Asiatic acid and maslinic acid attenuated kainic acid-induced seizure through decreasing hippocampal inflammatory and oxidative stress

Epilepsy Research 139 (2018) 28–34

Contents lists available at ScienceDirect

Epilepsy Research journal homepage: www.elsevier.com/locate/epilepsyres

Asiatic acid and maslinic acid attenuated kainic acid-induced seizure through decreasing hippocampal inflammatory and oxidative stress ⁎

Zhi-hong Wanga, , Mei-chin Monga, Ya-chen Yanga, Mei-chin Yina,b, a b

MARK

⁎

Department of Food Nutrition and Health Biotechnology, Asia University, Taichung City, Taiwan Department of Medical Research, China Medical University Hospital, China Medical University, Taichung City, Taiwan

A R T I C L E I N F O

A B S T R A C T

Keywords: Asiatic acid Maslinic acid Seizure Glutamine Calcium release

Seizure is a neurological disorder including hippocampal oxidative and inflammatory stress, and glutamate toxicity. Thus, any agent(s) that mitigate(s) these events in hippocampus might attenuate seizure severity. The effects of asiatic acid (AA) or maslinic acid (MA) pre-administration at 20 or 40 mg/kg body weight/day upon inflammatory, oxidative and apoptotic injury in hippocampus of kainic acid (KA)-treated mice were examined. KA induced seizure-like behavioral patterns, which was attenuated by AA or MA pre-administration. KA stimulated the release of interleukin (IL)-1beta, IL-6, tumor necrosis factor-alpha and prostaglandin E2 in hippocampus of mice. AA or MA pre-administration decreased the production of these inflammatory factors. AA or MA also diminished KA-induced increase in hippocampal cyclooxygenase-2 activity and relative NF-κB p50/65 binding activity. KA depleted glutathione content and promoted reactive oxygen species generation. AA or MA pre-administration reversed these alterations. KA lowered Bcl-2 mRNA expression and increased Bax mRNA expression. AA or MA treatments reduced Bax mRNA expression. AA or MA pre-administration enhanced glutamine synthetase activity, decreased glutamate level and increased glutamine level in hippocampus of KA treated mice. In addition, AA or MA pre-treatments at 10 and 20 μM increased viability and decreased plasma membrane damage in KA treated nerve growth factor (NGF)-differentiated PC12 cells. Both agents also lowered the release of calcium ion induced by KA in NGF-treated PC12 cells. These findings support that asiatic acid and maslinic acid are potent nutraceutical agents for seizure alleviation.

1. Introduction Seizure, a neurological disorder, is characterized by involuntary shaking of partial or entire body, and sometimes causes consciousness loss (Chen et al., 2017). Both experimental and clinical evidence indicate that inflammatory stress is involved in the etiopathogenesis of seizure, especially in the area of hippocampus (Friedman, 2011; Vezzani et al., 2011). The activation of crucial inflammatory mediators such as cyclooxygenase (COX)-2 and nuclear factor kappa B (NF-κB), and the over-production of the down-stream inflammatory factors including interleukin (IL)-1beta, IL-6, tumor necrosis factor (TNF)-alpha and prostaglandin E2 (PGE2) contributed to the progression of seizure (Huang et al., 2012; Teocchi et al., 2013). On the other hand, excessive generation of reactive oxygen species (ROS) promotes seizure related to neuronal membrane depolarization, and exacerbates neuron malfunctions (Kovac et al., 2014). Consequently, both inflammatory and oxidative injuries cause the death of brain nerve cells (Murashima et al., 2005). In addition, glutamate excitotoxicity is another important element in charge of seizure induction (Ravizza et al., 2011). Eid et al. ⁎

(2013) reported that increased extracellular glutamate level in hyperexcitable areas of brain trigged seizure through activating glutamate receptor and genes involved in synaptic plasticity. Glutamine synthetase (GS) could metabolize glutamate to glutamine, and facilitates glutamate clearance (Rosati et al., 2009). Therefore, any agent(s) with the capability to diminish inflammatory, oxidative and/or glutamate toxicity in hippocampus may attenuate seizure severity. Kainic acid (KA), a glutamate related compound, could induce neuronal excitability and neuronal membrane depolarization by releasing calcium ions to impair nerve impulse transmission (Malva et al., 2003). KA-induced seizure in rodents has been widely used as an experimental seizure model for the associated pathological, preventive and therapeutic researches (Gupta et al., 2002; Huang et al., 2012). Furthermore, KA-induced seizure has been considered as a model of lesional epilepsy because KA causes focal hippocampal lesion (Craig et al., 2008). Asiatic acid (AA) and maslinic acid (MA) are pentacyclic triterpenes (Fig. 1) naturally occurring in many edible plant foods such as centella (Centella asiatica L.), olive (Olea europaea L.), brown mustard (Brassica juncea) and gynura (Gynura bicolor DC) (Yin et al., 2012;

Corresponding authors at: Department of Food Nutrition and Health Biotechnology, Asia University, Taichung City, Taiwan. E-mail addresses: [email protected] (Z.-h. Wang), [email protected] (M.-c. Yin).

https://doi.org/10.1016/j.eplepsyres.2017.11.003 Received 13 July 2017; Received in revised form 13 October 2017; Accepted 11 November 2017 0920-1211/ © 2017 Elsevier B.V. All rights reserved.

