Enhanced accumulation of γ-aminobutyric acid by deletion of aminotransferase genes involved in γ-aminobutyric acid catabolism in engineered Halomonas elongata

ABSTRACT

Halomonas elongata OUT30018 is a moderately halophilic bacterium that synthesizes and accumulates ectoine as an osmolyte by activities of the enzymes encoded by the high salinity-inducible ectABC operon. Previously, we engineered a γ-aminobutyric acid (GABA)-producing H. elongata GOP-Gad (ΔectABC::mCherry-HopGadmut) from an ectoine-deficient mutant of this strain due to its ability to use high-salinity biomass waste as substrate. Here, to further increase GABA accumulation, we deleted gabT, which encodes GABA aminotransferase (GABA-AT) that catalyzes the first step of the GABA catabolic pathway, from the H. elongata GOP-Gad genome. The resulting strain H. elongata ZN3 (ΔectABC::mCherry-HopGadmut ΔgabT) accumulated 291 µmol/g cell dry weight (CDW) of GABA in the cells, which is a 1.5-fold increase from H. elongata GOP-Gad’s 190 µmol/g CDW. This result has confirmed the role of GABA-AT in the GABA catabolic pathway. However, redundancy in endogenous GABA-AT activity was detected in a growth test, where a gabT-deletion mutant of H. elongata OUT30018 was cultured in a medium containing GABA as the sole carbon and nitrogen sources. Because L-2,4-diaminobutyric acid aminotransferase (DABA-AT), encoded by an ectB gene of the ectABC operon, shares sequence similarity with GABA-AT, a complementation analysis of the gabT and the ectB genes was performed in the H. elongata ZN3 genetic background to test the involvement of DABA-AT in the redundancy of GABA-AT activity. Our results indicate that the expression of DABA-AT can restore GABA-AT activity in H. elongata ZN3 and establish DABA-AT’s aminotransferase activity toward GABA in vivo.

IMPORTANCE

In this study, we were able to increase the yield of GABA by 1.5 times in the GABA-producing H. elongata ZN3 strain by deleting the gabT gene, which encodes GABA-AT, the initial enzyme of the GABA catabolic pathway. We also report the first in vivo evidence for GABA aminotransferase activity of an ectB-encoded DABA-AT, confirming a longstanding speculation based on the reported in vitro GABA-AT activity of DABA-AT. According to our findings, the DABA-AT enzyme can catalyze the initial step of GABA catabolism, in addition to its known function in ectoine biosynthesis. This creates a cycle that promotes adequate substrate flow between the two pathways, particularly during the early stages of high-salinity stress response when the expression of the ectB gene is upregulated.

INTRODUCTION

Biotechnology for synthesizing medicine, chemicals, and food additives has gradually been developed to replace the traditional petroleum-based industry. Halomonas elongata is one of the bacterial strains used in microbial cell factories due to its ability to utilize different substrates and grow under high-salinity conditions (1–4).

H. elongata OUT30018 is a moderately halophilic eubacterium isolated from high-salinity soil in Northeastern Thailand (5). It grows in an environment containing 0.3% ~ 21% wt/vol NaCl by balancing osmotic equilibrium with the environment by producing, accumulating, and releasing a major osmolyte ectoine, which is synthesized by enzymes in ectoine biosynthesis pathway encoded by an ectABC operon (6). H. elongata OUT30018 can assimilate a variety of sugars and amino acids, including major putrefactive non-volatile amines, histamine, and tyramine, derived from biomass waste as carbon (C) and nitrogen (N) sources (3). Therefore, H. elongata OUT30018 is suitable for developing cell factories for the sustainable production of fine chemicals.

γ-Aminobutyric acid (GABA) is a non-protein amino acid widely found in plants, animals, and microorganisms. GABA is assimilated as carbon and nitrogen sources for growth by various fungi and bacteria, including Neurospora crassa (7), Rhizobium leguminosarum (8), Escherichia coli (9), and Corynebacterium glutamicum (10). GABA involves in the spore germination process in Bacillus megaterium (11) and Neurospora crassa (12), and through the function of the glutamate decarboxylase (GAD) system (13), acidic glutamate (Glu) can be converted to neutral GABA in response to acidic pH environment in E. coli (14), Lactobacillus reuteri (15), and Lactococcus lactis (16). GABA also functions as a major inhibitory neurotransmitter that produces a calming effect in humans (17, 18) and increases performance and physiological responses in farm animals, including poultry and swine (19, 20). Therefore, GABA is a commercially high-value compound of interest for the pharmaceutical and farming industries. Moreover, to mitigate global warming and marine plastic pollution, GABA is used to produce polyamide 4, which is recognized as a promising renewable plastic degradable in marine environments (21–23). To date, there are many successful applications of biotechnology for the production of GABA in recombinant microbial hosts, which are thoroughly summarized in the recent report by Son et al. (24). For example, a recombinant C. glutamicum strain CGY-PG-304, whose genome was modified for improved metabolic flux, could produce 41 g of GABA per 1 L of medium in shake flask cultivation (25). However, the growing concern about the rapidly degrading environment demands the creation of a GABA production process that is both sustainable and cost-effective.

Previously, we have explored the possibility of using H. elongata OUT30018 to biosynthesize valuable osmolytes, other than ectoine, by screening its ectoine-deficient salt-sensitive mutant strain H. elongata KA1 (ΔectABC) for spontaneous mutants with restored salt tolerance (26). As a result, we obtained a Glutamic acid Over-Producing (GOP) mutant strain H. elongata GOP, which attains higher salt tolerance than the H. elongata KA1 by overproducing and accumulating glutamic acid (Glu) as an osmolyte. However, the acidity of Glu impeded its cellular accumulation by interfering with cellular pH homeostasis. We created H. elongata GOP-Gad (ΔectABC::mCherry-HopGadmut), which expresses a glutamic acid decarboxylase (GAD) used in many bacteria to stabilize intracellular pH by converting the acidic Glu into a neutral GABA (13), thus turning the obstacle into an opportunity. We were able to restore cellular pH balance and enhance salt tolerance in H. elongata GOP-Gad cells by converting Glu into GABA, which was accumulated inside the cell as a major osmolyte. When cultured in an M63 minimal medium containing 7% wt/vol NaCl, the accumulation of GABA in H. elongata GOP-Gad was 177 µmol/g CDW (26).

Many bacteria can use GABA as a carbon and nitrogen source for growth through the enzymatic activities in the GABA catabolic pathway (9, 27, 28). In this pathway, GABA aminotransferase (GABA-AT) catalyzes the reversible interconversion between GABA and succinic semialdehyde (SSA), and succinic semialdehyde dehydrogenase (SSADH) converts SSA to succinate (SUC), which is taken up by the tricarboxylic acid (TCA) cycle (29). As shown previously, enhanced GABA accumulation can be achieved by inactivating the GABA catabolic pathway; for example, deletion of both the gabT gene and the putrescine-inducible puuE gene, which encode GABA-ATs, could improve GABA productivity in engineered E. coli (30, 31). Recently, based on structural modeling and biochemical validation, reaction mechanism of the L-2,4-diaminobutyric acid aminotransferase (DABA-AT) encoded by ectB gene for ectoine biosynthesis was reported, and it was shown that DABA-AT enzymes are mechanistically identical to the PLP-dependent GABA-AT enzymes and act in a way analogous to GABA-TA enzymes (32).

Here, we found that GABA-AT and SSADH are encoded by gabT and gabD genes on the genome of H. elongata OUT30018, and inactivation of the GABA catabolism could increase GABA accumulation in recombinant H. elongata strains. When the gabT gene was removed from the genome of H. elongata GOP-Gad (ΔectABC::mCherry-HopGadmut), the accumulation of GABA increased from 190 µmol/g CDW to 291 µmol/g CDW in H. elongata ZN3 (ΔectABC::mCherry-HopGadmut ΔgabT) when grown in a high-salinity minimal medium. We found that L-2,4-diaminobutyric acid aminotransferase (DABA-AT) has aminotransferase activity toward GABA, in addition to being the initial rate-limiting enzyme of the ectoine biosynthesis pathway (32). Our study provides in vivo evidence that supports the previously reported finding of in vitro DABA-AT’s aminotransferase activity toward GABA (33). Additionally, our work establishes the role of DABA-AT in GABA catabolism.

