Characterization of carboxylic acid reductases for biocatalytic synthesis of industrial chemicals

Article abstract of DOI:10.1002/cbic.201800157

Carboxylic acid reductases (CARs) catalyze the reduction of a broad range of carboxylic acids into aldehydes, which can serve as common biosynthetic precursors to many industrial chemicals. This work presents the systematic biochemical characterization of five carboxylic acid reductases from different microorganisms, including two known and three new ones, by using a panel of short-chain dicarboxylic acids and hydroxy acids, which are common cellular metabolites. All enzymes displayed broad substrate specificities. Higher catalytic efficiencies were observed when the carbon chain length, either of the dicarboxylates or of the terminal hydroxy acids, was increased from C₂ to C₆. In addition, when substrates of the same carbon chain length are compared, carboxylic acid reductases favor hydroxy acids over dicarboxylates as their substrates. Whole-cell bioconversions of eleven carboxylic acid substrates into the corresponding alcohols were investigated by coupling the CAR activity with that of an aldehyde reductase in Escherichia coli hosts. Alcohol products were obtained in yields ranging from 0.5 % to 71 %. The de novo stereospecific biosynthesis of propane-1,2-diol enantiomer was successfully demonstrated with use of CARs as the key pathway enzymes. E. coli strains accumulated 7.0 mm (R)-1,2-PDO (1.0 % yield) or 9.6 mm (S)-1,2-PDO (1.4 % yield) from glucose. This study consolidates carboxylic acid reductases as promising enzymes for sustainable synthesis of industrial chemicals.

Full text of DOI:10.1002/cbic.201800157

Accepted Article
Title: Characterization of Carboxylic Acid Reductases for Biocatalytic
Synthesis of Industrial Chemicals
Authors: Levi Kramer, Erome Hankore, Yilan Liu, Kun Liu, Esteban
Jimenez, Jiantao Guo, and Wei Niu
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: ChemBioChem 10.1002/cbic.201800157
Link to VoR: http://dx.doi.org/10.1002/cbic.201800157
A Journal of
www.chembiochem.org

10.1002/cbic.201800157
ChemBioChem
FULL PAPER
Characterization of Carboxylic Acid Reductases for Biocatalytic
Synthesis of Industrial Chemicals
Levi Kramer,[a]† Erome Hankore, [b]† Yilan Liu,[a] Kun Liu,[b] Esteban Jimenez,[c] Jiantao Guo, [b]* and Wei
Niu[a]*
Abstract: Carboxylic acid reductases (CARs) catalyze the reduction
of a broad range of carboxylic acids into aldehydes, which can serve
as common biosynthetic precursors to many industrial chemicals.
This work presents a systematic biochemical characterization of five
carboxylic acid reductases from different microorganisms, including
two known and three new ones, using a panel of short-chain
dicarboxylic acids and hydroxyacids, which are common cellular
metabolites. All enzymes displayed broad substrate specificities.
Higher catalytic efficiencies were observed when the carbon chain
length of either the dicarboxylates or the terminal hydroxyacids was
increased from C2 to C6. In addition, when substrates of the same
carbon chain length are compared, carboxylic acid reductases favor
hydroxyacids over dicarboxylates as their substrates. Whole-cell
bioconversions of twelve carboxylic acid substrates into
corresponding alcohols were investigated by coupling the CAR
activity with an aldehyde reductase in E. coli hosts. Alcohol products
were obtained at yields ranging from 0.5% to 71%. The de novo
stereospecific biosynthesis of 1,2-propanediol isomers was
successfully demonstrated with CAR being used as the key pathway
enzyme. E. coli strains accumulated 7.0 mM R-1,2-PDO (1.0%
yield) or 9.6 mM S-1,2-PDO (1.4% yield) from glucose. This study
consolidates carboxylic acid reductases as promising enzymes for
sustainable synthesis of industrial chemicals.
alternate platform of chemical production.[1-2] Besides key
advantages over chemical syntheses including reaction
selectivity and smaller environmental impact, microbial
syntheses use renewable feedstocks, such as plant-based
cellulosic material, instead of hydrocarbon as the starting
material. Long-term implementations of microbial syntheses
therefore will inherently address many sustainability challenges
that are facing the chemical industry.
Because commercial chemicals are often ‘man-made’
molecules, finding suitable enzymes are critical to achieving
highly efficient and economical biosynthesis of these non-natural
molecules.[3-5]
The present work focuses on the use of
carboxylic acid reductases (CARs) to catalyze the formation of
common biosynthetic precursors for industrial chemical
production. More specifically, facile reductions of carboxylic
acids by CARs yield aldehydes, a class of common and versatile
biosynthetic intermediates that can be readily transformed into
target molecules by subsequent enzymatic reactions, such as
decarbonylation, reduction, or transamination.[6-7] Aldehydes as
biosynthetic intermediates bridge the functional group mismatch
between short-chain aliphatic acid/diacid metabolites in naturally
high-flux pathways and desirable biosynthetic targets. Current
industrial production of ethanol by fermentation is an excellent
example where an aldehyde (acetaldehyde) serves as a key
intermediate.
