Efficient production of enantiomerically pure chiral amines at concentrations of 50 g/L using transaminases

Article abstract of DOI:10.1021/op900303q

Two methods for the efficient (50 g/L) production of optically pure amines from their corresponding ketones using transaminases have been developed. The first method utilizes an ion-exchange resin for in situ product removal allowing the reaction to be carried out a substrate concentration of 50 g/L. The second approach relies upon conversion of the initially formed amine, via spontaneous cyclisation, to a noninhibitory product. Both methods have been demonstrated at 50 mL scale. (R)- and (S)-methylbenzylamine, and (R)- and (S)-6-methyl-2- piperidone have been produced in >90% isolated yield and >99% ee.

Full text of DOI:10.1021/op900303q

Organic Process Research & Development 2010, 14, 234–237
Efficient Production of Enantiomerically Pure Chiral Amines at Concentrations of 50
g/L Using Transaminases
,†
‡
,†
Matthew D. Truppo,* J. David Rozzell, and Nicholas J. Turner*
School of Chemistry, UniVersity of Manchester, Manchester Interdisciplinary Biocentre, 131 Princess Street,
Manchester M17DN, U.K. and Codexis Inc., 129 North Hill AVenue, Pasadena, California 91106, U.S.A.
Abstract:
Scheme 1. Overview of transaminase-catalysed reaction
methods
Two methods for the efficient (50 g/L) production of optically pure
amines from their corresponding ketones using transaminases have
been developed. The first method utilizes an ion-exchange resin
for in situ product removal allowing the reaction to be carried
out a substrate concentration of 50 g/L. The second approach relies
upon conversion of the initially formed amine, via spontaneous
cyclisation, to a noninhibitory product. Both methods have been
demonstrated at 50 mL scale. (R)- and (S)-methylbenzylamine, and
(R)- and (S)-6-methyl-2-piperidone have been produced in >90%
isolated yield and >99% ee.
Introduction
In view of the importance of enantiomerically pure chiral
amines as building blocks for pharmaceuticals, there is currently
considerable effort underway to develop efficient catalytic
production on scale.8,9 First, the equilibrium constant for the
reaction often favors the ketone starting material requiring the
development of methods to drive the reaction to completion.
Second, transaminases typically suffer significant product
inhibition by both the desired amine product and the keto acid
byproduct. Although methods have been reported which address
1
methods for their preparation. Approaches based on biocatalysis
have largely relied upon kinetic resolution of racemic amides
or amines using lipases, and indeed such methods have been
2
successfully commercialized at scale. Some groups have
recently addressed the issue of in situ racemization of the
unreactive enantiomer by developing catalysts for amine race-
mization permitting dynamic kinetic resolutions to be carried
10
the issue of product inhibition, no general procedures have
been reported for carrying out productive, scaleable reactions.
