Angewandte
Chemie
DOI: 10.1002/anie.200705046
Biocatalysis
Rapid Determination of Both the Activity and Enantioselectivity of
Ketoreductases**
Matthew D. Truppo,* Franck Escalettes, and Nicholas J. Turner*
The application of biocatalysis for the production of key
chiral intermediates in the pharmaceutical and fine chemicals
industries is rapidly growing.[1,2] As the number of biocatalysts
increases, screening these enzymes for activity and enantio-
selectivity against target molecules becomes a major bottle-
neck in the process-development timeline.[3] The current state
of the art for resolving enantiomers of chiral products
involves the use of chiral stationary phases in HPLC or GC
format.[4,5] While these methods are very precise, assays
generally require sample preparation and run times of
approximately 20 minutes per sample.
Scheme 1. Screening for KREDs using an alcohol oxidase.
Efforts have been made to devise rapid assays with a
variety of technologies, including fluorescence in reaction
microarrays,[6] infrared thermography,[7] mass spectrometry,[8]
capillary electrophoresis,[9] circular dichroism,[10] and pH
indicators.[11] One drawback of many of these techniques is
that they require reference reactions involving pure enantio-
mers of the compound of interest to calibrate the system.
Additionally, it would be beneficial to obtain not only
selectivity information from the screen, but activity informa-
tion as well.
Herein we report two new high-throughput, 96-well
microtiterplate-format screening methods that are based on
employing an enantioselective oxidase enzyme for determin-
ing both the rate and the enantioselectivity of asymmetric
ketone reduction. Ketoreductases (KREDs) were selected as
ideal targets for initial development in view of their increasing
importance and application in the asymmetric reduction of
ketones.[12] Although a large number of KREDs (more than
100) are available, difficulty remains in predicting which
particular enzyme will provide the best enantioselectivity
simply based on substrate structure. Minor substituent
changes can lead to different “best hits”, thus highlighting
the need for extensive screening.[13]
alcohols (R)-and ( S)-2, the ratio of which can be determined
by addition of an alcohol oxidase of known enantioselectivity.
The alcohol oxidase catalyzes oxidation of (R/S)-2 back to
ketone 1 with concomitant production of H2O2, which can be
detected spectrophotometrically. The alcohol oxidase
employed herein is derived from galactose oxidase and has
been evolved to have high activity and R enantioselectivity
towards substituted 1-phenylethanol analogues.[14]
Using acetophenone (R1 = Ph, R2 = Me) as substrate, the
rate of ketone reduction by a KRED was initially determined
by monitoring the consumption of NADPH through the
decrease in absorbance at 340 nm. A limiting quantity of
NADPH (0.15 mm) was present in the assay mixture along
with an excess (2 mm) of ketone 1 substrate to ensure rapid
reduction and hence a short assay time. Upon complete
consumption of NADPH, the R-selective alcohol oxidase was
added, resulting in oxidation of any (R)-2 present in the
reaction mixture back to the ketone 1 with concomitant
production of one mole of H2O2 for every mole of alcohol
molecule oxidized. The precise amount of hydrogen peroxide,
and hence (R)-alcohol, produced in the reaction was moni-
tored at 400 nm through the colorimetric HRP/ABTS
(HRP = horse radish peroxidase, ABTS = 2,2’-azino-di(3-eth-
ylbenzthiazoline-6-sulfonic acid) assay.[15] Ensuring that
NADPH is the limiting reagent is essential, as the cofactor
is known to be oxidized by HRP/H2O2.[16]
Scheme 1 outlines the basis of an assay for detecting the
activity and enantioselectivity of KREDs. Reduction of
ketone 1 by the NADPH-dependent KRED generates
[*] M. D. Truppo, Dr. F. Escalettes, Prof. N. J. Turner
School of Chemistry, University of Manchester
Manchester Interdisciplinary Biocentre
131 Princess Street, Manchester M17DN (UK)
Fax: (+44)161-275-1311
Monitoring this assay at two different wavelengths (340
and 400 nm) generates all of the required data, namely 1) the
rate of reduction of the ketone, 2) the total amount (R + S) of
alcohol produced [Eq. (1)], 3) the amount of (R)-and ( S)-
alcohol produced [Eqs. (2) and (3)], and 4) the enantioselec-
tivity of the KRED, expressed in terms of the enantiomeric
excess (ee) of alcohol produced [Eq. (4)].
E-mail: matthew_truppo@merck.com
[**] This work was supportedby the Engineering andPhysical Sciences
Research Council (F.E.) andMerck Doctoral Study Program
(M.D.T.)
½ðRÞ-2 þ ½ðSÞ-2 ¼ ½NADPHoꢀ½NADPH ðat 340 nmÞ
ð1Þ
ð2Þ
f
Supporting information for this article (experimental details) is
the author.
½ðRÞ-2 ¼ ½ABTS cationð2Þꢀ1 ðat 400 nmÞ
Angew. Chem. Int. Ed. 2008, 47, 2639 –2641
ꢀ 2008 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
2639