Overexpression of IbGAD4 enhances salt tolerance in sweetpotato by modulating the GABA pathway

Abstract

γ-Aminobutyric acid (GABA) is a non-proteinogenic amino acid that plays a pivotal role in plant development and adaptation to environmental stressors. Although sweetpotato (Ipomoea batatas) is a nutritionally valuable crop, the regulatory mechanisms governing the GABA biosynthetic pathway and its role in salt stress tolerance remain poorly understood. In this study, we identified and functionally characterized IbGAD4, a glutamate decarboxylase gene from sweetpotato. IbGAD4 was predominantly expressed in roots and developmentally regulated, with expression peaking at 90 days in storage roots. Meanwhile, IbGAD4 expression was induced by salt stress. Overexpression of IbGAD4 in transgenic sweetpotato significantly elevated endogenous GABA levels and upregulated key genes in the GABA shunt and tricarboxylic acid (TCA) cycle. Under salt stress conditions, IbGAD4 overexpression enhanced plant tolerance, as evidenced by improved growth, increased antioxidant enzyme activity, reduced malondialdehyde (MDA) levels, and transcriptional activation of salt-responsive genes, including components of the SOS pathway and ABA biosynthesis. Yeast one-hybrid screening using fragments of the IbGAD4 promoter identified IbMYB44 as a transcriptional repressor of IbGAD4. Notably, IbMYB44 expression was downregulated by salt stress, suggesting a stress-responsive regulatory circuit. This study provides a foundation for developing sweetpotato cultivars with improved resilience to abiotic stress by modulating the GABA pathway.

Supplementary Information

The online version contains supplementary material available at 10.1186/s12870-026-08247-w.

Introduction

Sweetpotato (Ipomoea batatas [L.] Lam.), a trailing herbaceous plant in the family Convolvulaceae, originated in Central and South America and was introduced to China during the Ming Dynasty, where it gradually became widely cultivated and utilized [1]. As the seventh most important food crop globally [2], sweetpotato is rich in nutrients such as proteins, anthocyanins, and dietary fiber [3], making it a valuable resource for developing functional foods. However, sweetpotato growth, development, and yield are often compromised by both biotic and abiotic stresses. Among which, salt stress is a major limiting factor. Therefore, improving salt tolerance of sweetpotato is a key objective in current breeding and molecular research.

γ-aminobutyric acid (GABA) is a functional non-proteinogenic amino acid widely found in living organisms [4]. In plants, GABA regulates intracellular pH, maintains carbon/nitrogen balance [5], influences growth and development [6], and participates in multiple physiological, biochemical, and metabolic processes associated with environmental adaptation and abiotic stress tolerance. GABA is also implicated in seed germination, adventitious root formation [7], fruit ripening, and various abiotic stress responses. For example, it promotes α-amylase expression during germination to facilitate starch breakdown and energy supply [8], and its levels fluctuate significantly during tomato fruit development [9]. Under abiotic stress conditions, GABA enhances plant tolerance by regulating intracellular pH, modulating ion transport, and activating antioxidant defense [10]. During salt stress, GABA supports seed germination by enhancing Na⁺/K⁺ transport, promoting osmolyte accumulation, and maintaining osmotic balance [11]. In addition, endogenous GABA enhances salinity tolerance in rice [12]. Under drought conditions, exogenous GABA induces the accumulation of abscisic acid (ABA) and activates ABA signaling to promote stomatal closure, thereby enhancing drought tolerance in apple [13].

In plants, GABA is primarily synthesized through the GABA shunt and polyamine degradation pathways. Most studies focus on the GABA shunt, in which glutamate decarboxylase (GAD) catalyzes the decarboxylation of glutamate (Glu) to produce GABA [14]. GABA is then transaminated by GABA transaminase (GABA-T) to succinic semialdehyde (SSA), which is oxidized by SSA dehydrogenase (SSADH) to succinate [15], feeding into the tricarboxylic acid (TCA) cycle.

GAD, the rate-limiting enzyme in the GABA biosynthesis pathway, was first characterized in Petunia hybrida in 1993 [16]. GAD plays critical roles in plant development and stress responses. Studies have shown that its expression and enzymatic activity are induced by various abiotic stresses, contributing to stress responses and adaptation [17]. For example, hypoxic treatment significantly upregulates OsGAD genes in germinating rice (Oryza sativa) seeds [18]; GAD6 is involved in Cd²⁺ stress responses in cotton (Gossypium hirsutum) [19]; the gad1 mutant in Arabidopsis thaliana exhibits defective root hair development [20]; and PgGAD expression is elevated under multiple abiotic stresses in ginseng (Panax) [21]. Overexpression of GAD1 in tomato (Lycopersicon esculentum) increases GABA and glutamate accumulation, alleviating salt stress-induced damage [22]. As the rate-limiting enzyme in GABA biosynthesis, GAD is central to controlling GABA levels in plants. Its activity is often upregulated under environmental stresses such as salinity, resulting in enhanced GABA accumulation. Therefore, developing sweetpotato cultivars with high GABA content and strong stress tolerance holds significant promise for both food and agricultural innovation. However, despite its importance, the functional roles of sweetpotato GAD genes, particularly IbGAD4, in GABA biosynthesis and salt stress response remain largely uncharacterized.

