Induced axial chirality in biocatalytic asymmetric ketone reduction

Article abstract of DOI:10.1021/ja3092517

Catalytic asymmetric reduction of prochiral ketones of type 4-alkylidene cyclohexanone with formation of the corresponding axially chiral R-configurated alcohols (up to 99% ee) was achieved using alcohol dehydrogenases, whereas chiral transition-metal catalysts fail. Reversal of enantioselectivity proved to be possible by directed evolution based on saturation mutagenesis (up to 98% ee (S)). Utilization of ketone with a vinyl bromide moiety allows respective R- and S-alcohols to be exploited as key compounds in Pd-catalyzed cascade reactions.

Full text of DOI:10.1021/ja3092517

Communication
pubs.acs.org/JACS
Induced Axial Chirality in Biocatalytic Asymmetric Ketone Reduction
Ruben Agudo,†,‡ Gheorghe-Doru Roiban,†,‡ and Manfred T. Reetz*,†,‡
́
†Department of Chemistry, Philipps-Universitat Marburg, Hans-Meerwein Str., 35032 Marburg, Germany
̈
‡Max-Planck-Institut fur Kohlenforschung, Kaiser-Wilhelm-Platz 1, 45470 Mulheim an der Ruhr, Germany
̈
̈
S
* Supporting Information
members,10 as catalysts in ketones 1a−f reduction. Stereo-
selective reduction of ketone 1a appeared particularly attractive
because R- and S-2a could subsequently serve as pivotal
compounds in transition-metal catalyzed C−C bond-forming
reactions with formation of a variety of structurally different
axially chiral compounds. Since companies do not reveal the
ADHs identity (i.e., some kit members may be identical), in all
cases we used one standard protocol (glucose dehydrogenase/
glucose as the cofactor regeneration system with either NAD+
or NADP+, pH 7, 30 °C, 16 h; see Supporting Information, SI).
Typical data are summarized in Table 1 (full data in Table S1),
including the most promising results. High enantioselectivity
amounting to ≥96% ee resulted in the reduction of all ketones
except 1b (maximally 86% ee). In all high enantioselectivity
cases, except alcohol 2c, R-selectivity was observed, showing
that the preferred direction of hydride attack by NADPH
occurs consistently from the Re side of the carbonyl function.
Enantioselectivity reversal proved to be possible in some cases
but not with high ee values (SI).
ABSTRACT: Catalytic asymmetric reduction of prochiral
ketones of type 4-alkylidene cyclohexanone with formation
of the corresponding axially chiral R-configurated alcohols
(up to 99% ee) was achieved using alcohol dehydro-
genases, whereas chiral transition-metal catalysts fail.
Reversal of enantioselectivity proved to be possible by
directed evolution based on saturation mutagenesis (up to
98% ee (S)). Utilization of ketone with a vinyl bromide
moiety allows respective R- and S-alcohols to be exploited
as key compounds in Pd-catalyzed cascade reactions.
ype 2 axially chiral molecules1 characterized by a
T
substituted 4-alkylidene cyclohexane structure are of
interest in circular dichroism exciton chirality studies2,3 and
as precursors to enantiomerically pure liquid crystals.4 Several
synthetic routes to this class of chiral compounds have been
described, all involving traditional antipode separation or
utilization of stoichiometric amounts of nonracemic chiral
reagents or auxiliaries.3,5 An attractive synthetic approach is the
catalytic enantioselective reduction of corresponding prochiral
ketones 1 using chiral transition-metal catalysts. Unfortunately,
using one of the most efficient chiral Ru-based catalysts
((TsDPEN)(p-cymene)RuCl2)6 in ketone 1b reduction,
formation of alcohol 2b as a racemate was observed. This is
not surprising because high stereoselectivity in ketone
reductions can be achieved only when α- and α′- positions
flanking the carbonyl function differ sterically to a notable
degree.6,7 So we turned to biocatalytic reduction using alcohol
dehydrogenases (ADHs).8
Absolute configuration assignment of R/S-2d was made by
comparison with the known absolute configuration of S-2d
determined by Walborsky.3 For the other axially chiral
compounds, chemical correlation was performed, e.g., by Pd-
catalyzed carbonylation and Suzuki coupling using the vinyl
bromide 2a (SI).
