Coupled Immobilized Amine Dehydrogenase and Glucose Dehydrogenase for Asymmetric Synthesis of Amines by Reductive Amination with Cofactor Recycling

Article abstract of DOI:10.1002/cctc.201601446

The amine dehydrogenase (AmDH) engineered from the phenylalanine dehydrogenase of Rhodococcus sp. M4 was directly immobilized on magnetic nanoparticles (MNP) from the cell-free extract containing his-tagged AmDH through affinity attachment to give AmDH-MNPs with high yield, enzyme loading efficiency, and specific enzyme loading. AmDH-MNPs showed higher activity and productivity than the free enzyme for the asymmetric reductive amination of 4-phenyl-2-butanone 1 a and phenylacetone 1 b, producing the corresponding amines (R)-2 a,b in 99 % ee and 99 % yield, and with recycling of NADH for up to 3956 times. AmDH-MNPs were easily recycled, retaining 91 % of the original productivity in the third cycle of the reductive amination of 1 a. Coupling of immobilized AmDH and immobilized glucose dehydrogenase (GDH) for the asymmetric reductive amination of 1 a gave (R)-2 a in 99 % ee and 74 % yield, with a total turnover number (TTN) of 2940 for NADH recycling. Both immobilized enzymes showed good recyclability, retaining 81 % productivity in the third reaction cycle. The developed method with coupled immobilized AmDH and immobilized GDH for the asymmetric reductive amination of ketones is useful for the synthesis of enantiopure amines, superior to the use of coupled isolated enzymes with enhanced catalytic performance and reduced enzyme cost through catalyst recycling.

Full text of DOI:10.1002/cctc.201601446

Accepted Article
Title: Coupled immobilized amine dehydrogenase and glucose
dehydrogenase for asymmetric synthesis of amines via reductive
amination with cofactor recycling
Authors: Ji LIU, Bryan Q. W. Pang, Joseph P Adams, Radka
snajdrova, and Zhi Li
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: ChemCatChem 10.1002/cctc.201601446
Link to VoR: http://dx.doi.org/10.1002/cctc.201601446
A Journal of
www.chemcatchem.org

10.1002/cctc.201601446
ChemCatChem
FULL PAPER
Coupled immobilized amine dehydrogenase and glucose
dehydrogenase for asymmetric synthesis of amines via reductive
amination with cofactor recycling
Ji Liu,[a] Bryan Q. W. Pang,[a] Joseph P. Adams,[b] Radka Snajdrova,[b] and Zhi Li*[a]
Abstract: The amine dehydrogenase (AmDH) engineered from the
phenylalanine dehydrogenase of Rhodococcus sp. M4 was directly
immobilized on magnetic nanoparticles (MNP) from the cell-free
extract containing his-tagged AmDH via affinity attachment to give
AmDH-MNP with high yield, enzyme loading efficiency, and specific
enzyme loading. AmDH-MNP showed higher activity and productivity
than the free enzyme for the asymmetric reductive amination of 4-
phenyl-2-butanone 1a and phenylacetone 1b, producing the
amine through enzymatic reductive amination of a ketone with
ammonia is highly desirable.[11-13] While the reductive amination is
naturally used to convert a keto acid to an amino acid, Bommarius
et al. engineered an amine dehydrogenase (AmDH) by directed
evolution of the leucine dehydrogenase of Bacillus
stereothermophilus for the asymmetric reductive amination of
methyl isobutyl ketone, to give (R)-1,3-dimethylbutylamine in
99.8% ee.[14] An AmDH was also engineered from the
corresponding amines (R)-2a-b in 99% ee and 99% yield, respectively, phenylalanine dehydrogenase of Bacillus badius for the amination
and recycling NADH for up to 3956 times. AmDH-MNP was easily
recycled, retaining 91% of the original productivity in the 3rd cycle of
the reductive amination of 1a. Coupling of immobilized AmDH and
immobilized GDH for the asymmetric reductive amination of 1a gave
(R)-2a in 99% ee and 74% yield, with a TTN of 2940 for NADH
recycling. Both immobilized enzymes showed good recyclability,
retaining 81% productivity in the 3rd reaction cycle. The developed
method with coupled immobilized AmDH and immobilized GDH for the
asymmetric reductive amination of ketones is useful for the synthesis
of enantiopure amines, being advantageous over the use of coupled
isolated enzymes with enhanced catalytic performance and reduced
enzyme cost through catalyst recycling.