The transcriptional regulation of plant GAD is a complex process influenced by developmental cues, tissue specificity, and various environmental stresses, especially those related to abiotic stresses such as salt, drought, and hypoxia. The GAD gene promoters typically harbor cis-regulatory elements associated with stress responses and hormonal regulation, such as ABRE (ABA-responsive element) that is activated during drought and salt stress and other elements that are recognized by bHLH, MYB, and WRKY transcription factors [23]. Some MYB transcription factors (e.g., MYB2, MYB15, MYB96 in Arabidopsis) are known to function under abiotic stress [24]. In Arabidopsis, overexpression of AtMYB44 results in enhanced salt and drought resistance [25]. In lettuce, LsMYB44 and LsWRKY12 downregulate the transcription of key genes of GABA shunt and polyamine metabolism [26]. However, it remains unclear which specific MYB subfamilies target the sweetpotato GAD genes.

Building upon prior analysis of the ItfGAD gene family in the wild sweetpotato relative Ipomoea triloba, we identified 15 ItfGAD members in Ipomoea triloba. Among the members, the expression of ItfGAD4 was induced significantly under NaCl treatment. In addition, ItfGAD4 exhibited induction or inhibition under various hormone treatments and abiotic stress treatments. We assumed that ItfGAD4 homologs may play a vital function in NaCl stress in sweetpotato. Thus, this study successfully cloned and functionally characterized the IbGAD4 gene. The IbGAD4 expression was induced by salt stress. Transgenic sweetpotato plants overexpressing IbGAD4 were generated and assessed for GABA accumulation and salt stress response. IbGAD4 overexpression induced GABA pathway gene expression and GABA accumulation, upregulated the Salt Overly Sensitive (SOS) pathway and enhanced salt tolerance. Additionally, yeast one-hybrid screening was performed to identify transcription factors interacting with the IbGAD4 promoter, aiming to elucidate its upstream regulatory network. This screening identified IbMYB44 as a candidate regulator. IbMYB44 represses the transcriptional activity of IbGAD4 promoter and function as a negative regulator. Notably, IbMYB44 expression was downregulated under salt stress. Taken together, this study provides the first functional characterization of IbGAD4 in sweetpotato and establishes its role in regulating both GABA accumulation and salt stress tolerance. Additionally, we identified a MYB regulator IbMYB44 as a repressor of IbGAD4, revealing a previously unrecognized link between MYB-mediated transcriptional control and the GABA shunt in sweetpotato.

Materials and methods

Experimental materials and planting conditions

The sweetpotato variety used in this study, Lizixiang (LZX), was maintained in a tissue culture room under a 16 h light/8 h dark photoperiod. Field-grown materials were cultivated at the experimental base of the College of Agronomy, Shanxi Agricultural University, located at 37°25′ N latitude and 112°33′ E longitude. The average annual temperature is 12 °C, and approximate annual rainfall in 17.9 inches.

DNA and RNA extraction and cDNA synthesis

Genomic DNA was extracted using the CTAB method, and RNA was isolated using the Huayueyang Nucleic Acid Extraction Kit (Beijing Huayueyang, 0416 − 100 GK). First-strand cDNA was synthesized using the MonScript™ RTIII All-in-One Mix with dsDNase kit (Suzhou Monan, MR05101M). Reverse transcription quantitative PCR (RT-qPCR) was performed on a BIO-RAD CFX96 Real-Time qPCR system using the MonAmp™ ChemoHS Specificity Plus qPCR Mix (None ROX) (Suzhou Monan, MQ00501S). IbActin (AY905538) served as the internal reference gene. Each sample included three biological replicates. Relative gene expression levels were calculated using the 2^−ΔΔCt method. Statistical analyses, including t-tests, one-way ANOVA, and two-way ANOVA, were conducted using GraphPad Prism (version 10.4.1). Results are presented as mean ± SD (n = 3).

