The Protective Effect of Cerebralcare Granule�� on BrainEdema, Cerebral Microcirculatory Disturbance, andNeuron Injury in a Focal Cerebral Ischemia Rat Model
FANG WANG,*,� QIN HU,� CHUN-HUA CHEN,� XIANG-SHUN XU,*,� CHANG-MAN ZHOU,� YA-FANG
ZHAO,*,� BAI-HE HU,*,� XIN CHANG,*,� PING HUANG,*,� LEI YANG,� YU-YING LIU,*,� CHUAN-SHE
WANG,*,�,§ JING-YU FAN,*,� KE ZHANG,– GUO-YU LI,– JING-HUI WANG,– AND JING-YAN HAN,*,�,§
*Tasly Microcirculation Research Center, Peking University Health Science Center, Beijing, China; �Key Laboratory of Microcirculation, State
Administration of Traditional Chinese Medicine, China; �Department of Anatomy, School of Basic Medical Sciences, Peking University, Beijing,
China; §Department of Integration of Chinese and Western Medicine, School of Basic Medical Sciences, Peking University, Beijing, China; –School
of Pharmacy, Shihezi University, Xinjiang, China
Address for correspondence: Jing-Yan Han, M.D., Ph.D., Professor and Chairman, Department of Integration of Chinese and Western Medicine,
School of Basic Medical Sciences, Peking University, 38 Xueyuan Road, Beijing 100191, China. E-mail: [email protected]
Received 25 April 2011; accepted 13 December 2011.
ABSTRACT
Objective: The purpose of the present study was to explore the
protective effects of CG on rat cerebral injury after focal cerebral I ⁄ R.
Methods: Male Sprague–Dawley rats were subjected to right
middle cerebral artery occlusion for 60 minutes followed by
reperfusion for 60 minutes or 24 hours. CG (0.4 or 0.8 g ⁄ kg) was
administrated 90 minutes before ischemia. Brian edema was
evaluated by Evan’s blue dye extravasations and brain water
content, leukocyte adhesion, and albumin leakage were
determined with an upright fluorescence microscope, and neuron
damage was assessed by 2,3,5-triphenyltetrazolium chloride
staining, terminal deoxynucleotidyl transferase-mediated dUTP
nick end labeling, and immunohistochemistry of caspase-3, p53,
p53 upregulated modulator of apoptosis.
Results: Focal cerebral I ⁄ R elicited a prominent brain edema, an
increase in leukocyte adhesion, and albumin leakage, as well as
neuron damage. All the insults after focal cerebral I ⁄ R were
significantly attenuated by pretreatment with CG.
Conclusions: Pretreatment with CG significantly reduced focal
cerebral I ⁄ R-induced brain edema, cerebral microcirculatory
disturbance, and neuron damage, suggesting the potential of CG
as a prophylactic strategy for patients in danger of stroke.
Key words: ischemia and reperfusion, middle cerebral artery
occlusion, neuron injury, cerebral microcirculatory disturbance,
brain edema
Abbreviations used: BBB, blood–brain barrier; CG, Cerebralcare
Granule; dUTP, 2¢-deoxyuridine, 5¢-triphosphate; EB, Evan’s blue;
FITC, fluorescein isothiocyanate; I ⁄ R, ischemia and reperfusion;
MCAO, middle cerebral artery occlusion; PBS, phosphate-buffered
saline; PUMA, p53 upregulated modulator of apoptosis;
TTC, 2,3,5-triphenyltetrazolium chloride; TUNEL, terminal
deoxynucleotidyl transferase-mediated dUTP nick end labeling;
UPLC-TOF-MS, ultra performance liquid chromatography-time-
of-flight-mass spectrometer
Please cite this paper as: Wang F, Hu Q, Chen C-H , Xu X-S, Zhou C-M, Zhao Y-F, Hu B-H, Chang X, Huang P, Yang L, Liu Y-Y, Wang C-S, Fan J-Y,
Zhang K, Li G-Y, Wang J-H, Han J-Y. The protective effect of cerebralcare granule� on brain edema, cerebral microcirculatory disturbance and neuron
injury in a focal cerebral ischemia rat model. Microcirculation 19: 260–272, 2012.
INTRODUCTION
Stroke, as a serious neurological disease, inflicts a major
impact on the public health in almost every country. Ische-
mic stroke frequently results in brain edema causing severe
or even fatal outcome. However, several available strategies,
such as decompression craniotomy, hypothermia, and
osmotic diuresis therapy, are applied only as a rescue ther-
apy [6,19,24], while options for preventing brain edema are
limited.
CG was approved in 1996 by the China State Food and
Drug Administration for treatment of headache and dizzi-
ness associated with cerebrovascular diseases. CG is com-
posed of 11 herbs, including Radix angelicae sinensis
(Dang Gui), Rhizoma chuan xiong (Chuan Xiong), Raidix
paeoniae alba (Bai Shao), Ramulus uncariae cum uncis
(Gou Teng), Caulis spatholobi (Ji Xue Teng), Spica prunel-
lae (Xia Ku Cao), Concha margaritifera usta (Zhen Zhu
Mu), Radix rehmanniae preparata (Di Huang), Semen cas-
siae (Jue Ming Zi), Rhizoma corydalis yanhusuo (Yan Hu
DOI:10.1111/j.1549-8719.2011.00155.x
Original Article
260 ª 2012 John Wiley & Sons Ltd
The Official Journal of the Microcirculatory Society, Inc. and the British Microcirculation Society
Suo), and Herba asari (Xi Xin). Each of these herbs has
previously been characterized and commonly used in tra-
ditional Chinese medicine. A number of studies have been
conducted to identify the active constituent contained in
these herbs and the mechanism behind their actions. For
example, ligustrazine (the major constituent of Rhizoma
chuan xiong) was reported to inhibit neutrophil adhesion
to endothelial cells [15,17], and peoniflorin (the major
constituent of Raidix paeoniae alba), ligustrazine, and feru-
lic acid (the major constituent of Radix angelicae sinensis)
were found to decrease the infarct size and brain tissue
damage [12,15,17,18,29].
Our previous studies undertaken in Mongolian gerbils
demonstrated that CG attenuates the bilateral common
carotid artery occlusion-elicited cerebral microcirculatory
disturbance, hippocampal neuron injury, as well as BBB dis-
ruption, indicative of the potential of CG in protecting brain
from edema after ischemia and reperfusion (I ⁄ R) [28,30].
The purpose of the present study was to verify the ability of
CG as a prophylactic management in the development
of brain edema following I ⁄ R, in addition to attenuation of
microcirculation disturbance and neuron injury, using a focal
cerebral I ⁄ R model in which rats were subjected to MCAO.
