Skip to main content

Improve gamma-aminobutyric acid production in Corynebacterium glutamicum by optimizing the metabolic flux

Abstract

Gamma-aminobutyric acid is an important nonprotein amino acid and has been extensively applied in pharmaceuticals, livestock, food additives, and so on. It is important to develop Corynebacterium glutamicum strains that can efficiently produce gamma-aminobutyric acid from glucose. In this study, production of gamma-aminobutyric acid in C. glutamicum CGY700 was improved by construction of CO2 anaplerotic reaction and overexpression of citrate synthase. The co-expression of ppc encoding phosphoenolpyruvate carboxylase and gltA encoding citrate synthase was constructed and optimized in the chromosome to compensate carbon loss and conquer metabolic bottleneck. The expression of ppc and gltA were controlled by promoters Ptac and PtacM, and the optimal mode of PtacM-ppc-Ptac-gltA was determined. Simultaneously, the genes pknG encoding serine/threonine protein kinase G and ldh encoding l-lactate dehydrogenase were deleted, and glnA2 encoding glutamine synthase was overexpressed in the chromosome. The final strain CGY-PG-304 constructed in this study could produce 41.17 g/L gamma-aminobutyric acid in shake flask cultivation and 58.33 g/L gamma-aminobutyric acid via Fed-Batch fermentation with a yield of 0.30 g/g glucose. CGY-PG-304 was constructed by genome editing; therefore, it is stable and not necessary to add any antibiotics and inducer during fermentation.

Introduction

Gamma-aminobutyric acid (GABA) is a natural nonprotein amino acid existed in plants, animals and microorganism [1]. Plants can secrete GABA under certain conditions such as oxygen deficiency, thermal shock, mechanical injury, and salt stress against pathogenic fungal infection [2]. Some bacteria adjust intracellular pH in acid environment through GABA shunt [3], which could improve acid resistance and make the cells survive under low pH environment. GABA is an inhibitory neurotransmitter and plays an important role in neural development in mammals [4, 5]. GABA can be applied in livestock [6, 7], food additives [8, 9], pharmaceuticals and clinical research [10, 11]. GABA can also be used as chemical intermediate to synthesize renewable bio-based products [12, 13].

Currently, chemical synthetic GABA is forbidden in the commercial utilization of food and drinks [14], GABA production is mainly produced by enzymatic synthesis [15, 16]. Glutamic acid decarboxylase (GAD) catalyzes l-glutamic acid (l-Glu) to GABA using pyridoxal-5′-phosphate (PLP) as cofactor [17]. GADs with high enzymatic activity are mainly originated from E. coli and lactic acid bacteria. GABA can also be produced by whole cell catalysis, using l-Glu or monosodium glutamate (MSG) as substrate and cells of E. coli and some species of lactobacillus [1]. The engineered E. coli and lactobacillus can obtain high GAD activity and productivity by enzymatic synthesis or whole cell catalysis, however, addition of the abundant l-Glu and cofactor PLP results in high production costs.

In recent years, some researches have been done to produce GABA using microbial fermentation from cheap and renewable food-based materials such as glucose, starch [18], and xylose [19]. There are two pathways of GABA biosynthesis in microorganisms. Putrescine pathway is rarely used to synthesize GABA because of its complex metabolic regulation and extremely poisonous intermediates [20, 21]. The GABA shunt using the key enzyme GAD has been considered a potential pathway to synthesize GABA. Natural GAD usually functions only under acidic condition (pH 4.0–5.0) and would be inactive when pH is higher than 5.5 [22,23,24]. The proper pH for l-Glu synthesis is 6.8–7.5. To date, the highest GABA titer was 6.16 g/L by E. coli fermentation [25].

Corynebacterium glutamicum (C. glutamicum) has considerable capacity for l-Glu synthesis, so it should be an ideal candidate for GABA production [16]. Some researches on GABA production in C. glutamicum fermentation from glucose have been reported [26,27,28]. An industrial l-Glu production strain C. glutamicum G01 has been engineered by introducing L. plantarum-derived GAD and PLK gene, resulting in 70.6 g/L GABA after Fad-Batch fermentation [29]. GABA shunt construction in wild type C. glutamicum ATCC 13032 resulted in 38.6 g/L GABA [30]. However, GABA shunt construction in C. glutamicum ATCC 13032 was conducted using plasmid-dependent expression system, which needs adding antibiotics in media to maintain the plasmid, resulting in extra costs and contamination to the environment.

In this study, a global genomic expression strategy of GABA production in C. glutamicum was constructed from a GABA-producing strain CGY700, which was derived from C. glutamicum ATCC 13032 by introducing Lactobacillus brevis GAD in the chromosome [26]. The metabolic flux in C. glutamicum was enhanced by optimizing the co-expression of ppc encoding phosphoenolpyruvate carboxylase and gltA encoding citrate synthase. In addition, the effects of ldh, pknG, alaT, and glnA2 on GABA synthesis in C. glutamicum were also investigated. The overall strategy used for GABA synthesis in C. glutamicum in this study is shown in Fig. 1.

Fig. 1
figure1

Genetic engineering design to improve GABA production in C. glutamicum. PEP phosphoenolpyruvic acid, Pyr pyruvic acid, Acetyl-CoA acetyl-coenzyme A, (S)-Lac lactic acid, Cit citric acid, AKG alpha-ketoglutaric acid, l-Ala l-alanine, l-Gln l-glutamine, TCA cycle tricarboxylic acid cycle, SSA succinate semialdehyde, Suc succinic acid, OAA oxaloacetic acid

Materials and methods

Strains, plasmids, media, and culture conditions

The strains and plasmids used in this study are listed in Table 1. E. coli DH5α is used as cloning host for plasmid construction. GABA producer C. glutamicum CGY700 derived from C. glutamicum ATCC 13032 is used as the starting strain in this study. Plasmids pCCG1 and pBS-sgRNA [31] are used for CRISPR/Cas9-assisted genomic editing in C. glutamicum. Luria–Bertani (LB) media (g/L): NaCl 10, tryptone 10, yeast extract 5, pH 7.0–7.2. LB agar supplemented with 15 g/L agar. LBHIS media (g/L): Brain heart infusion 18.5, NaCl 5, tryptone 5, yeast extract 2.5, d-sorbitol 91, pH 7.2–7.4. LBHIS agar supplemented with 15 g/L agar. E. coli DH5α was cultivated in LB media at 37 °C. C. glutamicum was cultivated in LBHIS media at 30 °C. C. glutamicum harboring pCCG1 was cultured at 28 °C for genomic editing, and 37 °C for plasmid curing. Supplement kanamycin 30 mg/L for E. coli DH5α and 20 mg/L for C. glutamicum when needed.

