【作者】 王天俊；

【导师】 吕佩源；

【作者基本信息】 河北医科大学 ， 神经病学， 2009， 博士

【摘要】 目的:血管性痴呆(vascular dementia,VD)是由各种脑血管疾病导致的获得性、持续性智能障碍综合征,以学习、记忆功能损害为主要症状,可伴有语言、运动、视空间及人格障碍。随着社会人口日益老龄化和脑血管病后生存率的提高,由脑血管病造成的精神及智能损害日益突出,给社会和家庭带来沉重负担。因此探讨其可能的发病机制,制定合理的治疗方案已成为当前医学界的紧迫任务之一。近年研究显示:谷氨酸的兴奋毒性参与了血管性痴呆的病理机制。因此,减轻兴奋性氨基酸(excitatory amino acids, EAA)的毒性作用,对于减轻缺血性脑损伤及其继发的学习与记忆障碍,具有极为重要的意义。谷氨酸转运体(excitator aminoacid transporters, EAATs)作为清除兴奋性氨基酸的主要物质,在全脑缺血的发生机制和治疗中发挥了重要作用。脑缺血时,细胞外液谷氨酸浓度持续大量升高,过度激活谷氨酸受体,导致Ca2+大量内流,细胞内Ca2+超载可激活中性蛋白酶Calpain,使其对细胞酶产生过度降解,引起细胞结构与功能异常及细胞死亡。细胞周期依赖性蛋白激酶5(Cyclin dependent kinase 5, Cdk5)参与突触小泡循环、离子通道和细胞内信号转导通路的调节,在突触可塑性、学习与记忆过程中发挥关键作用。Cdk5的激动亚基p35也是Calpain的作用底物之一,病理条件下,Calpain可使p35蛋白降解成p25蛋白,后者使Cdk5激酶活性失调,导致包括凋亡、兴奋毒性等多种形式的细胞死亡。由活性氧(reactive oxygen species, ROS)引起的氧化损伤是导致认知功能障碍的重要原因。超氧化物歧化酶(superoxide dismutase, SOD)和丙二醛(malondialdehyde, MDA)是评价氧化和抗氧化程度的重要指标。神经元内的氧化损伤和其他危害可引起Calpain介导的p35向p25的裂解。另外, Cdk5对自由基的产生具有调节作用。已有证据表明,在缺血、缺氧等条件下,自由基引起的氧化损伤参与了线粒体破坏、细胞坏死和凋亡等过程。目前有关EAATs、Calpain-Cdk5和活性氧通路在VD发生中的作用尚鲜有报道。盐酸多奈哌齐是胆碱酯酶抑制剂,可选择性抑制中枢神经系统中乙酰胆碱(Acetylcholine, Ach)的降解,增加神经细胞突触间隙Ach的浓度,提高记忆脑区的神经传导功能,从而改善大脑学习和记忆能力。有资料表明,多奈哌齐可增加VD患者受损脑区局部脑血流量,激活细胞外信号调节激酶(Extracellular signal-regulated kinases, ERK)的表达,降低海马神经元N-甲基-D-天门冬氨酸(N - methyl - D - aspartate receptor, NMDA)受体亚单位R1(NMDAR1, NR1)的表达,通过影响p38丝裂原活化蛋白激酶(mitogen activated protein kinase, MAPK)的mRNA表达而改善学习和记忆功能,提示该药除发挥胆碱酯酶抑制剂的作用外,还通过多通道、多靶点参与学习和记忆功能的调节。但该药对EAATs和Calpain-Cdk5/p25通路以及ROS在VD中的作用尚未见报道。基于此,本文通过探讨EAATs和钙信号转导机制,观察该药治疗VD的效果并探讨其作用机制。方法:1采用双侧颈总动脉反复三次缺血-再灌注方法,制备小鼠VD模型,并设立假手术组和药物组(盐酸多奈哌齐);通过跳台试验和水迷宫试验进行学习和记忆成绩测试,HE染色观察海马组织病理变化。2采用免疫组化方法测定海马CA1区神经细胞膜上EAAT1和EAAT2的表达。3免疫组化方法测定海马CA1区神经元内Calpain I的表达。4反转录-聚合酶链反应(reverse transcription PCR, RT-PCR)和Western-blot方法测定小鼠海马组织Cdk5的mRNA表达和蛋白表达,Western-blot方法测定各组小鼠海马组织p25表达。5紫外分光光度法测定海马神经元SOD活性和MDA含量。结果:1 VD小鼠模型的行为学评价和海马病理学特征1.1跳台试验学习成绩模型组术后第29d,43d和57d的反应时间分别是(120.46±30.25),(133.16±29.77)和(121.05±31.43),较假手术组的(29.38±3.76),(28.33±4.28)和(28.72±2.67)明显延长(P<0.05);多奈哌齐治疗组的反应时间分别是(48.12±11.23),(56.12±12.44)和(42.17±11.26),较模型组显著缩短(P<0.05)。模型组术后第29d,43d和57d的错误次数分别是(3.88±1.29),(4.38±1.13)和(3.14±1.29),假手术组分别是(1.96±0.46),(2.16±0.43)和(1.58±0.63),模型组错误次数较假手术组显著增多(P<0.05);多奈哌齐治疗组的错误次数分别是(2.27±0.64),(2.86±0.87)和(1.96±0.78),较模型组显著减少(P<0.05),与假手术组无明显差异(P>0.05)。1.2跳台试验记忆成绩术后第30d,44d和58d模型组的潜伏时间分别是(79.97±26.38),(80.78±26.12)和(79.87±20.17),较假手术组的(195.78±36.55),(186.78±36.21)和(178.26±35.23)均明显缩短(P<0.05),多奈哌齐治疗组的潜伏时间分别是(156.13±31.25),(161.22±33.15)和(129.58±22.43),较模型组显著延长(P<0.05);模型组术后第30d、44d和58d错误次数分别是(4.14±1.25),(4.04±0.98)和(3.98±1.14),假手术组分别是(1.93±1.01),(1.91±0.56)和(1.76±1.06),模型组错误次数较假手术组显著增多(P<0.05),多奈哌齐治疗组的错误次数分别是(2.56±1.08),(2.74±0.87)和(1.89±0.86),较模型组显著减少(P<0.05);药物组与假手术组之间无明显差别(P>0.05)。1.3水迷宫试验学习成绩术后第29d、43d和57d模型组小鼠游完全程时间分别为(124.33±53.22)、(116.21±50.31)和(119.31±55.41) ,较假手术组的(66.32±51.61)、(61.63±38.86)和(65.73±51.96)均明显延长(P<0.05),多奈哌齐治疗组小鼠游完全程的时间分别是(80.56±53.28)、(76.43±43.86)和(77.49±48.86),较模型组显著缩短(P<0.05);模型组术后第29d、43d和57d的错误次数分别是(11.68±1.36)、(15.76±1.68)和(13.78±1.77),假手术组分别是(5.76±1.55)、(5.65±1.28)和(5.85±1.39),模型组错误次数较假手术组显著增多(P<0.05),多奈哌齐治疗组的错误次数分别是(7.48±1.67)、(7.38±1.34)和(7.38±1.49),较模型组显著减少(P<0.05);提示药物组的学习成绩得到改善;并且与假手术组相比无明显差别(P>0.05)。