TK1211 Encodes an Amino Acid Racemase towards Leucine and Methionine in the Hyperthermophilic Archaeon Thermococcus kodakarensis

Phylogenetic analysis of aminotransferase class III proteins from all domains of life reveals numerous groups of protein sequences. One of these groups includes a large number of sequences from Thermococcales species and can be divided into four subgroups.

ABSTRACT

Members of Thermococcales harbor a number of genes encoding putative aminotransferase class III enzymes. Here, we characterized the TK1211 protein from the hyperthermophilic archaeon Thermococcus kodakarensis. The TK1211 gene was expressed in T. kodakarensis under the control of the strong, constitutive promoter of the cell surface glycoprotein gene TK0895 (P_csg). The purified protein did not display aminotransferase activity but exhibited racemase activity. An examination of most amino acids indicated that the enzyme was a racemase with relatively high activity toward Leu and Met. Kinetic analysis indicated that Leu was the most preferred substrate. A TK1211 gene disruption strain (ΔTK1211) was constructed and grown on minimal medium supplemented with l- or d-Leu or l- or d-Met. The wild-type T. kodakarensis is not able to synthesize Leu and displays Leu auxotrophy, providing a direct means to examine the Leu racemase activity of the TK1211 protein in vivo. When we replaced l-Leu with d-Leu in the medium, the host strain with an intact TK1211 gene displayed an extended lag phase but displayed cell yield similar to that observed in medium with l-Leu. In contrast, the ΔTK1211 strain displayed growth in medium with l-Leu but could not grow with d-Leu. The results indicate that TK1211 encodes a Leu racemase that is active in T. kodakarensis cells and that no other protein exhibits this activity, at least to an extent that can support growth. Growth experiments with l- or d-Met also confirmed the Met racemase activity of the TK1211 protein in T. kodakarensis.

IMPORTANCE Phylogenetic analysis of aminotransferase class III proteins from all domains of life reveals numerous groups of protein sequences. One of these groups includes a large number of sequences from Thermococcales species and can be divided into four subgroups. Representatives of three of these subgroups have been characterized in detail. This study reveals that a representative from the remaining uncharacterized subgroup is an amino acid racemase with preference toward Leu and Met. Taken together with results of previous studies on enzymes from Pyrococcus horikoshii and Thermococcus kodakarensis, members of the four subgroups now can be presumed to function as a broad-substrate-specificity amino acid racemase (subgroup 1), alanine/serine racemase (subgroup 2), ornithine ω-aminotransferase (subgroup 3), or Leu/Met racemase (subgroup 4).

INTRODUCTION

The d-amino acids are widely distributed in microbes, plants, and animals. In addition to serving as building blocks for important biomolecules, some d-amino acids are inherently bioactive and act as specific signal molecules (1, 2). Peptides containing d-amino acid have enhanced stability against proteolysis and can serve as important recognition elements in mammalian systems (3). It has been reported that fermented foods using lactic acid bacteria as starters contained significant amounts of d-amino acids (4). d-Amino acids are primarily provided by the amino acid racemases, which catalyze the interconversion between the d- and l-enantiomers of amino acids (5). Amino acid racemases can be classified into two types, the pyridoxal 5′-phosphate (PLP)-independent enzymes and the PLP-dependent enzymes (6). The PLP-independent amino acid racemases, including aspartate racemase, proline racemase, and glutamate racemase, contain two Cys residues and use a thiol/thiolate pair for deprotonation and reprotonation (7). In contrast, most amino acid racemases catalyze the reaction via imine formation with PLP as a cofactor, exemplified by alanine racemase, serine racemase, arginine racemase, and the racemases with broad substrate specificity (8). PLP-dependent enzymes identified thus far are classified by structure into fold types I, II, III, IV, and V, and the PLP-dependent racemases are included in fold types I, II, and III (9).

Several archaeal species, including Desulfurococcus sp. strain SY, Pyrobaculum islandicum, Methanosarcina barkeri, and Halobacterium salinarum, have been shown to accumulate peptidyl and free d-amino acids, such as d-Ala, d-Asp, d-Glu, and d-Ser (10, 11), and a number of amino acid racemases from archaea have been studied. Alanine racemase activity was genetically confirmed in Methanococcus maripaludis (12). An amino acid racemase identified from the hyperthermophilic archaeon Desulfurococcus sp. strain SY was PLP independent and highly specific for aspartate (13). Homologs of this enzyme were later found to be ubiquitous among members of the hyperthermophilic genera Thermococcus and Pyrococcus (14–16). A proline racemase/hydroxyproline epimerase was identified from Thermococcus litoralis, and its crystal structure has been elucidated (17, 18). A PLP-dependent aspartate racemase was discovered in Thermoplasma acidophilum HO-62, which contained high proportions (up to 39.7%) of d-Asp to total Asp (19). Another aspartate racemase from Picrophilus torridus has also been characterized (20). A PLP-dependent serine racemase from Pyrobaculum islandicum was responsible for high concentrations of free d-Ser in the cells (21). Recently, broad-substrate-specificity amino acid racemase (BAR) encoded by PH0138 (22, 23) and Ala/Ser racemase (ASR) encoded by PH0782 (24) on the Pyrococcus horikoshii OT-3 genome have been biochemically characterized.

