assay kit. This was followed by protein
estimation and treatment with the substrate conjugate. Each
experimental group consisted of four replicates. The resulting
fluorescent intensity was measured in a fluorescent plate
reader capable of measuring excitation at 400 nm and
emission at 505 nm.
Statistical AnalysisData are expressed as the mean � SEM (%). Statistical
differences among groups were determined by one-way
ANOVA followed by Bonferroni post hoc test to determine
significant differences between specific groups. These proce-
dures were followed for analysis of results from monolayer
permeability, ROS, cytochrome c and caspase-3 studies. A
value of p < 0.05 was considered a statistically significant
difference.
RESULTS
OGD-R Induces BBB Endothelial Cell MonolayerHyperpermeabilityOGD-R (one hour and three hours) induced significant
increase in permeability of the monolayer, evidenced by
increased FITC-dextran-10 leakage to the lower chamber
(p < 0.05; Figure 1). Pretreatment with caspase-3 inhibitor,
Z-DEVD (10 lM; one hour) or L-AA (10 mM; one hour)
attenuated this effect (p < 0.05; Figure 1). This shows that
caspase-3 and ROS play important roles in promoting BBB
hyperpermeability. Caspase-3 siRNA-transfected monolayers
showed significant decrease in permeability following expo-
sure to OGD-R compared to OGD-R alone group (p < 0.05;
Figure 2).
OGD-R Induces Disruption of BBB Endothelial CellTight JunctionsThe immunofluorescence study demonstrated junctional
continuity of ZO-1, whereas cells exposed to OGD-R showed
discontinuity of junctional staining demonstrating the loss of
barrier integrity. Pretreatment with Z-DEVD (10 lM; one
hour) and L-AA (10 mM; one hour) decreased OGD-R-
induced tight junction disruption (Figure 3A). This study
demonstrates that ROS and caspase-3 promotes monolayer
permeability by targeting the tight junctions.
Rhodamine phalloidin staining for f-actin in normal control
cells demonstrated clear cytoskeletal assembly with no stress
fiber formation. Cells exposed to OGD-R showed an increase
in stress fiber formation. Pretreatment ofOGD-R exposed cells
with Z-DEVD (10 lM; one hour) or L-AA (10 mM; one hour)
showed decrease in f-actin stress fiber formation (Figure 3B).
OGD-R Induces ROS Formation in BBB EndothelialCellsRBMECs exposed to OGD-R (one hour and three hours)
resulted in a significant increase in ROS formation evidenced
by increased fluorescent intensity (p < 0.05). Pretreatment of
cells exposed to OGD-R with Z-DEVD showed no significant
change in ROS formation compared to OGD-R alone group.
Treatment of cells with L-AA (10 mM; one hour) prior to
OGD-R resulted in a significant decrease in ROS formation
compared to OGD-R alone group (Figure 4).
OGD-R Induces Mitochondrial Release ofCytochrome c in BBB Endothelial CellsOGD-R (one hour and three hours) induced significant
increase in mitochondrial release of cytochrome c evidenced
by significant increase in its cytosolic levels (p < 0.05).
0
50
100
150
200
FIT
C-d
extr
an fl
uore
scen
t int
ensi
ty(%
con
trol
)
Control OGD-R 1 hour OGD-R 3 hours Z-DEVD+OGD-R1 hour
Z-DEVD+OGD-R3 hours
L-AA+OGD-R1 hour
L-AA+OGD-R3 hours
*a
*c*b*b
*a
*c
Figure 1. Caspase-3 inhibitor (Z-DEVD) and L-AA attenuates OGD-R-induced RBMEC monolayer hyperpermeability. Monolayer permeability is expressed
as a percentage of the change in FITC-dextran-10 fluorescent intensity (30 minutes accumulation). ‘*’ indicates statistical significance (p < 0.05; n = 5);
‘a’ indicates significant change compared to control group; ‘b’ indicates significant change compared to OGD-R one hour; ‘c’ indicates significant change
compared to OGD-R three hours.
H. Alluri et al.
190 ª 2013 John Wiley & Sons Ltd
Pretreatment with L-AA (10 mM; one hour) attenuated
OGD-R-induced increase in cytosolic cytochrome c levels
significantly (p < 0.05; Figure 5A). Caspase-3 siRNA-
transfected monolayers showed significant decrease in
permeability following OGD-R compared to OGD-R alone
group (p < 0.05; Figure 2).
