Increasing γ-Aminobutyric Acid Content in Dwarf Cherry Tomato Using CRISPR/Cas9-Mediated Gene Editing

Abstract

Gamma-aminobutyric acid (GABA) is considered an important bioactive compound that improves sleep quality and regulates blood pressure. Tomatoes are an ideal horticultural crop that can accumulate a high level of GABA in fruits. The development of higher-GABA tomatoes has significant market potential. In this study, we edited the SlGAD3 gene to increase GABA content in the dwarf cherry tomato, WEIMEI T102. After transformation using the Agrobacterium-mediated method, we identified several SlGAD3 mutation lines, which showed changed GABA levels compared to the recipient line. Molecular characterization showed stable trait inheritance for multiple generations. The GABA level in fruits also stably accumulated for multiple generations, which significantly increased up to about 1.9 mg/g FW in E13-13. These results indicate that it is feasible to increase the GABA content in dwarf cherry tomatoes by using gene editing technology.

1. Introduction

Gamma-aminobutyric acid (GABA) is widely found in the plant kingdom and mammalian brains. Studies of its distribution and metabolism pathway have shown that free GABA is abundant in nerve tissues, and GABA is an intermediate product of glutamate metabolism [1]. GABA has been shown to be a typical inhibitory neurotransmitter in the central nervous system, playing a vital role in transmitting information, regulating neuronal development, and improving sleep and mood [2,3,4]. Studies have shown that GABA exhibits several other important health benefits, such as anti-hypertensive, anti-diabetes, and anti-inflammatory properties [5]. In order to prevent the occurrence of these diseases, more GABA-rich foods are needed in our daily lives. Microbial activities during the fermentation process of fermented foods, such as kimchi, miso, and soy sauce, produce GABA [6]. In addition, some grains like brown rice have higher GABA content after germination, and green tea may also contain a certain amount of GABA. Legumes such as soy and foods after fermentation such as natto are also rich in GABA [7]. However, GABA intake from these foods alone is not enough. GABA has been blended into many functional foods and is sold worldwide [8].

Tomatoes are an ideal horticultural crop that can accumulate high levels of GABA in fruits. Although the natural content of GABA in tomatoes may be lower than that in some other foods, the GABA retention rate of tomatoes is more than 90% after steaming at 100 °C for 5 min, while other vegetables such as spinach retain only 60% under the same conditions. Tomatoes can retain GABA when eaten raw or cooked for a short time [9]. Therefore, tomatoes, which are a high-intake vegetable in people’s daily lives, have attracted researchers’ attention. Scientists are working to increase the GABA content in tomatoes in order to raise daily dietary levels of GABA.

In recent years, many studies have tried to identify the key genes to regulate the GABA metabolism in tomato [10,11,12,13,14,15,16]. The elucidations of the GABA biosynthesis pathway in tomato lay a theoretical foundation for breeding new varieties of tomato with high GABA. In plants, GABA homeostasis is important for plant growth, and it is regulated by a GABA shunt [11]. In this shunt, GABA is first synthesized from its precursor glutamate via glutamate decarboxylase (GAD) catalyzation, and it is then reversibly converted to succinic semialdehyde (SSA) via GABA transaminase (GABA-T) catalyzation. GAD plays a key role in the biosynthesis of GABA. Several studies have identified five key genes encoding GADs. Upregulation of SlGAD2 and SlGAD3 can effectively improve the accumulation of GABA in tomato. A self-inhibition region in the C-terminal of GAD3 that reduces the enzyme activity has been identified. Increasing GAD activity when the domain was deleted was directly related to the accumulation of GABA content [12,13,14,15,16].

