【作者】 李先辉；

【导师】 金满文；

【作者基本信息】 华中科技大学 ， 药理学， 2013， 博士

【摘要】 研究目的：糖尿病是由于胰岛素抵抗或者胰岛素分泌不足所致的以高血糖为特征的代谢性疾病,可以引起多种并发症。糖尿病脑病是糖尿病最复杂的慢性并发症之一,临床上以认知功能障碍为主要表现,轻者学习记忆能力明显下降、空间定向困难,重者可发展为痴呆。迄今为止,其确切的病理损伤机制依然是迷雾重重,给糖尿病脑病的防治带来很大的困难。因此,深入研究糖尿病脑病的发生机制和防治措施,对提高糖尿病脑病患者生存质量具有重要意义。cAMP反应元件结合蛋白(CREB)是一种真核生物细胞核内调控因子,激活CREB可以增加突触可塑性蛋白、记忆相关蛋白的表达及新的突触生成,在神经元再生、突触形成及学习记忆等方而具有重要的调节作用。Akt/CREB是CREB激活通路中的一种。由于糖尿病状态下,胰岛素信号通路活性下降,首要影响下游的Akt的信号通路。因此很自然想到,糖尿病诱导的认知功能障碍可能是由于海马神经元Akt的活性下降,使得CREB的磷酸化减少,影响多种与记忆有关的靶基因转录,最终导致学习记忆功能障碍。因此,改善Akt/CREB信号通路的活性可能成为治疗糖尿病脑病有效靶点。多甲氧基黄酮(PMFs)是一类富含甲氧基的酚类化合物,广泛的存在自然界中。现有研究发现PMFs具有广泛的药理作用,如改善胰岛素抵抗,抗炎,抗氧化,改善认知功能障碍,可作为防治糖尿病诱导的认知功能障碍的候选药物。五甲基槲皮素(PMQ)是典型的多甲氧基黄酮类化合物,从结构式上可以推断也应该有上述功能。然而迄今为止,有关PMQ改善糖尿病诱导的认知功能障碍的研究还未见报道。有鉴于此,本课题通过采用自发性2型糖尿病GK大鼠和新生鼠STZ诱导的糖尿病大鼠模型,研究认知功能、海马神经元树突棘形态学及Akt/CREB信号通路在糖尿病大鼠中的改变,探讨Akt/CREB信号通路在糖尿病所致的海马神经元损伤和学习记忆功能障碍中的作用。同时,本课题也通过PMQ治疗自发性2型糖尿病GK大鼠,干预高糖诱导的海马神经元损伤,探讨PMQ保护糖尿病诱导的脑损伤的机制。研究方法：1、选择雄性新出生2天后的乳鼠,腹腔注射STZ (90mg/kg),对照组给以等体积生理盐水注射,在4周龄时,按空腹血糖随机分为对照组(SD组)和STZ处理组(STZ组),每组12只,连续喂养29周。12周龄雄性GK大鼠,12只,12周龄雄性Wistar大鼠,9只,常规饲料喂养20周。大鼠自由进食饮水,每天常规观察两次,每周定时称量大鼠体重,每天称量水食消耗,于32周龄做水迷宫实验,实验后检测大鼠空腹血糖,血清胰岛素水平,计算胰岛素抵抗指数(HOMA-IR)。取大鼠海马组织,一部分做尼氏染色、高尔基染色及免疫组化等形态学实验。一部分做Western blot实验分析Akt、CREB、SYP和BDNF蛋白表达。2、GK大鼠适应性喂养约4周后,按照空腹血糖随机分成：GK正常饲料组(GK组,n=12),GK+含PMQ饲料给药组(PMQ2.5、5、10mg/kg, n=8), GK(?)含二甲双胍饲料300mg/kg/d (MET组,n=12),以及Wistar正常饲料组(Wistar组,n=9)共6组,GK组和Wistar组的大鼠给予正常饲料喂养,其它组的大鼠分别给予相应含药饲料。记录每周大鼠的饮食、饮水及体重。于给药后第16周进行Morris水迷宫测试。水迷宫实验后检钡OGTT、大鼠空腹血糖、血清胰岛素水平,计算胰岛素抵抗指数(HOMA-IR);取大鼠海马组织,一部分做尼氏染色、高尔基染色及免疫组化等形态学实验。一部分做Western blot分析Akt、CREB、SYP和GluR1蛋白表达。3、成熟的神经元细胞加入不同浓度的葡萄糖(50、75、100、125mM)分别诱导24、48、72和96小时后。每孔加入20μlMTT工作液,孵育4h后,弃去各孔培养基,每孔加入200μl DMSO溶解。酶标仪570nm检测各孔吸光度值。选择诱导72小时使海马神经元活力下降到50%左右的葡萄糖浓度,作为最佳葡萄糖浓度和最佳作用时间,用于后续实验。实验分组如下：正培养液组,甘露醇组,最佳高糖组,最佳高糖+溶剂组,最佳高糖+PMQ(0.1、0.3、1、3、10μM)组,最佳高糖+PMQ (1μM)+LY294002(50μM)组。培养72小时后,MTT检测细胞活力,Western blot分析Akt、CREB、SYP和GluRl蛋白表达。实验结果：1.与正常Wistar和SD大鼠相比,32周龄的GK和STZ大鼠出现严重的高血糖、高胰岛素血症和高胰岛素抵抗指数,并伴有体重获得减少和水食消耗增加,同时在Morris水迷宫实验中,GK和STZ大鼠潜逃时间显著地延长,在目标象限逗留的时间百分比明显缩短。而且胰岛素抵抗指数越高,潜逃时间越长。在可见目标平台的实验中,GK和STZ大鼠潜逃时间无显著性改变,而且,游泳速度没有差异性。GK和STZ大鼠海马CA1区神经元细胞排列稀疏,细胞形态模糊,正常形态细胞明显减少,细胞活力显著降低,树突棘的密度也显著减少,长度变化不明显。而且GK和STZ大鼠海马神经元活力及树突棘密度与血糖浓度成负相关。同时,GK和STZ大鼠海马神经元p-Akt (Ser473). p-CREB (Ser133)、SYP和BDNF蛋白的表达均出现显著下调,总的Akt和CREB蛋白表达无明显改变。2.PMQ能够改善GK大鼠食水摄取指数,降低餐后血糖,降低空腹胰岛素水平。同时,PMQ也能改善葡萄糖耐量受损,提高胰岛素的敏感性,且呈一定的剂量依赖性,与GK组比较,具有显著性差异。PMQ干预能够显著缩短潜逃时间和游泳距离,同时增加在目标现象停留时间的百分比例,呈现一定的剂量依赖性,PMQ10mg/kg作用尤为明显。而且胰岛素抵抗指数越小,潜逃时间越短。PMQ干预后GK大鼠海马CA1区神经元细胞排列紧密,细胞形态清晰,正常形态细胞明显增多,细胞活力显著增加,树突棘的密度也显著增加。同时上调P-Akt、P-CREB、SYP和GluRl的蛋白表达。3.海马神经元细胞成熟后,在24h之内不同浓度的高糖处理,对神经元细胞活力及数量无显著影响,处理48h,50mM和75mM高糖组细胞活力下降,但无统计学意义,100mM和125mM的高糖组细胞活力下降出现显著性差异,处理72h,100mM的高糖使细胞活力下降到58.8%。而100mM的甘露醇处理对细胞活力无明显影响。PMQ干预高糖处理海马神经元后,神经元细胞活力和数量增加,呈现剂量依赖性,PMQ1μM时,作用达到平台,增加浓度,作用不再增加。LY294002干预后,可部分抵消PMQ的保护作用。同时,PMQ可以增加海马神经元p-Akt、p-CREB、SYP和GluRl蛋白的表达,但对Akt和CREB,总的蛋白表达无明显影响,LY294002干预后,p-Akt、p-CREB、SYP和GluRl蛋白的表达上调不明显,部分抵消了PMQ的作用。实验结论：1.糖尿病大鼠出现了认知功能障碍、海马神经元损伤及树突棘密度减少,其程度与胰岛素抵抗呈正相关。2.海马神经元细胞存活信号转导通路Akt/CREB活性降低在糖尿病认知功能障碍中起关键作用。3.PMQ能够改善糖尿病认知功能障碍,保护海马神经元,其作用机制部分是通过激活Akt/CREB信号通路来实现的。更多还原

【Abstract】 Objective:Diabetes is a metabolic disease characterized by hyperglycemia due to a lack of insulin secretion or insulin resistance, which can cause many complications. Diabetes encephalopathy is one of the most complicated chronic complications of diabetes, which main clinical symptoms presented cognitive dysfunction, the ability of learning and memory of the light symptoms is significantly decreased, spatial orientation is difficult, the person whith weigh can develop dementia. So far the exact pathological damage mechanism is