TaGW2-TaVOZ1 module regulates wheat salt tolerance via both E3 ligase–dependent and –independent pathways

Abstract

Wheat production is limited by the rapid expansion of salinized arable land worldwide. Identification of the molecular mechanisms that underlie the salt stress response is of great importance. Here, we uncovered the NAC-type transcription factor, TaVOZ1, as a positive regulator of wheat salt tolerance. Its overexpression could enhance yield and biomass production under salt stress, while TaVOZ1 knockdown attenuates salt tolerance. TaVOZ1 transcriptionally activates stress-responsive genes, especially HKT1-family transporters, decreasing shoot Na⁺ accumulation. However, the RING-type E3 ligase, TaGW2, directly interacts with and ubiquitinates TaVOZ1, promoting its ubiquitin/26S proteasomal degradation. TaGW2 overexpression reduces salt tolerance, while its knockdown or knockout enhances wheat response to salt stress. Moreover, we found a moonlight function of TaGW2 wherein it binds the same DNA motifs as TaVOZ1 to block its up-regulation of HKT1-family genes while coordinately governing both the salt tolerance and grain yield. This study highlights the functional versatility of TaGW2 and defines an antagonistic TaGW2-TaVOZ1 regulatory module in wheat salt tolerance.

TaGW2 mediates TaVOZ1 clearance and blocks transcription of HKT1-family transporter genes, attenuating salt tolerance in wheat.

INTRODUCTION

Saline soil conditions affect roughly 20 million hectares of cultivated land globally, and this proportion expands yearly due to the effects of climate change on water availability and extensive irrigation practices. As a result, salt stress is among the most prevalent abiotic stresses, presenting a major challenge for maintaining agricultural production, especially in staple crops grown in arid or semiarid climates, such as wheat (1–3). Ion toxicity due to excessive cation (e.g., Na⁺) uptake from saline soils can increase the osmotic potential of the tissue to facilitate water uptake, leading to downstream consequences, including oxidative stress, and ultimately reduced grain quality and yields (4–6). Although wheat is cultivated across more regions worldwide than any other crop (3, 7), it shows relatively high sensitivity to salt conditions (1, 8), posing a challenge for food security and presenting a need for wheat lines with improved salt tolerance.

Salt tolerance in plants typically depends on mechanisms of Na⁺ exclusion from aerial tissues and/or subcellular compartmentalization mediated by Na⁺-preferential transporters (4, 5). For example, the Na⁺/H⁺ exchangers (NHX) family sodium/proton antiporter, SALT OVERLY SENSITIVE 1 (NHX7), was recently shown to localize to vacuoles and mediate vacuolar sequestration of Na⁺ in root meristems (4, 9), while the high-affinity K⁺ (HAK)–family transporter, ZmHAK4, removes Na⁺ from xylem sap (10). However, the HAK transporter (HKT) family, comprising class 1 (HKT1) and class 2 (HKT2) proteins, may be the most well studied of these salt transporters. Among them, HKT1 proteins preferentially remove Na⁺ from xylem sap to prevent toxicity arising from source-to-sink accumulation in shoots and leaves, while HKT2 proteins preferentially remove K⁺ from xylem sap (8, 11, 12). HKT1-family genes have been shown to positively regulate salt tolerance in numerous crops by removing Na⁺ from xylem sap for sequestration in nonphotosynthetic tissues. Some previously reported HKT1 Na⁺ transporters associated with salt tolerance include Arabidopsis HKT1;1 (13, 14), SKC1/OsHKT1;5 in rice (15), Nax1/HKT1;4 and Nax2/HKT1;5 in durum wheat (16, 17), ZmNC1/ZmHKT1 in maize (18), and Kna1/TaHKT1;5-D in bread wheat (8).

Despite this mechanistic understanding, relatively few proteins have been identified as upstream regulators of HKT1 genes, many of which negatively affect salt tolerance. For instance, AtABI4 reportedly binds the promoter of AtHKT1;1, the only HKT1 gene in Arabidopsis, to suppress its transcription (13, 19); OsWRKY53 functions as a transregulator of OsHKT1;5 (20); together with OsSDG721, the OsSUVH7-OsBAG4-OsMYB106 complex epigenetically reprograms the OsHKT1;5 promoter region to modulate its expression (21), and the TaSPL6-D^Del variant binds the TaHKT1;5-D promoter to inhibit its transcription in wheat (8). Alternatively, the OsMYBc transcription factor up-regulates OsHKT1;1 to enhance salt tolerance in rice (22). These studies present substantial and highly valuable insights into the transcriptional regulation of HKT1 genes and the corresponding salt response in plants. In addition, VASCULAR PLANT ONE-ZINC FINGER (VOZ) genes, which encode NAC (NAM, ATAF, and CUC)–type transcription factors, have also been shown to regulate drought and salt stress responses (23–26). For example, VOZ-family transcription factors positively regulate salt tolerance but negatively regulate drought resistance in Arabidopsis (23, 24). In soybean, GmVOZ1G positively regulates both drought and salt tolerance (25). However, the detailed molecular mechanisms by which VOZs regulate the salt stress response, especially the transcriptional regulation of HKT genes, are still poorly understood.

In addition to transcriptional regulation, the salt stress response in plants also relies on precise control of relevant protein levels to maintain stress-responsive functions, particularly in the relative activation of signaling pathways. Among such posttranscriptional or posttranslational regulatory mechanisms, protein degradation by the ubiquitin/26S proteasome (UPS) pathway, which requires sequential activity of ubiquitin-activating enzyme E1, ubiquitin-conjugating enzyme E2, and E3 ubiquitin ligase, is well known to play essential roles in plant stress responses, among numerous other cellular processes (27–29). In particular, E3 ligases such as Really Interesting New Gene (RING) finger proteins mediate key functions in plant hormone signaling, stress response, and developmental pathways (30–34). Among them, the C5HC2-type E3 ubiquitin ligase, TaGW2, has been studied extensively due to its regulatory contributions to kernel size and weight in wheat (35, 36). A recent study revealed that TaGW2 positively regulates drought resistance by promoting the degradation of the type-B ARR transcription factor TaARR12 (37). Other studies also reported that TaGW2 attenuates resistance to wheat leaf rust and stripe rust through degradation of TaSGT1, TaSnRK1γ, and TaVPS24 (38, 39). However, the mechanism through which TaGW2 coordinates growth with stress response pathways remains uncertain, and the full scope of its targets has not been defined.

In the current study, we observed that transgenic wheat cv. Fielder plants overexpressing TaVOZ1 exhibited stronger growth and higher yield than the wild type following salt stress treatment. Subsequent DNA affinity purification sequencing (DAP-seq) and RNA sequencing (RNA-seq) assays identified a suite of salt-responsive genes that are up-regulated by TaVOZ1. Among them, we found that HKT1-family transporters were essential for a salt-tolerant phenotype and prevented Na⁺ accumulation in shoots. In addition, we found that the E3 ubiquitin ligase, TaGW2, could interact with TaVOZ1 and reduce its stability through UPS-mediated degradation, while functional assays in TaGW2 overexpression (OE), knockdown, or knockout wheat lines confirmed that it acts as a negative regulator of salt tolerance. Unexpectedly, our experiments uncovered a moonlight function of TaGW2 in which it directly represses transcription of diverse genes and binds the same promoter motif as TaVOZ1, resulting in antagonistic regulatory effects in wheat response to salt stress. This study thus demonstrates that TaGW2 and its ubiquitination substrate, TaVOZ1, antagonistically modulate wheat salt tolerance, suggesting that these proteins, especially TaGW2, could serve as versatile and potent molecular breeding targets for improving wheat yields in high-saline growing regions.

RESULTS

TaVOZ1 positively regulates salt tolerance in wheat

Previous studies have revealed that VOZ transcription factors are positive regulators of the salt stress response in plants (23–25). To understand the role of TaVOZ1 in wheat salt tolerance, we first examined the expression pattern of TaVOZ1 following salt stress (100 mM NaCl) treatment by reverse transcription quantitative polymerase chain reaction (RT-qPCR). The results showed that TaVOZ1 expression in leaves of 20-day-old wheat seedlings significantly increased within 3 hours following salt stress treatment, peaking by 12 hours and remaining elevated for 72 hours (Fig. 1A). In contrast, TaVOZ1 reached its highest expression levels in root tissue at 6 hours post–NaCl treatment (Fig. 1B). To identify other genes that might be conjointly regulated by salt exposure along with TaVOZ1, we conducted bulk RNA-seq analysis of 20-day-old seedlings from wheat cv. Chinese Spring collected at 0.5 through 72 hours of 100 mM NaCl treatment (Fig. 1C and fig. S1A). This analysis indicated that all subgenome homoeologs of TaVOZ1 exhibited a similar, significant increase in transcription at 3 hours posttreatment, which peaked at 6 hours posttreatment (fig. S1B). To further focus on TaVOZ1, we constructed a gene regulatory network (GRN) for salt stress response based on the above RNA-seq data (Fig. 1C), which identified 552 differentially expressed genes (DEGs) that were significantly correlated with TaVOZ1 expression (Fig. 1D and table S1). These coexpressed genes included ion transporters such as TaHKT1, TaHKT1;1, TaHKT1;5, TaHKT1;3, TaHKT2;3, and TaNHK7 (Fig. 1D). Gene Ontology (GO) analysis suggested that these 552 genes were enriched in functional terms related to “transcriptional regulation,” “response to salt stress,” “response to osmotic stress,” “ion transport,” and “ion homeostasis” (Fig. 1E). Together, these results confirmed TaVOZ1 as a salt stress–responsive gene in wheat.

Fig. 1. A TaVOZ1-associated coregulation network in wheat response to salt stress.

