【作者】 谷海刚；

【导师】 宋存先； 龙大宏；

【作者基本信息】 中国协和医科大学 ， 生物医学工程， 2007， 博士

【摘要】 老年性痴呆（Senile Dementia，SD），亦称阿尔茨海默病（Alzheimer Disease，AD）是一种以进行性认知障碍和记忆能力损害为主的中枢神经系统退变性疾病，伴有不同程度的运动、认知、语言和人格等多方面的异常。AD的发病率随着人口年龄的增大而增高。目前，我国正步入老龄社会，老年人口不断增多，AD患者大幅度增加。AD发病率高，病程长，又缺乏有效的治疗方法，众多的AD病人给家庭和社会带来严重的经济和精神负担。因此，寻找有效的AD治疗方法已成为当今医学界的紧迫课题。基底前脑胆碱能神经元的营养匮乏或者丢失是AD的一个重要病理变化。神经营养因子能够很好地阻止或者减少神经元的营养匮乏或者丢失；在众多的神经营养因子家族中，神经生长因子是最典型的，迄今为止研究最深入的神经营养因子。基底前脑中胆碱能神经元表达低亲和力的p75^NTF和TrkA受体，神经生长因子通过这些受体来提高基底前脑胆碱能神经元的功能，神经生长因子已被广泛地应用于老年性痴呆的实验性治疗。神经生长因子是大分子蛋白类物质，很难透过血脑屏障；生物半衰期很短。将神经生长因子成功地投递到基底前脑仍是一个很大的难题。神经生长因子的应用多数采用埋管后连续的侧脑室灌注。虽然埋管后连续的侧脑室灌注显示出一定的疗效，长期的侧脑室灌注引起很大的副作用。一种行之有效的方法是用微球的方式将神经生长因子投递到脑内。脑组织细胞移植对AD模型鼠的学习记忆能力有一定的改善作用，但临床供体问题难以解决。神经干细胞在中枢神经系统中的存在并培养成功，尤其是神经干细胞系的建立，解决了脑细胞移植供体不足的难题，成为细胞移植的理想供体。神经干细胞在体内和体外的成活、分化都离不开神经营养因子的支持和诱导；在中枢神经系统中的成活、迁移和分化离不开局部微环境中各种营养的支持。神经干细胞能够在特定的脑组织微环境的作用下分化成这一区域特定的神经细胞；其中神经营养因子对神经干细胞的成活、迁移和分化起着至关重要的作用。本文就神经生长因子缓释微球和神经干细胞联合治疗AD进行了实验研究。一方面神经干细胞分化成成熟的神经细胞补充溃变的神经元；另一方面，神经生长因子缓释微球所提供的神经生长因子可促进幸存的神经细胞的存活和功能的恢复，同时为神经干细胞的成活、迁移和分化提供营养的支持。本课题研究内容包括三个方面：1、神经生长因子缓释微球的制备及评价；2、神经干细胞移植后在AD模型鼠基底前脑内的迁移和分化；3、神经生长因子缓释微球、神经干细胞联合治疗AD模型鼠。第一章神经生长因子缓释微球的制备及评价一、材料和方法采用水／油／水（W1／O／W2）的双乳化技术来制备神经生长因子缓释微球。将5mg蛋白（rhNGF／FITC-BSA 1／2000，w／w）溶于100μl去离子水中，100mg PLGA溶于4ml二氯甲烷／丙酮（3∶1，v／v）的混合溶液中，将蛋白溶液加到PLGA溶液中，在冰浴条件下超声1min（超声功率为20W），即形成油包水（W1／O）型的乳液。然后，将蛋白和PLGA的混合溶液加入25ml 1％PVA的水溶液中，在冰浴的条件下匀速搅拌5min（1000rpm），即形成水包油包水（W1／O／W2）型的复乳液。在常压下，磁力搅拌3-4h，挥发有机溶剂，固化的微球通过离心（12000rpm，10min）收集，用去离子水清洗三遍除掉微球表面的PVA和没有包进的药物，最后冷冻干燥24h，去除微球中的水分。并研究不同条件和添加剂对蛋白包封率的影响。将神经生长因子缓释微球注射入鼠的基底前脑，观察其在体内的释放。二、结果蛋白／多聚物比率越高，载药量也越高，但是蛋白包封率则越低；当蛋白／PLGA比率由5／100（w／w）增加到15／100（w／w）时，包封率则由89.1％下降到56.5％。乳化前将水溶性添加剂加入内水相能够明显提高蛋白的包封率。内水相中加入聚乙二醇能够将包封率从89.1％提高到97.5％。神经生长因子缓释微球能够持续释放有生物学活性的神经生长因子达35天，且初时爆破释放较低。神经生长因子缓释微球能够在基底前脑持续释放神经生长因子达4周以上。第二章神经干细胞移植后在AD模型鼠基底前脑内的迁移和分化一、材料和方法新生的SD大鼠的基底前脑，机械分离后胰酶消化，制成单细胞悬液，加入含2％B27、表皮生长因子（EGF）和碱性成纤维生长因子（FGF-2）（终浓度均为10ng／ml）的DMEM／F12培养基，放置在37℃的二氧化碳培养箱内培养。待原代克隆形成后机械分离成单细胞悬液，按上述条件继续培养。以后每5～7天分离克隆传代一次，方法同前。用10％的小牛血清诱导贴壁的神经干细胞分化。用BrdU（6μg／ml）标记传代培养的细胞。用免疫细胞化学方法，行Nestin、NF、GFAP、BrdU染色，鉴定培养细胞的增殖能力和分化潜能。雄性SD大鼠（250-300g），单侧切断穹隆—海马伞（FF），以制备隔-海马通路损伤的AD模型。大鼠切断FF，将神经干细胞4ul（2.5x10⁴个／μl）移植入基底前脑，坐标为：前囟+0.6mm，外侧+0.6mm插入，腹侧-5.5mm。分别在1、2、3、4周检测神经干细胞在体内的存活、迁移和分化情况。二、结果神经干细胞能够在EGF和FGF-2的刺激下分裂增殖，1周左右形成由数十到数百个细胞组成的细胞球。传代培养的神经干细胞具有与原代培养的神经干细胞相同的生物学特性，并能够分化为神经元和神经胶质细胞，分别呈NF、GFAP阳性。克隆细胞球染色，显示Nestin和BrdU标记阳性反应。免疫荧光标记检测，贴壁的细胞为DAPI+BrdU+GFAP阳性反应或者DAPI+BrdU+NF阳性反应。移植后1周，大部分移植细胞聚集在移植的针道附近，结果显示Nestin阳性，细胞胞体较小，向四周伸出突起。移植后2周，细胞向附近的脑组织迁移，部分细胞迁移到相对较远的地方，部分细胞与周围组织整合，Nestin阳性细胞数明显减少。移植后3周、4周，看不到Nestin阳性细胞；免疫荧光双标染色显示，移植的针道附近以及基底前脑散在有许多抗BrdU+抗NF和抗BrdU+抗GFAP免疫荧光双标阳性细胞。第三章、神经生长因子缓释微球、神经干细胞联合移植治疗AD模型鼠一、材料和方法青年健康SD雄性大鼠，体重250-310克，共分为3组：①正常对照组8只；②FF损伤组即模型组8只；③神经生长因子缓释微球治疗组12只；④NSC移植治疗组12只；⑤神经生长因子缓释微球+NSC移植治疗组12只。单侧切断FF，以toFF通路损伤的AD模型。动物模型的建立后，即刻行同侧神经干细胞移植。