Screening of 31 Lactic Acid Bacteria Strains Identified Levilactobacillus brevis NCTC 13768 as a High-Yield GABA Producer

Abstract

Gamma-aminobutyric acid (GABA) is the main inhibitory neurotransmitter in the vertebrate central nervous system, known for its role in promoting sleep, reducing anxiety, regulating blood pressure, and modulating stress, cognition, and behavior. Microbial fermentation offers an effective method for GABA production, with certain lactic acid bacteria (LAB) strains recognized as efficient producers. This study assessed the GABA-producing potential of 31 LAB strains, including isolates from traditional Bulgarian foods and plants. The strains were cultivated in an MRS medium supplemented with 1% monosodium glutamate (MSG), and GABA production was quantified using HPLC after derivatization with dansyl chloride. Most strains produced between 200 and 300 mg/L of GABA. However, Levilactobacillus brevis NCTC 13768 showed much higher productivity, reaching 3830.7 mg/L. To further evaluate its capacity, L. brevis NCTC 13768 was cultivated for 168 h in MRS medium with and without MSG. Without MSG, GABA production peaked at 371.0 mg/L during the late exponential phase. In contrast with MSG, GABA levels steadily increased, reaching 3333.6 mg/L after 168 h. RT-qPCR analyses of the glutamic acid decarboxylase (GAD) system showed that the genes of glutamate decarboxylase (gadB), glutamate-GABA antiporter (gadC), and transcriptional regulator (gadR) are significantly overexpressed when the culture reaches the late stationary phase of growth (96 h after the beginning of cultivation). These results identify L. brevis NCTC 13768 as a high-yield GABA producer, with potential applications in the production of fermented functional foods and beverages.

1. Introduction

Gamma-aminobutyric acid (GABA) is a four-carbon free amino acid that functions as the primary inhibitory neurotransmitter in the vertebrate central nervous system. It is synthesized from L-glutamic acid by glutamate decarboxylase (GAD) and plays a crucial role in behavior, cognition, and stress response [1]. The widespread adoption of GABA stems from a growing understanding of its diverse physiological roles. It is recognized for a range of physiological benefits, including improved sleep, anxiety relief, and blood pressure regulation [2]. Beyond the central nervous system, GABA and its receptors have also been identified in the peripheral nervous system, the endocrine system, and other non-neural organs involved in oxidative metabolism [3]. GABA’s therapeutic mechanisms in various diseases are primarily linked to its presence in the central nervous system and the innervation of different organs. It regulates human functions by influencing nerve signal transmission and interacting with various receptors [4]. GABA is recognized as a powerful pain reliever and offers benefits for cardiovascular function. It also shows promise in treating several neurological disorders, including Alzheimer’s disease and other dementias [5,6]. Furthermore, GABA demonstrates significant health advantages, such as anti-hypertensive, anti-diabetic, and anti-inflammatory properties. Its potential anticancer effects, which involve stimulating cancer cell death and inhibiting growth, hold promise for future applications in cancer treatment [7].

Namely, for these health benefits, GABA’s use is widespread as a dietary supplement, and there is considerable interest in developing GABA-enriched functional foods. Despite the established benefits, our recent research determined that the endogenous GABA content in staple plant foods, such as fruits, vegetables, cereals, and legumes, as well as in various medicinal plants, is notably low, which requires other approaches to obtain GABA-containing foods and beverages [8]. Microbial biosynthesis offers an efficient method for GABA production, with various micro-organisms, including yeasts, fungi, and bacteria, capable of synthesizing it [2]. Lactic acid bacteria (LAB) are particularly significant, having been used for centuries in fermented foods. Their Generally Recognized as Safe (GRAS) status, rapid growth, and minimal space requirements make them excellent candidates for enhancing GABA production [9,10]. Many Lactobacillus species are notable for their ability to produce GABA, including prominent examples such as Lactiplantibacillus plantarum, Lacticaseibacillus paracasei, Lactobacillus brevis, and Lactobacillus delbrueckii subsp. bulgaricus, Limosilactobacillus fermentum, and Lactobacillus helveticus [11,12,13,14]. It was demonstrated that GABA production by LAB is regulated by the glutamic acid decarboxylase GAD system, and these strains are distinguished by their active GAD systems and capacity to tolerate gastrointestinal-like stress conditions, enabling efficient conversion of glutamate into high levels of GABA [1,15]. The gad operon includes the genes encoding glutamate decarboxylase (gadA or gadB), glutamate-GABA antiporter (gadC), and the transcriptional regulator gadR [16]. The activity of the gad operon varies significantly among different LAB genera and species. In general, the expression levels of gadB, gadC, and gadR can be significantly influenced by pH, fermentation conditions, or the addition of monosodium glutamate (MSG), which can affect the GABA production.