OGD-R Induces Caspase-3 Activity in BBBEndothelial CellsOGD-R (one hour and three hours) induced significant
increase in caspase-3 activity (p < 0.05). Pretreatment with
L-AA (10 mM; one hour) attenuated OGD-R-induced increase
in caspase-3 activity significantly (p < 0.05; Figure 5B).
0
50
100
150
200
FIT
C-d
extr
an fl
uore
scen
t int
ensi
ty(%
con
trol
)
Control OGD-R 1 hour OGD-R 3 hours si RNA+OGD-R1 hour
siRNA+OGD-R3 hours
*a*a
*c*b
Figure 2. OGD-R induced hyperpermeability is significantly decreased by caspase-3 siRNA in RBMEC monolayers. ‘*’ indicates statistical significance
(p < 0.05; n = 5); ‘a’ indicates significant change compared to control group; ‘b’ indicates significant change compared to OGD-R one hour; ‘c’ indicates
significant change compared to OGD-R three hours. ‘Control group’ indicates an untreated group.
Control OGD-R 1 hour Z-DEVD + OGD-R Z-DEVD L-AA+OGD-R
Control OGD-R Z-DEVD + OGD-R Z-DEVD L-AA+OGD-R
A
B
Figure 3. (A) Immunofluorescence localization of tight junction protein ZO-1 demonstrating disruption of the tight junctions following OGD-R in
RBMECs. Z-DEVD and L-AA provides protection against OGD-R-induced disruption of the tight junctions. Arrows indicate tight junction disruption. (B)
Rhodamine phalloidin staining for f-actin stress fiber formation demonstrating changes in cytoskeletal assembly following OGD-R in RBMECs. OGD-R
increased the formation of f-actin stress fibers. Z-DEVD and L-AA decreased the formation of f-actin stress fibers. Arrows indicate actin stress fiber
formation. Each experimental group consisted of two replicates.
Blood-Brain Barrier and Hyperpermeability
ª 2013 John Wiley & Sons Ltd 191
0
Contro
l
OGD-R 1
hour
OGD-R 3
hours
L-AA+OGD-R
-
1 hou
r
L-AA+OGD-R
-
3 hou
rsL-A
A
Z-DEVD
Z-DEVD+OGD-R
1 hou
r
Z-DEVD+OGD-R
3 hou
rs
20406080
100120140160180
RO
S Fo
rmat
ion
(% C
ontr
ol) *a *a
*b
*a *a
*c
Figure 4. L-AA attenuates OGD-R-induced ROS formation in RBMECs whereas Z-DEVD showed no significant effect. ‘*’ indicates statistical significance(p < 0.05; n = 4); ‘a’ indicates significant change compared to control group; ‘b’ indicates significant change compared to OGD-R one hour; ‘c’ indicates
significant change compared to OGD-R three hours.
0
10
20
30
40
Cyt
ochr
ome
c le
vels
(ng/
mL
)
*a
*c*b
*a
Control OGD-R1 hour
OGD-R3 hours
Z-DEVD+OGD-R1 hour
L-AA+OGD-R3 hours
0
50
100
150
200
*a
Cas
pase
-3 a
ctiv
ity (%
con
trol
)
Control OGD-R1 hour
OGD-R3 hours
L-AA+OGD-R1 hour
L-AA+OGD-R3 hours
L-AA
*b*a
*c
A
B
Figure 5. OGD-R induced mitochondrial release of cytochrome c and caspase-3 activation is decreased by L-AA in RBMECs. (A) OGD-R induced
significant increase in cytosolic cytochrome c levels and this effect was decreased by L-AA. (B) OGD-R induces significant increase in caspase-3 activity.
This effect was significantly reduced by L-AA pretreatment. ‘*’ indicates statistical significance (p < 0.05; n = 5); ‘a’ indicates significant change compared
to control group; ‘b’ indicates significant change compared to OGD-R one hour; ‘c’ indicates significant change compared to OGD-R three hours.
