An esterase with superior activity and enantioselectivity towards 1,2-O-isopropylideneglycerol esters obtained by protein design

Article abstract of DOI:10.1002/adsc.201200211

The Escherichia coli esterase YbfF displays high activity towards 1,2-O-isopropylideneglycerol (IPG) butyrate and IPG caprylate, and prefers the R-enantiomer of these substrates, producing the S-enantiomer of the IPG product in excess. To improve the pote

Full text of DOI:10.1002/adsc.201200211

FULL PAPERS
DOI: 10.1002/adsc.201200211
An Esterase with Superior Activity and Enantioselectivity
towards 1,2-O-Isopropylideneglycerol Esters Obtained by
Protein Design
Luis F. Godinho,a Carlos R. Reis,a Ronald van Merkerk,a Gerrit J. Poelarends,a
and Wim J. Quaxa,*
a
Department of Pharmaceutical Biology, Groningen Research Institute of Pharmacy, University of Groningen, Antonius
Deusinglaan 1, 9713 AV Groningen, The Netherlands
Fax : (+31)-50-363-3000; phone: (+31)-50-363-7660; e-mail: w.j.quax@rug.nl
Received: March 14, 2012; Revised: July 5, 2012; Published online: November 4, 2012
Supporting information for this article is available on the WWW under http://dx.doi.org/10.1002/adsc.201200211.
Abstract: The Escherichia coli esterase YbfF displays 235 yielded several best YbfF variants with enhanced
high activity towards 1,2-O-isopropylideneglycerol activity and enantioselectivity towards IPG esters.
(IPG) butyrate and IPG caprylate, and prefers the The best YbfF mutant, W235I, exhibited a 2-fold
R-enantiomer of these substrates, producing the S- higher enantioselectivity than wild-type YbfF, with
enantiomer of the IPG product in excess. To improve an E=38 for IPG butyrate and an E=77 for IPG
the potential of the enzyme for the kinetic resolution caprylate. Molecular docking experiments further
of racemic esters of IPG, an enhancement of the ac- support the enhanced enantioselectivity shown ex-
tivity and enantioselectivity would be highly desira- perimentally and the structural effects of this amino
ble. Molecular docking of the R-enantiomer of both acid substitution on the active site of YbfF are pro-
IPG esters into the active site of YbfF allowed the vided. The engineered W235I mutant is an attractive
identification of proximal YbfF active site residues. catalyst for practical applications in the kinetic reso-
Four residues (25, 124, 185 and 235) were selected as lution of IPG esters.
targets for mutagenesis, in order to enhance YbfF ac-
tivity and enantioselectivity towards IPG esters. Keywords: enantioselectivity; Escherichia coli; ester-
Random mutagenesis at positions 25, 124, 185 and ases; kinetic resolution; protein engineering
Introduction
erate enantioselectivity (E=10) towards IPG butyrate
(1) and a high enantioselectivity (E=29) towards IPG
Esterases represent a class of hydrolytic enzymes that
are highly useful for organic synthesis, as they have
broad substrate specificity and exhibit often high ste-
reoselectivity.[1,2] The development of an esterase-
based biocatalytic process for the kinetic resolution of
racemic 1,2-O-isopropylideneglycerol (IPG) esters
(Scheme 1) has exciting potential for organic synthesis
because the optically pure product (S)-IPG is an im-
portant building block for the preparation of b-block-
ers, prostaglandins, and leukotrienes.[3,4]
In an ideal case, the esterase converts only the R-
enantiomer of the IPG ester, yielding the optically
pure product (S)-IPG.
Previously, we have reported that the cytoplasmic
esterase YbfF from Escherichia coli has high hydro-
lytic activity and enantioselectivity towards IPG
esters with different alkyl chains. YbfF displays mod- Scheme 1. Esterase-catalyzed conversion of IPG esters.
Adv. Synth. Catal. 2012, 354, 3009 – 3015
ꢀ 2012 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
3009

Luis F. Godinho et al.
