Method for producing optically active amine compounds by deracemization

Abstract

Disclosed are methods for producing optically active amino acids and amines. According to the methods, α-keto acids are generated as reaction intermediates, and as a result, ω-transaminase-catalyzed kinetic resolution of racemic amino acids or amines as racemic amine compounds enables the production of optically active amine compounds without the need to use expensive α-keto acids as starting materials. Therefore, the optically active amine compounds are produced at greatly reduced costs. In addition, the optically active amine compounds have high enantiomeric excess.

Claims

1. A one-pot method for producing an optically pure amino acid by deracemizing a racemic amino acid with two transaminases of opposite enantioselectivity and a co-substrate dedicated for each transaminase, comprising: 1) providing a racemic amino acid, an amino acceptor for an amino acid transaminase, and an amino donor for a co-transaminase, to the amino acid transaminase and the ω-transaminase whose enantioselectivities are opposite to each other; 2) contacting the racemic amino acid, the amino acceptor, and the amino acid transaminase provided in step 1) to convert one amino acid enantiomer of the racemic amino acid to an α-keto acid; and 3) contacting the α-keto acid produced in step 2), the amino donor provided in step 1), and the ω-transaminase provided in step 1) to convert the α-keto acid to the amino acid enantiomer whose chirality is opposite to the amino acid enantiomer reacted in step 2), wherein an (R)-amino acid transaminase and a (S)-selective ω-transaminase are used as the amino acid transaminase and the ω-transaminase, respectively, to deracemize the racemic amino acid into optically pure (S)-amino acid, or wherein a (S)-amino acid transaminase and an (R)-selective ω-transaminase are used as the amino acid transaminase and the ω-transaminase, respectively, to deracemize the racemic amino acid into optically pure (R)-amino acid.

2. The one-pot method according to claim 1, wherein the (R)-amino acid transaminase is provided in step 1), the (R)-amino acid of the racemic amino acid and the amino acceptor are contacted with the R-amino acid transaminase to produce an α-keto acid in step 2), and the α-keto acid is contacted with the (S)-selective ω-transaminase to produce a (S)-amino acid of the α-keto acid in step 3).

3. The one-pot method according to claim 1, wherein the (S)-amino acid transaminase is provided in step 1), the (S)-amino acid of the racemic amino acid and the amino acceptor are contacted with the (S)-amino acid transaminase to produce an α-keto acid in step 2), and the a-keto acid is contacted with the (R)-selective ω-transaminase to produce an (R)-amino acid of the a-keto acid in step 3).

4. The one-pot method according to claim 1, wherein the (S)-selective ω-transaminase is an enzyme derived from Ochrobactrum anthropi and encoded by the nucleic acid sequence set forth in SEQ ID NO: 1 or an enzyme derived from Paracoccus denitrificans and encoded by the nucleic acid sequence set forth in SEQ ID NO: 2.

5. The one-pot method according to claim 1, wherein the (R)-selective co-transaminase is an enzyme derived from Aspergillus terreus and encoded by the nucleic acid sequence set forth in SEQ ID NO: 3 or an enzyme derived from Arthrobacter sp. and encoded by the nucleic acid sequence set forth in SEQ ID NO: 4.

6. The one-pot method according to claim 1, wherein the racemic amino acid is selected from the group consisting of racemic alanine, racemic serine, racemic homoserine, racemic norvaline, racemic norleucine, racemic leucine, and mixtures thereof.

7. The one-pot method according to claim 1, wherein the amino acceptor is selected from the group consisting of a-ketoglutarate, phenylglyoxylate, phenylpyruvate, and mixtures thereof.

8. The one-pot method according to claim 1, wherein the amino donor is selected from the group consisting of isopropylamine, methylbenzylamine, benzylamine, and mixtures thereof.

9. The one-pot method according to claim 1, wherein the (R)-selective amino acid transaminase is an enzyme derived from Bacillus sphaericus and encoded by the nucleic acid sequence set forth in SEQ ID NO: 5.

10. The one-pot method according to claim 1, wherein the (R)-selective amino acid transaminase is an enzyme derived from Escherichia coli and encoded by the nucleic acid sequence set forth in SEQ ID NO: 6.

Description

BRIEF DESCRIPTION OF THE DRAWINGS

(1) These and/or other aspects and advantages of the invention will become apparent and more readily appreciated from the following description of the embodiments, taken in conjunction with the accompanying drawings of which:

(2) FIG. 1 is a graph showing the inactivation of a D-amino acid transaminase toward L-amino acid;

(3) FIG. 2 is a graph showing the conversion to an L-amino acid when isopropylamine was used as an amino donor;

(4) FIG. 3 is a graph showing the throughput of L-homoalanine from racemic homoalanine according to one embodiment of the present invention.

(5) FIG. 4 is a graph showing changes in the concentration of α-ketoglutarate and isopropylamine as substrates and D-glutamate as a final product as a function of reaction time according to one embodiment of the present invention;

(6) FIG. 5 is a graph comparing the activities of pyruvate and propanal as amino acceptors for an ω-transaminase; and

(7) FIG. 6 is a graph showing changes in the concentration of (S)-α-methylbenzylamine and (R)-α-methylbenzylamine as a function of reaction time when an (R)-selective ω-transaminase was used in accordance with the present invention.

DETAILED DESCRIPTION OF THE INVENTION

(8) The present inventors have earnestly conducted research to develop methods for producing optically active amine compounds at reduced costs and as a result found that asymmetric, optically active amine compounds can be produced by deracemization. The present invention has been accomplished based on the finding.

(9) Generally, resolution of optically pure amine compounds from racemic amine compounds can be performed by suitable methods such as kinetic resolution, asymmetric synthesis, and deracemization. Particularly, the production of an optically pure amine compound via deracemization can be achieved by deracemizing a racemic amine compound as a substrate and reacting the deracemized substrate with an α-keto acid and an (S)- or (R)-selective ω-transaminase. The present invention is directed to methods for producing optically active amine compounds with high purity at reduced costs by generating α-keto acids as intermediate without the need to provide expensive α-keto acids as substrates, which are currently used for the optical resolution of racemic amine compounds. The racemic amine compounds may be racemic amines or amino acids. Optically active amine compounds produced by the methods of the present invention may be amino acid and amine compounds with high enantiomeric excess.

