Stabilization of mitochondrial function by tetramethylpyrazine protects against kainate-induced oxidative lesions in the rat hippocampus

Article abstract of DOI:10.1016/j.freeradbiomed.2009.12.004

Mitochondria are critical regulators of cell death, a key feature of neurodegeneration. Reactive oxygen species (ROS) are crucial to Ca²⁺-mediated effects of glutamate receptor activation leading to neuronal degeneration. Tetramethylpyrazine (TMP) is a principal ingredient of Ligusticum wallichi Franchat (a Chinese herb), used for treatment of cardiovascular and cerebrovascular ischemic diseases. However, its protection against oxidative brain injury associated with excessive activation of glutamate receptors is unknown. In this study, we demonstrate TMP neuroprotection against kainate-induced excitotoxicity in vitro and in vivo. We found that TMP could partly alleviate kainate-induced status epilepticus in rats and prevented and rescued neuronal loss in the hippocampal CA3 but not the CA1 region. The partial prevention and rescue of neuronal loss by TMP were attributable to the preservation of the structural and functional integrity of mitochondria, evidenced by maintaining the mitochondrial membrane potential, ATP production, and complex I and III activities. Stabilization of mitochondrial function was linked to the observation that TMP could function as a reductant/antioxidant to quench ROS, block lipid peroxidation, and protect enzymatic antioxidants such as glutathione peroxidase and glutathione reductase. These results suggest that TMP may protect against oxidative brain injury by stabilization of mitochondrial function through quenching of ROS.

Full text of DOI:10.1016/j.freeradbiomed.2009.12.004

Free Radical Biology & Medicine 48 (2010) 597–608
Contents lists available at ScienceDirect
Free Radical Biology & Medicine
journal homepage: www.elsevier.com/locate/freeradbiomed
Original Contribution
Stabilization of mitochondrial function by tetramethylpyrazine protects against
kainate-induced oxidative lesions in the rat hippocampus
Shu-Yan Li a, , Yu-Hong Jia a,1, Wen-Ge Sun a,2, Yuan Tang a, Guo-Shun An a, Ju-Hua Ni a, Hong-Ti Jia a,b
⁎
a
Department of Biochemistry and Molecular Biology, Peking University Health Science Center, Beijing 100191, People's Republic of China
Department of Biochemistry and Molecular Biology, Capital Medical University, Beijing 100069, People's Republic of China
b
a r t i c l e i n f o
a b s t r a c t
Article history:
Received 22 July 2009
Revised 4 November 2009
Accepted 7 December 2009
Available online 23 December 2009
Mitochondria are critical regulators of cell death, a key feature of neurodegeneration. Reactive oxygen
species (ROS) are crucial to Ca2+-mediated effects of glutamate receptor activation leading to neuronal
degeneration. Tetramethylpyrazine (TMP) is a principal ingredient of Ligusticum wallichi Franchat (a Chinese
herb), used for treatment of cardiovascular and cerebrovascular ischemic diseases. However, its protection
against oxidative brain injury associated with excessive activation of glutamate receptors is unknown. In this
study, we demonstrate TMP neuroprotection against kainate-induced excitotoxicity in vitro and in vivo. We
found that TMP could partly alleviate kainate-induced status epilepticus in rats and prevented and rescued
neuronal loss in the hippocampal CA3 but not the CA1 region. The partial prevention and rescue of neuronal
loss by TMP were attributable to the preservation of the structural and functional integrity of mitochondria,
evidenced by maintaining the mitochondrial membrane potential, ATP production, and complex I and III
activities. Stabilization of mitochondrial function was linked to the observation that TMP could function as a
reductant/antioxidant to quench ROS, block lipid peroxidation, and protect enzymatic antioxidants such as
glutathione peroxidase and glutathione reductase. These results suggest that TMP may protect against
oxidative brain injury by stabilization of mitochondrial function through quenching of ROS.
Keywords:
Tetramethylpyrazine
Neuroprotection
Kainate
Reactive oxygen species
Mitochondrial dysfunction
Free radicals
© 2009 Elsevier Inc. All rights reserved.
Oxidative stress is a causal, or at least an ancillary, factor in the
neuropathology of several adult neurodegenerative disorders and in
stroke, trauma, and seizures. Glutamate receptors mediate excitatory
neurotransmission, and excessive or persistent activation of gluta-
Apoptosis is implicated in neurodegenerative disorders such as
Alzheimer disease, Parkinson disease, amyotrophic lateral sclerosis,
and Huntington disease [4]. Eliciting a rise in intracellular free Ca2+ is
believed to play a particularly important role in glutamate-induced
neuronal death [5]. The abnormal influx of intracellular Ca2+ activates
a number of proteases, phospholipases, and endonucleases by
generating ROS that attack the cellular membrane [6,7] and induce
apoptosis [5,6]. The levels of Ca2+ and ROS in the cells are regulated by
mitochondria and are closely linked to cell survival and death. Cell
survival under oxidative stress depends on the levels of intracellular
ROS, whose formation is closely related to the balance between
oxidants and antioxidants [8,9]. Disruption of the balance, for
instance, depletion of glutathione (GSH), may result in increased
ROS, promoting cellular damage [9]. By contrast, ROS-quenching
agents can effectively protect neurons against excitotoxicity [2].
2,3,5,6-Tetramethylpyrazine (TMP; also known as ligustrazine) is
a principal ingredient of Ligusticum wallichi Franchat (Chuan Xiong), a
promising traditional Chinese herb that is widely used for treatment
of patients with cardiovascular diseases and cerebrovascular ischemic
diseases in China and Japan [10–12]. Animal experiments have shown
that TMP suppresses epilepsy discharge induced by pentylenetetrazol
[13] and protects against spinal cord ischemia [14]. The study of
mechanisms reveals that TMP may decrease platelet aggregation,
inhibit thrombus formation, and improve brain microcirculation in
rats [15]. Recent studies have demonstrated that alleviation of the
mate receptors may cause neuronal degeneration through Ca2+
-
mediated effects that may induce oxidative stress in a number of
pathways [1,2]. In the models of excitotoxicity and seizures, kainate,
an exogenous glutamate analogue, increases production of reactive
oxygen species (ROS). ROS can damage a variety of biomolecules such
as lipids, proteins, carbohydrates, and nucleic acids, leading to various
types of cellular dysfunction. Of the cellular dysfunctions, mitochon-
drial dysfunction has a causal role in all major neurodegenerative
diseases [3].
Abbreviations: AMPA, α-amino-3-hydroxy-5-methyl-4-isoxazolepropionic acid;
CTZ, cyclothiazide; D-AP5, D-2-amino-5-phosphonopentanoic acid; DHE, dihydroethi-
dium; GPX, glutathione peroxidase; GR, glutathione reductase; GSH, glutathione; HRP,
horseradish peroxidase; MTT, 3-(4,5-dimethylthiozol-2-yl)-2,5-diphenyltetrazolium
bromide; MDA, malondialdehyde; NBQX, 2,3-dihydroxy-6-nitro-7-sulfamoylbenzo(f)
quinoxaline; NMDA, N-methyl-D-aspartate; SOD, superoxide dismutase; ROS, reactive
oxygen species; TMP, tetramethylpyrazine.
⁎
Corresponding author. Fax: +86 010 82801434.
E-mail address: shuyanli@bjmu.edu.cn (S.-Y. Li).
Present address: Department of Pathophysiology, Dalian Medical University, Dalian
1
116027, People’s Republic of China.
2
Present address: Department of Biochemistry, Medical School of Chifeng College,
Chi Feng 024001, People’s Republic of China.
0891-5849/$ – see front matter © 2009 Elsevier Inc. All rights reserved.
doi:10.1016/j.freeradbiomed.2009.12.004