FULL PAPERS
caprylate (3).[5] The crystal structure of YbfF has been Results and Discussion
solved and the enzyme is composed of an a/b hydro-
lase-fold domain with a three-helical bundle cap that In previous work reported by our group,[5] YbfF was
is linked by a twist helix to the a/b hydrolase-fold identified as the major enzyme responsible for the
domain.[6] The sequence motif GxSxG, characteristic pronounced activities and enantioselectivities in cell-
for a/b hydrolase-fold superfamily of enzymes, is also free extracts prepared from E. coli cells toward sub-
conserved in YbfF. However, YbfF has less than 20% strates 1 and 3 (Scheme 1), with formation of (S)-2 in
sequence identity with most known esterases and excess over (R)-2 for both substrates. YbfF exhibits
likely to contain a catalytic tetrad of Ser89-His234- moderate enantioselectivity (E=10) towards 1 and
Asp113-Ser206 (with the first Ser being the nucleo- a high enantioselectivity (E=29) towards 3.[5] Based
phile) instead of the conserved catalytic triad.[6] It was on previous molecular docking experiments, where
reported that the spatial position of the conserved the R-enantiomer of 1 or 3 was modeled into the
catalytic Asp113 in YbfF differs from the spatial posi- active site of YbfF, we obtained insight in the binding
tion observed for the catalytic acid residue from other mode of these IPG esters. The docking model sug-
hydrolases, and that the Ser206 takes up the spatial gests that the short alkyl chain of substrate 1 can be
position of the catalytic acid residue.[6] Evidence for accommodated in the thin hydrophobic pocket, while
a fourth catalytic residue such as Ser206 has also been for substrate 3 the longer alkyl chain is accommodat-
obtained in other members of the hydrolase family.[7] ed along the crevice. These observations set the stage
Although wild-type YbfF exhibits significant activi- for the mutagenesis study reported here. A detailed
ty and selectivity towards IPG esters, it still requires examination of the most favorable binding poses of
further improvement towards these substrates for (R)-1 and (R)-3 in the active site of YbfF (Figure 1),
practical applications. In the last two decades novel revealed several active site residues that are in close
laboratory evolution methods have been devised in proximity to these substrates. Two residues (L25 and
order to improve enzyme properties (such as activity, I185) that are part of the catalytic cavity were select-
stability and selectivity) towards non-natural sub-
strates.[8–10] Random evolution approaches have been
the preferred method of choice, since they do not re-
quire detailed knowledge of the enzyme structure or
reaction mechanism. However, with the increased
number of 3D structures solved and the growth of
computational methods, structure-based enzyme engi-
neering strategies are becoming more in reach for im-
proving enzyme properties. The major advantage of
this structure-based approach is that it allows for the
generation of smaller libraries, reducing the time of
screening or selection, with higher functional diversi-
ty.[8–12]
Herein, we describe molecular docking experiments
of (R)-1 into the YbfF active site in order to identify
active site residues that are in close proximity to the
substrate. These docking experiments helped to iden-
tify four positions (25, 124, 185 and 235) in the YbfF
active site that might be good targets for engineering
the enantioselectivity of YbfF towards 1 and 3. These
positions were randomized by site-saturation muta-
genesis and four single-site libraries were created. We
used activity screening (colorimetric screen) to select Figure 1. Close-up view of the catalytic residues (green
sticks) and mutated active site surrounding residues of YbfF
(white sticks) in complex with the most favorable docked
pose for (R)-IPG-C4 (R-1) (yellow sticks). The carbonyl
oxygen of (R)-IPG-C4 is at hydrogen bonding distance to
the nitrogen atoms of Leu-25 and Met-90, suggesting that
these two residues compose the oxyanion hole of YbfF and
trap the carbonyl oxygen atom of the substrate, in agree-
ment with the mechanism suggested for other hydrolases of
this family.[5] Docking of substrate and minimization runs
were performed using available tools in Discovery Studio
2.5.
for YbfF clones with improved enantioselectivity to-
wards substrate 1.[13] Several single mutants at posi-
tions 185 and 235 affected YbfF enantioselectivity to-
wards substrates 1 and 3. The best mutant (W235I)
showed significantly higher activity and enantioselec-
tivity towards both substrates 1 and 3, when com-
pared to the wild-type enzyme. A possible mechanism
is devised to explain the experimental observations
obtained for the best YbfF mutant.
