Identification and characterization of wheat drought-responsive MYB transcription factors involved in the regulation of cuticle biosynthesis

Abstract

A plant cuticle forms a hydrophobic layer covering plant organs, and plays an important role in plant development and protection from environmental stresses. We examined epicuticular structure, composition, and a MYB-based regulatory network in two Australian wheat cultivars, RAC875 and Kukri, with contrasting cuticle appearance (glaucousness) and drought tolerance. Metabolomics and microscopic analyses of epicuticular waxes revealed that the content of β-diketones was the major compositional and structural difference between RAC875 and Kukri. The content of β-diketones remained the same while those of alkanes and primary alcohols were increased by drought in both cultivars, suggesting that the interplay of all components rather than a single one defines the difference in drought tolerance between cultivars. Six wheat genes encoding MYB transcription factors (TFs) were cloned; four of them were regulated in flag leaves of both cultivars by rapid dehydration and/or slowly developing cyclic drought. The involvement of selected MYB TFs in the regulation of cuticle biosynthesis was confirmed by a transient expression assay in wheat cell culture, using the promoters of wheat genes encoding cuticle biosynthesis-related enzymes and the SHINE1 (SHN1) TF. Two functional MYB-responsive elements, specifically recognized by TaMYB74 but not by other MYB TFs, were localized in the TdSHN1 promoter. Protein structural determinants underlying the binding specificity of TaMYB74 for functional DNA cis-elements were defined, using 3D protein molecular modelling. A scheme, linking drought-induced expression of the investigated TFs with downstream genes that participate in the synthesis of cuticle components, is proposed.

Keywords: Abiotic stress; MYB and SHINE1 transcription factors; cuticle; drought; molecular model; water deficit; wax; wheat.; β-diketone.

Fig. 1.

The visual appearance, ultrastructure, and the wax composition of cuticle on flag leaves of wheat. (A) The appearance of abaxial sides of flag leaves detached from RAC875 and Kukri wheat cultivars grown under well-watered conditions. (B and C) Scanning electron micrographs of the abaxial sides of flag leaves derived from RAC875 and Kukri plants grown under well-watered conditions. (D and E) Scanning electron micrographs of the abaxial side of leaves derived from RAC875 and Kukri plants grown under the conditions of limited watering (mild drought). (F and G) Total wax loads and amounts of the four most abundant wax components on the flag leaves of RAC875 (R) and Kukri (K) under well-watered (WW) and mild drought (DR) conditions. Wax loads were calculated as µg of wax per dry leaf weight (DW). Dik, C31 β-diketones; Alk, alkane; Alc, primary alcohol. Means and SEs were calculated from three replicates. Two-way ANOVA with the Fisher’s least significant difference post-hoc test was conducted using GenStat. The same lower case letters on top of error bars indicate differences that are not significant at the 5% level. Scale bars=5 µm.

Fig. 2.

A phylogenetic tree of MYB TFs. We analysed 141 sequences including six sequences of wheat MYB TFs (indicated by dots) derived from cDNAs cloned in this work, five query Arabidopsis and tomato MYB TFs (GenBank accessions shown in the figure), 103 wheat MYB factors from the Plant Transcription Factor Database (annotated as Tae with a six-figure number) and 27 wheat R2R3 MYB TFs (Zhang et al., 2012). The branches, to which wheat MYB protein sequences studied in this work belong, are indicated with thick grey lines. The number near a scale indicates a residue difference per site. The tree was constructed using the Neighbour–Joining method in MEGA6. (This figure is available in colour at JXB online.)

Fig. 3.

Transcriptional activation assays and the localization of activation domains of cloned MYB genes. The assay was performed in yeast using full-length and C-terminal truncated MYB TFs fused to a binding domain (BD) of yeast GAL4 TF. An empty pGBKT7 plasmid was used as a negative control. -Trp represents the synthetic defined (SD) medium lacking tryptophan (selection for plasmid presence) and -Trp/-His refers to the SD medium without tryptophan and histidine (selection for activation of the yeast HIS3 gene). Drought-responsive MYBs and their truncations are shown in bold. Domain structures and positions of truncations are indicated in the right part of the figure. SANT: Swi3, Ada2, N-Cor, and TFIIIB DNA-binding domains. D1 and D2 represent removed protein fragments; D2 truncation included the removal of D1. The residue positions of truncations are indicated. (This figure is available in colour at JXB online.)

Fig. 4.

