【作者】 周冬；

【导师】 孙若鹏；

【作者基本信息】 山东大学 ， 儿内科， 2008， 博士

【摘要】 研究背景及目的:持续不断的血糖供应是保持脑功能正常的基本要求。由于多种原因使得生后能量代谢适应过程发生异常而使新生儿阶段成为人一生中最易发生低血糖的阶段。反复的、严重的新生儿低血糖可以造成不可逆的脑损伤,导致严重的神经系统后遗症,但至目前为止我们对反复、严重的新生期低血糖造成不可逆脑损伤的机制了解甚少。在临床实践中单纯研究新生儿低血糖是受限制的,因为低血糖常伴发在其它疾病中,而且我们尚没有一个可靠的新生期低血糖脑损伤的动物模型来进行深入的研究。因此建立一个可靠的新生期低血糖脑损伤的动物模型是十分必要和迫切的。多年来对成熟脑低血糖损害的研究表明,严重低血糖时,脑组织得不到足够葡萄糖的供应,使脑电活动减少,游离脂肪酸和氨基酸代谢障碍,使谷氨酸的水平增加,低血糖脑损伤是谷氨酸的中枢神经系统兴奋性中毒的最终结果。低血糖所致神经元死亡是由于糖供应下降而触发的一系列事件的结果,一旦被启动,恢复血糖浓度也不能阻断或逆转细胞的死亡过程。严重低血糖时,由于代谢的异常和ATP的不足,使得谷氨酸的产生增多而回摄取减少,使脑组织细胞外液中谷氨酸的浓度明显增高,导致谷氨酸受体的过渡激活。而且低血糖还可增加N—甲基天门冬氨酸(N-methyl-D-asparate,NMDA)型的谷氨酸受体对谷氨酸激活作用的敏感性,导致了谷氨酸激活浓度的阈值降低。NMDA受体与进出细胞的钾、钠、钙的通道有关。谷氨酸过度激活NMDA受体,使胞质内的钠离子和钙离子的浓度过度增加,超过神经元内环境平衡机制的调节范围,使跨膜离子梯度改变。低血糖时,引起的ATP和磷酸肌酸的缺乏使依赖能量的钠和钙离子正常跨膜浓度梯度恢复机制不能运作,终致细胞内钙超载,线粒体功能异常,活性氧簇(reactive oxygen species,ROS)的产生,DNA的破坏,聚腺苷二磷酸核糖多聚酶-1(poly(ADP-ribose)polymerase-1,PARP-1)被激活。钙离子的分布异常和PARP-1的过渡激活均可导致细胞线粒体的通透性转变,线粒体跨膜电位(mitochondrial transmembrane potent,△Ψm)的崩解,线粒体内促进凋亡的蛋白释放,包括细胞色素C(cytochrome c,cyt c)。释放进入胞质的cyt c在ATP/dATP的参与下,与凋亡蛋白酶活化因子-1(apoptotic protease activating factor-1,Apaf-1)以2:1的比例结合,形成cyt c/Apaf-1寡聚体,Apaf-1通过其氨基端和procaspase-9的功能区相互作用,形成cyt c/Apaf-1/caspase-9复合体,此复合体继续作用于下游的caspase-3,使caspase-3激活启动凋亡。研究认为,各种凋亡诱导因素最后都通过caspase起作用,caspase-8和caspase-9被认为是caspase系统的始动环节,caspase-3是终末步骤,caspase-3的活化现已被作为凋亡的标记。严重低血糖时,PARP-1的过渡激活大量消耗胞浆氧化型辅酶Ⅰ(oxidizedform of nicotinamide-adenine dinucleotide,NAD~+),造成细胞内NAD~+耗竭,使得葡萄糖酵解过程被阻断,导致即便恢复葡萄糖供应,由于NAD~+的耗竭组织也无法利用葡萄糖。研究证实在严重低血糖后额外提供PARP-1抑制剂,NAD~+或丙酮酸,α酮戊二酸等不需NAD~+参与的代谢底物可以减轻低血糖对神经元及其功能的损害。其中丙酮酸因其价格低廉,容易获得,无毒副作用越来越被研究人员重视,已在体外实验、成熟脑低血糖损伤和缺氧脑损伤的模型中得到研究,而丙酮酸对新生期低血糖脑损伤是否具有同样的保护性尚未得到认识。本课题首先建立一个可靠稳定的新生期低血糖的动物模型,然后在此动物模型的基础上探讨线粒体功能异常在低血糖导致神经元死亡的机制中的作用,最后给予低血糖动物额外提供丙酮酸钠,探讨丙酮酸钠对反复严重新生期低血糖脑损伤动物神经元及其功能是否具有保护性,为临床治疗新生儿低血糖,防止低血糖脑损伤后遗症建立新的思路和提供理论依据。研究方法:1.建立反复严重新生期低血糖脑损伤的大鼠模型。实验采用新生Wistar大鼠,胰岛素15 U/kg皮下注射和饥饿的方法诱导低血糖。160只实验动物随机分为胰岛素处理S组(insulin-treated rats with short hypoglycemia,INS-S,n=40)、胰岛素处理P组(insulin-treated rats with prolonged hypoglycemia,INS-P,n=40)、饥饿组(fasted rats,FAS,n=40)和对照组(control rats,CON,n=40)。INS-S组新生鼠低血糖持续1.5h,INS-P组新生鼠低血糖持续2.5h,FAS组新生鼠饥饿12h,处理中用微量血糖仪对新生鼠血糖进行检测。新生鼠低血糖的诱导在生后第二天(postnatal day,P2)、P4和P6进行3次,第三次低血糖处理后2h、6h、1d、3d、7d、14d取脑,制作脑组织切片经Fluoro-Jade B染色,计数大脑旁矢状面皮层、海马CA1区(cornu ammonis sector 1,CA1)、CA3、海马齿状回(dentate gyrus ofhippocampus,DG)、丘脑、下丘脑和梨状皮层7个区域变性神经元的数目。并在最后一次低血糖后6h取海马CA3区进行电镜检查。2.反复严重新生期低血糖脑损伤机制的研究:158只实验动物随机分为INS-P组(n=79)和CON(n=79)组。①线粒体膜电位的检测:低血糖处理后6h取新鲜大脑皮层制作单个神经元的悬液,用罗丹明123(Rhodamine123,Rho123)染色后流式细胞仪检测神经元悬液Rho 123的荧光强度。②胞浆cyt c水平的检测:低血糖处理后2h、6h、1h、3h、7h、14h取新鲜大脑皮层组织,低温差速离心法提取神经元的胞浆蛋白,western blot法检测大脑皮层神经元胞浆内cyt c的水平。③活化caspase-3的检测:低血糖处理后2h、6h、1h、3h、7h、14h取脑,用荧光免疫组化的方法检测caspase-3 P20在细胞中的表达,同时应用特异性神经核抗原(neuronal nuclear protein,NeuN)鉴定细胞类型,并观察神经元形态的变化。3.丙酮酸钠给药对反复严重新生期低血糖脑损伤的神经保护作用:36只实验动物随机分为胰岛素处理P组(insulin-treated rats with prolonged hypoglycemia,INS-P,n=12)、胰岛素处理+丙酮酸钠组(insulin-treated rats with prolongedhypoglycemia+pyruvate,INS-PP,n=12)和对照组(control rats,CON,n=12)。