Magnolol Reduces Glutamate-Induced Neuronal Excitotoxicity and Protects against Permanent Focal Cerebral Ischemia Up to 4 Hours

Neuroprotective efficacy of magnolol, 5,5′-dially-2,2′-dihydroxydiphenyl, was investigated in a model of stroke and cultured neurons exposed to glutamate-induced excitotoxicity. Rats were subjected to permanent middle cerebral artery occlusion (pMCAO). Magnolol or vehicle was administered intraperitoneally, at 1 hr pre-insult or 1–6 hrs post-insult. Brain infarction was measured upon sacrifice. Relative to controls, animals pre-treated with magnolol (50–200 mg/kg) had significant infarct volume reductions by 30.9–37.8% and improved neurobehavioral outcomes (P<0.05, respectively). Delayed treatment with magnolol (100 mg/kg) also protected against ischemic brain damage and improved neurobehavioral scores, even when administered up to 4 hrs post-insult (P<0.05, respectively). Additionally, magnolol (0.1 µM) effectively attenuated the rises of intracellular Ca2+ levels, [Ca2+](i), in cultured neurons exposed to glutamate. Consequently, magnolol (0.1–1 µM) significantly attenuated glutamate-induced cytotoxicity and cell swelling (P<0.05). Thus, magnolol offers neuroprotection against permanent focal cerebral ischemia with a therapeutic window of 4 hrs. This neuroprotection may be, partly, mediated by its ability to limit the glutamate-induced excitotoxicity.


Introduction
Ischemic stroke is characterized by overstimulation of glutamate receptors of the N-methyl-D-aspartate type (NMDARs) and increased inflow of intracellular Ca 2+ , [Ca 2+ ](i) [1][2][3][4][5]. NMDAR overactivation disrupts antioxidant defenses and critical survival pathways, which not only increase the susceptibility of neurons and glia to ischemic damage but also trigger numerous ischemic cascades, leading to further neuronal degeneration, swelling, or even deaths [1,2]. Nonetheless, efforts to inhibit NMDARs have generally failed, mainly due to critical roles of these receptors in neuronal survival and synaptic plasticity [2,3]. Other strategies to improve neuronal defense against the glutamate-induced excitotoxicity and/or to decrease the [Ca 2+ ](i) inflow into the ischemic neurons have, therefore, been suggested [4][5][6][7]. Several different classes of antioxidants and/or neuroprotectants such as calpain inhibitors have been shown to protect against ischemia-induced excitotoxicity and, thus, decrease brain damage caused by experimental stroke [6][7][8][9][10].
Magnolol, a blood-brain barrier permeable phenolic constituent (5, 59-dially-2, 29-dihydroxydiphenyl) of magnolia bark, is known to be a central nervous system depressant agent and potent antioxidant [11][12][13]. Magnolol has been shown to protect against brain damage in an experimental heatstroke model [14]. We have previously shown that magnolol protects against hind limb ischemic-reperfusion injury in rats by reducing post-ischemic rises in the levels of nitrite/nitrate (NOX), malondialdehyde (MDA) and myeloperoxidase (MPO) [13]. Alternatively, magnolol is an inhibitor of voltage-dependent Ca 2+ channel and can reduce necrotic cell deaths in mixed neuron-astrocyte cultures exposed to chemical hypoxia [15][16][17]. Accordingly, we suspected that magnolol might protect the brain against ischemic stroke.
In the study, we evaluated the protective effects of magnolol against cell damage and swelling as well as increased inflow of [Ca 2+ ](i) in cultured neurons exposed to glutamate. Additionally, we investigated neuroprotective efficacy of magnolol in rats subjected to permanent focal cerebral ischemia.

Results
Neurotoxicity of magnolol was observed with a concentration beyond 100 mM. Nine animals (7.8%) died prior to completing the protocol following pMCAO and were excluded: 4 (8.2%) were in the vehicle-treated groups and 5 (5.4%) were in the magnolol-treated groups. Following the ischemic onset, the ipsilateral LCBP declined to 14-22% and 32-38% of the baseline data in the ischemic core and penumbral areas, respectively. The LCBP was not significantly different among experiment groups, and was independent of magnolol treatments (P.0.05; data not shown). The other physiological parameters were kept within normal limits and did not differ significantly among experiment groups, except that high-dose (200 mg/kg) magnolol-treated animals had arterial pCO 2 retention along with reduced heart rate and arterial pH (Table 1). Ischemic animals invariably experienced spontaneous hyperthermia throughout the recovery period. Animals treated with magnolol at 200 mg/kg, but not at lower doses, had modest temperature reductions by <3uC, and this temperature-lowering effect remained effective up to 2 hrs post-insult (Table 2).
In the delayed treatment paradigm, our results indicated that magnolol (100 mg/kg) resulted in significant infarct volume reductions when administrated within 4 hrs after the ischemic onset (P,0.05; Figs. 3B, C). Relative to controls (n = 30), infarction were reduced by 42.5, 28.5 and 20.6%, respectively, when magnolol was given at 1 (n = 7), 2 (n = 7), and 4 hrs (n = 9) post-insult (P,0.05; Figures 3C). Magnolol treated at 6 hrs postinsult (n = 9) did not significantly reduced brain infarction. However, delayed treatment with magnolol significantly improved sensory neurologic scores, even when administered up to 6 hrs post-insult (P,0.05), and effectively reduced post-ischemic body weight loss, when administered up to 2 hrs post-insult (P,0.05), but did not affect post-ischemic motor scores (P.0.05; Table 3). The physiological parameters were kept within normal limits and did not differ statistically between study and control animals (data not shown).

