Development of elite germplasm with high ɣ-aminobutyric acid (GABA) content by targeted mutation of ZmGAD3 and ZmGABA-TP1 in maize

Abstract

γ-Aminobutyric acid (GABA) plays a crucial role in human health. Although tomatoes with enhanced GABA levels have been engineered through single-gene editing, GABA-fortified maize has not previously been developed. In this study, CRISPR/Cas9 editing of the biosynthetic gene ZmGAD3, the degradation pathway gene ZmGABA-TP1-1, or both genes simultaneously, increased the GABA content of maize kernels 0.59–2.15 fold, 1.28–2.99 fold, or 2.63–4.87 fold, respectively. We thus developed an elite germplasm resource with significantly fortified GABA content and demonstrated that coordinated editing of ZmGAD3 and ZmGABA-TP1-1 could effectively enhance GABA accumulation. This study reveals important insight into the functional regulation of GABA content through maize metabolic engineering.

Dear editor,

Maize (Zea mays L.) is one of the most abundant and versatile crops produced worldwide, contributing significantly to human nutrition, food security, and industrial applications. Maize kernels are a rich source of essential nutrients, including carbohydrates (e.g., starch and dietary fiber), oil, and micronutrients (e.g., vitamins, minerals, and phytonutrients) [1]. Although awareness of the critical importance of micronutrients for human health is increasing, over 800 million people worldwide are currently malnourished, with about 2 billion affected by deficiencies in one or more micronutrients [2]. Therefore, the majority of the global population requires better access to nutrient-rich food. Biofortification is a sustainable strategy for addressing micronutrient deficiencies, and micronutrient-biofortified maize has been shown to improve the nutrition and health status of people in certain regions of Africa.

γ-Aminobutyric acid (GABA) is a four-carbon non-proteinogenic amino acid found in bacteria, fungi, animals, and plants. Because GABA exists mainly as a free monomer in maize and does not form macromolecular complexes with other compounds, its native structure and function remain intact [3]. As the principal inhibitory neurotransmitter in the central nervous system, GABA is effective in preventing certain lifestyle-related diseases such as hypertension, depression, insomnia, and diabetes, as well as regulating the endocrine system [4]. Although GABA is naturally present in some foods, its concentration is typically insufficient to provide protection against such diseases. Thus, biofortification of foods with GABA is necessary to obtain its health-promoting benefits.

In vascular plants, GABA is biosynthesized through the “GABA shunt”, a metabolic bypass of the tricarboxylic acid (TCA) cycle. In this pathway, glutamate is irreversibly catabolized by glutamate decarboxylase (GAD) to form GABA, which is then reversibly converted to succinic semialdehyde (SSA) by GABA transaminase (GABA-T). SSA is irreversibly oxidized by succinate semialdehyde dehydrogenase (SSADH) to form succinate, which ultimately participates in the TCA cycle (Fig. 1A). GABA-rich elite germplasms have been developed through metabolic engineering of SlGAD, SlGABA-T, and SlSSADH in tomato [5,6], but similar improved germplasms have not been reported for maize. This study provides the first evidence that mutations in ZmGAD3 and ZmGABA-TP1 can individually enhance GABA accumulation and have a synergistic effect on GABA levels in maize.

Fig. 1.

Precise editing of both ZmGAD3 and ZmGABA-TP1 increased ɣ-aminobutyric acid (GABA) content in maize. A GABA metabolism and biosynthetic pathway in maize. B Homozygous mutant genotypes of ZmGAD3 in the T₂ generation. C GABA content in the ZmGAD3 mutants. D Plant height and hundred-kernel weight in the ZmGAD3 mutants. E Homozygous mutant genotypes of ZmGABA-TP1-1 in the T₂ generation. F GABA content in the ZmGABA-TP1-1 mutants. G Plant height and hundred-kernel weight in the ZmGABA-TP1-1 mutants. H Homozygous mutant genotypes of both ZmGABA-TP1-1 and ZmGABA-TP1-2 in the T₂ generation. I GABA content in the ZmGABA-TP1-1 and ZmGABA-TP1-2 double mutants. J Plant height and hundred-kernel weight in the ZmGABA-TP1-1 and ZmGABA-TP1-2 double mutants. K Homozygous mutant genotypes of both ZmGAD3 and ZmGABA-TP1-1 in the T₂ generation. L GABA content in the ZmGAD3 and ZmGABA-TP1-1 double mutants. M Plant height and hundred-kernel weight in the ZmGAD3 and ZmGABA-TP1-1 double mutants. The data were analyzed by ANOVA and Duncan's test using SPSS 23.0. All results are presented as mean ± standard deviation (SD).

