Fig 1.
A summary of differentially expressed genes of B. distachyon seedlings under heat, cold, drought and salt stresses.
(A) The number of up-/down-regulated genes under four abiotic stresses. Vertical axis indicated the differentially expressed gene numbers. Horizontal axis indicates four stresses. Red, green, and blue bars represented the total, decreased, and increased probe sets, respectively. (B) Venn diagram of differentially expressed genes under four abiotic stresses. Three circles indicate the differentially expressed genes under cold, drought and salt stresses. Numbers in red circles indicated gene numbers co-regulated by heat stress and other stresses, while numbers in green circles indicated the number of BdHsf and BdHsp70 genes co-regulated by heat stress and other stresses in our qPCR experiment.
Fig 2.
Chromosomal locations and gene duplication for B. distachyon Hsf and Hsp70 genes.
The chromosomal position of each BdHsf and BdHsp70 gene was mapped according to the B. distachyon genome. The chromosome number is indicated at the top of each chromosome. The CpG island distribution maps shown in each chromosome depended on the CpG density in B. distachyon genome. The dotted line showed the gene duplication events among these BdHsf and BdHsp70 genes. The scale bar was show on the left.
Fig 3.
Neighbor-Joining phylogenetic trees of Hsf (A) and Hsp70 (B) proteins from Arabidopsis, rice and B. distachyon.
The phylogenetic trees were constructed using the neighbor-joining method in the MEGA version 5 software, with bootstrap values from 1,000 replicates indicated at each node with the following parameters: p-distance and pairwise deletion.
Fig 4.
Phylogenetic relationships, protein structures and tissue specific expression heatmap of Hsfs and Hsp70s in B. distachyon.
(A) The phylogenetic tree was constructed from the nucleotide sequences using the NJ program from MEGA 5, representing relationships among BdHsfs and BdHsp70s, respectively. The numbers beside the branches represent bootstrap support values (>50%) from 1000 replications. (B) Structure of Hsf and Hsp70 proteins were analyzed by MEME web server, and pictures were generated by Expasy web server. The details of sequence logo of motifs were shown in S1 Fig. (C) Quantitative RT-PCR analysis of the expression levels of BdHsf and BdHsp70 genes in different tissues. The expression profile was shown by a green-red gradient using the PermutMatrix program. Results were normalized using BdActin (Bradi2g24070) gene expression as the internal control. Numbers under heatmap indicated the –ΔCT value [–ΔCT = –(CTgene–CTactin)].
Fig 5.
Expression profiles of BdHsf and BdHsp70 genes.
(A-B) Expression patterns of Hsf genes in B. distachyon in response to HS challenge (A) and multiple abiotic stresses (B). (C-D) Expression patterns of Hsp70 genes in B. distachyon in response to HS challenge (C) and multiple abiotic stresses (D). The expression values of the BdHsf and BdHsp70 genes were assessed upon the qPCR result analysis. The expression profile was shown by a green-red gradient using the PermutMatrix program. Numbers under heatmap indicated the –ΔΔCT value [–ΔΔCT = (CTcontrol.gene–CTcontrol.actin)–(CTtreat.gene–CTtreat.actin)].
Fig 6.
Expression correlations of Hsf and Hsp70 genes between any two of heat, cold, drought, salt and oxidative stresses in B. distachyon.
For each family of Hsf and Hsp70 (Hsp110s were separately analyzed), responsive expression levels of gene members under any two stress were plotted. At each top right of subfigure, H stands for heat stress, C for cold stress, O for oxidative stress, D for drought stress, S for salt stress. For example, ‘H Vs C’ meant that x-axis and y-axis represented the response fold change under heat and cold stresses, respectively. Pearson correlation coefficient (R) represents the expression pattern similarity of a given gene family under two stresses.
Fig 7.
Co-expression networks of Hsfs and Hsp70s in B. distachyon.
Nodes represent Hsfs and Hsp70s in B. distachyon, edges indicate pairwise correlation constructed by Cluster 3.0 software. Blue edges indicate the positive correlation and green edges indicate the negative correlation, while thin edges indicate moderate co-expressions and thick edges indicate strong co-expressions between the two nodes. The network picture which represented the co-expression regulatory network was created using Cytoscape.