598
S.-Y. Li et al. / Free Radical Biology & Medicine 48 (2010) 597–608
damage in brain and spinal cord by TMP is linked to the scavenging of
free radicals [14]. In addition, TMP neuroprotection may involve anti-
inflammatory potential [16] and increased transcription of thiore-
doxin [17]. Therefore, the mechanisms of TMP actions are complicated
and uncertain. In this study, we focus on the effect of TMP on
mitochondria and demonstrate that TMP administration can alleviate
kainate-induced status epilepticus and partially prevent and rescue
neuronal loss in the rat hippocampus. The neuroprotective effective-
ness of TMP is attributed to the stabilization of mitochondrial function
in the cells through scavenging ROS.
containing 1% sodium hydroxide in 80% alcohol for 5 min. After a 2-
min rinse in 70% (v/v) alcohol and 2-min rinse in dH2O, sections were
transferred to a solution of 0.06% (w/v) KMnO4 for 15 min. After a 2-
min rinse in dH2O, slides were incubated in 0.0004% (w/v) Fluoro-
JadeB solution in 0.1% (v/v) acetic acid for 20 min at 4°C. After three
washes in dH2O, sections were air-dried and mounted. Sections were
examined with a laser confocal microscope (Leica Microsystems,
Germany).
Primary neuronal cultures and kainate exposure
Materials and methods
The primary neuronal cultures from hippocampus were prepared
as described previously [21,22], with modification. Briefly, hippocam-
pi were removed from newborn Wistar rat pups (postnatal day 1),
and dissected hippocampi were digested at 37°C for 15 min in 0.25%
trypsin. Hippocampi were then dissociated using fire-polished
pipettes and plated at a density of 5×105 cells/cm2 in DMEM with
20% fetal bovine serum onto 96-well or 100-mm dishes precoated
with poly-L-lysine and grown in a 37°C incubator with 5% CO2. After
the cells attached to the plates, the culture medium was replaced with
Neurobasal medium supplemented with 2% (v/v) B27. Hippocampal
neurons were cultured for 12–14 days before kainate (100 μM)
exposure. Vitamin E (50 μM) was added twice at 2 (20 μM) and 0.5 h
(50 μM) before kainate exposure. TMP was added at 2 (25 μM) and
0.5 h (50 μM) before kainate treatment.
Reagents and materials
Kainate, 2,3-dihydroxy-6-nitro-7-sulfamoylbenzo(f)quinoxaline
(NBQX), D-2-amino-5-phosphonopentanoic acid (D-AP5), and
cyclothiazide (CTZ) were from Tocris Cookson (Bristol, UK). α-
Tocopherol (vitamin E) and horseradish peroxidase (HRP) were
from Sigma (St. Louis, MO, USA). TMP was from Sigma–Aldrich
(Shanghai, China). Fetal bovine serum, Dulbecco's modified Eagle
medium (DMEM), Neurobasal medium, and B27 were from Invitrogen
(Grand Island, NY, USA). Culture dishes and plates were purchased
from Corning Glass (Corning, NY, USA). Other chemicals were
obtained from Sigma, unless stated otherwise.
Animal experiments
Cell viability assays
Animal experimental protocols were approved by the Institutional
Committee on Animal Research and were carried out in accordance
with the National Institutes of Health guidelines for animal use and
care (National Institutes of Health Publication No. 96-23, revised
1996). Male Wistar rats (7 weeks of age, 180 and 200 g) were supplied
by the Experimental Animal Center, Peking University Health Science
Center, and randomly divided into four groups for the preventive
experiments. Rats in the kainate-treated group (KA group) received
intraperitoneal (ip) injection of a single dose of kainate (10 mg/kg) in
0.9% NaCl as described [18]. Rats in the control group (saline group)
received ip injection of 0.9% (w/v) NaCl of an equal volume. Rats in
the TMP and kainate-treated group (TMP + KA group) were
pretreated with a single ip injection of TMP (20 mg/kg/day) [17] 12
h before kainate injection, followed by ip injection of the same dose
TMP, once a day for 1 week after kainate injection. Rats in the TMP
group were injected with TMP in the same way. The severity of
seizures was scored for 4 h after kainate injection, according to a
previous classification [19]. In the rescue experiments, the rats in the
KA + TMP group were treated with TMP (20 mg/kg/day) after 4 h of
kainate injection, once a day for 1 week. One week after kainate
injection, the animals were anesthetized with pentobarbital sodium
(30 mg/kg ip) and transcardially perfused with 4% paraformaldehyde
in 0.1 M phosphate buffer. Brains were removed and postfixed
overnight in the same fixative at 4°C. Coronal brain sections (20 μm
thickness) were cut on a freezing microtome and stained with cresyl
violet for histological examination of neuronal damage. Nissl-positive
undamaged neurons were counted in five coronal brain sections per
animal (sections were chosen by unbiased sampling), and the mean
number of cells per section was determined such that the value
obtained for each rat represents an average total number of neurons
counted per section (0.04 mm2 in the CA1/CA3 regions). Counts were
performed by an investigator blinded to treatment status.
Cell viability assays were performed as described [22]. After
exposure for the given number of hours, neurons were assayed for
viability, using 3-(4,5-dimethylthiozol-2-yl)-2,5-diphenyltetrazolium
bromide (MTT; Sigma), which was added at a final concentration of
1 mg/ml for 4 h. MTT was removed, and neurons were lysed in 200 μl
of dimethyl sulfoxide. Absorbance was measured at 570 nm on a Bio-
Rad 680 microplate reader (Hercules, CA, USA). The data are
expressed as the percentage of unexposed neurons that remained in
the presence of kainate.
Isolation of mitochondria from the rat hippocampus
For analyses of ATP levels and complex I and III activities, isolation
of functional mitochondria from the rat hippocampus was performed
as described previously [23] with some modifications. Hippocampi
were dissected and washed with ice-cold Ca2+-free and Mg2+-free
phosphate-buffered saline solution supplemented with 2 mM EGTA.
Hippocampi were rapidly immersed in 1 ml of ice-cold mitochondrial
isolation buffer containing 210 mM mannitol, 70 mM sucrose, 10 mM
Hepes–KOH, pH 7.4, 2 mM EGTA, and 0.1% fatty acid-free bovine
serum albumin and then homogenized using a Dounce homogenizer
(10 passes with a loose pestle and 10 passes with a tight pestle). After
homogenate centrifugation for 10 min at 1060 g at 4°C, the
supernatant was collected. The pellet was resuspended in the
isolation buffer and centrifuged at 1060 g for 5 min. The first and
second supernatants were pooled together and centrifuged at
14,600 g for 10 min at 4°C. The mitochondrial pellet was resuspended
in the isolation buffer lacking EGTA and albumin and centrifuged at
14,600 g for 10 min. All of the isolation steps were performed on ice.
Protein concentration was measured using a BCA protein assay kit
(Pierce, Rockford, IL, USA).
Enzyme assays
Fluoro-JadeB staining
The activities of respiratory chain complexes were determined as
described previously [24] with modifications using the respective assay
kit (Genmed, Shanghai, China). NADH:ubiquinol reductase (complex I)
was measured by the decylquinazolinamine-sensitive oxidation of
To evaluate neuronal degeneration, Fluoro-JadeB (Chemicon,
Temecula, CA, USA) staining was performed as described [20].
Sections were dried on a slide warmer and dipped in solution

