Research article

Unbiased characterization of genotype-dependent metabolic regulations by metabolomic approach in Arabidopsis thaliana

Miyako Kusano^{* 1} , Atsushi Fukushima^{* 1} , Masanori Arita^1,2 , Pär Jonsson³ , Thomas Moritz⁴ , Makoto Kobayashi¹ , Naomi Hayashi¹ , Takayuki Tohge^1,6 and Kazuki Saito^1,5

¹RIKEN Plant Science Center, 1-7-22, Yokohama, Kanagawa 230-0045, Japan

²Department of Computational Biology, Graduate School of Frontier Sciences, University of Tokyo, Kashiwa, Chiba 277-8561, Japan

³Group for Chemometrics, Organic Chemistry, Department of Chemistry, Umeå University, SE-901 87 Umeå, Sweden

⁴Umeå Plant Science Centre, Department of Forest Genetics and Plant Physiology, Swedish University of Agricultural Sciences, SE-901 87 Umeå, Sweden

⁵Department of Molecular Biology and Biotechnology, Graduate School of Pharmaceutical Sciences, Chiba University, Chiba 263-8522, Japan

⁶Max-Planck Institut für Molekulare Pflanzenphysiologie, Am Mühlenberg 1, 14476 Golm-Potsdam, Germany

author email corresponding author email* Contributed equally

BMC Systems Biology 2007, 1:53doi:10.1186/1752-0509-1-53

Published:	21 November 2007

Abstract

Background

Metabolites are not only the catalytic products of enzymatic reactions but also the active regulators or the ultimate phenotype of metabolic homeostasis in highly complex cellular processes. The modes of regulation at the metabolome level can be revealed by metabolic networks. We investigated the metabolic network between wild-type and 2 mutant (methionine-over accumulation 1 [mto1] and transparent testa4 [tt4]) plants regarding the alteration of metabolite accumulation in Arabidopsis thaliana.

Results

In the GC-TOF/MS analysis, we acquired quantitative information regarding over 170 metabolites, which has been analyzed by a novel score (ZMC, z-score of metabolite correlation) describing a characteristic metabolite in terms of correlation. Although the 2 mutants revealed no apparent morphological abnormalities, the overall correlation values in mto1 were much lower than those of the wild-type and tt4 plants, indicating the loss of overall network stability due to the uncontrolled accumulation of methionine. In the tt4 mutant, a new correlation between malate and sinapate was observed although the levels of malate, sinapate, and sinapoylmalate remain unchanged, suggesting an adaptive reconfiguration of the network. Gene-expression correlations presumably responsible for these metabolic networks were determined using the metabolite correlations as clues.

Conclusion

Two Arabidopsis mutants, mto1 and tt4, exhibited the following changes in entire metabolome networks: the overall loss of metabolic stability (mto1) or the generation of a metabolic network of a backup pathway for the lost physiological functions (tt4). The expansion of metabolite correlation to gene-expression correlation provides detailed insights into the systemic understanding of the plant cellular process regarding metabolome and transcriptome.