Highly enantioselective mutant carbonyl reductases created via structure-based site-saturation mutagenesis

Article abstract of DOI:10.1021/jo101541n

A carbonyl reductase from Sporobolomyces salmonicolor reduced para-substituted acetophenones with low enantioselectivity. Enzyme-substrate docking studies revealed that residues M242 and Q245 were in close proximity to the para-substituent of acetophenones in the substrate binding site. Site-saturation mutagenesis of M242 or Q245, and double mutation of M242 and Q245 were performed in order to enhance the enzyme's enantioselectivity toward the reduction of para-substituted acetophenones. Three Q245 mutants were obtained, which inverted the enantiopreference of product alcohols from (R)- to (S)-configuration with high ee values (Org. Lett. 2008, 10, 525-528). Four M242 mutant enzymes also showed greater preference for the formation of (S)-enantiomeric alcohols than the wild-type enzyme, but to a much less extent than Q245 mutants. M242/Q245 double variations not only greatly affect the enantiomeric purity of the product alcohols, but also invert the enantiopreference, demonstrating that these residues play a critical role in determining the enantioselectivity of these ketone reductions. The kinetic parameters of these mutant enzymes indicated that residues 242 and 245 also exert an effect on the catalytic activity of this carbonyl reductase. Highly enantioselective mutant carbonyl reductases were created by site-saturation mutagenesis, among which the one bearing double mutations, M242L/Q245P, showed the highest enantioselectivity that catalyzed the reduction of the tested para-substituted acetophenones to give (S)-enantiomeric products in ≥99% ee with only one exception of p-fluoroacetophenone (92% ee).

Full text of DOI:10.1021/jo101541n

pubs.acs.org/joc
Highly Enantioselective Mutant Carbonyl Reductases Created via
Structure-Based Site-Saturation Mutagenesis
Hongmei Li,‡ Yan Yang,‡ Dunming Zhu,*,†,‡ Ling Hua,‡,§ and
Katherine Kantardjieff^
†State Engineering Laboratory for Industrial Enzymes, Tianjin Institute of Industrial Biotechnology,
Chinese Academy of Sciences, Tianjin 300308, China, ‡Department of Chemistry, Southern Methodist
University, Dallas, Texas 75275, United States, §China Research Center, Genencor, A Danisco Division,
Shanghai 200335, China, and ^Department of Chemistry, California State Polytechnic University, Pomona,
California 91768, United States
zhu_dm@tib.cas.cn
Received August 6, 2010
A carbonyl reductase from Sporobolomyces salmonicolor reduced para-substituted acetophenones with low
enantioselectivity. Enzyme-substrate docking studies revealed that residues M242 and Q245 were in close
proximity to the para-substituent of acetophenones in the substrate binding site. Site-saturation mutagen-
esis of M242 or Q245, and double mutation of M242 and Q245 were performed in order to enhance the
enzyme’s enantioselectivity toward the reduction of para-substituted acetophenones. Three Q245 mutants
were obtained, which inverted the enantiopreference of product alcohols from (R)- to (S)-configuration
with high ee values (Org. Lett. 2008, 10, 525-528). Four M242 mutant enzymes also showed greater
preference for the formation of (S)-enantiomeric alcohols than the wild-type enzyme, but to a much less
extent than Q245 mutants. M242/Q245 double variations not only greatly affect the enantiomeric purity of
the product alcohols, but also invert the enantiopreference, demonstrating that these residues play a critical
role in determining the enantioselectivity of these ketone reductions. The kinetic parameters of these mutant
enzymes indicated that residues 242 and 245 also exert an effect on the catalytic activity of this carbonyl
reductase. Highly enantioselective mutant carbonyl reductases were created by site-saturation mutagenesis,
among which the one bearing double mutations, M242L/Q245P, showed the highest enantioselectivity that
catalyzed the reduction of the tested para-substituted acetophenones to give (S)-enantiomeric products in
g99% ee with only one exception of p-fluoroacetophenone (92% ee).
Introduction
other approaches.1-5 The information about the substrate
specificity and enantioselectivity of enzymes is very valuable
Enantiomerically pure chiral alcohols are important inter-
mediates in the fine chemical and pharmaceutical industries.