H. Alluri et al.
192 ª 2013 John Wiley & Sons Ltd
DISCUSSION
In this study, our primary focus was to investigate the
mechanistic relationship between ROS and caspase-3 and
their downstream effect on BBB integrity and microvas-
cular permeability following oxygen–glucose deprivation
and reoxygenation, an in vitro model of IR Injury. The
important findings of the study are (i) brain microvascular
endothelial cell monolayer hyperpermeability that occurs
following OGD-R, is promoted by mitochondria-mediated
activation of caspase-3; (ii) mitochondria-mediated regu-
lation of microvascular hyperpermeability is associated
with tight junction disruption and cytoskeletal disorgani-
zation; (iii) ROS plays an important role in promoting
microvascular permeability via caspase-3 induced tight
junction disruption and cytoskeletal disorganization;
(iv) L-AA, an endogenous antioxidant has protective
functions against OGD-R-induced brain microvascular
hyperpermeability.
Previous studies have shown that mitochondrial oxidative
stress caused by ROS, after reperfusion injury promotes
mitochondrial release of cytochrome c [6]. Cytochrome c is a
highly conserved heme protein loosely attached to the inner
mitochondrial membrane and its release occurs depending
on the changes in MPT pore that is a key event in the
activation of caspase-3. MPT is known to alter in response to
oxidative stress induced by mitochondrial ROS [41]. The
exogenous ROS also induce MPT pore opening and
cytochrome c release in isolated mitochondria, which could
be blocked by drugs that protect mitochondrial transmem-
brane potential [36,37,41]. In this study, an increase in ROS
formation and mitochondrial release of cytochrome c and
their relationship to endothelial cell monolayer permeability
following OGD-R was observed. In addition to ROS
production by mitochondrial respiratory chain, ROS gener-
ation by cytoplasmic enzymes also has been reported in
different cell types [28,33]. However, this study does not
clearly discriminate between the two different sources of
ROS. Cytochrome c release and activation of caspase-3 are
the key events in the apoptotic signaling pathway. Produc-
tion of ROS during apoptosis can also be caspase dependent
[36]. Activation of caspase-3 as well as neuroprotection
following caspase-3 inhibition has been reported in cerebral
ischemia [11,29]. In this study OGD-R induced ROS
formation, mitochondrial release of cytochrome c, and
caspase-3 activation followed by BBB endothelial cell hyp-
erpermeability. L-AA a powerful antioxidant attenuated all
these effects. Thus, our studies demonstrate ROS formation
and activation of caspase-3 as major mechanisms that control
endothelial cell barrier dysfunction and hyperpermeability
following OGD-R.
OGD-R as an in vitromodel of ischemia reperfusion injury
has been well established [1,17,43]. However, the specific
effects of various combinations of oxygen and glucose levels
are not clearly known. We have used an anoxia-reperfusion
strategy using an oxygen–glucose deprivation and reoxygen-
ation system with complete deprivation of oxygen and
glucose followed by reoxygenation. For this purpose, we used
95% nitrogen/5% CO2 and a glucose-free medium as
described previously [1,17]. Anoxia reperfusion has been
previously known to generate ROS formation in vitro [1,43].
Also, the glucose-free medium used in this study does not
contain L-AA as a basic component. However, L-AA may be
present as a constituent of the fetal bovine serum. This model
may not be a true replication of human hypoxia in vivo, but
is a highly reproducible anoxia-reperfusion model to study
various signaling pathways involved in ischemia/anoxia-
reperfusion conditions.
To evaluate the integrity of BBB endothelial cells, we used
ZO-1 protein localization and visualization of actin cyto-
skeleton based on f-actin staining. Our studies demonstrated
tight junction (TJ) disruption evidenced by discontinuous
localization of ZO-1 and increased formation of f-actin stress
fibers. ZO-1 is a scaffolding protein that plays a major role in
determining the integrity of the BBB, as its carboxyl terminal
binds to the actin cytoskeleton, while its N-terminal binds to
the TJPs such as claudin-5 and occludin [16,20]. Further-
more, recent studies suggest that, ZO-1 and ZO-2 are
decreased following ischemic brain injury [7]. Our evalua-
tion of results from the immunofluorescence studies are
focused on tight junction integrity based on ZO-1 localiza-
tion along the cell-cell junctions. It is not clear if the
cytoplasmic staining observed in some of the groups (Z-
DEVD, Z-DEVD + OGD-R one hour, and L-AA + OGD-R
one hour) is definitely related to ZO-1 redistribution or due
to any nonspecific background staining. The precise regula-
tion of the structure and dynamics of the actin cytoskeleton
is essential for many developmental and physiological
processes including cell-cell adhesion. Actin stress fibers are
contractile actomyosin bundles found in many cultured
nonmuscle cells, where they play a major role in cell adhesion
and morphogenesis [38]. In several occasions, in endothelial
cells, an increase in actin stress fiber formation has been
associated with barrier dysfunction and hyperpermeability
[35,37].
