Comparative transcriptome analysis revealed that tomato response to salt stress induced by exogenous GABA and functional verification of SlGAD1 Authors: Xiaolei Wu, Ding Yuan, Lida Chen, Yukun Sun, Yening Wang, Shiqing Jiang, Xule Sun, Xinyue Zhang, Jingrui Li, Hongbo Gao Published: 2025 ABSTRACT AbstractBackgroundGamma-aminobutyric acid (GABA) can act as a signalling molecule and metabolite to increase plant salt tolerance. However, it is currently unclear which genes play a major role in the defence system under the sustained effects of salt stress and how different metabolic pathways work synergistically to regulate them. The transcriptomic data and metabolite analysis of 33 leaf samples from tomato (Solanum lycopersicum L.) seedlings under 175 mmol·L− 1 salt stress revealed that the GABA + Na (GN) group presented strong salt tolerance.ResultsAmino acid synthesis, especially that of GABA, and some transporter-related genes played important roles in the GN group. Phenylalanine accumulation subsequently induced flavonoid biosynthesis via upregulating key genes involved in the flavonoid biosynthetic pathway, including the gene encoding enzyme protein PAL, 4CL, C4H, CHS and CHI. The molecular mechanism by which exogenous GABA enhances salt tolerance involves the induction of gene expression in pathways such as amino acid synthesis and transport and phenylpropanoid and flavonoid biosynthesis, increasing the content of intracellular small molecules with osmotic regulation function, thereby regulating ion transporter genes to accelerate Na+ transfer and maintain relatively stable osmotic pressure. SlGAD1 overexpression alleviated salt stress by increasing the GABA content and the flavonoid biosynthesis pathway.ConclusionsThese findings provide direct evidence that exogenous GABA affects the phenylpropanoid and flavonoid pathways and ion transporters, which will be helpful for the development of GABA as an agent for inducing salt tolerance.Supplementary InformationThe online version contains supplementary material available at 10.1186/s12870-025-07986-6.Keywords: Solanum lycopersicum L., NaCl stress, GABA, Transcriptome, SlGAD1 Introduction Soil salinization is a soil environmental problem that has attracted increasing attention worldwide. According to incomplete Food and Agriculture Organization (FAO) statistics, the global coverage of saline‒alkali land area is approximately 9.5%, amounting to more than 500 million hm2. Approximately 10% of the total area in China is saline‒alkali land, which poses a serious threat to the normal growth of plants and sustainable economic development [1, 2]. Owing to the increasing secondary salinization of greenhouse soils caused by the closed environment and the excessive use of chemical fertilizers, investigations have shown that 42.9% of protected soil is salinized [3, 4]. In some areas, the soil electrical conductivity (EC) even exceeds the critical value for vegetable fertility barriers by more than twofold, resulting in ion toxicity, osmotic stress, and other harmful effects at levels that far exceed those observed under moderate salt damage, severely affecting the yield and quality of cultivated vegetables. Among these toxic ions, Na+ and Cl− are the main ions that cause salt stress injury to plants, and the Na+ and Cl− contents increase by 140% and 58%, respectively, in salinized soil [5, 6]. The exploration of effective methods and mechanisms for improving crop salt tolerance has been a research hotspot in the field of protected cultivation and is also a key problem for the utilization of saline soil and the development of protected cultivation. Gamma-aminobutyric acid (GABA) is an important free amino acid in plant cells and is involved in plant salt tolerance as a signalling molecule and metabolite [7]. GABA, an exogenous substance, can be directly absorbed by plant roots and can effectively improve salt tolerance [8–10]. The increase in the endogenous GABA content under salt stress was much greater than the increase in the amount synthesized by the plant itself [8, 9, 11]. Therefore, GABA has been identified as an important substance that enhances plant salt tolerance. In recent years, many breakthroughs have been made in understanding the mechanisms by which GABA enhances plant salt tolerance. Although the mechanism by which GABA enhances plant salt tolerance has not yet been established, it is generally believed that GABA is involved in ion transport, active oxygen regulation and substance metabolism. GABA enhances plant salt tolerance mainly by altering the ion plasma membrane potential difference and specific anion transporters, which in turn affect ion transport [7]. It has also been shown that the enhancement of salt tolerance by GABA is associated with the active oxidase/nonenzymatic metabolic clearance pathway [12]. Salt tolerance in muskmelon is strongly associated with decreased reactive oxygen species metabolism and avoidance of chlorophyll damage [13]. GABA can induce polyamine accumulation in muskmelon plants under salt stress [14], providing nutrients such as sugars and amino acids for clover seed germination by promoting starch decomposition [15]. Our previous results revealed that the increase in salt tolerance caused by GABA was closely related to a decrease in the Na+ flow rate in the roots and leaves, an increase in amino acid content, an increase in antioxidant metabolism and an increase in photosynthesis [10, 11, 16]. Tomato is the most widely cultivated vegetable in greenhouses and is moderately salt sensitive. A high soil salt concentration leads to stunted plant growth, decreased yield, and poor quality [3]. Although several traditional stress-related pathways are involved in the salt tolerance induced by GABA, it is not clear how the different metabolic pathways work together in the GABA response to improve salt tolerance. Therefore, identifying the metabolic processes and genes involved in the response of exogenous GABA to salt stress is highly important. Transcriptional-level regulation is an important regulatory process involved in the regulation of plant life activity, and transcriptome sequencing is a powerful tool for understanding the regulation of plant gene expression. Currently, high-throughput transcriptome sequencing techniques are widely applied in the study of the mechanisms by which plants respond to biotic and abiotic stresses [17, 18]. Most of these key salt tolerance-related genes are involved in biological processes such as the response to oxidative stress, osmotic stress and abiotic stress [19, 20]. Our previous research revealed that adding exogenous GABA to nutrient solutions can directly and quickly affect the salt stress response and induce salt tolerance in tomato plants by increasing the concentration of endogenous GABA [10]. SlGAD expression plays an important role in the resistance of tomato plants to NaCl stress in response to GABA treatment [10]. However, the mechanism by which exogenous GABA induces salt tolerance at the molecular level is not yet clear. Previous studies have shown that in addition to common pathways such as reactive oxygen species metabolism and photosynthesis, many metabolic pathways can effectively alleviate salt stress in plants. In high-salt environments, amino acids and sugars in plant cells can serve as osmoregulatory substances, helping to maintain water balance and reduce osmotic stress caused by