Here, we describe a detailed characterization of the TK1211 protein from the hyperthermophilic archaeon Thermococcus kodakarensis. The protein is a member of the class III PLP-dependent aminotransferase family, which is included in the fold type I proteins of PLP-dependent enzymes. We found that TK1211 protein was not an aminotransferase but exhibited amino acid racemase activity toward Ala, Phe, Val, Ile, Leu, and Met, with a clear preference toward Leu and Met. Genetic analysis of TK1211 confirmed that the protein functions as a leucine and methionine amino acid racemase in this archaeon.

RESULTS

Overexpression of the TK1211 gene and purification of the recombinant protein.

The class III aminotransferase family includes a wide range of aminotransferases, including ornithine aminotransferase, N-acetylornithine aminotransferase, 4-aminobutyrate (GABA) aminotransferase, and 5-aminovalerate aminotransferase. Within the class III aminotransferase family is a large group of proteins that is predominantly occupied by proteins from Archaea (25). This group includes four phylogenetically distinguishable subgroups (subgroups 1 to 4) (Fig. 1). TK1211 and the previously characterized TK2101 are two genes on the T. kodakarensis genome annotated as a γ-aminobutyrate (GABA) aminotransferase (GABA-AT) (26). TK2101 lies in subgroup 3, and TK1211 is a member of subgroup 4. Biochemical and genetic characterization of TK2101 revealed that the gene encodes an ornithine ω-aminotransferase required for the biosynthesis of proline in T. kodakarensis (25).

FIG 1.

Phylogenetic tree of selected class III aminotransferase sequences from archaea. Bootstrap values above 50 are shown. Abbreviation of source organism: Tko, Thermococcus kodakarensis KOD1; Tgg, Thermococcus gorgonarius W-12; Tpep, Thermococcus peptonophilus OG-1; Tnu, Thermococcus nautili 30-1; Tga, Thermococcus gammatolerans EJ3; Tha, Thermococcus sp. strain AM4; Tprf, Thermococcus profundus DT 5432; Tsl, Thermococcus siculi RG-20; Tpaf, Thermococcus pacificus P-4; Tgy, Thermococcus guaymasensis DSM 11113; Teu, Thermococcus eurythermalis A501; Thm, Thermococcus cleftensis CL1; The, Thermococcus sp. strain 4557; Tbs, Thermococcus barossii SHCK-94; Ttd, Thermococcus thioreducens OGL-20P; Ton, Thermococcus onnurineus NA1; Thh, Thermococcus sp. strain 5-4; Trl, Thermococcus radiotolerans EJ2; Tce, Thermococcus celer Vu 13; Tba, Thermococcus barophilus MP; Tpie, Thermococcus piezophilus CDGS; Ths, Thermococcus paralvinellae; Thy, Thermococcus sp. strain P6; Pch, Pyrococcus chitonophagus DSM 10152; Tlt, Thermococcus litoralis DSM 5473; Tsi, Thermococcus sibiricus MM 739; Tsl, Thermococcus siculi RG-20; Thg, Thermogladius calderae 1633; Tag, Thermosphaera aggregans DSM 11486; Ppac, Palaeococcus pacificus DY20341; Pho, Pyrococcus horikoshii OT3; Pfu, Pyrococcus furiosus DSM 3638; Pys, Pyrococcus sp. strain ST04; Pab, Pyrococcus abyssi GE5; Pya, Pyrococcus yayanosii CH1; Pyc, Pyrococcus kukulkanii NCB100; Pyn, Pyrococcus sp. strain NA2; Dka, Desulfurococcus amylolyticus 1221n; Dfd, Desulfurococcus amylolyticus DSM 16532; Dmu, Desulfurococcus mucosus DSM 2162; Ffo, Fervidicoccus fontis Kam940.

Here, we examined the function of the TK1211 gene. The gene was initially expressed in Escherichia coli, but we found that the protein was produced in an insoluble form with no activity. Thus, the gene was overexpressed in its native host T. kodakarensis under the control of the strong, constitutive promoter of the cell surface glycoprotein gene TK0895 (P_csg). The gene was modified so that the recombinant protein included a His₆ tag at its N terminus. Cell-free extracts of T. kodakarensis ETK1211 harboring the TK1211 protein were subjected to nickel chelate affinity and gel filtration chromatography. An SDS-PAGE analysis of the eluate after gel filtration indicated that the protein was purified to apparent homogeneity. The molecular mass of the protein was approximately 50 kDa (Fig. 2A), corresponding to that (49.92 kDa) calculated from the amino acid sequence of the TK1211 protein fused to a His₆ tag. The subunit assembly of the recombinant enzyme was estimated by gel filtration chromatography. The protein eluted as a single peak with a retention time corresponding to 200 kDa, indicating that the protein exists in buffer solution as a tetramer (Fig. 2B).

FIG 2.

Purification of the recombinant TK1211 protein. (A) Three micrograms of the TK1211 protein sample was applied, and the gel was stained with Coomassie brilliant blue. M, molecular weight marker. (B) Molecular weight of the TK1211 protein was estimated by gel filtration chromatography with a Superdex 200 10/300 column. Standard proteins were ferritin (440,000), aldolase (158,000), conalbumin (75,000), ovalbumin (44,000), and RNase (13,700).

The TK1211 protein displays amino acid racemase activity.