still unclear, which bring great difficulty in getting diabetes encephalopathy in the prevention and control.The cAMP-responsive element-binding protein (CREB) is a member of a large family of structurally related transcription factors, which regulated the expression of memory-related proteines in the healthy brain. A number of signal pathways regulate CREB activity, in which Akt signaling pathways may be important one in the regulation of CREB phosphorylation. In diabetic state, the activity of Akt was inhibited, so it has been natural thought that diabetes-induced cognitive dysfunction may be because the Akt pathway was inhibited in the hippocampus, the expression of phosphorylated CREB was significantly downregulated and ultimately led to behaviour deficits. Thus, improving the Akt/CREB pathway activity may be a useful therapeutic strategy in treating humans with diabetic encephalopathy.Polymethoxylated flavones (PMF) are a group of highly methoxylated phenolic compounds existing widely in the natural world. Numerous reports have demonstrated that a few PMF family members possess the health beneficial effects including improving insulin resistance, anti-inflammatory, antioxidative and radical-scavenging capacities and ameliorating cognitive dysfunction. Pentamethylquercetin (PMQ) is a typical member of the polymethoxylated flavone family in the natural world. Previous studies have demonstrated that PMQ possesses many health beneficial effects including anti-diabetic, anti-carcinogenic and cardioprotective properties. However, there is no report whether PMQ can improve diabetes-induced cognitive dysfunction.Therefore, we have conducted studies using neonatally streptozotocin-induced diabetic rats (n-STZ rats) and genetic diabetic Goto-Kakizaki rats (GK rats) to assess the effects of insulin resistance and hyperglycemia on synaptic plasticity and behaviour, explore the mechanisms of Akt/CREB signaling pathways in diabetes-induced cognitive dysfunction and hippocampal neuron damage. At the same time, we investigated the protective role of PMQ on diabetes-induced cognitive dysfunction, using GK rats.Methods:1、The2-day-old male SD pups were treated with a single intraperitoneal injection of streptozotocin (90mg/kg, i.p.).Age-matched control SD pups were injected with sodium citrate solution. In4weeks, twenty-four rats were were randomly divided into control group (n=12) and STZ group (n=12), continuous fed for29weeks. Twelve-week-old (male) GK rats (n=12) and age-matched healthy Wistar rats (n=9) were continuous fed for20weeks. Tap water was provided ad libitum. Food and water intakes were measured on every2days, and rats were weighed weekly. On32weeks, the learning and memory of the rats were evaluated by water maze. After the evaluation of learning and memory, the fasting blood glucose and serum insulin were determined. Homeostasis model assessment of insulin resistance (HOMA-IR) was calculated. Then the rats were sacrificed under phenobarbital anesthesia and transcardial perfusion with ice-cold normal saline was performed. The hippocampus were rapidly removed and weighed. One part of the hippocampus were selected for further Golgi, Nissl and immunohistochemical staining, the other part were assayed for Akt, CREB, SYP and BDNF protein expression by western blot analysis.2-. The rats were randomized into six groups as follows (n=9-12):GK group; PMQ groups (GK rats given PMQ at2.5,5,10mg kg-1day-1, respectively); Metformin (MET) group (GK rats given MET at300mg kg-1day-1); Wistar group. All rats were acclimatized to their environment for4week prior to the beginning