We also tested Thermoethanolicus brockii (TbSADH*)
ADH,11,12 which is of industrial interest because it is unusually
thermostable, tolerates organic solvents, and exhibits high
activity and stereoselectivity toward many structurally different
prochiral ketones. It is a zinc-dependent ADH in which the
substrate carbonyl moiety binds to the metal, thereby
positioning and activating the carbonyl function for hydride
attack to occur from NADPH. Wild-type TbSADH generally
obeys Prelog’s rule in the normal case of ketones leading to
central chirality,13,14 although anti-Prelog products were
observed in some cases, leading Keinan to propose a model
based on small and large hydrophobic pockets assessment.13
Subsequently, several crystal structures were determined,15 one
a binary complex with cofactor NADP+ bound to the zinc15a
and a different one with sec-butanol as guest in the binding
pocket bound to the zinc.15b Unfortunately for ketone 1a, this
robust ADH is only moderately R-selective (66% ee).
To obtain highly R- and S-selective mutants of the robust
TbSADH*, we turned to directed evolution16 using the
combinatorial active-site saturation test (CAST)16g which
In exploratory experiments, we first tested two commercially
Received: September 18, 2012
Published: October 17, 2012
available ADH kits, one with 35 enzymes9 and the other 12
© 2012 American Chemical Society
1665
dx.doi.org/10.1021/ja3092517 | J. Am. Chem. Soc. 2013, 135, 1665−1668

Journal of the American Chemical Society
Communication
a b
,
Table 1. Reduction of Model Ketones 1a−f with Commercial ADHs
substrate
d
1a→2a
1b→2b
1c→2c
1d→2d
1e→2e
1f→2f
conv.
(%)
conv.
(%)
conv.
(%)
conv.
(%)
conv.
(%)
conv.
(%)
c
entry
ADH
cofactor
% ee
% ee
% ee
% ee
% ee
% ee
1
2
A1
A2
NADP+
NADP+
NAD+
NAD+
NAD+
NAD+
NAD+
NAD+
NAD+
NAD+
NAD+
NAD+
NAD+
NAD+
NADP+
NAD+
NADP+
NAD+
NAD+
NAD+
NADP+
39
46(S)
>99
>99
>99
>99
>99
>99
>99
60
47(S)
23(S)
1(S)
>99
>99
>99
87
97(S)
96(S)
97(R)
98(R)
98(R)
98(R)
98(R)
96(R)
97(R)
96(R)
93(R)
93(R)
87(R)
84(R)
96(S)
96(R)
96(S)
89(R)
95(R)
97(R)
94(R)
>99
>99
>99
>99
>99
>99
>99
29
60(S)
58(S)
98(R)
95(R)
95(R)
95(R)
97(R)
92(R)
97(R)
96(R)
>99
>99
>99
>99
>99
>99
>99
82
27(S)
27(S)
99(R)
>99(R)
>99(R)
>99(R)
99(R)
98(R)
99(R)
>99(R)
90(R)
95(R)
96(R)
96(R)
27(S)
98(R)
27(S)
97(R)
96(R)
98(R)
97(R)
>99
>99
>99
>99
>99
>99
>99
>99
>99
>99
>99
>99
>99
>99
>99
>99
>99
>99
>99
>99
>99
39(R)
38(R)
96(R)
99(R)
99(R)
99(R)
96(R)
>99(R)
96(R)
>99(R)
99(R)
96(R)
99(R)
98(R)
40 (R)
95(R)
40 (R)
97(R)
93(R)
92(R)
92(R)
e
≤20
>99
>99
>99
>99
>99
54
nd
3
A7
64(R)
96(R)
96(R)
95(R)
74(R)
91(R)
4
A8
62(R)
73(R)
67(R)
27(R)
68(R)
2(R)
5
A10
>99
97
6
A13
7
A14
96
8
A15
81
e
9
A18
≤20
>99
41
nd
>99
>99
≤20
>99
≤20
>99
>99
>99
>99
90
97
>99
>99
≤20
>99
≤20
>99
>99
>99
>99
>99
>99
>99
>99
>99
>99
27
10
11
12
13
14
15
16
17
18
19
20
21
A21
93(R)
73(R)
61(R)
80(R)
87(R)
81(S)
65(R)
82(S)
91(R)
83(R)
11(R)
51(R)
74(R)
95
e
e
A22
nd
60
nd
A25
>99
41
0
88
97(R)
>99
61
e
e
A29
nd
67
nd
A30
>99
>99
>99
>99
>99
>99
>99
>99
72(R)
42(S)
11(R)
46(S)
86(R)
5(R)
0
88
91(R)
59(S)
98(R)
59(S)
95(R)
96(R)
94(R)
97(R)
>99
>99
>99
>99
>99
>99
>99
>99
A32
>99
94
A33
A34
97
1.1.020
1.1.030
1.1.210
1.1.260
96
>99
>99
94
>99
91
>99
50(R)
a
b
Average of at least two independent measurements. Deviation in values obtained did not exceed 10% of average ee value shown. Entries 1−13 are
