Inverting the Enantiopreference of Nitrilase-Catalyzed Desymmetric Hydrolysis of Prochiral Dinitriles by Reshaping the Binding Pocket with a Mirror-Image Strategy

Article abstract of DOI:10.1002/anie.202012243

A mirror-image strategy, that is, symmetry analysis of the substrate-binding pocket, was applied to identify two key amino acid residues W170 and V198 that possibly modulate the enantiopreference of a nitrilase from Synechocystis sp. PCC6803 towards 3-isobutyl glutaronitrile (1 a). Exchange of these two residues resulted in the enantiopreference inversion (S, 90 % ee to R, 47 % ee). By further reshaping the substrate-binding pocket via routine site-saturation and combinatorial mutagenesis, variant E8 with higher activity and stereoselectivity (99 % ee, R) was obtained. The mutant enzyme was applied in the preparation of optically pure (R)-3-isobutyl-4-cyanobutanoic acid ((R)-2 a) and showed similar stereopreference inversion towards a series of 3-substituted glutaronitriles. This study may offer a general strategy to switch the stereopreference of other nitrilases and other enzymes toward the desymmetric reactions of prochiral substrates with two identical reactive functional groups.

Full text of DOI:10.1002/anie.202012243

A Journal of the Gesellschaft Deutscher Chemiker
Angewandte
Chemie
International Edition
www.angewandte.org
Accepted Article
Title: Inverting the enantiopreference of nitrilase-catalyzed
desymmetric hydrolysis of prochiral dinitriles by reshaping the
binding pocket with a “mirror-image” strategy
Authors: Shanshan Yu, Jinlong Li, Peiyuan Yao, Jinhui Feng, Yunfeng
Cui, Jianjiong Li, Xiangtao Liu, Qiaqing Wu, Jianping Lin, and
Dunming Zhu
This manuscript has been accepted after peer review and appears as an
Accepted Article online prior to editing, proofing, and formal publication
of the final Version of Record (VoR). This work is currently citable by
using the Digital Object Identifier (DOI) given below. The VoR will be
published online in Early View as soon as possible and may be different
to this Accepted Article as a result of editing. Readers should obtain
the VoR from the journal website shown below when it is published
to ensure accuracy of information. The authors are responsible for the
content of this Accepted Article.
To be cited as: Angew. Chem. Int. Ed. 10.1002/anie.202012243
Link to VoR: https://doi.org/10.1002/anie.202012243

10.1002/anie.202012243
Angewandte Chemie International Edition
RESEARCH ARTICLE
Inverting the enantiopreference of nitrilase-catalyzed desymmetric
hydrolysis of prochiral dinitriles by reshaping the binding pocket
with a “mirror-image” strategy
Shanshan Yu+, Jinlong Li+, Peiyuan Yao+, Jinhui Feng, Yunfeng Cui, Jianjiong Li, Xiangtao Liu, Qiaqing
Wu,* Jianping Lin,* and Dunming Zhu*
Abstract: Nitrilase-catalyzed desymmetric hydrolysis of prochiral
dinitriles is an attractive method for the synthesis of chiral
cyanocarboxylic acids, the precursors to pharmaceutically important
-amino acids. However, the tested nitrilases towards 3-alkyl- or aryl
substituted glutaronitriles mostly generate (S)-configurated
efficient enzyme catalysts remain greatly desired for these
biotransformations.
The active pharmaceutical ingredients (S)-Pregabalin ((S)-4a)
or (R)-Baclofen ((R)-4b) can be prepared from chiral
cyanocarboxylic acid (2a or 2b) through two routes dependent on
the absolute configuration of 2a or 2b: hydration and subsequent
Hofmann rearrangement of (R)-2a or (R)-2b (Scheme 1, Route
1),[7b] or Curtius rearrangement followed by hydrolysis of (S)-2a or
(S)-2b (Scheme 1, Route 2).[8a] In terms of reaction conditions and
atomic economy, the Hofmann rearrangement route has some
advantages,[10] searching for the (R)-selective nitrilase was thus
pursued.
cyanocarboxylic acids. Herein
a “mirror-image” strategy, i.e.
symmetry analysis of substrate-binding pocket, was applied to identify
two key amino acid residues W170 and V198 that possibly modulate
the enantiopreference of a nitrilase from Synechocystis sp. PCC6803
towards 3-isobutyl glutaronitrile (1a). Exchange of these two residues
resulted in the enantiopreference inversion (S, 90% ee to R, 47% ee).
By further reshaping the substrate-binding pocket via routine site-
saturation and combinatorial mutagenesis, variant E8 with higher
activity and stereoselectivity (99% ee, R) was obtained. The mutant
enzyme was applied in the preparation of optically pure (R)-3-isobutyl-
4-cyanobutanoic acid ((R)-2a), and showed similar stereopreference
inversion towards a series of 3-substituted glutaronitriles. This study
may offer a general strategy to switch the stereopreference of other
nitrilases and other classes of enzymes toward the desymmetric
reactions of prochiral substrates with two identical reactive functional
groups, that are often desirable in asymmetric synthesis.
