002C-3 protects the brain against ischemia-reperfusion injury by inhibiting autophagy and stimulating CaMKK/CaMKIV/HDAC4 pathways in mice 
Jingliang Zhang1#, Tao Hu1#, Xiaoyan Liu1, Yuanjun Zhu1, Xiaoling Chen1, Ye Liu2, Yinye Wang1* 
1. Department of Molecular and Cellular Pharmacology, School of Pharmaceutical Sciences, Peking University Health Science Center, Beijing 100191, China
2. Beijing Honghui New Medical Technology Co.Ltd., Beijing Daxing Biological Medicine Industry Base, Beijing 102600,China 

Abstract: This study was designed to investigate the effect of 002C-3, a derivative of magnolol, on transient cerebral middle occlusion (tMCAO) in a mice model and to identify the underlying mechanisms. 002C-3 (100 and 150 μg/kg, i.v. after ending occlusion) significantly reduced neurological deficit scores, infarct volumes, and brain water contents after 1.5 h MCAO and 24 h reperfusion. 002C-3 (75–150 µg/kg) decreased the exudation of Evans blue from brain capillaries. 002C-3 (100 μg/kg) significantly inhibited the activity of MMP-9 and MMP-2 in the injured hemisphere. 002C-3 decreased the expression of autophagy-associated proteins, Beclin-1 and LC3B-II, and increased the level of p62 in injured hemisphere. 002C-3 (100 μg/kg)significantly increased the expression of p-CaMKIV and p-HDAC4 in injured hemisphere. In conclusion, 002C-3 shows a neuroprotective effect on tMCAO injury in mice, and its mechanisms may be associated with alleviation of blood-brain barrier damage caused by the activation of MMPs, inhibition of autophagy, and stimulation of calcium signals related to cell survival. These findings suggest that 002C-3 is a neuroprotective agent that acts on multiple pathways.         
Keywords: 002C-3, Cerebral ischemia-reperfusion, Microvascular permeability, Autophagy, CaMKK/CaMKIV/HDAC4 pathway   
CLC number: R962                Document code: A                 Article ID: 10031057(2016)859807
1. Introduction
Neuro-protection is a vital part of ischemic stroke therapy. To date, more than 100 clinical trials of promising neuroprotective compounds have ended in failure[1]. Despite these tremendous efforts, there remains a lack of effective neuroprotective agents clinically. One of potential reasons for clinical trial failure likely is that the mechanisms of ischemic injury are so complicated that a single agent acting on a single target cannot block all of the injury pathways. The protection of neurons alone is not sufficient because ischemia and reperfusion also cause vascular damage, which induces inflammation and edema[2]. Thus, the concept of protecting the neurovascular unit (NVU) has been put forward and recognized because it expands the protection effect to the interaction of neuron-astrocyte-capillary pathophysiology[3]. This concept provides a new approach for the research and development of neuroprotectants. The concurrent effects aimed at multiple pathways in the NVU likely provide more benefit compared with those aimed at a single target in neuronal protection.
Magnolol is a compound isolated from the bark of Magnolia officinalis and has been demonstrated to reduce cerebral ischemia and reperfusion injury in rats[4].The action mechanisms of magnolol include antioxidant activity[5], anti-inflammation[6] , inhibition of cytotoxic NO production[7], improvement of inhibitory amino acids levesl[8], inhibition of platelet aggregation and thrombosis[9], and anti-apoptosis[10]. Obviously, magnololis a compound that confers its protective effects throughmultiple pathways. Despite the benefits of this compound, the small space between effective dose and toxic dose limits its application. 002C-3, a new derivative of magnolol, has a greatly improved water-solubility and much better safety profiles than magnolol. It has shown potent effects on transient middle cerebral occlusion (tMCAO) in rats. The mechanism of action of 002C-3 is associated with inhibition of both apoptosis and autophagy[11]. Because 002C-3 is magnolol derivative, it is likely to act on multi-pathway in ischemic stroke pathophysiology. Mice are most used mammal in biomedical research because their genetic similarity with human is about 99%. Therefore, we sought to investigate whether 002C-3 would be effective in treating MCAO and cerebral microvascular injury in mice and its mechanisms of action.
2. Material and methods
2.1. Chemicals and agents
002C-3 (purity: 93.8%) was provided by Beijing Honghui New Medical Technology Co. Ltd. (Beijing, China). It was dissolved with normal saline (NS) before application. Edaravone injection was obtained from Simcere (Nanjing, China).
2.2. Animals
Male Kunming mice (30–35 g) were obtained from the Department of Laboratory Animal Science, Peking University Health Science Center. The protocol was approved by the Committee on the Ethics of Animal Experiments of the Peking University Health Science Center (Permit number: LA2013-69).