ADHs from X-zymes kit;9 entries 14−17 are ADHs from Evocatal kit.10 Experimental conditions are detailed in SI. Cofactor used upon
recommendation of ADH supplier. In 1b→2b, none of commercial ADHs led to ee values >86%. Not determined (nd) due to low conversion.
c
d
e
constitutes a systematization of the first example of random-
ization at the active site of an enzyme to control stereo-
selectivity.17 Accordingly, sites composed of one or more amino
acids positioned around the binding pocket are subjected to
saturation mutagenesis, if necessary in an iterative manner.16g
Guided by the crystal structure of TbSADH15 and site-directed
mutagenesis, reported by Phillips,18 six amino acid positions
were chosen for saturation mutagenesis: S39, A85, I86, W110,
Y267, and C295 (Figure 1), using ketone 1a as the model
substrate.
To minimize screening effort (oversampling) while ensuring
full library coverage,16g we generated six CAST libraries at the
defined residues using a highly reduced amino acid alphabet
defined by RNG codon degeneracy (R: adenine/guanine; N:
adenine/cytosine/guanine/thymine; G: guanine). It encodes
eight amino acids of different steric “size” irrespective of
Figure 1. TbSADH active site model harboring compound 1a
(yellow) as a rough guide for choosing six CAST sites. 1a coordinates
were generated using MAESTRO program and subsequently manually
fitted in the active site of TbSADH based on PDB 1BXZ.15b NADP+
(green) and zinc (gray) were modeled by superimposition using
PyMOL software (SI). Randomization sites Y267, S39, and W110
(blue) align the large part of the binding pocket, while A85, I86, and
C295 (red) are next to the small part.
electronic properties, viz., Ala, Arg, Glu, Gly, Lys, Met, Thr, and
Val, and requires oversampling of only 31 transformants for
98% library coverage.16g,19 This choice was inspired by Prelog’s
diamond lattice mnemonic device for predicting the stereo-
selectivity of ADH-mediated ketone reduction following
consideration of relative size of the groups flanking the
carbonyl moiety;20 a useful guide that was also employed by
Phillips in interpreting certain mutations introduced in
TbSADH by site-directed mutagenesis.18 It should be noted
that in the present case of ketone 1, the steric properties of the
respective first three C atoms flanking the carbonyl function are
identical. Therefore, application of Prelog’s rule goes beyond
conventional considerations. The best results originated from
libraries generated at single-residue sites 85, 86, 110, and 295
(Table 2). Remarkably, both R- and S-selective mutants were
identified having ee values >97% in both stereochemical
regimes. Thus, iterative rounds of CASTing were not necessary
for obtaining useful mutants. Interestingly, mutants I86A and
W110A (Table 2, entries 3 and 8, respectively) were prepared
previously by site-specific mutagenesis as catalysts in asym-
metric reduction of structurally completely different ketones.14f
RNG-limited randomization libraries generated by at positions
39 and 267 failed to harbor mutants with notable
1666
dx.doi.org/10.1021/ja3092517 | J. Am. Chem. Soc. 2013, 135, 1665−1668

Journal of the American Chemical Society
Communication
Table 2. Performance of Best TbSADH* Mutants Specifically Evolved for 1a as Catalysts in Asymmetric Reduction of Ketones
a b
,
1a−f
substrate
1a→2a
conv.
