Enantioselective Bioamination of Aromatic Alkanes Using Ammonia: A Multienzymatic Cascade Approach

Article abstract of DOI:10.1002/cctc.201902253

Chiral amines are common drug building blocks and important active pharmaceutical ingredients. Preparing these functionalized compounds from simple materials, such as alkanes, is of great interest. We recently developed an artificial bioamination cascade for the C?H amination of cyclic alkanes by combining P450 monooxygenase, alcohol dehydrogenase, and amine dehydrogenase. Herein, this system has been extended to the synthesis of chiral aromatic amines. In the first hydroxylation step, process optimization increased the conversion to 77 %. Two stereoselectively complementary alcohol dehydrogenases and an amine dehydrogenase were selected for the bioconversion of aromatic hydrocarbons to amines. The amination reaction was optimized with respect to cofactor addition and enzyme dosage. Isopropanol was added to decrease ketone intermediate accumulation in the amination step, which further enhanced the overall conversion. This cascade system converted a panel of hydrocarbon substrates into the corresponding amines with excellent optical purity (>99 % ee) and moderate conversion ratios (13–53 %).

Full text of DOI:10.1002/cctc.201902253

Accepted Article
Title: Enantioselective Bioamination of Aromatic Alkanes Using
Ammonia: A Multienzymatic Cascade Approach
Authors: Hui Wang, Yu-Cong Zheng, Fei-Fei Chen, Jian-He Xu, and
Hui-Lei Yu
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.201902253
Link to VoR: http://dx.doi.org/10.1002/cctc.201902253
A Journal of
www.chemcatchem.org

10.1002/cctc.201902253
ChemCatChem
FULL PAPER
Enantioselective Bioamination of Aromatic Alkanes Using
Ammonia: A Multienzymatic Cascade Approach
Hui Wang,[a] Yu-Cong Zheng,[a] Fei-Fei Chen,[a] Jian-He Xu[a] and Hui-Lei Yu*[a]
Abstract: Chiral amines are common drug building blocks and
important active pharmaceutical ingredients. Preparing these
functionalized compounds from simple materials, such as alkanes, is
of great interest. We recently developed an artificial bioamination
cascade for the C–H amination of cyclic alkanes by combining P450
However, owing to equilibrium issues in the amine transfer
reactions, excess amounts of organic amino donors had to be
supplied, leading to a low atom economy.[6] This system was
further expanded to reductive coupling of a broad set of carbonyl
compounds with a variety of primary and secondary amines,
monooxygenase, alcohol dehydrogenase, and amine dehydrogenase. catalyzed by reductive aminases.[7] Compared to enzymatic
Herein, this system has been extended to the synthesis of chiral
aromatic amines. In the first hydroxylation step, process optimization
increased the conversion to 77%. Two stereoselectively
amination using organic amine donors, Bommarius and
coworkers pioneered biocatalytic amination of carbonyl
compounds using inorganic ammonium as amino donor by a
“amine dehydrogenase” (AmDH) engineered from amino acid
dehydrogenase.[8] This class of innovative enzymes have been
applied in generating varied of amines from the corresponding
alcohols, with the water as the only by-product.[9] As AmDHs
showed great synthetic potential as well as environmental
acceptability, we explored a multienzymatic cascade reaction for
the C–H amination of alkanes, employing AmDH for the amination
step at almost the same time.[10]
complementary
alcohol
dehydrogenases
and
an
amine
dehydrogenase were selected for the bioconversion of aromatic
hydrocarbons to amines. The amination reaction was optimized with
respect to cofactor addition and enzyme dosage. Isopropanol was
added to decrease ketone intermediate accumulation in the amination
step, which further enhanced the overall conversion. This cascade
system converted a panel of hydrocarbon substrates into the
corresponding amines with excellent optical purity (>99% ee) and
moderate conversion ratios (13–53%).
There are, however, still some practical problems in these
systems. Usually, the cascade reactions stall either at the
hydroxylation step (about 30% conversion for hydroxylation) or at
the amination step, with a limited amount of amine produced.