Epilepsy Research 139 (2018) 28–34

Z.-h. Wang et al.

cages under a 12-h light/dark cycle and a constant temperature, 22 ± 2 °C. Mice were fed by water and standard diet. Mice with body weight at 24.9 ± 1.4 g were used in all experiments. Use of these mice was approved by China Medical University Animal Care Committee, and the approval number was 103-27-N. All procedures were handled carefully in order to minimize mice suffering within experiments.

2.3. Experimental design AA or MA at 20 or 40 mg/kg body weight (BW)/day was orally administrated once per day for 7 consecutive days, and followed by KA treatment at 20 mg/kg BW intraperitoneally for seizure induction. Mice in control and KA groups received MC oral administration for 7 days. Then, mice in control groups were treated by intraperitoneal injection of MC, and mice in KA groups were intraperitoneally injected by KA (20 mg/kg BW). Each group had 7 mice. After KA challenge, seizure behavior was monitored and rated according to the modified Racine’s scale (Wang et al., 2014): stage 0, normal behavior; stage 1, facial automatisms; stage 2, facial and head clonus; stage 3, forelimb clonus; stage 4, rearing; stage 5, rearing, loss of balance and falling. Seizure score was recorded at 0, 30, 60, 90, 120, 150 and 180 min after KA challenge. One day after the KA treatment, mice were sacrificed by decapitation under CO2 asphyxia. The whole brain was dissected and the hippocampus of each mouse was collected for analyses. AA, MA or KA treatment did not cause mouse loss during experimental period.

2.4. Measurement of inflammatory factors

Fig. 1. Structure of asiatic acid and maslinic acid.

Ten mg hippocampus was homogenized in a Tris solution (10 mM, pH 7.4) composted of 1 mM EDTA, 1 mM phenylmethylsulfonyl fluoride and 0.01% Tween 80. After centrifuging at 9000 × g for 30 min at 4 °C, IL-1beta, IL-6 or TNF-alpha content in supernatant was assayed by cytoscreen kits (BioSource International, Camarillo, CA, USA). PGE2 level was determined by an EIA kit obtained from Cayman Chemical Co. (Ann Arbor, MI, USA).

James et al., 2013; Sánchez-Quesada et al., 2013). Lee et al. (2014) indicated that AA displayed neuroprotective activities for focal embolic stroke in rats via lowering the release of apoptosis-inducing factor from brain mitochondria. Our previous animal study revealed that dietary AA intake alleviated the progression of Parkinson’s disease by suppressing striatal expression of α-synuclein and increasing striatal dopamine level (Chao et al., 2016). Huang et al. (2011) reported that MA limited NF-κB expression and ameliorated inflammatory injury in cortical astrocytes. Qian et al. (2015) indicated that MA benefited synaptogenesis in cerebral ischemia model through activating Akt/GSK3β. Those previous studies suggest that AA and MA are potent protective agents for brain. Therefore, both animal and cell line studies were designed to evaluate the anti-seizure activities of AA and MA. In our present study, the effects of AA and MA pre-administrations at various doses upon inflammatory, oxidative and apoptotic injury in hippocampus of KA-treated mice were examined. The impact of these agents upon seizure behavior, and hippocampal variation of inflammatory and oxidative factors, glutamine level and GS activity was measured. In addition, the activities of AA or MA against KA induced apoptosis and calcium release in nerve growth factor (NGF) differentiated PC12 cells were determined.

2.5. Assay of caspase activity The activity of caspase-3 and caspase-8 was assayed by fluorometric kits (Upstate, Lake Placid, NY, USA). The values of coefficient of variability for inter-assay and intra-assay were 4.7–6.1% and 4.2–5.3%, respectively. Ten mg hippocampus was lysed by lysis buffer. Lysates were then mixed with reaction buffer and specific fluorogenic substrates for caspase-3 or -8. After 1 h incubation at 37 °C, fluorescence was monitored by a Hitachi fluorophotometer (F-4500, Tokyo, Japan), excitation wavelength was 400 nm and emission wavelength was 505 nm. Protein content was measured by a commercial assay kit (Pierce, Rockford, IL, USA). Activity was showed as fluorescence unit/ mg protein.

2. Materials and methods 2.6. NF-κB p50/65 binding activity assay 2.1. Materials Nuclear extract was prepared according to the method of Schilling et al. (1999). Ten μg nuclear protein extract was used for measuring NFκB p50/65 binding activity. NF-κB p50/65 binding activity was assayed by a commercial kit purchased from Chemicon International Co. (Temecula, CA, USA). It was processed by adding the substrate, 3, 3′, 5, 5′tetramethylbenzidine, a primary NF-κB p50/p65 antibody. After 1 h incubation at room temperature and washing by PBS twice, a second horseradish peroxidase-conjugated antibody was added for another 1 h. The variation of absorbance at 450 nm was monitored by a microtiter plate reader (Model 550, Bio-Rad, Hercules, CA, USA). Data are expressed as optical density (OD) value/mg protein.

