Untargeted Metabolomic Liquid Chromatography High-Resolution Mass Spectrometry Fingerprinting of Apple Cultivars for the Identification of Biomarkers Related to Resistance to Rosy Apple Aphid.
Rosa María Alonso-SalcesLuis A BerruetaBeatriz Abad-GarcíaAndrea Sasía-ArribaCarlos Asensio-RegaladoEnrique DapenaBlanca GalloPublished in: Journal of agricultural and food chemistry (2022)
Liquid chromatography high-resolution mass spectrometry fingerprinting together with pattern recognition techniques was used to determine the metabolites involved in the susceptibility of apple cultivars to rosy apple aphid (RAA). Preprocessing of ultra-high-performance liquid chromatography coupled to electrospray ionization and quadrupole time-of-flight mass spectrometry raw data of resistant and susceptible apple cultivars was carried out with XCMS and CAMERA packages. Univariate statistical tools and multivariate data analysis highlighted significant different profiles of the apple metabolomes according to their tolerance to RAA. Optimized and cross-validated Partial least squares discriminant analysis and orthogonal projections to latent structures discriminant analysis models confirmed trans -4-caffeoylquinic acid and 4- p -coumaroylquinic acid as biomarkers for the identification of resistant and susceptible apple cultivars to RAA and disclosed that only hydroxycinnamic acids are involved in the disease susceptibility of cultivars. In this sense, the final steps of the biosynthesis of caffeoylquinic acid (CQA) and p -coumaroylquinic acid ( p -CoQA) become decisive because the isomerization of 5-CQA to 4-CQA is favored in resistant cultivars, whereas the isomerization of 5- p -CoQA to 4- p -CoQA is favored in susceptible cultivars.
Keyphrases
- high resolution mass spectrometry
- liquid chromatography
- ultra high performance liquid chromatography
- tandem mass spectrometry
- mass spectrometry
- data analysis
- gas chromatography
- simultaneous determination
- high performance liquid chromatography
- solid phase extraction
- high resolution
- ms ms
- electronic health record
- bioinformatics analysis
- deep learning
- convolutional neural network