A new d-threonine aldolase as a promising biocatalyst for highly stereoselective preparation of chiral aromatic β-hydroxy-α-amino acids

Article abstract of DOI:10.1039/c7cy01774j

d-Threonine aldolase is an enzyme belonging to the glycine-dependent aldolases, and it catalyzes the reversible aldol reaction of glycine and acetaldehyde to give d-threonine and/or d-allo-threonine. In this study, a putative d-threonine aldolase gene from Delftia sp. RIT313 was cloned and expressed in Escherichia coli BL21 (DE3). The purified enzyme (DrDTA, 47 KDa) exhibited 21.3 U mg^-1 activity for the aldol addition of glycine and acetaldehyde in MES-NaOH buffer (pH 6.0) at 50 °C. Both pyridoxal 5′-phosphate and metal ions were needed for the reaction, and the existence of the metal ions enhanced the stability of the enzyme. It was found that the conversion and C_β-stereoselectivity were dramatically influenced by the reaction temperature, co-solvent, amount of enzyme and reaction time, and it is possible to enable the reaction under kinetic control to retain suitable conversion and high stereoselectivity at the β-carbon, thus tackling the "C_β-stereoselectivity problem". DrDTA showed high activity toward aromatic aldehydes with electron-withdrawing substituents. Under the optimized reaction conditions, phenylserines with a 2′-fluoro- or 3′-nitro-substituent were obtained with >90% conversion and >90% de. In addition, dl-threo-phenylserine and dl-threo-4-(methylsulfonyl)phenylserine were efficiently resolved with an excellent enantiomeric excess value (ee, >99%) using a whole cell biocatalyst in a two-phase system at 1.0 M and 0.3 M, respectively, the highest substrate concentration reported so far. These results suggested that DrDTA might be a promising biocatalyst for producing chiral aromatic β-hydroxy-α-amino acids.

Full text of DOI:10.1039/c7cy01774j

View Article Online
View Journal
Catalysis
Science &
Technology
Accepted Manuscript
This article can be cited before page numbers have been issued, to do this please use: Q. Chen, X. Chen,
Y. Cui, J. Ren, W. Lu, J. Feng, Q. Wu and D. Zhu, Catal. Sci. Technol., 2017, DOI: 10.1039/C7CY01774J.
This is an Accepted Manuscript, which has been through the
Volume 6 Number 1 7 January 2016 Pages 1–308
Royal Society of Chemistry peer review process and has been
accepted for publication.
Catalysis
Science &
Technology
Accepted Manuscripts are published online shortly after
www.rsc.org/catalysis
acceptance, before technical editing, formatting and proof reading.
Using this free service, authors can make their results available
to the community, in citable form, before we publish the edited
article. We will replace this Accepted Manuscript with the edited
and formatted Advance Article as soon as it is available.
You can find more information about Accepted Manuscripts in the
author guidelines.
Please note that technical editing may introduce minor changes
to the text and/or graphics, which may alter content. The journal’s
ISSN 2044-4753
standard Terms & Conditions and the ethical guidelines, outlined
in our author and reviewer resource centre, still apply. In no
event shall the Royal Society of Chemistry be held responsible
PAPER
Qingzhu Zhang et al.
Catalytic mechanism of C–F bond cleavage: insights from QM/MM
analysis of fluoroacetate dehalogenase
for any errors or omissions in this Accepted Manuscript or any
consequences arising from the use of any information it contains.
rsc.li/catalysis

Page 1 of 30
Catalysis Science & Technology
View Article Online
DOI: 10.1039/C7CY01774J
1
2
A new Dꢀthreonine aldolase as promising biocatalyst for highly stereoselective
preparation of chiral aromatic βꢀhydroxyꢀαꢀamino acids
3
Qijia Chen,a,b Xi Chen,b Yunfeng Cui,b Jie Ren,b Wei Lu,b Jinhui Feng,b Qiaqing
4
Wu,a,b Dunming Zhu a,b
*
5
University of Chinese Academy of Sciences, Beijing, China a;
6
National Engineering Laboratory for Industrial Enzymes and Tianjin Engineering
Center for Biocatalytic Technology, Tianjin Institute of Industrial Biotechnology,
Chinese Academy of Sciences, Tianjin, China b
7
8
9
10
11
12
13
14
15
Corresponding Author:
Dunming Zhu, zhu_dm@tib.cas.cn
Tel. 86ꢀ22ꢀ84861962, Fax 86ꢀ22ꢀ84861996
1

Catalysis Science & Technology
Page 2 of 30
View Article Online
DOI: 10.1039/C7CY01774J
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
Abstract
ꢀThreonine aldolase is an enzyme belongs to glycineꢀdependent aldolases, and
catalyzes reversible aldol reaction of glycine and acetaldehyde to give ꢀthreonine
and/or ꢀalloꢀthreonine. In this study, a putative ꢀthreonine aldolase gene from
D
D
D
D
Delftia sp. RIT313 was cloned and expressed in Escherichia coli BL21 (DE3). The
purified enzyme (DrDTA, 47 KDa) exhibited 21.3 U/mg activity for the aldol
addition of glycine and acetaldehyde in MESꢀNaOH buffer (pH 6.0) at 50°C. Both
pyridoxal 5'ꢀphosphate and metal ion were needed for the reaction, and existence of
the metal ion enhanced the stability of the enzyme. It was found that the conversion
and Cβꢀstereoselectivity were dramatically influenced by the reaction temperature,
coꢀsolvent, the amount of enzyme and reaction time, and it is possible to enable the
reaction under kinetic control to retain suitable conversion and high stereoselectivity
at βꢀcarbon, thus tackling the “Cβꢀstereoselectivity problem”. DrDTA showed high
activity toward aromatic aldehydes with electronꢀwithdrawing substituent. Under the
optimized reaction conditions, phenylserines with 2′ꢀfluoroꢀ or 3′ꢀnitroꢀsubstituent
were obtained with >90% conversion and >90% de. In addition,
DLꢀthreoꢀphenylserine and DLꢀthreoꢀ4ꢀ(methylsulfonyl)phenylserine were efficiently
resolved with excellent enantiomeric excess value (ee, >99%) using whole cell
biocatalyst in twoꢀphase system at 1.0 M and 0.3 M, respectively, the highest
substrate concentration reported so far. These results suggested that DrDTA might
be a promising biocatalyst for producing chiral aromatic βꢀhydroxyꢀαꢀamino acids.
37
38
39
Keywords:
DꢀThreonine aldolase, Aldol addition reaction, βꢀHydroxyꢀαꢀamino acid,
Enzymatic resolution, Cβꢀstereoselectivity
2