While lactobacilli are key players, certain strains of Streptococcus thermophilus and Lactococcus lactis also produce GABA. Moreover, recent research has expanded the list of known GABA producers to include various species within the genera Pediococcus, Enterococcus, Leuconostoc, Propionibacterium, and Weissella [2]. The development of GABA-enriched fermented foods and the isolation of GABA-producing LAB from traditional sources (i.e., yogurt, cheese, kefir, sauerkraut, kimchi, tempeh, sourdough bread, etc.) underscore their potential as functional food ingredients. Bacterial GABA production typically initiates during the logarithmic growth phase and intensifies with GAD activity as the culture moves into stabilization. Key factors that influence this fermentation process include the substrate type, pH, temperature, and culture time [17]. The most significant contributors to GABA production are the carbon source and MSG, which serve as essential substrates for both bacterial growth and the conversion into GABA. Numerous studies report varying levels of GABA production across different LAB species. For example, L. brevis PM17, L. plantarum C48, L. paracasei PF6, L. bulgaricus PR1, and L. lactis PU1 demonstrated the highest GABA concentrations (15–63 mg/kg) [18]. Reports indicate that mulberry juice fermented with a single culture of Saccharomyces cerevisiae SC125 yielded 1.45 g/L GABA, while fermentation with L. plantarum BC114 produced 1.03 g/L GABA. Significantly, the co-culture of these yeasts was observed to enhance the GABA content to 2.42 g/L [19]. Another study investigated GABA production by two L. lactis strains. Maximum GABA production (13.20 mM and 13.35 mM) was observed after 40 h when these strains were cultured in MRS medium supplemented with 5% monosodium glutamate (MSG) at 30 °C [20]. Additional research demonstrated GABA production by L. plantarum, Companilactobacillus futsaii (former name Lactobacillus futsaii), and Levilactobacillus namurensis (former name Lactobacillus namurensis). One L. plantarum strain was particularly effective, producing the highest GABA concentration (20.34 mM), followed by another L. plantarum strain at 16.47 mM [21].

While significant research has been conducted globally on the screening of LAB isolated from plants and foods for the production of GABA, such specific and comprehensive studies are largely unexplored in Bulgaria. Bulgaria possesses a rich repository of LAB strains within its traditional fermented products, including diverse yogurts, cheeses, and fermented fruits and vegetables. This represents a significant, yet underexplored, resource for the isolation of novel LAB. Furthermore, the country’s native flora offers considerable potential for the direct isolation of unique LAB strains from various plant sources. Of particular interest is L. bulgaricus, an emblematic species intrinsically linked to Bulgarian yogurt and recognized for its established probiotic and beneficial properties [22]. A comprehensive investigation into the GABA-producing capabilities of L. bulgaricus and other indigenous LAB strains from both traditional food matrices and plant environments could provide substantial advancements for functional food development and various biotechnological applications. The current study aimed to investigate the GABA-producing potential of thirty-one LAB strains isolated from traditional Bulgarian foods and plants. Identifying a high-producing GABA strain is crucial, as it is the first step towards potentially developing GABA-enriched functional foods and beverages.

2. Materials and Methods

2.1. Micro-Organisms

Thirty-one lactic acid bacteria strains in total were used in this study (

Table 1. LAB strains used in this study.

