Figure 1 shows a phylogenetic tree of Class I aminotransferase homologs from four selected Thermococcales species, T. kodakarensis, P. furiosus, P. horikoshii, and T. litoralis, from which multiple aminotransferases have been experimentally studied. A clade including TK1990, a representative of a Class IV protein encoding cysteine desulfurase (Hidese et al., 2014), was used as the outgroup. There are eight groups (G1–G8) in which homologs occur in ten or more genomes of the 40 Thermococcales species examined. Among the eight groups, members of G1 (39 genomes among the 40 genomes examined harbor homologs; 39/40), G2 (40/40), G3 (40/40), G6 (40/40), and G8 (35/40) are present in most or all of the Thermococcales genomes used in our analysis. Members of G4 (12/40), G5 (10/40), and G7 (20/40) are less distributed. T. litoralis harbors a higher number of Class I aminotransferase homologs compared to the other three species. Homologs of OCC_03517 are not found on any of the other Thermococcales genomes, but homologs of OCC_10965 (3/40), OCC_02240 (10/40, G5), and OCC_11814 (5/40) can be identified in other species. A detailed phylogenetic tree based on sequences of all Class I aminotransferase homologs from Thermococcales species is shown in Supplementary Figure S1.

We considered the physiological roles of these groups of aminotransferases by first taking into account their gene location and previous biochemical studies on aminotransferases from members of Thermococcales. In terms of gene location, we observed that TK0250 (G7) is included within the His biosynthesis gene cluster (TK0242-TK0251) in T. kodakarensis. We also observed a complete co-occurrence between the TK0250 homologs and the His biosynthesis operon in Thermococcales members (Supplementary Table S1). A similar situation was observed for TK0260 (G4). The gene is situated in a gene cluster (TK0259-TK0261) involved in the biosynthesis of Phe/Tyr in T. kodakarensis, and co-occurrence is observed between TK0260 homologs and the gene cluster in Thermococcales species (Supplementary Table S2). These observations suggest that the physiological roles of members of G4 and G7 are related to Phe/Tyr and His biosynthesis, respectively. The members of G8 from Thermococcales have not been characterized, but display 30% identity to Thr decarboxylase (encoded by MM2060) from Methanosarcina mazei (Tavares et al., 2018, 2019) and are clustered with genes related to cobalamin salvage. By contrast, members of G1, G2, G3, G5, and G6 did not show a tendency to be included in a particular biosynthesis operon or gene cluster.

Members of G1, G2, G3, and G6 are widely distributed among Thermococcales species, and a number of members have been biochemically characterized. In particular, members of G1 and G6 have been relatively well studied. In the case of G1, the structure of the PH1371 protein from P. horikoshii has been elucidated, and the amino donors most recognized were Tyr, Phe, Glu, His, and Trp (Matsui et al., 2000). The PF1253 protein from P. furiosus, referred to as AroAT II, displayed preference to Phe, Tyr and Trp, with highest k_cat/K_m toward Phe (Ward et al., 2002). The OCC_04737 protein from T. litoralis, referred to as ArAT II, also displayed similar properties, preferring Phe, Tyr, and Trp (Andreotti et al., 1994). In the case of G6, the PF0121 protein from P. furiosus, referred to as ArAT or AroAT I, displayed significant activity toward Phe, Trp, and Tyr (Andreotti et al., 1995). The OCC_04335 protein from T. litoralis, referred to as ArAT I, also recognized Phe, Tyr, and Trp (Andreotti et al., 1994). The PH0207 protein displays kynurenine aminotransferase activity which is involved in the degradation of Trp (Okada et al., 2012), and the crystal structure has been determined (Chon et al., 2005; Okada et al., 2014). Concerning other groups, the PF1497 protein from P. furiosus (G3), referred to as AlaAT, displays highest activity with Ala. The enzyme did not display activity toward Phe and Tyr (Ward et al., 2000). A genetic analysis on TK1094 from T. kodakarensis (G3) suggested a role in the conversion between pyruvate and Ala (Kanai et al., 2015). The PF1702 protein (G4), referred to as AspAT, displays highest activity toward Asp and Glu, with low activities to a broad range of amino acids (Ward et al., 2002).