O
ATP
R
Introduction
O
OH
SH
R
PPi
A
OH
R
Chemical industry faces sustainability challenges that arise from
both the dwindling starting materials derived from fossil fuels
and increasingly unaffordable environmental costs to the human
society. As a promising solution to these problems, biocatalytic
syntheses, either in the form of enzymatic conversion or
microbial whole-cell synthesis, are increasingly explored as the
O
R
O
AMP
SH
A
SH
R
A
R
O
O
R
R
H
A
S
NADP+
NADPH
R
AMP
Scheme 1. Catalytic mechanism of CAR enzymes. Abbreviations: A,
adenylation domain; R, reduction domain; -SH, phosphopantetheinyl group.
[a]
[b]
L. Kramer, Y. Liu, Prof. W. Niu
Department of Chemical & Biomolecular Engineering
University of Nebraska-Lincoln
Lincoln, NE 68588, USA
E-mail: wniu2@unl.edu
E. Hankore, K. Liu, Prof. J. Guo
Department of Chemistry
University of Nebraska-Lincoln
Lincoln, NE 68588, USA
E-mail: jguo4@unl.edu
Chemical reduction of a carboxylate to an aldehyde group is
an energetically demanding reaction, which is also hard to
control because of the low energy barrier for the further
reduction of the aldehyde product into an alcohol. Three
biological routes are currently known to accomplish this
challenging transformation: (a) a single enzyme-catalyzed direct
reduction;[8-9] (2) a two-enzyme cascade reaction through a
coenzyme A thioester intermediate;[10-12] (3) a two-enzyme
cascade reaction through a phosphoester intermediate.[13] In
comparison to the two-enzyme systems, single enzyme-
[c]
[†]
E. Jimenez
Department of Chemical Engineering, University of Arizona, Tucson,
Arizona, 85721, USA
These authors contribute equally to this work.
Supporting information for this article is given via a link at the end of
the document.
1
This article is protected by copyright. All rights reserved.

10.1002/cbic.201800157
ChemBioChem
FULL PAPER
catalyzed reactions present a simpler and more practical option,
even though the two-enzyme route (2) has been well explored in
microbial production of energy and industrial molecules such as
n-butanol and1,4-butanediol.[14-15] The CAR enzymes catalyze
an irreversible reduction of a carboxylate through in situ
activation and covalent linkage of the substrate to its
posttranslationally added phosphopantetheinyl arm at the
expense of an ATP, followed by the reduction of the resulting
thioester using NADPH (Scheme 1).[9,16] The overall reaction
sequence is orchestrated by three distinct domains of CAR, the
adenylation (A) domain, the phosphopantetheinylation (PPT)
domain, and the reduction (R) domain (Scheme 1). A recent
structural biology study of CARs demonstrated the importance of
domain dynamics during the catalytic cycle.[17] Although the
native substrate is generally identified as benzoic acid, studies
lb/year.[25]
1,2-PDO is mainly used in the production of
unsaturated polyester resin and as deicer in anti-freeze products.
Due to its unique solubility for non-water-soluble molecules and
its GRAS statue, 1,2-PDO is also widely used in formulations of
food, cosmetics, and medicines.
Expression of the CAR
enzyme in engineered E. coli host strains that are capable of
homo-lactate fermentation led to the conversion of lactic acids
into corresponding aldehydes, which were further reduced to
1,2-PDOs. The implementation of the CAR enzyme replaced a
two-enzyme reaction sequence (path b in Scheme 2) with a
single-step direct reaction (path a in Scheme 2) in a previously
reported artificial pathway.[26]
Under fermenter-controlled
microaerobic conditions, R-1,2-PDO was accumulated at 7.0
mM and S-1,2-PDO was accumulated at 9.6 mM. As a proof-of-
concept investigation, this report demonstrates a de novo
biosynthesis of an industrially important chemical using a CAR
showed that CAR enzymes could tolerate
a range of
substituents of various properties on the phenyl ring.[18] More
interestingly, CARs were reported to be active on aliphatic
mono-, di- or tricarboxylic acid metabolites, such as citric acid
and malic acid.[18-20] Recent studies also highlighted the use of a
CAR from Mycobacterium marinum for the biosynthesis of
alkanes as biofuel, due to its appreciable activity on C4-C18 fatty
acids.[21-22] Additional examples of incorporating CAR activities
in (bio)syntheses were recently reviewed.[23-24]
enzyme.
Further improving CAR activity through protein
engineering will likely enable a biocatalytic process that is
complementary to the current commercial route of 1,2-PDO
production (high pressure, high temperature, noncatalytic
hydrolysis of propylene oxide)[27] in feedstock selection.
Results and Discussion
Identification of New CAR Enzymes.
OH
NADPH NADP+
ATP AMP, PPi
a
OH
OH
O
NADPH NADP+
YahK
O
O
Biochemical and structural characterization of CAR enzymes
from Nocardia iowensis (NiCAR), Mycobacterium marinum
(MmCAR), and Segniliparus rugosus (SrCAR) have been
CAR
OH
H
OH
OH
OH
D- or L-lactaldehyde
OH OH
OH
R- or S-1,2-PDO
D- or L-lactic acid
D-glucose
O
b
reported.[17,20-21]
Based on their protein sequences,
a
SCoA
Pct
PduP
phylogenetic tree was constructed from potential CAR enzymes
that were identified in the GenBank database using BLAST
search (Figure S1 of Supporting Information). Three CAR
candidates from Kutzneria albida (KaCAR), Mycobacterium
aromaticivorans (MarCAR), and Mycobacterium avium K-10
(MavCAR), which have no previous literature precedent, were
chosen for further biochemical analysis. Phylogenetic analysis
shows that the three Mycobacterium CARs form one clad, while
the KaCAR and NiCAR form another clad (Figure 1). The two
clads are evolutionarily more closely related than to the SrCAR.