(10) Specifically, the present invention provides a method for producing an L-amino acid, including 1) providing a racemic amino acid, an amino acceptor, and an amino donor as substrates to a D-amino acid transaminase and an (S)-selective ω-transaminase as enzymes, 2) reacting the racemic amino acid substrate and the amino acceptor substrate provided in step 1) with the D-amino acid transaminase to generate an α-keto acid, and 3) reacting the α-keto acid with the (S)-selective ω-transaminase to produce an optically active L-amino acid.

(11) According to the method of the present invention, the L-amino acid may be produced via Reaction Scheme 1:

(12) ##STR00001##

(13) wherein each R.sub.1 is independently selected from the group consisting of methyl, ethyl, propyl, butyl, hydroxymethyl, hydroxyethyl, 2-methylpropyl and combinations thereof, and each R.sub.2 may be 2-carboxyethyl, phenyl or phenylmethyl.

(14) As depicted in Reaction Scheme 1, the D-amino acid of the racemic amino acid serves as a substrate for the D-amino acid transaminase, the amino group of the D-amino acid is transferred to the amino acceptor by the action of the D-amino acid transaminase, and as a result, the α-keto acid can be generated as an intermediate. The intermediate α-keto acid receives the amino group from the amino donor by the action of the (S)-selective ω-transaminase to produce the desired L-amino acid as a final product. That is, deracemization of the racemic amino acid by the action of the D-amino acid transaminase and the (S)-selective ω-transaminase gives the pure L-amino acid.

(15) Specifically, the present invention also provides a method for producing a D-amino acid, including 1) providing a racemic amino acid, an amino acceptor, and an amino donor as substrates to an L-amino acid transaminase and an (R)-selective ω-transaminase as enzymes, 2) reacting the racemic amino acid substrate and the amino acceptor substrate provided in step 1) with the L-amino acid transaminase to generate an α-keto acid, and 3) reacting the α-keto acid with the (R)-selective ω-transaminase to produce an optically active D-amino acid.

(16) According to the method of the present invention, the D-amino acid may be produced via Reaction Scheme 2;

(17) ##STR00002##

(18) wherein each R.sub.1 is independently selected from the group consisting of methyl, ethyl, propyl, butyl, hydroxymethyl, hydroxyethyl and combinations thereof, and each R.sub.2 may be 2-carboxyethyl, phenyl or phenylmethyl.

(19) As depicted in Reaction Scheme 2, the L-amino acid of the racemic amino acid serves as a substrate for the L-amino acid transaminase, the amino group of the L-amino acid is transferred to the amino acceptor by the action of the L-amino acid transaminase, and as a result, the α-keto acid can be generated as an intermediate. The intermediate α-keto acid receives the amino group from the amino donor by the action of the (R)-selective ω-transaminase to produce the desired D-amino acid as a final product. That is, the coupled enzyme reactions of the L-amino acid transaminase and the (R)-selective ω-transaminase give the pure D-amino acid.

(20) As depicted in Reaction Scheme 1 or 2, the α-keto acid is not provided as a substrate and is generated as an intermediate. The D-amino acid transaminase is selective for the D-amino acid of the racemic amino acid or the L-amino acid transaminase is selective for the L-amino acid of the racemic amino acid. Due to this selectivity, the amino group of the selected amino acid is transferred to the amino acceptor to generate the α-keto acid.

(21) In Reaction Scheme 1, the L-amino acid of the racemic amino acid does not participate in the reaction and only the D-amino acid of the racemic amino acid is isomerized to the L-amino acid as its enantiomer. As a result of this enantiomeric isomerization, the L-amino acid is optically resolved from the racemic amino acid. Likewise, in Reaction Scheme 2, the D-amino acid of the racemic amino acid does not participate in the reaction and only the L-amino acid of the racemic amino acid is isomerized to the D-amino acid as an enantiomer. As a result of this enantiomeric isomerization, the D-amino acid is optically resolved from the racemic amino acid. The enantiomeric excess (ee) values of the optically active amino acids resolved from the racemic amino acids by the methods of the present invention may be 99% or greater.

(22) As depicted in Reaction Scheme 1, the reaction product of the amino acceptor and the amino group of the amino acid, together with the product L-amino acid, remains as a final product without participating in the subsequent reaction. The reaction product may be D-glutamate, D-phenylglycine or D-phenylalanine. The amino donor loses its amino group by the action of the ω-transaminase to generate the ketone, which may also remain as a final product.

(23) As depicted in Reaction Scheme 2, the reaction product of the amino acceptor and the amino group of the amino acid, together with the product D-amino acid, remains as a final product without participating in the subsequent reaction. The reaction product may be L-glutamate, L-phenylglycine or L-phenylalanine. The amino donor loses its amino group by the action of the ω-transaminase to generate the ketone, which may also remain as a final product.

(24) The optically active amino acids as the final products produced by the methods of the present invention are preferably selected from the group consisting of optically active D- and L-configurations of alanine, homoalanine, serine, homoserine, norvaline, norleucine, and leucine.

(25) The racemic amino acid as one of the substrates provided in step 1) is not particularly limited so long as it possesses activity toward the D- or L-amino acid transaminase and the (S)- or (R)-selective ω-transaminase. The racemic amino acid is preferably selected from the group consisting of alanine, serine, homoserine, norvaline, norleucine, leucine, and mixtures thereof.

(26) The amino acceptor as one of the substrates provided in step 1) should possess activity toward the D- or L-amino acid transaminase but be inactive toward the (S)- or (R)-selective ω-transaminase. α-ketoglutarate, phenylglyoxylate, and phenylpyruvate that have no activity toward the ω-transaminase may be provided as preferable amino acceptors for the reactions in the methods of the present invention.

(27) The amino donor as one of the substrates provided in step 1) should be inactive toward the D- or L-amino acid transaminase but should provide an amino group while being active toward the (S)- or (R)-selective ω-transaminase. The amino donor is preferably selected from the group consisting of isopropylamine, methylbenzylamine, benzylamine, and mixtures thereof.

(28) Any D-amino acid transaminase may be used without particular limitation so long as it actively reacts with the amino acceptor while enabling deracemization of the racemic amino acid. An enzyme derived from Bacillus sphaericus and encoded by a base sequence set forth in SEQ ID NO: 5 is preferably used as the D-amino acid transaminase.

(29) Any L-amino acid transaminase may be used without particular limitation so long as it actively reacts with the amino acceptor while enabling deracemization of the racemic amino acid. An enzyme derived from Escherichia coli and encoded by a base sequence set forth in SEQ ID NO: 6 is preferably used as the L-amino acid transaminase.

