A new mechanism of enantioselectivity toward chiral primary alcohol by lipase from Pseudomonas cepacia

Article abstract of DOI:10.1016/j.molcatb.2014.08.014

The stereo-recognition of chiral primary alcohols by lipase from Pseudomonas cepacia was found to deviate from earlier observations. Enantioselectivity toward 14 pairs of chiral primary alcohol esters by this lipase was dependent on the existence of an O^non-α(oxygen at non-α-position of the acyloxy group) in the alcohol moiety, and decreased as the size of the acyl moiety increased. Chemical modification on the lipase and molecular dynamics simulations indicated that Tyr²⁹located within the catalytic cavity forms a hydrogen bond with the O^non-αof the preferred enantiomer of the primary alcohol ester. However, a larger acyl moiety suffered stronger hindrance from the catalytic cavity wall of the lipase, pushing the O^non-αaway from Tyr²⁹, and thus weakening the stereo-recognition.

Full text of DOI:10.1016/j.molcatb.2014.08.014

Journal of Molecular Catalysis B: Enzymatic 109 (2014) 109–115
Contents lists available at ScienceDirect
Journal of Molecular Catalysis B: Enzymatic
journal homepage: www.elsevier.com/locate/molcatb
A new mechanism of enantioselectivity toward chiral primary alcohol
by lipase from Pseudomonas cepacia
∗
∗
Xiao Meng, Li Guo, Gang Xu, Jian-Ping Wu , Li-Rong Yang
Institute of Biological Engineering, Department of Chemical and Biological Engineering, Zhejiang University, 310027 Hangzhou, China
a r t i c l e i n f o
a b s t r a c t
Article history:
Received 26 March 2014
Received in revised form 22 August 2014
Accepted 22 August 2014
Available online 1 September 2014
The stereo-recognition of chiral primary alcohols by lipase from Pseudomonas cepacia was found to deviate
from earlier observations. Enantioselectivity toward 14 pairs of chiral primary alcohol esters by this lipase
non-˛
was dependent on the existence of an O
(oxygen at non-˛-position of the acyloxy group) in the alcohol
moiety, and decreased as the size of the acyl moiety increased. Chemical modification on the lipase and
2
9
molecular dynamics simulations indicated that Tyr located within the catalytic cavity forms a hydrogen
non-˛
bond with the O
of the preferred enantiomer of the primary alcohol ester. However, a larger acyl
Keywords:
non-˛
moiety suffered stronger hindrance from the catalytic cavity wall of the lipase, pushing the O
from Tyr , and thus weakening the stereo-recognition.
away
Chemical modification
Molecular dynamics
Enantioselectivity
2
9
©
2014 Elsevier B.V. All rights reserved.
Chiral primary alcohol
Pseudomonas cepacia lipase
1
. Introduction
computational tools such as molecular docking and QSAR (quanti-
tative structure–activity relationship) calculation [9,10].
Chiral alcohols, including secondary and primary alcohols, are
Nonetheless, the success of these studies was only based on
secondary alcohols, whose chiral center is the hydroxyl ˛-C of
a common type of optically active chemicals. As chiral build-
ing blocks or chiral sources, secondary and primary alcohols play
important roles in medicine, cosmetic production, and food chem-
istry. Lipase-catalyzed resolution is an effective method to produce
optically pure enantiomers [1–3]. Therefore, the mechanism of
chiral-recognition of alcohols by lipases has become a research
hotspot over the last few years. Kazlauskas et al. [4] investigated the
resolution of 94-pairs of enantiomers of chiral alcohols catalyzed
by lipases including PcL (Pseudomonas cepacia lipase), and pro-
posed an empirical law of enantiopreference: the stereo-specific
catalytic cavity of the enzyme contains large and small pockets
that fit, respectively, the large and small group on the chiral cen-
ter of the preferred enantiomer. This groundbreaking law was
considered to be the origin of enantioselectivity by lipases [5,3].
Based on this law and using various genetic approaches, a number
of studies on improving, manually altering lipase’s enantioselec-
tivity and even reversing the enantiopreference were reported
˛
the alcohol, the C . Research, either genetic or computational, on
alc
stereo-recognition of primary alcohols, whose chiral center is not
˛
alc
the C , has been limited. These alcohols could only be resolved
by a limited number of lipases including PcL [11,12]. Although by
these lipases, only a small number of primary alcohols could be
recognized, and even toward this small number of alcohols, the
lipases’ enantiopreference did not obey Kazlauskas’ Rule [13–17].
In the present contribution, enantioselectivity by PcL toward 14
pairs of different chiral primary alcohol esters was investigated.
The mechanism behind the deviant behavior of stereo-recognition
of these substrates was studied by chemical modification of PcL,
MALDI-TOF mass spectrometry, and molecular dynamics simula-
tion. As a result, a new mechanism for enantioselectivity of lipase
is proposed.