S.-Y. Li et al. / Free Radical Biology & Medicine 48 (2010) 597–608
599
NADH (340–400 nm, and ∈340–400nm =5.5 mM−1 cm−1) using
decylubiquinone as electron acceptor. The reaction mixture contained
60 μmol NaPHO4, (pH 7.5), 30 μmol NaN3, 0.15 μmol EDTA, 1.3 mg
bovine serum albumin, 0.033 μmol NADH, and 0.063 μmol coenzyme Q
in a final volume of 1 ml. Ubiquinol:cytochrome c reductase (complex
III) was measured by the antimycin-sensitive reduction of cytochrome c
(550–540 nm, ∈550–540nm =21.84 mM−1 cm−1). The reaction mixture
was the same as in the assay of NADH:ubiquinol reductase. The
substrate was 0.063 μmol coenzyme QH, and the acceptor was
0.058 μmol cytochrome c.
Determination of glutathione peroxidase (GPX) and glutathione
reductase (GR) levels in rat hippocampus and primary hippocampal
neurons was performed as described previously [25] using the
respective assay kits (bioEol, Nanjing, China). Protein contents of
samples were determined with a BCA protein assay kit on aliquots of
homogenate before centrifugation. Determination of GR activity was
based on the equation A340 min/6220 mol/L−1 cm−1 =mU/ml,
where A340 is the change in absorbance per minute at 340 nm, and
6220 mol/L−1 cm−1 is the molar extinction coefficient of NADPH.
Determination of GPX activity was achieved by the equation
A340 min/6220 mol/L−1 cm−1 =mU/ml, where A340 is the change
in absorbance per minute at 340 nm, and 6220 mol/L−1 cm−1 is the
molar extinction coefficient of NADPH. Results for both GR and GPX
assays were then adjusted to give data as mU/mg protein. One GR
unit reduces 1 μmol of oxidized glutathione (GSSG) per minute at
25°C and pH 7.6. One GPX unit consumes 1 μmol NADPH per minute
at 25°C and pH 8.0.
measured by flow cytometry (excitation, 488 nm; emission, 525 nm)
using a FACSCalibur flow cytometer (BD Biosciences, San Jose, CA,
USA). Data are expressed as the mean linear channel number from
three to five experiments, with triplicate tubes within one
experiment.
Wavelength spectra of TMP and TMP-oxide
Wavelength spectra of TMP and oxidative TMP within 200–
700 nm were recorded on a UV/Vis 2800H photometer (UNICO
Instrument, Shanghai, China). TMP (0.2 mM) was dissolved in
100 mM phosphate buffer (pH 6.8) and incubated with hydrogen
peroxide (3.0, 6.0, 12.0 mM) and HRP at 37°C for 60 min. The buffer
used for dilutions and for the blanks was 100 mM sodium phosphate
(pH 6.8). Spectra were recorded at room temperature (25°C). In a 1-
ml reaction mix, the final concentrations were 100 mM sodium
phosphate, 0.2 mM TMP, 3.0–12 mM hydrogen peroxide, and 50 units
of HRP.
TMP oxidation by hydrogen peroxide
In the reaction catalyzed by HRP, hydrogen peroxide was
reduced (to H2O) at the expense of TMP as a substance that is
oxidized to TMP-N,N′-dioxide (Fig. 5A). Based on this reaction, the
oxidation of TMP by hydrogen peroxide in the presence of HRP was
monitored by the decrease in TMP. Briefly, hydrogen peroxide
solution ranged from 3 to 15 mM mixed with 0.2 mM TMP and HRP
(50 units/ml) dissolved in 100 mM phosphate buffer (pH 6.8). The
reaction containing hydrogen peroxide and HRP was used as a blank
(adjusted to 0), and the reaction containing TMP and HRP was used
as control. The mixture was shaken up vigorously and left at 37°C
for 60 min in the dark. The UV absorbance of reactions was measured
at 294 nm. The oxidative percentage of TMP was calculated by
dividing the absorbance difference between the control (AC) and the
reaction (AR) by the absorbance of the control, as the in the following
equation:
Adenosine triphosphate (ATP) level measurements
ATP was determined luminometrically as described previously
[26] using an ATP bioluminescence assay kit (bioEol). Following the
provided protocol, the ATP amount of mitochondrial samples was
determined by a concentration standard curve, and the values of ATP
content were normalized according to the protein concentration for
each sample (mol ATP/μg protein) and converted to percentage of
untreated controls.
TMP oxidation ð%Þ = ½ðAC − ARÞ= ACꢀ × 100
Determination of lipid peroxidation
Data analyses
Lipid peroxidation in rat hippocampus and cultured hippocampal
neurons was assessed by measuring the concentration of malondial-
dehyde (MDA) as previously described [27]. Briefly, hippocampus or
cell extracts were prepared and incubated with SDS, acetic acid, and
thiobarbituric acid at 95°C for 1 h. 1,1,3,3-Tetraethoxypropane was
used as a standard. The protein concentration of hippocampus and cell
extracts was determined using the BCA protein assay kit, and lipid
peroxidation was calculated as nanomoles of MDA per milligram of
protein.
Multiple comparisons were examined for significant differences
using a one-way analysis of variance, followed by individual
comparisons with the Tukey posttest. Statistical significance was set
at Pb0.05.
ROS measurement in cultured hippocampal neurons
Production of ROS in neurons was determined as described [28] by
using dihydroethidium (DHE; Molecular Probes, Eugene, OR, USA).
Cultures were incubated in 10 μM DHE at 37°C for 30 min, washed
with 1× D-Hanks', and detected with flow cytometry. Hydrogen
peroxide (250 μM) was added directly to the bathing medium, and the
cells were incubated for 10 min at 37°C as positive controls.
Fig. 1. Alleviation of kainate-induced convulsive behavior in rats by TMP. Male Wistar
rats were grouped (n=5 or 6/group) and pretreated (ip injection) with TMP
(20 mg/kg) or NaCl (0.9%). Twelve hours after TMP or saline pretreatment, seizures
in kainate (KA) and TMP + KA group rats were induced by injection of kainate
(10 mg/kg); the rats in the saline group received 0.9% NaCl in the equal volume. The
Measurement of mitochondrial transmembrane potential
Mitochondrial transmembrane potential was measured as de-
scribed previously [29]. Primary hippocampal neurons were incubat-
ed for 30 min at 37°C with PBS containing 5 μM rhodamine 123
(R123). After being washed with PBS, cells were trypsinized at room
temperature and resuspended in PBS, and the fluorescence was
preconvulsive behavior was scored within
4 h after kainate injection. Scores at
various time points after seizure induction were analyzed by two independent
observers and averaged and statistically compared with one another by two-tailed t
tests. Data represent means SD. ⁎Pb0.05.

600
S.-Y. Li et al. / Free Radical Biology & Medicine 48 (2010) 597–608
Results
pretreatment, seizures were induced by kainate injection (10 mg/kg)
or 0.9% NaCl (saline group). The preconvulsive behavior was scored
within 4 h after kainate injection. Compared with the saline group rats
in which no neurologic anomaly could be observed after saline
induction, the preconvulsive behavior in kainate-induced rats
occurred initially within 30 min; the seizure scores reached 2.88 at
1 h and substantially increased to 3.7–4 within 3.5 h after kainate
injection. Obviously, a delayed onset and decreased score of seizure
were observed in the rats receiving TMP pretreatment and kainate
TMP alleviates kainate-induced convulsive behavior in rats
To evaluate protection of TMP against kainate-induced status
epilepticus, male Wistar rats (180–200 g) in the TMP + KA group
were pretreated with a single ip injection of TMP (20 mg/kg); saline
and saline + KA group rats were injected with equal volumes of 0.9%
NaCl solution instead of TMP. Twelve hours after TMP or saline
Fig. 2. Kainate-induced neuronal loss in the rat hippocampus. Seizures in male Wistar rats were induced by kainate injection (10 mg/kg, ip). Animals were anesthetized with pentobarbital
sodium and transcardially perfused. Brains were removed and postfixed. Coronal brain sections (20-μm thickness) were cut on a freezing microtome and stained with cresyl violet (Nissl)
or Fluoro-JadeB. (A) Representative histological examples of cresyl violet- (images a–l) and Fluoro-JadeB- (images m–p) stained brain sections. Saline, saline “induction;” 12 h, 36 h, and 1
w indicate the time after kainate induction. Bar, 1 mm or 100 μm as indicated. (B) Histogram showing neuronal densities in the CA1 and CA3 regions of rat hippocampus measured by Nissl
staining and cell counting (cell number/0.04 mm2). Data represent means SD (n=5 or 6). ⁎Pb0.05 (kainate vs saline group); ⁎⁎Pb0.05 (CA3 36 h vs 1 week).