1.4水迷宫试验记忆成绩术后第30d、44d和58d模型组小鼠游完全程时间分别为(111.58±48.32)、(109.77±51.28)和(111.32±58.38) ,较假手术组的(51.23±45.75)、(49.34±36.36)和(49.56±40.33)均明显延长(P<0.05),多奈哌齐治疗组小鼠游完全程的时间分别是(76.29±41.27)、(66.27±40.86)和(57.27±40.11),分别较模型组显著缩短(P<0.05);模型组术后第30d、44d和58d的错误次数分别是(11.69±2.34)、(14.65±1.39)和(14.95±2.19),假手术组分别是(5.31±1.22)、(5.35±1.04)和(5.68±1.13),模型组错误次数较假手术组显著增多(P<0.05),多奈哌齐治疗组的错误次数分别是(6.33±1.46)、(7.01±1.26)和(7.07±1.35),较模型组显著减少(P<0.05);提示药物组的记忆成绩得到改善;并且与假手术组相比无明显差别(P>0.05)。1.5 HE染色显示海马CA1区病理变化(1)假手术组:HE染色示各组小鼠海马CA1区轮廓清楚、锥体细胞为3～5层,数量丰富,呈圆形或椭圆形,排列规则致密,细胞核圆而大,核仁清晰,染色质丰富,神经纤维密集。术后4、6、8周各组单位面积正常神经元计数分别为(101.58±15.32)、(108.24±11.33)和(101.24±12.26)。(2)模型组:各组小鼠海马CA1区的锥体细胞层次减少,排列松散,细胞数目减少,细胞核体积变小,结构不清,呈核固缩表现,神经纤维排列紊乱。术后4周、6周和8周时各组单位面积内正常神经元计数分别为(65.33±19.28)、(21.23±3.12)和( 50.52±11.28),较假手术组均显著降低(P<0.05),其中以术后6周损害最严重,绝大部分神经元死亡,仅有少量细胞存活,正常神经元计数较模型组术后4、8周时也显著减少(P<0.05)。(3)治疗组:术后4周和8周时,小鼠海马CA1区锥体细胞排列较整齐,层次较清楚;无明显核固缩现象。正常神经元计数分别是(88.22±19.88)和(83.36±12.21),较模型组显著增高(P<0.05),与假手术组相比无明显差别(P>0.05)。术后6周小鼠海马组织CA1区锥体细胞排列较整齐,层次较清楚;仍有部分核固缩现象。正常神经元计数为(46.38±5.17),较模型组显著增高(P<0.05),但仍显著低于假手术组(P<0.05)。上述结果表明,海马是影响学习和记忆功能的重要结构,脑缺血-再灌注可引起VD的发生,伴海马神经元严重受损,以术后6周最明显。盐酸多奈哌齐可减轻VD小鼠的病理改变和学习与记忆能力。2 VD小鼠海马CA1区EAATs表达及盐酸多奈哌齐的影响取出小鼠脑组织,经4 %多聚甲醛灌注固定后,石蜡包埋,冠状切片,进行免疫组化染色,观察EAAT1和EAAT2的平均光密度值(mean optical density,OD)表达。结果显示:假手术组在术后4周、6周和8周时小鼠海马CA1区细胞膜上EAAT1表达的OD值分别是(0.030±0.003)、(0.031±0.002)和(0.030±0.003);模型组细胞膜上EAAT1表达的OD值分别是(0.099±0.009)、(0.122±0.009)和(0.112±0.010),较假手术组均显著升高(P<0.05),尤以术后6周时增高明显;盐酸多奈哌齐治疗组在上述三个时间点EAAT1表达的OD值分别是(0.047±0.004)、(0.046±0.003)和(0.044±0.004),显著低于模型组(P<0.05),与假手术组无明显差异(P>0.05);术后4、6、8周假手术组海马CA1区细胞膜上EAAT2表达的平均OD值分别是(0.031±0.003)、(0.030±0.002)和(0.030±0.003);模型组的平均OD值为(0.098±0.008)、(0.113±0.008)和(0.101±0.009),较假手术组均显著增高(P<0.05),药物组EAAT2表达的平均OD值分别是(0.039±0.003)、(0.047±0.005)和(0.042±0.004),均较模型组显著降低(P<0.05),与假手术组无显著差异(P>0.05)。提示EAAT1和EAAT2参与了VD的发生,多奈哌齐改善学习和记忆的能力可能与其降低EAAT1和EAAT2表达有关。3 VD小鼠海马CA1区Calpain I的表达及盐酸多奈哌齐的影响小鼠脑组织经4 %多聚甲醛灌注固定后,石蜡包埋,冠状切片,行免疫组化染色,观察海马CA1区Calpain I的OD值。结果显示:假手术组术后4周、6周和8周时小鼠海马CA1区细胞层次清晰,胞浆内仅有微量Calpain I的表达(分别是0.030±0.003、0.031±0.003和0.029±0.003);模型组术后4周、6周和8周时海马CA1区神经细胞结构松散,神经元胞浆内Calpain I的平均OD值分别是(0.099±0.009)、(0.121±0.010)和(0.115±0.010)较假手术组均显著增高(P<0.05),尤以术后6周时增高明显;盐酸多奈哌齐治疗组在上述三个时间点Calpain I的表达分别是(0.040±0.004)、(0.044±0.004)和(0.042±0.003),均较模型组显著降低(P<0.05),与假手术组接近(P>0.05)。说明脑缺血-再灌注可引起神经元内Ca2+超载,激活Calpain I,出现持续高表达,引起神经元损害和VD的发生。盐酸多奈哌齐可能通过降低Calpain I的表达而发挥神经保护作用。4 VD小鼠海马Cdk5的蛋白表达和mRNA表达及盐酸多奈哌齐的影响4.1 Western-blot显示海马组织Cdk5的蛋白表达脑缺血-再灌注后4周、6周和8周时模型组海马组织Cdk5的蛋白表达为(0.54±0.05)、(0.73±0.07)和(0.70±0.06) ,分别较假手术组的(0.23±0.02)、(0.31±0.02)和(0.33±0.02)显著增高(P<0.05),提示Cdk5的蛋白表达在脑缺血-再灌注8周内呈持续性增高;盐酸多奈哌齐治疗组在上述三个时间点Cdk5蛋白表达分别为(0.28±0.02)、(0.33±0.03)和(0.38±0.02),均较模型组显著降低,差异有统计学意义(P<0.05)。4.2 RT-PCR检测Cdk5的mRNA表达术后4周、6周和8周时模型组海马组织Cdk5的mRNA表达分别为(0.80±0.07)、(0.91±0.08)和(0.92±0.08),分别较假手术组的(0.35±0.02)、(0.38±0.04)和(0.36±0.03)显著增高(P<0.05);盐酸多奈哌齐治疗组在上述三个时间点mRNA表达分别为(0.33±0.02)、(0.40±0.04)和(0.39±0.08),较模型组均显著降低,差异有统计学意义(P<0.05)。以上结果提示,在脑缺血-再灌注后4周时海马组织Cdk5蛋白表达和mRNA表达均显著增高,且持续至术后8周仍未恢复,此与行为学改变和神经元损害基本一致,并与Cdk5的上游激活物Calpain I的表达相一致,提示Calpain I-Cdk5的异常增多参与了脑缺血后VD小鼠海马神经元的死亡,并可能与记忆损害相关。盐酸多奈哌齐可能通过降低Calpain I的表达和Cdk5的生成,进而改善学习和记忆能力。5 VD小鼠海马组织p25蛋白表达及盐酸多奈哌齐的影响Western-blot检测海马组织p25蛋白的表达水平:术后4周、6周和8周时模型组海马组织p25蛋白的表达分别为(0.44±0.04)、(0.51±0.04)和(0.55±0.06),分别较假手术组的(0.19±0.02)、(0.24±0.02)和(0.20±0.02)显著增高(P<0.05),提示p25蛋白表达在脑缺血-再灌注8周内呈持续性增高;盐酸多奈哌齐治疗组在上述三个时间点p25蛋白表达分别为(0.26±0.02)、(0.25±0.03)和(0.21±0.02),较模型组显著降低,差异有统计学意义(P<0.05)。