L-AA, commonly known as vitamin C, is an endogenously
occurring, small molecular weight, water soluble vitamin
[39]. Our studies demonstrated that L-AA effectively inhib-
ited the release of mitochondrial cytochrome c, a major step
to the regulation of caspase-3 activation. It has been shown
that vitamins such as vitamin E, C and A possess antioxidant
properties and play an important role in inhibiting ROS
production [22]. While our studies support the antioxidant
properties of L-AA, our results also demonstrate how L-AA
regulates barrier integrity and permeability by targeting
mitochondrial release of cytochrome c and subsequent
Blood-Brain Barrier and Hyperpermeability
ª 2013 John Wiley & Sons Ltd 193
caspase-3 activation. We did not study the effect of caspase-3
inhibition using Z-DEVD on cytochrome c levels, as caspase-
3 activation is a downstream event to mitochondrial
cytochrome c release. This was further supported by the
observation that Z-DEVD had no effect on OGD-R induced
ROS formation.
In conclusion, the present findings suggest that OGD-R
induces ROS formation and cytochrome c release leading to
caspase-3-mediated disruption of BBB tight junctions and
microvascular hyperpermeability. Furthermore, our studies
demonstrate that the endogenous antioxidant L-AA has
protective benefits against barrier disruption and hyperper-
meability by preserving tight junction integrity as well as the
actin cytoskeletal assembly. Our results thus provide a
mechanistic relationship between ROS formation, caspase-3
activation, and BBB endothelial hyperpermeability following
OGD-R.
PERSPECTIVE
Blood-Brain Barrier (BBB) dysfunction is a major contri-
butor to the vasogenic brain edema that occurs, following
ischemia-reperfusion (IR) injury. The present findings
suggest that OGD-R, an in vitro model of IR injury induces
ROS formation and cytochrome c release leading to caspase-
3 mediated disruption of BBB tight junctions leading to
microvascular hyperpermeability. Inhibition of this pathway
may have therapeutic benefit against IR injury.
ACKNOWLEDGMENTS
We acknowledge Scott and White Hospital Research Grants
Program for financial support and Texas A&M Health
Science Center College of Medicine Integrated Imaging
Laboratory for the use of the confocal laser microscope.
REFERENCES
1. Abramov AY, Scorziello A, Duchen MR.
Three distinct mechanisms generate oxy-
gen free radicals in neurons and contribute
to cell death during anoxia and reoxygen-
ation. J Neurosci 27: 1129–1138, 2007.
2. Baburamani AA, Ek CJ, Walker DW,
Castillo-Melendez M. Vulnerability of the
developing brain to hypoxic-ischemic dam-
age: contribution of the cerebral vascula-
ture to injury and repair? Front Physiol 3:
424, 2012.
3. Ballabh P, Braun A, Nedergaard M. The
blood–brain barrier: an overview: struc-
ture, regulation, and clinical implications.
Neurobiol Dis 16: 1–13, 2004.
4. Ballabh P, Hu F, Kumarasiri M, Braun A,
Nedergaard M. Development of tight junc-
tion molecules in blood vessels of germinal
matrix, cerebral cortex, and white matter.
Pediatr Res 58: 791–798, 2005.
5. Chan PH. Reactive oxygen radicals in
signaling and damage in the ischemic
brain. J Cereb Blood Flow Metab 21: 2–
14, 2001.
6. Chen Q, Chai YC, Mazumder S, Jiang C,
Macklis RM, Chisolm GM, Almasan A.
The late increase in intracellular free
radical oxygen species during apoptosis
is associated with cytochrome c release,
caspase activation, and mitochondrial
dysfunction. Cell Death Differ 10: 323–
334, 2003.
7. Chen X, Threlkeld SW, Cummings EE, Juan
I, Makeyev O, Besio WG, Gaitanis J, Banks
WA, Sadowska GB, Stonestreet BS. Ische-
mia-reperfusion impairs blood–brain bar-
rier function and alters tight junction
protein expression in the ovine fetus.
Neuroscience 226: 89–100, 2012.
8. Childs EW, Tharakan B, Hunter FA, Isong
M, Liggins ND. Mitochondrial respiratory
chain complex III is involved in BAK-
induced microvascular endothelial cell hyp-
erpermeability. Shock 29: 636–641, 2008.