S.-Y. Li et al. / Free Radical Biology & Medicine 48 (2010) 597–608
601
induction, and the seizure values reached a maximal score of 2.6 at 3
h; thereafter the seizure score dropped to 1.8 (Fig. 1). These results
indicate that TMP may alleviate kainate-induced status epilepticus.
excitotoxicity, we first examined the cell lesions in the CA1 and CA3
regions of kainate-injected rats by histochemistry. The cresyl violet
staining (Fig. 2A) and the counts of neuron numbers in the two regions
(Fig. 2B) showed that kainate caused a marked loss of pyramidal
neurons in the CA1 region at 36 h (images c and g in Fig. 2A) and kept a
steady decrease by 1 week (images d, h) after kainate-induced seizure.
Unlike the CA1 region, neuronal loss in the CA3 region was not so serious
at 36 h, and a substantial loss of neurons was detected on day 7 after
kainate-induced seizure (images d and l), which was also evidenced by
TMP prevents kainate-induced neuronal loss in the CA1 and CA3 regions
Kainate-induced status epilepticus leads to delayed selective death
of pyramidal neurons in the rat hippocampus [30]. To further evaluate
the role of TMP in preventively protecting neurons against kainate
Fig. 3. Partial prevention of kainate-induced neuronal loss in the hippocampus by TMP. Wistar rats were pretreated with TMP (20 mg/kg, ip) or NaCl (0.9%), and seizures were
induced by kainate (10 mg/kg, ip) at 12 h after pretreatment, followed by TMP injection (20 mg/kg, ip), once a day for 1 week. Brain sections (20-μm thickness) were prepared and
stained as for Fig. 2. (A) Representative histological examples of cresyl violet- (images a–l) or Fluoro-JadeB- (images m–p) stained brain sections. Saline (control), saline
pretreatment/saline “induction,” n=5; KA, saline pretreatment/kainate induction, n=6; TMP + KA, TMP pretreatment/kainate induction, n=6; TMP, TMP pretreatment/saline
“induction,” n=5. Samples were from rats 1 week after kainate induction. Bar, 1 mm or 100 μm as indicated. (B) Histogram showing relative neuron densities in the CA1 and CA3
regions of the hippocampus measured by Nissl staining and cell counting (cell number/0.04 mm2). The saline group number is normalized as 1 (100%). Data represent means SD.
⁎Pb0.05 (vs control group).

602
S.-Y. Li et al. / Free Radical Biology & Medicine 48 (2010) 597–608
Fluoro-JadeB staining (images m to p). These results indicate that
kainate injection may damage pyramidal neurons in the CA1 and CA3
regions, of which the CA1 region seems to be more vulnerable than CA3.
Next, we compared neuronal loss in the hippocampus among the
four groups of rats. The section staining was performed using rats 1
week after induction. The saline (control) and TMP pretreatment did
not affect the distribution of neurons in tested areas (images a and d in
Figs. 3A and 3B). Compared with the control, kainate treatment led to
a loss of the pyramidal neuron numbers in the CA1 (images b and f)
and CA3 (images b and j) regions after 1 week of 79.3 and 67.3% of the
control, respectively. Notably, TMP pretreatment plus kainate injec-
tion resulted in only 39.2 and 0% of kainate-induced neuronal loss in
these two regions (images g and k). Similar effects of the reagents on
the CA3 region were also observed by Fluoro-JadeB staining (images
m to p). These results reveal that TMP pretreatment can partially
prevent kainate-induced cell death in the hippocampus.
neuronal loss in the CA1 region and 67.3% loss in CA3 (images b, f, and
j in Figs. 4A and 4B), compared with that in the hippocampus of
control (saline treated) rat (images a, e, and i). It was noteworthy that
administration of TMP after kainate treatment for 4 h rescued about
27.2% of the neuronal loss in the CA3 region, but did not reverse the
neuronal loss in the CA1 (images c, g, k). These results indicate that
TMP may rescue kainate-induced neuronal loss in the CA3 but not the
CA1 region of rat hippocampus.
TMP can function as a reductant to destroy hydrogen peroxide
TMP can be oxidized in vitro to TMP-1,4-(N,N′)-dioxide in the
presence of hydrogen peroxide at high temperature (70°C) [31]. To
investigate whether TMP might directly destroy hydrogen peroxide in
living cells, we mimicked the intracellular conditions for developing
an oxidation–reduction reaction catalyzed by HRP at 37°C for 60 min,
in which hydrogen peroxide was reduced at the expense of TMP (Fig.
5A). The spectra of the reactions in the presence of hydrogen peroxide
at various concentrations (3–12 mM) were recorded at the wave-
lengths 200–700 nm. In the reaction containing TMP and HRP, a UV
spectrum with double peaks (at 282 and 294 nm) was detected within
250–320 nm (Fig. 5B). When hydrogen peroxide was added, the
maximal absorption was markedly decreased in a hydrogen peroxide
dose-dependent manner, such that the decrease in UV absorption
might be due to oxidation of TMP to TMP-dioxide by hydrogen
TMP rescues kainate-induced neuronal loss in the CA3 region
To ascertain whether TMP could rescue kainate-induced neuronal
loss in the hippocampus, we administered TMP (20 mg/kg/d, ip) after
4 h of kainate injection, once a day for 1 week, and examined the
neuronal loss in the CA1 and CA3 regions. The cresyl violet staining
sections (Fig. 4A) and the counts of neuron numbers (Fig. 4B) in rat
brain sections showed that kainate treatment led to about 79.3%
Fig. 4. Rescue of kainate-induced neuronal loss in the CA3 but not the CA1 region by TMP. Seizures in Wistar rats were induced by kainate (10 mg/kg, ip), followed by treatment with
or without TMP (20 mg/kg/day, ip), once a day for 1 week. (A) Brain sections were prepared and stained with cresyl violet staining. Bar, 500 or 100 μm as indicated. (B) The neuronal
numbers per 0.04 mm2 in the CA1 and CA3 regions were calculated. Saline (control), saline treatment, n=5; KA, kainate induction/saline treatment, n=6; KA + TMP, kainate
induction/TMP treatment, n=6; TMP, TMP treatment, n=5. Samples were from rats 1 week after kainate induction. Data represent means SD. ⁎Pb0.05.

S.-Y. Li et al. / Free Radical Biology & Medicine 48 (2010) 597–608
603
Fig. 5. Direct reduction of hydrogen peroxide by TMP. (A) The oxidation–reduction reaction catalyzed by HRP. The reaction was carried out at 37°C for 60 min, and hydrogen
peroxide was reduced at the expense of TMP. (B) The spectra of the reactions in the presence of hydrogen peroxide at various concentrations (3, 6, 12 mM) were recorded at the
wavelengths 200–700 nm. (C) Oxidation of TMP by hydrogen peroxide. After the reaction catalyzed by HRP (50 unit), the UV absorption maximum at 294 nm (A294) was
determined, and the oxidative rate (%) of TMP was calculated by dividing the absorbance difference between the control (AC) and the reaction (AR) by the absorbance of the
control. The reaction containing hydrogen peroxide and HRP was used as a blank (0). Data represent means SD (n=3).
peroxide in the presence of HRP. This enabled us to use the decrease in
TMP in the reactions to detect the capability of TMP to destroy
hydrogen peroxide by determining the UV absorption at 294 nm
(A294). As shown in Fig. 5C, a hydrogen peroxide concentration-
dependent TMP oxidation was observed. These results imply that TMP
can function as a reductant to destroy hydrogen peroxide directly.
protection against kainate-induced neuronal loss was correlated
with the role of TMP antioxidation, we examined the effect of TMP
on primary cultures of hippocampal neurons. Exposure of hippocam-
pal neurons to 100 μM kainate for 3–24 h elicited a significant
decrease in cell survival (Fig. 6B), whereas kainate-induced neuronal
loss was prevented by adding 50 and 100 μM TMP, as well as the ROS-
quenching agent α-tocopherol (50 μM; Fig. 6C). These results indicate
that TMP can protect neurons against kainate-induced cytotoxicity
through its antioxidant effect.
TMP prevents kainate-induced neuronal death in primary hippocampal
neurons
Initially, we examined the effects of TMP on cell viability using
primary hippocampal neurons in culture. The MTT method revealed
that exposure of hippocampal neurons to 25–100 μM TMP for 24 h had
no effect on cell viability (Fig. 6A). Thus, a dose of 50 μM TMP was
employed in this study. To demonstrate whether the in vivo
TMP reduces ROS accumulation and protects antioxidant enzymes
in vitro and in vivo
Excessive activation of glutamate receptors induces oxidative
stress in which Ca2+-mediated ROS accumulation plays a critical
Fig. 6. TMP prevention of kainate-induced neuronal loss in culture. Primary cultured hippocampal neurons were plated at a density of 5×105 cells/cm2 in DMEM with 20% fetal bovine
serum onto 96-well dishes and grown in a 37°C incubator with 5% CO2. Hippocampal neurons were cultured for 12–14 days before experiments (see Materials and methods). (A)
Examination of the dose effect of TMP on neuronal viability by MTT assay. Neurons were exposed to TMP at concentrations of 25, 50, 100, or 200 μM for 24 h. (B) Analysis of kainate-
induced neuronal loss by MTT assay. Neurons were exposed to 100 μM kainate for 0, 3, 6, 12, or 24 h. (C) Protection of hippocampal neurons against kainate-induced cell loss by TMP.
Neurons were exposed to 100 μM kainate (lanes 2–5) with 50 μM vitamin E (lane 3) or 50 (lane 4) or 100 μM (lane 5) TMP. Neurons were exposed to 0.9% NaCl as negative control (lane
1) and to 250 μM H2O2 for 10 min as positive control (lane 6). Cell viability at 24 h after kainate exposure was measured by MTT assay. All data represent means SD (n=5). ⁎Pb0.05.