of para-fluorophenylacetone to produce (R)-1-(4-fluorophenyl)-
propyl-2-amine in 99.8% ee,[15] and an enzyme chimera was
created with a distinctive substrate scope.[16] In our previous study,
a triple mutant of the phenylalanine dehydrogenase from
Rhodococcus sp. M4 was developed as an AmDH for the
asymmetric reductive amination of 4-phenyl-2-butanone 1a and
phenylacetone 1b to give (R)-1-methyl-3-phenylpropylamine 2a
and (R)-amphetamine 2b in >98% ee, respectively.[17] In addition
to the direct use of AmDH for the synthesis of a chiral amine from
a ketone, AmDH was also recently integrated in the cascade
biotransformations to synthesize chiral amines from alcohols.[18,
19]
Introduction
Chiral amines are an important class of compounds for
pharmaceutical synthesis. Compared with chemical approach,
biocatalytic preparation of chiral amines is green and highly
enantioselective.[1]
Enzymatic
kinetic
resolution
via
enantioselective acylation or oxidation produces a desired
enantiomer from racemic amines, but with maximal 50%
3]
theoretical yield.[2, Higher yields could be achieved through
dynamic kinetic resolution or deracemization using an amine
oxidase.[4, 5] Asymmetric synthesis of an enantiopure amine from
a ketone using a transaminase has received great attention and
been recently developed for many synthetic applications.[6-9]
Nevertheless, it requires the use of an amino donor in excess with
poor atom efficiency and also encounters equilibrium problem.[10]
From green chemistry point of view, the synthesis of a chiral
Scheme 1. a) Coupled immobilized AmDH and GDH for asymmetric reductive
amination of ketone 1a-b to (R)-2a-b with cofactor recycling. b) Immobilization
of his-tagged AmDH on Ni-NTA-MNP directly from cell-free extract by highly
specific affinity attachment.
[a]
[b]
J. Liu, B. Q. W. Pang, and Prof. Z. Li*
Department of Chemical & Biomolecular Engineering, National
University of Singapore, 4 Engineering Drive 4, Singapore 117585.
E-mail: chelz@nus.edu.sg; Fax: +65 6779 1936; Tel: +65 6516 8416
Dr. J. P. Adams and Dr. R. Snajdrova
Medicines Research Centre, GlaxoSmithKline R&D Ltd, Gunnels
Wood Road, Stevenage, Hertfordshire, SG1 2NY U.K.
AmDH-catalyzed reductive amination often uses high
concentration of NH3/NH4 and requires expensive NADH
+
cofactor,[11-19] which presents problems for practical synthesis. As
high concentration of ammonia is difficult to enter the cells for
whole-cell biocatalysis, free AmDH was used for the reductive
amination, and NADH was regenerated by using another free
enzyme, such as glucose dehydrogenase (GDH).[20-22] However,
Supporting information for this article is available on the WWW
under
For internal use, please do not delete. Submitted_Manuscript
This article is protected by copyright. All rights reserved.

10.1002/cctc.201601446
ChemCatChem
FULL PAPER
enzymes are expensive and also less stable at high salt
concentration.[23] To tackle these problems, we intended to
develop coupled immobilized AmDH and immobilized GDH for the
reductive amination with cofactor recycling. The reductive
amination of 1a-b catalyzed by the AmDH from Rhodococcus sp.
M4 to produce (R)-2a-b was selected as the target reaction
(Scheme 1a).[17] Enzyme immobilization could improve enzyme
stability and reduce enzyme cost through recycling. Many
immobilization techniques have been developed.[24-29] Among
them, the immobilization of enzymes on magnetic nanoparticles
(MNP) is very attractive, since the large surface area, low mass
transfer limitation, and magnetic properties of MNP could give
high specific enzyme loading, retain high enzyme activity, and
enable easy magnetic separation.[30-35] We previously developed
a nickel-nitrilotriacetic acid (Ni-NTA) functionalized iron oxide
MNP for enzyme immobilization via affinity attachment.[31, 33-35]
This method is chosen for convenient and efficient immobilization
of AmDH and GDH on MNP.
loading efficiency of 82%, a specific enzyme loading of 151 mg
protein g-1 MNP, and a total activity recovery of 80%. SDS-PAGE
showed a significantly reduced AmDH band in CFE after the
immobilization (Figure 1a), suggesting the attachment of the
majority of AmDH on MNPs. AmDHCFE-MNP was washed with
200 mM imidazole, and a single AmDH band was observed in the
SDS-PAGE of the eluent, indicating a high purity of the
immobilized enzyme on MNP and the excellent selectivity of the
immobilization method based on affinity attachment.
Results and Discussion
Figure 1. SDS-PAGE. a) Direct immobilization of AmDH on Ni-NTA-MNP from
the CFE of E. coli (AmDH) encoding his-tagged AmDH. Lane 1: CFE before
immobilization; Lane 2: CFE after immobilization; Lane 3: eluent from washing
AmDHCFE-MNP with 200 mM imidazole; M: Protein ladder; b) Immobilization of
GDH on Ni-NTA-MNP. Lane 1: GDH before immobilization; Lane 2: GDH after
immobilization; M: Protein ladder.
Immobilization of AmDH on Ni-NTA-MNPs
Ni-NTA-MNP was synthesized according to the published method
in 83% yield with a mean size of 138±40 nm.[31] An isolated his-
tagged AmDH fraction containing 8.5 mg protein mL-1 with an
activity of 39 U g-1 protein for the reductive amination of 1a to (R)-
2a was prepared for the immobilization (supporting information).