We carried out further analyses on TK2268 and TK0548. The TK2268 protein is a member of G2, which does not have any member that has been experimentally characterized. Determining its biochemical properties would contribute to our understanding of amino acid metabolism in T. kodakarensis. We also examined the TK0548 protein, the most closely related protein to the TK2268 protein (45.4% identical). None of the G1 and G2 genes have been examined genetically.

In order to obtain recombinant proteins, the TK0548 and TK2268 genes were expressed in Escherichia coli. In the case of TK0548, soluble protein was obtained, and the recombinant TK0548 protein was purified through heat treatment, anion exchange chromatography and gel-filtration chromatography. In the case of TK2268, expression in E. coli resulted in the formation of inclusion bodies. The gene was thus expressed in the native host T. kodakarensis under the control of a strong constitutive promoter of the cell surface glycoprotein gene (csg, TK0895; Yokooji et al., 2009). Sequences to incorporate a His₆-tag on the C-terminus of the TK2268 protein were introduced. The soluble TK2268 protein obtained in the cell extract of T. kodakarensis cells was purified using a nickel affinity column and gel filtration chromatography. Both proteins were subjected to SDS-PAGE, confirming their apparent homogeneities (Figure 2).

The molecular masses of the purified, recombinant TK0548 and TK2268 proteins were examined with gel-filtration chromatography. The estimated molecular mass of the TK0548 protein was 85 kDa, and considering that the calculated molecular mass of the monomer was 43,608 Da, this suggested that the TK0548 protein was a dimer. In the case of the TK2268 protein, we consistently observed two peaks, corresponding to estimated molecular masses of 98 kDa and 327 kDa. As the calculated molecular mass of the monomer was 45,116 Da, the result suggested that the TK2268 protein forms a dimer unit, which may then further assemble to form an octamer.

The purified TK0548 and TK2268 proteins were examined for aminotransferase activity. We used Glu (10 mM) as the amino group donor and pyruvate (10 mM) as the amino acceptor. We observed aminotransferase activity in both proteins. When PLP was omitted from the reaction, we observed a partial decrease in activity in both cases (TK0548: 13% decrease, TK2268: 50% decrease) compared to those observed with the addition of PLP, suggesting that the TK0548 and TK2268 proteins are PLP-dependent aminotransferases. The only partial decrease in activity is most likely due to PLP bound to the proteins when they were produced in their respective host cells, E. coli (TK0548) and T. kodakarensis (TK2268). To further support the necessity of PLP for the reactions, we added hydroxylamine, a PLP inhibitor (Kito et al., 1978) to the enzymes in the absence of supplemental PLP. In this case, we observed a further 93% decrease in activity of the TK0548 protein and a 78% decrease in activity of the TK2268 protein with the addition of hydroxylamine (Supplementary Figure S2).

We first carried out an initial screening to identify the amino acids recognized by the TK0548 and TK2268 proteins. 2-Oxoglutarate or pyruvate was used as the amino acceptor, and the production of Glu or Ala, respectively, was examined after 15 min. The substrates were present in the reaction mixture at a concentration of 10 mM. As shown in Figure 3A, the TK0548 protein utilized Phe, Tyr, His, and Trp as an amino donor, as well as Met, Glu, and Leu to a lower extent. In the case of the TK2268 protein, Asp and Glu were utilized, along with Tyr, Leu, Ala, Met, and Cys (Figure 3B).