2. Materials and methods
[
6–8]. Moreover, researchers were able to quantitatively predict
with reasonable precision the enantioselectivity ratio of particu-
lar lipase-catalyzed resolution of chiral compounds using modern
2.1. Materials
PcL (Pseudomonas cepacia lipase in powder form, commer-
cially Lipase PS) was purchased from Amano Enzyme Inc. Chiral
primary alcohols including 2, 3, 3-trimethyl-1-butanol, 2-phenyl-
1-propanol, 2-phenyl-1-butanol, 2-methoxymethyl-1-butanol,
2-methoxy-1-butanol and 2-phenoxy-1-propanol together with
∗ Corresponding authors. Tel.: +86 0571 87952363; fax: +86 0571 87952363.
E-mail addresses: wjp@zju.edu.cn (J.-P. Wu), lryang@zju.edu.cn (L.-R. Yang).
http://dx.doi.org/10.1016/j.molcatb.2014.08.014
1
381-1177/© 2014 Elsevier B.V. All rights reserved.

1
10
X. Meng et al. / Journal of Molecular Catalysis B: Enzymatic 109 (2014) 109–115
analytical samples of their optically pure enantiomers were
purchased from Tokyo Chemical Industry Co. Ltd. Modifica-
tion reagents, including CHD (1, 2-cyclohexanedione); DEPC
AKTA Prime protein purifier (GE Healthcare, Little Chalfont, Buck-
inghamshire, UK) equipped with a HiTrap desalting column (GE
Healthcare, Little Chalfont, Buckinghamshire, UK).
(
diethyl pyrocarbonate); EDA (ethylenediamine); EDC (1-(3-
The NAI-modified Tyr can be recovered by hydroxylamine: the
reaction conditions were: 1 mL of the desalted PcL was mixed with
30 ␮L hydroxylamine (50% aqueous solution) and reacted for 1 h.
The resulting mixture was vacuum frozen for 12 h.
dimethylaminopropyl)-3-ethylcarbodiimide hydrochloride);
HNBB (2-hydroxy-5-nitrobenzyl bromide); IAA (iodoacetic acid);
ICl (iodine monochloride); NAI (N-acetylimidazole); TNBS (2, 4,
6
-trinitrobenzenesulfonic acid); TNM (tetranitromethane), were
In addition to the standard reaction conditions of the chemical
modification method referred to above, n-hexane was employed as
an activator of PcL, because lipases are widely known to be activated
at the aqueous-hydrophobic interface [28–30]. In the modification
reaction coupled with activation, 10% (V/V) of n-hexane was added
to the mixture under the same reaction condition.
also purchased from Tokyo Chemical Industry Co. Ltd. All other
chemicals referred in this paper were of analytical grade.
2.2. Synthesis of substrates
Esters of chiral primary alcohols were synthesized from
their corresponding alcohols. Alcohols and triethylamine (molar
ratio = 1:1.2) were mixed in an ice bath, and acetyl chloride was
slowly added. The reaction mixture was stirred overnight. The end
of reaction was determined by TLC (thin layer chromatography).
The product was washed with a saturated sodium bicarbonate solu-
tion several times, and residual water was removed by anhydrous
sodium sulphite.
2.5. Protein mass spectrum
Chemically modified PcL was isolated by SDS-PAGE according
to Vandahl et al. [31]. The resulting denatured pure PcL was then
◦
digested by trypsin at 37 C overnight, according to Chen et al. [32].
The digested peptide slices were estimated by ExPASy Peptide-
Mass (http://web.expasy.org/peptide mass/). Finally, the sample
was lyophilized using a Labconco FreeZone 18 freeze drier (Lab-
conc Corporation, Kansas City, MO, USA) and afterwards dissolved
in 20 ␮L of acetonitrile with 0.1% trifluoroacetic acid (1: 9, V/V) for
subsequent mass spectrometric analysis according to Zehl et al.
[33]. ˛-Cyano-4-hydroxycinnamic acid was used as the matrix
and ions were extracted into the linear UltraFleXtreme MALDI-
TOF/TOF spectrometer (Bruker Corporation, Boston, MA, USA) using
an extraction potential of 20 kV in the high-mass detection mode
according to Chang et al. [34]. Peptide mass fingerprinting data was
also employed to search in NCBI non-redundant protein databases
using the Mascot software (Matrix Science Inc., Boston, MA, USA).