Amination by transaminase suffered from inhibition by the
byproduct acetone,[11] resulting in a very poor titer of the final
product with 5-26% overall conversion. Another issue for low
conversion was attributed to the poor performance of cytochrome
P450 monooxygenase in the presence of ammonium or amines,
Although hydroxylation step was separated from the amination
step to facilitate C–H activation, namely “one-pot–two-step”
cascade reaction to secure the overall conversions, amination of
cyclohexane to N-propargylcyclohexylamine was achieved with
only a 19.6% conversion, with reductive aminases usually
showing quantitative conversions of carbonyl compounds.[7,12]
Similar results were also obtained in our previous reported
cascade, the overall conversion (29.8%) from cyclohexane to the
corresponding amine was still very low.[10] In addition, although
the formation of optically pure (R)-1-phenylethylamine was
demonstrated, only 2.2% conversion of ethylbenzene was
achieved. As a broad substrate scope is a prerequisite for wide
Introduction
Amines are important intermediates widely used in the production
of pharmaceuticals, fine chemicals, agrochemicals, polymers,
dyestuffs, pigments, emulsifiers, and plasticizing agents.[1]
However, their industrial production mainly depends on the
expensive metal-catalyzed hydrogenation of enamides produced
from ketone precursors.[2] Therefore, in the last decade, the direct
amination of alcohols has been developed to avoid tedious
protection and deprotection processes, and the use of transition
metal complexes. However, some challenges remain to be
overcome, such as low efficiency, harsh conditions, and low
chemoselectivity.[3] In addition to the concept of single-step
amination, the synthesis of high-value amines, especially chiral
amines, from unfunctionalized alkanes using multiple enzymatic
cascades has received increasing interest.[4]
To implement C–H amination, current multienzymatic
cascade systems usually comprise three modules, namely, C–H
hydroxylation involving heme proteins, oxidoreductase-catalyzed
oxidation, and amination. Flitsch and coworkers first
demonstrated a cascade involving transaminase for the C–H
amination of ethylbenzene and several substituted substrates.[5]
[a]
Hui Wang, Yu-Cong Zheng, Dr. Fei-Fei Chen, Prof. Dr. Jian-He Xu
and Prof. Dr. Hui-Lei Yu
State Key Laboratory of Bioreactor Engineering
East China University of Science and Technology
130 Meilong Road, Shanghai 200237 (P. R. China)
E-mail: huileiyu@ecust.edu.cn
Scheme 1. Biocatalytic cascade for asymmetric C-H amination of
ethylbenzene and its substituted derivatives.
Supporting information for this article is given via a link at the end of
the document.
This article is protected by copyright. All rights reserved.

10.1002/cctc.201902253
ChemCatChem
FULL PAPER
Table 1. Characterization of component enzymes needed in the cascade reaction.
Spec. activity[a]
[b]
ee
t1/2
Enzyme
Substrate
Cofactor
[U/mg]
[%]
[h]
n.d.[c]
160
29
Monooxygenase
Alcohol dehydrogenase
Amine dehydrogenase
P450BM3 19A12
ScCR
Ethylbenzene
1-Phenylethanol
1-Phenylethanol
Acetophenone
NADPH
NAD+
NAD+
2.35 ± 0.07
50 (R)
96 (R)
> 99 (S)
> 99 (R)
0.45 ± 0.003
2.91 ± 0.04
(26 ± 1.9) × 10-3
PpADH
187
EsAmDH
NADH
[a] Analytical buffer: P450BM3 19A12 in potassium phosphate buffer (100 mM, pH 8.0); ScCR, PpADH, AmDH in NH4OH/NH4Cl buffer (2 M, pH 9.5);
Substrate concentration: 1 mM ethylbenzene for P450; 2 mM 1-phenylethanol for ScCR and PpADH; 10 mM acetophenone for EsAmDH, All
measurements were performed in triplicate. [b] Determined in NH4OH/NH4Cl buffer (2 M, pH 9.5). [c] n.d.: not detected.
synthetic utility, especially for chiral amine synthesis, this issue
must be further explored.
Table 2. Optimization of filling volume for the
hydroxylation reaction
Herein, we report a further optimized system capable of
producing a panel of chiral aromatic amines with excellent ee
values and moderate overall conversions (Scheme 1). Using the
amination of ethylbenzene as a model reaction, we reconstructed
the C–H amination cascade by combining a cytochrome P450
monooxygenase, two stereoselective complementary alcohol
dehydrogenases, and an amine dehydrogenase. The reaction
conditions of each step were carefully investigated, providing
further insight into multienzymatic cascade C–H amination
reaction.
Product conc. [mM]
Filling volume
[%]
1-Phenylethanol
3.85 ± 0.45
5.51 ± 0.32
Acetophenone
0.56 ± 0.09
0.37 ± 0.01
10
20
40
5.36 ± 0.04
0.22 ± 0.00
Reactions were performed in potassium phosphate buffer (100 mM
pH 8.0) containing 0.11 μM active P450BM3 19A12, 2 U/mL BmGDH,
10 mM ethylbenzene in 50 mL shaking flasks at 25 °C, 180 rpm, 4
h. All reactions were performed in triplicate.
Table 3. Optimization of shaking speed for hydroxylation
reaction.