(30) The (S)-selective ω-transaminase is not particularly limited so long as it can react with the α-keto acid to produce the L-amino acid. An enzyme derived from Ochrobactrum anthropi and encoded by a base sequence set forth in SEQ ID NO: 1 or an enzyme derived from Paracoccus denitrificans and encoded by a base sequence set forth in SEQ ID NO: 2 is preferably used as the (S)-selective ω-transaminase.

(31) The (R)-selective ω-transaminase is not particularly limited so long as it can react with the α-keto acid to produce the D-amino acid. An enzyme derived from Aspergillus terreus and encoded by a base sequence set forth in SEQ ID NO: 3 or an enzyme derived from Arthrobacter sp. and encoded by a base sequence set forth in SEQ ID NO: 4 is preferably used as the (R)-selective ω-transaminase.

(32) Specifically, the present invention also provides a method for producing an optically active amine, including 1) providing a racemic amine and L-threonine as substrates to a threonine deaminase and an (S)- or (R)-selective ω-transaminase as enzymes, 2) deaminating the L-threonine by the threonine deaminase to generate an α-keto acid as an amino acceptor, and 3) kinetically resolving the racemic amine provided in step 1) and the α-keto acid generated in step 2) by the (S)- or (R)-selective ω-transaminase to produce an optically active amine.

(33) According to the method of the present invention, the optically active amine may be produced via Reaction Scheme 3:

(34) ##STR00003##

(35) As depicted in Reaction Scheme 3, the racemic amine and the L-threonine are provided as substrates, and the threonine deaminase and the (S)- or (R)-selective ω-transaminase (ω-TA) as enzymes are provided thereto. The threonine deaminase can deaminate the L-threonine to generate 2-oxobutyrate. The 2-oxobutyrate can act as an amino acceptor and can be thus used as an expensive α-keto acid. The 2-oxobutyrate acting as an amino acceptor has considerably improved effects compared to propanal as a conventional amino acceptor.

(36) In one embodiment of the present invention, when the (S)-selective ω-transaminase is provided in step 1), the (R)-amine and L-homoalanine can be produced from the (S)-amine of the racemic amine and the α-keto acid generated in step 2) by the action of the (S)-selective ω-transaminase in step 3). Kinetic resolution via the coupled enzyme reactions of the threonine deaminase and the (S)-selective ω-transaminase leads to production of the (R)-amine and L-homoalanine.

(37) Alternatively, when the (R)-selective ω-transaminase is provided in step 1), the (S)-amine and D-homoalanine can be produced from the (R)-amine of the racemic amine and the α-keto acid generated in step 2) by the action of the (R)-selective ω-transaminase in step 3). Kinetic resolution via the coupled enzyme reactions of the threonine deaminase and the (R)-selective ω-transaminase leads to production of the (S)-amine and D-homoalanine.

(38) Accordingly, when the (S)-selective ω-transaminase is provided, the (R)-amine, the ketone and the L-homoalanine may remain as final products after the reactions. Alternatively, when the (R)-selective ω-transaminase is provided, the (S)-amine, the ketone and the D-homoalanine may remain as final products after the reactions.

(39) The threonine deaminase is not particularly limited but is preferably an enzyme derived from Escherichia coli and encoded by a base sequence set forth in SEQ ID NO: 7.

(40) The (S)-selective ω-transaminase is not particularly limited so long as it can react with the α-keto acid to produce the L-amino acid. An enzyme derived from Ochrobactrum anthropi and encoded by a base sequence set forth in SEQ ID NO: 1 or an enzyme derived from Paracoccus denitrifican and encoded by a base sequence set forth in SEQ ID NO: 2 is preferably used as the (S)-selective ω-transaminase.

(41) The (R)-selective ω-transaminase is not particularly limited so long as it can react with the α-keto acid to produce the D-amino acid. An enzyme derived from Aspergillus terreus and encoded by a base sequence set forth in SEQ ID NO: 3 or an enzyme derived from Arthrobacter sp. and encoded by a base sequence set forth in SEQ ID NO: 4 is preferably used as the (R)-selective ω-transaminase.

(42) The racemic amine is preferably selected from the group consisting of α-methylbenzylamine, 4-fluoro-α-methylbenzylamine, α-ethylbenzylamine, 1-methyl-3-phenylpropylamine, 1-aminoindane, sec-butylamine, cyclopropylethylamine, 2-aminopentane, 2-aminooctane, 1-methoxy-2-propylamine, alaninol, and mixtures thereof.

(43) In one preferred embodiment of the present invention, α-methylbenzylamine 1a and L-threonine 2 may be provided as substrates. In this embodiment, L-homoalanine may be produced via Reaction Scheme 4:

(44) ##STR00004##

(45) As depicted in Reaction Scheme 4, when the threonine deaminase (TD) deaminates the L-threonine 2 to generate 2-oxobutyrate 3, the (S)-selective ω-transaminase (ω-TA) selectively deaminates (S)-α-methylbenzylamine of the racemic α-methylbenzylamines 1a as a substrate and aminates the 2-oxobutyrate 3. As a result of this transamination, acetophenone 5 and L-homoalanine 4 are produced. This indicates that (R)-α-methylbenzylamine (R)-1a, L-homoalanine and acetophenone can be obtained as final products.

(46) According to a modification of Reaction Scheme 4, an (R)-selective ω-transaminase may be used instead of the (S)-selective ω-transaminase (ω-TA). In this modification, the threonine deaminase (TD) deaminates the L-threonine to generate 2-oxobutyrate 3, the (R)-selective ω-transaminase (ω-TA) selectively deaminates the (R)-α-methylbenzylamine of the racemic α-methylbenzylamines 1a as a substrate and aminates the 2-oxobutyrate 3. As a result of this transamination, acetophenone and D-homoalanine are produced. This indicates that (S)-α-methylbenzylamine, D-homoalanine and acetophenone can be obtained as final products.

(47) Another aspect of the present invention provides a ligand, a pharmaceutical drug or a physiologically active substance which includes an L- or R-amino acid or an L- or R-amine with high purity (≧99% ee) produced by the corresponding method. Such ligands, pharmaceutical drugs and physiologically active substances include those known in the art and those produced by all methods known in the art.