MATERIALS AND METHODS
AnimalsMale Sprague–Dawley rats weighing 270–300 g were pur-
chased from the Animal Center of Peking University
Health Science Center. The animals were fasted for
12 hours before each experiment and allowed free access to
water. All animals were handled according to the guidelines
of the Peking University Animal Research Committee. The
surgeries and protocols were approved by the Committee
on the Ethics of Animal Experiments of the Health Science
Center of Peking University.
Cerebralcare GranuleCG was produced by Tianjin Tasly Pharmaceutical Co. Ltd
(Tianjin, China). The batch number of the granules used in
this study was Z10960082. No steroid is included in the con-
tent of CG. The processing of the product followed a strict
quality control, and the ingredients were subjected to stan-
dardization. The herbs were manufactured as granules after
dynamic cycle extraction and concentration by evaporating
and spray drying. Then the CG was packed with aluminum
foil composite, 4 g per bag. The compound was dissolved in
water to a concentration of 80 or 160 mg ⁄ mL before use [30].
UPLC–TOF-MS Analysis of CG Components inRat PlasmaRats were fasted overnight before experiment, and adminis-
trated by gavage with 0.8 g ⁄ kg CG dissolved in saline. Ani-
mals that received equivalent volume of saline were used as
control. Blood was harvested from orbit veins one, two,
three, and four hours after drug administration, and anti-
coagulated with heparin. Plasm was separated by centri-
fugation for 10 minutes at 3000 rpm (956 rcf), and pro-
cessed for analysis by UPLC–TOF-MS (UPLC Waters
ACQUITY and LCT Premier XE TOF-MS) [31]. MassLynx
4.1 Software (Waters Corporation, Milford, MA, USA) was
used to acquire data. Data were collected over the mass
range of 50–1500 m ⁄ z.
Animal Model and Drug AdministrationFocal cerebral ischemia was induced by MCAO, as
described elsewhere [3]. Briefly, animals were anesthetized
using 20% urethane. The right common carotid artery,
including its bifurcation, was exposed. The external carotid
artery was divided, leaving a stump of 3–4 mm. The inter-
nal carotid artery was then isolated, and the pterygopalatine
artery was ligated close to its origin. The internal carotid
artery was then clamped with a small vascular clip. The
common carotid artery was also clamped. The stump of
the external carotid artery was reopened, and a 4-0 mono-
filament nylon suture with a slightly enlarged and round
tip was inserted up through the internal carotid artery.
When a small resistance was felt, insertion was stopped.
The distance from bifurcation of the common carotid
artery to the tip of the suture was 18–22 mm. After occlu-
sion for one hour, the suture was withdrawn through the
internal carotid artery into the external artery, allowing
reperfusion. The animals in the sham group underwent the
same surgical procedures, but without arterial occlusion.
For evaluating the CG effect, animals in the CG + I ⁄ Rgroups received the drug orally 90 minutes before MCAO
at either 0.4 g ⁄ kg (CG 0.4 + I ⁄ R group) or 0.8 g ⁄ kg (CG
0.8 + I ⁄ R group).
Evan’s Blue Dye Extravasation and Brain WaterContentDisruption of the BBB was analyzed 24 hours after the
MCAO using EB dye. Briefly, rats were anesthetized with
pentobarbital sodium (0.1 g ⁄ kg body weight, i.p.), and EB
dye (Sigma-Aldrich, St. Louis, MO, USA; 4%, 3 mL ⁄ kg) in
saline was injected within two minutes into the left femoral
vein and allowed to circulate for 120 minutes. Rats were
transcardially perfused with PBS until colorless perfusion
fluid drained from the right atrium. The amount of extrava-
sated EB in the brain was determined by spectrofluropho-
tometry at an excitation wavelength of 620 nm. The brain
was removed and weighed immediately for determination of
wet weight, and weighed after drying in an oven at 105�C for
24 hours for determination of dry weight. The brain water
content was presented as [(wet weight ) dry weight) ⁄ wet
weight] · 100% [13].
Chinese Medicine and Cerebral Reperfusion Injury
ª 2012 John Wiley & Sons Ltd 261
Observation of MicrocirculationAfter being anesthetized with 20% urethane (2.0 g ⁄ kg
body weight, i.m.), the rats were tracheotomized and
mechanically ventilated with a breathing machine designed
for small animals (ALC-V8). With a hand-held drill, a
skull window of 4 · 6 mm was created 1 mm behind the
coronal suture, and 1 mm on the right side of the sagittal
suture. The dura was superfused contiguously with 37�C
warm physiological saline. A single unbranched venule was
selected for determination, which was without obvious
bend with diameter ranging between 30 and 50 lm and
length of about 200 lm. The dynamic of cerebral microcir-
culation was continuously observed for 60 minutes using a
biological microscope (DM-LFS, Leica, Germany) equipped
with a color monitor (J2118A, TCL, Huizhou, China), a
video timer (VTG-55B, For-A, Tokyo, Japan), and a DVD
recorder (DVR-R25, Malata, China). The microvessel
images were recorded through a high speed video camera
system at a rate of 1000 frames ⁄ sec (FASTCAM-ultima
APX, Photron, Tokyo, Japan), and the recordings were
replayed at a rate of 25 frames ⁄ sec from the stored images.
The RBC velocity in venule was measured with Image Pro
Plus software (IPP, Media Cybernetic, Bethesda, MD,
USA) before ischemia and 1, 20, 40, and 60 minutes after
reperfusion, respectively. Using the same software, the
inner diameter of cerebral venules was measured and pre-
sented as the mean of three measurements at one location
[30]. To assess leukocyte adhesion in venules, the fluores-
cence tracer Rhodamine 6G (Sigma, St Louis, MO, USA)
was administrated (5 mg ⁄ kg body weight) to the animal
via the femoral vein 10 minutes before ischemia. After
craniotomy, the cerebral microvessels were observed under
an upright fluorescence microscope (DM-LFS, Leica,
Germany) in combination with a CCD camera (USS-301,
Uniq, Santa Clara, CA, USA) using a helium-neon laser
beam for illumination. The adherent leukocytes were identi-
fied as those that attached to the venular walls for more than
30 seconds. The number of adherent leukocytes was scored
under basal condition and 1, 20, 40, and 60 minutes after
reperfusion. For evaluation of albumin leakage from venular
wall, FITC-albumin (Sigma-Aldrich) was infused (50 mg ⁄ kg
body weight) slowly through femoral vein 10 minutes before
ischemia, and the upright fluorescence microscopy was
employed. Using IPP software, the fluorescence intensities of
FITC-albumin inside the lumen of selected venules (Iv) and
in the surrounding interstitial area (Ii) were measured. The
ratio Ii ⁄ Iv was calculated and compared with the baseline as
an indication of albumin leakage.