Treatment of the enzyme fraction with Ni-NTA-MNP at 4 °C for 12
h gave the immobilized enzyme, AmDHpure-MNP, with 95% yield
and 93% enzyme loading efficiency (Table 1 and Figure S1,
supporting information). The specific enzyme loading of
AmDHpure-MNP reached 188 mg protein g-1 MNP.
Direct immobilization of AmDH from CFE without enzyme
purification was then explored. Cell-free extract of E. coli (AmDH)
expressing his-tagged AmDH was prepared to contain 37 mg mL-
1 protein with an activity of 0.34 U mL-1 for the reductive amination
of 1a to (R)-2a. 10 mL CFE, which contained 37 mg his-tagged
AmDH, was treated with 200 mg Ni-NTA-MNPs giving AmDHCFE
MNP in 95% yield. Such immobilization afforded an enzyme
Specific activity of the immobilized AmDHs
The specific activity of AmDHpure-MNP and AmDHCFE-MNP was
determined for 30 min’s reductive amination of 1a (Figure S2,
supporting information). Compared with free AmDH (39 U g-1
protein), AmDHpure-MNP had a higher specific activity of 49 U g-1
protein, with a total activity recovery of 116% (Table 1). Similarly,
AmDHCFE-MNP obtained from the direct immobilization using the
CFE of E. coli (AmDH) also exhibited a higher specific activity (41
U g-1 protein) than the free enzyme. The high catalytic activity over
30 min’s reaction of the immobilized AmDHs was probably caused
by enhanced enzyme operational stability through immobilization,
similar to the immobilization of other enzymes.[34]
-
Table 1. Immobilization of AmDH and GDH on Ni-NTA-MNP, respectively, via affinity attachment.
Enzyme[a]
Enzyme
concentration
(mg prot. mL-1)
Ni-NTA-MNP
concentration
(mg mL-1)
Nano-catalyst
Yield
Spec. enzyme
loading
(mg prot. g-1
MNP)
Enzyme
loading
efficiency
(%)
Specific
activity
(U g-1 prot.)
Retained
enzyme
activity
(%)
Total
based
on MNP
(%)
activty
recovery
(%)
AmDHpure
AmDHCFE
GDHpure
8
40
20
50
AmDHpure-MNP
AmDHCFE-MNP
GDHpure-MNP
95±0
95±0
95±0
188±15
151±5[d]
147±7
93±8
82±3
71±4
49±6[b]
125±10
104±2
93±8
116±9
80±3
66±6
3.7[c]
10
41±3[b]
18700±1700[e]
[a] Enzyme immobilization was carried out at 4 °C and 20 rpm on a tube tumbler for 12 h. AmDHpure, isolated AmDH; AmDHCFE, AmDH from cell-free extract;
GDHpure, isolated GDH; [b] Specific activity was determined at 30 min in the reaction of 2 mmol NH4Cl, 19 mg AmDHpure-MNP or 23 mg AmDHCFE-MNP, 10
µmol 1a, 1 µmol NAD+, and 2 mg GDH in 1 mL Tris buffer (pH 9.0, 100 mM) at 30 °C and 250 rpm. [c] Concentration was estimated based on SDS-PAGE
using Quantity One software. [d] Specific enzyme loading was measured based on the eluted AmDH from AmDH-MNP using 200 mM imidazole. [e] Specific
activity was determined for 30 seconds reaction of 30 µg GDHpure-MNP, 1 µmol NAD+, and 20 mg glucose in 1 mL Tris buffer (pH 8.0, 50 mM) at room
temperature. The data with standard deviations are the mean values from triplicate experiments.
For internal use, please do not delete. Submitted_Manuscript
This article is protected by copyright. All rights reserved.

10.1002/cctc.201601446
ChemCatChem
FULL PAPER
Table 2. Reductive amination of 1a-b to (R)-2a-b with cofactor recycling using AmDHCFE-MNP and GDH.
Catalyst[a]
Substrate
Sub. conc.
(mM)
NAD+
(mM)
Time
(h)
Specific activity[b]
(U g-1 protein)
Product
Prod. conc.
