Characterization of the natural variation in Arabidopsis thaliana metabolome by the analysis of metabolic distance

Abstract

Metabolite fingerprinting is widely used to unravel the chemical characteristics of biological samples. Multivariate data analysis and other statistical tools are subsequently used to analyze and visualize the plasticity of the metabolome and/or the relationship between those samples. However, there are limitations to these approaches for example because of the multi-dimensionality of the data that makes interpretation of the data obtained from untargeted analysis almost impossible for an average human being. These limitations make the biological information that is of prime importance in untargeted studies be partially exploited. Even in the case of full exploitation, current methods for relationship elucidation focus mainly on between groups variation and differences. Therefore, a measure that is capable of exploiting both between- and within-group biological variation would be of great value. Here, we examined the natural variation in the metabolome of nine Arabidopsis thaliana accessions grown under various environmental conditions and established a measure for the metabolic distance between accessions and across environments. This data analysis approach shows that there is just a minor correlation between genetic and metabolic diversity of the nine accessions. On the other hand, it delivers so far in Arabidopsis unexplored chemical information and is shown to be biologically relevant for resistance studies.Electronic supplementary materialThe online version of this article (doi:10.1007/s11306-011-0375-3) contains supplementary material, which is available to authorized users.

Discovering variation of secondary metabolite diversity and its relationship with disease resistance inCornus floridaL.

“…To compare metabolic distances between subpopulations, we constructed an ANOSIM (Analysis of Similarities) R matrix from intersample Euclidean distances following the approach of Houshyani et al. (2012) and Kabouw, Biere, van der Putten, and van Dam (2009), 377 log‐transformed metabolites, and the program PAST (Hammer, Harper, & Ryan, 2001). An ANOSIM matrix is a reduced‐dimension matrix describing the similarity between pairs of subpopulations based on differences in abundances of multiple metabolites.…”

“…Promising work has investigated broader patterns of natural metabolome variation in the context of natural genetic variation, but most analyses of variation in chemical diversity focus on distance‐based measures versus explicit measurement of chemical richness. For example, significant correlations between metabolic and genetic distances were detected in nine Arabidopsis thaliana accessions exposed to different environments (Houshyani et al., 2012). In a second example, multigenerational lines inbred from different Drosophila melanogaster populations were found to remain distinguishable in general lipid composition, and approximately one‐fifth of the lipid compounds had clear concentration differences between male and female genotypes (Scheitz, Guo, Early, Harshman, & Clark, 2013).…”

Arabidopsis myrosinases link the glucosinolate-myrosinase system and the cuticle

“…The in-house script, MetAlign Output Transformer59 was used for filtering out low and irreproducible signals and noise value imputation. The resulting mass peak lists were subjected to MSClust60 to group the mass signals originating from the same compound based on their similar retention time and intensity patterns across samples.…”

Three-step pathway engineering results in more incidence rate and higher emission of nerolidol and improved attraction of Diadegma semiclausum

“…Detrended Correspondence Analysis (DCA) was used to check the length of the gradient (L) and due to an L < 4, the linear ordination techniques, principal component analysis (PCA) and redundancy analysis (RDA) were selected to visualize variation across chromatograms and correlations between metabolites using CANOCO (Smilauer, 2003). The procedure of Houshyani et al (2011) was followed for putative identification of selected LC-MS metabolites.…”