Introduction
Chiral -amino acids are important building blocks used for the
synthesis of pharmaceuticals.[1] Among them, the -aminobutyric
acid derivatives, (S)-3-aminomethyl-5-methylhexanoic acid and
(R)-4-amino-3-(4’-chlorophenyl) butyric acid, are effective
components of anticonvulsant Pregabalin (Lyrica®)[2] and the
muscle relaxant and antispastic agent Baclofen (Lioresal®)[3]
They are also valuable intermediates for synthesizing other
related drugs such as arbaclofen placarbil[4] and (R)-rolipram.[5]
Enzymatic desymmetrization involving lipase,[6] amidase,[7]
nitrilase[8] and one-pot cascade reaction including artificial
“Michaelase”[9] has been applied in the synthesis of -
aminobutyric acid derivative. However, highly stereoselective and
Scheme 1. Synthesis of the active pharmaceutical ingredients (S)-Pregabalin
((S)-4a) and (R)-Baclofen ((R)-4b) via enzymatic desymmetrization.
Nitrilase-catalyzed desymmetric hydrolysis of 3-substituted
glutaronitrile is an important method for the preparation of chiral
cyanocarboxylic acids. The desymmetric hydrolysis of 3-
benzoyloxy glutaronitrile with high enantiomeric excess (>99% ee,
S) was reported by Ohta et al. using Rhodococcus rhodochrous
IFO 15564 in 1991.[11] Since then, numbers of microorganisms
and nitrilases have been applied for the selective hydrolysis of
prochiral dinitrile compounds. Wang and co-workers reported
desymmetrization of 3-substituted glutaronitriles by Rhodococcus
sp. AJ270 with low enantiomeric excess,[8b,12] and the ee value of
(S)-2b was improved from 26% to 63% by adding acetone.[13] By
screening a large number of nitrilases from various environmental
[*]
S. Yu, J. Li, Prof. P. Yao, Prof. J. Feng, Y. Cui, J. Li, X. Liu, Prof. J.
Lin, Prof. Q. Wu, Prof. D. Zhu
National Technology Innovation Center of Synthetic Biology,
National Engineering Laboratory for Industrial Enzymes and Tianjin
Engineering Research Center of Biocatalytic Technology, Tianjin
Institute of Industrial Biotechnology, Chinese Academy of Sciences,
32 Xi Qi Dao, Tianjin Airport Economic Area, Tianjin 300308, PR
China.
E-mail: zhu_dm@tib.cas.cn, wu_qq@tib.cas.cn, lin_jp@tib.cas.cn
These authors have contributed equally to this work
[+]
Supporting information for this article is given via a link at the end of
the document.
1
This article is protected by copyright. All rights reserved.

10.1002/anie.202012243
Angewandte Chemie International Edition
RESEARCH ARTICLE
samples, the scientists at Diversa (Verenium) identified numerous
simulated the HEI state of the hydrolysis reaction by restriction
nitrilases for the desymmetric hydrolysis of 3-hydroxyglutaronitrile, covalent docking based on the reaction mechanism of nitrilase
providing (S)-4-cyano-3-hydroxybutyric acid with 98% ee or (R)-
4-cyano-3-hydroxybutyric acid with 95% ee.[14] Then, a variant
A190H of the most effective (R)-nitrilase from environmental
samples was obtained to catalyze the hydrolysis of 3-
hydroxyglutaronitrile to give (R)-4-cyano-3-hydroxybutyric acid in
98% ee at high substrate concentrations,[15] but generate (S)-2a
and (S)-2b in 85% ee and 67% ee, respectively.[8a] Furthermore,
we also tested the desymmetrization of various 3-substituted
glutaronitriles using other nitrilases.[8a,16] Among them, (S)-
configuration products were obtained except AtNIT3, which
produced (R)-2b in 78% ee but no activity for 1a. To the best of
our knowledge, the (R)-selective nitrilase for desymmetric
hydrolysis of 3-alkyl- or aryl substituted glutaronitriles remains
rare.
and the crystal structure of nitrilase SsNIT (Figure 1A, SI section
1.9).[19,21] Since 1a is a symmetrical dinitrile, the ligand molecule
was rotated by immobilizing the prochiral center to exchange the
positions of two cyano groups, leading to the HEI state for the
formation of (R)-2a (Figure 1B). Herein, we found that the isobutyl
of 1a was blocked by W170.
The enantiopreference of an enzyme can be successfully
inverted by enzyme engineering, [17] but it is still quite challenging
to rapidly achieve it with excellent enantioselectivity and without
activity trade-off. Enantiocomplementary enzymes exist in nature
to serve as biocatalysts for the synthesis of opposite enantiomers.
Although these enzyme pairs are not mirror-image molecules, the
active sites of some enantiocomplementary enzymes such as
subtilisin Carlsberg and lipase from Candida rugosa are identified
to be functionally mirror images of one another.[18] By analyzing
the crystal structures of vanillyl-alcohol oxidase (VAO) and its
enantiocomplementary enzyme p-cresol methylhydroxylase
(PCMH), D170, a critical residue for the hydration of the initially
formed p-quinone methide intermediate, was relocated to the
opposite face of the active site cavity, creating a D170S/T457E
for
the
transformation of
4-ethylphenol
into
1-(4'-
Figure 1. The substrate molecule was restricted covalent docking to the
catalytic pocket based on the reaction mechanism. The dotted green line
indicates hydrogen bonding interaction, the dotted yellow line indicates
hydrophobic interaction, and the solid red line indicates conflict between atoms.