Table 1 The strains and plasmids used in this study

Plasmid construction

The CRISPR/Cas9-assisted method was used to edit genome in C. glutamicum. The primers used in this study are listed in Table 2. The vector pCCG1 was digested with BamHI-AflII and the expression vector pJYW-4/5 was digested with BamHI-PstI, the linearized vectors were purified by SanPrep Column PCR Product Purification Kit (Sangon Biotech, China), The ClonExpress® II One Step Cloning Kit (Vazyme, China) was used to construct the plasmids, and the ligated product was chemically transformed into E. coli DH5α. The PCR products were purified by SanPrep Column DNA Gel Extraction Kit (Sangon Biotech, China) and the plasmid was purified from E. coli DH5α by TIANprep Mini Plasmid Kit (TIANGEN, China). The method of plasmids construction for genomic editing was described in previous report [31].

Table 2 The primers used in this study

The plasmid pCCG1-pknG was constructed for gene pknG deletion, the fragment of sgRNA, upstream homologous arm and downstream homologous arm was amplified using primers pknGsg-F/sgRNA-R, pknG-U-F/pknG-U-R, and pknG-D-F/pknG-D-R from plasmid pBS-sgRNA and C. glutamicum ATCC 13032 genome, respectively. Fused three fragments above using primers pknG-F/pknG-D-R by overlap PCR, the resulting fragment was ligated with BamHI-AflII-digested vector pCCG1 using the ClonExpress® II One Step Cloning Kit, and constructed it in E. coli DH5α. Other plasmids construction for gene deletion/insertion used the same method.

In order to construct the co-expression plasmid of genes ppc and gltA, primers ppc1-F/ppc1-R, gltA-F1/gltA-R were used to amplify ppc1 and gltA from E. coli W3110 genome, and then two genes were ligated with BamHI-PstI-digested pJYW-5 by the ClonExpress® II One Step Cloning Kit, constructed it in E. coli DH5α. Other expression plasmids construction was used the same method. ppc1 and ppc2 were used to distinguish ppc from E. coli and C. glutamicum, respectively.

Strain construction

The recombinant plasmid was electroporated into C. glutamicum to edit genome. Electroporation-competent cell of C. glutamicum was prepared as previous report [31]. For genomic editing, 1 μg plasmid was mixed with 80 μL competent cells and added into pre-cooled 1 mm Gene Pulser cuvette (Bio-Rad, USA), keeping in ice-bath for 10–15 min. Electroporation in micropulser (Bio-Rad, USA) at 1.8 kV for twice, the cells were immediately transferred into 0.9 mL pre-cooled liquid LBHIS medium and cultivated at 30 °C, 200 rpm for 1–1.5 h in shaking incubator, then spread on LBHIS agar (supplementing 20 mg/L kanamycin, 0.01 mM IPTG) and cultured for about 3 days at 28 °C. The transformant was examined by colony PCR method. The plasmids in cells were cured by cultivating at 37 °C, 200 rpm for 16 h in shaking incubator based on temperature-sensitive replicon pBL1TS in plasmid pCCG1.

The GABA fermentation in shake flask

Seed media (g/L): glucose·H2O 25, urea 5, corn steep liquor 20, yeast extract 2, KH2PO4 1, MgSO4·7H2O 0.4, PPE 0.01, adjusted pH 7.3 using 5 M NaOH. GABA fermentation media (g/L): glucose·H2O 110 (sterilized separately), corn steep liquor 1, yeast extract 0.2, MgSO4·7H2O 0.8, KH2PO4 2, MnSO4·H2O 0.01, FeSO4·7H2O 0.02, PPE 0.1, adjust pH to 7.3 using 5 M NaOH. The method of GABA production in shake flasks was programmed as follows: The candidate strain was transferred into 4 mL liquid LBHIS media and cultivated at 30 °C, 200 rpm for 12–14 h in shaking incubator, and then 0.2 mL pre-culture was transferred into 30 mL seed media in 500-mL baffled shake flask. The strains were cultivated at 30 °C, 200 rpm for 9–12 h until the optical density reached OD562nm 40 ± 3, 3 mL culture was transferred into 500-mL baffled shake flask containing 30 mL GABA fermentation media, cultivating at 30 °C, 200 rpm for 72 h in incubator. Urea was used as nitrogen source and acidity regulator in 24 h, and supplemented 0.4, 0.24, 0.24, 0.24, 0.24, and 0.24 mL of urea (300 g/L) at 0, 10, 13 h, 16, 19, and 22 h, respectively.

The Fed-Batch cultivation in 2.4L-bioreactor

For GABA production by Fed-Batch cultivation in bioreactor (T&J-Minibox 2.4L, T&J Bio-engineering, Co., Ltd., China), the strains activation and seed culture method were same as in shake flask. the fermentation broth was prepared as follows: weighing the following reagents and dissolving them in 800 mL purified water: yeast extract 0.2 g, corn steep liquor 1 g, KH2PO4 2 g, MgSO4·7H2O 0.8 g, FeSO4·7H2O 0.02 g, MnSO4·H2O 0.01 g, natural pH value, sterilized at 115 °C for 20 min. 200 mL glucose solution (550 g/L glucose·H2O) was supplemented in the media, and adjust pH to 7.2 using 12.5% ammonium hydroxide by the pumps. Cultivation temperature: 30 °C; ventilation rate: 1 L/min (1 atm); dissolved oxygen control: 20%—30% (coupled with the speed of stirrer). The glucose and l-Glu concentration in broth was traced by biosensor analyzer during fermentation. After the glucose concentration decreased, controlled it at 15 ± 5 g/L using 550 g/L glucose·H2O during the fermentation; pH value: control the pH in 7.0–7.2 by 12.5% ammonium hydroxide, and in 5.1–5.2 by 6 M HCl when l-Glu concentration reached maximum.