Focusing on the amino acids that were recognized by the proteins, we examined enzyme activity. The amino acids and amino acceptors were constant at 10 mM. As shown in Figure 4A, the TK0548 protein exhibited highest activity toward Tyr, followed by Phe, Trp, and His. Activity toward Leu, Met, and Glu were much lower. On the other hand, the TK2268 protein displayed highest activity toward Glu, followed by Asp. Lower levels of activity were observed using Cys, Leu, Ala, Met, and Tyr (Figure 4B).

Kinetic analyses were performed on the substrates that resulted in relatively high levels of activity. For the TK0548 protein, activities were first measured for varying concentrations of pyruvate or 2-oxoglutarate in the presence of 10 mM Phe. As shown in Table 1, the TK0548 protein exhibited a higher V_max and lower K_m toward 2-oxoglutarate when compared to pyruvate. The k_cat/K_m value toward 2-oxoglutarate was over 70-fold higher than that for pyruvate. We next examined activity with varying concentrations of Phe, Tyr, Trp, His, and Met in the presence of 10 mM 2-oxoglutarate (Table 1). The highest k_cat/K_m value was observed with Phe, with high values also observed for Tyr and Trp. A lower k_cat/K_m value was observed with His, and that for Met was extremely low. The results suggested that the TK0548 protein mainly utilizes the aromatic amino acids Phe, Tyr, and Trp as substrates, with His also a possible substrate.

Concerning the TK2268 protein, Leu (50 mM) was used as the amino donor to measure activity with varying concentrations of pyruvate and 2-oxoglutarate. The k_cat/K_m value toward 2-oxoglutarate was higher than that toward pyruvate (Table 2). This was mainly due to differences in the K_m value, as their V_max values were comparable. As the product of the aminotransferase reaction with 2-oxoglutarate would not be possible with Glu as the amino donor, 10 mM pyruvate was used as the amino acceptor to examine the activities with varying concentrations of Glu, Asp, and Leu. As a result, relatively high k_cat/K_m values were observed for both Asp and Glu compared to Leu and Tyr. The results suggest that the TK2268 protein is an aminotransferase with specificity toward the acidic amino acids Glu and Asp.

To understand the contribution of the TK0548 and TK2268 genes to amino acid catabolism in T. kodakarensis, we disrupted each gene using T. kodakarensis KU216 (ΔpyrF) as a host strain. Five transformants were chosen for each gene disruption and their loci were examined by PCR. Examples of transformants whose TK0548 gene (Figure 5A) or TK2268 gene (Figure 5B) was disrupted are shown. The respective loci were sequenced, confirming that gene disruption had occurred as intended.

The ΔTK0548 and ΔTK2268 disruption strains were first grown in the nutrient-rich ASW-YT-m1-S⁰ medium containing yeast extract and tryptone (Figure 6A). The growth of both disruption strains did not display significant differences to that of the host strain KU216. Therefore, in order to increase the dependency of growth on amino acid catabolism, the strains were grown in ASW-AA-m1-S⁰(+Ura) medium (Figure 6B). This medium is a synthetic medium with amino acids as the only major carbon and energy source. In this case, we observed a retardation in growth in both gene disruption strains. Cell yields of the cultures eventually reached similar levels, but the gene disruption strains took a 4-h longer period of time to reach maximum cell density. The results suggest that both proteins are involved in the utilization of amino acids for growth of T. kodakarensis.