Expression analysis of IbGAD4 and IbMYB44

Based on the IbGAD4 and IbMYB44 gene sequence (Ibat.Brg.07A_G004580.1 and Ibat.Brg.03B_G012990.1), RT-qPCR primers were designed using NCBI Primer-BLAST (https://www.ncbi.nlm.nih.gov/tools/primer-blast/) (Table S1). Total RNA was extracted from sweetpotato storage roots at different developmental stages (60 d, 75 d, 90 d, and 105 d), and from roots, stems, and leaves of 90-day-old plants (var. LZX). Following reverse transcription to cDNA, IbGAD4 expression was analyzed across developmental stages and tissues, with three biological replicates for each sample. For salt stress, 30-day-old sweetpotato plants (var. LZX) were grown in 1/2 strength Hoagland solution containing 200 mM NaCl and total RNA was extracted from leaves collected at different treatment stages (0 h, 1 h, 3 h, 6 h, 12 h and 24 h). IbGAD4 and IbMYB44 expression was analyzed across treatment stages, with three biological replicates for each sample.

Sequence analysis and overexpression vector construction of IbGAD4 and IbMYB44

Cloning primers for IbGAD4 were designed based its gene sequence (Table S1). PCR amplification was performed using cDNA and DNA from 90-day-old LZX storage root as templates. The amplified cDNA (ORF) and genomic DNA were ligated into the pMD19-T vector to generate plasmids pMD-IbGAD4 and pMD-IbGAD4-DNA, respectively. Gene structure was analyzed using the GSDS online tool (http://gsds.gao-lab.org/). To construct the overexpression vector, the stop codon of the IbGAD4 ORF was removed, and primers containing Kpn I and Xba I restriction sites were designed (Table S1). Using pMD-IbGAD4 as the template, the target fragment was amplified, purified, and digested with Kpn I and Xba I, along with the pCAMBIA1300-GFP vector. The digested fragments were ligated to pCAMBIA1300-GFP generate the recombinant overexpression plasmid.

Cloning primers for IbMYB44 were designed based its gene sequence (Table S1). The amplified CDS and genomic DNA sequence of IbMYB44 was ligated into the pGEM-T vector to obtain pGEMT-IbMYB44 and pGEMT-IbMYB44-DNA. Gene structure was analyzed by GSDS, and sequence alignment was analyzed by DNAMAN, phylogenetic tree carried out by MEGA7.0.

Generation of IbGAD4 transgenic sweetpotato

Agrobacterium harboring the recombinant plasmid pCAM1300-IbGAD4-GFP was used to infect the LZX embryogenic suspension cells. Following 27–35 weeks of culture and regeneration, genomic DNA was extracted from regenerated plants. PCR identification of positive transformants was performed using 35 S-PF as the forward primer and IbGAD- Xba I-R as the reverse primer. Total RNA was extracted from PCR-positive lines to detect IbGAD4 expression levels, with three biological replicates per sample.

Pathway gene expression and GABA content analysis in transgenic sweetpotato

Total RNA was extracted from the leaves of PCR-confirmed positive transgenic plants and reverse-transcribed to cDNA. Expression levels of GABA biosynthetic pathway genes were analyzed using RT-qPCR. IbActin (AY905538) served as the internal reference gene. Primers used in this study were listed in Table S1. Three biological replicates were included for each sample. GABA content in both LZX and transgenic lines (leaf and storage root) were quantified using an ELISA kit (Wuhan Elabscience, E-BC-K852-M), with each sample (50 mg/sample) measured in triplicate.

Salt stress treatment of IbGAD4 transgenic sweetpotato

One-month-old wild-type (LZX) and overexpression (OE) sweetpotato plantlets were cut into ~ 1 cm stem segments and cultured on MS solid media supplemented with either 0 or 60 mM NaCl. Phenotypic changes were documented periodically through photography. Samples were collected at 0 and 30 days, then immediately frozen in liquid nitrogen for subsequent analysis. Expression levels of SOS pathway genes such as Salt Overly Sensitive 1 (IbSOS1), IbSOS2, and IbSOS3, Reactive Oxygen Species (ROS) scavenging genes, Superoxide dismutase (IbSOD), Peroxidase (IbPOD), Ascorbate peroxidase (IbAPX), the abiotic stress-responsive gene Cold-inducible protein 2 (IbKIN2), and the abscisic acid (ABA) biosynthesis-related gene 9-cis-epoxycarotenoid dioxygenase 3 (IbNCED3) were analyzed. In addition, activities of antioxidant enzymes such as SOD, POD, catalase (CAT) and malondialdehyde (MDA) content were measured. All experiments were repeated in triplicate, and results are presented as mean values.

For further analysis under hydroponic conditions, 90-day-old field-grown LZX and IbGAD4-OE lines exhibiting consistent growth were selected. Apical cuttings (~ 10 cm) were placed in 1/2 strength Hoagland solution containing either 0 or 85 mM NaCl [27]. Phenotypic changes were recorded photographically over time, and samples were collected at 0, 1, 3, 6, 12, 24, and 36 h, frozen in liquid nitrogen, and used for gene expression analysis. Each time point was analyzed in triplicate.