OH
D- or L-lactoyl-CoA
Scheme 2. De novo biosynthesis of 1,2-PDO. a. Direct reduction of lactic acid
by CAR; b. Two-enzyme reduction of lactic acid by Pct and PduP. CAR,
carboxylic acid reductase; Pct, lactoyl-CoA transferase; PduP, CoA-acylating
propionaldehyde dehydrogenase; YahK, lactaldehyde reductase.
Here, we report the identification and characterization of
three new bacterial CAR enzymes with an emphasis on the
potential application of CAR-catalyzed reactions in the
biosynthesis of industrial chemicals. A panel of short-chain (C2-
C6) dicarboxylic acids and hydroxyacids were used as
substrates in full kinetic studies of the three new and two known
CARs. We found that the carbon chain length of the substrates
is the main determinant on the catalytic efficiency of CAR
enzymes.
A
distinct activity difference between the
dicarboxylate and the hydroxyacid of the same carbon chain
length was also observed. The utility of CAR enzymes was
demonstrated by in vivo bioconversions of the dicarboxylic acid
and hydroxyacid substrates into corresponding alcohols using
recombinant E. coli strains that overexpressed a CAR and a
broad substrate specificity aldehyde reductase. We further
integrated the CAR-catalyzed reaction into a de novo microbial
biosynthesis of 1,2-propanediol (1,2-PDO), which is a bulk
industrial chemical with a global demand of around 3 billion
Figure 1. The phylogeny is constructed using MEGA7.0.[28] * The structure of
the adenylation and/or the reduction domain of the enzyme is available.
To evaluate their catalytic activities, each of the three
potential CAR candidates (KaCAR, MarCAR, and MavCAR) was
2
This article is protected by copyright. All rights reserved.

10.1002/cbic.201800157
ChemBioChem
FULL PAPER
Figure 3. Effect of temperature on CAR variants. A. The specific activities of
CARs at various reaction temperatures; B. The thermostability of CAR
enzymes. The relative residual activity is calculated using specific activity of
enzymes stored at 4 °C as the benchmark. Data are shown as the averages
of three measurements. Error bars represent standard deviations.
expressed as fusion proteins with a C-terminal 6xHis affinity tag
in E. coli. The phosphopantetheinyl transferase (Sfp) from either
Bacillus subtilis or Nocardia iowensis was co-expressed to
ensure
efficient
activation
of
CAR
enzymes
by
phosphopantetheinylation. Activities of purified enzymes were
assayed on benzoic acid, a commonly accepted CAR substrate
(Figure 2). Reduction of the carboxylates were followed by
monitoring the oxidation of NADPH at 340 nm.[9] Blank controls
either without the carboxylic acid substrate or without the
purified enzyme were included for each set of assay. Two
known CARs (NiCAR and MmCAR) were also included in the
experiment to help gauge the activities of the three candidate
enzymes. As shown in Figure 2, all three candidate CARs could
indeed catalyze the reduction of benzoic acid. They were
confirmed as true carboxylic acid reductases. Among all the
enzymes tested, the MavCAR displayed the highest catalytic
activity, which was nearly six times more active than the
previously characterized MmCAR and slightly more active than
the best reported CAR from Mycobacterium abscessus.[29]
Next, we examined how the temperature and the pH affected
the activities of all five CAR enzymes with benzoic acid as the
substrate. The temperature screen was performed at 25, 30, 37,
40, 42, and 45 °C in phosphate buffer (pH 7.5; Figure 3A).
A
general trend of temperature-dependent increase of activities
was observed for all CARs. A more pronounced activity
increase was detected for MavCAR and KaCAR between 40 and
45 °C. The thermostability of CARs was tested using enzymes
that were incubated at indicated temperatures for 30 min (Figure
3B) followed by enzyme assays at room temperature under
standard conditions. After treatment at 45 °C, all CAR enzymes
except the MavCAR were completely inactivated, while KaCAR
is the least thermally stable enzyme. The MavCAR retained
over 90% of the activity following heating at 45 °C. It rapidly lost
activity between 47.5 °C and 50 °C, and became inactive at
52.5 °C. The higher specific activity at elevated reaction
temperature and the good thermostability make the MavCAR an
attractive choice as a biocatalyst. The pH profiles of the CAR
enzymes were tested between pH 5.5 and 9.0 at 25 °C. All
enzymes were most active when the pH was near neutral.
While MarCAR showed an optimum activity at pH 6.5, KaCAR
and MmCAR displayed an optimum activity at pH 7.5 (Figure 4).
Both MavCAR and NiCAR showed a broad optimum pH range
between 6.5 and 7.5. Significant reductions in catalytic activity
were observed at lower or higher pH values (Figure 4).