Discussion
Our results indicated that magnolol (50-200 mg/kg) reduced infarct volumes and improved neurobehavioral outcomes in rats subjected to permanent focal cerebral ischemia. Additionally, we found that magnolol (100 mg/kg) was effective in reducing brain infarction and improving neurobehavioral outcomes even when administrated up to 4 hrs post-insult. Moreover, we demonstrated that magnolol not only effectively attenuated both glutamate-and NMDA-induced neurotoxicity, but also reduced the glutamate-induced increases in the [Ca 2+ ](i) inflow and neuronal swelling. This neuroprotection cannot be accounted for by changes in glucose, hemodilution (as measured by blood hematocrit), or differences in mean arterial blood pressure, since these were not significantly different when compared between vehicle-injected and magnolol-treated animals. The changed physiologic parameters were decreases in arterial pH and heart rate, associated with a rise in pCO 2 , seen in the animals treated with magnolol at 200 mg/kg. These findings suggested that magnolol (200 mg/kg) might have induced a cardiopulmonary suppression, probably due to its centrally-acting muscular relaxant effect [12].
Exactly by which mechanisms in the glutamate-stimulated cultured neurons the dose-responsive regimen seen with magnolol for cell swelling inhibition was inconsistent with the ''U-shaped'' hormetic response observed for inhibiting the rises of [Ca 2+ ](i) remains to be elucidated [18,19]. Curiously, hormetic neuroprotective responses were also observed in the magnolol-treated stroke animals in which a low-dosing regimen was ineffective whereas high dosage (200 mg/kg) induced adverse effects along with a temperature-lowering action [12,20,21]. Thus, the in vitro dosing response might not represent the trend of dosing response observed in vivo [18]. It was very likely that magnolol actually had multiple mechanisms acted, independently or in combined, to exhibit neuroprotection observed here [12][13][14][15][16][17].
A therapeutic window of 4 hrs seen with magnolol in reducing brain infarction compares favorably with those of glutamate receptor antagonist and other anti-oxidant and radical-scavenging agents, but not as well as that reported with a calpain inhibitor [7,8,22]. Perhaps using multiple effective, smaller doses of magnolol, combined with an intravenous administration route, the therapeutic window may be extended and/or the degree of neuroprotection improved [8,19]. Further studies are needed to determine whether magnolol can protect against reperfusion damage and late-onset ischemic insults following cerebral ischemia/reperfusion after a prolonged reperfusion period [23]. In additional, more mechanisms underlying neuroprotection observed here need to be elucidated.
In conclusion, magnolol protects against permanent focal cerebral ischemia with a therapeutic window up to 4 hrs postinsult. This neuroprotection may be partly mediated by its ability to attenuate the glutamate and NMDA-induced neurotoxicity.

Materials and Methods
All procedures performed were approved by the Subcommittee on Research Animal Care of the University. All chemicals were purchased from Sigma-Aldrich Co. (St Louis, MO) unless

Neuronal Cultures and Cytotoxicity Assay
According the method described previously [18], cultured neurons were obtained from cerebral cortices of 1-day-old Sprague-Dawley rats. Cytotoxicity was determined at 24 hrs after treatment by using a LDH assay kit (Promega, Madison, WI) [18,24,25]. Experiments were undertaken on cultured neurons between 10 and 14 days in vitro (DIV). Neurons were incubated magnolol (0-300 mM) or vehicle (0.1% DMSO). The LD 50 value was defined as the concentration of compound required to induce 50% of cell deaths in 24 hrs at 37uC.