(A) The HEI state of 1a (cyan) for formation of (S)-2a. (B) The HEI state of 1a
(yellow) for the formation of (R)-2a in the catalytic reaction. (C) A cartoon of the
“mirror-image” strategy: A, A', C, and D respectively represent four substituent
groups of the prochiral substrate, wherein A and A' are the two identical CN
group, and C, D are two different substituents. The rotation axis (dotted green
line) is formed by a tetrahedral structure in accordance with the vertical line of
the A-A' line and the C-D line. Blue blocks, red balls, orange triangles and silver
balls represent protein residues.
hydroxyphenyl)ethanol with an inverted stereopreference.[17q]
Thus designing a functional mirror image of an enzyme active site
should greatly facilitate the creation of its enantiocomplementary
counterpart. Recently, the crystal structure of nitrilase (PDB ID:
3WUY) from Synechocystis sp. PCC6803 (SsNIT) has been
reported.[19] We found that wild-type enzyme SsNIT exhibited high
stereoselectivity toward 1a (90% ee, S) and 1b (97% ee, S),[8a]
and its triple mutant (SsNIT-P194A/I201A/F202V) displayed
significantly enhanced catalytic activity and stereoselectivity
toward 1b (99% ee, S) and other 3-substituted glutaronitriles.[16]
Herein, the stereoselectivity of SsNIT was inverted from S to R
toward the desymmetrization of 3-substituted glutaronitriles by a
“mirror-image” strategy, i.e. symmetry analysis of substrate-
binding pocket, followed by site-saturation mutagenesis and
combinatorial mutagenesis.
As such, a “mirror-image” strategy was proposed as follows:
the substrate is centered on the chiral carbon, and its four
substituent groups are numbered A, A', C, and D (as shown in
Figure 1C, A and A' are two identical cyano groups, C and D are
isobutyl and hydrogen atom). We speculated that when we
interchanged the positions of A and A', the other two substituent
groups C and D will also exchange their positions, leading to the
other enantiomeric product after CN hydrolysis. We thus
envisioned that the configuration of product 2a would be inverted
by exchanging the two residues (W170 and V198) which
interacted with C and D, respectively. Therefore, W170 and V198
were doubly mutated to give W170V/V198W. The resulting
mutant enzyme catalyzed the hydrolysis of 1a to give (R)-2a with
47% ee. Further reducing the bulkiness of residue W170 and
adjusting V198 led to mutants W170A/V198W, W170G/V198W
(E2) and W170G/V198F, which hydrolyzed 1a to afford (R)-
enantiomer in 17% ee, 60% ee and 82% ee, respectively (SI
Table S6). Variants W170G/V198W (E2) and W170G/V198F
were thus selected for further studies.
Results and Discussion
In order to invert the enantioselectivity of SsNIT toward 1a,
alanine scanning of the amino acid residues in the substrate-
binding pocket, and subsequent semi-saturated and
combinatorial mutagenesis were performed, giving a best (R)-
selective mutant with 50% ee (SI section 1.3, Figure S1, Table S3,
S4 and S5). The unsatisfactory results promoted us to seek for
different mutation strategies.
High-energy intermediate (HEI) state can reflect early
stereoselective recognition of enzymes, which can be treated as
the proxies for the typically unknown transition state structure for
its chemical similarity with the transition state.[20] Hence, we
2
This article is protected by copyright. All rights reserved.

10.1002/anie.202012243
Angewandte Chemie International Edition
RESEARCH ARTICLE
improved the affinity between mutants and 1a (SI Table S8). In
addition, the F202A provided more space for isobutyl of 1a,
leading to the orientation of the large steric hindrance group of
W198 inclining into binding-pocket and the unreacted cyano
group and isobutyl adjusting their orientation. Thus, compared
with E2, mutant E5 showed higher stereoselectivity (95% ee
values) and 8.3- and 23.6-fold improvement in kcat and kcat/Km,
respectively (SI Table S8). Furthermore, two distances of sulfydryl
of C169 with cyano carbon of 1aHEI and amino of K135 with
hydroxyl of 1aHEI, which support catalysis, in the MD trajectory
were calculated as shown in Figure 3C. For wt-1a, the proportions
of stable HEI-S states with both d(SGC169-C11a-HEI) ≤ 2.0 Å and
d(HZK135- O11a-HEI) ≤ 3.5 Å were 0.460%, more than 1.7-fold than
that observed in stable HEI-R state (0.272%). On the contrary, the
99
98
96
96
95
90
100
80
3500
3000
2500
2000
1500
1000
500
60
3232
60
40
20
90
WT E2
0
E3
E4
54
E5
E6
E7
E8
-20
-40
-60
-80
-100
1303
1031
512
0
100
29
119
Figure 2. Activity and enantioselectivity of mutants toward the hydrolysis of 1a.