EXAMPLES

Example 1: Construction of Recombinant DNA Encoding (S)-Selective ω-Transaminase from Ochrobactrum anthropi

(48) After culturing Ochrobactrum anthropi in an LB broth (peptone 10 g/L, yeast extract 5 g/L, sodium chloride 5 g/L, pH 7) at 37° C. for 12 h, a gene expressing an (S)-selective ω-transaminase was synthesized from a single colony of the culture. The gene was obtained by PCR amplification using DNA primers. The obtained DNA fragments were cloned into DNA pET28a(+) expression vector using Nde1 and Xho1 restriction enzymes and ligase. The primers used are shown in Table 1.

(49) TABLE-US-00001 TABLE 1 Primer Base sequence Forward primer 5′-GATATACCATGGNNACTGCTCAG CCAAACTCT-3′ (SEQ ID NO: 14) Reverse primer 5′-CGAGTGCGGCCGTCCTGGTGAGG CTTGC-3′ (SEQ ID NO: 15)

Example 2: Construction of Recombinant DNA Encoding (S)-Selective ω-Transaminase from Paracoccus denitrificans

(50) After culturing Paracoccus denitrificans in an LB broth (peptone 10 g/L, yeast extract 5 g/L, sodium chloride 5 g/L, pH 7) at 37° C. for 12 h, a gene expressing an (S)-selective ω-transaminase was synthesized from a single colony of the culture. The gene was obtained by PCR amplification using DNA primers. The obtained DNA fragments were cloned into DNA pET28a(+) expression vector using Nde1 and Xho1 restriction enzymes and ligase. The primers used are shown in Table 2.

(51) TABLE-US-00002 TABLE 2 Primer Base sequence Forward primer 5′-GATATACATATGAACCAACCGCA AAGC-3′ (SEQ ID NO: 16) Reverse primer 5′-GTGGTGCTCGAGGGCCACCTCGG CAAA-3′ (SEQ ID NO: 17)

Example 3: Construction of Recombinant DNA Consisting of DNA Encoding (R)-Selective ω-Transaminase from Aspergillus terreus and Vector DNA

(52) After an (R)-selective ω-transaminase gene sequence (NCBI gene ID: 115385557) derived from Aspergillus terreus was synthesized in pGEM-T vector, a gene expressing the (R)-selective ω-transaminase was synthesized from a plasmid. The gene was obtained by PCR amplification using DNA primers. The obtained DNA fragments were cloned into DNA pET28a(+) expression vector using Nco1 and Xho1 restriction enzymes. The primers used are shown in Table 3.

(53) TABLE-US-00003 TABLE 3 Primer Base sequence Forward primer 5′-AGAAGGAGATATACCATGGCCTC CATGGACAAAGTCT-3′ (SEQ ID NO: 18) Reverse primer 5′-GGTGGTGGTGGTGCTCGAGGTTC CTCTCGTTATAATC-3′ (SEQ ID NO: 19)

Example 4: Construction of Recombinant DNA Encoding D-Amino Acid Transaminase Derived from Bacillus sphaericus

(54) After a D-amino acid transaminase gene sequence (NCBI gene ID: 849138) derived from Bacillus sphaericus was synthesized in pGEM-T vector, a gene expressing the D-amino acid transaminase was synthesized from a plasmid. The gene was obtained by PCR amplification using synthetic DNA primers. The obtained DNA fragments were cloned into DNA pET28a(+) expression vector using Nco1 and Xho1 restriction enzymes. The primers used are shown in Table 4.

(55) TABLE-US-00004 TABLE 4 Primer Base sequence Forward primer 5′-GATATACCATGGCATACTCATTA TGG-3′ (SEQ ID NO: 20) Reverse primer 5′-GTGGTGCTCGAGGGCATTAATTG AAATTGG-3′ (SEQ ID NO: 21)

Example 5: Construction of Recombinant DNA Encoding L-Amino Acid Transaminase Derived from Escherichia coli

(56) After a branched-chain L-amino acid transaminase gene sequence derived from Escherichia coli was synthesized in pGEM-T vector, a gene expressing the branched-chain L-amino acid transaminase was obtained from a plasmid by PCR amplification using synthetic DNA primers. The obtained DNA fragments were cloned into DNA pET28a(+) expression vector using Nco1 and Xho1 restriction enzymes.

Example 6: Construction of Recombinant DNA Consisting of DNA Encoding Threonine Deaminase from Escherichia coli and Vector DNA

(57) After culturing Escherichia coli in an LB broth (peptone 10 g/L, yeast extract 5 g/L, sodium chloride 5 g/L, pH 7) at 37° C. for 12 h, a gene expressing a threonine deaminase was synthesized from a single colony of the culture. The gene was obtained by PCR amplification using DNA primers. The obtained DNA fragments were cloned into DNA pET28a(+) expression vector using Nde1 and Xho1 restriction enzymes and ligase. The primers used are shown in Table 5.

(58) TABLE-US-00005 TABLE 5 Primer Base sequence Forward primer 5′-GATATACATATGGCTGACTCGCA ACCCCTG-3′ (SEQ ID NO: 22) Reverse primer 5′-GTGGTGCTCGAGACCCGCCAAAA AGAA-3′ (SEQ ID NO: 23)

Example 7: Overexpression and Purification of Enzymes from Transformed Strain

(59) E. coli BL21 (DE3) was transformed with each of the plasmids obtained in Examples 1-6. After culturing DE3 in 300 mL of an LB broth containing kanamycin, IPTG (final concentration 1 mM) at an optical density (OD) of 0.5 was added thereto. Thereafter, the mixture was incubated at 37° C. for 6 h or more and centrifuged at 10000×g and at 4° C. for 20 min. The bacterial cells were suspended in 15 mL of resuspension buffer (50 Mm Tris-HCl, 50 mM calcium chloride, 1 mM β-mercaptoethanol, 0.1 mM PMSF, 20 μM PLP, pH 7). The suspension was sonicated under ice cooling and centrifuged at 17000×g and 4° C. for 30 min. Purification of the supernatant as a crude extract by affinity chromatography afforded a solution of the desired enzyme, i.e. (S)-selective ω-transaminase, (R)-selective ω-transaminase, D-amino acid transaminase, L-amino acid transaminase, or threonine deaminase.

Example 8: Activity and Concentration Determination of the Purified Enzymes

(60) The activity and concentration of each enzyme were determined in 50 mM potassium phosphate at 37° C. and pH 7. 10 min after initiation of the enzyme reaction for activity measurement, 600 μl of acetonitrile was added to 100 μl of the reaction solution to stop the reaction.