(mM)
Yield
(%)
TTN[c]
AmDH
1a
1a
1a
1a
1a
1a
1a
1b
1b
1b
1b
1b
1b
1b
40
40
40
40
40
40
40
20
20
20
20
20
20
20
1
24
24
24
24
24
24
48
24
24
24
24
24
24
48
68.8
37.3
26.3
73.8
32.7
18.5
16.7
31.1
17.2
13.6
50.1
18.3
15.6
10.3
(R)-2a
(R)-2a
(R)-2a
(R)-2a
(R)-2a
(R)-2a
(R)-2a
(R)-2b
(R)-2b
(R)-2b
(R)-2b
(R)-2b
(R)-2b
(R)-2b
32.0
25.3
15.9
39.7
39.4
35.3
39.6
14.0
13.4
8.1
80
63
40
99
98
88
99
70
67
41
100
99
86
99
32
AmDH
0.1
0.01
1
253
1590
40
AmDH
AmDHCFE-MNP
AmDHCFE-MNP
AmDHCFE-MNP
AmDHCFE-MNP [d]
AmDH
0.1
0.01
0.01
1
394
3530
3956
14
AmDH
0.1
0.01
1
134
810
20
AmDH
AmDHCFE-MNP
AmDHCFE-MNP
AmDHCFE-MNP
AmDHCFE-MNP [d]
20.0
19.8
17.4
19.9
0.1
0.01
0.01
198
1740
1985
[a] Reaction was conducted in 1 mL Tris buffer (pH 9.0, 100 mM) containing 2 mmol NH4Cl, 3 mg AmDH or 23 mg AmDHCFE-MNP, substrate 1a or 1b, 2
mg GDH, 20 mg glucose, and NAD+ at 30 °C and 250 rpm. [b] Specific activity was measured at 30 min. [c] TTN for NADH recycling. [d] Reaction was
performed on 5 mL scale.
Reductive amination of 1a-b with cofactor recycling using
AmDHCFE-MNP and GDH
MNP was thus clearly demonstrated, which was possibly due to
the enhanced enzyme operational stability through the
immobilization.
AmDHCFE-MNP was used for the reductive amination of 1a-b (10
mM) in the presence of NAD+ (1 mM) and GDH at different pH
(7.0-10.0) and different concentrations (0.5-4 M) of NH4Cl (Figure
S3, supporting information). pH of 9.0 and 2 M NH4Cl were found
to be the optimal conditions. Although the use of 4M NH4Cl gave
slightly higher conversion, it degraded the stability of AmDH-MNP
and GDH, thus not being chosen for the reductive amination.
Reductive amination of 1a-b with AmDHCFE-MNP and GDH was
then performed under these conditions with different
concentrations of NAD+. Based on the specific activity of
AmDHCFE-MNP towards 1a-b and the biotransformation results
with a range of substrate concentrations, 40 mM 1a and 20 mM
1b was chosen for the reductive amination for achieving high
conversions. In the presence of 1 mM NAD+, AmDHCFE-MNP
showed higher activity and yield than free AmDH and fully
converted 1a and 1b to (R)-2a-b in 24 h (Table 2). The reductive
amination with lower NAD+ concentrations (0.1 mM and 0.01 mM)
was examined to improve the TTN of NADH recycling. In all cases,
the immobilized AmDH gave much higher yield than the free
enzyme. By using 0.01 mM NAD+, the reductive amination of 1a-
b for 24 h with AmDHCFE-MNP afforded (R)-2a-b in 88% and 87%
yield with TTN of 3530 and 1740, respectively. Under the same
conditions, free AmDH achieved only 40% and 41% yield and
TTN of 1590 and 810 for the amination of 1a and 1b, respectively,
for the recycling of NADH. The enhanced productivity of AmDH-
Figure 2. Time course of the reductive amination of 1a-b using AmDHCFE-MNP
and GDH. a) Reductive amination of 40 mM 1a to (R)-2a; b) reductive amination
of 20 mM 1b to (R)-2b. Reactions were conducted in 5 mL Tris buffer (pH 9.0,
100 mM) containing 10 mmol NH4Cl, 115 mg AmDHCFE-MNP, substrate 1a or
1b, 10 mg GDH, 100 mg glucose, and 0.05 µmol NAD+ at 30 °C and 250 rpm.
For internal use, please do not delete. Submitted_Manuscript
This article is protected by copyright. All rights reserved.

10.1002/cctc.201601446
ChemCatChem
FULL PAPER
Table 3. Reductive amination of 1a to (R)-2a with cofactor recycling using AmDHCFE-MNP and GDHpure-MNP.
Sub.[a]
Sub. conc.
(mM)
GDH-MNP
(mg mL-1)
AmDHCFE-MNP
(mg mL-1)
NAD+
(mM)
Spec. activity[b]
(U g-1)
Prod.
Prod. conc.
(mM)
Yield
(%)
ee[c]
(%)
TTN[d]
1a
1a
40
40
0.44
2.2
23
23
0.01
0.01
12.7
16.3
(R)-2a
(R)-2a
23.0
29.4
58
74
99
99
2304
2940
[a] Reaction was performed in 2 mL Tris buffer (pH 9.0, 100 mM) containing 4 mmol NH4Cl, AmDHCFE-MNP, GDHpure-MNP, 1a, NAD+, and 40 mg
glucose at 30 °C and 250 rpm for 24 h. [b] Specific activity was measured at 30 min. [c] Product ee was determined using chiral HPLC. [d] TTN for
NADH recycling.
