The heat shock factor family from Triticum aestivum in response to heat and other major abiotic stresses and their role in regulation of heat shock protein genes

Abstract

Heat shock factors (Hsfs) play a central regulatory role in acquired thermotolerance. To understand the role of the major molecular players in wheat adaptation to heat stress, the Hsf family was investigated in Triticum aestivum. Bioinformatic and phylogenetic analyses identified 56 TaHsf members, which are classified into A, B, and C classes. Many TaHsfs were constitutively expressed. Subclass A6 members were predominantly expressed in the endosperm under non-stress conditions. Upon heat stress, the transcript levels of A2 and A6 members became the dominant Hsfs, suggesting an important regulatory role during heat stress. Many TaHsfA members as well as B1, C1, and C2 members were also up-regulated during drought and salt stresses. The heat-induced expression profiles of many heat shock protein (Hsp) genes were paralleled by those of A2 and A6 members. Transactivation analysis revealed that in addition to TaHsfA members (A2b and A4e), overexpression of TaHsfC2a activated expression of TaHsp promoter-driven reporter genes under non-stress conditions, while TaHsfB1b and TaHsfC1b did not. Functional heat shock elements (HSEs) interacting with TaHsfA2b were identified in four TaHsp promoters. Promoter mutagenesis analysis demonstrated that an atypical HSE (GAACATTTTGGAA) in the TaHsp17 promoter is functional for heat-inducible expression and transactivation by Hsf proteins. The transactivation of Hsp promoter-driven reporter genes by TaHsfC2a also relied on the presence of HSE. An activation motif in the C-terminal domain of TaHsfC2a was identified by amino residue substitution analysis. These data demonstrate the role of HsfA and HsfC2 in regulation of Hsp genes in wheat.

Keywords: Gene expression; gene regulation; heat shock factors; heat shock proteins; heat stress; transcription factors; wheat..

Fig. 1.

The Neighbor–Joining phylogenetic tree of Hsf proteins from wheat, rice, and Arabidopsis. The N-proximal regions (from the start of the DNA-binding domain to the end of the HR-A/B region) of Hsf proteins were used for construction of the phylogenetic tree using the MEGA 5.10 program. Unrooted Neighbor–Joining analysis was performed with pairwise deletion and Poisson correction. For rice (prefixed by Os) and Arabidopsis (prefixed by AT) Hsf proteins, both locus ID and subclass number were used (e.g. AT4G17750A1a=AtHsfA1a with a locus ID AT4G17750). TaHsf proteins are marked with asterisks. Bootstrap values >50 are shown.

Fig. 2.

Relative mRNA abundance of TaHsf members in wheat organs. Values are means ±SD of three biological replicates and are expressed as apparent expression levels relative to a control gene TaRP15. 1Mo, 1-month-old; 20DAA, 20 d after anthesis; 27DAA, 27 d after anthesis; 42DAA, 42 d after anthesis.

Fig. 3.

Expression profiles of TaHsp genes. Values are means ±SD of three biological replicates and are expressed as apparent expression levels relative to a control gene TaRP15. (A) Organ mRNA distribution. (B) Response to heat stress in the leaves and roots. Statistical significance in differences between control and heat-treated groups (36 °C for 1.5h, 5h, or 3 d with 5h d^–1) is indicated by an asterisk.

Fig. 4.

Expression changes of TaHsf members during heat stress. Values are means ±SD of three biological replicates and are expressed as apparent expression levels relative to a control gene TaRP15. Statistical significance in differences between control and heat-treated groups (36 °C for 1.5h, 5h, or 3 d with 5h d^–1) is indicated by a black (up-regulated) or blue (down-regulated) asterisk. (Note: the expression levels of some TaHsfs are too low to be seen in this comparative presentation, but statistical significance in the changes of their transcript levels in response to heat stress is shown.)

Fig. 5.

Drought responsiveness of TaHsf genes in the flag leaves and glumes of T. aestivum Chinese Spring (CS) and T. durum Creso. Raw data are derived from an Affymetrix wheat genome array data set at http://www.plexdb.org (accession # TA23; Aprile et al., 2009). Values are means ±SD of three biological replicates and relative expression levels within each genotype (stress groups versus control group and each control group was arbitrarily set as 1). 5AL, Chinese Spring 5A deletion line. Hybridization signal <20 is considered not detectable (nd). *P < 0.05.

Fig. 6.

