Asymmetric synthesis of chiral cyclic amine from cyclic imine by bacterial whole-cell catalyst of enantioselective imine reductase

Article abstract of DOI:10.1039/c0ob00353k

Streptomyces sp. GF3587 and 3546 were found to be imine-reducing strains with high R- and S-selectivity by screening using 2-methyl-1-pyrroline (2-MPN). Their whole-cell catalysts produced 91 mM R-2-methylpyrrolidine (R-2-MP) with 99.2%e.e. and 27.5 mM S-2-MP (92.3%e.e.) from 2-MPN at 91-92% conversion in the presence of glucose, respectively.

Full text of DOI:10.1039/c0ob00353k

View Article Online / Journal Homepage / Table of Contents for this issue
COMMUNICATION
www.rsc.org/obc | Organic & Biomolecular Chemistry
Asymmetric synthesis of chiral cyclic amine from cyclic imine by bacterial
whole-cell catalyst of enantioselective imine reductase†
Koichi Mitsukura,* Mai Suzuki, Kazuhiro Tada, Toyokazu Yoshida and Toru Nagasawa
Received 30th June 2010, Accepted 12th August 2010
DOI: 10.1039/c0ob00353k
Streptomyces sp. GF3587 and 3546 were found to be imine-
reducing strains with high R- and S-selectivity by screening
using 2-methyl-1-pyrroline (2-MPN). Their whole-cell cata-
lysts produced 91 mM R-2-methylpyrrolidine (R-2-MP) with
99.2%e.e. and 27.5 mM S-2-MP (92.3%e.e.) from 2-MPN at
91–92% conversion in the presence of glucose, respectively.
enantioselectivity. We also examined the possibility of the efficient
production of R- and S-2-MP using whole-cell catalyst.
2-MPN, which is barely hydrolyzed at neutral pH, is suitable
for the screening of imine reductase activity in microorganisms
(Scheme 1). In the screening of imine reductase, we initially ex-
pected that yeasts such as Candida, Rhodotorula, Saccharomyces,
Sporobolomyces and Pichia might catalyze the reduction of 2-
MPN in a similar way to various carbonyl compounds9,10 or
Chiral amines have broad utility for the syntheses of various
pharmaceutical and agrochemical intermediates. Numerous syn-
theses of optically active amines have been studied for many years.
Recently, catalytic asymmetric syntheses of amines have attracted
attention and have been extensively studied. In the reduction of
imine in organic solvent, chiral amine can be synthesized with
high enantioselectivity by well-designed transition metal catalysts
or organocatalysts.1
C
C double bonds.11,12 However, no 2-MPN-reducing activity
was found among 226 yeast strains tested. Furthermore, we
attempted to carry out a survey of 2-MPN-reducing activity in
our stocked microorganisms (261 strains of bacteria, 117 strains of
actinomycetes and 84 strains of fungi). The genera of strains used
were shown in the ESI.† Surprisingly, 2-MPN-reducing activity
was found in only five filamentous bacteria (Fig. 1). Each of them
was identified to be Streptomyces sp. based on 16S ribosomal DNA
sequence analysis. Streptomyces sp. GF3501, 3585, 3587 and 4415
reduced 2-MPN to R-2-MP with 72%e.e., 80%e.e. 99.2%e.e. and
90%e.e., respectively. Streptomyces sp. GF3546 was the sole strain
with high S-selectivity (81%e.e.). Among them, Streptomyces sp.
GF3587 showed the highest conversion and enantioselectivity
(Fig. 1).
For enzymatic syntheses of chiral amines, several approaches
have been suggested. They include kinetic or dynamic kinetic
resolutions with hydrolases, enantioselective desymmetrization
with lipase, asymmetric synthesis with amine dehydrogenase,
asymmetric synthesis with transaminase and deracemization with
monoamine oxidase.2 In contrast, enzymatic reduction of imines
seemed to be difficult because most of them are very unstable
in water and rapidly degraded into aldehyde or ketone and
ammonia or amine. Stephens et al. proposed a novel screening
method of imine reductase activity in a biphasic system of
water-tetradecane. They found that Acetobacterium woodii DSM
1030 catalyzed not only the reduction of C C double bond of
phenylacrylate, such as caffeate,3,4 but also the reduction of aryl
or alkyl imines.5 Recently, asymmetric reduction of aryl imines
synthesized from acetophenone and aniline was examined using
whole cells of Candida parapsilosis ATCC 7330, having carbonyl
reductase activity.6 Although the imine reduction yielded R-amine
with moderate conversion and high optical purity, the activity was
very low and the addition of large amounts of cells was required
to respond to the imine reduction.7
Scheme 1 Microbial asymmetric reduction of 2-MPN.
We attempted to search for a novel imine-reducing enzyme
exhibiting high activity and high enantioselectivity. We used a
cyclic imine, 2-methyl-1-pyrroline (2-MPN), which is very stable
in water. An optically active 2-methylpyrrolidine (2-MP), R-2-MP
can be useful for chiral building block of H3 histamine receptor
antagonist, ABT-239.8 In the present study, we have reported the
occurrence of bacterial cyclic imine-reducing activity with high
Department of Biomolecular Science, Gifu University, 1-1 Yanagido, Gifu
501-1193, Japan. E-mail: mitukura@gifu-u.ac.jp; Fax: +81-58-293-2794;
Tel: +81-58-293-2649
Fig. 1 Screening of imine reductase activity.
† Electronic supplementary information (ESI) available: Experimental
details, cultivation, derivatization and HPLC chromatogram, 1H, 13C
NMR and mass spectra for the prepared compounds. See DOI:
10.1039/c0ob00353k
We also examined 2-MPN-reducing activity of Candida para-
psilosis ATCC 7330, which has been reported to catalyze
This journal is
The Royal Society of Chemistry 2010
Org. Biomol. Chem.
©