Blue and green bars represent ee values with (S)- or (R)-enantiomer,
respectively and red triangle means relative activity. The relative activity of wild-
type defined as 100%. The detail results were presented in SI Table S6 and
Table S7.
stable HEI states (0.540%) in E5-1aHEI- were 2.6-fold of that
R
observed in E5-1aHEI- (0.208%). The results indicate that wt-
S
1aHEI-S and E5-1aHEI-R are favorable in the formation of HEI states
to give corresponding (S)- and (R)-2a, respectively.
To further improve the activity and stereoselectivity of variant
E5 towards 1a, 14 amino acid residues (Y59, F64, T137, P138,
T139, Y140, E142, W146, A168, E171, H172, Y173, F193, P194)
located within a distance of 6Å of 1a in the docking structure of
E5-1a (SI Figure S2) were selected for site-saturation
mutagenesis. Screening of these libraries using 1a as a substrate
by phenol hypochlorite colorimetry method,[22] followed by
measuring the ee value and the amount of by-product with GC
analysis, resulted in three beneficial mutants E5-T137C (E6, 4.3-
fold more active relative to E5, 96% ee, R), E5-A168E (3.23-fold
relative to E5, 95% ee, R) and E5-F193W (2.93-fold relative to E5,
98% ee, R). Subsequent combinatorial mutagenesis of these
beneficial mutants resulted in E5-T137C/A168E (9.7-fold relative
to E5, 97% ee, R), E5-T137C/F193W (E7, 11-fold relative to E5,
98% ee, R) and mutant E5-T137C/A168E/F193W (E8) which
exhibited 27.2-fold activity improvement relative to E5 and
generated (R)-2a in 99% ee (Figure 2 and SI Table S7). It is worth
noting that variant E8 not only improved the activity and
stereoselectivity, but also reduced by-product 3-(cyanomethyl)-5-
methylhexanamide (5a) to only 0.9% yield.
The kcat and kcat/Km of mutant E8 were 55.22 min-1 and 9.34
min-1·mM-1 respectively, which exhibited a 23- and 13-fold
improvement compared to E5 (SI Table S8). The excellent
stereoselectivity and activity of E8 could be attributed to a
synergistic effect of the three residues. In order to shed light on
the enhanced catalytic activity of E8, the binding modes of E8-1a
and E5-1a were simulated by molecular dynamics, and the HEI
state binding energy[23] of E8-1a and E5-1a was -43.81 kcal/mol
and -39.94 kcal/mol, respectively. It is notable that the computed
binding energy of E8-1a is approximately 3.87 kcal/mol smaller
In order to further improve the stereoselectivity, mutants
W170G/V198W (E2) and W170G/V198F were respectively
combined with N118A, H141A and F202A, which showed
preference to the formation of (R)-enantiomer compared to wild-
type enzyme in the alanine scanning studies. Six mutants were
constructed and evaluated for the hydrolysis of 1a (Figure 2 and
SI Table S6). The best mutant W170G/V198W/F202A (E3) led to
pronouncedly inverted stereoselectivity in favor of the (R)-
configuration with 90% ee. Other variants included
N118A/W170G/V198W (no activity), N118A/W170G/V198F (79%
ee,
R),
H141A/W170G/V198W
(76%
ee,
R),
H141A/W170G/V198F (34% ee, R), and W170G/V198F/ F202A
(85% ee, R). The results indicated positive synergistic effects
between mutants H141A, F202A and W170G/V198W (E2), and
negative effects between variants N118A, H141A and
W170G/V198F. Further combinatorial mutants N118A/W170G/
V198W/F202A (E4), H141A/W170G/V198W/F202A, and N118A/
H141A/W170G/V198W/F202A (E5) were constructed and tested
for the hydrolysis of 1a, and (R)-2a was obtained in 96%, 93%
and 95% ee, respectively (Figure 2 and SI Table S6).
To better understand the evolution of stereoselective inversion,
molecular dynamic simulations of wt-1a, E2-1a and E5-1a were
performed based on HEI states. The steric hindrance exchange
at W170G and V198W made a major contribution to the stereo-
preference inversion (Figure 3A). As shown in Figure 3B, residues
A141 stabilize the unreacted cyano group by hydrogen bond
interactions, while A118 led to the disappearance of hydrogen
bonding with E53 enhancing the swing of the catalytic triplet and
Figure 3. Comparison of catalytic
cavity conformations of E2
(yellow) with (A) wt SsNIT (blue)
and (B) E5 (cyan) based on MD
analysis. The dotted green line
indicates
hydrogen
bonding
interaction. The carbon atoms of
substrate in wt-1a, E2-1a and E5-
1a are represented by magenta,
orange and green respectively.
(C) Conformation maps of wt-1a
and E5-1a in dynamical trajectory
simulations.
3
This article is protected by copyright. All rights reserved.

10.1002/anie.202012243
Angewandte Chemie International Edition
RESEARCH ARTICLE
than that of E5-1a. Through further binding energy analysis, we
found that the binding energy of catalytic triplet with 1a in mutant
E8 was approximately 4.5 kcal/mol smaller than that of E5-1a,
implying that mutation of residues T137, A168 and F193 in mutant
E8 play a significant role in the difference of binding energy (SI
Table S9). Additionally, Q192 could form a hydrogen bond with
C169 to stabilize catalytic triplet during MD simulation (SI Figure
S3), but the respective probabilities to form the hydrogen bond in
mutant E8 (74.3%) and E5 (22.7%) were dramatically different.