RESULTS

H. elongata OUT30018 can assimilate GABA as carbon and nitrogen sources

It is known that various bacteria can utilize GABA as the sole carbon (C) and nitrogen (N) sources via the GABA catabolic pathway (9, 27, 28). To test whether H. elongata OUT30018 can utilize GABA, we cultivated the strain in the M63 minimal medium containing 4% wt/vol GABA as the sole C and N sources. As shown in Fig. 1, the wild-type H. elongata OUT30018 could grow in the M63 medium containing glycerol as a C source and ammonium (NH₄⁺) as an N source and in the M63 medium containing GABA as the sole C and N sources. However, no growth was detected when the strain was grown in an M63 medium containing no C and N sources. Therefore, we conclude that H. elongata OUT30018 has an active GABA catabolic pathway (Fig. 2) and can rely solely on GABA as C and N sources for growth.

Fig 1.

Growth of H. elongata OUT30018 cultured in a M63 minimal medium containing 3% wt/vol NaCl and supplemented with different carbon (C) and nitrogen (N) sources. H. elongata OUT30018 was precultured in M63 medium containing 3% wt/vol NaCl with 4% wt/vol glycerol and 15 mM (NH₄)₂SO₄ as C and N sources until OD₆₀₀ reached around 0.7, then the cells were washed three times with a C, N-free M63 medium containing 3% wt/vol NaCl before it was used as a 5% vol/vol inoculum for the test cultures in C, N-free M63 medium containing 3% wt/vol NaCl (⚫), C, N-free M63 medium supplemented with 4% wt/vol glycerol and 15 mM (NH₄)₂SO₄ (▲), or with 4% wt/vol GABA (◆) as C and N sources.

Fig 2.

GABA catabolic and ectoine biosynthesis pathways in H. elongata OUT30018. H. elongata OUT30018 can utilize GABA as sole C and N sources for growth through TCA and GS/GOGAT cycles. Ac-CoA, acetyl coenzyme A; ADABA, Nγ-acetyl-2,4-diaminobutyric acid; α-KG, α-ketoglutaric acid; ASA, L-aspartate-β-semialdehyde; ASD, L-aspartate-semialdehyde dehydrogenase; Asp, L-aspartic acid; Asp-AT, L-aspartic acid aminotransferase; ASK, aspartate kinase; CoA, coenzyme A; DABA, L-2,4-diaminobutyric acid; DABA-AcT, L-2,4-diaminobutyric acid acetyltransferase; DABA-AT, L-2,4-diaminobutyric acid aminotransferase; Ect, ectoine; ES, ectoine synthase; GABA, γ-aminobutyric acid;GABA-AT, γ-aminobutyric acid aminotransferase; Glu, L-glutamic acid; GDH, L-glutamic acid dehydrogenase; GOGAT, L-glutamic acid synthetase; Gln, L-glutamine; GS: L-glutamine synthetase; OAA, oxaloacetate; 4PAsp, 4-phospho-L-aspartate; SSA, succinic semialdehyde; SSADH, succinic semialdehyde dehydrogenase; and SUC, succinate.

Identification of essential genes involved in GABA catabolic pathway in H. elongata OUT30018

In bacteria, genes involved in the GABA catabolic pathway were identified as various gab gene clusters, such as the gabDTPC cluster of E. coli (9) and the gabTRD cluster of Bacillus thuringiensis (34). As essential gab genes, gabT and gabD encode the enzymes working in the first two steps of the GABA catabolic pathway, a GABA aminotransferase (GABA-AT), which catalyzes a reversible interconversion between GABA and succinic semialdehyde (SSA), and a succinic semialdehyde dehydrogenase (SSADH), which converts SSA to succinate (SUC) (Fig. 2) (29).

By homology search with E. coli’s gabT sequence, we found a gabT gene on the genome of H. elongata OUT30018. On the 3’ end of the gabT gene, we found a gabD gene, forming a gabDT gene cluster (Fig. 3A ). To establish the involvement of the gabDT gene cluster in the GABA catabolic pathway, we generated H. elongata ZN1 (ΔgabDT) and ZN2 (ΔgabT) strains by deleting both the gabD and the gabT genes or the gabT gene alone from the genome of the H. elongata OUT30018 (Fig. 3B). We tested the ability of these strains to grow in an M63 minimal medium containing 4% wt/vol GABA as the sole C and N sources. As shown in Fig. 4, photos of the cultures taken on day 2 indicated that deletion of the gabT gene alone only partially affected the growth of the H. elongata ZN2 cells. In contrast, double deletion of gabD and gabT genes severely inhibited the growth of the H. elongata ZN1 cells. Moreover, as shown in Fig. 5, changes in the concentration of GABA in the culture media confirmed that H. elongata OUT30018 and ZN2 cultures can efficiently utilize GABA and grow with shorter lag time, whereas H. elongata ZN1 culture has a significantly longer lag time. These results indicate that H. elongata OUT30018 has a functional GABA catabolic pathway, in which SSADH encoded by gabD is a key enzyme responsible for GABA catabolism in the initial growth stage. Notably, ability of the H. elongata ZN2 strain to eventually use GABA as the sole C and N sources for growth indicates that GABA-AT encoded by the gabT gene is not the only enzyme with aminotransferase activity toward GABA in H. elongata OUT30018.

Fig 3.

Schematic diagram of the gabDT gene cluster on H. elongata OUT30018 genome and schematic diagram of the genomic structure at the ectABC and the gabDT loci in H. elongata OUT30018, ZN1, ZN2, GOP-Gad, and ZN3 strains. (A) Schematic diagram of the gabDT gene cluster on H. elongata OUT30018 genome. (B) Schematic diagram of the genomic structure at the ectABC and the gabDT loci in H. elongata OUT30018, ZN1, ZN2, GOP-Gad, and ZN3 strains. U_ectA: A 1 kb upstream region of the ectA gene, which contains an ectA promoter with putative binding sites for the osmotically induced sigma factor σ38, and the vegetative sigma factor σ70. This region was used as a target for homologous recombination at the ectABC locus. D_ectC: A 1 kb downstream region of the ectC gene, which contains an ectC terminator. This region was used as a target for homologous recombination at the ectABC locus. ectA: A gene encoding an L-2,4-diaminobutyric acid (DABA) acetyltransferase (DAA). ectB: A gene encoding a DABA transaminase (DABA-AT). ectC: A gene encoding an ectoine synthase (ES). U_gabD: A 994 bp upstream region of the gabD gene. This region was used as a target for homologous recombination at the gabDT locus. U_gabT: A 1,158 bp upstream region of the gabT gene. This region was used as a target for homologous recombination to delete the gabT gene from H. elongata genome. D_gabT: A 1,091 bp downstream region of the gabT gene. This region contains a gabT terminator and was used as a target for homologous recombination at the gabDT locus. mCherry: A gene encoding a red fluorescent reporter protein mCherry. HopGadBmut: A synthetic H. elongata’s codon-usage optimized gadB mutant gene encoding a mutant glutamate decarboxylase (GAD) with activity across a broader pH range than the wild-type GAD. gabT: A gene encoding GABA aminotransferase (GABA-AT). gabD: A gene encoding succinic semialdehyde dehydrogenase (SSADH).

Fig 4.