GAD mediates the dominant GABA biosynthetic pathway, although GABA can also be generated through polyamine-dependent metabolism. The calmodulin (CaM)-binding domain (CaMBD) binds the C-terminal autoinhibitory domain (30–50 aa) of GAD3, thereby suppressing GAD3 activity [7]. Editing the GAD3 inhibitory domain increased GABA accumulation by 0.3–1.3 fold in rice and 2–3 fold in tomato, prompting us to hypothesize that editing maize GAD3 (ZmGAD3) would enhance the GABA content of maize kernels. Through homology analysis, we identified the ZmGAD3 gene and its corresponding protein sequence in maize B104 (Table S1). Using CRISPR/Cas9 editing to target ZmGAD3, we obtained 9 transgenic-positive lines and 7 gene-editing events in the T₀ generation and 4 homozygous genotypes in the T₂ generation (Fig. 1B). Analysis of field-grown homozygous T₂ lines at 20 days after pollination (DAP) (Fig. S1A) revealed that their kernel GABA content was 0.59-fold (GAD3-1), 1.39-fold (GAD3-2), 1.82-fold (GAD3-3), and 2.15-fold (GAD3-4) higher than that of wild-type plants (Fig. 1C). This result confirmed the successful enhancement of GABA accumulation. Although overall plant architecture was unaffected in the gene-edited lines, plant height decreased by 2.13%–6.16% and hundred-grain weight by 2.99%–5.11% (Fig. 1D and Fig. S1B and C). These results indicated that GABA negatively regulated plant growth and yield, consistent with its reported effects on plant height and fruit set in tomato [8]. Further analysis showed that glutamate content was approximately 11.51%–22.44% lower in the edited lines (Fig. S2). Analysis of protein sequences in the different mutants revealed that their mutations abolished the binding site for the inhibitory protein (Table S2). Thus, removal of inhibition at the binding site enhanced the activity of ZmGAD3 in converting glutamate to GABA.

The function of GABA-T has been confirmed in potato (Solanum tuberosum), tobacco (Nicotiana tabacum), and tomato (Solanum lycopersicum). After RNAi-mediated suppression of SlGABA-TP1, SlGABA-TP2, and SlGABA-TP3 in tomato, only SlGABA-TP1^RNAi significantly increased GABA content (1.5–10.7 fold) [9]. However, the regulatory role of ZmGABA-TP1 in maize has not previously been characterized. We identified maize homologs of SlGABA-TP1 with high homology to the SlGABA-TP1 protein and designated them ZmGABA-TP1-1 and ZmGABA-TP1-2 (Table S1). We targeted ZmGABA-TP1-1 and ZmGABA-TP1-2 using CRISPR/Cas9 gene editing (Fig. 1E) and obtained three homozygous ZmGABA-TP1-1 mutant lines in the T₂ generation. Analysis of kernels at 20 DAP revealed that the GABA content was 1.28-fold, 1.91-fold, and 2.99-fold higher in the mutant lines than in the wild type (Fig. 1F and Fig. S3A). However, plant height was significantly reduced by 16.52%–20.30% and hundred-kernel weight by 7.11%–9.83% (Fig. 1G and Fig. S3B and C). This phenotype was similar to that reported in tomato, in which plant height was reduced by over 45% [9]. These results suggest that ZmGABA-TP1-1 knockout promotes GABA accumulation.