Variation in amide 5a formation was studied by MD sampling of
the wt SsNIT and three mutants. The results showed that NH2
group in the tetrahedral intermediate for the mutant was more
easily protonated and resulted in less amide 5a formation, being
consistent with the studies by Jiang et al[24] (SI Table S10).
Specific activity and enantioselectivity of wt SsNIT, and
mutants E2, E5 and E8 toward sixteen 3-substituted
enantioselectivity were both greatly affected by the amino acid
residues in the substrate-binding pocket and the 3-substitutient of
glutaronitriles. As shown in Table 1, the variants exhibited similar
effects on enantiopreference inversion towards all the tested
substrates although they had lower stereoselectivity towards 3-
aryl substituted glutaronitriles. Compared to wt SsNIT, variant E8
showed substrate preference to aliphatic dinitrile substrates since
E8 exhibited high activity and (R)-stereoselectivity toward 1c-1f.
The HEI docking study of 3-(4-(trifluoromethyl)phenyl)
glutaronitrile (1p) showed that for mutant E8, the aromatic ring of
(R)-configuration interacts with W198, while that of (S)-
configuration has interaction with Y140 and Y173 (SI Figure S4),
thus possibly leading to low enantioselectivity. This is reasonable
because the substrate-binding pocket was reshaped using 1a as
the model substrate. Therefore, reshaping the substrate-binding
pocket to its “mirror-image” to some extent could switch the
enzyme stereopreference for a range of similar substrates.
glutaronitriles
were
determined.
The
activity
and
Table 1. Activity and enantioselectivity of wt SsNIT and its variants in the asymmetric hydrolysis of various 3-substitued glutaronitriles 1a−1p
Specific activity[a] [mU·mg-1] and ee[b] [%]
E2 E5
Entry
Substrate
R
WT
E8
1
2
1a
1b
1c
1d
1e
1f
((CH3)2CHCH2)
(4-Cl-C6H4)
22
210
103
141
106
16
91S
97S
95S
96S
96S
98S
96S
98S
95S
99S
97S
74S
93S
98S
97S
97S
2
3
60R
48R
28R
51R
80R
29R
52R
38R
ND[c]
54R
47R
35R
32R
71R
50R
31R
15
15
28
63
109
41
38
20
4
95R
16R
77R
93R
99R
99R
64R
8R
457
63
99R
23R
91R
97R
99R
99R
50R
29R
70R
34R
11R
38R
31R
76R
66R
49R
3
(CH3CH2CH2)
(CH3CH2CH2CH2)
(CH3CH2CH2CH2CH2)
((CH2)5CH)
8
310
682
565
278
122
124
45
4
4
5
3
6
10
6
7
1g
1h
1i
(C6H5)
457
208
10
8
(4-CH3-C6H4)
(4-(CH3)2CH-C6H4)
(4-F-C6H4)
5
9
0
91R
49R
12R
49R
41R
88R
94R
53R
10
11
12
13
14
15
16
1j
443
137
142
397
87
3
13
10
167
60
36
7
79
1k
1l
(4-Br-C6H4)
2
46
(2-Cl-C6H4)
10
4
205
169
129
66
1m
1n
1o
1p
(3-Cl-C6H4)
(4-CH3O-C6H4)
(4-CH3S-C6H4)
(4-CF3-C6H4)
5
14
1
281
1
3
705
[a] The activity was determined by the phenol hypochlorite colorimetry reaction. [b] The ee value was determined by HPLC on a chiral stationary phase. The absolute
configurations of 2a and 2b were determined by converting them into the corresponding amino acids and then comparing their retention times with those of the
standard samples. The configurations of other products were estimated by comparing their elution sequence in HPLC analysis with those of 2b. [c] ND means not
detected.
4
This article is protected by copyright. All rights reserved.

10.1002/anie.202012243
Angewandte Chemie International Edition
RESEARCH ARTICLE
[3]
[4]
a) P. Hudgson, D. Weightman, Br. Med. J. 1971, 4, 15-17; b) N. G.
Bowery, D. R. Hill, A. L. Hudson, A. Doble, D. N. Middlemiss, J. Shaw,
M. Turnbull, Nature 1980, 283, 92-94.
The synthesis of (R)-2a was performed by using whole E.coli
cells overexpressing variant E8. To investigate the substrate
inhibition, various concentrations of 1a from 100 mM to 500 mM
were examined. A complete conversion of 400 mM 1a was
achieved with 99% ee (SI Figure S6). The synthesis of (R)-2a was
performed at 40 mL scale with substrate loading of 400 mM (60
g·L-1). The time course showed that the conversion reached 99%
within 24 h and only 0.9% by-product 5a was detected (SI Figure
S7). After simple centrifugation and extraction, the desired
product (R)-2a was isolated as yellow oil in >99% ee and 97%
yield.
a) A. Man, T. Boulanger, B. Brandau, F. Durant, G. Evrard, M. Heaulme,
E. Desaulles, C.-G. Wermutht, J. Med. Chem. 1991, 34, 1307-1313; b)
R. Lal, J. Sukbuntherng, E. H. L. Tai, S. Upadhyay, F. Yao, M. S. Warren,
W. Luo, L. Bu, S. Nguyen, J. Zamora, G. Peng, T. Dias, Y. Bao, M.
Ludwikow, T. Phan, R. A. Scheuerman, M. G. Hui Yan, Q. Q. Wu, T.