2.3. Resolution of primary alcohol esters
1
0 mL of a mixed solution of crude PcL powder (10 mg mL−1),
−
1
substrate (ester of chiral primary alcohol, 100 mmol L ) and PBS
(
phosphate buffer solution, 100 mmol L−1, pH 7.0, including 10%
V/V isopropanol as co-solvent) was added into a 20 mL flask incu-
◦
bated at 27 C (300 K) under 200 rpm stirring. The reaction time was
controlled as the conversion ratio was between 20 and 30%. The
product was extracted by an equal volume of n-octanol. Substrates
with aryl groups (2–7, 10–14) were analyzed by an Agilent 1100
HPLC equipped with an Chiralpak AD-H column, whereas those
without an aryl group (1, 8 and 9) were analyzed by an HP 1890 GC
equipped with a Varian CP-Cyclodextrin-B-2, 3, 6-M-19 capillary
2.6. Simulation
GC column. The initial reaction rate = c·[Ester] ·V/ꢀt, where c is the
The XRD (X-ray distraction) crystalline structure of PcL (PDB
ID: 2NW6) was used as a starting structure of the modeled
lipase. The ligand ((phenoxymethyl)propyl methylphosphono-
chloridoate, named POT in the file) and crystal water molecules
(named HOH in the file) in the PDB file, were manually deleted
in the Tripos SYBYL-X 1.1 (Tripos Associates, St. Louis, MO, USA)
workspace window. Then a pKa calculation for the titratable
residues of the protein at pH 7 was performed using the online
0
conversion of the substrate, herein c should be less than 5%; [Ester]
0
−
1
is the initial molar concentration of the substrate (100 mmol L );
V is the volume of the reactant (10 mL); and ꢀt is the reac-
tion time. The enantioselectivity ratio (E value) was calculated
according to Chen et al. [18]: E = ln{1 − c(1 + eep)}/ln{1–c(1 − eep)},
where
c is the conversion of the substrate; and eep is
the enantiomer excess, eep = |[(S)-alcohol] − [(R)-alcohol]|/{[(S)-
alcohol] + [(R)-alcohol]} × 100%. In the formulas, the quantity in the
square brackets denotes molar concentration.
2
tool PCE [35]. Accordingly, the side-chain carboxyl groups of Asp ,
2
1
36
55
56
102
121
130
159
228
118
Asp , Asp , Asp , Asp , Asp , Asp , Asp , Asp , Asp
,
,
Asp , Asp , Asp , Asp288, Asp303, Glu28, Glu35, Glu63, Glu
236
242
264
197
289
and Glu302 were deprotonated and charged with
2.4. Chemical modification of the lipase
Glu , Glu
1.00; phenoxy oxygen atoms (O ) of Tyr , Tyr , Tyr , Tyr , Tyr
ꢁ
4
9
23
274
29
31
−
,
45
68
95
129
175
179
207
282
The lipase powder adopted in the article was a semi-purified
Tyr , Tyr , Tyr , Tyr , Tyr , Tyr , Tyr , Tyr
and Tyr
were protonated; His15, His86, His114, His
204
and His
311
were set to
form and contains only PcL protein and some salts. Before modifica-
tion, the powder was dissolved in deionized water, and added into a
centrifuge tube (10 mL) with ultrafiltration membrane (molecular
ı
ε
the HIE state (each with a deprotonated N and a protonated N );
side-chain amino groups of Lys , Lys , Lys , Lys , Lys , Lys
22
70
80
165
269
283
◦
316
and guanidyl groups of Arg , Arg , Arg , Arg , Arg115,
8
40
61
94
weight cut off = 10 kDa), centrifuged at 12,000 rpm at 4 C for 5 min
and Lys
Arg258, Arg297, Arg , Arg
309
314
were protonated and charged with
to remove the salts. The purified lipase was afterwards dissolved
in buffer solutions for modification study. Arg (Arginine residue)
was modified by CHD according to Patthy et al. [19]; Asp (aspartic
acid residue), Glu (glutamic acid residue) and Lys (lysine residue)
was modified by EDA, EDA and TNBS according to Barbosa et al.
+1.00; the N-terminal amino group was protonated and charged
with +1.00 and the C-terminal carboxyl group was deprotonated
2
86
and charged with −1.00. Additionally, the catalytic His
was set
ı
ε
to the HIP state (with both N and N protonated) with a charged
state of +1.00, and the Oꢂ of the catalytic Ser was deprotonated.
The tetrahedral intermediates of the chiral primary alcohol
esters’ enantiomers were generated by SYBYL-X. Starting with the
resulting substrate structure and the enzyme structure prepared
above, the substrate-enzyme complex in which acyloxy carbon
atom of the tetrahedron covalently bonded to the catalytic group
(deprotonated Ser87-O ) was elucidated by the covalent docking
module FlexX of SYBYL-X according to Chen et al. [32]
87
[
20] and Geoghegan [21]; His (histidine residue) was modified by
DEPC according to Miles et al. [22]; Met (methionine residue) was
modified by IAA according to Gundlach et al. [23]; Trp (tryptophan
residue) was modified by HNBB according to Hortan et al. [24];
Tyr (tyrosine residue) was modified by NAI, ICl and TNM accord-
ing to Cacace et al. [25], Koshland et al. [26] and Riordan et al.