Determination of biochemical parameters

Cell growth was monitored by measuring the optical density at 562 nm using UV-1800 spectrophotometer (MAPADA instrument, China). The glucose and l-Glu concentration in fermentation broth was measured by biosensor analyzer (SBA-40E, Biology Institute of Shandong Academy of Sciences, China) during the Fed-Batch cultivation, the precise yield of l-Glu and other amino acids were measured by high performance liquid chromatography (HPLC) on Agilent 1260 instrument (Agilent Technologies, USA) when the fermentation is over.

The analysis of amino acids in media

The method of HPLC was used to analyze the amino acids in fermentation broth, HPLC column: Hypersil ODS-2, 250 × 4.6 mm, 5 μm (ThermoFisher Scientific, USA). The sample was prepared as follows: 0.2 mL supernatant of broth was mixed with 0.2 mL 10% trichloroacetic acid and incubated at 4 °C for 4 h, centrifuge it at 13,000×g for 10 min to remove the impurities in supernatant, and dilute the supernatant to a suitable concentration. Amino acids were derivatized by OPA reagent (Agilent Technologies, USA). Mobile phase A: 3.01 g sodium acetate, 5 mL tetrahydrofuran, and 0.2 mL triethylamine were dissolved in 995 mL water, adjust pH 7.2 with 5% acetic acid; mobile phase B: 3.01 g sodium acetate was dissolved in 200 mL water, adjust pH 7.2 with 5% acetic acid, then add 400 mL methanol and 400 mL acetonitrile. Gradient program: 0 min: 92% A, 8% B; 15 min: 63.1% A, 36.9% B; 16 min: 0% A, 100% B; 21 min: 0% A, 100% B; 22 min: 92% A, 8% B; 24 min: stop; flow rate: 0.8 mL/min; column temperature: 40 °C; UV detection wavelength: 338 nm.

Results

The deletion of ldh or pknG could increase GABA biosynthesis in C. glutamicum

Lactic acid metabolism is an important bypass of amino acid biosynthesis, and ldh encoding l-lactate dehydrogenase is the key gene referring to lactic acid biosynthesis in C. glutamicum. The gene ldh was deleted from CGY700, resulting in the strain CGY705, and the GABA production in CGY705 reached 22.40 g/L after 60 h fermentation (Fig. 2b), which is 17.0% increase compared to the starting strain CGY700 (19.14 g/L). Serine/threonine protein kinase G (PknG) could phosphorylate OdhI, the dephosphorylated ODhI can binds to E1 subunit of 2-oxoglutarate dehydrogenase complex (ODHC) and inhibit its activity [32], therefore, ΔpknG mutant could improve the GABA production in C. glutamicum [33, 34]. The pknG was deleted from CGY700, resulting in the strain CGY707. The GABA production in CGY707 reached 20.84 g/L after 60 h fermentation (Fig. 2b), however, the l-Glu titer in CGY707 decreased 21.9% after 24 h compared to CGY700 (Fig. 2a). These results indicate that ldh or pknG deletion improves GABA production in C. glutamicum.

Fig. 2
figure2

Comparison of CGY700, CGY705, and CGY707 fermentations in shake flasks. a l-Glu titer; b GABA titer; c glucose concentration; d Biomass, the optical density OD562nm. Data were shown as mean ± SD (standard deviation), and three independent experiments were performed for statistics

Co-expression of ppc, pck, pyc, and gltA could improve GABA biosynthesis in C. glutamicum

Citric acid biosynthesis is a critical rate-limiting step of the TCA cycle in C. glutamicum, and citrate synthase (CS) encoded by gltA is involved in the reaction. Theoretically, the enhancement of PEP/pyruvate carboxylation through overexpressing phosphoenolpyruvate carboxylase (PEPC) encoded by ppc, phosphoenolpyruvate carboxykinase (PEPCK) encoded by pck or pyruvate carboxylase (PYC) encoded by pyc could drive the metabolic flux to the TCA cycle. In this study, ppc1, pck or pyc was co-expressed with gltA in CGY700, resulting in strains CGY700-ppc1-gltA, CGY700-pck-gltA, and CGY700-pyc-gltA, respectively. Comparing with CGY700, the maximum titer of l-Glu increased 41.8% in CGY700-ppc1-gltA (32.11 g/L), but decreased 14.2% in CGY700-pyc-gltA (19.93 g/L) and 0.4% in CGY700-pyc-gltA (23.15 g/L), respectively (Figs. 2a, 3a). Comparing with CGY700, the GABA titer decreased 39.8% and 7.4% in CGY700-ppc1-gltA and CGY700-pyc-gltA, respectively, but increased 7.9% in CGY700-pck-gltA (Figs. 2b, 3b) after 72 h fermentation. A mole of l-Glu decarboxylated by GAD, resulting in a mole of GABA synthesis, therefore, the total molar productions of l-Glu and GABA are showed in Fig. 3c. Comparing with CGY700, molar production of GABA reached 267.27, 205.52 and 208.56 mmol/L in CGY700-ppc1-gltA, CGY700-pck-gltA, and CGY700-pyc-gltA, respectively. GADs functions at acidic condition and turns active when pH < 5.5, the low GABA production in CGY700-ppc1-gltA owing to the low pH declining rate in media (Fig. 3f), but this strain showed potential highest productivity because of its higher conversion rate from glucose (Fig. 3d).