(61) One unit of the activity of each (S)- or (R)-selective ω-transaminase purified in Examples 1-3 is defined as the enzyme amount catalyzing the formation of 1 μmole of acetophenone in 1 min at 20 mM pyruvate and 20 mM (S)-α-methylbenzylamine. One unit of the activity of each D- or L-selective transaminase purified in Examples 4-5 is defined as the enzyme amount catalyzing the formation of 1 μmole of pyruvate in 1 min at 20 mM D-alanine and 2-oxobutyrate. One unit of the activity of the threonine deaminase purified in Example 7 is defined as the enzyme amount catalyzing the formation of 1 μmole of 2-oxobutyrate in 1 min at 0.1 mM PLP and 50 mM L-threonine.

(62) The activities of the (S)- or (R)-selective ω-transaminase, the D- or L-amino acid transaminase, and the threonine deaminase were determined by measuring the amounts of acetophenone, pyruvate, and 2-oxobutyrate using HPLC, respectively.

Example 9: Analysis of the Amines and Amino Acids

(63) In this example, the concentrations of the products were analyzed and the enantiomeric excess (ee) values of the products were calculated. To this end, each of the chiral amines and amino acids was derivatized with GITC and Marfey reagents and analyzed by high-performance liquid chromatography (HPLC). The GITC derivatization was performed by the following general procedure. First, a reaction sample of each amine or amino acid (≦0.5 mM) was mixed with a solution of 1.5 mM GITC and 1.5 mM triethylamine in acetonitrile as a derivatization solution in a ratio of 1:1, and the reaction was allowed to proceed at room temperature for at least 35 min. Then, analysis was conducted using C18 column and HPLC under the following conditions: flow rate=1 ml/min, detection=UV at 254 nm. The Marfey reagent derivatization was performed by the following procedure. First, 10 μl of each sample was mixed with 8 μl of 1 M sodium bicarbonate and 40 μl of 1% Marfey reagent. The mixture was allowed to react at 40° C. for 1 h. The reaction mixture was cooled and mixed with 8 μl of 1 M hydrochloric acid. The resulting solution was mixed with 434 μl of a 40% acetonitrile solution. Then, analysis was conducted using C18 column and HPLC under the following conditions: flow rate=1 ml/min, temperature=40° C., detection=UV at 320 nm. The results are shown in Table 6. The enantiomeric excess values were calculated from the concentrations of the (S)-enantiomers (C.sub.s) and (R)-enantiomers (C.sub.R) measured above by the following equation:

(64) $\%{\mathrm{ee}}^R=\frac{\left(C_R-C_S\right)}{\left(C_R+C_S\right)}\times 100$

(65) In Table 6, 1a, 1b, 1c, 1d, 1e, 1f, 1g, 1h, 1i, 1j, and 1k represent α-methylbenzylamine, 4-fluoro-α-methylbenzylamine, α-ethylbenzylamine, 1-methyl-3-phenylpropylamine, 1-aminoindane, sec-butylamine, cyclopropylethylamine, 2-aminopentane, 2-aminooctane, 1-methoxy-2-propylamine, and alaninol, respectively.

(66) TABLE-US-00006 TABLE 6 Reaction time (min) Elution condition.sup.a (R)-form (S)-form GITC derivatization 1a Isocratic elution with A55%, B45% 14.5 16.2 1b Isocratic elution with A55%, B45% 18.7 21.5 1c Isocratic elution with A55%, B45% 24.5 26.4 1d Isocratic elution with A55%, B45% 39.4 42.0 1e Gradient elution from A60%, B40% 14.1 16.6 to A50% 1f Isocratic elution with A50%, B50% 13.3 14.4 1h Isocratic elution with A50%, B50% 27.5 30.6 1i Isocratic elution with A65%, B35% 23.6 25.4 1j Gradient elution from A60%, B40% 24.9 27.0 from A65%, B35% 4 Isocratic elution with A55%, B45% 5.3 4. Alanine Isocratic elution with A50%, B50% 5.6 4.6 Marfey reagent derivatization 1g Isocratic elution with A50%, B50% 26.9 28.5 1k Isocratic elution with A55%, B45% 4.9 3.3 Note - A: methanol (0.1% TFA), B: TDW (0.1% TFA)

Example 10: Substrate Specificities of the Purified D-Amino Acid Transaminase Toward Keto Acids and Conversion of Optically Active Amino Acids by Keto Acids

(67) Each of amino acceptors shown in Tables 7 and 8 and D-alanine or D-homoalanine as substrates were allowed to react with the D-amino acid transaminase purified in Example 7. Specifically, each amino acceptor (20 mM), D-alanine or D-homoalanine (20 mM), and potassium phosphate (50 mM, pH 7) were allowed to react with 0.1 U/ml of the D-amino acid transaminase at 37° C. for 10 min, and the amount of a keto acid (pyruvate or 2-oxobutyrate) generated was measured. The D-amino acid transaminase was shown to have activity toward the amino acceptors, including 2-oxobutyrate, β-hydroxypyruvate, mercaptopyruvate, fluoropyruvate, 3-methyl-2-oxobutyrate, 2-oxopentanoate, 4-methyl-2-oxopentanoate, 2-ketohexanoate, α-ketoglutarate, phenylpyruvate, and pyruvate. The results are shown in Tables 7 and 8.

(68) TABLE-US-00007 TABLE 7 Substrates Products Amino Pyruvate Amino acceptor donor D-amino acid (mM) 2-Oxobutyrate D-Alanine Homoalanine 1.375 β-Hydroxypyruvate D-Alanine Serine 0.079 Mercaptopyruvate D-Alanine Cysteine 0.022 Fluoropyruvate D-Alanine Fluoroalanine 0.053 Bromopyruvate D-Alanine Bromoalanine 0.000 3-Methyl-2-oxobutyrate D-Alanine Valine 0.014 Trimethylpyruvate D-Alanine tert-leucine 0.000 2-Oxopentanoate D-Alanine Norvaline 0.882 4-Methyl-2-oxopentanoate D-Alanine Leucine 0.073 2-Ketohexanoate D-Alanine Norleucine 0.531 α-Ketoglutarate D-Alanine Glutamate 0.971 Phenyl glyoxylate D-Alanine Phenylglycine 0.000 Phenylpyruvate D-Alanine Phenylalanine 0.075 2-(3-Hydroxy-1-adamantyl)- D-Alanine 3-hydroxy- 0.000 oxoethanoate adamantylglycine