Page 3 of 30
Catalysis Science & Technology
View Article Online
DOI: 10.1039/C7CY01774J
40 1. Introduction
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
Asymmetric carbonꢀcarbon bond formation based on catalytic aldol addition
reaction has been recognized as one of the most useful reactions in synthetic organic
chemistry.1, 2 Among the catalysts available for aldol reaction, aldolase is becoming
increasingly important due to its ability to catalyze the formation of carbonꢀcarbon
bond under mild reaction conditions with high regioꢀ and stereoꢀselectivity.3ꢀ5 As
such, many aldolases have been isolated and characterized as biocatalysts for
synthetic applications and kinetic resolution.6
Threonine aldolase (TA) is a pyridoxal 5'ꢀphosphate (PLP)ꢀdependent aldolase
which catalyzes the aldol addition of glycine and aldehyde to furnish
βꢀhydroxyꢀαꢀamino acid with two new stereogenic centers.7 According to their
stereospecificity at the αꢀcarbon, TAs are classified into two types:
Lꢀtype and
D
ꢀtype. ꢀType is further divided into three subꢀtypes depending on the
L
stereospecificity toward the βꢀcarbon: (1)
LꢀTA (EC 4.1.2.5); (2) LꢀalloꢀTA (EC
4.1.2.49); (3) low specificity
L
ꢀTA (
L
ꢀlowꢀTA, EC 4.1.2.48).8 Similarly,
Dꢀtype is
classified into three subꢀtypes, however, only
DꢀlowꢀTAs (EC 4.1.2.42) have been
reported so far.9 Up to now ꢀlowꢀTAs were identified from Arthrobacter sp.,8
D
Alcaligenes (Achromobacter) xylosoxidans,10, 11 Xanthomonas oryzae,12 Silicibacter
pomeroyi,13 Singularimonas variicoloris,14 Pseudomonas,14, 15 and Chlamydomonas
reinhardtii16.
βꢀHydroxyꢀαꢀamino acids are useful chiral building blocks of pharmaceuticals
and agrochemicals.17,18 Threonine aldolases catalyze the formation of these
important compounds from simple and inexpensive starting materials. As such, these
enzymes may serve as promising alternatives to the existing chemical methods for
the synthesis of βꢀhydroxyꢀαꢀamino acids. In this context, a handful of TAs have
3

Catalysis Science & Technology
Page 4 of 30
View Article Online
DOI: 10.1039/C7CY01774J
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
been examined for the synthetic applications, and the substrate and product profiles
article.19
of
these
TAs
were
summarized
in
a
review
(2R,3S)ꢀ2ꢀAminoꢀ3ꢀhydroxyꢀ3ꢀ(pyridinꢀ4ꢀyl)ꢀ propanoic acid, a key intermediate in
the synthesis of the active pharmaceutical ingredient, was synthesized by using
recombinant DꢀlowꢀTAs to craft the two chiral centers in one step with high de and
ee.20, 21 On the other hand, TAs can also be used as a biocatalyst for the resolution of
βꢀhydroxyꢀαꢀamino acids to obtain their single stereomers.11, 22
However, in most cases, TAs catalyze the formation of βꢀhydroxyꢀαꢀamino
acids with excellent stereoselectivity (>99% ee) at the αꢀcarbon chiral center, but
low to moderate stereoselectivity at βꢀcarbon.17,18 The lack of stereochemical control
at the βꢀcarbon has hampered the synthetic application of TAs. This “Cβ
diastereoselectivity problem” may be overcome by examining more wildꢀtype
threonine aldolases or by engineering existing enzymes.23 In our effort, the genomic
mining approach was adopted to tackle this problem. Herein we report the
expression, purification and biochemical characterization of a new DꢀlowꢀTA from
Delftia sp. RIT313. We also examined its substrate profile, stereoselectivity, and
applied it in the resolution of some important βꢀhydroxyꢀαꢀamino acids. These
results indicate that the new
DꢀlowꢀTA may be a promising biocatalyst for the
preparation of the single stereomers of these chiral compounds.
84 2. Results and Discussion
85
86
87
88
2.1 Identification and sequence analysis of
D
ꢀthreonine aldolase
DꢀLowꢀTAs from A. xylosoxidans and Arthrobacter sp. were selected as
templates to identify new DꢀlowꢀTAs. Eight putative threonine aldolase genes from
different microorganisms were cloned (data not shown), and only the one from
4

Page 5 of 30
Catalysis Science & Technology
View Article Online
DOI: 10.1039/C7CY01774J
89
90
Delftia sp. RIT313 (DrDTA) was characterized as a DꢀlowꢀTA. The results of amino
acid sequence alignment show that DrDTA has 76%, 44%, 41%, 40%, 18%, 17%,
ꢀlowꢀTAs from Arthrobacter sp.,8 A. xylosoxidans,11 C.
91
17% and 16% identity with D
92
reinhardtii,16 S. pomeroyi,13 S. variicoloris,14 P. aeruginosa,23 P. protegens23 and
Pseudomonas sp.15 (Table S1). The crystal structure from A. xylosoxidans shows
that PLPꢀbinding sites include Lys59, Gln81, Arg157, Tyr187, Thr233, Gln249,
Gly251, Ser252 and Tyr260.24 A multiple sequence alignment, which was generated
using ClustalX25 and visualized using ESPript,26 shows that these PLPꢀbinding sites
of DrDTA are conserved (Fig. 1). DrDTA comprises an alanineꢀracemaseꢀlike
domain with an eightꢀstranded α/βꢀbarrel and is similar to an alanine racemase
mutant from G. stearothermophilus which exhibits retroꢀaldolase activity.27ꢀ29
93
94
95
96
97
98
99
100
101
Several studies have shown that
with alanine racemases.30, 31
DꢀlowꢀTAs share a common evolutionary origin
102
103
104
105
106
107
108
109
110
111
2.2 Cloning and heterogeneous expression of DrDTA
The ꢀlowꢀTA gene from Delftia sp. RIT 313 was cloned and expressed in E. coli
D
BL21 (DE3) at the optimum conditions (1 mM IPTG, 30°C). Cellꢀfree extracts lost
activity after storing at 4°C overnight. This is similar to the observation reported by
Kimura et al12 and Schirchet al32, and the reason may be the oxidation of sulfhydryl
groups of the protein. The recombinant enzyme was purified with Hisꢀtag affinity
chromatography and could be stable at ꢀ80°C for 6 months when 2 mM DTT was
added during the purification procedure. The cellꢀfree extracts and purified enzyme
had specific activity of 4.9 U/mg and 10.5 U/mg, respectively. The purified enzyme
gave a single band judged by SDSꢀPAGE (Fig. S1). The molecular mass was
5

Catalysis Science & Technology
Page 6 of 30
View Article Online
DOI: 10.1039/C7CY01774J
112
113
determined to be 47 kDa by exclusion chromatography, suggesting that the enzyme
was a monomeric protein.
114
115
116
117
118
119
120
121
122
123
124
125
126
127
128
129
130
131
132
133
134
2.3 Enzymatic properties
As shown in Fig. 2, the optimal pH for the retroꢀaldol and aldol addition reaction
were at pH 8.0 and pH 6.0, respectively. After being incubated at various pH for 12
h, the enzyme remained the highest activity at pH 6.0 for both retroꢀaldol and aldol
addition reactions (Fig. 2b). The optimum temperature for aldol addition reaction
was between 50ꢀ60°C (Fig. 2c). The enzyme retained about 70% activity from 25 to
60°C after it was heated for 30 min. However, the enzyme activity decreased rapidly
when being heated at 70ꢀ80°C (Fig. 2d).
Metal ionꢀfree enzyme solution showed no activity, suggesting that the
PLPꢀdependent enzyme required metal ion. When different metal ions was
individually added into the enzyme solution, the enzyme activity was restored to
different degrees, and the highest activity was recovered with Mn2+ (Fig. 3a). The
enzyme activity was most recovered at 0.05 mM Mn2+ and increased slowly as more
metal ion was added (Fig. 3b). Similarly, the enzyme activity almost reached the
maximum with 0.5 ꢁM PLP and more PLP exerted minimum effect on the activity
(Fig. 3d). Furthermore, the thermostability was enhanced when the enzyme solution
was incubated in the presence of Mn2+, suggesting Mn2+ was crucial for the catalytic
activity and thermal stability (Fig. 3c). Previous studies indicated that DꢀlowꢀTAs
from Arthrobacter sp., A .xylosoxidans, X. oryzae, C. reinhardtii and S. pomeroyi
demanded metal ions.8,11ꢀ13 However,
DꢀlowꢀTAs from Pseudomonas and S.
variicoloris had activity without metal ions.14 The metalꢀdependent
DꢀlowꢀTAs have
6