Clustered regularly interspaced short palindromic repeats/CRISPR-associated protein (CRISPR/Cas) mediated gene editing has been widely used for crop improvement. Currently, seven countries (Canada, Japan, the Philippines, the United States, Brazil, China, Colombia) have approved genome-edited crops regulated as conventional plants with no additional restrictions or on a case-by-case basis. Other countries such as South Korea, United Kingdom, and the EU are still debating about regulations for gene editing (https://crispr-gene-editing-regs-tracker.geneticliteracyproject.org/, accessed on 11 October 2025). This website also shows that a total of fourteen crops developed by new breeding techniques were approved for sale. A gene-edited tomato using CRISPR to contain more GABA submitted by Sanantech Seed was approved in 2021 in Japan. Studies showed increased GABA content by gene-editing the related genes, GADs or GABA-Ts [17,18,19,20]. Nonaka et al. showed that the highest-accumulation gene-edited Micro-Tom tomato line presented 2.329 ± 0.074 mg/g FW in T0, and the highest value corresponding to 1.7901 ± 0.2516 mg/g FW in T1 plants [17]. Li et al. found that the GABA level was increased to 89.88 mg/100 g FW in the GABA-T gene-edited lines [18]. GABA levels from two gene-edited hybrid tomato red fruits were 1.905 ± 0.22 mg/g FW and 1.314 ± 0.11 mg/g FW RED stages, respectively [19]. In another study, GABA levels could be as high as 3.269 mg/g FW in fruits of slgad3-SFT1-3, and GABA levels of SFT2 and SFT3 reached nearly 1 mg/g in SFT2 fruits, and even GABA content increased to over 6 mg/g in some slgad3-SFT2 mutant fruits [20].

In this study, we tried to edit the self-inhibitory domains of SlGAD3 using the CRISPR/Cas9 system to improve the activity of GAD and then increase GABA accumulation in the dwarf cherry tomato WEIMEI T102. The tomato was developed by our laboratory and has been launched on the market. This variety could be cultivated in greenhouses in pots and offered to customers as both a consumable and ornamental plant for office or house window settings, adding a touch of natural food and beauty. Free amino acid content, multiple-generation inheritance, plant height, fruit weight, and Soluble Solids Content (SSC) were tested to assess the difference between the edited lines and the wild-type tomato. This study provided a basis for rapidly increasing GABA content in the dwarf cherry tomato variety.

2. Materials and Methods

2.1. Plant Materials and Growth Conditions

Seeds of WEIMEI T102, a dwarf cherry tomato variety, were germinated on plates with moistened filter paper, then transplanted into soil and grown in the greenhouse with a 16 h/8 h light/dark photoperiod. Fruits were harvested for about 10 days after the breaker stage.

For lines that were planned for the field experiments, seeds were germinated in soil in the greenhouse and then transplanted to the field. One-month-old plants were collected to measure the plant height. Red ripe fruit was harvested to determine the GABA content, free amino acid level, SSC, and fruit weight.

2.2. Molecular Cloning

pHSE401, a vector of the CRISPR/Cas9 system, was used to express gRNA following the previous study [21]. For the construct, in brief, the T-DNA region contains an Arabidopsis U6-26 promoter that controls the single-guide RNA (sgRNA), a cauliflower mosaic virus 35S promoter-regulated Cas9, and a hygromycin phosphotransferase (hptII) selection marker gene under the control of a dual 35S promoter (Figure 1A). A 20 bp DNA sequence (CCCGAATGCCAAAAAAGTGGAGG) of SlGAD3 (Solyc01g005000) was selected as the gRNA for targeting. Two synthesized oligos were annealed and inserted into the two Bsa I recognition sites. The final construct pDN077 was confirmed by sequencing (Table S1).

Figure 1. CRISPR/Cas9-mediated gene editing of SlGAD3. (A) Structures of SlGAD3 gene in tomato. Blue letters are gRNA for targeting. Red letters indicate PAM site. (B) Schematic of the vector. Target site (sgRNA) of SlGAD3 activated by the U6 promoter, and the screening agent was hygromycin. (C) Sequence alignment between T102 and gene-edited lines. Red nucleotides indicate PAM region, blue nucleotides are sgRNA, purple nucleotides are insertion, and dashed lines indicate deletion.

