Transcriptomic Insights into GABA Accumulation in Tomato via CRISPR/Cas9-Based Editing of SlGAD2 and SlGAD3

Abstract

Background: γ-Aminobutyric acid (GABA) is a non-proteinogenic amino acid with key roles in plant metabolism, stress responses, and fruit nutritional quality. In tomato (Solanum lycopersicum), GABA levels are dynamically regulated during fruit development but decline in the late ripening stages.

Methods: To enhance GABA accumulation, we used CRISPR/Cas9 to edit the calmodulin-binding domain (CaMBD) of SlGAD2 and SlGAD3, which encode glutamate decarboxylases (GADs). The resulting truncated enzymes were expected to be constitutively active. We quantified GABA content in leaves and fruits and performed transcriptomic analysis on edited lines at the BR+7 fruit stage.

Results: CaMBD truncation significantly increased GABA levels in both leaves and fruits. In gad2 sg1 lines, GABA levels increased by 3.5-fold in leaves and 3.2-fold in BR+10 fruits; in gad3 sg3 lines, increases of 2.8- and 2.5-fold were observed, respectively. RNA-seq analysis identified 1383 DEGs in gad2 #1-5 and 808 DEGs in gad3 #3-8, with 434 DEGs shared across both lines. These shared DEGs showed upregulation of GAD, GABA-T, and SSADH, and downregulation of stress-responsive transcription factors including WRKY46, ERF, and NAC. Notably, total free amino acid content and fruit morphology remained unchanged despite elevated GABA.

Conclusions: CRISPR/Cas9-mediated editing of the CaMBD in SlGAD genes selectively enhances GABA biosynthesis in tomato without adverse effects on development or fruit quality. These lines offer a useful platform for GABA-centered metabolic engineering and provide insights into GABA's role in transcriptional regulation during ripening.

Keywords: CRISPR/Cas9; GABA; glutamate decarboxylase (GAD); metabolic engineering; tomato fruit; transcriptome.

Figure 1

CRISPR/Cas9-mediated editing of SlGAD2 and SlGAD3 in tomato. (A) Schematic representation of the gene structures of SlGAD2 and SlGAD3, showing the positions of the sgRNA target sites (blue arrows) near the 3′ end of the coding sequences. Exons are indicated by boxes, with dark gray representing coding sequences (CDS) and white indicating untranslated regions (UTRs). The target sequences for each sgRNA are shown in blue. (B) Deep sequencing analysis of the CRISPR target regions in transgenic plants. Blue text indicates sgRNA sequences; hyphen (-) represent deleted nucleotides; red letters indicate inserted bases. (C) Translated amino acid sequences of the corresponding mutant alleles. Asterisks (*) denote premature stop codons, and italicized residues indicate altered amino acid sequences resulting from frameshift mutations.

Figure 2

Phenotypic analysis of CRISPR/Cas9-edited SlGAD2 and SlGAD3 tomato lines. (A) Representative images of WT and SlGAD-edited lines at the mature vegetative stage (upper panels) and of harvested fruits (lower panels). Scale bars = 2.5 cm. (B) Quantitative analysis of morphological traits, including plant height, leaf width and length, fruit width, length, and fresh weight. Bars represent means ± SD (n = 6). Asterisks indicate significant differences from the wild type (* p < 0.05, ** p < 0.01).

Figure 3

Quantification of GABA and total free amino acids in SlGAD edited lines. (A) GABA content in leaf and fruit tissues from wild-type and SlGAD-edited lines. (B) Total free amino acid levels in the same samples. Fruits were harvested at MG and 10 days after breaker stage. Bars represent means ± SD (n = 3). Asterisks indicate significant differences from the wild type (* p < 0.05, ** p < 0.01, *** p < 0.001).

Figure 4

Shared transcriptomic response in SlGAD2 and SlGAD3 edited tomato fruits at the BR+7 stage. (A) Principal component analysis (PCA) of RNA-seq data derived from pericarp tissues of red ripe fruits at the BR+7 stage. Samples from gad2 #1–5, gad3 #3–8, and WT cluster distinctly, indicating genotype-dependent transcriptional profiles. (B) Venn diagrams showing the number of differentially expressed genes (DEGs) that are commonly upregulated (top) or downregulated (bottom) in both gad2 #1–5 and gad3 #3–8 lines compared to WT. DEGs were identified using DESeq2 with thresholds of |log₂FC| ≥ 1 and adjusted p-value < 0.05. (C) Heatmap of the 434 shared DEGs (192 upregulated, 242 downregulated), visualized via hierarchical clustering. Red indicates relatively upregulated genes, while green indicates downregulation relative to wild type. Samples cluster by genotype, reflecting consistent expression patterns across biological replicates. (D) Volcano plot depicting the distribution of the 434 shared DEGs by log₂ fold change and statistical significance. Genes meeting the cutoff criteria (adjusted p-value < 0.05 and |log₂FC| ≥ 1) are highlighted in red (upregulated) and green (downregulated). Selected DEGs with strong statistical significance and high expression changes are labeled by gene ID for reference.

Figure 5

Validation of GABA Pathway and Stress-Responsive Genes by qRT-PCR. (A) Expression levels of GABA metabolism-related genes including (B) Expression of stress-responsive transcription factor genes WRKY46, NAC, and ERF. qRT-PCR analysis was performed using fruit samples harvested at the BR+7 stage. Bars represent means ± SD (n = 3). Asterisks indicate significant differences from the wild type (* p < 0.05, ** p < 0.01).