O-succinyl-L-homoserine and sodium methyl mercaptan were purchased from J&K Chemical Technology (Shanghai, China). L-methionine and PLP were purchased from Aladdin reagent Co. Ltd. (Shanghai, China). Kanamycin and isopropyl-β-D-thiogalactopyranoside (IPTG) were obtained from Sigma Chemical Co. (St. Louis, MO, USA). All other chemicals were of analytical grade and were commercially available.

E. coli DH5α (TSINGKE Biotech Co., Hangzhou, China) was used as the host for gene cloning, and E. coli BL21(DE3) was the host for protein expression. The plasmid pET28b was used as the expression vector in E. coli. The E. coli was grown in Luria Bertani (LB) medium (yeast extract 5 g/L, peptone 10 g/L, NaCl 10 g/L) with 50 mg/L kanamycin.

Twelve genes were synthesized artificially according to the sequence deposited in the NCBI Protein database under accession number WP_016715902.1 (Kim et al., 2015). For expression purpose, the genes were subcloned to the expression vector pET28b between restriction sites NcoI and XhoI, and then the recombinant plasmids were transformed into the E. coli strain BL21(DE3). After the recombinant E. coli cells were cultivated in LB medium containing kanamycin (50 mg/L) at 37°C for 4 h, IPTG was added for induction and the cells were cultured at 28°C for 12 h. Recombinant OSHS was expressed intracellularly, and the pellet was collected by centrifugation at 12,000 rpm and 4°C for 10 min. To purify the recombinant OSHS, the centrifuged cell pellet was washed twice with 0.85% NaCl solution, and then resuspended in 50 mM sodium phosphate buffer (pH 7.0), and sonicated in ice bath. After sonication, the cell lysate was centrifuged at 8,000 rpm for 30 min at 4°C to remove cell debris. Then the supernatant was filtered through a 0.22 μm filter. The filtrate was further purified by a chelating agarose fast-flow resin column (1.0 × 10.0 cm) containing Ni2+, and the binding buffer (50 mM sodium phosphate buffer, pH 7.0) was used for balance. Then, proteins were eluted with a linear imidazole gradient of 20–500 mM in 50 mM sodium phosphate buffer (pH 7.0). The protein was analyzed by sodium dodecyl sulfate polyacrylamide gel electrophoresis (SDS-PAGE) to determine the protein purity. The protein concentration was determined by the protein determination kit (Keygen Biotech, China). Thereafter, the purified enzyme was used in characterization experiments. Its kinetics and catalytic properties were determined.

The enzyme activity of recombinant OSHS was determined by measuring the conversion rate of OSH to L-methionine and the process was depicted by the reaction equation (Scheme 1). The system was carried out in a 500 μL system, which contained 50 mM Phosphate Buffered Saline (PBS) (pH 7.0), 50 mM OSH, 40 μL sodium methyl mercaptan and 15 μg purified enzyme at 30°C for 5 min. One unit (1U) is defined as the amount of enzyme catalyzing the production of 1 μM L-methionine from OSH per minute at 30°C. The L-methionine content was determined by amino acid analyzer.

The influence of pH on the activity of the purified enzyme was performed at different pH (4.5–9.0). 100 mM OSH and 5% (v/v) sodium methyl mercaptan was biocatalyzed at 30°C for 10 min in three buffers with different pH range, citric acid-sodium citrate (pH 4.5–6.0), Na₂HPO₄-NaH₂PO₄ (pH 6.0-8), and Tris-HCl (pH 8.0–9.0). The pH stability was determined by measuring the residual activity after incubating the enzyme in buffers at pH 4.5–9.0 at 4°C for up to 10 h.

To investigate the optimal temperature, the reactions were carried out with 100 mM OSH, 5% (v/v) sodium methyl mercaptan, pH 7.0 adjusted with 50 mM PBS, and the temperature range was 20–60°C. The thermostability was analyzed by measuring the residual enzyme activity of the purified OSHS, which was incubated at 20 to 60°C for 25 h. The half-life of the enzyme was determined by calculating the change in Ln (relative activity) over time.