Number	Strain	Medium	Temperature of Cultivation, °C	Source	References
Bulgarian National Bank for Industrial Microorganisms and Cell Cultures (NBIMCC)
1	Lactobacillus delbrueckii subsp. bulgaricus NBIMCC 1132	MRS	37	Collection	[23]
2	Streptococcus thermophilus NBIMCC 3916	MRS	37	Collection	[23]
3	Lacticaseibacillus rhamnosus NBIMCC 1013	MRS	37	Collection	[23]
4	Limosilactobacillus helveticus NBIMCC 275	MRS	37	Collection	[23]
5	Lentilactobacillus kefiri NBIMCC 3446	MRS	30	Collection	[23]
6	Levilactobacillus brevis NBIMCC 8426	MRS	30	Collection	[23]
7	Liquorilactobacillus mali NBIMCC 3316	MRS	30	Collection	[23]
8	Lactococcus lactis subsp. lactis NBIMCC 350	MRS	30	Collection	[23]
9	Lactococcus lactis subsp. lactis NBIMCC 355	MRS	30	Collection	[23]
Cryobiotica Ltd.
10	Lactobacillus helveticus 13	MRS	37	Cheese
11	Lactobacillus helveticus 14	MRS	37	Cheese
12	Lactobacillus rhamnosus 1	MRS	37	Cheese
13	Lactobacillus rhamnosus 2	MRS	37	Cheese
14	Lactobacillus paracasei 8	MRS	37	Cheese
15	Lactobacillus paracasei	MRS	37	Cheese
16	Lactobacillus plantarum Y6	MRS	37	Cheese
17	Levilactobacillus brevis NCTC 13768	MRS	37	Collection	[24]
18	Lactobacillus delbrueckii subsp. delbrueckii NBIMCC 2449	MRS	37	Collection	[23]
19	Lactobacillus delbrueckii subsp. bulgaricus LB6	MRS	37	Yogurt
20	Lactobacillus delbrueckii subsp. bulgaricus S10	MRS	37	Yogurt
21	Lactobacillus rhamnosus S25-1	MRS	37	Cheese
Laboratory of Lactic Acid Bacteria and Probiotics, Institute of Microbiology—BAS
22	Lactiplantibacillus plantarum AC131	MRS	37	Artisanal white brined cheese	[25]
23	Lactiplantibacillus plantarum LC2	MRS	37	Artisanal white brined cheese
24	Lactiplantibacillus plantarum L3	MRS	37	Home-made katak
25	Limosilactobacillus fermentum LB4	MRS	37	Vaginal sample
26	Limosilactobacillus helveticus 611	MRS	37	Baby feces	[26]
Laboratory of Cell Biosystems, Institute of Microbiology—BAS
27	Lacticaseibacillus rhamnosus LR2	MRS	37	Home-made yogurt
28	Lacticaseibacillus rhamnosus PglGER32	MRS	37	Isolated from Panax ginseng	[27]
29	Lacticaseibacillus rhamnosus PglGER1	MRS	37	Isolated from Panax ginseng	[27]
30	Lacticaseibacillus rhamnosus PglGER26	MRS	37	Isolated from Panax ginseng	[27]
31	Lacticaseibacillus rhamnosus PglGER3	MRS	37	Isolated from Panax ginseng	[27]

). The strains were obtained from the Bulgarian National Collection for Microorganisms and Cell Cultures (NBIMCC), the laboratory collection of The Laboratory of Lactic Acid Bacteria and Probiotics, Institute of Microbiology—BAS, the research collection of Laboratory of Cell Biosystems, Institute of Microbiology—BAS, or provided by Cryobiotica Ltd. (Voivodinovo, Bulgaria). The lab strains belong to the eight genera as follows: Lacticaseibacillus, Limosilactobacillus, Lactiplantibacillus, Lactobacillus, Streptococcus, Lactococcus, Levilactobacillus, and Lentilactobacillus. The strains were received in freeze-dried form. Before experiments, the strains were reactivated following the provided protocols. Activated LAB strains were propagated three times for 48 h each in MRS broth (Merck KGaA, Burlington, MA, USA) at 30 or 37 °C (see Table 1), without aeration, and then used as inoculum for the next experiments.