AA (98%), MA (97%), KA (99.5%) and NGF (99.5%) were purchased from Sigma Chemical Co. (St. Louis, MO, USA). AA or MA was suspended in 0.8% methyl cellulose (MC). KA was dissolved in phosphate-buffered saline (PBS, pH 7.2). Antibodies were purchased from Boehringer Ingelheim Co. (Ingelheim am Rhein, Germany). 2.2. Animals Male C57BL/6 mice at 4-week old were obtained from the National Laboratory Animal Center (Taipei City, Taiwan). Mice were housed in 29

Epilepsy Research 139 (2018) 28–34

Z.-h. Wang et al.

48 h at 37 °C, which led to 95.4–97.3% incorporation of AA and MA into NGF-differentiated PC12 cells. Cells treated by AA or MA were washed by PBS twice. Those PBS solutions were collected, and AA or MA concentration was detected by HPLC according to the method of Lin et al. (2011). AA or MA left in NGF-differentiated PC12 cells was thought as incorporated. Cells without AA, MA and KA treatments were control groups. NGF-differentiated PC12 cells with or without AA or MA treatments were further treated by 150 μM KA.

2.7. Measurement of oxidative factors An oxidation sensitive dye, 2′,7′-dichlorofluorescein diacetate (DCFH-DA), was used to measure ROS level. Briefly, 100 μl hippocampus homogenate was mixed with 100 μl DCFH-DA at 2 mg/ml. Fluorescence value was recorded after 30 min incubation at 37 °C by a fluorescence plate reader with excitation and emission wavelengths at 488 nm and 525 nm, respectively. Relative fluorescence unit (RFU) per mg protein was shown as result. Hippocampal levels of reduced glutathione (GSH) and oxidized glutathione (GSSG) were measured by assay kits purchased from OxisResearch (Portland, OR, USA). Commercial kits obtained from EMD Biosciences, Inc. (San Diego, CA, USA) were used to analyze the activity (U/mg protein) of glutathione peroxide (GPX) or glutathione reductase (GR).

2.11. Cell viability and plasma membrane damage assays Cell viability was examined by 3-(4,5-dimethylthiazol-2-yl)-2,5-diphenyltetrazolium bromide (MTT) assay. Twenty-four hr after KA treatment, MTT at 0.25 mg/ml was added and incubated for 3 h at 37 °C. The absorbance at 570 nm was monitored to determine MTT formazan product by a Bio-Rad microplate reader (Hercules, CA, USA). Cell viability was shown as a percentage of control groups. Cell plasma membrane damage was analyzed by measuring the activity (U/l) of lactate dehydrogenase (LDH). Fifty μl of supernatant was processed by a LDH kit (Sigma Chemical Co., St. Louis, MO, USA).

2.8. Determination of glutamate, glutamine and GS activity Hippocampus homogenate was mixed with sodium citrate buffer or lithium citrate buffer to quantify glutamate or glutamine content, respectively. After centrifugation, the level of these amino acids in supernatant was quantified by a Hitachi L-8800 amino acid analyzer (Hitachi, Tokyo, Japan). Each amino acid peak was identified by its retention time compared with an external standard. Amino acid concentration was determined according to the relative peak height, and expressed in ng/mg protein. GS activity (U/mg protein) was determined by a commercial kit purchased from Jiancheng Institute of Biotechnology (Nanjing, China). Ten mg hippocampus was homogenized in lysis buffer composed of 100 mM Tris-HCl, 1 mM ethyleneglycoltetraacetic acid, 1 mM ethylenediaminetetraacetic acid and protease inhibitor. GS activity was quantified by using a standard curve made with pre-determined glutamyl-thydroxamate.

2.12. Calcium release Fura-2AM, a Ca2+-sensitive dye, was used to quantify intracellular Ca concentration according to the method of Lenart et al. (2004). A solution composed of Fura-2AM (5 mmol/l), 1% BSA and 0.1% DMSO was added. After incubating at 37 °C for 30 min in dark condition, fluorescence value was detected by a Shimadzu spectrofluorimeter (Model RF-5000, Kyoto, Japan). The absorbance at 340 and 380 nm wavelengths for excitation, and 510 nm wavelength for emission were recorded. Calcium concentration was calculated according to the equation: [Ca2+] (nM) = Kd × [(R − Rmin)/(Rmax − R)] × FD/FS. Kd was 224 nM, R was the ratio of 340:380, Rmin or Rmax was measured by treating cells with ethylene glycol tetra-acetic acid or triton X100, respectively. FD and FS were the fluorescence values of Ca2+-free form and Ca2+-bound form detected at 380 and 340 nm, respectively. 2+