Page 7 of 30
Catalysis Science & Technology
View Article Online
DOI: 10.1039/C7CY01774J
135
136
137
138
139
140
141
142
143
144
higher sequence identity with each other, and are significantly different from the
ones which do not require metal ions (Table S1).
Kinetic parameters were determined for the retroꢀaldol reaction using threonine
as the substrate. The Km toward
Dꢀthreonine was approximately 2.5ꢀfold higher than
that toward
the known
Dꢀalloꢀthreonine (Table 1). The Km of DrDTA was lower than those of
D
ꢀlowꢀTAs from prokaryotes, suggesting that DrDTA had better affinity
ꢀalloꢀthreonine.8, 11, 14 The catalytic efficiencies of DrDTA
ꢀthreonine and D
ꢀalloꢀthreonine were 28.3 and 42.6 mMꢀ1·minꢀ1,
with
Dꢀthreonine and
D
toward
D
respectively, while the enzyme was not active toward
ꢀalloꢀthreonine. As such, DrDTA can be classified as a ꢀlowꢀTA.
Lꢀthreonine and
L
D
145
146
147
148
149
150
151
152
153
154
155
156
157
158
2.4 Effects of different conditions on conversion and stereoselectivity
It has been reported that the de value of the aldol addition product is dependent on
whether the reaction is under kinetic control or thermodynamic control,12 that in turn
is affected by the reaction conditions such as reaction temperature, amount of
enzyme, reaction medium, reaction time, etc. In general, the de value of the product
decreases as the reaction shifts from kinetic control to thermodynamic control.12
Therefore, in order to achieve reasonable conversion and high de value under kinetic
contol, the aldol addition reaction of glycine and benzaldehyde was chosen as the
model reaction to evaluate the βꢀcarbon stereoselectivity by measuring the
conversion and de of the product under different reaction conditions. First,
DrDTAꢀcatalyzed aldol addition reaction of glycine and benzaldehyde was carried
out at different temperature. As shown in Table S2, the erythro isomer was obtained
as the major product when the temperature was above 15°C and lowering the
temperature resulted in an increase of threo isomer. This might be because that at
7

Catalysis Science & Technology
Page 8 of 30
View Article Online
DOI: 10.1039/C7CY01774J
159
160
161
162
163
164
165
166
167
168
169
170
171
172
173
174
175
176
177
178
179
lower temperature the reaction rate is reduced and the reaction at the time interval is
under kinetic control, which favors the formation of threo isomer.18 The effects of
the coꢀsolvents on the conversion and de value were then investigated (Table 2).
Waterꢀmiscible organic solvents exert limited effects on the conversion, but greatly
enhanced de of the product, with 10% CH3CN giving the highest de. For other
organic solvents (THF, ethyl acetate, MTBE, 1,4ꢀdioxane, dichloromethane and
toluene), lower stereoselectivity at βꢀcarbon was observed. This indicated that
coꢀsolvents had obvious impact on the stereoselectivity at βꢀcarbon, even leading to
the inverse of the stereopreference in some cases.
Reducing the amount of enzyme from 57 U to 19 U resulted in an increase of
de from 42.2% (Table S2, entry 3) to 53.8% (Table 2, entry 6), suggesting the
product de was affected by the amount of enzyme. Therefore, the aldol addition was
performed in the buffer with 10% CH3CN with 5 U or 10 U of DrDTA, and the
conversion and de of the product were measured at different time intervals. As
shown in Table S3, although less enzyme usage resulted in slower reaction, as the
reaction proceeded, the conversion reached 68.3% with 5 U of enzyme at 12 h,
which was close to those with 10 U of enzyme. Surprisingly, 91% de was obtained
in the former case, significantly higher than those with more enzyme. Therefore,
although the reaction rate was reduced, lower temperature and less usage of enzyme
enabled the reaction under kinetic control to retain suitable conversion and high
stereoselectivity at βꢀcarbon.
180
181
182
2.5 Donor and acceptor specificity
To explore the donor and acceptor specificity of DrDTA, the aldol addition was
performed as follows: 5 U DrDTA in the buffer with 10% (v/v) CH3CN at 10°C for
8

Page 9 of 30
Catalysis Science & Technology
View Article Online
DOI: 10.1039/C7CY01774J
183
184
185
186
187
188
189
190
191
192
193
194
195
196
197
198
199
200
201
202
203
204
205
206
207
12 h. No consumption of acetaldehyde was observed when
D
ꢀalanine or
D
ꢀserine was
used as the donor, suggesting high donor specificity of DrDTA. Several
DꢀlowꢀTAs
can accept alanine and serine as amino acid donors and they all don′t require metal
ions as coꢀfactor.14 The acceptor substrate specificity was then evaluated with
glycine as the donor. The results showed that DrDTA accepted a wide range of
aromatic aldehydes as acceptors (Table 3). The type and position of the substituent
at the phenyl ring of aromatic aldehydes exerted effect on the activity and
stereoselectivity at βꢀcarbon. DrDTA showed high activity toward benzaldehydes
with electronꢀwithdrawing substituent (CH3SO2, NO2, F and Cl) on the phenyl ring,
while the electronꢀdonating substituents such as CH3, OCH3 and OH dramatically
decreased the enzyme activity so that the de values of the products were not
determined due to low conversion. This is consistent with the expected tendency of
the activation of the aldehyde group by the substituent on the phenyl ring. Among
the tested substrates, high conversion and de value were obtained for
3ꢀnitrobenzaldehyde and 2ꢀfluorobenzaldehyde (entries 3 and 5). Lower conversion
and de were observed for 2ꢀ and 4ꢀnitrobenzaldehyde (entries 2 and 4). For
fluoroꢀsubstituted benzaldehydes, significantly lower conversion was obtained for
4ꢀfluorobenzaldehyde, while lower de was observed for 3ꢀfluorobenzaldehyde
(entries 5ꢀ7). High stereoselectivity at βꢀcarbon was also retained for
3ꢀchlorobenzaldehyde and 4ꢀ(methylsulfonyl)benzaldehyde (entries 9ꢀ10). To the
best of our knowledge, these results represented a few examples of
DꢀlowꢀTAꢀcatalyzed aldol condensation reactions with high conversion and
Cβꢀstereoselectivity.
The stereoselectivity at βꢀcarbon was particularly worthy noting. External
aldimine intermediates of phenylserine, 2ꢀfluorophenylserine, 2ꢀbromophenylserine
9