604
S.-Y. Li et al. / Free Radical Biology & Medicine 48 (2010) 597–608
role in neurotoxicity [1,2]. To explore the mechanisms of TMP
neuroprotection, ROS production in kainate-exposed hippocampal
neurons was evaluated by DHE staining. DHE is oxidized by
intracellular ROS to produce fluorescent ethidium that is subse-
quently intercalated in DNA, further amplifying its fluorescence
[28]. Thus, an increase in DHE oxidation and the subsequent
increase in fluorescence are highly suggestive of ROS generation.
Flow cytometry analysis of ROS-positive cells (Fig. 7A) showed that,
compared with kainate-unexposed neurons (control), the levels of
ROS in kainate-exposed neurons were significantly increased
(Pb0.05), as in H2O2-exposed neurons (positive control; Fig. 7B).
Notably, TMP (50 μM) as well as vitamin E (50 μM) inhibited
kainate-induced ROS accumulation markedly. These results indicate
that TMP can reduce kainate-induced ROS accumulation in primary
hippocampal neurons.
GR in cultured hippocampal neurons. Kainate, like hydrogen
peroxide, significantly decreased the activities of GPX (10.55 mU/
mg; Fig. 7C) and GR (14.22 mU/mg; Fig. 7D) in cultured
hippocampal neurons, compared with the normal (basal) levels
(21.8 and 23.79 mU/mg in kainate-unexposed neurons). The
decrease in both enzyme activities could be reversed by TMP as
well as vitamin E. To evaluate the TMP effects in vivo, hippocampi
from the four groups of rats (treated as in Fig. 3) were removed 2
days after kainate treatment, and the activities of GPX and GR were
measured. Similarly, kainate-suppressed enzyme activities and TMP-
preserved enzymes were observed in KA- and KA + TMP-treated
rat hippocampus (Figs. 7E and 7F). These results suggest that TMP
can protect antioxidant enzyme activities.
TMP alleviates mitochondrial impairment in vitro and in vivo by
suppressing lipid peroxidation
Enzymatic and nonenzymatic antioxidants such as GSH, super-
oxide dismutase (SOD), GPX, and GR scavenge excessive ROS in the
cells. In turn, ROS are able to damage cellular antioxidants. To
determine whether ROS accumulation was attributed to suppression
of the antioxidant defense system during kainate exposure and
whether TMP prevented it, we examined the activities of GPX and
ROS induce lipid peroxidation, destroying cellular and subcellular
membranes. The level of MDA, a lipid peroxidation end-product, is
commonly known as a marker of oxidative stress and the antioxidant
status. We thus determined the effects of kainate and TMP on MDA
Fig. 7. Prevention of ROS generation and preservation of GPX and GR by TMP. (A) Flow cytometry analysis of ROS generation. Primary hippocampal neurons were cultured as for
Fig. 6. Neurons were exposed to 100 μM kainate for 24 h with or without TMP (50 μM) or vitamin E (50 μM) as indicated. Neurons were stained with DHE and analyzed by flow
cytometry. Neurons were exposed to 0.9% NaCl as negative control and to 250 μM H2O2 for 10 min as positive control. (B) Histogram showing the relative levels of ROS in the
experiments show in (A). The ROS levels in control cells were normalized to 1. Data represent means SD (n=3). ⁎Pb0.05. (C) Preservation of GPX and (D) GR in hippocampal
neurons in culture. Primary hippocampal neurons were treated as in (A) and lysed. After protein determination, enzyme activities were measured using the respective assay kits.
Specific activity of GPX is shown as μmol NADPH min−1 mg−1 protein. Specific activity of GR is shown as μmol GSSG min−1 mg−1 protein. Data represent means SD (n=3).
⁎Pb0.05. (E) Prevention of GPX and (F) GR in the rat hippocampus. Homogenates of rat hippocampi from the animals in the Fig. 3 experiments were prepared and lysed. After
protein determination, enzyme activities were measured as in (C) and (D). Data represent means SD (n=3). ⁎Pb0.05.

S.-Y. Li et al. / Free Radical Biology & Medicine 48 (2010) 597–608
605
levels in primary hippocampal neurons and the rat hippocampus. Fig.
8A shows that kainate, like hydrogen peroxide, significantly increased
the MDA level, to 4.96 nmol/mg on the basal level (without kainate
exposure) of 3.51 nmol/mg; Fig. 8B shows that the MDA level in
hippocampus from kainate-injected rats was 3.05 nmol/mg, whereas
the MDA level in hippocampus from saline (control) group rats was
only 2.09 nmol/mg. These results suggest that kainate treatment can
induce lipid peroxidation, because of suppressed antioxidant defense
and ROS accumulation. Notably, in the same experiments, addition of
TMP (50 μM) as well as vitamin E (50 μM) substantially decreased
MDA levels to normal levels in primary cultures of hippocampal
neurons (Fig. 8A) and rat hippocampus (Fig. 8B), indicating that TMP
can inhibit kainate-induced lipid peroxidation in vitro and vivo.
ROS-induced lipid peroxidation may disrupt the structural and
functional integrity of neurons. Mitochondria, containing multiple
electron carriers capable of producing ROS as well as an extensive
network of antioxidant defenses, are both regulators and targets of
ROS. Therefore, mitochondrial insults, including oxidative damage
itself, can cause an imbalance between ROS production and removal,
resulting in net ROS production [32]. We thus quantitatively
determined mitochondrial damage in primary hippocampal neurons
by flow cytometry. As shown in Figs. 8C and 8D, kainate caused a
significant decrease in mitochondrial membrane potential in exposed
neurons, which was promoted by CTZ, a desensitization blocker of
AMPA receptor, whereas D-AP5, a selective N-methyl-D-aspartate
(NMDA) receptor antagonist, did not affect it. By contrast, NBQX, a
selective and competitive AMPA/KA receptor antagonist, completely
blocked the decrease in the membrane potential in exposed neurons.
These results indicate that the decrease in mitochondrial membrane
potential in exposed neurons is attributable to overactivation of
AMPA/KA receptors by kainate. As expected, the antioxidant vitamin
E blocked the decrease in the membrane potential. Notably, TMP
Fig. 8. Prevention of lipid peroxidation and stabilization of mitochondrial functions by TMP. (A) Prevention of lipid peroxidation by TMP in kainate-exposed hippocampal neurons.
Primary hippocampal neurons were treated as in Fig. 7A and lysed. MDA levels in cell extracts were determined using 1,1,3,3-tetraethoxypropane as a standard. Lipid peroxidation
was calculated as nanomoles of MDA per milligram of protein. Data represent means SD (n=3). ⁎Pb0.05. (B) Prevention of lipid peroxidation by TMP in the rat hippocampus.
Homogenates of rat hippocampi from the animals in the Fig. 3 experiments were prepared and lysed 2 days after kainate induction. MDA levels in hippocampal extracts were
determined. Data represent means SD (n=3). ⁎Pb0.05. (C) Preservation of mitochondrial membrane potential by TMP. Primary hippocampal neurons were exposed to 100 μM
kainate for 24 h with or without TMP (50 μM) or vitamin E (50 μM). To identify AMPA/kainate receptor activation, CTZ (30 μM), D-AP5 (10 μM), and NBQX (5 μM) were used. After
R123 staining, mitochondrial membrane potential was analyzed by flow cytometry. (D) Histogram showing the quantitative changes in mitochondrial membrane potential in the
experiments shown in (C). Data represent means SD (n=3). ⁎Pb0.05; ⁎⁎Pb0.001. (E) Preservation of complexes I and III in the rat hippocampus by TMP. Homogenates of rat
hippocampi were prepared and lysed as in (B), followed by isolation of mitochondria. After protein determination, the activities of NADH:ubiquinol reductase (complex I) and
ubiquinol:cytochrome c reductase (complex III) were measured by the respective kits. Specific activities of complexes I and III were adjusted to give data as mU/mg protein. Data
represent means SD (n=5). ⁎Pb0.05. (F) Protection of ATP production in the rat hippocampus by TMP. Isolation of mitochondria of rat hippocampi was the same as for (E). After
protein determination, the levels of ATP were quantitatively determined by a bioluminescence assay kit and calculated from a concentration standard curve, normalized as mol ATP/
μg protein and converted to percentage of untreated controls. Data represent means SD (n=4). ⁎⁎Pb0.001.

