Transcriptomic and metabolic regulatory network characterization of drought responses in tobacco

Abstract

Drought stress usually causes huge economic losses for tobacco industries. Drought stress exhibits multifaceted impacts on tobacco systems through inducing changes at different levels, such as physiological and chemical changes, changes of gene transcription and metabolic changes. Understanding how plants respond and adapt to drought stress helps generate engineered plants with enhanced drought resistance. In this study, we conducted multiple time point-related physiological, biochemical,transcriptomic and metabolic assays using K326 and its derived mutant 28 (M28) with contrasting drought tolerance. Through integrative analyses of transcriptome and metabolome,we observed dramatic changes of gene expression and metabolic profiles between M28 and K326 before and after drought treatment. we found that some of DEGs function as key enzymes responsible for ABA biosynthesis and metabolic pathway, thereby mitigating impairment of drought stress through ABA signaling dependent pathways. Four DEGs were involved in nitrogen metabolism, leading to synthesis of glutamate (Glu) starting from NO-3 /NO-2 that serves as an indicator for stress responses. Importantly, through regulatory network analyses, we detected several drought induced TFs that regulate expression of genes responsible for ABA biosynthesis through network, indicating direct and indirect involvement of TFs in drought responses in tobacco. Thus, our study sheds some mechanistic insights into how plant responding to drought stress through transcriptomic and metabolic changes in tobacco. It also provides some key TF or non-TF gene candidates for engineering manipulation for breeding new tobacco varieties with enhanced drought tolerance.

Keywords: Nicotiana tabacum L.; drought stress; metabolome; regulatory network; transcriptome.

Figure 1

Biochemical comparisons between tobacco K326 and M28 at different time points after exposing to ND and SD, respectively. (A, B) SOD activity at 0 D,7 D,14 D post ND, and 0h,4h,24h between tobacco K326 and M28. (C, D) POD activity at 0 D,7 D,14 D post ND, and 0h,4h,24h between tobacco K326 and M28. (E, F) H2O2 content at 0 D,7 D,14 D post ND, and 0h,4h,24h between tobacco K326 and M28. (G, H) MDA content at 0 D,7 D,14 D post ND, and 0h,4h,24h between tobacco K326 and M28. Different letters represent significant differences (P< 0.05, one-way ANOVA).The difference is not significant if there is a letter with the same marker, and significant if there is a letter with a different marker.

Figure 2

Physiological comparisons between tobacco K326 and M28 at different time points after exposing to ND. (A, B) Relative leaf water content at 0 D,7 D,14 D post ND, and 0h,4h,24h between tobacco K326 and M28. (C) Net photosynthetic rate at 0 D,7 D,14 D post ND between tobacco K326 and M28. (D) Relative conductivity at 0 D,7 D,14 D post ND between tobacco K326 and M28. (E) Stomatal conductance at 0 D,7 D,14 D post ND between tobacco K326 and M28. (F) Transpiration rate at 0 D,7 D,14 D post ND between tobacco K326 and M28.Different letters represent significant differences (P< 0.05, one-way ANOVA).The difference is not significant if there is a letter with the same marker, and significant if there is a letter with a different marker.

Figure 3

Characterization of differentially expressed genes before and after drought treatment in M28 or K326. (A) Pairwise comparisons of down-regulated genes at the same time point between M28 and K326 after drought treatment. (B) Pairwise comparisons of up-regulated genes at the same time point between M28 and K326 after drought treatment. (C) 19 DEGs were randomly selected for RT-qPCR assay. The red line represents the relative expression, and the black line represents the FPKM values. (D) GO term enrichment analyses of down-regulated genes at the same time point between M28 and K326 after drought treatment. (E) GO term enrichment analyses of up-regulated genes at the same time point between M28 and K326 after drought treatment.

Figure 4

Differential metabolic changes in response to drought treatment between M28 and K326. (A) Histogram showing the number of differential metabolic changes in response to time-point-related drought-induced in M28 or K326 or between M28 and K326. (B) KEGG pathway enrichment analyses of up- and down-regulated metabolites in K326 or M28 under drought treatment. (C) KEGG pathway enrichment analyses of metabolites that were up- or down-regulated before (CK) or under drought treatment. (D) Pairwise comparisons of down- (left) and up-regulated (right) metabolites at the same time point between M28 and K326 after drought treatment.

Figure 5

Integrated analyses of transcriptome and metabolome. (A) KEGG pathway enrichment analyses of genes that were up- or down-regulated before (CK) or under drought treatment. (B) KEGG pathway enrichment analyses of up- and down-regulated genes in K326 or M28 under drought treatment. (C) Relative expression levels (up) and phenotypes (down) of VIGS-mediated knock-down of gene NCED1, PP2C-37L and P450 84A1L before or post drought treatment. (D) Overview of DEGs and metabolites involved in “Phenylpropanoid biosynthesis” and “Flavonoid biosynthesis”. The red representing differentially expressed genes and differential metabolites that are significantly enriched in the pathways of Phenylpropanoid biosynthesis terms and Flavonoid biosynthesis pathway. Four differentially expressed genes with relative expression levels (FPKM value) in the M28 and K326 are shown in the heatmap.

Figure 6

Involvement of Abscisic acid in drought responses in tobacco. (A) Overview of the KEGG pathway of CoA biosynthesis. The red indicating significantly up-regulated metabolites between M28 and K326 after drought treatment. (B) Overview of the KEGG pathway of Citrate cycle (TCA cycle). The red indicating significantly up-regulated metabolites between M28 and K326 after drought treatment. (C) Involvement of DEGs in combination with metabolites in synthesis and decomposition pathways of Abscisic acid in tobacco.

Figure 7

WGCNA analyses. (A) Module gene expression pattern in the brown module, red representing up-regulated genes, the green representing down-regulated genes. (B) The correlation network of genes in the brown module. A gene network is constructed by WGCNA, in which each node represents a gene, and the connecting line between genes represents the co-expression correlation, the blue representing not significant genes and the yellow representing down-regulated genes between M28 and K326 before and after drought treatment. The genes with weights >0.1 are visualized by Cytoscape. (C) The correlation network of genes in the blue module. (D) The correlation network of genes in the blue module. The red representing up-regulated genes and the blue representing not significant genes. The diamond representing the TF in the co-expression network. (E) GO term enrichment analyses of down-regulated genes in the brown module.

Figure 8

Prediction of TF regulatory networks using the co-expression correlation network of three TFs in the blue module. (A) The correlation network of genes related to regulation of the three TFs. The red representing up-regulated genes and the yellow representing down-regulated genes between M28 and K326 before and after drought treatment. The direction of the arrow indicates the regulatory relationship. The genes with weights > 0.1 are visualized by Cytoscape. (B) GO term enrichment analyses of down- and up-regulated genes related to regulation of the three TFs.