3010
asc.wiley-vch.de
ꢀ 2012 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Adv. Synth. Catal. 2012, 354, 3009 – 3015

An Esterase with Superior Activity and Enantioselectivity towards 1,2-O-Isopropylideneglycerol Esters
ed for mutagenesis along with two aromatic residues three clones from library 235) of them were selected
(H124 and W235) that form the neck in the bottom based on their apparent increased enantioselectivity
region of the l-shaped sack (crevice).
towards substrate 1. The plasmids from the five se-
lected clones were isolated and the YbfF mutant gene
was sequenced. Sequence analysis revealed that both
clones from library 185 contained the same mutation
I185V, whereas the clones from library 235 contained
Library Construction and Activity Screening
To investigate the effect of different amino acid side the following mutations: W235V, W235I, and W235E.
chains at the four selected positions (L25, H124, I185 These mutants and wild-type YbfF were produced
and W235) on YbfF enantioselectivity, each selected in the cytoplasm of E. coli BL21 (DE3) and purified
residue was replaced by 12 other residues (Phe, Leu, to >90% homogeneity (as assessed by SDS/PAGE)
Ile, Val, Cys, Arg, Ser, Gly, Tyr, His, Asn, Asp). Four using an Ni-based immobilized metal affinity chroma-
single-site libraries were generated with the help of tography procedure. The purification procedure used
degenerate NDT primers, comprising 12 codon var- to purify YbfF wild-type and mutants is highly specif-
iants, with a balanced mix of aliphatic and aromatic, ic for proteins containing a His-tag, so that the possi-
polar and non-polar, positively and negatively bility of contamination from native YbfF from the
charged residues.[14] The resulting libraries were host organism is eliminated. The purified enzymes
screened for clones that exhibited higher enantiose- were stored at À808C until further use.
lectivity towards 1. The screening assay used was
based on the colorimetric method developed by Janes
et al.[13] Cells of the YbfF clones from each library Kinetic Resolution of IPG Esters
were first grown in deep-well plates, after which cells
were washed, lysed and assayed separately for their The activity and enantioselectivity of YbfF wild-type
ability to hydrolyse (R)-1 and (S)-1. The clones initial and mutants toward substrates 1 and 3 were investi-
rates of hydrolysis of (R)-1 and (S)-1, were measured gated by kinetic resolution experiments (Table 1).
separately using a pH indicator (4-nitrophenol).[13] Mutant I185V exhibited higher activity and showed
Overall, 27 out of 128 clones from the four libraries the same enantioselectivity (E=12) towards substrate
were pre-selected due to their apparent higher enan- 1 when compared to YbfF wild-type. For substrate 3,
tioselectivity towards 1, when compared to wild-type the mutant I185V was also more active than wild-type
YbfF. Under the assay conditions used, no hydrolysis YbfF, however its enantioselectivity was reduced by
of 1 by the cell free extract (CFE) of E. coli BL21 2-fold. Mutants W235V, W235I, and W235E displayed
(DE3) cells transformed with pET26b (an empty both higher activity and enantioselectivity towards
vector not harboring the YbfF gene) was detected, substrate 1 when compared to wild-type YbfF, with
implying that the activity of the YbfF protein encoded mutant W235I exhibiting the most pronounced enan-
by the endogenous chromosomal ybfF gene is not de- tioselectivity (E=38). Similarly, for substrate 3 all
tectable.
mutants showed higher activity when compared to
Clones that exhibited apparent higher enantioselec- wild-type YbfF, but only mutant W235I showed
tivity towards 1 were selected for further analysis by higher enantioselectivity (E=77) than wild-type
chiral GC. The cells of the 27 selected clones were re- YbfF.