606
S.-Y. Li et al. / Free Radical Biology & Medicine 48 (2010) 597–608
(50 μM) also abolished the kainate-reduced mitochondrial membrane
potential, indicating that TMP can protect neurons against kainate-
induced mitochondrial impairment.
stabilization of mitochondrial function by TMP is linked to the
function of TMP as a reductant/antioxidant to quench ROS. These
results indicate that stabilization of mitochondrial function by TMP
can protect against kainate-induced oxidative lesions in the rat
hippocampus. Our data support the notion that mitochondrial
dysfunction and oxidative stress occur early in neurodegenerative
diseases, and this dysfunction has a causal role in disease pathogen-
esis [3]. In addition, our data suggest that TMP and its derivatives may
have the potential to be developed as a therapeutic drugs for
neurodegenerative diseases.
Multiple roles for TMP have been reported. In a model of ischemia
brain injury in animals, TMP neuroprotection is correlated with
increased transcription of thioredoxin [17] and anti-inflammatory
potential [16]. In spinal cord ischemia/reperfusion injury, TMP
reduces apoptosis by regulating Bcl-2 and Bax expression [14]. During
rat subarachnoid hemorrhage, TMP inhibits the activation of the
caspase-dependent apoptosis pathway [33]. TMP blocks H2O2-
induced apoptosis of PC12 cells by regulating Bcl-2 family members,
suppressing cytochrome c release and caspase activation [34]. Anti-
apoptosis by Ligusticum chuanxiong (the origin of TMP) is mediated by
activating the PKA/CREB-dependent pathway in serum deprivation-
induced PC12 death [35]. Early reports suggest that TMP can increase
cerebral blood flow [36] and improve the blood viscosity [15] and
microcirculation [37] of experimental animals. Although a variety of
mechanisms in various models have been suggested, we believe that
the core of TMP neuroprotection against kainate-induced oxidative
lesions is linked to the preservation of mitochondrial function.
Mitochondria are critical regulators of cell death, a key feature of
neurodegenerative diseases [3]. The inner mitochondrial membrane
harbors the four major respiratory complexes (I–IV) and the F0F1-ATP
synthase (complex V). Complexes I (NADH:ubiquinone oxidoreduc-
tase), III (ubiquinol:cytochrome c oxidoreductase), and IV (cyto-
chrome c oxidase) transduce the energy of nutritional compounds,
To further demonstrate that ROS-induced lipid peroxidation
disrupted the metabolic integrity of neurons and TMP protects
hippocampal neurons against kainate-induced mitochondrial dys-
function, mitochondria were isolated from the hippocampus of the
four groups of rats (Fig. 3), followed by analyzing the activities of
respiratory chain complexes I and III from mitochondrial membranes.
The activity of NADH:ubiquinol reductase (complex I) was measured
by the decylquinazolinamine-sensitive oxidation of NADH, and
ubiquinol:cytochrome c reductase (complex III) was measured by
the antimycin-sensitive reduction of cytochrome c. As shown in Fig.
8E, compared with the complexes I (10.61 mU/mg) and III (5.91 mU/
mg) activities in the saline group (control) rats, the enzymatic
activities of both complexes in kainate-treated rat hippocampus were
reduced to 3.79 and 3.19 mU/mg, respectively, whereas no decreases
in the complex activities could be observed in TMP- and kainate +
TMP-treated rat hippocampus. Supporting these results, kainate-
treated rats showed a decrease in the generation of ATP by 48%,
whereas TMP and kainate + TMP group rats showed a reduced
generation of ATP that was only about 4.4 and 15.87% that of saline
group rats, respectively (Fig. 8F). The results from Fig. 8 suggest that
TMP can protect neurons against kainate-induced mitochondrial
dysfunction.
Discussion
In this study we demonstrate that TMP can partially alleviate
kainate-induced status epilepticus in rats and prevent and rescue
pyramidal neuron loss in the hippocampus. The prevention and
rescue of partial neuronal loss in the hippocampus by TMP are
attributed to the preservation of mitochondrial function. The
Fig. 9. Schematic presentation of proposed signaling pathways in TMP protection of hippocampal neurons against kainate-induced oxidative stress. Overactivation of AMPA/KA
receptors induces mitochondrial and cellular Ca2+ overload, leading to the production of ROS, which ultimately causes oxidative stress [1,2]. Increased ROS cause lipid peroxidation
and protein (and DNA) damage and disrupt the structural and functional integrity of mitochondria (mitochondrial dysfunction), which is associated with cell death. Production of
free radicals by lipid peroxidation and damage of enzymatic antioxidants such as GPX and GR by protein oxidation further aggravate ROS accumulation. An imbalance between ROS
formation and ATP production due to the suppression of complexes I, III, and V can also increase ROS accumulation. Because mitochondria are critical regulators [3], mitochondrial
dysfunction fails to regulate Ca2+ homeostasis and ROS metabolism, further aggravating oxidative stress. TMP prevents ROS production by scavenging O2⋅−, H2O2, and hydroxyl
radicals (OH⋅), thereby elevating internal antioxidant GSH defenses, maintaining ATP production, and enhancing cell survival. ↑, promotion; ⊺, suppression.