2.3. tMCAO model in mice
After anaesthetized with chloral hydrate (500 mg/kg, i.p.), the mouse tMCAO model was prepared according to the previous method[12]. In brief, a nylon monofilamentwith blunted tip and coated with 1% of silicone oil was introduced into the internal carotid artery from the external carotid artery to occlude MCA. Relative regional cerebral blood flow (rCBF) monitored with Laser Doppler Measurement (PeriFlux System 5000) dropped to less than 80% of the baseline upon occlusion[13]. Mortality rate in operation was less than 10%. Mice were randomly assigned to the sham group, vehicle group, and three doses of 002C-3-treated groups, and edaravone-treated group. Vehicle or 002C-3 was administered immediatelyafter ending occlusion (i.v.), and edaravone was immediately administered twice (i.v.) after the occlusionof MCA and ending occlusion, respectively. Body temperature was maintained at (37±0.5) ºC on a heating operating table during surgery.
Neurologic deficit scores were evaluated 24 h after reperfusion. An expanded seven-point scale is used on the basis of the evaluation criterion reported previously[14]. The observer was blind to animal treatment.
Mice were transcardially perfused with NS to remove blood in the tissue under anesthesia 24 h after ending occlusion. The brains were quickly removed, sectioned coronally into 5 slices at 2-mm intervals, stained by immersion in 1% TTC according to previous reports[12], and fixed in 4% paraformaldehyde. The sections were photographed, and the normal and infarcted areas were analyzed using Adobe Photoshop CC. The percentage of the infarct volume was calculated after correcting for edema.
The brain water content was determined according to a previously described method[15]. In brief, brains were immediately weighed to obtain wet weight, then dried in an oven at 110 ºC for 8 h and weighed again to obtain the dry weight. Brain water content was calculated by taking the difference between the weights.
2.4. Measurement of Evans blue amount
tMCAO mice received Evans blue (80 mg/kg, i.v.) 3 h before sacrifice, and then were transcardially perfusedwith NS to remove the intravascular dye under anesthesia. Brains were collected, the left and right hemisphere were weighed, and the dye was extracted in 1.0 mL formamide at 45 ºC for 72 h[16]. The supernatants were analyzed by measuring the absorbance at 620 nm.
2.5. Gelatin zymography and Western blot
Gelatin zymography was performed as the previously described[17]. Proteins were loaded on a 10% SDS-PAGE containing 0.1% gelatin (G1890, Sigma) for electrophoresis and stained with 0.1% of Coomassie brilliant blue R-250. Signals were scanned in a ChemiDoc XRS System and quantitatively analyzed by densitometry with Quantity One 4.6.9 software (Bio-Rad).
The levels of autophagy-related proteins Beclin-1, LC3B-II, and p62, calmodulin dependent protein kinase IV (CaMKIV), and Histone deacetylase 4 (HDAC4) phosphorylation proteins were determined by Western blotting. Proteins were extracted from injuredcerebral hemispheres in cold RIPA Lysis Buffer (MP015, Macgene) or with a nuclear and cytoplasmic Extraction Kit (CW0199B, CWBIO). Western blot was performed as described previously[18].
2.6. Statistical analysis
All data are expressed as mean±SEM. Statistical evaluation was performed using one-way analysis of variance. Significant differences between groups were assessed by Tukey’s test and *P<0.05 was considered statistically significant.
3. Results
3.1. 002C-3 protected brain against tMCAO injury
In the tMCAO model group, the infarct volume was 58.5%±2.7%. 75 μg/kg, 100 μg/kg, and 150 μg/kg of 002C-3 reduced the infarct volume to 31.8%±6.3%, 33.6%±6.7% and 22.6%±5.1%, respectively (Fig. 1A, white region, Fig. 1B). The reduction of infarct volume for 75 μg/kg, 100 μg/kg, and 150 μg/kg of 002C-3 was about 45.6%, 42.6% and 61.4%, respectively. In addition, 100 μg/kg and 150 μg/kg of 002C-3 significantly reduced the neurological deficit scores (Fig. 1C),which was consistent with the effect on the infarct volume. 002C-3 (75–150 μg/kg) significantly decreased brain water content (Fig. 1D). 

Figure 1.
The protective effect of 002C-3 on cerebral I-R in mice. Representatives of TTC-strained brain slices 24 h after reperfusion (A); Infarct volume (B); Neurological deficit scores (C); Brain water content (D). Data were expressed as mean±SEM, n = 11 for sham group; n = 14–15 for other groups. ^^P<0.01 versus sham group; *P<0.05, **P<0.01 versus vehicle (NS) group; #P<0.05, ##P<0.01 versus edaravone (Eda) group.  
