褪黑素对脑缺血-再灌注损伤大鼠大脑皮质细胞毒性水肿的影响研究

Effect of melatonin on cytotoxic edema in the cerebral cortex of cerebral ischemia-reperfusion injury rats

目的 探讨褪黑素对大鼠脑缺血-再灌注损伤(CIRI)后大脑皮质细胞毒性水肿的影响及其机制。方法 将60只成年健康无特定病原体雄性Sprague Dawley大鼠完全随机分为假手术组(Sham组,12只)、CIRI组24只、褪黑素治疗组(24只)。采用改良的Zea-Longa线栓法建立大鼠CIRI模型,其中褪黑素治疗组在建模前、后30 min腹腔注射褪黑素(5 mg/kg);CIRI组在建模前、后30 min经腹腔注射等量的等渗盐水。Sham组分离右侧颈总动脉后缝合皮肤,并于假手术前、后30 min注射等量的等渗盐水。建模后2、6、12、24 h及72 h, 3组大鼠(每组6只)均行活体MRI T2加权成像(T2WI)、扩散加权成像(DWI)检查,结合表观扩散系数(ADC)图,动态分析褪黑素干预后CIRI大鼠脑梗死体积百分比、脑水肿百分比及大脑皮质相对ADC(rADC)值的变化。建模后6、24 h及72 h,采用脱氧核苷酸末端转移酶介导的2′-脱氧尿苷5′-三磷酸盐介导的缺口末端标记(TUNEL)法观察褪黑素对CIRI大鼠缺血侧大脑皮质细胞凋亡的影响;免疫荧光双标法观察褪黑素干预对星形胶质细胞水通道蛋白4(AQP4)和钠-钾-氯协同转运蛋白1(NKCC1)的影响;免疫荧光染色结合蛋白质印记,探讨褪黑素对离子钙接头蛋白抗原(Iba-1)标记的小胶质细胞活化的影响。建模后24 h,苏木素-伊红(HE)染色观察褪黑素对CIRI大鼠缺血侧大脑皮质细胞形态学的影响,酶联免疫吸附测定(ELISA)方法测定褪黑素对炎性因子白细胞介素1β(IL-1β)和肿瘤坏死因子α(TNF-α)的影响。采用Pearson相关性分析结合一般线性回归分析对建模后各时点CIRI组与褪黑素治疗组大鼠缺血侧大脑皮质Iba-1+细胞数差值与胶质纤维酸性蛋白(GFAP+)/AQP4+及GFAP+/NKCC1+细胞数差值进行相关性分析。结果 T2WI和DWI扫描结果显示,Sham组大鼠脑组织信号均匀,未见异常信号。建模后24 h, CIRI组见大范围异常信号脑损伤区,缺血侧大脑半球水肿,大脑中线向对侧移位;褪黑素治疗组脑损伤区范围较CIRI组减小,大脑中线向对侧移位减轻。建模后2、6、12、24 h及72 h,与CIRI组比较,褪黑素治疗组大鼠各时点脑梗死体积百分比、脑水肿百分比显著减少,大脑皮质rADC值显著增高(均P<0.05)。HE染色结果显示,Sham组大脑皮质细胞形态规则、排列整齐;建模后24 h, CIRI组缺血侧大脑皮质细胞形态不规则、排列紊乱,而褪黑素治疗组大脑皮质细胞形态较规则、排列较整齐。TUNEL检测及免疫荧光染色结果显示,Sham组大脑皮质见少量TUNEL+、GFAP+/AQP4+和GFAP+/NKCC1+、Iba-1+细胞表达,小胶质细胞胞体大小适中,呈高度分枝状;建模后24 h, CIRI组缺血侧大脑皮质TUNEL+、GFAP+/AQP4+、GFAP+/NKCC1+、Iba-1+细胞数较Sham组增多,小胶质细胞突起消失,形态呈现阿米巴样,褪黑素治疗组缺血侧大脑皮质TUNEL+、GFAP+/AQP4+、GFAP+/NKCC1+、Iba-1+细胞数较CIRI组减少,小胶质细胞胞体变小,突起较长。建模后6、24 h及72 h,与CIRI组比较,褪黑素治疗组大鼠各时点缺血侧大脑皮质TUNEL+、GFAP+/AQP4+、GFAP+/NKCC1+、Iba-1+细胞数均显著减少(均P<0.05)。蛋白质印迹及ELISA方法分析结果显示,建模后24 h,褪黑素治疗组Iba-1蛋白表达、IL-1β及TNF-α水平较CIRI组显著降低(均P<0.01)。建模后6、24、72 h,褪黑素治疗组与CIRI组缺血侧大脑皮质GFAP+/AQP4+及GFAP+/NKCC1+细胞数差值与Iba-1+细胞数差值呈极强正相关(r值分别为0.937、0.963),差异均具有统计学意义(均P<0.01)。结论 褪黑素可下调CIRI大鼠缺血侧大脑皮质星形胶质细胞AQP4及NKCC1蛋白的表达,改善细胞毒性水肿,减轻CIRI,其机制可能与抑制小胶质细胞活化有关。

