Manual annotation combined with untargeted metabolomics for chemical characterization and discrimination of two major crataegus species based on liquid chromatography quadrupole time-of-flight mass spectrometry

https://doi.org/10.1016/j.chroma.2019.460628Get rights and content

Highlights

  • A total of 78 compounds were characterized in two hawthorn species.

  • Two hawthorn species were successfully distinguished by untarget metabolomics.

  • 47 differential compounds and 17 false positive ions were exposited and discussed.

  • Most compounds were higher in crataegus pinnatifida bunge except triterpenoids.

  • Five chemical markers were obtained to perform a powerful prediction model.

Abstract

Hawthorn is a popular functional food. In China, Crataegus pinnatifida Bunge. and C. pinnatifida Bge. var. major N. E. Brown are two major species that are used for the preparation of hawthorn products. Accordingly, it is crucial to explore the chemical differences between these two species for the market standardization of hawthorn products. In this study, we integrated manual annotation with untargeted metabolomics based on ultra-high performance liquid chromatography coupled with quadrupole time-of-flight tandem mass spectrometry for compound characterization and discrimination of the two hawthorn species. We characterized 78 compounds in the two species including saccharides, glycosides, organic acids, phenols, flavonoids and triterpenoids. Moreover, 47 differential compounds and 17 false positive ions were recognized and fully reviewed. The 47 identified compounds were then used to build a partial least squares discriminant analysis model that successfully discriminated C. pinnatifida Bge. and C. pinnatifida Bge. var. major N.E.Br.

Introduction

Hawthorn, the trivial name of the genus Crataegus, has been reported to encompass over 200 species that are widely distributed in the northern hemisphere, mostly in China, Europe, and North America [1], [2]. Hawthorn is considered to be a functional food with many health benefits, and its pharmacological actions include antioxidative, anti-atherosclerosis, and lipid-lowering effects [3], [4], [5]. In China, hawthorn products are prepared from two major species, namely C. pinnatifida Bge. and C. pinnatifida Bge. var. major N.E.Br. [6], [7]. Morphologically, these two plants are similar, with slight differences in leaf sinus, pericarp color, and fruit size [8]. Proanthocyanidins (procyanidin B2, procyanidin C) and pentacyclic triterpenic acids (maslinic acid, oleanolic acid) are two major classes of compounds found in both species [9], [10], [11]. However, a smattering of publications are concerned with the chemical differences of between the two species, which is important for proposals regarding the market standardization of hawthorn products.

The advent of mass spectrometry (MS) has allowed for the determination of more chemical information from plants. MS provides a high sensitivity and resolution capable of distinguishing between two similar but different chemical entities [12], [13]. Accordingly, MS fingerprint and untargeted metabolomics have been proposed as effective methods for species authentication and discrimination. Ultra-high performance liquid chromatography (UHPLC) coupled with high-resolution MS has been widely used for such applications [14], [15]. MS fingerprint depends on manual annotation and quantitation of major compounds, which mainly focus on the visible peaks in the total ion chromatogram [16]. In contrast, metabolomics extracts all chemical components using an automatic algorithm and then determines the differential chemical markers based on multivariate statistical analysis [17]. Metabolomics has been widely applied in different fields such as mechanism study [18], [19], disease diagnosis and prediction [20], [21], food classification [22], [23], and plant species authentication [24], [25]. Despite their strengths, these two methods also have some limitations. Manual annotation relies on reference substances to identify compounds and may neglect trace or differential compounds. On the other hand, untargeted metabolomics can uncover all compound information, but unaccountable features can result in a small number of valuable differential chemical entities. Recognition of these false positive ions is a bottleneck in species discrimination.

Here, we performed both manual annotation and untargeted metabolomics on two hawthorn species using UHPLC-quadrupole time-of-flight (QTOF) MS. We found that the two methods were complimentary and together contributed to reliable compound characterization. Forty-seven differential compounds and 17 false positive ions were fully examined. The differential compounds were subsequently used to build a partial least squares discriminant analysis (PLS-DA) model, which achieved the successful discrimination of C. pinnatifida Bge. and C. pinnatifida Bge. var. major N.E.Br. Our data provides experimental evidence to support the market standardization of hawthorn products.

Section snippets

Chemicals and reagents

HPLC-grade acetonitrile and formic acid were purchased from ROE (Newark, New Castle, DE, USA), and HPLC-grade methanol was purchased from Jiangsu Hanbon Sci. & Tech. (Nanjing, China). Deionized water (18.2 MΩ cm−1) was prepared from a Milli-Q water purification system (Millipore, Bedford, MA, USA).