p25蛋白是Cdk5的异常激活因子,其显著增高提示Cdk5活性失调,是导致脑缺血-再灌注后神经元死亡的重要因素,盐酸多奈哌齐可能通过降低Calpain 1的表达而减少p35的裂解,使p25生成减少,缓解Cdk5的活性失调,从而减轻神经元死亡。6 VD小鼠海马组织SOD活性和MDA含量变化及盐酸多奈哌齐的影响6.1 SOD活性测定结果脑缺血-再灌注后4、6和8周时,模型组SOD活性分别是(7.17±1.66 U/mg*prot)、(7.07±1.43U/mg*prot)和(7.23±1.33U/mg*prot),分别高于假手术组的(3.10±0.46 U/mg*prot)、(3.22±0.52 U/mg*prot)和(3.19±0.40 U/mg*prot),差异有统计学意义(P<0.05);药物组术后4、6和8周的SOD活性分别为(4.39±0.38 U/mg*prot)、(4.88±0.47 U/mg*prot)和(4.76±0.55 U/mg*prot),显著低于模型组(P<0.05),与假手术组无明显差异(P>0.05)。6.2 MDA含量测定结果术后4、6和8周时,模型组MDA含量分别是(15.26±3.23 nmol/mg *prot)、(16.33±3.76 nmol/mg*prot)和(16.67±3.75 nmol/mg*prot),高于假手术组的(6.33±1.16nmol/mg*prot)、(6.87±1.06nmol/mg*prot)和(6.63±1.11 nmol/mg*prot),差异有统计学意义(P<0.05);药物组术后各模型组的MDA含量分别为(7.42±1.89nmol/mg*prot)、(8.32±1.68nmol/mg*prot)和(8.58±1.7 6nmol/mg*prot),显著低于模型组(P<0.05),与假手术组无差异(P>0.05)。本研究通过对脑缺血后MDA含量和SOD活性的检测发现氧化损伤参与了该病的发生。盐酸多奈哌齐可通过降低MDA含量和SOD活性而发挥神经保护作用。结论:1本实验成功建立了小鼠VD模型,经HE染色显示海马神经元受损严重,以缺血再灌注后6周时最明显,持续至术后8周仍未恢复正常。提示本模型能够基本模拟临床上VD的智能障碍,是比较理想的VD动物模型;盐酸多奈哌齐通过减轻小鼠海马病理损害而改善其学习和记忆能力,取得了相应的治疗效果。2 EAAT1和EAAT2作为对谷氨酸转运起决定作用的兴奋性氨基酸转运体,在脑缺血-再灌注后8周内小鼠海马出现持续性高表达,与神经元损伤和认知能力下降相一致,提示EAATs异常继而导致谷氨酸转运失调,及其后的Ca2+信号系统功能障碍可能是VD发生的重要因素。3 Calpain I是Ca2+信号通路的重要组成部分,在脑缺血-再灌注后8周内呈持续高表达,伴海马神经元受损,RT-PCR和Western-blot技术进一步证实海马组织Cdk5表达增加,其病理性激活因子p25的生成也相应增高,提示Calpain I-Cdk5/p25通路的表达增加参与了VD的发生。4 ROS的代表性物质SOD和MDA参与了VD的发生,提示氧化应激在脑缺血-再灌注后痴呆的发生中扮演了重要角色。5盐酸多奈哌齐可以通过减少Calpain I-Cdk5/p25通路的表达、EAAT1和EAAT2的表达和ROS的产生而改善VD的临床症状。更多还原

【Abstract】 Objective: As an acquired syndrome of intelligent impairment, vascular dementia（VD）, which is a kind of cerebral dysfunction caused by various kinds cerebral vascular disease, demonstrates mainly as learning and memory dysfunction, accompanied with the possible disorder of tongue, motion, direction and personality. With the increase of the proportion of the elderly in the population and that of the curing rate of cerebral vascular disease, morbidity of VD mounts continuously and hence brings about heavy loads to the society and family. However, both the pathogenesis of VD and its specific treatment remain unknown up to now. It is therefore significant to study the pathophysiological mechanism of VD and to find some effective treatments .Evidence emerging within the last decade has shown that glutamate excitotoxicity is presumed to be involved in pathological mechanisims in vascular dementia. It is very important to attenuate the injury of excitatory amino acid as well as ischemic damage and learning and memory. The excitator aminoacid transporters （EAATs） provide the primary mechanism for the reuptake of glutamate from the extra-cellular space, which will help to prevent over-activation and cytotoxicity. Hypoxia and/or ischemia can induce the release of glutamate. Such increased levels of glutamate, and the subsequent activation of ionotropic NMDA receptors, are primarily responsible for an increase in intracellular Ca2+, Ca2+ overload intracellular can active Calpain, which induce overactivation of enzyme and dysfunction of cells. Recent advances suggest roles for cyclin dependent kinase 5（Cdk5） in vesicle cycling, ion channel modulation and intracellular signalling. These data are correlated with others implicating Cdk5 in synaptic plasticity, learning and