1b→2b
1c→2c
1d→2d
1e→2e
1f→2f
conv.
conv.
(%)
conv.
conv.
(%)
conv.
c
entry site
mutation
(%)
% ee
(%)
% ee
% ee
(%)
% ee
% ee
(%)
% ee
1
2
-
TbSADH*
A85V
≥95
≥95
≥95
≥95
≥95
≥95
≥95
≥99
≥99
≥99
≥99
16
66(R)
95(R)
98(S)
98(S)
95(S)
92(S)
92(S)
82(R)
91(R)
97(R)
97(R)
84(R)
92
91(R)
96(R)
74(S)
82(S)
84(S)
25(S)
61(S)
93(R)
92(R)
95(R)
92(R)
79(R)
38
28
77(R)
63(R)
65(R)
66(R)
88(R)
73(R)
74(R)
99(R)
89(R)
99(R)
97(R)
88(R)
26
18
89(R)
94(R)
75(R)
24(S)
58(R)
84(R)
40(R)
98(R)
91(R)
98(R)
98(R)
65(R)
33
22
87(R)
91(R)
85(R)
8(R)
42
26
92(R)
93(R)
95(R)
93(R)
92(R)
90(R)
93(R)
99(R)
91(R)
99(R)
99(R)
72(R)
85
90
3
86
I86A
≥99
≥99
≥99
≥99
89
33
13
18
29
4
86
I86G
30
24
25
24
5
86
I86E
22
18
30
89(R)
86(R)
57(R)
99(R)
92(R)
99(R)
99(R)
31
6
86
I86M
27
35
52
46
7
86
I86T
15
21
23
27
8
110
110
110
110
295
W110A
W110E
W110M
W110T
C295E
≥99
≥99
≥99
≥99
27
≥99
≥99
≥99
≥99
10
≥99
≥99
≥99
≥99
8
≥99
≥99
≥99
≥99
10
≥99
≥99
≥99
≥99
14
9
10
11
12
64(R)
a
b
Average of two independent measurements. Deviation in values obtained did not excess 3% of average ee value shown. Only hits from plate
c
screening confirmed in medium scale reactions (60 mg of 1a, 63 mM final concn) are presented here. See SI for exptl details. ≤5% side products in
all cases.
enantioselectivity (ee ≥ 80%). However, noteworthy is mutant
S39T which induces enantioselectivity reversal (38% ee in favor
of 2a).
enantioselectivity (≥99% ee); absolute configuration is
presently unknown (Table S3). While testing ketones 5a and
5b reduction, the best mutant, W110T, showed respectable
enantioselectivity, 79 and 62% ee, respectively (see SI for other
mutants performance). This mutant is therefore a good starting
point for future genetic optimization.
Reaction 1a→2a was upscaled employing two different
mutants: R-selective variant W110T using 500 mg (2.64 mmol)
of 1a at 67 mM concn and S-selective variant I86A using 630
mg (3.33 mmol) of 1a at 67 mM concn. Isolated yields were
81% and 84% for (R)- and (S)-2a, respectively (SI), which
underscores practical utility of the mutants. Indeed, with access
to both enantiomers, many different kinds of transition-metal
catalyzed transformations can be envisioned.
We then tested the best hits specifically evolved for ketone
1a as catalysts in asymmetric reduction of the other substrates
1b−f. Table 2 shows mostly excellent enantioselectivity and
high conversion to R-configurated alcohols are possible, but
asymmetric induction reversal with formation of respective S-
alcohols was not achieved.