(A and B) RT-qPCR analysis of TaVOZ1 expression in wheat leaves (A) and roots (B) following exposure to 100 mM NaCl. Data show means ± SDs from three independent experiments. h, hours. (C) Heatmap of the transcriptomic changes in 20-day-old wheat cv. Chinese Spring seedlings collected at 0.5 through 72 hours of exposure to 100 mM NaCl under hydroponic conditions. Color gradient represents z-standardized transcripts per million values for all expressed genes. (D) A TaVOZ1-based gene coregulation network in wheat. The network was constructed by determining pairwise Pearson correlation coefficients between transcript levels of all coexpressed genes in TaVOZ1-containing gene modules (identified by WGCNA) and each TaVOZ1 homoeolog across all samples. Circles represent genes, pink lines show positive gene-gene associations, and green lines are negative correlations among genes (false discovery rate < 0.01). (E) GO term analysis of genes coexpressed with TaVOZ1.

To further investigate its possible functions in wheat salt tolerance, we generated transgenic lines in the wheat cv. Fielder background for TaVOZ1 OE or knockdown by RNA interference (RNAi). We then selected three independent lines for each construct that respectively displayed the greatest increase or decrease in TaVOZ1 expression at the mRNA and protein levels (Fig. 2, A and B). Gross phenotypic analysis of ~30-day-old TaVOZ1 OE, RNAi, and wild-type (WT) seedlings grown under nonsaline soil conditions showed no obvious alterations in morphology or plant architecture associated with increased or decreased TaVOZ1 expression (Fig. 2C, top), whereas exposure to high-salt soil conditions (150 mM NaCl treatment) resulted in an obviously shorter stature of TaVOZ1 RNAi lines compared to WT Fielder, while OE lines had the tallest stature (Fig. 2C, bottom). Shoot biomass measurements indicated that TaVOZ1 OE plants had significantly greater biomass than WT Fielder, while RNAi lines had significantly lower biomass than WT under high-salt soil conditions, whereas the biomass of transgenic lines did not differ from WT under normal growth conditions (Fig. 2, D and E). Similarly, measurements of shoot Na⁺ contents indicated that TaVOZ1 RNAi plants accumulated higher Na⁺ concentrations, while OE plants contained lower Na⁺ contents than WT under high-salt soil conditions but showed no such difference in shoot Na⁺ levels under normal growth conditions (Fig. 2, F and G). In addition, TaVOZ1 RNAi seedlings had significantly shorter root lengths than WT Fielder after 20 days of growth in 150 mM NaCl solution, while TaVOZ1 OE seedlings had significantly longer roots than WT following this treatment; no difference in root length was observed between WT and transgenic nonsaline control groups (Fig. 2, H to K).

Fig. 2. Positive regulation of salt tolerance by TaVOZ1 in wheat.

(A and B) RT-qPCR analysis of TaVOZ1 expression in TaVOZ1 OE (A) and RNAi lines (B). Protein levels are shown at the bottom. Values are means ± SDs from four independent experiments. RI, RNAi. (C) Assessment of salt tolerance in OE and RNAi lines. Photographs were taken following 30 days of normal growth (top) or 150 mM NaCl treatment under saline soil conditions (bottom). (D and E) Statistical analysis of shoot biomass in OE and RNAi lines under normal growth (D) or 150 mM NaCl treatment under saline soil conditions (E). Values are means ± SDs (n = 20). (F and G) Na⁺ contents in shoots of WT Fielder, OE, and RNAi lines under normal growth (F) or 150 mM NaCl treatment under saline soil conditions (G). Values are means ± SDs (n = 4). DM, dry matter. (H and I) Representative images of root development in TaVOZ1 OE and RNAi lines grown for 20 days under 150 mM NaCl or deionized water hydroponic conditions. (J and K) Statistical comparison of root lengths between OE and RNAi lines grown in deionized water (J) or 150 mM NaCl (K) hydroponic conditions. Values are means ± SDs (n = 20). (L) Grain morphology of WT and TaVOZ1 transgenic lines under normal growth conditions. Scale bars, 5 mm. (M) Grain morphology of WT and TaVOZ1 transgenic lines under 150 mM NaCl treatment under saline soil conditions. Scale bars, 5 mm. (N and O) Statistical analysis of plant height, kernel length, kernel width, and thousand kernel weight (TKW) for wheat plants grown under nonsaline soil (N) or saline soil (O) conditions. Values are means ± SDs; statistical significance was determined by a two-sided t test (**P < 0.01).

To further characterize the effects of altered TaVOZ1 expression on yield, we compared grain production between lines grown in saline (150 mM NaCl treatment) and nonsaline soils under greenhouse conditions. Measurements of plant height, kernel length, kernel width, and thousand kernel weight (TKW) indicated that there were no significant differences between transgenic and WT Fielder plants under nonsaline conditions (Fig. 2, L and N). However, when grown under saline soil conditions, TaVOZ1 OE plants had significantly greater height, kernel size, and grain weight compared to WT, while RNAi lines had reduced plant height, smaller kernels, and lower grain weight (Fig. 2, M and O). These results collectively suggested that TaVOZ1 functions as a positive regulator of wheat plant growth and grain production under high-salt conditions, indicating a salt-tolerant phenotype.

TaVOZ1 is a key regulator governing the stress-responsive transcriptomic changes

As TaVOZ1 is annotated as a NAC transcription factor with three subgenome homoeologs (i.e., A, B, and D) in wheat that share 90 to 94% similarity in full-length protein sequence (fig. S2A), we performed a phylogenetic analysis to assess its conservation with its homologs in other species, including OsVOZ1 in rice (40), SlVOZ1 in tomato (26), and AtVOZ1 in Arabidopsis (23) (Fig. 3A). Subcellular localization assays using fluorescence microscopy of a TaVOZ1–green fluorescent protein (GFP) reporter in wheat protoplasts showed that it mainly localized in the nucleus, with a weak fluorescent signal in the cytoplasm (Fig. 3B). Transactivation assays in Saccharomyces cerevisiae cells grown on synthetic dropout medium indicated that TaVOZ1 function as a transcription factor (fig. S2B). Furthermore, similar to BD-VP16–positive control, TaVOZ1 fused to a GAL4 binding domain (BD) could induce significantly enhanced luciferase (LUC) expression in Nicotiana benthamiana leaves compared to the BD-TaActin1–negative control (Fig. 3C), further supporting its likely activity as a transcription activator.

Fig. 3. Genome-wide overview of the regulatory network downstream of TaVOZ1.

(A) Phylogenetic analysis of TaVOZ1 and its homologs from Arabidopsis, rice, Brachypodium distachyon, maize, and wheat. Bootstrap values (from 1000 replicates) are indicated at each node. Branch lengths represent scale. (B) TaVOZ1 subcellular localization assays in wheat protoplasts expressing 35S:TaVOZ1-GFP or 35S:GFP. Scale bars, 15 μm. (C) Dual-LUC assays of TaVOZ1 transcriptional activity in which TaVOZ1 was fused to the GAL4 BD or a BD-VP16 fragment, with BD alone, BD-VP16, and BD-VP16–actin serving as controls. Transcriptional activity of each construct was normalized to that of LUC driven by BD. (D and E) DAP-seq identification of high-confidence TaVOZ1 binding peaks (D) and their distribution in different gene regions (E) across the wheat genome. (F) Sequence logos for enriched motifs within the TaVOZ1 binding peaks. The CTTCTT and AAGAAG motifs showed the greatest enrichment among core sequences. E values were calculated by MEME–chromatin immunoprecipitation (ChIP). (G) Venn diagram of overlapping genes between RNA-seq and DAP-seq results. Up- and down-regulated candidate target genes (CTGs) are also shown. (H and I) GO analysis of differentially up-regulated (H) and down-regulated (I) CTGs. P values were adjusted by Benjamini-Hochberg correction; only significant categories (P < 0.01) are displayed.

To define the regulatory network required for TaVOZ1 function in wheat response to salt stress, we first compared the transcriptomes of 20-day-old TaVOZ1 OE and WT Fielder seedlings following 12 hours of exposure to 150 mM NaCl solution or deionized water. This analysis identified 6485 total DEGs [fold change (FC) > 2 or FC < 0.5 and P < 0.01] in the OE lines relative to their expression in WT under nonsaline conditions, including 4166 up- and 2319 down-regulated DEGs (fig. S3, A to D, and table S2). Alternatively, we found 4461 total DEGs in comparisons of OE versus WT under salt stress conditions, encompassing 1939 up-regulated and 2522 down-regulated genes (fig. S4, A to D, and table S3). GO analysis showed significant enrichment in terms associated with stress responses, such as transcriptional regulation, “response to stimulus,” “response to water deprivation,” “response to abscisic acid,” response to osmotic stress, and response to salt stress among the up-regulated DEGs (figs. S3E and S4E). In contrast, down-regulated DEGs showed enrichment in functional terms such as “phosphorylation,” “defense response,” “response to salicylic acid,” “response to hypoxia,” “response to auxin,” and “lipid metabolism” (figs. S3F and S4F). These results implied that TaVOZ1 may contribute to multiple, potentially overlapping GRNs that coordinately regulate a salt-tolerant phenotype in wheat.