Annamalai, S. P. Raillard, K. Koller, M. A. Gallop, K. C. Cundy, J.
Pharmacol. Exp. Ther. 2009, 330, 911-921.
[5]
[6]
G.-S. Lee, N. Subramanian, A. I. Kim, I. Aksentijevich, R. Goldbach-
Mansky, D. B. Sacks, R. N. Germain, D. L. Kastner, J. J. Chae, Nature
2012, 492, 123-127.
a) C. A. 5, S. H. Y. Dumond, J. T. P. Kelleher, L. Tully, Org. Process Res.
Dev. 2008, 12, 392-398; b) J.-H. Jung, D.-H. Yoon, P. Kang, W. K. Lee,
H. Euma, H.-J. Ha, Org. Biomol. Chem. 2013, 11, 3635-3641; c) R.
Zheng, T. Wang, D. Fu, A. Li, X. Li, Y. Zheng, Appl. Microbiol. Biotechnol.
2013, 97, 4839-4847.
Conclusion
The nitrilase SsNIT was engineered for inverted
stereoselectivity from S to R toward 3-substituted glutaronitriles
by directed evolution. A “mirror-image” strategy was performed to
identify key residues responsible for stereo-recognition in the
substrate binding pocket, and a simple switch of the two key
residues resulted in variant with inverted stereopreference (47%
ee, R). Further reshaping the substrate-binding cavity by site-
saturation mutagenesis and combinatorial mutagenesis led to
variant E8 with high catalytic efficiency and stereoselectivity (99%
ee, R). MD simulations showed that the mutations changed the
probabilities of catalytically active conformations leading to (S)- or
(R)-enantiomer. This is in agreement with the experimental
observations. This study provides valuable information for our
understanding on how nitrilase controls the enantioselectivity for
the desymmetric hydrolysis of various prochiral dinitriles, and the
“mirror-image” strategy may be applicable to other nitrilases and
other enzymes for selective desymmetric reactions of prochiral
substrates with two identical active functional groups, that are
currently tested in our continuing pursue.
[7]
[8]
[9]
a) L. Zhang, D. Wang, L. Zhao, M. Wang, J. Org. Chem. 2012, 77, 5584-
5591; b) M. Nojiri, K. Uekita, M. Ohnuki, N. Taoka, Y. Yasohara, J. Appl.
Microbiol. 2013, 115, 1127-1133.
a) Y. Duan, P. Yao, J. Ren, C. Han, Q. Li, J. Yuan, J. Feng, Q. Wu, D.
Zhu, Sci. China. Chem. 2014, 57, 1164-1171; b) M. Wang, Acc. Chem.
Res. 2015, 48, 602-611.
L. Biewenga, T. Saravanan, A. Kunzendorf, J. Y. van der Meer, T. Pijning,
P. G. Tepper, R. van Merkerk, S. J. Charnock, A. W. H. Thunnissen, G.
J. Poelarends, ACS Catal. 2019, 9, 1503-1513.
[10] M. S. Hoekstra, D. M. Sobieray, M. A. Schwindt, T. A. Mulhern, T. M.
Grote, B. K. Huckabee, V. S. Hendrickson, L. C. Franklin, E. J. Granger,
G. L. Karrick, Org. Process Res. Dev. 1997, 1, 26-38.
[11] H. Kakeya, N. Sakai, A. Sano, M. Yokoyama, H. T. Sugai, H. Ohta, Chem.
Lett. 1991, 20, 1823-1824.
[12] M. Wang, C. Liu, J. Li, O. Meth-Cohn, Tetrahedron Lett. 2000, 41, 8549-
8552.
[13] M. Wang, C. Liu, J. Li, Tetrahedron: Asymmetry 2001, 12, 3367-3373.
[14] a) G. DeSantis, Z. Zhu, W. A. Greenberg, K. Wong, J. Chaplin, S. R.
Hanson, B. Farwell, L. W. Nicholson, C. L. Rand, D. P. Weiner, D. E.
Robertson, M. J. Burk, J. Am. Chem. Soc. 2002, 124, 9024-9025; b) D.
E. Robertson, J. A. Chaplin, G. DeSantis, M. Podar, M. Madden, E. Chi,
T. Richardson, A. Milan, M. Miller, D. P. Weiner, K. Wong, J. McQuaid,
B. Farwell, L. A. Preston, X. Q. Tan, M. A. Snead, M. Keller, E. Mathur,
P. L. Kretz, M. J. Burk, J. M. Short, Appl. Environ. Microbiol. 2004, 70,
2429-2436.
Acknowledgements
[15] G. DeSantis, K. Wong, B. Farwell, K. Chatman, Z. Zhu, G. Tomlinson, H.