在终止实验动物低血糖的同时给予INS-PP组新生鼠丙酮酸钠(500mg/kg)。于第三次低血糖处理后1d取脑制作脑组织切片,Fluoro-Jade B染色计数各脑区变性神经元。第三次低血糖处理后六周Morris水迷宫测试大鼠空间学习和记忆能力。研究结果:1.反复严重新生期低血糖脑损伤的大鼠模型的建立:饥饿和胰岛素皮下注射均可诱导出新生大鼠低血糖的发生。但新生大鼠饥饿12h的方法在生后第二天更为有效,而在生后四、六天是血糖下降的程度较轻。新生大鼠对胰岛素十分敏感,单剂量胰岛素即可诱导出理想的低血糖。INS-S组和INS-P组新生鼠在胰岛素注射后0.5h血糖可低至2.5mmol/L以下,1.5h时至＜1.5mmol/L,而INS-P组2.5h时可低至1.2mmol/L以下。经FJB染色发现INS-S组、FAS组和CON大鼠脑切片中未见FJB阳性细胞出现,而INS-P组大鼠脑切片中见大量FJB阳性细胞。INS-P组在损伤后6h,1d,3d,5d,7d和14d旁矢状面,梨状皮层,海马齿状回,丘脑和下丘脑见大量广泛的FJB阳性细胞,FJB阳性细胞的数目及分布不随时间的变化而变化,在低血糖后7d和14d FJB阳性细胞的荧光信号不断变弱并逐渐消失。但海马CA1和CA3区始终未见FJB阳性细胞出现。损伤后6h电镜检查各组CA3区微观结构变化不大,仅见部分线粒体肿胀。2.反复严重新生期低血糖脑损伤机制的研究:①线粒体△Ψm的检测:流式细胞仪检测显示INS-P组大鼠神经元悬液Rho 123荧光强度明显降低(p＜0.01),仅为CON组的38.36%。②胞浆cyt c水平的检测:western blot检测示INS-P组在低血糖处理后2h、6h、1d、3d、7d、14d神经元胞浆内的水平明显升高(p值均小于0.01),2h、6h时为表达的高峰,3d后逐渐下降,14d时明显下降,但仍比CON组为高。③活化caspase-3的检测:免疫组化检查显示CON组大鼠脑组织切片中有少量caspase-3 P20阳性细胞散在分布,在第三次低血糖处理后6h INS-P组切片中即见有大量caspase-3 P20阳性细胞出现,1d时达高峰,7d后减弱,14d时明显减弱,但仍比对照组为强。Caspase-3 P20阳性细胞主要分布在旁矢状面,梨状皮层,海马齿状回,丘脑和下丘脑,而海马CA1,CA3仅见少量阳性细胞。Caspase-3 P20阳性细胞与NeuN荧光共定位,经DAPI复染,处理后14d,caspase-3 P20阳性细胞呈现凋亡的特征,包括染色质浓集到核周,呈空壳状,并可见凋亡小体。3.丙酮酸钠给药对反复严重新生期低血糖脑损伤的保护作用:经FJB染色,INS-PP组脑组织切片各脑区中FJB阳性细胞较INS-P组均见明显减少(p值均小于0.01),而CON组未见到FJB阳性细胞出现。低血糖处理后6w,Morris水迷宫测试结果显示INS-PP组大鼠找到平台的潜伏期和游泳的距离均明显低于INS-P组(p值均小于0.01),而与CON组无明显差异。在探索实验中,INS-PP组大鼠在平台所在象限的逗留时间和穿越平台位置的次数均明显高于INS-P组(p值均小于0.01),而与CON组无明显差异。研究结论:1.胰岛素(15 U/kg)皮下注射可诱导出新生大鼠低血糖的发生,且新生大鼠对胰岛素的敏感性和耐受性良好,饥饿12h也可诱导出新生大鼠轻度的低血糖。2.反复严重的新生期低血糖可造成大鼠明显的脑损伤,皮层、丘脑、下丘脑、海马齿状回是反复严重新生期低血糖的易损区。3.Fluoro-Jade B染色是一种有效的新生鼠低血糖脑损伤神经元变性的检测办法。4.反复严重的新生期低血糖脑损伤中,存在神经元线粒体△Ψm下降,cyt c释放,caspase-3激活,神经元呈现凋亡的形态学特征,线粒体介导的细胞死亡程序在反复严重的新生鼠低血糖脑损伤发生机制中起到一定的作用。5.丙酮酸钠对反复严重新生鼠低血糖时神经元及其功能具有明显的保护作用。创新和意义:1.本研究首次建立反复严重新生期低血糖脑损伤的大鼠模型,并应用变性神经元的特异性荧光染料证实其脑损伤的存在,发现反复严重新生期低血糖脑损伤的易损区,与成熟脑略有不同。2.研究证实了线粒体介导的细胞死亡程序在反复严重新生鼠低血糖脑损伤发生机制中起到一定的作用;反复严重新生鼠低血糖可导致神经元的凋亡。3.研究证明丙酮酸钠对反复严重新生鼠低血糖脑损伤神经元及其功能具有保护作用,为反复严重新生期低血糖脑损伤及其后遗症的防治提供了新的思路。更多还原

【Abstract】 Background and objectivesContinuous blood glucose availability is the basic requirement to maintain the function of the brain. The neonate suffers the most risk of hypoglycemia because of the deviations or perturbations in the adaptive responses of energy metabolism in the postnatal period. Hypoglycemia is a common disease of new born infants. Repititive and profound neonatal hypoglycemia can induce irreversible brain injury and result in severe neurologic sequelae, of which the mechanism was not elucidaled by hitherto. Studies on isolated neonatal hypoglycemia are limited in clinical practice, because hypoglycemia is often accompanied by other conditions, making it difficult to separate the effects of hypoglycemia from those of concurrent pathophysiologic states. Moreover no reliable animal model of brain injury induced by neonatal