memory. p35, an activator of Cdk5, is also a substrate of Calpain. Under pathological condition, p35 can be cleavaged by Calpain to p25 protein, the latter causes hyperactivation of Cdk5, leading to neurodegeneration via several cell death paradigms including apoptosis, excitotoxicity and neurifila- ment hyperphosphorylation. Oxidative damage caused by reactive oxygen species（ROS） has been proposed to be critically involved in cognitive dysfun- ction. The oxidant-antioxidant status of neurons was assessed by determine- ing the levels of superoxide dismutase（SOD） activity and malondialdehyde （MDA）. Oxidative stress and other perturbation to neurons can result in the calpain-mediated cleavage of p35 to p25. On the other hand, deregulated Cdk5 may cause oxidative stress by compromising the cellular anti-oxidant defense system. Recent evidence indicates that under ischemic and hypoxic episodes, the oxidative stress induced by free radicals take part in the dysfun- ction of mitochondria and cell death. To date much less is known about the role of EAATs, Calpain-Cdk5 and ROS signal transduction cascades in VD.Donepezil hydrochloride has been used widely in the treatment of VD, but the pharmacological mechanism of treating VD is still in discussion. Donepezil can inhibit acetylcholinesterase（AchE） to degrade to acetylcholine （Ach） reversibility, increase quantity of Ach, which affect neuronal conduction and increase the ability of learning and memory . Extensive research revealed that donepezil can increase cerebral blood flow, activates the express of ERK, and decrease express of subtype of NMDAR1-NR1, increase NR2B express of hippocampus, affect expression of MAPK mRNA to adjust learning and memory. These results indicates that donepezil can increase learning and memory function by multiple pathway besides inhibiting AchE. To date the role of donepezil on EAATs, Calpain-Cdk5 pathway and oxidative stress in VD have not been studied in detail. The present study aims to examine the involvement of the EAATs, Calpain-Cdk5 pathway and oxidative stress in VD mice, in order to elucidate the potential effect of donepezil in treating VD.Method:（1） Kunming mice were subjected for continuously three repeated times ischemia-reperfusion through the ligation of the bilateral common carotid arteries, accompanied by sham group and drug group （donepezil hydrochloride）. The capability of learning and memory of mice were investigated by the stepdown test and water maze test, the pathological change of hippocampus in each group were observed by HE staining. （2） The tissue were dyed by immunohistochemistry technique to observe the expression of EAAT1 and EAAT2. （3）The expression of CalpainⅠin hippocampus were studied by immunohistochemistry. （4） The expression of Cdk5mRNA and protein were studies by RT-PCR and Western-blot technique. Western-blot was used to observe the expression of p25 protein. （5） Ultraviolet method was used to evaluate the activity of SOD and the content of MDA.Results:1 Behavioral test and pathology change of VD mice1.1 Learning ability of step-down test:The response time of model group 29d, 43d and 57d postsurgery were （120.46±30.25）,（133.16±29.77）and（121.05±31.43） respectively, longer than sham-operated