接着进行同侧神经生长因子缓释微球注射。对内侧隔核、斜角带核的胆碱能神经元进行计数和形态学参数的测定。并用SPSS统计软件包进行统计学处理。二、结果1、神经生长因子缓释微球、神经干细胞联合移植治疗对AD模型鼠基底前脑胆碱能神经元的影响穹窿-海马伞切断4周后，损伤组损伤侧的MS和VDB的胆碱能阳性神经元都大量减少。神经生长因子缓释微球治疗组损伤侧的胆碱能神经元得到明显的保护，明显高于损伤组损伤侧的胆碱能神经元存活数（p＜0.01），与正常组相比，低于正常组（p＜0.05）。NSC移植组损伤侧的胆碱能神经元得到补充和保护，明显高于损伤组损伤侧的胆碱能神经元存活数（p＜0.01），与正常组相比，低于正常组（p＜0.05）。神经生长因子缓释微球、NSC移植联合治疗组损伤侧的胆碱能神经元得到最为明显的保护和补充，明显高于神经生长因子缓释微球组或者NSC移植组损伤侧的胆碱能神经元存活数（p＜0.01），基本达到正常组水平（P＞0.05）。2、神经生长因子缓释微球、神经干细胞联合移植治疗对AD模型鼠空间学习记忆能力的影响Y型迷宫测试5组大鼠空间学习记忆能力，结果显示损伤组大鼠空间学习能力明显下降（P＜0.05），记忆能力显著下降（P＜0.01）；神经生长因子缓释微球治疗组空间学习能力得到改善（P＜0.05），记忆能力得到显著提高（P＜0.01）；NSC移植组空间学习能力得到改善（P＜0.05），记忆能力得到显著提高（P＜0.01）；神经生长因子缓释微球、NSC移植联合治疗组空间学习能力得到显著改善（P＜0.05），记忆能力得到显著提高（P＜0.01）；与正常组相比未见差异（P＞0.05）。总之，本实验结果显示，采用水／油／水（W1／O／W2）的双乳化技术来制备神经生长因子缓释微球，蛋白／多聚物比率越高，载药量也越高，蛋白包封率则越低。乳化前将水溶性添加剂加入内水相能够明显提高蛋白的包封率。神经生长因子缓释微球能够持续释放有生物学活性的神经生长因子达35天，且初时爆破释放较低。体内注射结果显示神经生长因子缓释微球能够在基底前脑持续释放神经生长因子达4周以上。神经干细胞能够利用无血清的培养基在EGF和FGF-2刺激下从基底前脑组织中培养获得，并且能够在宿主脑内存活、迁移和分化，较好地与宿主脑组织整合。单侧切断FF后，基底前脑MS、VDB的胆碱能阳性神经元大量丢失。MS、VDB的胆碱能神经元的丢失与AD模型大鼠的空间学习记忆障碍密切相关。神经生长因子缓释微球和神经干细胞单独或者联合应用能够对AD模型鼠基底前脑胆碱能神经元有明显的补充和保护作用；联合应用的疗效明显优于单独应用。神经生长因子缓释微球和神经干细胞单独或者联合应用能够显著提高AD模型鼠的空间学习记忆能力；联合应用疗效更佳。更多还原

【Abstract】 Alzheimer’s disease （AD） is an irreversible, progressive disorder due to brain cells （neurons） deterioration, resulting in cognitive impairment, such as abnormal primarily memory, judgment and reasoning, movement coordination, and pattern recognition. A lot of investigations have showed that the prevalence of AD is strongly associated with aging. As the life span of human beings increases and more people live beyond the age of 65, the number of people with AD is also progressively increased. Contrasting with the enormous toll that this disease puts on patient, his or her caregivers, society as a whole is the lack of an effective therapy at present.One of the neuropathological characteristics of AD is atrophy or loss of cholinergic neuron observed in the basal forebrain （BF）. Neurotrophic factors are a good strategy to prevent or reduce the neuronal atrophy or loss. Among the neurotrophic factor family, nerve growth factor （NGF） is the most highly characterized neurotrophic factor for peripheral sympathetic neurons and a subpopulation of sensory neurons. The cholinergic neurons of BF express both the low affinity receptor (P75^NTF) and TrkA receptor, and respond to NGF by increased activity levels of the choline acetyltransferase （CHAT）. NGF is widely used for therapeutic studies in the experimental models of AD. However, NGF is a large molecular protein that does not easily cross the blood-brain barrier and has a short biologic half-life. Although infusing NGF solution into the cerebroventricular space via osmotic minipumps showed some therapeutic effects, long-term ICV NGF administration