Infarct VolumeTwenty-four hours after reperfusion, the rat was anesthe-
tized with 20% urethane (2.0 g ⁄ kg body weight, i.m.) and
the brain was rapidly excised and sliced coronally into five
sections (2 mm thick) beginning from optic chiasma. The
slices of the brain were incubated for 30 minutes at 37�C
in a 2% solution of TTC in PBS and then photographed as
digital images (Digital Sight DS-5M-U1; Nikon, Tokyo,
Japan). The infarct volume was calculated based on the
ratio of the infarct area of the ipsilateral hemisphere to
total non-infarct area from both the ipsilateral and contra-
lateral hemispheres to avoid the influence of tissue edema
[3].
Neurological ScoresTwenty-four hours after reperfusion, the neurological
scores were performed in a blinded fashion, as previously
reported with some modifications [13]. Each rat was sub-
jected to 6 tests (spontaneous activity, symmetry in the
movement of four limbs, forepaw outstretching, climbing,
body proprioception, and response to vibrissae touch) with
the minimum neurological score being 3 and the maximum
18. The score given to each rat at the completion of the
evaluation was the summation of the 6 individual test
scores [13].
Nissl and Immunohistochemistry StainingFor this propose, the brain samples were collected 24 hours
after reperfusion. After transcardiac perfusion with 250 mL
of 4% paraformaldehyde under anesthesia, rat brain was
removed and postfixed in the same fixative for 48 hours.
The brain tissue located in the middle between the optic
chiasma and the cerebral caudal end was cut into blocks,
embedded in paraffin, and sectioned at 10 lm. The
sections were deparaffinized and rehydrated, sequentially,
and processed for either Nissl staining or immunohisto-
chemistry. For immunohistochemistry, the sections were
incubated with the following primary antibodies: mouse
anti-caspase-3, goat anti-p53 (1:1000, Santa Cruz Biotech-
nology, Santa Cruz, CA, USA), and rabbit anti-PUMA
(1:1000, Cell Signaling, Boston, MA, USA). After being
washed, the samples were incubated with a biotinylated
secondary antibody followed by avidin–biotin–peroxidase
complex, and visualized with diaminobenzidine. As control,
a consecutive section was treated similarly except that the
primary antibody was omitted. IPP software was used to
assess the number of positive cells of caspase-3, p53, and
PUMA in the cortex penumbral region of the ischemic side
of brain, as described previously [3,13].
TUNEL AssayThe animals under anesthesia were infused via the left ven-
tricle with 4% paraformaldehyde in 0.01 M PBS (pH 7.4)
24 hours after reperfusion. Brain was removed and post-
fixed with the same fixative for six hours, and cryoprotect-
ed in 30% sucrose in PBS for at least 48 hours at 4�C. The
coronal brain sections (10 lm thick) were cut from the
F. Wang et al.
262 ª 2012 John Wiley & Sons Ltd
middle between the optic chiasma and the cerebral caudal
end on a cryostat (Leica CM1800, Bensheim, Germany). To
evaluate the apoptotic neuron cells in the cortex penumbral
region, TUNEL was conducted using an in situ cell death
detection kit (Fluorescein dUTP Kit; Roche Inc., Indiana-
polis, IN, USA), according to the instruction of manufac-
turer. The nuclei were counterstained with hoechst33342
(2 lg ⁄ mL) for five minutes. The slides were rinsed with
PBS, coverslipped with mounting medium and observed
under a laser confocal microscope (Axiovert 2000, Zeiss,
Germany) with a 63· objective at an excitation wavelength
of 480 nm and emission wavelength of 530 nm. Five fields
were randomly selected for each rat. The number of TUN-
EL-positive nuclei and the total number of nuclei in each
field were scored, and the ratio of the two values was auto-
matically calculated with IPP 5.0 software. A similarly trea-
ted consecutive section without addition of TdT was used
as control [31].
Ultrastructure of Cerebral CortexAnother set of rats were anesthetized 24 hours after reper-
fusion, and perfused with a fixative composed of 4% form-
aldehyde and 2% glutaraldehyde in 0.1 M phosphate buffer
at a speed of 3 mL ⁄ min through the left ventricle. The ani-
mals were decapitated and a cerebral slice of 1 mm thick
was taken from the middle between the optic chiasma and
the cerebral caudal end, and tissue samples smaller than
1 mm3 in the cortex penumbral region were prepared as
routine for transmission electron microscopy. Ultrathin
sections of cortex were examined by an electron micro-
scope (JEM 1230, Jeol, Japan).
Statistical AnalysisAll parameters were expressed as mean ± SE. Statistical
analysis was performed using ANOVA followed by Tukey
test for multiple comparisons. A p-value <0.05 was consid-
ered statistically significant.
RESULTS
Identification and Kinetics of CG Components inRat PlasmaUPLC–TOF-MS system is characterized by high sensitivity
and specificity. By this approach, two components, coryda-
line and cassialactone, were detected in the plasma of CG
treated rats, which are obviously distinct from that in
plasma of control animals (Figures 1 and 2). These two
components appeared one hour after CG administration,
and maintained detectable over the next three hours.
EB Dye Extravasation and Brain Water ContentAfter 24 hours of reperfusion, EB was clearly observed in
the ipsilateral hemisphere of the coronal sections (Fig-
ure 3A). The EB content of the brain tissue increased sig-
nificantly in the I ⁄ R group compared with the sham group,
while CG treatment mitigated the I ⁄ R-caused increase in
EB content of the brain tissue (Figure 3B). Likewise, a sig-
nificant increase in brain water content was revealed in the
ipsilateral hemisphere 24 hours after the MCAO, when
compared with the sham group, which was alleviated sig-
nificantly by CG treatment at 0.8 g ⁄ kg (Figure 3C). No sig-
nificant differences in brain water content were observed
among the groups in the contralateral hemisphere.
Albumin LeakageCG pretreatment prevented FITC-labeled albumin leakage
from cerebral venules evoked by I ⁄ R, and the representative
images are presented in Figure 4A. Transvascular efflux of
FITC-labeled albumin was quantified for different groups
(Figure 4B). There was no detectable leakage under the
basal condition in all groups. In the sham group, only a
small amount of albumin was found to have leaked from
the venular wall during reperfusion. By contrast, albumin
leakage in the I ⁄ R group increased markedly after one
minute of reperfusion and kept increasing with time. The
I ⁄ R-induced albumin leakage was diminished significantly
in the animals receiving CG at both doses tested.