group（29.38±3.76）, （28.33±4.28）and （28.72±2.67） （P<0.05）; The response time of donepezil-treated group were（48.12±11.23）, （56.12±12.44）and（42.17±11.26） respectively, shorter than model group （P <0.05）. The number of error of model group 29d、43d and 57d were （3.88±1.29）, （4.38±1.13）and（3.14±1.29） respectively, with that of sham- operated group were（1.96±0.46）、（2.16±0.43） and （1.58±0.63）. The number of error of model group were much more than sham-operated group （P<0.05）; The number of error of donepezil-treated group were（2.27±0.64）, （2.86±0.87） and （1.96±0.78）respectively, lower than model group（P <0.05）.1.2 Memory ability of step-down test:The latent time of model group 30d、44d and 58d postsurgery were（79.97±26.38）, （80.78±26.12） and （79.87±20.17） respectively, shorter than that of sham-operated group （195.78±36.55）, （186.78±36.21） and （178.26±35.23）（P<0.05）. The latent time of donepezil-treated group were （156.13±31.25）, （161.22±33.15） and （129.58±22.43） respectively, longer than model group（P <0.05）. The number of error of model group at 30d, 44d and 58d were（4.14±1.25）, （4.04±0.98） and （3.98±1.14）respectively, more than sham-operated group （1.93±1.01）, （1.91±0.56） and （1.76±1.06） （P<0.05）. The number of error of donepezil-treated group were （2.56±1.08）, （2.74±0.87） and （1.89±0.86） respectively, lower than that of model group （P <0.05）.1.3 Learning ability of Morris water maze test:The swimming time of model group 29d、43d and 57d post surgery were（124.33±53.22）, （116.21±50.31） and （119.31±55.41）respectively, longer than that of sham-operated group （66.32±51.61）, （61.63±38.86） and （65.73±51.96） （P<0.05）. The swimming time of donepezil-treated group were（80.56±53.28）, （76.43±43.86） and （77.49±48.86） respectively, shorter model group（P <0.05）. The number of error of model group 29 d、43d and 57d post surgery were（11.68±1.36）, （15.76±1.68） and （13.78±1.77） respectively, more than that of sham-operated group （5.76±1.55）, （5.65±1.28） and （5.85±1.39）（P<0.05）. The number of error of donepezil- treated group were（7.48±1.67）, （7.38±1.34） and （7.38±1.49） respectively, lower than model group（P <0.05）. There was no significant difference between sham-operated group and donepezil-treated group（P>0.05）.1.4 Memory ability of Morris water maze test:The swimming time of model group 30d、44d and 58d post surgery were（111.58±48.32）, （109.77±51.28） and （111.32±58.38） respectively, longer than sham-operated group （51.23±45.75）, ,（49.34±36.36） and （49.56±40.33） （P<0.05）. The swimming time of donepezil-treated group were （76.29±41.27）, （66.27±40.86） and （57.27±40.11） respectively, shorter than that of model group（<0.05）. The number of error of model group 30d、44d and 58d post surgery were（11.69±2.34）, （14.65±1.39） and （14.95±2.19） respectively, more than that of sham-operated group（5.31±1.22）, （5.35±1.04） and （5.68±1.13）（P<0.05）. The number of error of donepezil- treated group were（6.33±1.46）, （7.01±1.26） and （7.07±1.35） respectively, lower than model group（<0.05）. There was no significant difference between sham-operated group and donepezil-treated group （P>0.05）.1.5 HE staining of hippocampus: （1） Sham-operated group:The hippocampal profile was clear, there were 3 to 5 layers pyramidal cells, circinal and big nucleus, clear nucleolus. The many pyramidal cells ranged tightly, and dense nerve fibers were observed. The number of normal neuron 4, 6 and 8 weeks post surgery were （101.58±15.32）, （108.24±11.33） and （101.24±12.26）.