Enzymatic synthesis of N-methyl-L-phenylalanine by a novel enzyme, N-methyl-L-amino acid dehydrogenase, from Pseudomonas putida

Article abstract of DOI:10.1016/j.tetasy.2004.06.030

An enzymatic system for the synthesis of N-methyl-l-phenylalanine from phenylpyruvic acid and methylamine with a novel enzyme, N-methyl-l-amino acid dehydrogenase from Pseudomonas putida ATCC12633, using NADP⁺ and glucose dehydrogenase from Bacillus subtilis as a co-factor-recycling system is described. Analysis of the product on a laboratory preparative scale process revealed N-methyl-l-phenylalanine in 98% yield and over 99% ee. N-Methyl-l-phenylalanine can be used as chiral building blocks for the synthesis of several products with pharmacological activity.

Full text of DOI:10.1016/j.tetasy.2004.06.030

Tetrahedron:
Asymmetry
Tetrahedron: Asymmetry 15 (2004) 2841–2843
Enzymatic synthesis of N-methyl-L-phenylalanine by a
novel enzyme, N-methyl-L-amino acid dehydrogenase, from
Pseudomonas putida
Hisashi Muramatsu,a Hisaaki Mihara,a Ryo Kakutani,a Mari Yasuda,b Makoto Ueda,b
Tatsuo Kuriharaa and Nobuyoshi Esakia,*
aInstitute for Chemical Research, Kyoto University, Uji, Kyoto 611-0011, Japan
bScience and Technology Research Center, Mitsubishi Chemical Group, 1000 Kamoshida-cho, Aoba-ku, Yokohama 227-8502, Japan
Received 30 April 2004; accepted 18 June 2004
Available online 30 July 2004
Abstract—An enzymatic system for the synthesis of N-methyl-L-phenylalanine from phenylpyruvic acid and methylamine with a
novel enzyme, N-methyl-L-amino acid dehydrogenase from Pseudomonas putida ATCC12633, using NADP+ and glucose dehydro-
genase from Bacillus subtilis as a co-factor-recycling system is described. Analysis of the product on a laboratory preparative scale
process revealed N-methyl-L-phenylalanine in 98% yield and over 99% ee. N-Methyl-L-phenylalanine can be used as chiral building
blocks for the synthesis of several products with pharmacological activity.
ꢀ 2004 Elsevier Ltd. All rights reserved.
1. Introduction
have been reported, none have been adapted for the
modification of amino acids with other N-alkyl groups.
The ability to synthesize a variety of enantiomerically
pure N-alkyl-a-amino acids from a common intermedi-
ate using enzymatic catalysis would be useful in the
development of pharmaceutical agents.
Optically active N-methyl-a-amino acids occur as func-
tionally critical components of several biologically ac-
tive natural peptides, such as dolastatins1 and
didemnins.2 These peptides have been studied because
they inhibit certain aspects of viral proliferation, tumor
cell growth, and proliferative immune responses.1,2
Therefore, enantiomerically pure N-methyl-a-amino
acids are useful building blocks in enantiospecific syn-
theses of medicines and pesticides. Several methods for
the preparation of optically active N-methyl amino acids
have already been reported.3–8 However, each procedure
has its drawbacks, which include an excessive number of
steps needed to prepare the substrate,5 problems of race-
mization,6 harsh reaction conditions,7 and the instability
of acid-sensitive substrates.8 Thus, there is no satisfac-
tory method available for industrial production of enan-
tiomerically pure N-methyl amino acids under mild
conditions. Although bacteria producing N-methyl-L-
phenylalanine9 and enzymes, which synthesize N-
methyl-L-alanine10 and N-methyl-L-glutamic acid,11
Recently, we identified a gene encoding a novel N-
methyl-L-amino acid dehydrogenase (NMAADH) in
Pseudomonas putida ATCC12633 and succeeded in the
functional expression of the enzyme in host Escherichia
coli BL21(DE3) cells.12 The enzyme converts an a-keto
acid and methylamine into the corresponding N-
methyl-L-amino acid in an NADPH-dependent manner
(Scheme 1). NADPH is a very expensive reagent how-
ever and so the economic viability of this reaction re-
quires limiting the use of this co-enzyme to catalytic
amounts. Thus, a convenient method to recycle the co-
enzyme is of fundamental importance.13,14 In anticipa-
tion of possible scale-up, we decided to demonstrate
the feasibility of a NADPH regeneration system in an
enzyme-coupled system employing glucose dehydroge-
nase (GDH) from Bacillus subtilis15 (Scheme 2). Herein,
we report the use of NMAADH in the synthesis of opti-
cally active N-methyl-L-phenylalanine from phenylpyru-
vic acid and methylamine.