2.3. Tomato Transformation

The CRISPR/Cas9 construct pDN077 was transformed into the Agrobacterium strain EHA105. Construct was transformed into the tomato genome with an Agrobacterium-mediated method. Pre-culture Stage: Sterile tomato seedlings were grown for approximately 7 days; then, the newly expanded cotyledons were excised at both ends and cut into 5 mm cotyledon segments. These segments were subjected to culturing at 28 °C in the dark for 2 days. Infection: The pre-cultured explants were immersed in Agrobacterium suspension for 15 min with gentle shaking in the dark. Co-culture Stage: After infection, the explants were blotted dry and placed with the cut surface facing upward for co-culture. They were maintained at 28 °C under dark conditions for 2 days. Selection Culture Stage: The explants were transferred to selection medium (M519 4.43 g/L, Sugar 30 g/L, Phytagel 3.75 g/L, IAA 0.1 mg/L, TZT 1 mg/L, hygromycin 5 mg/L, Timentin 200 mg/L, pH 5.8) with the cut-surface in contact with the medium and cultured under light at 26 °C. Subculturing was performed every 3 weeks until new shoots emerged. Shoot Elongation Stage: The newly formed shoots, along with attached callus, were excised and transferred to elongation medium (M519 4.43 g/L, Sugar 30 g/L, Phytagel 3.75 g/L, TZT 0.2 mg/L, hygromycin 6 mg/L, Timentin 200 mg/L, pH 5.8) for further growth. Rooting Stage: When the shoots reached approximately 1 cm in length, excess callus was removed, and the shoots were transferred to rooting medium (M519 4.43 g/L, Sugar 30 g/L, Phytagel 3.75 g/L, IBA 1 mg/L, hygromycin 5 mg/L, Timentin 200 mg/L, pH 5.8). Root initiation typically occurred within 3 to 5 days. The regenerated lines were transplanted into the soil under greenhouse conditions for further analysis.

2.4. Identification of Target Mutations

T0 generation plants were used to identify target SlGAD3 mutations. The genomic regions including the target sites were amplified with gene-specific primers (Table S1). The PCR products were directly sent for sequencing by Tsingke Biotech Co., Ltd. (Beijing, China). DECODR (https://decodr.org/, accessed on 2 April 2024) and Clustal Omega Multiple Sequence Alignment (https://www.ebi.ac.uk/jdispatcher/msa/clustalo, accessed on 20 August 2024) were used for sequence analysis.

Transgene-free individuals were selected from the T1 generation. Leaves from T1 plants were screened by PCR of hptII and Cas9 fragments from the CRISPR/Cas9 construct with the specific primers (Table S1). T2 plant testing was confirmed with Cas9 primers.

2.5. Off-Target Analysis

CRISPR-P 2.0 (http://crispr.hzau.edu.cn/CRISPR2/, accessed on 10 April 2025) was used to select the gRNA and the potential off-target sites in the tomato genome sequence (Solanum lycopersicum SL2.50). According to the genome location of the potential off-target sites, specific primers were designed based on the genomic sequence. Then the fragments including the putative off-target sites were amplified. PCR fragments were sequenced by Tsingke Biotech Co., Ltd. (Beijing, China). Clustal Omega Multiple Sequence Alignment (https://www.ebi.ac.uk/jdispatcher/msa/clustalo, accessed on 10 May 2025) was used for sequence alignment.

2.6. Determination of GABA and Other Free Amino Acid Content

Amino acids were measured using an Ultra High-Performance Liquid Chromatography and Tandem Mass Spectrometry (UHPLC-MS/MS) (UHPLC system, vanquish; MS/MS system, TSQ Altis Plus triple quadrupole; Thermo Fisher Scientific, Waltham, MA, USA). A total of 73 free amino acids were tested following the instrumental protocol with the analysis method developed in the lab. All amino acids are listed in Table S2.

Sample extraction. Tomato samples were ground in a knife mill Grindomix 200 (Retsch, Haan, Germany). The powder was transferred to a 5 mL centrifuge tube, and then a total of 10 mg of powder was weighed into a 1.5 mL centrifuge tube. Then, 1 mL pre-cooled mass spectrometry water on ice was added to make a 10 mg/mL extraction solution and swirled until thoroughly mixed. An accurately weighed aliquot of the homogenized sample was extracted with pre-cooled ultrapure water at a sample-to-solvent ratio of 1:5 (w/v). Extraction was conducted by ice-bath ultrasonication for 15 min, followed by centrifugation at 12,700 rpm for 10 min at 4 °C. The supernatant was collected, then filtered through a 0.22 μm membrane and subsequently diluted 100- and 1000-fold prior to UHPLC-MS/MS analysis.