Fig. 3
figure3

Comparison of CGY700-ppc1-gltA, CGY700-pck-gltA and CGY700-pyc-gltA fermentations in shake flasks. a l-Glu titer; b GABA titer; c the total mole production of l-Glu and GABA; d glucose concentration; e biomass; f pH value. Data were shown as mean ± SD, and three independent experiments were performed for statistics

Optimizing ppc and gltA co-expression to further improve GABA biosynthesis in C. glutamicum

Since co-expression of E. coli-derived ppc1 and gltA improves GABA production, the expression levels of these two genes were optimized by putting them under different promoters Ptac and PtacM. Six modes were programmed to optimized ppc1 and gltA co-expression (Fig. 4a). Plasmids pJYW-Ptac-ppc1-gltA, pJYW-PtacM-ppc1-gltA, pJYW-Ptac-ppc1-Ptac-gltA, pJYW-Ptac-ppc1-PtacM-gltA, pJYW-PtacM-ppc1-Ptac-gltA, and pJYW-PtacM-ppc1-PtacM-gltA were prepared and introduced to CGY700, resulting in the strains CGY700A, CGY700B, CGY700C, CGY700D, CGY700E, and CGY700F, respectively. The results of fermentation in shake flasks (Fig. 4) showed that the highest l-Glu titer was obtained in CGY700B (32.95 g/L), while the lowest l-Glu titer was obtained in CGY700E (28.00 g/L). However, CGY700E revealed the highest GABA productivity and produce 16.86 g/L after 72 h fermentation, while CGY700B showed the lowest GABA production (10.94 g/L). The lowest pH in CGY700E media was contributed to this result in 24 h (Fig. 4e), because the GAD activity is higher under low pH value. The l-Glu titer increased 20.5% compared to CGY700 as well. Hence, PtacM-ppc1-Ptac-gltA should be the optimal expression mode and more conducive to GABA synthesis in C. glutamicum. Therefore, CGY700E was used for further study.

Fig. 4
figure4

Optimization of ppc and gltA expression for GABA synthesis. a The design of ppc1 and gltA expression mode. b l-Glu titer. c GABA titer. d glucose concentration. e pH value. Data were shown as mean ± SD, and three independent experiments were performed for statistics

Construction of ldh, pknG, alaT, and glnA2 mutants for higher GABA production in C. glutamicum

Deletion of ldh or pknG could improve GABA synthesis in C. glutamicum (Fig. 2b). The genes alaT encoding PLP-dependent aminotransferase and glnA2 encoding glutamine synthase are involved in l-Ala and l-Gln synthesis from l-Glu, respectively, and they are important bypass of l-Glu metabolism. GlnA2 mainly catalyzes the degradation of l-Gln to l-Glu under normal physiological conditions [31], hence, enhancing glnA2 expression might improve l-Glu accumulation. Ptac-glnA2 was introduced into the chromosome to enhance its expression. Gene ldh, pknG and alaT were deleted from the chromosome. The strains CGY705E, CGY707E, CGY708E, CGY709E, and CGY710E were constructed and their genomic features are showed in Table 1.

l-Glu accumulation in all these five mutants decreased, and CGY709E decreased the most (Fig. 5). However, GABA production has a significant increase. GABA titer in CGY705E and CGY707E increased 67.2% and 16.4% after 72 h, respectively, GABA titer in CGY708E increased 86.7%. alaT deletion in CGY708E resulted in GABA titer increased 3.1% to reach maximum 32.83 g/L (CGY710E), while GABA titer in CGY709E decreased 40.0%. Compared to CGY700E, the total glucose consumption increased 36.5% and 54.6% in CGY705E and CGY708E, respectively (Figs. 4d, 5c). The results indicate that co-deletion of ldh and pknG benefits glucose consumption and GABA synthesis in C. glutamicum.

Fig. 5
figure5

Comparison of CGY705E, CGY707E, CGY708E, CGY709E, and CGY710E fermentations in shake flasks. a l-Glu titer; b GABA titer; c glucose concentration; d biomass. Data were shown as mean ± SD, and three independent experiments were performed for statistics

Co-expression of ppc and gltA in the chromosome of C. glutamicum for GABA synthesis

The co-expression of ppc1 and gltA using PtacM-ppc1-Ptac-gltA operon in plasmid pJYW-4/5 could improve GABA synthesis in C. glutamicum. In this study, the plasmids pCCG1-Ptac-ppc and pCCG1-PtacM-ppc were constructed and used to replace the native ppc2 promoter with Ptac or PtacM in CGY700 chromosome, the plasmids pCCG1-Ptac-gltA and pCCG1-PtacM-gltA were constructed and used to insert Ptac-gltA/PtacM-gltA fragment following ppc2 downstream sequence in the chromosome (Fig. 6a), resulting strains CGY-PG-100, CGY-PG-200, CGY-PG-300, and CGY-PG-400, respectively.

Fig. 6
figure6

Comparison of CGY-PG-100, CGY-PG-200, CGY-PG-300, and CGY-PG-400 fermentations in shake flasks. a l-Glu titer; b GABA titer; c glucose concentration; d biomass. Data were shown as mean ± SD, and three independent experiments were performed for statistics

Based on the fermentation results in shake flasks, the strain CGY-PG-300 generated the highest l-Glu (26.36 g/L) and GABA (23.63 g/L) production, increased 9.1% and 10.9%, respectively, compared to the lowest titer strain CGY-PG-400. Therefore, PtacM-ppc2-Ptac-gltA chromosomal expression mode should be the optimal for GABA production, similar to its expression in plasmid. GABA titer increased 23.5% in CGY-PG-300 compared to CGY700. The results indicate that the chromosomal co-expression of ppc and gltA is beneficial to GABA production in C. glutamicum.

Construction of ldh, pknG, alaT, and glnA2 mutants for more GABA production in CGY-PG-300

Constructions of ldh, pknG, alaT, and glnA2 mutants were programmed in the strain CGY-PG-300, resulting in strains CGY-PG-301, CGY-PG-302, CGY-PG-303, CGY-PG-304, CGY-PG-305, and CGY-PG-306, respectively, and their genomic features are listed in Table 1.