（2） Model group:There were decreased layers pyramidal cells that ranged loosely. Their nucleus that had not clear structure became small, were dyed strongly, nerve fibers ranged unregularly. The number of normal neuron 4, 6 and 8 weeks post surgery were（65.33±19.28）, （21.23±3.12） and （50.52±11.28）, lower than sham-operated group（P<0.05）, among those the most damaged group were 6 weeks post surgery, there are only e few neuron alive, the number of normal cell were lower than that of 4 and 8 weeks post surgery （P<0.05）.（3） Donepezil-treated group:Above mentioned pathologic changes eased off in drug group. The pyramidal cells ranged tightly at 4 and 8 weeks post surgery. The number of normal neuron were（88.22±19.88） and （83.36±12.21）, higher than that of model group（P<0.05）. There was no significant difference between sham-operated group and donepezil-treated group（P>0.05）. At 6 weeks post surgery, the pyramidal cells ranged tightly, but still with some nucleus shrinkage. The number of normal neuron were（46.38±5.17）, higher than that of model group （P<0.05）, but still lower than sham-operated group （P<0.05）. These results suggest that hippocampus is an important region relating to learning and memory. Cerebral ischemia and reperfusion can induce the occurrence of VD, with hippocampus neurons damaged severely, among which the worst is emerging at 6 weeks post surgery. Donepezil can reduce the pathological changes and the ability of learning and memory of VD.2 Expression of EAATs in hippocampus of VD mice and the effect of donepezilThe mice’s cerebral tissue were fast taken out, fixed up via perfusion of 4% paraformaldehyde solution. Then paraffin–embedded, coronal and serial sections were taken from each brain and EAAT1 and EAAT2 were stained through immunohistochemistry. The result revealed that:（1） In model group, the expression of EAAT1 on membrane of cells 4 weeks, 6 weeks and 8 weeks post surgery were （0.099±0.009）, （0.122±0.009）and（0.112±0.010）, higher than sham-operated group（0.030±0.003）, （0.031±0.002）and（0.030±0.003）（P<0.05）, with the highest expression was at 6 weeks post surgery; the expression of EAAT1 in donepezil-treated group at the three time points were （0.047±0.004）, （0.046±0.003）and（0.044±0.004）, lower than model group （P<0.05）, there was no significant difference between sham-operated group and donepezil-treated group （P>0.05）. （2） In model group, the expression of EAAT2 at 4 weeks、6 weeks and 8 weeks post surgery were （0.098±0.008）,（0.113±0.008） and （0.101±0.009）, higher than sham-operated group （0.031±0.003）, （0.030±0.002）and（0.030±0.003） （P<0.05）, the expression of EAAT2 in donepezil-treated group at the three time points were （0.039±0.003）, （0.047±0.005） and （0.042±0.004）, lower than model group（P<0.05）, there was no significant difference between sham-operated group and donepezil-treated group（P>0.05）. These results suggest that EAAT1 and EAAT2 were involved in the occurrence of VD, donepezil may improve learning and memory ability through reducing the expression of EAAT1 and EAAT2.3 Expression of Calpain I in hippocampus