Leukocyte AdhesionLeukocytes in the microcirculation are easily visualized and
quantified after rhodamine injection (Figure 5A). The
changes in the number of leukocytes adherent to the venu-
lar wall in the four groups are shown in Figure 5B. In the
sham group, the number of adherent leukocytes increased
slightly after reperfusion but showed no significance when
compared with baseline. The number of adherent leuko-
cytes in the I ⁄ R group significantly increased immediately
after reperfusion and remained increasing until the end of
observation. Pretreatment with CG attenuated I ⁄ R-elicited
enhancement of leukocyte adhesion with the higher dose
being more significant than the lower dose.
RBC Velocity and Venular DiameterWith the use of the microscope equipped with a high-speed
video camera, cerebral microvessels and RBC moving inside
the venules were clearly visible in the cerebral cortex (Fig-
ure 6A). Impressively, the number of open capillaries after
I ⁄ R reduced considerably (a2) in comparison with the
sham group (a1), indicating a decreased blood supply in
this area. CG pretreatment, especially at 0.8 g ⁄ kg, obviously
attenuated this alteration by I ⁄ R (a4).
No significant difference in venular diameter was
observed among the sham, I ⁄ R, and CG-treated groups at
baseline, nor over the 60 minutes reperfusion (Figure 6B).
The time courses of RBC velocity in cerebral venules in dif-
ferent groups are depicted in Figure 6C. The mean RBC
Chinese Medicine and Cerebral Reperfusion Injury
ª 2012 John Wiley & Sons Ltd 263
velocity in cerebral venules in the sham group did not alter
significantly during the period of observation. RBC velocity
declined temporarily after I ⁄ R compared with baseline with
the minimum at one minute of reperfusion and then grad-
ually returned to the basal level. The CG treatment had no
significant effect on RBC velocity in comparison with the
I ⁄ R group.
Cerebral Infarct and Neurological ScoreRepresentative TTC-stained brain sections in different
groups are shown in Figure 7A, wherein the areas of gray
color represent the infarct regions. The ratios of cerebral
infarct volume in four groups are shown in Figure 7B. No
infarct was found in the sham group. The ratio of infarct vol-
ume in the I ⁄ R group reached to 30%, which was diminished
significantly by pretreatment with CG with a reduction of
10% and 15% at 0.4 and 0.8 g ⁄ kg, respectively.
The neurological scores of the animals in each
group were 17.6 ± 0.2 (sham group), 9.3 ± 0.6 (I ⁄ Rgroup), 10.8 ± 0.6 (CG0.4 + I ⁄ R group), and 12.0 ± 0.5
(CG0.8 + I ⁄ R group). Pretreatment with CG at 0.8 g ⁄ kg
was found to increase the neurological score significantly in
comparison with the I ⁄ R group (p < 0.05) (Figure 7C).
Nissl StainingResults of Nissl staining at 24 hours after reperfusion
showed that neurons of the sham group exhibited normal
morphological features, while diverse neuron damages
occurred in the I ⁄ R group, such as cell loss, cellular swell-
ing, nuclear pyknosis, and karyorrhexis. CG treatment par-
tially prevented cell loss and protected neuron cells from
injury after MCAO (Figure 8).
Immunohistochemistry and TUNEL Staining ofCerebral CortexTUNEL assay was conducted to evaluate neuron apoptosis
24 hours after MCAO, and the result revealed the presence
of a large number of TUNEL-positive cells in the cortex
region of rats subjected to I ⁄ R challenge, although TUNEL-
positive cells were hardly observed in the sham group. The
Corydaline
H3CO
CH2OH
OH OH O
O
Time1.00 2.00 3.00 4.00 5.00 6.00 7.00 8.00 9.00 10.00 11.00 12.00
%
0
100
6.58
6.57
9.41
A
100
1.00 2.00 3.00 4.00 5.00 6.00 7.00 8.00 9.00 10.00 11.00 12.00
%
0
100B
100 6.58C
1.00 2.00 3.00 4.00 5.00 6.00 7.00 8.00 9.00 10.00 11.00 12.00
%
0
1006.58
D
%
100 6.56
1.00 2.00 3.00 4.00 5.00 6.00 7.00 8.00 9.00 10.00 11.00 12.00
%
0
E
1.00 2.00 3.00 4.00 5.00 6.00 7.00 8.00 9.00 10.00 11.00 12.000
Figure 1. Identification and kinetics of CG component (corydaline) in rat plasma (1). Rats were administrated 0.8 g ⁄ kg CG dissolved in saline.
Control animals received equivalent volume of saline. The rat plasma was analyzed by UPLC–TOF-MS. The figure is a representative of three
experiments, showing a prominent distinct component appeared in the plasma of CG-treated rat starting from one hour after administration of CG
and maintained over the next three hours, which is identified as corydaline. From bottom to top, control (A), and one hour (B), two hours (C), three
hours (D), and four hours (E) after CG administration.
F. Wang et al.
264 ª 2012 John Wiley & Sons Ltd
number of TUNEL-positive cells was significantly reduced
in the CG treatment groups compared with the I ⁄ R group,
with 10% and 30% reduction at 0.4 and 0.8 g ⁄ kg, respec-
tively (Figure 9A and C). Moreover, immunohistochemistry
was carried out to address the possible contribution of
caspase-3, p53, and PUMA in neuron apoptosis, and
the representative pictures and quantification of the imm-
unolabeling positive cells in different conditions are
presented in Figure 9B and D, respectively. The result
showed that the expressions of caspase-3, p53, and PUMA
increased noticeably in response to I ⁄ R challenge, all of
which were attenuated significantly by CG pretreatment in
a dose-dependent manner.
Ultrastructure of Microvessels and Neurons inCerebral CortexFigure 10 shows the representative transmission electron
micrographs of capillaries and neurons in cerebral cortex
24 hours after reperfusion in the sham, I ⁄ R, and CG
0.8 + I ⁄ R groups. Compared with the sham group (a1, a2),
I ⁄ R elicited a remarkable alteration in the cortical capillary,
manifested as narrowing lumen, rough inner surface, and
swelling endothelial cells. In addition, swelling perivascular
astrocyte processes containing dilated organelles were
observed frequently (b1, b2). These ultrastructural altera-
tions were alleviated by pretreatment with 0.8 g ⁄ kg CG
(c1, c2). The neurons in the sham group displayed normal
ultrastractural features (a3). In contrast, in the I ⁄ R group,
the neurons became obviously irregular in shape and
packed loosely, and some neurons were even missing.
Swelling processes of glial cells and death of neurons were
often observed as well (b3). CG treatment apparently atten-
uated I ⁄ R-induced changes (c3).