We first examined the ω-aminotransferase activity of the TK1211 protein, as it was annotated as GABA-AT and the structurally related TK2101 protein exhibited ornithine ω-aminotransferase activity (25). Activity toward GABA, l-ornithine, and l-Lys as amino donors and 2-oxoglutarate as an amino accepter was examined, but we could not detect any aminotransferase activity.

It has been shown that the putative GABA-AT genes from P. horikoshii (PH0138 and PH0782) and Lactobacillus buchneri (GenBank accession no. KC413940) displayed amino acid racemase activity (22–24, 27). The PH0138 and PH0782 proteins represent members of subgroup 1 and subgroup 2, respectively (Fig. 1). Thus, we examined the amino acid racemase activity of the TK1211 protein. Using d- or l-amino acids as substrates, we measured the generation of the corresponding enantiomer by high-performance liquid chromatography (HPLC) after derivatizing the product with o-phthalaldehyde (OPA) and N-acetyl-l-cysteine (NAC). We observed a significant level of activity with Leu or Met as the substrate. The specific activities toward all d- or l-amino acid substrates examined in this study are shown in Table 1. The results suggest that the TK1211 protein is an amino acid racemase with preference toward Leu and Met.

TABLE 1.

Racemase activity of the recombinant TK1211 protein^a

l- to d-form reaction			d- to l-form reaction
l-Form substrate	Sp act (μmol mg−1 ml−1)	Relative activity (%)	d-Form substrate	Sp act (μmol mg−1 ml−1)	Relative activity (%)
Leu	31.5 ± 2.6	100	Leu	48.1 ± 0.7	100
Met	14.5 ± 0.4	46	Met	22.3 ± 1.1	46.2
Phe	2.53 ± 0.06	8.04	Phe	3.42 ± 0.25	7.10
Ala	1.73 ± 0.02	5.50	Ala	2.51 ± 0.04	5.21
Ile*	0.23 ± 0.01	0.74	Allo-Ile*	0.28 ± 0.01	0.57
Val	0.20 ± 0.01	0.64	Val	0.24 ± 0.02	0.51
Trp	ND		Trp	ND
Glu	ND		Glu	ND
Arg	ND		Arg	ND
Asp	ND		Asp	ND
Gln	ND		Gln	ND
Thr**	ND		Thr**	ND
Tyr	ND		Tyr	ND
Asn	ND		Asn	ND
Ser	ND		Ser	ND
Pro	ND		Pro	ND
Lys	ND		Lys	ND
His	ND		His	ND
Cys	ND		Cys	ND
Gly

^aThe data are averages from three independent measurements along with their standard deviations. ND, not detected. *, l-Ile (2S,3S), d-allo-Ile (2R,3S); **, l-Thr (2S,3R), d-Thr (2R,3R).

l-Leu, d-Leu, l-Met, and d-Met were individually incubated at 75°C with recombinant TK1211 protein and PLP. When the substrate was added at 10 mM, in all four cases we observed a complete racemization of the substrate, resulting in equivalent concentrations (5 mM) of l- and d-forms of the amino acids (Fig. 3).

FIG 3.

Time course of amino acid racemization with the recombinant TK1211 protein. The substrates applied were d-Leu (A), l-Leu (B), d-Met (C), or l-Met (D). The reactions were carried out with 50 mM acetate-NaOH buffer (pH 5.0), 10 mM amino acid, and 4 μg ml⁻¹ of purified enzyme at 75°C. Filled squares and open squares represent concentrations of d-Leu and l-Leu, respectively. Filled circles and open circles represent concentrations of d-Met and l-Met, respectively. The data represent averages from three measurements and are shown with standard deviations.

Effect of pH and temperature on enzyme activity.

The effects of pH and temperature on the reaction catalyzed by the TK1211 protein were examined using d-Leu and l-Leu as substrates. In terms of pH, the highest activities for the d-Leu to l-Leu reaction and the l-Leu to d-Leu reaction were obtained at pH 5.0 in 50 mM acetate-NaOH buffer (Fig. 4A and B). In terms of temperature, the maximum activity of the d-Leu-to-l-Leu reaction was detected at 70°C within the examined temperature range (30 to 90°C), and the maximum activity of the l-Leu-to-d-Leu reaction was obtained at 75°C (Fig. 4C). To investigate the thermostability of TK1211, the purified enzyme was incubated at 60, 70, 80, and 90°C for various periods of time, and residual activity was measured at 75°C. As shown in Fig. 4D, the TK1211 protein displayed typical inactivation behavior at 60 and 70°C, with half-lives of 26.6 h and 22.7 h, respectively. However, at 80 and 90°C, we observed an initial sharp drop in activity, followed by a decrease in inactivation rate. As we observed that the enzyme solution became slightly cloudy at these temperatures, the initial sharp drop might be due to inactivation caused by protein aggregation, which is dependent on protein concentration. As precipitation would also lower the protein concentration of the solution, the situation afterwards might reflect the intrinsic denaturation of the protein, which might explain the relatively stable nature of the protein after 2 h of incubation at 80 and 90°C.

FIG 4.