Growth comparison between H. elongata OUT30018, ZN1, and ZN2 strains cultures in C, N-free M63 medium containing 3% wt/vol NaCl with or without GABA as C and N sources. H. elongata OUT30018 was precultured in M63 medium containing 3% wt/vol NaCl with 4% wt/vol glycerol and 15 mM (NH₄)₂SO₄ as C and N sources until OD₆₀₀ reached 0.8, then the cells were washed three times with C, N-free M63 medium containing 3% wt/vol NaCl before it was used as a 5% vol/vol inoculum for test cultures in C, N-free M63 medium containing 3% wt/vol NaCl and supplemented with different substrates as C and N sources. Photos of the cultures were taken on day 3 to show differences in cell density.

Fig 5.

Correlation between GABA consumption and growth of H. elongata OUT30018, ZN1, and ZN2 strains cultured in M63 medium containing 3% wt/vol NaCl with GABA as C and N sources. H. elongata OUT30018 was precultured in M63 medium containing 3% wt/vol NaCl with 4% wt/vol glycerol and 15 mM (NH₄)₂SO₄ as C and N sources until OD₆₀₀ reached 0.8, then the cells were washed three times with C, N-free M63 medium containing 3% wt/vol NaCl before it was used as a 5% vol/vol inoculum for test cultures in C, N-free M63 medium containing 3% wt/vol NaCl and supplemented with 4% wt/vol GABA as C and N sources. GABA concentration of the medium was analyzed by HPLC, and OD₆₀₀ of the cultures was measured every 24 h. ◾, GABA concentration; ⚫, OD₆₀₀. Data were normalized with internal standard norvaline. Values are mean ± standard deviation (n = 5). The average initial OD₆₀₀ values of H. elongata OUT30018, ZN1, and ZN2 cultures were 0.04, 0.03, and 0.03, respectively. (A) Correlation between GABA consumption and growth of H. elongata OUT30018. (B) Correlation between GABA consumption and growth of H. elongata ZN1. (C) Correlation between GABA consumption and growth of H. elongata ZN2.

The impact of gabT gene deletion was studied by analyzing changes in the ectoine concentration in the cells of H. elongata OUT30018 and H. elongata ZN2 (ΔgabT) during the initial phase of high-salinity stress adaptation in an M63 medium containing 4% wt/vol GABA as the sole C and N sources. In H. elongata OUT30018 cells, ectoine accumulation increased over 2-fold within 6 h of transfer to high-salinity minimal media containing 4% wt/vol GABA and 6% or 12% wt/vol NaCl (Fig. 6A and C). This result shows that ectoine was efficiently synthesized using C and N sources from GABA (Fig. 6A and C). In H. elongata ZN2 (ΔgabT), ectoine accumulation levels decreased to one-third of the initial level within 6 h of transfer to a no-salt stress medium containing 3% wt/vol NaCl. This outcome reveals that ectoine is utilized as a nutrient in cells with inhibited GABA-AT activity (Fig. 6B and D).

Fig 6.

Effect of the deletion of the gabT gene on the amount of ectoine accumulated in H. elongata OUT30018 and H. elongata ZN2 (ΔgabT) during the early stage of high-salinity-stress adaptation. H. elongata OUT30018 and H. elongata ZN2 (ΔgabT) were precultured in an M63 minimal medium containing 3% wt/vol NaCl, 4% wt/vol glycerol as a C source, and 15 mM (NH₄)₂SO₄ as an N source until OD₆₀₀ reached 1.0. Cell pellet harvested from 2 mL aliquot of the precultures was transferred into a fresh M63 minimal medium containing 3%, 6%, or 12% wt/vol NaCl and 4% wt/vol GABA as the sole C and N sources and cultured for 6 h. Ectoine extracted from cell pellets harvested at 0, 3, and 6 h was analyzed by HPLC. (A and B) Ectoine concentration in H. elongata OUT30018 (A) and H. elongata ZN2 (ΔgabT) (B) cells grown in M63 minimal medium containing 3% (light-green bars), 6% (medium-green bars), or 12% (dark-green bars) wt/vol NaCl and 4% wt/vol GABA as the sole C and N sources for 0, 3, and 6 h. The experiments were performed in triplicate (n = 3). Bars display means ± standard deviations with individual data points indicated with black dots. (C and D) Relative ectoine accumulation level in H. elongata OUT30018 (C) and H. elongata ZN2 (ΔgabT) (D) cells grown in M63 minimal medium containing 3% (light-green bars), 6% (medium-green bars), or 12% (dark-green bars) wt/vol NaCl and 4% wt/vol GABA as the sole C and N sources for 0, 3, and 6 h. The relative concentration of ectoine was calculated as a percentage of that obtained at the start of the culture (0 h = 100%).

Interestingly, within 6 h after being transferred to a high-salinity medium containing 12% wt/vol NaCl, ectoine accumulated in H. elongata ZN2 (ΔgabT) increased more than 1.7-fold. This result indicates that ectoine was efficiently synthesized in H. elongata ZN2 (ΔgabT) using C and N sources from GABA, even in the absence of the gabT gene. It was observed that an increase in ectoine accumulation occurred only when cells of H. elongata ZN2 (ΔgabT) were cultured in a high-salinity medium containing 12% wt/vol NaCl. This increase could be due to the higher expression of the salt-inducible ectABC operon or the activation of high-salinity-inducible GABA-AT activity, encoded by a gene other than gabT, which is yet to be identified.

Deletion of the gabT gene from the H. elongata GOP-Gad (ΔectABC::mCherry-HopGadmut) genome confirmed its involvement in GABA catabolism and conferred higher GABA accumulation to the resulting H. elongata ZN3 (ΔectABC::mCherry-HopGadmut ΔgabT)

To address a possibility of redundancy in aminotransferase activity toward GABA in H. elongata OUT30018, we used the peptide sequence of GABA-AT encoded by the gabT gene of H. elongata OUT30018 as a query for homology search against the protein database (BlastP). One interesting match from the result is an L-2,4-diaminobutyric acid aminotransferase (DABA-AT), which is encoded by an ectB gene of the ectoine biosynthetic ectABC operon (6). DABA-AT is the initial rate-limiting enzyme of the ectoine biosynthesis pathway (32), which converts aspartate β-semialdehyde (ASA) to DABA (Fig. 2). Among the three enzymes of the ectoine biosynthesis pathway, DABA-AT was detected at a higher amount than the other two enzymes under a high-salinity condition of 12% wt/vol NaCl, and a higher amount of DABA-AT protein was found to improve ectoine production in bacterial cells (35, 36). Furthermore, there is evidence that DABA-AT in the thermos-tolerant Gram-positive bacterium Paenibacillus lautus has aminotransferase activity toward GABA in vitro (33).

We observed that H. elongata ZN2 (ΔgabT, Fig. 4) and its ectoine-deficient mutant H. elongata ZN4 (ΔectABC ΔgabT; data not shown) could grow at the same rate as the wild-type H. elongata OUT30018 in M63 minimal medium containing 4% wt/vol GABA as the sole C and N sources. Therefore, the effect of the gabT or the ectB gene deletion cannot be demonstrated in H. elongata OUT30018. Instead, we decided to study the effect of gabT deletion in the GABA-producing H. elongata GOP-Gad (ΔectABC::mCherry-HopGadmut, Fig. 7A) previously generated in our laboratory (26) based on three reasons. First, the ectB gene is already absent in H. elongata GOP-Gad (Fig. 3), which was generated by engineering a GAD system that converts Glu to GABA into H. elongata GOP, which is an ectoine-deficient (ΔectABC) Glu-overproducing mutant of H. elongata OUT30018; second, H. elongata GOP-Gad produces and accumulates GABA in the cells. Therefore, changes in GABA catabolism in the cell can be analyzed quantitatively; and third, If the gabT-encoded GABA-AT has aminotransferase activity toward GABA, the deletion of gabT would suppress GABA catabolism and result in increased GABA accumulation, making it a better GABA-producing cell factory. By deleting the gabT gene from the H. elongata GOP-Gad genome, we generated H. elongata ZN3 (ΔectABC::mCherry-HopGadmutΔgabT) (Fig. 3), and the amount of GABA accumulated in H. elongata GOP-Gad and ZN3 cells grown in an M63 medium with 6% wt/vol NaCl and 4% wt/vol glycerol were compared. As expected, GABA accumulation was improved from 58 ± 5 µmol/g CFW (equivalent to 190 ± 29 µmol/g CDW) in H. elongata GOP-Gad to 97 ± 15 µmol/g CFW (equivalent to 291 ± 30 µmol/g CDW) in H. elongata ZN3 strain (Fig. 7B). This result provides in vivo evidence that the GABA-AT enzyme of H. elongata OUT30018 can catalyze the first step of the GABA catabolic pathway.

