Fruit metabolite networks in engineered and non-engineered tomato genotypes reveal fluidity in a hormone and agroecosystem specific manner

INTRODUCTION: Metabolomics provides a view of endogenous metabolic patterns not only during plant growth, development and senescence but also in response to genetic events, environment and disease. The effects of the field environment on plant hormone-specific metabolite profiles are largely unknown. Few studies have analyzed useful phenotypes generated by introducing single or multiple gene events alongside the non-engineered wild type control at field scale to determine the robustness of the genetic trait and its modulation in the metabolome as a function of specific agroecosystem environments. OBJECTIVES: We evaluated the influence of genetic background (high polyamine lines; low methyl jasmonate line; low ethylene line; and isogenic genotypes carrying double transgenic events) and environments (hairy vetch, rye, plastic black mulch and bare soil mulching systems) on the metabolomic profile of isogenic reverse genetic mutations and selected mulch based cropping systems in tomato fruit. Net photosynthesis and fruit yield were also determined. METHODS: NMR spectroscopy was used for quantifying metabolites that are central to primary metabolism. We analyzed both the first moment (means) of metabolic response to genotypes and agroecosystems by traditional univariate/multivariate methods, and the second moment (covariances) of responses by creating networks that depicted changes in correlations of paired metabolites. This particular approach is novel and was necessary because our experimental material yielded highly variable metabolic responses that could not be easily understood using the traditional analytical approaches for first moment statistics. RESULTS: High endogenous spermidine and spermine content exhibited strong effects on amino acids, Krebs cycle intermediates and energy molecules (ADP + ATP) in ripening fruits of plants grown under different agroecosystem environments. The metabolic response to high polyamine genotypes was similar to the response to hairy vetch cover crop mulch; supported by the pattern of changes in correlation between metabolites. Changes in primary metabolites of genotypes mutated for the deficiency of ethylene or methyl jasmonate were unique under all growth conditions and opposite of high polyamine genotype results. The high polyamine trait was found to dominate the low ethylene and low jasmonate mutations under field conditions. For several metabolites low ethylene and low methyl jasmonate genotypes had an inverse relationship. Collectively, these results affirm that interactions between metabolite pathways and growth environments are affected by genotype, and influence the metabolite quality of a crop. CONCLUSION: This study portrays how metabolite relationships change, both in mean and in correlation, under different genotypic and environmental conditions. Although these networks are surprisingly dynamic, we also find examples of selectively conserved associations. ELECTRONIC SUPPLEMENTARY MATERIAL: The online version of this article (doi:10.1007/s11306-016-1037-2) contains supplementary material, which is available to authorized users.