*
Corresponding author. Tel.: +81-774-38-3240; fax: +81-774-38-
3248; e-mail: esaki@scl.kyoto-u.ac.jp
0957-4166/$ - see front matter ꢀ 2004 Elsevier Ltd. All rights reserved.
doi:10.1016/j.tetasy.2004.06.030

2842
H. Muramatsu et al. / Tetrahedron: Asymmetry 15 (2004) 2841–2843
COOH
COOH
R
R
H+
NADP+
H2O
+
+
+
+
+
CH3NH2
NADPH
NH
O
Scheme 1. Reaction catalyzed by N-methyl-L-amino acid dehydrogenase.
R
COOH
NH
Glucose
NADP+
NADPH
NMAADH
GDH
Gluconolactone
R
COOH
+
CH3NH2
O
Scheme 2. The NMAADH-catalyzed production of N-methyl-L-amino acid in combination with co-factor-recycling catalyzed by GDH.
2. Results and discussion
MA) with an ULTRON ES-PhCD column
(6.0·150mm, Shinwa Chemical Industries, Kyoto,
Japan) and 20mM KH2PO4–H3PO4–CH3CN (2000:1:
500, v/v) as eluent at a flow rate of 0.85mL/min at
40ꢁC, monitored at 210nm. The concentration of
glucose was determined by a glucose assay kit (Glucose
CII-Test Wako; Wako Pure Chemical Industries,
Osaka, Japan).
Recombinant P. putida NMAADH can be efficiently
produced in host E. coli BL21(DE3) cells induced with
isopropyl-b-D-thiogalactopyranoside (IPTG).12 The
purified NMAADH converted the following a-keto
acids into their corresponding N-methyl-L-amino acids
in the presence of methylamine and NADPH, as deter-
mined by photometric assay monitoring the oxidation
of NADPH at 340nm using a Shimadzu UV2450 spec-
trophotometer (relative activity in parentheses): pyruvic
acid (100%) > a-ketohexanoic acid (52%) > phenylpyru-
vic acid (30%) > a-ketobutyric acid (30%) > fluoropyru-
vic acid (27%) > a-ketovaleric acid (16%) > a-
ketoisocaproic acid (9.0%). The broad substrate specific-
ity of the enzyme indicates that the enzyme would be
useful in the production of a wide variety of N-methyl-
L-amino acids. Furthermore, the enzyme was able to
catalyze reactions using ethylamine, isopropylamine, di-
ethylamine, cyclopropylamine, and n-propylamine,
although the activities for these amines were less than
2% of that for methylamine. Therefore, the enzyme could
be used for the synthesis of various N-alkyl-L-amino
acids. We have confined our bioconversion studies
here to the production of N-methyl-L-phenylalanine,
because no enzyme affecting this catalysis has yet been
reported.
Initially, we attempted to synthesize N-methyl-L-phenyl-
alanine in a 10mL standard reaction mixture at 30ꢁC
with reciprocal shaking. In order to prevent a decrease
in pH during the reaction, the reaction mixture was ad-
justed to pH8.0–9.0 with 10M NaOH. After a 24h reac-
tion, 75mM (13g/L) of N-methyl-L-phenylalanine was
produced, with a yield of 75%. Phenylpyruvic acid and
glucose were almost completely depleted after 10h
(Fig. 1A).