Fig 7.

GABA metabolic pathway in H. elongata GOP-Gad and profiles of major osmolytes accumulated in the cells of H. elongata GOP-Gad and ZN3 strains grown in M63 medium containing 6% wt/vol NaCl. (A) GABA metabolic pathway in H. elongata GOP-Gad. This strain is engineered to contain a salt-inducible artificial mCherry-HopGadBmut operon, which encodes a mCherry reporter protein and a wide pH-range glutamic acid decarboxylase (GAD) that converts Glu accumulated in H. elongata GOP-Gad into GABA. GABA is catabolized by the enzymes GABA-AT and SSADH into SUC, which enters the TCA cycle as a C source, and Glu, which enters the GS/GOGAT cycle as the N source. Ac-CoA, acetyl coenzyme A; Ala, L-alanine; Ala-AT, L-alanine aminotransferase; α-KG, α-ketoglutaric acid; ASA, L-aspartate-β-semialdehyde; Asp, L-aspartic acid; Asp-AT, L-aspartic acid aminotransferase; CoA, coenzyme A; DHAP, dihydroxyacetone phosphate; GABA, γ- aminobutyric acid;GABA-AT, γ-aminobutyric acid aminotransferase; GAD, L-glutamic acid decarboxylase; Glu, L-glutamic acid; GDH, L-glutamic acid dehydrogenase; GOGAT, L-glutamic acid synthetase; Gln, L-glutamine; GS: L-glutamine synthetase; Gly3P, glycerol 3-phosphate.OAA, oxaloacetate; PYR, pyruvate; SSA, succinic semialdehyde; SSADH, succinic semialdehyde dehydrogenase; and SUC, succinate. (B) Profiles of major osmolytes in the cells of H. elongata GOP-Gad and ZN3 strains grown in M63 medium containing 6% wt/vol NaCl. The intracellular concentration of the major osmolytes, glutamate (Glu), alanine (Ala), and GABA in H. elongata GOP-Gad and ZN3 (GOP-Gad-ΔgabT) strains cultured in M63 medium containing 6% wt/vol NaCl and 4% wt/vol glycerol were profiled. Precultures grew in M63 medium containing 6% wt/vol NaCl and 4% wt/vol glycerol until OD₆₀₀ reached 1.0 to 1.2 was used as a 5% vol/vol inoculum for main cultures in M63 medium containing 6% NaCl and 4% glycerol. When OD₆₀₀ of the main cultures reached 1.0 to 1.2, major osmolytes in the cells were extracted and analyzed by HPLC. Data were normalized with internal standard norvaline. Five cultures were grown for each condition (n = 5). Bars display means ± standard deviations with individual data points indicated with black dots. CDW, cell dry weight; Glu, blue bars; Ala, pink bars; GABA, purple bars. Asterisk (*) indicates a statistically significant result with P ≤ 0.005.

DABA-AT exhibits aminotransferase activity toward GABA in vivo

Although DABA-AT shares sequence similarity with GABA-AT, and their substrates have similar structural features (Fig. 9B), there is still no proof that DABA-AT has aminotransferase activity toward GABA. To obtain the proof, we employed a gain-of-function approach in a complementation assay, in which H. elongata ZN3 was transformed with pHS15N (empty shuttle plasmid; negative control), pHS-gabT-HA (shuttle plasmid that expresses GABA-AT-HA; positive control), and pHS-ectB-HA (shuttle plasmid that expresses DABA-AT-HA) to generate H. elongata ZN3-1, ZN3-2, and ZN3-3 strains. To confirm the correct expression and translation of the HA-tagged recombinant GABA-AT-HA and DABA-AT-HA proteins, crude protein extracts of H. elongata ZN3-1, ZN3-2, and ZN3-3 cells were separated by sodium dodecyl-sulfate polyacrylamide gel electrophoresis (SDS-PAGE), and amount of protein loaded in each lane was visualized with Coomassie Brilliant Blue staining; HA-tagged proteins were visualized by immunodetection with HRP-conjugated anti-HA antibody and 3'-diaminobenzidine (DAB) chromogen substrate. As shown in Fig. 8A, both GABA-AT-HA (47.9 kDa) and DABA-AT-HA (48.1 kDa) proteins were correctly expressed, whereas HA-tagged protein was not detected in the lane loaded with crude protein extracts of the negative control H. elongata ZN3-1. Subsequently, the amount of GABA accumulated in H. elongata ZN3-1, ZN3-2, and ZN3-3 grown in high-salinity LB medium containing 12% wt/vol NaCl and 60 mg/L streptomycin for maintenance of the plasmid were compared. As shown in Fig. 8B, the amount of GABA accumulated in H. elongata ZN3-2 (21 ± 4 µmol/g CFW) and ZN3-3 (24 ± 1.5 µmol/g CFW) strains were significantly lower than that of the negative control H. elongata ZN3-1 strain (43 ± 4.6 µmol/g CFW). This result indicates that both GABA-AT-HA and DABA-AT-HA recombinant proteins can restore GABA catabolic activity in H. elongata ZN3-2 and ZN3-3 cells (Fig. 8). This is probably due to their aminotransferase activities, which catalyze the initial step in the GABA catabolic pathway.

Fig 8.

Complementation analysis to verify GABA aminotransferase activity of GABA-AT and DABA-AT enzymes. H. elongata ZN3 was transformed with an empty pHS15N E. coli-H. elongata shuttle plasmid vector (negative control), a pHS-gabT-HA plasmid containing gabT-HA expression cassette, or a pHS-ectB-HA plasmid containing ectB-HA expression cassette to generate H. elongata ZN3-1, ZN3-2, and ZN3-3 strains. Aminotransferase activity toward GABA of GABA-AT and DABA-AT, encoded by gabT-HA and ectB-HA genes, was verified by measuring the amount of GABA accumulated in H. elongata ZN3-2 and ZN3-3 strains in comparison to the amount accumulated in the H. elongata ZN3-1 strains. Precultures were grown in LB medium containing 12% wt/vol NaCl and 60 mg/L streptomycin until OD₆₀₀ reached more than 1.0 and used as a 5% vol/vol inoculum for the main culture in the same medium. Cells were harvested when OD₆₀₀ of the main culture reached more than 1.0 for protein extraction and analysis of major osmolytes. (A) Immunodetection of GABA-AT-HA and DABA-AT-HA proteins produced in the H. elongata ZN3-2 and ZN3-3 strains. Crude proteins extracted from each strain were electrophoresed in 2 identical 5%–20% wt/vol gradient SDS-Polyacrylamide gels. One gel was stained with Coomassie Brilliant Blue (CBB; left panel) for visualization of total proteins separated on each lane, whereas proteins on the other gel were transferred to PVDF membrane and probed with Anti-HA-HRP conjugated antibody (IB, right panel). Protein extracted from H. elongata ZN3-1 was used as a negative control. GABA-AT-HA and DABA-AT-HA protein bands were detected at 47.9 and 48.1 kDa as indicated. (B) Profile of major osmolytes in H. elongata ZN3-1, ZN3-2, and ZN3-3 cells analyzed by HPLC. Data were normalized with internal standard norvaline. Five cultures were grown in parallel for each condition (n = 5). Bars display means ± standard deviations with individual data points indicated with black dots. CDW, cell dry weight; Glu, blue columns; Ala, pink columns; GABA, purple columns. Asterisks (*) indicate statistically significant results with P ≤ 0.005.