The substrate specificities of multiple aminotransferases from members of Thermococcales have been determined in vitro. However, the contribution of each protein in the conversion of a particular amino acid in vivo cannot be determined by biochemical properties alone, as multiple aminotransferases, in some cases with overlapping substrate specificities or different expression levels, are present in the cell. We thus measured and compared aminotransferase activity in the cell extracts of T. kodakarensis host strain and gene disruption strains. As in vitro studies indicated that the TK0548 protein preferred Trp, Phe, Tyr, and His, while the TK2268 protein recognized Asp and Glu, activities in the cell extracts toward Trp, Phe, Tyr, His, Asp, and Glu were measured. We also used Ile as an amino donor. In vitro studies indicated that the recognition of both TK0548 and TK2268 proteins toward Ile was minimal (Figure 3), so the effects of TK0548 and TK2268 disruption on intracellular aminotransferase activity toward Ile would be expected to be low. As shown in Figure 7, aminotransferase activity for all seven amino acids was clearly observed in KU216 cell extracts. We observed that the levels of aminotransferase activity toward different amino acids greatly differed. Activities toward Phe or Glu were particularly high, whereas that toward Asp was notably low, two orders of magnitude lower than those observed for Phe or Glu. When we examined the effects of gene disruption, the disruption of TK0548 and TK2268 had no effect on intracellular aminotransferase activity toward Ile, consistent with our in vitro results that neither enzyme recognized Ile (Figure 3). We next examined differences observed upon gene disruption, focusing on those with p values below 0.01. We observed that the disruption of TK2268 resulted in a 35% decrease in aminotransferase activity toward Asp, one of the substrates preferred by the protein in in vitro experiments. Disruption of TK2268 did not affect the Glu aminotransferase activity in cell extracts, suggesting that the contribution of the protein to this activity is relatively small. This is not surprising though, as in vitro analysis indicated that the activity of the TK2268 protein toward Glu and Asp are comparable (Table 2). As the total Glu aminotransferase activity in T. kodakarensis cell extracts is over 100-fold higher than that toward Asp, a decrease in Glu aminotransferase activity comparable to the levels toward Asp would correspond to only a very small fraction of the total activity toward Glu. We also observed a decrease in His aminotransferase activity, but this could not be explained by the results of in vitro analysis. When TK0548 was disrupted, decreases in activity toward Trp (32%), Tyr (64%), and His (89%) were observed. The TK0548 protein seems to be a major contributor for Tyr and His aminotransferase activity in T. kodakarensis.

Thermoccocus kodakarensis was isolated from Kodakara Island, Kagoshima, Japan (Morikawa et al., 1994; Atomi et al., 2004). T. kodakarensis KU216 (Sato et al., 2003, 2005) and derivative strains were cultivated under strictly anaerobic conditions at 85°C in nutrient-rich medium (ASW-YT-m1-S⁰ or ASW-YT-m1-pyruvate) or synthetic medium (ASW-AA-m1-S⁰). ASW-YT-m1-S⁰, ASW-YT-m1-pyruvate, and ASW-AA-m1-S⁰ are modified versions of ASW-YT-S⁰, ASW-YT-pyruvate, and ASW-AA-S⁰ media, respectively. ASW-YT-S⁰ was composed of 0.8 × artificial seawater (ASW) (Robb and Place, 1995), 5 g L⁻¹ yeast extract, 5 g L⁻¹ tryptone, and 2 g L⁻¹ elemental sulfur. In ASW-YT-m1-S⁰, 20 μM KI, 20 μM H₃BO₃, 10 μM NiCl₂, and 10 μM Na₂WO₄ were supplemented. In ASW-YT-m1-pyruvate medium, elemental sulfur was replaced with 5 g L⁻¹ sodium pyruvate. ASW-AA-S⁰ was composed of 0.8 × ASW, a mixture of 20 amino acids, modified Wolfe’s trace minerals and a mixture of vitamins (Sato et al., 2003). In ASW-AA-m1-S⁰, 20 μM KI, 20 μM H₃BO₃, 10 μM NiCl₂, and 10 μM Na₂WO₄ were supplemented, and the concentrations of l-arginine hydrochloride and l-valine were increased (from 125 mg L⁻¹ to 250 mg L⁻¹ and from 50 mg L⁻¹ to 200 mg L⁻¹, respectively). When cells without a pyrF gene were grown, 10 μg mL⁻¹ uracil was added to make ASW-AA-m1-S⁰(+Ura). To remove oxygen in the medium, 5% (w/v) Na₂S solution was added until the medium became colorless. Resazurine (0.5 mg L⁻¹) was also added to all media as an oxygen indicator. For solid medium used to isolate transformants, 10 g L⁻¹ gelrite, 7.5 g L⁻¹ 5-fluoroorotic acid (5-FOA), 10 μg mL⁻¹ uracil, 4.5 mL of 1 M NaOH and 0.2% (v/v) polysulfide solution (10 g Na₂S 9H₂O and 3 g sulfur flowers in 15 mL H₂O) rather than elemental sulfur was supplemented to ASW-AA-m1 medium. Escherichia coli DH5α (Takara Bio, Kusatsu, Japan) and BL21-Codonplus(DE3)-RIL strains (Agilent Technologies, Santa Clara, CA) were cultivated at 37°C in Lysogeny broth (LB) medium supplemented with ampicillin (100 mg L⁻¹). E. coli DH5α was used for recombinant plasmid construction and E. coli BL21-Codonplus (DE3)-RIL was used for heterologous gene expression. Chemicals were purchased from Wako Pure Chemicals (Osaka, Japan) or Nacalai Tesque (Kyoto, Japan) unless mentioned otherwise.