Expression levels of cloned MYB genes in rapidly dehydrating leaves of Kukri and RAC875. Expression of TaMYB24, TaMYB31, TaMYB74, and TaMYB77 was studied by Q-PCR. Flag leaf samples were sampled at awn emergence. Dehydration was performed at room temperature for 0, 2, 4, and 7h, after which leaves were snap-frozen in liquid nitrogen. Two-way ANOVA with the Fisher’s least significant difference post-hoc test was conducted using GenStat. Error bars indicate the SE of three replicates.

Fig. 5.

Expression levels of cloned MYB genes under cyclic drought in Kukri and RAC875. Expression of TaMYB24, TaMYB31, TaMYB74, and TaMYB77 was studied by Q-PCR. Expression of genes was examined after 5, 9, 14, 23, and 25 d, using either well-watered (WW) or cyclic drought-exposed (DR) plants. Three cycles of drought (after watering points 1–3) indicated by arrows were applied at 0, 15, and 24 d as shown in Supplementary Fig. S1. Two-way ANOVA with the Fisher’s least significant difference post-hoc test was conducted using GenStat. Error bars indicate the SE of three replicates.

Fig. 6.

Expression profiles of TaMYB24, TaMYB31, TaMYB74, and TaMYB77 in wheat tissues revealed by Q-PCR. germ., germinating seed; Emb., embryo; Immat. Inflor., immature inflorescence; b.a., before anthesis; Car., caryopsis; DAP, days after pollination. Error bars indicate the SE of three replicates.

Fig. 7.

Activation of promoters of cuticle-related genes TaATT1, TaKCS1, and TdSHN1 by drought-responsive MYB TFs. The data were obtained by a transient expression assay in a wheat suspension culture. (A) Schematic showing DNA constructs used in the transient expression assay. The reporter GUS gene was driven by one of three promoters of cuticle biosynthesis genes, TaATT1, TaKCS1, and TdSHN1. In effector constructs, wheat MYB genes were cloned under the control of the ubiquitin promoter. GFP served as a negative control. (B) Activation of GUS expression fused with promoters of TaATT1, TaKCS1, and TdSHN1 by drought-responsive MYB factors. Each reporter construct was co-bombarded with each effector and GFP construct into a wheat suspension culture. One-way ANOVA with the Fisher’s least significant difference post-hoc test was conducted using GenStat. Error bars indicate the SE of three replicates.

Fig. 8.

Identification of functional MYBR cis-elements in the TdSHN1 promoter using a transient expression assay. The full-length TdSHN1 promoter (F) and six 5'-deletions (D1–D6) were cloned upstream of the GUS reporter gene, and co-transformed by biolistic bombardment with either a negative control (pUbi–GFP) or pUbi–TaMYB74 constructs. Promoter deletions and existing MYBR cis-elements (MYBCORE, MYB1AT, and MYBATRD22) within 696bp upstream of the start codon are shown in the left panel. GUS expression quantifications are shown in the right side of the figure. One-way ANOVA with the Fisher’s least significant difference post-hoc test was conducted using GenStat. Error bars indicate the SE of three replicates. Functional cis-elements are circled.

Fig. 9.

Protein sequence analyses and a molecular model of TaMYB74 in complex with the MYBR1 cis-element. (A) The protein alignments of the DNA-binding domains of wheat MYB and protozoan TvMYB2 proteins; the latter was used as a template for molecular modelling. Tandem imperfect amino acid repeats R2 and R3 are indicated by lines above the sequences. The conserved residues that form a hydrophobic core and the residues that interact with the DNA cis-element are denoted by filled circles and filled inverted triangles, respectively. (B) A cartoon of the TaMYB74 model (cyan) in complex with MYBR1 (orange) (left panel). Predicted residues interacting with DNA (distances between 2.8 Å and 3.6 Å) are shown in magenta sticks (right panel). (C) The orientations and positions of conserved tryptophan and phenylalanine residues, which form a hydrophobic core of TaMYB74. (D) Energy gains (kcal mol^–1) upon mutation (into alanine or lysine) of Lys14, Trp17, Lys51, Ser90, Lys105, Asn106, and Arg115, involved in MYBR1 DNA binding, as determined by Fold-X (Schymkowitz et al., 2005).

Scheme 1.

The proposed roles of TaMYB31 and TaMYB74 in the regulation of cuticle biosynthesis under drought. The dashed lines reflect the roles of TaSHN1 in regulating TaATT1 and TaKCS1 genes, and consequently the biosynthesis of cuticular wax components, based on our own and other data.