UHPLC-MS conditions. Chromatographic separation was performed on an Acquity UPLC BEH Amide column (2.1 × 100 mm, 1.7 μm) operated at 40 °C. The mobile phases consisted of 0.1% formic acid in water containing 10 mM ammonium formate (A) and 0.1% formic acid in acetonitrile (B). The gradient elution was as follows: 0–3 min, 90–80% B; 3–4 min, 80–60% B; 4–5 min, 60–50% B; 5–8 min, 50% B; 8–8.1 min, 50–40% B; 8.1–11 min, 40% B; 11–12 min, 40–90% B; 12–15 min, 90% B. Flow rate was 0.3 mL/min and injection volume was 2 μL.

The MS with a heated electrospray ionization (H-ESI) source was operated in positive and negative ion modes. The spray voltage was 4000 V and 3000 V in positive and negative modes. The flow rates of sheath gas and auxiliary gas were 35 Arb and 10 Arb. The ion transfer tube temperature was 350 °C, and the vaporizer temperature was 350 °C. Analytes were detected via multiple reaction monitoring (MRM), and the detailed MRM parameters for the analytes are shown in Table S2.

2.7. Determination of Soluble Solid Content (SSC)

SSC was measured with the refractometer PAL-1 (ATAGO, Tokyo, Japan) according to the manufacturer’s protocol. Sample preparation: The sample was ensured to be a homogeneous liquid. Calibration (Zero Setting): A few drops of distilled water were placed on the prism, the “ZERO” button was pressed to calibrate, and then the prism was cleaned thoroughly. Measurement: The daylighting plate was lifted to expose the main prism. Placed 2–3 drops (approximately 0.3 mL) of the sample liquid onto the prism surface. The daylighting plate was closed to ensure the sample spreads evenly across the prism. The “START” button was pressed. The measurement result (Brix value) and the sample temperature were displayed on the screen within approximately 3 s. Cleaning: After measurement, the prism was cleaned with water immediately.

2.8. Statistical Analysis

Experiments were performed with at least three independent replicates, and the data were presented as mean ± SD. Statistical significance between groups was determined using an unpaired Student’s t-test, with p < 0.05 considered significant. One-way ANOVA and Tukey’s test were used to compare multiple edited lines. All statistical analyses were conducted using Microsoft Excel (Microsoft, Redmond, WA, USA) and GraphPad Prism (version 8.4.3).

3. Results

3.1. CRISPR/Cas9-Mediated Mutation of SlGAD3 in Tomatoes

A 20 bp sequence in the exon 6 of the SlGAD3 (Solyc01g005000) was selected as the target site (Figure 1A). The CRISPR/Cas9 construct was introduced into WEIMEI T102 via Agrobacterium-mediated transformation (Figure 1B). Target site sequencing analysis of the T0 regeneration lines revealed that mutations occurred upstream of the PAM. Through analysis with the online DECODR software, we identified 13 lines that carried a mutation in the target site, including different mutation patterns. For example, 1 bp insertion was found in lines E5, E13, and E36, 1 bp insertion and 77 bp deletion in line E12, 2 bp deletion in line E46, and other complex mutations (Table 1). In order to facilitate the segregation of transgene-free lines, lines E12, E13, and E46 were chosen for advancing to the next generation.

Table 1. Target mutation in the regeneration T0 lines.

T0 Lines	Target Mutation
E5	95% + 1 bp, 5%WT
E10	39.8%WT, 25.5% + 1 bp, 15.9% − 16 bp, 18.8% − 77 bp
E11	24.8%WT, 42.4% − 11 bp, 32.8% − 70 bp
E12	52.5%WT, 28.2% + 1 bp, 19.2% − 77 bp
E13	100% + 1 bp
E17	30.9% + 1 bp, 28.8% − 1 bp, 27.8% − 4 bp, 12.6%WT
E19	8.7%WT, 17.7% + 15 bp, 20.6% − 11 bp, 16% − 12 bp, 37.1% − 70 bp
E34	55.5% + 1 bp, 20% − 3 bp, 24.4%WT
E36	86.5% + 1 bp, 13.5%WT
E39	46.3% + 1 bp, 53.7% − 5 bp
E41	49.7% + 1 bp, 24.1% − 1 bp, 26.2% − 2 bp
E46	100% − 2 bp
E48	62.5%WT, 17.3% + 1 bp, 20.2% − 1 bp

T1-generation homozygous lines were identified through amplification of the target site (Table S1). Homozygous mutation lines were screened for transgene-free analysis (Figure S1). Among these lines, five lines were 1 bp insertion (E12-7, E12-37, E12-54, E13-2, E13-13), five lines were 2 bp deletion (E46-1, E46-3, E46-4, E46-5, E46-7), three lines were 141 bp deletion and 45 bp insertion (E12-4, E12-33, E12-55), and one line was 91 bp deletion and 21 bp insertion (E12-24) (Figure 1C and Figure S2).