DISCUSSION

Previously, we have generated a salt-inducible GABA-producing H. elongata GOP-Gad strain (ΔectABC::mCherry-HopGadmut) as a prototype for developing a GABA-producing H. elongata cell factory (26). To further increase GABA accumulation in H. elongata GOP-Gad, we took an approach to suppress GABA catabolism in the cells (Fig. 7). Although it is known that many kinds of bacteria including Gram-negative bacterium E. coli (9) and Gram-positive bacterium C. glutamicum (27, 28) can utilize GABA as carbon (C) and nitrogen (N) sources for growth, it was unclear whether H. elongata has that ability. Therefore, we started with an experiment to test the GABA assimilation ability of the wild-type H. elongata OUT30018 strain, and the result shown in Fig. 1 confirms that H. elongata OUT30018 can efficiently assimilate GABA as the sole C and N sources for growth. By sequence homology search, we found gabT and gabD genes, which encode GABA-AT and SSADH in the H. elongata OUT30018 genome. These genes are clustered together in a gabDT operon (Fig. 3A). As shown in Fig. 4 and 5, double deletion of both gabT and gabD genes suppressed the growth of H. elongata ZN1 in an M63 minimal media, which contained GABA as the sole C and N sources. However, the H. elongata ZN2 strain with a gabT deletion and an intact gabD could assimilate GABA as efficiently as the wild-type H. elongata OUT30018 strain. This result indicates that H. elongata OUT30018 has a functional GABA catabolic pathway, in which SSADH encoded by the gabD gene is a crucial enzyme; however, GABA-AT encoded by the gabT gene could be dysfunctional, or there could be redundancy in aminotransferase activity toward GABA in H. elongata OUT30018. There are other GABA-AT encoding genes, such as the PuuE gene, which is involved in the putrescine degradation pathway in E. coli (9, 37), and the BioA gene in C. glutamicum (27).

To identify the enzymes responsible for the redundancy of GABA-AT activity observed in H. elongata OUT30018, we searched the protein database with the peptide sequence of GABA-AT. As a result, we found that DABA-AT, which is encoded by the ectB gene of the high-salinity inducible ectoine biosynthetic ectABC operon, shares sequence similarity with the gabT-encoded GABA-AT. Interestingly, previous biochemical characterization of P. lautus’s DABA-AT suggested that this enzyme could also function as a GABA-AT in vitro (33). Therefore, we decided to investigate whether DABA-AT has aminotransferase activity toward GABA.

Because the H. elongata GOP-Gad (ΔectABC::mCherry-HopGadmut) produces and accumulates GABA in the cell, it has a unique character that allows for easy detection of changes in intracellular GABA concentration. Moreover, H. elongata GOP-Gad is generated from H. elongata GOP that lacks the ectABC operon (26), making it a suitable host strain for the functional analysis of gabT and ectB genes. Here, we generated a gabT and ectB double-deletion strain by deleting the gabT gene from the H. elongata GOP-Gad genome, creating H. elongata ZN3 (ΔectABC::mCherry-HopGadmutΔgabT). Using the H. elongata ZN3 strain, we attempt to verify the involvement of gabT in GABA catabolism. Moreover, if gabT functions as a major GABA-AT that catalyzes the first step of the GABA catabolic pathway, the resulting H. elongata ZN3 would have improved GABA accumulation. As a result, GABA-AT’s function in the GABA catabolic pathway was confirmed as the H. elongata ZN3 accumulated GABA to 291 µmol/g CDW, which is over 1.5-fold increase from 190 µmol/g CDW accumulated in the H. elongata GOP-Gad strain (Fig. 6). Result from complementation analysis, where the HA-fusion gabT-HA and ectB-HA genes separately expressed in the H. elongata ZN3-2 and ZN3-3, further confirms the function of GABA-AT in GABA catabolism and provides the first in vivo evidence for DABA-AT’s aminotransferase activity toward GABA (Fig. 8). Remarkably, our result suggests that DABA-AT might be as active as GABA-AT in catalyzing the first step of the GABA catabolic pathway as comparable expression of the recombinant GABA-AT-HA and DABA-AT-HA proteins (Fig. 8A) caused comparable decrease in GABA accumulation in H. elongata ZN3-2 and ZN3-3 strains (Fig. 8B). With this result, we proposed a hypothetical model, which suggests that DABA-AT could function in the GABA catabolic pathway in addition to its primary function in the ectoine biosynthetic pathway. (Fig. 9A). In H. elongata OUT30018, the expression of the ectB gene is controlled by a high-salinity-inducible ectA promoter of the ectoine biosynthetic ectABC operon (Fig. 3) (35, 38). Therefore, in a high-salinity medium that contains GABA, DABA-AT activity in both GABA catabolism and ectoine biosynthesis would be further activated to allow an efficient flow of substrates between the two pathways for a more active ectoine biosynthesis. This model could explain the increase in ectoine concentration in H. elongata ZN2 (ΔgabT) cells grown in the M63 minimal medium containing 12% wt/vol NaCl and 4% wt/vol GABA as the sole C and N sources.

Fig 9.

Schematic diagram of a model depicting the dual function of DABA-AT in GABA catabolism and ectoine biosynthesis in H. elongata. (A) Schematic diagram of a model depicting the dual function of DABA-AT in GABA catabolism and ectoine biosynthesis in H. elongata. Based on the result of the complementation test shown in Fig. 8, we propose that DABA-AT has aminotransferase activity that can catalyze the first step of both the GABA catabolic and the ectoine biosynthesis pathways. Because the expression of the ectoine biosynthetic ectABC operon is inducible by high-salinity stress, the activity of DABA-AT would be upregulated during the high-salinity stress response, allowing the efficient flow of substrates between the two pathways for active ectoine biosynthesis.Ac-CoA, acetyl coenzyme A; ADABA, Nγ-acetyl-2,4-diaminobutyric acid; Ala, L-alanine; Ala-AT, L-alanine aminotransferase; α-KG, α-ketoglutaric acid; ASA, L-aspartate-β-semialdehyde; ASD, L-aspartate-semialdehyde dehydrogenase; Asp, L-aspartic acid; Asp-AT, L-aspartic acid aminotransferase; ASK, aspartate kinase; CoA, coenzyme A; DABA, L-2,4-diaminobutyric acid; DABA-AcT, L-2,4-diaminobutyric acid acetyltransferase; DABA-AT, L-2,4-diaminobutyric acid aminotransferase; Ect, ectoine; ES, ectoine synthase; GABA, γ-aminobutyric acid; GABA-AT, γ-aminobutyric acid aminotransferase; Glu, L-glutamic acid; GDH, L-glutamic acid dehydrogenase; GOGAT, L-glutamic acid synthetase; Gln, L-glutamine; GS: L-glutamine synthetase; OAA, oxaloacetate; 4PAsp, 4-phospho-L-aspartate; SSA, succinic semialdehyde; SSADH, succinic semialdehyde dehydrogenase; and SUC, succinate. (B) Amino transferase reactions catalyzed by GABA-AT and DABA-AT. The red color highlights the similarities between their substrates' chemical structures.

Although H. elongata ZN4 (ΔectABC ΔgabT) lacks both the ectB and the gabT genes, it is still able to grow in an M63 minimal medium that contains 4% wt/vol GABA as the sole source of carbon and nitrogen (data not shown). These unpublished data suggest that other enzymes may be responsible for the remaining redundancy. Identifying these enzymes could improve both GABA yield and understanding of the GABA catabolic pathway. Although H. elongata GOP (Glu Over Producing) accumulates more Glu than the WT strain, the amount of Glu in H. elongata GOP was still relatively low. Because Glu is the substrate for GABA biosynthesis, we are also interested in increasing Glu supply in H. elongata cells via metabolic engineering. Due to its genetic background, H. elongata ZN3 can be developed to upcycle high-salinity nitrogen-rich waste biomass into a GABA-rich single-cell eco feed, thus contributing to the sustainable development of the feedstock industry.