Promoter cloning and cis-element analysis of IbGAD4

The promoter sequence of IbGAD4 (Pro-IbGAD4) was retrieved based on its gene ID from the sweetpotato reference genome database (https://sweetpotato.uga.edu/) and PCR-amplified using LZX genomic DNA as the template. The amplified promoter fragment was submitted to the PlantCARE database (http://bioinformatics.psb.ugent.be/webtools/plantcare/html/) for the identification of cis-acting elements.

Screening and expression analysis of pro-IbGAD4 interacting genes

Based on promoter analysis, the Pro-IbGAD4 region was divided into two fragments, Pro-IbGAD4-1 (-1739 to -1478 bp) and Pro-IbGAD4-2 (-238 bp to -1 bp) for yeast one-hybrid (Y1H) screening. Primers containing Hind III and Sal I restriction sites were designed to construct bait vectors. The resulting bait plasmids pAbAi-IbGAD4 were digested with BstBI, purified, transformed into Y1HGold yeast (Coolaber, CC308). To determine the optimal concentration of Aureobasidin A (AbA) for suppressing autoactivation, positive clones were diluted in 154 mM NaCl to OD₆₀₀ of 0.002 and plated on SD/-Ura medium supplemented with different concentrations of AbA.

A secondary cDNA library derived from sweetpotato storage roots was transformed into Y1HGold strains harboring either pAbAi-Pro-IbGAD4-1 or pAbAi-Pro-IbGAD4-2. Initial screening was conducted on SD/-Leu medium containing AbA. We were only able to get positive colonies using the pAbAi-Pro-IbGAD4-1 as bait. Colonies that grew were re-streaked for secondary screening. Yeast cells from positive clones were lysed, and PCR was performed using primers T7-F and 3’AD-R to confirm insert presence. Plasmids from confirmed positive clones were extracted, transformed into E. coli DH5α, and sequenced. The resulting sequences were annotated using NCBI BLAST analysis to identify candidate interacting genes.

To validate interactions, candidate genes were co-transformed with pAbAi-Pro-IbGAD4-1 into Y1HGold strains and plated on SD/-Leu medium containing AbA. Positive clones were cultured in SD/-Leu liquid medium, serially diluted (1:10, 1:100, 1:1000) in 154 mM NaCl, and spotted onto SD/-Leu and SD/-Leu/AbA plates. Colony growth was monitored over 3–5 d at 30 °C to assess interaction strength, and expression levels were evaluated in both LZX and IbGAD4 transgenic lines.

Promoter assay

For promoter assay, the reporter plasmid was generated by cloning IbGAD4 promoter in a modified pUC vector containing a firefly luciferase (LUC) and rbcs terminator. The effector plasmid was constructed by cloning IbMYB44 into a modified pBS vector under the control of the mirabilis mosaic virus (MMV) promoter and rbcs terminator. β-glucuronidase (GUS) reporter, driven by CaMV 35 S promoter and rbcs terminator, was used as an internal control in the protoplast assay. Protoplast isolation from Nicotiana tabacum cell suspension cultures and electroporation were performed as previously described [28]. Luciferase and GUS activities in transfected protoplasts were measured to assess transcriptional activity.

Results

Characteristics and expression of IbGAD4 in sweetpotato

The ORF of IbGAD4 was 1158 bp in length and encoded a protein of 385 amino acids (Fig. S1A) The corresponding genomic DNA sequence was 1719 bp (Fig. S1B). Gene structure analysis revealed that IbGAD4 consists of four exons and three introns (Fig. S1D). A total of five IbGAD4 homologs were identified in the sweetpotato genome, distributed across chromosomes Chr07A, 07C, 07D, 07E, and 07F (Fig. S1C). Expression analysis showed that IbGAD4 expression was highest in the roots, approximately 82-fold and 27-fold higher compared to that of stems and leaves, respectively (Fig. 1A). To analyze the temporal expression patterns of IbGAD4, storage roots at different developmental stages were analyzed by RT-qPCR. The results indicated that IbGAD4 expression peaked at 90 days and was nearly undetectable by 105 days (Fig. 1B). To identify the response of IbGAD4 under salt stress, different treatment stage leaves were analyzed by RT-qPCR. The results showed that IbGAD4 expression was induced by 10-fold within 1 h of NaCl (200 mM) treatment and then declined (Fig. 1C).

Fig. 1.