(69) TABLE-US-00008 TABLE 8 Substrates Products Amino donor D-amino 2-Oxobutyrate Amino acceptor (D-amino acid) acid (mM) Pyruvate D-Homoalanine Alanine 0.399 2-Oxopentanoate D-Homoalanine Norvaline 0.651

Example 11: Confirmation Whether L-Amino Acid Concentration Deteriorates the Activity of the Purified D-Amino Acid Transaminase

(70) For continuous deracemization of a racemic amino acid with the D-amino acid transaminase purified in Example 7, L-amino acid as a final product should not affect the activity of the D-amino acid transaminase. Deterioration in the activity of the enzyme by the final product makes it difficult to induce continuous reactions. The activities of the D-amino acid transaminase were confirmed at different concentrations (0, 1, 3, 5, 7, 10, 15, and 20 mM) of L-homoalanine in 20 mM D-homoalanine, 20 mM pyruvate, and 50 mM potassium phosphate (pH 7). After the reactions were carried out at 37° C. for 10 min, the amounts of 2-oxobutyrate produced were measured. The results are shown in FIG. 1. FIG. 1 shows that the D-amino acid transaminase did not lose its activity by the final product L-amino acid.

Example 12: Substrate Specificities of the Purified ω-Transaminases Toward Keto Acids and Conversion of Optically Active Amino Acids by Keto Acids

(71) An investigation was made as to the substrate specificities of each of the (S)-selective ω-transaminases derived from O. anthropi and P. denitrificans, which were purified in Example 7, toward keto acids. The substrate specificity of the (S)-selective ω-transaminase was confirmed in each keto acid (20 mM), (S)-α-methylbenzylamine (20 mM) and potassium phosphate (50 mM, pH 7). After the reactions were carried out at 37° C. for 10 min, the amounts of acetophenone produced were measured. The (S)-selective ω-transaminase was shown to have activity toward glyoxylate, pyruvate, 2-oxobutyrate, β-hydroxypyruvate, fluoropyruvate, 2-oxopentanoate, 4-methyl-2oxopentanoate (P. denitrificans), and 2-ketohexanoate. The results are shown in Table 9.

(72) TABLE-US-00009 TABLE 9 Substrates Products Amino acceptor Amino donor Acetophenone (mM) (keto acid) ((S)-amine) L-amino acid O. anthropi P. denitrificans Glyoxylate (S)-α-methylbenzylamine Glycine 1.733 1.076 Pyruvate (S)-α-methylbenzylamine Alanine 1.284 1.216 2-Oxobutyrate (S)-α-methylbenzylamine Homoalanine 0.173 0.821 β-Hydroxypyruvate (S)-α-methylbenzylamine Serine 0.183 0.066 Mercaptopyruvate (S)-α-methylbenzylamine Cysteine 0.012 0.008 Fluoropyruvate (S)-α-methylbenzylamine Fluoroalanine 0.554 0.339 Bromopyruvate (S)-α-methylbenzylamine Bromoalanine 0.010 −0.001 3-Methyl-2-oxobutyrate (S)-α-methylbenzylamine Valine −0.001 −0.002 Trimethylpyruvate (S)-α-methylbenzylamine tert-leucine −0.002 −0.004 2-Oxopentanoate (S)-α-methylbenzylamine Norvaline 0.078 1.032 4-Methyl-2- (S)-α-methylbenzylamine Leucine 0.005 0.175 oxopentanoate 3-Methyl-2- (S)-α-methylbenzylamine Isoleucine 0.001 −0.004 oxopentanoate 2-Ketohexanoate (S)-α-methylbenzylamine Norleucine 0.006 0.713 α-Ketoglutarate (S)-α-methylbenzylamine Glutamate −0.001 −0.003 Phenylglyoxylate (S)-α-methylbenzylamine Phenylglycine −0.005 −0.008 2-(3-hydroxy-1- (S)-α-methylbenzylamine 3-hydroxyadamantyl −0.002 −0.004 adamantyl)-oxoethanoate glycine

(73) The results in Table 9 show that α-ketoglutarate had no activity toward the ω-transaminase. In contrast, the results in Tables 7 and 8 show that α-ketoglutarate had activity toward the D-amino acid transaminase. In conclusion, α-ketoglutarate can be considered as an example of preferred amino acceptors.

Example 13: Substrate Specificities of the Purified ω-Transaminases Toward Amines and Conversion of L-Alanine from Pyruvate

(74) An investigation was made as to the substrate specificities of each of the (S)-selective ω-transaminases derived from O. anthropi and P. denitrificans, which were purified in Example 7, toward amines. The substrate specificity of the (S)-selective ω-transaminase was confirmed in each amine (20 mM), (S)-amine (20 mM, in 40 mM racemic sec-butylamine) and potassium phosphate (50 mM, pH 7). After the reactions were carried out at 37° C. for 10 min, the amounts of L-alanine produced were measured. The (S)-selective ω-transaminase was shown to have activity toward (S)-α-methylbenzylamine, isopropylamine, and sec-butylamine. The results are shown in Table 10.

(75) TABLE-US-00010 TABLE 10 Substrates Products Amino donor Amino acceptor L-alanine (mM) (Amine) (Keto acid) Ketone O. anthropi P. denitrificans (S)-α-methylbenzylamine Pyruvate Acetophenone 0.715 ± 0.082 0.451 ± 0.077 sec-butylamine Pyruvate 2-Butanone 0.164 ± 0.019 0.000 ± 0.000 Isopropylamine Pyruvate Acetone 0.309 ± 0.050 0.033 ± 0.002

EXPERIMENTAL EXAMPLES

Experimental Example 1: Production of High Concentration of L-Amino Acid from Keto Acid by Using the Purified ω-Transaminase and Isopropylamine as Amino Donor

(76) An experiment was carried out to confirm whether the use of the ω-transaminase purified in Example 7 led to the production of a high concentration of an L-amino acid from a keto acid. Specifically, after 1 U/ml ω-transaminase was allowed to react in 100 mM 2-oxobutyrate as a keto acid, 100 mM isopropylamine as an amino donor, and 50 mM potassium phosphate (pH 7) at 37° C., the amounts of L-homoalanine produced were measured as a function of time. The reaction of the purified ω-transaminase and the use of the amino donor were confirmed to lead to the production of a high concentration of L-homoalanine from the keto acid with ≧80% conversion. The results are shown in FIG. 2.