hypoglycemia is available to carry out more researches. So it is very necessary to establish a reliable animal model of brain injury induced by neonatal hypoglycemia. Studies on mature brain injury resulted from severe hypoglycemia indicated that decreased availability of glucose induces energy crisis in the brain that alters electrical activity, results in dysbolism of free fatty acid and amino acid, elevates the glutamate concentrations in brain extracellutar. Hypoglycemic brain injury is the end-result of excitotoicity in central nervous system induced by sustained glutamate receptor activation.Neuronal death resulting from hypoglycemia is the result of a series of events triggered by reduced glucose availability, and the normalization of blood glucose levels does not necessarily block or reverse this cell death process once it has begun. Because of the abnormality of metabolism and the decline of ATP level, hypoglycemia causes several-fold elevations in brain extracellular glutamate concentrations, resulted from the increased generation and the decreased uptake of glutamate, inducing sustained glutamate receptor activation. Sustained action of glutamate on neuronal NMDA （N-methyl-D-asparate） receptors, which concerned with channels of potassium, sodium and calcium, induces a massive influx of calcium ions into brain cells and the loss of ion homeostasis. Furthermore, hypoglycemia also can increase the sensibility of NMDA receptor to the activation of glutamate. The threshold of glutamate activation concentreation declines. The energy dependent revert mechanism of normal ion concentreation gradient across membrane lose efficacy because of the deficiency of ATP and phosphagen. It results intracellular calcium increases, mitochondrial dysfunction, production of reactive oxygen species （ROS）, DNA damage, PARP-1（poly （ADP-ribose） polymerase-1） activation. The abnormality of Ca⁺ distribution and the extensive activation of PARP-1 can induce mitochondrial permeability transition （MPT）, mitochondrial membrane potential （ΔΨm） disaggregate and the protein which promote apoptosis released, including cytochrome c （cyt c）. With the participation of ATP/dATP, cyt c, released into cytoplasm, combines apoptotic protease activating factor-1 （Apaf-1） to form cyt c/Apaf-1oligomer in the ratio of 2:1. The N-terminal of Apaf-1 interacts with the domain of procaspase-9 and form cyt c/Apaf-1/caspase-9 complex. Then the complex activates caspase-3 to switch on apoptosis. And all the inducing factors have been demonstrated that they work through caspase. Caspase-8 and caspase-9 are the initiators that trigger the apotosis pathway, and caspase-3 is the executor. Active caspase-3 has been used as apoptosis marker now.Activated PARP-1 consumes cytosolic oxidized form of nicotinamide-adenine dinucleotide （NAD⁺）, and because NAD⁺ is required for