Untargeted metabolomics analysis based on LC-IM-QTOF-MS for discriminating geographical origin and vintage of Chinese red wine

Highlights

• An untargeted metabolomics approach for vintage and origin discrimination of Chinese red wine.

• MS-based untargeted metabolomics was firstly applied to identify Chinese red wine vintages.

• 42 and 48 metabolites in ESI+ and ESI− modes were identified as vintage markers.

• 37 and 35 metabolites in ESI+ and ESI− modes were identified as origin markers.

• Reliable OPLS-DA models using markers were built to predict wine vintages and origins.

Abstract

Identifying wine geographical origin and vintage is vital due to the abundance of fraudulent activity associated with wine mislabeling of region and vintage. In this study, an untargeted metabolomic approach based on liquid chromatography/ion mobility quadrupole time-of-flight mass spectrometry (LC-IM-QTOF-MS) was used to discriminate wine geographical origin and vintage. Wines were well discriminated according to region and vintage with orthogonal partial least squares-discriminant analysis (OPLS-DA). The differential metabolites subsequently were screened by OPLS-DA with pairwise modeling. 42 and 48 compounds in positive and negative ionization modes were screened as differential metabolites for the discrimination of different wine regions, and 37 and 35 compounds were screened for wine vintage. Furthermore, new OPLS-DA models were performed using these compounds, and the external verification trial showed excellent practicality with an accuracy over 84.2%. This study indicated that LC-IM-QTOF-MS-based untargeted metabolomics was a feasible tool for wine geographical origin and vintage discrimination.

Introduction

Wine is a complex mixture that consists of thousands of chemical compositions (Sherman et al., 2020). This complexity means that wine is susceptible to adulteration. Especially taking into consideration its economic importance, the abundance of economically interest-driven frauds associated with mislabeling of vintage and geographical origins to counterfeit high-value wines, has become of great concern among the wine industry, researchers, and consumers. Therefore, many studies have been focused on the urgent issue of wine authentication (Pan et al., 2022, Phan and Tomasino, 2021, Ranaweera et al., 2021, Solovyev et al., 2021).

Metabolomics, also generally regarded as “systems biology”, seeks an analytical description of complex biological samples, aims to characterize all the small molecules in such a sample and focuses on the object as a whole for analysis (Nicholson & Lindon, 2008). Up to now, metabolomics has been successfully applied in food, agriculture, medicine and other fields (Oyedeji et al., 2021, Pinu, 2015, Reher et al., 2022, Wurtzel and Kutchan, 2016). The term “wine-omics”, a novel analysis of wine data based on metabolomics, was first proposed by Wohlgemuth (2008) in the journal Nature. With the continuous development of metabolomics, wine metabolomics combined with powerful statistical techniques has been widely applied in wine authentication.

In general, metabolomics can be probably divided into two categories: targeted and untargeted analysis approaches. The targeted analysis is used for the detection or quantification of specific compounds, such as polyphenols, organic acids, carbohydrates, amino acids, mineral elements, aromatic substances, etc (Sun et al., 2021). Combining these characteristic constituents with multivariate statistical techniques proved to be a potential tool for wine authentication. For example, LC–MS determination of 37 anthocyanin derivatives was performed by Zhang et al. (2021) for wine classification according to vintages and aging stages. Wu et al. (2021) verified the geographical origins of red wines imported into China using 16 mineral elements. Nevertheless, it has to be noted that targeted metabolomic approaches are not preferable in food authentication unless the suspected targets are specific markers as the targeted metabolomic method detects only limited compounds at a time (Ballin & Laursen, 2019).

On the contrary, untargeted analyses are therefore gaining territory in food authentication, especially in highly complex authentication issues such as vintage or geographic origin discrimination (Ballin and Laursen, 2019, McGrath et al., 2018). Untargeted analyses, also known as fingerprinting, can simultaneously detect numerous unspecified targets, providing high sensitivity and good resolution. In the last decade, untargeted metabolomic techniques have been widely applied in food authentication (Li et al., 2021, Rubert et al., 2014, Springer et al., 2014). In the early days of metabolomics development, nuclear magnetic resonance (NMR) was the most-used technique for the detection of metabolites in foods (Cassino et al., 2019, Consonni and Cagliani, 2019). For example, a previous study reported that the authenticity, the grape variety, the geographical origin, and the year of the vintage of wines produced in Germany were successfully distinguished by ¹H NMR spectroscopy in combination with multivariate data analysis (Mannu et al., 2020, Mannu et al., 2020). In recent years, with the continual development and spread of mass spectrometry (MS), it gradually substituted the use of NMR (Zhou et al., 2019). On the one hand, MS is more suitable for combination with a separation technique such as Liquid chromatography/gas chromatography. On the other hand, for most researchers, developing or improving detection methods matching high-resolution MS (HRMS) equipment is feasible and affordable, which can greatly enhance the capabilities for the identification of unknown metabolites (Lacalle-Bergeron et al., 2021). Therefore, based on the aforementioned facts, the MS-based untargeted metabolomic analysis has been regarded as an attractive method for the discriminative, predictive, informative objectives involving food quality, safety, and authentication (Cavanna et al., 2018, Hu et al., 2022, Sherman et al., 2020, Zhang et al., 2015). However, the information regarding the application of LC-MS-based untargeted metabolomic analysis approaches in the identification and analysis of differential metabolites in Chinese red wines (i) from different geographical origins in the present literature are far from adequate (Pan et al., 2022), (ii) from different vintages have yet to be reported.

Therefore, this study aimed to develop a suitable approach for screening differential metabolites and predicting the vintages and regions of Chinese red wines by combining the untargeted metabolomics based on LC-IM-QTOF-MS with chemometrics. For this purpose, untargeted metabolomic fingerprints of 114 Chinese red wine samples from four famous geographical origins in China with vintages ranging from 2011 to 2019 were investigated. The differences in metabolites among different wine sample groups were revealed by multivariate chemometric techniques. Principal component analysis (PCA) models were initially performed to roughly discriminate wine samples from different origins and vintages. OPLS-DA models were subsequently constructed to screen the differential metabolites. Based on these differential metabolites, effective OPLS-DA models were re-constructed and used successfully for predicting the wine sample geographical origins and vintages. This work indicated that LC-IM-QTOF-MS-based untargeted metabolomics in combination with chemometrics showed strong potential for screening reliable markers and predicting the Chinese red wine geographical origins and vintages.