To further investigate the mechanism(s) of TaVOZ1 in alleviating salt stress, we sought to identify potential TaVOZ1 binding sites and target genes in the wheat genome by DAP-seq (41, 42). This analysis uncovered 6574 candidate binding sites across 4436 genes, located within both gene bodies and the 2-kb upstream and 2-kb downstream flanking sequences, presumably harboring the promoter and transcriptional terminator, respectively (Fig. 3D and table S4). Further detailed examination indicated that 33.6, 29.1, 16.3, 8.9, 2.7, 3.4, and 5.9% of the predicted TaVOZ1 binding sites were located in distal intergenic regions, promoter regions, introns, exons, 5′ untranslated region (5′-UTR), 3′-UTR, and terminator regions, respectively (Fig. 3E). Subsequent predictions of potential TaVOZ1 binding motifs, based on consensus sequence of binding sites obtained by MEME–chromatin immunoprecipitation (ChIP), identified 5′-CTTCTT-3′, or its reverse complement, 5′-AAGAAG-3′, as the most significantly enriched core recognition motifs (Fig. 3F). Through combining our DAP-seq and RNA-seq results, we identified 2347 candidate target genes that TaVOZ1 could potentially bind to and transcriptionally regulate (Fig. 3G and table S5), including 1402 up-regulated and 945 down-regulated genes in TaVOZ1 OE plants (Fig. 3G). GO analysis showed that the up-regulated predicted target genes were primarily enriched in terms associated with stress response, such as response to salt stress, response to water deprivation, ion homeostasis, transcriptional regulation, and response to abscisic acid (Fig. 3H), whereas the down-regulated predicted targets of TaVOZ1 showed enrichment in “protein autophosphorylation,” response to salicylic acid, defense response, “salicylic acid catabolicsm,” and response to auxin (Fig. 3I). These cumulative results suggested that TaVOZ1 contributes to salt tolerance by activating the expression of stress-responsive genes.

HKT1-family genes are direct targets of TaVOZ1

As TaVOZ1 RNAi plants exhibited higher Na⁺ accumulation and OE plants had lower Na⁺ in shoots compared to WT Fielder, we next focused on predicted gene targets functionally annotated to participate in ion transport or homeostasis. This analysis uncovered several up-regulated HKT1-type transporters, such as TaHKT1;1-B, TaHKT1;3-B, and TaHKT1;5-D, under TaVOZ1 OE. These predicted target genes also exhibited relatively higher expression in wheat, especially during salt stress, compared to other HKT1 family members (fig. S5). After confirming by RT-qPCR that TaHKT1;1-B, TaHKT1;3-B, and TaHKT1;5-D were transcriptionally activated in TaVOZ1 OE plants and suppressed in TaVOZ1 RNAi lines compared to that in WT (Fig. 4, A to C), we interrogated the promoter sequences of 14 HKT1-family genes (fig. S6). This analysis revealed that the promoter regions of all 14 genes harbored the 5′-CTTCTT-3′ or 5′-AAGAAG-3′ core motif (Fig. 4, D to F, and table S6), supporting the role of HKT1-family genes as regulatory targets of TaVOZ1. DAP-qPCR assays of representative HKT1 genes (TaHKT1;1-B, TaHKT1;3-B, and TaHKT1;5-D) in 20-day-old TaVOZ1 OE seedlings to verify whether TaVOZ1 could directly bind these promoter motifs revealed that P1, P2, and P3 sites (Fig. 4, D to F) were significantly enriched in DAP samples compared to input (Fig. 4G). Electrophoretic mobility shift assays (EMSA) further demonstrated that TaVOZ1 could directly bind to P1, P2, and P3 sites in the TaHKT1;1-B, TaHKT1;3-B, and TaHKT1;5-D promoters, and this binding could be blocked by mutating the core motifs (Fig. 4H). These results demonstrated that TaVOZ1 could indeed directly bind 5′-CTTCTT-3′/5′-AAGAAG-3′ core promoter motifs of HKT1-family genes.

Fig. 4. TaVOZ1-mediated transcriptional activation of HKT1 genes in wheat.

(A to C) RT-qPCR analysis of TaHKT1;1-B (A), TaHKT1;5-D (B), and TaHKT1;3-B (C) expression in the WT Fielder, TaVOZ1 OE, and RNAi lines. Values are means ± SDs (n = 3). (D to F) Enriched TaVOZ1 binding peaks in the TaHKT1;1-B (D), TaHKT1;5-D (E), and TaHKT1;3-B (F) promoters in DAP-seq data. The probe sequences used for EMSA are shown below with the core motif and the corresponding mutated motif indicated in red. (G) DAP-qPCR validation of the TaVOZ1 binding sites in HKT1 gene promoters from (D to F). Values are means ± SDs (n = 4). (H) EMSA of TaVOZ1–hemagglutinin (HA) or HA-only binding to biotin-labeled probes containing the HKT1 promoter core motifs. (I to K) Transactivation assays in N. benthamiana leaves measuring TaVOZ1-dependent activation of HKT1 gene expression. Values are means ± SDs (n = 4). (L) Assessment of salt tolerance in WT Fielder and TaHKT1;3-B OE lines. Photographs were taken following 30 days of normal growth or 150 mM NaCl treatment under saline soil conditions. EV, empty vector. (M and N) Statistical analysis of shoot biomass in WT and OE lines under normal growth (M) or 150 mM NaCl treatment under saline soil conditions (N). Data show means ± SDs (n = 20). (O and P) Na⁺ contents in WT and OE lines under normal growth (O) or 150 mM NaCl treatment under saline soil conditions (P). Values are means ± SDs (n = 4). (Q) Representative images of root development in WT and OE lines grown for 20 days under 150 mM NaCl or deionized water hydroponic conditions. (R and S) Statistical comparison of root lengths between WT and OE lines grown under deionized water (R) or 150 mM NaCl conditions (S). Data show means ± SDs (n = 20); statistical significance was determined by a two-sided t test (**P < 0.01).

To further verify TaVOZ1 regulation of these candidate targets, we conducted transactivation assays in a dual-LUC reporter system using 1.5-kb promoter fragments of TaHKT1;1-B, TaHKT1;3-B, or TaHKT1;5-D, harboring the original 5′-AAGAAG-3′ core motifs or mutated 5′-CCCCCC-3′ motifs. For these experiments, we transformed the pGreen II-0800 vector carrying a LUC reporter driven by the normal or mutant promoter into N. benthamiana leaves expressing either the pGreenII 62-SK-TaVOZ1 vector construct or an empty pGreenII 62-SK vector (Fig. 4, I to K). We found that the original, but not mutated, promoter could significantly up-regulate LUC expression in the presence of TaVOZ1 compared to the vector controls (Fig. 4, I to K), indicating that TaVOZ1 could activate TaHKT1;1-B, TaHKT1;3-B, or TaHKT1;5-D transcription.

To further verify the role of HKT1-family transporters in TaVOZ1-dependent salt stress response, we next characterized HKT1-family transporter function during salt stress. As TaHKT1;5-D has been shown to primarily govern Na⁺ exclusion from above-ground tissues in wheat by mediating its removal from xylem sap, consequently avoiding sodium toxicity in leaves (8, 43), we then generated transgenic wheat lines in the cv. Fielder background that overexpressed TaHKT1;3-B under control of the ubiquitin promoter. After confirming by RT-qPCR and Western blotting that TaHKT1;3-B transcription and protein level were significantly increased relative to that in WT Fielder plants (fig. S7), we compared the gross morphology of ~30-day-old TaHKT1;3-B OE plants with WT seedlings grown in high-salt (150 mM NaCl treatment) or nonsaline control soil. Under high-salt conditions, the TaHKT1;3-B transgenic lines showed obviously stronger salt tolerance compared to WT Fielder plants, which displayed clear growth retardation, whereas no such visible differences were observed between groups under nonsaline conditions (Fig. 4L). Subsequent shoot biomass measurements showed that TaHKT1;3-B OE lines indeed had higher biomass than WT under salt stress but did not significantly differ under normal growth conditions (Fig. 4, M and N). Similarly, TaHKT1;3-B transgenic lines had significantly lower shoot Na⁺ contents than WT following growth under saline soil but did not differ under nonsaline normal conditions (Fig. 4, O and P). In addition, TaHKT1;3-B OE plants grown in 150 mM NaCl solution had significantly longer roots than the corresponding WT treatment group but had similar root lengths in deionized water (Fig. 4, Q to S). These results collectively indicated that TaVOZ1 could directly bind HKT1-family gene promoters to enhance their expression, reducing Na⁺ accumulation in shoots, enhancing the salt tolerance of wheat.

TaGW2 mediates TaVOZ1 ubiquitination and proteasomal degradation

To identify potential TaVOZ1 interaction partners, we used yeast two-hybrid (Y2H) assays to screen a cDNA library obtained from wheat cv. Chinese Spring seedlings. These Y2H assays were conducted in triplicate with TaVOZ1 serving as bait on SD/-Trp-Leu-His-Ade medium supplemented with 25 mM 3-AT to account for possible autoactivation by TaVOZ1 (fig. S2B). This screen confirmed that TaGW2 was a likely TaVOZ1 interaction partner (Fig. 5A) among 186 total candidate proteins (table S7). Subsequent LUC complementation imaging (LCI) assays in N. benthamiana leaves showed that coexpression of TaGW2-nLUC and cLUC-TaVOZ1 reporters resulted in strong LUC signal, while no LUC signal was generated by vector controls (Fig. 5B). Anti–hemagglutinin (HA) pull-down assays using recombinant maltose-binding protein (MBP)–TaGW2 and HA-HIS-TaVOZ1 proteins purified from Escherichia coli verified their interaction, while immunoblot detection of MBP further showed the presence of a band corresponding to the predicted size of MBP-TaGW2 (Fig. 5C). Moreover, coimmunoprecipitation (Co-IP) assays in N. benthamiana leaves coexpressing 35S:mCherry-TaVOZ1 and 35S:GFP-TaGW2 confirmed that mCherry-TaVOZ1 could be immunoprecipitated by GFP-TaGW2 in planta (Fig. 5D). These results collectively demonstrated that TaGW2 could directly interact with TaVOZ1.