glycolysis, hypoglycemia-induced PARP-1 activation may render cells unable to use glucose even when glucose availability is restored. Administration of PARP inhibitors, NAD⁺, or pyruvate,α-ketoglutarate and other substrates, which can be metabolized in the absence of cytosolic NAD⁺, at the termination of hypoglycemia substantially reduced neuronal death in vulnerable brain regions and prevented cognitive impairment. In recent years researchers paid more attention on pyruvate because it is inexpensive, readily available, and innocuous. Many studies were carried out in the models of mature brain injury induced by hypoglycemia and hypoxia. Whether pyruvate can offer a neuroprotectective effect in repetitive and profound neonatal hypoglycemic brain injury still keep unknown.In the present study, we established a reliable animal model of brain injury induced by neonatal hypoglycemia first, then to study the mechanism of hypoglycemic brain injury induced by mitochondrial dysfunction. Fanily, the neuroprotectection of pyruvate to repetitive and profound neonatal hypoglycemic brain injury was investigated by administration of pyruvate at the termination of hypoglycemia and also provide new ideas and theories for treating repetitive and profound neonatal hypoglycemia.Methods:1. To establish a repetitive and profound neonatal hypoglycemic brain injury rat model. Neonatal hypoglycemia was induced by insulin （15 U/kg） hypodermic injection and fasting for 12h on postnatal day 2 （P 2）, P 4 and P 6. One hundred and sixty experimental animals were randomly divided into four groups: insulin-treated rats with short hypoglycemia （INS-S, n=40）, insulin-treated rats with prolonged hypoglycemia （INS-P, n=40）, fasted rats （FAS, n=40） and control rats （CON, n=40）. The period of hypoglycemia in INS-S was 1.5h and in INS-P was 2.5h. New born rats in FAS were fasting for 12h. The blood glucose was measured with a One Touch II blood glucose meter. FJB staining was used to quantify cell death and determine cellular localization of neuronal degeneration 2h, 6h, 1d, 3d, 7d and 14d after the last challenge of hypoglycemia. The dead neurons were counted in parasagittal cortex, piriform cortex, cornu ammonis sector 1 （CA1）, CA3 and dentate gyrus （DG） of hippocampus, thalamus and hypothalamus within each section, and compared among groups. At 6h after the last challenge of hypoglycemia, CA3 of hippocampus of each group was separated for electron microscopic examination.2. To study the mechanism of brain injury induced by repetitive and profound neonatal hypoglycemia. One hundred and fifty-eight experimental animals were randomly divided into two groups: insulin-treated rats with prolonged hypoglycemia （INS-P, n=79） and control rats （CON, n=79）.①To detect the mitochondrialΔΨm. Suspensions of neuron of each group were prepared 6h after the last challenge of hypoglycemia. The suspensions were detected by flow cytometry after Rho 123 stainning.