Results and Discussion
Product conc. [mM]
Shaking
speed [rpm]
1-Phenylethanol
Acetophenone
0.43 ± 0.02
0.51 ± 0.04
0.51 ± 0.03
Enzyme screening and characterization
100
120
150
6.42 ± 0.24
7.21 ± 0.36
6.37 ± 0.13
Initially, the amination of ethylbenzene (1a) was selected as the
model reaction, for which only very poor conversion (2.2%) was
observed in our previous study. To facilitate the hydroxylation step,
the catalytic performances of well-known P450 monooxygenase
P450BM3[13] and two variants, P450BM3 F87V[14] and P450BM3 19A12
were compared (Table S1). Mutant P450BM3 19A12 was found to
have higher activity toward ethylbenzene. Therefore, this variant
was selected for further studies.
By coupling with the appropriate enantioselective nicotine
cofactor-dependent oxidoreductases, intermediate alcohols were
supposed to be converted into the corresponding ketones, while
unmatched oxidoreductases might result in alcohol accumulation.
Accordingly, the enantioselectivity of all biocatalysts in the current
study was determined. The results indicated that the
enantioselectivity of P450BM3 19A12 was moderate, with both (R)-
and (S)-alcohols produced (Table 1). In contrast, our previously
selected dehydrogenation module, namely, the oxidoreductase
from Streptomyces coelicolor (ScCR), exhibited reductive activity
toward acetophenone (2a) with excellent (R)-selectivity (96%
ee).[16] To further facilitate the transformation of residual (S)-2a
into the corresponding ketone, another complementary alcohol
dehydrogenase that tolerated high concentrations of
NH4OH/NH4Cl buffer was needed. After screening a series of
oxidoreductases (Table S2), one candidate originating from
Paracoccus pantotrophus (PpADH) was selected by
consideration of both the activity and stability.
180
6.11 ± 0.08
0.50 ± 0.02
Reaction conditions: 10 mL reaction mixture containing potassium
phosphate buffer (100 mM, pH 8.0) in a 50 mL shaking flask, 0.11
μM purified P450BM3 19A12, BmGDH: 2 U/mL; substrate loading: 10
mM; 25 °C, 4 h. All measurements were performed in triplicate.
[15]
,
dehydrogenases toolbox. EsAmDH (amine dehydrogenase from
Exiguobacterium sibiricum), which was developed by our group,
was finally selected owing to its high activity among other amine
dehydrogenases (Table S3).
Hydroxylation of ethylbenzene by P450BM3 19A12
The hydroxylation step is the prerequisite to maintaining an
efficient multienzymatic cascade C–H amination system.
However, reported alkane hydroxylation examples generally have
low conversion. These results suggested that the limiting step
inthe cascade reaction was hydroxylation catalyzed by P450,
indicating that optimization of this step was imperative.
In the first set of experiments, as ethylbenzene and other
alkanes are highly volatile, the reaction was performed in an air-
tight flask to reduce evaporation of the alkane substrate. Both the
evaporation of volatile substances and oxygen mass transfer
were affected by the filling volume and shaking speed of the
reaction flask. In this hydroxylation reaction, we found that
different filling volumes in the range of 10%–20% significantly
affected the reaction conversion, while reaction performance did
Meanwhile, amine dehydrogenases, which also played a vital
role in the amination step, were screened from our amine
This article is protected by copyright. All rights reserved.

10.1002/cctc.201902253
ChemCatChem
FULL PAPER
For the conversion of alcohol to amine, we have previously
reported using a redox-neutral cascade.[9] This reaction was
conducted in NH4OH/NH4Cl buffer, with alcohol oxidation and
ketone amination proceeding consecutively. First, the cofactor
concentration was optimized, with amine formation found to
not further improve when the filling volume was further increased
to 40% (Table 2). As increased liquid volume led to a decrease in
substrate evaporation as well as gas-liquid mass transfer (Table
S4). The shaking frequency was subsequently optimized (from
100 rpm to 180 rpm) to facilitate the reaction, on the base of 20%
filling volume.
To our surprise, optimal conversion was observed at a
shaking speed of 120 rpm, with a 17% increase compared to that
at 180 rpm (Table S5). For the further analysis of the
hydroxylation reaction, since the increase of shaking speed
elevated the initial hydroxylation rate (Table S5), and the OTR is
far beyond the hydroxylation rate, the limiting factor in the
beginning of the reaction was the mass transfer of substrate in the
liquid phase. According to the Henry's law and saturated vapor
pressure of ethylbenzene (1.33 KPa, 25.9 °C), about 21%
ethylbenzene distributed into the gas phase theoretically, which is
consistent to our measured data (78% residual in the liquid phase).
The increase of shaking speed also increased the substrate
volatility, and decreased the final reaction conversion, which thus
mainly be attributed to the slow transfer efficiency of substrate
from the gas phase into the liquid phase.