Next, a large-scale production was performed in a
100mL standard reaction mixture at pH8.0–9.0 at
30ꢁC with stirring. After a 7h reaction, 91mM (16g/
L) of N-methyl-L-phenylalanine was produced, with a
yield of 98%. Phenylpyruvic acid and glucose were
added again at 7h, but the rate of the formation of N-
methyl-L-phenylalanine had significantly decreased after
this point, largely due to the inactivation of GDH dur-
ing the reaction: GDH activity was completely lost after
15h, whereas NMAADH was still active. After 22h of
reaction, the overall yield of N-methyl-L-phenylalanine
was 54% (Fig. 1B).
Recombinant E. coli BL21(DE3) cells harboring the
genes encoding NMAADH and GDH were grown in a
Luria–Bertani medium containing 100lg/mL ampicillin
and 25lg/mL chloramphenicol at 37ꢁC for 16h. Three
hours after the induction of gene expression with
1mM IPTG at 37ꢁC for 3h, the cells were harvested
by centrifugation, re-suspended in 20mM Tris–HCl
(pH7.0) and washed twice with the same buffer. The
resulting cells were used directly for enzymatic synthesis.
The standard reaction mixture for the N-methyl-L-
phenylalanine synthesis contained 700mM methyl-
amine–HCl, 100mM phenylpyruvic acid, 100mM glu-
cose, 0.2mM NADP+, and washed cells (OD660 =20)
at pH9.0. Aliquots of the reaction mixture were
removed for analysis of the product and the substrate.
Both N-methyl-L-phenylalanine and phenylpyruvic acid
were detected by high-performance liquid chromato-
graphy (HPLC) (Waters Alliance; Waters, Milford,
The product was isolated from a 24h reaction mixture
(100mL) prepared with a protocol identical to that em-
ployed for the reaction shown in Figure 1A, as follows.
Cells were removed by centrifugation, and the superna-
tant adjusted to pH10.0 with 10M NaOH and concen-
trated about 2.5-fold with an evaporator. The solution
was adjusted to pH7.0 with 12M HCl and left overnight
at 4ꢁC. The solids formed were collected by filtration
and dissolved in 50mL 0.7M NaOH. The solution was
then adjusted to pH7.0 with 12M HCl. Crystals, which
had formed by addition of acetone, were collected,
washed with 20mL acetone, and dried under vacuum
1
to obtain 1.1g of N-methyl-L-phenylalanine. H NMR
(400MHz, D2O) spectrum of the product showed the

H. Muramatsu et al. / Tetrahedron: Asymmetry 15 (2004) 2841–2843
2843
120
100
80
60
40
20
0
120
100
80
60
40
20
0
0
4
8
12
16
20
24
0
4
8
12
16
20
24
28
(A)
(B)
Time (h)
Time (h)
Figure 1. Time course of the small-scale (A) and large-scale (B) production of N-methyl-L-phenylalanine by N-methyl-L-amino acid dehydrogenase
(NMAADH) from P. putida ATCC12633. The reactions were carried out at 30ꢁC at pH9.0 in 700mM methylamine–HCl, 100mM phenylpyruvic
acid, 100mM glucose, 0.2mM NADP+ and the E. coli cells (OD660 =20) expressing NMAADH and glucose dehydrogenase. N-Methyl-L-
phenylalanine (s) and phenylpyruvic acid (d) were determined by HPLC. Glucose (m) was determined with a glucose assay kit.
following signals: d=2.55 (s, 3H); 3.09 (d, 2H,
J=6.3Hz); 3.72 (t, 1H, J=6.3Hz); 7.14–7.21 (m, 2H);
7.21–7.33 (m, 3H), which agree with those of authentic
N-methyl-L-phenylalanine.
Ministry of Education, Culture, Sports, Science, and
Technology, Japan (21st Century COE on Kyoto Uni-
versity Alliance for Chemistry). We thank Mr. Hisatoshi
Uehara of the Science and Technology Research Center,
Mitsubishi Chemical Group, for the NMR analysis.