2.2. Screening for GABA Overproducing Strain

The LAB strains were cultivated in 50 mL test tubes on MRS broth (Merck KGaA, Burlington, MA, USA), supplemented with 2% (w/v) MSG (Merck KGaA, Burlington, MA, USA), and pH = 6.3. Cultivation was without shaking, at 30 or 37 °C (see Table 1) for 120 h. 48 h—old bacterial suspension, grown on MRS without MSG, was used as inoculum (2%, v/v). At the end of cultivation, the tubes were centrifuged for 20 min at 5000 rpm, and the supernatant was filtered through a 0.2 µm sterile syringe filter (Corning, New York, NY, USA) and used for GABA and GA quantification.

2.3. Dynamics of Growth and GABA Production by L. brevis NCTC 13768

L. brevis NCTC 13768 was cultivated in sealed 50 mL flasks (2% v/v inoculum), both on MRS broth and MRS broth supplemented with 1% (w/v) MSG. The cultivation was performed at 37 °C, without shaking, for 168 h. Samples were taken at 6, 12, 24, 48, 72, 96, 120, 144, and 168 h of fermentation. The OD600, pH, and conductivity were measured at each point. The biomass was separated by centrifugation (20 min at 5000 rpm) and used to determine fresh cell weight (g/L), gene expression analyses, and dry cell weight (g/L) after drying at 90 °C for 24 h. The supernatant was filtered by a 0.2 µm sterile syringe filter (Corning, New York, NY, USA) and used for measuring pH, conductivity (INOLAB, WTW, Weilheim, Germany), and for quantification of GABA and GA. The experiment was performed in triplicate.

2.4. High-Performance Liquid Chromatography Analysis of GABA and GA

The determination of GABA and GA was carried out as described in our previous study [8]. The filtered cultural media (100 µL) were mixed with 0.1 M sodium hydrogencarbonate buffer (pH 8.7) and further derivatized (55 °C, 1 h) with a freshly prepared acetone solution of dansyl chloride. The quantification of GABA and GA was performed using the UHPLC system Nexera i LC 2040 C Plus (Shimadzu Corporation, Kyoto, Japan). The system was equipped with a binary pump, an Accucore (Thermo Fisher Scientific, Waltham, MA, USA) C18 (21 mm × 150 mm, 26 µm) column, and a UV-VIS detector (Shimadzu Corporation, Kyoto, Japan).

2.5. Real-Time qPCR Assay

The fresh L. brevis NCTC 13768 biomass (about 50 mg) was immediately fixed with RNAlater Solutions (Thermo Fisher Scientific Inc., Waltham, MA, USA) and used for total RNA extraction by using RNeasy Mini Kit (Qiagen, Hilden, Germany) according to the manufacturer’s instructions. The iScript™ Reverse Transcription Supermix for RT-qPCR (Bio-Rad Laboratories, Inc., Hercules, CA, USA) was used to convert RNA to cDNA. The used primers are listed in

Table 2. Primers used in this study.

Gene	Forward Primer 5′-3′	Reverse Primer 5′-3′	Reference
gadB	TGGCTAAGTATGGTTGGCAAGTTCC	TCATCGGCAATCGTCATGGTCATG	[20]
gadC	TACCTCGTACAAGGAAACCCA	GATAAAC-GGAACAAATCCCACT	[20]
gadR	GTCGTCGATTCTCATGCTTATT	TGCCTGCTTCAGACTCTGTTT	[20]
16S rRNA	CACATTGGGACTGAGACACG	AGCCGAAACCCTTCTTCACT	[20]

.

SsoAdvanced Universal SYBR Green Supermix (Bio-Rad Laboratories, Inc., Hercules, CA, USA) was used to prepare RT-qPCR reactions on CFX Opus 96 Real-Time PCR System (Bio-Rad Laboratories, Inc., Hercules, CA, USA) following the standard protocol. The relative gene expressions were calculated by the 2^−ΔΔCt method by using 16S rRNA as a reference gene and cDNA from 0 h as a control.