2.9. Real-time polymerase chain reaction for mRNA expression Total RNA of 10 mg hippocampus was isolated by Trizol reagent (Invitrogen, Life Technologies, Carlsbad, CA, USA). One μg RNA was used for cDNA generation, and further amplified by using Taq DNA polymerase. Reaction mixture at 50 μl consisted of 2.5 U Taq DNA polymerase, 0.5 mM of each primer, 20 mM Tris-HCl, 50 mM KCl, 2.5 mM MgCl2 and 200 mM dNTP was applied for PCR process. The used oligonucleotide primers are as follow: Bcl-2, forward, 5′-GTG GAT GAC TGA GTA CCT GAA C-3′, reverse, 5′-GAG ACA GCC AGG AGA AAT CAA-3′; Bax, forward, 5′-GCT GAT GGC AAC TTC AAC TG-3′, reverse, 5′-ATC AGC TCG GGC ACT TTA G-3′; glyceraldehyde-3-phosphate dehydrogenase (GAPDH), forward, 5′-AGA GGC AGG GAT GTT CTG-3′, reverse, 5′-GAC TCA TGA CCA CAG TCC ATG C-3′. The condition for cDNA amplification was as follow, 95 °C for 3 min, 95 °C for 10s, and 56 °C for 30 s. Forty cycles were performed for Bcl-2 and Bax, and 28 cycles were performed for GAPDH, the housekeeping gene. Real-time sequence detection Taqman system was used to quantify the produced fluorescence. mRNA level was shown as Bcl-2/GAPDH or Bax/GAPDH.

2.13. Statistical analyses The effect of each treatment was measured from 7 mice or 5 different preparations for NGF-treated PC12 cells in each group. All data were expressed as mean ± standard deviation (SD). Statistical analysis was processed by using one-way analysis of variance. Post-hoc comparison was also carried out by using Dunnett’s t-test. p value < 0.05 was considered as significant. 3. Results 3.1. AA and MA decreased seizure scores KA induced seizure-like behavioral patterns, the scoring of which appears in Fig. 2 (p < 0.05). AA or MA pre-administration at both doses decreased seizure scores at each time point within 30–180 min when compared with KA treatment alone (p < 0.05). Seizure scores were lower than 1 in mice with AA or MA pre-administration at 40 mg/ kg BW/day at 180 min after KA challenge.

2.10. Cell culture and treatments PC12 cells, obtained from ATCC (Rockville, MD, USA), were cultured in Dulbecco’s modified Eagle’s medium (DMEM) under 95% air and 5% CO2 at 37 °C. PC12 cells, through reacting with NGF, could be differentiated to neuronal phenotype. This NGF-differentiated PC12 cell line is often applied as a sympathetic neurons model (Lipman et al., 2006). In present study, PC12 cells were reacted with NGF at 50 ng/ml, and followed by 5-day incubation at 37 °C for differentiation. Cells were sub-cultured every 7 days, and medium was refreshed every three days. After washing with serum-free DMEM, cells were planted in 96 well plates. Cell number was adjusted by PBS to 105/ml for experiments. AA or MA, suspending in 0.8% methyl cellulose, was diluted with DMEM. Cells (105 cells/ml) were treated with AA or MA at 0, 5, 10 or 20 μM for

3.2. AA and MA attenuated inflammatory and apoptotic injury As shown in Table 1, KA stimulated the release of IL-1beta, IL-6, TNF-alpha and PGE2, and increased COX-2 activity in hippocampus of mice (p < 0.05). AA or MA pre-administration dose-dependently decreased the production of these inflammatory factors (p < 0.05). MA pre-administration at 40 mg/kg BW/day led to lower IL-1beta, TNFalpha, PGE2 levels and COX-2 activity than AA pre-administration at equal dose (p < 0.05). KA increased the activity of caspase-3, caspase30

Epilepsy Research 139 (2018) 28–34

Z.-h. Wang et al.

Fig. 2. Effects of AA or MA on seizure behavior in KA treated mice. Mice with AA or MA supplements at 0, 20 or 40 mg/kg BW/day for 7 days and followed by KA treatment. Seizure score was recorded at 0, 30, 60, 90, 120, 150 and 180 min after KA challenge. Values are mean ± SD, n = 7. a–eValues among treatments at the same time point without a common letter differ, p < 0.05.

8 and relative NF-κB p50/65 binding (Fig. 3, p < 0.05). AA or MA preadministration reduced caspase-3 and relative NF-κB binding activities (Fig. 3a and c, p < 0.05). Both agents only at 40 mg/kg BW/day decreased caspase-8 activity (Fig. 3b, p < 0.05). MA at 40 mg/kg BW/ day exhibited greater reduction in relative NF-κB binding activity than AA at equal dose (Fig. 3c, p < 0.05). KA reduced Bcl-2 mRNA expression and increased Bax mRNA expression in mice hippocampus (Fig. 4a and b, p < 0.05). AA or MA pre-administration at both doses suppressed Bax mRNA expression (Fig. 4b, p < 0.05). 3.3. AA and MA alleviated oxidative injury As shown in Table 2, KA depleted GSH level, increased GSSG and ROS levels, and decreased GPX and GR activities in mice hippocampus (p < 0.05). AA or MA at both doses reversed these changes (p < 0.05). AA was greater than MA in lowering GSSG and ROS formation, and raising GPX and GR activities (p < 0.05). 3.4. AA and MA altered glutamate and glutamine levels, and GS activity AA or MA pre-administration at both doses reduced hippocampal glutamate level, and increased glutamine level and GS activity when compared with KA treatment alone (Fig. 5a–c, p < 0.05), in which MA at 40 mg/kg BW/day was greater than AA at equal dose in deceasing glutamate and increasing glutamine levels (Fig. 5a and b, p < 0.05).