The effect of metal ions (CaCl₂, MgSO₄, FeCl₃, FeCl₂, CuSO₄, MnCl₂. ZnSO₄, NiCl₂, AlCl₃, CoCl₂) on purified OSHS activity were investigated at final concentrations of 1 and 5 mM, respectively. Effects of SDS, EDTA were investigated using the final concentration of 1 and 5 mM, respectively. All reactions were performed with 50 mM OSH, 10 mM PLP and 5% (v/v) sodium methyl mercaptan at 30°C, and the pH was adjusted to 6.5 with 50 mM PBS. The concentration of purified OSHS in the reaction was 30 mg/L.

The kinetic parameters K_m (Michaelis-Menten constant) and V_max (maximum initial reaction rate) of OSHS were determined by using OSH as the substrate in the PBS at an initial rate ranging from 1 to 500 mM (pH 6.5). Before adding purified tsOSHS (30 mg/L), the reaction mixture was pre-incubated at 30°C for 10 min. The catalytic reaction was performed as a standard enzyme activity determination. K_m and k_cat were calculated by a non-linear regression method based on a function of the initial velocity data plotted against the substrate concentration.

Enzymatic synthesis of L-methionine was carried out in 200 of 500 mL shake flask under the optimum conditions. Samples were taken regularly to measure the concentration of L-methionine until the reaction reached equilibrium.

Purified OSHS enzyme with a final concentration of 30 mg/L was used to study the effect of substrate concentration on the enzymatic hydrolysis reaction in various concentrations of OSH (10–400 mM) in a reaction system with 10 mM PLP and 10% (v/v) sodium methyl mercaptan. The reactions were carried out at 30°C for 240 min. Under the condition of 400 mM OSH, 10 mM PLP and 10% (v/v) sodium methyl mercaptan, 30–100 mg/L purified OSHS was used to evaluate the effect of catalyst dosage on the enzymatic reaction. The reactions were carried out at 30°C for 240 min. Under the condition of 400 mM OSH, 100 mg/L purified OSHS and 10% (v/v) sodium methyl mercaptan, 1–20 mM PLP was used to investigate the effect of coenzyme on the catalyst. The reactions were carried out at 30°C for 140 min.

All the reactions were terminated using 100 μL 6 M HCl (10% v/v) for 10 min. The samples were diluted 100-fold and then centrifuged for 5 min at 12,000 rpm. Supernatant was collected and used for assay. Amino acids were detected using an automatic amino acid analyzer (SYKAM S-433D, SYKAM, München, BY, Germany) (El-Naggar et al., 2017; Huang et al., 2017). All experiments were replicated three times.

In order to obtain an enzyme could convert OSH into L-methionine, gene mining approach was used to identify novel enzyme with high yield and the potential for industrial application. A reported OSHS from Pseudomonas could convert OSH into L-methionine (Kim et al., 2015) was used to perform a BLAST search in the nation center for biotechnology information database. The identity and conversion rate were listed in Table 1. Among them, WP_095206569.1 showed the highest conversion rate. Therefore, the protein sequence of this enzyme was aligned with the reported OSHS (WP_016715902.1) and one of the eCGS (KFF54031.1) (Figure 1). It was found that the three sequences shared the same conservative site in Y46, R48, R361, K198, and showed different in Y101 which was not directly involved in catalysis (Jaworski et al., 2012). Besides, the sequence analysis indicated that tsOSHS had a predicted molecular weight of 42736.1 Da and a predicted isoelectric point of 5.85.