Generally, for a biocatalyst to be considered for an industrial
application, the process must tolerate substrate concentrations
3-5
out. Alternatively, racemic amines can be deracemized in
high yield and ee by the combined use of an enantioselective
11
6
of at least 50 g/L. Herein, we report a simple procedure for
conducting transaminase-catalysed reactions at substrate con-
centrations of 50 g/L.
amine oxidase and a nonselective reducing agent. However,
the emergence of transaminases, which catalyse the direct
amination of ketones to chiral amines using ammonia, has
7
provided an alternative and highly attractive additional option.
Results
Transaminases typically possess high turnover rate, stability,
and a tightly bound pyridoxal cofactor. However, two factors
have limited the application of transaminases for chiral amine
Scheme 1 shows two alternative approaches for carrying out
12
transaminase catalysed reactions. In Method 1, alanine is
employed as the amine donor, and the pyruvate generated is
*
Author for correspondence. E-mail: nicholas.turner@manchester.ac.uk. Fax:
13
reduced by lactate dehydrogenase (LDH) to L-lactate. Remov-
+
44 161 275 1311. Telephone: +44 161 306 5173.
†
‡
ing the pyruvate serves the dual purpose of driving the reaction
University of Manchester.
Codexis.
(
1) Turner, N. J.; Carr, R. In Biocatalysis in the Pharmaceutical and
Biotechnology Industries; Patel, R. N., Ed.; CRC Press: Boca Raton,
FL, 2007, pp 743-755.
(7) Faber, K.; Kroutil, W. Curr Opin. Chem. Biol. 2005, 9, 181. H o¨ hne,
M.; Robins, K.; Bornscheuer, U. T. AdV 2008, 350, 807. Stewart, J. D.
Curr. Opin. Chem. Biol. 2001, 5, 120. Koszelewski, D.; Lavandera,
I.; Clay, D.; Guebitz, G. M.; Rozzell, D.; Kroutil, W. Angew. Chem.,
Int. Ed 2008, 47, 9337. Kaulmann, U.; Smithies, K.; Smith, M. E. B.;
Hailes, H. C.; Ward, J. M. Enz. Microb. Technol. 2007, 41, 628.
(8) Patel, R. N. Enz. Microb. Technol. 2000, 27, 376.
(
2) Ismail, H.; Lau, R. M.; van Rantwijk, F.; Sheldon, R. A. AdV. Synth.
Catal. 2008, 350, 1511. Breuer, M.; Ditrich, K.; Habicher, T.; Hauer,
B.; Kesseler, M.; St u¨ rmer, R.; Zelinski, T. Angew. Chem., Int. Ed.
2
004, 43, 788.
(
(
3) Paetzold, J.; B a¨ ckvall, J. E. J. Am. Chem. Soc. 2005, 127, 17620.
4) Blacker, A. J.; Stirling, M. J.; Page, M. I. Org. Process Res. DeV.
(9) Cameron, M.; Cohen, D.; Cottrell, I. F.; Kennedy, D.; Roberge, C.;
Chartrain, M. J. Mol. Catal. B: Enzym. 2001, 14, 1.
2
007, 11, 642.
(10) H o¨ hne, M.; K u¨ hl, S.; Robins, K.; Bornscheuer, U. T. ChemBioChem
2008, 9, 363.
(
(
5) Kim, M.-J.; Kim, W.-H.; Han, K.; Choi, Y. K.; Park, J. Org. Lett.
2
007, 9, 1157.
(11) Pollard, D. J.; Woodley, J. M. Trends Biotechnol. 2007, 25, 66.
(12) Truppo, M. D.; Rozzell, J. D.; Moore, J. C.; Turner, N. J. Org. Biomol.
Chem. 2009, 395.
6) Dunsmore, C. J.; Carr, R.; Fleming, T.; Turner, N. J. J. Am. Chem.
Soc. 2006, 128, 2224.
2
34
•
Vol. 14, No. 1, 2010 / Organic Process Research & Development
10.1021/op900303q  2010 American Chemical Society
Published on Web 12/28/2009