Effects of pH and temperature on the racemase activity of the TK1211 protein. (A and B) Effects of pH with d-Leu (A) or l-Leu (B) as the substrate. Reactions were carried out with 50 mM buffers, 20 mM d-Leu or 20 mM l-Leu, and 2 μg ml⁻¹ of purified enzyme at 80°C. Squares, circles, triangles, hexagons, and diamonds represent the activities in citrate-NaOH (pH 3.0 to 4.0), acetate-NaOH (pH 4.0 to 5.5), MES-NaOH (pH 5.5 to 7.0), HEPES-NaOH (pH 7.0 to 8.0), and Tricine-NaOH (pH 8.0 to 9.0), respectively. (C) Effects of temperature on amino acid racemase activity. Reactions were carried out with 50 mM acetate-NaOH buffer (pH 5.0), 20 mM d-Leu or l-Leu, and 2 μg ml⁻¹ purified enzyme at different temperatures. Filled squares and circles represent the activities with d-Leu and l-Leu, respectively, as the substrate. (D) Thermostability of the TK1211 protein. Residual activity was determined with 50 mM acetate-NaOH buffer (pH 5.0), 20 mM d-Leu, and 2 μg ml⁻¹ purified enzyme at 75°C. Squares, circles, triangles, and diamonds represent residual activities after incubation at 90, 80, 70, and 60°C, respectively. The data represent averages from three measurements and are shown with standard deviations.

Effect of PLP on enzyme activity.

Amino acid racemases are divided into two groups, the PLP-dependent racemase and the PLP-independent racemase. To investigate the effect of PLP on enzyme activity, the TK1211 protein was first dialyzed against 50 mM sodium phosphate buffer (pH 7.4) for 16 h at 4°C. When PLP was not added to the reactions with d-Leu as the substrate, the activity decreased to approximately 15% compared to that in the presence of PLP. Furthermore, when hydroxylamine, a known specific PLP inhibitor, was added to the reaction mixture, the amino acid racemase activity was completely abolished (data not shown). The results clearly indicate that the TK1211 protein is a PLP-dependent racemase.

Kinetic analysis of the TK1211 protein reaction.

Initial velocities of the amino acid racemase activity were examined with various concentrations of Leu or Met. The temperature was set at 75°C, and reactions were carried out for 5 min to minimize enzyme inactivation. Substrate concentrations were varied from 0 to 40 mM. At higher concentrations, complete separation of the l- or d-amino acids using HPLC became difficult. The reactions with Leu and Met both followed Michaelis-Menten kinetics (Fig. 5). The kinetic parameters are shown in Table 2. The k_cat/K_m value with Leu was 2-fold higher than that with Met, suggesting that Leu is the preferred substrate of the TK1211 protein.

FIG 5.

Kinetic studies of the TK1211 protein. (A) Initial velocities of the racemase activity of the TK1211 protein with various concentrations of Leu as the substrate. Filled squares and open squares represent initial velocities with d-Leu and l-Leu, respectively. (B) Initial velocities of the racemase activity of the TK1211 protein with various concentrations of Met as the substrate. Filled squares and open squares represent initial velocities with d-Met and l-Met, respectively.

TABLE 2.

Kinetic parameters of TK1211 protein toward enantiomers of Leu and Met

Substrate	Vmax (μmol mg−1 min−1)	Km (mM)	kcat (s−1)	kcat/Km (s−1 · mM−1)
d-Leu	133 ± 7	16.3 ± 1.9	108 ± 6	6.64
l-Leu	53 ± 3	6.98 ± 1.33	43.3 ± 2.5	6.20
d-Met	105 ± 6	26.8 ± 3.1	86.0 ± 5.3	3.21
l-Met	66.6 ± 5.7	18.0 ± 4.7	54.4 ± 4.7	3.02

Construction of a TK1211 gene disruption strain.

To examine the physiological role of the TK1211 protein, a TK1211 gene disruption strain was constructed using T. kodakarensis KU216 (ΔpyrF) as the host strain. A plasmid was constructed with the 5′- and 3′-flanking regions of TK1211 directly fused to one another. The plasmid harbored a pyrF marker gene so that cells that had undergone single-crossover insertion of the plasmid could be selected/enriched by uracil prototrophy, and those that had undergone a second, pop-out recombination could be selected by resistance to 5-fluroorotic acid (5-FOA). A number of transformants were selected, and PCR analysis was carried out to examine the deletion of the TK1211 gene (Fig. 6). DNA sequencing confirmed the complete deletion of the TK1211 gene from the genome of T. kodakarensis.

FIG 6.

Gene disruption and PCR analysis of the TK1211 locus. Amplification was carried out with primers annealing at the termini of the homologous regions (A/B) used for gene disruption (1211F1/1211R1) and primers within the coding region (1211F3/1211R3). The predicted loci and lengths of the amplified products are illustrated under the gel. Abbreviations: M, marker; host, T. kodakarensis host strain KU216; Δ, TK1211 gene disruption strain.

Growth characteristics of the TK1211 gene disruption strain.

The TK1211 gene disruption strain (ΔTK1211) and the host strain KU216 were grown in a number of media to examine whether TK1211 was responsible for Leu and/or Met racemase activity in T. kodakarensis cells. Wild-type T. kodakarensis KOD1 and the host strain KU216 do not have the capability to synthesize Leu and display Leu auxotrophy. On the other hand, both strains can synthesize Met and can grow in a medium without Met. The experiments described below all were carried out in a synthetic medium based on amino acids supplemented with sodium pyruvate and uracil (ASW-AA-S⁰-Pyr-Ura-W). Only the presence or absence of relevant amino acids (Leu and Met) will be indicated.

