A digital compendium of genes mediating the reversible phosphorylation of proteins in fe-deficient Arabidopsis roots

Abstract

Post-translational modifications of proteins such as reversible phosphorylation provide an important but understudied regulatory network that controls important nodes in the adaptation of plants to environmental conditions. Iron (Fe) is an essential mineral nutrient for plants, but due to its low solubility often a limiting factor for optimal growth. To understand the role of protein phosphorylation in the regulation of cellular Fe homeostasis, we analyzed the expression of protein kinases (PKs) and phosphatases (PPs) in Arabidopsis roots by mining differentially expressed PK and PP genes. Transcriptome analysis using RNA-seq revealed that subsets of 203 PK and 39 PP genes were differentially expressed under Fe-deficient conditions. Functional modules of these PK and PP genes were further generated based on co-expression analysis using the MACCU toolbox on the basis of 300 publicly available root-related microarray data sets. Results revealed networks comprising 87 known or annotated PK and PP genes that could be subdivided into one large and several smaller highly co-expressed gene modules. The largest module was composed of 58 genes, most of which have been assigned to the leucine-rich repeat protein kinase superfamily and associated with the biological processes "hypotonic salinity response," "potassium ion import," and "cellular potassium ion homeostasis." The comprehensive transcriptional information on PK and PP genes in iron-deficient roots provided here sets the stage for follow-up experiments and contributes to our understanding of the post-translational regulation of Fe deficiency and potassium ion homeostasis.

Keywords: RNA-seq; co-expression; iron deficiency; potassium homeostasis; protein phosphorylation.

Figure 1

Flowchart for mining differentially expressed PK and PP genes and subsequent co-expression analysis in Arabidopsis iron-deficient roots.

Figure 2

Differentially expressed PK and PP genes in Fe-deficient Arabidopsis roots. (A,B) Number and expression levels of PK genes. (C,D) Number and expression levels of PP genes.

Figure 3

Co-expression relationships of the differentially expressed PK and PP genes. Red nodes indicate up-regulated genes, green nodes denote genes that are repressed by Fe deficiency. Round-shaped nodes represent protein kinase genes, rectangles indicate protein phosphatase genes.

Figure 4

Gene ontology (GO) enrichment analysis of the genes forming the network. GO enrichment of the 87 PK and PP genes which involved in the co-expression network was performed with the Gene Ontology Browsing Utility (GOBU) (Lin et al., 2006) using the TopGo “elim” method (Alexa et al., 2006) with P < 0.01. (A) GO of the biological process and (B) GO of the subcellular localization.

Figure 5

Gene ontology (GO) enrichment analysis of the genes from FEPKPP1. GO enrichment of the 58 PK and PP genes in the module FEPKPP1 was performed with the GOBU toolbox using the TopGo “elim” method with P < 0.01. (A) GO biological process and (B) GO subcellular localization.

Figure 6

Co-expression relationships of Fe-responsive genes with fold-changes greater than 1.5-fold in Arabidopsis roots. (A) Module FEPKPP2 and (B) module FEPKPP3. Red nodes indicate up-regulated PK genes, green nodes denote PK genes that are repressed by Fe deficiency, white nodes indicate fished genes.

Figure 7

Gene ontology (GO) enrichment analysis of the genes from FEPKPP2. GO enrichment of the 64 Fe-responsive genes in the module FEPKPP2 was performed with the GOBU toolbox using the TopGo “elim” method with P < 0.01. (A) GO biological process and (B) GO subcellular localization.

Figure 8

Gene ontology (GO) enrichment analysis of the genes from FEPKPP3. GO enrichment of the 36 Fe-responsive genes in the module FEPKPP3 was performed with the GOBU toolbox using the TopGo “elim” method with P < 0.01. (A) GO biological process and (B) GO subcellular localization.