T2 generation plants, E13-2, E13-13, E46-3, and E46-4, were further tested for the construct random integration using the Cas9 primers. No corresponding fragments were amplified in six individual plants for the four gene-edited lines (Figure 2). These results confirmed that these lines were transgene-free. One T102 and three edited plants from T2 progeny were shown without a significant difference (Figure 3).

Figure 2. Transgene detection by PCR with Cas9-specific primers. Lane M, DNA ladder. Lane 1, wild-type T102. Lane 2, plasmid pDN077 DNA. Lane 3~8, six individual plants from four gene-edited lines, E13-2, E13-13, E46-3, and E46-4, respectively.

Figure 3. The 10-week-old plants of T2 lines. One T102 and three 1 bp insertion edited lines are shown for an example.

In the four gene-edited lines, a premature stop codon was introduced following the deletion of a self-inhibitory domain. This genetic alteration resulted in a truncated protein product. Specifically, the edited sequences encoded proteins that were 36 and 37 amino acids shorter than the original wild-type protein, respectively. Furthermore, the editing events led to subtle changes at the C-terminus of the truncated protein: two amino acids “Gly-Gly” were added in the case of the 1 bp insertion event, while “Asn-Gly-Gly” was added in the case of the 2 bp deletion event (Figure 4).

Figure 4. The deduced amino acid sequence alignment of SlGAD3 for the WT and gene-edited lines. WT, GAD3 amino acid sequence from wild type T102. I1, GAD3 amino acid sequence from 1 bp insertion lines, E13-2 and E13-13. D2, GAD3 amino acid sequence from 2 bp deletion lines, E13-2 and E13-13.

3.2. Stable Inheritance of Edited Gene

To test if the mutated target sites can be inherited by multiple generations, we employed the PCR and sequencing of the target region using gene-specific primers. We sequenced the target of leaf tissue from T0 to T2 for the gene-edited line E13. All the results showed 1 bp insertion compared to WT T102. We also tested the target sites of a 2 bp deletion in the gene-edited line E46. The sequences showed an identical mutation (Figure S3). All of these results confirmed that the mutation sites were stably inherited by the next generation.

3.3. No Off-Target Site Edits Detected

To assess the potential off-target effects of the CRISPR/Cas9 system targeting SlGAD3, we evaluated nine predicted off-target sites across chromosomes 1, 2, 3, 5, 6, 8, and 9 using the CRISPR-P algorithm. These sites were selected based on sequence homology to the designed sgRNA, with up to four mismatches allowed (Table S3). Specific primers were designed to amplify genomic regions encompassing each candidate off-target locus (Table S4). Amplicons spanning all nine loci were subjected to high-fidelity PCR amplification followed by Sanger sequencing.

Sequence alignment between the amplified products and the reference genome revealed no detectable mutations at any of the examined loci, demonstrating complete sequence identity under stringent alignment criteria (Figure S4). Notably, even the most homologous site on chromosomes 6 and 9 (with three mismatches relative to the sgRNA) exhibited no unintended edits. These results conclusively indicated the absence of off-target activity within the putative regions. These findings collectively demonstrated that the high specificity of the CRISPR/Cas9 system is guided by the sgRNA. The lack of detectable off-target mutations across multiple chromosomal loci underscored the precision of the designed targeting strategy, supporting its reliability for gene editing applications in tomato.

3.4. Content of GABA and Other Free Amino Acids

GABA levels in red ripe fruits of wild-type (WT) and mutant lines from T1 to T3 were tested (Tables S5–S7).

