Introduction

LC–MS based untargeted metabolomics are the main untargeted methods used for juice metabolomics to solve the authentication problem faced in fruit juice industry.

Objectives

To evaluate the performances of different untargeted metabolomics methods on fruit juices metabolomics and authentication, orange and apple fruit juices were selected for this study.

Methods

IDA-MS and SWATH-MS based on UHPLC-QTOF were used for the metabolomics and authenticity determination of apple and orange juices, including the lab-made samples of oranges (Citrus sinensis Osb.) from Jiangxi Province, apples (Malus domestica Borkh) from Shandong Province, and different brands of commercial orange and apple juice samples from markets.

Results

IDA-MS and SWATH-MS could both acquire numerous MS1 features and MS2 information of juice components, while SWATH-MS excels at the acquisition rate of MS2. Distinctive separation between authentic orange juice and not authentic orange juice could be seen from principal component analysis and hierarchical clustering analysis based on both IDA-MS and SWATH-MS. After analysis of variance, fold change analysis and orthogonal projection to latent structures discriminant mode, 53 and 46 potential markers were defined by IDA-MS and SWATH-MS (with 77.4% and 100% MS2 acquisition rate) separately. Subsequently, these potential markers were putatively annotated using general chemical databases with 6 more annotated by SWATH-MS. Furthermore, 7 of the potential markers, l-asparagine, umbelliferone, glucosamine, phlorin, epicatechin, phytosphingosine and chlorogenic acid, were identified with standards. For the consideration of model simplicity, two determined makers (umbelliferone and chlorogenic acid) were selected to construct the DD-SIMCA model in commercial samples because of their good correlation with apple adulteration proportion, and the sensitivity and specificity of the model were 100% and 95%.

Conclusion

SWATH-MS excels at the MS2 acquisition of juice components and potential markers. This study provides an overall performance comparison between IDA-MS and SWATH-MS, and guidance for the method selection on fruit juice metabolomics and juice authenticity determination. Two of the potential markers determined, umbelliferone and chlorogenic acid, could be used as apple juice indicators in orange juice.

中文翻译：

---

使用信息相关的采集质谱图和对所有理论碎片离子质谱图进行连续窗口采集的果汁代谢组学和验证方法。

介绍

基于LC–MS的非靶向代谢组学是用于果汁代谢组学解决果汁行业面临的认证问题的主要非靶向方法。

目标

为了评估不同的非靶向代谢组学方法对果汁代谢组学和认证的性能，本研究选择了橙汁和苹果汁。

方法

基于UHPLC-QTOF的IDA-MS和SWATH-MS用于苹果汁和橙汁的代谢组学和真实性测定，包括江西省实验室制作的橙子（Citrus sinensis Osb。），苹果（Malus domestica Borkh）来自山东省，以及来自市场的不同品牌的商业橙汁和苹果汁样品。

结果

IDA-MS和SWATH-MS都可以获取许多MS1功能和果汁成分的MS2信息，而SWATH-MS在MS2的获取率方面表现出色。从基于IDA-MS和SWATH-MS的主成分分析和层次聚类分析可以看出，纯正的橙汁和非纯正的橙汁之间有明显的区别。经过方差分析，倍数变化分析和对潜在结构判别模式的正交投影，分别通过IDA-MS和SWATH-MS定义了53个和46个潜在标记（MS2采集率分别为77.4％和100％）。随后，使用通用化学数据库推定了这些潜在的标记，另6个通过SWATH-MS进行了标记。此外，还有7种潜在的标志物，即l-天冬酰胺，伞形酮，氨基葡萄糖，phlorin，表儿茶素，用标准品鉴定了植物鞘氨醇和绿原酸。考虑到模型的简单性，选择了两个确定的制造商（伞形酮和绿原酸）来构建商业样品中的DD-SIMCA模型，因为它们与苹果掺假比例具有良好的相关性，该模型的敏感性和特异性分别为100％和95％。

结论

SWATH-MS在获得果汁成分和潜在标记物的MS2方面表现出色。这项研究提供了IDA-MS和SWATH-MS之间的总体性能比较，并为果汁代谢组学和果汁真实性确定方法的选择提供了指导。确定的两个潜在标志物伞形酮和绿原酸可用作橙汁中的苹果汁指示剂。