Fig. 5. TaGW2 mediates the ubiquitination and degradation of TaVOZ1.

(A) Y2H assays of potential TaGW2 interactions with TaVOZ1. (B) LCI assays of potential TaGW2-TaVOZ1 interactions in N. benthamiana leaves coexpressing TaGW2-nLUC and cLUC-TaVOZ1 fusion constructs, or their corresponding LUC-only vector controls. (C) Pull-down assays of HA-HIS-TaVOZ1 interaction with MBP-TaGW2 immobilized on amylose resin beads. (D) Co-IP assays of potential interactions between TaGW2-GFP and TaVOZ1-mCherry. (E) HA-HIS-TaVOZ1 ubiquitination assays measuring E3 ligase activity by MBP-TaGW2. Ubiquitinated HA-HIS-TaVOZ1 was detected by anti-MBP and anti-HA antibodies. (F) Cell-free assays of TaGW2-mediated TaVOZ1 degradation in which HA-HIS-TaVOZ1 protein was purified and incubated with equal amounts of crude protein extract from Kenong199 (KN199) or tagw2 mutant plants. After incubation, HA-HIS-TaVOZ1 was measured by immunoblotting with anti-HA antibody at the indicated time points. Degradation was halted for each time point using the proteasome inhibitor, N-carbobenzyloxy-l-leucyl-l-leucyl-l-leucinal (MG132). (G and H) Relative luciferase activity assays in N. benthamiana leaves coinfiltrated with TaVOZ1-LUC and TaGW2-GFP. GFP was used as a negative control. (I and J) Inhibition of TaGW2-GFP–mediated TaVOZ1-LUC degradation by MG132. Blue, weakest luciferase activity; red, strongest luciferase activity. (K) In vivo degradation assays in N. benthamiana leaves transiently expressing TaVOZ1. Protein levels were detected by Western blotting with anti-GFP antibody in the presence of 50 μM MG132 or an equal volume of dimethyl sulfoxide (control). (L) Immunoblot detection of TaVOZ1 in WT Fielder, TaGW2 OE, and RNAi plants using anti-TaVOZ1 antibody, and anti-actin antibody serving as a control. (M) The relative abundance of TaVOZ1 in WT Fielder, TaGW2 OE, and RNAi plants. (N) Immunoblot detection of TaVOZ1 protein levels in tagw2 and KN199 plants. (O) The relative abundance of TaVOZ1 in KN199 and tagw2 plants. Values are means ± SD from at least three independent experiments; statistical significance was determined by a two-sided t test (**P < 0.01).

Because TaGW2 is a RING-type E3 ubiquitin ligase, we tested whether TaGW2 could ubiquitinate TaVOZ1 through in vitro ubiquitination assays with purified MBP-TaGW2 and HA-HIS-TaVOZ1. This analysis showed that TaVOZ1 was indeed ubiquitinated by TaGW2 in the presence of ubiquitin-activating enzyme E1, ubiquitin-conjugating enzyme E2, and Myc–ubiquitin (Ub) (Fig. 5E). However, TaVOZ1 was not ubiquitinated by the E3 ligase–dead variant, mTaGW2, although TaVOZ1 could still interact with mTaGW2 (fig. S8). We then examined TaVOZ1 protein stability in the tagw2 knockout mutant that we previously generated by CRISPR-Cas9–mediated editing in wheat cv. Kenong199 (37). In cell-free degradation assays, incubating purified HA-HIS-TaVOZ1 with total protein extracts of Kenong199 (KN199) resulted in faster degradation of TaVOZ1 than incubation with tagw2 mutant protein extracts, which had little effect on TaVOZ1 stability (Fig. 5F), while the actin or glutathione S-transferase (GST) control proteins showed no such degradation. In addition, this TaVOZ1 clearance by KN199 protein extracts could be blocked by the proteasome inhibitor, N-carbobenzyloxy-l-leucyl-l-leucyl-l-leucinal (MG132) (Fig. 5F). These results demonstrated that TaGW2 could specifically mediate TaVOZ1 degradation.

To confirm that TaGW2 mediates ubiquitin-dependent TaVOZ1 degradation, we also conducted in vivo protein degradation assays in N. benthamiana leaves infiltrated with GFP-tagged TaGW2 (GFP-TaGW2) or TaVOZ1-LUC. After confirming that TaVOZ1 expression was consistent among infiltrations by RT-qPCR and that GFP-TaGW2 was well expressed through Western blots (fig. S9), we quantified LUC activity at 48 hours postinfiltration. We observed that GFP-TaGW2 coexpression with TaVOZ1-LUC resulted in significantly decreased LUC signal compared to that in TaVOZ1-LUC plus GFP-only vector control regions (Fig. 5, G and H), indicating that TaGW2 could mediate TaVOZ1 degradation in planta. Furthermore, LUC signal intensity was restored following MG132 treatment in leaf regions coexpressing GFP-TaGW2 plus TaVOZ1-LUC compared to the corresponding vehicle control sites (Fig. 5, I and J). Consistent with these results, Western blot detection of GFP in N. benthamiana leaves coexpressing GFP-TaVOZ1 with HA-TaGW2 or HA-mTaGW2 confirmed that TaVOZ1 was degraded in the presence of TaGW2, but not the loss of ubiquitination activity variant mTaGW2 (Fig. 5K and fig. S8). These results suggested that TaGW2 could interact with and ubiquitinate TaVOZ1 to mediate its clearance via the 26S proteasomal pathway.

Last, we examined how TaGW2 activity affects TaVOZ1 stability in wheat by quantifying TaVOZ1 protein accumulation in TaGW2 OE, RNAi, and WT Fielder seedlings. Immunoblots showed that TaVOZ1 accumulated to higher levels in TaGW2 RNAi plants but had lower levels in TaGW2 OE lines compared to the WT Fielder controls (Fig. 5, L and M). Further immunoblotting in KN199 and tagw2 seedlings indicated that TaVOZ1 levels were significantly higher in the tagw2 mutant compared to KN199 (Fig. 5, N and O). These results indicated that TaGW2 is indispensable for regulating TaVOZ1 accumulation in wheat.

TaGW2 negatively regulates wheat salt tolerance

Our RNA-seq analysis in 20-day-old wheat seedlings revealed that TaGW2 expression significantly decreased from 0.5 to 6 hours of salt stress (100 mM NaCl) treatment and then gradually increased, peaking at 72 hours (fig. S10A). To investigate the possible role of TaGW2 in salt stress response, we examined the salt tolerance of ~30-day-old TaGW2 OE, RNAi, and WT Fielder seedlings. There were no obvious phenotypic differences between WT and transgenic lines when grown in nonsaline control soil (Fig. 6A). However, following 30 days of growth in high-saline soil (150 mM NaCl), TaGW2 RNAi lines had obviously taller stature than WT, while TaGW2 OE lines had shorter stature compared to WT Fielder plants (Fig. 6B). Similarly, TaGW2 RNAi plants had significantly greater biomass, and OE plants had lower biomass than WT Fielder under high-salt conditions (P < 0.01), while no significant difference in biomass was observed between transgenic and WT lines in control soil (Fig. 6, C and D). TaGW2 OE plants also accumulated higher Na⁺ concentrations in shoots compared to WT, while TaGW2 knockdown plants had lower shoot Na⁺ compared to WT controls, but all lines displayed similar shoot Na⁺ contents under nonsaline conditions (Fig. 6, E and F). After 20 days of hydroponic growth in 150 mM NaCl solution, TaGW2 OE seedlings had significantly shorter roots than WT Fielder plants, while TaGW2 RNAi seedlings showed the opposite trend relative to controls, whereas all lines had similar root lengths following growth in deionized water (Fig. 6, G to J).

Fig. 6. The effects of TaGW2 on wheat salt tolerance.

(A and B) Assessment of salt tolerance in TaGW2 OE and RNAi lines. (C and D) Shoot biomass in TaGW2 OE and RNAi lines under normal growth (C) or 150 mM NaCl treatment under saline soil conditions (D). Values are means ± SDs (n = 20). (E and F) Shoot Na⁺ contents in WT Fielder, TaGW2 OE, and RNAi lines under normal growth (E) or 150 mM NaCl treatment under saline soil conditions (F). Values are means ± SDs (n = 4). (G and H) Root development in TaGW2 OE and RNAi lines following 20 days of growth under deionized water (G) or 150 mM NaCl (H) hydroponic conditions. (I and J) Statistical analysis of root length in TaGW2 OE and RNAi lines. Values are means ± SDs (n = 20). (K) Salt tolerance phenotype in tagw2 mutant and KN199 plants. (L) Statistical analysis of shoot biomass in tagw2 mutant and KN199 plants. Values are means ± SDs (n = 20). (M) Na⁺ contents in tagw2 mutant and KN199 plants grown under normal growth or 150 mM NaCl treatment under saline soil conditions. Data show means ± SDs (n = 4). (N and O) Root development in tagw2 mutant and KN199 plants following 20 days of growth in deionized water (N) or 150 mM NaCl hydroponic conditions (O). (P) Statistical comparison of root lengths between tagw2 mutant and KN199 plants. Values are means ± SDs (n = 20). (Q) Grain morphology of tagw2 mutant and KN199 plants grown under normal growth or 150 mM NaCl treatment under saline soil conditions. (R) Statistical analysis of plant height, kernel length, kernel width, and TKW in plants. Statistical significance was determined by a two-sided t test (**P < 0.01).