DISCUSSION
Brain edema is a major determinant in the progression of
cerebral dysfunction following I ⁄ R, and the development
Cassialactone
100A
Time1.00 2.00 3.00 4.00 5.00 6.00 7.00 8.00 9.00 10.00 11.00 12.00
%
03.09
1003.12
B
1.00 2.00 3.00 4.00 5.00 6.00 7.00 8.00 9.00 10.00 11.00 12.00
%
0
1003.12
C
1.00 2.00 3.00 4.00 5.00 6.00 7.00 8.00 9.00 10.00 11.00 12.00
%
0
100D
1.00 2.00 3.00 4.00 5.00 6.00 7.00 8.00 9.00 10.00 11.00 12.00
%
0
3.12
100E
1.00 2.00 3.00 4.00 5.00 6.00 7.00 8.00 9.00 10.00 11.00 12.00
%
0
3.12
H3CO
OCH3
OCH3
H3COH
N
Figure 2. Identification and kinetics of CG component (cassialactone) in rat plasma (2). Rats were administrated with 0.8 g ⁄ kg CG dissolved in
saline. Control animals received equivalent volume of saline. The rat plasma was analyzed by UPLC–TOF-MS. The figure is a representative of three
experiments, showing a prominent distinct component that appeared in the plasma of CG-treated rat starting from one hour after administration of
CG and maintained over the next three hours, which is identified as cassialactone. From bottom to top, control (A), and one hour (B), two hours (C),
three hours (D), and four hours (E) after CG administration.
Chinese Medicine and Cerebral Reperfusion Injury
ª 2012 John Wiley & Sons Ltd 265
of strategy for blunting brain edema is considered to be of
crucial importance in improving outcome of the patients
after stroke. With the focal cerebral I ⁄ R model, the present
study demonstrated that pretreatment with CG significantly
attenuated I ⁄ R-induced brain edema, as indicated by the
decline in EB dye extravasation and brain water content.
Furthermore, the I ⁄ R-imposed insults on cerebral microcir-
culation and neuron were alleviated by pretreatment with
A a1 a2 a3 a4
Sham I/R CG 0.4 + I/RB
C
2.5sue)
0.82 ††
CG 0.8 + I/R
#1.5
2.0 †
of b
rain
tis
†
er c
onte
nt
0.79
0.82 †
#†
0.5
1.0#†
s blu
e (μ
g/g
Bra
in w
at
0.73
0.76
0
Eva
n’
0.70Right Left
Figure 3. EB extravasation and brain water content. (A) Representatives of EB-stained brain sections from rats 24 hours after MCAO. No dye
leakage was observed in the sham group (a1), while intensive staining was seen in ischemic area in the I ⁄ R group (a2). In contrast, the brains from
the CG treatment groups showed weak EB staining in infarct areas (a3 for CG0.4 + I ⁄ R and a4 for CG0.8 + I ⁄ R, respectively). (B) Statistical analysis
of EB extravasation. (C) The brain water content of the ipsilateral ischemic hemisphere (right) and the contralateral hemisphere (left). The ipsilateral
hemisphere of CG 0.8 + I ⁄ R group had a significantly decreased water content compared with the I ⁄ R group. Data are means ± SE from six animals.�p < 0.05 vs. sham group, #p < 0.05 vs. I ⁄ R group.
ba
c
V
d
500
ShamI/RCG 0.4 + I/RCG 0.8 + I/R
400
300 **
**
†
†
††
200 **
* *
* *#
# #
#
#
#
Alb
umin
leak
age
(%)
#
Baseline 1 20 40 60 (minutes)100 * * *
I R
e f
A B
Figure 4. The effect of pretreatment with CG on the albumin leakage from the venule. (A) Representative images of the sham group (a and b), I ⁄ Rgroup (c and d), and CG 0.8 g ⁄ kg + I ⁄ R group (e and f). Before I ⁄ R injury (baseline), no obvious albumin leakage was detected in the three groups
(a, c, e). This situation persisted to the end of observation in the sham group (b). The albumin leakage from venular wall was observed at 60 minutes
after I ⁄ R injury (d). The I ⁄ R injury-induced albumin leakage from cerebral venule was apparently suppressed by the pretreatment with 0.8 g ⁄ kg CG
(f). Bar = 50 lm. (B) Time course of albumin leakage in the sham, I ⁄ R, CG0.4 + I ⁄ R and CG0.8 + I ⁄ R groups. Data are mean ± SE from six animals.
*p < 0.05 vs. baseline, �p < 0.05 vs. sham group, #p < 0.05 vs. I ⁄ R group.
F. Wang et al.
266 ª 2012 John Wiley & Sons Ltd
CG as well, which is in accordance with our previous find-
ings using gerbil as a global I ⁄ R model [28,30]. Finally, the
result of neurological score proved the prospective benefi-
cial action of CG on the outcome. The results by UPLC–
TOF-MS analysis showed that the protocol used can ensure
the appearance of the CG components in the plasma when
ShamI/R
8CG 0.4 + I/RCG 0.8 + I/R
6
*
**
*†
†
2
4
* * * *
* * * *
# #
# # #
††† #
Baseline 201 40 60 (minutes)
Num
ber
of a
dher
ent l
euko
cyte
s(p
er 2
00 μ
m v
enul
e)
0
I R
ba
dc
V
e fe f
A B
Figure 5. The effect of pretreatment with CG on the number of leukocytes adhering to venular wall. (A) Representative images of the sham group (a
and b), I ⁄ R group (c and d), and CG 0.8 g ⁄ kg + I ⁄ R group (e and f) acquired at baseline (a, c, e) and 60 minutes after I ⁄ R injury (b, d, f). Dotted line
arrows: adherent leukocytes; Actual line arrow: rolling leukocyte. Bar = 50 lm. (B) The time course of the number of adherent leucocytes in different
conditions. Compared with the I ⁄ R group, the number of adherent leukocytes in the CG administration groups decreased significantly starting from
20 minutes after reperfusion. Data are mean ± SE from six animals. I ⁄ R, ischemia and reperfusion; V, venule. *p < 0.05 vs. baseline, �p < 0.05 vs. sham
group, #p < 0.05 vs. I ⁄ R group.
A
a31 2 4a3a1 a2 a4
50 μm
Sham
CB
45
50
μm)
/s)
1.6
I/R CG 0.4 + I/R CG 0.8 + I/R
40
diam
eter
(
eloc
ity (m
m/
0.8
1.2
*
†
†
30
35
Venu
lar
d
RB
C v
e
0.4*
††
Baseline 1 20 40 60 (minutes)Baseline 1 20 40 60 (minutes)0
Figure 6. The effect of pretreatment with CG on rat cerebral cortical venular diameter and RBC velocity in venules. (A) The representative images of
venules observed by the high-speed video camera system. Arrows show the venules in the sham group (a1), I ⁄ R group (a2), CG 0.4 + I ⁄ R group (a3),
and CG 0.8 + I ⁄ R (a4). (B) The time course of the cerebral venular diameter. (C): The time course of the velocity of RBC in cerebral venule. Data are
mean ± SE from six animals. *p < 0.05 vs. baseline, �p < 0.05 vs. sham group.