We first examined the effects of TK1211 gene disruption on Leu racemization. When KU216 and ΔTK1211 cells were inoculated in medium with l-Leu {[l-Leu(+)d-Leu(–)]}, ΔTK1211 cells showed a slight delay in entering the exponential phase but displayed growth rates and growth yield similar to those of the KU216 host strain (Fig. 7A). When KU216 cells were grown in [l-Leu(–)d-Leu(+)] medium, the initiation of cell growth was delayed significantly, but eventually growth rates and cell yield were similar to those of the host strain grown on [l-Leu(+)d-Leu(–)]. In the case of ΔTK1211 cells, however, growth was not observed in [l-Leu(–)d-Leu(+)] medium, indicating that the function of the TK1211 gene, i.e., the Leu racemase activity of the TK1211 protein, was necessary to provide l-Leu from d-Leu for cell growth. The result strongly suggests that the TK1211 protein functions as a Leu racemase in T. kodakarensis cells and that no other protein displays this activity, at least to a degree that is sufficient to support growth.

FIG 7.

Growth properties of the host strain KU216 and the TK1211 gene disruption strain in the presence of l- or d-Leu (A) or l- or d-Met (B). (A) To examine the effects of TK1211 gene on Leu racemization, the host and ΔTK1211 strains were incubated in [l-Leu(+)d-Leu(–)] and [l-Leu(–)d-Leu(+)] medium. Filled squares and open squares represent the host and ΔTK1211 strains in [l-Leu(+)d-Leu(–)] medium, respectively, and filled circles and open circles represent the host and ΔTK1211 strains in [l-Leu(–)d-Leu(+)] medium, respectively. (B) The host and ΔTK1211 strains were incubated in [l-Met(–)d-Met(–)], [l-Met(-)d-Met(+)], or [l-Met(+)d-Met(–)] medium. Filled squares and open squares represent the host and ΔTK1211 strains in [l-Met(–)d-Met(–)] medium, filled triangles and open triangles represent the host and ΔTK1211 strains in [l-Met(+)d-Met(–)] medium, and filled circles and open circles represent the host and ΔTK1211 strains in [l-Met(-)d-Met(+)] medium, respectively. The data represent averages from three independent culture experiments and are shown with standard deviations.

We next examined the contribution of TK1211 to Met metabolism. When cells were grown in standard medium with all l-amino acids {[l-Met(+)d-Met(–)]}, KU216 and ΔTK1211 strains both grew with similar growth rates (Fig. 7B). We did, however, notice a slight delay in the initiation of growth in ΔTK1211 cells compared to KU216 cells, and the maximum cell yield was also slightly lower. When we grew KU216 cells in medium without Met {[l-Met(–)d-Met(–)]}, cell growth was observed, but only after a significant elongation of the lag phase. Surprisingly, this delay in the initiation of growth was not observed in ΔTK1211 cells grown in [l-Met(–)d-Met(–)]. When d-Met was added to the medium in the absence of l-Met, we found that both strains displayed similar growth curves. As both KU216 and ΔTK1211 can synthesize l-Met, the results here are not as clear as those obtained with the growth experiments with Leu. However, the stark difference between KU216 and ΔTK1211 grown in [l-Met(–)d-Met(–)] does suggest the function of TK1211 as a Met racemase. The presence of Met racemase activity can be presumed to release a major portion of the l-Met synthesized in the cell by converting it into d-Met, whereas this activity is absent in ΔTK1211 and, thus, cells can maintain a higher concentration of l-Met. In addition, d-Met may have a moderate inhibitory effect on the growth of T. kodakarensis.

DISCUSSION

In this study, through biochemical and genetic analyses, we have elucidated the function of TK1211, a member of the class III aminotransferases, representing a phylogenetically related group of proteins (Fig. 1, subgroup 4) that had not been functionally characterized in detail. The protein displayed Leu and Met racemase activity, with relatively lower levels of activity toward Phe and Ala and trace activities toward Ile and Val (Table 1).

Among the archaeal species, an abundant number of amino acid racemases from P. horikoshii have been studied. An Asp racemase encoded by PH0670 has been structurally examined, along with a closely related homolog, PH1733 (15, 28, 29). An amino acid racemase induced in the presence of d-amino acids was also identified and shown to be the product of PH0138 (22, 23). The enzyme was PLP dependent and displayed broad substrate specificity toward Met, Leu, Phe, Trp, Tyr, Thr, Ala, Ser, Ile, and Val; thus, it was designated BAR (broad-substrate-specificity amino acid racemase). PH0138 is a representative of subgroup 1 proteins presented in Fig. 1.

