FTIR结合化学计量学对三七地下部位鉴别及皂苷含量预测
|
摘要
当今中药市场上掺假现象屡见不鲜,不良商贩利用三七须根粉末假冒主根和剪口粉末,严重影响三七的质量与药效。通过傅里叶变换红外光谱(FTIR)结合化学计量学建立三七主根、剪口和须根粉末鉴别及四种皂苷含量快速预测模型,为快速三七质量控制提供基础。采集三七主根、剪口和须根红外光谱,超高效液相色谱(UPLC)测量样品中三七皂苷R1、人参皂苷Rg1、人参皂苷Rb1和人参皂苷Rd含量。采用纵坐标归一化及二阶导数对原始红外光谱进行预处理;Kennard-stone算法将60个样本分为2/3训练集与1/3预测集。训练集数据结合支持向量机(SVM)判别建立三七主根、剪口和须根粉末鉴别模型,最佳核函数c和g采用交叉验证进行网格式搜索,预测集数据用于对判别模型进行外部验证。正交信号校正偏最小二乘回归(OSC-PLSR)建立三七中四种皂苷含量预测模型,红外光谱采用一阶、二阶导数及Savitsky-Golay平滑5点、7点、9点、11点预处理。60个样本分为2/3训练集与1/3预测集,训练集数据建立OSC-PLSR模型,预测集数据对OSC-PLSR模型的预测结果进行外部验证。结果显示:(1)二阶导数可有效的分离原始谱图的叠合隐蔽谱峰,并提高谱图的分辨率;(2)交叉验证网格式搜索计算出最佳核函数c=2.828 43,g=4.882 81×10~(-4),此时训练集判别正确率为100%;(3)SVM判别模型核函数设置为最佳核函数,预测集数据外部验证正确率为100%,所有样本均被正确鉴别;(4)三七皂苷R1、人参皂苷Rg1、人参皂苷Rb1和人参皂苷Rd最优含量预测模型预测值与UPLC检测值接近,预测效果良好。FTIR结合SVM判别能对三七主根、剪口和须根粉末快速鉴别,结合OSC-PLSR能对四种皂苷含量进行准确预测。该方法准确可靠,可为中药材三七提供快速有效的质量控制。
Phenomenon of adulterated traditional Chinese medicine(TCM)are still common in TCM market today.Unscrupulous traders used fibrous root powder pretending to be main root and rhizome powder of Panax notoginseng,and such behavior has serious influence on the quality and efficacy of Panax notoginseng.In this study,we have established a rapid method to discriminate the main root,rhizome and fibrous root powder and detect saponins content of Panax notoginsengin order to provide some research bases for rapid quality assessment of Panax notoginseng.A total of 60 Fourier transform infrared(FTIR)spectra of the main root,rhizome and fibrous root powder of Panax notoginseng were collected,and ultra-high performance liquid chromatography(UPLC)was used for measuring the content of notoginsenoside R1,ginsenoside Rg1,ginsenoside Rb1 and ginsenoside Rd of samples.The origin data of identify model were processed by ordinate normalization and second derivative,and 2/3of the 60 individuals were selected to form the calibration set by using Kennard-stone algorithm as well as the other 1/3were used as validation set.Calibration set data were used to establish the discriminant model of support vector machine(SVM)and the cross-validation was used for screening optimal parameters c and g,and validation set data were used to verify the results of SVM discriminant model for external validation.The origin data used to predict saponins content were calculated by first(1D)and second derivative(2D),Savitsky-Golay smoothing with five,seven,nine,and eleven points.2/3of the 60 individuals were selected to form the calibration set and the rest were used as validation set.The orthogonal signal correction-partial least squares regression(OSC-PLSR)model was established by calibration set and the validation set was utilized to verify the results of the model for external validation.Results showed that,(1)with second derivative processing,the overlapped peak of FTIR spectra were efficiently separated and the resolution of the spectra has been improved.(2)The optimal parameters c and gof support vector machine calculated by cross-validation were 2.828 43 and 4.882 81×10~(-4) respectively and the optimal accuracy rate of calibration set was 100%.(3)The parameter of support vector machine model was set as the optimal parameter and the accuracy rate of validation set was 100%,and all samples in validation set have been identified correctly.(4)The prediction content of greatest model of notoginsenoside R1,ginsenoside Rg1,ginsenoside Rb1 and ginsenoside Rd was close to the content measured by UPLC.The result indicated that,FTIR combined with support vector machine could effectively identify the main root,rhizome and fibrous root powder of Panax notoginseng.OSC-PLSR could accurately predict the content of four saponins of Panax notoginseng.In summary,the FTIR spectroscopy could provide a rapid and effective method for the quality control of Panax notoginseng.