2.00
1.50
1.00
0.50
0.00
KaCAR MarCAR MavCAR MmCAR NiCAR
Figure 2. Specific activities of CAR candidates on benzoic acid at 25 °C and
pH 7.5. Data are shown as the averages of three measurements. Error bars
represent the standard deviations.
2.00
KaCAR
MavCAR
NiCAR
MarCAR
MmCAR
1.50
1.00
0.50
0.00
A
4.00
3.50
KaCAR
MavCAR
NiCAR
MarCAR
MmCAR
3.00
2.50
2.00
1.50
1.00
0.50
0.00
5.0
5.5
6.0
6.5
7.0
7.5
pH
8.0
8.5
9.0
9.5
Figure 4. Effect of reaction pH on the activity of CAR variants. Data are
shown as the averages of three measurements. Error bars represent standard
deviations.
20
25
30
35
40
45
50
Temperature (oC)
B
1.20
1.10
1.00
0.90
0.80
0.70
0.60
0.50
0.40
0.30
0.20
0.10
0.00
Kinetic Studies of CAR Enzymes on C2-C6 Dicarboxylic
Acids and Hydroxyacids.
Short chain carboxylic acids, such as lactic acid and succinic
acid, are common metabolites that can be accumulated as
bioproducts. Further enzymatic reduction of the carboxylates
leads to aldehyde products that can serve as precursors to
various industrial chemicals, such as 1,2-propanediol and 1,4-
butanediol. We therefore examined the kinetic properties of
KaCAR
MavCAR
NiCAR
MarCAR
MmCAR
20
25
30
35
40
45
50
55
Temperature (oC)
3
This article is protected by copyright. All rights reserved.

10.1002/cbic.201800157
ChemBioChem
FULL PAPER
Table 2. Kinetic studies of CARs on hydroxyacid substrates (C2-C6). a. CARs,
NiCAR, MmCAR, KaCAR, MarCAR, and MavCAR by using a
panel of C2-C6 dicarboxylic acid (Table 1) and hydroxyacid
(Table 2) substrates. The straight-chain carbon skeleton of
above molecules deviates significantly from the aromatic
scaffold of benzoic acid, the native CAR substrate. In addition, a
second hydrophilic functional group, carboxylate in dicarboxylic
acids or hydroxyl group in hydroxyacids, exists in these
substrates. Nevertheless, all the five CAR enzymes were able
to tolerate the structural changes and showed broad substrate
promiscuity to different extents, while benzoic acid remained as
the preferred substrate.
b.
1, KaCAR; 2, NiCAR; 3, MmCAR; 4, MarCAR; 5, MavCAR.
n.d., Activity
was not detected. Data are the average of three measurements with standard
deviation.
CARa
1
2
3
4
5
1
2
3
4
5
1
2
3
4
5
1
2
3
4
5
1
2
3
4
5
1
2
3
4
5
1
2
3
4
5
Km (mM)b
kcat (min-1)
13.5±0.6
4.5±0.5
kcat /Km (mM-1min-1)
0.10±0.01
133.4±13.4
163.7±48.7
76.8±11.9
0.03±0.01
glycolate
6.0±0.3
0.08±0.01
O
OH
-O
n.d.
227.9±31.4
267.3±29.0
143.6±22.4
245.3±38.5
111.5±19.1
67.3±11.8
61.0±6.1
21.1±1.4
13.8±0.6
19.7±1.1
15.5±1.3
34.4±2.4
46.3±3.3
66.7±2.6
34.8±1.6
23.9±0.9
61.7±4.1
42.6±1.7
132.5±10.0
43.3±1.8
33.8±1.3
65.8±2.8
59.3±2.7
109.2±4.1
46.2±2.5
27.3±1.7
79.9±6.0
93.5±3.3
17.9±1.4
5.5±0.9
0.09±0.01
0.05±0.01
0.14±0.02
0.06±0.01
0.31±0.06
0.69±0.13
1.1±0.1
Table 1. Kinetic studies of CARs on dicarboxylate substrates (C2-C6). a. CARs,
b.
1, KaCAR; 2, NiCAR; 3, MmCAR; 4, MarCAR; 5, MavCAR.
n.d., Activity
3-hydroxy
propionate
was not detected. Data are the average of three measurements with standard
deviation.
O
-O
OH
CARa
1-5
Km (mM)b
kcat (min-1)
kcat /Km (mM-1min-1)
oxalate
O
4-hydroxy
butyrate
31.0±4.3
1.1±0.2
n.d.
O-
-O
66.9±6.5
0.36±0.04
0.48±0.08
2.7±0.4
O
O
OH
-O
128.2±18.5
16.0±2.1
1-4
5
n.d.
n.d.
malonate
O
O
160.5±49.7
1.5±0.2
0.009±0.003
26.6±4.9
5.0±1.0
-O
O-
5-hydroxy
pantanoate
10.1±1.7
4.3±0.8
1-3
4
5
1
2
3
4
5
1
2
3
4
5
1
2
3
4
5
succinate
O
39.9±4.0
0.85±0.09
1.8±0.2
281.3±84.3
13.2±3.3
5.9±0.8
3.4±0.2
6.3±0.6
0.02±0.01
0.26±0.07
0.14±0.05
0.05±0.03
0.03±0.01
O
O-
-O
-O
OH
36.0±3.8
O
12.8±1.8
4.6±0.7
44.3±15.2
314.2±172.3
273.3±43.0
11.7±1.2
9.3±1.0
16.8±4.4
7.3±0.5
glutarate
6-hydroxy
hexanoate
O
4.6±0.7
10.1±1.6
1.2±0.2
O
O
22.9±3.3
-O
O-
n.d.