For the T1 homozygous lines, eleven lines were screened for the determination of GABA content. The average GABA contents in RED-stage fruits of E12-54, E12-17, E13-2, E12-37, E13-13 lines, with 1 bp insertion, were 0.58 ± 0.01 mg/g FW, 0.68 ± 0.02 mg/g FW, 0.91 ± 0.02 mg/g FW, 1.62 ± 0.22 mg/g FW, 1.66 ± 0.07 mg/g FW, respectively. For those lines with 2 bp deletion, E46-1, E46-3, E46-4, E46-5 and E46-7 lines, they are 0.34 ± 0.03 mg/g, 0.65 ± 0.02 mg/g FW, 0.66 ± 0.06 mg/g FW, 0.26 ± 0.02 mg/g FW, and 0.21 ± 0.01 mg/g FW, respectively, while the GABA of T102 was 0.30 ± 0.01 mg/g FW (Figure 5a).

Figure 5. Content of GABA in red fruit of gene-edited mutants. (a) GABA level of the T1 lines and T102. (b) GABA level of the T2 lines and T102. (c) GABA level of the T3 line and T102. Different lower-case letters indicate statistically significant differences based on an ANOVA followed by Tukey’s test, p < 0.05; **, p < 0.01.

Large-fragment deletion mutants, E12-24, E12-4, E12-55, and E12-33, showed significantly reduced GABA level, which is 0.01 ± 0.00 mg/g FW, 0.02 ± 0.00 mg/g FW, 0.02 ± 0.00 mg/g FW, and 0.07 ± 0.01 mg/g FW, compared to WT (Figure 5a). These results indicated that small mutations in the C-terminal auto-inhibitory domain of GAD enhanced GABA accumulation, while large-fragment deletions of SlGAD3, which could disrupt the gene, failed to promote GABA biosynthesis.

Four lines with increasing GABA content, E13-2, E13-13, E46-3, and E46-4, were chosen for further analysis. In the T2 generation, mutant lines consistently maintained elevated GABA levels compared to WT. The average content was 1.64 ± 0.34 mg/g FW, 1.90 ± 0.45 mg/g FW, 1.02 ± 0.47 mg/g FW, and 0.67 ± 0.29 mg/g FW, but the wild-type T102 was about 0.45 ± 0.11 mg/g FW (Figure 5b). GABA contents of E13-2 and E13-13 lines were significantly higher than T102, but E46-3 and E46-4 lines showed no significant difference from T102.

Based on its highest GABA production, the T3 generation of the E13-13 line was chosen for the field test. Similarly to the T1 and T2, E13-13 T3 lines accumulated higher GABA (1.93 ± 0.41 mg/g FW), which was substantially greater than the value in the control line T102 (0.81 ± 0.37 mg/g FW) (Figure 5c). Together, GABA content analysis demonstrated the stable inheritance of the gene-edited mutation.

Previous studies showed that the extremely high accumulation of GABA in fruits could affect the content of other free amino acids. Therefore, we measured a total of 73 free amino acid levels in the red fruit stage of the mutant lines by UHPLC-MS/MS. Among these 73 amino acids, 40 were detected (Tables S6 and S7). We manually divided these amino acids into two groups, one group including 14 amino acids which were more than 50 ug/g FW, and another one including 26 amino acids which were less than 50 ug/g FW (Figure 6).

Figure 6. Content of free amino acid in the gene-edited mutants. (a,b) Free amino acid content of T2 red fruit harvested from greenhouse. (c,d) Free amino acid content of T3 red fruit harvested from field test.

In the T2 generation, GABA(AA-47) levels of three gene-edited lines were significantly higher than T102. Other amino acids did not show a significant difference between T102 and gene-edited lines, although there were some changes in content compared to T102. AA-59 (L-Glutamic acid), AA-60 (L-Glutamine), and AA-75 (L-Tryptophan) accumulated more in T102 than in the gene-edited lines (Figure 6a,b).

For the field test of T102E13-13, the level of GABA was significantly increased compared to T102 (Figure 6c,d). These results suggested that the target mutation regulated the GABA level with a minimal impact on other free amino acids.

3.5. Analysis of the Plant Height, the Fruit Weight, and SSC

Plant heights of eighteen plants of T102 and ten plants of E13-13 from the field test were measured at 1 month after transplanting (Table S8). The average plant height was 23.39 ± 3.11 cm and 23.10 ± 3.00 cm, for T102 and E13-13, respectively. There was no significant difference observed between them (Figure S5A).