3.2. 002C-3 decreased exudation amount of Evans blue
In the tMCAO group, the dye amounts in the injured and normal cerebral hemispheres were increased compared with that in the sham group. The amount in injured hemisphere was much higher than that in the normal hemisphere. 002C-3 significantly reduced the dye amount compared with the vehicle group both in the normal and in the injured hemisphere (Fig. 2).

2. Effect of 002C-3 on Evans blue exudation in tMCAO mice. Data were expressed as mean±SEM, n = 10. ^P<0.05, ^^P<0.01versus sham group; *P<0.05, **P<0.01 versus corresponding hemisphere in vehicle group.
3.3. 002C-3 decreased the activity of MMP-2 and MMP-9 in injured hemisphere
MMPs hydrolyze tight junction proteins of the BBB and are responsible for BBB disruption after ischemia injury. The results showed that tMCAO increased the activities of MMP-2 and MMP-9. In contrast, 002C-3 reduced their activities in injured hemisphere (Fig. 3A and 3B) by 100% and 89.8%, respectively. 

Figure 3. The influence of 002C-3 on MMP-9 and MMP-2 activities in in injured hemisphere. Gelatin zymography analysis of the activity (A). Statistical analysis of activity of MMP-9 and MMP-2 (B). Data were expressed as mean±SEM, n = 5. ^^P<0.01 versus sham group; **P<0.01 versus vehicle group.
3.4. 002C-3 inhibited autophagy in tMCAO injured hemisphere
The increase of Beclin-1 and LC3B-II and the decrease of p62 signify the activation of autophagy. The results showed that the relative expression of both Beclin-1 and LC3B-II were significantly increased in the injured hemisphere, and 002C-3 significantly reduced the expression of the two proteins by 79.4% and 95.1%, respectively (Fig. 4B and 4C). Additionally, the relative expression of p62 protein was decreased in I-R hemisphere, and 002C-3 recovered the level of p62 by 74.6% (Fig. 4D). These results imply that 002C-3 may also confer neuroprotective effects via suppressing autophagy activation by I-R injury. 

4. The influence of 002C-3 on autophagy related proteins in injured hemisphere. Western blot analysis for the expression of proteins in cerebral hemisphere 24 h after reperfusion (A). Statistical analysis of Beclin-1 (B), LC3B-II (C) and p62 (D) expression level. Data were expressed as mean±SEM, n = 10. ^^P<0.01 versus sham group; *P<0.05, **P<0.01 versus vehicle group.
3.5. 002C-3 upregulated CaMKK/CaMKIV/HDAC4 pathway
The nuclear translocation of HDAC4 is detrimental for neuronal survival. CaMKIV inhibits HDAC4 translocation in neurons under stress via phosphorylation of HDAC4[19]. We observed that p-CaMKIV and p-HDAC4levels were significantly reduced in vehicle-treated mice (Fig. 5B and 5C), and 002C-3 completely recovered p-CaMKIV levels and reversed p-HDAC4 levels by 69.1%.

5. The influence of 002C-3 on CaMKK/CaMKIV/HDAC4 pathway in injured hemisphere. Western blot analysis for the expression of proteins associated with CaMKK/CaMKIV/HDAC4 pathway (A). Statistical analysis of p-CaMKIV (B) and p-HDAC4 expression (C). Data were expressed as mean±SEM, n = 5. ^^P<0.01 versus sham group; *P<0.05, **P<0.01 versus vehicle group. 
4. Discussion
This study demonstrates 002C-3 is a potent protective agent against ischemia and reperfusion in mice. This effect may be associated with decreasing the permeabilityof cerebral microvessels mediated by MMPs, inhibiting autophagy, and upregulating survival related calcium signals.
Reperfusion with rtPA injures cerebral vessels and thus aggravates neuronal injury. rtPA induces cerebral hemorrhage, which is linked to the excessive activation of MMPs after reperfusion. The activation of MMP-9 causes BBB late-onset permeability increase (24–48 h after reperfusion), brain edema, and hemorrhagic transformation[20]. In this study, 002C-3 reduced brain edema (Fig. 1D) and Evans blue exudation (Fig. 2) caused by tMCA. We believe that these effects were related to the activities of MMP-9 and MMP-2. As expected, 002C-3 significantly lowered their activities in injured hemisphere (Fig. 3), suggesting that 002C-3 protects BBB through the inhibition of activities of MMPs. The band of MMP-2 was thin and indistinct (Fig. 3B), which may be due to the differential expression of MMP-2 and MMP-9 at varying times after cerebral ischemia[20].