OH
-O
10.6±2.0
7.5±1.5
55.1±4.0
46.4±10.7
36.2±7.0
31.6±2.2
44.4±13.0
3.7±0.6
13.5±0.4
15.6±1.3
14.1±0.9
10.9±0.2
39.7±4.2
13.5±0.4
173.6±7.9
104.1±3.1
35.5±1.4
209.6±18.1
242.1±14.0
0.24±0.02
0.34±0.08
0.39±0.08
0.34±0.02
0.89±0.28
3.64±0.6
10.0±1.4
9.3±1.4
100.4±20.2
165.6±64.3
229.5±23.5
340.2±71.8
217.3±26.6
100.3±21.7
142.0±41.3
240.9±51.8
310.4±48.1
162.9±23.3
0.18±0.04
0.03±0.01
0.06±0.01
0.09±0.02
0.30±0.04
0.05±0.01
0.04±0.01
0.02±0.01
0.05±0.01
0.14±0.02
adipate
O
L
-lactate
O-
-O
14.8±0.8
30.0±3.7
65.3±3.9
4.6±0.3
O
O
-O
OH
0.24±0.06
0.78±0.08
0.30±0.06
4.2±0.8
731.7±188.8
133.2±14.6
118.0±25.7
50.3±10.6
162.1±26.8
benzoate
5.8±0.6
O
D
-lactate
O-
5.6±0.6
O
-O
17.0±1.3
23.4±1.5
1.4±0.2
OH
Two general trends of CAR activity were observed: (1) When
the carbon chain length of either the dicarboxylates (Table 1) or
the terminal hydroxyacids (Table 2) was increased from C2 to C6,
all CARs showed lower Km and higher kcat values, which
correlates with an increase in catalytic efficiency. In an extreme
case of C2 dicarboxylate, oxalate, none of the CAR enzymes
showed any detectable activity. The observed preference for
substrates of longer carbon chain length resonates with previous
report where fatty acids were examined as substrates of
MmCAR.[21] (2) When substrates of the same carbon chain
length were compared, all CARs favored hydroxyacids over
4
This article is protected by copyright. All rights reserved.

10.1002/cbic.201800157
ChemBioChem
FULL PAPER
Figure 5. Reduction of short-chain carboxylates using E. coli cells expressing
MavCAR and YahK. The x-axis represents the substrates spiked in the
culture media. (A) Dicarboxylic acid substrates. (B) Hydroxyacid substrates.
Cells were cultured in media (pH 7.0) containing 100 mM of indicated
carboxylate. Data are the average of three measurements with standard
deviation.
dicarboxylates as their substrates.
This trend was more
prominent for KaCAR, NiCAR, and MmCAR, which showed no
detectable activity on any of the C3 and C4 dicarboxylic acids,
but were still active on the corresponding hydroxyacids. Among
all the examined enzymes, MavCAR showed the best activities
on dicarboxylic acids and was the only CAR that could utilize
malonate as the substrate. The MavCAR was also highly active
on hydroxyacids, while KaCAR and NiCAR displayed
appreciable activities on this group of substrates as well. We
speculate that varied activities on dicarboxylate and hydroxyacid
of the same chain length are resulted from the active site
properties of CAR enzymes. With benzoic acid as the native
substrate, a highly hydrophobic active site with limited ability to
engage hydrogen bond or ionic interaction was observed for
As an initial test, we examined the in vivo bioconversion of
lactic acids into 1,2-PDO. A previously constructed E. coli strain
PDO-Δ4 was used.[26] This strain lacks the activities of D- and L-
lactate
dehydrogenases,
the
fermentative
D-lactate
dehydrogenase, and an alcohol dehydrogenase due to
chromosomal deletions of dld, lldD, ldhA, and adhE genes. As a
result, PDO-Δ4 cannot use lactic acid isomers as the carbon
source and is also incapable of producing lactic acid. It was
used in previous report for the bioconversion of lactic acids into
1,2-propanediol through a CoA acylating pathway.[26] Since the
genetic deletions of ldhA and adhE eliminated reactions that
compete for reducing equivalent and reduced accumulation of
byproducts, the PDO-Δ4 strain is generally beneficial to the
reduction reaction of other carboxylic acids. Therefore, the
PDO-Δ4 strain was employed in all of our in vivo bioconversion
studies except for glycolic acid. The consumption of glycolic
acid by E. coli significantly affected the bioconversion efficiency.
