TaWRKY33 Positively Regulates TaERF1‐A, Thereby Activating TaP5CS ₂ ⁻ Mediated Proline Biosynthesis, Which Enhances Drought Tolerance in Wheat (Triticum aestivum L.)

ABSTRACT

Drought stress severely limits global wheat productivity. Herein, the function of TaERF1‐A, encoded by a previously undescribed wheat ERF gene, in enhancing drought stress tolerance was analyzed. TaERF1‐A exhibited nuclear localisation and transcriptional activation activity in planta, although no transactivation was detected in yeast. Notably, TaERF1‐A was upregulated under drought and salt stress. TaERF1‐A overexpression (OE) in Arabidopsis enhanced drought stress resistance, resulting in increased transcription levels of stress‐ and antioxidant‐enzyme–related genes. BSMV‐TaERF1‐A–silenced wheat plants exhibited exacerbated wilting under drought, accompanied by elevated malondialdehyde levels, reduced proline accumulation, and decreased antioxidant enzyme activity levels. In contrast, TaERF1‐A OE wheat lines exhibited enhanced drought tolerance, with opposite physiological trends. Notably, RNAi induced a weakened response in wheat compared to wild‐type wheat. Yeast one‐hybrid, electrophoretic mobility shift, dual‐luciferase, and qRT‐PCR assays demonstrated that TaWRKY33 directly binds the TaERF1‐A promoter, thereby activating its transcription. TaWRKY33 overexpression also enhanced drought tolerance in wheat. TaERF1‐A subsequently regulated TaP5CS2 expression, which stimulated proline biosynthesis, thereby contributing to improved drought resistance. This study elucidates the mechanism by which the TaWRKY33–TaERF1‐A–TaP5CS2 module positively regulates drought tolerance and highlights the candidate genes that can be explored for developing elite drought‐resistant wheat varieties.

Summary statement

We delineate a hierarchical TaWRKY33–TaERF1‐A–TaP5CS2 transcriptional cascade that enhances drought tolerance in wheat through activation of proline biosynthesis and improved osmotic adjustment.

1. Introduction

Crops often experience adverse environmental conditions, including elevated salinity, insufficient water, and extreme heat and cold, throughout their lifecycle (Rivero et al. ²⁰²²). These conditions negatively affect crop yield. Notably, drought is a major environmental stressor inhibiting plant development and productivity. Plants are particularly vulnerable to dryness during their early development (Dietz et al. 2021). The range of yield losses on a farm experiencing drought is between 30% and 90% (Dietz et al. 2021). A large suite of stress response transcription factors (TFs) form transcription cascades that control the transcriptional level of related genes, enhance associated physiological characteristics, and alleviate drought stress (DS) to help the plant physiologically adapt to DS (Munemasa et al. 2015; Wang et al. ^2018a; Wang et al. ^2018b). Genetic engineering is a proven strategy (Rong et al. 2014) that utilises TFs, such as ERF, NF‐Y, HD‐ZIP, and WRKY, to enhance drought tolerance (DT) in crops by expressing numerous stress‐responsive genes (Bhatnagar‐Mathur et al. ²⁰⁰⁸).

The APETALA2/ethylene‐responsive factor (AP2/ERF) superfamily is one of the most extensive groups of TFs in plants. Its functions in numerous developmental processes, as well as stress responses, have been extensively researched (Yu et al. 2022). Notably, a minimum of 147, 164, 200, and 322 AP2/ERF members have been found in Arabidopsis, poplar (Populus spp.), rice (Oryza sativa), and wheat (Triticum aestivum L.), respectively (Feng et al. 2020; Riaz et al. ²⁰²¹). AP2/ERF TFs are characterised by the presence of one to two highly conserved AP2 domains, and they constitute an important multigene family that can be categorised into four subfamilies, AP2, RAV, ERF, and DREB, based on their sequence similarities and number of domains. The AP2 subfamily is distinguished by the inclusion of two replicated AP2 domains and is involved in various developmental processes in plants, thus attracting substantial research attention (Chuck et al. 1998; Boutilier et al. ²⁰⁰²; Qi et al. ²⁰¹¹). RAV members contain a single AP2 domain alongside a B3 domain. Each domain fulfills unique biological functions and engages in various forms of transcriptional regulation (Fu et al. 2014; Matías‐Hernández et al. ²⁰¹⁴). ERF and DREB members contain a single AP2 domain, which features conserved amino acids at positions 14 and 19 within the domain. ERF members attach to the GCCGCC (GCC‐box) motif located in the promoters of genes associated with pathogenesis and induced by ethylene (Ohme‐Takagi and Shinshi ¹⁹⁹⁵). In contrast, the DREB family members possess C‐repeats that engage with the A/GCCGAC (core DRE/CRT) element (Kasuga et al. 2004). Positions 14 valine (14 V) and 19 glutamic acid (19E) within DREB members are particularly conserved. In contrast, these positions are conserved as alanine (A) and aspartic acid (D), respectively, in ERF members (Sakuma et al. 2002).

Plant ERF TFs influence numerous developmental processes and responses to both biotic and abiotic stresses. ERFs are known for their conserved DNA‐binding domain and are essential in multiple facets of plant growth, including within roots and leaves (Tan et al. 2018; Zhuang et al. ²⁰²¹). Plants face various biotic stresses from a range of organisms, including insects, fungi, and bacteria (Wu et al. 2020). Notably, the intricate signalling networks mediated by ERF TFs enable plants to physiologically adapt and survive under adverse conditions, highlighting their importance in plant biology. For instance, ERF TFs BrERF11b and ORA59 serve crucial roles in plant resistance against insects and the fungus Botrytis cinerea (Pré et al. 2008; Wu et al. ²⁰²⁰). Some ERF proteins participate in biotic reactions and regulate responses to abiotic stimuli (Xu et al. 2007; Dong et al. ²⁰¹²; Cheng et al. ²⁰¹³; Rong et al. ²⁰¹⁴). Notably, the ERF subfamily has more members, which play vital roles in plant DS responses. Overexpression (OE) of ERF subfamily members, such as AtERF1 (Cheng et al. 2013), AtERF019 (Scarpeci et al. 2017), AtERF74 (Yao et al. 2017), HARDY (Abogadallah et al. 2011), TINY (Xie et al. 2019), and TRANSLUCENT GREEN (Zhu et al. 2014), significantly improved drought resistance (DR) in Arabidopsis. In rice, OsERF48 (Jung et al. 2017), OsERF83 (Jung et al. 2021), OsLG3 (Xiong et al. ²⁰¹⁸), OsERF71 (Lee et al. 2016), and HYR (Ambavaram et al. 2014) genes regulate drought resistance. In wheat, TaERF1 (Xu et al. 2007), TaERF3 (Rong et al. 2014), TaERF‐6‐3A (Yu et al. 2022), and TaERF87 (Du et al. 2023) are involved in regulating the DS response. Despite several studies demonstrating that wheat ERF genes regulate the DS response, the molecular mechanisms through which ERF members regulate DR in wheat remain unclear. Further exploration of genes associated with these responses is thus necessary to elucidate the DR regulatory network.

The WRKY family members in plants are characterised by one to two highly conserved WRKY domains (Ross et al. 2007). The N‐terminus of the WRKY domain features a well‐conserved motif (WRKYGQK), while the C‐terminus contains a zinc‐finger motif (Hinderhofer and Zentgraf ²⁰⁰¹). WRKY proteins can attach to the W‐box (TTGACC/T) or analogous sequences found in the downstream regions of genes because of the configuration of the WRKY domain, thereby stimulating the expression of downstream targets (Rushton et al. 2012; Yu et al. ²⁰²³). A growing body of evidence suggests that WRKYs serve a crucial role in helping plants physiologically adapt to DS. OE of WRKY genes, such as AtWRKY1, AtWRKY46, AtWRKY57, and AtWRKY63, leads to significant improvement in DT in Arabidopsis (Ren et al. 2010; Jiang et al. ²⁰¹²; Ding et al. ²⁰¹⁴; Qiao et al. ²⁰¹⁶). In wheat, WRKY genes, such as TaWRKY1, TaWRKY1‐2D, TaWRKY2, TaWRKY10, TaWRKY19, TaWRKY33, TaWRKY40, and TaWRKY75, respond to water deficiency (Niu et al. 2012; He et al. ²⁰¹⁶; Gao et al. ²⁰¹⁸; Yu et al. ²⁰²³). Notably, similar observations have been made with regard to OsWRKY11 and OsWRKY45 in rice (Tao et al. 2011; Wu et al. ²⁰⁰⁹), SbWRKY30 and SbRD19 in sorghum (Yang et al. 2020), and SlWRKY8 in tomato (Gao et al. 2020). These studies collectively reveal the potential of WRKY TFs in crop engineering to improve water efficiency, especially under DS.