Glutamate-and N-methyl-D-aspartate (NMDA)-induced Cell Cytotoxicity
Cultured neurons were pre-treated with magnolol (0.1-1 mM) or vehicle (0.1% DMSO) for 30 min and, then, were exposed to glutamate (300 mM) or NMDA (100 mM) for 24 hrs. The ED 50 value was defined as the concentration of compound required to reduce 50% of cell deaths of controls in 24 hrs at 37uC. Physiologic data obtained from control and pre-treated animal groups are represented as the mean6standard deviation (SD). Hct -hematocrit; Gluc -blood glucose; MABP -mean arterial blood pressure; HR -heart rate; n -number of animals. All animals were maintained at 3760.5uC. Paired Students' t tests were used to evaluate the response to a change in conditions, and one-way Analysis of Variance (one-way ANOVA) with Dunnett's posthoc comparison was used to evaluate differences between groups. The symbol * and { mean P,0.05, compared to preischemic and control data, respectively. doi:10.1371/journal.pone.0039952.t001

Intracellular Ca 2+ Measurement
The level of [Ca 2+ ](i) were measured on a single cell fluorimeter [26,27]. Briefly, neuronal cultures were incubated with 3 mM fura 2-acetoxymethylester (Fura-2 AM) and 10 mM ionomycin in a standard buffer (composition in mM: NaCl, 140; KCl and Fo/Fs is the fluorescence emitted at 380 nm excitation at minimum Ca 2+ level divided by the same emission fluorescence at the fura-saturated concentration [28]. R is the ratio fluorescence intensity recorded at 340 and 380 nm, and Rmin and Rmax are the rations of 340/380 nm fluorescence intensity recorded at minimum Ca 2+ and the fura-saturated Ca 2+ concentrations, respectively. We used the Fura-2 Calcium Imaging Calibration Kit (F-6774; Invitrogen Molecular Probes, Eugene, OR) to detect the Kd level under conditions. Measurements of Fo and Rmin were performed in nominally Ca 2+ -free isotonic solution containing 10 mM EGTA. Cells were then superfused with isotonic solution containing 1 mM thapsigargin, 10 mM ionomycin and 10 mM Ca 2+ to evaluate Fs and Rmax.

Cell Swelling Measurements
The glutamate (300 mM)-induced neuronal morphologic changes were measured by time-lapse imaging techniques in a microscope equipped with a thermo-controllable heating stage, differential interference contrast (DIC) lens and an image analyzer (MCID Elite) by the method described previously [29,30]. DIC images of pyramid-shaped neurons were measured and compared over time. Three randomly selected fields were counted and averaged per culture (approximately 12 to 15 neurons per culture). Data are expressed as a percentage relative to the baseline values.

Animal Preparation, Anesthesia, and Monitoring
Male Sprague-Dawley rats, weighting 220-270 g, were supplied by the University Laboratory Animal Center, and were allowed free access to food and water before and after surgery. Animals were anesthetized with 1-2% halothane in 70% N 2 O/30% O 2 . During surgery, body temperature was maintained at 3760.5uC using a thermostatically controlled heating blanket and rectal probe (Harvard Apparatus, South Natick, MA). The right femoral artery was cannulated for measuring arterial blood gases, glucose, hematocrit and blood pressure [8,18,24,25].

Neurobehavioral Testing
Neurologic and body weight measurements were conducted by an investigator unaware of treatment protocol at 24 hrs post-insult [8,18,24,25,32]. Five categories of motor neurologic findings were scored: 0, no observable deficit; 1, forelimb flexion; 2, forelimb flexion and decreased resistance to lateral push; 3, forelimb flexion, decreased resistance to lateral push and unilateral circling; 4, forelimb flexion, unable or difficult to ambulate. The affected forelimb also received forward and sideways visual placing tests which were scored as follows: 0, complete immediate placing; 1, incomplete and/or delayed placing (,2 seconds); 2, absence of placing.

Animal Sacrifice and Quantification of Ischemic Damage
Sacrifice was performed at 24 hrs post-insult by decapitation under anesthesia. The brain was cut into 2 mm coronal sections using a rat brain matrix (RBM 4000 C, ASI Instrument, Inc., Warren, MI) and stained according to standard 2, 3, 5triphenyltetrazolium chloride (TTC) method [8,24]. Briefly, the brain was cut into 2 mm coronal sections using a rat brain matrix (RBM 4000 C, ASI Instrument, Inc., Warren, MI) and stained according to the standard 2, 3, 5-triphenyltetrazolium chloride (TTC) method [8,19,24]. A total of 7 brain sections were traced and measured using a computerized image analyzer (MCID Elite). The calculated infarction areas were then compiled to obtain the infarct volumes per brain (in mm 3 ). Brain Infarct volumes were expressed as a percentage of the contralateral hemisphere volume [8,24].

Statistical Analysis
All data were expressed as the mean6standard deviation (S.D.). Paired Students' t test was used to evaluate the response to a change in conditions, and unpaired Students' t test/one-way analysis of variance (one-way ANOVA) with Fisher's protected least significant difference (LSD) posthoc comparison was used to evaluate differences between groups. Neurobehavioral scores were analyzed by the Kruskal-Wallis/Mann-Whitney U test. P,0.05 was selected for statistical significance.