80
Ⅱ
Ⅰ
1- Phenylalamine
Acetophenone
60
40
20
0
1- Phenylethanol
0
10
20
30
40
50
60
Reaction time [h]
Figure 2. Time course of C–H amination of ethylbenzene. The reaction was
performed in the “20 mL-scale One-pot–two-step” cascade described in the
experimental section.
2.0
1.0 mM NAD+
0.5 mM NAD+
0.2 mM NAD+
1.5
1.0
0.5
0.0
0.1 mM NAD+
increase with increasing NAD+ concentration from 0.1 mM to 0.5
mM. However, further increasing the NAD+ concentration did not
enhance conversion (Figure 1). The reaction time course showed
that a high initial NAD+ concentration resulted in more ketone
remaining in the reaction mixture (Figure S2). Finally, 0.5 mM
NAD+ was selected for subsequent cascade reactions. To achieve
maximum conversion and avoid catalyst waste, the enzyme
dosages of ScCR, PpADH, and EsAmDH were optimized. The
enzyme dosages were set as 0.2 U/mL for ScCR and PpADH,
and 0.04 U/mL for EsAmDH (Table 4), which afforded the
maximum conversion of 82%. Determination of the kinetic
parameters of EsAmDH (Table S7) showed that its low catalytic
efficiency (kcat/KM = 3.2710−3 s-1 mM-1) was the limiting factor of
0
10
20
30
40
50
Reaction time [h]
Figure 1. Optimization of cofactor addition for amination of 1-phenylethanol by
a redox-neutral reaction. 0.5 mL reaction mixture containing NH4OH/NH4Cl
buffer (2 M, pH 9.5), 2 mM rac-1-phenylethanol, various NAD+, 25 °C.
Table 5. Optimization of isopropanol concentration to
Table 4. Optimization of enzyme dosage for amination of 1-
increase generation of the final amine product.
phenylethanol.
Isopropanol
[mM]
Alcohol
[%]
4
4
22
27
Ketone
[%]
23
14
6
Amine
[%]
73
82
72
67
ScCR &
EsAmDH
Alcohol
[%]
Ketone
[%]
Amine
[%]
Entry
PpADH
[U mL -1
]
[U mL -1
0.05
0.1
]
0
1
2
4
1
2
3
4
5
6
7
0.04
0.04
0.04
0.04
0.01
0.02
0.10
11
10
11
11
11
10
11
14
14
7
75
76
82
82
75
80
82
0.2
0.4
0.2
0.2
6
7
Hydroxylation reaction: 10 mL reaction mixture containing potassium
phosphate buffer (100 mM, pH 8.0), 0.11 μM active P450BM3 19A12, 2
U/mL BmGDH and 10 mM ethylbenzene was performed in a 50 mL
shaking flask at 25 °C for 4 h. 0.5 mL hydroxylation reaction mixture
was mixed with 0.5 mL of NH4OH/NH4Cl buffer (4 M, pH 9.5)
containing optimized amount of enzymes and cofactor for amination
reaction. Isopropanol was added into the reaction system at the
beginning of the amination reaction.
14
10
7
0.2
Reaction condition: 0.5 mL mixture containing NH4OH/NH4Cl buffer
(2 M, pH 9.5), 2 mM rac-1-phenylethanol, 0.5 mM NAD+, 25 °C, 48 h.
Amination of 1-phenylethanol by a redox-neutral cascade
This article is protected by copyright. All rights reserved.

10.1002/cctc.201902253
ChemCatChem
FULL PAPER
concentration led to a decrease in amine composition, which
might be attributed to the excess amount of reductive cofactor
being adopted by alcohol dehydrogenases to afford 1-
phenylethanol, which would slow the conversion of ketone to
amine.
(A)
3.0
(B)
3.0
2.5
2.0
1.5
1.0
0.5
0.0
2.5
2.0
1.5
1.0
0.5
Table 6. Substrate scope expansion for C–H amination of
0.0
aromatic alkanes.
0
10 20 30 40 50
Reaction time [h]
0
10 20 30 40 50
Reaction time [h]
Alkane
Chemical
formula
Amine[a]
[%]
Conv.[b]
[%]
ee
[%]
Figure 3. Time course of amination step (A) without the addition of isopropanol
or (B) with the addition of 1 mM isopropanol at 12 h; 1-Phenylethanol (●);
Acetophenone (●); 1-Phenylethylamine (●). Reaction time was measured from
the start of the amination step. 0.5 mL hydroxylation reaction mixture was mixed
with 0.5 mL of NH4OH/NH4Cl buffer (4 M, pH 9.5) containing optimized amount
of enzymes and cofactor for amination reaction. Isopropanol was added into the
reaction system to meet the final concentration of 1 mM at different reaction
time.