Huang, X. Tan, L. Bibbs, P. Chen, K. Kretz, M. J. Burk, J. Am. Chem.
Soc. 2003, 125, 11476-11477.
The work was financially supported by the National Natural
Science Foundation of China (Grant No. 21572261), the Youth
Innovation Promotion Association of the Chinese Academy of
Sciences (Grant No. 2016166), Tianjin Synthetic Biotechnology
Innovation Capacity Improvement Project (TSBICIP-KJGG-001)
and National Key Research and Development Plan Special
Project for “Synthetic biology” (2018YFA0901402).
[16] S. Yu, P. Yao, J. Li, J. Feng, Q. Wu, D. Zhu, Catal. Sci. Technol. 2019,
9, 1504-1510.
[17] For changing the entrance tunnel, see: a) A. Shehzad, S. Panneerselvam,
M. Linow, M. Bocola, D. Roccatano, J. Mueller-Dieckmann, M. Wilmanns,
U. Schwaneberg, Chem. Commun. 2013, 49, 4694-4696; b) H. Ma, X.
Yang, Z. Lu, N. Liu, Y. Chen, PLoS One 2014, 9, e103792; c) J. Zhou, Y.
Wang, G. Xu, L. Wu, R. Han, U. Schwaneberg, Y. Rao, Y. L. Zhao, J.
Zhou, Y. Ni, J. Am. Chem. Soc. 2018, 140, 12645-12654; For changing
substrate binding site, see: d) Y. Ensari, G. V. Dhoke, M. D. Davari, M.
Bocola, A. J. Ruff, U. Schwaneberg, Chem. Eur. J. 2017, 23, 12636-
12645; e) L. Skalden, C. Peters, J. Dickerhoff, A. Nobili, H. J. Joosten, K.
Weisz, M. Hohne, U. T. Bornscheuer, ChemBioChem 2015, 16, 1041-
1045; f) X. Chen, H. Zhang, M. A. Maria-Solano, W. Liu, J. Li, J. Feng,
X. Liu, S. Osuna, R.-T. Guo, Q. Wu, D. Zhu, Y. Ma, Nature Nat. Catal.
2019, 2, 931; g) D. Zhang, X. Chen, J. Chi, J. Feng, Q. Wu, D. Zhu, ACS
Catal. 2015, 5, 2452-2457; h) D. Zhu, Y. Yang, S. Majkowicz, T. H.-Y.
Pan, K. Kantardjieff, L. Hua, Org. Lett. 2008, 10, 525-528; i) E. Eger, A.
Simon, M. Sharma, S. Yang, W. B. Breukelaar, G. Grogan, K. N. Houk,
W. Kroutil, J. Am. Chem. Soc. 2020, 142, 792-800; j) F. Qin, B. Qin, W.
Zhang, Y. Liu, X. Su, T. Zhu, J. Ouyang, J. Guo, Y. Li, F. Zhang, J. Tang,
X. Jia, S. You, ACS Catal. 2018, 8, 6012-6020; k) P. Yao, P. Cong, R.
Gong, J. Li, G. Li, J. Ren, J. Feng, J. Lin, P. C. K. Lau, Q. Wu, D. Zhu,
ACS Catal. 2018, 8, 1648-1652; l) X. Ren, N. Liu, A. L. Chandgude, R.
Keywords: directed evolution • desymmetric hydrolysis •
nitrilase • stereopreference inversion • mirror-image
[1]
a) R. B. Silverman, R. Andruszkiewicz, S. M. Nanavati, C. P. Taylor, M.
G. Vartanian, J. Med. Chem. 1991, 34, 2295-2298; b) J. Farrera-Sinfreu,
E. Giralt, S. Castel, F. Albericio, M. Royo, J. Am. Chem. Soc. 2005, 127,
9459-9468; c) M. Ordóñez, C. Cativiela, Tetrahedron-Asymmetry 2007,
18, 3-99; d) M. Ordóñez, C. Cativiela, I. Romero-Estudillo, Tetrahedron:
Asymmetry 2016, 27, 999-1055; e) P. Ramesh, D. Suman, K. S. N.
Reddy, Synthesis 2018, 50, 211-226; ; f) S. Wu, R. Snajdrova, J. C.
Moore, K. Baldenius, U. T. Bornscheuer, Angew. Chem. Int. Ed. 2020,
59,
10.1002/anie.202006648;
Angew.
Chem.
2020,
10.1002/ange.202006648; g) B. Hauer, ACS Catal. 2020, 10, 8418-8427.
a) J. E. Frampton, CNS Drugs 2014, 28, 835-854; b) R. Patel, A. H.
Dickenson, Pharmacol. Research Perspect. 2016, 4, e00205.
[2]
5
This article is protected by copyright. All rights reserved.

10.1002/anie.202012243
Angewandte Chemie International Edition
RESEARCH ARTICLE
Fasan, Angew. Chem. Int. Ed. 2020, 10.1002/anie.202007953; Angew.
Chem. 2020, 10.1002/ange.202007953; m) Z. T. Sun, R. Lonsdale, L.