In this study, pET21a(+) was used as an expression vector for TK0548 which was amplified from genomic DNA of T. kodakarensis KU216 using the primer set TK0548F/TK0548R (Table 4). The restriction enzyme sites NdeI and BamHI were incorporated into the 5′- and 3′-termini of the fragments, respectively, during PCR. The amplified product and pET21a(+) were digested with NdeI and BamHI, ligated using Ligation high Ver. 2, and introduced into E. coli DH5α cells. Positive colonies were selected by PCR analysis and confirmed by DNA sequencing. Plasmids were introduced into E. coli BL21-Codonplus (DE3)-RIL for gene expression.

The TK2268 gene was amplified from the genomic DNA of T. kodakarensis KU216 using the primer set TK2268F/R (Table 4). NdeI and SalI sites as well as a sequence to introduce a C-terminal His₆-tag were incorporated during amplification. The amplified product was digested with NdeI and SalI and inserted into a T. kodakarensis-E. coli shuttle plasmid previously used for heterologous expression of TK2101 and TK1211 (Zheng et al., 2018, 2021). After confirming the absence of unintended mutations by DNA sequencing, the plasmid was introduced into T. kodakarensis KPD2 (ΔpyrF, ΔpdaD, ΔchiA) for gene expression. pdaD corresponds to TK0149, and disruption of this gene results in agmatine auxotrophy (Fukuda et al., 2008). For transformation, T. kodakarensis KPD2 was grown in ASW-YT-m1-S⁰ medium supplemented with agmatine (1.0 mM) at 85°C for 12 h. Cells were harvested by centrifugation (12,000 × g, 5 min, 4°C), and resuspended in 200 μL 0.8 × ASW-m1, followed by incubation on ice for 30 min. After mixing with 3.0 μg of the expression plasmid, the mixture was further incubated on ice for 1 h. Cells were inoculated into 20 mL ASW-YT-m1-S⁰ medium. After incubation at 85°C for 24 h, a 200-μL aliquot was further inoculated into 20 mL ASW-YT-m1-S⁰ medium. After incubation at 85°C 24 h, cells were spread onto solid ASW-YT-m1-S⁰ medium. After incubation at 85°C for 24 h, transformants displaying agmatine prototrophy were isolated and cultivated in ASW-YT-m1-S⁰. The presence of recombinant plasmids and absence of unintended mutation were confirmed by PCR and DNA sequencing, respectively.