In recent years, biocatalytic reduction of ketones has in-
creasingly grown as a general method of choice for the
synthesis of chiral alcohols and efficiently complements
(1) Moore, J. C.; Pollard, D. J.; Kosjek, B.; Devine, P. N. Acc. Chem. Res.
2007, 40, 1412.
(2) De Wildeman, S. M. A.; Sonke, T.; Schoemaker, H. E.; May, O. Acc.
Chem. Res. 2007, 40, 1260.
(3) Kroutil, W.; Mang, H.; Edegger, K.; Faber, K. Curr. Opin. Chem.
Biol. 2004, 8, 120.
*To whom correspondence should be addressed. Tel: (þ86)-22-84861962.
Fax: (þ86)-22-84861996.
(4) Huisman, G. W.; Liang, J.; Krebber, A. Curr. Opin. Chem. Biol. 2010,
14, 122.
DOI: 10.1021/jo101541n
r
Published on Web 10/22/2010
J. Org. Chem. 2010, 75, 7559–7564 7559
2010 American Chemical Society

Li et al.
JOCArticle
and instructive in guiding the selection of appropriate bio-
catalysts for a specific transformation.6-9 Correlating the
substrate profile of an enzyme with its sequential and
structural information not only provides valuable insight
into the understanding of how enzymes control activity and
enantioselectivity but also facilitates our efforts to (semi)
rationally design novel enzymes with desired substrate spec-
ificity and enantioselectivity.10-13 Holding these in mind, we
have studied substrate specificity and enantioselectivity of
several carbonyl reductases from different origins that cat-
alyze reduction of ketones of diverse structures. To our
delight, these carbonyl reductases are capable of converting
aryl/alkyl ketones and R-/β-ketoesters to the corresponding
chiral alcohols with excellent enantiomeric purity.14-20 In
the course of our studies on a carbonyl reductase from
Sporobolomyces salmonicolor (SSCR), it has been found that
SSCR showed an unusually broad substrate range including
aliphatic, aromatic ketones, R- and β-ketoesters, and steri-
cally bulky aryl alkyl ketones and diaryl ketones.15,16,20 This
enzyme showed low enantioselectivity for the reduction of
para-substituted acetophenones (14-59% ee), although it
catalyzed the reduction of other ketones to the corre-
sponding chiral alcohols with excellent enantiomeric
purity.15,16 Recently, the X-ray structures of this carbonyl
reductase and its complex with a coenzyme, NADPH, have
been determined.21 A substrate-enzyme docking study of
p-methoxyacetophenone into the crystal structure of SSCR
was performed with ICM-Pro 3.4.9d22-24 in order to better
understand the enantioselective versatility in this ketone
reduction. During these simulations, two opposite orienta-
tions of p-methoxyacetophenone, which yield (S)-alcohol
isomer or the (R)-counterpart, respectively, have been found
to be energetically close to each other in the high-scoring
docking conformations. In both conformations, residues
SCHEME 1. Enzymatic Reduction of Para-Substituted Aceto-
phenones Catalyzed by SSCR Mutant Enzymes with a Cofactor
Regeneration System of D-Glucose Dehydrogenase and D-Glucose
M242 and Q245 are in close proximity to the para-substit-
uent of the acetophenones with hydrogen bonding or hydro-
phobic interactions.25,26 These simulation results are quali-
tatively consistent with the observed low enantio selectivity
in the SSCR-catalyzed reduction of para-substituted acet-
ophenones, and such close interaction was conjectured to
have played significant roles in determining the enzyme’s
enantioselectivity. Therefore, the residues M242 and Q245 in
the catalytic site were identified as the mutation targets to
improve the enzyme enantioselectivity. Single and double
saturation mutagenesis of residues M242 and Q245 were
performed, and the resulting mutant libraries were screened
for enhanced enantioselectivity toward the reduction of
para-substituted acetophenones. The preliminary results of
site-saturation mutagenesis of residue Q245 were previously
communicated,26 herein a detailed account of these studies
are presented.