CGY-PG-306 produced 26.89 g/L l-Glu at 24 h, considerable to CGY-PG-300, while other strains decreased 17.6%, 35.5%, 28.1%, 27.4%, and 32.4%, respectively (Figs. 6a, 7a). Meanwhile, GABA titer increased 58.2% to 37.39 g/L in CGY-PG-301, but slight decreased 8.0% to 21.74 g/L in CGY-PG-302. Interestingly, pknG deletion still could increase the production rate of GABA, and co-deletion of ldh and pknG (CGY-PG-303) resulted in GABA titer increased 67.4% to 39.56 g/L. GABA titer in CGY-PG-304, CGY-PG-305, and CGY-PG-306 increased to 41.17, 40.79, and 40.28 g/L, respectively. The maximum titer of l-Gln decreased 13.5% in CGY-PG-304 at 48 h (Fig. 7c) compared to CGY-PG-303, while the l-Glu only increased 1.0% (Fig. 7a). Co-deletion of ldh and pknG in CGY-PG-300 also caused the total glucose consumption increased 31.9% and biomass increased 16.6% (Figs. 6, 7). The results indicate that synergistic effect of ldh and pknG co-deletion significantly promotes GABA production in C. glutamicum.

Fig. 7
figure7

Comparison of CGY-PG-301, CGY-PG-302, CGY-PG-303, CGY-PG-304, CGY-PG-305, and CGY-PG-306 fermentations in shake flasks. a l-Glu titer; b GABA titer; c l-Gln titer in CGY-PG-303 and CGY-PG-304; d glucose concentration; e biomass. Data were shown as mean ± SD, and three independent experiments were performed for statistics

The Fed-Batch fermentation of C. glutamicum CGY-PG-304 in 2.4-L bioreactor

The proper pH for l-Glu synthesis is 6.8–7.5, but for GABA synthesis is 4.0–5.5, because GAD would turn to inactive when pH is higher than 5.5, therefore, two-stage fermentation based on pH control was employed for GABA production in C. glutamicum. The Fed-Batch fermentation results of C. glutamicum CGY-PG-304 in 2.4-L bioreactor are shown in Fig. 8. In the earlier stage, pH was controlled at 7.0–7.2 for cell growth and l-Glu accumulation, when the l-Glu concentration reached maximum at 48 h, supplementation of ammonium hydroxide was stopped to make pH naturally drop to less than 6, then 6 M HCl was supplemented to adjust pH to 5.2. The pH was controlled higher than 5.0 to make cells survive. The glucose levels dropped below 20 g/L after 18 h, and then glucose supplementation was started. l-Glu concentration reached maximum 72.32 g/L after 48 h (tracked by biosensor analyzer) and GABA titer reached maximum 58.33 g/L after 72 h fermentation. After adjusted pH to 5.2, l-Glu can still be slightly synthesized, the total mole production of l-Glu and GABA increased 10.6% from 48 to 72 h. GABA yield reached 0.30 g/g glucose.

Fig. 8
figure8

Fed-batch fermentation. The pH was controlled at 7.2 using 12.5% ammonium hydroxide before 48 h, and changed to pH 5.2 using 6 M HCl after 48 h

Discussion

Acetyl-CoA and citrate biosynthesis are the bottleneck reactions in TCA cycle. PEPC catalyzes the irreversible carboxylation of PEP to OAA, and PEPCK catalyzes the reversible carboxylation of PEP to OAA [35], accompanying ATP/ADP conversion. Pyruvate carboxylation by PYC to OAA requires energy from ATP. PEPC and PEPCK exist in E. coli but not PYC. PEPC, PEPCK, and PYC all exist in C. glutamicum. The reactions involved PEPC, PEPCK, and PYC compose the anaplerotic reaction of carbon source in microorganism. Though this pathway is dispensable for cell growth and metabolism [36], it is widely used as compensation to reprogram the flux in industrial strains to produce bioproducts [37,38,39] and to skip the obstacle of pyruvate dehydrogenase complex (PDHC). Furthermore, CO2 anaplerotic reaction is more efficient to drive CO2 back to TCA cycle, generating the theoretical maximum l-Glu yield of 81.7% from glucose.

In this study, E. coli-originated PEPC (PEPCEc), PEPCK, and CS were co-expressed in C. glutamicum to improve GABA synthesis, and the GABA production increased 71.5% (CGY710E). Co-expression of C. glutamicum-originated PEPC (PEPCCg) and E. coli-originated CS in C. glutamicum also resulted in 115.1% increase of GABA production (CGY-PG-304). However, GABA productivity was very low when overexpressed PEPCCg in plasmid pJYW-4/5 with medium copy numbers in C. glutamicum ATCC 13032 [40]. PEPCCg enzymatic activity might be too high and nitrogen consumption capacity decreased, resulting in high pH in media (pH > 8.5 in 32 h under this fermentation condition) and GAD inactivity [41]. The biomass was often < 25 (OD562nm), obviously lower than regular state, the tolerance of C. glutamicum to urea was reduced. As a matter of fact, overexpression of PEPCEc also made the pH higher (Fig. 3f). The PEPCCg activity should be proper when overexpressed in the chromosome because only one copy of ppc existed. To date, GABA production using fermentation from food-grade sugar (glucose or xylose) has been reported in C. glutamicum and E. coli, the main achievements in recent years are listed in Table 3. The highest GABA production (70.6 g/L) was obtained from an industrial l-Glu producer which produced 100–110 g/L l-Glu [29]. The highest GABA productions obtained in E. coli and wild-type C. glutamicum ATCC 13,032 were 6.16 g/L [25] and 38.6 g/L [30], respectively. These studies used plasmid expression systems, which need to supplement antibiotics in media, resulting in high production cost and environmental pollution [16]. In this study, we constructed an antibiotic-independent GABA-producing C. glutamicum strain using genome editing.