1a
85
53
> 99 (R)
1b
47
27
> 99 (R)
1c
1d
1e
1f
44
41
26
13
> 99 (R)
> 99 (R)
n.d.[c]
the whole cascade system with respect to higher substrate flux
and faster product formation. We attempted to increase either the
enzyme activity or lower the KM value through direct evolution of
EsAmDH by random mutagenesis, but failed to achieve any
improved results.
n.d.[c]
n.d.[c]
n.d.[c]
n.d.[c]
C–H amination of ethylbenzene by multienzymatic cascade
The C–H amination of ethylbenzene was performed using a one-
pot–two-step model, as reported previously.[10] After completion
of the first hydroxylation step, BmGDH was inactivated by heat
treatment to avoid affecting cofactor regeneration in the
subsequent amination reactions, then an equal volume of
NH4OH/NH4Cl buffer (4 M, pH 9.5) containing enzymes and
cofactor for amination reaction was added to the mixture. The time
course of the cascade C–H amination of ethylbenzene showed
that 1-phenylethylamine reached its analytical yield of 45% in the
amination step after 48 h, accounting for 69% of the total products.
Notably, a certain ratio of acetophenone remained in the reaction
mixture, which corresponded to 23% of the total products. Ketone
accumulation was attributed partly to overoxidation in the first
hydroxylation step and partly to untimely transformation of the
ketone by amine dehydrogenase.
n.d.[c]
Reaction conditions were the same with the “20 mL-scale one-pot–two-
step” cascade procedure described in experimental section. [a] Amine
proportion was calculated based on peak area, except for amination
reaction of ethylbenzene based on substrate concentration counting in
extinction coefficient. [b] Conversion was calculated based on the
concentration of amine product formed. [c] n.d.: not detectable.
Investigation of the addition time showed that the amine ratio did
not differ much (Figure S3), with isopropanol addition at 12 h
affording the best performance, reaching a reaction equilibrium
with amine accounting for 85% of the total products (Figure 3).
With isopropanol addition, the accumulated ketones were
transformed into alcohols, allowing the redox-neutral reaction
equilibrium to be manipulated to achieve an obvious ketone
decrease for a period of the reaction.
Isopropanol has been used for reductive cofactor
regeneration in reductase-catalyzed reactions. A popular strategy
involves using isopropanol as both cosubstrate and cosolvent,
with no extra enzymes required for cofactor regeneration.[17] To
minimize the amount of intermediates accumulated during the
cascade reaction, we manipulated the reaction process by adding
a certain amount of isopropanol as cosubstrate to generate extra
reductive cofactor.
Substrate scope exploration
Under the current optimized reaction conditions, we investigated
the versatility of our C–H amination system. Several substituted
ethylbenzenes (1b–1f) were applied to our established reaction
system. Four substrates were successfully converted into amine
Optimization of the cosubstrate concentration was achieved
with 1 mM isopropanol, which increased the amine composition
by 9% (Table 5). However, further increasing the isopropanol
This article is protected by copyright. All rights reserved.

10.1002/cctc.201902253
ChemCatChem
FULL PAPER
products, with overall conversions of 13%–53%. However,
intermediate accumulation was clearly observed (see SI for GC
spectra). Since the activity of P450BM3 19A12 was on the decrease
toward 1a to 1f (Table S8), which was consistent to the
conversion results. The low conversions of alkanes to alcohols
might be mainly attributed to the low catalytic efficiency of P450
monooxygenase. Besides, poor conversion in the amination of
alcohols was mainly attributed to the low specific activity of the
amine dehydrogenase. Future research on multienzymatic C–H
amination should focus on further improving the conversion of
alkanes with respect to increase the activity of P450. Improving
the efficiency of the redox-neutral reaction, such as through
protein engineering, is another future research direction. Finally,
the cascade product of ethylbenzene was isolated from a 200-mL-
Protein expression and purification
Recombinant E. coli BL21 (DE3) cells harboring the target
plasmids were grown in lysogeny broth medium (10% trypton, 5%
yeast extract, and 10% sodium chloride) containing kanamycin
(50 g/mL) at 37 C and 180 rpm. When the OD600 value reached
0.6–0.8, 0.2 mM isopropyl -D-1-thiogalactopyranoside (IPTG)
was added to induce enzyme expression at 16 C for 20 h. Cells
were harvested by centrifugation (10,000  g, 3 min) and
resuspended in sodium phosphate buffer (20 mM, pH 7.8). A clear
lysate was obtained by ultrasonication in an ice bath and
subsequently centrifuged at 10,000  g for 30 min to remove cell
debris. The supernatant was subjected to standard Ni-NTA affinity
chromatography, eluting with 20–500 mM imidazole solution (1
mL/min). Fractions containing the purified target proteins were
pooled and desalted by ultrafiltration.
