摘要目的:探讨Cre-loxP重组酶系统构建的兴奋性神经元腺苷酸活化蛋白激酶α1( AMPKα1)基因特异性敲除模型小鼠大脑能量代谢及认知功能的改变。 方法:将杂交繁育获得的16只基因型为 AMPKα1 flox/flox/Camk2a-Cre/ERT2 的6月龄小鼠按随机数字表法分为 AMPKα1敲除组( n=8)与 AMPKα1野生组( n=8)， AMPKα1敲除组小鼠每天腹腔注射0.1 mL他莫昔芬(20 mg/mL，溶于玉米油)以控制 AMPKα1基因在兴奋性神经元中的敲除， AMPKα1野生组小鼠每天腹腔注射等量玉米油以作对照。连续注射5 d，并等待7 d后分别采用Morris水迷宫和T迷宫实验检测2组小鼠的空间学习记忆及空间工作记忆能力，采用化学交换饱和转移成像(CEST)观察2组小鼠海马及海马周围皮层的葡萄糖代谢情况，采用Western blotting实验检测2组小鼠海马及海马周围皮层的AMPKα1、谷氨酸受体1(GluR1)蛋白表达情况。 结果:与 AMPKα1野生组比较， AMPKα1敲除组小鼠第3、4天的逃避潜伏期明显延长[(13.90±3.72) s vs. (22.40±6.28) s；(11.95±3.86) s vs. (22.39±9.77) s]，穿越平台次数明显减少[(5.25±1.83)次 vs. (1.75±1.28)次]，自由交替率明显降低[(73.21±9.16)% vs. (48.21±11.29)%]，海马及海马周围皮层的葡萄糖代谢水平明显降低(2.77±0.67 vs. 1.51±0.81；2.42±0.95 vs. 1.31±0.83)，海马及海马周围皮层AMPKα1、GluR1蛋白表达明显降低(AMPKα1：0.70±0.05 vs. 0.49±0.03，0.98±0.04 vs. 0.64±0.06；GluR1：1.22±0.18 vs. 0.60±0.11，0.96±0.08 vs. 0.79±0.04)，差异均有统计学意义( P<0.05)。 结论:特异性敲除兴奋性神经元 AMPKα1基因可导致小鼠大脑葡萄糖代谢异常，从而引起其认知功能障碍，其机制可能与能量代谢障碍引起兴奋性突触障碍有关。

abstractsObjective:To investigate the changes of brain energy metabolism and cognitive function in mice with specifically knocking out AMP-activated protein kinase α1 subunit ( AMPKα1) gene in the excitatory neurons by Cre-loxP recombination system. Methods:Sixteen 6-month-old mice with genotype AMPKα1 flox/flox/Camk2a-Cre/ERT2 obtained by hybrid breeding were randomly divided into AMPKα1 knockout group ( n=8) and AMPKα1 wild-type group ( n=8). Mice in the AMPKα1 knockout group were intraperitoneally injected 0.1 mL tamoxifen (20 mg/mL, dissolved in corn oil) daily for a consecutive 5 d to control AMPKα1 gene knockout in the excitatory neurons; and mice in the AMPKα1 wild-type group were intraperitoneally injected 0.1 mL corn oil daily for a consecutive 5 d. Seven d after that, Morris water maze and T maze experiments were employed to detect the spatial learning and memory abilities and spatial working memory of these mice; chemical exchange saturation transfer imaging (CEST) was used to observe the glucose metabolism in the hippocampus and cortex surrounding the hippocampus; Western blotting was used to detect the AMPKα1 and glutamate receptor 1 (GluR1) protein expressions in the hippocampus and cortex surrounding hippocampus of two groups. Results:(1) Morris water maze showed that, as compared with those in the AMPKα1 wild-type group, mice in the AMPKα1 knockout group had significantly prolonged escape latency ([13.90±3.72] s vs. [22.40±6.28] s; [11.95±3.86] s vs. [22.39±9.77] s]) on the 3 rd and 4 th d of experiment, statistically decreased times crossing the platform ([5.25±1.83] times vs. [1.75±1.28] times, P<0.05). (2) T-maze experiment showed that as compared with that of the AMPKα1 wild-type group, the free alternation rate in mice of the AMPKα1 knockout group was significantly decreased ([73.21±9.16]% vs. [48.21±11.29]%, P<0.05). (3) CEST showed that the glucose metabolism levels in the hippocampus and cortex surrounding the hippocampus of AMPKα1 knockout group were significantly lower than those in AMPKα1 wild-type group (1.51±0.81 vs. 2.77±0.67; 1.31±0.83 vs. 2.42±0.95, P<0.05). (4) Western blotting showed that the AMPKα1 and GluR1 protein expressions in the hippocampus and cortex surrounding the hippocampus of the AMPKα1 wild-type group were significantly higher than those of the AMPKα1 knockout group (AMPKα1: 0.70±0.05 vs. 0.49±0.03, 0.98±0.04 vs. 0.64±0.06; GluR1: 1.22±0.18 vs. 0.60±0.11, 0.96±0.08 vs. 0.79±0.04, P<0.05). Conclusion:Specifically knocking out AMPKα1 in excitatory neurons can result in abnormal glucose metabolism in the brain of mice, and thus cause cognitive dysfunction, whose mechanism may be related to excitatory synaptic disorder caused by energy metabolism disorder.