Figure 1. Inhibition of the transaminase enzyme by the product
methylbenzylamine (MBA).
Figure 2. Effect of ion-exchange resin on aqueous concentration
of methylbenzylamine (MBA). MBA concentrations corre-
sponding to 80% and 50% transaminase inhibition levels are
shown with red lines.
equilibrium and also eliminating pyruvate inhibition of the
transaminase. Glucose dehydrogenase (GDH) is used to recycle
the NADH cofactor. In Method 2, a large excess of isopropy-
lamine (1 M) is employed as the amine donor, instead of
14
alanine.
Using these approaches we recently reported the conversion
of acetophenone to (R)- or (S)-R-methylbenzylamine (MBA)
(99% ee) at substrate concentrations of up to 6 g/L (Method 1)
and 2.5 g/L (Method 2). Significant inhibition of the transami-
nase by the amine product prevented running the reactions at
higher substrate loading (Figure 1). For example, 50% inhibition
of the transaminase was observed at an MBA concentration of
6
g/L, with >80% inhibition at a concentration of 12 g/L.
To alleviate product inhibition, in situ product removal using
an ion-exchange resin (Amberlite XAD 1180) was examined.
To a solution of MBA (50 g/L) in 100 mM potassium phosphate
buffer (pH 7.5) were added various concentrations of ion-
exchange resin from 0 to 400 g/L. Each solution was stirred
for 5 min at 30 °C, after which the ion-exchange resin was
removed by filtration and the reaction buffer then sampled to
determine the aqueous MBA concentration by HPLC analysis.
Figure 2 shows that increasing the resin loading leads to a
decrease in the aqueous MBA concentration as expected. A
resin loading of 200 g/L resulted in an MBA concentration of
Figure 3. Transamination of acetophenone with and without
ion-exchange resin and lactate dehydrogenase (LDH).
to ∼10% conversion highlighting inhibition of the transaminase
by the MBA product. Pleasingly, the reaction containing both
LDH and the ion-exchange resin proceeded to >99% conver-
sion. Both (S)- and (R)-MBA were produced using the ion-
exchange resin process in >99% ee using transaminases ATA-
4
g/L, below the 50% transaminase inhibition threshold, and
hence this loading was selected for further experiments.
Fifty-milliliter scale reactions were run with automated pH
control and 50 g/L of acetophenone. Control of pH is necessary
because of acid production from recycling of the NADH
cofactor. Three separate reactions were run to demonstrate the
effectiveness of the system: (i) no lactate dehydrogenase (LDH)
to consume the pyruvate ketoacid byproduct and drive the
reaction equilibrium, (ii) with LDH but no ion-exchange resin
to alleviate transaminase inhibition, (iii) with both LDH and
1
13 and ATA-117, respectively. The desired MBA product was
easily recovered in >90% yield by washing the resin with buffer
pH 11).
As expected, the ion-exchange resin was not effective in
(
selectively binding the product MBA in reactions containing a
high concentration (1 M) of isopropylamine as the amine donor
(Figure 1, Method 2). The isopropylamine competed with the
MBA for binding to the resin, rendering the in situ product
removal ineffective. An alternative solution was envisaged in
which the product of the transamination reaction underwent
spontaneous and irreversible conversion to a secondary product,
thereby simultaneously driving the reaction to completion and
also eliminating product inhibition. Ethyl 4-acetylbutyrate 1 was
chosen as a model substrate. Upon transamination of the ketone,
the amine 2 spontaneously cyclizes to the 6-methyl-2-piperidone
2
00 g/L ion-exchange resin (Figure 3).
The reaction without LDH proceeded to <1% conversion
as a result of the equilibrium favoring ketone over amine. The
reaction containing LDH but no ion-exchange resin proceeded
(
13) Addition of LDH to whole-cell mediated transaminations has been
reported. Shin, J.-S.; Kim, B.-G. Biotechnol. Bioeng. 1999, 65, 206.
14) Wu, W.; Bhatia, M. B.; Lewis, C. M.; Lang, W.; Wang, A.; Matcham,
G. W. U.S. Patent 6,133,018, 2000.
(
Vol. 14, No. 1, 2010 / Organic Process Research & Development
•
235