1
scale reaction in 25% overall isolation yield and subjected to H
1
and 13C NMR analysis. The H and 13C spectra were consistent
with the structure of (R)-1-phenylethylamine.
Determination of active P450 concentration
The concentration of active P450 enzyme was determined using
ferrous carbon monoxide binding difference spectra.[20] Briefly,
two tubes containing diluted P450 enzyme solution (2 mL) were
prepared, followed by the addition of 1 M Na2S2O4 (20 L) and
gentle mixing. Carbon monoxide gas was bubbled through the
solution in one tube at a rate of one bubble per second, while the
solution in the untreated tube was used as the control. Aliquots of
the two samples (200 L) were then subjected to spectroscopy at
wavelengths ranging from 400 to 500 nm. The active P450
content was calculated using a previously reported extinction
coefficient ( = 91,000 M−1 cm−1).
Conclusions
In summary, we have successfully constructed a multienzymatic
cascade reaction comprising a P450 monooxygenase (P450BM3
19A12),
two
stereoselective
complementary
alcohol
dehydrogenases (ScCR and PpADH), and an amine
dehydrogenase (EsAmDH). After reaction optimization of each
step and the whole cascade, the cascade C–H amination of
ethylbenzene was successfully implemented on a laboratory
scale. A panel of aromatic alkanes was investigated using this
established cascade, affording conversions of 13%–53%. This C–
H amination of ethylbenzene represents the highest biocatalytic
C–H amination conversion of ethylbenzene reported to date. We
also demonstrated the biocatalytic preparative scale synthesis of
(R)-1-phenylethylamine, which gave 53% substrate conversion
and 25% isolated product yield. This study further demonstrates
the feasibility of C–H amination of inert aromatic alkanes using a
multienzymatic cascade approach, although the substrate scope
and enzyme efficiency require further improvement.
Enzyme assay
The enzymatic activity of P450BM3 variants, oxidoreductases, and
amine dehydrogenases was measured by monitoring the initial
change in the absorbance of nicotinamide cofactors at 340 nm (
= 6220 M−1cm−1) and 25 C using a UV–Vis spectrophotometer
(Shimadzu, Japan). For P450 monooxygenases, reactions were
performed in a cuvette supplemented with potassium phosphate
buffer (1 mL, 100 mM, pH 8.0) containing 1 mM ethylbenzene (1%
DMSO (v/v)), 0.1 mM NADPH, and an appropriate amount of
enzyme. For other enzymes, reactions were performed in a
cuvette supplemented with NH4Cl/NH4OH buffer (1 mL, 2 M, pH
9.5) containing 10 mM ketone substrate (or 2 mM alcohol
substrate), 0.1 mM NAD(H), and an appropriate amount of
enzyme. One unit of the enzymatic activity was defined as the
amount of enzyme catalyzing the reduction (or oxidation) of 1
mol of substrate in 1 min.
Experimental Section
Strains and plasmids
All plasmids used in this study were either obtained from our
laboratory or gene synthesis. All sequences were ligated on pET-
28(a)+ vector and subsequently transformed into E. coli BL21
(DE3) for expression. Recombinant plasmids pET-28(a)+-
GC analysis
P450BM3 WT, pET-28(a)+-P450BM3 F87V, pET-28(a)+-P450BM3 19A12
,
pET-28(a)+-ScCR, pET-28(a)+-PpADH, pET-28(a)+-CoADH,
pET-28(a)+-GcAR, pET-28(a)+-Pden_4730, pET-28(a)+-CoCR,
pET-28(a)+-RrADH, pET-28(a)+-LkADH, and pET-28(a)+-
EsAmDH were preserved in our laboratory. AaADH,[18]
Ch1AmDH,[19] and BsAmDH[8] were obtained through gene
synthesis with prior codon optimization, and ligated in pET-28(a)+
between EcoR I and Hind III.
GC analysis was performed on
a Shimadzu-2014 gas
chromatograph equipped with a flame ionization detector (FID)
and an Agilent J&W DB-1701 capillary column (30 m  0.25 mm,
0.25 m), using nitrogen as the carrier gas. To determine the
enantiomeric excess, an Agilent J&W CP-Chiralsil-DEX CB
capillary column (25  0.25 mm, 0.25 m) was equipped instead.
This article is protected by copyright. All rights reserved.