To minimize the loss of glycolic acid to cellular metabolism, E.
coli strain EG-Δ4 was constructed and used in the bioconversion
of glycolic acid to ethylene glycol. It was derived from MG1655
through the deletion of glcD, gcl, aceB, and glcB genes, which
encodes the glycolate oxidase, glyoxylate carboligase, malate
synthase A, and malate synthase G, respectively. Above
manipulations abolish cells’ ability to use glycolic acid as the
carbon source. Preliminary experiment for 1,2-PDO synthesis
from isomers of lactic acid also showed that the expression of a
yahK-encoded broad substrate range aldehyde reductase was
beneficial for the reduction of aldehydes into alcohols (Figure S5
of Supporting Information).[30] This enzyme was therefore
overexpressed in all the whole-cell reduction experiments
(Figure 5).
selected CARs.[17]
In comparison to hydroxyl group, the
negatively charged carboxylate group may lead to even greater
unfavorable interaction(s). We also tested two non-terminal
hydroxyacids, stereoisomers of lactic acid, and observed that all
the examined CAR enzymes were active with slight preference
to the L-isomer. The activities were generally lower than when
3-hydroxy propionate was used as the substrate.
Whole-Cell Reduction of Short-Chain Carboxylic Acids
using MavCAR.
Activities of CAR enzymes on short-chain carboxylic acids were
further investigated using E. coli strains that expressed MavCAR,
which was shown to be the most promiscuous CAR variant. E.
coli cells were cultured in media (pH 7.0) spiked with an
indicated short-chain carboxylic acid of interest to the final
concentration of 100 mM. Glucose was included in the media as
the carbon source for cell growth and to supply ATP and
NADPH as co-substrates for CAR activities.
O
n
O
n
MavCAR
YahK
R
R
R
OH
n
OH
H
R = -COOH, -OH
5
4.5
4
A.
4.48
hydroxyacid
diol
Suitable E. coli hosts were transformed with an F plasmid
that encoded the Sfp protein and a pSC101-based plasmid,
which contained a MavCAR-yahK gene cassette. Following 72
h of aerobic cultivation, metabolites in the media were analyzed
by HPLC. For each of the C3-C6 dicarboxylic acid substrate,
both the hydroxyacid, the product from the reduction of a single
carboxylate, and the diol, the fully reduced products, were
observed (Figure 5A). For each of the C2-C6 hydroxyacid
substrate, the corresponding diol product was detected (Figure
5B). By using the formation of the alcohol product as a measure
of the in vivo activity of MavCAR, a similar trend of substrate
preference as that observed in the kinetic characterization was
detected. With the increase of carbon chain length, higher
yields of the corresponding alcohol products were obtained with
either the dicarboxylic acid or the hydroxyacid substrates. In
addition, when substrates of the same carbon chain length were
used, significantly higher yields were obtained with hydroxyacid
substrates. The only deviation from this general trend was the
3.5
3
2.32
2.5
2
1.77
1.38
1.5
1
0.69
0.61
0.55
0.5
0
0.17
malonate
succinate
glutarate
adipate
B.
80
70
60
50
40
30
20
10
0
71.2
60.15
54.25
5.24
2.78
0.51
0.49
L-lactate D-lactate glycolate 3-hydroxy- 4-hydroxy- 5-hydroxy- 6-hydroxy-
propionate butyrate pentanoate hexanoate
5
This article is protected by copyright. All rights reserved.

10.1002/cbic.201800157
ChemBioChem
FULL PAPER
higher bioconversion yield of malonate than those of its C4 and
C5 analogs. It is possible that higher activity of YahK enzyme on
malonate semialdehyde leads to faster overall kinetics of the
bioconversion from malonate to 3-hydroxypropionate. Possible
toxicity caused by the carboxylate substrates was evaluated by
monitoring the cell growth (Figure S4 and Table S4 of
Supporting Information). Although extended lag phase was
observed in the presence of several compounds, minimal effect
on the doubling time was observed (less than 32%) for all
substrates except 3-hydroxypropionate. Products formed in the
above whole-cell experiments are all industrial chemicals with
varied market volumes and values. In particular, ethylene glycol
(production of glycolate), 1,3-propanediol (product of
malonate/3-hydroxypropionate), 1,4-butanediol (product of
succinate/4-hydroxybutyrate), and 1,2-propanediol (product of
lactates) are important monomers in polymer industry. The
whole-cell bioconversion results further emphasized the great
biosynthetic potential of CAR enzymes.
coli hosts lost the native pathway of ethanol biosynthesis
because of the adhE deletion, a possible source of ethanol
production is the reduction of acetate by CAR and YahK.
A.
12
10
8
36
30
24
18
12
6
[R-1,2-PDO] (mM)
OD600
6
4
2
0
0
0
4
12
16
Time (h)
20
22
B.
12
10
8
36
30
24
18
12
6
[S-1,2-PDO] (mM)
OD600
De novo Biosynthesis of 1,2-Propanediol (1,2-PDO) using
MavCAR
6
We further demonstrated the utility of CAR enzymes in the de
novo biosynthesis of 1,2-PDOs through the reduction of lactates.
Two previously constructed host strains were used in the
synthesis of each stereoisomer.[31] E. coli, PDO-Δ7, which has
chromosomal deletions of dld, lldD, adhE, frdA, pflB, mgsA, and
aldA genes,[32-33] was used in the production of R-1,2-PDO
through the reduction of D-lactic acid. E. coli, PDO-Δ7lldh, which
has an additional chromosomal modification by replacing the
native ldhA gene with lldh gene from Pediococcus acidilactici,
was used in the production of S-1,2-PDO through the reduction
of L-lactic acid.[34] Each host was transformed with plasmids
encoding MavCAR, YahK, and Sfp. Resulting strains were
4
2
0
0
0
4
12
16
20
Time (h)
24
40
44
48
Figure 6. De novo biosynthesis of 1,2-PDO isomers by E. coli. (A) R-1,2-PDO
accumulation by PDO-Δ7 strain. (B) S-1,2-PDO accumulation by PDO-Δ7lldh
strain. Open circles, OD600 nm; filled bars, 1,2-PDO. Data are the average of
two measurements with standard deviation.