Chinese Medicine and Cerebral Reperfusion Injury
ª 2012 John Wiley & Sons Ltd 267
A
CG 0.8 + I/R
I/R CG 0.4 + I/R
Sham
B C
volu
me
(%)
30
40
#
†
†15
20
l sco
res
#†
††
of i
nfar
ct v
10
20 #†
5
10
Neu
rolo
gica
l † †
Rat
io
0 0
N
CG 0.8 + I/R
I/R CG0.4 + I/R
Sham CG CG 0.8 + I/R
I/R0.4 + I/R
Sham
Figure 7. The influence of CG pretreatment on neuronal injury of rats subjected to focal cerebral I ⁄ R. (A) Representatives of TTC-stained brain
sections 24 hours after cerebral I ⁄ R. The white areas represent the infarct regions. (B) A quantitative evaluation of the influence of CG on infarct
volume. (C) Neurological scores 24 hours after MCAO. Grades of 3–18 were used. Administration of CG at 0.8 g ⁄ kg significantly improved
neurological function compared with the I ⁄ R group. Data are mean ± SE from six animals. �p < 0.05 vs. sham group, #p < 0.05 vs. I ⁄ R group.
D1A1 B1 C1
A2 B2 C2 D2
A3 B3 C3 D3
20 μm
10 μm
Figure 8. Nissl staining of cerebral cortex in penumbral region of rats. The arrows indicate Nissl-positive neurons. The double arrows indicate nuclear
pyknosis with karyorrhexis. a1, a2, a3: sham group; b1, b2, b3: I ⁄ R group; c1, c2, c3: CG 0.4 + I ⁄ R group; d1, d2, d3: CG 0.8 + I ⁄ R group. Data are
mean ± SE from six animals.
F. Wang et al.
268 ª 2012 John Wiley & Sons Ltd
I ⁄ R is initiated, and the CG components do exist in the rat
plasma over the observation of microcirculation when
given 90 minutes before ischemia. Brain edema following
I ⁄ R is the consequence of BBB disruption. Various mecha-
nisms are implicated in maintaining the integrity of BBB,
among which endothelium, basal lamina, and glial cell pro-
cesses are thought to be the prominent contributors
[1,11,25,27]. We did not undertake an experiment in the
present study to discriminate at which target CG exerted
action to attenuate brain edema. However, increasing evi-
dence demonstrates the close link between leukocyte
recruitment and breakdown of BBB [8,14]. The accumu-
lated leukocytes in postcapillary venules not only increase
the hydrostatic pressure within the vessels but also release
active oxygen species, which, in turn, impair the vessel wall
[2,9,10,20,26]. Thus, the preventing effect of CG on I ⁄ R-
induced leukocyte adhesion in venules found in the present
study, as well as in previous work [30], may account, at
least in part, for its favorite role in mitigating brain edema.
The CG is originally developed for treatment of cerebro-
vascular disturbance, such as headache and dizziness, which
is mostly associated with global cerebral ischemia. In an
animal model of global cerebral ischemia, we recently
reported the potential of CG pretreatment on I ⁄ R-induced
cerebral cortex microcirculatory disturbance [30]. Microcir-
culatory disturbance after I ⁄ R triggers the disruption of
BBB, and is exaggerated by the subsequent brain edema. In
the present study, we showed the protective effect of CG
on the cerebral vasculature disorder in a focal cerebral I ⁄ Rmodel that is more relevant to ischemic stroke in clinic,
suggesting a potential use of CG for improving the out-
come of the patients suffering from stroke.
The intravital microscopy in the present study revealed
that I ⁄ R caused only a transient decrease in blood velocity
while it had no effect on the venular diameter. On the
other hand, however, the number of nonperfused capillaries
in the area under investigation increased considerably in
animals subjected to I ⁄ R, in comparison with control,
implying a reduced blood perfusion in the affected terri-
tory. The rationale for the occurrence of nonperfused capil-
laries in the reperfusion phase remains unclear. One
possible mechanism is the activated leukocytes built up in
AShamI/R60B
a1 a2 a3 a4
NE
L po
sitiv
e ds
(%)
I/RCG 0.4 + I/RCG 0.8 + I/R40
†
†#
Num
ber
of T
UN
cells
/5 fi
eld
20#
†#
†
m2
N
0C
b4b3b2b1400
D
†
sitiv
e ce
lls/m
m
c3 c4c1 c2200
300
†
†
†#
umbe
r of
pos
d2 d3 d4d1
0
100†
†
†# #
#
#
#†
†N
Caspase-3 P53 Puma
Figure 9. The effects of CG pretreatment on the number of TUNEL-positive cells and expression of apoptosis-related proteins. (A) Representative
photomicrographs of TUNEL-positive cells in the sham group (a1), I ⁄ R (a2), CG 0.4 + I ⁄ R group (a3), and CG 0.8 + I ⁄ R group (a4). (B) The
quantitative analysis of TUNEL-positive cells. (C) Representative images of immunohistochemistry staining for caspase-3 (b), p53 (c), and PUMA (d) in
the sham group (1), I ⁄ R group (2), CG 0.4 + I ⁄ R group (3), and CG 0.8 + I ⁄ R group (4). The arrows indicate positive immunohistochemically stained
cells. (D) A quantification of immunohistochemistry staining positive cells for caspase-3, p53, and PUMA in different groups. Bar = 50 lm. Data are
mean ± SE from six animals. �p < 0.05 vs. sham group, #p < 0.05 vs. I ⁄ R group.
Chinese Medicine and Cerebral Reperfusion Injury
ª 2012 John Wiley & Sons Ltd 269
the capillaries, obstructing the blood flux. Nonetheless,
these nonperfused capillaries will exaggerate ischemia, and,
along with the adherent leukocytes in venules and BBB dis-
ruption, lead to neuron damage. Interestingly, CG pretreat-
ment attenuated I ⁄ R-evoked capillary nonperfusion as well,
particularly at the dose of 0.8 g ⁄ kg.