OSHS showed 71% amino acid sequence identity with OSHS from Thioalkalivibrio sulfidiphilus (GenBank accession number WP_095206569.1). Blast analysis showed that tsOSHS belong to cystathionine synthase family and PLP-dependent enzyme superfamily which used PLP as the coenzyme for enzymatic reaction (Figure 2). The cystathionine synthase family can catalyze the α, β elimination and α, γ elimination. However, there is no corresponding report on the characteristics of OSHS in this reaction. Thirty sequences belonged to the cystathionine synthase were collected to analyze their phylogenetic relationship. The phylogenetic tree is divided into two obvious clusters, namely β elimination and γ elimination.

α, β-elimination mechanism is usually carried out through a quinone intermediate, in which the carbanion formed upon abstraction of the C_α-proton is stabilized by delocalization into the pyridine ring of the PLP (Aitken and Kirsch, 2005). This transient intermediate can be detected when the elimination of the group is slow. All enzymes of the transsulfuration and reverse transsulfuration pathway except cysteine β-synthase are members of this subfamily. The structures of E. coli cysthionine-γ-Synthase (eCGS), yeast cystathionine c-lyase (yCGL) may originated from O-acetyl-L-homoserine sulfhydrylase, which were similar that their active sites located at the interface of subunits, and are composed of residues from two subunits (Messerschmidt et al., 2003). Phylogenetic analysis showed that tsOSHS is a member of the PLP-dependent enzyme superfamily cysteine γ-synthase.

In order to functionally characterize the target enzyme, the synthesized tsOSHS gene was cloned into the pET28b expression vector. The 6 × His tag was added to the carbon end of the protein for affinity-based enzyme purification. The recombinant tsOSHS was purified from the supernatant of the cell lysate on a Ni-NTA column by one-step affinity chromatography. SDS-PAGE of the purified enzyme gave a single band showed a molecule of ~40 kDa (Supplementary Figure 1), which corresponded to the molecular weight calculated from the amino acid sequence obtained. The purified tsOSHS (318.2 U/mg) was then used for enzymatic characterization.

tsOSHS has high activity and stability in slight acid neutral buffer solution from pH 6.5 to 7.5, and the relative activity reached optimum at pH 6.5 (Figure 3), which indicated that tsOSHS has high activity in a certain pH range and has the potential of industrial application. However, the enzyme activity of tsOSHS decreased significantly when pH was higher than 7.5. In addition, the enzyme activity of tsOSHS in PBS was higher than that in other buffers. As shown in Figure 3, tsOSHS still maintained more than 80% activity in the pH range of 6.5–8.5 after 10 h incubation. The highest stability was achieved at pH 7.0 and the stability decreased significantly when pH was lower than 6.5, which indicated that the enzyme was unstable in acidic conditions. It could be found from the above results that the optimum pH and the pH maintain enzyme stability were more biased toward alkaline conditions. However, the enzyme activity reached the highest under neutral conditions. It could be explained that tsOSHS is derived from T. sulfidiphilus which is an alkaliphilic strain can maintain good stability in alkaline environment. In this study, tsOSHS showed better activity in neutral environment when it catalyzing OSH convert to L-methionine, as the substrate hydrolyzed spontaneously in alkaline environment which would cause that the detected enzyme activity lower than the actual enzyme activity.

As shown in Figure 4, tsOSHS had high enzyme activity at 25–45°C. The optimal temperature was 35°C, and there was still 91% of the optimal enzyme activity at 40°C. When the temperature increased over 45°C, the enzyme activity decreased rapidly, and almost no enzyme activity was detected at 60°C. The residual enzyme activity of tsOSHS was detected after incubation at 25–45°C in PBS (pH6.5). Accordingly, the half-life (t_1/2) values were determined to be 21.72, 18.68, 16.05, and 7.32 h, respectively (Figure 4). These results indicated that tsOSHS had an excellent thermostability at 30–40°C, which make it a potential candidate for biocatalytic process development.

