Chemoenzymatic asymmetric total synthesis of (S)-Rivastigmine using ω-transaminases

Article abstract of DOI:10.1039/c0cc00585a

A straightforward, high-yielding, chemoenzymatic total synthesis of enantiopure (S)-Rivastigmine was developed using various ω-transaminases for the asymmetric amination of appropriate acetophenone precursors. Optimisation of the biotransformation allowed scale-up and the total synthesis of (S)-Rivastigmine. The Royal Society of Chemistry 2010.

Full text of DOI:10.1039/c0cc00585a

View Article Online / Journal Homepage / Table of Contents for this issue
This article is part of the
2010 ‘Enzymes & Proteins’
web themed issue
This issue showcases high quality research in the field of enzymes and
proteins.
Please visit the website to access the other papers in this issue:-
http://www.rsc.org/chemcomm/enzymesandproteins

COMMUNICATION
www.rsc.org/chemcomm | ChemComm
Chemoenzymatic asymmetric total synthesis of (S)-Rivastigmine
using x-transaminaseswz
Michael Fuchs,a Dominik Koszelewski,b Katharina Tauber,a Wolfgang Kroutila and
Kurt Faber*a
Received 25th March 2010, Accepted 26th April 2010
First published as an Advance Article on the web 11th May 2010
DOI: 10.1039/c0cc00585a
A straightforward, high-yielding, chemoenzymatic total synthesis
of enantiopure (S)-Rivastigmine was developed using various
x-transaminases for the asymmetric amination of appropriate
acetophenone precursors. Optimisation of the biotransformation
allowed scale-up and the total synthesis of (S)-Rivastigmine.
pyruvate. Enzyme cascade reactions using either pyruvate
dehydrogenase9,10 or pyruvate decarboxylase have been
employed,11 and the recycling of alanine using an amino acid
dehydrogenase in the presence of ammonium to directly
regenerate the amine donor alanine has been also demon-
strated as a formal reductive amination protocol.10,12
Due to the facile substrate-inhibition of transaminases,13
acetophenone derivatives are generally poor substrates, but we
envisaged that a combination of the appropriate type of
substituent(s), optimization of the reaction conditions and a
suitable enzyme might give access to the enantiopure amine
precusor 3 and the (S)-enantiomer of Rivastigmine (1).
(S)-Rivastigmine {(S)-3-[1-(dimethylamino)-ethyl]phenyl ethyl-
(methyl)carbamate (1)} represents one of the most potent
drugs for the treatment of Alzheimer’s disease at early stages1,2
and it also exerts a beneficial influence on dementia of
Parkinson patients.3 Trials proved that the (S)-enantiomer
exhibits the desired cholinesterase inhibition, which requires
the drug in enantiomerically pure form.1
For an initial screening of unprotected 3-hydroxyaceto-
phenone (2a) seven different transaminases were used:
o-transaminases from Bacillus megaterium SC6394 (BM-oTA),14
from Chromobacterium violaceum DSM 30191 (CV-oTA),15
and from Arthrobacter species CNB05-01 (ArS-oTA)16 as
well as 4 commercial enzymes [Vibrio fluvialis o-transaminase
(Vf-oTA), ATA-113, ATA-114, and ATA-117] obtained
from Codexis. In order to shift the equilibrium towards the
amine product by removal of the co-product pyruvate,
we chose lactate dehydrogenase combined with the glucose
dehydrogenase/glucose recycling system to regain the nicotin-
amide adenine dinucleotide cofactor (NADH, see Scheme 1).
Initial results showed that the stereoselectivities for
m-hydroxyacetophenone (2a) were high, but the conversions
were low. Only Vf-oTA gave amine 3a in 64% (Table 1,
entries 1–4). In order to improve the reaction rates, we focused
on the employment of cosolvents to increase the solubility of
the substrates (all substrates in this study were insoluble in the
reaction buffer) and substrate-engineering as the two major
strategies. Since dimethylsulfoxide (DMSO) exerted a negative
effect (Table 1, entry 5), we focused on substrate engineering
To date, enantiomerically pure (S)-Rivastigmine is obtained
via racemate resolution using various chiral acids,4 through
transition metal catalysis5,6 or via lipase-catalyzed kinetic
resolution of hydroxy-precursors.7 All of these methods have
certain drawbacks, such as complex sequential operations,
trace impurities of metals, or multiple crystallisation steps
involving diastereomeric salts.
Alternatively, newly developed biocatalytic processes based
on o-transaminases have demonstrated to be an efficient tool
for the preparation of enantiopure amines.8 In this process,
transamination of ketones gives direct access to the corres-
ponding a-chiral primary amines, thereby breaking the limits
of 50% maximum yield of racemate resolution. The challenge
in enzymatic transamination is to shift the unfavourable
equilibrium, which is on the substrate side, i.e. the ketone
substrate and the amine donor alanine.
Various approaches have proven to be successful, mainly by
pulling the equilibrium via removal of the formed coproduct
a Department of Chemistry, Organic & Bioorganic Chemistry,
University of Graz, Heinrichstrasse 28, A-8010 Graz, Austria.
E-mail: Kurt.Faber@Uni-Graz.at; Fax: +43 316-380-9840;
Tel: +43 316-380-5332
b Austrian Centre of Industrial Biotechnology, Heinrichstrasse 28,
A-8010 Graz, Austria
w This article is part of the ‘Enzymes and Proteins’ web-theme issue for
ChemComm.
z Electronic supplementary information (ESI) available: Detailed
results from screening experiments, experimental data, NMR spectra
and selected chromatograms. See DOI: 10.1039/c0cc00585a
Scheme
1
Transamination of acetophenone derivatives using
o-transaminases.
ꢀc
This journal is The Royal Society of Chemistry 2010
5500 | Chem. Commun., 2010, 46, 5500–5502