Table 3 GABA production in C. glutamicum and E. coli reported in recent years

Glutamine synthase (GS) encoded by glnA is the key enzyme that catalyzes the conversion of l-Glu to l-Gln. There are two isoenzymes of GS and labeled as glnA1 and glnA2 in C. glutamicum. The deletion of glnA1 turns the cells to auxotroph of l-Gln, while deletion of glnA2 improves l-Gln synthesis [31]. Overexpression of glnA2 might convert l-Gln to l-Glu in theory, however, overexpression of glnA2 only made l-Gln decreased 13.5% in CGY-PG-304 and l-Glu synthesis was basically unchanged (Fig. 7). Deletion of ldh improves GABA synthesis, and deletion of pknG increases the productivity (Fig. 2). Deletion of both ldh and pknG significantly improved GABA synthesis after ppc and gltA were overexpressed in C. glutamicum genome (Figs. 5, 7). Meanwhile, glucose consumption and biomass were increased, the glucose would be exhausted in 36–42 h when using 110 g/L glucose·H2O for GABA fermentation in shake flasks after deletion of ldh and pknG, therefore, 140 g/L glucose·H2O was used under this situation.

Conclusion

First, chromosomal expression of ppc and gltA was optimized in a GABA-producing strain CGY700, resulting in 23.5% increase of GABA production. Second, deletion of ldh and pknG increased GABA titer 67.4%. Finally, a high GABA producer CGY-PG-304 was constructed, which can produce 58.33 g/L GABA from glucose by Fed-Batch fermentation. GABA yield reached 0.30 g/g glucose.

Availability of data and material

The data that support the findings of this study are available from the corresponding author upon reasonable request.

Code availability

Not applicable.

References

  1. 1.

    Nikmaram N, Dar BN, Roohinejad S, Koubaa M, Barba FJ, Greiner R, Johnson SK. Recent advances in gamma-aminobutyric acid (GABA) properties in pulses: an overview. J Sci Food Agric. 2017;97(9):2681–9.

    CAS  PubMed  Article  Google Scholar 

  2. 2.

    Yang J, Sun C, Zhang Y, Fu D, Zheng X, Yu T. Induced resistance in tomato fruit by γ-aminobutyric acid for the control of alternaria rot caused by Alternaria alternata. Food Chem. 2017;221:1014–20.

    CAS  PubMed  Article  Google Scholar 

  3. 3.

    Seitanidou M, Franco-Gonzalez JF, Sjostrom TA, Zozoulenko I, Berggren M, Simon DT. pH Dependence of gamma-aminobutyric acid iontronic transport. J Phys Chem B. 2017;121(30):7284–9.

    CAS  PubMed  Article  PubMed Central  Google Scholar 

  4. 4.

    Jacobson LH, Vlachou S, Slattery DA, Li X, Cryan JF. The gamma-aminobutyric acid B receptor in depression and reward. Biol Psychiatry. 2018;83(11):963–76.

    CAS  PubMed  Article  Google Scholar 

  5. 5.

    Wong CG, Bottiglieri T, Snead OC 3rd. GABA, gamma-hydroxybutyric acid, and neurological disease. Ann Neurol. 2003;54(Suppl 6):S3-12.

    CAS  PubMed  Article  PubMed Central  Google Scholar 

  6. 6.

    Tang J, Chen Z. The protective effect of γ-aminobutyric acid on the development of immune function in chickens under heat stress. J Anim Physiol Anim Nutr. 2016;100(4):768–77.

    CAS  Article  Google Scholar 

  7. 7.

    Cheng JB, Bu DP, Wang JQ, Sun XZ, Pan L, Zhou LY, Liu W. Effects of rumen-protected γ-aminobutyric acid on performance and nutrient digestibility in heat-stressed dairy cows. J Dairy Sci. 2014;97(9):5599–607.

    CAS  PubMed  Article  PubMed Central  Google Scholar 

  8. 8.

    Kim JY, Lee MY, Ji GE, Lee YS, Hwang KT. Production of γ-aminobutyric acid in black raspberry juice during fermentation by Lactobacillus brevis GABA100. Int J Food Microbiol. 2009;130(1):12–6.

    CAS  PubMed  Article  PubMed Central  Google Scholar 

  9. 9.

    Nakamura H, Takishima T, Kometani T, Yokogoshi H. Psychological stress-reducing effect of chocolate enriched with γ-aminobutyric acid (GABA) in humans: assessment of stress using heart rate variability and salivary chromogranin A. Int J Food Sci Nutr. 2009;60(sup5):106–13.

    CAS  PubMed  Article  PubMed Central  Google Scholar 

  10. 10.

    Inoue K, Shirai T, Ochiai H, Kasao M, Hayakawa K, Kimura M, Sansawa H. Blood-pressure-lowering effect of a novel fermented milk containing γ-aminobutyric acid (GABA) in mild hypertensives. Eur J Clin Nutr. 2003;57:490–5.

    CAS  PubMed  Article  Google Scholar 

  11. 11.

    Leonte A, Colzato LS, Steenbergen L, Hommel B, Akyurek EG. Supplementation of gamma-aminobutyric acid (GABA) affects temporal, but not spatial visual attention. Brain Cogn. 2018;120:8–16.

    PubMed  Article  Google Scholar 

  12. 12.

    Park SJ, Kim EY, Noh W, Oh YH, Kim HY, Song BK, Cho KM, Hong SH, Lee SH, Jegal J. Synthesis of nylon 4 from gamma-aminobutyrate (GABA) produced by recombinant Escherichia coli. Bioprocess Biosyst Eng. 2013;36(7):885–92.

    CAS  PubMed  Article  Google Scholar 

  13. 13.

    Lammens TM, Franssen MCR, Scott EL, Sanders JPM. Synthesis of biobased N-methylpyrrolidone by one-pot cyclization and methylation of γ-aminobutyric acid. Green Chem. 2010;12(8):1430.