The time courses of the reductive amination of 1a (40 mM)
and 1b (20 mM) using AmDHCFE-MNP and GDH with cofactor
recycling were shown in Figure 2. At the early stage of the
reaction, AmDHCFE-MNP rapidly converted 1a and 1b to the
corresponding product 2a and 2b. The specific activity reached
16.7 U g-1 protein and 10.3 U g-1 protein, respectively. 47%-49%
of conversion was reached at 8 h, and the reactions became
slower afterwards. At 48 h, (R)-2a and (R)-2b were produced both
in 99% yield and 99% ee, with TTN for NADH recycling of 3956
and 1985, respectively. AmDHCFE-MNP also showed high
catalytic specificity to give a clean reaction with no by-product
formation. Further improvement of product concentration could be
explored by process engineering via in situ removal of amine
products, which showed inhibition effect at the concentration
higher than 30 mM.
activity of 18.7 U mg-1 protein in 30 seconds of reaction, retaining
93% of the free enzyme activity (Figure S4, supporting
information). The reductive amination of 10 mM 1a using 0.7 mg
GDHpure-MNP (0.1 mg GDH) and 3 mg AmDH gave 4.2 mM (R)-
2a in 30 min, displaying 95% productivity of the catalysis by free
GDH and AmDH under the same conditions (supporting
information).
Reductive amination of 1a with cofactor recycling using
coupled immobilized AmDH and immobilized GDH
For efficient cofactor recycling in the AmDH-catalyzed reductive
amination, GDH was often used in excess.[15, 17] As given in Table
2, 2 mg GDH (40 U mL-1) was coupled with 23 mg AmDHCFE-MNP
Recycling of immobilized AmDH for the reductive amination
of 1a-b
The recyclability of AmDHCFE-MNP was examined for the
reductive amination of 40 mM 1a or 20 mM 1b in the presence of
0.01 mM NAD+ and GDH. After 2 h reaction in each cycle,
AmDHCFE-MNP was fully recovered under a magnetic field. After
washing with Tris buffer (pH 8.0, 50 mM), the catalyst was reused
for the next cycle of reaction under the same conditions. As shown
in the Figure 3a, AmDHCFE-MNP retained 73%-83% of the original
productivity in the 5th cycle. As the AmDH was attached to the
MNPs through a strong affinity binding (dissociation constant =10-
13
M),[31] no enzyme leakage was detected at each cycle by
measuring the protein content of the reaction mixture and washing
fractions. The recyclability of AmDHCFE-MNP was also examined
for the reductive amination of 20 mM 1a to (R)-2a with 12 h
reaction time in each cycle. The immobilized AmDH retained 91%
of the original productivity in cycle 3 (Figure 3b), giving 86%
overall yield.
Figure 3. a-b) Recycling of AmDHCFE-MNP for the reductive amination of 1a-b.
a) with 40 mM 1a or 20 mM 1b for 2 h reaction in each cycle. b) with 20 mM 1a
for 12 h reaction in each cycle. Reaction condition for each cycle in a-b): 1 mL
Tris buffer (pH 9.0, 100 mM) containing 1a or 1b, 2 mmol NH4Cl, 23 mg
AmDHCFE-MNP, 10 U GDH, 20 mg glucose, and 0.01 µmol NAD+ was incubated
at 30 ºC and 250 rpm. c-d) Recycling of both AmDHCFE-MNP and GDHpure-MNP
for the reductive amination of 1a. c) 2 h reaction in each cycle. d) 12 h reaction
each cycle. Reaction condition for each cycle in c-d): 1 mL Tris buffer (pH 9.0,
100 mM) containing 40 µmol 1a, 2 mmol NH4Cl, 23 mg AmDHCFE-MNP, 2.2 mg
GDHpure-MNP, 20 mg glucose, and 0.01 µmol NAD+ was incubated at 30 ºC and
250 rpm.
Immobilization of GDH on Ni-NTA-MNPs
Ni-NTA-MNP was also used to immobilize GDH. Treatment of 20
mg his-tagged GDH with 100 mg Ni-NTA-MNP at 4 °C for 12 h
gave GDH-MNP in 95% yield with 71% enzyme loading efficiency
(Table 1). The specific enzyme loading was 147 mg protein g-1
MNP, indicating a good loading capacity of the nano-carrier. The
obtained GDHpure-MNP was examined for the NADH generation
in the presence of 1 mM NAD+ and 20 mg mL-1 glucose in Tris
buffer (pH 8.0, 50 mM). The nano-catalyst showed a specific
For internal use, please do not delete. Submitted_Manuscript
This article is protected by copyright. All rights reserved.

10.1002/cctc.201601446
ChemCatChem
FULL PAPER
(0.2 U mL-1) in the presence of 0.01 mM NAD+ to convert 40 mM
1a to (R)-2a in 88% yield at 24h and 99% yield at 48h. To reduce
the amount of GDH used for this reaction, the initial experiment of
reductive amination with coupled immobilized GDH and
recyclability, retaining 91% of the original productivity in the 3rd
cycle of the reductive amination of 1a to (R)-2a.