View Article Online
Table 1 Effect of hydroxy compounds on 2-MPN reduction
Compound
Activity/%a
None
100
130
103
91
Glucose
Glycerol
Sorbitol
2-Propanal
62
a Activity is shown as 100% (354 nmol h-1 mg-1 dry cell) in the absence
of compound. The reaction was carried out at 30 ◦C for 12 h: 2-MPN
(20 mM) and cells (11.1 mg).
the reduction of aryl imines such as (E)-N-(1-phenylethyl-
idene)benzenamine.7 Conversion of 2-MPN was conducted using
whole cells of C. parapsilosis (from 4 mL culture broth) at 25 ◦C
for 72 h in 100 mM phosphate buffer (pH 7.0) supplemented with
2% (w/v) glucose. However, no reductase activity for 2-MPN was
detected. These results indicated that the imine reductases found
in Streptomyces sp. GF3587 and 3546 seemed to be novel and
they were promising and most suitable as a whole-cell catalyst
for the production of R- and S-2-MP, respectively. Their culture
conditions were also optimized. To recycle an oxidative-reductive
coenzyme in the cell, the addition of glycerol, sorbitol or 2-
propanol was also examined at 2% (w/v or v/v), as well as glucose.
The addition of glucose resulted in 1.3-fold higher activity than in
the case without additive, but the others were relatively ineffective
or negative effect on the activity (Table 1). The imine reduction
by Streptomyces sp. GF3587 or 3546 proceeded almost linearly up
to 50 mM and 40 mM, respectively. On the basis of these results,
we performed an efficient production of optically active 2-MP at
30 or 25 mL scale using whole cells. Streptomyces sp. GF3587
produced 91 mM R-2-MP (99.2% e.e.) from total 100 mM 2-MPN
in the presence of 4% (w/v) glucose after 84-h incubation (Fig.
2). Streptomyces sp. GF3546 produced 27.5 mM S-2-MP from
30 mM 2-MPN with 92.3% e.e. in the presence of 2% (w/v) glucose
after 72-h incubation (Fig. 3). In both cases the optical purity
of 2-MP formed remained during the reduction reaction. Their
productivities of R- and S-2-MP were 392 and 21 nmol h-1 mg-1
dry cell, respectively. R- and S-2-MP in the reaction solution were
Fig. 3 S-2-MP production by whole cells of GF3546. Reduction of
2-MPN (30 mM) using whole cells (455 mg as dry cell weight) was
conducted at 25 ◦C in 25 mL of 100 mM phosphate buffer (pH 7.0)
containing 2% (w/v) glucose.
converted to the corresponding amine hydrochloride in 73% and
67% isolated yield via both reactions of amino group protection
and N-Boc group deprotection, respectively.13
In conclusion, we elucidated the occurrence of two kinds of
novel imine reductase of prochiral 2-MPN in filamentous bacteria
for the first time. Streptomyces sp. GF3587 and 3546 showed
high enantioselectivity for 2-MPN. Under optimized conditions,
efficient syntheses of R- and S-2-MP were achieved using whole-
cell catalysts in the presence of glucose. For further application of
the microbial imine reduction system to a range of chiral amine
production, more detailed information on R- or S-selective imine
reductase is essential. Further studies are in progress.
References
1 (a) M. Terada and N. Momiyama, Enantioselective synthesis of amines
by chiral Brønsted acid catalysts in Chiral amine synthesis, ed. T. C.
Nugent, Wiley-VCH, Weinheim, 2010, pp. 109–113; (b) Kocovsky and
S. Stoncˇius, Reduction of imines with trichlorosilane catalyzed by chiral
Lewis bases in Ibid., pp 131–156; (c) T. L. Church and P. G. Andersson,
Chiral amines from transition-metal-mediated hydrogenation and
transfer hydrogenation in Ibid., pp. 179–223; (d) G. Shang, W. Li
and X. Zhang, Transition metal-catalyzed homogeneous asymmetric
hydrogenation in Catalytic asymmetric synthesis, ed. I. Ojima, John
Wiley and Sons, Inc., Hoboken, New Jersey, 2010, pp. 414–436.
2 (a) N. J. Turner, Curr. Opin. Chem. Biol., 2010, 14, 115; (b) M. Ho¨hne
and U. T. Bornscheuer, ChemCatChem, 2009, 1, 42; (c) V. Gotor-
Frena´ndez and V. Gotor, Curr. Opin. Drug Discovery Dev., 2009, 12, 784;
(d) N. J. Turner and M. D. Truppo, Biocatalytic routes to nonracemic
chiral amines in Chiral amine synthesis, ed. T. C. Nugent, Wiley-
VCH, Weinheim, 2010, pp. 431–459; (e) H. Gro¨ger, Enzyme-catalyzed
asymmetric synthesis in Catalytic asymmetric synthesis, ed. I. Ojima,
John Wiley and Sons, Inc., Hoboken, New Jersey, 2010, pp. 269–341.
3 R. Bache and N. Pfennig, Arch. Microbiol., 1981, 130, 255.
4 A. Tschech and N. Pfennig, Arch. Microbiol., 1984, 137, 163.
5 H. Li, P. Williams, J. Micklefield, J. M. Gardiner and G. Stephens,
Tetrahedron, 2004, 60, 753.
ˇ
´
6 B. Basker, N. G. Pandian, K. Priya and A. Chadha, Tetrahedron:
Asymmetry, 2004, 15, 3961.
7 T. Vaijayanthi and A. Chadha, Tetrahedron: Asymmetry, 2008, 19, 93.
8 M. Cowart, J. K. Pratt, A. O. Stewart, Y. L. Bennani, T. A. Esbenshade
and A. A Hancock, Bioorg. Med. Chem. Lett., 2004, 14, 689.
9 M. Kataoka, K. Kita, M. Wada, Y. Yasohara, J. Hasegawa and S.
Shimizu, Appl. Microbiol. Biotechnol., 2003, 62, 437.
10 K. Goldberg, K. Schroer, S. Lu¨tz and A. Liese, Appl. Microbiol.
Biotechnol., 2007, 76, 249.
11 M. Kataoka, A. Kotaka, A. Hasegawa, M. Wada, A. Yoshizumi, S.
Nakamori and S. Shimizu, Biosci., Biotechnol., Biochem., 2002, 66,
2651.
Fig. 2 R-2-MP production by whole cells of GF3587. Reduction of
2-MPN using whole cells (83 mg as dry cell weight) was conducted at
25 ◦C in 30 mL of 100 mM phosphate buffer (pH 7.0) containing 4% (w/v)
glucose. Initial concentration of 2-MPN was at 50 mM. With monitoring
of 2-MP formation on HPLC, 2-MPN was added to the reaction mixture
three times (0.6, 0.6 and 0.3 mmol) during 72-h incubation. Arrows
represent the addition of 2-MPN.
Org. Biomol. Chem.
This journal is
The Royal Society of Chemistry 2010
©