Objective To investigate the effect of melatonin on cytotoxic edema in the cerebral cortex of cerebral ischemia-reperfusion injury(CIRI)rats and its mechanism.Methods Sixty adult healthy male Sprague Dawley rats without specific pathogens were completely randomized into Sham-operated group(12 rats),CIRI group(24 rats)and melatonin-treated group(24 rats).CIRI group and melatonin-treated group were established CIRI rat models using the modified Zea-Longa′s method.Melatonin(5 mg/kg)was injected intraperitoneally 30 min before and after modeling in the melatonin-treated group,and an equal amount of isotonic saline was injected 30 min before and after modeling in the CIRI group.In the Sham-operated group,the right common carotid artery was isolated and the skin was sutured,and an equal amount of isotonic saline was injected 30 min before and after Sham surgery.At 2,6,12,24 h and 72 h after modeling,in vivo T2-weighted imaging(T2WI)scans and diffusion-weighted imaging(DWI)combined with apparent diffusion coefficient(ADC)maps were performed by MRI to dynamically analyze the changes in percentage of brain infarct volume and percentage of brain edema,and the changes in relative apparent diffusion coefficient(rADC)values in the cerebral cortex of CIRI rats after melatonin intervention.At 6,24 h and 72 h after modeling,the effects of melatonin on apoptosis in the ischemic side of the cerebral cortex of CIRI rats were observed by deoxynucleotide end-transferase-mediated 2′-deoxyuridine 5′-triphosphate-mediated nick end-labeling(TUNEL)method,and the effect of melatonin on water channel protein 4(AQP4)and sodium-potassium-chloride co-transport protein 1(NKCC1)co-labeled with glial fibrillary acidic protein(GFAP)were observed by immunofluorescent double-labeling staining.The effect of melatonin on the activation of microglia labeled with Ionized calcium binding adapter molecule(Iba-1)was investigated by immunofluorescence staining combined with Western Blot(WB).At 24 h after modeling,hematoxylin-eosin(HE)staining was performed to observe the effects of melatonin on the morphology of the cells and the inflammatory factors interleukin-1β(IL-1β)and tumor necrosis factor-α(TNF-α)was determined by enzyme-linked immunosorbent assay(ELISA).The correlation between the number of Iba-1+cells and the number of GFAP+/AQP4+and GFAP+/NKCC1+cells in the cerebral cortex on the ischemic side of the rats was analyzed in each group.Results T2WI and DWI showed that,in the Sham-operated group,there was no abnormal signal in the brain tissue.At 24 h after modeling,the CIRI group showed a large area of abnormal signal brain damage,edema was visible in the cerebral ischemic hemisphere and the midline of the brain was shifted to the contralateral side.Compared with the CIRI group,the extent of brain damage was reduced in the melatonin-treated group,and the midline shift to the contralateral side of the brain was reduced.At 2,6,12,24 h and 72 h after modeling,the percentage of cerebral infarct volume and percentage of cerebral edema were significantly reduced,and the rADC values were significantly higher in all melatonin-treated groups compared with the CIRI group(all P<0.05).HE staining results showed that,at 24 h after modeling,the morphology of cerebral cortex cells on the ischemic side of the melatonin-treated group was regular and relatively neatly arranged compared with that of the CIRI group.TUNEL and immunofluorescence staining results showed that,in the Sham-operated group,cortical nerve cells were regular in morphology and neatly arranged.At 24 h after modeling,cortical nerve cells on the ischemic side of the CIRI group were irregular and disordered,while nerve cells were comparatively regular and neatly in the melatonin-treated group.Small amounts of TUNEL+,GFAP+/AQP4+and GFAP+/NKCC1+,Iba-1+cells were seen in the cerebral cortex in the Sham-operated group,microglia cell bodies are moderate in size and highly branched.At 24 h after modeling,the number of these cells in the cerebral cortex on the ischemic side was increased in the CIRI group compared with the Sham-operated group,and the microglia protrusions disappeared and showed an amoeboid phagocytes morphology.While the number of these cells in the ischemic side of the cerebral cortex was reduced in the melatonin-treated group compared with the CIRI group,and the microglia bodies became smaller and had longer protrusions.At 6,24 h and 72 h after modeling,the number of TUNEL+,GFAP+/AQP4+,GFAP+/NKCC1+and Iba-1+cells in the ischemic side of the cerebral cortex were significantly decreased in the melatonin-treated group compared with the CIRI group(all P<0.05).WB and ELISA results showed that,at 24 h after modeling,Iba-1 protein expression,IL-1βand TNF-αlevels in the ischemic side of the cerebral cortex were significantly lower in the melatonin-treated group compared with the CIRI group(all P<0.01).Correlation analysis showed that the number difference of Iba-1+cells in the cerebral cortex on the ischemic side was highly positively and linearly correlated with the number difference of GFAP+/AQP4+and GFAP+/NKCC1+cells between the melatonin-treated and CIRI groups,and the differences were statistically significant(all P<0.01).Conclusion Melatonin can down-regulate the expression of AQP4 and NKCC1 proteins in astrocytes in the ischemic side of the cerebral cortex of CIRI rats,ameliorate cytotoxic edema and reduce CIRI brain injury by a mechanism related to the inhibition of microglia activation.