Furthermore, a novel concept of using coupled immobilized
AmDH and immobilized GDH for reductive amination is
successfully demonstrated, with enhanced catalytic performance,
effective recycling of the cofactor, and reduced enzyme cost by
catalyst recycling. Coupled AmDHCFE-MNP and GDHpure-MNP for
the asymmetric reductive amination of 1a gave (R)-2a in 99% ee
and 74% yield, with the recycling of NADH for 2940 times. Both
AmDHCFE-MNP and GDHpure-MNP were fully recovered and
demonstrated good reusability. As shown in the reductive
amination of 1a to (R)-2a, 81% productivity was retained in the
third reaction cycle for the recycled catalysts. Thus, the reductive
amination of ketones by the coupled immobilized enzymes
provides a useful approach for the synthesis of enantiopure
amines.
immobilized AmDH was carried out at 0.44 mg mL-1 GDHpure
-
MNPs (1.1 U mL-1), 23 mg mL-1 AmDHCFE-MNP (0.2 U mL-1), 40
mM 1a, and 0.01 mM NAD+ for 24 h. A specific activity of 12.7 U
g-1 and yield of 58% were obtained. These values were lower than
those obtained with the experiment using 2 mg free GDH. The
less efficient NADH regeneration in comparison with the reductive
amination in such coupled enzymes system is possibly caused by
the higher Km of GDH to NAD+ (0.23 mM)[36] compared with the
Km of the AmDH to NADH (0.03 mM)[37]. Moreover, the real GDH
activity in the amination condition could be much lower than the
activity determined in Tris buffer with 1 mM NAD+ within 30
seconds of the reaction. To achieve higher yield, more GDH-MNP
(5.4 U mL-1) was hence used. As shown in Table 3, reaction of 23
mg mL-1 AmDHCFE-MNP (0.2 U mL-1) and 2.2 mg mL-1 GDHpure
-
MNP (5.4 U mL-1) for 24 h converted 74% of 1a (40 mM) to (R)-
2a, with a TTN of 2940. Thus, much less immobilized GDH was
used and the yield was only slightly lower than that using 2 mg
mL-1 of free GDH.
Experimental Section
Immobilization of AmDH on Ni-NTA-MNPs
The isolated AmDH (40 mg) was mixed with Ni-NTA-MNP (200 mg) in 5
mL of Tris buffer (pH 8.0, 50 mM) and incubated at 4 °C for 12 h to give
the immobilized AmDHpure-MNP. Ni-NTA-MNPs (200 mg) were added to
10 mL of the CFE of E. coli (AmDH) (370 mg protein; 37 mg His-tagged
AmDH) in Tris buffer (pH 8.0, 50 mM). The mixture was incubated on a
tube tumbler at 4 °C and 20 rpm for 12 h to give the immobilized enzyme
AmDHCFE-MNP. The catalyst was obtained through magnetic separation
and washed with Tris buffer (pH 8.0, 50 mM) containing 0.5 M NaCl. The
amount of the immobilized protein was deduced from the protein amount
in the enzyme fraction before and after immobilization as well as in the
washing buffer, determined by using Bradford method. AmDHCFE-MNP (10
mg) were treated with 200 mM imidazole at room temperature and 20 rpm
for 30 min to elute the immobilized protein. The purity of the protein was
checked by using SDS-PAGE. Specific enzyme loading and enzyme
loading efficiency of AmDHCFE-MNP was calculated based on the protein
amount in the eluent.
Recycling of immobilized AmDH and immobilized GDH for
the reductive amination of 1a
The recyclability of both immobilized enzymes were then
examined in the reductive amination of 40 mM 1a in the presence
of 0.01 mM NAD+ with 2 h reaction time in each cycle. Both nano-
catalysts were fully recovered by magnetic separation. They were
washed with Tris buffer (pH 8.0, 50 mM) and then used for the
next cycle of the reaction. No enzyme leakage was recorded
during the recycling due to the strong affinity attachment of the
enzymes to MNPs. As shown in Figure 3c, the two immobilized
enzymes retained 63% of their original activity in the 5th cycle.
Both immobilized enzymes were also examined for the recycling
in the reductive amination of 40 mM 1a to (R)-2a with 12 h reaction
time in each cycle (Figure 3d). 81% of the original productivity was
retained for both enzymes in the third cycle. These results clearly
demonstrated the excellent recyclability of the immobilized
AmDHCFE-MNP and GDHpure-MNP.
Specific activity of the immobilized AmDH
AmDHpure-MNP (19 mg) or AmDHCFE-MNP (23 mg) was added to 1 mL
Tris buffer (pH 9.0, 100 mM, 2 M NH4Cl) containing 1a (10 µmol), GDH (2
mg), NAD+ (1 µmol), and glucose (20 mg). The mixture was incubated at
30 °C and 250 rpm. The specific activity of the immobilized AmDH was
calculated in U g-1 protein (1 U = 1 µmol 2a min-1) based on the amount of
the produced 2a at 30 min, measured by HPLC analysis.