may cause negative side effects. The delivery of NGF to the BF poses a major challenge. New approach for drug formulation is required. One possibility is using controlled release formulations for the targeted delivery of NGF to brain.Neural stem cells have been successfully cultured, which solved the clinical problem of human fetal donors. Neural stem cells can generate neurons, astroglia and oligodendroglia in response to environmental signals, such as neurotrophic factors, retinoic acid and growth factors. The survival and differentiation of neural stem cells are associated with the inducement of the neurotrophic factors in vitro and in vivo. In central nervous system （CNS）, the survival, migration and differentiation of neural stem cells is not lack of the microenvironment of the brain. Neural stem cells can differentiate specific neurons of specific areas of the brain under the inducement of the microenvironment.In this study, we treated the rat model of AD by combining rhNGF microspheres with neural stem cells. On the one hand, neural stem cells will differentiate into neurons, which will supplie the degenerative neurons. On the other hand, the rhNGF released from the rhNGF microspheres will promote an/or ameliorate the survival and function of degenerative neurons. Furthermore, the released rhNGF from the microspheres will supply the nourishment for the survival, migration and differentiation of neural stem cells.The following three parts were included in this study:1. The preparation and evaluation of rhNGF microspheres. Microspheres, containing rhNGF and BSA, were prepared by （water-inooil）-in-water （W/O/W） emulsion and solvent evaporation technique with some modification. Briefly, 5 mg protein mixture of rhNGF and BSA （1/2000, w/w） in 100μl distilled water was emulsified in Poly （D, L-lactic-co-glycolic acid） （PLGA） （100mg） solution （3 ml of methylene dichloride and 1 ml of acetone） using sonication for lmin at 30 W over ice-bath. The first w/o emulsion was added to 25 ml of 1% PVA aqueous solution and homogenized at 1000rpm for 5 min over ice-bath. The resulting w/o/w double emulsion was stirred in a hood for 3-4 h to evaporate the organic solution at room temperature. The microspheres were collected by centrifugation, washed three times with distilled water and freeze-dried to obtain a free flowing powder. The rhNGF and FITC-BSA （1/2000, w/w） microspheres were prepared in the same way for in vivo studies. The following variables were designed to assess the effect of technical parameters on particle size, protein loading, and encapsulation efficiency. （1） Microspheres were prepared with a series of protein/PLGA ratios of 5%, 10% and 15% respectively. （2） Microspheres were prepared by adding different additives such as