Conclusions
examined for 1,2-PDO synthesis from glucose under
a
fermentor-controlled two-phase cultivation scheme.[14] Following
the initial aerobic phase of cell mass accumulation, expression
of MavCAR and YahK was induced by IPTG addition. After an
additional hour of growth, the airflow was reduced for
microaerobic cultivation and the glucose feed was initiated. This
point of transition was defined as t = 0 h. The microaerobic
condition was chosen to enable the production of ATP through
oxidative phosphorylation, while still allows the de novo
biosynthesis of lactic acid as the substrate for the CAR enzyme.
Sampling was done at indicated time points to monitor the cell
growth and metabolites accumulation until the glucose
concentration in the media was below 5 mM. The two E. coli
strains produced 7.0 mM R-1,2-PDO from D-lactate at 22 h and
9.6 mM S-1,2-PDO from L-lactate at 48 h, which corresponds to
1.0% and 1.4% yield from glucose, respectively (Figure 6).
Optical purity analysis showed that each isomer was produced
at over 99% ee (Figure S6 of Supporting Information). In
addition, the PDO-Δ7 strain also accumulated 695 mM D-lactate,
while the PDO-Δ7lldh strain produced 407 mM L-lactate (Figure
S7 and S8 of Supporting Information). Other metabolites,
including pyruvate, acetate, and ethanol, were also detected
(Figure S7 and S8 of Supporting Information). Since the two E.
Biocatalytic synthesis is becoming an indispensible choice for
chemical productions. Fulfilment of its full potential requires
relentless efforts into the discovery and the characterization of
enzymes that can catalyze the transformation of common
cellular metabolites into desirable industrial products. In this
study, we characterized five CAR enzymes on a panel of C2-C6
short chain dicarboxylic acids and hydroxyacids, which can be
synthesized by microbial hosts through endogenous or
engineered pathways. The data showed that CAR enzymes
were active on a broad range of substrates and could tolerate
the presence of additional functional groups (i.e., hydroxyl and
carboxyl) in the substrates. The carbon chain length is a major
factor that influences the activity of CARs. Using the most
promiscuous CAR variant from our study, we completed whole-
cell bioconversions of dicarboxylic acid and hydroxyacid
substrates into corresponding alcohols, including industrially
important platform chemicals ethylene glycol, 1,2-propanediol,
1,3-propanediol, 1,4-butanediol, and 1,6-hexanediol. The utility
of CAR enzyme in the de novo biosynthesis from a renewable
starting material was further demonstrated through the
production of R- and S-1,2-PDO stereoisomers from glucose by
6
This article is protected by copyright. All rights reserved.

10.1002/cbic.201800157
ChemBioChem
FULL PAPER
protein expression host E. coli BL21(DE3). Following cultivation
in LB media to the exponential growth phase, protein expression
was induced with IPTG at 0.5 mM. CAR proteins with C-
terminal Hisx6 tags were purified using Ni Sepharose 6 Fast
Flow resin (GE Healthcare) by following manufacture’s protocol.
Purified proteins were dialyzed against buffer containing
potassium phosphate (25 mM, pH 7.5), NaCl (150 mM), glycerol
(5%) and DTT (1 mM) prior to biochemical assays. Purity of
proteins was analyzed by SDS-PAGE (Figure S2 of Supporting
Information).
engineered E. coli strains. The present work lays a solid
foundation to employ CAR enzymes in biocatalytic syntheses.
In light of recent advance in the structural studies of CARs, our
efforts currently focus on protein engineering and pathway
engineering to enable and optimize the de novo biosynthesis of
industrial chemicals (e.g., ethylene glycol and propanediols).
Overall, CARs’ ability to catalyze a single-step reduction of acid
metabolites positions them as promising enzymes for the
sustainable synthesis of chemicals.
Biochemical Assays.
Experimental Section
All carboxylic acid substrates were purchased either as free acid
or salt except 5-hydroxypentanoic acid and 4-hydroxybutyric
acid, which were obtained from the hydrolysis of the
commercially available corresponding lactones. CAR enzyme
assays followed reported procedures at indicated temperature,
pH, and substrate concentration.[9] The enzyme assay solution
(0.15 mL) contained NADPH (0.15 mM), ATP (1 mM), MgCl2 (10
mM), NaCl (150 mM), glycerol (5%), DTT (1 mM), appropriate
amount of purified protein, and a substrate of interest. In 2-(N-
morpholino)ethanesulfonic acid buffer (MES, 50 mM) with pH
range of 5.5-6.5 and Tris-HCl buffer (50 mM) with pH range of
7.0-9.0 were used in the experiments to study the pH effects on
CAR enzymes. All other assays were carried out in phosphate
General.