DR is a complex trait controlled by multiple genes in plants. Notably, modularised regulation often occurs. The WRKY–ERF module is crucial for regulating drought resistance in plants. For instance, GhWRKY59 influences the ABA‐independent gene GhDREB2 by attaching to the W‐box within the promoter, thereby responding to drought signals in cotton (Li et al. 2017). Similarly, the BdWRKY36‐NtDREB3 module in tobacco reduces reactive oxygen species accumulation, thereby significantly enhancing plant drought resistance (Sun et al. 2015). These reports suggest that the WRKY–ERF module may also regulate drought resistance in wheat, though follow‐up studies are needed.

Wheat is primarily grown in arid and semi‐arid areas. Approximately 50% of global wheat production areas experience DS, resulting in yield reductions of 10%–70% (Curtis and Halford 2014; Piao et al. ²⁰¹⁰; Dhakal ²⁰²¹). The identification and analysis of crucial genes involved in DR are thus necessary for DR improvement in wheat. Although research on the ERF subfamily has advanced plant DR efforts, these studies have primarily focused on Arabidopsis and rice. Notably, only 19 of the 112 ERF subfamily members in the wheat genome have been characterised as responding to DS (Magar et al. 2022; Yu et al. ²⁰²²). As such, research on DR‐related members of the wheat ERF subfamily is relatively scarce. Moreover, exploration of upstream regulators and targets of ERFs involved in DT in wheat is still in its early stages, which limits the understanding of ERF genes and their regulatory networks. RNA‐seq data suggest that the ERF gene TaERF1‐A is markedly upregulated under DS. Its promoter contains numerous abiotic stress‐related cis‐acting elements, suggesting its role in DS response (Yu et al. 2022). The present study aimed to identify and characterise wheat TaERF1‐A and to investigate its potential DS response roles. To this end, transgenic Arabidopsis and wheat lines were generated and systematically evaluated under DS to assess the biological function of TaERF1‐A. This study also aimed to explore the molecular mechanisms underlying TaERF1‐A‐mediated drought responses, with particular emphasis on its possible involvement in proline biosynthesis. Specifically, the study examined whether TaERF1‐A regulates the expression of TaP5CS2 by interacting with cis‐regulatory elements in its promoter region. Additionally, the upstream transcriptional regulation of TaERF1‐A was investigated, with a focus on the potential role of TaWRKY33 and its interaction with W‐box motifs. This study integrated functional analyses and regulatory relationships to elucidate a regulatory pathway involving TaWRKY33, TaERF1‐A, and TaP5CS2, which potentially contributes to the adaptive DR in wheat. The findings provide key mechanistic insights into transcriptional regulation of proline metabolism under DS and empirical guidance for wheat drought‐resistance improvement and future molecular breeding strategies.

2. Materials and Methods

2.1. Plant Materials and Abiotic Stress Treatments

The wheat variety JW1 was used for tissue‐specific and abiotic stress‐related gene expression analyses. Arabidopsis (ecotype Columbia) was used to create overexpression (OE) lines. The wheat variety XN979 was also used to generate wheat gene‐silencing lines, while the wheat variety Fielder was used to create overexpression (OE) and RNAi (RI) lines. Etiolated seedlings, that is, seeds germinated in the darkness for 2 days, of the wheat variety JW1 were divided into two groups. One group was grown in a greenhouse maintained at 15°C–20°C, with a 16/8 h light/dark cycle until the seventh day after flowering, when roots, leaves, stems, grain, and spikes were harvested for tissue‐specific gene expression analysis under normal growth conditions. For PEG‐induced dehydration and salt treatments, the second group was grown in a climate‐controlled incubator set at 20°C–25°C under a 16/8 h light/dark cycle for 14 days, after which they were treated with 20% PEG6000 and 200 mM NaCl to ensure effective induction of stress responses. For the natural drought treatment, 14‐day‐old JW1 seedlings grown under well‐watered conditions were subjected to rapid dehydration by placement on dry filter paper under 40%–60% relative humidity. Leaf and root samples were subsequently collected at 1, 6, 12, and 24 h after treatment. Leaves from untreated plants (0 h) served as controls. The collected samples were subjected to RNA extraction for quantitative expression analysis and for use as templates for subsequent gene cloning. To further investigate gene expression patterns under PEG and natural drought stress across different wheat varieties, besides JW1, Xinong979 (XN979, drought‐tolerant) and Yangmai 9 (YM9, drought‐sensitive) varieties were included (Wang et al. 2019; Liu et al. ²⁰²¹), and the sampling procedure was consistent across all varieties.

2.2. Gene Isolation, qRT‐PCR, Sequence Alignment, and Phylogenetic Analysis

The cDNA of the untreated JW1 leaf tissue (0 h) was used as a template for gene cloning. Wheat TaERF1‐A (TraesCS6A02G243300) and TaWRKY33 (TraesCS6B02G175100) genes were amplified separately using their respective sequence‐specific primers (Table S1). RNA extraction, cDNA synthesis, and gene expression assays were performed following the methods outlined by Yu et al. (2023). DNAMAN 9.0 was employed to analyze multiple‐sequence alignment, while a phylogenetic tree was constructed using the neighbour‐joining algorithm implemented in MEGA 7.0.

2.3. Subcellular Localisation and Transcriptional Activity Assays of TaERF1‐A

The full‐length sequence of TaERF1‐A, but without a termination codon, was cloned into a 16318‐35S‐GFP plasmid. The 16318:GFP‐TaERF1‐A construct and control vector (16318:GFP‐empty) were then introduced into wheat protoplasts for transient expression. The fluorescence signal from GFP was detected using a laser confocal microscope.

The complete coding sequence (272 aa), including truncated segments representing the unique N‐terminal (1–127 aa) and C‐terminal (128‐272 aa) regions of TaERF1‐A, was cloned into pGBKT7 to create recombinant plasmids. These vectors were then transformed into AH109 yeast cells. The cells were subsequently grown on synthetic dropout (SD) medium plates (‐Trp and ‐Trp/‐His/‐Ade) to evaluate self‐activation.

For transient expression assays in tobacco (Nicotiana benthamiana), the coding sequence of TaERF1‐A was inserted into a modified GAL4‐DBD vector in order to generate the effector construct. The reporter construct consisted of a firefly luciferase gene with expression driven by a promoter containing GAL4 upstream activating sequences (GAL4 UAS:LUC), along with a Renilla luciferase (REN) gene under the control of the CaMV35S promoter as an internal reference. The effector and reporter plasmids were co‐infiltrated into tobacco leaves via Agrobacterium‐mediated transformation. After incubation, LUC and REN luciferase activities were quantified using a dual‐luciferase assay system. Each treatment was performed with three independent biological replicates.