Conclusions
For the first time, an AmDH is immobilized for catalyzing reductive
amination. The immobilized AmDH showed higher productivity
than the free AmDH due to the stabilization of enzyme via
immobilization. The direct immobilization of AmDH on MNP from
cell-free extract of E. coli (AmDH) expressing his-tagged AmDH
was achieved with excellent selectivity as well as high yield,
enzyme loading efficiency, specific enzyme loading, and catalytic
activity. The immobilization method without the pre-purification of
the enzyme is simple and practical. The asymmetric reductive
amination of 1a-b catalyzed by AmDHCFE-MNP in the presence of
0.01 mM NAD+ and GDH gave 99% yield with a TTN of NADH
recycling of 3956. The AmDH-MNP also showed good
Reductive amination of 1a-b with cofactor recycling using
immobilized AmDH and GDH
AmDHCFE-MNP (115 mg) were added to 5 mL Tris buffer (pH 9.0, 100 mM,
2 M NH4Cl) containing 1a (200 µmol) or 1b (100 µmol), GDH (10 mg),
NAD+ (0.05 µmol), and glucose (100 mg) in a 25 mL conical flask. The
reaction mixture was incubated at 30 °C and 250 rpm for 48 h. Samples
were taken at 0.5, 2, 4, 8, 12, 24, 36, and 48 h, respectively, for HPLC
analysis of the concentration of 1a-b and 2a-b. The ee value of (R)-2a-b
was measured by chiral HPLC analysis of the samples taken at 48 h.
Recycling of immobilized AmDH for the reductive amination of 1a-b
For internal use, please do not delete. Submitted_Manuscript
This article is protected by copyright. All rights reserved.

10.1002/cctc.201601446
ChemCatChem
FULL PAPER
A mixture of AmDHCFE-MNP (23 mg), 1a (40 µmol) or 1b (20 µmol), GDH
(2 mg), NAD+ (0.01 µmol), and glucose (20 mg) in 1 mL of Tris buffer (pH
9.0, 100 mM, 2 M NH4Cl), was shaken at 30 °C and 250 rpm for 2 h. The
nano-catalyst was separated under a magnet, washed with Tris buffer (pH
8.0, 50 mM) and then used for another cycle of reductive amination under
the same conditions as described above. The recycling of AmDHCFE-MNP
was also performed in the reductive amination of 20 mM 1a for 12 h for 3
cycles by using the same procedures.
[4]
[5]
M. T. Reetz, S. Klaus, Chimia 1996, 50, 668-669.
D. Ghislieri, A. P. Green, M. Pontini, S. C. Willies, I. Rowles, A. Frank, G.
Grogan, N. J. Turner, J. Am. Soc. Chem. 2013, 135, 10863-10869.
D. Koszelewski, K. Tauber, K. Faber, W. Kroutil, Trends Biotechnol. 2010,
28, 324-332.
[6]
[7]
[8]
M. S. Malik, E. S. Park, J. S. Shin, Appl. Microbiol. Biotechnol. 2012, 94,
1163-1171.
C. K. Savile, J. M. Janey, E. C. Mundorff, J. C. Moore, S. Tam, W. R.
Jarvis, J. C. Colbeck, A. Krebber, F. J. Fleitz, J. Brands, P. N. Devine, G.
W. Huisman, G. J. Hughes, Science 2010, 329, 305-309.
S. K. Wu, Y. Zhou, T. W. Wang, H. Too, D. I. C. Wan, Z. Li, Nat. Commun.
2016, 7, 11917.
Immobilization of GDH on Ni-NTA-MNPs
[9]
Ni-NTA-MNPs (50 mg) were added to 1 mL Tris buffer (pH 8.0, 50 mM)
containing isolated GDH (10 mg). The mixture was incubated on a tube
tumbler at 4 °C and 20 rpm for 12 h, and the immobilized GDHpure-MNP
was separated by using a magnet and washed with Tris buffer (pH 8.0, 50
mM) containing 0.5 M NaCl. The amount of the immobilized protein was
calculated from the protein amount in the immobilization buffer before and
after the immobilization as well as in the washing buffer, measured by
using Bradford reagent.
[10] P. Tufvesson, J. Lima-Ramos, J. S. Jensen, N. Al-Haque, W. Neto, J. M.
Woodley, Biotechnol. Bioeng. 2011, 108, 1479-1493.
[11] M. Breuer, K. Ditrich, T. Habicher, B. Hauer, M. Kesseler, R. Sturmer, T.
Zelinski, Angew. Chem. Int. Ed. 2004, 43, 788-824.
[12] D. Koszelewski, I. Lavandera, D. Clay, G. M. Guebitz, D. Rozzell, W.
Kroutil, Angew. Chem. Int. Ed. 2008, 47, 9337-9340.
[13] X. Gao, F. Huang, J. Feng, X. Chen, H. Zhang, Z. Wang, Q. Wu, D. Zhu,
Appl. Environ. Microbiol. 2013, 79, 5078-5081.
[14] M. J. Abrahamson, E. Vazquez-Figueroa, N. B. Woodall, J. C. Moore, A.
S. Bommarius, Angew. Chem. Int. Ed. 2012, 51, 3969-3972.
[15] M. J. Abrahamson, J. W. Wong, A. S. Bommarius, Adv. Synth. Catal.