Autophagy is involved in the pathological process of cerebral ischemia and affects the neuron survival and death[21]. But the conclusions of the influence of autophagy on cerebral ischemia are controversial[2123]. It may play different roles at different stages after cerebral ischemia[24] or in different degrees of injury[25].
In this study 002C-3 significantly suppressed the levels of Beclin-1 and LC3B-II, and elevated the levels of p62 (Fig. 4) in injured hemisphere of mice, indicatedthat 002C-3 protects the brain against injury by inhibiting autophagy, which is similar to the results from studies in rats[11]. Observations of autophagy in our tMCAO model differ from previous reports[25] that report expedient activation of autophagy in a tMCAO model. The disaccord between the two cases is likely due to different degrees of autophagy.
Increasing evidences indicates that stimulation of calcium signals can play an important role in protecting against I-R injury through the activation of endogenous neural protection mechanism[18]. CaMKK is activated by calcium/calmodulin signaling, and Akt phosphorylationmediated by CaMKK can inhibit the apoptosis of neuroblastoma and protect neurons[26]. CaMKK phosphorylates its downstream kinase CaMKIV[27]. p-CaMKIV mainly exists in the nucleus[28], and upon stressful stimuli (such as the excitoxicity of glutamic acid), HDAC4 translocates from the cytoplasm to the nucleus. p-CaMKIV can transport HDAC4 from nucleus to cytoplasm by catalyzing its phosphorylation, thereby restore the transcription of key endogenous survival genes such as cAMP response element binding protein (CREB), which would have been inhibited by HDAC4[19]. In this study 002C-3 significantly increased the levels of p-CaMKIV and p-HDAC4 (Fig. 5), suggesting that the protective effect of 002C-3 is maybe mediated via the up-regulation of CaMKK/CaMKIV/HDAC4 pathway.
In summary, 002C-3 has protective effect on tMCAOin mice. This effect may be associated with the inhibitionof MMPs and autophagy, and stimulation of cell survival related calcium signals.
This study was supported by National Natural ScienceFoundation of China (Grant No. 81302763, 81573333) and Beijing Natural Science Foundation (Grant No. 7144218).
[1] Turner, R.C.; Dodson, S.C.; Rosen, C.L.; Huber, J.D. J. Neurosurg.2013, 118, 1072–1085.
[2] Fisher, M.; Feuerstein, G.; Howells, D.W.; Hurn, P.D.; Kent, T.A.; Savitz, S.I.; Lo, E.H. Stroke. 2009, 40, 2244–2250.
[3] Lo, E.H.; Dalkara, T.; Moskowitz, M.A. Nat. Rev. Neurosci.2003, 4, 399–415.
[4] Lee, W.T.; Lin, M.H.; Lee, E.J.;Hung, Y.C.; Tai, S.H.; Chen, H.Y.; Chen, T.Y.; Wu, T.S. PloS One.2012, 7, e39952.
[5] Chen, Y.H.; Huang, P.H.; Lin, F.Y.; Chen, W.C.; Chen, Y.L.;Yin, W.H.; Man, K.M.; Liu, P.L. Eur. J. Integ. Med. 2011, 3, e317–e324.
[6] Liang, C.J.; Lee, C.W.; Sung, H.C.; Chen, Y.H.; Wang, S.H.; Wu, P.J.; Chiang, Y.C.; Tsai, J.S.; Wu, C.C.; Li, C.Y.; Chen, Y.L. Am. J. Chin. Med.2014, 42, 619–637.
[7] Huang, X.; Wang, C.; Chen, K.; Xiang, T.; Han, Z. Chin. J. Neuroimmunol. Neurol.2007, 14, 118–119.
[8] Lin, Y.R.; Chen, H.H.; Ko, C.H.; Chan, M.H. Eur. J. Pharmacol.2006, 537, 64–69.
[9] Pyo, M.K.; Lee, Y.; Yun-Choi, H.S. Arch. Pharmacol. Res.2002, 25, 325–328.
[10] Yan, Y.; Lin, Q.; Xu, Y.; Huang, X. Chin. J. Exp. Tradit. Med. Formul. 2011, 17, 223–225.
[11] Li, H.; Liu, X.; Zhu, Y.; Liu, Y.; Wang, Y. Neurosci. Lett.2015, 588, 178–183.
[12] Chiang, T.; Messing, R.O.; Chou, W.H. J. Visual. Exp.2011. JoVE (48).
