|Lange, B. Markus|
Submitted to: Archives Of Biochemistry and Biophysics
Publication Type: Peer Reviewed Journal
Publication Acceptance Date: 10/24/2005
Publication Date: 4/15/2006
Citation: Ghassemian, M., Lutes, J., Tepperman, J.M., Chang, H., Zhu, T., Wang, X., Quail, P.H., Lange, B. 2006. Integrative analysis of transcript and metabolite profiling data sets to evaluate the regulation of biochemical pathways during photomorphogenesis. Archives Of Biochemistry and Biophysics. 448(1-2):45-59. Interpretive Summary: Using a post-genomics strategy, we compared the genome-wide transcriptional changes induced by red and far-red light with the metabolite profiles in the same seedlings of Arabidopsis. The data reveal a high degree of correlation between the gene expression patterns regulated by the phytochrome system and the metabolite levels in the cognate biochemical pathways. The data illustrate the general importance of integrative approaches to correlate postgenomic data sets with phenotypic outcomes.
Technical Abstract: One of the key developmental processes during photomorphogenesis is the differentiation of prolamellar bodies of proplastids into thylakoid membranes containing the photosynthetic pigment–protein complexes of chloroplasts. To study the regulatory events controlling pigment–protein complex assembly, including the biosynthesis of metabolic precursors and pigment end products, etiolated Arabidopsis thaliana seedlings were irradiated with continuous red light (Rc), which led to rapid greening, or continuous far-red light (FRc), which did not result in visible greening, and subjected to analysis by oligonucleotide microarrays and targeted metabolite profiling. An analysis using BioPathAt, a bioinformatic tool that allows the visualization of post-genomic data sets directly on biochemical pathway maps, indicated that in Rc-treated seedlings mRNA expression and metabolite patterns were tightly correlated (e.g., Calvin cycle, biosynthesis of chlorophylls, carotenoids, isoprenoid quinones, thylakoid lipids, sterols, and amino acids). K-means clustering revealed that gene expression patterns across various biochemical pathways were very similar in Rc- and FRc-treated seedlings (despite the visible phenotypic differences), whereas a principal component analysis of metabolite pools allowed a clear distinction between both treatments (in accordance with the visible phenotype). Our results illustrate the general importance of integrative approaches to correlate post-genomic data sets with phenotypic outcomes.