We have previously reported that administration of CG to
gerbils ameliorated hippocampal CA neuron damage caused
by global I ⁄ R [28]. Likewise, CG was found to protect cortex
neurons from focal cerebral I ⁄ R injury in the current study,
as documented by the results of TTC staining, Nissl staining,
and electron microscopy. In the acute phase of ischemia, cell
death in the ischemic core is commonly considered necrotic,
whereas after a short period of cerebral ischemia followed by
reperfusion, the neurons in the penumbral regions undergo
another wave of delayed cell death, which is dominated by
apoptosis [4,32]. Consistent with others, we observed a dras-
tic increase in apoptosis 24 hours after reperfusion in the
penumbral regions of cortex in response to I ⁄ R challenge, as
evidenced by the increase in the number of both TUNEL-
positive and caspase-3-positive cells. Noticeably, the
enhanced apoptosis in the cerebral cortex evoked by I ⁄ R was
ameliorated significantly by pretreatment with CG. Neuron
death is the consequence of a cascade of reactions in
response to I ⁄ R challenge, which take place sequentially and
interplay with each other. The complexity of neuron death
process is superimposed by the multiple components of CG,
which renders it extremely difficult to elucidate the mecha-
nism underlying the effect of CG on I ⁄ R-induced neuron
injury. However, some clues emerge that may plausibly shed
light on this issue. In view of the critical importance of oxi-
dative stress in the initiation of neuron death process, the
antioxidant potential of CG may contribute most to the
observed attenuation effect on neuron death. To this end,
we have previously reported that CG is able to ameliorate
I ⁄ R-induced hydrogen peroxide production [30]. Consistent
with this finding, available evidence from in vitro studies
revealed that at least six chemicals derived from the com-
posed herbs exhibit antioxidant potential; they are: paeoni-
florin from Radix paeoniae alba [22], ligustrazine from
Rhizoma chuan xiong [7], ferulic acid from Radix angelicae
sinensis [23], rehmannioside from Radix rehmanniae prepara-
A1 B1 C1
A2 B2 C2
A3 B3 C3
E
,
E
S
N
N
Figure 10. The effect of CG pretreatment on the ultrastructure of cerebral cortex in penumbral region of rats subjected to focal cerebral I ⁄ R. The
brains were processed for electron microscopy 24 hours after reperfusion. a1, a2, a3: sham group; b1, b2, b3: I ⁄ R group; c1, c2, c3: CG 0.8 + I ⁄ Rgroup. The electron micrographs of low magnification for different conditions are shown in Panel 1, where the capillaries and neurons enclosed in
rectangles are further enlarged in Panels 2 and 3. N, nuclei of neurons; S, swelling perivascular astroglials; E, swelling endothelial cells.
F. Wang et al.
270 ª 2012 John Wiley & Sons Ltd
ta [21], genistein from Caulis spatholobi [16], and alaternin
from Semen cassiae [5]. In the present circumstance, these
antioxidants most likely work coordinately, yielding an addi-
tive effect. Secondary, the CG-caused attenuations in cere-
bral microcirculation disturbance, BBB disruption and the
resultant brain edema, as observed in present study, may
lend themselves to blunt the development of neuron death,
as obstruction of these insults improves the cerebral oxygen
supply, on the one hand, and alleviates the inflammatory
reaction, on the other hand. Finally, the fact that CG attenu-
ated neuron apoptosis after I ⁄ R and, meanwhile, depressed
the I ⁄ R-enhanced expression of p53 and PUMA suggests
that the pathway of p53 through PUMA to caspases is at
least one of the targets at which CG acts to protect neuron
from apoptosis after I ⁄ R. We admit that much more works
are required to elucidate the mechanism. The speculations
above are to point to the likelihood for future exploration of
the rationale behind the CG action observed in the present
study.
In conclusion, CG is able to alleviate the brain edema
induced by focal cerebral I ⁄ R in addition to attenuation of
microcirculatory disturbance and neuron damage, provid-
ing a potentially alternative prophylactic management for
the patients in imminent danger of stoke.
PERSPECTIVE
Cerebral microcirculatory disturbance, neuron injury, and
brain edema following I ⁄ R is a complicated pathological
process involving multiple insults. To interfere with this
process, the efforts so far made are targeting a single insult,
and commonly with a poor outcome in clinic. The fact that
CG, a compound medicine consisting of numerous ingredi-
ents, is able to ameliorate I ⁄ R-induced cerebral microcircu-
latory disturbance, neuron injury, and brain edema
suggests a possibility of using a medicine containing multi-
ple components as a protective strategy for the patients in
imminent danger of stoke in clinical practice.
ACKNOWLEDGMENTS
This study was supported financially by the Production of
New Medicine Program of Ministry of Science and Tech-
nology of the People’s Republic of China (2008ZX09401).
REFERENCES
1. Abbruscato TJ, Davis TP. Combination of
hypoxia ⁄ aglycemia compromises in vitro
blood-brain barrier integrity. J Pharmacol
Exp Ther 289: 668–675, 1999.
2. Cardio MC, Lucchesi BR, Werns SW, Fan-
ton JC. The role of the neutrophil and
free radicals myocardial injury in ischemic.
Cell 1251: 1241–1251, 1989.
3. Chen CH, Hu Q, Yan JH, Lei JL, Qin L, Shi
XZ, Luan L, Yang L, Wang K, Han JY,
Nanda A, Zhou CM. Multiple effects of
2ME2 and D609 on the cortical expres-
sion of HIF-1alpha and apoptotic genes in
a middle cerebral artery occlusion-induced
local ischemia rat model. J Neurochem
102: 1831–1841, 2007.
4. Choi DW. Ischemia-induced neuronal apop-
tosis. Curr Opin Neurobiol 6: 667–672,
1996.
5. Choi JS, Lee HJ, Kang SS. Alaternin, cas-
siaside and rubrofusarin gentiobioside,
radical scavenging principles from the
seeds of Cassia tora on 1,1-diphenyl-2-
picrylhydrazyl (DPPH) radical. Arch Pharm
Res 17: 462–466, 1994.
6. Churchill M, Grimm S, Reding M. Risks
of diuretic usage following stroke. Neu-
rorehab Neural Repair 18: 161–165,
2004.
7. Fan LH, Wang KZ, Cheng B, Wang CS,
Dang XQ. Anti-apoptotic and effects of
tetramethylpyrazine following spinal cord
ischemia in rabbits. BMC Neurosci 7: 48,
2006.
8. Ferrari CC, Depino AM, Prada F, Muraro
N, Campbell S, Podhajcer O, Perry VH,
Anthony DC, Pitossi FJ. Reversible demye-
lination, blood-brain barrier breakdown,
and pronounced neutrophil recruitment
induced by chronic IL-1 expression in the
brain. Am J Pathol 165: 1827–1837,
2004.
9. Granger DN. Role of xanthine oxidase and
granulocytes in ischemia-reperfusion
injury. Am J Physiol 255: H1269–H1275,
1998.
10. Harlan JM. Leukocyte-endothelial interac-
tions. Blood 65: 513–525, 1985.
11. Hawkins BT, Davis TP. The blood-brain
barrier ⁄ neurovascular unit in health and
disease. Pharmacol Rev 57: 173–185,
2005.