    CAS  Article  Google Scholar 

  14. 14.

    Boonstra E, de Kleijn R, Colzato LS, Alkemade A, Forstmann BU, Nieuwenhuis S. Neurotransmitters as food supplements: the effects of GABA on brain and behavior. Front Psychol. 2015;6:1520.

    PubMed  PubMed Central  Article  Google Scholar 

  15. 15.

    Ke C, Yang X, Rao H, Zeng W, Hu M, Tao Y, Huang J. Whole-cell conversion of l-glutamic acid into gamma-aminobutyric acid by metabolically engineered Escherichia coli. Springerplus. 2016;5(1):1–8.

    Article  Google Scholar 

  16. 16.

    Luo H, Liu Z, Xie F, Bilal M, Liu L, Yang R, Wang Z. Microbial production of gamma-aminobutyric acid: applications, state-of-the-art achievements, and future perspectives. Crit Rev Biotechnol. 2021;41:491–512.

    PubMed  Article  Google Scholar 

  17. 17.

    Shan Y, Man CX, Han X, Li L, Guo Y, Deng Y, Li T, Zhang LW, Jiang YJ. Evaluation of improved γ-aminobutyric acid production in yogurt using Lactobacillus plantarum NDC75017. J Dairy Sci. 2015;98(4):2138–49.

    CAS  PubMed  Article  Google Scholar 

  18. 18.

    Wang H, Huang J, Sun L, Xu F, Zhang W, Zhan J. An efficient process for co-production of γ-aminobutyric acid and probiotic Bacillus subtilis cells. Food Sci Biotechnol. 2018;28(1):155–63.

    PubMed  PubMed Central  Article  CAS  Google Scholar 

  19. 19.

    Zhao A, Hu X, Wang X. Metabolic engineering of Escherichia coli to produce gamma-aminobutyric acid using xylose. Appl Microbiol Biotechnol. 2017;101(9):3587–603.

    CAS  PubMed  Article  PubMed Central  Google Scholar 

  20. 20.

    Jorge JMP, Leggewie C, Wendisch VF. A new metabolic route for the production of gamma-aminobutyric acid by Corynebacterium glutamicum from glucose. Amino Acids. 2016;48(11):2519–31.

    CAS  PubMed  Article  Google Scholar 

  21. 21.

    Jorge JMP, Nguyen AQD, Perez-Garc F, Kind S, Wendisch VF. Improved fermentative production of gamma-aminobutyric acid via the putrescine route: systems metabolic engineering for production from glucose, amino sugars, and xylose. Biotechnol Bioeng. 2017;114(4):862–73.

    CAS  PubMed  Article  Google Scholar 

  22. 22.

    Thu Ho NA, Hou CY, Kim WH, Kang TJ. Expanding the active pH range of Escherichia coli glutamate decarboxylase by breaking the cooperativeness. J Biosci Bioeng. 2013;115(2):154–8.

    CAS  PubMed  Article  Google Scholar 

  23. 23.

    Yu K, Lin L, Hu S, Huang J, Mei L. C-terminal truncation of glutamate decarboxylase from Lactobacillus brevis CGMCC 1306 extends its activity toward near-neutral pH. Enzyme Microb Technol. 2012;50(4–5):263–9.

    CAS  PubMed  Article  Google Scholar 

  24. 24.

    Shi F, Xie Y, Jiang J, Wang N, Li Y, Wang X. Directed evolution and mutagenesis of glutamate decarboxylase from Lactobacillus brevis Lb85 to broaden the range of its activity toward a near-neutral pH. Enzyme Microb Technol. 2014;61–62:35–43.

    PubMed  Article  CAS  Google Scholar 

  25. 25.

    Park S-H, Sohn YJ, Park SJ, Choi J-I. Effect of DR1558, a Deinococcus radiodurans response regulator, on the production of GABA in the recombinant Escherichia coli under low pH conditions. Microb Cell Fact. 2020;19(1):1–12.

    Article  CAS  Google Scholar 

  26. 26.

    Shi F, Luan M, Li Y. Ribosomal binding site sequences and promoters for expressing glutamate decarboxylase and producing gamma-aminobutyrate in Corynebacterium glutamicum. AMB Expr. 2018;8(1):61.

    Article  CAS  Google Scholar 

  27. 27.

    Takahashi C, Shirakawa J, Tsuchidate T, Okai N, Hatada K, Nakayama H, Tateno T, Ogino C, Kondo A. Robust production of gamma-amino butyric acid using recombinant Corynebacterium glutamicum expressing glutamate decarboxylase from Escherichia coli. Enzyme Microb Technol. 2012;51(3):171–6.

    CAS  PubMed  Article  PubMed Central  Google Scholar 

  28. 28.

    Baritugo K-A, Kim HT, David Y, Khang TU, Hyun SM, Kang KH, Yu JH, Choi JH, Song JJ, Joo JC, Park SJ. Enhanced production of gamma-aminobutyrate (GABA) in recombinant Corynebacterium glutamicum strains from empty fruit bunch biosugar solution. Microb Cell Fact. 2018;17(1):1–12.

    Article  CAS  Google Scholar 

  29. 29.

    Zhang R, Yang T, Rao Z, Sun H, Xu M, Zhang X, Xu Z, Yang S. Efficient one-step preparation of γ-aminobutyric acid from glucose without an exogenous cofactor by the designed Corynebacterium glutamicum. Green Chem. 2014;16(9):4190.

    CAS  Article  Google Scholar 

  30. 30.

    Choi JW, Yim SS, Lee SH, Kang TJ, Park SJ, Jeong KJ. Enhanced production of gamma-aminobutyrate (GABA) in recombinant Corynebacterium glutamicum by expressing glutamate decarboxylase active in expanded pH range. Microb Cell Fact. 2015;14(1):1-11.

    Article  CAS  Google Scholar 

  31. 31.

    Yao C, Hu X, Wang X. Construction and application of a CRISPR/Cas9-assisted genomic editing system for Corynebacterium glutamicum. AMB Expr. 2021;11(1):1–15.