Catalysis Science & Technology
Page 10 of 30
View Article Online
DOI: 10.1039/C7CY01774J
208
209
210
211
212
213
214
215
216
217
218
219
220
221
222
223
224
225
226
227
228
229
230
231
and 3ꢀnitrophenylserine were docked into the homology model structure to analyze
the effect of different substituentes on the βꢀstereoselectivity. Suitable binding
modes of threo enantiomers were obtained, whereas corresponding erythro
enantiomers did not result in good binding poses. It is consistent with the
observation that threo enantiomer is the preferred product. In the active center,
manganese ion coordinates with His348 (3.9Å), Asp350 (3.1Å) and 3ꢀhydroxyl of
phenylserine (3.3Å) (Fig. 4a). The crystal structural and modeling studies of a
Dꢀthreonine aldolase from Alcaligenes xylosoxidans indicate that the coordination of
the βꢀOHꢀgroup of the substrate to the manganese ion very likely activates this group,
thus facilitating its deprotonation in the retroꢀaldol reaction and the CꢀC bond formation in
the aldol reaction.24 The hydrogen bond between fluorine atom at 2ꢀposition of phenyl
ring and hydrogen atom on imidazole of His348 (3.6Å) makes the βꢀOHꢀgroup of
the substrate closer to manganese ion (2.6Å) compared to phenylserine (Fig. 4b).
For 2ꢀbromophenylserine, the 2ꢀsubstituent is more steric and less electron negative,
preventing the hydrogenꢀbond formation and resulting in the 3ꢀhydroxyl of substrate
away from manganese ion (6.0 Å) (Fig. 4c). As such, the reaction conversion and de
of the 2ꢀsubstituted substrates were decreased in the order of F > H > Cl > Br (Table
3). The nitro group at 3ꢀpositon of phenyl ring interacts with His194 (4.5Å), and the
βꢀOHꢀgroup of the substrate is positioned toward the manganese ion with short
distance of 2.7Å (Fig. 4d), leading to high conversion and diastereoselectivity for
3ꢀnitrobenzaldehyde (Entry 3 in Table 3). Therefore, the type of the substituting
groups and the position on the benzene ring played an important role in the enzyme
activity and stereoselectivity by affecting the interaction between the βꢀOHꢀgroup of
the substrate and the manganese ion during the process of substrate recognition.
10

Page 11 of 30
Catalysis Science & Technology
View Article Online
DOI: 10.1039/C7CY01774J
232
233
234
235
236
237
238
239
240
241
242
243
244
245
246
247
248
249
250
251
252
253
254
255
256
2.6 Stereoselective resolution using whole cell biocatalyst in twoꢀphase system
ꢀthreoꢀ(4ꢀMethylsulfonylphenyl)serine ( ꢀthreoꢀMPS) is an important chiral
L
L
building block in the synthesis of veterinary antibiotic, thiamphenicol and
florfenicol. DrDTA catalyzed the aldol addition of 4ꢀ(methylsulfonyl)benzaldehyde
with glycine to give
kinetic resolutions
DLꢀthreoꢀ(4ꢀmethylsulfonylphenyl)serine (DLꢀthreoꢀMPS) were studied to explore its
potential as a biocatalyst for the preparation of optically pure ꢀthreoꢀPS and
ꢀthreoꢀMPS. Taking into account low solubility of aldehyde in water, various
DꢀthreoꢀMPS with high de (Table 3, entry 10). Therefore, the
of DLꢀthreoꢀphenylserine DLꢀthreoꢀPS) and
(
L
L
solvents were tested as the coꢀsolvent for the kinetic resolution of DLꢀthreoꢀMPS
(Table S4). The best result was obtained in the biꢀphasic system containing 1:1
volume ratio of 1,2ꢀdichloroethane and buffer. Considering the feasibility of the
solvents used in the industrial application33,34, dichloromethane and cyclohexanone
were also tested for further studies. As such, the resolution of DLꢀthreoꢀPS and
DLꢀthreoꢀMPS was then investigated at different substrate concentration in this
twoꢀphase system, respectively. As shown in Table 4, 1.0 M DLꢀthreoꢀPS and 0.3 M
DLꢀthreoꢀMPS were resolved with >99% ee and the conversion reached nearly 50%
with 1,2ꢀdichloroethane or dichloromethane. In the case with cyclohexanone as
coꢀsolvent, the conversion reached up to 50% for 0.6 M DLꢀthreoꢀPS or 0.3 M
DLꢀthreoꢀMPS. Lower conversion and ee value of LꢀthreoꢀPS or LꢀthreoꢀMPS were
obtained when the substrate concentration was further increased. The resolution at
high concentration in aqueous phase without organic solvent resulted in lower ee
value of product.
To assess the applicability of DrDTA for the kinetic resolution, the reaction
was conducted in 100 mL reaction solution which contained 10 ꢁM PLP, 1 mM
11

Catalysis Science & Technology
Page 12 of 30
View Article Online
DOI: 10.1039/C7CY01774J
257
258
259
260
261
262
263
264
265
MnCl2, 12 g wet cells and 1 M DLꢀthreoꢀPS or 0.3 M DLꢀthreoꢀMPS with 50% (v/v)
dichloromethane. The reaction was also carried out under the same reaction
conditions with 0.6 M DLꢀthreoꢀPS or 0.3 M DLꢀthreoꢀMPS using 50% (v/v)
cyclohexanone as coꢀsolvent. As expected,
LꢀthreoꢀPS and LꢀthreoꢀMPS were
produced with excellent ee vales (>99%) and the products were obtained with 34.8%
and 37.2% isolated yield with 50% (v/v) dichloromethane, and 35.2% and 34.6%
isolated yield in the case of 50% (v/v) cyclohexanone, respectively. To the best of
our knowledge, these results represent the highest substrate concentration reported
so far for Dꢀthreonine aldolaseꢀcatalyzed stereospecific resolution.
266 3. Conclusions
267
268
269
270
271
272
273
274
275
276
277
278
279
280
The dta gene from Delftia sp. RIT313 was cloned and recombinantly expressed in
E.coli BL21(DE3). The substrate profiling indicated that the ꢀlowꢀTA (DrDTA)
D
accepted a wider range of acceptor substrates and was more active toward aromatic
aldehydes bearing electronꢀwithdrawing groups than those with electronꢀdonating
substituents. Molecular docking studies suggest that the substituent on the benzene
ring of the substrate is criticle in determining the enzyme activity and
stereoselectivity by affecting the interaction between the βꢀOHꢀgroup of the
substrate and the manganese ion. Through optimization of the reaction condition, the
reaction can be under kinetic control to retain suitable conversion and high
stereoselectivity at βꢀcarbon. Under such conditions,
Dꢀthreoꢀphenylserine with
2′ꢀfluoroꢀ or 3′ꢀnitroꢀsubstituent were obtained with >90% conversion and >90% de.
Furthermore, enzymatic resolution were performed with whole cell in twoꢀphase
system to prepare optically pure
LꢀthreoꢀPS and LꢀthreoꢀMPS at the highest
substrate concentrations reported so far. Synthesis of optically pure
12