Transformants with TK0548 expression plasmid were cultivated in LB medium (100 mg L⁻¹ ampicillin and 30 mg L⁻¹ chloramphenicol) at 37°C until the OD₆₆₀ reached 0.6. Heterologous gene expression was induced by adding isopropyl-1-thio-β-D-galactopyranoside (IPTG) to a final concentration 0.1 mM followed by cultivation at 18°C for 20 h. Cells were harvested via centrifugation (12,000 × g, 15 min, 4°C), and resuspended in 50 mM HEPES buffer (containing 150 mM NaCl, pH 7.5). Sonication was used to lyse the cells and the insoluble cell debris was separated by centrifugation at 12,000 × g for 15 min at 4°C. The soluble cell extract was heat treated at 85°C for 15 min, and the thermolabile proteins derived from the host were removed by centrifugation (12,000 × g, 15 min, 4°C). The supernatant was loaded onto an anion exchange column (Resource Q) equilibrated with 50 mM HEPES buffer (pH 7.5). Protein was eluted with a linear gradient of 0 to 1.0 M NaCl. Fractions that contain TK0548 protein were collected and concentrated with an Amicon Ultra centrifugal filter unit (MWCO 10000). The resulting protein solution was applied to a Superdex 200 10/300 Gl gel-filtration column (GE Healthcare) with a mobile phase of 50 mM HEPES buffer (containing 150 mM NaCl, pH7.5) at a flow rate of 0.7 mL min⁻¹.

The TK2268 gene expression strain was cultivated in ASW-YT-m1-pyruvate medium at 85°C for 20 h, and cells were collected by centrifugation (6,000 × g, 15 min, 4°C). After washing with 0.8 × ASW-m1, cells were resuspended in binding buffer (50 mM HEPES buffer, 20 mM imidazole, 500 mM KCl, 10% (v/v) glycerol, pH 7.5), then disrupted by sonication. Insoluble cell debris was removed by centrifugation (12,000 × g, 15 min, 4°C). The soluble cell extract was applied to a His GraviTrap column (GE Healthcare) which had been equilibrated with binding buffer. The TK2268 protein with a His₆-tag at its C terminus was eluted by elution buffer (50 mM HEPES buffer, 500 mM imidazole, 500 mM KCl, 10% (v/v) glycerol, pH 7.5). After concentrating the eluate, it was applied to a Superdex 200 10/300 Gl gel-filtration column (GE Healthcare). The TK2268 protein was eluted with a mobile phase of 50 mM HEPES buffer (pH7.5) containing 500 mM KCl and 10% (v/v) glycerol, at a flow rate of 0.7 mL min⁻¹.

For examining the molecular mass of proteins, Blue 2000 was used to examine the void volume of the column, and ribonuclease A (13.7 kDa), carbonic anhydrase (29 kDa), ovalbumin (44 kDa), conalbumin (75 kDa), aldolase (158 kDa), and ferritin (440 kDa; GE Healthcare) were used as standards. The mobile phase was 50 mM HEPES buffer (pH7.5) containing 500 mM KCl, 10% (v/v) glycerol, and the flow rate was 0.7 mL min⁻¹. Protein concentrations were determined with a Protein Assay kit (Bio-Rad, Hercules, CA) using bovine serum albumin as standard.

Gene disruption strains were constructed using T. kodakarensis KU216 (ΔpyrF), which shows uracil auxotrophy, as a host strain. For disrupting the TK0548 gene, the region from the start codon to base number 1080 of TK0548 gene was deleted instead of the stop codon to avoid disturbing expression of the overlapping downstream gene. In the case of TK2268, the entire coding region, along with 9 bases of its 3′-flanking region, was deleted. The TK0548 and TK2268 genes along with their 5′- and 3′-flanking regions (~1.0 kbp) were amplified from the genome of T. kodakarensis KU216 using the primer sets dTK0548F1/R1 and dTK2268F1/R1 (Table 4). The amplified products were inserted in the HincII site of the plasmid pUD3 which contains the pyrF gene of T. kodakarensis inserted in the ApaI site of pUC118 (Yokooji et al., 2009). Inverse PCR was performed with the primer sets dTK0548F2/0548R2 and dTK2268F2/2268R2 (Table 4) to remove sequences from the recombinant plasmid. The sequences of relevant regions were confirmed by DNA sequencing.