2.4. Arabidopsis Overexpression and Wheat Virus‐Induced Gene Silencing of TaERF1‐A

Generation of TaERF1‐A transgenic Arabidopsis plants was performed as described by Yu et al. (2022). The TaERF1‐A coding sequence (CDS) was cloned into a modified pBI121 expression vector under the control of the CaMV35S promoter. The recombinant plasmid was introduced into wild‐type Arabidopsis ecotype Columbia‐0 (Col‐0) via the Agrobacterium strain GV3101 using the floral dip method. Transgenic T₁ plants were selected on ½‐strength Murashige and Skoog (MS) medium containing 50 mg/L kanamycin, and resistant seedlings were transferred to soil and grown to maturity to obtain T₂ seeds. T₂ plants showing a 3:1 segregation ratio for kanamycin resistance were identified as single‐locus insertions, and homozygous T₃ lines were subsequently obtained and used for functional analyses. Sterilised Arabidopsis seeds were first stratified at 4°C for 3 days and then sown on solid ½‐strength MS medium with or without 250 mM mannitol. The plates were transferred to a light incubator, and germination (defined as visible radicle emergence) was recorded daily for 7 days. Three independent biological replicates were performed for the germination assay, each consisting of 50 dry mature seeds. For root growth analysis, seedlings grown on solid MS medium containing 250 mM mannitol were vertically cultured for 7 days in a light incubator, after which root length was measured. In another assay, 3‐week‐old transgenic Arabidopsis expressing TaERF1‐A were not watered for 14 days, followed by a 3‐d rewatering phase, to analyze their drought survival rates.

For barley stripe mosaic virus (BSMV)–virus‐induced gene silencing (VIGS), pCaBS (α, β, and γ) vectors were constructed as described by Yuan et al. (2011). A conserved TaERF1‐A cDNA segment (134 bp) was inserted into pCaBS‐γ, with the intention of conferring the simultaneous silencing of TaERF1‐A (Figure S1A). The construction of the recombinant vector BSMV‐TaERF1‐A, its application in the VIGS system, transformation into Agrobacterium strains, and inoculation of plants were conducted following the methodology outlined by Yuan et al. (2011). An elite drought‐tolerant wheat variety, XN979, was used in the VIGS trial owing to its well‐established suitability for VIGS‐mediated gene silencing (Wang et al. 2019). The leaves of XN979 at the two‐leaf stage were inoculated with BSMV‐TaERF1‐A Agrobacterium. BSMV‐PDS and BSMV: γ were set as the positive and negative controls, respectively. Symptoms were noted and photographed at 10 days postinoculation. TaERF1‐A‐silenced wsheat plants were not watered for 9 days to analyze their phenotypic response to drought conditions.

2.5. Wheat Transformation and Drought Tolerance Assays

The coding sequences of TaERF1‐A (TraesCS6A02G243300), TaWRKY33 (TraesCS6B02G175100), and TaP5CS2 (TraesCS3B03G0987700) were amplified from the wheat cultivar XN979 and cloned into pWMB006. The resulting recombinant plasmids were digested using HindIII and EcoRI, followed by subcloning of the released inserts into the corresponding sites of pCAMBIA3301 to generate the corresponding overexpression (OE) constructs. A 245‐bp fragment of the TaERF1‐A coding region was amplified and inserted into the BamHI/KpnI and SacI/SpeI sites of pWMB006. The resulting plasmid was digested with HindIII and EcoRI. The RNAi (RI) fragment was subsequently cloned into pCAMBIA3301 to generate a TaERF1‐A RI construct. The constructs were introduced into Agrobacterium tumefaciens strain EHA105 and used to transform immature embryos of Fielder wheat based on a modified Agrobacterium‐mediated transformation protocol (Wang et al. 2017). The (RNAi) RI lines and OE lines derived from T₂ plants for each transgene, along with wild‐type (WT) plants, were grown in an artificial climate chamber with a 16 h light/8 h dark cycle and 16°C/14°C day/night temperature. Seedlings at the two‐leaf stage were subjected to DS by withholding water for 25 days, after which they were re‐watered and evaluated for survival. Images of the seedlings were captured 4 days after re‐watering. DT was evaluated based on the survival rate.

2.6. Transcriptional Levels of TaERF1‐A in Transgenic Plants Under Drought Stress

TaERF1‐A transgenic Arabidopsis seedlings were harvested before (21 days) and after drought treatment (7 days) and measured for AtRD29A, AtP5CS1, AtDREB2A, AtHSP70, AtPOD1, AtCAT1, and AtSOD (Cu/Zn) expression. TaERF1‐A transgenic wheat plants were not watered for 9 days in order to quantify the expression of TaDREB1, TaERF3, TaPOD, TaCAT, and TaSOD(Fe) genes.

2.7. Determination of Physiological Indexes

Chlorophyll content was determined following the method described by Porra et al. (1989). Chlorophyll was extracted from fresh leaf tissue using 80% acetone at 4°C. The absorbance of the extracts was measured at 645 and 663 nm using a spectrophotometer. Three biological replicates were analyzed, with each replicate consisting of leaves from four individual plants. Peroxidase (POD), catalase (CAT), and superoxide dismutase (SOD) activity levels were measured as described by Li et al. (2020). Malondialdehyde (MDA), proline, and O₂‐ levels were determined using their respective detection kits sourced from Solarbio Life Sciences (Beijing, China) (Ma et al. 2022).

2.8. Inference of Protein–Protein Associations

The protein–protein interaction network was constructed using STRING based on the Arabidopsis model. Notably, the wheat wGRN database further confirmed that wheat TaERF1‐A is an upstream regulatory factor.

2.9. Protein Interaction Assay

The yeast two‐hybrid (Y2H) screening assay was conducted as described by Yu et al. (2023). Briefly, the full‐length TaERF1‐A CDS was inserted into pGBKT7, while the CDSs of TaG6PE, TaUn, TaFBA, TaCSP41b, and TaNF‐YC6 were inserted into pGADT7 vectors. The resulting BD‐TaERF1‐A plasmids were co‐transformed into yeast with the resulting AD recombinant plasmids (TaG6PE, TaUn, TaFBA, TaCSP41b, and TaNF‐YC6).

2.10. Electrophoretic Mobility Shift Assay

The CDS of TaWRKY33/TaERF1‐A was cloned into pCold‐MBP using a Biomed Seamless Cloning and Assembly Kit to generate an MBP‐tagged fusion protein. An empty vector was used to express and purify MBP solely as a negative control. Biotin‐labelled DNA probes containing the putative cis‐elements recognised by TaWRKY33/TaERF1‐A were synthesised for an electrophoretic mobility shift assay (EMSA). In contrast, unlabelled probes were used as competitors to verify the binding specificity. Mutated probes in which the W‐box/GCC‐box motifs were replaced by “AAAAAA” sequences were used as non‐competitive controls to further verify the binding specificity. EMSA was performed using the Thermo Fisher Scientific LightShift Chemiluminescent EMSA Kit (20148; Thermo Fisher Scientific, Waltham, MA, USA) (Chu et al. 2024).

2.11. Dual‐Luciferase Assay

The 1500 bp upstream sequence of the start codon (ATG) was defined as the putative promoter region based on common practice in plant promoter studies (Liu et al. 2026). This region was retrieved from the wheat reference genome and analyzed using the PlantCARE database. Multiple stress‐responsive cis‐elements were identified, including W‐box elements, which are recognised by WRKY transcription factors. Accordingly, this fragment was used for subsequent dual‐luciferase and yeast one‐hybrid assays. A promoter region of 1500 bp from TaERF1‐A was inserted into the pGreenII 0800‐LUC plasmid to investigate the effect of TaWRKY33 on the promoter activities of TaERF1‐A. The reconstructed pGreenII 62‐SK harbouring the entire TaWRKY33 CDS was used to measure the transcriptional activity of TaWRKY33. TaERF1‐A‐LUC and 62SK‐TaWRKY33 plasmids were transformed into the GV3101 strain and then injected into N. benthamiana. A luciferase assay was performed 3 d after transfection using a double‐luciferase system of Promega.

2.12. Yeast One‐Hybrid Assay

The CDSs of TaWRKY33 and TaERF1‐A were individually cloned into pGADT7, while the promoter regions of TaERF1‐A and TaP5CS2 were inserted into pHIS2, respectively. Of note, 3‐AT was added to the medium to inhibit background expression from the pHIS2–TaERF1‐A/TaP5CS2‐Apro constructs. Yeast Y187 was subsequently transformed with the fusion constructs. Co‐transformation with empty vectors served as the negative control, while 3‐AT was used to suppress background growth.