of VD mice and the effect of donepezilThe mice’s cerebral tissue were fast taken out, fixed up via perfusion of 4% paraformaldehyde solution, then paraffin–embedded, coronal and serial sections were taken from each brain and Calpain I was stained through immunohistochemistry. The result revealed that: in model group, the expression of Calpain I in plasm of neurons 4 weeks, 6 weeks and 8 weeks post surgery were respectively（0.099±0.009, 0.121±0.010 and 0.115±0.010）, higher than sham-operated group（0.030±0.003, 0.031±0.003 and 0.029±0.003） （P<0.05）, with the highest expression at 6 weeks post surgery. The expression of Calpain I in donepezil-treated group at the three time points were （0.040±0.004, 0.044±0.004and0.042±0.003）, lower than model group （P<0.05）, there was no significant difference between sham-operated group and donepezil-treated group（P>0.05）. These results suggest that cerebral ischemia can induce Ca2+ overload intracellular, with sustain high expression of Calpain I, paralleled with hippocampus neuronal damage. Donepezil may protect neurons from injury by reducing the expression of Calpain I.4 The Cdk5 mRNA and protein expression in hippocamous of VD mice and the effect of donepezil4.1 Protein expression of Cdk5 in hippocamous by Western-blotThe protein expression of Cdk5 in model group 4, 6 and 8 weeks post surgery were （0.54±0.05）, （0.73±0.07） and （0.70±0.06）, higher than sham- operated group （0.23±0.02）, （0.31±0.02） and （0.33±0.02） （P<0.05）, suggesting that the protein expression of Cdk5 was increased within 8 weeks after cerebral ischemia and reperfusion. In donepezil-treated group, the protein expression of Cdk5 were （0.28±0.02）, （0.33±0.03） and （0.38±0.02）, lower than that of model group and there was significant difference（P<0.05）.4.2 Cdk5 mRNA expression in hippocamous by RT-PCRThe Cdk5 mRNA expression in model group 4, 6 and 8 weeks post surgery were （0.80±0.07）, （0.91±0.08）and（0.92±0.08）, higher than sham- operated group （0.35±0.02）, （0.38±0.04） and （0.36±0.03） （P<0.05）, suggesting that Cdk5 mRNA expression was increased within 8 weeks after cerebral ischemia and reperfusion. In donepezil-treated group, the mRNA expression of Cdk5 were （0.33±0.02）, （0.40±0.04）and（0.39±0.08）, lower than that of model group and there was significant difference （P<0.05）.The result mentioned above indicate that the mRNA and protein expression of Cdk5 were increased from 4 weeks to 8 weeks after cerebral ischemia and reperfusion, this is accordance with behavioral test and neuronal damage, and with the expression of Calpain I, which is upstream of Cdk5. It is reasonable to consider that abnormal increase expression of Calpain I-Cdk5 pathway take part in the hippocampus neuronal death of VD mice after cerebral ischemia, and maybe correlate to memory injury. Donepezil may improve learning and memory ability by reduce the expression of Calpain I and Cdk5.5 p25 expression of VD mice and the effect of donepezilThe expression of p25 in model group 4, 6 and 8 weeks post surgery were （0.44±0.04）, （0.51±0.04）and（0.55±0.06）, higher than that of sham- operated group （0.19±0.02）, （0.24±0.02） and （0.20±0.02） （P<0.05）, suggesting that the protein expression of p25 was increase within 8 weeks after cerebral ischemia and