grown, cell free extract prepared, and each CFE incu-
To demonstrate the synthetic potential of mutant
bated at 308C in the presence of 1. Out of the 27 W235I for the preparation of (S)-IPG, the kinetic res-
clones screened, five (two clones from library 185 and olution of racemic 3 (5 mM) by YbfF wild-type
Table 1. Results of kinetic resolution experiments with wild-type and mutants of YbfF using substrates 1 or 3.[a]
Substrate 1
ees [%][c]
Substrate 3
ees [%]
Enzyme
eep [%][b]
C [%][d]
E[e]
eep [%]
C [%]
E
wild-type
I185V
W235I
W235V
W235E
76
69
88
77
77
57
76
79
80
89
43
52
47
51
54
12
12
38
18
22
92
83
95
84
88
52
64
66
82
74
36
44
41
49
46
40
20
77
30
34
[a]
Kinetic resolution assays were performed in 70 mM MOPS buffer (pH 7.5) containing Tween 80 (14.3%, w/v) at 308C.
eep: enantiomeric excess of product.
ees: enantiomeric excess of substrate.
[b]
[c]
[d]
[e]
C: substrate conversion.
E: enzyme enantioselectivity, the five enzymes prefer the R-enantiomer of 1 and 3, producing S-IPG.
Adv. Synth. Catal. 2012, 354, 3009 – 3015
ꢀ 2012 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
asc.wiley-vch.de
3011

Luis F. Godinho et al.
FULL PAPERS
(0.3 mM) and mutant W235I (0.3 mM) is demonstrated
in Figure 2. After 10 min, the W235I mutant had
reached 93% conversion of (R)-3 while the wild-type
showed only 12% conversion. For substrate (S)-3, the
W235I mutant had reached 20% conversion after
60 min, while the wild-type showed only 6% conver-
sion. Both YbfF wild-type and mutant showed high
enantioselectivity in the formation of (S)-2 (wild-
type: 92% ee; W235I: 95% ee). Overall, the mutant
showed a ~7.5-fold increase in the catalytic activity
towards the R-substrate whereas for the S-substrate
the increase was only ~3.3-fold, when compared to
the wild-type enzyme. These results are in accordance
with the data obtained from the kinetic parameters
determined in the next section.
The effect of pH and temperature on enzyme activ-
ity (Figure S13, Supporting Information) and temper-
ature on enzyme stability (Figure S14, Supporting In-
formation) of mutant W235I were found to be similar
to the values determined for the YbfF wild-type.[5]
Kinetic Analysis of Wild-Type and W235I
To establish whether the enhanced enantioselectivity
of the mutant W235I is caused by changes in binding
affinity and/or turnover rate of the individual enantio-
mers, kinetic parameters for the enzyme-catalyzed
conversion of (R)-1, (S)-1, (R)-3 and (S)-3 were deter-
mined (Table 2). For substrate 1, the 2-fold increase
in enantioselectivity of the mutant W235I (E=20)
was due to a ~6-fold increase in catalytic efficiency
(kcat/Km) for (R)-1, combined with a ~3-fold increase
in catalytic efficiency for (S)-1, when compared to
YbfF wild-type (E=10). The mutant increased cata-
lytic efficiency for the preferred R-enantiomer, was
mainly due to a lower Km, whereas for the S-enantio-
mer the improvement observed in catalytic efficiency
was mainly the result of a higher kcat.
Figure 2. Kinetic resolution of 5 mM racemic IPG caprylate
(3) with 0.3 mM of YbfF wild-type (A) and with 0.3 mM of
mutant W235I (B). The substrates are represented by round
*
*
symbols, where ( ) represents the S-enantiomer and ( )
represents the R-enantiomer. The products are represented
~
~
by triangles, where ( ) represents the R-enantiomer and ( )
represents the S-enantiomer.
For substrate 3, the results showed that the mutant
W235I exhibits slightly lower Km values for both
Table 2. Steady-state kinetic parameters of wild-type YbfF and mutant W235I for substrate 1 and 3.[a]
Enzyme
WT
Substrate
Km [mM]
kcat [sÀ1]
kcat/Km [MÀ1 sÀ1]
E[b]
(R)-1
(S)-1
(R)-1
(S)-1
(R)-3
(S)-3
(R)-3
(S)-3
13.6Æ4.5
9.7Æ0.9
2.6Æ0.7
12.9Æ2.4
5.8Æ1.1
7.3Æ1.2
4.7Æ0.4
4.1Æ0.8
62Æ10
4559
474
27443
1695
2651
90
10
4.6Æ0.2
71.9Æ5.6
21.9Æ2
W235I
WT
16
29
76
15.4Æ1.2
0.66Æ0.05
113Æ3.9
1.3Æ0.1
W235I
24042
317
[a]
The steady-state kinetic parameters were determined in 70 mM MOPS buffer (pH 7.5) at 308C. Errors represent standard
deviations.