salt. Therefore, amino sugars such as glucose and galactose also play important roles in the plant salt stress response [21]. The amino acid pathway, especially proline, as an osmotic regulator, can stabilize the protein structure in plants under high-salt environments and protect cells from damage [22]. Some metabolites in the phenylpropanoid pathway may be involved in osmotic regulation under salt stress, helping plants maintain the intracellular water balance in high-salt environments [23]. Transcription factors can bind to the promoter regions of specific genes to regulate their transcription. For example, transcription factors of the WRKY family play important roles in the plant response to salt stress, and they can regulate gene expression by binding to W-box elements [24]. However, further research is needed to investigate the synergistic effects of different metabolic pathways on GABA-induced salt tolerance and the role of differentially expressed genes (DEGs). Previous studies have focused mainly on the accumulation of GABA under tomato stress and the analysis of related physiological salt tolerance mechanisms, with little research on the molecular mechanisms by which GABA enhances plant tolerance to salt stress. GAD is a key enzyme that regulates GABA synthesis, and its expression is related to GABA accumulation. The tomato genome consists of five GAD genes. Research has shown that the SlGAD1 gene has the highest expression level during the ripening stage of fruit ripening, after which its expression level gradually decreases. In contrast, SlGAD2 and SlGAD3 presented relatively high transcription levels during the early stages of fruit development, but their mRNA content tended to decrease as the fruit matured. The experimental data further revealed that SlGAD2 and SlGAD3 play important roles in regulating the biosynthesis of GABA in fruits, whereas SlGAD1 is not significantly correlated with changes in GABA content [25]. Therefore, SlGAD1 is likely related to fruit growth and development but is not significantly correlated with the GABA content in fruit. SlGAD4 is not related to tomato growth and development, and the function of SlGAD5 has not yet been reported. We have shown that GABA accumulation is influenced mainly by GAD enzyme activity [10]. The results of transcriptome data analysis of grafted seedlings revealed significant differences in salt tolerance between grafted and nongrafted seedlings, with only SlGAD1 gene expression showing significant differences in leaves, whereas other SlGAD genes presented no clear pattern of change. It is speculated that SlGAD1 is involved in the response of tomato to salt stress. However, studies on the molecular mechanisms governing the regulatory role of GABA in enhancing tolerance to saline–alkali stress in plants are scarce. Here, we report the results of a transcriptome analysis of 33 leaf samples, which included data collected at 5 different time points after the treatment of salt-sensitive tomato plants with GABA under salt stress and the measurement of key enzyme activities and metabolite levels associated with metabolic pathways. By constructing materials overexpressing the key GABA synthesis gene SlGAD1 and evaluating salt tolerance and measuring salt tolerance indicators, the function of SlGAD1 in regulating tomato salt tolerance was clarified. The objective of this study was to clarify the unique regulatory mechanism associated with GABA-induced salt tolerance in plants for the first time and to elucidate the mechanism by which exogenous GABA and the key gene SlGAD1 increase salt tolerance in tomato. Materials and methods Plant growth conditions The tomato variety ‘Zhongza 9’ (previously selected and reported by our research group) was used as the salt-sensitive experimental material and was purchased from the Vegetable and Flower Institute of the Chinese Academy of Agricultural Sciences (Beijing, China). When the tomato plants reached the four-leaf and one-heart stage, they were transplanted into 20 L grey plastic basins and cultured with 18 L Hoagland’s solution via hydroponics (pH 6.5; EC 2.0 ~ 2.2). The daytime temperature is 25 ℃~28 ℃, and the night temperature is 15 ℃~18 ℃. The photoperiod is 16 h/8 h, and the light intensity is 140 ~ 150 µ mol·m− 2·s− 1. The air humidity is 65%, and the air pump is used for 30 min·h− 1, with the dissolved oxygen concentration (DO) of the nutrient solution is maintained at 8.0 mg·L− 1. Salt stress was applied after 3 days of preculture. Treatments and sampling method GABA solution was added to the nutrient mixture for hydroponic treatment of the tomato seedlings. The treatments were as follows: (1) salt treatment (N): 175 mmol·L− 1 NaCl + Hoagland’s solution; (2) salt treatment with GABA (GN): 5 mmol·L− 1 GABA + 175 mmol·L− 1 NaCl + Hoagland’s solution. All the chemicals used were of analytical grade and were obtained from Sigma Aldrich Co., LLC (USA). The leaves of the treated plants were collected at 0 (control), 6, 12, 24, 48 and 96 h for transcriptomic sequencing (three biological replicates per treatment), and the same samples were used for physiological index determination (four biological replicates per treatment). For each treatment, the second leaf from the top of the plants was used. Physiological measurements Due to the focus of this study on the growth status of plants under salt stress after GABA addition, physiological and molecular indicators were not measured at the 0 h time point. The salt damage index was determined via the method of Liu et al. [26]. The electrolyte leakage of leaves was measured according to the methods of Dionisio-Sese and Toita [27]. The fresh weight of 15 plants in each treatment was determined, after which the material was dried at 80 °C for 10 min and subsequently heated at 105 °C in an oven. The dry weight of the plants was subsequently determined. The tissue moisture content was calculated as (fresh weight‒dry weight)/fresh weight×100%, and the root activity was determined via triphenyltetrazolium chloride (TTC) colorimetry [28]. RNA-seq and data analysis The 33 samples of frozen leaves were subsequently sent to Novogene Bioinformatics Technology Co. (Beijing, China) for RNA-seq. The RNA-seq libraries were sequenced on an Illumina HiSeq X Ten platform. The data were analysed via the filter software SOAPnuke, and the low-quality reads, adaptor contaminants and reads with high levels of unknown N bases were removed. Clean reads were aligned to the reference tomato SL 2.0 genome (https://www.ncbi.nlm.nih.gov/genome/7?genome_assembly_id=393272) via HISAT (NCBI). Use FPKM (fragments per kill of transcript per million mapped reads) values to represent the expression abundance of corresponding Unigenes. Differential expression analysis was performed using DESeq2 (v1.20.0) software based on the read count values of gene expression levels, with a corrected P < 0.05 as the screening criterion for DEG. According to the Kyoto Encyclopedia of Genes and Genomes (KEGG) pathway. Adopt mainstream hierarchical clustering to cluster the FPKM values of genes, and perform Z-score normalization on rows. FPKM values were used to calculate the expression levels of genes and transcripts. Gene expression is expressed as the FPKM value, and the screening criteria for significantly DEGs were a fold change of more than 2 (| log2fc | ≥1) and a false discovery rate (FDR) < 0.05 [29]. Perform GO enrichment and KEGG