②To detect the level of cyt c in neuronal endochylema. Tissue of cerebral cortex of each group was removed at 2h, 6h, 1d, 3d, 7d and 14d after the last challenge of hypoglycemia. Neuronal plasmosin was extracted by differential centrifugation in subambient temperature and detected by western bloting.③To detect activated caspase-3. Immunohistochemistry was performanced to detect activated caspase-3 P20 2h, 6h, 1d, 3d, 7d and 14d after the last challenge of hypoglycemia. Cell phenotype was determined using double-label immunofluorescent staining with caspase-3 P20 and neuronal nuclear protein （NeuN）, and the morphous of caspase-3 P20 positive neurons were observed.3. To investigate the neuroprotectection of pyruvate to repetitive and profound neonatal hypoglycemic brain injury. Thirty-six experimental animals were randomly divided into three groups: insulin-treated rats with prolonged hypoglycemia （INS-P, n=12）, insulin-treated rats with prolonged hypoglycemia + pyruvate （INS-PP, n=12） and control rats （CON, n=12）. Pyruvate was administrated to INS-PP new born rats at the termination of hypoglycemia. FJB staining was used to quantify cell death of each brain region, and compared among groups. Morris water maze test was performed to evaluate rat’s ability of spatial learning and memory.Results:1. The establishment of the repetitive and profound neonatal hypoglycemic brain injury rat model. Insulin injection and fasting both could induce consistent hypoglycemia in newborn rats. But on P 2, fasting could induce considerable hypoglycemic events, but on P 4 and P 6, the blood glucose levels were not very low. Newborn Wistar rats are very responsive to insulin. Only single dose insulin given by hypodermic injection can induce consistent hypoglycemia. The blood glucose of rats in INS-S and INS-P was <2.5mmol/l at 0.5h after injection and <1.5 mmol/l at 1.5h, and <1.2 mmol/1 at 2.5 h of rats in INS-P. FJB staining of brains of CON, FAS and INS-S rats did not result in any detectable fluorescence, while FJB+ cells were seen in all brains of INS-P 6h after injection of insulin. FJB+ cells were numerous in all sections of brains of INS-P. The regions with the greatest numbers of FJB+ cells are the parasagittal cortex, piriform cortex, hypothalamus and thalamus, then is DG. There was no time-dependent trend in numbers of FJB+ cells observed in brains collected from 6 h to 7 days after the last hypoglycemic insult (F_3,20=1.3600, p=0.2836). The fluorescence intensity of positive cells decreased from 7d after injection of insulin. The ultrastructural changes in brains of INS-S, INS-P and FAS animals were minimal.2. The mechanism of brain injury induced by repetitive and profound neonatal hypoglycemia.