The enzyme is selective towards the R-substrate, producing the S-product in large excess.
[b]
3012
asc.wiley-vch.de
ꢀ 2012 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Adv. Synth. Catal. 2012, 354, 3009 – 3015

An Esterase with Superior Activity and Enantioselectivity towards 1,2-O-Isopropylideneglycerol Esters
enantiomers of 3, when compared to wild-type. How- type YbfF, we modeled the R- and S-enantiomers into
ever, in terms of kcat the mutant W235I showed the active site of YbfF wild-type and the best mutant
a ~7.3-fold increase for (R)-3, whereas for (S)-3 the W235I. Analysis of the most favorable docking poses
increase was only ~2-fold. Taken together, these dif- of YbfF revealed striking differences in the position-
ferences in Km and kcat of mutant W235I are reflected ing of the enantiomeric substrates 1 and 3 into the
in a ~9-fold increase in the catalytic efficiency for active site, also with respect to the distance between
(R)-3 and a ~3.5-fold increase in catalytic efficiency the Og atom of the catalytic nucleophile (Ser-89) and
for (S)-3. This accounts for the observed increase the carbonyl carbon of the substrates (Figure 3). The
enantioselectivity of the mutant (E=76) when com- most favorable poses obtained for (R)-1 reveal that
pared to YbfF wild-type (E=29).
the carbonyl carbon of the substrate is proximal to
the Og atom of Ser-89 (3.4 ꢁ) for both wild-type
YbfF and W235I (Figure 3, A). For substrate (S)-1,
the most favorable pose places the carbonyl carbon of
the substrate further away from the Og atom of Ser-
89, with an observed distance of 4.0 ꢁ between the
Characterization of YbfF Wild-Type and W235I by
Molecular Modeling
In order to unravel differences in activity and enan- substrate carbonyl carbon and the Ser-89 Og
tioselectivity between the best YbfF variant and wild- (Figure 3, B), resulting in a less optimal positioning of
Figure 3. Molecular docking of (R)-IPG-C4 (R-1) (A), (S)-IPG-C4 (S-1) (B), (R)-IPG-C8 (R-3) (C) and (S)-IPG-C8 (S-3)
(D) into the active site of YbfF wild-type (yellow sticks) and mutant YbfF W235I (green sticks). The catalytic serine 89 and
mutated tryptophan 235 are depicted in yellow and green sticks for YbfF wild-type and YbfF W235I, respectively. The
cavity and crevice of YbfF are represented by a grey surface. Depicted are some of the distances (ꢁ) between the catalytic
Ser-89 Og, and the carbonyl carbon of the substrate. Docking of substrates and minimization runs were performed using
available tools in Discovery Studio 2.5. The figure was generated using Discovery Studio 2.5.
Adv. Synth. Catal. 2012, 354, 3009 – 3015
ꢀ 2012 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
asc.wiley-vch.de
3013

Luis F. Godinho et al.
FULL PAPERS
the substrate for nucleophilic attack in both YbfF enzyme, which renders the YbfF mutant to a very
wild-type and W235I. This change in conformation of suitable candidate for practical applications in the ki-
the substrate in the active site of W235I may partially netic resolution of IPG esters.
explain the increased kcat observed for (S)-1 (Table 2).
