目的 建立一测多评结合化学模式识别法及加权TOPSIS与灰色关联度融合模型评价不同产地凤尾草Pteris multifida质量差异的方法，以提高和完善凤尾草质控水平。方法 对18批不同产地的凤尾草用70%甲醇提取后，以木犀草素为内参物，采用QAMS法同时测定凤尾草中蕨素B、蕨素Z、蕨素C 3-O-β-D-葡萄糖苷、新西兰牡荆苷、忍冬苷、木犀草素-7-O-葡萄糖苷、野漆树苷、木犀草素、芹菜素、胡萝卜苷、β-谷甾醇含量，参照《中国药典》2020年版检测方法对其醇溶性浸出物、总灰分和酸不溶性灰分进行测定。检测结果利用SIMCA 14.1软件进行化学模式识别分析，在此基础上运用加权TOPSIS与灰色关联度融合模型对18批凤尾草进行质量差异评价。结果 外标法与QAMS法结果较接近，18批不同产地凤尾草的70%甲醇提取液中蕨素B、蕨素Z、蕨素C 3-O-β-D-葡萄糖苷、新西兰牡荆苷、忍冬苷、木犀草素-7-O-葡萄糖苷、野漆树苷、木犀草素、芹菜素、胡萝卜苷、β-谷甾醇含量差异较大，质量分数分别为0.214～0.342、0.255～0.458、0.119～0.332、0.556～1.096、0.048～0.146、0.701～1.197、1.460～2.553、0.940～1.609、0.085～0.217、0.034～0.079和0.084～0.163 mg/g。化学模式识别法将18批样品分为3类，木犀草素-7-O-葡萄糖苷、野漆树苷、蕨素C 3-O-β-D-葡萄糖苷、忍冬苷、木犀草素和芹菜素是影响凤尾草质量的主要差异性成分。18批样品的综合相对贴近度为0.310 1～0.657 7，证明了不同产地凤尾草存在地域差异。结论 通过多指标定量、化学模式识别结合加权TOPSIS与GRA融合模型方法，评价了不同产地凤尾草的质量差异性，为其质控水平的提高和资源开发提供了依据。

Objective To establish a method for evaluating the quality difference of Pteris multifida from different producing areas by QAMS combined with chemical pattern recognition and weighted TOPSIS and grey correlation degree fusion model, in order to improve and perfect the quality control level of P. multifida. Method A total of 18 batches of P. multifida from different habitats were extracted with 70% methanol, the contents of pterosin B, pterosin Z, pterosin C 3-O-β-D-glucoside, vicenin-2, lonicerin, luteolin-7-O-glucoside, rhoifolin, luteolin, apigenin, daucosterol and β-sitosterol in P. multifida were simultaneously determined by QAMS with luteolin as the internal reference substance, and the alcohol-soluble extract, total ash and acid-insoluble ash were determined according to the detection method of Chinese Pharmacopoeia (2020 edition). The detection results were analyzed by chemical pattern recognition using SIMCA 14.1 software. On this basis, the quality difference of 18 batches of P. multifida was evaluated by the fusion model of weighted TOPSIS and grey correlation degree. Results The results of external standard method were close to those of QAMS method. The contents of pterosin B, pterosin Z, pterosin C 3-O-β-D-glucoside, vicenin-2, lonicerin, luteolin-7-O-glucoside, rhoifolin, luteolin, apigenin, daucosterol and β-sitosterol in 70% methanol extract of 18 batches of P. multifida from different habitats were significantly different, and the mass fractions were 0.214—0.342, 0.255—0.458, 0.119—0.332, 0.556—1.096, 0.048—0.146, 0.701—1.197, 1.460—2.553, 0.940—1.609, 0.085—0.217, 0.034—0.079 and 0.084—0.163 mg/g, respectively. The 18 batches of samples were divided into three categories by chemical pattern recognition, luteolin-7-O-glucoside, rhoifolin, pterosin C 3-O-β-D-glucoside, lonicerin, luteolin and apigenin were the main differential components affecting the quality of P. multifida. The comprehensive relative closeness of 18 batches of samples was 0.310 1—0.657 7, which proved that there were regional differences in P. multifida from different producing areas. Conclusion The quality difference of P. multifida in different producing areas was evaluated by multi-index quantification, chemical pattern recognition combined with weighted TOPSIS and grey relation analysis (GRA) fusion model, which provided a basis for the improvement of quality control level and resource development.

R286.2

江苏省中医药科技发展计划面上项目(MS2021008)