Expression pattern of IbGAD4. A: Expression pattern of IbGAD4 in different tissues (root, stem and leaf); B: Expression pattern of IbGAD4 at different developmental stages; C: Expression pattern of IbGAD4 in sweet potato leaves under NaCl (200 mM) stress at different treatment stages (0 h, 1 h, 3 h, 6 h, 12 h and 24 h). Gene expression was measured using reverse transcription quantitative PCR (RT-qPCR). ns indicates no significant difference. Results from three independent experiments are presented as mean ± SD. One-way ANOVA was used for calculating statistical significance *, P < 0.05; **, P < 0.01; ***, P < 0.001

Overexpression of IbGAD4 enhances GABA content and expression of GABA biosynthesis-related genes in sweetpotato

To investigate the role of IbGAD4 in GABA biosynthesis, we generated 11 IbGAD4 overexpression (OE) lines in LZX (Fig. S2A-C). Independent IbGAD4 OE lines were confirmed by PCR (Fig. S2D) and RT-qPCR (Fig. S2E). Three representative OE lines (OE-3, OE-5, and OE-6) were selected for further analysis of GABA content and expression of GABA biosynthetic pathway genes (Fig. 2A). In storage roots, GABA content was significantly higher in OE lines compared to wild-type (LZX), with OE-3 showing the greatest increase (Fig. 2B; Fig. S3). Similarly, GABA levels in leaves of OE lines remained elevated relative to LZX (Fig. 2C). Expression analysis revealed that in OE lines, genes encoding enzymes involved in the GABA shunt (GABA-T, SSADH) and the TCA cycle were upregulated by 1.5 to 2.0-fold compared to LZX (Fig. 2D). These results suggested that IbGAD4 overexpression promotes GABA accumulation in sweetpotato by coordinately upregulating multiple key genes in both the GABA shunt and TCA cycle, thus contributing to a metabolically integrated regulatory network.

Fig. 2.

Detection of GABA content and gene expression levels in IbGAD4 overexpressing lines. A: Detection of the expression level of IbGAD4 in overexpression lines; OE-3, OE-5 and OE-6 represents three independent overexpression (OE) lines; B: Detection of GABA content in transgenic sweet potato storage roots; GABA level was measured in µmol/g wet weight; C: Detection of GABA content in transgenic sweet potato leaves; GABA level was measured at µmol/g wet weight; D: Detection of the expression levels of genes related to the GABA biosynthetic pathway in IbGAD4 OE lines. Gene expression was measured using RT-qPCR; ns indicates no significant difference; Results from three independent experiments are presented as mean ± SD. One-way ANOVA was used for calculating statistical significance: *, P < 0.05; **, P < 0.01; ***, P < 0.001

Overexpression of IbGAD4 improves salt tolerance in sweetpotato

To evaluate the role of IbGAD4 in salt stress tolerance, we subjected the three OE lines to salt stress treatment. One-month-old seedlings were cut into ~ 1 cm stem segments and transferred to MS medium with 0 and 60 mM NaCl. After 30 days of growth, both wild-type LZX and OE lines exhibited healthy growth under control conditions (0 mM NaCl), with no significant phenotypic differences observed (Fig. 3A). However, under salt stress conditions, LZX seedlings displayed clear signs of growth inhibition, including reduced growth rate, smaller leaves, shorter shoots, and underdeveloped roots. In contrast, OE lines exhibited enhanced salt tolerance, with larger leaves, taller shoots, and more robust root development, indicating improved salt resistance (Fig. 3A).

Fig. 3.

Phenotype, antioxidant enzyme activity and expression analysis of salt stress related genes in IbGAD4 overexpressing (OE) sweet potato plants. A: The phenotype of transgenic sweet potato under salt stress; Phenotypic analysis of LZX and the transgenic sweet potato lines grown in MS medium with 0 mM and 60 mM NaCl for 30 days, respectively; The activity of (B) SOD, (C) POD, (D) CAT and (E) MDA content of each line after 30 days of different treatments; F: Expression levels of salt stress-related genes in IbGAD4 overexpressing (OE) sweet potato lines in normal (untreated) and after NaCl treatment for 30 days; Gene expression was measured using RT-qPCR ns indicates no significant difference; Results from three independent experiments are presented as mean ± SD. Two-way ANOVA was used for calculating statistical significance: *, P < 0.05; **, P < 0.01; ***, P < 0.001

To assess biochemical responses, antioxidant enzyme activities for SOD, POD, and CAT, as well as MDA content were measured in LZX and OE lines after 30 days of salt treatment. OE lines exhibited higher antioxidant enzyme activities than LZX under both normal and salt stress conditions (Fig. 3B-D). Conversely, MDA levels, indicative of lipid peroxidation, were significantly lower in OE lines compared to LZX, with a more pronounced difference under salt stress (Fig. 3E). To explore the molecular mechanism of salt stress response in IbGAD4 OE lines, we analyzed the expression of eight stress-related genes, IbSOS1, IbSOS2, IbSOS3, IbSOD, IbPOD, IbAPX, IbKIN2 and IbNCED3 in OE-3, -5, and -6 lines. After 30 days salt treatment in MS medium, expression levels of all eight genes were significantly higher (1.5-8.0 fold) in OE lines than in LZX under both control and salt stress conditions, with differences more pronounced under stress (Fig. 3F).