2.6. Statistical Analysis

The experiments were performed at three independent replicates (n = 3). All measurements and RT-qPCR reactions were run with four technical repeats each. For the calculation of relative gene expression, the 2^−ΔΔCt method was used. The data was analyzed by one-way ANOVA analysis of variance, and the significance of the differences between the means was determined by Tukey post hoc test at p < 0.05—*; p < 0.01—**; and p < 0.001—***). The SigmaPlot 12.3 (Grafiti LLC, Palo Alto, CA, USA) was used to perform nonlinear regression analyses and to define the equations describing the growth dynamics and to calculate the 95% prediction and 95% confidence intervals. The other data were processed at p ≤ 0.05 and presented as mean values ± standard deviations (SD).

3. Results

3.1. GABA-Producing Capacity of Investigated Strains

In this study, 31 strains of LAB were investigated for their potential to produce GABA. For this purpose, bacteria were cultivated in static culture for 120 h in MRS broth, supplemented with 2% MSG. Results are shown in

Table 3. GABA-producing capacity of 31 lactic acid bacteria strains in MRS broth, supplemented with 2% MSG.

№	Strain	GABA, mg/L
1	Lactobacillus delbrueckii subsp. bulgaricus NBIMCC 1132	236.4 ± 2.6
2	Streptococcus thermophilus NBIMCC 3916	225.4 ± 4.9
3	Lacticaseibacillus rhamnosus NBIMCC 1013	269.3 ± 1.7
4	Limosilactobacillus helveticus NBIMCC 275	344.1 ± 10.3
5	Lentilactobacillus kefiri NBIMCC 3446	239.4 ± 2.6
6	Levilactobacillus brevis NBIMCC 8426	196.8 ± 2.2
7	Liquorilactobacillus mali NBIMCC 3316	328.9 ± 4.4
8	Lactococcus lactis subsp. lactis NBIMCC 350	259.6 ± 6.6
9	Lactococcus lactis subsp. lactis NBIMCC 355	315.2 ± 3.8
10	Lactobacillus helveticus 13	269.2 ± 2.3
11	Lactobacillus helveticus 14	293.9 ± 7.5
12	Lactobacillus rhamnosus 1	232.6 ± 4.7
13	Lactobacillus rhamnosus 2	256.6 ± 1.7
14	Lactobacillus paracasei 8	257.3 ± 3.2
15	Lactobacillus paracasei	269.4 ± 6.8
16	Lactobacillus plantarum Y6	343.4 ± 5.7
17	Levilactobacillus brevis NCTC 13768	3830.7 ± 12.1
18	Lactobacillus delbrueckii subsp. delbrueckii NBIMCC 2449	304.6 ± 8.9
19	Lactobacillus delbrueckii subsp. bulgaricus LB6	337.5 ± 0.5
20	Lactobacillus delbrueckii subsp. bulgaricus S10	286.7 ± 1.6
21	Lactobacillus rhamnosus S25-1	418.1 ± 1.2
22	Lactiplantibacillus plantarum AC131	n.d.
23	Lactiplantibacillus plantarum LC2	n.d.
24	Lactiplantibacillus plantarum L3	27.9 ± 0.2
25	Limosilactobacillus fermentum LB4	27.7 ± 0.2
26	Limosilactobacillus helveticus 611	n.d.
27	Lacticaseibacillus rhamnosus LR2	165.5 ± 0.5
28	Lacticaseibacillus rhamnosus PglGER32	234.0 ± 3.2
29	Lacticaseibacillus rhamnosus PglGER1	246.4 ± 0.5
30	Lacticaseibacillus rhamnosus PglGER26	245.5 ± 0.80
31	Lacticaseibacillus rhamnosus PglGER3	247.8 ± 0.40

Results are presented as mean values ± standard deviation; n.d.—not detected.

, and the distribution of GABA-producing capacity of the investigated strains is presented in Figure 1. Data reveal considerable diversity in the GABA production potential of the investigated LAB strains. Most of the strains (16) produced GABA in the range of 200–300 mg/L, and six strains produced between 300 and 400 mg/L, whereas three strains (L. plantarum AC131, L. plantarum LC2, and L. helveticus 611) did not produce any detectable GABA under the tested conditions. Notably, only one strain, L. brevis NCTC 13768, produced a markedly higher amount of GABA, in comparison to other strains, reaching 3830.7 mg/L. As a result of this initial screening, L. brevis NCTC 13768 was selected as a highly productive GABA strain and used to study the dynamics of growth and GABA accumulation in the next stages of the study.