The mutation W235I seems to contribute mostly to
an enhanced enantioselectivity towards (R)-3. In Experimental Section
mutant W235I where the Trp-235 is replaced by the
smaller side chain aliphatic residue Ile, the docking General
results show that (R)-3 further sinks in the pocket of
YbfF, resulting in a distance reduction between the
All reagents were of analytical grade unless specified other-
wise. (R)-IPG butyrate, (S)-IPG butyrate, (R)-IPG caprylate
carbonyl carbon of (R)-3 and the Ser-89 Og from
3.5 ꢁ (wild-type) to 2.8 ꢁ (W235I) (Figure 3, C). This
re-orientation of (R)-3 in the active site of W235I
seems to contribute to the enhanced kcat observed ex-
perimentally, which is a determinant for the increased
enantioselectivity of the mutant W235I when com-
pared to YbfF wild-type (Table 2). In contrast, the
binding of (S)-3 is less favorable in both active sites
of YbfF wild-type and W235I, with observed distances
between the carbonyl carbon of (S)-3 and the nucleo-
philic serine Og of 4.9 ꢁ and 4.8 ꢁ for YbfF wild-
type and W235I, respectively (Figure 3, D).
The increase in enantioselectivity promoted by the
mutation W235I was mainly due to the improvement
of the enzyme catalytic efficiency towards the pre-
ferred enantiomer (R-3). To the best of our knowl-
edge, the intracellular carboxyl esterase A from B. co-
agulans exhibited the higher enantioselectivity to-
wards IPG caprylate esters reported in literature. Mo-
linari et al. showed that the CesA from B. coagulans
hydrolyzed IPG caprylate esters producing (S)-IPG
with 94% enantiomeric excess at 23% conversion,
corresponding to an E value of 43.[15,16]
and (S)-IPG caprylate were a kind gift of Dr. Frank Dekker
(Department of Pharmaceutical Gene Modulation, Univer-
sity of Groningen, The Netherlands). The sources for the
media components, buffers, and molecular biology reagents,
including PCR, purification, gel extraction, and Miniprep
kits, are reported elsewhere.[5] Oligonucleotides for DNA
amplification were synthesized by Operon Biotechnologies
(Cologne, Germany).
Plasmids, Bacterial Strains and Media
E. coli strain BL21 (DE3) (Stratagene, La Jolla, CA) was
used for cloning and isolation of plasmids, and, in combina-
tion with the pET26b vector (Novagen), for recombinant
protein production. E. coli cells were grown in Luria-Bertani
(LB) medium containing bactotrypton (8%, w/v), yeast ex-
tract (0.5%, w/v) and sodium chloride (0.5%, w/v). When re-
quired, kanamycin (Kan; 30 mgmLÀ1), Difco agar (15 gLÀ1),
and IPTG (1 mM) were added to the medium. DNA se-
quencing was performed by Macrogen (Seoul, Korea). Re-
combinant DNA techniques were performed as described
by Sambrook et al.[19] YbfF wild-type and mutants were pro-
duced in E. coli BL21 (DE3) and purified to homogeneity
using a previously published protocol.[5]
Although the YbfF selectivity towards substrate 3
is high enough for practical applications, further im-
provement might still be possible. A plausible ap-
Library Construction and Screening
The construction of single-site mutagenesis libraries (NDT)
proach would pass by promoting a second round of consisting of 12 naturally occurring amino acids[11] at posi-
tions 25, 124, 185 and 235 in YbfF was performed by mega-
primer PCR, using pET26bYbfFwt as the template, accord-
ing to Miyazaki et al.[20] The resulting plasmids harboring
the mutant ybfF gene were transformed into electrocompe-
tent BL21 (DE3) cells. Transformants were selected on LB/
Kan plates. Colonies were picked with sterile toothpicks
into 96-wells plates (flat bottom, Greiner Bio-one) contain-
ing LB/Kan medium (200 mL per well). Cells were grown
overnight at room temperature with agitation. ZYP-5052
medium[21] (1.3 mL) containing kanamycin was then added
to the wells and the plates were incubated for 1 day at room
temperature, with agitation, to allow protein production. A
portion (50 mL) of each culture was diluted in demi water
(150 mL) and the A600 was determined in a 96-well plate
reader (Multiscan accent, Labsystems). The cells of the re-
maining cultures were harvested by centrifugation (2000ꢂg,
48C, 15 min). The cell pellet was then resuspended in Bug-
buster solution (Novagen, Merck, Germany) (300 mL) and
the plates were incubated at room temperature with agita-
tion for 30 min. Subsequently, 200 mL of BES buffer (5 mM,
pH 7.2) was added to each well and centrifugation was per-
mutagenesis in an iterative manner, according to the
iterative saturation mutagenesis approach (ISM) de-
scribed by Reetz. Another approach would be the se-
lection and mutation of second-sphere residues, resi-
dues that are not in direct contact to the substrate but
still make part of the enzyme binding pocket.[17,18,26]
Conclusions
In summary, the enantioselectivity of YbfF from E.
