目的 研究青礞石对戊四唑（pentylenetetrazol，PTZ）点燃癫痫大鼠肠道菌群的影响，从肠道微生态环境视角探讨青礞石治疗癫痫疾病可能的作用机制。方法 以青礞石粉末为研究对象，用PTZ点燃法建立大鼠癫痫动物模型，实验分为对照组、模型组、卡马西平组和青礞石组（高剂量组为临床用量的20倍、低剂量组为临床用量的5倍）。给药4周后，取各组大鼠结肠内容物，通过16S rRNA测序对各组大鼠肠道菌群的多样性进行分析，利用UPARSE、SPSS 18.0等对结果进行生物信息学及统计学分析。结果 α多样性分析可知，模型组大鼠肠道菌群的Chao 1、Ace丰富度指数与对照组相比没有显著性变化；与对照组、模型组相比，青礞石组尤其是高剂量组对细菌群落丰富度有显著影响（P<0.05）。Shannon、Simpson群落多样性指数分析结果显示，模型组与对照组无显著性差异；青礞石组对群落多样性干预不明显。青礞石主要干预肠道菌群的丰富度、卡马西平主要干预肠道菌群的多样性。β多样性各组层级聚类均能较好地分开，其中青礞石高剂量组与低剂量组部分样本不能明显分开，2组样本可归为一类，对照组、模型组可归为一类，卡马西平组单独归为一类。3D-主成分分析（3D-principal component analysis，3D-PCA）、3D-主坐标分析（3D-principal co-ordinates analysis，3D-PCoA）结果显示，各组菌群结构轮廓均能明显分开，青礞石低剂量组介于对照组和模型组之间，菌群结构轮廓部分与模型组、对照组有重合交叉，且有偏向对照组的明显趋势。非度量多维尺度（nonmetric multidimensional scaling，NMDS）分析结果显示，对照组、模型组、青礞石组集中在一个区域，青礞石组与模型组、对照组有重合交叉，青礞石高剂量组、低剂量组对肠道菌群结构有相似的干预效果。物种组成门水平分析可知，各组大鼠的优势菌群主要有厚壁菌门（Firmicutes）、拟杆菌门（Bacteroidetes）、疣微菌门（Verrucomicrobia）、变形菌门（Proteobacteria）和放线菌门（Actinobacteria），其中厚壁菌门占比最大，其次是拟杆菌门。模型组疣微菌门相对丰度较对照组显著增加（P<0.01），模型组变形菌门的相对丰度最高；青礞石干预后疣微菌门相对丰度下降，变形菌门相对丰度有不同程度降低。属水平分析可知，与对照组相比，模型组中norank_f__Muribaculaceae（P<0.05）、Akkermansia菌属（P<0.01）相对丰度显著增加，乳杆菌属显著减少（P<0.05）；青礞石干预后可降低Akkermansia菌属、升高乳杆菌属Lactobacillus的相对丰度。另外，线性判别分析效应大小（linear discriminant analysis effect size，LEfSe）分析可知厚壁菌门的瘤胃球菌科（Ruminococcaceae）在模型组中显著富集，给予青礞石干预后，瘤胃球菌科相对丰度显著降低。结论 青礞石对PZT点燃癫痫大鼠肠道菌群物种组成的丰富度、多样性具有明显的调节作用，可有效干预肠道微生态的重建。

Objective To study the effect of Chloriti Lapis on the gut microbiota in pentylenetetrazol (PTZ)-kindled epileptic rats, and explore the possible mechanism of Chloriti Lapis in the treatment of epilepsy from the perspective of intestinal micro-ecological environment. Methods Taking Chloriti Lapis powder as the research object, the rat epilepsy animal models were established by the PTZ kindling method. The experiment was divided into blank group, model group, carbamazepine group, and Chloriti Lapis group (the high-dose group was 20 times of the clinical drug, the low-dose group was five times). After 4 weeks of treatment, the colon contents of each group of rats were taken, and the diversity of the gut microbiota of each group of rats were analyzed by 16S rRNA sequencing, and the results were analyzed in bioinformatics and statistics using UPARSE, SPSS 18.0, etc. Results The α-diversity analysis showed that, compared with the blank group, the Chao 1 and Ace richness index of the gut microbiota flora in the model group had no significant change; Compared with the blank group and the model group, the Chloriti Lapis group, especially the high-dose group, had a significant impact on the richness of gut microbiota flora (P<0.05). The results of Shannon and Simpson index analysis showed that there was no significant difference between the model group and the blank group. Chloriti Lapis mainly interfered with the richness of gut microbiota flora, and carbamazepine mainly interfered with the diversity of gut flora. The results showed that the hierarchical clustering of each group of β-diversity could be well separated. Some of the samples in Chloriti Lapis high-dose group and low-dose group could not be separated obviously, and the two groups of samples could be classified into one group. The blank group and the model group could be classified into one group, and the carbamazepine group could be classified into one group. The results of 3D-PCA and 3D-PCoA analysis showed that the bacterial structure outline of each group could be obviously separated. The low-dose Chloriti Lapis group was between the blank group and the model group, and the part of the outline of the bacterial structure diversity outline overlapped with the model group and the blank group, and tended to the blank group. The NMDS analysis results showed that the blank group, the model group and the Chloriti Lapis group are located in one region, with the Chloriti Lapis group overlapping with the model group and the blank group. The high-dose and low-dose of Chloriti Lapis groups had similar intervention effects on the gut microbiota flora diversity. The composition analysis on phylum level showed that Firmicutes, Bacteroidetes, Verrucomicrobia, Proteobacteria, and Actinobacteria were the dominant bacteria in each group. Firmicutes accounted for the largest proportion, followed by Bacteroidetes. The richness ratio of Verrucomicrobia in the model group was significantly higher than that in the blank group (P < 0.01), and the relative richness of Proteobacteria in the model group was the highest. After Chloriti Lapis intervention, the richness of Verrucomicrobia decreased and the relative richness of Proteobacteria decreased in some degrees. Analysis on genus level showed that compared with the control group, the relative richness of norank_f_Muribaculaceae (P < 0.05) and Akkermansia (P < 0.01) in epileptic rats was significantly increased, and Lactobacillus was significantly reduced (P < 0.05). The intervention of Chloriti Lapis could reduce the relative richness of Akkermansia and increase that of the Lactobacillus. In addition, the linear discriminant analysis effect size (LEfSe) analysis showed that the Ruminococcaceae of Firmicutes was significantly enriched in the model group, and the relative richness of that was significantly reduced after the intervention of Chloriti Lapis. Conclusion Chloriti Lapis has a significant regulatory effect on the richness and diversity structure and composition of the gut microbiota of PTZ-kindled epileptic rats, and can effectively interfere with intervene in the reconstruction of the intestinal micro-ecology.

R285

国家自然科学基金项目（81673566）；全国矿物药资源普查项目（2019）；江苏省研究生科研与实践创新计划项目（KYCX18-1608）；2017年中医药公共卫生服务补助专项"全国中药资源普查项目"（财社[2017]66号）