9. Choi BH. Oxygen, antioxidants and brain
dysfunction. Yonsei Med J 34: 1–10, 1993.
10. Cino M, Del Maestro RF. Generation of
hydrogen peroxide by brain mitochondria:
the effect of reoxygenation following post-
decapitative ischemia. Arch Biochem Bio-
phys 269: 623–638, 1989.
11. Crack PJ, Taylor JM. Reactive oxygen spe-
cies and the modulation of stroke. Free
Radic Biol Med 38: 1433–1444, 2005.
12. ElAli A, Doeppner TR, Zecharah A, Her-
mann DM. Increased blood–brain barrier
permeability and brain edema after focal
cerebral ischemia induced by hyperlipid-
emia: role of lipd peroxidation, calpain 1/2,
matrixmetalloproteinase-2/9, and RhoA
overactivation. Stroke 42: 3238–3244,
2011.
13. Elias-Mir�o M, Jim�enez-Castro MB, Rod�es J,
Peralta C. Current knowledge on oxidative
stress in hepatic ischemia/reperfusion. Free
Radic Res 47: 555–568, 2013.
14. Flamm ES, Demopoulos HB, Seligman ML,
Poser RG, Ransohoff J. Free radicals in
cerebral ischemia. Stroke 9: 445–447,
1978.
15. Gasche Y, Soccal PM, Kanemitsu M, Copin
JC. Matrix metalloproteinases and diseases
of the central nervous system with a special
emphasis on ischemic brain. Front Biosci
11: 1289–1301, 2006.
16. Gu Y, Dee CM, Shen J. Interaction of free
radicals, matrix metalloproteinases and
caveolin-1 impacts blood–brain barrier per-
meability. Front Biosci (Schol Ed) 3: 1216–
1231, 2011.
17. Gundimeda U, McNeill TH, Elhiani AA,
Schiffman JE, Hinton DR, Gopalakrishna R.
Green tea polyphenols precondition
against cell death induced by oxygen-
glucose deprivation via stimulation of lam-
inin receptor, generation of reactive oxy-
gen species and activation of protein
kinase Ce. J Biol Chem 87: 34694–34708,
2012.
18. Hagberg H, Mallard C, Rousset CI, Wang
X. Apoptotic mechanisms in the immature
brain: involvement of mitochondria. J Child
Neurol 24: 1141–1146, 2009.
19. Hall ED, Vaishnav RA, Mustafa AG. Anti-
oxidant therapies for traumatic brain injury.
Neurotherapeutics 7: 51–61, 2010.
20. Hawkins BT, Davis TP. The blood-brain
barrier/neurovascular unit in health and
disease. Pharmacol Rev 57: 173–185, 2005.
21. Hensley K, Floyd RA. Reactive oxygen
species and protein oxidation in aging: a
look back, a look ahead. Arch Biochem
Biophys 397: 377–383, 2002.
22. H€uttemann M, Helling S, Sanderson TH,
Sinkler C, Samavati L, Mahapatra G, Va-
rughese A, Lu G, Liu J, Ramzan R, Vogt S,
Grossman LI, Doan JW, Marcus K, Lee I.
Regulation of mitochondrial respiration
and apoptosis through cell signaling: cyto-
chrome c oxidase and cytochrome c in
ischemia/reperfusion injury and inflamma-
tion. Biochim Biophys Acta 1817: 598–
609, 2012.
23. Jung JE, Kim GS, Chen H, Maier CM,
Narasimhan P, Song YS, Niizuma K, Katsu
M, Okami N, Yoshioka H, Sakata H, Goe-
ders CE, Chan PH. Reperfusion and neuro-
vascular dysfunction in stroke: from basic
mechanisms to potential strategies for
neuroprotection. Mol Neurobiol 41: 172–
179, 2010.
H. Alluri et al.
194 ª 2013 John Wiley & Sons Ltd
24. Kaur C, Ling EA. Antioxidants and neuro-
protection in the adult and developing
central nervous system. Curr Med Chem
15: 3068–3080, 2008.
25. Kaur C, Ling EA. Blood brain barrier in
hypoxic-ischemic conditions. Curr Neuro-
vasc Res 5: 71–81, 2008.
26. Leonardo CC, Pennypacker KR. Neuroin-
flammation and MMPs: potential thera-
peutic targets in neonatal hypoxic-ischemic
injury. J Neuroinflammation 6: 13, 2009.