To further confirm the salt stress-resistance phenotype of the OE line in another cultural condition, one-month-old seedlings of LZX and the representative OE-3 line were transferred to 1/2 Hoagland nutrient solution containing 85 mM NaCl. Over time, the growth of LZX declined significantly, showing symptoms such as yellowing and overall wilting after 15 days of treatment. In comparison, OE-3 plants maintained a healthier appearance with green leaves and upright shoots, indicating a higher level of salt tolerance (Fig. 4A). Although gene expression patterns varied over time, all genes were consistently more highly expressed in OE-3 than in LZX (Fig. 4B-I), correlating with observed enhancement in salt tolerance. These results demonstrate that overexpression of IbGAD4 enhanced salt tolerance in sweetpotato by upregulating salt stress-responsive genes, boosting antioxidant defense mechanisms, and reducing oxidative damage.

Fig. 4.

IbGAD4 overexpression improves the salt tolerance of sweet potato. (A) Phenotype of the control (LZX) and IbGAD4-overexpressing (OE-3) sweetpotato under salt stress; apical cuttings of LZX and the IbGAD4 overexpression line (OE-3) showing consistent growth in the field were collected and subjected to salt stress treatment in 1/2 strength Hoagland nutrient solution supplemented with NaCl (85 mM) for 3 d and 15 d. Expression of SOS pathway genes IbSOS1 (B), IbSOS2 (C), and IbSOS3 (D); ROS scavenging-related genes IbSOD (E), IbPOD (F), and IbAPX (G); stress-related gene IbKIN2 (H); and the ABA biosynthesis-related gene IbNCED3 (I) in control (LXZ) and IbGAD4-overexpressing (OE-3) lines under short-term NaCl stress. Gene expression was measured using RT-qPCR. ns indicates no significant difference. Results from three independent experiments are presented as mean ± SD. Two-way ANOVA was used for calculating statistical significance**:** *, P < 0.05; *, P < 0.01; , P < 0.001

Identification of IbMYB44 as a transcription factor binding to the IbGAD4 promoter

To identify transcription factors that regulate IbGAD4, a sweetpotato storage root cDNA Y1H library was screened using IbGAD4 as bait. We amplified an 1850 bp promoter fragment of IbGAD4 (Fig. S4A), and based on cis-element analysis (Table S2), constructed a bait vector, pAbAi-IbGAD4-1 (Fig. S4B-C). The vector exhibited self-activation, which was suppressed by adding 100 ng/mL Aureobasidin A (AbA) (Fig. S4D). Using this system, we obtained 13 candidate genes from the cDNA library (Table S3). Among them, many are not DNA-binding proteins. We thought that they were identified due to either false positives or interacting with other DNA-binding proteins. Therefore, we focused only on DNA-binding proteins. From these, one candidate protein (IbMYB44) from pAbAi-IbGAD4-1 was selected for verification through yeast point-to-point interaction assays (Fig. 5B). The candidate, IbMYB44 (gene ID: Ibat.Brg.03B_G012990.1), which encodes an MYB transcription factor, was confirmed to bind specifically to the IbGAD4 promoter. The promoter assay demonstrated that IbMYB44 could repress IbGAD4 transcription (Fig. 5C-D). Subsequent expression analysis showed that IbMYB44 transcript levels were significantly higher in IbGAD4 OE lines compared to LZX (Fig. 5E), suggesting a potential feedback activation of IbMYB44 by IbGAD4. Besides, IbMYB44 expression was downregulated under salt treatment (Fig. 5F).

Fig. 5.

Interaction of IbMYB44 with IbGAD4 promoter and IbMYB44 expression level under NaCl stress. A: Schematic diagram showing the IbGAD4 promoter; The full length fragment of the promoter of IbGAD4 were shown; the bait sequence IbGAD4-1 (-1739 bp to -1478 bp) contains a MYB binding site; B: Verification of pAbAi⁃IbGAD4-1(-1739 bp to -1478 bp) and IbMYB44 interaction using Y1H assay; C: Schematic diagrams of the constructs used in protoplast assay; D: Relative LUC activity in tobacco protoplast; Results from three independent experiments are presented as mean ± SD. Student’s t-test was used for calculating statistical significance *, P < 0.05; E: Expression level of IbMYB44 in IbGAD4 overexpressing (OE-3, OE-5, OE-6) lines; F: Expression level of IbMYB44 under 200 mM NaCl treatment; Gene expression was measured using RT-qPCR; ns indicates no significant difference; Results from three independent experiments are presented as mean ± SD. One-way ANOVA was used for calculating statistical significance *, P < 0.05; **, P < 0.01; ***, P < 0.001