S.-Y. Li et al. / Free Radical Biology & Medicine 48 (2010) 597–608
607
leading to ATP generation. Normally, the electron flow in mitochon-
dria may produce ROS. On the other hand, mitochondria contain an
extensive antioxidant defense system to detoxify the ROS, which
depends mainly on GSH, SOD, and GPX [32]. GPX can convert
hydrogen peroxide into water and oxygen in the presence of GSH, and
GR catalyzes the reduction of GSSG to GSH (Fig. 9). Oxidative stress is
an imbalance between ROS production and the antioxidant defense
system. The pro-oxidant–antioxidant imbalance in kainate-induced
oxidative stress may be due to Ca2+ increases in the cytosol and the
mitochondria. Elevated intraneuronal Ca2+ activates peptidases such
as calpain I, which can catalyze the enzymatic conversion of xanthine
dehydrogenase to xanthine oxidase; the metabolism of purine bases
by xanthine oxidase yields O⋅2− and H2O2. This reaction may become
quite prominent, because kainate receptor agonists cause a depletion
of ATP [2]. Like ROS generation, antioxidant defenses are tied to the
redox and energetic state of the mitochondrion. We did find the
decrease in ATP generation to be coupled with the decrease in
complex I and III activities (Figs. 7E and 7F) and GPX and GR activities
(Figs. 6C–6F) in kainate-treated hippocampal neurons and the rat
hippocampus, suggesting mitochondrial dysfunction during kainate-
induced neurotoxicity. Decreased activities of GPX and GR result in
ROS accumulation during AMPA/kainate receptor activation. In-
creased ROS can damage lipids, proteins, and DNA and disrupt
membrane integrity [6,7], leading to cellular and mitochondrial
dysfunctions that are associated with cell death [3]. Lipid peroxidation
produces toxic aldehyde products and free radicals, which further
aggravate oxidative stress and degenerative processes. Antioxidants
can protect cells from excitotoxicity by scavenging ROS and increasing
intracellular cysteine levels [38]. We found that TMP could directly
function as a reductant (Fig. 8) to quench ROS and reduce ROS
accumulation, thereby protecting antioxidant enzymes (Figs. 6C–6F),
preventing lipid peroxidation (Figs. 7A and 7B), and preserving the
structural and functional integrity of mitochondria (Figs. 7C–7F).
Subsequently, maintaining mitochondrial function by TMP may
regulate cellular Ca2+ homeostasis and attenuate Ca2+-mediated
ROS generation, thereby protecting biomolecules, including mem-
brane lipids, essential cellular proteins, and DNA. In these ways, TMP
improves cellular redox status, protecting from neuronal death in
kainate-induced oxidative stress (Fig. 9). Non-NMDA-receptor-medi-
ated neurotoxicity is more complex [2]. In addition, superoxide
activates inducible nitric oxide synthase, leading to the production of
nitric oxide and the formation of peroxynitrites, which ultimately
causes oxidative/nitrosative stress [2,39]. Furthermore, ROS originat-
ing in mitochondria are thought to be a major source of endogenous
nuclear DNA damage [40]. Therefore, further study of the detailed
signaling pathways may provide a better understanding of TMP
protection against kainate-induced oxidative stress.
pathways. The CA1 and CA3 regions and the hilus of the dentate gyrus
are particularly sensitive to the excitotoxicity of kainate [41–43]. We
found that the loss of pyramidal neurons in the CA1 region was
greater than that in the CA3 (Figs. 2–4). Consistent with this,
protection of CA3 neurons by TMP was more effective than that of
CA1 (Figs. 3 and 4), indicating that CA1 neurons are more sensitive to
kainate-induced excitotoxicity. Probably, there are several reasons for
the different sensitivities of CA1 and CA3 neurons. First, the different
sensitivities of the two regions might be associated with the cell death
pathways. For instance, p53 promotes cell death via multiple
pathway-apoptotic cell death in CA1 neurons and necrotic cell death
in CA3 in kainate-treated mice [44]. Second, our observation that TMP
rescued kainate-induced neuronal loss in the CA3 but not CA1 region
suggests earlier occurrence of irreversible injury in CA1 neurons. The
irreversible injury to CA1 might involve specific changes in gene
expression. This inference is supported by a previous finding that
GluR2 antisense combined sublethal ischemia produces virtually
complete loss of CA1 pyramidal cells and partial loss of CA3 pyramidal
cells [45]. Third, CA3 and CA1 regions have distinct anatomical and
physiological features. The CA3 region has dense excitatory recurrent
synaptic connections between pyramidal cells; CA1 pyramidal cells
receive a large number of excitatory synaptic inputs from CA3
pyramidal cells through Schaffer collaterals. This may allow CA1
cells to be sequence-sensitive [46]. The precise explanation for the
different injuries to CA1 and CA3 during kainate-induced excitotoxi-
city remains to be found.
Acknowledgments
We thank Miss Yang Lei (Department of Anatomy, Peking
University Health Science Center, China) for her assistance in brain
section management. This work was supported in part by National
Natural Science Foundation of PR China grants (30771225, 30671062,
30971449) and the Education Committee of Beijing Teaching
Reinforcing Plan.
References
[1] Choi, D. W. Glutamate neurotoxicity and diseases of the nervous system. Neuron
1:623–634; 1988.
[2] Coyle, J. T.; Puttfarcken, P. Oxidative stress, glutamate, and neurodegenerative
disorders. Science 262:689–695; 1993.
[3] Lin, M. T.; Beal, M. F. Mitochondrial dysfunction and oxidative stress in
neurodegenerative diseases. Nature 443:787–795; 2006.
[4] Vila, M.; Przedborski, S. Targeting programmed cell death in neurodegenerative
diseases. Nat. Rev. Neurosci. 4:365–375; 2003.
[5] Tanaka, H.; Grooms, S. Y.; Bennett, M. V.; Zukin, R. S. The AMPAR subunit GluR2:
still front and center-stage. Brain Res. 886:190–207; 2000.
[6] Choi, D. W. Calcium and excitotoxic neuronal injury. Ann. N. Y. Acad. Sci. 747:
162–171; 1994.
Although the basicity of pyrazine is weak, TMP (2,3,5,6-tetra-
methylpyrazine) has an increased basicity, because of the repelling of
electrons by the four methyl groups (Fig. 5A). Therefore, the two
nitrogen atoms with an unshared electron pair can function as an
electron donor (Lewis base). TMP can be oxidized in vitro to TMP-1,4-
(N,N′)-dioxide in the presence of hydrogen peroxide and at 70°C [31].
This, however, is not the case in living cells. To mimic the intracellular
environment, we developed a reduction–oxidation reaction catalyzed
by HRP (Fig. 5A). The reduction state of TMP exhibited a UV-
absorption maximum at 294 nm (A294), while the A294 of the
oxidation state was decreased (Fig. 5B), which can be used to
determine the reducing capability of TMP. In the presence of HRP,
TMP could directly be oxidized at 37°C in a hydrogen peroxide dose-
dependent manner (Fig. 5C), indicating that TMP is a hydrogen
peroxide scavenger. This can explain how TMP decreases ROS
accumulation, protecting against oxidative damage in the rat
hippocampus.
[7] Siesjö, B. K.; Bengtsson, F.; Grampp, W.; Theander, S. Calcium, excitotoxins, and
neuronal death in the brain. Ann. N. Y. Acad. Sci. 568:234–251; 1989.
[8] Herrlich, P.; Bohmer, F. D. Redox regulation of signal transduction in mammalian
cells. Biochem. Pharmacol. 59:35–41; 2000.
[9] Hall, A. G. The role of glutathione in the regulation of apoptosis. Eur. J. Clin. Invest.
29:238–245; 1999.
[10] Lu, C. S.; Yin, M. H.; Huang, Y. H.; Pao, C.; Sung, C. F. Radix salviae miltiorrhizae and
Rhizoma ligustici wallichii in coronary heart disease. Chin. Med. J. (English) 4:
43–46; 1978.
[11] Guo, S. K.; Chen, K. J.; Qian, Z. H.; Weng, W. L.; Qian, M. Y. Tetramethylpyrazine in
the treatment of cardiovascular and cerebrovascular diseases. Planta Med. 47:89;
1983.
[12] Jiao, F.; Cong, K.; Li, Y. Observation on ischemic cerebrovascular disease (113
cases) treated by injection of ligustrazine. J. Pract. Tradit. Chin. Med. 20:174–175;
2004.
[13] Zhu, X. Q.; Liu, G. H.; Hu, J. X. The anti-epileptic effect of tetramethylpyrazine and
its relationship to interleukin-1 receptor. Lishizhen Med. Materia Med. Res. 18:
2927–2928; 2007.
[14] Fan, L. H.; Wang, K. Z.; Cheng, B.; Wang, C. S.; Dang, X. Q. Anti-apoptotic and
neuroprotective effects of tetramethylpyrazine following spinal cord ischemia in
rabbits. BMC Neurosci. 7:48; 2006.
[15] Chang, F. C.; Huang, Y. T.; Lin, H. C.; Hong, C. Y.; Lin, J. G.; Chen, K. J. Beneficial
effects of combined terlipressin and tetramethylpyrazine administration on portal
hypertensive rats. Can. J. Physiol. Pharmacol. 77:618–624; 1999.
Systemic administration of kainate in rodents increases ROS
production and induces cell death by both necrotic and apoptotic