[13]Liu, F.; McCullough, L.D. Methods Mol. Biol.2014, 1135, 81–93.
[14] Rousselet, E.; Kriz, J.; Seidah, N.G. J. Visual. Exp.2012, JoVE(69). 
[15] Chao, X.; Zhou, J.; Chen, T.; Liu, W.; Dong, W.; Qu, Y.; Jiang, X.; Ji, X.; Zhen, H.; Fei, Z. Brain Res. 2010, 1363, 206–211. 
[16] Cevik, N.G.; Orhan, N.; Yilmaz, C.U.; Arican, N.; Ahishali, B.; Kucuk, M.; Kaya, M.; Toklu, A.S. Brain Res. 2013, 1531, 113–121.
[17] Toth, M.; Sohail, A.; Fridman, R. In Metastasis Research Protocols, 2012, Springer. 121–135.
[18] McCullough, L.D.; Tarabishy, S.; Liu, L.; Benashski, S.;Xu, Y.; Ribar, T.; Means, A.; Li, J. Stroke. 2013. 44, 2559–2566.
[19] Bolger, T.A.; Yao, T.P. J. Neurosci: the official journal of the Society for Neuroscience. 2005, 25, 9544–9553.
[20] Liu, Y.; Liu, K.; Liu, J.; Li, J. Chin. J. Gerontol.2014, 1, 135.
[21] Xu, F.; Gu, J.; Qin, Z. Neurosci. Bullet.2012, 28, 658–666.
[22] Carloni, S.; Buonocore, G.; Balduini, W. Neurobiol Dis. 2008, 32, 329–339.
[23] Zheng, Y.; Liu, J.; Li, X.; Xu, L.; Xu, Y. Acta Pharmacol. Sin. 2009, 30, 919–927. 
[24] Wang, J.; Xia, Q.; Chu, K.; Pan, J.; Sun, L.; Zeng, B.; Zhu, Y.; Wang, Q.; Wang, K.; Luo, B. J. Neuropathol. Exp. Neurol.2011, 70, 314–322.
[25] Zhang, X.; Yan, H.; Yuan, Y.; Gao, J.; Shen, Z.; Cheng, Y.; Shen, Y.; Wang, R.R.; Wang, X.; Hu, W.W.; Wang, G.; Chen, Z. Autophagy.2013, 9,1321–1333.
[26] Yano, S.; Tokumitsu, H.; Soderling, T.R. Nature.1998, 396, 584–587.
[27] Means, A.R. Mol. Endocrinol.2008, 22, 2759–2765.
[28] Wayman, G.A.; Lee, Y.S.; Tokumitsu, H.; Silva, A.J.; Soderling, T.R. Neuron.2008, 59, 914–931.

张精亮1#, 胡涛1#, 刘晓岩1, 朱元军1, 陈晓玲1, 刘晔2, 王银叶1*
1. 北京大学医学部 药学院 分子与细胞药理学系,北京 100191
2. 北京红惠新药科技有限公司, 北京102600
摘要: 研究厚朴酚衍生物002C-3对小鼠中脑动脉缺血(MCAO)再灌注损伤的作用及可能的机制。与模型对照组相比, 在缺1.5 h后单次静注002C-3 (100150 μg/kg)可明显减小神经缺陷评分, 缩小脑梗死体积和降低脑水含量。002C-3 (75–150 µg/kg)可明显减少脑毛细血管伊文氏蓝的渗出量。与模型对照组相比, 002C-3 (100 μg/kg)可明显抑制缺血半脑组织中的基质金属蛋白酶MMP-9MMP-2的活性; 减少自噬相关蛋白Beclin-1LC3B-II的表达, 增加p-62蛋白的表达; 002C-3还可增加细胞存活相关的钙信号蛋白p-CaMKIVp-HDAC4的表达。这些结果表明: 002C-3在小鼠脑缺血再灌注损伤模型上有明显的神经保护作用, 其作用机制可能与抑制MMPs活性, 保护血脑屏障, 抑制缺血诱导的细胞自噬以及激活细胞存活相关的钙信号通路有关。
关键词: 002C-3; 脑缺血再灌注; 脑微血管通透性; 自噬; CaMKK/CaMKIV/HDAC4通路

Received: 2016-02-27, Revised:2016-04-15, Accepted: 2016-05-10.
Foundation items: National Natural Science Foundation of China (Grant No. 81302763, 81573333) and Beijing Natural Science Foundation (Grant No. 7144218).
#Jingliang Zhang and Tao Hu contributed equally to this work.
*Corresponding author. Tel.: +86-010-82802652/+86-010-62015584, E-mail: wangyinye@bjmu.edu.cn