metabolic pathway analysis using online software agriGO based on DGEs. Using the WGCNA package in R language, a gene module is constructed for the whole genome genes that have been filtered out of low expression genes (FPKM ≤ 1.0). Weighted parameter soft thresholds are set to cluster genes based on their pairwise dissimilarity, resulting in a systematic clustering tree of genes and branch modules. The module eigenvalue (ME) of each module is calculated, and the ME is clustered. The high similarity ME is used to merge the branch modules, ultimately obtaining an integrated gene module. Determine the expression of the module in the sample by calculating the correlation and significance level between ME and the sam-ple phenotype. qRT‒PCR validation The same RNA samples used for the transcriptomic analysis were used to verify the transcriptomic data via qRT‒PCR. The relative expression levels of the candidate genes were analysed via the use of the Actin-7 gene as an internal reference, the sequence of which was retrieved from GenBank (accession number X58253). The gene-specific primers used are shown in Table S1. qRT‒PCR was performed according to the instructions for the Fast Super EvaGreen qPCR Master Mix Kit (US Everbright®, Inc.). Six biological replicates were set per treatment sample. Metabolite and enzyme activity assays The treatment and sampling time were the same as those used for transcriptome sequencing. The soluble sugar content was determined via anthrone colorimetry [30]. The contents of sucrose, glucose and fructose were determined via high-performance liquid chromatography (HPLC) with a Shimadzu RID-10 A differential detector and a YMC-Pack Polyamine II column [31]. The mobile phase was acetonitrile: water (75:25, V: V), the column temperature was 30 °C, the cell temperature was 35 °C, the injection volume was 10 µL, and the flow rate was 1 mL·min− 1. Phenylalanine aminotransferase (PAL), cinnamate-4-hydroxylase (C4H), 4-Coumarate coenzyme A ligase (4CL) and chalcone isomerase (CHI) activities were determined via the methods of Ge et al. [32]. The chalcone synthase (CHS) activity was determined via an ELISA kit (J&L Biological Co.). The methods of Zheng et al. were used as references for HPLC-based identification of flavonoid components [33]. The amino acid content was determined by precolumn derivatization with DNFB and RP-HPLC [34]. The plant leaves were subsequently dried, crushed, and passed through a 60-mesh sieve. One hundred milligrams of dried sample was subsequently boiled in 25 mL of deionized water for 2 h and then filtered after cooling [35]. After digestion with HNO3-HClO4, the contents of Na+, K+ and Ca2+ in the filtrate were determined via atomic absorption spectrophotometry (Beijing, TAS-990). Salt tolerance validation and physiological index analysis of overexpressed plants Two overexpressing SlGAD1 tomato lines (SlGAD1OE # 3, SlGAD1OE # 7) and wild-type plant ‘Ailsa Craig’ were selected as test materials and subjected to salt stress (175 mmol · L− 1 NaCl) treatment at the stage of four leaves and one center in seedlings, with normal nutrient solution treatment as the control. Salt damage index, GABA content, flavonoid components, and key gene changes measure refers to the existing methods in this article. The design of SlGAD1 overexpression vector is shown in Fig. S1, with resistance of screening to hygromycin. The Agrobacterium strain is GV3101. Positive seedling screening PCR was shown in Fig. S2. The screening fragment size is 519 bp, and the primer sequence is shown in Table S2. Statistical analysis Fisher’s least significant difference (LSD) test was used for analysis of variance (SAS 8.1 statistical procedure). The difference was assessed at a significance level of 0.05. Results Physiological indicator changes in tomato plants treated with exogenous GABA Based on previous experiments, we evaluated the phenotype of GABA in alleviating salt stress injury in tomato seedlings by screening for different salt concentration phenotypes and the optimal concentration of GABA [36] (Fig. S3). The salt damage index and electrolyte leakage of the plants subjected to NaCl increased with increasing treatment time, and the application of GABA significantly reduced the salt damage index and electrolyte leakage (Fig. 1A and B). Compared with the N treatments, the GN treatments decreased the salt damage index by 50.0%, 37.7%, and 26.4% and the electrolyte permeability by 40.2%, 52.0%, and 35.3%, respectively. The fresh weight of the plants in the GN treatment group was significantly greater than that of the plants in the N treatment group. However, there was no significant increase in dry weight, except at 96 h (Fig. 1C). The application of GABA also effectively reduced the downwards trend and increased the tissue water content by 2.3% at 96 h (Fig. 1D). Exogenous GABA treatment significantly improved root activity under NaCl conditions, with an 8.9%~20.3% increase observed at 6 ~ 96 h (Fig. 1E). Effects of exogenous GABA on salt tolerance of tomato seedlings. A Effect of GABA addition on salt injury index of tomato seedlings under salt stress. B Effect of GABA addition on leaf electrolyte leakage of tomato seedlings under salt stress. C Effect of GABA addition on dry and fresh weight of tomato seedlings under salt stress. D Effect of GABA addition on water content in tissue of tomato seedlings under salt stress. E Effect of GABA addition on root activity of tomato seedlings under salt stress Comparison and enrichment analysis of the DEGs identified at the transcriptomic level in response to GABA under salt stress Table S2 shows that 249 Gb of clean bases were obtained from the quality-filtered sequence data (error rate ≤ 0.03) via RNA-seq; the mapping rate ranged from 82.53% to 97.40%, and the base coverage was > 92.65%. The unique mapping rate ranged from 80.63% to 95.73%, and the percentage of reads aligned to positive and negative chains exceeded 40.30% (Table S3). All the sample correlations indicated that the biological replicates in this study were reliable (Fig. S4). By comparing the two treatments at the same time during salt stress, we found that the number of DEGs varied greatly at different time points, and the changes occurred mainly during the early stages (6 h) and late stages (48–96 h) of salt stress. In total, 1118, 43, 29, 854 and 324 genes were differentially expressed at various time points from 6 to 96 h (Fig. 2A). We compared the different time points for the GN and N treatment groups and found that more than 1000 genes were differentially expressed at different time points within the same treatment group (Fig. 2B and C). The above results indicated that the two treatments elicited relatively similar gene expression changes at 12 and 24 h, and Gene Ontology (GO) and Kyoto Encyclopedia of Genes and Genomes (KEGG) enrichment analyses also confirmed these findings in the GN vs. C and N vs. C comparison groups at 12 and 24 h (Fig. S5). These findings can explain why there were so few DEGs at 12 and 24 h, indicating that the other time points represented key periods for the effect of the DEGs in the two treatment groups. The expression patterns of 23 genes were analysed via qPCR to evaluate the reliability of the RNA-seq data in this study (Fig. 3A and B). The correlation coefficients (R2 = 0.9267, R2 = 0.9072) suggested that the RNA-seq data were reliable (Fig. 3C and D). Comparison of