3. Results

3.1. Identification and Characterisation of the TaERF1‐A Gene

The TaERF10‐6A (TraesCS6A02G243300) gene, which was markedly upregulated under PEG‐induced dehydration conditions, was identified in transcriptome databases documenting DS (Yu et al. 2022). Its promoter contains ABRE, W‐box, LTR, and other abiotic stress response elements, suggesting that it potentially serves a vital role in cold and DS responses. The ORF of TaERF‐10‐6A was cloned from the total RNA extracted from the wheat variety JW1 by RT‐PCR. Sequence analysis conducted in a previous study indicated that the ORF spans 819 bp and encodes a polypeptide consisting of 272 amino acids (Yu et al. 2022). Phylogenetic analysis revealed that TaERF1‐A is similar to gene family members in other monocots (e.g., OsERF1), less similar to dicot gene family members (e.g., AtERF1) (Figure 1A). The coding sequences of TaERF1‐A were determined to be identical among 15 wheat varieties, including JW1 and XN979, and all are located on chromosome 6 A (Figure S1B). We thus renamed TaERF‐10‐6A to TaERF1‐A based on their evolutionary relationships (Figure 1A). Sequence analysis revealed that TaERF1‐A possesses a conserved AP2 domain comprising 58 amino acids. Notably, the domain is segmented into the conserved YRG and RAYD regions. Moreover, the AP2 domain of TaERF1‐A exhibited conserved amino acids at positions 14 (A, alanine) and 19 (D, aspartic acid) (Figure S2). Notably, this finding was consistent with the ERF subfamily profile. These findings suggest that wheat TaERF1‐A may have the basic functions of plant ERF1 members. The highest transcript levels of TaERF1‐A were observed in the roots, while lower levels were observed in the leaves, stems, spikes, and grains of JW1 (Figure 1B). TaERF1‐A demonstrated minor upregulation at 1 h, increased by 6 h, fell back to baseline by 12 h, and then reached the peak at 24 h in response to salt stress (Figure S3). To further characterise the expression dynamics of TaERF1‐A under PEG‐simulated DS, its transcript levels were examined in the roots and leaves of three wheat cultivars (JW1, XN979, and YM9). Under PEG treatment, TaERF1‐A expression in both roots and leaves of JW1 and XN979 displayed a transient induction pattern, with an initial increase followed by a decline (Figure S4A). In contrast, YM9 exhibited the opposite trend, characterised by initial downregulation followed by a gradual recovery (Figure S4A). Under natural DS, TaERF1‐A expression was rapidly induced in the leaves of JW1 and XN979, with expression levels continuing to increase more slowly after 12 h. In the roots of JW1 and XN979, TaERF1‐A expression increased slowly before 12 h and more rapidly thereafter (Figure S4B). In contrast, TaERF1‐A expression in the roots and leaves of YM9 was initially downregulated and then gradually increased (Figure S4B).

Figure 1.

Expression and protein characterisation of TaERF1‐A. (A) Phylogenetic analysis of wheat TaERF1‐A with other plant AP2/ERF proteins from Triticum aestivum (Ta), Oryza sativa (Os), Zea mays (Zm), and Arabidopsis thaliana (At). (B) Tissue‐specific expression levels of TaERF1‐A in wheat root, stem, leaf, spike, and grain tissues. (C) Subcellular localisation of TaERF1‐A protein in wheat protoplasts (scale bars, 10 μM). The mCherry signal indicates the position of the nucleus. The co‐localisation of TaERF1‐A–GFP with the mCherry signal confirms its nuclear localisation. Statistical significance was determined by one‐way ANOVA followed by Tukey's test. Different letters indicate significant differences at p < 0.05.

A transient TaERF1‐A expression vector was constructed and transformed into wheat mesophyll protoplasts to confirm that TaERF1‐A is localised to the nucleus, as is expected for a TF. Confocal microscopy revealed that TaERF1‐A GFP fluorescence was mainly distributed in the nucleus, clearly indicating nuclear localisation of TaERF1‐A (Figure 1C). Yeast transcriptional activity assays were performed to determine whether TaERF1‐A functions as a transcriptional activator. The full‐length (TaERF1‐A_{1‐272 aa}), N‐terminal (TaERF1‐A_1‐127aa), and C‐terminal (TaERF1‐A_128‐272aa) sections of the TaERF1‐A cDNA were fused to pGBKT7 (GAL4BD) to generate recombinant plasmids (pGBKT7‐TaERF1‐A _1‐272aa, pGBKT7‐TaERF1‐A _1‐127aa, and pGBKT7‐TaERF1‐A _128‐272aa). Notably, the three transformants and the control grew well on 1 × Dropout medium (Figure S5A). In contrast, only the yeast cells harbouring the pGBKT7‐TaERF1‐A _1‐127aa grew well on × Dropout (Figure S5A). This finding indicates that TaERF1‐A lacks a transcriptional activation effect in yeast, while the N‐terminal region functions as an activator. Additionally, dual‐luciferase reporter assays demonstrated that fusion of TaERF1‐A with the yeast GAL4 DNA‐binding domain significantly enhanced the transcriptional activity of a LUC reporter driven by a minimal cauliflower mosaic virus CaMV35S promoter containing GAL4 upstream activating sequences (Figure S5B). These results indicate that TaERF1‐A possesses transcriptional activation activity in planta.

3.2. Functional Characterisation of Drought Resistance in TaERF1‐A Transgenic Arabidopsis and Transiently Silenced Wheat Plants

Agrobacterium integrated the TaERF1‐A cDNA segment between the RB and LB of the pBI121 vector into Arabidopsis (Figure 2A). Three Arabidopsis OE lines, that is, OE1, OE2, and OE3, had detectable transcription of TaERF1‐A based on RT‐PCR results. In contrast, TaERF1‐A transcripts were not detectable in WT plants (Figure 2B), indicating that TaERF1‐A had been successfully transformed into Arabidopsis. Under normal conditions, germination rates did not significantly differ between WT and TaERF1‐A‐overexpressing Arabidopsis plants. However, under mannitol treatment, the germination rate of TaERF1‐A‐overexpressing lines was significantly lower than that of WT plants (Figure S6). Notably, the transgenic OE lines (7‐day‐old seedlings) grew better at the seedling stage than WT plants under 250 mM mannitol treatment (Figure 2C). The root lengths of OE plants were greater than those of the WT plants under normal conditions or mannitol treatment (Figure 2D–E). Three‐week‐old Arabidopsis plants exhibited severe wilting in most WT specimens after 14 days of water deprivation followed by 3 days of re‐watering (Figure 2F). The survival rate (SR) of transgenic OE lines following a DS period of 14 days and 3 days of re‐watering was 68.6%, which significantly exceeded the SR among WT plants, 45.0% (Figure 2G). Under DS, chlorophyll content decreased in all lines, whereas overexpression plants retained significantly higher levels than the WT plants (Figure S7). This finding suggests that the OE lines exhibit greater DT compared to WT plants.

Figure 2.

Analysis of the drought tolerance (DT) of transgenic Arabidopsis. (A) Schematic diagram of pBI121 (CaMV35S). CDS of TaERF1‐A was cloned into pBI121 between its BamHI and SacI restriction sites. (B) RT‐PCR verification of Arabidopsis TaERF1‐A overexpression (OE) and wild‐type (WT) lines. Root length phenotype (C) and data statistics (D–E) of transgenic and WT Arabidopsis (scale bars, 1 cm). Phenotypic (F) and survival rate (SR) statistics (G) of transgenic Arabidopsis under drought stress (DS). Images of Arabidopsis plants captured before DS and after 14 days of drought followed by 3 days of re‐watering. Gene expression levels of stress‐responsive and antioxidant‐related gene family members in Arabidopsis leaves of TaERF1‐A OE and WT lines before (H) and after (I) DS exposure. Data are presented as mean ± SD of three biological replicates (n = 3), with 40 plants per line used in each replicate. Two asterisks (**p < 0.01) denote significant differences based on 2‐sided t‐test. [Color figure can be viewed at wileyonlinelibrary.com]

Several well‐known stress and antioxidant‐related members were analyzed to investigate the potential tolerance mechanism by which TaERF1‐A exerts regulatory effects under DS. Notably, there was no difference in the transcript levels of AtPOD1, AtCAT1, and AtSOD (Cu/Zn) between the WT and OE lines before DS. In contrast, the transcript levels of AtRD29A, AtP5CS1, AtDREB2A, and AtHSP70 were all higher in the Arabidopsis OE lines (Figure 2H). The transcript levels of AtRD29A, AtP5CS1, AtDREB2A, AtHSP70, AtPOD1, AtCAT1, and AtSOD (Cu/Zn) were significantly higher in the transgenic Arabidopsis plants than in the WT after DS (Figure 2I). These results suggested that the enhanced DT of TaERF1‐A transgenic Arabidopsis was potentially associated with the expression of corresponding stress‐ and antioxidant‐related members.