Scheme 2. Transamination of ethyl 4-acetylbutyrate
followed by spontaneous cyclisation to 6-methyl-2-piperidone
using isopropylamine as the amine donor
Experimental Section
Conversion Analysis. Reaction conversions were monitored
using reverse phase high performance liquid chromatography
(HPLC) at 220 nm using an Agilent 1100 series HPLC and a
Zorbax Eclipse XDB-C18 (50 mm × 4.6 mm) column with a
flow rate of 1 mL/min (60% acetonitrile/40% water) for 3 min.
Enantiomeric Excess Analysis. Enantiomeric excess of
methylbenzylamine was determined by normal phase high
performance liquid chromatography (HPLC) at 220 nm using
an Agilent 1100 series HPLC and a Chiralpak OD-H (250 mm
×
4.6 mm) column with a flow rate of 1 mL/min (90% hexanes/
0% 2-propanol) for 12 min. The retention times of the (S)-
1
and (R)-enantiomers were 8.7 and 7.2 min, respectively. Specific
rotation of the methylbenzylamine product was established by
comparison to known standards that were purchased from
Sigma.
The enantiomeric excess of 6-methyl-2-piperidone was
determined by normal phase high performance liquid chroma-
tography (HPLC) at 220 nm using an Agilent 1100 series HPLC
and a Chiralpak AD-H (250 × 4.6 mm) column with a flow
rate of 0.5 mL/min (80% hexanes/20% ethanol) for 20 min.
The retention times of the (S)- and (R)-enantiomer were 17 and
3
product (Scheme 2). Optically pure intermediates of this type
15
are found in a variety of natural products.
Fifty-milliliter scale reactions were run with 50 g/L ethyl
-acetylbutyrate and 1 M isopropylamine as the amine donor.
4
The reactions proceeded to >99% conversion in <15 h (Figure
). Both (S)- and (R)-6-methyl-2-piperidone were produced with
99% ee using transaminases ATA-113 and ATA-117, respec-
4
15 min, respectively. The absolute configuration of the 6-meth-
>
yl-2-piperidone products was established by comparison to that
tively. The desired product was isolated in >90% yield via
extraction into dichloromethane.
16
of authentic standards using chiral HPLC.
Reagents and Enzymes. Commercial-grade reagents and
solvents were purchased from Sigma-Aldrich and used without
further purification. All enzymes, including transaminases
(ATA-113 and ATA-117), glucose dehydrogenase (GDH CDX-
In summary, we have developed transamination procedures
that can be run at industrially acceptable substrate concentrations
of 50 g/L. The approach is based on the elimination of product
inhibition that commonly plagues transaminase reactions by the
use of an ion-exchange resin to selectively remove the desired
product amine in situ. The method has been successfully applied
to the transamination of acetophenone yielding (S)- or (R)-
methylbenzylamine in >90% isolated yield and >99% ee.
A second approach, which is substrate dependent, relies upon
spontaneous cyclisation of the initially formed amine to a
secondary product. This method has been successfully applied
to the transamination of ethyl 4-acetylbutyrate at 50 g/L yielding
9
01), and lactate dehydrogenase (LDH-102) are catalogue
enzymes of Codexis and were generously supplied by Codexis
Redwood City, CA).
(
Transamination of Acetophenone. Transaminations of
acetophenone were conducted at 50-mL scale using the lactate
dehydrogenase/glucose dehydrogenase system under the fol-
lowing conditions: 100 mM potassium phosphate buffer, 1 g/L
+
NAD , 0.5 g/L pyridoxal-5-phosphate, 90 g/L glucose, 90 g/L
alanine, 50 g/L acetophenone, 1 g/L glucose dehydrogenase
(GDH), 1 g/L lactate dehydrogenase (LDH), 5 g/L transaminase
(
>
S)- or (R)-6-methyl-2-piperidone in >90% isolated yield and
99% ee.
(ATA), and 200 g/L ion-exchange resin (Amberlite XAD 1180).
Reactions were run at 30 °C and pH 7.5 in a Multimax reactor
system with overhead mechanical stirring at 400 rpm. Reaction
pH was controlled through the automated addition of 2 M
NaOH.
Upon complete conversion, the ion-exchange resin was
recovered from the reaction mixture by filtration and washed
with pH 11 buffer to recover the methylbenzylamine product.
Transamination of Ethyl 4-Acetylbutyrate. Transamina-
tions of ethyl 4-acetylbutyrate were conducted at 50 mL scale
using the isopropylamine amino donor system under the
following conditions: 100 mM potassium phosphate buffer, 0.5
g/L pyridoxal-5-phosphate, 1 M isopropylamine, 50 g/L ethyl
4-acetylbutyrate, and 5 g/L transaminase (ATA). Reactions were
run at 20 °C and pH 9.5 in a Multimax reactor system with
overhead mechanical stirring at 400 rpm.
(
(
15) Freville, S.; Celerier, J. P.; Thuy, V. M.; Lhommet, G. Tetrahedron:
Asymmetry 1995, 6, 2651.
Figure 4. Transamination of ethyl 4-acetylbutyrate using
isopropylamine transamination system.
16) Urbanczyk-Lipkowska, Z.; Fukuda, N.; Tanaka, K. Tetrahedron:
Asymmetry 2007, 18, 1254.
2
36
•
Vol. 14, No. 1, 2010 / Organic Process Research & Development

Upon complete conversion, the reaction was extracted with
volumes of dichloromethane followed by removal of the
solvent by via vacuum distillation.
Supporting Information Available
2
HPLC chromatograms showing enantiomeric excess analysis
of methylbenzylamine products are available. This information
is available free of charge via the Internet at http://pubs.acs.org/.
Acknowledgment
This research was supported through the Merck & Co., Inc.
Doctoral Study program. We gratefully acknowledge Codexis
Inc. for providing all enzymes used in this research.
Received for review November 16, 2009.
OP900303Q
Vol. 14, No. 1, 2010 / Organic Process Research & Development
•
237