ꢂ
[
27]. All modification processes were carried out in an ice bath to
avoid thermal inactivation. The modified PcL was desalted using an

X. Meng et al. / Journal of Molecular Catalysis B: Enzymatic 109 (2014) 109–115
111
Molecular dynamic simulation procedures were performed
by the AMBER 11 program [36] using the ff99SB force field.
A TIP3P (transferable intermolecular potential 3P) [37,38] cubic
periodic boundary water box was set up using a minimal dis-
+
tance 10 A˚ between the box boundary and the protein. 22 Na
−
and 16 Cl ions were added to give a net charge of zero. Pro-
cedures of energy minimization, heating (from 0 to 300 K) and
density equilibrium were performed in turn. Subsequently, a
1
2.5 ns dynamics simulation was performed at constant temper-
ature (300 K) and constant pressure (1 bar) using a time step
of 2 fs. The trajectories and conformations of the system were
recorded every 2 ps. The first 2.5 ns was for dynamics equilibra-
tion, and the remaining 10 ns was taken for further interaction
analysis. During the simulation, the pressure was controlled using
the Berendsen algorithm [39]. SHAKE algorithm [40] was applied
to all bonds. Electrostatic interactions were calculated using the
Particle Mesh Ewald [41] method with a non-bonded cutoff of
1
2 A˚ .
3
. Results
Fig. 1. Deacylation of NAI modified PcL.
3.1. Resolution of primary alcohol esters
PcL-catalyzed resolution of 14 pairs of different chiral primary
the chiral center of the substrate, and there was no (R)-preference
commonly observed in lipase resolution of secondary alcohols
either. What should be mentioned is, according to 8–14 vs. 1–7
in Table 1, enantioselectivity of PcL toward primary alcohol esters
alcohols via hydrolysis of their corresponding esters was investi-
gated (Table 1). The results indicate that the enantioselectivity was
not related to the difference between groups directly connected to
Table 1
PcL’s enantioselectivity toward chiral primary alcohols with different structure.
.
Entry
L
M
Acyl
Initial reaction rate (␮mol min−1)
E
Enantio-preference
1
2
3
4
5
6
7
8
9
0
1
2
3
4
t-Bu
Ph
Ph
Ph
Ph
CH3
CH3
CH3
CH3
CH3
CH3
CH2CH3
CH2CH3
CH2CH3
CH3
COCH3
COCH3
0.32 ± 0.01
0.60 ± 0.02
3.1 ± 0.1
(S)
(S)
(S)
(S)
(S)
(S)
(S)
(S)
1.7 ± 0.1
1.2 ± 0.1
1.0 ± 0.1
1.0 ± 0.1
1.0 ± 0.1
2.0 ± 0.1
42 ± 2
CO(CH2)2CH3
CO(CH2)4CH3
CO(CH2)6CH3
CO(CH2)10CH3
COCH3
COCH3
COCH3
COCH3
0.16 ± 0.01
0.072 ± 0.004
0.021 ± 0.002
0.010 ± 0.001
0.34 ± 0.01
Ph
Ph
CH3OCH2
CH3O
PhO
PhO
PhO
PhO
PhO
0.86 ± 0.02
a
1.45 ± 0.03
16 ± 1
(S)
a
1
1
1
1
1
0.52 ± 0.02
33 ± 1
(S)
a
CH3
CH3
CH3
CH3
CO(CH2)2CH3
CO(CH2)4CH3
CO(CH2)6CH3
0.11 ± 0.01
7 ± 1
(S)
a
0.025 ± 0.002
0.012 ± 0.001
0.011 ± 0.001
1.7 ± 0.1
1.2 ± 0.1
1.1 ± 0.1
(S)
a
(S)
1
0
a
CO(CH2) CH3
(S)
a
Due to the oxygen at the stereo center, the shape of enantiomers of 8–14 denoted as (S)-conformation was opposite to the shape of (S)-1–7.
Table 2
Chemical modification on PcL.