In agreement with these findings, examination of tagw2 knockout plants revealed that this mutant line had a taller stature (Fig. 6K), with greater biomass (Fig. 6L), lower shoot Na⁺ contents (Fig. 6M), and longer roots (Fig. 6, N to P) compared to KN199 controls after growth under saline conditions, akin to the TaGW2 RNAi phenotype, while no differences in these traits were observed between tagw2 and KN199 plants under normal growth conditions (Fig. 6, K to P). Evaluation of grain-related traits in greenhouse experiments showed that growth in saline soil could inhibit kernel size and weight compared to growth under nonsaline conditions. However, the tagw2 mutant line had significantly greater kernel size and weight than KN199 controls following growth in saline-treated soil, as well as nonsaline conditions (Fig. 6, Q and R). These results strongly suggested that TaGW2 negatively regulates salt tolerance in wheat, opposite to the regulatory effects of TaVOZ1.

TaGW2 antagonizes TaVOZ1 function in regulating TaHKT expression

Given that several regulatory enzymes are known to exhibit moonlight functions as transcription factors (44–48), and our previous transactivation assays in yeast expressing truncated TaGW2 variants identified a conserved C-terminal transactivation domain (37), we next attempted to investigate the additional mechanisms by which TaGW2 attenuates salt tolerance. Fluorescence microscopy of wheat protoplasts expressing a TaGW2-GFP fusion protein showed that TaGW2-GFP mainly localized in the nucleus and cytoplasm (fig. S10B), and we hypothesized that TaGW2 might function as a transcriptional regulator. We therefore conducted dual-LUC activity assays to test whether TaGW2 could regulate transcription of a LUC reporter. We found that expression of TaGW2 fused to BD-VP16 (BD-VP16–TaGW2) decreased LUC activity compared to that of the BD-VP16– and BD-VP16–actin–positive controls in N. benthamiana leaves (fig. S10C). These results suggested that TaGW2 could potentially act as a transcriptional repressor.

To investigate whether HKT1-family genes (i.e., TaHKT1;1-B, TaHKT1;3-B, and TaHKT1;5-D) could serve as direct targets of TaGW2, we performed DAP-qPCR assays in 20-day-old TaGW2 OE and WT Fielder seedlings. The results showed significant enrichment with sites P1, P2, and P3 (see Fig. 4, D to F) in DAP samples relative to that in input (Fig. 7A). Further examination by EMSA indicated that MBP-TaGW2 could indeed directly bind to DNA fragments containing the P1, P2, or P3 sites from the TaHKT1;1-B, TaHKT1;3-B, and TaHKT1;5-D promoters, and this binding activity could be abolished by mutation of their 5′-AAGAAG-3′ core motif (Fig. 7B). To verify this regulatory activity of TaGW2, we performed dual-LUC transactivation assays using the 1.5-kb promoter fragments of TaHKT1;1-B, TaHKT1;3-B, or TaHKT1;5-D harboring the 5′-AAGAAG-3′ core motif or a mutated 5′-CCCCCC-3′ variant (see Fig. 4, D to F). Upon transforming N. benthamiana leaves expressing pGreenII 62-SK–TaGW2 or the pGreenII 62-SK vector only with a pGreen II-0800 vector harboring LUC driven by a normal or mutant promoter (Fig. 7, C and D), we observed that transient TaGW2 expression decreased LUC signal driven by the normal promoters compared to the vector-only positive controls. However, this negative regulatory effect of TaGW2 was abolished in leaves expressing mutant promoter–driven LUC (Fig. 7, C and D). These results thus demonstrated that TaGW2 functioned as a negative transcriptional regulator of TaHKT1;1-B, TaHKT1;3-B, and TaHKT1;5-D.

Fig. 7. Genome-wide identification of TaGW2 binding sites and target genes.

(A) DAP-qPCR validation of TaGW2 binding sites in HKT1-family gene promoters. The P1, P2, and P3 fragments used for DAP-qPCR are indicated in Fig. 4 (D to F). Values are means ± SDs (n = 4). (B) EMSA of TaGW2-MBP or MBP-only binding to biotin-labeled probes containing core motifs from HKT1-family gene promoters. (C) Schematic of effector and reporter constructs used in transactivation assays. (D) Transactivation assays of TaGW2 antagonism of TaVOZ1 function, resulting in HKT1 repression. Values are means ± SDs (n = 4). n.s., not significant. (E) Competitive EMSA assay of TaGW2-MBP and TaVOZ1-GST binding to biotin-labeled probe containing CTTCTT core motif. (F) Volcano plots of differentially expressed proteins in tagw2 mutant plants relative to KN199 plants (|FC| > 2 and P < 0.01). RPKM, reads per kilobase of transcript per million mapped reads. (G) Genome-wide distribution of high-confidence TaGW2 binding peaks in different gene regions from DAP-seq data. (H) Venn diagram of overlapping genes between RNA-seq and DAP-seq analyses. (I) Distribution of overlapping up- and down-regulated genes from (H). (J) Sequence logos for enriched motifs within the TaGW2 binding peaks. The TCTTCTTCTT and GAAGAAGAAG motifs showed the greatest enrichment among core sequences. E values were calculated by MEME. (K and L) GO analysis of up-regulated (K) and down-regulated (L) genes in (I). P values were adjusted by Benjamini-Hochberg correction; only significant categories (P < 0.01) are displayed.

Moreover, we performed EMSA experiments examining competitive binding between TaGW2 and TaVOZ1 for DNA fragments containing the CTTCTT/AAGAAG motif. These assays yielded distinct TaVOZ1- and TaGW2-bound DNA bands. Among them, TaGW2-MBP exhibited obviously stronger binding affinity for the labeled probe compared to TaVOZ1-GST, with TaGW2-MBP binding to the probe increasing with the addition of more protein, while TaVOZ1-GST binding to the labeled probe decreased with further supplementation of TaGW2-MBP. These results indicated that TaGW2 and TaVOZ1 competitively bind to the CTTCTT/AAGAAG motif (Fig. 7E). To further determine whether TaGW2 functions coordinately or antagonistically with TaVOZ1 in regulating TaHKT1;1-B, TaHKT1;3-B, and TaHKT1;5-D, we conducted further dual-LUC assays. To block TaGW2 function in TaVOZ1 ubiquitination and degradation, we used an E3 ligase–dead variant, mTaGW2 (fig. S8), that retained the normal DNA binding activity and therefore could still repress the transcription of TaHKT1;1-B, TaHKT1;3-B, and TaHKT1;5-D (fig. S11). In N. benthamiana leaves coexpressing mTaGW2 with TaVOZ1, LUC signal driven by the TaHKT1;1-B, TaHKT1;3-B, and TaHKT1;5-D promoters was significantly lower than that in leaves expressing TaVOZ1 alone (Fig. 7, C and D). Subsequent RT-qPCR analysis revealed that TaHKT1;1, TaHKT1;3, and TaHKT1;5 were expressed at significantly higher levels in the tagw2 mutant and TaGW2 RNAi lines compared to WT KN199 or Fielder, while their expression in TaGW2 OE lines was lower than that in WT control plants (fig. S12). These cumulative findings demonstrated that TaGW2 could function as a transcriptional repressor to antagonize TaVOZ1-mediated activation of TaHKT1;1, TaHKT1;3, and TaHKT1;5, potentially underpinning its negative regulatory effects on wheat salt tolerance.

TaGW2 regulates the expression of a wide array of genes

In light of its function as an apparent transcriptional repressor of HKT genes, we next searched for other possible regulatory targets of TaGW2. Our previous RNA-seq analysis identified 9166 DEGs in tagw2 mutant plants compared to KN199 (FC > 2 or < 0.5 and P < 0.01) (37), including 3812 up-regulated and 5354 down-regulated genes (Fig. 7F). To define the possible regulatory network controlled by TaGW2, we applied DAP-seq (41, 42) to identify DNA binding sites and target genes of TaGW2. DAP-seq data from two biological replicates identified 6340 candidate TaGW2 binding sites spanning 4260 genes (table S8). Further analysis indicated these binding sites were located within gene bodies and 2-kb upstream and downstream flanking sequences likely to contain the promoters and transcriptional terminators, respectively. Specifically, 31.96, 22.14, 22.72, 9.26, 9.18, 3.49, and 1.25% of the predicted TaGW2 binding sites were respectively detected in distal intergenic regions, promoter regions, introns, exons, 5′-UTR, 3′-UTR, and terminator regions (Fig. 7G).

By examining the DAP-seq results together with our RNA-seq analyses, we found 1419 genes that could potentially serve as direct binding targets for TaGW2 (Fig. 7H and table S9), including 615 up-regulated and 804 down-regulated genes in the tagw2 mutant (Fig. 7I). Further MEME-ChIP identified several significantly enriched TaGW2 binding motifs in the consensus sequences of binding sites, among which 5′-GAAGAAGAAG-3′ and 5′-TCTTCTTCTT-3′ showed the highest enrichment (Fig. 7J). GO analysis of target genes harboring these promoter motifs showed that up-regulated genes were enriched in terms related to “cell cycle,” “cell division,” “response to stress,” “cytoplasmic translation,” “embryo development,” and “leaf morphogenesis” (Fig. 7K), while down-regulated candidate target genes were enriched in “response to wounding,” response to water deprivation, response to abscisic acid, response to salt stress, response to osmotic stress, transcriptional regulation, and “regulation of stomatal movement” (Fig. 7L). These results suggested that TaGW2 might participate in overlapping regulatory networks responsible for developmental and stress-responsive processes in wheat.

