1H-NMR metabolomics: Profiling method for a rapid and efficient screening of transgenic plants

• Published in: African journal of biotechnology Vol. 11; no. 52; pp. 11386 - 11399
• Main Authors: Ben Akal-Ben Fatma, Yosra, Pianelli, Katia, Dieuaide Noubhani, Martine, Le Menn, Aline, Deborde, Catherine, Maucourt, Mickael M., Andrieu, Marie-Hélène, Rolin, Dominique D., Rothan, Christophe, Moing, Annick, Bouzid, Sadok, Cabasson, Cécile
• Format: Journal Article
• Language: English
• Published: Academic Journals 28.06.2012
• Subjects: Life Sciences; Vegetal Biology; GMO; Metabolome; 1H-NMR profiling; screening; tomato; fruit
• ISSN: 1684-5315; 1684-5315
• DOI: 10.5897/AJB12.088

Metabolomics-based approaches are methods of choice for studying changes in fruit composition induced by environmental or genetic modulation of biochemical pathways in the fruit. Owing to enzyme redundancy and high plasticity of the metabolic network, transgenic alteration of the activity of the enzymes from the central metabolism very often results in only slight modifications of the fruit composition. In order to avoid costly and time-consuming plant analysis, we used a fast and sensitive 1H-NMR-based metabolomic profiling technique allowing discovery of slight metabolite variations in a large number of samples. Here, we describe the screening of transgenic tomato plants in which two genes from the central metabolism, phosphoenolpyruvate carboxylase (EC.3.4.1.1) and malate synthase (EC 2.3.3.9) were silenced by antisens and RNAi strategy. 1H-NMR metabolomic profiles of methanol-d4 D2O buffer extracts of tomato fruit flesh were acquired and subjected to unsupervised multivariate statistical analysis. 1H-NMR spectra were binned into variable-size spectral domains, making it possible to get an overall analysis of a large number of resonances, even in the case of uncontrolled variation of the chemical shift. Principal component analysis was used to separate groups of samples and to relate known and unknown metabolites to transgenic events. The screening of 100 samples, from extraction to data mining, took 36 h. Thus, this procedure allows the rapid selection of metabolic phenotypes of interest among about 30 transgenic lines.