3.2. Dynamics of Growth and GABA Production by L. brevis NCTC 13768

The growth dynamics of L. brevis NCTC 13768, cultivated on MRS broth and MRS broth supplemented with 1% MSG for 168 h, were investigated in order to evaluate its potential to produce GABA. The results show a tenfold increase in accumulated GABA when L. brevis NCTC 13768 was cultivated in medium supplemented with MSG (Figure 2). The strain reached a maximal growth at 72 h (OD₆₀₀ = 1.08 ± 0.02) when cultured on MRS (Figure 2a) and at 120 h (OD₆₀₀ = 1.16 ± 0.05) when on medium, supplied with MSG (Figure 2b). However, the medium with MSG was enriched with gas bubbles, which could affect the accuracy of optical density measurements. The analyses of accumulated fresh and dry biomass showed that the maximal amounts of accumulated bacterial biomass were at 72 h in both media (Figure 3). In addition, L. brevis NCTC 13768 accumulated significantly more (p ≤ 0.01) fresh (3.53 ± 0.12 g/L) and dry (0.78 ± 0.03 g/L) biomass when grown on MRS with 1% MSG medium, compared to the amounts accumulated on pure MRS (2.88 ± 0.03 g/L fresh and 0.70 ± 0.01 g/L dry biomass) (Figure 3).

The dynamics of GABA accumulation and GA depletion are presented in Figure 4. The data show that the amount of GA decreased rapidly until 24 h and remained constant until the end of fermentation in both media. However, there were very different patterns in GABA accumulation. When cultivated in MRS broth, GABA accumulation by L. brevis NCTC 13768 followed the dynamics of biomass accumulation, and the maximal amount (364.60 ± 9.05 mg/L) of GABA was reached at 72 h (Figure 4a). In contrast, when grown on MRS supplied with 1% MSG, the amount of GABA continued to grow until the end of the fermentation, reaching 3333.6 ± 146.42 mg/L at 168 h (Figure 4b).

3.3. Gene Expression Analysis

In order to understand the mechanism responsible for the increased GABA production by L. brevis NCTC 13768, we investigated the gene expression of the genes encoding glutamate decarboxylase (gadB), glutamate—GABA antiporter (gadC), and the transcriptional regulator gadR, which are parts of the gad operon, responsible for GABA biosynthesis in LAB (Figure 5). The data show that the genes gadB, gadC, and gadR were upregulated in a time-dependent manner during cultivation of L. brevis NCTC 13768 in both MRS and MRS supplemented with 1% MSG (Figure 5). After 12 h of cultivation, up to 72 h, the expression of gadB, gadC, and gadR in bacteria grown in medium with MSG was significantly higher than that in the cells without MSG. When the cultures enter into the stationary phase (72 h to 144 h), the expression levels of the three genes remained high in both media. After that, at 168 h, the gadB, gadC, and gadR were significantly downregulated in cells grown in medium with MSG compared to the standard MSG medium (Figure 5).