Experimental Example 2: Conversion of High Concentrations of Racemic Amino Acids to L-Amino Acids by Coupling Reactions of the Purified D-Amino Acid Transaminase and (S)-Selective ω-Transaminase

(77) An experiment was carried out to confirm whether the use of the D-amino acid transaminase and (S)-selective ω-transaminase purified in Example 7 led to the production of L-amino acids from high concentrations of racemic amino acids. Specifically, coupling reactions of the D-amino acid transaminase (4 U/ml) and the (S)-selective ω-transaminase (20 U/ml) were allowed to proceed in each racemic amino acid (100 mM), α-ketoglutarate, phenylglyoxylate or phenylpyruvate (70 mM) as an amino acceptor, isopropylamine (70 mM) as an amino donor, PLP (0.1 mM), and potassium phosphate (50 mM, pH 7). In the case of leucine and norleucine, each racemic amino acid (50 mM), α-ketoglutarate, phenyl glyoxylate or phenylpyruvate (35 mM), and isopropylamine (35 mM) as substrates were used for the reaction. In the case of deracemization for alanine production, the amounts of the substrates and products were measured as a function of reaction time. In the case of homoalanine, serine, homoserine, norvaline, leucine, and norleucine, their D-amino acids and L-amino acids were analyzed. L-amino acids with ≧99% enantiomeric excess (ee) were produced from 100 mM racemic amino acids (each 50 mM for leucine and norleucine) via deracemization. The results are shown in FIG. 3, FIG. 4 and Table 11.

(78) TABLE-US-00011 TABLE 11 Final concentration Amino acid of L-amino acid (mM) ee (%) Reaction time (h) Alanine 100.15 99.34 10 Homoalanine 99.24 99.94 6 Serine 101.84 98.92 25 Norvaline 103.86 99.16 25 Norleucine 55.29 99.30 20 Leucine 49.66 99.81 10 Homoserine 98.51 98.87 25

Experimental Example 3: Conversion of High Concentrations of Racemic Amino Acids to D-Amino Acids by Coupling Reactions of the Purified Branched-Chain L-Amino Acid Transaminase and (R)-Selective ω-Transaminase

(79) An experiment was carried out to confirm whether the use of the branched-chain L-amino acid transaminase and (R)-selective ω-transaminase purified in Example 7 led to the production of D-amino acids from high concentrations of racemic amino acids. Specifically, coupling reactions of the branched-chain L-amino acid transaminase (4 U/ml) and the (S)-selective ω-transaminase (20 U/ml) were allowed to proceed in each racemic amino acid (100 mM), α-ketoglutarate (70 mM), isopropylamine (70 mM), PLP (0.1 mM), and potassium phosphate (50 mM, pH 7). In the case of norleucine, the racemic amino acid (50 mM), α-ketoglutarate (35 mM), and isopropylamine (35 mM) as substrates were used for the reaction. In the case of alanine, homoalanine, serine, homoserine, norvaline, leucine, and norleucine, their D-amino acids and L-amino acids were analyzed. L-amino acids with ≧99% enantiomeric excess (ee) were produced from 100 mM racemic amino acids (50 mM for norleucine) via deracemization. The results are shown in Table 12.

(80) TABLE-US-00012 TABLE 12 Final concentration Amino acid of L-amino acid (mM) ee (%) Reaction time (h) Alanine 102.88 99.10 20 Homoalanine 99.54 99.82 15 Serine 100.71 99.01 20 Norvaline 98.42 99.01 20 Norleucine 48.58 99.51 20 Homoserine 97.78 98.89 25

Experimental Example 4: Comparison of Activities of Pyruvate and Propanal as Amino Acceptors for Kinetic Resolution of Optically Active Chiral Amines Using the Purified ω-Transaminase

(81) Racemic α-methylbenzylamine and pyruvate or propanal as substrates were allowed to react using the (S)-selective ω-transaminase purified in Example 7. After 5 U/ml of the (S)-selective ω-transaminase derived from O. anthropi was added to 100 mM α-methylbenzylamine, 80 mM pyruvate or propanal, 0.1 mM PLP, and 50 mM phosphate buffer (pH 7.5) at 37° C., the amounts of the (S)-α-methylbenzylamine and (R)-α-methylbenzylamine produced were measured with time and the enantiomeric excess values thereof were calculated. The enatiomeric excess (ee) of the (R)-α-methylbenzylamine reached ≧99% with pyruvate at 3 h, whereas that of the (R)-α-methylbenzylamine was only 57% with propanal even after the reaction for 24 h. The results are shown in FIG. 5.

Experimental Example 5: Investigation of Substrate Specificities of the Purified ω-Transaminase Toward Amines

(82) An investigation was made as to the substrate specificities of the (S)-selective ω-transaminase derived from O. anthropi, which was purified in Example 7, toward 11 amines: α-methylbenzylamine 1a, 4-fluoro-α-methylbenzylamine 1b, α-ethylbenzylamine 1c, 1-methyl-3-phenylpropylamine 1d, 1-aminoindane 1e, sec-butylamine 1f, cyclopropylethylamine 1g, 2-aminopentane 1h, 2-aminooctane 1i, 1-methoxy-2-propylamine 1j, and alaninol 1k. 0.125 U/ml of the (S)-selective transaminase derived from O. anthropi was added to each racemic amine (10 mM), pyruvate (10 mM), PLP (0.1 mM), and phosphate buffer (50 mM, pH 7.5) at 37° C. After the mixture was allowed to react for 10 min, the amount of L-alanine produced was measured via GITC derivatization. All activities were measured in the initial reaction rate range of ≦5% conversion. The initial reaction rate for α-methylbenzylamine was measured to be 0.033 mM/min. When the activity of the ω-transaminase toward α-methylbenzylamine was defined as 100%, the activities of the ω-transaminase toward 4-fluoro-α-methylbenzylamine 1b, α-ethylbenzylamine 1c, 1-methyl-3-phenylpropylamine 1d, and 1-aminoindane 1e as arylalkylamines were 91%, 18%, 70%, and 64%, respectively. The activity of the ω-transaminase toward (S)-1-aminoindane was 153% of that toward (S)-α-methylbenzylamine. The activities of the ω-transaminase toward sec-butylamine 1f, cyclopropylethylamine 1g, 2-aminopentane 1h, 2-aminooctane 1i, 1-methoxy-2-propylamine 1j, and alaninol 1k as alkylamines were 54%, 63%, 29%, 146%, 60%, and 63%, respectively. The results are shown in Table 13.