different sample DEGs. GNvsN (6-96h). A Number of different genes in different treatments. B Number of genes was different between GN treatments. C Number of genes was different between N treatments.DEGs were identified setting the false discovery rate at 10% (p < 0.1) and the minimum fold change at ±1.1 qPCR verification of differential gene expression in GN and N treatments. A qPCR verification of GNs. B qPCR verification of Ns. C Correlation analysis between transcriptome data and qPCR validation of GNs. D Correlation analysis between transcriptome data and qPCR validation of Ns GO terms such as ‘response to biotic stimulus’, ‘response to auxin’, and ‘response to stress’ were commonly identified (Fig. S6). After the DEGs at the three main time points (6, 48 and 96 h) were analysed, we found that they were enriched in transcription factor (TF)-related terms, such as ‘transmembrane transporter activity’ and ‘nucleic acid-binding transcription factor activity’. KEGG enrichment analysis was performed on the DEGs detected in the GN vs. N comparison group (Fig. S7). The pathways related to stress that were most significantly enriched included ‘plant hormone signal transduction’, ‘MAPK signalling pathway-plant’, and ‘photosynthesis-antenna proteins’. Notably, the ‘amino sugar and nucleotide sugar network’, ‘phenylpropanoid biosynthesis’, ‘flavonoid biosynthesis’, and ‘zeatin biosynthesis’ pathways, which are not common in the salt resistance response, were also identified. Moreover, the ‘biosynthesis of amino acids’ pathway was also detected at 12 h and 48 h, and this pathway was thought to be directly related to the presence of excess GABA. Using a coefficient of variation of 0.5 to filter the data, 31 different gene co expression modules were obtained from transcriptome data, and each module was represented by a different color (Fig. S8). There are differences in gene expression patterns between GN and N treatments at different time periods, and the GN group shows a greater trend of upregulation or downregulation. Among them, the turquoise module contains the highest number of genes, with 3187, while the steelblue module contains the lowest number of genes, with only 34. To further explore the modules and co expressed genes related to GABA effects, correlation analysis was conducted on 31 gene modules (Fig. S9). We found that modules MEblue, MEbrown, MEgreen, MElightcyan, and MEmidnightblue have high specificity and may contain key genes regulating tomato salt stress response by GABA. Among them, 2488 genes in the MEbrown module showed a significant upregulation trend between GN treatment and N treatment at 6 ~ 12 h (Fig. S10). This module contains the key gene SlGAD1 in the GABA synthesis pathway. Changes in gene expression and metabolite levels in the GABA and amino acid synthesis pathways We carried out a comprehensive analysis of genes related to the ‘biosynthesis of amino acids’ metabolic pathway and the levels of 17 main amino acids (Fig. 4A), focusing on genes associated with the GABA metabolic pathway. Compared with those in NS, the contents of free amino acids in the GN treatments increased, and these amino acids appeared mainly at 12 ~ 96 h. The levels of proline, GABA, and phenylalanine, which are considered closely related to stress, exhibited the most dramatic increase at 48 ~ 96 h after treatment. Similarly, the content of glutamic acid, a raw material for GABA synthesis, increased by 39.5% and 33.5% at 24 and 48 h, respectively. The synthesis of the polyamine raw material putrescine for GABA synthesis via another metabolic pathway also increased significantly, with the putrescine level increasing up to 2.3-fold compared with that in the N treatment group. The expression of genes related to GABA synthesis and transport also changed significantly (Fig. 4A). First, the expression levels of the GABA transmembrane transporter (GAT) gene and the gene encoding the Al-activated malate transporter (ALMT), which is considered a GABA transport receptor, were significantly elevated after the addition of GABA. GAD is the major GABA synthase gene, and four homologous genes were upregulated at 6 ~ 12 h. The bidirectional amino acid transporter (BAT1) gene encodes the major transporter of GABA into mitochondria and was upregulated at 48 ~ 96 h. Two metabolic enzymes of the GABA shunt, SSADH and GABA-TP, were downregulated in the GN treatment group. Changes of Genes and GABA and amino acid metabolic pathway. A Changes of GABA and amino acid metabolic pathway related genes of GNs and Ns. B Gene changes related to amino acid transporters of GNs and Ns The expression of genes related to GABA synthesis and transport also changed significantly. First, the expression level of the GABA transmembrane transporter (GAT) gene was significantly elevated after the addition of GABA, and the expression of the gene encoding the Al-activated malate transporter (ALMT), which is considered a GABA transport receptor, changed. GAD is the major GABA synthase gene, and four homologous genes were upregulated at 6 ~ 12 h. The bidirectional amino acid transporter (BAT1) gene encodes the major transporter of GABA into mitochondria and was upregulated at 48 ~ 96 h. Two metabolic enzymes of the GABA shunt, SSADH and GABA-TP, were downregulated in the GN treatment group. Because the accumulation of amino acids in leaves is related not only to metabolism but also to transport, we analysed 38 amino acid transporter (AAT) genes from different amino acid transporter families (Fig. 4B), including 5 functional families, namely, amino acid permeases (AAPs), proline transporters (ProTs), lysine‒histidine transporters (LHTs), cationic amino acid transporters (CATs), and amino acid vacuolar transporters (AVTs). The expression of AAP, ProT, and LHT family genes did not significantly change after treatment, whereas the expression of 3 CAT genes and 10 AVT family transporter genes significantly increased after treatment, especially at 48 ~ 96 h after treatment. Changes in gene expression and metabolite levels associated with the phenylpropanoid and flavonoid biosynthesis pathways ‘Phenylpropanoid biosynthesis’ and ‘flavonoid biosynthesis’ are two closely related metabolic pathways. Phenylpropanoid metabolism occurs downstream of phenylalanine metabolism and is associated with amino acid metabolism. We comprehensively analysed the expression levels of the major enzyme-coding genes and the related enzyme activities and metabolite levels in the two metabolic processes (Fig. 5). We detected 31 different genes belonging to 13 gene boxes. Most of the genes tended to be upregulated after 24 ~ 96 h of treatment. These DEGs were enriched in the ‘phenylpropanoid biosynthesis’ pathway. The total expression level in the GN treatment group was slightly greater than that in the N treatment group. The activities of key enzymes involved in two metabolic pathways, such as PAL, 4CL, C4H, CHS and CHI, were promoted when GABA was added. The corresponding results revealed that the levels of intermediate products, such as cinnamic acid, naringenin, kaempferol, and quercetin, significantly increased after 24 ~ 96 h. The levels of phenylalanine and total flavonoids also significantly increased at 12 ~ 96 h. Metabolic chart of ‘phenylpropanoid biosynthesis’ and ‘flavonoid biosynthesis’. Pathway-related gene expression, key enzyme activity and metabolite content in the leaves of plants from the GN