Expression changes of TaHsf genes in the shoot and roots of bread wheat plants in response to long-term salt stress. Raw data are derived from Affymetrix wheat genome array at http://www.ebi.ac.uk/arrayexpress/ (accession #: E-MEXP-971; Mott and Wang, 2007). Values are means ±SD of 15 biological replicates and are expressed as relative expression level (stress group versus control group and each control group was arbitrarily set as 1). The Affymetrix probe sets for these genes are in the same order as shown in Fig. 5. Hybridization signal <20 is considered not detectable (nd). *P < 0.01.

Fig. 7.

Transactivation of TaHsp promoter-driven reporter genes by TaHsfs. (A) Reporter and effector gene constructs. (B) Transactivation of the TaHsp17 or TaHsp90.1-A1 promoter-driven gfp reporter gene by a constitutively expressed TaHsfA2b, A4e, or C2a effector gene. The maize Ubi1 promoter-driven GUS reporter was co-bombarded with test constructs to check transformation events. GFP foci (green) indicate the expression of the TaHsp promoter-driven gfp reporter gene, and blue foci resulted from the expression of the co-introduced GUS reporter as indication that tissue sections were transformed with these constructs. The red background is the red fluorescence of shoot chlorophyll. The TaHsfA4e or TaHsfC2a effector gene also activated the expression of the Hsp90gfp reporter gene (data not shown). Co-bombardment of a TaHsfB1b or TaHsfC1b effector gene with a Hsp17gfp or Hsp90gfp reporter gene did not produce GFP foci (data not shown).

Fig. 8.

TaHsfA2b binding to elements present in the promoters of TaHsp17, TaHsp26.6, TaHp70d, and TaHsp90.1-A1 genes. (A) Relative binding activity (RBA) of TaHsfA2b to nGAAn or nTTCn sequences. Values are means ±SD of three assays and are relative to the binding activity of TaHsp90.1E1, which is arbitrarily set as 1. GAA or TTC sequences are in bold. TaHsp sequences are selected from the available promoter region sequences within 1.5kb upstream of the translation start of these TaHsp genes. Three artificial sequences (GAAn2GTCn2GAA, TCCn2GCAn2TCC, and AGAAn2TTCT) were also tested for tolerance of nucleotide sequences deviating from the perfect HSE (nGAAnnTCCnnGAA or nTCCnnGAAnnTCCn). (B) Illustration of TaHsfA2b binding activity as the 4-methylumbelliferyl group fluorescence produced through hydrolysis of methylumbelliferyl-β-d-cellobioside by TaHsfA2b–CELD that bound to immobilized HSE-containing oligonucleotides (TaHsp70dE1 and TaHsp90.1E1). Control is the oligonucleotide containing no HSE elements.

Fig. 9.

The heat-induced expression of gfp reporter genes driven by a truncated TaHsp90.1-A1 promoter or a minimal promoter with an addition of TaHsp90.1E1 (GAAGCTTCGGGAA) or TaHsp17E1 (GAACATTTTGGAA). (A) Reporter gene constructs. A short TaHsp90.1-A1 promoter (sHsp90) has a 328bp fragment upstream of the translational start codon, which contains the TaHsp90.1E1 element. The sequence upstream the TATA box of the TaHsp90.1-A1 promoter is shown. The ΔHsp90 promoter starts immediately downstream of the TaHsp90.1E1 element. HSE90 and HSE17 constructs contain three TaHsp90.1E1 or TaHsp17E1 repeats, which are added immediately upstream of the ΔHsp90 promoter (a TaHsp90.1-A1 minimal promoter). (B) Transient expression assays of reporter genes. The Ubi1GUS reporter gene was also co-introduced, and GUS foci are illustrated when gfp expression was essentially undetectable.

Fig. 10.

TaHsfA2b and TaHsfC2a transactivate the gfp reporter driven by the TaHsp90.1-A1 minimal promoter containing TaHsp90.1E1 (HSE90) or TaHsp17E1 (HSE17). Constructs used are shown in Fig. 9. The Ubi1GUS reporter gene was also co-introduced, and GUS foci are illustrated when gfp expression was essentially undetectable. Ubi1 promoter-driven HvCBF1, which is a transcriptional activator for cold-inducible genes, was used as a negative control.

Fig. 11.

Loss of the transactivation activity of TaHsfC2a by mutation of hydrophobic and acidic residues in its C-terminal domain. (A) TaHsfC2a and its mutant constructs. The substituted amino residues in the mutants (C2a-mC1 and C2a-mC2) are in red. (B) Transactivation of the TaHsp17 promoter-driven gfp reporter (Hsp17gfp) by TaHsfC2a and its mutants. The Ubi1GUS reporter gene was also co-introduced, and GUS foci are illustrated when gfp expression was essentially undetectable.