10.1002/cctc.201902253
ChemCatChem
FULL PAPER
The aqueous mixture was then extracted with dichloromethane (3
 200 mL). The combined organic layer was dried over anhydrous
Na2SO4, filtered, and concentrated under reduced pressure to
afford 4a as a yellow oil. (R)-1-phenylethylamine (4a): 30.2 mg
(25% yield), 1H NMR (400 MHz, CDCl3)  7.29–7.22 (m, 4H),
7.18–7.13 (m, 1H), 4.03 (q, J = 6.6 Hz, 1H), 1.31 (d, J = 6.6 Hz,
3H). 13C NMR (100 MHz, CDCl3)  148.5, 129.2, 127.6, 126.4,
52.1, 26.4.
The enantiomeric excess values of the amine products were
determined by GC analysis using a chiral stationary phase after
derivatization by comparison with commercially available
standard materials. For derivatization, samples in methyl tert-
butyl ether (MTBE; 500 L) were mixed with acetic anhydride (20
L) and pyridine (10 L), then incubated for 30 min (40 C).
Saturated CuSO4 aqueous solution (500 L) was then added to
remove excess pyridine, and the organic phase was washed with
saturated NaHCO3 to remove remaining acetic anhydride. The
organic phase was dried over anhydrous Na2SO4 and analyzed
by GC. The detailed GC analytical conditions are provided in the
supporting information.
Acknowledgements
The work was financially supported by the National Natural
Science Foundation of China (Nos. 21672063, 21871085 &
21922804) and the Fundamental Research Funds for the Central
Universities (No. 22221818014). The authors are sincerely
grateful to Mr. Tuo Li (ECUST) for his helpful discussions.
Determination of substrate evaporation
In the hydroxylation step, the reaction mixture (without cofactor)
was incubated and shaken under the selected reaction conditions
for 4 h. Samples (500 L) were taken, an equal amount of MTBE
(internal standard cyclopentanone) was added for extraction, and
the organic phase was dried over anhydrous Na2SO4 and
subjected to GC analysis. The residual substrate was determined
based on the substrate concentration before and after incubation.
Keywords: bioamination • cascade reaction • C-H activation •
asymmetric synthesis • optimization
[1]
H. A. Wittcoff, B. G. Reuben, J. S. Plotkin, Industrial Organic Chemicals,
Second Edition, 2005, pp. 223-254.
[2]
[3]
T. C. Nugent, M. El-Shazly, Adv. Synth. Catal. 2010, 352, 753-819.
a) E. Fabiano, B. T. Golding, M. M. Sadeghi, Synthesis 1987, 190-
191; b) K. Weissermel, H. J. Arpe, Industrial Organic Chemistry,
Fourth Edition, 2008, pp.15-57.
20 mL-scale One-pot–two-step cascade procedure
[4]
a) P. Anastas, J.C. Warner, Green Chemistry: Theory and Practice,
Oxford University Press, Oxford, 1999; b) R. A. Sheldon, I. Arends, J.
Am. Chem. Soc. 2008, 130, 8, 2704-2704.
The 10 mL hydroxylation reaction mixture contained potassium
phosphate buffer (100 mM, pH 8.0), ethylbenzene (10 mM), 2%
DMSO (v/v), NADP+ (0.2 mM), glucose (15 mM), BmGDH (2
U/mL), and purified P450BM3 19A12 (0.11 M). The mixture was
shaken at 25 C and 120 rpm for 4 h. After hydroxylation was
complete, the reaction mixture was incubated in boiling water for
3 min to terminate the biocatalytic reaction from the first step. For
the subsequent amination step, calculated amount of enzymes
and cofactor were dissolved in NH4OH/NH4Cl (4 M, pH 9.5) buffer.
Then 10 mL of NH4OH/NH4Cl (4 M, pH 9.5) buffer containing
ScCR, PpADH, EsAmDH and NAD+ was added to initiate the
amination reaction. In this amination reaction stage, enzyme
dosages for ScCR, PpADH and EsAmDH were 0.2 U/mL, 0.2
U/mL, and 0.04 U/mL, respectively, and NAD+ addition was 0.5
mM in final concentration. 20 μmol Isopropanol was added after
the amination step proceeded for 12 h with a final concentration
of 1 mM.. At the scheduled time, a 200-L sample of the reaction
mixture was taken, quenched with NaOH aqueous solution (10 M,
80 L) and extracted with MTBE (200 L). The organic phase was
dried with anhydrous Na2SO4 and analyzed by GC.
[5]
[6]
P. Both, H. Busch, P. P. Kelly, Angew. Chem. Int. Ed. 2016, 55, 1511-
1513.
a) A. Gomm, E. O'Reilly, Curr. Opin. Chem. Biol. 2018, 43,106-112; b)
A. P. Green, N. J. Turner, E. O'Reilly, Angew. Chem. Int. Ed. 2014, 53,
10714-10717; c) S. E. Payer, J. H. Schrittwieser, W. Kroutil, Eur. J.