MATERIALS AND METHODS

Bacterial strains, plasmids, and growth conditions

Bacterial strains and plasmids used in this study are listed in Tables 1 and 2, respectively. For routine bacterial cultures, construction of plasmids, and generation of recombinant strains, Luria-Bertani (LB) medium (39) was used. LB medium contains 10 g/L Bacto-tryptone, 5 g/L Bacto-yeast extract, and 10, 20, or 60 g/L NaCl to generate media with different salt-stress levels. LB medium containing 1% wt/vol NaCl was used for routine culturing of E. coli. LB medium containing 2% wt/vol NaCl was used for triparental mating to generate H. elongata ZN1, ZN2, and ZN3 strains. LB medium containing 6% wt/vol NaCl was used for routine culture of H. elongata strains. Solid LB medium was supplemented with 15 g/L agar. The antibiotics, kanamycin (Kan), ampicillin (Amp), or streptomycin (Stm) were added to culture media as selection markers and for maintaining the plasmids containing the marker genes in recombinant E. coli strains (50 mg/L for Kan, 100 mg/L for Amp, 60 mg/L for Stm) and recombinant H. elongata strains (100 mg/L Kan, 60 mg/L for Stm). For the selection of the counter-selectable marker gene (sacB) in H. elongata, 150 g/L sucrose was added to a solid LB medium containing 6% wt/vol NaCl. For the selection of H. elongata harboring pHS15N, pHS-ectB-HA, or pHS-gabT-HA plasmids, 120 mg/L Stm was added to a solid LB medium containing 6% wt/vol NaCl. E. coli and H. elongata strains were cultured in the liquid or on the solid media for 17–24 h at 37°C. Liquid cultures were aerated by shaking in a water bath at 120 rpm.

TABLE 1.

List of bacterial strains used in this study

Strain	Relevant phenotype or description	Source or reference
Halomonas elongata
OUT30018	Wild-type strain (Osaka University Type culture formerly designated as KS3); salt-tolerant phenotype due to production and accumulation of ectoine as a major osmolyte	(5)
ZN1	OUT30018 strain with deletion of gabDT operon, which encodes GABA aminotransferase (GABA-AT encoded by gabT) and succinic semialdehyde dehydrogenase (SSADH encoded by gabD) involved in GABA catabolism	This study
ZN2	OUT30018 strain with the deletion of gabT gene encoding GABA-AT involved in catabolism	This study
GOP-Gad	Ectoine-deficient, Glu-overproducing strain with artificial mCherry-HopGadBmut operon encoding a red fluorescent protein mCherry and a mutant glutamate decarboxylase; salt-tolerant phenotype due to the ability to covert Glu into GABA, which is accumulated as a major osmolyte	(26)
ZN3	GOP-Gad strain with the deletion of gabT gene encoding GABA-AT involved in GABA catabolism	This study
ZN3-1	ZN3 strain harboring the empty pHS15N plasmid	This study
ZN3-2	ZN3 strain harboring the pHS-gabT-HA plasmid for expression of HA-tagged GABA-AT	This study
ZN3-3	ZN3 strain harboring the pHS-ectB-HA plasmid for expression of HA-tagged DABA-AT	This study
Escherichia coli
DH5α	Host for pUC57-Kan-based and pK18mobsacB-based plasmids, pHS15N plasmid; F-, Φ80dlacZΔM15, Δ(lacZYA-argF) U169, hsdR17(rk- mk+), recA1, endA1, relA, deoR, supE44, thi-1, gyrA96, λ-	(40)
HB101	Used as host for pRK2013 plasmids, F-, hsd S20(rB-, mB-), recA13, ara-14, proA2, lacY1, galK2, rpsL20 (str), xyl-5, mtl-1, supE44, leuB6, thi-1	(41)

TABLE 2.

List of plasmids used in this study

Plasmid	Description	Source or reference
pK18mobsacB	Suicide plasmid vector, which allows for selection of double crossover in H. elongata; Kanr, mob, sacB	(42)
pK18mobsacB-UgabD-DgabDT	pK18mobsacB with UgabD and DgabDT fragments for deletion of gabDT operon from the genome of H. elongata OUT30018; Kanr	This study
pK18mobsacB-UgabT-DgabT	pK18mobsacB with UgabT and DgabT fragments for deletion of gabT gene from the genome of H. elongata OUT30018; Kanr	This study
pUC19	Standard cloning vector for E. coli; Ampr	(43)
pUC19-gabT	pUC19 containing a PCR-amplified NdeI-gabT-SpeI-NheI fragment at SmaI site; Ampr	This study
pUC19-ectB	pUC19 containing a PCR-amplified NdeI-ectB-SpeI-NheI fragment at SmaI site; Ampr	This study
pET-Lipop5-HA	pET15b derivative plasmid containing a NcoI-NdeI-HeLipop5-HA-BamHI fragment; used for sub-cloning of the gabT or ectB gene; Ampr	(44)
pET-gabT-HA	Subcloning plasmid, in which the NdeI-Lipop5- SpeI fragment of pET-Lipop5-HA was replaced with NdeI-gabT-SpeI fragment from pUC19-gabT; contains a chimeric gabT-HA gene; Ampr	This study
pET-ectB-HA	Subcloning plasmid, in which the NdeI-Lipop5- SpeI fragment of pET-Lipop5-HA was replaced with NdeI-ectB-SpeI fragment from pUC19-ectB; contains a chimeric ectB-HA gene; Ampr	This study
pHS15N	E. coli-H. elongata shuttle vector Ampr and Stmr	(44)
pHS-gabT-HA	pHS15N plasmid digested with SpeI and BamHI and ligated with XbaI-gabT-HA-BamHI fragment from pET-gabT-HA; containing a chimeric gabT-HA gene for production of HA-tagged GABA-AT in H. elongata; Ampr, Stmr	This study
pHS-ectB-HA	pHS15N plasmid digested with SpeI and BamHI and ligated with XbaI-ectB-HA-BamHI fragment from pET-ectB-HA; containing a chimeric ectB-HA gene for production of HA-tagged DABA-AT in H. elongata; Ampr, Stmr	This study
pRK2013	Used as a mobilizing helper plasmid in triparental conjugation; Kanr	(45)

For growth analysis and osmolyte profiling, H. elongata strains were cultured in M63 minimal medium (46), which consisted of 100 mM KH₂PO₄, 15 mM (NH₄)₂SO₄, 1 mM MgSO₄, 3.9 µM FeSO₄, 3% wt/vol or 6% wt/vol NaCl, and 4% wt/vol glycerol as the sole carbon (C) source. For C, N-free medium, 4% wt/vol glycerol and 15 mM (NH₄)₂SO₄ were omitted. For the medium with GABA as the sole C and N sources, 4% wt/vol GABA was added to the C, N-free medium. pH of the media was adjusted to 7.2 with 1 M KOH. H. elongata strains were precultured in M63 medium supplemented with 3% or 6% wt/vol NaCl and 4% wt/vol glycerol until OD₆₀₀ reached 0.8 or higher before being used as a 5% vol/vol inoculum for the main cultures. For culturing of recombinant H. elongata harboring pHS15N, pHS-ectB-HA, or pHS-gabT-HA plasmid, 60 mg/L Stm was added to M63 containing 3% or 6% wt/vol NaCl.

Recombinant DNA constructions

Primers used in this study are synthesized by Eurofins Genomics K. K., Tokyo, Japan, and sequences of the primers are listed in Table 3. PCR reactions were performed with PrimeSTAR HS DNA Polymerase (Takara Bio Inc., Shiga, Japan) or Quick Taq HS DyeMix (Toyobo, Osaka, Japan).