sucrose, PEG, and glycerol in the inner aqueous phase at concentrations of 5 and 10% （w/w） respectively. The results showed rhNGF in PLGA microspheres provided a sustained release formulation with low initial burst （11.4 %） for at least 35 days in vitro. The results showed that the higher the protein/polymer ratio, the higher the protein loading into the microspheres, and the lower the efficiency of protein encapsulation in the microspheres. The encapsulation efficiency could be increased with adding water-soluble additives in the inner aqueous layer prior to the emulsification. The efficiency of protein encapsulation was about 89.7 % without using additive and increased to 97.5 % with the using of PEG. The microencapsulation technique allowed an entrapment of biologically active rhNGF. This is the first report so far of rhNGF-loaded microspheres implanted into BF. The biodegradable rhNGF-loaded microspheres maintained a sustained release of rhNGF for at least 4 weeks in brain tissue.2. The migration and differentiation of neural stem cells in the basal forebrain of the rat model of AD. The single cell suspension from brain tissues of the basal brain of new-born （less than 1 day） rats by gentle mechanical dissociation with the use of trypsin were cultured in DMEM/F12 medium containing 2% B27, EGF and FGF-2 （10 ng/ml）. The cell were maintained for 5-7 day at 37℃in 5% CO₂. After 5-7 days, neurospheres were generated and floated in the culture. Unilateral fimbria-fornix （FF） of SD rats was transected to simulate the impairment of cholineric neurons of AD by the lesion of the pathway of septohippocampus. Neural stem cells （25,000 cells/μl×4μl） were injected into rats brain: AP+0.6 ram, LL+0.6 ram, DV-5.5 mm。The results showed neural stem cells could divide and proliferate under the inducement of EGF and FGF-2. After 5-7 d in vitro, large self-renewing and expandable spheres were generated. The proliferating and multipotential differentiation properties of neural stem cells were identified by immunochemistry. BrdU labeling and immunochemistry test were used to confirm the proliferation potential. In vivo studies, after 1 week, most of grafted cells stayed in the needle pathway, which survived well in the host brain, expressing Nestin antigen. After 2 weeks, most of grafted cells migrated neighbouring brain tissue, less of grafted cells expressing Nestin antigen. After 3 or 4 weeks, non-grafted cells express Nestin antigen, most of grafted cells migrated into whole basal forebrain and showed BrdU+NF or BrdU+GFAP.3. The therapeutic effect of combining rhNGF microspherewith neural stem cells on rat AD model. Adult male SD rats （12-20 weeks old, 250-310g） were used in this study. The rats were divided into five groups:①Normal control groups rats （Normal group, n=8）;②FF-lesion groups rats （LES group, n=8）;③RhNGF microspheres group rats （MIC group, n=12）;④Neural stem cells group rats （NSC group, n=12）;⑤The combination of rhNGF microspheres and neural stem cells group rats （MIC+NSC group, n=12）. Neural stem cells （25,000 cells/μl×4μl） were injected into rats brain: AP+0.6mm, LL+0.6mm, DV-5.5mm。