coli toward caprylate esters of 1,2-O-isopropylidene-
glycerol (IPG) could be improved (up to E=77) by
a structure-based rational design approach. The best
mutant (W235I) has been identified based on molecu-
lar docking experiments, followed by the use of
a screening assay based on the hydrolytic activities to-
wards the substrate (1). The YbfF mutant W235I dis-
plays higher enantioselectivity in hydrolyses of IPG
caprylate esters than the reported B. coagulans CesA formed (2000ꢂg, 48C, 1 hour) to remove unbroken cells
3014
asc.wiley-vch.de
ꢀ 2012 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Adv. Synth. Catal. 2012, 354, 3009 – 3015

An Esterase with Superior Activity and Enantioselectivity towards 1,2-O-Isopropylideneglycerol Esters
and cell debris. Aliquots (20 mL) of the cell free extracts
were transferred into new 96-well plates containing BES
buffer (100 mL, 5 mM, pH 7.2), 4-nitrophenol (1 mM,
pH 7.2), and IPG butyrate (R or S) stock solutions of 5 mM.
The plate was placed in the microplate reader and shaken
for 15 s, the decrease in absorbance at 405 nm was moni-
tored each minute over 30 min, at 308C. Cell-free extracts
prepared from E. coli BL21 (DE3) cells transformed with
empty pET26b vector were used as references. Activities
were determined from slopes in the linear section of the
curve.[13] YbfF mutants of which the enantioselectivity was
potentially increased were selected and purified for further
characterization, and the corresponding genes were se-
quenced.
Dekker (University of Groningen) for the kind gift of the
IPG esters.
References
[1] U. T. Bornscheuer, FEMS Microbiol. Rev. 2002, 26, 73–
81.
[2] D. Bottcher, U. T. Bornscheuer, Curr. Opin. Microbiol.
2010, 13, 274–282.
[3] J. Jurczak, S. Pikul, T. Bauer, Tetrahedron 1986, 42,
447–488.
[4] E. Maconi, R. Gualandris, F. Aragozzini, Biotechnol.
Lett. 1990, 12, 415–418.
Kinetic Resolution of IPG Esters
[5] L. F. Godinho, C. R. Reis, P. G. Tepper, G. J. Poelar-
ends, W. J. Quax, Appl. Environ. Microbiol. 2011, 77,
6094–6099.
[6] S. Y. Park, S. H. Lee, J. Lee, K. Nishi, Y. S. Kim, C. H.
Jung, J. S. Kim, J. Mol. Biol. 2008, 376, 1426–1437.
[7] K. P. Hopfner, E. Kopetzki, G. B. Kresse, W. Bode, R.
Huber, R. A. Engh, Proc. Natl. Acad. Sci. USA 1998,
95, 9813–9818.
[8] F. H. Arnold, Nature 2001, 409, 253–257.
[9] Y. L. Boersma, M. J. Drçge, A. M. van der Sloot, T.
Pijning, R. H. Cool, B. W. Dijkstra, W. J. Quax, Chem-
biochem 2008, 9, 1110–1115.
[10] Y. L. Boersma, M. J. Drçge, W. J. Quax, FEBS J. 2007,
274, 2181–2195.
The procedure for the determination of the enantioselectivi-
ty of YbfF wild-type and mutants towards IPG esters is re-
ported elsewhere.[5] The samples were analyzed by chiral
GC on a Hewlett Packard 5890 series II gas chromatograph,
as described by Drçge et al.[22] The non-enzymatic hydrolysis
of the substrates was negligible under the conditions used.