Last, we searched for other proteins that share substantial homology with TaGW2 outside of the RING domain in Arabidopsis and other crop plants (fig. S13), which uncovered a set of proteins with 41.5 to 90% similarity to TaGW2. Among them, GW2 from rice shared 86.9% amino acid sequence similarity with TaGW2, while Arabidopsis DA2 shared 41.5% similarity. As previous studies reported that transient expression of GFP-GW2 or GFP-DA2 in N. benthamiana leaves resulted in both nuclear and cytoplasmic GFP signals (33, 49, 50), we tested whether DA2 or GW2 exhibited similar functions to those of TaGW2 in repressing transcription. EMSA results showed that neither DA2 nor GW2 could bind the 5′-CTTCTT-3′ or 5′-AAGAAG-3′ core motifs in vitro (fig. S14), suggesting that TaGW2’s transcriptional repressor activity was a distinct role in wheat response to salt stress.

DISCUSSION

High soil salinity poses a serious challenge for grain production in many growing regions (1–8). Characterizing the key genes and coordinated mechanisms of salt tolerance can facilitate the development of high-yielding, salt-tolerant wheat varieties essential for global food security and sustainable agriculture. In the present study, we found that TaVOZ1 positively regulates salt tolerance in wheat by activating the transcription of stress response genes, especially HKT1-family genes (Figs. 2 to 4) and that TaGW2 mediates TaVOZ1 ubiquitination to promote its 26S proteasomal degradation (Fig. 5). While TaVOZ1 enhances plant growth and kernel morphology under salt stress, TaGW2 negatively affects wheat growth and grain production under high-salt conditions (Fig. 6). We found a previously unreported moonlight function of TaGW2, wherein it functions as a transcriptional repressor that can directly bind the same DNA recognition motifs as TaVOZ1 to block up-regulation of stress-responsive genes, such as HKT1 genes (Fig. 7 and fig. S10). These results suggest a potentially viable strategy for breeding salt-tolerant wheat, and point to TaVOZ1 and especially TaGW2, as strong candidate targets for molecular breeding of elite wheat varieties.

Although VOZ genes were first found because of their role in initiating flowering in Arabidopsis (51, 52), further examination uncovered functions in plant responses to drought, salt, and biotic stresses (23–26, 40, 53, 54). Results in our study also show that TaVOZ1 positively regulates biomass production, root length, plant height, kernel size, and grain yield while reducing Na⁺ accumulation in shoots during salt stress, whereas TaVOZ1 knockdown abolishes this salt-tolerant phenotype (Fig. 2). Specifically, our assays show that TaVOZ1 is a transcriptional activator (Fig. 3, A to C), with 4436 putative targets and numerous significantly enriched recognition motifs detected in DAP-seq assays, which included 2347 DEGs identified by RNA-seq in TaVOZ1 OE versus WT Fielder plants (Fig. 3, D to G, and figs. S3 and S4). Enrichment for “stress response” terms in these up-regulated DEGs supports a scenario in which TaVOZ1 can positively regulate salt tolerance through transcriptional regulation of stress-responsive genes and pathways, especially the Na⁺ transporter genes and ion homeostasis pathway (Fig. 3H).

HKT proteins, including HKT1, are membrane-associated transporters known to export Na⁺ ions from root xylem cells to limit translocation, accumulation, and toxicity in aerial tissues, such as stems and leaves (11–17). Targeted breeding of crop lines with increased HKT1 expression has been well investigated as an approach for developing elite, salt-tolerant varieties (8, 43). In wheat, the TaSPL6-D^Del allele confers a salt-sensitive phenotype due to its direct interactions and, consequently, inhibitory regulatory effects on TaHKT1;5-D. Alternatively, the TaSPL6-D^In allele harbors a 47–base pair (bp) insertion in the first exon, leading to translation of a truncated protein that cannot repress TaHKT1;5-D expression. Thus, introducing the TaSPL6-D^In allele into a salt-sensitive, high-performance wheat cultivar could improve grain yields under saline soil conditions (8). In the present study, we found that promoter regions of all 14 HKT1-family genes in wheat contain at least one TaVOZ1-binding motif (5′-CTTCTT-3′ or 5′-AAGAAG-3′; table S6), indicating that TaVOZ1 can directly regulate their expression. In TaVOZ1 OE wheat plants, TaHKT1;1-B, TaHKT1;3-B, and TaHKT1;5-D were all up-regulated, whereas their expression could be suppressed by RNAi targeting TaVOZ1 (Fig. 4, A to C). These HKT1 genes were also confirmed as downstream targets for transcriptional up-regulation by TaVOZ1 through EMSA, DAP-qPCR, and dual-LUC assays (Fig. 4, D to K). Furthermore, wheat plants overexpressing TaHKT1;3-B displayed significantly increased salt tolerance compared to their corresponding vector controls (Fig. 4, L to S).

While our findings thus illustrate a TaVOZ1-mediated mechanism for enhancing wheat response to salt stress via activation of HKT1 expression, previous studies have shown that VOZ-family proteins are suppressed during drought stress, cold stress, and plant immune response through E3 ubiquitin ligase–mediated degradation (40, 55). Our results also indicate that the E3 ligase, TaGW2, can directly interact with TaVOZ1 to positively regulate its clearance by 26S proteasomal degradation (Fig. 5). In wheat, TaGW2 was found to serve as an essential regulator in modulating kernel size and responses to drought stress, leaf rust, and stripe rust infection (35–39). Further multiple lines of evidence in our current study indicate that TaGW2 positively regulates salt stress tolerance, as transgenic TaGW2 OE lines generally display lower biomass production, shorter roots, and increased Na⁺ accumulation in shoots compared to WT Fielder following growth in high-saline soil. In addition, wheat TaGW2 knockdown lines, as well as the tagw2 knockout mutant, show decreased shoot accumulation of Na⁺ compared to WT, along with increased biomass and longer roots, typical of a salt tolerance phenotype (Fig. 6). These findings, together with previous reports, illustrate a major function of TaGW2 in coordinately regulating wheat yield and response to salt stress. It warrants mention that the ubiquitin-dependent clearance of RING-type E3 ligase substrates has been previously established as an important mechanism for regulating plant response to salt (56–58). The identification and characterization of these E3 ligase–transcription factor regulatory modules reveal the molecular basis of salt tolerance in crops and thus provide helpful information for molecular breeding.

In addition, many proteins have been shown to mediate secondary or moonlight functions apart from their canonical activity to participate in adaptive mechanisms underlying responses to complex or stress-related conditions (59–61). Among such moonlight functions, several proteins have been shown to act as transcription regulators in different model organisms (44–48), although it has remained uncertain whether and how such proteins might contribute to stress response in plants. Our work provides the first evidence, to our knowledge, that the E3 ubiquitin ligase TaGW2 can act as a transcriptional regulator to modulate multiple downstream developmental and stress response processes (Fig. 7 and fig. S10). We observed that TaGW2 localizes to both the nucleus and cytoplasm (fig. S10B) and could directly bind to DNA fragments carrying the 5′-CTTCTT-3′ or 5′-AAGAAG-3′ core motifs (Fig. 7, A and B). Dual-LUC assays showed that TaGW2 can inhibit TaVOZ1 activity in up-regulating target genes such as TaHKT1;1-B, TaHKT1;3-B, and TaHKT1;5-D by binding to the same DNA recognition sites as TaVOZ1 (Fig. 7, C and D). Closer examination of TaVOZ1 expression patterns across various tissues in publicly available transcriptomic data (table S10) revealed that TaVOZ1 is highly expressed in kernels and seedling leaves and roots, similar to the pattern of TaGW2 expression in different tissues (37). Moreover, we observed that TaGW2 and TaVOZ1 exhibited inverse expression patterns in the same tissue during salt stress (fig. S1 and fig. S10A), wherein TaGW2 showed significantly suppressed expression, which could allow activation of a TaVOZ1-mediated salt stress response. Furthermore, our combined DAP-seq and RNA-seq analyses identified 1419 candidate target genes that are potentially regulated by DNA binding activity of TaGW2 (Fig. 7, F to H, and table S9), which were enriched in various biological processes, such as cell cycle, cell division, photosynthesis, response to salt stress, response to water deprivation, response to abscisic acid, response to osmotic stress, and regulation of stomatal movement (Fig. 7, K and L). The broad diversity of these potentially affected pathways implies a versatile function of TaGW2 in mediating the trade-off between stress response and yield in wheat.

The extensive genetic, biochemical, and phenotypic evidence obtained in our study supports a working model in which the TaGW2-TaVOZ1 regulatory module functions antagonistically to regulate wheat salt stress response through both E3 ligase–dependent and –independent mechanisms (Fig. 8). Specifically, TaGW2 activation promotes TaVOZ1 ubiquitination and subsequent degradation in WT wheat plants. Further, TaGW2 displays moonlight function as a transcriptional repressor, binding to the same DNA recognition motifs as TaVOZ1 to hinder its up-regulation of HKT1-type genes (i.e., TaHKT1;1-B, TaHKT1;3-B, and TaHKT1;5-D). Alternatively, TaVOZ1 accumulates to high levels in tagw2 mutant plants and thus maintains high transcription of stress-responsive genes, ultimately enhancing salt tolerance. These cumulative results illustrate a molecular genetic basis for salt tolerance in wheat through a TaGW2-TaVOZ1 regulatory module, suggesting potentially effective targets for improving salt tolerance in wheat.

Fig. 8. A working model of the TaGW2-TaVOZ1-TaHKT1 family regulatory module function in wheat response to salt stress.

In WT plants, TaGW2 activation promotes TaVOZ1 ubiquitination and subsequent degradation. TaVOZ1 can directly bind recognition motifs in promoters of HKT1-family genes (e.g., TaHKT1;1-B, TaHKT1;5-D, and TaHKT1;3-B) to activate their expression. In addition, TaGW2 displays a moonlight function as a transcriptional repressor that antagonizes positive regulation by TaVOZ1 through competitive binding to the same TaVOZ1 recognition motifs. In the tagw2 knockout mutant, TaVOZ1 is not ubiquitinated or functionally suppressed by TaGW2, resulting in its accumulation, and consequently, increased transcription of stress-responsive genes, ultimately enhancing the salt tolerance phenotype of the tagw2 mutant.