Results and Discussion
Focused libraries of mutants were created by means of
site-saturation mutagenesis at residues M242 and Q245 in
the catalytic cavity of the carbonyl reductase from S. salmo-
nicolor. The resulting mutant libraries were screened with
p-methoxyacetophenone as substrate to select colonies
which showed close or higher activity than the wild-type
SSCR enzyme. The enantioselectivity of these selected co-
lonies was then measured by chiral gas chromatography. The
preliminary results obtained from the screening of the site-
saturation mutagenesis library at residue Q245 were pre-
viously reported in a communication26 and are not repeated
here. In screening the saturation mutagenesis library of
residue M242, four mutants were selected with higher
activity than others in the NADPH-adjuvant reduction of
p-methoxyacetophenone. Among these mutant enzymes,
M242Y, M242C, and M242G maintained the same stereo
preference (R) with the wild-type SSCR but they catalyzed
the reduction with lower enantioselectivity than the latter.
However, M242D produced (S)-1-(p-methoxyphenyl)ethanol
with ee values of 39%.
ꢀ €
(5) Martın-Matute, B.; Edin, M.; Bogar, K.; Kaynak, F. B.; Backvall,
´
J.-E. J. Am. Chem. Soc. 2005, 127, 8817.
(6) Burton, S. G.; Cowan, D. A.; Woodley, J. M. Nat. Biotechnol. 2002,
20, 37.
(7) Ema, T.; Yagasaki, H.; Okita, N.; Nishikawa, K.; Korenaga, T.;
Sakai, T. Tetrahedron: Asymmetry 2005, 16, 1075.
(8) Inoue, K.; Makino, Y.; Itoh, N. Tetrahedron: Asymmetry 2005, 16,
2539.
(9) Kaluzna, I. A.; Matsuda, T.; Sewell, A. K.; Stewart, J. D. J. Am.
Chem. Soc. 2004, 126, 12827.
(10) Phillips, R. S. Can. J. Chem. 2002, 80, 680.
(11) Reetz, M. T.; Carballeira, J. D.; Peyralans, J.; Hobenreich, H.;
Maichele, A.; Vogel, A. Chem.;Eur. J. 2006, 12, 6031.
(12) Paradisi, F.; Conway, P. A.; Maguire, A. R.; Engel, P. C. Org.
Biomol. Chem. 2008, 6, 3611.
(13) Babtie, A.; Tokuriki, N.; Hollfelder, F. Curr. Opin. Chem. Biol. 2010,
14, 200.
(14) Zhu, D.; Yang, Y.; Hua, L. J. Org. Chem. 2006, 71, 4202.
(15) Zhu, D.; Yang, Y.; Buynak, J. D.; Hua, L. Org. Biomol. Chem. 2006,
4, 2690.
(16) Zhu, D.; Hua, L. J. Org. Chem. 2006, 71, 9484.
(17) Zhu, D.; Malik, H. T.; Hua, L. Tetrahedron: Asymmetry 2006, 17,
3010.
(18) Yang, Y.; Zhu, D.; Piegat, T. J.; Hua, L. Tetrahedron: Asymmetry
2007, 18, 1799.
(19) Zhu, D.; Stearns, J. E.; Ramirez, M.; Hua, L. Tetrahedron 2006, 62,
4535.
(20) Li, H.; Zhu, D.; Hua, L.; Biehl, E. R. Adv. Synth. Catal. 2009, 351,
583.
(21) Kamitori, S.; Iguchi, A.; Ohtaki, A.; Yamada, M.; Kita, K. J. Mol.
Biol. 2005, 352, 551.
These mutant SSCR enzymes were further studied to
determine their enantioselectivity toward the reduction of
other para-substituted acetophenones (Scheme 1). The co-
factor NADPH was regenerated with D-glucose dehydro-
genase and D-glucose. The results are summarized in Table 1.
It can be seen that when compared to the wild-type SSCR,
(22) Schapira, M.; Abagyan, R.; Totrov, M. J. Med. Chem. 2003, 46,
3045.
(23) Schapira, M.; Raaka, B. M.; Das, S.; Fan, L.; Totrov, M.; Zhou, Z.;
Wilson, S. R.; Abagyan, R.; Samuels, H. H. Proc. Natl. Acad. Sci. U.S.A.
2003, 100, 7354.
(25) Cundari, T. R.; Dinescu, A.; Zhu, D.; Hua, L. J. Mol. Model. 2007,
13, 685.
(26) Zhu, D.; Yang, Y.; Majkowicz, S.; Pan, T. H.-Y.; Kantardjieff, K.;
Hua, L. Org. Lett. 2008, 10, 525.
(24) An, J.; Totrov, M.; Abagyan, R. Mol. Cell. Proteomics 2005, 4, 752.