We also obtained three homozygous edited lines for ZmGABA-TP1-2 (Fig. S4A). No significant increase in GABA content was observed in these mutants at 20 DAP (Fig. S4B). Furthermore, no significant effects were observed on plant height and yield (Fig. S4C–E). It is possible that we failed to identify effective editing sites within ZmGABA-TP1-2 for enhancement of GABA content, and saturation mutagenesis of ZmGABA-TP1-2 could be performed to identify such functional sites [10].

We obtained two homozygous double-mutant lines with simultaneous mutations in ZmGABA-TP1-1 and ZmGABA-TP1-2 (Fig. 1H). These lines showed 1.92-fold and 2.51-fold increases in GABA content (Fig. 1I and Fig. S5A), together with reductions of 10%–20% in plant height and 5.79%–7.71% in hundred-kernel weight (Fig. 1J and Fig. S5B and C). These phenotypes are essentially similar to those observed in the ZmGABA-TP1-1 single mutant. However, this does not demonstrate that ZmGABA-TP1-2 synergistically enhances GABA content. Saturation mutagenesis of ZmGABA-TP1-2 in the ZmGABA-TP1-1 mutant background could be performed to investigate potential cooperativity between ZmGABA-TP1-1 and ZmGABA-TP1-2.

Given the established functional relationship between ZmGABA-TP1-1 and ZmGABA-TP1-2, we next asked whether ZmGAD3 and ZmGABA-TP1-1 could synergistically enhance GABA content. To this end, we simultaneously edited both genes and obtained four homozygous double-mutant genotypes (ZmGAD3/ZmGABA-TP1-1) in the T₂ generation (Fig. 1K). These lines exhibited significant increases in GABA content relative to the wild type (2.63-, 3.12-, 3.55-, and 4.87-fold; P < 0.05) (Fig. 1L and Fig. S6A). However, unfavorable phenotypic traits were also observed: plant height decreased by 19.88%–55.74% and hundred-kernel weight by 9.56%–49.57% (Fig. 1M and Fig. S6B and C). This represented the first demonstration that ZmGAD3 and ZmGABA-TP1-1 act synergistically to enhance GABA content. These findings demonstrated that repression of ZmGAD3 was abolished and, simultaneously, the degradative function of ZmGABA-TP1-1 was suppressed. Determining an appropriate regulatory balance between ZmGAD3 and ZmGABA-TP1-1 is challenging, as an inappropriate balance could reduce plant height and hundred-kernel weight by 50%. To address this issue, we developed a set of germplasms that exhibited a range of GABA concentrations and yield performances. This approach enabled us to identify a more optimal balance. The high-GABA germplasm developed here shows excellent promise for application in fresh corn breeding.

Clinical recommendations suggest that a daily intake of 100 mg GABA can significantly improve relevant symptoms, and the novel maize germplasm developed in this study contains at least ∼50 mg of GABA per ear [11]. Because future hybrid varieties are expected to accumulate even higher GABA levels, the consumption of just two ears per day would be sufficient to meet these nutritional and therapeutic requirements. Steaming and microwaving are widely recognized by consumers as the best methods for preserving the taste of fresh corn. These cooking methods also preserve 70%–90% of its GABA content (Table S3), which remains within the range (50–300 mg) recommended by the Natural and Non-prescription Health Products Directorate (NNHPD). Because storage conditions have a significant effect on GABA retention, we recommend the following storage guidelines to preserve both the soft, waxy texture and the GABA content of fresh corn: (1) refrigeration at 4 °C for consumption within 6 days; (2) storage at room temperature for consumption within 3 days; and (3) freezing at −20 °C for long-term storage up to 6 months (Table S4).

In summary, independent editing of ZmGAD3 and ZmGABA-TP1-1 enhanced GABA content, and editing of both genes had a synergistic effect. The GABA-enriched maize germplasm developed by editing these two genes provides a novel breeding strategy for nutritional enhancement of fresh corn. From the perspectives of both breeding and dietary intake, the kernel (comprising the embryo and endosperm) is the primary edible part of maize. Consequently, the total GABA content in the kernel not only facilitates hybrid breeding programs but also provides precise dietary guidance for the development of biofortified maize. Although these edits introduced manageable negative effects on plant height and yield, these phenotypic alterations may provide valuable insight for the identification of dwarfing-associated genes.