①Changes of the mitochondrialΔΨm. The fluorescence intensity of Rho 123 in suspensions of neuron of INS-P rats was declined significantly （p<0.01）. The fluorescence intensity only was 38.36% of that of CON.②The level of cyt c in neuronal endochylema. The level of cyt c in neuronal plasmosin of the brain following repetitive neonatal hypoglycemia expressed highly at 2h and 6h, decreased at 3d after the third insulin injection.③To detect activated caspase-3. Sporadic of caspase-3 P20 positive cells, with normal appearance of cell nucleus stained by DAPI, were observed in brains of INS-P rats at 2h after hypoglycemia insult and CON rats at every time point. In the brains collected 6 h or more after hypoglycemic insult of INS-P rats, the number and the fluorescence intensity of caspase-3 P20 positive cells increased at 6h after treatments, reached the peak at 3d and decreased at 7d. The regions with the greatest numbers of caspase-3 P20 positive cells are the parasagittal cortex, piriform cortex, DG of hippocampus, hypothalamus and thalamus. There were only small amounts of caspase-3 P20 positive cells detected in CA1 and CA3. The caspase-3 P20 positive cells co-located with NeuN. At 14d after hypoglycemic insult, many caspase-3 P20 positive cells showed morphological signs of apoptosis.3. The neuroprotectection of pyruvate to repetitive and profound neonatal hypoglycemic brain injury. The numer of FJB positive dead neuron in each brain region of INS-PP rats decreased significantly compared with that of INS-P rats, and no FJB positive cell was detected in brain sections of CON rats. Six weeks after the last challenge of hypoglycemia, in Morris water maze test, rats of INS-P showed a significant impairment in their ability to locate the platform when compared with rats of INS-PP and CON. Performance of the rats of INS-P was significantly better than rats of INS-P （P <0.05） and not significantly different than the CON group.Conclusions:1. Insulin （15 U/kg） hypodermic injections can induce consistent hypoglycemia in newborn rats. New born rat has good sensitiveness and tolerance to insulin injection. Fasting for 12h also can induce mild hypoglycemia in new born rats. 2. Repetitive and profound neonatal hypoglycemia can result in extensive neurodegeneration, and the neurons of cortex, thalamus, hypothalamus and dentate gyrus are more vulnerable to hypoglycemic insult in newborn rats.3. FJB staining is a useful method of marking neuronal degeneration in neonatal rat following hypoglycemic brain damage.4. The procedure of apoptosis induced by mitochondria plaies a key role in the mechanism of brain injury induced by repetitive and profound neonatal hypoglycemia.5. Pyruvate administration can provid a neuroprotectective effect to rat suffered repetitive and profound neonatal hypoglycemia. Pyruvate may be an effective intervention for patients with severe hypoglycemia.更多还原