Three milligrams of rhNGF microspheres （suspended in 10μl of dispersing medium） were stereotaxically implanted into the BF （coordinates: AP +0.6 mm, LL +0.6 mm, DV-5.5 mm from bregma）. The rats survived for four week after FF-lesion, discrimination learning and memory in Y-maze were observed. The morphologic data and percentages of cholinergic neurons of medial septum （MS） and vertical diagonal branch （VDB） were analyzed statistically by SPSS. The results showed after FF lesion, the percentages of cholinergic neurons at the lesion side to the intact side of MS and VDB were significantly lost. In MIC group, the percentages of cholinergic neurons at the lesion side to the intact side of MS and VDB were protected, which were significantly higher than that in the lesion group （p＜0.01）. In NSC group, the percentages of cholinergic neurons at the lesion side to the intact side of MS and VDB were protected, which were significantly higher than that in the lesion group （p＜0.01）. In MIC+NSC group, the percentages of cholinergic neurons at the lesion side to the intact side of MS and VDB were protected, which were significantly higher than that in MIC or NSC group （p＜0.01）. In the behavioral changes, the learning and memory of rats were significantly improved in the MIC+NSC group.Summary: The rhNGF-containing PLGA microspheres were preparaed by a W/O/W emulsion and solvent evaporation technique with some modifications. Using higher protein/polymer ratios in primary emulsions resulted in higher protein contents in the microspheres. The encapsulation efficiency could be increased with adding water-soluble additives in the inner aqueous layer prior to the emulsification. The in vitro rhNGF release lasted for more than 5 weeks. The rhNGF- microspheres maintained a sustained release of rhNGF for at least 4 weeks in brain. Neural stem cells can be obtained from basal forebrain by using free-serum medium adding EGF and FGF-2 （10 ng/ml）. After FF lesion the percentages of cholinergic neurons at the lesion side to the intact side of MS and VDB were significantly lost. The loss of cholinergic neurons was associated with the impairment of learning and memory of the rat AD model. The combinative therapy of rhNGF microspheres and neural stem cells could significantly ameliorate, rescue and supply the degenerative neurons of the basal forebrain, and significantly improve the spatial learning of the rat AD model.更多还原