Three other proteins encoded by genes on the P. horikoshii genome display similarity to BAR and are encoded by PH0782, PH1423, and PH1501. The product of PH0782, which is a member of subgroup 2, has been characterized in detail (24). The enzyme displayed strong activity toward Ala and Ser, with lower levels of activity toward Val and Thr. Site-directed mutagenesis indicated that Lys291 and Asp234 were essential for enzyme activity, most likely due to the involvement of these residues in interaction with PLP. In contrast to PH0138, PH0782 expression does not seem to be induced when d-amino acids are added to the medium, raising the possibility of a role other than assimilation of d-Ser and/or d-Ala. We would like to note a statement concerning PH1501 in the study on PH0782 (24). The authors state that slight peaks corresponding to d-Met/Phe/Leu were generated when l-Met/Phe/Leu were incubated in E. coli cell extracts harboring the PH1501 gene. We are not aware of any data or further studies. However, as PH1501, along with TK1211, is a representative of subgroup 4 and our results indicate that the TK1211 protein is an amino acid racemase preferring Leu and Met and, to a lesser extent, Phe and Ala, we acknowledge this statement. On the contrary, the product of PH1423 did not show any racemase activity (24). This may be because PH1423 resides in subgroup 3 along with the TK2101 protein, which has been shown to function as an ornithine ω-aminotransferase for proline biosynthesis in T. kodakarensis (25).

If we presume that each member within the four individual subgroups displays similar enzyme activity, P. horikoshii harbors three different racemases and an ornithine ω-aminotransferase, as a protein from this archaeon is found in all four subgroups. It also possesses the Asp racemase that does not fall into any of these subgroups. Thus, the organism can potentially catalyze the conversion of the d-enantiomers of Asp, Met, Leu, Phe, Trp, Tyr, Thr, Ala, Ser, Ile, and Val to their l-enantiomers. In contrast, T. kodakarensis does not have members of subgroups 1 and 2. It does have a closely related homolog (67% identical) of the PH0670 Asp racemase, encoded by TK0504, and, thus, should be able to convert the d-enantiomers of Asp, Leu, Met, Phe, and Ala to their l-counterparts or vice versa. The reason for this difference among organisms is unknown, but P. horikoshii does not seem to represent the majority of members of the Thermococcales. Those that possess the BAR enzymes of subgroup 1 are a minority and may reflect horizontal gene transfer from bacteria, as proteins from bacterial members are also found in this subgroup.

One question that remains is the physiological role of the racemases. In the case of the TK1211 protein, we have clearly shown that it can function for the assimilation of d-amino acids. In P. horikoshii, the expression of PH0138 is induced upon addition of d-amino acids to the medium, supporting a role in d-amino acid assimilation (23). However, these results alone do not rule out a role of amino acid racemases for biosynthetic purposes to generate d-amino acid-related compounds. The presence of d-amino acids has been examined in a range of archaea, including P. islandicum, Sulfolobus acidocaldarius, M. barkeri, and H. salinarum, and d-Ala, d-Ser, d-Pro, d-Glu, and d-Asp have been detected (10, 11). Studies have also been carried out on Thermoplasma acidophilum, Sulfolobus sp. strain 7, Natronomonas (Natronobacterium) pharaonis, Haloarcula vallismortis, Halobacterium salinarum strain S9, and Desulfurococcus sp. strain SY, revealing the presence of d-Ala, d-Glu, d-Asp, d-Leu, d-Lys, and d-Phe (16, 19). In terms of members of Thermococcales, d-Ala, d-Asp, d-Leu, d-Lys, and d-Phe were detected (16). In particular, the proportion of free d-Asp among total Asp exceeded 40% in Thermococcus sp. strains KS-1 and KS-8. The proportions of d-Leu in Thermococcus sp. strains KS-1 and KS-8 were also significant at 28.4% and 6.3%, respectively. Although information on the genome sequences of these strains is not available, it is possible that a homolog of TK1211 is involved in the generation of d-Leu in these organisms. In the case of T. kodakarensis, we have demonstrated that the TK1211 protein is the only protein that can catalyze the conversion of d-Leu and l-Leu, at least to an extent that can support cell growth.

At present, whether these d-amino acids are utilized for biosynthetic purposes in archaea is still an open question. Biomolecules from archaea that include d-amino acids have not yet been reported, and further studies will be needed for their identification. The results of TK1211 disruption show that d-Leu generation via TK1211 function is not essential for growth of T. kodakarensis and did not result in retarded growth under standard culture conditions. Thus, the biosynthesis of compounds containing d-Leu, if any, might only be necessary under growth conditions not examined under our laboratory growth conditions.

MATERIALS AND METHODS

Strains, media, and culture conditions.

T. kodakarensis is a hyperthermophilic archaeon isolated from Kodakara Island, Japan (30, 31). The cultivation of T. kodakarensis KU216 (32) and its derivative strains was performed under anaerobic conditions at 85°C in nutrient-rich ASW-YT medium or minimal ASW-AA medium. ASW-YT medium was composed of 0.8× artificial seawater (ASW), 5 g liter⁻¹ yeast extract, and 5 g liter⁻¹ tryptone. Elemental sulfur (2 g liter⁻¹) and sodium pyruvate (5 g liter⁻¹) were added prior to inoculation to prepare ASW-YT-S⁰ and ASW-YT-Pyr medium, respectively. ASW-AA medium consists of 0.8× ASW, a mixture of 20 amino acids, modified Wolfe’s trace minerals, a vitamin mixture, and 2 g liter⁻¹ elemental sulfur (ASW-AA-S⁰ medium) (33). In the case of plate culture, elemental sulfur was replaced with 2 ml polysulfide solution (10 g Na₂S·9H₂O and 3 g sulfur flowers in 15 ml H₂O) per liter, and 1% gelrite was added to solidify the medium. Escherichia coli strain DH5α and plasmids pUC118 and pUD3 were used for general DNA manipulation. E. coli strains were cultivated in lysogeny broth medium at 37°C containing ampicillin (100 mg liter⁻¹). Unless otherwise indicated, all chemicals were purchased from Fujifilm Wako Pure Chemicals (Osaka, Japan) or Nacalai Tesque (Kyoto, Japan).