27. Lew�en A, Matz P, Chan PH. Free radical
pathways in CNS injury. J Neurotrauma 17:
871–890, 2000.
28. Li JM, Mullen AM, Yun S, Wientjes F,
Brouns GY, Thrasher AJ, Shah AM. Essen-
tial role of the NADPH oxidase subunit
p47 (phox) in endothelial cell superoxide
production in response to phorbol ester
and tumor necrosis factor-alpha. Circ Res
90: 143–150, 2000.
29. Mehta SL, Manhas N, Raghubir R. Molec-
ular targets in cerebral ischemia for devel-
oping novel therapeutics. Brain Res Rev 54:
34–66, 2007.
30. Powell JT, Tsapepas DS, Martin ST, Hardy
MA, Ratner LE. Managing renal transplant
ischemia reperfusion injury: novel therapies
in the pipeline. Clin Transplantation 27:
484–491, 2013.
31. Pun PB, Lu J, Moochhala S. Involvement of
ROS in BBB dysfunction. Free Radic Res 43:
348–364, 2009.
32. Ricci JE, Gottlieb RA, Green DR. Caspase-
mediated loss of mitochondrial function
and generation of reactive oxygen species
during apoptosis. J Cell Biol 160: 65–75,
2003.
33. Rieger JM, Shah AR, Gidday JM. Ischemia-
reperfusion injury of retinal endothelium by
cyclooxygenase- and xanthine oxidase-
derived superoxide. Exp Eye Res 74: 493–
501, 2002.
34. Saito A, Maier CM, Narasimhan P, Nishi T,
Song YS, Yu F, Liu J, Lee YS, Nito C,
Kamada H, Dodd RL, Hsieh LB, Hassid B,
Kim EE, Gonz�alez M, Chan PH. Oxidative
stress and neuronal death/survival signaling
in cerebral ischemia. Mol Neurobiol 31:
105–116, 2005.
35. SawantDA, Tharakan B, Adekanbi A, Hunter
FA, Smythe WR, Childs EW. Inhibition of
VE-cadherin proteasomal degradation atten-
uates microvascular hyperpermeability.
Microcirculation 18: 46–55, 2011.
36. Sugawara T, Noshita N, Lewen A, Gasche
Y, Ferrand-Drake M, Fujimura M, Morita-
Fujimura Y, Chan PH. Overexpression of
copper/zinc superoxide dismutase in trans-
genic rats protects vulnerable neurons
against ischemic damage by blocking the
mitochondrial pathway of caspase activa-
tion. J Neurosci 22: 209–217, 2002.
37. Sun H, Breslin JW, Zhu J, Yuan SY, Wu
MH. Rho and ROCK signaling in VEGF-
induced microvascular endothelial hyper-
permeability. Microcirculation 13: 237–
247, 2006.
38. Tojkander S, Gateva G, Lappalanian P.
Actin stress fibers-asesembly, dynamics
and biological roles. J Cell Sci 125: 1–10,
2012.
39. Traystman RJ, Kirsch JR, Koehler RC. Oxy-
gen radical mechanisms of brain injury
following ischemia and reperfusion. J Appl
Physiol 71: 1185–1195, 1991.
40. Volpe JJ. Intraventricular hemorrhage in
the premature infant-current concepts.
Part I. Ann Neurol 25: 3–11, 1989.
41. Zhao K, Zhao GM, Wu D, Soong Y, Birk
AV, Schiller PW, Szeto HH. Cell-permeable
peptide antioxidants targeted to inner
mitochondrial membrane inhibit mitochon-
drial swelling, oxidative cell death, and
reperfusion injury. J Biol Chem 279:
34682–34690, 2004.
42. Zhu H, Wang Z, Xing Y, Gao Y, Ma T, Lou
L, Lou J, Gao Y, Wang S, Wang YJ. Baicalin
reduces the permeability of the blood–
brain barrier during hypoxia in vitro by
increasing the expression of tight junction
proteins in brain microvascular endothelial
cells. J Ethnopharmacol 141: 714–720,
2012.
43. Zulueta JJ, Sawhney R, Yu FS, Cote CC,
Hassoun PM. Intracellular generation of
reactive oxygen species in endothelial cells
exposed to anoxia-reoxygenation. Am J
Physiol 272: L897–L902, 1997.
Blood-Brain Barrier and Hyperpermeability
ª 2013 John Wiley & Sons Ltd 195