7560 J. Org. Chem. Vol. 75, No. 22, 2010

Li et al.
JOCArticle
TABLE 1. Reduction of Para-Substituted Acetophenones with M242 Mutant Enzymes of SSCRa
substrate
SSCR-WT ee (%)
M242Y ee (%)
M242D ee (%)
M242C ee (%)
M242G ee (%)
p-H
p-F
p-Cl
p-Br
p-CH3
p-OCH3
p-C(CH3)3
p-CF3
42 (R)
46 (R)
14 (R)
42 (R)
59 (R)
57 (R)
31 (R)
17 (S)
12 (S)
41 (S)
36 (S)
22 (S)
21(R)
7 (R)
82 (S)
90 (S)
77 (S)
61 (S)
43 (S)
39 (S)
>99 (S)
37 (S)
13 (S)
54 (S)
27 (S)
4 (R)
38 (R)
18 (R)
96 (S)
5 (S)
54 (S)
70 (S)
62 (S)
52 (S)
4 (S)
6 (R)
90 (S)
17 (S)
93 (S)
28 (S)
aThe ee values were determined by chiral GC analysis.27
TABLE 2. Reduction of Para-Substituted Acetophenones with M242/Q245 Double Mutant Enzymes of SSCRa
substrate
SSCR-WT ee (%)
M242L/Q245P ee (%)
M242F/Q245T ee (%)
M242C/Q245L ee (%)
M242L/Q245T ee (%)
p-H
p-F
p-Cl
p-Br
p-CH3
p-OCH3
p-C(CH3)3
p-CF3
42 (R)
46 (R)
14 (R)
42 (R)
59 (R)
57 (R)
31 (R)
17 (S)
74 (S)
92 (S)
99 (S)
>99 (S)
99 (S)
99 (S)
42 (S)
90 (S)
94 (S)
89 (S)
72 (S)
92 (S)
95 (S)
98 (S)
17 (S)
16 (S)
50(S)
24 (S)
20 (S)
36 (S)
99 (S)
94 (S)
58 (S)
94 (S)
98 (S)
99 (S)
95 (S)
97 (S)
99 (S)
99 (S)
>99 (S)
>99 (S)
aThe ee values were determined by chiral GC analysis.27
which catalyzed the reduction to give (R)-enantiomer, these
mutant enzymes (M242Y, M242D, M242C, and M242G)
disfavored the formation of (R)-configurated chiral alcohols
for reductions of all para-substituted acetophenones to such
an extent that the (S)-enantiomer of product alcohols was
obtained in most cases. Especially for p-tert-butylacetophe-
none, these mutant SSCR enzymes exhibited inverted en-
antiopreference and enhanced enantioselectivity. It is clear
that the residue 242 in the catalytic cavity plays an important
role in determining the enantioselectivity for the reduction of
the para-substituted acetophenones, but to a less extent than
residue Q245. Among the four mutants, M242D exerts the
greatest effect on the enzyme enantioselectivity. If the sub-
stituent amino acids are lined up according to their hydro-
phobicity as M > C, Y, G . D, it is intriguing to observe that
the replacement of hydrophobic methionine with polar
amino acids cysteine, tyrosine, and glycine has altered the
enzyme spatial preference toward the S configuration. This
tendency is first reflected on the decrease in the number of
substrates with R preference (SSCR-WT = 7, M242C = 3,
M242Y = 2, M242G = 1 and M242D = 0). Second, this
hydrophobicity effect was also reflected in enantiomeric
values, e.g., for p-methoxyacetophenone, the ee value toward
the R enantiomer dropped from 57% (the wild-type) to 6%
(M242G). The replacement with negative charged aspartic
acid then led S isomer alcohols as predominant products for
all eight substrates. For p-trifluoromethylacetophenone, the
major product was (S)-enantiomer with wild-type and mu-
tant enzymes as biocatalyst; mutation of residue M242 did
not greatly affect the enzyme enantioselectivity.
The library was screened with p-methoxyacetophenone as
the substrate for higher activity than the wild-type enzyme,
(S)-1-(p-Methoxyphenyl)ethanol was the major enantiomer of
the product alcohol for all four mutant enzymes in which three of
them exhibited high enantioselectivity with greater than 92% ee.