METHODS

Plant materials and growth conditions

The CRISPR-Cas9–mediated tagw2 mutant, TaGW2 OE, and TaGW2 RNAi transgenic lines were obtained from our previous study (37). TaVOZ1 and TaHKT1;3-B OE vector constructs were built by amplifying and cloning their respective coding regions into the pCAMBIA3301 vector, driven by the ubiquitin promoter. To induce TaVOZ1 knockdown through RNAi, a 205-bp fragment with the highest homology among three subgenome TaVOZ1 homoeologs was amplified and inserted into the pC336 (Ubi:GWRNAi:Nos) plasmid using the Gateway cloning method. All Sanger sequencing confirmed constructs were introduced into immature embryos of wheat cv. Fielder using Agrobacterium tumefaciens–mediated transformation (62). All primers used for plasmid construction are listed in table S11.

To cultivate wheat plants, seeds were first surface sterilized and stored in the dark for 3 days at 4°C, then moved to room temperature, and exposed to white light. After germination, seeds were planted in pots [7 cm by 10 cm (diameter × depth)]. The wheat seedlings were grown in a controlled environment chamber with the following settings: 14°C day and 12°C night, a 16-hour light/8-hour dark cycle, and light intensity of 200 μmol m⁻² s⁻¹. For reproductive stages, the wheat plants were grown in a greenhouse at a relative humidity of 60%, 22°/20°C day/night temperatures, and a 16-hour light/8-hour dark photoperiod, with a light intensity of 3000 lux.

Salt tolerance assay

Salt tolerance assays were performed using a previously established method (63). Briefly, TaVOZ1 OE, TaVOZ1 RNAi, WT Fielder, tagw2, and KN199 seeds were planted in 7 cm–by–10 cm plastic pots (12 per pot) containing well-mixed growth substrate and irrigated to complete saturation with either deionized water (control) or 150 mM NaCl solution. After 30 days of growth in the controlled environment chamber, the differences between the experimental groups and the control groups were photographed. Subsequently, the shoot biomass of each genotype was measured individually.

In addition, to measure the primary root length, sterilized seeds of TaVOZ1 OE, TaVOZ1 RNAi, WT Fielder, tagw2, and KN199 were grown in deionized water with or without 150 mM NaCl for 20 days. The primary root length of each genotype was measured using 20 independent seedlings. Three independent biological experiments were carried out with similar results, and the representative results are shown in the figures.

Measurement of Na⁺ concentration

To measure shoot Na⁺ contents, the samples of each genotype were desiccated for 48 hours at 80°C for 2 days until they reached a constant weight. The samples were then calcined at 300°C in a muffle furnace for 3 hours, and after which, heat was increased to 575°C for an additional 6 hours. The ash from each sample was dissolved in 10 ml of 1% hydrochloric acid, and the solution was diluted with 1% hydrochloric acid as needed for the determination of ion concentration. The Na⁺ concentrations were measured by a 4100-MP AES device (Agilent, Santa Clara, CA). The experiment was conducted with four biological replicates.

Yield trait evaluation under salt stress condition

Evaluation of yield-related traits in TaVOZ1 OE, TaVOZ1 RNAi, WT Fielder, tagw2, and KN199 plants under salt stress conditions was conducted in plastic pots [0.3 m by 0.35 m (diameter × depth)] using a 3:1 soil:vermiculite mixture. Each line was grown in groups of five plants per pot in 15 identical plastic pots. For salt stress treatment, plants were regularly watered until the head sprouting period; then, 150 mM NaCl was added. Yield characteristics for all lines were recorded, and statistical data were based on at least 20 plants per line.

RNA extraction and RT-qPCR

Total RNA extraction was performed with HiPure Plant RNA Mini Kit (Vazyme, R4151-02), and after which, 1 μg of RNA was used as template for single-strand cDNA synthesis with HiScript II Q RT SuperMix (Vazyme, R223-01), following instructions in the manual. A CFX Connect Real-Ttime system (Bio-Rad, California, USA) was used for qPCR assays, using TaActin1 as the internal reference. Relative expression levels were calculated with the 2^–ΔΔCt method (64) using at least three biological replicates for each sample.

Subcellular localization

To conduct subcellular localization assays, TaVOZ1-GFP and TaGW2-GFP fusion reporters were generated by individually cloning the TaVOZ1 and TaGW2 coding sequences, lacking a stop codon, into the N-terminal region of the pJIT163-GFP vector. The vector constructs or GFP-only vector control were subsequently transformed into wheat protoplasts via polyethylene glycol–mediated transfection. The florescence signal was observed by confocal microscopy (Olympus, FV3000, Japan) after incubation in darkness for 16 hours. The laser wavelength for GFP was set at 488 nm, and the detection wavelength ranged from 500 to 540 nm, while the autofluorescence of chloroplasts was set at 488 nm, and the detection wavelength ranged from 650 to 750 nm. Three independent biological experiments were carried out with similar results, and the representative results are shown in the figures.

Transactivation assay

To identify the transactivation activity of TaVOZ1 in yeast, the full-length coding sequence was constructed into the pGBKT7 vector and then transformed into the yeast strain AH109. Positive clones were screened and transformed to SD/-Trp-His-Ade medium containing X-α-Gal to assess their transactivation activity. To assess the transcriptional regulatory functions of TaGW2 and TaVOZ1, a well-established approach was used. Specifically, the transcriptional activation region of the herpes simplex virus VP16 was fused with the GAL4 DNA BD to construct the pGBKT7-VP16 vector. Subsequently, TaGW2 and TaVOZ1 were respectively fused with pGBKT7 and pGBKT7-VP16 to construct the pGBKT7-TaGW2, pGBKT7-TaVOZ1, pGBKT7-TaGW2-VP16, and pGBKT7-TaVOZ1-VP16 vectors. The constructs were transformed into Y2HGold cells, and the diluted yeast cells were then inoculated on SD/-Trp, SD/-Trp-His, and SD/-Trp-His-Ade media. pGBKT7 and pGBKT7-VP16 were used as negative and positive controls, respectively.

Y2H assay

For Y2H screens of potential TaVOZ1 interaction with TaGW2, we respectively cloned full-length or progressively truncated fragments of the TaGW2 protein coding sequence or the E3 ligase–dead variant, mTaGW2, into the pGBKT7 vector to serve as baits, while the TaVOZ1 protein coding region was cloned into pGADT7. The constructed bait and prey were cotransformed into the yeast strain AH109, and the cells were grown on SD/-Trp-Leu. Positive clones were diluted (1:10) and plated on SD/-Trp-Leu-His-Ade medium containing X-α-Gal and 25 mM 3-AT. Images were captured after a 3-day incubation at 30°C.

LCI assay

For LCI assays, we fused the TaGW2 or mTaGW2 protein coding sequence to the N-LUC reporter in the pCAMBIA1300-nLuc vector, and the coding region of TaVOZ1 was fused downstream of C-LUC in the pCAMBIA1300-cLuc vector. The constructs were then transformed into the Agrobacterium strain GV3101. Subsequently, TaGW2-nLUC or mTaGW2-nLUC was coinjected with cLUC-TaVOZ1 into 30-day-old N. benthamiana leaves in a proportion of 1:1, respectively. GUS-nLUC and cLUC-GUS were used as controls. After 48 hours, luminescent signals were detected using the PlantView100 Multispectral Dynamic Fluorescence Microscopy Imaging System (BLT Photon Technology, Guangzhou, China). Three independent biological experiments were carried out with similar results, and the representative results are shown in the figure.

In vitro pull-down assay

The full-length coding region of TaGW2 was inserted into the pMAL-c5X vector to generate the TaGW2-MBP fusion construct. Similarly, the coding region of TaVOZ1-HA was cloned into the pET-15b vector, creating the TaVOZ1-HA-HIS recombinant construct. Subsequently, these vectors were transformed into the E. coli BL21 (DE3) strain for the expression of the corresponding proteins. The proteins were purified using amylose resin beads (New England Biolabs) and Ni–nitrilotriacetic acid resin beads (Thermo Fisher Scientific), respectively. For the in vitro pull-down assay, mixtures of equal amounts of TaVOZ1-HA-HIS protein were respectively incubated with MBP and TaGW2-MBP and then subjected to IP using amylose resin beads at 4°C for 3 hours. The resin was collected by centrifugation at 500g for 5 min and washed five times with tris-buffered saline [10 mM tris (pH 7.4), 150 mM NaCl, 1 mM EDTA, and 1% Triton X-100]. The proteins were separated by 10% (w/v) SDS–polyacrylamide gel electrophoresis (SDS-PAGE) and detected with anti-HA (Beyotime, AH158-IgG1) and anti-MBP (Proteintech Group, 66003-1-Ig) antibodies.

In vivo Co-IP assay

The Co-IP experiment was conducted following the protocols previously described (37). Briefly, GFP-tagged TaGW2 and mCherry-tagged TaVOZ1 were transformed into Agrobacterium GV3101 and infiltrated into 30-day-old N. benthamiana leaves. After 2 days, total proteins were extracted from the infected leaves by homogenization in radioimmunoprecipitation assay (RIPA) buffer (Beyotime, P0013D). After incubation at 4°C for 3 hours with 20 μl of GFP-Trap beads (ChromoTek, GTA-20), the immunoprecipitated samples were washed five times with washing buffer [0.5 mM EDTA (pH 8.0), 10 mM tris/HCl (pH 7.5), 150 mM NaCl, 1 mM dithiothreitol (DTT), 1 mM phenylmethylsulfonyl fluoride, 1% Triton X-100, and 1× protease inhibitor] and subjected to immunoblot analysis. The total protein extracts as inputs and the immunoprecipitated proteins bound to the beads were detected by Western blotting with anti-GFP (Beyotime, AF0159) or anti-mCherry (Abbkine, 40125) antibodies.