TABLE 3.

List of oligonucleotide primers used in this study

Primer name	Sequence (5'−3')a	Description	Restriction enzyme sites	PCR product size (bp)
UgabT-F	CTCTAGAGGATCCCCGCGCCAGCTTCGTCGAGTTCTAC	Forward primer for amplifying upstream region of gabT of H. elongata OUT30018 genome (UgabT)		1,180
UgabT-R	TGGCTCACATGGCTGTCTCCTGGGATCGAAAATATG	Reverse primer for amplifying UgabT
DgabT-F	CAGCCATGTGAGCCAGGATCTGATCTTGCTGCC	Forward primer for amplifying downstream region of gabT of H. elongata OUT30018 genome (DgabT)		1,114
DgabT-R	TCGAGCTCGGTACCCAGCCGTGCAAGCAGATGCCATAC	Reverse primer for amplifying DgabT
Confirmation-ΔgabT-F	GGTAGGCAATGGGCTGGAAGACG	Forward primer for amplifying partial fragment of UgabT +DgabT for confirmation of ΔgabT genotype of H. elongata ZN2 and ZN3		952
Confirmation-ΔgabT-R	CAGCCAGGGTCTTCCTTCTGACATC	Reverse primer for amplifying partial fragment of UgabT +DgabT for confirmation of ΔgabT genotype of H. elongata ZN2 and ZN3
UgabD-F	CTCTAGAGGATCCCCCTTTCCACGAAGCCCAGCAGGAC	Forward primer for amplifying upstream region of gabD of H elongata OUT30018 genome (UgabD)		1,016
UgabD-R	CTGGCTCACATGGCAGCGGGCTCGC	Reverse primer for amplifying UgabD
DgabDT-F	TGCCATGTGAGCCAGGATCTGATCTTGCTGCC	Forward primer paired with DgabT-R primer for amplifying downstream region of gabT (DgabDT)		1,113
Confirmation-ΔgabDT-F	GCCACAGGAGCAGACGAATCAGC	Forward primer paired with Confirmation-ΔgabT-R for amplifying partial fragment of UgabDT +DgabDT for confirmation of ΔgabDT genotype of H. elongata ZN1		1,261
gabT-F	CATATGAACAACGCCCAACTCAACGAACTCAAGCAGCG	Forward primer for amplifying gabT of H. elongata OUT30018 genome	NdeI	1,314
gabT-R	GCTAGCTCAACTAGTGGCGGTGGCGGCGCTC	Reverse primer for amplifying gabT of H. elongata OUT30018 genome	SpeI, NheI
ectB-F	CATATGCAGACCCAGATTCTCGAACGC	Forward primer for amplifying ectB of H. elongata OUT30018 genome	NdeI	1,281
ectB-R	GCTAGCTCAACTAGTGCTAAAGGCCTGCTTGGTGCTGG	Reverse primer for amplifying ectB of H. elongata OUT30018 genome	SpeI, NheI

^a Restriction endonuclease sites within the primer sequence are underlined.

For construction of pK18mobsacB-U_gabT-D_gabT and pK18mobsacB-U_gabD-D_gabDT for deletion of the entire coding regions of the gabT gene alone and the gabDT gene cluster from H. elongata OUT30018 or H. elongata GOP-Gad strain, U_gabT, D_gabT, U_gabD, and D_gabDT fragments were PCR amplified from genomic DNA of H. elongata OUT30018 using U_gabT-F +U_gabT R, D_gabT-F +D_gabT R, U_gabD-F +U_gabD R, and D_gabDT-F +D_gabT R primer pairs. PCR cycles undergo initial denaturation at 98°C for 2 min, followed by 30 cycles of denaturation at 98°C for 10 seconds and annealing/extension at 68°C for 1 min. The fragments U_gabT +D_gabT, or U_gabD +D_gabDT were fused into SmaI digested pK18mobsacB plasmid using Infusion kit (In-Fusion HD EcoDryTM Cloning Kit from Takara Bio Inc., Shiga, Japan) to create pK18mobsacB-U_gabT-D_gabT and pK18mobsacB-U_gabD-D_gabDT.

For the construction of pHS-gabT-HA and pHS-ectB-HA plasmids used in complementation analysis, coding sequence (CDS) of gabT and ectB genes were PCR amplified from genomic DNA of H. elongata OUT30018 using gabT-F +gabT R or ectB-F +ectB R primer pairs. PCR cycles undergo initial denaturation at 98°C for 2 min, followed by 30 cycles of denaturation at 98°C for 10 seconds, and annealing/extension at 68°C for 1 min 30 seconds. The amplified NdeI-gabT-SpeI-NheI (1,314 bp) and NdeI-ectB-SpeI-NheI (1,281 bp) fragments were individually inserted into the SmaI site of the pUC19 plasmid. The fragments were subsequently cut out of the plasmids by NdeI and SpeI enzymes and ligated into the corresponding sites on pET-Lipop5-HA to construct pET-gabT-HA and pET-ectB-HA. Finally, to construct pHS-gabT-HA and pHS-ectB-HA plasmids, XbaI-gabT-HA-BamHI and XbaI-ectB-HA-BamHI fragments from pET-gabT-HA and pET-ectB-HA were inserted into SpeI and BamHI digested sites on pHS15N plasmids (44), which is an E. coli–H. elongata shuttle vector derivative of the original pHS15 plasmids (47). Expressions of the gabT-HA and the ectB-HA transgenes from the resulting plasmids are controlled by a mob promoter, which is a promoter of the mobilization region of the native pHE1 plasmid of H. elongata (48).

Generation of the recombinant H. elongata ZN1, ZN2, and ZN3 strains

For the generation of H. elongata ZN1 and ZN2 strains, pK18mobsacB-U_gabD-D_gabDT and pK18mobsacB-U_gabT-D_gabT plasmids used for double deletion of gabD and gabT genes or deletion of gabT gene alone from H. elongata OUT30018 genome were introduced into H. elongata OUT30018 by E. coli HB101/pRK2013-mediated tri-parental conjugation method (49), as mentioned previously (26). The resulting recombinant strains were identified by genomic PCR using the DgabDT confirmation primer set (Confirmation-ΔgabDT-F and Confirmation-ΔgabT-R, Table 3), which amplifies 1,261 bp fragments that confirm double deletion of gabD and gabT genes in H. elongata ZN1 or ΔgabT confirmation primer set (Confirmation-ΔgabT-F and Confirmation-ΔgabT-R, Table 3), which amplifies 952 bp fragments that confirm deletion of gabT gene in H. elongata ZN2. PCR cycles for these amplifications are as follows: initial denaturation at 94°C for 2 min, followed by 30 cycles of denaturation at 94°C for 30 seconds, and annealing/extension at 68°C for 1 min 30 seconds for H. elongata ZN1 or 1 min for H. elongata ZN2.

The same method was used to generate H. elongata ZN3 as for H. elongata ZN2, except the host strain for H. elongata ZN3 was the H. elongata GOP-Gad strain.

Generation of the recombinant H. elongata ZN3-1, ZN3-2, and ZN3-3 strains

pHS-gabT-HA and pHS-ectB-HA plasmids, which contains expression cassettes of DABA-AT-HA or GABA-AT-HA proteins, were introduced into H. elongata ZN3 to generate H. elongata ZN3-1, ZN3-2, and ZN3-3 strains. Empty pHS15N plasmid was also introduced into H. elongata ZN3 to generate a negative control H. elongata ZN3-1. Introduction of these plasmids into H. elongata ZN3 cells was done using E. coli HB101/pRK2013-mediated conjugation method (49). After plasmid transfer, transformed H. elongata were isolated by plating on an LB medium containing 120 mg/L Stm (selection marker for the pHS15N plasmid) and 6% wt/vol NaCl (for elimination of E. coli).