Table
1
Stereo-complementary enzymatic amination of aceto-
entries 12 and 13), although it considerably increased the
enantiomeric excess (ee) from 66 to 99% in the case of
Vf-oTA (Table 1, entry 12). On the other hand, ATA-117
showed slightly reduced enantioselectivity under these condi-
tions (Table 1, entry 13). Changes in the enantioselectivity of
biocatalysts induced by cosolvents have been frequently
observed for various enzymes, such as nitrile- and epoxide-
hydrolases18 and lipases.19
phenones 2a–f yielding (R)- or (S)-a-methylbenzylamines 3a–fa
Entry Enzyme
Substrate R = Productb Conv.c (%) eed (%)
1
2
3
CV-oTAj m-OH (2a)
Vf-oTA m-OH (2a)
ATA-114 m-OH (2a)
3a
3a
3a
3a
3a
29
64
15
33
499
98
499
98
4e
5f
6
ATA-117 m-OH (2a)
m-OH (2a)
Vf-oTA
Vf-oTA
24
98
m-OTMS (2b) 3b
n.c.
n.c.
60
86
87
58
89
45
20
10
499
13
96
81
63
99 (80)i
76 (56)i
n.d.
n.d.
499
66
499
93
Next, we investigated the substitution pattern on the aryl
moiety by using p-methoxy- (2d) and o-methoxyacetophenone
(2e). All o-transaminases showed significantly reduced
activities for the para-substituted substrate 2d (Table 1,
entries 14 and 15). On the other hand, the o-substrate 2e gave
full conversion with excellent (S)-selectivity using Vf-oTA,
whereas ATA-114 showed again a significant drop in activity
(Table 1, entries 16 and 17). This strong impact of subtle
changes in the substitution pattern of the aryl moiety is
presumably due to steric effects: the less sterically demanding
substrate 2c gave good conversions and high enantio-
selectivities with ATA-114 and Vf-oTA in the presence of
DMSO. The stereo-complementary (R)-selective ATA-117
gave the mirror image product in 93% ee (Table 1, entry 11).
Encouraged by these results, we envisaged protective groups
that can be cleaved under milder conditions than the methyl-
group.5–7 In this context, the methoxymethyl ether (MOM)
group proved to be perfectly suitable (Table 1, entries 18–20),
giving higher conversions and perfect enantioselectivities in
case of Vf-oTA (Table 1, entries 9–18). The origin of this
selectivity-enhancement is most probably due to the increased
steric requirements of the MOM-group. Again, (S)-3f was
predominantly formed, only ATA-117 gave the (R)-enantiomer
(detailed data are given in the ESIz). With the optimized
conditions in hand, the reaction was performed at 50 mmol
scale using (S)-selective Vf-oTA and (R)-selective ATA-117 in