and N treatment groups Analysis of aminosugar metabolism pathways we measured the sugar content in the leaves of plants subjected to two treatments targeting the amino sugar and nucleotide sugar metabolism pathways (Fig. 6A). In general, the contents of sucrose, glucose and soluble sugars in the leaves of plants treated with N tended to increase, while the level of fructose first decreased but then increased. The sugar content in the GN treatment group was greater than that in the N treatment group. The difference in sucrose content was not notable during most of the study period. The increase in sugar content also affected the expression of sugar transport genes. Fig. 6B shows that 19 sugar transporter genes presented significant changes in expression at different treatment time points. Most of the genes tended to be upregulated, but the amplitude of the upregulated genes in the GN treatment was significantly greater than that in the N treatment, especially for UDP-galactose/UDP-glucose transporters (LOC101264861, LOC101247050, LOC101257753, and LOC101263578) and nucleotide/sugar transporter family proteins (LOC101262574). Soluble sugar content and sugar transporter related genes changes in leaves of GNs and Ns. A Soluble sugar content in leaves of GNs and Ns. B Heatmap of sugar transporter related genes in leaves of GNs and Ns Analysis of the gene expression patterns of ion transporters and determination of leaf ion content We measured the ion content in the leaves. With increasing treatment time, the contents of Na+, K+, and Ca2+ tended to increase (Fig. 7A). By comparing the N and GN treatment groups, we found that GABA application significantly increased the contents of K+ and Ca2+ in each treatment, with the most significant effect occurring at 12 h, at which time the contents of these ions increased by 31.9% and 48.1%, respectively. However, after 96 h of treatment, the Na+ content decreased significantly, and the most significant decrease was 26.4%. The K+/Na+ and Ca2+/Na+ ratios increased significantly, with the K+/Na+ ratio increasing most significantly at 96 h, reaching 61.9%, and the Ca2+/Na+ ratio increasing most significantly at 12 h, reaching 76.4% (Fig. 7B). Effect of GABA addition on ion content,ratio and gene changes related to ion transport. A Effect of GABA addition on Na+, K+, Ca2+content in leaves of tomato seedlings under salt stress. B Effect of GABA addition on K+/Na+, Ca2+/Na+ in leaves of tomato seedlings under salt stress. C Gene changes related to ion transport Next, we examined the abovementioned ion-associated and hydrogen ion transporter genes, such as Na+/H+ antiporters (NHXs), cation/proton antiporters (CPAs), K+ efflux antiporters (KEAs) and Na+/H+ exchangers (NHEs) (Fig. 7C). A comparison of the changes in gene expression after GABA addition revealed that most of the genes were upregulated, and the changes occurred mainly at 6 ~ 48 h. The expression levels of NHX, CPA and KEA family genes were significantly greater in the GN treatment group than in the N treatment group. There was no significant difference in the expression of most genes, such as NHEs, CCXs, NCXs and CAXs, between the two treatments. Identification of key TFs regulating multiple metabolic processes On the basis of the transcriptomic data, we obtained DEGs enriched in TF-related pathways at different time points (Fig. 8A). A total of 494 TF-encoding genes were screened after the data from the two genotypes were analysed; 244 upregulated and 250 downregulated DEGs were found to encode TFs. After classifying these TFs, we found that they belonged to 15 main families. The MYB family was the most abundant family, followed by the HLH family and the AP2/ERF family. There were more downregulated DEGs than upregulated DEGs in the above three TF families. WRKYs ranked fourth, but most WRKY genes were upregulated. In addition, members of the bHLH, bZIP, GATA, HSF and other major stress response-related TF families were identified. Changes of TFs of GNs and Ns. A The number of significantlychanging transcription factors in the transcriptome data. B Heatmap of transcription factor expression at different time points We selected 23 TF-encoding genes whose expression changed most significantly (Fig. 8B). The expression levels of 25 of these genes significantly differed: AP2/ERF, bHLH, HSF, MYB, MADS, bZIP, HLH, CBFB_NFYA, NAM and TCP TF-encoding genes. The expression of different TFs, such as MYB TFs, changed at different time points, occurring mainly at 24 ~ 96 h, and that of HSF, BHLH, TCP and HLH changed at 6 ~ 24 h. The patterns of differentially expressed TF-encoding genes, which may play a regulatory role in the molecular mechanism through which GABA improves plant salt tolerance, differed significantly between the GN and N treatment groups. The effect of SlGAD1 overexpression on salt tolerance in tomato Compared with wild-type plants, SlGAD1-overexpressing plants presented slight dwarfism, but there were no significant differences in root or leaf growth. After salt stress treatment, the degree of yellowing and wilting of the plants whose leaves were overexpressing was relatively mild, indicating increased salt tolerance (Fig. 9A). In addition, the expression level of SlGAD1 in different tissues of SlGAD1-overexpressing plants was significantly greater than that in wild-type plants. The expression level of SlGAD1 in SlGAD1OE #7 was even greater, with an increase of 2.6 ~ 3.2 times under normal conditions and 2.8 ~ 3.5 times after salt stress (Fig. 9B), indicating that SlGAD1 enhances tomato salt tolerance. Phenotype and tissue expression patterns of tomato plants overexpressing SlGAD1. Leaf and root system status of different types of plants. A SlGAD1 relative expression in different tissues (B) As shown in Fig. 10, the overexpression of SlGAD1 significantly reduced the salt damage index of the plants. Although both overexpression lines presented a decrease in the salt damage index, the salt damage index of SlGAD1OE #7 was significantly lower than that of SlGAD1OE #3 on the 2nd and 6th days of treatment, indicating greater salt tolerance. The results of the membrane permeability analysis also revealed that the overexpression of SlGAD1 significantly reduced membrane permeability, with a decrease ranging from 22.5%~33.6%. Salt damage index and membrane permeability of SlGAD1OE tomato plants under salt stress. A represents the salinity stress index. B represents the membrane permeability As shown in Fig. 11, the overexpression of SlGAD1 significantly increased the GABA content of the plants, with SlGAD1OE #7 resulting in a relatively high GABA content and relatively high SlGAD1 expression. Under the control treatment, the GABA content was relatively high, increasing by 0.81 ~ 1.19 times compared with that in the wild type. The increase under salt stress also reached 0.91 ~ 0.95-fold, indicating that the expression of SlGAD1 directly affects the GABA content in tomato plants. GABA content in the leaves of SlGAD1OE tomato plants under salt stress Owing to the significant impact of GABA on flavonoid synthesis and metabolism identified via transcriptome data analysis, the expression levels of major flavonoid components and key genes involved in flavonoid synthesis and metabolism were evaluated in the leaves of the overexpression and wild-type plants. As shown in Figs. 