of the study. The Wistar group and GK group were fed on a standard rat chow (60%carbohydrate,20%protein,10%vitamin and mineral mix,5%fat and5%cellulose), while PMQ groups and MET group were given a diet containing2.5,5,10mg kg-1day-1of PMQ and300mg kg-1day-1of MET for16consecutive weeks. Tap water was provided ad libitum. Food and water intakes were measured on every2days, and rats were weighed weekly. On16weeks after PMQ treatment, the learning and memory of the rats were evaluated by water maze. After the evaluation of learning and memory, the fasting blood glucose and serum insulin were determined. Oral glucose tolerance test (OGTT) was performed. Homeostasis model assessment of insulin resistance (HOMA-IR) was calculated. Then the rats were sacrificed under phenobarbital anesthesia and transcardial perfusion with ice-cold normal saline was performed. The hippocampus were rapidly removed and weighed. One part of the hippocampus were selected for further Golgi, Nissl and immunohistochemical staining, the other part were assayed for Akt, CREB, SYP and GluR1protein expression by western blot analysis.3、Primary cultures of rat hippocampal neurons were prepared from the hippocampus of E18-E19Wistar rat embryos. In brief, the rats were decapitated and hippocampus was carefully isolated. After being washed twice using D-Hank’s buffer, hippocampus was digested with equal volumes of0.125%trypsin/ethylene diamine tetraacetic acid (EDTA) for approximately15min at37℃. The hippocampus was then washed with Hank’s solution containing10%fetal bovine serum (Biochrom, Cambridge, UK) to stop digestion. Then, neurons were plated in six-well plates (8.75×104cells/cm2) or in96well plate plates (0.85×104cells/cm2). The cultured neurons were incubated with DMEM in a humidified5%CO2incubator at37℃for6h, and the medium was then replaced with Neurobasal/B27medium. Neurobasal/B27medium containing25mM glucose (control condition) is intended to give optimal growth and long-term survival to rat embryonic hippocampal neurons. After7days in culture, cells were incubated with50,75,100,125mM of glucose or with75mM mannitol (plus75mM mannitol in normal medium), which was used as an osmotic control, and maintained for further24,48,72and96h. The optimal time point and glucose concentration were selected for further study as a representation of hyperglycemic condition. At day8, hippocampal neurons were incubated with various concentrations of PMQ (0.1,0.3,1,3and10μM, PMQ were dissolved in dimethylsulphoxide.) or vehicle under this optimal time point and glucose concentration. To study the mechanism involved in PMQ effects on high glucose-induced neurotoxicity, hippocampal neurons were incubated under this optimal time point and glucose concentration with PMQ (1μM), either with or without LY294002(a specific inhibitor of PI3kinase;50μM; Beyotime Institute of Biotechnology, China).Results:1. GK and n-STZ rats exhibited severe hyperglycemia, hyperinsulinemia and significant insulin resistance, as well as significantly reduced weight gain and increased food and water uptake compared with Wistar rats and SD rats. At28weeks of age, GK and n-STZ rats displayed significantly increased escape latencies in the hidden platform paradigm, while spending less time in the target quadrant in a probe trial (Fig.1G) compared with Wistar rats and SD rats. The performance of all groups in the trial with the visible platform was not significantly different. Furthermore, velocities were similar. The escape latency time varied in proportion to the insulin resistance. In the CA1sectors of GK and n-STZ rats, cells were