NAC transcription factors ATAF1 and ANAC055 affect the heat stress response in Arabidopsis

Abstract

Pre-exposing (priming) plants to mild, non-lethal elevated temperature improves their tolerance to a later higher-temperature stress (triggering stimulus), which is of great ecological importance. 'Thermomemory' is maintaining this tolerance for an extended period of time. NAM/ATAF1/2/CUC2 (NAC) proteins are plant-specific transcription factors (TFs) that modulate responses to abiotic stresses, including heat stress (HS). Here, we investigated the potential role of NACs for thermomemory. We determined the expression of 104 Arabidopsis NAC genes after priming and triggering heat stimuli, and found ATAF1 expression is strongly induced right after priming and declines below control levels thereafter during thermorecovery. Knockout mutants of ATAF1 show better thermomemory than wild type, revealing a negative regulatory role. Differential expression analyses of RNA-seq data from ATAF1 overexpressor, ataf1 mutant and wild-type plants after heat priming revealed five genes that might be priming-associated direct targets of ATAF1: AT2G31260 (ATG9), AT2G41640 (GT61), AT3G44990 (XTH31), AT4G27720 and AT3G23540. Based on co-expression analyses applied to the aforementioned RNA-seq profiles, we identified ANAC055 to be transcriptionally co-regulated with ATAF1. Like ataf1, anac055 mutants show improved thermomemory, revealing a potential co-control of both NAC TFs over thermomemory. Our data reveals a core importance of two NAC transcription factors, ATAF1 and ANAC055, for thermomemory.

Figure 1

Expression of NAC transcription factors during the memory phase, after heat priming. (a) Schematic representation of the heat stress (HS) regime applied to asses HS memory. (b) Heat map and k-means clustering (k = 5 clusters) of NACs, based on their change in expression (HS samples compared to unstressed controls) during the memory phase at the time points indicated in the representation in (a). The color indicates the expression ratio as log2-fold change for the different time points, where ‘yellow’ denotes an increased expression and ‘purple’ a reduced expression. The list of genes in each cluster and their expression values are shown in Supplementary Table S1.

Figure 2

Expression of NACs after a triggering stimulus. (a) Schematic representation of the time points used for expression analysis of NACs in triggered plants (T), and in plants that were primed before triggering (P + T). In triggered (T) plants, the expression was calculated as a ratio of HS samples compared to unstressed controls. In primed and triggered (P + T) plants, the expression was calculated as primed and triggered (P + T), compared to triggered only (T). (b) K-means clustering of NAC expression profiles of primed and triggered (P + T) and triggered (T) plants after the triggering stimulus (k = 10 clusters). The expression of genes in each cluster is displayed as a heat map. The color indicates the expression as log2-fold change for the different time points, where yellow denotes an increased expression and purple a reduced expression. The list of genes in each cluster and their expression values are shown in Supplementary Table S2.

Figure 3

Improved thermomemory in ataf1 mutants. (a) Thermomemory phenotype of ATAF1 transgenic plants. Seedlings of ATAF1-OE, ataf1-2, ataf1–4, and WT were exposed to the HS regime schematically shown in Fig. 1a. Photos were taken 14 days after the second HS (triggering stimulus). The phenotype of one representative replicate of at least three independent biological replicates is shown. (b,c) Quantification of the results shown in (a). (b) Percentage of seedlings in different phenotypic classes; the thermotolerance phenotype was classified into three classes depending of the extent of plant recovery. The phenotype classes are illustrated in panel (a). (c) Seedling fresh mass in HS-primed and triggered plants compared to unstressed control plants. Error bars represent the standard deviation, calculated from three biological replicates; each replicate is the average mass of 13 seedlings. Significant differences between transgenic and WT plants were calculated using Student’s t-test; *indicates P-value < 0.05.

Figure 4

Clustering of RNA-seq data. (a) Schematic presentation of the time points used for expression analysis by RNA-seq. (b) Top: Hierarchical clustering of the samples according to their expression patterns. “+” denotes ATAF1-OE lines; “0” denotes wild-type Col-0; and “−” denotes ataf1–4. The sizes of the circles indicate the time elapsed since HS (0, 1, and 4 h after heat stress). The black and red colors indicate whether the plants underwent control or heat temperature treatment for 90 min. Legend: colored heat map of Pearson’s coefficients of expression profiles between all possible pairs of the samples. A partial rainbow color scheme, from red to blue, indicates the range of correlation values. (c) Multidimensional scaling (MDS) plot. Overall similarities and differences between samples, based on a multidimensional scaling analysis. Briefly, the Euclidean distance between all samples was calculated according to the expression levels of all genes, and represented in a two-dimensional space.

Figure 5

Global transcriptomic changes of thermomemory-associated genes potentially regulated by ATAF1. (a) Venn diagrams of heat-induced and heat-repressed genes, after heat priming (heat relative to control) in WT, ataf1–4 mutant, and ATAF1-OE transgenic plants. (b) Venn diagram of genes upregulated in ATAF1-OE and downregulated in ataf1 mutant, and vice versa, compared to WT, in response to a priming HS, after 0 h, 1 h or 4 h. (c) Schematic representation of the position of ATAF1-binding sites in the upstream regions of genes identified in (b).

Figure 6

Potential targets of ATAF1 in response to heat stress. (a) The bar graph shows the number of genes in each co-expression cluster. A clustering analysis was performed by weighted gene correlation network analysis, which resulted in 40 clusters of genes co-expressed together. (b) The bar graph shows the proportion of ATAF1 and ANAC055 targets in each cluster, including genes targeted by both; clusters with significant enrichment of ATAF1 and ANAC055 target genes are shaded (P < < 0.001); only two clusters (10 and 13) are highly enriched with ATAF1 targets; clusters 10 and 21 are significantly enriched for ANAC055, and clusters 10, 13 and 21 are enriched for genes that are targeted by both transcription factors.

Figure 7

Thermomemory is improved in the ataf1–4/anac055 double-knockout mutant compared to WT. (a) Schematic representation of HS regime. (b) Thermomemory phenotype of ANAC055-OE, anac055-1 and anac055-2 mutants, and WT plants. (c) Thermomemory phenotype of ATAF1-OE, ataf1–4 mutant, ataf1–4/anac055 double-mutant, and WT plants. Seedlings were exposed to the HS regime shown in (a). Left panels: seedlings photographed 14 days after exposure to the triggering heat stress. The phenotype of one representative replicate of at least three (b) or five (c) independent biological replicates is shown. Right panels: seedling fresh mass upon heat stress, compared to control plants (no heat stress). Error bars represent standard deviation, which was calculated from three biological replicates, where each replicate was the average mass of 18–19 seedlings. Significant differences between transgenic lines and WT plants were calculated using Student’s t-test; * if P < 0.05; *** if P < 0.001.