The ORF of IbMYB44 gene was 1173 bp in length and it encoded a protein of 391 amino acids (Fig. S5A). The amino acid sequence alignment showed that IbMYB44 contains two SANT domains and two specific motifs of subgroup 22, the 22.1 motif and 22.2 motif, which indicated that IbMYB44 is an R2R3-MYB transcription factor belonging to the MYB subfamily 22, and the sequence also contained a putative EAR transcriptional repressor motif (LSLSL) (Fig. S5B). A total of 17 subgroup 22 R2R3-MYBs were identified in sweetpotato (Fig. S5C).

Discussion

This study was designed to test the hypothesis that IbGAD4, a glutamate decarboxylase gene in sweetpotato, functions not only as a rate-limiting enzyme in GABA biosynthesis but also plays a central role in modulating salt stress tolerance through metabolic and transcriptional reprogramming. While GABA is widely recognized as a multifunctional metabolite involved in stress responses, its biosynthetic regulation and physiological role in sweetpotato, especially under abiotic stress conditions, have not been previously characterized. Our findings provide new evidence that positions IbGAD4 at the intersection of primary metabolism and environmental stress adaptation, and highlight its potential utility in crop improvement.

We began by identifying and cloning IbGAD4 from the LZX variety of sweetpotato. Structural analysis revealed that IbGAD4 encodes a 385-amino acid protein with four exons and three introns, and that it is one of five homologs located on chromosome 7 subgenomes (Chr07A, 07 C, 07D, 07E, and 07 F), reflecting the hexaploid nature of the sweetpotato genome (Fig. S1). Gene expression profiling indicated that IbGAD4 is induced by salt treatment and primarily expressed in roots and is developmentally regulated, peaking at 90 days and nearly absent by 105 days in storage roots (Fig. 1). These observations suggested its likely involvement in root metabolism and adaptive developmental programming.

It was reported that GAD played a crucial role in GABA biosynthesis in higher plants, with the activation of GAD considered a key regulatory step in the GABA biosynthetic pathway [29]. To assess its function, IbGAD4 was overexpressed in transgenic sweetpotato plants. Three independent OE lines (OE-3, OE-5, OE-6) exhibited significantly elevated levels of IbGAD4 transcript and accumulated higher levels of GABA in both leaves and storage roots (Fig. 2A-C). Concurrently, key genes in the GABA shunt such as GABA-T and SSADH, as well as TCA cycle-associated enzymes, were upregulated in OE lines, suggesting that GABA metabolism was broadly enhanced (Fig. 2D). Together, these results confirm the role of IbGAD4 in promoting GABA accumulation and suggest that it acts in coordination with downstream metabolic pathways to facilitate GABA flux and integration into core carbon metabolism.

GABA accumulation was involved in plant tolerance to various abiotic stresses [30–32] and played an essential role in ROS scavenging. Downregulation of GhGAD6 in cotton reduced endogenous GABA levels, intensified ROS accumulation, and enhanced Cd²⁺ toxicity [19]. Overexpression of SlGAD1 increased endogenous GABA content and ROS scavenging capacity, reducing membrane lipid peroxidation in tomato [33]. A major focus of the study was to determine whether IbGAD4-mediated GABA production could influence plant responses to salt stress. Under 60 mM NaCl in MS medium, all OE lines showed better growth performance than LZX plants, with more vigorous shoots, healthier leaves, and longer roots. These phenotypes were corroborated in hydroponic systems (1/2 Hoagland solution + 85 mM NaCl), where OE-3 lines maintained upright, green foliage while LZX plants exhibited leaf yellowing, wilting, and growth retardation (Figs. 3A and 4A). This clear divergence in salt tolerance was associated with the transcriptional upregulation of multiple stress-responsive genes. Specifically, the SOS pathway genes, including IbSOS1, IbSOS2, and IbSOS3, were highly expressed in OE lines, alongside ROS-scavenging genes (IbSOD, IbPOD, IbAPX), IbKIN2, and the ABA biosynthetic gene IbNCED3 (Figs. 3F and 4B-I). Notably, these stress-related genes were already expressed at higher levels in OE lines under non-stress conditions, suggesting a primed state of enhanced stress readiness.

In parallel, biochemical analyses revealed higher enzymatic activities of SOD, POD, and CAT in OE lines compared to WT, consistent with enhanced antioxidant capacity (Fig. 3B-D). Moreover, MDA levels, a marker for lipid peroxidation and oxidative stress, were significantly reduced in OE lines, especially under salt treatment (Fig. 3E). These results collectively indicate that IbGAD4-mediated GABA production enhances salt tolerance by reinforcing both transcriptional and physiological defense systems, including ionic homeostasis, hormone signaling, and ROS detoxification.