Entry
Modifier
Target residue
E toward 10
E toward 2
With n-hexane
Without n-hexane
With n-hexane
Without n-hexane
1
2
3
4
5
6
7
8
9
Control
CHD
DEPC
EDA
EDC
HNBB
IAA
ICl
NAI
TNBS
TNM
–
Arg
His
Asp/Glu
Asp/Glu
Trp
Met
His/Tyr
Tyr
33 ± 1
33 ± 1
30 ± 1
28 ± 1
33 ± 1
35 ± 1
30 ± 1
Inactivated
34 ± 1
18 ± 1
35 ± 1
36 ± 2
1.7 ± 0.1
1.2 ± 0.1
Inactivated
1.8 ± 0.1
1.8 ± 0.1
1.3 ± 0.1
Inactivated
1.6 ± 0.1
1.7 ± 0.1
1.7 ± 0.1
1.8 ± 0.1
1.7 ± 0.1
1.1 ± 0.1
1.2 ± 0.1
1.8 ± 0.1
1.8 ± 0.1
1.3 ± 0.1
Inactivated
1.7 ± 0.1
1.7 ± 0.1
1.4 ± 0.1
1.8 ± 0.1
30 ± 1
Inactivated
33 ± 1
35 ± 1
32 ± 1
Inactivated
36 ± 1
9 ± 0.7
35 ± 1
1
1
0
1
Lys
Tyr
42 ± 2

1
12
X. Meng et al. / Journal of Molecular Catalysis B: Enzymatic 109 (2014) 109–115
Table 3
MALDI-TOF protein mass spectrum of NAI-modified PcL.
Entry
Z/A
Matched slice
Modified Tyr
Measured
Predicted
Error
1
2
3
4
5
6
7
8
9
882.1
924.7
881.4
924.4
0.7
0.3
0.8
0.7
1.0
0.1
0.2
0.8
0.3
0.4
1.0
0.9
001–008
001–008 + Ac
259–269
298–309
270–283
284–297
041–061
041–061 + Ac
095–115
095–115 + Ac
–
004
–
–
–
–
–
045
–
095
023, 029, 031
023, 029, 031
a
1048.3
1212.3
1520.7
1706.8
2125.2
2168.8
2138.5
2180.8
2318.0
2360.9
1047.5
1211.6
1519.7
1706.9
2125.0
2168.0
2138.2
2181.2
2317.0
2360.0
1
1
1
0
1
2
023–040 + 2 Ac
023–040 + 3 Ac
a
Ac, acetyl group.
with Onon-˛ (oxygen atom at non-˛-position of the acyloxy group)
was far better than those without. Besides, according to 10–14
in Table 1, as the acyl moiety of the substrate increased in size
enantioselectivity decreased while acyl moiety growing, which
is opposite to that observed for secondary alcohols. Substrate 10
and 2 were taken as model substrates for further investigation
because they have similar structures yet large differences in the
enantioselectivity.
(
2–6, 10–14), the initial reaction rates decreased accordingly. This
tendency is consistent with many studies before. However, the
Fig. 2. Conformation of substrate–enzyme complex resulting from MD simulation. In the figure, all non-polar hydrogen atoms have been hidden to avoid visual disturbance.

X. Meng et al. / Journal of Molecular Catalysis B: Enzymatic 109 (2014) 109–115
113
Table 4
Determination of the distance between atoms of the potential {PcL-Tyr -O -H· · ·O
2
9
ꢁ
non-˛
-substrate} hydrogen bond.
Entry
Substrate
Enzyme
Measured atoms
Distance (Å)
R)
(
(S)
1
2
1
2
Native PcL
Native PcL
10.4
8.6
8.0
7.3
Tyr29-O -acetylated PcL
ꢁ
5.2
5.8
9.1
6.5
3
4
2
7
Native PcL
5
6
8
9
Native PcL
Native PcL
6.1
6.1
3.5
3.8
7
8
10
10
Native PcL
5.5
7.8
3.5
7.6
Tyr29-O -acetylated PcL
ꢁ
3.2. Chemical modification of PcL
3.3. Molecular dynamics simulations
To define regions related to the enantioselectivity toward 10
The complexes of PcL and the substrates were simulated using
molecular dynamics (MD) computing. According to the result-
ing conformations generated by 10 ns of MD, Oꢂ of (S)-8, 9 and
10 (which were the preferred enantiomers according to Table 1)
formed a hydrogen bond with Tyr29-O H (Fig. 2 row 3, column
1–3), but (R)-enantiomers did not (Fig. 2 row 4, column 1–3). Both
(R)- and (S)-10 enantiomers did not form a hydrogen bond with
acetylated-Tyr29-O of NAI-modified PcL (Fig. 2 row 3 and 4, col-
and 2, PcL was treated with 10 different common protein modifiers
respectively (Table 2). It is widely accepted that hydrophobic
environment induces the exposing of lipase’s cavity of catalytic
center by removing the ˛-helix covering it [23]. Therefore during
the modification process, n-hexane was adopted as an activator of
PcL to modify residues inside the catalytic cavity normally isolated
from the outer solvent environment. In the absence of n-hexane,
almost none of the modifiers influenced the enantioselectivity of
PcL toward either 10 or 2. In contrast, in the presence of n-hexane,
enantioselectivity toward 2 remained hardly affected by any modi-
fication, but enantioselectivity toward 10 was affected significantly
by modifications targeting tyrosine residues. ICl and TNM (by
ortho-substitution of the phenolic hydroxyl group) increased PcL’s
enantioselectivity toward 10 from 33 to 36 and 42 respectively. NAI
ꢁ
ꢁ
29
ꢁ
non-˛
umn 4). Determination of the {PcL-Tyr -O -H· · ·O
-substrate}
distance (Table 4) showed that only in the case of (S)-8, 9 and
2
9
ꢁ
non-˛
10 + native PcL, was the {PcL-Tyr -O -H· · ·O
-substrate} dis-
tance in the acceptable range to form a hydrogen bond (2.7∼3.7 A˚ ).