Some other para-substituted acetophenones were also
studied with these double mutant enzymes to examine their
enantioselectivity (Table 2). As shown in Table 2, all four
double mutant enzymes greatly increased the (S)-enantio-
meric preference to an extent that the absolute configuration
of major products was inverted from (R) to (S). Among
them, mutant M242L/Q245P possessed the highest enantio-
selectivity in the reduction of these ketones, which was higher
than those of the mutant Q245P, suggesting that replace-
ments of M242 with L and Q245 with P had a synergic effect
on preferable formation of (S)-enantiomeric products. Com-
pared to Q245L, mutant M242C/Q245L showed much lower
enantioselectivity in most cases. The preference for the (S)-
enantiomer formation of these related mutant enzymes
followed the order wild-type < M242C < M242C/Q245L
<Q245L, indicating that M242C had a negative effect on the
(S)-enantiomeric preference of Q245L. In contrast to this,
M242C had a positive effect on the (S)-enantiomeric pre-
ference of wild-type enzyme as described above. Mutant
M242L/Q245T exhibited higher enantioselectivity than
mutant M242F/Q245T. The synthetic application of these
mutant enzymes was demonstrated by the reduction of
p-chloroacetophenone with mutant M242L/Q245P at prepara-
tive scale, giving (S)-1-(p-chlorophenyl)ethanol in 92% yield.
The kinetic parameters of these mutant carbonyl reduc-
tases were measured and their overall catalytic efficiencies
(kcat/Km) toward these para-substituted acetophenones are
presented in Figures 1-3. From Figure 1, it can be seen that
the overall catalytic efficiencies of mutant carbonyl reduc-
tases M242Y, M242D, and M242G were lower than wild-
type enzymes, while for mutant M242C, the kcat/Km values
were comparable with those of the wild-type enzyme for
most substrates with the exceptions of p-chloro (twice
higher) and the unsubstituted acetophenone (half high).
It has been demonstrated that mutation of either residue
Q245 or M242 greatly tuned the enantioselectivity of car-
bonyl reductase SSCR toward para-substituted acetophe-
nones. A question has arisen as to whether a highly enantio-
selective mutant carbonyl reductase could be created by
double mutation of residues M242 and Q245? With interest
in searching for novel highly enantioselective carbonyl re-
ductases, we constructed a focused double mutant library by
site-saturation mutagenesis at both residues 242 and 245.
J. Org. Chem. Vol. 75, No. 22, 2010 7561

Li et al.
JOCArticle
FIGURE 1. The catalytic efficiency (kcat/Km, min-1 mM-1) of
3
FIGURE 3. The catalytic efficiency (kcat/Km, min-1 mM-1) of
SSCR wild-type and M242 mutant enzymes toward reduction
of acetophenones: (1) unsubstituted, (2) p-fluoro, (3) p-chloro, (4)
p-bromo, (5) p-methyl, (6) p-methoxy, (7) p-tert-butyl, and (8)
p-trifluoromethyl.
3
SSCR wild-type and M242/Q245 double mutant enzymes for the
reduction of acetophenones: (1) unsubstituted, (2) p-fluoro, (3)
p-chloro, (4) p-bromo, (5) p-methyl, (6) p-methoxy, (7) p-tert-butyl,
and (8) p-trifluoromethyl.
corresponding “anti-Prelog” (R)-alcohols. As a comparison,
the wild-type enzyme was not able to reduce these ketones
and provided (S)-alcohols as the major products for the
reduction of smaller ketones.28 Other examples are the
substrate-binding residue exchange of stereocomplementary
ketoreductase domains leading to the reversion of stereo-
chemistry. For example, mutation of residues in motifs I and
II in ketoreductase domains eryKR1 and eryKR2 from the
erythromycin polyketide synthase altered the stereochemical
outcome in reduction of (2R,S)-2-methyl-3-oxopentanoic
acid N-acetyl-cysteamine thioester.29,30 In the study to con-
firm the hypothetical tropinone binding mode, replacement
of five substrate-binding residues of one tropinone reductase
(TR-1) with those found in the corresponding positions of
the other tropinone reductase (TR-2) in the biosynthetic
pathway of tropane alkaloids resulted in the switch of
stereospecificity of TR-1 into that of TR-2 and vice versa.31
These structure-based mutageneses revealed some insights
into how these enzymes control their stereoselectivity. The
scarcity of such study might be due to the difficulty of
implementing a high-throughput method to determine en-
antioselectivity in the ketone reduction that allows the rapid
screening of a large number of mutants. The present study
demonstrates that enzyme-substrate docking-guided site-
mutagenesis is a practical approach to access highly enan-
tioselective carbonyl reductase enzymes.