Fig. 3. Activity of caspase-3 (fluorescence unit/mg protein, a), caspase-8 (fluorescence unit/mg protein, b), and NF-κB p50/65 binding (OD value/mg protein, c) in hippocampus of mice with AA or MA supplements at 0, 20 or 40 mg/kg BW/day for 7 day and followed by KA treatment. Values are mean ± SD, n = 7. a–eValues among bars without a common letter differ, p < 0.05.

3.5. AA and MA increased viability and reduced Ca2+ release in NGFdifferentiated PC12 cells

4. Discussion

As shown in Table 3, AA and MA pre-treatments at 10 and 20 μM increased cell viability and decreased LDH activity in KA treated NGFdifferentiated PC12 cells (p < 0.05). These agents also at 10 and 20 μM reduced the release of calcium ion induced by KA (p < 0.05), in which MA displayed greater effect than AA at equal dose (p < 0.05).

In our present study, KA challenge induced seizure-like behavior, which was evidenced by greater seizure scores. However, AA or MA pre-administration lowered seizure scores at test time points in those KA-treated mice. Apparently, these two compounds effectively improved seizure related abnormal behavior. Furthermore, our animal

Table 1 Level of IL-1beta (pg/ml), IL-6 (pg/ml), TNF-alpha (pg/ml) and PGE2 (pg/g protein), and COX-2 activity (U/mg protein) in hippocampus of mice with AA or MA supplements at 0, 20 or 40 mg/kg BW/day for 7 day and followed by KA treatment. control IL-1beta IL-6 TNF-alpha PGE2 COX-2

KA a

13 ± 3 10 ± 2a 12 ± 3a 995 ± 34a 0.23 ± 0.05a

Values are mean ± SD, n = 7.

AA, 20 +KA e

191 ± 17 163 ± 10d 217 ± 19e 2603 ± 102e 2.41 ± 0.19e

d

142 ± 13 105 ± 11c 161 ± 16d 2231 ± 95d 1.91 ± 0.15d

a–e

Values in a row without a common letter differ, p < 0.05.

31

AA, 40 +KA c

98 ± 6 75 ± 7b 112 ± 14c 1793 ± 72c 1.29 ± 0.12c

MA, 20 +KA d

132 ± 18 113 ± 9c 123 ± 10c 2154 ± 86d 1.75 ± 0.1d

MA, 40 +KA 70 ± 8b 83 ± 4b 73 ± 7b 1404 ± 58b 0.88 ± 0.09b

Epilepsy Research 139 (2018) 28–34

Z.-h. Wang et al.

Fig. 4. mRNA expression of Bcl-2 (Bcl-2/GAPDH, a) and Bax (Bax/GAPDH, b) in hippocampus of mice with AA or MA supplements at 0, 20 or 40 mg/kg BW/day for 7 day and followed by KA treatment. Values are mean ± SD, n = 7. a–eValues among bars without a common letter differ, p < 0.05.

data indicated that 7-day AA or MA pre-administration at 20 or 40 mg/ kg BW/day markedly protected hippocampus against subsequent KA induced inflammatory, oxidative and apoptotic stress, increased glutamine level and glutamine synthetase activity. In addition, our results from NGF-differentiated PC12 cells revealed that AA or MA pre-treatments overcame KA toxicity, which was evidenced by increasing cell viability and decreasing calcium release. These novel findings suggest that AA and MA, through providing multiple activities, are potent preventive agents against seizure. Brain inflammatory injury has been considered as a crucial pathological contributor toward the development and recurrence of seizure (Legido and Katsetos, 2014). Both caspase-3 and caspase-8 are apoptotic executors, and the increased activity of these two caspases promotes seizure related cell death (Fujikawa et al., 2007). In our present study, KA enhanced COX-2, caspase-3, caspase-8 and relative NF-κB p50/65 binding activities in hippocampus, which consequently raised apoptotic stress, and stimulated the release of inflammatory factors such as IL-1beta, IL-6, TNF-alpha and PGE2. However, AA or MA pre-

Fig. 5. Level of glutamate (ng/mg protein, a), glutamine (ng/mg protein, b) and GS activity (U/mg protein, c) in hippocampus of mice with AA or MA supplements at 0, 20 or 40 mg/kg BW/day for 7 day and followed by KA treatment. Values are mean ± SD, n = 7. a–eValues among bars without a common letter differ, p < 0.05.

Table 2 Level of GSH (nmol/mg protein), GSSG (nmol/mg protein), ROS (RFU/mg protein), and activity (U/mg protein) of GPX and GR in hippocampus of mice with AA or MA supplements at 0, 20 or 40 mg/kg BW/day for 7 day and followed by KA treatment. control GSH GSSG ROS GPX GR

KA d

89 ± 3 0.37 ± 0.06a 0.19 ± 0.04a 32.7 ± 2.2e 22.5 ± 1.3e

Values are mean ± SD, n = 7.