12. Hsiao G, Chen YC, Lin JH, Lin KH, Chou
DS, Lin CH, Sheu JR. Inhibitory mecha-
nisms of tetramethylpyrazine in middle
cerebral artery occlusion (MCAO) induced
focal cerebral ischemia in rats. Planta Med
72: 411–417, 2006.
13. Hu Q, Chen CH, Yan JH, Yang XM,
Shi XZ, Zhao J, Lei JL, Yang L, Wang K,
Chen L, Huang HY, Han JY, Zhang JH.,
Zhou CM. Therapeutic application of
gene silencing MMP-9 in a middle
cerebral artery occlusion-induced focal
ischemia rat model. Exp Neurol 216:
35–46, 2009.
14. Jin AY, Tuor UI, Rushforth D, Kaurl J, Mul-
ler RN, Petterson JL, Boutry S, Barberl PA.
Reduced blood brain barrier breakdown in
P-selectin deficient mice following tran-
sient ischemic stroke: a future therapeutic
target for treatment of stroke. BMC Neu-
rosci 11: 12, 2010.
15. Kao TK, Ou YC, Kuo JS, Chen WY, Liao
SL, Wu CW, Chen CJ, Ling NN, Zhang
YH, Peng WH. Neuroprotection by
tetramethylpyrazine against ischemic brain
injury in rats. Neurochem Int 48: 166–
176, 2006.
16. Liang HW, Qiu SF, Shen J, Sun LN, Wang
JY, Bruce IC, Xia Q. Genistein attenuates
oxidative stress and neuronal damage fol-
lowing transient global cerebral ischemia
in rat hippocampus. Neurosci Lett 438:
116–120, 2008.
17. Liao SL, Kao TK, Chen WY, Lin YS, Chen
SY, Raung SL, Wu CW, Lu HC, Chen CJ.
Tetramethylpyrazine reduces ischemic
brain injury in rats. Neurosci Lett 372: 40–
45, 2004.
18. Liu DZ, Xie KQ, Ji XQ, Ye ???, Jiang CL,
Zhu XZ. Neuroprotective effect of paeoni-
florin on cerebral ischemic rat by activat-
ing adenosine A1 receptor in a manner
different from its classical agonists. Br J
Pharmacol 146: 604–611, 2005.
19. Liu L, Yenari MA. Clinical application of
therapeutic hypothermia in stroke. Neurol
Res 31: 331–335, 2009.
20. Matsuo Y, Onodera H, Shiga Y, Shozuha-
ra H, Ninomiya N, Kihara T, Tamatani T,
Chinese Medicine and Cerebral Reperfusion Injury
ª 2012 John Wiley & Sons Ltd 271
Miyasaka M, Kogure K. Role of cell adhe-
sion molecules in brain injury after tran-
sient middle cerebral artery occlusion in
the rat. Brain Res 656: 344–352, 1994.
21. Miao J, Wang W, Yao S, Navaratnam S,
Parson BJ. Antioxidative properties of
Martynoside: pulse radiolysis and laser
photolysis study. Free Radic Res 37: 829–
833, 2003.
22. Okubo T, Nagai F, Seto T, Satoh K, Ushiy-
ama K, Kano I. The inhibition of phenylhy-
droquinone-induced oxidative DNA
cleavage by constituents of moutan cortex
and paeoniae radix. Biol Pharm Bull 23:
199–203, 2000.
23. Patro BS, Rele S, Chintalwar GJ, Chatto-
padhyay S, Adhikari S, Mukherjee T.
Protective activities of some phenolic 1,3-
diketones against lipid peroxidation:
possible involvement of the 1,3-diketone
moiety. ChemBioChem 3: 364–370, 2002.
24. Rabadan AT, Sposato L, Mazia C. Update
on interventional treatment of acute
ischemic stroke. Medicina 70: 463–468,
2002.
25. Rosell A, Cuadrado E, Ortega-Aznar A,
Hernandez-Guillamon M, Lo EH, Monta-
ner J. MMP-9-positive neutrophil infiltra-
tion is associated to blood-brain barrier
breakdown and basal lamina type IV colla-
gen degradation during hemorrhagic
transformation after human ischemic
stroke. Stroke: J Cereb Circ 39: 1121–
1126, 2008.
26. Siesjo BK, Zhao Q, Pahlmark K, Siesjo P,
Katsura K, Folbergrova J. Glutamate, cal-
cium, and free radicals as mediators of
ischemic brain damage. Ann Thorac Surg
59: 1316–1320, 1995.
27. Stoll G, Jander S, Schroeter M. Inflamma-
tion and glial responses in ischemic brain
lesions. Prog Neurobiol 56: 149–171,
1998.
28. Sun K, Hu Q, Zhou CM, Xu XS, Wang F,
Hu BH, Zhao XY, Chang X, Chen CH, Hu-
ang P, An LH, Liu YY, Fan JY, Wang CS,
Yang L, Han JY. Cerebralcare Granule�, a
Chinese herb compound preparation,
improves cerebral microcirculatory disor-
der and hippocampal CA1 neuron injury
in gerbils after ischemia-reperfusion. J Eth-
nopharmacol 130: 398–406, 2010.
29. Wang Q, Chen SY, Xiong LZ, Jin WL,
Yang J. Neuroprotective effect of sodium
ferulate on transient focal cerebral ische-
mia by weakening activation of postsyn-
aptic density-95 in rats. Chin J Traumatol
8: 297–302, 2005.
30. Xu XS, Ma ZZ, Wang F, Hu BH, Wang CS,
Liu YY, Zhao XR, An L, Chang X, Liao FL,
Fan JY, Niimi H, Han JY. The antioxidant
Cerebralcare Granule(R) attenuates cere-
bral microcirculatory disturbance during
ischemia-reperfusion injury. Shock 32:
201–209, 2008.
31. Yu K, Little D, Plumb R, Smith B. High-
throughput quantification for a drug mix-
ture in rat plasma-a comparison of Ultra
Performance liquid chromatography ⁄ tan-
dem mass spectrometry with high-perfor-
mance liquid chromatography ⁄ tandem
mass spectrometry. Rapid Commun Mass
Spectrom 20(4): 544–52, 2006.
32. Yuan JY. Neuroprotective strategies tar-
geting apoptotic and necrotic cell death
for stroke. Cell 14: 469–477, 2009.
33. Zhao N, Liu YY, Wang F, Hu BH, Sun K,
Chang X, Pan CS, Fan JY, Wei XH, Li X,
Wang CS, Guo ZX, Han JY. Cardiotonic
pills, a compound Chinese medicine, pro-
tects ischemia-reperfusion-induced micro-
circulatory disturbance and myocardial
damage in rats. Am J Physiol Heart Circ
Physiol 298(4): H1166–H1176, 2010.
F. Wang et al.
272 ª 2012 John Wiley & Sons Ltd