1H NMR, GC-EI-TOFMS, and data set correlation for fruit metabolomics: Application to spatial metabolite analysis in melon

A metabolomics approach combining 1H NMR and gas chromatography- electrospray ionization time-of-flight mass spectrometry (GC-EI-TOFMS) profiling was employed to characterize melon (Cucumis meloL.) fruit. In a first step, quantitative 1H NMR of polar extracts and principal component analyses (PCA) of the corresponding data highlighted the major metabolites in fruit flesh, including sugars, organic acids, and amino acids. In a second step, the spatial localization of metabolites was investigated using both analytical techniques. Direct 1H NMR profiling of juice or GC-EI-TOFMS profiling of tissue extracts collected from different locations in the fruit flesh provided information on advantages and drawbacks of each technique for the analysis of a sugar-rich matrix such as fruit. 1H NMR and GC-EI-TOFMS data sets were compared using independently performed PCA and multiblock hierarchical PCA (HPCA), respectively. In addition a correlation-based multiblock HPCA was used for direct comparison of both analytical data sets. These data analyses revealed several gradients of metabolites in fruit flesh which can be related with differences in metabolism and indicated the suitability of multiblock HPCA for correlation of data from two (or potentially more) metabolomics platforms. © 2009 American Chemical Society.