AA, 20 +KA a

32 ± 4 2.88 ± 0.21f 2.57 ± 0.19f 18.7 ± 0.7a 11.4 ± 0.5a

b

51 ± 2 1.96 ± 0.17d 1.68 ± 0.09d 24.5 ± 1.2c 14.0 ± 0.8b

a–f

Values in a row without a common letter differ, p < 0.05.

32

AA, 40 +KA c

77 ± 5 1.16 ± 0.11b 0.91 ± 0.13b 28.3 ± 1.5d 19.0 ± 0.6d

MA, 20 +KA b

48 ± 4 2.37 ± 0.14e 1.97 ± 0.15e 21.9 ± 0.8b 13.3 ± 0.2b

MA, 40 +KA 73 ± 3c 1.62 ± 0.08c 1.29 ± 0.1c 25.1 ± 1.6c 16.9 ± 1.0c

Epilepsy Research 139 (2018) 28–34

Z.-h. Wang et al.

toxicity. Because the analysis of intracellular Ca2+ concentration in hippocampus was more difficult, we used those cells to examine the effects of AA or MA upon KA induced Ca2+ release. It is known that released calcium ion interferes nerve impulse transmission and stimulates neuronal excitability, which subsequently stimulates the recurrence of seizure (Gupta et al., 2002; Malva et al., 2003). As reported by others (Takemiya et al., 2011; Zheng et al., 2013), our data revealed that KA led to massive release of Ca2+ in NGF-differentiated PC12 cells. However, we found that the addition of AA or MA substantially reduced Ca2+ release in KA treated NGF-differentiated PC12 cells. These results implied that these two agents were able to attenuate seizure through limiting Ca2+ release in neuronal cells. It is reported that dietary AA and/or MA intake increased their bioavailability in mice brain (Yin et al., 2012). The novel findings from our present work suggested that AA and MA supplement benefited the stability of mice hippocampus. These results suggested that AA and MA might be able to penetrate blood brain barrier and execute their protective actions. The intake of AA or MA at 40 mg/kg BW/day for mice is approximately equal to 2.8 g/day for an adult with 70 kg BW. This dose might be safe and feasible because AA and MA are pentacyclic triterpenoic acids naturally presented in many edible vegetables, fruits or herbs. This study had at least two limitations, one was that electroencephalogram was not used for diagnosis or treatment of seizures. Another was that the therapeutic effects (post-administration) of AA and/or MA for seizures remained unsure. Therefore, further in vivo studies are definitely necessary to use electroencephalogram and examine the therapeutic effects of both agents in order to confirm the effects and safety of these two agents upon seizures improvement. In conclusion, asiatic acid or maslinic acid pre-administration at 20 or 40 mg/kg BW/day improved seizure-like behavior, and attenuated hippocampal inflammatory, oxidative and apoptotic injury in kainic acid treated mice. These agents also increased survival and decreased Ca2+ release in kainic acid treated NGF-differentiated PC12 cells. These findings support that asiatic acid and maslinic acid are potent nutraceutical agents for seizure alleviation.

Table 3 Effects of AA and MA upon cell viability (% of control), determined by MTT assay, plasma membrane damage, determined by LDH activity (U/l), and calcium release (Ca2+, nM) in NGF-differentiated PC12 cells. Cell number in control group was 105/ml. Cells were pretreated by AA or MA at 0, 5, 10 or 20 μM and followed by 150 μM KA treatment.

control KA AA, 5 +KA AA, 10 +KA AA, 20 +KA MA, 5 +KA MA, 10 +KA MA, 20 +KA

Cell viability

LDH activity

Ca2+

100d 34 ± 38 ± 56 ± 74 ± 39 ± 60 ± 78 ±

40 ± 3a 223 ± 24e 208 ± 21e 163 ± 9d 128 ± 13c 210 ± 17e 151 ± 12d 97 ± 8b

254 ± 28a 1602 ± 124e 1531 ± 112e 1132 ± 73d 847 ± 50c 1519 ± 89e 924 ± 66c 517 ± 48b

Values are mean ± SD, n = 5. p < 0.05.