FIGURE 2. The catalytic efficiency (kcat/Km, min-1 mM-1) of
3
SSCR wild-type and Q245 mutant enzymes toward reduction
of acetophenones: (1) unsubstituted, (2) p-fluoro, (3) p-chloro, (4)
p-bromo, (5) p-methyl, (6) p-methoxy, (7) p-tert-butyl, and (8)
p-trifluoromethyl.
The Q245 mutant and M242/Q245 double mutant en-
zymes showed higher overall catalytic efficiencies than the
wild-type enzyme, especially for the acetophenones with
electron-withdrawing groups such as chloro, bromo, and
trifuoromethyl (Figures 2 and 3). These results suggest that
residues 242 and 245 not only play an important role in
determining the enzyme’s enantioselectivity, but also affect
the catalytic activity toward the reduction of para-substi-
tuted acetophenones, and residue 245 exerts a much greater
influence on both enantioselectivity and catalytic activity.
Naturally occurring enzymes are not always highly active
and/or stereoselective for unnatural substrates. Improve-
ment in enzyme activity and enantioselectivity is often
necessary to meet the synthetic needs. Up to now, reports
on the enantioselectivity engineering of carbonyl reductase
are still very limited. Recently, Philips’s group has reported
that secondary ADH from Thermoanaerobacter ethanolicus
(TeSADH) bearing a single mutation I86A catalyzed the
reduction of benzylic and heteroaryl ketones to give the
Conclusions
Although a carbonyl reductase from Sporobolomyces
salmonicolor catalyzed the reduction of various ketones
with diverse structures to highly enantiomerically pure
products, it reduced para-substituted acetophenones with
low enatioselectivity. Enzyme-substrate docking studies
(28) Musa, M. M.; Lott, N.; Laivenieks, M.; Watanabe, L.; Vieille, C.;
Phillips, R. S. ChemCatChem 2009, 1, 89.
(29) Baerga-Ortiz, A.; Popovic, B.; Siskos, A. P.; O’Hare, H. M.; Spiteller,
D.; Williams, M. G.; Campillo, N.; Spencer, J. B.; Leadlay, P. F. Chem. Biol.
2006, 13, 277.
(30) O0Hare, H. M.; Baerga-Ortiz, A.; Popovic, B.; Spencer, J. B.;
Leadlay, P. F. Chem. Biol. 2006, 13, 287.
(27) Zhu, D.; Rios, B. E.; Rozzell, J. D.; Hua, L. Tetrahedron: Asymmetry
2005, 16, 1541.
(31) Nakajima, K.; Kato, H.; Oda, J. i.; Yamada, Y.; Hashimoto, T. J.
Biol. Chem. 1999, 274, 16563.
7562 J. Org. Chem. Vol. 75, No. 22, 2010

Li et al.
JOCArticle
with p-methoxyacetophenone revealed that the observed low
enantioselectivity was consistent with two energetically close
conformations of substrate in the binding site. Residues M242
and Q245 in the substrate binding site were identified as the key
residues for enzyme engineering to create highly enantioselective
carbonyl reductases for the reduction of para-substituted acet-
ophenones. Site-saturation mutagenesis of single point mutation
at M242 or Q245 and of double point mutation of these two
residues was performed. Four M242 variants, three Q245 var-
iants, and four M242/Q245 double mutant enzymes were ob-
tained by screening of these focused mutant libraries. All these
mutant enzymes showed stronger preference for the formation of
(S)-configurated product alcohols in the reduction of para-
substituted acetophenones than the wild-type SSCR, leading
to the configuration invertion of the major product alcohols
from (R) to (S) in most cases. Especially for the Q245 mutant
and the double mutant enzymes, (S)-alcohols were obtained in a
very high enantiomeric purity. Similarly, residues M242 and
Q245 also alter the enzyme activity, but M242 has less effect than
Q245, as shown in Figures 1-3. These studies indicated that
residues M242 and Q245 not only play a crucial role in determin-
ing the enantioselectivity of these ketone reductions, but also
exert some effect on the enzyme’s catalytic activity. The struc-
ture-based site-saturation mutagenesis created highly enantiose-
lective mutant carbonyl reductases. Among them, double
mutant enzyme M242L/Q245P showed the highest enantioselec-
tivity, which catalyzed the reduction of the tested para-substi-
tuted acetophenones to give (S)-enantiomeric products in
g99% ee, and the only exception was p-fluoroacetophenone
with 92% ee. As such, the in silico docking guided semirational
approach should be a very valuable methodology for discovery
of new enzymes with synthetic advantages.