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14 and 15, the contents of the five main flavonoid components in the leaves of SlGAD1-overexpressing plants were significantly greater than those in the leaves of wild-type plants, and the total flavonoid content increased by 1.6 ~ 2.1 times, indicating that the overexpression of SlGAD1 can induce flavonoid synthesis and promote intracellular flavonoid accumulation under salt stress. Flavonoid components in the leaves of SlGAD1OE tomato plants under salt stress Expression of major genes involved in flavonoid synthesis in the leaves of SlGAD1OE tomato plants under salt stress Pattern diagram of different metabolic pathways involved in the salt tolerance of tomato seedlings induced by exogenous GABA. Note: the red border shows upregulated genes and substances, whereas the green border shows downregulated genes and substances. The square represents the substance, and the oval represents the gene Pattern diagram of SlGAD1 promotes salt tolerance in tomatoes by increasing GABA content and flavonoid metabolism As shown in Fig. 13, the expression levels of four key genes involved in flavonoid synthesis in SlGAD1-overexpressing plant leaves were significantly increased under salt stress. CHS, CHI, and F3H are important catalytic enzymes in the upstream pathway of flavonoid synthesis, whereas FLS is an important gene in the downstream pathway. Under normal conditions, the overexpression of SlGAD1 does not affect the expression of these genes. However, under salt stress conditions, the expression levels of SlCHS and SlCHI in SlGAD1-overexpressing plants increased by more than 36.9%, whereas the expression levels of SlF3H and SlFLS increased by more than 1.41-fold. These findings indicate that the SlGAD1 gene can positively regulate the expression of flavonoid synthesis pathway genes and promote flavonoid accumulation, with a more significant effect on catalytic enzyme-encoding genes such as SlF3H and SlFLS. Discussion Exogenous GABA alleviates the phenotypic symptoms of salt damage Exogenous application of GABA can alleviate the inhibitory effects of salt stress on plant growth by increasing the antioxidant capacity [10], reducing chlorophyll degradation, and maintaining leaf photosynthesis [37]. The results of the present experiments revealed that, compared with the plants in the N group, those in the GN group were more tolerant to salt in terms of their phenotype, biomass and physiology. This finding is consistent with the results of our previous research. Compared with other studies, this study included more time points. Owing to the relatively high concentration of salt used, the plants presented symptoms of salt damage at the early stage of salt stress, such as at 6 h, during which the physiological indicators also changed. This system also provides a reference for determining the early phenotype and internal physiological and biochemical responses caused by the application of GABA to plants under salt stress. Although salt ions are usually absorbed through the root system, the aboveground part of plants is the main site for evaluating salt tolerance phenotypes, such as leaf discoloration, wilting, and stem lodging. Therefore, in order to comprehensively evaluate the physiological and molecular mechanisms of GABA affecting tomato salt tolerance, we selected leaves as the main site for measurement. Of course, this may have certain limitations in the analysis of salt tolerance mechanisms, such as neglecting root specific genes or metabolic pathways. The next step requires in-depth analysis of the differences in salt tolerance mechanisms among different tissues. Plants in the GN group presented different gene expression patterns, specifically at 6, 48 and 96 h Salt stress is a complex and continuous process. In the past, the effects of exogenous substances on the plant transcriptome under salt stress were usually assessed at certain time points [38, 39]. In this study, we compared plants in the GN and N groups at 6 ~ 96 h and tested a total of 5 time points and 33 samples. Our aim was to observe the GABA-mediated response to high-salt stress throughout the experimental period. After analysing the data, we found that the number of DEGs in the plants in the GN group and among the different time points were greater than those in the N group. In the middle stage of salt stress, i.e., at 12 ~ 24 h, very few genes were differentially expressed between the two treatments. That is, during this period, the two treatments resulted in relatively consistent changes in transcript levels. Data suggest that the changes in gene expression of the two treatment groups may have led to the development of a trimming state after salt stress. Therefore, the same expression pattern was detected. However, with increasing stress duration, i.e., at 48 ~ 96 h, the GABA treatment and N treatment resulted in enormous differences in transcript levels. Therefore, we speculate that 6, 48 and 96 h are the key periods for the changes in gene expression observed in these two treatment groups. Similarly, the comparison between the GO and KEGG functional enrichment results revealed that TF-encoding genes and stress-related gene pathways were enriched during these three periods. Therefore, on the one hand, these three time points may be the focus of the analysis of salt tolerance induced by exogenous GABA; on the other hand, the time points associated with small changes in gene expression may be more meaningful for the study of plant adaptation to salt stress. Amino acid transport may be more critical than amino acid synthesis for plants in the GN treatment group In the present study, the application of GABA to the root system induced GABA enrichment in the leaves. Amino acid metabolism plays an important role in plant salt stress adaptation, as indicated by the amino acid contents in leaves. In addition to GABA and proline, other amino acids also exhibited significant changes in abundance. Compared with those in the N group, the abundance of stress-related amino acids significantly increased in the GN group. However, except for several major synthesis- and metabolism-related genes, such as those involved in GAD and proline synthesis, most of the changes in gene expression were not significantly different between the two treatments. However, exogenous GABA induced the expression of genes encoding major amino acid transporters. The transport of amino acids is mediated mainly by the amino acid transporter (AAT) family [40]. This study identified AAT genes from 9 families. GATs are considered the major transporters of exogenous GABA into cells. The upregulated expression of GATs in the early stage of salt stress provides a basis for the transcription of exogenous GABA into cells. The functions of AVT family genes are thought to mediate the transport of various neutral amino acids and histidine into vacuoles to control amino acid flux in vacuolar-like organelles [41, 42]. Most transporter genes have been identified only in yeast and Arabidopsis thaliana, and only a few have been studied in tomato. For example, SlCAT9 has been shown to transport L-aspartic acid and L-glutamine into vacuoles, and the exchange of GABA is involved in the accumulation of amino acids during tomato fruit development [43]. For the first time, we report that the expression of AVT6C, AVT3C and SlCAT9 was upregulated after GABA treatment. Therefore, we speculate that amino acid transport is stimulated by exogenous GABA