2a 4a 3b 4c 3a 5b 4c a–e

Values in a column without a common letter differ,

administration decreased caspase-3 activity, which in turn declined brain apoptotic stress. Furthermore, AA or MA pre-administration at both doses markedly diminished the activity of COX-2 and relative NFκB binding. Since the activity of these up-stream mediators has been restricted, the observed less production of inflammatory factors such as TNF-alpha and PGE2 in hippocampus could be explained. In addition, KA reduced Bcl-2 mRNA expression and increased Bax mRNA expression, which definitely enhanced apoptotic stress in mice hippocampus. However, AA and MA pre-treatments at test doses limited mRNA expression of Bax, an apoptotic factor, which subsequently attenuated apoptotic injury of hippocampus. These findings suggest that AA or MA at test doses effectively ameliorated KA induced inflammatory and apoptotic injury, which consequently benefited neuronal stability. Thus, the improvement in seizure-like behavior in KA treated mice as we observed could be partially ascribed to AA or MA mitigating hippocampal inflammatory stress. Hippocampal oxidative stress is another important element responsible for seizure severity (Azam et al., 2012). In our present study, KA decreased the activity of GPX and GR, which in turn depleted GSH content, raised GSSG level and increased ROS generation in hippocampus. Obviously, KA was able to weaken hippocampal anti-oxidative defense. However, AA or MA pre-administration at both doses substantially maintained the activity of GPX and GR, which in turn decreased GSSG generation and reserved GSH. Since hippocampal antioxidative defensive capability has been improved, it seems reasonable to observe lower ROS level in AA or MA treated mice hippocampus. Thus, the attenuation in hippocampal oxidative injury could be partially due to AA or MA benefiting GSH homeostasis. Glutamate is a neurotransmitter involved in synaptic excitation of central nervous system. Glutamate toxicity is mainly mediated by ROS because greater extracellular glutamate blocks cellular cysteine uptake, which subsequently decreases intracellular level of cysteine, a GSH precursor (Rosin et al., 2004). Consequently, GSH loss favored ROS or GSSG accumulation, and even caused cell death. In our present study, KA treatment markedly raised hippocampal glutamate content, which partially explained the observed greater oxidative injury in hippocampus. However, AA or MA pre-administration effectively lowered ROS production and enhanced the activity of glutamine synthase, an enzyme responsible for the conversion of glutamate to glutamine. The less glutamate content and greater glutamine level in hippocampus of AA or MA treated mice agreed that these two compounds were efficient agents for glutamate clearance in injured nervous tissue. Since glutamate level has been decreased, it is reasonable to observe the alleviation in glutamate induced oxidative stress and seizure-like behavior. In our cell lines study, KA treatment caused apoptosis and impaired plasma membrane in NGF-differentiated PC12 cells. However, AA or MA at test concentrations increased viability and decreased LDH activity of those NGF-differentiated PC12 cells. These results suggest that these AA or MA could protect those cells against KA induced cellular

Conflict of interest statement All authors declare that there is no conflict of interest. Acknowledgement This study was partially supported by a grant from China Medical University (CMU105-ASIA-012). References Azam, F., Prasad, M.V., Thangavel, N., 2012. Targeting oxidative stress component in the therapeutics of epilepsy. Curr. Top. Med. Chem. 12, 994–1007. Chao, P.C., Lee, H.L., Yin, M.C., 2016. Asiatic acid attenuated apoptotic and inflammatory stress in the striatum of MPTP-treated mice. Food Funct. 7, 1999–2005. Chen, T., Giri, M., Xia, Z., Subedi, Y.N., Li, Y., 2017. Genetic and epigenetic mechanisms of epilepsy: a review. Neuropsychiatr. Dis. Treat. 13, 1841–1859. Craig, L.A., Hong, N.S., Kopp, J., McDonald, R.J., 2008. Reduced cholinergic status in hippocampus produces spatial memory deficits when combined with kainic acid induced seizures. Hippocampus 18, 1112–1121. Eid, T., Tu, N., Lee, T.S., Lai, J.C., 2013. Regulation of astrocyte glutamine synthetase in epilepsy. Neurochem. Int. 63, 670–681. Friedman, A.D.R., 2011. Molecular cascades that mediate the influence of inflammation on epilepsy. Epilepsia 52, 33–39. Fujikawa, D.G., Shinmei, S.S., Zhao, S., Aviles Jr., E.R., 2007. Caspase-dependent programmed cell death pathways are not activated in generalized seizure-induced neuronal death. Brain Res. 1135, 206–218. Gupta, Y.K., Briyal, S., Chaudhary, G., 2002. Protective effect of transresveratrol against kainic acid-induced seizures and oxidative stress in rats. Pharmacol. Biochem. Behav. 71, 245–249. Huang, L., Guan, T., Qian, Y., Huang, M., Tang, X., Li, Y., Sun, H., 2011. Anti-inflammatory effects of maslinic acid, a natural triterpene, in cultured cortical astrocytes via suppression of nuclear factor-kappa B. Eur. J. Pharmacol. 672, 169–174. Huang, H.L., Lin, C.C., Jeng, K.C., Yao, P.W., Chuang, L.T., Kuo, S.L., Hou, C.W., 2012. Fresh green tea and gallic acid ameliorate oxidative stress in kainic acid-induced

33

Epilepsy Research 139 (2018) 28–34

Z.-h. Wang et al.