Thermococcus kodakarensis KU216 was cultivated in ASW-YT-m1-S⁰ medium for 12 h. Cells were harvested and resuspended in 200 μL of 0.8 × ASW-m1, then kept on ice for 30 min. After addition of 3.0 μg of the disruption plasmid and further incubation on ice for 1 h, cells were cultivated in ASW-AA-m1-S⁰ medium without uracil for 48 h at 85°C. A 200 μL aliquot was inoculated into fresh ASW-AA-m1-S⁰ medium and further cultivated under the same conditions to enrich transformants displaying uracil prototrophy. The culture was spread onto ASW-YT-m1 solid medium supplemented with 7.5 g L⁻¹ 5-FOA and 60 mM NaOH. Only cells that have undergone a pop-out recombination can grow in the presence of 5-FOA. After cultivation at 85°C for 48 h, colonies were selected, and their genotypes were analyzed by PCR using primer sets dTK0548outF/0548outR and dTK2268outF/2268outR (Table 4). Transformants that led to amplification of DNA products with the expected size were chosen and cultivated in ASW-YT-m1-S⁰ medium. Gene disruption was also confirmed by DNA sequencing.

Initial examination of the aminotransferase activity of TK0548 and TK2268 proteins were carried out with Glu and pyruvate as amino donor and amino acceptor, respectively. Aminotransferase activity was measured at 80°C unless mentioned otherwise. The standard reaction mixture of TK0548 protein contained 20 μM PLP, 10 mM Glu, 10 mM pyruvate, 6.24 mM NaCl, and 0.4 μg mL⁻¹ recombinant protein in 50 mM HEPES buffer (pH7.4). The standard reaction mixture of TK2268 protein contained 20 μM PLP, 10 mM Glu, 10 mM pyruvate, 8.8 mM KCl, 0.18% (v/v) glycerol and 4 μg mL⁻¹ recombinant protein in 50 mM HEPES buffer (pH7.4). When the PLP-dependency of activity was measured, PLP was omitted and activity with or without 10 mM hydroxylamine was measured. When PLP was omitted without addition of hydroxylamine, the amino donor and acceptor were 10 mM Leu and 10 mM 2-oxoglutarate, respectively. In the presence of hydroxylamine, the TK0548 protein reaction was measured with Phe and 2-oxoglutarate, while the TK2268 protein reaction was measured with Asp and pyruvate. The reaction mixture without amino donor and amino acceptor was pre-incubated at 80°C for 2 min, and the amino donor and acceptor were added to start the reaction. After further incubation at 80°C for 5 or 15 min, the reaction was stopped through cooling the reaction mixture on ice for 10 min. Proteins were removed with an Amicon Ultra-0.5 centrifugal filter unit with an Ultracel-10 membrane (Millipore). The formation of Glu and Ala was detected and quantified by HPLC after derivatization. The derivatization mixture (100 μL) contained 10 μL of reaction mixture, 70 μL of solution B [borate sodium hydroxide buffer (0.4 M, pH 10.4)] and 20 μL of solution A (8 mg o-phthalaldehyde and 10 mg N-acetylcysteine were dissolved in 1 mL methanol). After derivatization for 5 min at room temperature, an aliquot (10 μL) of the solution was applied to a COSMOSIL 5C18-PAQ packed column (4.6ID × 250 mm) using a Nexera X2 liquid chromatography system with a fluorescence detector RF-20A XS (Shimadzu, Kyoto, Japan). Compounds were eluted with a solution of 20 mM sodium acetate (pH 5.6) and methanol at a flow rate of 0.7 mL min⁻¹. The excitation and emission wavelength were 350 and 450 nm, respectively.