All plants of the wheat variety XN979 infected with BSMV _PDS exhibited bleaching 10 days post inoculation. In contrast, those treated with BSMV₀ and BSMV _TaERF1‐A exhibited stripe chlorotic symptoms (Figure S8). BSMV _TaERF1‐A ‐treated wheat exhibited leaf wilting and grew slowly after drought treatment. In contrast, BSMV₀‐treated wheat plants exhibited mild symptoms only and were upright (Figure S9A). Notably, TaERF1‐A transcript level was downregulated in TaERF1‐A‐silenced wheat plants (Figure S9B). No physiological indices differed between BSMV₀‐treated and TaERF1‐A‐silenced wheat plants under normal conditions (Figure S9C). However, the content of proline and the activities of the POD, CAT, and SOD enzymes in TaERF1‐A‐silenced wheat plants were considerably reduced, while the MDA level was elevated compared to the corresponding content and activities in BSMV₀‐treated wheat plants under DS (Figure S9C).

3.3. Functional Investigation of Drought Resistance in TaERF1‐A Stable Transgenic Wheat Lines

TaERF1‐A RNAi (RI) and OE wheat lines were generated through Agrobacterium‐mediated transformation to further investigate the role of TaERF1‐A in DT. The OE lines exhibited a pronounced increase in TaERF1‐A transcript abundance, whereas the RI plants displayed a substantial reduction compared to the WT plants (Figure 3A). Notably, both transgenic lines exhibited growth and morphology comparable to WT plants under well‐watered (WW) conditions despite the contrasting gene expression levels (Figure 3B). OE plants demonstrated significantly enhanced tolerance, achieving an 83%–84.4% seedling survival rate (SR), compared to just 30% in WT plants under water‐deficit (WD) conditions. In contrast, the RI lines were significantly more susceptible, exhibiting a considerable decrease in SR (Figure 3C). The proline and chlorophyll contents in TaERF1‐A OE lines were markedly higher compared to the corresponding content in WT plants under WW and drought conditions. In contrast, RI lines exhibited reduced proline and chlorophyll levels specifically under DS (Figure 3D, Figure S10). Notably, antioxidant enzyme activities were markedly higher in OE lines compared to WT plants under DS, whereas MDA content was markedly lower despite there being no differences in MDA, POD, CAT, or SOD levels between OE plants and WT plants under full irrigation. In contrast, RI lines exhibited the opposite trend (Figure 3E–H). These results collectively demonstrate that altered TaERF1‐A expression significantly influences proline accumulation, thereby affecting DT in wheat.

Figure 3.

Drought resistance evaluation of TaERF1‐A stable transgenic wheat plants. (A) Expression profiles of TaERF1‐A in RNAi (RI), overexpression (OE), and wild‐type (WT) wheat plants. (B) Drought tolerance (DT) of RI, OE, and WT plants, photographed pre‐ and post‐drought treatment and after 4 days of re‐watering. (C) Post‐drought survival rates (SRs) of RNAi, OE, and WT plants. (D–H) proline and Leaf malondialdehyde (MDA) levels, along with the peroxidase (POD), catalase (CAT), and superoxide dismutase (SOD) activities, in WT and transgenic plants under well‐watered (WD) and water‐deficient (WD) conditions. Data are presented as mean ± SD of three biological replicates (n = 3), with 45 plants per line used in each replicate. The asterisks (*p < 0.05; **p < 0.01) denote significant differences based on two‐sided t‐test. [Color figure can be viewed at wileyonlinelibrary.com]

3.4. Analysis of Stress‐ and Antioxidant‐Related Genes in Transgenic Wheat

TaDREB1, TaERF3, TaPOD, TaCAT, and TaSOD (Fe) are important marker genes that respond to DS in wheat. Notably, their variation may influence the drought resistance of wheat. The expression of stress‐response members may cause changes in physiological indices. The transcriptional levels of stress response (TaDREB1 and TaERF3) and antioxidant enzyme (TaCAT, TaPOD, and TaSOD (Fe)) genes in TaERF1‐A transgenic wheat plants under DS were thus analyzed. Notably, there was a decrease in the transcription levels of TaDREB1, TaERF3, TaPOD, TaCAT, and TaSOD (Fe) in TaERF1‐A‐RI wheat lines under WD conditions compared to the WT plants. In contrast, these members were upregulated in the OE lines (Figure 4A). The LUC/REN ratios of TaDREB1, TaERF3, TaCAT, and TaSOD (Fe) were significantly increased compared to those of the control (Figure 4B–C), revealing that TaERF1‐A can bind to the promoters of these members and activate their expression. These results indicate that the expression of these genes is directly or indirectly regulated by TaERF1‐A.

Figure 4.

Transactivation assays of TaERF1‐A in tobacco and screening of TaERF1‐A‐interacting proteins. (A) Gene expression levels of TaDREB1, TaERF3, TaPOD, TaCAT, and TaSOD(Fe) in transgenic TaERF1‐A overexpression (OE), TaERF1‐A RNAi (RI), and wild‐type (WT) wheat lines under water deficit (WD) conditions. (B) Schematic diagram showing the reporter and effector constructs utilised in the co‐transfection. (C) Dual‐luciferase assay in tobacco leaves demonstrating the transactivation effect of TaERF1‐A on the promoter of TaDREB1, TaERF3, TaPOD, TaCAT, and TaSOD(Fe). Data are shown as mean ± SD values from three independent replicates, with statistical significance determined using a two‐tailed Student's t‐test (*p < 0.05, **p < 0.01). [Color figure can be viewed at wileyonlinelibrary.com]

3.5. Screening of TaERF1‐A‐Interacting Proteins

The TaERF1‐A (1–272 aa) protein, which harbours a single ERF domain, was used as a bait protein to screen the cDNA library of yeast to reveal the potential molecular pathways by which wheat TaERF1‐A regulates drought resistance. Five candidate proteins containing complete ORFs were identified as potential interaction partners of TaERF1‐A and were subsequently annotated in the NCBI database: TaG6PE (XM_044541187.1), TaUn (XM_044511767.1), TaFBA (XM_044508333.1), TaCSP41b (XM_044539082.1), and TaNF‐YC6 (XM_044533290.1) (Table S2). The recombinant vectors pGADT7‐TaG6PE, pGADT7‐TaUn, pGADT7‐TaFBA, pGADT7‐CSP41b, and pGADT7‐TaNF‐YC6 were co‐transformed individually into the AH109 yeast strain harbouring pGBKT7‐TaERF1‐A. Notably, all the co‐transformed strains grew on SD/‐Trp/‐Leu and SD/‐Trp/‐Leu/‐His/‐Ade/media containing X‐α‐Gal (Figure S11A), indicating that TaERF1‐A interacted with the five candidates in the yeast system. Among the five interacting proteins, nuclear factor Y (NF‐Y) plays a particularly important role in plant responses to DS. Thus, wheat TaNF‐YC6 protein was selected for further analysis. Bimolecular fluorescence complementation was used to further verify the interaction between TaERF1‐6A and TaNF‐YC6. YFP fluorescence was observed when TaERF1‐6A and TaNF‐YC6 were transferred into tobacco, whereas no YFP fluorescence was observed in the control combination (Figure S11B). Thus, TaERF1‐6A and TaNF‐YC6 were observed to interact with each other in tobacco cells. These results indicate that TaERF1‐A may cooperate with the identified interacting proteins during DS as part of a complex regulatory network that orchestrates multiple pathways involved in the drought response. These findings collectively suggest that TaERF1‐A integrates transcriptional and protein‐level regulation within a complex regulatory network, thereby coordinating multiple pathways involved in abiotic stress tolerance.