The enantiomeric excess (ee) of the product obtained by
a 7h reaction in the large-scale production, was deter-
mined by HPLC with a CHIRALPAK WE column
(4.6·250mm, Daicel Chemical Industries, Tokyo,
Japan) and 2mM CuSO4 as eluent at a flow rate of
0.75mL/min at 50ꢁC, monitored at 254nm. The reten-
tion time of N-methyl-L-phenylalanine was 21.5min,
while that of N-methyl-D-phenylalanine was 41.2min.
N-Methyl-L-phenylalanine was obtained with a satisfac-
torily high enantiomeric excess (>99% ee).
References
1. Rinehart, K. L., Jr.; Gloer, J. B.; Cook, J. C., Jr. J. Am.
Chem. Soc. 1981, 103, 1857–1859.
2. Pettit, G. R.; Kamano, Y.; Herald, C. L.; Fujii, Y.; Kizu,
H.; Boyd, R. R.; Boettner, F. E.; Doubek, D. L.; Schmidt,
J. M.; Chapuis, J.-C.; Michel, C. Tetrahedron 1993, 49,
9151–9170.
3. Dorow, R. L.; Gingrich, D. E. J. Org. Chem. 1995, 60,
4986–4987.
4. Reddy, G. V.; Rao, G. V.; Iyengar, D. S. Tetrahedron
Lett. 1998, 39, 1985–1986.
3. Conclusion
5. Quitt, P.; Hellerbach, J.; Vogler, K. Helv. Chim. Acta
1963, 46, 327.
6. Effenberger, F.; Burkard, U.; Willfahrt, J. Liebigs Ann.
Chem., 1986, 314.
7. McDermott, J. R.; Benoiton, N. L. Can. J. Chem. 1973,
51, 1915.
8. Freidinger, R. M.; Hinkle, J. S.; Perlow, D. S.; Arison, B.
H. J. Org. Chem. 1983, 48, 77.
In conclusion, we have demonstrated the coupled enzy-
matic synthesis of N-methyl-L-phenylalanine with good
yield and high ee by using NMAADH from P. putida
and GDH from B. subtilis. Thus, due to its catalytic effi-
ciency and broad substrate specificity, NMAADH may
be an attractive reagent in the development of biocata-
lytic routes to N-substituted L-amino acids for use as
pharmaceutical intermediates.
9. Tsuda, S.; Kato, T.; Yasohara, R.; Hasegawa, J. Japan
Kokai Patent, 2001, 2001-190298.
10. Lin, M. C. M.; Wagner, C. J. Biol. Chem. 1975, 250,
3746–3751.
11. Shaw, W. V.; Tsai, L.; Stadtman, E. R. J. Biol. Chem.
1966, 241, 935–945.
Acknowledgements
12. Mihara, H.; Kakutani, R.; Muramatsu, H.; Yasuda, M.;
Ueda, M.; Kurihara, T.; Esaki, N. (in preparation).
13. Goncalves, L. P. B.; Antunes, O. A. C.; Pinto, G. F.;
Oestreicher, E. G. Tetrahedron: Asymmetry 1996, 5,
1237–1240.
14. van der Donk, W. A.; Zhao, H. Curr. Opin. Biotechnol.
2003, 14, 421–426.
15. Lampel, K. A.; Uratani, B.; Chaudhry, G. R.; Ramaley,
R. F.; Rudikoff, S. J. Bacteriol. 1986, 166, 238–243.
This work was supported in part by a Grant-in-Aid for
Scientific Research on Priority Areas (B) 13125203 (to
N.E.) from the Ministry of Education, Culture, Sports,
Science, and Technology of Japan, by Grant-in-Aid
for Encouragement of Young Scientists 15780070 (to
H.M.) from the Japan Society for the Promotion of Sci-
ence, by the National Project on Protein Structural and
Functional Analyses and by Grant-in-Aid from the