To screen the amino acids recognized by TK0548 and TK2268 proteins, 20 amino acids were used as amino donor, and 2-oxoglutarate or pyruvate was used as amino acceptor. To analyze the aminotransferase activity of TK0548 protein, Tyr, Phe, Trp, His, Leu, Met, and Glu were chosen as amino donor, and 2-oxoglutarate or pyruvate was used as amino acceptor. For the TK2268 protein, Glu, Asp, Cys, Leu, Ala, Met, and Tyr were chosen as amino donor, and 2-oxoglutarate or pyruvate was used as amino acceptor. The standard reaction mixture was incubated at 80°C for 3, 5, and 7 min (reaction mixture with His and Glu were incubated at 80°C for 1, 2, 3 min) to confirm that product formation was linear with time.

For kinetic analysis of the TK0548 protein reaction, reaction rates with various concentrations of Phe, Trp, Tyr, Met, and His were examined with 10 mM 2-oxoglutarate. Reaction rates with various concentrations of 2-oxoglutarate and pyruvate were examined with 10 mM Phe. For analysis of the TK2268 protein, reaction rates with various concentrations of Asp, Glu, Tyr, and Leu were examined with 10 mM pyruvate or 10 mM 2-oxoglutarate. Reaction rates with various concentrations of 2-oxoglutarate and pyruvate were examined with 10 mM Leu. Kinetic parameters were obtained by fitting the data to the Michaelis–Menten equation using IGORPRO, version 6.03 (Wave-Metrics, Lake, Oswego, OR).

Growth properties of the host strain KU216 and the ΔTK0548 and ΔTK2268 gene disruption strains were examined in ASW-YT-m1-S⁰ and ASW-AA-m1-S⁰(+Ura) media. Cells were precultured in the nutrient-rich medium ASW-YT-m1-S⁰ for 15 h until the stationary phase and inoculated into ASW-YT-m1-S⁰ medium or synthetic medium ASW-AA-m1-S⁰(+Ura). The OD₆₆₀ of the culture was monitored.

Thermococcus kodakarensis KU216, ΔTK0548 and ΔTK2268 disruption strains were cultivated at ASW-YT-m1-pyruvate medium for 20 h and cells were collected by centrifugation (6,000 × g, 15 min, 4°C). Cells were disrupted by sonication and insoluble cell debris was removed (12,000 × g, 30 min, 4°C). After exchanging the buffer with 50 mM HEPES (containing 150 mM NaCl, pH7.4) using Amicon Ultra centrifugal filter unit (MWCO 10000), the aminotransferase activity in the cell-free extract was measured. For the aminotransferase activity toward His, Tyr, and Asp, the reaction mixture contained 20 μM PLP, 10 mM amino donor (the final concentration of Tyr was 6 mM), 10 mM amino acceptor (2-oxoglutarate or pyruvate), 5.76 mM NaCl and 0.384 mg mL⁻¹ cell-free extracts in 50 mM HEPES buffer (pH7.4). For Trp, the reaction mixture contained 20 μM PLP, 10 mM amino donor, 10 mM amino acceptor (2-oxoglutarate), 2.88 mM NaCl and 0.182 mg mL⁻¹ cell-free extracts in 50 mM HEPES buffer (pH7.4). In the case of Glu, Ile, and Phe, the reaction mixture contained 20 μM PLP, 10 mM amino donor, 10 mM amino acceptor (2-oxoglutarate or pyruvate), 0.576 mM NaCl and 0.0384 mg mL⁻¹ cell-free extracts in 50 mM HEPES buffer (pH7.4). The reaction mixture without amino acceptor was pre-incubated at 80°C for 2 min, and the amino acceptor was added to start the reaction. After further incubation at 80°C for 1, 2, and 3 min (When Asp and Ile were used as amino donor, the reaction mixtures were incubated at 80°C for 4, 6, and 8 min and 2, 4, and 6 min, respectively). The reaction was stopped through cooling the reaction mixture on ice for 10 min. Proteins were removed with an Amicon Ultra-0.5 centrifugal filter unit with an Ultracel-10 membrane (Millipore). The formation of Glu and Ala was determined by HPLC after derivatization.