4. Discussion

It is well established that GABA production varies considerably among LAB, as multiple studies have identified distinct bacterial species as efficient producers. For instance, Yogeswara et al. isolated thirty LAB strains from traditional Indonesian fermented foods and reported L. plantarum FNCC 260 as the most productive, yielding up to 1226 mg/L GABA in medium supplemented with 100 mM MSG [28]. Similarly, another screening revealed GABA synthesis in only 58 of 135 strains across five species, with Bifidobacterium emerging as the most efficient producers, reaching 6000 mg/L [29]. In our study, L. brevis NCTC 13768 generated markedly higher GABA levels than the other investigated strains, achieving 3830.7 mg/L of GABA. Consistent with these findings, numerous studies have shown that different L. brevis strains exhibit superior GABA-producing capacity [11,16,30,31], with some accumulating up to 260 mM GABA [32]. For example, Liu et al. demonstrated that among 110 LAB strains from 18 species isolated from traditional Chinese fermented foods, L. brevis strains displayed the highest productivity (2.97–767.67 μg/mL), with strain YSJ3 being the most efficient [16]. In contrast, other species yielded substantially lower amounts, including L. plantarum (1.05–81.64 μg/mL), L. rhamnosus (1.87–28.06 μg/mL), L. sakei (1.62–13.23 μg/mL), and L. reuteri (6.16–6.71 μg/mL) [16]. Importantly, even within the same species, GABA yields vary significantly. For instance, distinct L. brevis isolates from different sources show marked differences in productivity [11,16,33,34,35]. Collectively, all these observations underscore the need for large-scale screenings, such as the present study, to identify the most potent GABA producers, even within a single species. While global efforts have focused on LAB screening, this is the first systematic evaluation of Bulgarian isolates. Notably, 22 of the 31 strains examined here were isolated from diverse sources, including traditional Bulgarian foods and plants, and this is the first report of their GABA-producing potential. To date, no clear consensus exists regarding the threshold defining a “high” GABA-producing strain. For reference, a balanced diet rich in fruits and vegetables provides approximately 740 mg of GABA daily [36], whereas the NNHPD monograph for Cognitive Function Products recommends 50–3000 mg per day, with no more than 750 mg per single dose [37]. Considering these benchmarks and the potential application of the tested strains in GABA-enriched functional beverages, L. brevis NCTC 13768 can be regarded as a high-yield GABA producer, capable of generating dietary-relevant amounts of GABA.

To study the effect of MSG on GABA production, the dynamics of growth and GABA accumulation by L. brevis NCTC 13768 were investigated in both MRS broth and MRS broth with 1% MSG. The results showed 10 times more GABA when the L. brevis NCTC 13768 was cultivated on medium with added MSG, which was probably due to the upregulated expression of the genes from the gad operon, responsible for GABA biosynthesis. In our study, the amount of supplied MSG was 1% because it was demonstrated that the concentration of MSG could have a significant effect on GABA production by L. brevis [16,38,39] and that higher amounts of MSG may significantly decrease the GABA production [16]. Likewise, Tanamool et al. found that Lactobacillus plantarum L10-11 produced five times more GABA than the control when 1% MSG was added to the medium [40]. Similarly, Valenzuela et al. identified 24 LAB strains, including Lactobacillus brevis and Lactobacillus plantarum, that synthesized over 1 mM of GABA from 1% MSG [41], and L. brevis RK03 isolated from a fermented product yielded 1024 mg/L of GABA in an MRS medium with 1% MSG [42].