(83) TABLE-US-00013 TABLE 13 Substrate R.sup.1 R.sup.2 Relative reactivity (%) 1a C.sub.6H.sub.5 CH.sub.3 100.sup.c 1b p-F—C.sub.6H.sub.4 CH.sub.3  91 1c C.sub.6H.sub.5 CH.sub.2CH.sub.3  18 1d C.sub.6H.sub.5(CH.sub.2).sub.2 CH.sub.3  70 1e 64 (153).sup.d 1f CH.sub.3CH.sub.2 CH.sub.3  54 1g Cyclopropyl CH.sub.3  63 1h CH.sub.3(CH.sub.2).sub.2 CH.sub.3  29 1i CH.sub.3(CH.sub.2).sub.5 CH.sub.3 146 1j CH.sub.3OCH.sub.2 CH.sub.2  60 1k HOCH.sub.2 CH.sub.3  63

Experimental Example 6: Yields of Optically Active Chiral Amines According to Changes in the Concentration of Threonine Deaminase and ω-Transaminase when Kinetic Resolution was Performed Via Coupled Enzyme Reactions of the Enzymes

(84) Coupled enzyme reactions of the threonine deaminase and ω-transaminase purified in Example 7 were carried out at varying enzyme concentrations shown in Table 14. The reaction conditions were as follows: 100 mM racemic methylbenzylamine, 60 mM L-threonine, 0.1 mM PLP, and 50 mM phosphate buffer (pH 7.5) at 37° C. for 2.5 h. Increasing ω-transaminase concentrations (1.25 U/ml, 2.50 U/ml, and 3.75 U/ml) at a constant threonine deaminase (9 U/ml) led to an increase in the enantiomeric excess of (R)-α-methylbenzylamine (28%, 65%, and 90%). In contrast, increasing threonine deaminase concentrations (9 U/ml, 18 U/ml, and 27 U/ml) at a constant ω-transaminase concentration (1.25 U/ml) did not lead to an improvement in the enantiomeric excess of (R)-α-methylbenzylamine (28%, 24%, and 26%). This result indicates that the ω-transaminase reaction is a rate-determining step of the coupled enzyme reactions. Therefore, a higher ω-transaminase concentration is preferred for a higher enantiomeric excess of (R)-α-methylbenzylamine. The results are shown in Table 14.

(85) TABLE-US-00014 TABLE 14 ω-TA (U mL.sup.−1) TD (U mL.sup.−1) 4.sup.a (mM) ee (%) of (R)-1a.sup.b 1.25 9 15 28 2.50 9 33 65 3.75 9 47 90 1.25 18 19 24 1.25 27 13 26 Note: ω-TA: ω-transaminase, TD: threonine deaminase, .sup.aconcentration determined by a standard calibration method from HPLC analysis of the GITC derivative, .sup.bdetermined by HPLC analysis of the GITC derivatives

Experimental Example 7: Yields of (R)-Amines and Production of L-Homoalanine Via Coupled Enzyme Reactions of (S)-Selective ω-Transaminase and Threonine Deaminase Using L-Threonine and Racemic Amines as Substrates

(86) Referring to the enzyme proportions yielding the best outcome in Experimental Example 6, kinetic resolution of 11 amines was performed and L-homoalanine was produced via coupled enzyme reactions of the ω-transaminase and threonine deaminase purified in Example 7. The amines were α-methylbenzylamine 1a, 4-fluoro-α-methylbenzylamine 1b, α-ethylbenzylamine 1c, 1-methyl-3-phenylpropylamine 1d, 1-aminoindane 1e, sec-butylamine 1f, cyclopropylethylamine 1g, 2-aminopentane 1h, 2-aminooctane 1i, 1-methoxy-2-propylamine 1j, and alaninol 1k. After addition of the (S)-selective ω-transaminase (3.75 U/ml) and the threonine deaminase (9 U/ml) to each of the amines 1a-1k (100 mM), L-threonine (60 mM), PLP (0.1 mM), and phosphate buffer (50 mM, pH 7.5), the concentration and enantiomeric excess of each amine were measured. All the amines except α-ethylbenzylamine 1c, 2-aminooctane 1i, and alaninol 1k had enantiomeric excess values ≧99% within a reaction time of 10 h. For kinetic resolution of α-ethylbenzylamine 1c, 2-aminooctane 1i, and alaninol 1k, the ω-transaminase concentration, which is a rate determining step of the coupled enzyme reactions, was increased to 10 U/ml. As a result, enantiomeric excess values of α-ethylbenzylamine 1c and 2-aminooctane 1i higher than 99% were attained, and an enantiomeric excess of 66% was attained for alaninol 1k within 30 h. In all reactions, the production of L-homoalanine with 99% enantiomeric excess was confirmed. The results are shown in Table 15.

(87) TABLE-US-00015 TABLE 15 Chiral amine Reaction time (h) Conversion (%) ee of (R)-amine (%) 1a 3 49.9 >99 1b 3.5 50.5 >99 1c 28 35.1 52 1c 30 50.1 >99 1d 5 50.9 >99 1e 7 50.0 >99 1f 8 50.2 >99 1g 8 50.1 >99 1h 9 50.5 >99 1i 24 37.4 58 1i 24 50.2 >99 1j 9 50.6 >99 1k 30 40.8 66

Experimental Example 8: Yields of (S)-Amines and Production of D-Homoalanine Via Coupled Enzyme Reactions of (R)-Selective ω-Transaminase and Threonine Deaminase Using L-Threonine and Racemic Amines as Substrates

(88) Kinetic resolution of amines was performed and D-homoalanine was produced via coupled enzyme reactions of the (R)-selective ω-transaminase and threonine deaminase purified in Example 7. After addition of the (R)-selective ω-transaminase (1.5 U/ml) and the threonine deaminase (4.5 U/ml) to α-methylbenzylamine (30 mM), L-threonine (20 mM), PLP (0.1 mM), and phosphate buffer (50 mM, pH 7.5), the concentrations of (S)-α-methylbenzylamine and (R)-α-methylbenzylamine were measured as a function of reaction time. An enantiomeric excess ≧99% was obtained within a reaction time of 5 h. Likewise, D-homoalanine was produced with ≧99% enantiomeric excess. The results are shown in FIG. 6.

(89) According to the methods of the present invention, α-keto acids are generated as reaction intermediates, eliminating the need to use expensive α-keto acids as starting materials. In addition, optically active amine compounds produced by the methods of the present invention can be widely used in various industrial applications, including pharmaceutical and agrochemical industries that require chiral compounds, due to their high enantiomeric excess (≧99%).