cultures were induced with isopropyl-β-D-thiogalactopyrano-
side (IPTG, 0.1 mM), and incubated at 30 ꢀC for 6 h. The cells
were harvested by centrifugation at 4100 rpm for 30 min and
lysed with 400 U of Ready-Lyse lysozyme (Epicenter, Inc.) in
50 μL of TES buffer (100 mM NaCl, 1 mM EDTA, 10 mM Tris-
HCl, pH 7.5). The lysis reactions were incubated at room
temperature for 30 min, and 300 μL of potassium phosphate
buffer (100 mM, pH 6.5) was added to the mixture. The diluted
cell-free extract thus obtained after centrifugation was used in
the activity assays, which were carried out by spectrophotomet-
rically monitoring the disappearance of NADPH at 340 nm with
40-methoxyacetophenone as the substrate. The colonies, which
showed activity close to or higher than the wild-type SSCR
enzyme, were cultivated in 15 mL of LB media containing 100
μg/mL ampicillin at 37 ꢀC for 12 h. The cell-free lysates were
prepared as described above and used in the 1 mL scale reactions
to determine the enantioselectivity. The reaction mixtures con-
taining 300 μL of lysate, 0.5 mg of NADPH, 12.5 mM
40-methoxyacetophenone, 0.5 mg of GDH, and 4.0 mg of
glucose in 1 mL of potassium phosphate buffer (100 mM, pH
6.5) were shaken at room temperature for 12 h. The ee value
of the product alcohol was measured by chiral GC analysis.
The plasmid DNAs from the clones, which showed some
enantioselectivity, were isolated and sequenced. Confirmed
plasmid DNAs were transformed into E. coli competent cell
Rosetta 2(DE3)pLysS.
Screening of M242 and Q245 Double Mutant Library. Three
thousand colonies were picked up randomly from the double
mutant library. A similar procedure described above was fol-
lowed, using the wild type and mutant Q245P as references. The
plasmid DNAs from the clones, which showed higher enantios-
electivity, were isolated and sequenced. Confirmed plasmid
DNAs were transformed into E. coli competent cell Rosetta
2(DE3)pLysS.
Purification of SSCR Mutant Enzymes. SSCR mutants
were grown in LB medium containing 100 μg/mL ampicillin
and incubated at 37 ꢀC until the optical density reached 0.6 at
600 nm. The expression was induced by 0.1 mM IPTG and the
cell culture was incubated at 30 ꢀC for another 6 h. The cells were
harvested by centrifugation at 4100 nm at 4 ꢀC for 30 min. The
cell pellet was resuspended in potassium phosphate buffer (100
mM, pH 7.4) and the cells were disrupted by a High Pressure
Homogenizer. The cell-free extract was mixed with an equal
volume of PEI solution (0.25% polyethyleneimine, NW 40-60
K, 6% NaCl, 100 mM Borax, pH 7.4) to remove lipids. The
supernatant was precipitated with 55% ammonium sulfate. The
precipitate was resuspended in potassium phosphate buffer (10
mM, pH 7.4, 0.1 mM dithiothreitol), followed by diafiltration
with potassium phosphate buffer (10 mM, pH 7.4, 0.1 mM
dithiothreitol) to remove the excess of ammonium sulfate. The
resulting lysate was lyophilized to give the SSCR mutant
enzymes as white powders, which were used in the acetophenone
reduction to measure the activity and enantioselectivity.
Kinetic Assay. The enzyme activity for ketone reduction was
determined by spectrophotometrically monitoring the reduc-
tion of NADPH at 340 nm at room temperature with a molar
extinction coefficient of 6220 M-1 cm-1. The reaction was
carried out in potassium phosphate buffer (100 mM, pH 7.0)
with 0.25 mM NADPH. The substrate concentration range was
between 0.1 and 2.4 mM. A microplate reader was applied for
the measurements and SigmaPlot version11 software was used
for calculation of the kinetic parameters.