Gene disruption of TK1211.

The gene disruption plasmid for TK1211 was constructed by amplifying the gene along with its 5′- and 3′-flanking regions (approximately 1.0 kbp) using the primer set 1211F1/1211R1 (Table 3). The fragment was inserted into the plasmid pUD3 at the HincII site, which contains the pyrF marker gene inserted in the ApaI site of pUC118. Inverse PCR was performed with the primer set 1211F2/1211R2 (Table 3), and the obtained fragment was self-ligated. Sequences of the 5′- and 3′-flanking regions were confirmed.

TABLE 3.

Primers used in this study

Name	Sequencea
1211F1	5′-GGGCTTCCTGGCAATAGCGTTTGG-3′
1211R1	5′-GGAATAGCTCGTTGTTCCGCAGGC-3′
1211F2	5′-ATGGAGAATGTCGAAGTTGTTG-3′
1211R2	5′-GCCGGTTCTCACCTAACTGTTATATG-3′
1211F3	5′-AACATATGCACCACCACCACCACCACACTCCTGAAGAGGTCGTTGAGAG-3′
1211R3	5′-AAGTCGACTTACCATCCCTGCACTTTCTCAATG-3′

^aRecognition sites of restriction enzymes are underlined.

T. kodakarensis KU216 (ΔpyrF), which displays uracil auxotrophy, was used as the host strain for TK1211 gene disruption. KU216 cells grown in ASW-YT-S⁰ medium for 12 h were harvested, resuspended in 200 μl of 0.8× ASW, and kept on ice for 30 min. After adding 3.0 μg of plasmid DNA and further incubation on ice for an hour, cells were cultivated twice in ASW-AA-S⁰ medium without uracil for 48 h at 85°C to enrich the desired transformants exhibiting uracil prototrophy. Cells were then diluted with 0.8× ASW and spread onto ASW-YT solid medium supplemented with 10 g liter⁻¹ 5-fluoroorotic acid (5-FOA) and 60 mM NaOH. As cells with an intact pyrimidine biosynthesis pathway convert 5-FOA to the toxic 5-fluoroorotidine 5′-phosphate, only cells that have undergone a popout recombination can grow under these conditions. Cells were grown at 85°C for 2 days. Individual colonies were selected, and their genotypes were analyzed with the primer sets 1211F1/1211R1 and 1211F3/1211R3 (Table 3). Transformants whose amplified DNA products showed the expected size were chosen and cultivated in ASW-YT-S⁰ medium. All loci were analyzed by PCR and sequenced to confirm the absence of unintended mutations.

Overexpression of TK1211 and purification of the recombinant protein.

The coding region of the TK1211 gene was amplified from the genomic DNA of T. kodakarensis KU216 by PCR using the primer set 1211F3/1211R3 (Table 3), and a His₆ tag sequence was incorporated in the N terminus for use in purification. Using the NdeI-SalI restriction enzyme sites incorporated during PCR, the amplified fragment was inserted into a T. kodakarensis-E. coli shuttle plasmid previously used for overexpression of TK2101 (25, 34). The resulting TK1211 overexpression plasmid pRPETK1211 was confirmed by DNA sequencing.

T. kodakarensis KPD1 (ΔpyrF ΔpdaD), which shows uracil and agmatine auxotrophy (35, 36), was used as the host strain for TK1211 gene expression. T. kodakarensis KPD1 was cultivated in ASW-YT-S⁰ medium supplemented with 0.5 mM agmatine for 12 h at 85°C. Cells were harvested, resuspended with 200 μl of 0.8× ASW, and kept on ice for 30 min. After treatment of 3.0 μg of pRPETK1211 and further incubation on ice for 1 h, cells were spread onto solid ASW-YT-S⁰ medium and cultivated at 85°C for 24 h. Transformants displaying agmatine prototrophy were isolated and designated T. kodakarensis ETK1211. The presence of the plasmid was confirmed by PCR.

The ETK1211 strain was cultivated in ASW-YT-Pyr for 12 h at 85°C. Cells were resuspended in 20 mM sodium phosphate buffer (pH 7.4) containing 500 mM KCl, 20 mM imidazole, and 0.2 mM PLP. Cells were disrupted by sonication and centrifuged at 15,000 × g for 20 min, and the supernatant was loaded onto His GraviTrap (GE Healthcare, Chicago, IL). Proteins were eluted with elution buffer (pH 7.4) containing 20 mM sodium phosphate, 500 mM KCl, 500 mM imidazole, and 0.2 mM PLP. After exchanging the elution buffer to 50 mM sodium phosphate (pH 7.4) containing 150 mM NaCl and 0.2 mM PLP, the sample was applied to a Superdex 200 10/300 gel filtration column (GE Healthcare) with a mobile phase of 50 mM sodium phosphate buffer (pH 7.4) containing 150 mM NaCl and 0.2 mM PLP at a flow rate of 0.5 ml min⁻¹. The molecular mass of recombinant TK1211 protein was determined from a calibration curve with the standard proteins ferritin (440 kDa), aldolase (158 kDa), conalbumin (75 kDa), ovalbumin (44 kDa), and RNase (13.7 kDa). Protein concentrations were determined with a protein assay kit (Bio-Rad, Hercules, CA) with bovine serum albumin as the standard.