Page 13 of 30
Catalysis Science & Technology
View Article Online
DOI: 10.1039/C7CY01774J
281
282
283
284
285
βꢀhydroxyꢀαꢀamino acids have been attracting a great deal of attention due to their
potential application as chiral building blocks for the preparation of pharmaceuticals
and agrichemicals. This study provides a promising biocatalyst to tackle the
“Cβꢀstereoselectivity problem” in the enzymatic production of highꢀvalue chiral
aromatic βꢀhydroxyꢀαꢀamino acids.
286 4. Experimental
287
288
289
290
291
292
293
294
295
4.1 Materials
pETꢀ21a and Escherichia coli BL21 (DE3) were used as expression vector and host
strains, respectively. Disruption of the ltaE and glyA gene of E. coli BL21 (DE3)
were carried out according to the published method35 and the resulting strain was
used in the expression of the DrDTA gene to produce the whole cell biocatalyst for
kinetic resolution. Yeast alcohol dehydrogenase (ADH) was obtained from
SigmaꢀAldrich. NADH was purchased from Roche. Other reagents and solvents
were purchased from SigmaꢀAldrich, Alfa Aesar or Sinopharm Chemical.
Phenylserine derivatives were synthesized as previously described.36, 37
296
297
298
299
300
301
302
303
4.2 Gene mining of
ꢀLowꢀTAs from A. xylosoxidans (Gene bank accession number: BAA86032) and
Arthrobacter sp. (BAA31547) were selected as templates to identify new ꢀlowꢀTAs
Dꢀthreonine aldolase, expression and purification
D
D
by using BLASTP search in NCBI. The obtained putative gene sequence from
Delftia sp. RIT313 (EZP46516) was selected for further study and the codons were
optimized using the JCat algorithm.38 The optimized gene was synthesized by
Shanghai Xuguan Biotechnological Co., Ltd (China) and ligated into the pETꢀ21a
plasmid between the NdeI and XhoI restriction sites. The recombinant plasmid was
13

Catalysis Science & Technology
Page 14 of 30
View Article Online
DOI: 10.1039/C7CY01774J
304
305
306
307
308
309
310
311
312
313
314
315
316
317
318
319
320
321
322
transformed into E. coli BL21 (DE3). Cells were grown in 800 mL of LuriaꢀBertani
medium containing 100ꢁg/ml of ampicillin at 37°C with shaking (200 rpm). When
the OD600 reached approximately 0.6ꢀ0.8, isopropylꢀβꢀDꢀthioꢀgalactopyranoside
(IPTG, 1 mM) was added. After induction at 30°C overnight, the cells were
harvested by centrifugation (6000×g, 15 min).
All the purification operations were performed at 4°C in MESꢀNaOH buffer
containing 2 mM dithiothreitol (DTT). After being suspended, the cells pellets were
disrupted by highꢀpressure homogenizer. The cell lysate was centrifuged at 12,000
×g for 30 min, filtered through a 0.45ꢀꢁm filter, and then purified by Ni affinity
chromatography with ÄKTA purifier. The column was equilibrated with binding
buffer (20 mM MESꢀNaOH, 50 mM NaCl, 20 mM imidazole, pH 6.0) and then
washed with a linear gradient of elution buffer (20 mM MESꢀNaOH, 50 mM NaCl,
500 mM imidazole, pH 6.0). Enzyme fractions were analyzed by SDSꢀPAGE.
Fractions with
DꢀlowꢀTA activity were dialyzed. The pure enzyme solution was
concentrated by ultrafiltration device (Amicon Ultraꢀ15 30ꢀkDa, Millipore) and
stored in 10% glycerol at ꢀ80°C. Protein concentration was determined with the
BCA protein assay kit (CWBIO, China). Molecular weight of the holoenzyme was
determined by size exclusion chromatography with Gel Filtration HMW Calibration
Kit using Superdex 200 Increase 10/300 GL column (GE Healthcare).
323
324
325
326
327
4.3 Activity assays
The retroꢀaldol activity was measured by an NADH coupled assay with ADH. The
released acetaldehyde was reduced to ethanol with NADH being oxidized to NAD+.
Reaction rate was measured by monitoring the decrease of absorbance at 340 nm (ε
= 6.22 mMꢀ1 cmꢀ1) using SpectraMax® M2e at 25°C. The assay mixture contained 50
14

Page 15 of 30
Catalysis Science & Technology
View Article Online
DOI: 10.1039/C7CY01774J
328
329
330
331
332
333
334
335
336
mM Dꢀthreonine, 100 ꢁM PLP, 1 mM MnCl2, 10 U ADH, 200 ꢁM NADH and
enzyme solution in 100 mM HEPESꢀNaOH buffer (pH 8.0). One unit (U) of
retroꢀaldol activity was defined as the amount of enzyme able to catalyze the
cleavage of 1 ꢁmol of Dꢀthreonine per minute.
The aldol addition assay mixture contained 1 M glycine, 100 mM acetaldehyde,
10 ꢁM PLP, 1 mM MnCl2 and enzyme solution in 100 mM MESꢀNaOH buffer (pH
6.0). After the reaction mixture was incubated at 25°C for desirable time, the
mixture was diluted and analyzed by HPLC after derivatization with
oꢀphthaldialdehyde/Nꢀacetylꢀcysteine (OPA/NAC).39
337
338
339
340
341
342
343
344
345
346
347
348
349
350
351
4.4 Characterization of DrDTA
Unless otherwise specified, glycine and acetaldehyde or Dꢀthreonine were used as
substrates for the activity assay. Since metal compounds may serve as cofactor,
MESꢀNaOH buffer (pH 5.0‒6.5), HEPESꢀNaOH (pH 7.0‒8.5), and CHESꢀNaOH
(pH 9.0‒10.0) were used. The optimum reaction pH and temperature were
determined by measuring the activity at different pH values (pH 5.0 to 10.0) and
temperatures (25ꢀ80°C). To investigate the pH stability, the enzyme solutions in
various pH buffers were incubated at 4°C for 12 h, and the remaining activities were
determined 25°C. To study the thermostability, the enzyme solutions in 100 mM of
MESꢀNaOH buffer (pH 6.0) were incubated at various temperatures (25ꢀ80°C) for
30 minutes, and the remaining enzyme activities were measured 25°C after being
cooled.
Metal ionꢀfree enzyme solution was prepared by adding EDTA. The retroꢀaldol
activity of metal ionꢀfree enzyme solution and the solutions with 1 mM of each
metal were measured. The optimum concentration of metal ion and PLP were also
15