In order to explore whether tsOSHS has the properties of metalloenzymes, the effects of metal ions and organic solvents on enzyme activity were studied. The activity of purified tsOSHS affected by metal ions and organic solvents in the reaction were shown in Supplementary Figure 2. The activity of tsOSHS decreased with the increase of metal ion concentration. When the concentration of Cu²⁺ was 1 mM, the enzyme was seriously inhibited. When the concentration of Cu²⁺ and Al³⁺ was 5 mM, the enzyme was almost completely inhibited. Ni²⁺, Co²⁺, and Zn²⁺ also had a significant inhibitory effect on enzyme activity at 5 mM. In contrast, when the final concentration of Fe²⁺ was 1 mM, the activity increased 115%, but decreased when the concentration was 5 mM. And the effect of Fe²⁺ concentration on the relative activity of tsOSH was studied (Supplementary Figure 3). At the same time, the effects of 1 mM surfactant SDS and metal chelating agent EDTA on tsOSHS pre-treated with 1 mM Fe²⁺ were studied. The relative enzyme activity was 63 and 90%, respectively, which indicated that the enzyme activity showed inhibition under the reaction of organic solvents. Apparently, according to the above experiment results, various metal ions had almost no positive effect on tsOSHS, and the addition of metal chelating agents had no obvious inhibitory effect. Nowadays, there is no report showing that CGS has the characteristics of metalloenzymes. Therefore, it could be determined that tsOSHS is not a metalloenzyme.

As shown in Supplementary Figure 4, the kinetic parameters were calculated through non-linear fitting results by plotting the initial velocity and OSH concentration. It was found that the V_max, K_m, and k_cat values of tsOSHS were 44.57 μmol min⁻¹ mg⁻¹, 72.52 mM and 29.71 s⁻¹, respectively. It's higher than the K_m and k_cat of CGS from other sources producing cystathionine, such as 2.5 mM for A. thaliana CGS (Stephane et al., 1998), 3.02 mM for H. pylori CGS, and 1.29 mM for eCGS (Kong et al., 2008). It could be related to the structure of the sulfur donor. Compared with cysteine, methyl mercaptan used in this study leads to a smaller steric hindrance, so that the reaction of γ-elimination might be easier to happen.

The effects of substrate concentration, enzyme concentration and PLP concentration on L-methionine production were studied. The yield of L-methionine was increased with the increasing of OSH concentration, and 38.25 g/L of L-methionine was obtained at 400 mM OSH reaching maximum yield (Figure 5A). With the extension of reaction time, the accumulation of L-methionine gradually slowed down and stabilized. When the substrate concentration was 400 mM, the conversion rate of OSH was 64.25% at 240 min and L-methionine was accumulated slowly. The results showed that the 30 mg/L tsOSHS could not completely catalyze 400 mM OSH. Therefore, in the following experiment, the concentration of tsOSHS in 400 mM OSH was optimized. As shown in Figure 5B, when the enzyme concentration increased from 30 to 100 mg/L, the accumulation rate of L-methionine increased. After about 120 min, 42.3 g/L-methionine was obtained, and the conversion rate increased from 64.17 to 71%. In addition, PLP is the coenzyme of tsOSHS and an important rate-limiting factor, which could affect the catalytic activity of tsOSHS. As shown in Figure 5C, when the concentration of PLP was lower than 10 mM, the accumulation amount and accumulation speed of L-methionine varied with the PLP concentration, indicating that the coenzyme concentration was not enough to maintain the catalytic activity of tsOSHS at this concentration. Besides, PLP not only has a catalytic role but also participate in the stability of CGS (Stephane et al., 1998). Therefore, when the PLP concentration increased, the conversion rate was significantly increased. When the concentration of PLP was over 10 mM, the yield of L-methionine and accumulation speed was close and the highest yield of L-methionine was 42.63 g/L. This indicated that when the concentration of PLP reached 10 mM, the ratio of enzyme and coenzyme was suitable for catalyzing the L-methionine production, and continued to increase the coenzyme concentration has almost no effect on the catalytic activity of the enzyme.