Analysis of d- and l-amino acids by HPLC.

d,l-Amino acids were detected and quantified by HPLC after derivatization using methanolic solution containing o-phthalaldehyde (OPA) and N-acetyl-l-cysteine (NAC) (27). The derivatization mixture (250 μl) contained 25 μl of amino acid sample, 50 μl of methanolic solution, and 175 μl of borate-NaOH buffer (0.5 M, pH 10.0). After derivatization for 3 min at room temperature in the dark, an aliquot (1 μl) of the solution was applied to a Gemini 5-μm C₁₈ 150- by 4.60-mm column using an LC-VP system (Shimadzu, Kyoto, Japan). The excitation and emission wavelengths for fluorescent detection of the diastereoisomeric derivatives of amino acids formed with OPA-NAC were 350 nm and 450 nm, respectively. The system was operated at a flow rate of 0.8 ml min⁻¹ at 30°C. The mobile phase was composed of 50 mM sodium acetate (pH 5.9, A) and methanol (B). The ratio of mobile phase and gradient system for analysis of OPA-NAC derivatives were the following: Leu and His, 50% B for 20 min; Met, Phe, Ile, Val, and Trp, 40% B for 25 min; Tyr and Lys, 35% B for 30 min; Thr, 25% B for 30 min; Asp, 0% to 20% B for 16 min; Ser and Glu, 0% to 16% B for 16 min and then 20% B for 6 min; Arg, 0% to 20% B for 16 min and then 20% B for 6 min and 20% to 40% B for 18 min; Ala, 0% to 20% B for 16 min and then 20% B for 24 min; Asn, 0% to 20% B for 25 min; Gln, 0% to 20% B for 25 min and then 20% B for 15 min.

Enzyme assay.

The amino acid racemase activity of the TK1211 protein was assayed by monitoring the generation of the opposite enantiomer from d- or l-Leu as the substrate. The standard reaction mixture (300 μl) contained 20 mM d- or l-Leu, 0.2 mM PLP, 2 μg ml⁻¹ recombinant TK1211 protein, and 50 mM acetate-NaOH (pH 5.0 at 75°C). Unless mentioned otherwise, the reaction was performed at 75°C for 3, 5, and 10 min and quenched by the addition of 100 μl HCl (1.0 M). TK1211 protein was removed by ultrafiltration, and aliquots were applied for derivatization. The amount of product, l- or d-Leu, in the solution was then measured using HPLC as described above. One unit of enzyme activity was defined as the amount of enzyme catalyzing the formation of 1.0 μmol l- or d-Leu per min under standard conditions.

To determine the kinetic parameters, the initial velocity toward d- and l-forms of Leu or Met were measured by varying the concentration from 1 to 40 mM. Kinetic parameters were obtained by fitting the data to the Michaelis-Menten equation using IGORPRO, version 6.03 (Wave-Metrics, Lake, Oswego, OR).

Effects of pH and temperature on enzyme activity.

To examine the effect of pH on enzyme activity, the racemization reactions were carried out under standard reaction conditions using the following buffers at 50 mM: citrate-NaOH (pH 3.0 to 4.0), acetate-NaOH (pH 4.0 to 5.5), morpholine ethanesulfonic acid (MES)-NaOH (pH 5.5 to 7.0), HEPES-NaOH (pH 7.0 to 8.0), and Tricine-NaOH (pH 8.0 to 9.0). To investigate the effect of temperature, the reactions were performed at various temperatures (30 to 90°C) under standard reaction conditions. The thermostability of the recombinant TK1211 protein was examined at 60, 70, 80, and 90°C with enzyme preincubated for various times at the corresponding temperatures. After the protein solution was cooled on ice for 30 min, the residual activity was assayed under standard reaction conditions with d-Leu as the substrate. Measurements were carried out in triplicate.

Effect of PLP and PLP inhibitor on enzyme activity.

The effect of PLP on the TK1211 protein activity was investigated after dialysis of the recombinant enzyme solution. The TK1211 protein was dialyzed with 50 mM sodium phosphate buffer (pH 7.4) for 16 h at 4°C. Enzyme activity was then measured under the standard assay conditions with 2.0 μg ml⁻¹ enzyme and d-Leu as the substrate without PLP. In addition, the effects of a specific PLP inhibitor (15.0 mM hydroxylamine) on the TK1211 protein activity were examined under the standard assay conditions using d-Leu as a substrate. Measurements were carried out in triplicate.

Time course of Leu and Met catalyzed by recombinant enzyme.

The reaction mixture containing 10 mM Leu or Met, 0.2 mM PLP, 50 mM acetate-NaOH (pH 5.0), and 4 μg ml⁻¹ purified enzyme was incubated at 75°C. Aliquots were taken after various periods of time and measured. Measurements were carried out in triplicate.

Phylogenetic analysis.

Sequences were aligned by the multiple-sequence alignment software ClustalW and Molecular Evolutionary Genetic Analysis (MEGA) software, version 7.0.26. Phylogenetic trees were constructed with the method UPGMA by using the software. Bootstrap resampling was performed 1,000 times.

Data availability.

All data are included in the paper.