Catalysis Science & Technology
Page 16 of 30
View Article Online
DOI: 10.1039/C7CY01774J
352
353
354
355
356
357
tested. The role of metal ion on thermal stability was investigated after the enzyme
solution was incubated at 25ꢀ80°C for 30 minutes in the presence of different
concentrations of Mn2+ ion.
Kinetic parameters were measured in the presence of varying concentration of
substrate (0.2 to 20 mM) at 25°C. Kinetic parameters were calculated by fitting to
the MichaelisꢀMenten equation.
358
359
360
361
362
363
364
365
366
367
368
369
370
371
4.5 Reaction condition optimization and specificity for acceptor and donor
Glycine and benzaldehyde were used as substrates to investigate the effect of
temperature, coꢀsolvent, reaction time and the amount of enzyme on the conversion
and de value. The standard reaction mixture consisted of 1 M glycine, 100 mM
benzaldehyde, 10 ꢁM PLP and 1 mM MnCl2 in 1 mL MESꢀNaOH buffer (100 mM,
pH 6.0). After the reaction mixture was stirred at various temperatures (4 to 35°C)
for 6 h or 12h, the mixture was diluted with 9 mL MeOH and stored at 4°C
overnight. The conversion and de value were analyzed after chiral derivatization
with OPA/NAC. The assay mixture with 10% v/v of organic solvents were measured
as described above. In order to explore the influence of the amount of enzyme and
reaction time, 5 U or 10 U enzyme solution were added to the standard reaction
mixture with 10% (v/v) CH3CN. Substrate specificity was measured under the
optimal reaction conditions. The donor specificity was tested in the aldol
condensation of acetaldehyde with Dꢀalanine or Dꢀserine as the donor.
372
373
374
375
4.6 Homology modeling and molecular docking
A homology model of DrDTA was constructed based on the known crystal
structure of AxDTA from A. xylosoxidans (PDB entry 4V15) using Discovery
Studio V4.5 (Accelrys). During the modeling, coꢀfactor PLP and Mn2+ was copied
16

Page 17 of 30
Catalysis Science & Technology
View Article Online
DOI: 10.1039/C7CY01774J
376
377
378
379
380
381
382
383
from the template. The model structure was further evaluated using Ramachandran
plot. As docking highꢀenergy intermediate rather than ground state of substrate led
to striking improvements for substrate prediction,40 external aldimine intermediates
were docking into the active site with CDOKER program. The runs were performed
on each ligand using the same setting parameters. According to the binding model of
the PLP part and the known crystal structure, the unsuited docking poses were
discarded. The remaining poses were analyzed with CDOKER energy and the lower
energy poses were selected for comparison.
384
385
386
387
388
389
390
391
392
393
394
395
396
397
398
4.7 Stereoselective resolution of DLꢀthreoꢀphenylserine and its derivatives using
whole cell biocatalyst
In order to avoid any side reactions, E. coli BL21 (DE3)ꢂltaEꢂglyA was used as
host for expression of DrDTA gene and the recombinant whole cells were used as
the biocatalyst in the kinetic resolution. The effect of biphasic reactions with
nonꢀwaterꢀmiscible organic solvents was investigated with 50% (v/v) of ethyl acetate,
nꢀbutyl acetate, nꢀheptane, nꢀhexane, nꢀoctane, isooctane, toluene, dichloromethane,
1,2ꢀdichloroethane, MTBE, 2ꢀMethyltetrahydrofuran, cyclohexanone, MeOH, EtOH,
CNCH3, Acetone, DMSO and THF, respectively. Control experiment without
organic solvent was manipulated under the same conditions. Enzymatic resolution
were performed at 30°C in 1 mL of 100 mM HEPESꢀNaOH buffer (pH 8.0) that
contained 10 ꢁM PLP, 1 mM MnCl2, 0.08 g/mL wet cells, 50% (v/v) organic solvent
and DLꢀthreoꢀPS or DLꢀthreoꢀMPS. Increased substrate concentrations of 0.2 to1.2
M for DLꢀthreoꢀPS or 0.2 to 0.4 M for DLꢀthreoꢀMPS were tested with 50% (v/v) of
1,2ꢀdichloroethane, dichloromethane and cyclohexanone, respectively.
17

Catalysis Science & Technology
Page 18 of 30
View Article Online
DOI: 10.1039/C7CY01774J
399
400
401
402
403
404
405
406
407
408
409
410
411
412
413
414
415
416
In a solution of PLP (0.25 mg, 1 ꢁmol), MnCl2 (12.6 mg, 0.1 mmol), and wet
cells (12.0 g) in 50 mL buffer (HEPESꢀNaOH, pH 8.0) and 50 mL dichloromethane,
DLꢀthreoꢀPS (18.1 g , 0.1 mol) or DLꢀthreoꢀMPS (7.8 g, 0.03 mol) was added, and
the resulting reaction mixture was stirred at 30°C. Another preparative scale
experiments was performed with DLꢀthreoꢀPS (10.9 g, 0.06 mol) or DLꢀthreoꢀMPS
(7.8 g, 0.03 mol) in 50% (v/v) of cyclohexanone at 30°C. After 12 h, the reaction
solution was centrifuged at 12000×g for 5 min. The organic layer and wet cells were
washed with water and the water layers were combined, neutralized to a pH 6 and
concentrated by vacuum evaporation. The remaining solution was cooled to 4°C to
give crystalline precipitate, which was filtered and washed with cold water. The
crystalline solid was dried under vacuum to give the product as white powder: 6.3 g
(>99% ee, 34.8% yield) of
L
ꢀthreoꢀPS and 2.9 g (>99% ee, 37.2% yield) of
ꢀthreoꢀPS
ꢀthreoꢀPS:
LꢀthreoꢀMPS for dichloromethane, and 3.8 g (>99% ee, 35.2% yield) of
L
and 2.7 g (>99% ee, 34.6% yield) of
1H NMR (D2O, 400MHz) δ (ppm) 7.29ꢀ7.40 (m, 5H), 5.21 (d, J = 4.4 Hz, 1H), 3.82
ꢀthreoꢀMPS: 1H NMR (D2O, 400MHz) δ (ppm) 7.96 (d, J = 12
LꢀthreoꢀMPS for cyclohexanone. L
(d, J = 4.4 Hz, 1H).
L
Hz, 2H), 7.70 (d, J = 8.0 Hz, 2H), 5.35 (d, J = 8.0 Hz, 1H), 3.92 (d, J = 4.4 Hz, 1H),
3.23 (s, 3H).
417
418
Conflicts of interest
419
420
421
There are no conflicts of interest to declare.
Acknowledgements
18