In order to check the hypothesis that, namely, the genes from the gad operon were responsible in GABA biosynthesis, we analyzed the gene expression of the genes encoding glutamate decarboxylase (gadB), gluta-mate-GABA antiporter (gadC), and the transcriptional regulator gadR, which are parts of the gad operon, responsible for GABA biosynthesis in LAB (Figure 5). When cultivated on MRS medium, L. brevis NCTC 13768 accumulates the maximum amount of GABA at 72 h of cultivation and then decreases, whereas the gadB, gadC, and gadR genes remain upregulated till the end of cultivation. At the same time, the amount of glutamic acid dropped down below 100 mg/L at 96 h of cultivation when the decrease in GABA accumulation began. Said with other words, the expression levels of gadB, gadC, and gadR in the investigated strain, both in MRS and MRS with 1% MSG media, were synchronized during cultivation (Figure 5), but did not correlate well with the dynamics of GABA production (Figure 4) and biomass accumulation (Figure 3). Moreover, at the early lag phase (6 h), the expression levels of all GAD genes were significantly higher when cultivated in MRS broth than in MRS supplemented with MSG. This could be explained by the adaptation of the cells to the new conditions, since the GAD pathway is considered to be the major mechanism of lactic acid bacteria survival during acidic stress [43,44,45]. For both experiments, we used an adapted 48-hour-old bacterial culture grown on MRS as inoculum, and in the case of added MSG, the cells were exposed to higher stress. At this time point, a slight increase in pH up to 4.63 ± 0.01 (Figure 2), lack of depletion of glutamic acid, and almost no production of GABA (Figure 4) were recorded for L. brevis NCTC 13768 grown on MRS with MSG (Figure 4), which was probably due to the culture adaptation. Similar increases in levels of gadR, gadA, and gadC at the first hours of cultivation on medium without MSG compared to the medium with added MSG were reported for L. brevis NCL912 and explained by the content of glutamate in cells used as inoculum [46]. After that, at 168 h, the gadB, gadC, and gadR were significantly downregulated in cells grown in medium with MSG compared to the standard MSG medium (Figure 5). Similar decrease in GAD gene expression at the end of the cultivation process has been reported in L. brevis NCL912 and explained with product inhibition process, in which GABA competes with glutamate for the binding to the induction site of gadR protein and, thus, inhibits the expression of genes from the gad operon [45]. It is known that the transcription regulator gadR plays a crucial role in GABA production and acid resistance of LAB by activating the expression of gadB and gadC. The expression of gadR itself has been found to be strongly induced by glutamate [43,47]. However, the expression levels of gadR were found to vary significantly in different LAB species and even strains. In some L. brevis strains, its expression was found to be much higher than the expression of gadB and gadC, whereas in others, its expression was much lower than that of the other gad genes [48,49]. Liu et al. reported the existence of a connection between GABA accumulation and the concentration of glutamic acid in the medium [16]. They analyzed gene expression during cultivation of L. brevis on MRS medium with 1.25% glutamate and demonstrated that gadB and gadC were upregulated at 20 h and remained upregulated until the end of cultivation (40 h), whereas gadR reached the maximum at 20 and 28 h and then was slightly decreased [16]. However, their cultivation was limited to 48 h, which is considerably shorter than the duration used in our study (168 h). In contrast, we demonstrated that in L. brevis NCTC 13768, the dynamics of gadR expression were similar to the dynamics of gadB and gadC expression in both MSG free and MSG supplemented medium (Figure 5), which is a specific characteristic reported for the first time for this strain. Furthermore, Liu et al. investigated the impact of pH on GABA accumulation and examined the gad operon gene expression, though only in MRS medium supplemented with glutamate [16]. They identified pH 4.5 as optimal for GABA synthesis. In our experiments initiated at the same value, pH decreased to 4.0 in MRS and to 4.3 in MRS with MSG (Figure 2). It is well established that the activity of the gadC antiporter is strongly dependent on medium pH [50]. The gadC is activated at acidic conditions by starting to import glutamate and export GABA from the cell. The protein is inactivated at neutral pH, and at alkaline conditions, it starts to import glutamine instead of glutamate [50]. We can speculate that the slight increase in pH at the end of the cultivation process could also have some effects on the decrease in the gad gene expression, but additional research should be performed to confirm this. The analyses of our data demonstrated that the relative expression of the genes from the gad operon could not be considered as the only factor responsible for the GABA accumulation. Deeper study, including proteomics, should be performed to demonstrate the real presence of gadB, gadC, and gadR proteins and their activities, as well as the consideration of other factors, mainly the pH and substrate availability, should be studied.

5. Conclusions

The identification of GABA-producing lactic acid bacteria strains is of significant importance to the food industry, given that most LAB species are considered safe and are highly effective at fermenting diverse food matrices. This study represents the first comprehensive screening of Bulgarian LAB strains, some of which sourced from foods and plants, to assess their potential for GABA production. The screening identified for the first time L. brevis NCTC 13768 as a high-yield GABA producer, capable of producing more than 3300 mg/L of GABA. It was revealed that all genes related to GABA biosynthesis were upregulated during the cultivation of L. brevis NCTC 13768, but the concentration of MSG in the medium was identified as a key parameter for GABA production. Future research efforts will concentrate on optimizing fermentation parameters, including temperature, pH, and glutamate concentration, as well as scaling up production to facilitate industrial application of L. brevis NCTC 13768 in the production of GABA-enriched functional foods and beverages.