View Article Online
12 M. Wada, A. Yoshizumi, Y. Noda, M. Kataoka, S. Shimizu, H. Takagi
and S. Nakamori, Appl. Environ. Microbiol., 2003, 69, 933.
The crude product was treated at 25 ◦C for 1 h with 4M HCl-ethyl
acetate (6 mL) and evaporated. After washing of the residue with ethyl
acetate, the corresponding amine hydrochloride was obtained as white
solid (243 mg, 2.0 mmol, 99%e.e., 73% yield). In the same manner as R-
2-MP hydrochloride synthesis, S-2-MP hydrochloride was synthesized
from the solution (25 mL) including 27.5 mM S-2-MP (92%e.e.) to
give the corresponding amine hydrochloride as white solid (56 mg,
0.46 mmol, 92%e.e., 67% yield).
13 After removal of cells from reaction mixture by centrifugation (12,000
g, 5 min), di-tert-butyl dicarbonate (894 mg, 4.10 mmol), NaHCO3
(0.50 g, 5.95 mmol) and EtOH (20 mL) was added to the reaction
solution (30 mL) including 91 mM R-2-MP (99.2%e.e.). The reaction
was carried out at 25 ◦C for 15 h. The reaction product was extracted
with ethyl acetate (50 mL) thrice, dried over Na2SO4, and evaporated.
This journal is
The Royal Society of Chemistry 2010
Org. Biomol. Chem.
©