Most results summarized in Table 2 are in accord with
Prelog’s diamond lattice model,20 extended in the present
system. All hits arising from saturation mutagenesis at position
I86 using RNG codon degeneracy showed a preference for S-
2a, although the side chains of some of the introduced amino
acids are not substantially smaller than that of isoleucine (e.g.,
glutamic acid or methionine; Table 2, entries 5 and 6,
respectively). To study the residue 86 influence on reaction
1a→2a stereoselectivity, we prepared by site-directed muta-
genesis (QuikChange)21 the rest of the theoretically possible
mutants. Surprisingly, we observed S-selectivity in all new
mutants regardless of size or electronic nature of respective
amino acid side chains, e.g., ≥98% ee for I86D, I86C, I86Y, and
I86P (see Table S2 for mutants with lower S-selectivity).
Finally, we wanted to see if asymmetric reduction of
analogous four- and especially eight-membered cyclic ketones
is also possible, so ketones 3 and 5a−b were prepared and
subjected to reduction using some of the previously obtained
TbSADH* mutants without performing additional muta-
genesis. TbSADH* itself essentially does not accept these
substrates, which means that enantioselectivity could not be
determined reliably. For ketone 3, all mutants characterized by
position 110 point mutations led to high activity and complete
In addition to these substrates, we tested the performance of
some mutants as catalysts in reduction of a “normal” ketone,
specifically acetophenone leading to phenyl ethanol. WT
TbSADH* does not accept this compound. Mutant I86N
proved to be active and R-selective (99% ee), while mutant
W110M led to inversion of enantioselectivity (98% ee (S)),
(Table S5).
In conclusion, ketones of type 1a−f can be reduced
enantioselectively with formation of axially chiral alcohols
2a−f, being possible with appropriate alcohol dehydrogenases
but not with transition-metal catalysts. Various ADHs from two
commercial kits provide respective R-alcohols with essentially
complete stereoselectivity (≥98% ee), while robust T. brockii
(TbSADH*)11,12 proved to be only moderately R-selective
(66% ee). To solve this problem, directed evolution was
applied with generation of highly R- and S-selective TbSADH*
1667
dx.doi.org/10.1021/ja3092517 | J. Am. Chem. Soc. 2013, 135, 1665−1668

Journal of the American Chemical Society
Communication
Chem. Soc. Rev. 2009, 38, 2282. (e) Oxazaborolidine catalysts for
asymmetric ketone reductions: Corey, E. J.; Helal, C. J. Angew. Chem.,
Int. Ed. 1998, 37, 1986.
mutants. In the alcohol 2a case with a vinyl bromide moiety, a
variety of transition-metal catalyzed C−C bond forming
substitution reactions are possible, as demonstrated in many
Pd-catalyzed carbonylation and Suzuki coupling (SI) cases.
Moreover, alcohol function (or the respective tosylate)
reactions of can be envisioned with stereospecific introduction
of fluorine or alkyl groups using standard methods, which
means that a wide variety of structurally different axially chiral
compounds is now readily accessible in R- or S-form. Initial
work using four- and eight-membered cyclic ketones 3 and 5
shows that asymmetric reduction is likewise possible. Future
efforts will focus on biophysical characterization of key mutants
with emphasis on QM/MM-based elucidation of the stereo-
selectivity source on a molecular level.
(8) Alcohol dehydrogenases (ADHs) as catalysts in asymmetric
ketone reduction reviews: (a) Groger, H.; Hummel, W.; Borchert, S.;
̈
Kraußer, M. In Enzyme Catalysis in Organic Synthesis, 3rd ed.; Drauz,
K., Groger, H., May, O., Eds.; Wiley-VCH: Weinheim, 2012; pp
̈
1035−1110. (b) Gotz, K.; Hilterhaus, L.; Liese, A. In Enzyme Catalysis
̈
in Organic Synthesis, 3rd ed.; Drauz, K., Groger, H., May, O., Eds.;
Wiley-VCH: Weinheim, 2012; pp 1205−1223. (c) García-Urdiales, E.;
Alfonso, I.; Gotor, V. Chem. Rev. 2005, 105, 313. (d) Woodley, J. M.
̈
Trends Biotechnol. 2008, 26, 321. (e) Goldberg, K.; Schroer, K.; Lutz,
̈
S.; Liese, A. Appl. Microbiol. Biotechnol. 2007, 76, 237. (f) Moore, J. C.;
Pollard, D. J.; Kosjek, B.; Devine, P. N. Acc. Chem. Res. 2007, 40, 1412.