Org. Chem. 2017, 54, 2553-2559.
[7]
[8]
[9]
M. Tavanti, J. Mangas-Sanchez, S. L. Montgomery, M. P. Thompson,
N. J. Turner, Org. Biomol. Chem. 2017, 46, 9790-9793.
M. J. Abrahamson, E. Váquez-Figueroa, N. B. Woodall, J. C. Moore,
A. S. Bommarius, Angew. Chem. Int. Ed. 2012, 51, 3969-3972.
a) F.-F. Chen, Y.-Y. Liu, G.-W. Zheng, ChemCatChem 2015, 7, 3838-
3841; b) F. G. Mutti, T. Knaus, N. S. Scrutton, Science 2016, 47, 1525-
1529.
[10]
[11]
H.-L. Yu, T. Li, F.-F. Chen, X.-J. Luo, Metab. Eng. 2018, 47, 184-189.
a) B. Wang, H. Land, P. Berglund, Chem. Commun. 2013, 49, 161-
163; b) M. D. Truppo, J. D. Rozzell, N. J. Turner, Org. Process Res.
Dev. 2010, 14, 234-237.
[12]
G. A. Aleku, S. P. France, H. Man, J. Mangas-Sanchez, S. L.
Montgomery, M. Sharma, F. Leipold, S. Hussain, G. Grogan, N. J.
Turner, Nat. Chem. 2017, 9, 961-969.
[13]
[14]
L. O. Narhi, A. J. Fulco, J. Biol. Chem. 1986, 261, 7160-7169.
W. T. Sulistyaningdyah, J. Ogawa, Q. S. Li, Appl. Microbiol.
Biotechnol. 2005, 67, 556-562.
[15]
R. Fasan, M. M. Chen, N. C. Crook, F. H. Arnold, Angew. Chem. Int.
Ed. 2007, 46, 8414-8418.
Preparative-scale C–H amination reaction
[16]
[17]
L.-J. Wang, C.-X Li, Y. Ni, Bioresour. Technol. 2011, 102, 7023-7028.
a) F. Hollmann, I. W. C. E. Arends, D. Holtmann, Green Chem. 2011,
13, 2285-2313; b) Y. Ni, J.-H. Xu, Biotechnol. Adv. 2012, 30, 1279-
1288; c) F. Xu, B. Kosjek, F. L. Cabirol, H. -B. Chen, R. Desmond, J.
Park, A. P. Gohel, S. J. Collier, D. J. Smith, Z. -Q. Liu, J. M. Janey, J.
Y. L. Chung, O. Alvizo, Angew. Chem. Int. Ed. 2018, 57, 6863-6867;
The preparative-scale cascade reaction (200 mL) was performed
by scaling up the 20-mL-scale reaction. When the reaction was
complete, concentrated hydrochloric acid was added to acidify the
mixture to pH 2–3, and the mixture was extracted with
dichloromethane (3  200 mL). The aqueous phase was then
basified to pH 13–14 by adding 10 M NaOH aqueous solution.
This article is protected by copyright. All rights reserved.

10.1002/cctc.201902253
ChemCatChem
FULL PAPER
d) S. Staudt, E. Burda, C. Giese, Angew. Chem. Int. Ed. 2013, 52,
2359-2363.
[18]
[19]
H. W. Höffken, M. Duong, T. Friedrich, Biochemistry 2006, 45, 82-93.
B. Bommarius, M. Schurmann, A. Bommarius, Chem. Commun.
2014, 50, 14953-14955.
[20]
T. Omura, R. Sato, J. Biol. Chem. 1964, 239, 2370–2378.
This article is protected by copyright. All rights reserved.

10.1002/cctc.201902253
ChemCatChem
FULL PAPER
Entry for the Table of Contents (Please choose one layout)
FULL PAPER
Present here is a comprehensively
optimized multienzymatic cascade
reaction for enantioselective C–H
amination of aromatic alkanes, which
employed a P450 monooxygenase,
two stereoselectively complementary
alcohol dehydrogenases and an amine
Hui Wang, Yu-Cong Zheng, Fei-Fei
Chen, Jian-He Xu and Hui-Lei Yu
Page No. – Page No.
Enantioselective Bioamination of
Aromatic Alkanes Using Ammonia: A
Multienzymatic Cascade Approach
dehydrogenase. This cascade system
EsAmDH
converted a panel of hydrocarbon
substrates into the corresponding
P450BM3
PpADH
ScCR
chiral amines with excellent optical
purity (>99% ee) and moderate overall
conversions (13–53%).
This article is protected by copyright. All rights reserved.