In all other cases ((R)- or (S)-1, 2 and 7 + native PcL, (R)-8, 9 and
10 + native PcL, (R)- or (S)-8, 9 and 10 + Tyr29-O -acetylated PcL)
the distance was too large to form a hydrogen bond.
ꢁ
(
by acetylating the phenolic hydroxyl group) decreased the value
to 9, which is less than a third of that of native PcL. MALDI-TOF-MS
of the NAI-modified PcL in the presence of n-hexane (Table 3)
showed that the only tyrosine residue in the catalytic cavity of PcL,
Tyr , was actually acetylated. Hydroxylamine treatment of NAI-
modified PcL restored the modified Tyr to a free phenolic hydroxyl
hydrogen, and the enantioselectivity was recovered (Fig. 1).
4
. Discussion
The deviant behaviors from both alcohol and acyl moiety of the
29
primary alcohol ester indicated that stereo-recognition by PcL for
these type of substrates is decided by some non-Kazlauskas’ shape-
determined interaction between the O
non-˛
of the substrate and

1
14
X. Meng et al. / Journal of Molecular Catalysis B: Enzymatic 109 (2014) 109–115
above that PcL-Tyr29-OH and the Onon-˛ of the primary alcohol
ester to a great extent decided the enantioselectivity, we pos-
tulate that increasing the length of the acyl moiety negatively
2
9
ꢁ
non-˛
effects to the formation of the {PcL-Tyr -O -H· · ·O
-substrate}
hydrogen bond. As shown in Fig. 3a, in order to determine this
“
negative effect” quantitatively by MD simulations, the preferred
non-˛
enantiomer of chiral primary alcohol esters with O
and dif-
ferent acyl moiety lengths ((S)-10, 11, 12, 13 and 14, respectively
with C , C , C , C and C12 acyl chain lengths) were respectively
2
4
6
8
docked into the catalytic center of PcL. The resulting conforma-
tion of the substrate-enzyme complexes were simulated using
2
9
ꢁ
non-˛
MD, during which the {PcL-Tyr -O -H· · ·O
-substrate} dis-
tance was manually restricted to 2.7–3.7 A˚ so as to form the
hydrogen bond. Because the selected long acyl chain substrates
would not readily form a hydrogen bond with Tyr29, the manual
restriction was hypothesized to cause unreasonable structure of the
simulated substrate-enzyme complex, and therefore there must
be correlation between the unreasonability feature and the enan-
tioselectivity. The MD result did match the hypothesis. As shown
1
˛
ˇ
in Fig. 3b, the {C
-Oalc-C -C } dihedral displayed a clear con-
acyl
alc alc
sistent correlation to the enantioselectivity ratio, confirming that
the chiral primary alcohol ester with a larger acyl moiety suffered
stronger steric hindrance from the acyl side of the PcL catalytic cav-
non-˛
29
ity, thereby pushing the O
further away from Tyr -OH and
thus decreasing the enantioselectivity.
5. Conclusion
In summary, PcL stereo-recognition of chiral primary alcohols
2
9
(and their esters) is mostly decided by the possibility and extent
Fig. 3. Determination of the acyl moiety’s negative effort on forming the Tyr
1
˛
alc
ˇ
non-˛
hydrogen bond. (a) The {C -Oalc-C -C } dihedral to be determined. In the figure,
of the formation of a hydrogen bond between the O
of the
acyl
alc
“
Acyl Pocket” “Alcohol Pocket M” and “Alcohol Pocket L” are substructure domains
29
substrates and the Tyr -OH of the lipase. Moreover, a larger acyl
of PcL’s catalytic cavity according to Kazlauskas’ Law of lipase’s enantiopreference
toward chiral secondary alcohols. (b) Values of the dihedral during the 10 ns MD
simulation. In the chart, spheres denoted enantioselectivity ratio, squares denoted
non-˛
29
moiety was found to push the O
further away from Tyr -OH,
thereby reducing the likelihood of this key hydrogen bond, and thus
decreasing the enantioselectivity. This mechanism rather satisfac-
torily explains the lipase’s deviant behavior of stereo-recognition of
chiral primary alcohols which does not follow the former empirical
rules and laws for secondary alcohols.