sparsely arranged and the cell shapes were fuzzy; cells with eumorphism were significantly reduced; the%cell viability was significantly lower. The dendritic spines density was significantly decreased in the CA1hippocampus of the GK and n-STZ rats compared with Wistar rats and SD rats. The fasted glucose levels and the neuronal cell viability and dendritic spine density were observed a significant positive correlation between these parameters. The GK and n-STZ rats showed a significant decrease in phosphorylation of Akt and CREB, and the levels BDNF and synaptophysin also was a significant decrease.2. PMQ treatment alleviated weight loss, ameliorated the polyphagia and polydipsia symptom, reduced levels of fed blood glucose, fasting insulin levels and ameliorated glucose intolerance. PMQ intervention significantly dose-dependently improved insulin resistance. Treatment with MET in GK rats showed similar results.PMQ treatment significantly decreased escape latency and swimming distance and increased the time spent in target quadrant compared to GK rats group, Treatment with MET in GK rats showed similar results. The escape latency time varied in proportion to the insulin resistance.The administration of PMQ and MET significantly increased the dendritic spines density compared with the GK rats. However, the dendritic length of the pyramidal cells of the CA1hippocampus was not significantly different in all groups. In PMQ (10mg/kg group) and MET group, cell shapes were clear and the cell structure was compact; cells were relative large and had abundant cytoplasm and Nissl bodies. The%cell viability was significantly higher. PMQ significantly activated the Akt/cAMP response element-binding protein pathway and increased the expression of memory-related proteins in the downstream part of the Akt/cAMP response element-binding protein pathway, such as SYP and GluR1.3. When incubating cells for24hours, we did not find any detrimental effect of high glucose (HG) on cell viability at concentrations ranging from50to125mM; when incubated for48,72and96h, cells activities were inhibited. Dealt72h with100mM of glucose, the hippocampus neurons activity dropped to58.8%(P<0.01vs. control group). So this time point and glucose concentration were selected for further study as a representation of high glucose condition. To support the in vivo findings reported above, hippocampal neurons grown in HG media were treated with PMQ. As expected, the cells treated with0.1,0.3,1,3and10μM of PMQ in the presence of100mM glucose for72h exhibited higher viability than HG group. Similarly, we have found that the high glucose condition can lead to shortening dendrite of hippocampal neurons, decreasing the numbers of the branches, influencing the normal growth of neuronal cells, while PMQ (1μM) can ameliorate the pathological changes. Pretreatment with LY294002resulted in a part loss of PMQ-mediated protective effects in hippocampal neurons. Concurrent treatment with1μM PMQ resulted in elevations in p-Akt, p-CREB, SYP, GluRl protein. When hippocampal neurons were pretreated with LY294002for30min, PMQ did not increase the levels of p-Akt, p-CREB, SYP, GluRl protein.Conclusion:1. Insulin resistance could predominantly reduce Akt/cAMP response element-binding protein activation in the brain, which is associated with a higher risk of cognitive dysfunction.2. Pentamethylquercetin ameliorates cognitive deficits through enhancing of Akt/CREB signaling pathways in diabetic goto-kakizaki rats.3. Pentamethylquercetin could provide a new potential option for prevention of the cognitive dysfunction in diabetes.更多还原