Preparation of crude protein extracts

To prepare crude protein extracts for immunoblot analysis, H. elongata ZN3-1, ZN3-2, and ZN3-3 were cultured in M63 medium containing 3% wt/vol NaCl until OD₆₀₀ reached around 1 and used as a 5% vol/vol inoculum for the main culture in the same medium. When OD₆₀₀ of the main cultures reached 1.2, the cells were collected by centrifugation at 10,000 × g for 3 min, resuspended in 500 µL of cold phosphate-buffered saline, and disrupted on ice by five cycles of 15 seconds on and 45 seconds off sonication to yield cell lysates, which was swiftly centrifuged to separate cell debris from crude protein extracts.

Immunoblot analysis

20 µL aliquot of crude protein extracts of H. elongata ZN3-1, ZN3-2, or ZN3-3 strains were mixed with the same volume of Ez-Apply dye (Tris-HCl buffer, 1% wt/vol SDS, 10% wt/vol sucrose, bromophenol blue, and 50 mM dithiothreitol; ATTO Corporation, Tokyo, Japan) and incubated in 100°C heat block for 3 min. After cooling down to room temperature, identical sets of protein samples were separated by electrophoresis in two 5%-20% wt/vol gradient SDS-polyacrylamide gels (ATTO Corporation, Tokyo, Japan) in Ez-Run (25 mM Tris, 192 mM glycine, and 0.1% wt/vol SDS; ATTO Corporation, Tokyo, Japan) buffer with constant 20 mA current per gel for 70 min. Separated proteins on one of the gels were dyed with EzStain AQUA (Coomassie Brilliant Blue, ATTO Corporation, Tokyo, Japan) at room temperature for 1 h for visualization of total proteins separated in each lane, whereas separated proteins on the other gel were electrophoretically transferred onto a polyvinylidene difluoride membrane (ATTO Corporation, Tokyo, Japan) in EzFastBlot transfer buffer (ATTO Corporation, Tokyo, Japan) by semi-dry electroblotting (HorizeBLOT 2 M-R, ATTO Corporation, Tokyo, Japan) with constant current 2 mA/cm² of gel for 30 min. After the transfer, the membrane was rinsed twice with Tris-buffered saline with 0.1% wt/vol Tween (TBS-T) solution containing 0.1% wt/vol EzTween20 (ATTO Corporation, Tokyo, Japan) in EzTBS buffer (25 mM Tris, 150 mM NaCl; ATTO Corporation, Tokyo, Japan), and non-specific antibody binding sites on the membrane were blocked by incubation with gentle agitation in TBS-T solution containing 0.3% wt/vol skim milk for 1 h at room temperature. To detect HA-tagged proteins, the membrane was washed three times with TBS-T before it was incubated with gentle agitation in 20 mL TBS-T containing 0.3% wt/vol skim milk and 5 µL of anti-HA HRP-conjugated antibody (Anti-HA-Peroxidase High-Affinity Rat monoclonal, Roche Japan) for 2 h at room temperature. After three washes with TBS-T solution, immunoreactive bands were visualized by incubating the membrane in a reaction solution containing 3, 3’-diaminobenzidine of a Peroxidase Stain DAB Kit (Nacalai Tesque, Inc., Kyoto, Japan).

Hypo-osmotic extraction of free amino acids from H. elongata cells

Free amino acids were extracted from H. elongata cells as previously described (26). H. elongata strains were cultured in liquid M63 medium containing 6% wt/vol NaCl with 4% wt/vol glycerol or 4% wt/vol GABA (also supplemented with 60 g/L Stm to maintain shuttle expression plasmids in H. elongata ZN3-1, ZN3-2, and ZN3-3 strains) until OD₆₀₀ reached more than 1 when the cells were harvested by centrifugation at 10,000 × g for 3 min. The weight of the cell pellets was recorded as cell fresh weight (CFW). For the experiment, in which later conversion of the yield was applied to report the GABA concentration as per cell dry weight, cells were harvested from the cultures in duplicate, and weights of both wet cell pellets were recorded as cell fresh weight (CFW). One of the cell pellets was freeze-dried, and the weight of the dried cell pellet was recorded as cell dry weight (CDW). For major osmolytes extraction, pure water (20 µL per 1 mg CFW) was added to the pellet samples to extract major osmolytes accumulated inside the cells using the hypo-osmotic extraction method, and the osmolytes sample was analyzed by HPLC (50, 51). After centrifugation at 10,000 × g for 3 min, the supernatant containing the free amino acids was collected as a major osmolyte sample.

Amino acid dabsylation

An amino acid dabsylation reaction was performed as previously described (26). A 10 µL aliquot of major osmolyte sample or standard amino acid was mixed with 2 µL of 2.5 mM internal standard norvaline and 8 µL of 1 M NaHCO₃ pH adjustment solution. Then, the sample was mixed with 40 µL of dabsylation reagents containing 2 mg/mL dabsyl chloride dissolved in acetonitrile and incubated at 70°C for 15 min. After the incubation, 440 µL of 250 mM NaHCO₃ solution was added, and the samples were centrifuged at 10,000 × g for 3 min. The supernatant of each sample was collected and filtered through a polytetrafluoroethylene membrane filter vial with 0.2 µm pore size (SEPARA Syringeless filter, GVS Japan K.K., Tokyo, Japan) before HPLC analysis.

HPLC gradient system for determination of dabsyl amino acids

Determination of dabsyl amino acids derived from H. elongata cells was carried out as previously described (26), using high-performance liquid chromatography system (Shimadzu, Kyoto, Japan) equipped with a UV/VIS detector (SPD-10 A VP), an autosampler (SIL-10 AD VP), two pumps (LC-10 AD VP), degasser (DGU-14A), system controller (SCL-10A Vp), and column oven (CTO-10AC VP). LabSolutions LC software (Shimadzu, Kyoto, Japan) was used for system control and data acquisition. Chromatographic separation of dabsyl amino acids was achieved through an analytical C18 column (Poroshell 120 2.7 µm, EC-C18, 4.6 × 75 mm, Agilent Technologies Inc.) with C18 guard column (Poroshell 120 2.7 µm Fast Guard, EC-C18, 4.6 × 5 mm, Agilent Technologies Inc.) using a mobile-phase gradient system consisting of 15% vol/vol acetonitrile in 20 mM sodium acetate (pH 6.0) (mobile phase A) and 100% acetonitrile (mobile phase B). Dabsyl amino acids were determined by UV/VIS detector at 468 nm. The injection volume was 10 µL, the flow rate was 0.5 mL/min, and the column temperature was maintained at 27°C.

HPLC analysis of ectoine

Following the previously described method (2), 25 µL of 100 mmol/L KH₂PO₄ was added to 25 µL of ectoine extract sample, and the sample mixture was vortexed. Subsequently, 450 µL of Ectoine derivatizing solution, made by dissolving 2.5 mmol of 18-crown-6 and 50 mmol of 4-bromophenacyl bromide in 100 mL acetonitrile, was then added to the sample mixture, and the derivatizing mixture was vortexed then heated to 80°C for 60 min. After cooling down to room temperature, the derivatizing mixture was filtered through a polytetrafluoroethylene membrane filter vial with 0.2 µm pore size (SEPARA Syringeless filter, GVS Japan K.K., Tokyo, Japan) before HPLC analysis. The concentration of ectoine was determined using a Shimadzu HPLC system (Kyoto, Japan) equipped with a SUPELCOSIL^TM LC-SCX column (Sigma Aldrich Corp. St. Louis, MO, USA). The analyses were conducted at 28°C using 22 mM choline chloride dissolved in 90% acetonitrile as the mobile phase. The column was eluted at a flow rate of 1.5 mL/min, and the eluate was analyzed using an ultraviolet absorbance detector set at 254 nm.

AFTER EPUB

[This article was published on 12 August 2024 with an error in Fig. 6D. The figure was corrected in the current version, posted on 19 August 2024.]