A particularly novel finding of this study was the identification of IbMYB44 as a transcriptional regulator of IbGAD4. Using Y1H screening with a segmented 252 bp IbGAD4 promoter, IbMYB44 was among several candidates shown to bind the promoter directly (Fig. 5B). We identified and characterized the MYB transcript factor, it encoded 391 amino acid protein, with no intron, and belonged to the R2R3 MYB subgroup 22 and contained an EAR motif typical of transcriptional repressors. In sweetpotato, IbMYB44 has been implicated in anthocyanin biosynthesis [34]. Although previously associated with regulating secondary metabolism (anthocyanin biosynthesis) in sweetpotato by competitively inhibiting the MYB340-bHLH2-NAC56 complex, this study is the first to demonstrate its role in primary metabolism, specifically in repressing IbGAD4. The transcriptional repression function of IbMYB44 was supported by promoter assay and expression analyses, which showed lower LUC activity in pro-IbGAD4 with IbMYB44 (Fig. 5D) and significantly higher IbMYB44 levels in OE lines under normal conditions (Fig. 5E), suggesting potential negative feedback regulation. Importantly, IbMYB44 expression was downregulated in response to salt stress (Fig. 5F), which may serve as a derepression mechanism to allow upregulation of IbGAD4 under adverse conditions (Fig. 6).

Fig. 6.

Schematic diagram of IbGAD4 regulating salt stress. Salt stress repressed the expression of IbMYB44 and induced the expression of IbGAD4. IbMYB44 can bind to the promoter of IbGAD4 and restrain it transcriptional activity. Overexpression of IbGAD4 increased salt resistance in transgenic sweet potato by improved GABA level, salt response genes expression and promoted ROS scavenging system activity

The role of GABA in sweetpotato under salt stress has not been reported. This study is the first to functionally characterize IbGAD4 in sweetpotato and establish its role in both GABA accumulation and salt stress tolerance. It reveals that IbGAD4 overexpression not only enhances GABA production but also triggers coordinated upregulation of genes involved in the GABA shunt, TCA cycle, and multiple stress-responsive pathways including the SOS pathway, ROS-scavenging pathway, and the ABA biosynthetic pathway. Consistently, IbGAD4 sweetpotato OE lines showed enhanced antioxidant capacity and reduced MDA levels, a marker of lipid peroxidation and oxidative stress, especially under salt conditions. Collectively, these findings suggest that IbGAD4-mediated GABA accumulation activates multiple stress-responsive pathways that jointly contribute to improved salt tolerance in sweetpotato. Furthermore, we identified IbMYB44 as a potential repressor of IbGAD4, thereby linking MYB-mediated transcriptional regulation to the GABA shunt in sweetpotato. Additional experimental evidence will be required to further substantiate the role of MYB44 in regulating GABA biosynthesis and stress tolerance.

Our work opens new avenues for improving the stress resilience of sweetpotato. By genetically enhancing GABA accumulation through targeted manipulation of IbGAD4, it may be possible to develop sweetpotato cultivars with improved abiotic stress tolerance. Given the increasing challenges posed by climate-induced salinity stress, the IbGAD4 offers promising targets for future crop breeding and gene editing.

Conclusion

Salt stress repressed the expression of IbMYB44 and induced the expression of IbGAD4. IbMYB44 can bind to the promoter of IbGAD4 and restrain its transcriptional activity. Overexpression of IbGAD4 increased salt resistance in transgenic sweet potato by improving GABA levels, enhancing the expression of salt-responsive genes and promoting ROS scavenging system activity.

Authors’ contributions

XJ and LY: conceptualization; CZ, YZ, SS, XH, NB, RT, RY and XW: performed experiments and data analysis; CZ and YZ wrote the first draft and prepared the figures; XJ, LH, SP, and LY: reviewed and edited the manuscript and figures. All authors have read and approved the final version of the manuscript.

Funding

This work was supported by the National Key Research and Development Project of China (2023YFD1200700/2023YFD1200701), Special Project for Scientific and Technological Cooperation and Exchange in Shanxi Province (202304041101009), the earmarked fund for Modern Agro-industry Technology Research System (2025CYJSTX06-04), the Central Government Guides Local Science and Technology Development Fund (YDZJSX2024B007), the Innovative Talent International Cooperation Program in Crop Science of the China Scholarship Council (CXXM20240052), and the Shanxi Engineering Research Center for Genetics and Metabolism of Special Crops.

Data availability

All data are provided in the article and supplementary information.

Declarations

Ethics approval and consent to participate

This article does not contain any studies on human or animal subjects.

Consent for publication

Not applicable.

Competing interests

The authors declare that there are no competing interests.