High-Throughput Time Series Metabolomic Analysis of a Systematically Perturbed Plant System

Abstract

In the post-genomic era, availability of high-throughput profiling techniques enabled the measurement of entire cellular molecular fingerprints. Major characteristics of the high-throughput revolution were that (a) studying biological problems did not have to rely on prior hypotheses, while (b) parallel occurring phenomena, previously assumed disconnected, could now be simultaneously observed. Metabolomics is the newest of the "omics" techniques. It enables the quantification of hundreds of free metabolite pools, providing a metabolic fingerprint. Considering the importance of cellular metabolism, which is the net effect of changes at the genomic, transcriptomic and proteomic levels and of the cell with its environment, the metabolomic profile, is a fundamental determinant of cellular physiology.

Obtaining accurate and complete metabolomic profiles is thus of great importance. However, being recent technology, metabolomics is currently at its standardization phase. As part of my PhD thesis research, I focused on addressing several current challenges in metabolomics technology development. Specifically a novel data correction, validation and normalization strategy for gas chromatography-mass spectrometry (GC-MS) metabolomic profiling analysis was developed, which dramatically increased the accuracy and reliability of GC-MS metabolomic profiles. The optimized metabolomics protocol was applied to study the short-term dynamic response of systematically perturbed <i>Arabidopsis thaliana</i> liquid culture system to study regulation of its primary metabolism. The biological system was studied under conditions of elevated CO₂ stress, salt (NaCl) stress, sugar (trehalose) signal, and hormone (ethylene) signal, applied individually; the latter three stresses also applied in combination with the CO₂ stress. Analysis of the obtained results required the appropriate application of multivariate statistical analysis techniques, which are developed mainly in transcriptomic analysis, into metabolomics analysis for the first time.

The acquired results identified important new regulatory information about the biological systems resulting in new targets for metabolic engineering of plants. The large number of dynamic perturbation allowed re-construction of metabolic networks to identify possible novel metabolic pathways based on correlations between metabolic profiles. In addition, it demonstrates the advantages of dynamic, multiple-stress "omic" analysis for the elucidation of plant systems function. In this sense, it contributes in further advancing the computational and experimental metabolic engineering and systems biology toolbox.