3.6. Identification of Upstream Regulatory Proteins of TaERF1‐A

Proteins that potentially interact with or have other relationships with TaERF1‐A were screened using the STRING database to delve deeper into the possible molecular mechanisms associated with TaERF1‐A. Notably, a potential co‐expression gene named TaWRKY33, which is a homologue of AtWRKY40, was uncovered (Figure 5A). It has been previously postulated that TaWRKY33 is upregulated under DS, activates stress response genes (including AtRD29A, AtDREB2B, AtABA1/2, and AtABI1/5), promotes root growth, increases the germination rate, and enhances the DT of transgenic Arabidopsis. As such, TaWRKY33 could be essential to the DS response of wheat (He et al. 2016). Herein, 286 TFs that potentially regulate the transcription of TaERF1‐A were predicted using the wheat wGRN database. Notably, 80% of the top 10 TFs with the strongest correlations were members of the WRKY TF family, with TaWRKY33 ranked seventh (Table S3; Figure S12A). Two W‐box elements (GGTCAA) were identified in the TaERF1‐A promoter (Figure S12B), indicating that TaWRKY33 potentially acts as its transcriptional regulator.

Figure 5.

TaWRKY33 binds to the TaERF1‐A promoter and activates its expression. (A) TaERF1‐A protein interaction networks. The letter in parentheses represents Arabidopsis homologues in wheat. (B) Y1H assay results showing that TaWRKY33 binds to the TaERF1‐A promoter. (C) EMSA assay results showing that TaWRKY33 can specifically recognise and bind to the W‐box element. (D) Dual‐luciferase assay results of tobacco leaves demonstrating the transactivation effect of TaWRKY33 on the promoter of TaERF1‐A. (E) TaERF1‐A expression analysis of related genes in TaWRKY33 transgenic plants. Data are shown as mean ± SD values from three independent replicates, with statistical significance determined using a two‐tailed Student's t‐test (*p < 0.05). [Color figure can be viewed at wileyonlinelibrary.com]

Yeast one‐hybrid (Y1H), electrophoretic mobility shift, and dual‐luciferase assays were conducted to further validate this hypothesis. In the Y1H assay, the Y187 strain co‐transformed with pGADT7‐TaWRKY33 and pHIS2‐TaERF1‐A _pro grew on medium containing 75 mM 3‐amino‐1,2,4‐triazole (3‐AT). In contrast, the strain carrying pHIS2‐Empty did not grow under the same selection conditions (Figure 5B). EMSA confirmed that TaWRKY33 directly bound to the TaERF1‐A promoter, whereas the mutated probe failed to show any binding affinity (Figure 5C). The dual‐luciferase assay revealed a significant increase in the LUC/REN ratio compared to that of the control (Figure 5D), which indicated activation of the promoter. TaWRKY33 was subsequently overexpressed in wheat to investigate the expression levels of TaERF1‐A. Notably, TaERF1‐A was significantly upregulated in TaWRKY33 transgenic plants (Figure 5E). These results collectively demonstrate that TaWRKY33 activates and enhances the expression of TaERF1‐A in wheat via a W‐box‐dependent manner.

3.7. Characterisation of TaWRKY33 and Its Functional Analysis Under Drought Stress

Phylogenetic analysis revealed that TaWRKY33 is closely related to monocot homologues (ZmWRKY71), but is less similar to dicot homologues (AtWRKY18) (Figure 6A). TaWRKY33 contains a canonical WRKY domain and a C‐X_4–5‐C‐X₂₃‐H‐X₁‐H (C₂H₂) type zinc‐finger motif, and is classified into the Group II‐a subfamily, together with ZmWRKY71 and OsWRKY71 (Figure 6B). TaWRKY33 showed the highest transcript abundance in roots of JW1 wheat, while its expression in the leaves, stems, spikes, and grains was comparatively lower (Figure 6C). Under PEG‐induced stress, TaWRKY33 was rapidly upregulated and then gradually downregulated in both roots and leaves of both JW1 and XN979, whereas it showed an initial decrease followed by a gradual increase in YM9 (Figure S13A). Similarly, under natural DS, TaWRKY33 was progressively upregulated over time in both JW1 and XN979, while it exhibited a gradual downregulation in YM9 (Figure S13B). Notably, both its tissue‐specific expression pattern and drought‐responsive expression dynamics closely resembled those of TaERF1‐A.

Figure 6.

Expression and protein characterisation of TaWRKY33. (A) Phylogenetic analysis of wheat TaWRKY33 with other plant WRKY proteins from Triticum aestivum (Ta), Zea mays (Zm), Oryza sativa (Os), and Arabidopsis thaliana (At). (B) Protein sequence and domain analysis of TaWRKY33. (C) Tissue‐specific expression levels of TaWRKY33 in wheat. All experiments were performed with three biological replicates. Statistical significance was determined by one‐way ANOVA followed by Tukey's test. Different letters indicate significant differences at p < 0.05. [Color figure can be viewed at wileyonlinelibrary.com]

TaWRKY33 OE wheat lines were generated via Agrobacterium‐mediated transformation in order to evaluate the role of TaWRKY33 in DT. The OE plants exhibited significant elevation of TaWRKY33 transcript levels compared to WT plants (Figure 7A). However, there were no discernible morphological differences observed between OE and WT plants under well‐watered (WW) conditions (Figure 7B). In contrast, drought treatment results revealed clear phenotypic divergence. TaWRKY33 OE seedlings displayed substantially enhanced DT, achieving a survival rate of 74.1%–77.8%, whereas WT seedlings had a survival rate of only 31.1% (Figure 7C). In addition, there was an increase in proline levels in the TaWRKY33 OE plants relative to WT plants under WW and drought conditions (Figure 7D). In normal conditions, no disparities in chlorophyll concentration were observed between the WT and OE lines. Chlorophyll concentration decreased under DS, although the extent of reduction varied. Compared with WT plants, the chlorophyll concentration was significantly higher in the transgenic lines after DS (Figure S14). DS led to a pronounced reduction in MDA accumulation in the OE lines (Figure 7D), despite the comparable MDA content between OE and WT lines under full irrigation. These findings collectively indicate that TaWRKY33 enhances proline biosynthesis and acts as a positive regulator of DT in wheat, mirroring the functional role reported for TaERF1‐A.

Figure 7.

Drought resistance evaluation of wheat TaWRKY33. (A) Expression profiles of TaWRKY33 in RNAi (RI) and overexpression (OE) wheat plants as well as wild‐type (WT) wheat plants. (B) Drought tolerance (DT) of RI, OE, and WT plants, photographed pre‐ and post‐drought treatment and after 4 days of re‐watering. (C) Post‐drought survival rates (SRs) of RI, OE, and WT plants. (D) Leaf proline and malondialdehyde (MDA) levels in WT and transgenic plants under Well‐watered (WW) and water‐deficient (WD) conditions. Data are shown as mean ± SD values from three independent replicates, with statistical significance determined using a two‐tailed Student's t‐test (*p < 0.05, **p < 0.01). [Color figure can be viewed at wileyonlinelibrary.com]