Rosati, A., Marconi, S., Pollo, B., Tomassini, A., Lovato, L., Maderna, E., Maier, K., Schwartz, A., Rizzuto, N., Padovani, A., Bonetti, B., 2009. Epilepsy in glioblastoma multiforme: correlation with glutamine synthetase levels. J. Neurooncol. 93, 319–324. Rosin, C., Bates, T.E., Skaper, S.D., 2004. Excitatory amino acid induced oligodendrocyte cell death in vitro: receptor-dependent and -independent mechanisms. J. Neurochem. 90, 1173–1185. Sánchez-Quesada, C., López-Biedma, A., Warleta, F., Campos, M., Beltrán, G., Gaforio, J.J., 2013. Bioactive properties of the main triterpenes found in olives, virgin olive oil, and leaves of Olea europaea. J. Agric. Food Chem. 61, 12173–12182. Schilling, G., Wood, J.D., Duan, K., Slunt, H.H., Gonzales, V., Yamada, M., Cooper, J.K., Margolis, R.L., Jenkins, N.A., Copeland, N.G., Takahashi, H., Tsuji, S., Price, D.L., Borchelt, D.R., Ross, C.A., 1999. Nuclear accumulation of truncated atrophin-1 fragments in a transgenic mouse model of DRPLA. Neuron 24, 275–286. Takemiya, T., Matsumura, K., Sugiura, H., Yasuda, S., Uematsu, S., Akira, S., Yamagata, K., 2011. Endothelial microsomal prostaglandin E synthase-1 facilitates neurotoxicity by elevating astrocytic Ca2+ levels. Neurochem. Int. 58, 489–496. Teocchi, M.A., Ferreira É, A., da Luz de Oliveira, E.P., Tedeschi, H., D'Souza-Li, L., 2013. Hippocampal gene expression dysregulation of Klotho, nuclear factor kappa B and tumor necrosis factor in temporal lobe epilepsy patients. J. Neuroinflamm. 10, 53. Vezzani, A., French, J., Bartfai, T., Baram, T.Z., 2011. The role of inflammation in epilepsy. Nat. Rev. Neurol. 7, 31–40. Wang, D., Ren, M., Guo, J., Yang, G., Long, X., Hu, R., Shen, W., Wang, X., Zeng, K., 2014. The inhibitory effects of Npas4 on seizures in pilocarpine-induced epileptic rats. PLoS One 9, e115801. Yin, M.C., Lin, M.C., Mong, M.C., Lin, C.Y., 2012. Bioavailability, distribution, and antioxidative effects of selected triterpenes in mice. J. Agric. Food Chem. 60, 7697–7701. Zheng, H., Wang, X., Tang, Z., Zheng, W., Li, Z., 2013. The PI3K/Akt and ERK1/2 signaling pathways mediate the erythropoietin-modulated calcium influx in kainic acidinduced epilepsy. Neuroreport 24, 335–341.

status epilepticus. J. Agric. Food Chem. 60, 2328–2336. James, J.T., Tugizimana, F., Steenkamp, P.A., Dubery, I.A., 2013. Metabolomic analysis of methyl jasmonate-induced triterpenoid production in the medicinal herb Centella asiatica (L.) urban. Molecules 18, 4267–4281. Kovac, S., Domijan, A.M., Walker, M.C., Abramov, A.Y., 2014. Seizure activity results in calcium- and mitochondria-independent ROS production via NADPH and xanthine oxidase activation. Cell. Death. Dis. 5, e1442. Lee, K.Y., Bae, O.N., Weinstock, S., Kassab, M., Majid, A., 2014. Neuroprotective effect of asiatic acid in rat model of focal embolic stroke. Biol. Pharm. Bull. 37, 1397–1401. Legido, A., Katsetos, C.D., 2014. Experimental studies in epilepsy: immunologic and inflammatory mechanisms. Semin. Pediatr. Neurol. 21, 197–206. Lenart, B., Kintner, D.B., Shull, G.E., Sun, D., 2004. Na-K-Cl cotransporter-mediated intracellular Na+ accumulation affects Ca2+ signaling in astrocytes in an in vitro ischemic model. J. Neurosci. 24, 9585–9597. Lin, Y., Vermeer, M.A., Trautwein, E.A., 2011. Triterpenic acids present in hawthorn lower plasma cholesterol by inhibiting intestinal ACAT activity in hamsters. Evid. Based Complement. Alternat. Med. 2011. Lipman, T., Tabakman, R., Lazarovici, P., 2006. Neuroprotective effects of the stable nitroxide compound Tempol on 1-methyl-4-phenylpyridinium ion-induced neurotoxicity in the nerve growth factor-differentiated model of pheochromocytoma PC12 cell. Eur. J. Pharmacol. 549, 50–57. Malva, J.O., Silva, A.P., Cunha, R.A., 2003. Presynaptic modulation controlling neuronal excitability and epileptogenesis: role of kainate: adenosine and neuropeptide Y receptors. Neurochem. Res. 28, 1501–1515. Murashima, Y.L., Suzuki, J., Yoshii, M., 2005. Developmental program of epileptogenesis in the brain of EL mice. Epilepsia 46 (Suppl. 5), 10–16. Qian, Y., Huang, M., Guan, T., Chen, L., Cao, L., Han, X.J., Huang, L., Tang, X., Li, Y., Sun, H., 2015. Maslinic acid promotes synaptogenesis and axon growth via activation in cerebral ischemia model. Eur. J. Pharmacol. 764, 298–305. Ravizza, T., Balosso, S., Vezzani, A., 2011. Inflammation and prevention of epileptogenesis. Neurosci. Lett. 497, 223–230.

34