order to access both enantiomers (Table 1, entries 21 and 22).
In both cases conversion as well as enantioselectivity were
similar to those from screening experiments, while isolated
yields were acceptable.
7
8
9
10
11e
12g
ATA-114 m-OTMS (2b) 3b
CV-oTAj m-OMe (2c)
(S)-3c
Vf-oTA m-OMe (2c)
ATA-114 m-OMe (2c)
ATA-117 m-OMe (2c)
Vf-oTA
(S)-3c
(S)-3c
(R)-3c
(S)-3c
(R)-3c
3d
m-OMe (2c)
499
86
13e,g ATA-117 m-OMe (2c)
14
15
16
17
18
19
20e
21h
Vf-oTA
ATA-114 p-OMe (2d)
Vf-oTA o-OMe (2e)
ATA-114 o-OMe (2e)
Vf-xTA
ATA-114 m-OMOM (2f) (S)-3f
ATA-117 m-OMOM (2f) (R)-3f
Vf-xTA
p-OMe (2d)
499
499
499
499
499
499
94
3d
3e
3e
m-OMOM (2f) (S)-3f
m-OMOM (2f) (S)-3f
499
98
22e,h ATA-117 m-OMOM (2f) (R)-3f
a
Reaction conditions: substrates 2a–f (50 mM), L-alanine (250 mM),
crude o-transaminase (10 mg), LDH mix (30 mg) containing LDH,
GDH, glucose and NAD+, phosphate buffer (100 mM, pH 7.0, 1 mM
b
pyridoxal 50-phosphate), shaking at 30 1C for 24 h. Absolute
configuration is only given for compounds with determined [a]D20
c
d
value. Determined via GC-analysis; n.c. = no conversion. Deter-
mined via GC-analysis using a chiral stationary phase. n.d. = not
e
determined. D-Alanine was employed. DMSO (10 v/v%) was used
f
g
as cosolvent. DMSO (5 v/v%) was used as cosolvent. Experiment
h
i
was upscaled to 100 mg scale. Isolated yields are given in parentheses.
j
Whole cell system (30 mg) was used.
with respect to the protective group masking the phenolic
OH-moiety.17 Thus, various protective groups were tested,
including the trimethylsilyl- (TMS, 2b) and the methyl-
group (2c).
Substrate 2b was not accepted by all o-transaminases
(Table 1, entries 6 and 7, detailed data are given in the ESIz).
On the other hand, O-methyl-protected substrates gave good
to excellent conversions and moderate to excellent enantio-
selectivities (Table 1, entries 8–11). Again, DMSO did
not have any beneficial effects on the reaction rate (Table 1,
To complete synthesis of (S)-1, precursor (S)-3f (499% ee)
was N-methylated in the presence of formaldehyde and
sodium triacetoxyborohydride in 95% yield (Scheme 2).7
O-Deprotection was achieved quantitatively using acidic condi-
tions (aqueous hydrochloric acid, 5 M) at room temperature.
Scheme 2 Chemoenzymatic asymmetric synthesis of (S)-Rivastigmine (1).
ꢀc
This journal is The Royal Society of Chemistry 2010
Chem. Commun., 2010, 46, 5500–5502 | 5501