In vitro ubiquitination assay

For the in vitro ubiquitination assays, TaGW2, mTaGW2, TaVOZ1-HA, and TaVOZ1 were separately cloned into pMAL-c5X, pET-15b, or pGEX4T-1 vectors to purify TaGW2-MBP, mTaGW2-MBP, TaVOZ1-HA-HIS, and TaVOZ1-GST recombinant proteins. The purified proteins TaGW2-MBP and mTaGW2-MBP were mixed with TaVOZ1-HA-HIS or TaVOZ1-GST and added to 60 μl of ubiquitination reaction buffer containing 50 mM tris-HCl (pH 7.5), 5 mM MgCl₂, 5 mM adenosine 5′-triphosphate (ATP), 25 mM DTT, 2 μg of ubiquitin (Boston Biochem), 50 ng of E1 (recombinant human UBE1, Boston Biochem), and 100 ng of E2 (recombinant human UbcH5b/UBE2D2, Boston Biochem). The mixture was incubated at 30°C for 2 hours. The reaction was terminated by adding 5× protein loading buffer and boiling for 10 min. The ubiquitinated TaVOZ1-HA-HIS and nonubiquitinated TaVOZ1-GST proteins were detected with anti-ubiquitin (HUABIO, ET1609-21), anti-HA (Beyotime, AH158-IgG1), and anti-GST (Beyotime, AM926) antibodies, respectively, while the TaGW2-MBP or mTaGW2-MBP proteins were detected with anti-MBP (Proteintech Group, 66003-1-Ig) antibody.

In vivo ubiquitination assay

For in vivo ubiquitination assays, the recombinant proteins TaGW2-HA and TaVOZ1-GFP, mTaGW2-HA and TaVOZ1-GFP, and TaGW2-GFP and TaVOZ1-LUC were transiently expressed in N. benthamiana leaves through Agrobacterium-mediated transformation (65). The infiltrated leaves were injected with 50 μM MG132 for 10 hours before extracting total proteins or observing the luminescence. GFP was used as a negative control. The proteins were boiled for 10 min and detected via immunoblotting with anti-GFP (Beyotime, AF0159) and anti-HA (Beyotime, AH158-IgG1) antibodies. d-Luciferin potassium (Meilunbio) was infiltrated into the leaves, and the luminescence of leaves was observed using the PlantView 100 (BLT Photon Technology, Guangzhou, China). To determine the transcriptional level of TaVOZ1 in each sample, RT-qPCR was performed. Each experiment was conducted with at least three biological replicates, and similar results were obtained. Representative images were selected for presentation.

Cell-free protein degradation

The RIPA lysis buffer (Beyotime, P0045) was used to extract total proteins from 14-day-old seedlings of tagw2 mutant and WT KN199 plants. Soluble target protein TaVOZ1-HA-HIS was acquired through prokaryotic expression and purification. Subsequently, equal amounts of TaVOZ1-HA-HIS and the above total proteins were incubated at 30°C in a reaction buffer containing 10 mM ATP, 10 mM MgCl₂, 5 mM DTT, 100 mM NaCl, and 25 mM tris-HCl (pH 7.5) with or without 50 μM MG132. The reaction system was terminated by adding 5× protein loading buffer at different time points (0, 20, 40, 60, and 80 min), followed by boiling for 10 min. The degradation of TaVOZ1-HA-HIS protein was detected by immunoblotting with anti-HA (Beyotime, AH158-IgG1) antibody. The wheat housekeeping gene Actin was used as an internal reference and detected by immunoblotting with anti-actin (ABclonal, AC009) antibody.

DAP-seq analysis

A fragmented genomic DNA library of wheat cv. Chinese Spring was established according to the method described previously (42). The coding regions of TaGW2 or TaVOZ1 were cloned into the pFN19K HaloTag T7 SP6 Flexi vector. Subsequently, the fusion protein was expressed in vitro using the TNT SP6–coupled wheat germ extract system (Promega). The expressed proteins were directly captured using Magne HaloTag Beads (Promega) and then incubated with the genomic DNA library. The DNA samples bound to the proteins were tagged with dual-indexed multiplexing barcodes and sequenced on the Illumina HiSeq 2500, generating single-end reads of 150 bp in length. The quality of raw reads was inspected using FastQC (v.0.11.9), and then the reads were trimmed with Trimmomatic (66). The trimmed reads were aligned to the wheat reference genome (IWGSC RefSeq v.1.1), and peaks were detected using GEM v2.5 (67). The motifs with the highest protein enrichment levels within 100-bp upstream and downstream of each peak were identified using MEME-ChIP (68).

RNA-seq analysis

The RNA-seq analysis was carried out on 20-day-old wheat seedlings of TaVOZ1 OE and WT Fielder treated with 100 mM NaCl for 0 and 12 hours. The RNA samples were extracted using the TRIzol method, and sequencing was conducted on three biological replicates. For library construction, 5 μg of total RNA was used, and the libraries were quantified using the Qubit 2.0. Sequencing of libraries was performed via Illumina HiSeq 2500, and the quality control of the obtained raw data was conducted through FastQC (v.0.11.9). The data were then mapped to the wheat reference genome (IWGSC RefSeq v.1.1) using GEM v2.5 (67). Read counting and the measurement of DEGs were performed using DESeq2. Significant DEGs were defined as those with a log₂ absolute value of the fold change (FC) greater than 1 and a P value less than 0.01. Significantly up-regulated and down-regulated genes were subjected to GO enrichment analysis using the GOseq R package (69).

Construction of coexpression networks

To identify genes that might be conjointly regulated by salt exposure along with TaVOZ1, we conducted bulk RNA-seq analysis of 20-day-old seedlings from wheat cv. Chinese Spring collected at 0.5 through 72 hours of 100 mM NaCl treatment. Coexpression networks were constructed using the WGCNA R package to identify gene modules with similar expression patterns in response to salt stress. Topographical overlap matrices (TOM) were calculated by the blockwiseModules, with parameters “corType = pearson, TOMType = unsigned, and mergeCutHeight = 0.30.” The soft-thresholding power was set to 12, which was determined to be the lowest value that achieved a scale-free topology fit index > 0.8. We then calculated pairwise Pearson correlation coefficients between transcript levels of coexpressed genes in modules containing TaVOZ1 and each individual TaVOZ1 homoeolog across all samples. The resulting TaVOZ1-centric subnetwork was then visualized using Gephi (v.0.1).

Electrophoretic mobility shift assay

EMSA assays were performed by a LightShift Chemiluminescent EMSA Kit (Thermo Fisher Scientific, 20158) according to the manufacturer’s protocol. To investigate the binding ability of TaGW2, mTaGW2, or TaVOZ1 to the promoters of TaHKT1;1-B, TaHKT1;3-B, and TaHKT1;5-D, the recombinant proteins TaGW2-MBP, mTaGW2-MBP, and TaVOZ1-HA were purified. Approximately 30-bp sequences containing the motif AAGAAG were selected from the promoters of TaHKT1;1-B, TaHKT1;3-B, and TaHKT1;5-D to generate the nonlabeled or biotin-labeled probes. The AAGAAG motif was mutated to CCCCCC to generate mutant probes. Subsequently, the recombinant proteins were incubated with each biotin-labeled probe at the corresponding concentrations in 5× EMSA/gel-shift binding buffer at room temperature for 20 min. To determine binding specificity, unlabeled or mutant probes were added to the reaction mixture as controls. The products were separated by 6% native PAGE at 100 V in 0.5 × tris-borate EDTA buffer for 1 hour and then transferred to a positively charged nylon membrane (Millipore, INYC00010). Chemiluminescent nucleic acid detection modules were used to detect the binding bands on the membrane. The probe sequences used in the EMSA are listed in table S11.

Dual-LUC reporter assay

For the dual-LUC reporter assays, the ~1.5-kb promoter sequences or the CCCCCC-based mutated promoter sequences located upstream of TaHKT1;1-B, TaHKT1;3-B, and TaHKT1;5-D were cloned into the pGreen II-0800-LUC vector to serve as reporters. Meanwhile, the full-length coding sequences of TaGW2 and TaVOZ1 were cloned into the pGreenII 62-SK vector to act as effectors. Dual-LUC reporter assays were performed as previously described (70). The effector and reporter were individually introduced into Agrobacterium strain GV3101 and then cotransfected into N. benthamiana for 2 days. Meanwhile, GFP was used as a negative control in this experimental setup. The results were detected using the PlantView100 (BLT Photon Technology, Guangzhou, China). The activities of LUC and REN were detected with a Dual-Luciferase Reporter Assay Kit (Vazyme, DL101) following the manufacturer’s instructions.

Phylogenetic analysis

MEGA7.0 software was used to conduct alignment and analyze the phylogenetic relationships of protein sequences. The default parameters were used for both pairwise and multiple alignments, and the neighbor-joining method was applied to build phylogenetic trees. The following parameters were used: Poisson correction, complete deletion, uniform rates, and bootstrap (1000 replicates).

Statistical analysis

A two-tailed t test was used for significant difference analysis for two-group comparison. On the basis of the thresholds of P < 0.05 and P < 0.01, the mean values were considered to have significant differences, denoted by * and **, respectively.

Supplementary Materials

The PDF file includes:

Figs. S1 to S14

Legends for tables S1 to S11

Other Supplementary Material for this manuscript includes the following:

Tables S1 to S11