and that transporters under salt stress are equally important in regulating the intracellular amino acid content as they are in amino acid synthesis. The stimulation of phenylpropanoid and flavonoid metabolism in the later stage of stress increased salt stress resistance Plant phenylpropanoid metabolism is the main pathway for the synthesis of secondary metabolites such as phenols, flavonoids and lignin. These secondary metabolites constitute an important nonenzymatic antioxidant defence system in plants that can eliminate excessive ROS and slow membrane lipid peroxidation under stress [44]. The regulation of enzyme activity and metabolite accumulation by altering the expression levels of genes involved in the phenylpropanoid metabolic pathway is an important defence mechanism for plants to cope with abiotic stress [45]. However, the effects of GABA on phenylpropanoid and flavonoid metabolism under stress conditions have not been reported. In this study, the key gene expression, enzyme activity, and metabolite levels induced by GABA significantly changed in the middle and late stages of stress, namely, 24 ~ 48 h. Exogenous GABA induced significant accumulation of phenolic and flavonoid substances, which reflects the acceleration of antioxidant defence and lignification processes in plants, providing a new metabolic pathway for the mechanism of exogenous GABA-induced salt tolerance. Sugar transport are involved in GABA-induced salt tolerance The addition of GABA promoted the upregulation of several sugar transport genes in tomato plants under salt stress, and these genes basically progressed through the pre- and mid-stress stages, that is, 6 ~ 48 h. These findings provide a new direction for the study of the mechanism of plant salt tolerance. Exogenous GABA decreased the intracellular sodium content by increasing the expression of plasma transport genes High Na+ levels cause K+ deficiency and a K+/Na+ imbalance in cells [46, 47]. Na+ usually competes with K+ for binding sites and has important functions in cell membranes. Therefore, effective control of the intracellular Na+ concentration has become the main target for improving plant salt tolerance. In our study, plants in the N group presented greater Na+ contents in their leaves and lower K+/Na+ ratios than did those in the GN group did, indicating that the Na+ and K+ ions in plant cells are essential for salt tolerance. Moreover, their ratio is a critical parameter in investigations of salinity tolerance responses. Our research also revealed that the Ca2+ content in the leaves of plants in the N group was considerably greater than that in the leaves of the plants in the GN group. Compared with that of the N group, the Ca2+/Na+ ratio of the GN group was greater, especially at two successive time points after salt stress. We believe that maintaining cellular ion homeostasis is also one of the functions of GABA in the development of salt tolerance. We also observed changes in the expression of osmotic stress-related genes that encode transporters, such as NHXs, CPAs, and KEAs, which are generally believed to be the most important ion transporters involved in the response to salt stress. NHXs can transfer excess Na+ from the aboveground cytoplasm to vacuoles for isolation [48]. This not only reduces the concentration of Na+ in the cytoplasm and weakens the toxicity caused by Na+ but also leads to the use of Na+ as an osmotic regulator to prevent cell dehydration and mediate the absorption of K+ by vacuoles [49, 50]. CPAs exist widely and are located mainly in the plasma membrane, vacuolar membrane and organelle membrane in plants; CPAs can expel cations such as Na+, K+, and Li+ from cells and can maintain the dynamic equilibrium of ions and protons in cells [48, 51]. The expression of most CPAs and KEAs was significantly upregulated, and we speculate that the effect of GABA on the excretion of sodium ions in the cytoplasm directly regulates the expression of ion transport-related genes, including those involved in the transport of sodium ions, calcium ions, and potassium ions. These findings explain the results of the ion content assays and provide direct evidence that the addition of exogenous GABA affects the expression of intracellular ion transport genes to achieve intracellular sodium localization, which in turn induces salt tolerance. TFs May play an important role in regulating salt tolerance via GABA When plants are subjected to external stimulation or stress, TFs induce the expression of resistance genes and plant-related defence systems to respond to external stimuli or stress, generate a series of response signals in plants, transmit them, and ultimately shift the adaptability of plants to the environment [52]. The function of TFs in the response to stress has attracted much attention in recent years and has also been studied in depth by researchers. Numerous studies have shown that TFs such as AP2/ERF and MYB TFs play important regulatory roles in plant responses to abiotic stress [53, 54]. In this study, the changes in TF expression after GABA addition were consistent among members of the AP2, bHLH, HSF, and MYB families. The overexpression of the CaMYb78 TF enhances the resistance of chickpea to F. oxysporum [55]. MYB58 is a transcriptional activator of the lignin biosynthetic pathway during the formation of the secondary cell wall in Arabidopsis [56]. Furthermore, SlMYC2 is involved in MeJA-induced cold tolerance [57]. The overexpression of bHLH123 promotes salt resistance in tobacco plants [58]. Some TFs may play a role in regulating key genes involved in the molecular mechanism through which GABA enhances plant salt tolerance. However, owing to the limited identification and validation of the TFs involved in the tomato salt tolerance response, further validation of their function in salt tolerance is needed. The results of this study provide reference data for future research on the role of TFs in GABA-induced plant salt tolerance. Multiple metabolic pathways participate in the GABA-induced salt tolerance response in tomato plants and exhibit synergistic effects at different times The addition of exogenous GABA induced significant changes in key genes and products of several metabolic pathways during salt stress. The GN group presented a specific pattern of DEG enrichment, which indicated that GABA had a time-dependent effect during salt stress; in particular, the most dramatic genetic changes were observed at 6, 48 and 96 h after salt stress. The results of the transcriptomic and metabolite analyses led us to propose a potential mechanism by which amino acid, phenylpropanoid, flavonoid, and zeatin metabolic pathways and TF regulation play important synergistic roles in the response to GABA-induced salt stress at different stages (Fig. 14). The gene changes involved in different metabolic pathways in tomato plant leaves subjected to GABA presented the following temporal patterns. Changes in amino acid metabolism, zeaxanthin metabolism, and TF expression occur throughout the entire salt stress process (6 ~ 96 h). GABA transporters are sensitive to the exogenous GABA response, and genetic changes occur in the early stages (6 h) of salt stress. Afterwards, changes in GABA synthesis and catabolism become dominant (48 ~ 96 h), accompanied by changes in phenylpropanoid and flavonoid metabolism caused by amino acid accumulation. Sugar transport and ion transport are