Since the latter deprotection step could be easily incorporated
into the work-up of the N-methylation step (see ESIz), the
reaction sequence was shortened to furnish (S)-5 in 92% yield
over 2 steps. O-Carbamoylation of the latter was achieved by
treatment with N-ethyl-N-methylcarbamoyl chloride (7),
which was generated from triphosgene (6) and N-ethyl-N-
methylamine, giving (S)-Rivastigmine (1) in 97% yield and
enantiopure form (ee 4 99%).7 It is noteworthy that purifica-
tion of all intermediates was achieved by simple extraction and
washing operations without the necessity for chromatography.
In summary, a highly stereoselective and short chemo-
enzymatic synthesis of (S)-Rivastigmine (1) was developed
with an overall isolated yield of 71% via a four step procedure.
The key building block (S)-3b was derived via enzyme-
catalyzed asymmetric transamination of a structurally tuned
ketone using o-transaminase from V. fluvialis, whereas the
mirror-image product (R)-3b was accessed using the stereo-
complementary enzyme ATA-117. The presented method
towards (S)-Rivastigmine (1) represents the shortest route
published to date.
H. Stepankova, J. Hajicek, and S. Simek, PCT, WO 2004/037771
A1, 2004; H. Stepankova, J. Hajicek and S. Simek, Chem. Abstr.,
2004, 142, 6315; D. W. Ma, Q. B. Pan, and S. Pan, PCT, WO 2007/
025481, 2007; D. Ma, Q. Pan and S. Pan, Chem. Abstr., 2007, 146,
295621; M. J. V. Garrido, A. M. Montserrat, and M. J. Jua
PCT, WO 2007/014973, 2007; M. Garrido, J. Vicente,
A. M. Montserrat and M. J. Juarez, Chem. Abstr., 2007, 146,
206113.
5 A. A. Boezio, J. Pytkowicz, A. Cote
Chem. Soc., 2003, 125, 14260–14261.
´
rez,
´
´
and A. B. Charette, J. Am.
6 M. Hu, F.-L. Zhang and M.-H. Xie, Synth. Commun., 2009, 39,
1527–1533.
7 J. Mangas-Sanchez, M. Rodrıguez-Mata, E. Busto, V. Gotor-
´ ´
Fernandez and V. Gotor, J. Org. Chem., 2009, 74, 5304–5310.
´
8 D. J. Ager, T. Li, D. P. Pantaleone, R. F. Senkpeil, P. P. Taylor
and I. G. Fotheringham, J. Mol. Catal. B, 2001, 11, 199–205;
J. D. Stewart, Curr. Opin. Chem. Biol., 2001, 5, 120–129.
9 D. Koszelewski, I. Lavandera, D. Clay, D. Rozzell and W. Kroutil,
Adv. Synth. Catal., 2008, 350, 2761–2766; J.-S. Shin and
B.-G. Kim, Biotechnol. Bioeng., 1999, 65, 206–211.
10 M. D. Truppo, J. D. Rozzell, J. C. Moore and N. J. Turner, Org.
Biomol. Chem., 2009, 7, 395–398.
11 M. Hohne, S. Kuhl, D. Robins and U. T. Bornscheuer, ChemBioChem,
¨
2008, 9, 363–365.
¨
12 D. Koszelewski, I. Lavandera, D. Clay, G. M. Guebitz, D. Rozzell
and W. Kroutil, Angew. Chem., Int. Ed., 2008, 47, 9337–9340.
13 J.-S. Shin and B.-G. Kim, Biotechnol. Bioeng., 1997, 55, 348–358;
J.-S. Shin and B.-G. Kim, Biotechnol. Bioeng., 1998, 60, 534–540.
14 R. L. Hanson, B. L. Davis, Y. Chen, S. L. Goldberg, W. L. Parker,
T. P. Tully, M. A. Montana and R. N. Patel, Adv. Synth. Catal.,
2008, 350, 1367–1375.
15 U. Kaulmann, K. Smithies, M. E. B. Smith, H. C. Hailes and
J. M. Ward, Enzyme Microb. Technol., 2007, 41, 628–637.
16 S. Pannuri, S. V. Kamat, and A. R. M. Garcia, PCT, WO 2006/
063336, 2006; S. Pannuri, S. V. Kamat and A. R. M. Garcia, Chem.
Abstr., 2006, 145, 42156.
We would like to acknowledge Klaus Zangger and his
working group for recording the NMR spectra, Barbara
Grischek who supplied us with the freeze dried whole cell
systems and Francesco Mutti for fruitful discussions during
the work.
Notes and references
1 A. Enz, Ger. Offen., DE 3805744, 1988; A. Enz, Chem. Abstr.,
1989, 111, 587620.
2 M. R. Farlow and J. L. Cummings, Am. J. Med., 2007, 120,
388–397; Y. Pan, X. Xianghua and X. Wang, Br. J. Pharmacol.,
2003, 140, 907–912; R. J. Polinsky, Clin. Ther., 1998, 20, 634–647.
3 M. Emre, CNS Drugs, 2006, 20, 748–750.
17 M. Hohne, K. Robins and U. T. Bornscheuer, Adv. Synth. Catal.,
¨
2008, 350, 807–812.
18 Y. Simeo and K. Faber, Tetrahedron: Asymmetry, 2006, 17,
´
402–409; M.-X. Wang, C.-S. Liu, J.-S. Li and O. Meth-Cohn,
Tetrahedron Lett., 2000, 41, 8549–8552.
19 G. Carrea and S. Riva, Angew. Chem., Int. Ed., 2000, 39,
2226–2254.
4 A. Gaitonde, M. Mangle, and S. Pawar, PCT, WO 2005/061446,
2005; A. Gaitonde and M. Mangle, Chem. Abstr., 2005, 143, 77963;
ꢀc
This journal is The Royal Society of Chemistry 2010
5502 | Chem. Commun., 2010, 46, 5500–5502