1. Materials and methods

1.1. Plant material and Agrobacterium strain

The B104 seeds were stored at the Maize Research Institute, Beijing Academy of Agriculture and Forestry Sciences, Beijing, China. Agrobacterium tumefaciens strain LBA4404 was used for embryo infection. The vector was based on pCambia2300 and included the pmi gene as a selectable marker driven by the CaMV35S promoter and the cas9 gene driven by the ZmUbi promoter.

1.2. Plant growth conditions

Wild-type and edited plants were grown in the greenhouse or in Fangshan and Hainan Biological Breeding Base, Institute of Biology, Beijing Academy of Agriculture and Forestry Sciences. The greenhouse growth conditions included natural light supplemented with LED lighting (16-h light/8-h dark photoperiod), with temperatures of 28 °C during the day and 22 °C at night. For the field experiment, the wild-type and edited plants were grown in the field at a spacing of 25 × 60 cm.

1.3. Maize transformation

Ears were harvested 10–12 DAP and sterilized for 20 min in a 50% dilution (v/v) of sodium hypochlorite. Isolated embryos were collected in infection medium. The infection medium was removed, and the embryos were infected using infection medium supplemented with 100 μM acetosyringone for 5 min. The embryos were placed on co-culture medium for 2 days in the dark and then transferred to recovery medium for 7 days in the dark. To induce callus formation, the embryos were cultured on the first selection medium supplemented with 0.6 g/L mannose for 14 days in the dark. The resistant calli were transferred to the second selection medium supplemented with 1.5 g/L mannose for 14 days in the dark. To obtain regenerated plants, resistant calli were cultured on the first regeneration medium supplemented with 0.2 mg/mL NAA for approximately 14 days under a 16-h light/8-h dark cycle. The small plantlets were transferred to the second regeneration medium under the same light conditions for 14 days. To induce root growth, the regenerated plants were cultured in rooting medium for 21 days [12].

1.4. Analysis of transgenic plants

Genomic DNA was extracted from leaves of regenerated transgenic plants using the SDS method. The detected fragment was amplified using pmi primers (pmi-F: 5′-CGTGCAGAACTACGCCTGGGG-3′, pmi-R: 5′-GTCGGGCTGCTGCAGGAAGTG-3′) (Table S5).

1.5. GABA and glutamate analysis

The endosperm contents of 20 kernels were extracted at 20 DAP, and the concentrations of GABA and glutamate were analyzed using corresponding assay kits.

1.6. Statistical analysis

The transformation data were analyzed by ANOVA with Tukey's test (∗P < 0.05; ∗∗P < 0.01; ∗∗∗P < 0.001) using GraphPad Prism 10. The phenotype data were analyzed by ANOVA with Duncan's test using SPSS 23.0. All results are presented as mean ± standard deviation (SD).

CRediT authorship contribution statement

Yao Wang: Writing – review & editing, Writing – original draft, Visualization, Supervision, Investigation, Funding acquisition, Data curation. Lu Zhang: Writing – review & editing, Writing – original draft, Validation, Investigation. Mengyuan Liu: Writing – review & editing, Supervision, Methodology, Investigation. Xiaoxu Li: Methodology, Investigation. Xinxiang Liu: Supervision, Resources. Ziheng Song: Methodology, Investigation. Yue Fu: Investigation. Ya Liu: Writing – review & editing, Supervision, Resources, Project administration, Funding acquisition. Ronghuan Wang: Writing – review & editing, Supervision, Resources. Jiuran Zhao: Writing – review & editing, Validation, Supervision, Resources, Project administration, Funding acquisition, Conceptualization.

Declaration of competing interest

The authors declare that they have no known competing financial interests or personal relationships that could have appeared to influence the work reported in this paper.

Appendix A. Supplementary data

The following is the Supplementary data to this article:

Data availability

The data that support the findings of this study are available in the Supporting Information.