液相色谱-四极杆飞行时间质谱结合化学计量学方法筛查饲料中的未知药物

Screening on Non-targeted Comporcents in Feed Using High Performance Liquid Chromatography Time-of-Flight Mass Spectrometry Combined with Chemometrics

建立了液相色谱-四极杆飞行时间质谱（LC-Q-TOF/MS）对饲料中非目标污染物（兽药）的筛查方法,以及未知物的判定和确认技术。饲料样品经0.1%EDTA-乙腈-甲醇（1∶1∶1,含0.1%甲酸）混合溶液提取,Q-TOF全扫描模式采集数据,数据经过峰对齐（质量误差5 m Da以内）、噪音过滤、自动排除无用峰等处理后,进行正交偏最小二乘判别分析（OPLS-DA）处理,根据Score图、S-Plot图和VIP图判定出特征离子,结合二级信息与自建谱库比对,进行结构式确认,或通过同位素丰度比例和分子式生成功能,确认分子式,结合二级断裂信息,判定特征离子的结构式,从而判定未知物的成分。结果显示：16份饲料样品共筛查出27种兽药。采用液相色谱-三重四极杆串联质谱法对其中的12种兽药进行定性定量分析,验证结果仅出现2个假阳性,兽药含量在12.5~152.0 mg/kg之间。通过LC-Q-TOF/MS结合OPLS-DA的方法能够对饲料中的非目标物进行筛查分析,方法准确度高。

To screen feed for new or unexpected contaminants at the ppm level, a generic method was developed. The sample was extracted with 0. 1% EDTA - acetonitrile - methanol （ 1 ： 1 ： 1, containing 0. 1% formic acid） , and analyzed by high-performance liquid chromatography coupled to electrospray ionization time - of - flight mass spectrometry （ HPLC - Q - TOF/MS） in positive ioniza- tion mode. The data were collected after peak alignment （mass error within 50 mDa） , noise filtra- tion, elimination of useless peaks and statistical evaluation was performed based on Orthogonal Partial Least Squares Discriminant Analysis （ OPLS - DA） . According to the score plot, S - plot and VIP, blind statistical evaluation revealed that feed samples could be separated regarding unknown compo- nents by OPLS - DA and the m/z - retention time pair. With the assistance of MS/MS data, isotope abundance ratio, formula generation and self-buih spectral library, the structures of unknown com- pounds were confirmed. The results showed that 27 veterinary drugs were screened out from 16 feed samples. By quantitative analysis of 12 existing veterinary drugs using liquid chromatography tandem mass spectrometry（ LC -MS/MS） , only two false positives appeared in the verification results, and the concentration levels of 12 veterinary drugs were between 12.5 mg/kg and 152.0 mg/kg. From the perspective of the results, TOF - MS data and OPLS - DA could be applied in the non-targeted screening of veterinary drugs in feed samples with high accuracy.