Phenomenon of adulterated traditional Chinese medicine(TCM)are still common in TCM market today.Unscrupulous traders used fibrous root powder pretending to be main root and rhizome powder of Panax notoginseng,and such behavior has serious influence on the quality and efficacy of Panax notoginseng.In this study,we have established a rapid method to discriminate the main root,rhizome and fibrous root powder and detect saponins content of Panax notoginsengin order to provide some research bases for rapid quality assessment of Panax notoginseng.A total of 60 Fourier transform infrared(FTIR)spectra of the main root,rhizome and fibrous root powder of Panax notoginseng were collected,and ultra-high performance liquid chromatography(UPLC)was used for measuring the content of notoginsenoside R1,ginsenoside Rg1,ginsenoside Rb1 and ginsenoside Rd of samples.The origin data of identify model were processed by ordinate normalization and second derivative,and 2/3of the 60 individuals were selected to form the calibration set by using Kennard-stone algorithm as well as the other 1/3were used as validation set.Calibration set data were used to establish the discriminant model of support vector machine(SVM)and the cross-validation was used for screening optimal parameters c and g,and validation set data were used to verify the results of SVM discriminant model for external validation.The origin data used to predict saponins content were calculated by first(1D)and second derivative(2D),Savitsky-Golay smoothing with five,seven,nine,and eleven points.2/3of the 60 individuals were selected to form the calibration set and the rest were used as validation set.The orthogonal signal correction-partial least squares regression(OSC-PLSR)model was established by calibration set and the validation set was utilized to verify the results of the model for external validation.Results showed that,(1)with second derivative processing,the overlapped peak of FTIR spectra were efficiently separated and the resolution of the spectra has been improved.(2)The optimal parameters c and gof support vector machine calculated by cross-validation were 2.828 43 and 4.882 81×10~(-4) respectively and the optimal accuracy rate of calibration set was 100%.(3)The parameter of support vector machine model was set as the optimal parameter and the accuracy rate of validation set was 100%,and all samples in validation set have been identified correctly.(4)The prediction content of greatest model of notoginsenoside R1,ginsenoside Rg1,ginsenoside Rb1 and ginsenoside Rd was close to the content measured by UPLC.The result indicated that,FTIR combined with support vector machine could effectively identify the main root,rhizome and fibrous root powder of Panax notoginseng.OSC-PLSR could accurately predict the content of four saponins of Panax notoginseng.In summary,the FTIR spectroscopy could provide a rapid and effective method for the quality control of Panax notoginseng.
引文
[1] Xiao C,Wu M,Chen Y,et al.Phytochemical Analysis,2015,26(1):86.
[2] Le T H V,Lee G J,Vu H K L,et al.Chemical and Pharmaceutical Bulletin,2015,63(11):950.
[3] Pan Y,Zhang J,Shen T,et al.Biomedical Chromatography,2016,30(2):232.
[4] Uzayisenga R,Ayeka P A,Wang Y.Phytotherapy Research,2014,28(4):510.
[5] Wang Z,Chen Y,Pan H,et al.Chinese Herbal Medicines,2015,7(2):179.
[6] Yang C,Wang J,Zhao Y,et al.Journal of Ethnopharmacology,2010,130(2):231.
[7] Uzayisenga R,Ayeka P A,Wang Y.Phytotherapy Research,2014,28(4):510.
[8] Sun S,Wang C Z,Tong R,et al.Food Chemistry,2010,118(2):307.
[9] Sun S,Qi L W,Du G J,et al.Food Chemistry,2011,125(4):1299.
[10] Xu L,Liu J,Liu N,et al.Journal of Ethnopharmacology,2011,137(1):226.
[11] Fan J S,Liu D N,Huang G,et al.Journal of Ethnopharmacology,2012,142(3):732.
[12] Zhou N,Tang Y,Keep R F,et al.Phytomedicine,2014,21(10):1189.
[13] Ng T B.Journal of Pharmacy and Pharmacology,2006,58(8):1007.