    Article  CAS  Google Scholar 

  32. 32.

    Schultz C, Niebisch A, Gebel L, Bott M. Glutamate production by Corynebacterium glutamicum: dependence on the oxoglutarate dehydrogenase inhibitor protein OdhI and protein kinase PknG. Appl Microbiol Biotechnol. 2007;76(3):691–700.

    CAS  PubMed  Article  Google Scholar 

  33. 33.

    Okai N, Takahashi C, Hatada K, Ogino C, Kondo A. Disruption of pknG enhances production of gamma-aminobutyric acid by Corynebacterium glutamicum expressing glutamate decarboxylase. AMB Expr. 2014;4(1):1–8.

    CAS  Google Scholar 

  34. 34.

    Wang N, Ni Y, Shi F. Deletion of odhA or pyc improves production of gamma-aminobutyric acid and its precursor l-glutamate in recombinant Corynebacterium glutamicum. Biotechnol Lett. 2015;37(7):1473–81.

    CAS  PubMed  Article  Google Scholar 

  35. 35.

    Riedel C, Rittmann D, Dangel P, Möckel B, Petersen S, Sahm H, Eikmanns BJ. Characterization of the phosphoenolpyruvate carboxykinase gene from Corynebacterium glutamicum and significance of the enzyme for growth and amino acid production. J Mol Microbiol Biotechnol. 2001;3(4):573–83.

    CAS  PubMed  Google Scholar 

  36. 36.

    Peters-Wendisch PG, Eikmanns BJ, Thierbach G, Bachmann B, Sahm H. Phosphoenolpyruvate carboxylase in Corynebacterium glutamicum is dispensable for growth and lysine production. FEMS Microbiol Lett. 1993;112:269–74.

    CAS  Article  Google Scholar 

  37. 37.

    Xu D, Zhao J, Cao G, Wang J, Li Q, Zheng P, Zhao S, Sun J. Removal of feedback inhibition of Corynebacterium glutamicum phosphoenolpyruvate carboxylase by addition of a short terminal peptide. Biotechnol Bioproc E. 2018;23(1):72–8.

    CAS  Article  Google Scholar 

  38. 38.

    Mao Y, Li G, Chang Z, Tao R, Cui Z, Wang Z, Tang Y-J, Chen T, Zhao X. Metabolic engineering of Corynebacterium glutamicum for efficient production of succinate from lignocellulosic hydrolysate. Biotechnol Biofuels. 2018;11(1):1–7.

    Article  CAS  Google Scholar 

  39. 39.

    Wang Z, Moslehi-Jenabian S, Solem C, Jensen PR. Increased expression of pyruvate carboxylase and biotin protein ligase increases lysine production in a biotin prototrophic Corynebacterium glutamicum strain. Eng Life Sci. 2015;15(1):73–82.

    Article  CAS  Google Scholar 

  40. 40.

    Hu J, Li Y, Zhang H, Tan Y, Wang X. Construction of a novel expression system for use in Corynebacterium glutamicum. Plasmid. 2014;75:18–26.

    CAS  PubMed  Article  Google Scholar 

  41. 41.

    Zhu L, Mack C, Wirtz A, Kranz A, Polen T, Baumgart M, Bott M. Regulation of γ-aminobutyrate (GABA) utilization in Corynebacterium glutamicum by the PucR-type transcriptional regulator GabR and by alternative nitrogen and carbon sources. Front Microbiol. 2020;11:1–20.

    Article  Google Scholar 

  42. 42.

    Cho JS, Choi KR, Prabowo CPS, Shin JH, Yang D, Jang J, Lee SY. CRISPR/Cas9-coupled recombineering for metabolic engineering of Corynebacterium glutamicum. Metab Eng. 2017;42:157–67.

    CAS  PubMed  Article  Google Scholar 

  43. 43.

    Soma Y, Fujiwara Y, Nakagawa T, Tsuruno K, Hanai T. Reconstruction of a metabolic regulatory network in Escherichia coli for purposeful switching from cell growth mode to production mode in direct GABA fermentation from glucose. Metab Eng. 2017;43:54–63.

    CAS  PubMed  Article  Google Scholar 

  44. 44.

    Im DK, Hong J, Gu B, Sung C, Oh MK. 13C metabolic flux analysis of Escherichia coli engineered for gamma-aminobutyrate production. Biotechnol J. 2020;15(6):1900346.

    CAS  Article  Google Scholar 

Download references

Acknowledgements

This work is supported by the National Key Research and Development Program of China (2021YFC2100900) and the Key Technology Project of Inner Mongolia Autonomous Region, China (2019GG302).

Funding

Publication costs are funded by the National Key Research and Development Program of China (2021YFC2100900); the Key Technology Project of Inner Mongolia Autonomous Region, China (2019GG302).

Author information

Affiliations

Authors

Contributions

XW and CY conceived and designed research. XW, CY, YL, and XH conducted experiments. XW contributed new reagents or analytical tools. XW and CY analyzed data. CY wrote the manuscript. All authors read and approved the final manuscript.

Corresponding author

Correspondence to Xiaoyuan Wang.

Ethics declarations

Consent to participate

Not applicable.

Ethical approval

This article does not contain any studies with human participants or animals performed by any of the authors.

Conflict of interest

The authors declare that they have no conflict of interest.

Consent for publication

Written informed consent for publication was obtained from all participants.

Rights and permissions

Reprints and Permissions

About this article

Verify currency and authenticity via CrossMark

Cite this article

Yao, C., Liu, Y., Hu, X. et al. Improve gamma-aminobutyric acid production in Corynebacterium glutamicum by optimizing the metabolic flux. Syst Microbiol and Biomanuf (2021). https://doi.org/10.1007/s43393-021-00062-8

Download citation

Keywords

  • Gamma-aminobutyric acid
  • Metabolic engineering
  • Corynebacterium glutamicum
  • Phosphoenolpyruvate carboxylase
  • Citrate synthase