The enantiomeric excess (ee) of products and substrates was
calculated according to Chen et al.[23] The calculated values
of enantiomeric excess of products and substrates were used
to determine the E-values by using the program Selectivity
(K. Faber, H. Hoenig. ftp://borgc185.kfunigraz.ac.at/pub/
enantio/). The chromatographs data corresponding to the ki-
netic resolution of racemic IPG-caprylate (3) with YbfF
wild-type and mutant W235I are provided as Figure S1 to
Figure S12 in the Supporting Information.
[11] L. G. Otten, F. Hollmann, I. W. Arends, Trends Biotech-
nol. 2010, 28, 46–54.
[12] M. T. Reetz, J. J. Peyralans, A. Maichele, Y. Fu, M.
Maywald, Chem. Commun. 2006, 41, 4318–4320.
[13] L. E. Janes, A. C. Lowendahl, R. J. Kazlauskas, Chem.
Eur. J. 1998, 4, 2324–2331.
Determination of the Kinetic Parameters
The kinetic parameters Km and kcat were determined using
a previously published protocol.[5]
[14] M. T. Reetz, D. Kahakeaw, R. Lohmer, ChemBioChem
2008, 9, 1797–1804.
Molecular Docking
[15] F. Molinari, O. Brenna, M. Valenti, F. Aragozzini,
Enzyme Microb. Technol. 1996, 19, 551–556.
[16] D. Romano, F. Falcioni, D. Mora, F. Molinari, A.
Buthe, M. Schumacher, Tetrahedron: Asymmetry 2005,
16, 841–845.
[17] M. T. Reetz, J. D. Carballeira, Nat. Protoc. 2007, 2,
891–903.
[18] M. T. Reetz, Angew. Chem. 2011, 123, 144–182; Angew.
Chem. Int. Ed. 2011, 50, 138–174.
[19] J. Sambrook, D. W. Russell, Molecular cloning: a labo-
ratory manual, 3nd edn., Cold Spring Harbor Laborato-
ry Press, New York. 2001.
[20] K. Miyazaki, M Takenouchi, Biotechniques 2002, 33,
1033–1038.
[21] F. W. Studier, Protein Expres. Purif. 2005, 41, 207–234.
[22] M. J. Drçge, R. Bos, H. J. Woerdenbag, W. J. Quax, J.
Sep. Sci. 2003, 26, 771–776.
Substrates R-1, S-1, R-3 and S-3 were constructed and mo-
lecular docking simulations were performed in wild-type
YbfF (Protein Data Bank accession code: 3BF7) and the
structural model of the YbfF mutant (W235I), using the
grid-based approach CDOCKER.[24,25] The active site resi-
due W235 of YbfF was mutated to isoleucine in silico. The
mutant enzyme model was energy minimized consisting of
150 steps of steepest descent followed by 500 iteractions of
the adopted basis-set Newton-Raphson algorithm. All struc-
tures containing substrates were further energy minimized
using CHARMm, consisting of 5000 steps of steepest de-
scent followed by 5000 iterations of the adopted basis-set
Newton-Raphson algorithm using an energy tolerance of
0.01 kcalmolÀ1 ꢁÀ1.
[23] C. S. Chen, Y. Fujimoto, G. Girdaukas, C. J. Sih, J. Am.
Acknowledgements
Chem. Soc. 1982, 104, 7294–7299.
[24] J. A. Erickson, M. Jalaie, D. H. Robertson, R. A. Lewis,
M. Vieth, J. Med. Chem. 2004, 47, 45–55.
[25] G. S. Wu, D. H. Robertson, C. L. Brooks, M. Vieth, J.
Comput. Chem. 2003, 24, 1549–1562.
L.F.G. the recipient of an Ubbo Emmius fellowship. G.J.P.
was financially supported by VIDI grant 700.56.421 from the
Division of Chemical Sciences of the Netherlands Organisa-
tion for Scientific Research (NWO-CW). We thank Dr. Frank
[26] L. G. Otten, W. J. Quax, Biomol. Eng. 2005, 22, 1–9.
Adv. Synth. Catal. 2012, 354, 3009 – 3015
ꢀ 2012 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
asc.wiley-vch.de
3015