Standards of epicatechin, quinic acid, procyanidin B1, chlorogenic acid, procyanidin B2, rutin, vitexin, quercetin, maslinic acid, ursolic acid, idaein, hyperoside, isoquercitrin, and dihydromyricetin

Results and discussion

This study aimed to characterize the chemical components of C. pinnatifida Bge. and C. pinnatifida Bge. var. major N. E. Br., and to explore the differences in their components to discriminate the two species. The work consisted of both manual annotation and metabolomic profiling followed by multivariate statistical analysis. The workflow is shown in Fig. 1, and the detailed process is described in the following sections:

Conclusion

Untargeted metabolomics has been demonstrated as a powerful tool for discrimination, but the approach still has some limitations. Our study involved performing manual annotation followed by untargeted metabolomics to explore the false positive features in metabolomics. We focused on 47 differential compounds to build the prediction model. Our results indicated the successful discrimination of C. pinnatifida Bge. and C. pinnatifida Bge. var. major N. E. Br. with good predictive validation.

Declaration of Competing Interest

The authors declare that they have no known competing financial interests or personal relationships that could have appeared to influence the work reported in this paper.

Acknowledgements

This study was supported in part by the National Natural Science Foundation of China (Nos. 81722048, 81872998), and "Double First-Class" University project (CPU2018GY09). We sincerely thank our laboratory technician Hui-Ying Wang, State Key Laboratory of Natural Medicines, China Pharmaceutical University, for the assistance of the instrument administration and management in the study.

References (31)

Cited by (22)

  • Anti-inflammatory/anti-oxidant properties and the UPLC-QTOF/MS-based metabolomics discrimination of three yellow camellia species

    2022, Food Research International
    Citation Excerpt :

    Currently, chemical marker-based methods such as chromatographic fingerprint and untargeted metabolomics analysis have been commonly used for discovery of components for authenticating and discriminating herbal medicines (An et al., 2021; Liu et al.,2020; Pérez-Jiménez et al., 2021; Zhang et al., 2019). Providing an unsurpassed resolution and high sensitivity, MS fingerprint is becoming a preferred analytical method for authenticating closely-related species with similar but different chemical entities, while untargeted metabolomics coupled with multi-variate analysis could achieve simultaneous detection of near complete secondary metabolites and reveal an unbiased abundant metabolic profile (Li et al., 2020). Despite the strength, the selection of makers by either of the above method alone usually lacks sufficient chemical evidence due to non-ignorable limitations of each.

  • A comprehensive strategy integrating metabolomics with multiple chemometric for discovery of function related active markers for assessment of foodstuffs: A case of hawthorn (Crataegus cuneata) fruits

    2022, Food Chemistry
    Citation Excerpt :

    In this work, 43 chemical components were authenticated from mass spectrometry data in negative and positive ion modes (Fig. S4 in supplementary materials), great part of which belong to organic acids and flavonoids. Among them, fifteen compounds, including D-(-)-quinic acid, malic acid, gallic acid, protocatechuic acid, protocatechualdehyde, caffeic acid, procyanidin C1, orientin, rutin, hyperoside, isoquercitrin, quercitrin, luteolin, quercetin, (+)-balanophonin were identified by comparing with standard substances, and the other compounds were inferred by comparing with previous literature reports (Table S3 in supplementary materials) (Hu, et al., 2020; Li, et al., 2020; Li, et al., 2018; Qiao, et al., 2014; Zhou, 2020). Most compounds have excellent response in negative ion mode and have a wide range of bioactivity.

  • Liquid chromatographic study of two structural isomeric pentacyclic triterpenes on reversed-phase stationary phase with hydroxypropyl-β-cyclodextrin as mobile phase additive

    2022, Journal of Pharmaceutical and Biomedical Analysis
    Citation Excerpt :

    leaves, mainly including euscaphic acid, tormentic acid, maslinic acid, corosolic acid, oleanolic acid, ursolic acid and pomolic acid [1,2]. Numerous papers considering pharmacologic activity, analysis and separation of these compounds are available [3–6]. Analytical separation of maslinic acid and corosolic acid (Fig. 1) by high performance liquid chromatography has been reported [6,7].

View all citing articles on Scopus
1

These authors contributed equally to this work.

View full text