sensitive to the exogenous GABA response, and their changes occur in the middle and early stages of salt stress. Therefore, we propose a model of exogenous GABA-induced salt tolerance in tomato. After exogenous GABA enters leaf cells, it causes the accumulation of endogenous GABA, which first increases the amount of osmoregulatory substances, such as amino acids, sugars, and potassium ions, to maintain relatively stable osmotic pressure in the plant and generate sodium ions. Then, it induces antioxidant defence responses in phenylpropanoid and flavonoid metabolism, reducing the damage caused by sodium ions to leaves. During this process, transcription factors are involved in regulating various metabolic pathways, signal transduction, and changes in key genes, achieving synergistic effects on multiple metabolic processes and improving tomato salt tolerance. Positive regulation of salt tolerance and flavonoid content in tomato plants by SlGAD1 GABA enhances plant tolerance to various abiotic stresses and is closely related to its accumulation in the body [59]. The GAD enzyme plays a major role in the in vivo accumulation of GABA. In this study, we identified a GAD gene, SlGAD1, that responds positively to salt stress and constructed overexpression and knockout tomato materials via transgenic technology to evaluate the function of SlGAD1. The overexpression of SlGAD1 increases the GABA content and salt tolerance in tomato, confirming that SlGAD1 positively regulates tomato salt tolerance. Previous studies on GABA accumulation in tomatoes have focused mainly on fruits and have focused mainly on SlGAD2 and GABA-T, the main genes involved in GABA catabolism whose expression levels are high under normal conditions. This study demonstrated that, under stress conditions, GABA accumulation is related mainly to the activity of GADs, making SlGADs more meaningful for improving plant stress resistance. Zhang Yong reported that SlGAD2 is induced at low temperatures and that the overexpression of SlGAD2 significantly increases the content of endogenous GABA and antioxidant enzyme activity in tomato seedlings, indicating strong resistance to low temperatures [60]. These findings indicate that there are certain differences in the regulation of stress resistance function by GAD genes among different species. Previous studies on the low-temperature and salt-alkali response of SlGAD2 and SlGAD1 revealed that SlGAD2 alleviated low-temperature-induced oxidative stress damage by increasing the accumulation of flavonoids and anthocyanins. Under salt alkali stress, SlGAD1 actively regulates ethylene signalling pathway genes and ethylene content, indicating that GABA is involved in the ethylene signalling pathway [61]. Therefore, the direct function of GABA as an important signalling substance in reactive oxygen species metabolism and ion transport requires further exploration of the regulatory molecular mechanism of SlGAD1 in this process. As an important class of secondary metabolites, flavonoids play crucial roles in the response of plants to abiotic stress. Their main functions include clearing excess reactive oxygen species and enhancing plant stress resistance. At present, the biosynthetic pathway of flavonoids has been extensively studied. The synthesis process begins with phenylalanine and is catalysed by upstream key enzymes such as chalcone synthase (CHS) and chalcone isomerase (CHI). First, dihydroflavonol intermediates are formed, which are then converted into anthocyanins under the action of dihydroflavonol reductase (DFR) and anthocyanin synthase (ANS). Moreover, flavonol compounds are generated through the catalysis of flavonol synthase (FLS). Its downstream involvement in lignin metabolism is closely related to plant disease and stress resistance [62]. Reports indicate that both salt stress and exogenous GABA can induce the expression of flavonoid synthesis genes, promoting the accumulation of flavonoids [63, 64]. However, how abiotic stress regulates flavonoid synthesis, especially the relationship between the GABA metabolism pathway and the flavonoid synthesis pathway, is poorly understood. This study investigated the flavonoid content in roots and leaves. The results revealed that the levels of the five main flavonoid components in the leaves of SlGAD1-overexpressing plants were significantly greater than those in the leaves of wild-type plants. The expression levels of SlCHS and SlCHI in SlGAD1-overexpressing plants increased by more than 36.9%, and the expression levels of SlF3H and SlFLS increased by more than 1.41 times. However, this effect did not occur under normal conditions, which proves that the SlGAD1 gene can positively regulate the expression of flavonoid synthesis pathway genes and promote flavonoid accumulation and that SlGAD1 is involved in multiple pathways in response to salt stress. Conclusions This study on the synergistic effects of GABA administration on multiple metabolic pathways at different stages of salt tolerance enhancement in tomato plants contributes to a better understanding of the mechanisms of salt tolerance in plants and further screens and validates the key gene SlGAD1, which responds to salt tolerance by promoting GABA synthesis to increase salt tolerance in tomato plants (Fig. 15). Our results represent the first detailed analysis of transcriptomic data at multiple time points during salt stress and elucidate the synergistic process of multiple metabolic pathways. We propose a model of exogenous GABA-induced salt tolerance in tomato, which includes the accumulation and transport of osmoregulatory substances, an antioxidant defence response, synergistic effects on multiple metabolic processes, and improved salt tolerance. These findings provide direct evidence that exogenous GABA affects the phenylpropanoid and flavonoid pathways and ion transporters, which will be helpful for the development of GABA as an agent for inducing salt tolerance. These findings also provide new strategies for enhancing plant salt stress. Supplementary Information Supplementary Material 1. Acknowledgements Not applicable. Authors’ contributions Xiaolei Wu: Writing–review & editing, Writing–original draft. Ding Yuan: Writing–review & editing, Writing–original draft. Lida Chen: Writing–review & editing. Yukun Sun and Yening Wang: Formal analysis. Shiqing Jiang: Conceptualization. Xule Sun: Writing–original draft. Xinyue Zhang: Methodology. Jingrui Li: Writing–original draft. Hongbo Gao: Supervision, Project administration, Conceptualization. Funding This work was financially supported by Natural Science Foundation of Hebei Province (C2023204091), Hebei Facility vegetables Innovation Team of Modern Agro-industry Technology(HBCT2023100211) and S&T Program of Baoding(2494N014). Data availability The data presented in the study are deposited in NCBI(accession number GSE233233)and GSA(https://ngdc.cncb.ac.cn/gsa/browse/CRA033059). Declarations Ethics approval and consent to participate Not applicable. Consent for publication Not applicable. Competing interests The authors declare no competing interests. Footnotes Publisher’s Note Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations. References Associated Data Supplementary Materials Supplementary Material 1. Data Availability Statement The data presented in the study are deposited in NCBI(accession number GSE233233)and GSA(https://ngdc.cncb.ac.cn/gsa/browse/CRA033059).