A total of 33 fruits of T102 and 72 fruits of E13-13 were weighed (Table S9). The average fruit weight was 7.82 ± 2.83 g and 7.35 ± 2.53 g for the control line T102 and the gene-edited line E13-13, respectively. There was no significant difference found (Figure S5B).

SSC is a key determinant of tomato taste. The SSC between the T102 and E13-13 lines was measured using a PAL-1 pocket refractometer (Table S10). The average Brix was 6.12 ± 0.63% for T102 and 6.36 ± 0.54% for E13-13; the gene-edited line was significantly higher than T102, p = 0.0476 (Figure S5C).

4. Discussion

GABA is available from many vegetables and fruits, including tomatoes. Currently, studies show different ways to increase GABA content in tomatoes, and the targeted mutagenesis seems to be sufficient to increase GABA accumulation and promote health [17,18,19,20]. In this study, we also created the SlGAD3 gene-edited lines with increased GABA in the red ripe fruits, from a dwarf cherry tomato developed for pot-grown by home gardeners. The new high-GABA lines could be a new germplasm for breeding different types of dwarf or normal tomato varieties. The average GABA level of the E13-13 line in this study reached 1.9 mg/g FW in red fruit. Compared to the available data on gene-edited tomatoes, the GABA-3 mutant increased to 0.8988 mg/g FW [18], but the highest lines reached over 6 mg/g FW [20]. The study also confirmed that excessive accumulation of GABA inhibited the development of leaf and flower/fruit settings [18].

We found that there was higher GABA accumulation in the short sequence insertion or deletion at the C terminal of SlGAD3, which is the autoinhibitory domain. GABA could not be detected if there was a large deletion in the SlGAD3 and the protein structure was broken.

Samples from the field test showed higher Methionine Sulfoxide (AA-33) accumulation. It was significant in plant stress tolerance, particularly in response to abiotic stresses like salinity and drought [22,23]. The increased Methionine Sulfoxide could be the result of the continuing high temperatures in July this year. Methionine sulfoxide reductase (MSR) is an antioxidant enzyme that plays important roles in stress resistance by scavenging free radicals and repairing oxidized proteins [22,23,24]. Data available from higher plants revealed that MSRs fulfilled an essential physiological function during environmental constraints through a role in protein repair and in protection against oxidative damage [25]. Study also showed that overexpression of SlGAD1 increased GABA levels and decreased reactive oxygen species accumulation under saline-alkali stress [26]. More experiments are needed to be done to explain the mechanism between higher GABA and Methionine Sulfoxide accumulation in the open field environment.

WEIMEI T102 is a dwarf cherry tomato cultivar. It has been cultivated in greenhouses in pots and offered to customers as both a consumable and ornamental plant for office or house window settings, adding a touch of natural food and beauty. The edited lines further enhanced the value of this type of tomato. More physiological tests, RNA level, protein level, and nutrition analysis will be evaluated.

The high-GABA tomato developed by the University of Tsukuba in Japan was successfully approved for marketing. The development of a high-GABA variety improves the nutritional value of tomatoes. International Seed Federation stated that it is an important step in the implementation of the Japanese policy on genome editing, and it provides opportunities for the seed sector to continue its efforts on plant breeding innovation to contribute to sustainable food systems (www.worldseed.org, accessed on 11 October 2025). Gene editing and all the new breeding techniques contribute to sustainable agriculture and food security. The improved products could be beneficial for human health. In this study, the high-GABA tomato has its advantages and application scenarios and is expected to be available in the market in the future, bringing a new selection to consumers. Owing to their superior traits (e.g., higher nutritional value, ornamental plant for house and office window settings), gene-edited tomatoes are expected to yield promising economic benefits.

5. Conclusions

We successfully generated tomato lines with significantly enhanced GABA accumulation in this study. Multi-generation evaluations confirmed the stable inheritance of the intended genetic modifications and the associated high-GABA trait at the molecular level. The increased GABA content in tomato fruits is anticipated to offer potential health benefits, particularly due to its documented hypotensive effects. Furthermore, the dwarf growth habit and compact plant architecture of this line make it suitable for use as an ornamental plant in indoor environments, such as house and office window settings, potentially providing psychological well-being through visual interaction with the greenery.