3.8. Exploration of Downstream Targets Regulated by TaERF1‐A

Four hundred candidate target genes of TaERF1‐A were identified using the wheat wGRN database (Table S4; Figure 8A), including TraesCS3A03G0864700, TraesCS3B03G0987700, and TraesCS3D03G0793300, which encode delta‐1‐pyrroline‐5‐carboxylate synthase 2 and are hereafter referred to as TaP5CS2 genes. TaP5CS2 catalysers the conversion of glutamate to glutamate‐5‐semialdehyde (GSA), which spontaneously cyclizes to Δ¹‐pyrroline‐5‐carboxylate (P5C), providing a direct precursor for proline production (Rai and Penna 2013). In addition, under normal watering conditions, proline content was significantly increased in TaERF1‐A OE wheat lines (Figure 3D), suggesting that TaP5CS2 is involved in TaERF1‐A‐mediated regulation of proline biosynthesis. TaP5CS2s were thus subjected to functional analysis in the context of the roles of TaERF1‐A in regulating proline accumulation. Sequence analysis revealed the presence of one putative GCC‐box motif (GCCGCC) within the promoters of the TaP5CS2 homologues (Figure S15). In the Y1H assay, the Y187 strain co‐transformed with pGADT7‐TaERF1‐A and pHIS2‐TaP5CS2‐A _pro grew on medium containing 100 mM 3‐AT. In contrast, the strain carrying pHIS2‐Empty did not grow under the same selection conditions (Figure S16). EMSA using biotin‐labelled probes encompassing the GCC‐box–containing promoter fragments and purified MBP–TaERF1‐A protein was conducted to determine whether TaERF1‐A associates with the promoters of TaP5CS2 homologues (Figure 8B). Notably, distinct mobility‐shifted bands were observed only when the fusion protein was incubated with the labeled probes, confirming the formation of specific protein–DNA complexes (Figure 8B). These results demonstrate that TaERF1‐A can directly and specifically bind to the promoters of TaP5CS2 genes. Dual‐LUC assays performed in a transient expression system in N. benthamiana revealed that co‐infiltration of the TaERF1‐A effector with reporter constructs driven by the TaP5CS2 promoters significantly increased the LUC/REN ratio compared to the empty‐vector controls (Figure 8C). These results indicate that TaERF1‐A can activate the transcription of TaP5CS2s in wheat.

Figure 8.

Identification and functional analysis of TaERF1‐A targets. (A) Putative downstream targets of TaERF1‐A, that is, TaP5CS2‐A/B/C, which encode delta‐1‐pyrroline‐5‐carboxylate synthase 2, were identified using the wGRN database. (B) EMSA assay results demonstrating that TaERF1‐A specifically recognises and binds to the ERF‐binding site present in the promoters of TaP5CS2. (C) Dual‐luciferase assay results demonstrating that TaERF1‐A transcriptionally activates the promoters of TaP5CS2s. (D) The expression profiles of TaP5CS2‐B in OE and WT wheat plants. (E) Drought tolerance of TaP5CS2‐B OE transgenic lines, photographed before and after drought treatment and after a 4‐day re‐watering period. The numbers represent the SRs of OE and WT plants. (F) Proline accumulation in the leaves of WT and transgenic lines under WW and WD treatments. Data are shown as mean ± SD values from three independent replicates, with statistical significance determined using a two‐tailed Student's t‐test (*p < 0.05, **p < 0.01). [Color figure can be viewed at wileyonlinelibrary.com]

TaP5CS2 showed a clear tissue‐specific expression pattern, with the highest transcript abundance detected in spikes, followed by stems, whereas relatively lower expression levels were observed in roots and leaves; the lowest expression was found in grains (Figure S17). The expression pattern of TaP5CS2 was analyzed in the leaves and roots of three wheat cultivars (JW1, XN979, and YM9) under PEG‐induced and DS conditions. Under PEG treatment, TaP5CS2 exhibited a time‐dependent induction pattern in JW1 and XN979, particularly in leaves, where expression increased markedly at successively later time points. In roots, transient induction followed by fluctuation in expression level was observed (Figure S18A). In contrast, YM9 showed a weaker and more variable response under PEG stress (Figure S18A). Under natural DS, TaP5CS2 was strongly induced in both leaves and roots of JW1 and XN979, peaking in expression at mid to late time points before slightly declining thereafter (Figure S18B). However, YM9 displayed a delayed and less stable induction pattern, with overall lower expression levels compared with both JW1 and XN979 (Figure S18B). TaP5CS2‐B OE wheat lines were generated, and transgene integration and expression were confirmed by q‐PCR in order to assess whether TaP5CS2 functions similarly in wheat proline biosynthesis and to further evaluate its contribution to DT (Figure 8D). Three OE plants with the highest TaP5CS2‐B transcript levels were selected for subsequent analyses. Notably, TaP5CS2‐B OE plants exhibited enhanced DT and significantly increased SRs compared to WT plants (Figure 8E). In addition, OE plants accumulated markedly higher proline levels than WT plants under WW and WD conditions (Figure 8F) and also maintained significantly higher chlorophyll content under DS (Figure S19). These findings collectively demonstrate that TaERF1‐A acts as a transcriptional activator of TaP5CS2 expression. They also confirm that TaP5CS2 enhances proline biosynthesis, thereby improving DS tolerance in wheat.

4. Discussion

4.1. Basic Characteristics of TaERF1‐A

To date, the functions of ERF TFs have been extensively characterised in diverse plant species, including Arabidopsis (Cheng et al. 2013), rice (Ambavaram et al. 2014), wheat (Du et al. 2023), cucumber (Hu and Liu 2011), maize (Hao et al. 2020), alfalfa (Jin et al. 2019), pineapple (Huang et al. 2020), grapes (Zhu et al. 2019), and tomatoes (Yang et al. 2021). ERF TFs play various roles in biotic and abiotic stress responses in a range of plant species. Herein, an ERF TF‐encoding gene, TaERF1‐A, was identified in wheat. Sequence analysis revealed a KRRK nuclear localisation signal sequence in the C‐terminal region of TaERF1‐A (Figure S1A), suggesting that TaERF1‐A may be localised to the nucleus. Notably, TaERF1‐A was restricted to the nucleus based on the TF role of ERF proteins (Figure 1C). Notably, TaERF1‐A showed distinctly different expression patterns between drought‐tolerant and drought‐sensitive wheat cultivars. Under DS, TaERF1‐A was more rapidly and strongly induced in the drought‐tolerant cultivar XN979 compared with the drought‐sensitive cultivar YM9, suggesting that its transcriptional activation is closely associated with enhanced DT. This differential expression pattern indicates that TaERF1‐A may contribute to genotypic variation in drought response by promoting the activation of downstream stress‐responsive regulatory networks in tolerant genotypes. However, yeast‐based assays revealed that TaERF1‐A showed no transcriptional activation (Figure S5A), which aligns with the findings of earlier studies on ERF transcription factors, such as TaERF4 in wheat (Dong et al. 2012), CnERF1 in chrysanthemum (Gao et al. 2015), and IbERF5 in sweet potatoes (Meng et al. 2020). In another previous study, LkERF6 lacked transcriptional activation activity in the yeast system, but its overexpression in tobacco enhanced salt tolerance and DT by modulating the expression of LkSOD, LkCCS, and LkCAT (Tian et al. 2024). PagERF113 lacks transcriptional activation activity in yeast and cannot directly activate transcription; however, it mediates the auxin signalling pathway by activating the expression of PagGH3.6 (Wang et al. 2025). In our present study, although the full‐length TaERF1‐A did not exhibit transcriptional activation activity in yeast, dual‐luciferase assays showed that TaERF1‐A significantly enhanced reporter gene expression in planta (Figure S5B), indicating that it functions as a transcriptional activator within plant cells. This discrepancy may be owing to the requirement for plant‐specific cofactors or post‐translational modifications that are absent in yeast.

4.2. TaERF1‐A Enhances Drought Resistance in Genetically Modified Arabidopsis and Wheat

Drought is a primary environmental factor that impacts agriculture by restricting crop growth and yield (Zhang et al. 2018). Notably, TaERF1 and TaERF3 are activated by DS, and they improve DT in transgenic Arabidopsis and wheat (Xu et al. 2007; Rong et al. ²⁰¹⁴). It is hypothesised that the function of wheat ERF TFs in the regulation of resistance is conserved in Arabidopsis and wheat. In the present study, wheat TaERF1‐A was markedly upregulated under PEG‐induced dehydration stress. Moreover, the root length and SR of Arabidopsis TaERF1‐A OE plants were markedly higher than those of the WT plants (Figure 2E,F), indicating that TaERF1‐A OE improved DT in transgenic Arabidopsis.