1
˛
ˇ
the {C -Oalc-C -C } dihedral.
acyl
alc
alc
certain structure features of the lipase, and the growing of sub-
strate’s acyl moiety made a negative effort to such an interaction.
Modification of PcL (Table 3) suggested that, the weaken-
2
9
ing of Tyr ability as a hydrogen donor (via NAI modification)
decreased the enantioselectivity toward 10, while strengthening
the hydrogen donor capability of Tyr29 (via ICl or TNM modifica-
tion) increased the enantioselectivity toward 10. However, these
Acknowledgements
This work was financially supported by the National Basic
Research Program of China (973 Program, No. 2011CB710800), Hi-
Tech Research and Development Program of China (863 Program,
2011AA02A209), National Natural Science Foundation of China (No.
20936002), and the Science and Technology Planning Project of
Zhejiang Province (No. 2010C31127).
modifications had negligible effect on enantioselectivity toward
non-˛
2. The only structural difference between 10 and 2 is the O
,
which is also very important to the enantioselectivity of native PcL
toward primary alcohol esters (see entries 8–14 vs. entries 1–7 in
non-˛
Table 1). The significance of the presence of this O
in chiral
primary alcohol substrates, together with the influence by Tyr29’s
ability as a hydrogen donor was sufficient to infer that a hydrogen
References
2
9
non-˛
bond between Tyr -OH and the O
of the substrate enantiomer
[
1] A. Ghanem, H.Y. Aboul-Enein, Chirality 17 (2005) 1–15.
would aid significantly the stereo-recognition by PcL toward the
primary alcohol ester. This indication was strongly supported by
the MD simulation (Fig. 2 and Table 4), which showed that only
the PcL preferred enantiomers ((S)-conformation) of 8, 9 and 10
[2] A. Ghanem, H.Y. Aboul-Enein, Tetrahedron Asymmetry 15 (2004) 3331–3351.
[
[
3] V. Gotor-Fernández, R. Brieva, V. Gotor, J. Mol. Catal. B: Enzym. 40 (2006)
11–120.
4] R.J. Kazlauskas, A.N.E. Weissfloch, A.T. Rappaport, L.A. Cuccia, J. Org. Chem. 56
(1991) 2656–2665.
1
29
[5] R.J. Kazlauskas, A.N.E. Weissfloch, J. Mol. Catal. B: Enzym. 3 (1997) 65–72.
were able to form such a hydrogen bond with Tyr (Fig. 2 row 3,
columns 1–3, and Table 4 entries 5–7, column (S)). In contrast, the
non-preferred (R)-8, 9 and 10 did not form any hydrogen bond with
[
[
6] F. Haffner, T. Norin, Chem. Pharm. Bull. 47 (1999) 591–600.
7] G. Cedrone, A. Menez, E. Quemeneur, Curr. Opin. Struct. Biol. 10 (2000) 405–410.
[8] J.L. Harris, C.S. Craik, Curr. Opin. Chem. Biol. 2 (1998) 127–132.
[9] J. Liu, X.-L. Tang, H.-W. Yu, Biotechnol. Bioeng. 105 (2010) 687–696.
10] T. Schulz, J. Pleiss, R.D. Schmid, Protein Sci. 9 (2000) 1053–1062.
11] C. Orrenius, F. Hbffner, D. Rotticci, T. Norin, K. Hult, Biocatal. Biotransform. 16
(1998) 1–15.
29
Tyr (Fig. 2 row 4, column 1, 2 and 3, and Table 4 entries 5–7, col-
[
[
non-˛
umn (R)). Moreover, as enantiomers of 1, 2 and 7 lack the O
to form a hydrogen bond with Tyr29-OH, PcL could not distinguish
these pairs of primary alcohol esters and had poor enantioselectiv-
ity toward them.
[
[
12] J. Zuegg, H. Hönig, J.D. Schrag, M. Cygler, J. Mol. Catal. B: Enzym. 3 (1997) 83–98.
13] P. Ferraboschi, S. Casati, S. Degrandi, P. Grisenti, E. Santaniello, Biocatalysis 10
(
1994) 279.
One more thing left to discuss is the influence of the acyl moiety
of the substrate on enantioselectivity. Since we have mentioned
[14] G. Guanti, L. Banfi, E. Narisano, J. Org. Chem. 57 (1992) 1540–1554.
[15] J.B. Jones, Tetrahedron Lett. 33 (1992) 1399–1402.