Page 19 of 30
Catalysis Science & Technology
View Article Online
DOI: 10.1039/C7CY01774J
422
423
424
425
426
This work was financially supported by the National Natural Science Foundation of
China (No.21602246) and the STS program of Chinese Academy of Sciences
(KFJꢀSWꢀSTSꢀ164).
References
427
428
429
430
431
432
433
434
435
436
437
438
439
440
441
442
443
444
445
446
447
448
449
450
451
452
453
454
455
456
1.
T. D. Machajewski and C. H. Wong, Angew. Chem., Int. Ed, 2000, 39,
1352ꢀ1375.
2.
3.
S. Joly and U. Hanefeld, Cheml. Soc. Rev, 2005, 34, 530ꢀ542.
S. M. Dean, W. A. Greenberg and C. H. Wong, Adv. Synth. Catal, 2007, 349,
1308ꢀ1320.
4.
P. Clapés, W.ꢀD. Fessner, G. A. Sprenger and A. K. Samland, Curr. Opin.
Chem.Biol, 2010, 14, 154ꢀ167.
5.
6.
P. Clapés and X. Garrabou, Adv. Synth. Catal,2011, 353, 2263ꢀ2283.
N. Duckers, K. Baer, S. Simon, H. Groger and W. Hummel, Appl. Microbiol.
Biotechnol, 2010, 88, 409ꢀ424.
7.
M. Ikemi, T. Morikawa, T. Miyoshi, S. Shimizu, M. Kataoka and H.
Yamada, Biochemical Engineering for 2001, Springer, 1992, 3, 96ꢀ99.
J. Q. Liu, T. Dairi, N. Itoh, M. Kataoka, S. Shimizu and H. Yamada, J. Biol.
Chem, 1998, 273, 16678ꢀ16685.
8.
9.
J. Q. Liu, T. Dairi, N. Itoh, M. Kataoka, S. Shimizu and H. Yamada, J Mol.
Catal. B:Enzym, 2000, 10, 107ꢀ115.
10.
11.
12.
13.
14.
15.
16.
A. Soler, X. Garrabou, K. Hernandez, M. L. Gutierrez, E. Busto, J. Bujons,
T. Parella, J. Joglar and P. Clapés, Adv. Synth. Catal, 2014, 356, 3007ꢀ3024.
J. Q. Liu, M. Odani, T. Yasuoka, T. Dairi, N. Itoh, M. Kataoka, S. Shimizu
and H. Yamada, Appl. Microbiol. Biotechnol, 2000, 54, 44ꢀ51.
T. Kimura, V. P. Vassilev, G.ꢀJ. Shen and C.ꢀH. Wong, J. Am. Chem. Soc,
1997, 119, 11734ꢀ11742.
K. Fesko, C. Reisinger, J. Steinreiber, H. Weber, M. Schürmann and H.
Griengl, J Mol. Catal. B:Enzym, 2008, 52ꢀ53, 19ꢀ26.
K. Fesko, G. A. Strohmeier and R. Breinbauer, Appl. Microbiol. Biotechnol,
2015, 1ꢀ11.
K. Fesko, M. Uhl, J. Steinreiber, K. Gruber and H. Griengl, Angew. Chem.,
Int. Ed, 2010, 49, 121ꢀ124.
Y. Hirato, M. Tokuhisa, M. Tanigawa, H. Ashida, H. Tanaka and K.
Nishimura, Phytochemistry, 2017, 135, 18ꢀ23.
19

Catalysis Science & Technology
Page 20 of 30
View Article Online
DOI: 10.1039/C7CY01774J
457
458
459
460
461
462
463
464
465
466
467
468
469
470
471
472
473
474
475
476
477
478
479
480
481
482
483
484
485
486
487
488
489
490
491
492
493
17.
18.
J. Steinreiber, K. Fesko, C. Mayer, C. Reisinger, M. Schürmann and H.
Griengl, Tetrahedron, 2007, 63, 8088ꢀ8093.
J. Steinreiber, K. Fesko, C. Reisinger, M. Schürmann, F. van Assema, M.
Wolberg, D. Mink and H. Griengl, Tetrahedron, 2007, 63, 918ꢀ926.
S. E. Franz and J. D. Stewart, Adv Appl Microbiol, 2014, 88, 57ꢀ101.
S. Goldberg, A. Goswami, Z. Guo, Y. Y.Chan, E. T. Lo, A. Lee, V. C. Truc,
K. J. Natalie, C. Hang, L. T. Rossano and M. A. Schmidt, Org. Process. Res.
Dev, 2015, 19, 1308ꢀ1316.
19.
20.
21.
M. A. Schmidt, E. A. Reiff, X. Qian, C. Hang, V. C. Truc, K. J. Natalie, C.
Wang, J. Albrecht, A. G. Lee, E. T. Lo, Z. Guo, A. Goswami, S.
Goldberg, J. Pesti and L. T. Rossano, Org. Process. Res. Dev, 2015, 19,
1317ꢀ1322.
22.
J. Liu, M. Odani, T. Dairi, N. Itoh, S. Shimizu and H. Yamada, Appl.
Microbiol. Biotechnol, 1999, 51, 586ꢀ591.
23.
24.
K. Fesko, Appl. Microbiol. Biotechnol, 2016, 100, 2579ꢀ2590.
M. K. Uhl, G. Oberdorfer, G. Steinkellner, L. RieglerꢀBerket, D. Mink, F.
van Assema, M. Schurmann and K. Gruber, Plos One, 2015, 10
M. A. Larkin, G. Blackshields, N. Brown, R. Chenna, P. A. McGettigan, H.
McWilliam, F. Valentin, I. M. Wallace, A. Wilm and R. Lopez,
Bioinformatics, 2007, 23, 2947ꢀ2948.
25.
26.
27.
28.
X. Robert and P. Gouet, Nucleic Acids Res, 2014, 42, W320ꢀW324.
F. P. Seebeck and D. Hilvert, J. Am. Chem. Soc, 2003, 125, 10158ꢀ10159.
M. D. Toscano, M. M. Muller and D. Hilvert, Angew. Chem., Int. Ed, 2007,
46, 4468ꢀ4470.
29.
30.
31.
32.
33.
K. Fesko, L. Giger and D. Hilvert, Bioorg. Med. Chem. Lett, 2008, 18,
5987ꢀ5990.
A. Paiardini, R. Contestabile, S. D'Aguanno, S. Pascarella and F. Bossa, BBA
– Proteins. Proteom, 2003, 1647, 214ꢀ219.
G. Liu, M. Zhang, X. Chen, W. Zhang, W. Ding and Q. Zhang, J. Mol. Evol,
2015, 80, 102ꢀ107.
V. Schirch, S. Hopkins, E. Villar and S. Angelaccio, J. Bacteriol, 1985, 163,
1ꢀ7.
C. M. Alder, J. D Hayler, R. K. Henderson, A. M. Redman, L. Shukla , L. E.
Shuster and H. F. Sneddon, Updating and further expanding GSK's solvent
sustainability guide. Green Chem, 2016, 18, 3879ꢀ3890.
H. E. Eastman, C. Jamieson and A. J. B. Watson, Development of Solvent
Selection Guides. Aldrichimica Acta, 2015, 48, 51ꢀ55.
34.
20

Page 21 of 30
Catalysis Science & Technology
View Article Online
DOI: 10.1039/C7CY01774J
494
495
496
497
498
499
500
501
502
35.
K. A. Datsenko and B. L. Wanner, Proc. Natl. Acad. Sci. USA, 2000, 97,
6640ꢀ6645.
36.
37.
38.
K. N. Shaw and S. W. Fox, J. Am. Chem. Soc, 1953, 75, 3421ꢀ3424.
R. Mita, T. Katoh, C. Higuchi and A. Yamaguchi, US Pat., 4,605,759, 1986.
A. Grote, K. Hiller, M. Scheer, R. Münch, B. Nörtemann, D. C. Hempel and
D. Jahn, Nucleic. Acids. Res, 2005, 33, W526ꢀW531.
39.
40.
I. MolnarꢀPerl, J. Chromatograp A, 2001, 913, 283ꢀ302.
J. C. Hermann, E. Ghanem, Y. C. Li, F. M. Raushel, J. J. Irwin and B. K.
Shoichet, J. Am. Chem. Soc, 2006, 128, 15882ꢀ15891.
503
21

Catalysis Science & Technology
Page 22 of 30
View Article Online
DOI: 10.1039/C7CY01774J
504
Table 1 Kinetic parameters of DrDTA
Km
kcat
[minꢀ1]
kcat/ Km
[mMꢀ1minꢀ1]
Substrate
[mM]
0.83±0.07 23.50±0.37 28.3
0.35±0.04 14.90±0.21 42.6
D
ꢀthreonine
Dꢀalloꢀthreonine
L
ꢀthreonine
n.d.
n.d.
—
—
—
—
Lꢀalloꢀthreonine
505
506
n.d.Not detected.
22