(g) Matsuda, T.; Yamanaka, R.; Nakamura, K. Tetrahedron: Asymmetry
2009, 20, 513.
(9) X-Zyme kit; X-Zyme GmbH: Dusseldorf, 2011.
̈
ASSOCIATED CONTENT
* Supporting Information
Experimental details and characterization data. This material is
available free of charge via the Internet at http://pubs.acs.org.
■
(10) Evocatal kit; Evocatal GmbH: Dusseldorf, 2011.
̈
S
(11) (a) Lamed, R. J.; Keinan, E.; Zeikus, J. G. Enzyme Microb.
Technol. 1981, 3, 144. (b) Burdette, D. S.; Vieille, C.; Zeikus, J. G.
Biochem. J. 1996, 316, 115. (c) Burdette, D. S.; Secundo, F.; Phillips, R.
S.; Dong, J.; Scott, R. A.; Zeikus, J. G. Biochem. J. 1997, 326, 717.
(12) TbSADH was cloned from plasmid pADHB1M1-kan, provided
by Prof. J. G. Zeikus (Michigan State University). This plasmid
encodes the secondary ADH from Thermoanaerobacter ethanolicus
(TeSADH). However, it has been shown that TeSADH is identical to
TbSADH (unpublished data referred in ref 14f). We discovered that
the sequence of TbSADH from pADHB1M1-kan, stored in our
laboratory, actually contains an additional mutation (encoding A186D
amino acid substitution), leading us to designate it as TbSADH*.
Control experiments show that additional point mutation has no effect
on our study results (see SI).
AUTHOR INFORMATION
Corresponding Author
reetz@mpi-muelheim.mpg.de
Notes
■
The authors declare no competing financial interest.
ACKNOWLEDGMENTS
■
Financial support by the Max-Planck-Society and the Arthur C.
Cope Foundation is gratefully acknowledged. We thank Dr. Z.-
G. Zhang for preliminary experiments, Stephanie Dehn for GC
measurements, and Heike Hinrichs and Alfred Deege for
HPLC analyses. Thanks also to J. G. Zeikus for providing the
(13) Keinan, E.; Hafeli, E. K.; Seth, K. K.; Lamed, R. J. Am. Chem.
Soc. 1986, 108, 162.
(14) (a) Pham, V. T.; Phillips, R. S.; Ljungdahl, L. G. J. Am. Chem.
Soc. 1989, 111, 1935. (b) Zheng, C.; Pham, V. T.; Philipps, R. S. Catal.
Today 1994, 22, 607. (c) Secundo, F.; Philipps, R. S. Enzyme Microb.
Technol. 1996, 19, 487. (d) Burdette, D. S.; Tchernajencko, V.; Zeikus,
J. G. Enzyme Microb. Technol. 2000, 27, 11. (e) Heiss, C.; Philipps, R.
S. J. Chem. Soc., Perkin Trans. 1 2000, 2821. (f) Musa, M. M.; Philipps,
R. S. Catal. Sci. Technol. 2011, 1, 1311.
(15) (a) Korkhin, Y.; Kalb(Gilboa), A. J.; Peretz, M.; Bogin, O.;
Burstein, Y.; Frolow, F. J. Mol. Biol. 1998, 278, 967. (b) Li, C.;
Heatwole, J.; Soelaiman, S.; Shoham, M. Proteins: Struct. Funct. Genet.
1999, 37, 619.
(16) Recent directed evolution reviews: (a) Turner, N. J. Nat. Chem.
Biol. 2009, 5, 567. (b) Lutz, S.; Bornscheuer, U. T. Protein Engineering
Handbook; Wiley-VCH: Weinheim, 2009; Vol. 1−2. (c) Jackel, C.;
Kast, P.; Hilvert, D. Annu. Rev. Biophys. Biomol. Struct. 2008, 37, 153.