X. Meng et al. / Journal of Molecular Catalysis B: Enzymatic 109 (2014) 109–115
115
[
16] Z. Wimmer, A suggestion to the PPL active site model dilemma, Tetrahedron
[31] B.B. Vandahl, S. Birkelund, H. Demol, B. Hoorelbeke, G. Christiansen, J. Vande-
kerckhove, K. Gevaert, Electrophoresis 22 (2001) 1204–1223.
[32] H. Chen, J.-P. Wu, L.-R. Yang, G. Xu, Biochim. Biophys. Acta 1834 (2013)
2494–2501.
4
8 (1992) 8431–8436.
[17] Z.F. Xie, H. Suemune, K. Sakai, Tetrahedron Asymmetry 4 (1993) 973–980.
18] C.S. Chen, Y. Fujimoto, G. Girdaukas, C.-J. Sih, J. Am. Chem. Soc. 104 (1982)
[
7
294–7299.
[33] Z.M. Zehl, L. Ivana, A. Marija, R. Andreas, K.P. Biserka, J. Mass Spectrom. 39
(2004) 1474–1483.
[34] C. Chang, X. Liu, K. Chen, Adv. Biosci. Biotechnol. 9 (2011) 123–127.
[35] M.A. Miteva, P. Tuffery, B.O. Villoutreix, Nucleic Acids Res. 33 (2005)
W372–W375.
[
[
19] L. Patthy, E.L. Smith, J. Biol. Chem. 250 (1975) 565–569.
20] O. Barbosa, M. Ruiz, C. Ortiz, M. Fernández, R. Torres, R. Fernandez-Lafuente,
Process Biochem. 47 (2012) 867–876.
[
21] K.F. Geoghegan, D.T. Osuga, A.I. Ahmed, Y. Yeh, R.E. Feeney, J. Biol. Chem. 255
(
1980) 663–667.
[36] D.A. Case, T.A. Darden, T.E. Cheatham III, C.L. Simmerling, J. Wang, R.E. Duke,
R. Luo, R.C. Walker, W. Zhang, K.M. Merz, B. Roberts, S. Hayik, A. Roitberg, G.
Seabra, J. Swails, A.W. Goetz, I. Kolossváry, K.F. Wong, F. Paesani, J. Vanicek, R.M.
Wolf, J. Liu, X. Wu, S.R. Brozell, T. Steinbrecher, H. Gohlke, Q. Cai, X. Ye, J. Wang,
M.J. Hsieh, G. Cui, D.R. Roe, D.H. Mathews, M.G. Seetin, R. Salomon-Ferrer, C.
Sagui, V. Babin, T. Luchko, S. Gusarov, A. Kovalenko, P.A. Kollman, AMBER 11,
University of California, San Francisco, 2011.
[
[
[
[
[
22] E.W. Miles, Methods Enzymol. 47 (1977) 431–442.
23] H.G. Gundlach, S. Moore, W.H. Stein, J. Biol. Chem. 234 (1959) 1761–1764.
24] H.R. Hortan, D.E. Koshland Jr., J. Am. Chem. Soc. 87 (1965) 1126–1132.
25] M.G. Cacace, G.D. Prisco, R. Zito, FEBS Lett. 62 (1976) 338–341.
26] M.E. Koshland, F.M. Englberger, M.J. Erwin, S.M. Gaddone, J. Biol. Chem. 238
(
1963) 1343–1348.
[
[
27] J.F. Riordan, M. Sokolovsky, B.L. Vallee, Biochemistry 6 (1967) 358–361.
28] A.M. Brozozowski, U. Derewenda, Z.S. Derewenda, G.G. Dodson, D.M. Lawson,
J.P. Turkenburg, F. Bjorkling, B. Hugejensen, S.A. Patkar, L. Thim, Nature 351
[37] W.L. Jorgensen, J. Chandrasekhar, J.D. Madura, R.W. Impey, M.L. Klein, J. Chem.
Phys. 79 (1983) 926.
[38] E. Neria, S. Fischer, M. Karplus, J. Chem. Phys. 105 (1996) 1902.
[39] B. Hess, H. Bekker, H.J.C. Berendsen, J.G.E.M. Fraaije, J. Comput. Chem. 18 (1997)
1463–1472.
(
1991) 491.
29] U. Derewenda, A.M. Brzozowski, D.M. Lawson, Z.S. Derewenda, Biochemistry
1 (1992) 1532–1541.
[
[
3
[40] J.P. Rychaert, G. Ciccotti, H.J.C. Berendsen, J. Comput. Phys. 23 (1977) 327–341.
[41] T. Darden, D. York, L. Pedersen, J. Chem. Phys. 98 (1993) 10089.
30] P. Grochulski, Y. Li, J.D. Schrag, F. Bouthillier, P. Smith, D. Harrison, B. Rubin, M.
Cygler, J. Biol. Chem. 268 (1993) 12843–12847.