[14] Chinese Pharmacopoeia Commission(中华人民共和国药典委员会).Pharmacopoeia of the People’s Republic of China,Part One(中华人民共和国药典一部).Beijing:China Medical Science Press(北京:中国医药科技出版社),2015.
[15] Wan J B,Li P,Yang R L,et al.Journal of Liquid Chromatography&Related Technologies,2013,36(3):406.
[16] Wang J R,Yau L F,Gao W N,et al.Journal of Agricultural and Food Chemistry,2014,62(36):9024.
[17] Jia X H,Wang C Q,Liu J H,et al.Journal of Natural Medicines,2013,67(2):339.
[18] Cheng C,Liu J,Zhang C,et al.Applied Spectroscopy Reviews,2010,45(2):148.
[19] Luo C,He T,Chun Z.Journal of Molecular Structure,2013,1037:40.
[20] Li B,Wei Y,Duan H,et al.Vibrational Spectroscopy,2012,62:17.
[21] Kyomugasho C,Christiaens S,Shpigelman A,et al.Food Chemistry,2015,176:82.
[22] Luo Q,Yun Y,Fan W,et al.RSC Advances,2015,5(7):5046.
[23] Li D,Jin Z,Zhou Q,et al.Journal of Molecular Structure,2010,974(1):68.
[24] Shen F,Yang D,Ying Y,et al.Food and Bioprocess Technology,2012,5(2):786.
[25] Chang C C,Lin C J.ACM Transactions on Intelligent Systems and Technology(TIST),2011,2(3):27.
[2] Le T H V,Lee G J,Vu H K L,et al.Chemical and Pharmaceutical Bulletin,2015,63(11):950.
[3] Pan Y,Zhang J,Shen T,et al.Biomedical Chromatography,2016,30(2):232.
[4] Uzayisenga R,Ayeka P A,Wang Y.Phytotherapy Research,2014,28(4):510.
[5] Wang Z,Chen Y,Pan H,et al.Chinese Herbal Medicines,2015,7(2):179.
[6] Yang C,Wang J,Zhao Y,et al.Journal of Ethnopharmacology,2010,130(2):231.
[7] Uzayisenga R,Ayeka P A,Wang Y.Phytotherapy Research,2014,28(4):510.
[8] Sun S,Wang C Z,Tong R,et al.Food Chemistry,2010,118(2):307.
[9] Sun S,Qi L W,Du G J,et al.Food Chemistry,2011,125(4):1299.
[10] Xu L,Liu J,Liu N,et al.Journal of Ethnopharmacology,2011,137(1):226.
[11] Fan J S,Liu D N,Huang G,et al.Journal of Ethnopharmacology,2012,142(3):732.
[12] Zhou N,Tang Y,Keep R F,et al.Phytomedicine,2014,21(10):1189.
[13] Ng T B.Journal of Pharmacy and Pharmacology,2006,58(8):1007.
[14] Chinese Pharmacopoeia Commission(中华人民共和国药典委员会).Pharmacopoeia of the People’s Republic of China,Part One(中华人民共和国药典一部).Beijing:China Medical Science Press(北京:中国医药科技出版社),2015.
[15] Wan J B,Li P,Yang R L,et al.Journal of Liquid Chromatography&Related Technologies,2013,36(3):406.
[16] Wang J R,Yau L F,Gao W N,et al.Journal of Agricultural and Food Chemistry,2014,62(36):9024.
[17] Jia X H,Wang C Q,Liu J H,et al.Journal of Natural Medicines,2013,67(2):339.
[18] Cheng C,Liu J,Zhang C,et al.Applied Spectroscopy Reviews,2010,45(2):148.
[19] Luo C,He T,Chun Z.Journal of Molecular Structure,2013,1037:40.
[20] Li B,Wei Y,Duan H,et al.Vibrational Spectroscopy,2012,62:17.
[21] Kyomugasho C,Christiaens S,Shpigelman A,et al.Food Chemistry,2015,176:82.
[22] Luo Q,Yun Y,Fan W,et al.RSC Advances,2015,5(7):5046.
[23] Li D,Jin Z,Zhou Q,et al.Journal of Molecular Structure,2010,974(1):68.
[24] Shen F,Yang D,Ying Y,et al.Food and Bioprocess Technology,2012,5(2):786.
[25] Chang C C,Lin C J.ACM Transactions on Intelligent Systems and Technology(TIST),2011,2(3):27.