Wu, G. Y. Li, A. T. Li, J. B. Wang, J. H. Zhou, M. T. Reetz, ACS Catal.
2016, 6, 1590-1597; n) A. T. Li, A. Ilie, Z. T. Sun, R. Lonsdale, J. H. Xu,
M. T. Reetz, Angew. Chem. Int. Ed. 2016, 55, 12026-12029;
Angew.Chem. 2016, 128, 12205–12208; For moving key catalytic
residue, see: o) Y. Terao, Y. Ijima, K. Miyamoto, H. Ohta, J. Mol. Catal.
B: 2007, 45, 15-20; p) Y. Ijima, K. Matoishi, Y. Terao, N. Doi, H.
Yanagawa, H. Ohta, Chem. Commun 2005, 877-879; q) R. H. H. van
Den Heuvel, M. W. Fraaije, M. Ferrer, A. Mattevi, W. J. van Berkel, PNAS
2000, 97, 9455-9460.
[19] L. Zhang, B. Yin, C. Wang, S. Jiang, H. Wang, Y. A. Yuan, D. Wei, J.
Struct. Biol. 2014, 188, 93-101.
[20] a) D. N. Bolon, S. L. Mayo, PNAS 2001, 98, 14274-14279; b) J. C.
Hermann, E. Ghanem, Y. Li, F. M. Raushel, J. J. Irwin, B. K. Shoichet, J.
Am. Chem. Soc. 2006, 128, 15882-15891; c) J. C. Hermann, R. Marti-
Arbona, A. A. Fedorov, E. Fedorov, S. C. Almo, B. K. Shoichet, F. M.
Raushel, Nature 2007, 448, 775-779; d) V. nanda, R. l. Koder, Nat. Chem.
2010, 2, 15-24; e) M. J. Grisewood, N. P. Gifford, R. J. Pantazes, Y. Li,
P. C. Cirino, M. J. Janik, C. D. Maranas, PLoS One 2013, 8, e75358; f)
N. London, J. D. Farelli, S. D. Brown, C. Liu, H. Huang, M. Korczynska,
N. F. Al-Obaidi, P. C. Babbitt, S. C. Almo, K. N. Allen, B. K. Shoichet,
Biochemistry 2015, 54, 528-537.
[18] a) P. F. Mugford, U. G. Wagner, Y. Jiang, K. Faber, R. J. Kazlauskas,
Angew. Chem. Int. Ed. 2008, 47, 8782-8793; Angew. Chem. 2008, 120,
8912 – 8923; b) A. M. K. P. A. Fitzpatrick, J. Am. Chem. Soc. 1991, 113,
3166-3171; c) E. C. D. L. Ollis, M. Cygler, B. Dijkstra, F. Frolow, S. M.,
M. H. Franken, S. J. Remington, I. Silman, J. Schrag, J. L., K. H. G. V.
Sussman, A. Goldman, Protein Eng. 1992, 5, 197-211; d) R. J.
Kazlauskas, A. N. E. Weissfloch, J. Mol. Catal. B Enzymatic. 1997, 3, 65-
72; e) R. J. K. C. K. Savile, J. Am. Chem. Soc. 2005, 127, 12228-12229;
f) Q. Wu, P. Soni, M. T. Reetz, J. Am. Chem. Soc. 2013, 135, 1872-1881;
g) S. Bartsch, R. Kourist, U. T. Bornscheuer, Angew. Chem. Int. Ed. 2008,
47, 1508-1511; Angew. Chem. 2008, 120, 1531-1534; h) S. Kim, Y. K.
Choi, J. Hong, J. Park, M.-J. Kim, Tetrahedron Lett. 2013, 54, 1185-1188.
[21] B. C. M. Fernandes, C. Mateo, C. Kiziak, A. Chmura, J. Wacker, F. v.
Rantwijk, A. Stolz, R. A. Sheldon, Adv. Synth. Catal. 2006, 348, 2597-
2603.
[22] a) M. W. Weatherburn, Anal. Chem. 1967, 39, 971-974; b) T. T. Ngo, A.
P. H. Phan, C. F. Yam, M. Lenhoff, Anal. Chem. 1982, 54, 46-49; c) Y.
Xue, Y. Yang, S. Lv, Z. Liu, Y. Zheng, Appl. Microbiol. Biotechnol. 2016,
100, 3421-3432.
[23] I. Bill R. Miller, J. T. Dwight McGee, J. M. Swails, N. Homeyer, H. Gohlke,
A. E. Roitberg, J. Chem. Theory. Comput. 2012, 8, 3314-3321.
[24] S. Jiang, L. Zhang, Z. Yao, B. Gao, H. Wang, X. Mao, D. Wei, Catal. Sci.
Technol. 2017, 7, 1122-1128.
6
This article is protected by copyright. All rights reserved.

10.1002/anie.202012243
Angewandte Chemie International Edition
RESEARCH ARTICLE
Entry for the Table of Contents
Exchange to invert. A “mirror-image” strategy, i.e. symmetry
analysis of substrate-binding pocket, was applied to invert the
stereopreference of a nitrilase toward the desymmetrization of
prochiral dinitriles.
7
This article is protected by copyright. All rights reserved.