(d) Romero, P. A.; Arnold, F. H. Nat. Rev. Mol. Cell Biol. 2009, 10,
866. (e) Shivange, A. V.; Marienhagen, J.; Mundhada, H.; Schenk, A.;
Schwaneberg, U. Curr. Opin. Chem. Biol. 2009, 13, 19. (f) Goldsmith,
M.; Tawfik, D. S. Curr. Opin. Struct. Biol. 2012, 22, 406. (g) Reetz, M.
T. Angew. Chem., Int. Ed. 2011, 50, 138.
pADHB1M1-kan plasmid11 and to Evocatal (Dusseldorf/
̈
Germany) for a kit containing 12 ADHs.
REFERENCES
■
(1) Eliel, E. L.; Wilen, S. H. Stereochemistry of Organic Compounds;
Wiley-VCH: New York, 1994.
(2) (a) Duraisamy, M.; Walborsky, H. M. J. Am. Chem. Soc. 1983,
105, 3264.
For CD overviews and recent contributions, see
(b) Berova, N.; Nakanishi, K.; Woody, R. W. Circular Dichroism:
Principles and Applications; 2nd ed., Wiley-VCH: New York, 2000.
(c) You, L.; Pescitelli, G.; Anslyn, E. V.; Di Bari, L. J. Am. Chem. Soc.
2012, 134, 7117. (d) Taniguchi, T.; Monde, K. J. Am. Chem. Soc. 2012,
134, 3695. (e) Hosoi, S.; Kamiya, M.; Ohta, T. Org. Lett. 2001, 3,
3659.
(3) Walborsky, H. M.; Gawronska, K.; Gawronski, J. K. J. Am. Chem.
Soc. 1987, 109, 6719.
́
(4) (a) Solladie, G.; Zimmermann, G. Angew. Chem., Int. Ed. Engl.
1984, 23, 348. (b) Kitzerow, H.; Bahr, C. Chirality in Liquid Crystals;
Springer: New York, 2001. (c) Khoo, I.-C. Liquid Crystals; 2nd ed.,
Wiley-VCH: New York, 2007. (d) Eelkema, R.; Feringa, B. L. Org.
Biomol. Chem. 2006, 4, 3729.
(5) (a) Denmark, S. E.; Chen, C.-T. J. Am. Chem. Soc. 1992, 114,
10674. (b) Denmark, S. E.; Chen, C.-T. Heteroatom Chem. 1995, 6,
133. (c) Tanak, K.; Furuta, T.; Fuji, K. In Modern Carbonyl Olefination;
Takeda, T., Ed.; Wiley-VCH: Weinheim, 2004, pp 286−342.
(6) (a) Noyori, R. Angew. Chem., Int. Ed. 2002, 41, 2008.
(b) Sandoval, C. A.; Li, Y.; Ding, K.; Noyori, R. Chem. Asian J.
2008, 3, 1801.
(7) Transition-metal catalyzed asymmetric ketone reduction reviews:
(a) Blaser, H.-U.; Malan, C.; Pugin, B.; Spindler, F.; Steiner, H.;
Studer, M. Adv. Synth. Catal. 2003, 345, 103. (b) Clapham, S. E.;
Hadzovic, A.; Morris, R. H. Coord. Chem. Rev. 2004, 248, 2201.
(c) Tang, W.; Zhang, X. Chem. Rev. 2003, 103, 3029. (d) Morris, R. H.
̈
(17) Reetz, M. T.; Wilensek, S.; Zha, D.; Jaeger, K.-E. Angew. Chem.,
Int. Ed. 2001, 40, 3589.
(18) Musa, M. M.; Lott, N.; Laivenieks, M.; Watanabe, L.; Vielle, C.;
Philipps, R. S. ChemCatChem 2009, 1, 89.
(19) CASTER computer aid can be used for designing saturation
mutagenesis libraries, including calculation of degree of oversampling;
available free of charge: http://www.kofo.mpg.de/de/forschung/
organische-synthese).
(20) Prelog, V. Pure Appl. Chem. 1964, 9, 119.
(21) Hogrefe, H. H.; Cline, J.; Youngblood, G. L.; Allen, R. M.
BioTechniques 2002, 33, 1158.
1668
dx.doi.org/10.1021/ja3092517 | J. Am. Chem. Soc. 2013, 135, 1665−1668