中文摘要：

目的探讨刺五加干预对大鼠的潜在保护／毒性作用。方法应用随机数字表法将20只雄性SD大鼠分为刺五加组和对照组，每组10只。刺五加组大鼠给予刺五加提取物31．6mg／kg（相当于人类的生药量0．386g／kg）灌胃，灌胃药物的容积为10ml／kg，对照组给予等容积0．9％氯化钠溶液灌胃，均为1次／d，连续20d。分别于灌胃1、5、10、15、20d收集2组大鼠24h尿液，应用超高效液相色谱一串联质谱法检测尿液样本代谢产物的离子强度，并对2组大鼠不同时点尿液代谢产物的离子强度进行比较。将每个代谢产物的离子强度相对于总离子数进行归一化形成数据矩阵。使用EZinfo软件的偏最小二乘分析（PLS—DA）方法对不同时点检测的多个数据矩阵进行处理，根据PLS—DA得分图筛选特征性的代谢产物。对于刺五加组和对照组在PLS—DA得分图上显示相对位移较大的2组数据矩阵进一步使用正交PLS—DA（OPLS—DA）方法进行处理，获得变量投射重要性（VIP）值。从VIP〉1且P〈0．05的变量中筛选刺五加作用后表达发生变化的特征性代谢产物，检测其质荷比，通过与人类代谢组学数据库（HMDB）信息的比对做出这些代谢产物的鉴定，并通过查阅文献对这些代谢产物表征的刺五加对大鼠潜在的保护／毒性作用进行分析。结果PLS—DA得分图显示2组大鼠不同时点尿液样本代谢物轨迹在各自组内／时段内相似度较好，呈现各自聚类现象。刺五加组大鼠灌胃1、5、10、15d尿液中代谢产物表达轮廓间与对照组存在部分重叠，灌胃20d尿液中代谢产物表达轮廓则远离对照组，相对位移较大。对相对位移较大的2组数据应用OPLS—DA方法进一步分析，筛选出VIP〉1且P〈0．05的6种代谢产物，经与HMDB相关信息比对，分别鉴定为3-甲基鸟嘌呤、3-甲基戊二酰肉碱、3-羟基十二烷二酸、甲基巴豆酰肉碱、

英文摘要：

Objective To study the potential protective and toxic effects of Acanthopanax senticosus harms （AS） in rats. Methods Twenty male SD rats were divided into the AS-treated and control groups by random number table method. Each group comprised 10 rats. The rats in the AS-treated group were gavage-fed with AS extracts 31.6 mg/kg （equivalent to crude drug 0. 386 g/kg in human） once daily for 20 days and the volume of the drug was 10 ml/kg. The rats in the control group received an equal volume of saline once daily for 20 days. The urinary samples of 24 h from rats in the 2 groups which were gavage-fed for 1, 5, 10, 15, and 20 days were colleeted respeetively. Ultra-performance liquid ehromatography- quadrupole time-of-flight-mass spectrometry was used to deteet the ion intensities of metabolites from urinary samples. The ion intensities of urinary metabolites at different time points in the 2 groups were compared. The ion intensity of each metabolite was normalized with respect to the total ion intensities to generate a data matrix. The data matrix at different time points were processed by partial least-squares-discrimination analysis （PLS-DA） of EZinfo software. Characteristic metabolites were screened according to the PLS-DA score plot. Two sets of data matrix with relatively large distance were further processed by orthogonal partial least-squares-discrimination analysis （OPLS-DA） and the variable importance of projeet （VIP） scores were ealculated. Characteristic metabolites induced by AS intervention were screened from variables with VIP 〉 1 and P 〈 0.05 and their mass-eharge ratios were detected and eompared to the information in Human Metabolome Database （ HMDB）, and the corresponding metabolites were identified at last. The potential protective or toxic effects of AS in rats represented by these metabolites were analyzed through literature review. Results PLS-DA score plot showed that the urinary metabolic profiles at different time points in the 2 groups had good similarity within