reperfusion. In donepezil-treated group, the p25 expression of were （0.26±0.02）, （0.25±0.03） and （0.21±0.02）, lower than model group and there was significant difference（P<0.05）. As an abnormal activator of Cdk5, p25 high expression indicating irregulation of Cdk5, which is one of the most important case in evoking cell death, donepezil can reduce the expression of Calpain I and the Calpain-mediated generation of p25.6 The changes of SOD activity and MDA content in VD mice and the effect of donepezil6.1 The activity of SODSOD activity in model group 4, 6 and 8 weeks post surgery were （7.17±1.66 U/mg*prot）, （7.07±1.43 U/mg*prot） and （7.23±1.33 U/mg*prot）, higher than sham-operated group （3.10±0.86 U/mg*prot, 3.22±0.92 U/mg* Prot and 3.19±0.90 U/mg*prot） （P<0.05）. In drug group, the SOD activity were （4.39±0.98 U/mg*prot）, （4.88±0.97 U/mg*prot） and （4.76±0.95 U/mg* prot）, there was no significant difference between the two groups（P<0.05）. 6.2 The content of MDAMDA content in model group were（15.26±3.23, 16.33±3.76 and 16.67±3.75 nmol/mg*prot）, higher than sham-operated group （6.33±1.16, 6.87±1.06 and 6.63±1.11 nmol/mg*prot） （P<0.05）. In drugl- treated group, MDA content were （7.42±1.89, 8.32±1.68 and 8.58±1.76 nmol/mg*prot）, lower than model group and there was significant difference （P<0.05）.Both SOD activity and MDA content were increase in model groups, suggesting that oxidative stress is a key feature of the neurodegenerative process of VD. Donepezil , as a potent free radical scavenger, might protect neurons against oxidative stress by regulating the activity and expression of antioxidant enzymes.Conclusion:（1） This study has successfuly established VD model. The pathological features in hippocampus show that pyramidal cells are damaged, especially at 6 weeks post surgery. This model might mimic the cognitive deficiency of VD in clinic. Donepezil hydrochloride might improve learning and memory ability of mice by attenuate the injury of neurons.（2） It was confirmed by immunohistochemistry that sustain increased expression of EAAT1 and EAAT2 during 8 weeks post surgery in hippocampus were related to cognitive damage and neuronal damage in VD. This indicate that glutamate excitotoxicity is a main factor in pathogenesis of VD.（3） As an importmant content of Ca²⁺ -signal cascade, Calpain I expression is sustain high during 8 weeks post ischemia-reperfusion, accompanied by neuronal injury. Cdk5 expression is also increased demonstrated by immunohistochemistry method、RT-PCR and Western-blot technique, with high expression of its activator p29 protein, indicating that more p35 is cleavaged to p25 by Calpain, which induce dysregulation of Cdk5 and neuronal death. These results suggest an importmant role of Calpain I-Cdk5/p25 pathway in the occurrence of VD.（4）Ultraviolet results demonstrate that SOD and MDA might be involved in the occurrence of VD, indicating that oxidative stress play an important role in dementia after cerebral ischemia.（5） The cognitive abilities of mouse with VD could be improved by Donepezil hydrochloride through reducing the expression of Calpain I-Cdk5/p25 pathway, EAAT1 and EAAT2 and the production of ROS, this result suggest that Donepezil hydrochloride might have neuroprotective effects.更多还原