Experimental Section
M242 Mutant Library. The carbonyl reductase gene from
Sporobolomyces salmonicolor (SSCR) was cloned as described
before.15 The plasmid was used as the template for the Quick
Change site-directed mutagenesis polymerase chain reaction
(PCR). The forward primer was 50-TTCCCCGGCTCTGGC-
TCTGNNKCCACCGCAGT ACTACGTT-30 and the reverse
primer was 50-AACGTAGTACTGCGGTGGMNNCAGAG-
CCAGAGCCGGGGAA-30. PCR was carried out with pfx
polymerase under the following conditions: the reaction was
started at 94 ꢀC (2 min), followed by 30 cycles: 94 ꢀC (30 s), 55 ꢀC
(45 s), 68 ꢀC (6 min 48 s), with a final extension at 68 ꢀC (10 min).
The reaction was carried out in 100 μL reaction volume contain-
ing 0.5 pM of both primers, 210 ng genomic DNA, 1.0 mM
MgSO4, 0.2 mM dNTP, and 3.7 U of pfx polymerase. The PCR
product was digested with DpnI restriction enzyme and elec-
tronically transformed to E. coli competent cell BL21(DE3).
M242 and Q245 Double Mutant Library. The double mutant
library construction followed the similar procedure described
above for M242. The forward primer was 50-TTCCCCGG-
CTCTGGCTCTGNNKCCACCGNNKT-30, and the reverse
primer was 50-AACAGCGGAAACGTAGTAMNNCGGTG-
GMNNCA-30.
Screening of M242 Mutant Library. Each of the 92 mutant
colonies along with 4 colonies expressing wild-type SSCR gene
were grown in 150 μL of Luria-Bertani (LB) medium supple-
mented with 100 μg/mL ampicillin at 37 ꢀC for 12 h in a 96-well
microplate. For each colony, 4 μL of the overnight culture
was diluted into 2 mL of LB medium containing 100 μg/mL
ampicillin and the culture was incubated at 37 ꢀC with shaking at
150 rpm until the optical density reached 1.2 (2 h). The cell
Measurement of Enantioselectivity. The enantioselectivity
of the enzymatic reduction of ketones was studied by using
an NADPH recycle system. The general procedure was as
follows: D-glucose (4 mg), D-glucose dehydrogenase (0.5 mg),
NADPH (0.5 mg), the carbonyl reductase (SSCR, 0.5 mg), and
ketone solution in DMSO (50 μL, 0.25 M) were mixed in a
J. Org. Chem. Vol. 75, No. 22, 2010 7563

Li et al.
JOCArticle
potassium phosphate buffer (1 mL, 100 mM, pH 6.5) and the
mixture was shaken overnight at room temperature. The mix-
ture was extracted with methyl tert-butyl ether (1 mL). The
organic extract was dried over anhydrous sodium sulfate and
was subjected to chiral GC analysis to determine the enantio-
meric excess. The absolute configurations of product alcohols
were identified by comparing the chiral GC data with the
standard samples.27
reaction mixture was shaken at room temperature for 24 h. The
reduction was completed as shown by GC analysis. The mixture
was saturated with sodium chloride and extracted with methyl
tert-butyl ether (3 ꢀ 30 mL). The organic extract was dried over
anhydrous sodium sulfate and removal of solvent gave (S)-
1-(p-chlorophenyl)ethanol14 (71.1 mg, yield 92%).
Acknowledgment. D.Z. thanks the Chinese Academy of
Sciences for support from the Knowledge Innovation Program
(Grant No. KSCX2-YWG-031), Tianjin Municipal Science &
Technology Commission (No. 09ZCKFSH01000), and Minis-
try of Science and Technology of China from the National Key
Technology R&D Program (No. 2008BAI63B07).
Enzymatic Reduction of p-Chloroacetophenone (Preparative
Scale). D-Glucose (20g L-1), D-glucose dehydrogenase (2 g L-1),
NADPH (1 g L-1), and mutant M242L/Q245P enzyme (2 g L-1
)
were dissolved in potassium phosphate buffer (100 mM,
pH 7.0), and 38 mL of the resulting solution was mixed with
2 mL of p-chloroacetophenone solution in DMSO (0.25 M). The
7564 J. Org. Chem. Vol. 75, No. 22, 2010