Page 23 of 30
Catalysis Science & Technology
View Article Online
DOI: 10.1039/C7CY01774J
507
Table 2 Effects of coꢀsolvents on conversion and de
Entry
1
Coꢀsolvent
non
% (v/v) Conv. (%)
threo:erythro
61.4:38.6
83.4:16.6
78.0:22.0
78.9:21.1
77.5:22.5
76.9:23.1
74.6:23.4
66.4:33.6
53.6:46.4
51.6:48.4
49.6:50.4
49.3: 50.7
44.0:56.0
43.9:56.1
de (%)
—
10
20
10
10
10
10
10
10
10
10
10
10
10
69.9
77.3
60.6
63.9
79.7
66.3
81.6
79.9
75.4
46.7
83.1
76.9
47.6
30.9
22.8
66.8
56.0
57.8
55.0
53.8
49.2
32.8
7.2
2
CH3CN
3
CH3CN
4
DMF
5
Acetone
DMSO
6
7
MeOH
8
EtOH
9
THF
10
11
12
13
14
Ethyl acetate
MTBE
3.2
0.8
1,4ꢀDioxane
Dichloromethane
Toluene
1.4
12.0
12.2
508
509
510
Reaction conditions: 1 mL solution containing glycine (1 M), benzaldehyde (100
mM), PLP (10 ꢁM), MnCl2 (1 mM) and DrDTA (19 U) at 10°C for 6 h.
23

Catalysis Science & Technology
Page 24 of 30
View Article Online
DOI: 10.1039/C7CY01774J
511
512
Table 3 Substrate specificity of DrDTAꢀcatalyzed aldol addition reaction
Entry
X
H
Conv. (%) de (%) Entry
X
2ꢀBr
Conv. (%) de (%)
1
2
68.3
64.1
92.2
63.1
92.8
85.8
29.8
25.9
42.8
47.5
91.1
64.5
93.2
66.4
97.8
82.1
92.5
66.4
93.9
94.2
11
12
13
14
15
16
17
18
19
20
18.5
17.9
7.1
n.d.
n.d.
n.d.
n.d.
n.d.
n.d.
n.d.
n.d.
n.d.
n.d.
2ꢀNO2
3ꢀNO2
4ꢀNO2
2ꢀF
3ꢀBr
3
2ꢀOH
4
3ꢀOH
7.8
5
4ꢀOH
<1.0
4.7
6
3ꢀF
4ꢀCOOH
3,4ꢀOH
3,4ꢀO(Me)
4ꢀO(Me)
4ꢀMe
7
4ꢀF
8.2
8
2ꢀCl
9.3
9
3ꢀCl
6.3
10
4ꢀMeSO2
6.1
513
514
515
516
517
Reaction condition: 1 mL solution containing glycine (1 M), aldehyde (100 mM),
PLP (10 ꢁM), MnCl2 (1 mM), 10% (v/v) CH3CN and DrDTA (5U) at 10°C for 12h.
n.d. Not determined.
24

Page 25 of 30
Catalysis Science & Technology
View Article Online
DOI: 10.1039/C7CY01774J
518
Table 4 Kinetic resolution of DLꢀthreoꢀPS and DLꢀthreoꢀMPS
Entry
Substrate
Coꢀsolvent
Wet cells
Conc.
[M]
1.0
Conv.
(%)
50
ee
(%)
>99
Time
(h)
9
(g/mL)
0.08
0.08
0.12
0.12
0.12
0.12
0.08
0.08
0.12
0.12
0.12
0.12
1
2
DLꢀthreoꢀPS
DLꢀthreoꢀPS
DLꢀthreoꢀPS
DLꢀthreoꢀPS
DLꢀthreoꢀPS
DLꢀthreoꢀPS
DLꢀthreoꢀMPS
DLꢀthreoꢀMPS
DLꢀthreoꢀMPS
DLꢀthreoꢀMPS
DLꢀthreoꢀMPS
DLꢀthreoꢀMPS
1,2ꢀDichloroethane
1,2ꢀDichloroethane
Dichloromethane
Dichloromethane
Cyclohexanone
1.2
1.0
1.2
0.6
0.8
0.3
0.4
0.3
0.4
0.3
0.4
48.8 95.4
50 >99
48.5 94.2
50 >99
47.2 89.1
50 >99
36.2 56.9
50 >99
37.5 60.0
50 >99
35.6 55.3
12
12
12
12
12
9
3
4
5
6
Cyclohexanone
7
1,2ꢀDichloroethane
1,2ꢀDichloroethane
Dichloromethane
Dichloromethane
Cyclohexanone
8
12
12
12
12
12
9
10
11
12
Cyclohexanone
519
520
521
Reaction condition: 1 mL solution containing racemic substrates, 10 ꢁM PLP, 1 mM
MnCl2 and 50% (v/v) coꢀsolvent1,2ꢀdichloroethane.
25

Catalysis Science & Technology
Page 26 of 30
View Article Online
DOI: 10.1039/C7CY01774J
522
523
524
525
526
527
528
529
530
Fig. 1 Sequences alignment of DrDTA and other
D
ꢀlowꢀTAs. Amino acid sequences
of proteins and their accession number are as follows:
DꢀlowꢀTAs from A.
xylosoxidans (BAA86032), Arthrobacter sp. (BBA31547), S. pomeroyi
(WP_011241927) and C. reinhardtii (BAV59441). The PLP binding site ( ) and
metal binding site ( ) are indicated.
△
○
26

Page 27 of 30
Catalysis Science & Technology
View Article Online
DOI: 10.1039/C7CY01774J
531
532
533
534
535
536
537
538
Fig. 2 pH and temperature profile. (a) pH profile for retroꢀaldol and aldol addition
activity. (b) pH stability profile for retroꢀaldol and aldol addition activity after the
enzyme solutions incubated in various pH buffers at 4°C for 12 h. (c) Temperature
profile for the aldol addition activity. (d) Temperature stability profile for the aldol
addition activity after the enzyme solutions incubated at various temperatures
(25ꢀ80°C) for 30 minutes. All assays were performed three times.
27

Catalysis Science & Technology
Page 28 of 30
View Article Online
DOI: 10.1039/C7CY01774J
539
540
541
542
543
544
545
Fig. 3 The role of coꢀfactors in DrDTAꢀcatalyzed reactions. (a) Relative activities of
DrDTA with various metal ions. (b) Specific activity of DrDTA treated with Mn2+ at
different concentrations. (c) Effects of Mn2+ concentrations on the thermal stability.
(d) Specific activity of DrDTA with PLP at different concentrations. All assays were
performed three times.
28

Page 29 of 30
Catalysis Science & Technology
View Article Online
DOI: 10.1039/C7CY01774J
546
547
548
549
Fig. 4 Molecular docking. External aldimine intermediates of phenylserine (a),
2ꢀfluorophenylserine (b), 2ꢀbromophenylserine (c) and 3ꢀnitrophenylserine (d) were
docked into the homology model structure. Amino acid residues are shown as gray
sticks and manganese ion is shown as purple. Dashed lines indicate hydrogen bond
interactions or metal coordination, and their lengths (Å).
550
551
552
553
29