可溶性Aβ42寡聚物诱导的食蟹猴阿尔茨海默病拟临床模型的构建和评估

Modeling of Alzheimer's disease (AD) that could mostly mirror the clincal AD patients either in behavior manifestation or pathological features will play a critical role in new drug invesigaton, early stage diagnosis and some translational research, however, it is still limited in pathological observation and lack of effective methods for modeling and systemic validation. Based on our previous work, we currently have developed the technologies for modeling of AD and in vivo evaluation. The soluble Aβ42 oligomers will be continuously stereotaxic infused through osmotic pump into ipsilateral ventricle of the monkeys and then stereotaxic microinjected into the entorhinal cortex. during the 6 months of period before and after modeling, the alterations of cognitive function will be tested by CANTAB system, the formation and distribution of Aβ will be in vivo detected by a specific radiopharmaceutical 18F-AV45 PET imaging, and related biomarkers in CSF and blood such as Aβ40、Aβ42，P-tau 181、BDNF and NP1 will be examed by ELISA. The appearance of behaviral cognitive impairement and abnormally increased 18F-AV45 upake would be as the endpoint of soluble Aβ42 oligomers induction , then brain tissues will be collected after the euthanasia, the Aβ plaques, tau pathology in the hippocampus, entorhinal area, temporale lobe cortex will be performed by immunohistopathological analysis, meanwhile, the concentration of tau and BDNF in the hippocampus and entorhinal cortex will be measured by ELISA and western bolts respectively. So, we suppose the appearance of cognitive impairement and similar pathological changes could be the criteria for the success of AD model, and finally based on the changes of above mentioned in vivo evaluating indexes(cognitive impairement, concentration of biomarkers and the uptake of 18F-AV45), the AD model could be validated, so it will provide an effective carrier for investigating new drugs，early stage diagnosis, and related translational researches in AD.

在与人类接近的灵长类实验动物建立在认知记忆行为和病理特征上最大程度模拟阿尔茨海默病(AD)的动物模型是AD新药研发、早期诊断和转化医学研究的关键所在,但仍局限在简单的病理观察,缺乏有效的建模方法和系统性的评价鉴定体系。基于我们在前期工作基础已具备了构建AD拟临床模型的方法和在活体状态进行评价验证的技术手段。即在老年食蟹猴将可溶性的Aβ42寡聚物立体定向脑室内连续微渗透泵灌注和内嗅皮质区微注射，用CANTAB测试认知记忆行为学变化，用18F-AV45-PET显像评价脑内Aβ形成和分布，另用ELISA检测脑脊液中Aβ40、Aβ42，P-tau 181等生物标记物含量。安乐死后取相应脑组织，分析Aβ斑块、Tau病理改变和Tau蛋白含量，以出现认知障碍和类似AD的病理变化为模型成功的标志，根据上述活体状态评价指标可建立模型验证评价标准，从而为AD发病机制研究、早期诊断以及新的治疗方法研发提供载体。

本研究从转化医学应用的目的初步研究了建立灵长类拟临床阿尔茨海默病（AD）模型的方法和评价技术。首先收集37只食蟹猴脑脊液并检测分析了AD相关的特异性生物标记物Aβ40、Aβ42和p-tau 181含量，与4-6岁组比较，12-19岁组及＞20岁组Aβ42含量显著减低（(P＜0.05)，年龄与Aβ42存在统计学意义的负相关关系(r2=0.17, p=0.018)，而Aβ40和p-tau 181未见显著变化。本研究选取6只老年食蟹猴，首次应用对流增强给药技术（CED）立体定向手术脑内注射可溶性Aβ42寡聚体（3只）和单体（3只）构建AD模型，注射靶点为两侧海马外后方，靶点注射液体总容积可达150-250ul,MRI扫描显示注射的溶液在靶点周围呈球形弥散，且未造成局部损伤效应。本研究已建立了稳定的可溶性Aβ42寡聚体和单体的准备、合成的方法，并通过ELISA方法鉴定了在造模手术前后注射样本的稳定性。同时，为了动态评价灵长类动物认知记忆功能变化。本研究根据WGTA原理自主开发了两种工作记忆检测装置，分别为延迟位置比对测试（DMTP）和延迟物品比对测试（DMTS）装置。并用于中老年组和老年组食蟹猴的认知功能对比研究，中老年组DMTP测试，3s延迟使得正确率从100%下降到93%(P＜0.01)，10s延迟正确率进一步下降到84%(P＜0.001)；在老年组，3s延迟后仅显示有正确率下降的趋势(P＞0.05 but ＜0.1)，10s延迟则进一步下降到73%(P＜0.01)。DMTS在老年组3s延迟后，正确率从87%下降至71%(P＜0.001)，在10s延迟后进一步下降到58%(P＜0.01)。虽然在中老年组在3s和10s延迟后正确率均较无延迟时下降了13%(P＜0.01；P＜0.01)，但在3s和10s延迟之间未见显著性差异。研究显示年龄与DMTS正确率呈显著负相关关系(r2=0.67, p=0.0006)。因此表明工作记忆检测装置尤其是DMTS是评价灵长类认知功能的有效方法。造模前6只动物进行DMTS检测作为基线， 之后每隔1个月共3次进行造模手术并进行DMTS检测。 在随机0、3、10、20、30、60s延迟条件下应用DMTS检测，截至当前DMTS正确率数据尚未显示显著的工作记忆功能的减退趋势并仍需在后续实验中持续观察。