2013, 355, 1780-1786.
Reductive amination of 1a with cofactor recycling using coupled
immobilized AmDH and immobilized GDH
A mixture of AmDHCFE-MNP (46 mg), 1a (80 µmol), GDHpure-MNP (0.9 or
4.4 mg), NAD+ (0.02 mM), and glucose (40 mg) in 2 mL of Tris buffer (pH
9.0, 100 mM, 2 M NH4Cl) was incubated at 30 °C and 250 rpm for 24h.
The amount of 1a and 2a, and the ee value of the produced (R)-2a were
measured by HPLC analysis.
[16] B. R. Bommarius, M. Schurmann, A. S. Bommarius, Chem. Commun.
2014, 50, 14953-14955.
[17] L. J. Ye, H. H. Toh, Y. Yang, J. P. Adams, R. Snajdrova, Z. Li, ACS Catal.
2015, 5, 1119-1122.
[18] F. G. Mutti, T. Knaus, N. S. Scrutton, M. Breuer, N. J. Turner, Science
2015, 349, 1525-1529.
Recycling of immobilized AmDH and immobilized GDH for the
reductive amination of 1a
[19] F. F. Chen, Y. Y. Liu, G. W. Zheng, J. H. Xu, ChemCatChem 2015, 7,
3838-3841.
[20] J. Zhang, B. Withold, Z. Li, Chem. Commun. 2006, 398-400.
[21] W. Zhang, K. O’Connor, D. I. C. Wang, Z. Li, Appl. Environ. Microbiol.
2009, 75, 687-694.
AmDHCFE-MNP (23 mg), 1a (40 µmol), GDHpure-MNP (4.4 mg), NAD+ (0.01
µmol), and glucose (20 mg) in 1 mL of Tris buffer (pH 9.0, 100 mM, 2 M
NH4Cl) was incubatedat 30 °C and 250 rpm for 2 or 12 h. The catalyst was
then separated from the reaction mixture by using a magnet, and the
biotransformation was analyzed by using HPLC. The recovered catalyst
was rinsed with Tris buffer and then used for a new cycle of reductive
amination under the same reaction conditions as described above.
[22] J. Liu, Z. Li, ACS Catal. 2013, 3, 908-911.
[23] P. V. Iyer, L. Anathanarayan, Process Biochem. 2008, 43, 1019-1032.
[24] R. Sheldon, Adv. Synth. Catal. 2007, 349, 1289-1307.
[25] R. A. Sheldon, S. van Pelt, Chem. Soc. Rev. 2013, 42, 6223-6235.
[26] O. Barbosa, C. Ortiz, A. Berenguer-Muricia, R. Torres, R. C. Rodrigues,
R. Fernadez-Lafuente, Biotech. Adv. 2015, 33, 435-456.
[27] R. Huang, S. Wu, A. Li, Z. Li, J. Mater. Chem. A 2014, 2, 1672-1676.
[28] R. Huang, M. Wu, M. J. Goldman, Z. Li, Biotechnol. Bioeng. 2015, 112,
1092-1101.
Acknowledgements
[29] P. A. Limadinata, A. Li, Z. Li, Green Chem. 2015, 17, 1194-1203.
[30] W. Wang, Y. Xu, D. I. C. Wang, Li, Z. J. Am. Chem. Soc. 2009, 131,
12892-12893.
This work was financially supported by GlaxoSmithKline (GSK)
and Singapore Economic Development Board (EDB) through a
Green and Sustainable Manufacturing Grant (Project No. 279-
000-348-592).
[31] W. Wang, D. I. C. Wang, Li, Z. Chem. Commun. 2011, 47, 8115-8117.
[32] T. P. N. Ngo, W. Zhang, W. Wang, Z. Li, Chem. Commun. 2012, 48,
4585-4587.
[33] T. P. N. Ngo, A. Li, K. W. Tiew, Z. Li, Biores. Tech. 2013, 145, 233-239.
[34] A. K. Vahidi, Y. Yang, T. P. N. Ngo, Z. Li, ACS Catal. 2015, 5, 3157-3161.
[35] A. K. Vahidi, Z. S. Wang, W. S. Y. Wong, Z. Li, Catal. Sci. Technol. 2016,
6, 6286-6293.
Keywords: Biocatalysis • Biotransformations • Enzyme
immobilization
[36] Y. Fujita, R. Ramaley, E. J. Freese, Bacteriol. 1977, 132, 282-293.
[37] N. M. Brunhuber, J. B. Thoden, J. S. Blanchard, J. L. Vanhooke,
Biochemistry 2000, 39, 9174-9187.
[1]
[2]
[3]
R. N. Patel, Coord. Chem. Rev. 2008, 252, 659-701.
M. Höhne, U. T. Bornscheuer, ChemCatChem 2009, 1, 42-51.
D. Ghislieri, N. J. Turner, Top. Catal. 2014, 57, 284-300.
For internal use, please do not delete. Submitted_Manuscript
This article is protected by copyright. All rights reserved.