The applicability of the PureWheat transformation system remains highly dependent on genotype despite the system significantly improving the efficiency of Agrobacterium‐mediated transformation in wheat (Ishida et al. 2015). The optimised media and operational procedures are most effective in varieties with strong regenerative capacity, such as Fielder. However, many modern commercial wheat varieties, particularly the elite winter wheat varieties widely grown in China, cannot be efficiently transformed using the standard PureWheat protocol. For instance, Xinong 979 (XN979) is a variety largely grown in the Yellow and Huai River Valley regions and is recalcitrant to transformation. Such features highlight the persistent technical bottleneck in both functional genomic studies and trait improvement in these genotypes. Herein, a dual‐strategy approach was employed to overcome these constraints. Transient silencing of TaERF1‐A was performed in XN979 in order to circumvent its transformation recalcitrance. In contrast, stable transgenic overexpression and RI lines were generated in Fielder, which is compatible with PureWheat‐based transformation. This complementary strategy made it possible to overcome genotype‐dependent limitations and enabled functional validation of TaERF1‐A across distinct genetic backgrounds. BSMV _TaERF1‐A ‐silenced XN979 wheat plants exhibited a drought‐sensitive phenotype, with higher MDA contents and lower proline, SOD, POD, and CAT contents (Figure S9). Notably, BSMV _TaERF1‐A ‐silenced wheat plants exhibited weaker antioxidant capacity than BSMV₀ plants. Overexpression of TaERF1‐A in the wheat variety Fielder consistently enhanced DT, as reflected by its improved survival rates, reduced membrane damage, and elevated antioxidant enzyme activities under WD conditions. OE plants also accumulated higher levels of proline (Figure 3D), suggesting that TaERF1‐A activates osmoprotective pathways in addition to strengthening antioxidative defence. In contrast, TaERF1‐A RI lines exhibited increased sensitivity to DS, exhibiting higher MDA content, impaired antioxidant capacity, and reduced proline accumulation (Figure 3D–H). These reciprocal phenotypes between OE and RI lines validate the functional importance of TaERF1‐A and underscore its importance as a crucial upstream regulator that integrates multiple stress‐responsive pathways. The consistent phenotypic and molecular responses across the BSMV transient silencing and PureWheat stable transformation systems underscore the significant role of TaERF1‐A in stress adaptation. They also highlight the utility of integrating complementary transformation platforms to investigate gene functions in wheat with differing transformation capacities.

4.3. Potential Mechanism of TaERF1‐A in Regulating Drought Resistance in Wheat

TFs potentially control the expression of specific genes by favouring or discouraging the recruitment of RNA polymerases. TaWRKY33 activates stress‐related genes, including AtRD29A, DREB2B, ABA1/2, and ABI1/5, promotes root growth, and increases germination rates in Arabidopsis under drought conditions (He et al. 2016). Herein, TaWRKY33, which is a WRKY TF that has the potential to enhance the expression of TaERF1‐A by interacting with W‐box elements located in its promoter, was studied. Y1H, EMSA, and dual‐luciferase assays mediated by Agrobacterium were performed (Figures 5B, 5C, 5D). TaWRKY33 directly activated the expression of TaERF1‐A by specifically binding to the W‐box in response to drought stress. ERF TFs, which are DREB2‐like genes, participate in regulatory pathways involving WRKY proteins. For instance, this phenomenon is seen in the overexpression lines of wheat genes TaWRKY2 and TaWRKY19 in Arabidopsis, which trigger the activation of AtDREB2A, thereby enhancing tolerance to drought and salinity (Niu et al. 2012). In cotton, GhWRKY59 directly activates GhDREB2 to enhance DT (Li et al. 2017). In lily (Lilium longiflorum), the WRKY TF LlWRKY22 promotes thermotolerance by activating LlDREB2B. These reports suggest that WRKY TFs regulate ERF and participate in various stress responses (Wu et al. 2022).

Herein, the interaction between TaERF1‐A and five candidate proteins (TaG6P1E, TaUn, TaFBA, TaCSP41b, and TaNF‐YC6) was confirmed in yeast (Figure S11). G6PE is a glucose‐6‐phospho1‐differential isomerase (G6P1E) involved in the glucose metabolic pathway and serves a role in maintaining normal cell metabolism (Wurster and Hess 1972). Fructose diphosphate aldolase (FBA) is a crucial metabolic enzyme involved in glycolysis, gluconogenesis, and the Calvin cycle, and plays a role in carbohydrate metabolism and signal transduction (Lv et al. 2017). In Arabidopsis, CSP41b is essential in maintaining normal chloroplast function (Ariga et al. 2015). The atcsp41b mutant causes dwarfness, leaf paleness, and alters chloroplast morphology and photosynthetic performance (Ariga et al. 2015). Notably, OE of AtCSP41b and EsCSP41b increases heat and salt tolerance in transgenic Arabidopsis (Ariga et al. 2015). Photoinhibition and impaired chloroplast function can occur under salt and heat stress. CSP41b alleviates photoinhibition and restores chloroplast function in response to abiotic stress (Ariga et al. 2015). The nuclear factor Y (NF‐Y) comprises a heterotrimeric assembly consisting of its functional subunits: NF‐YA/B/C (Li et al. 2008; Panzade et al. ²⁰²²). Dimers consisting of NF‐YB/C subunits are initially formed within the cytoplasm, after which they migrate to the nucleus, where they engage with NF‐YA to create a mature heterotrimer complex (Panzade et al. 2022). Mature heterotrimer NF‐Y complexes then bind to promoter CCAAT sequences, causing transcriptional regulation (Panzade et al. 2022). DS and ABA treatment strongly induce the expression of NFYA5, an Arabidopsis NF‐YA family member. NF‐YA5 endows Arabidopsis with drought resistance (Ariga et al. 2015). NFYB1 substantially enhances DR and yield in Arabidopsis and maize under DS (Nelson et al. 2007). The Cdt‐NF‐YC1 protein boosts tolerance to drought and salt in transgenic rice by modulating stress‐response members via ABA‐dependent and independent pathways (Chen et al. 2015). However, these interactions are currently supported only by Y2H or BiFC assays and lack further in vivo validation. Therefore, their physiological relevance under drought conditions remains unclear, and these results should therefore be considered preliminary. It was thus hypothesised that the TaERF1‐A protein interacts with five proteins involved in glucose metabolism, signal transduction, and phytohormone signalling, thereby regulating downstream gene expression to influence DT in wheat.

Proline is a suitable osmolyte that helps plants develop resilience to biotic and abiotic stresses (Amini et al. 2015). The P5CS gene significantly contributes to proline accumulation in plants under osmotic stress. Herein, EMSA and dual‐luciferase assays confirmed that TaERF1‐A activates TaP5CS2s by directly binding to GCC‐box motifs in their promoter regions (Figure 8). Correspondingly, TaERF1‐A overexpression significantly elevated TaP5CS2 transcript levels, whereas TaERF1‐A knockdown in RI plants substantially decreased their mRNA abundance (Figure S20). The pronounced shifts in TaP5CS2 expression potentially contributed to the contrasting drought tolerance phenotypes observed between the TaERF1‐A OE and RI lines (Figure 3B). We thus propose a functional model for the TaWRKY33–TaERF1‐A‐TaP5CS2 regulatory cascade that enhances drought tolerance in wheat. In this model, TaWRKY33 activates the transcription of TaERF1‐A, which subsequently binds to the GCC‐box elements in the TaP5CS2 promoters to elevate their expression during drought stress (Figure 9). Glutamate (Glu) is converted to glutamate‐5‐semialdehyde (GSA) and is subsequently cyclized to Δ¹‐pyrroline‐5‐carboxylate (P5C) under the catalysis of the TaP5CS gene–encoded enzyme. P5C is then reduced to proline by P5C reductase (P5CR) (Rai and Penna 2013). This transcriptional module ultimately boosts proline biosynthesis, thereby promoting osmotic homoeostasis and mitigating drought‐induced damage.

Figure 9.

Schematic model showing TaWRKY33‐TaERF1‐A‐TaP5CS2 modulation of drought stress homoeostasis in wheat. [Color figure can be viewed at wileyonlinelibrary.com]

Conflicts of Interest

The authors declare no conflicts of interest.

Supporting information

Supporting File 1

Supporting File 2

Data Availability Statement

The data that support the findings of this study are available from the corresponding author upon reasonable request.