608
S.-Y. Li et al. / Free Radical Biology & Medicine 48 (2010) 597–608
[16] Liao, S. L.; Kao, T. K.; Chen, W. Y.; Lin, Y. S.; Chen, S. Y.; Raung, S. L.; Wu, C. W.; Lu,
H. C.; Chen, C. J. Tetramethylpyrazine reduces ischemic brain injury in rats.
Neurosci. Lett. 372:40–45; 2004.
[17] Zhu, X. L.; Xiong, L. Z.; Wang, Q.; Liu, Z. G.; Ma, X.; Zhu, Z. H.; Hu, S.; Gong, G.; Chen,
S. Y. Therapeutic time window and mechanism of tetramethylpyrazine on
transient focal cerebral ischemia/reperfusion injury in rats. Neurosci. Lett. 449:
24–27; 2009.
[18] Savaskan, N. E.; Bräuer, A. U.; Kühbacher, M.; Eyüpoglu, I. Y.; Kyriakopoulos, A.;
Ninnemann, O.; Behne, D.; Nitsch, R. Selenium deficiency increases susceptibility
to glutamate-induced excitotoxicity. FASEB J. 17:112–124; 2003.
[19] Zhang, X.; Gelowitz, D. L.; Lai, C. T.; Boulton, A. A.; Yu, P. H. Gradation of kainic
acid-induced rat limbic seizures and expression of hippocampal heat shock
protein-70. Eur. J. Neurosci. 9:760–769; 1997.
[20] Schmued, L. C.; Albertson, C.; Slikker, W. J. Fluoro-Jade: a novel fluorochrome for
the sensitive and reliable histochemical localization of neuronal degeneration.
Brain Res. 751:37–46; 1997.
[21] Lissin, D. V.; Carroll, R. C.; Nicoll, R. A.; Malenka, R. C.; von Zastrow, M. Rapid,
activation-induced redistribution of ionotropic glutamate receptors in cultured
hippocampal neurons. J. Neurosci. 19:1263–1272; 1999.
gradients inhibits mitochondrial hyperpolarization and cytochrome c release
during neural apoptosis. J. Neurosci. 21:4551–4563; 2001.
[30] Grooms, S. Y.; Opitz, T.; Bennett, M. V.; Zukin, R. S. Status epilepticus decreases
glutamate receptor 2 mRNA and protein expression in hippocampal pyramidal
cells before neuronal death. Proc. Natl. Acad. Sci. USA 97:3631–3636; 2000.
[31] Cheng, X. C.; Liu, X. Y.; Xu, W. F.; Guo, X. L.; Ou, Y. Design, synthesis, and biological
activities of novel Ligustrazine derivatives. Bioorg. Med. Chem. 15:3315–3320; 2007.
[32] Andreyev, A. Y.; Kushnareva, Y. E.; Starkov, A. A. Mitochondrial metabolism of
reactive oxygen species. Biochemistry (Moscow) 70:200–214; 2005.
[33] Gao, C.; Liu, X.; Liu, W.; Shi, H.; Zhao, Z.; Chen, H.; Zhao, S. Anti-apoptotic and
neuroprotective effects of tetramethylpyrazine following subarachnoid hemor-
rhage in rats. Auton. Neurosci. 141:22–30; 2008.
[34] Cheng, X. R.; Zhang, L.; Hu, J. J.; Sun, L.; Du, G. H. Neuroprotective effects of
tetramethylpyrazine on hydrogen peroxide-induced apoptosis in PC12 cells. Cell
Biol. Int. 31:438–443; 2007.
[35] Lin, Y. L.; Lee, Y. C.; Huang, C. L.; Lai, W. L.; Lin, Y. R.; Huang, N. K. Ligusticum
chuanxiong prevents rat pheochromocytoma cells from serum deprivation-
induced apoptosis through
a protein kinase A-dependent pathway. J.
Ethnopharmacol. 109:428–434; 2007.
[22] Li, S. Y.; Ni, J. H.; Xu, D. S.; Jia, H. T. Down-regulation of GluR2 is associated with
Ca2+-dependent protease activities in kainate-induced apoptotic cell death in
cultured rat hippocampal neurons. Neurosci. Lett. 352:105–108; 2003.
[23] Kushnareva, Y. E.; Wiley, S. E.; Ward, M. W.; Andreyev, A. Y.; Murphy, A. N.
Excitotoxic injury to mitochondria isolated from cultured neurons. J. Biol. Chem.
280:28894–28902; 2005.
[24] Tzagoloff, A.; Akai, A.; Needleman, R. B.; Zulch, G. Assembly of the mitochondrial
membrane system: cytoplasmic mutants of Saccharomyces cerevisiae with lesions
in enzymes of the respiratory chain and in the mitochondrial ATPase. J. Biol. Chem.
250:8236–8242; 1975.
[36] Feng, M. G.; Feng, G. H.; Zhou, Q. G. Effects of methylhesperidin on coronary, renal
and cerebral circulation in dogs [in Chinese]. Acta Pharmacol. 9:548–550; 1988.
[37] Dai, X. Z.; Bache, R. J. Coronary and systemic haemodynamic effect of
tetramethylpyrazine in the dog. J. Cardiovasc. Pharmacol. 7:841–849; 1985.
[38] Carriedo, S. G.; Sensi, S. L.; Yin, H. Z.; Weiss, J. H. AMPA exposures induce
mitochondrial Ca2+ overload and ROS generation in spinal motor neurons in vitro.
J. Neurosci. 20:240–250; 2000.
[39] Ansari, M. A.; Keller, J. N.; Scheff, S. W. Protective effect of Pycnogenol in human
neuroblastoma SH-SY5Y cells following acrolein-induced cytotoxicity. Free Radic.
Biol. Med. 45:1510–1519; 2008.
[25] White, A. R.; Collins, S. J.; Maher, F.; Jobling, M. F.; Stewart, L. R.; Thyer, J. M.;
Beyreuther, K.; Masters, C. L.; Cappai, R. Prion protein-deficient neurons reveal
lower glutathione reductase activity and increased susceptibility to hydrogen
peroxide toxicity. Am. J. Pathol. 155:1723–1730; 1999.
[26] Kumar, V.; Bal, A.; Gill, K. D. Impairment of mitochondrial energy metabolism in
different regions of rat brain following chronic exposure to aluminium. Brain Res.
1232:94–103; 2008.
[27] White, A. R.; Multhaup, G.; Maher, F.; Bellingham, S.; Camakaris, J.; Zheng, H.;
Bush, A. I.; Beyreuther, K.; Masters, C. L.; Cappai, R. The Alzheimer's disease
amyloid precursor protein modulates copper-induced toxicity and oxidative
stress in primary neuronal cultures. J. Neurosci. 19:9170–9179; 1999.
[28] Gross, E. R.; LaDisa Jr., J. F.; Weihrauch, D.; Olson, L. E.; Kress, T. T.; Hettrick, D. A.;
Pagel, P. S.; Warltier, D. C.; Kersten, J. R. Reactive oxygen species modulate
coronary wall shear stress and endothelial function during hyperglycemia. Am. J.
Physiol. Heart Circ. Physiol. 284:H1552–1559; 2003.
[40] Huang, M. E.; Kolodner, R. D. A biological network in Saccharomyces cerevisiae
prevents the deleterious effects of endogenous oxidative DNA damage. Mol. Cell
17:709–720; 2005.
[41] Ben-Ari, Y.; Cossart, R. Kainate, a double agent that generates seizures: two
decades of progress. Trends Neurosci. 23:580–587; 2000.
[42] Choi, D. W. Excitotoxic cell death. J. Neurobiol. 23:1261–1276; 1992.
[43] Frerking, M.; Nicoll, R. A. Synaptic kainate receptors. Curr. Opin. Neurobiol. 10:
342–351; 2000.
[44] Morrison, R. S.; Wenzel, H. J.; Kinoshita, Y.; Robbins, C. A.; Donehower, L. A.;
Schwartzkroin, P. A. Loss of the p53 tumor suppressor gene protects neurons from
kainate-induced cell death. J. Neurosci. 16:1337–1345; 1996.
[45] Oguro, K.; Oguro, N.; Kojima, T.; Grooms, S. Y.; Calderone, A.; Zheng, X.; Bennett,
M. V.; Zukin, R. S. Knockdown of AMPA receptor GluR2 expression causes delayed
neurodegeneration and increases damage by sublethal ischemia in hippocampal
CA1 and CA3 neurons. J. Neurosci. 19:9218–9227; 1999.
[29] Poppe, M.; Reimertz, C.; Düssmann, H.; Krohn, A. J.; Luetjens, C. M.; Böckelmann,
D.; Nieminen, A. L.; Kögel, D.; Prehn, J. H. Dissipation of potassium and proton
[46] Yoshida, M.; Hayashi, H. Emergence of sequence sensitivity in a hippocampal
CA3–CA1 model. Neural Netw 20:653–667; 2007.