All chemicals are of reagent grade or higher. All solutions were
prepared in deionized water that was further treated by
Barnstead Nanopure® ultrapure water purification system
(Thermo Fisher Scientific Inc). LB medium[35] (1 L) contained
Bacto tryptone (10 g), Bacto yeast extract (5 g), and NaCl (10 g).
M9 salts (1 L) contained Na2HPO4 (6 g), KH2PO4 (3 g), NH4Cl (1
g), and NaCl (0.5 g). M9 glucose medium contained glucose (10
g), MgSO4 (0.12 g), and thiamine hydrochloride (0.001 g) in 1 L
of M9 salts.
For whole-cell bioconversion experiments,
dicarboxylic acid or hydroxyacid substrates were included in M9
glucose medium at a final concentration of 100 mM. The pH of
the media was adjusted to 7.0 using KOH prior to inoculation.
Fed-batch fermentation media (850 mL) contained Na2HPO4
(6.78 g), KH2PO4 (3 g), NH4Cl (2 g), NaCl (0.5 g), and (NH4)2SO4
(1 g). Antibiotics were added where appropriate to the following
final concentrations: chloramphenicol (in methanol), 17 mg/L;
buffer (25 mM, pH 7.5).
Sodium benzoate at a final
concentration of 5 mM was used in assays where the specific
activities (Figure 2), the temperature profiles (Figure 3A), the
thermostability (Figure 3B), and the pH profiles (Figure 4) of
CAR enzymes were examined. In thermostability studies,
aliquots of an CAR enzyme were incubated at the indicated
temperature for 30 min, then stored on ice for 5 min prior to
experiments. Enzyme assays were initiated upon the addition of
kanamycin, 50 mg/L; ampicillin, 100 mg/L.
Protein
concentrations were determined using the Bio-Rad Bradford
protein assay kit (Bio-Rad Laboratories Inc). A Biotek Synergy
HTX platereader was used to record UV-vis spectra.
a
substrate of interest.
CAR activity was measured
spectrophotometrically by following the oxidation of NADPH at
340 nm. A molar extinction coefficient of 6,220 M-1cm-1 (340 nm)
was used for NADPH.
Blank controls either without the
CAR Cloning and Protein Purification.
carboxylic acid substrate or without the purified enzyme were
included for each set of assay. Triplicate experiments were
performed for each condition. Data of kinetics assays were
analyzed using Prism 7 (GraphPad Software).
Standard protocols were used for the construction, purification,
and analysis of plasmid DNA.[36] PCR amplifications were
carried out using KOD HotStart DNA polymerase by following
manufacture's protocol. Primer synthesis and DNA sequencing
services were provided by Eurofins MWG Operon. Table S1
(see Supporting Information) lists the sequences of all the
primers that were used in this study. Bacteria strains Kutzneria
Whole-Cell Bioconversions.
E. coli EG-Δ4 and PDO-Δ4 were constructed following the PCR
product-mediated gene deletion method using primers in Table
S1 of Supporting Information.[38] The appropriate host was
transformed with plasmid pZF-Sfp(Ni) and pZSC101-MavCAR-
YahK (Table S2 of Supporting Information). For bioconversions,
a single colony was inoculated in 5 mL LB media with
appropriate antibiotics and cultured at 37 ºC overnight. A 100
µL of the overnight culture was transferred into 5 mL of M9
minimal salts media (pH 7.0) containing glucose (10 g/L), IPTG
(0.25 mM), and a carboxylic acid substrate of interest (100 mM)
in 30 mL culture tubes. Triplicate cultures of each strain were
cultivated at 37 ºC with shaking at 250 rpm for 72 h. Products in
cell-free broth were identified and quantified by HPLC based on
previously established retention times (Table S3 of Supporting
Information) and calibration curves using authentic standards.
For products with low concentrations, including ethylene glycol
and 1,4-butanediol, 1H NMR was also used to confirm their
formation.
albida
(ATCC25243),
Mycobacterium
aromaticivorans
(ATCCBAA-1378), and Bacillus subtilis (ATCC6051) were
purchased from American Type Culture Collection (ATCC).
Strain Nocardia iowensis (DSM45197) was purchased from the
German Collection of Microorganisms and Cell Cultures GmbH.
All strains were cultured by following the venders’ instructions.
Genomic DNAs were isolated using the PureLink Genomic DNA
Mini Kit (Thermo Fisher Scientific Inc). The genomic DNAs of
Mycobacterium
avium
K-10
(ATCCBAA-968D)
and
Mycobacterium Marinum (ATCCBAA535D) were purchased
from ATCC. The Sfp-encoding gene of B. subtilis or N. iowensis
was cloned together with the PA1 promoter[37] into the XbaI site
of plasmid pET30b (Novagen). Genes encoding CARs were
amplified without the stop codon and cloned between the NdeI
and XhoI sites of the modified pET30b vector. Table S2 of
Supporting Information lists the all plasmids that were used in
this study. Sequence-verified plasmids were transformed into
7
This article is protected by copyright. All rights reserved.

10.1002/cbic.201800157
ChemBioChem
FULL PAPER
De novo Biosynthesis.
This work was supported by the National Science Foundation
(grant 1438332 to W.N. and J.G.; REU grant 1560163) and the
Nebraska Public Power District throutgh the Nebraska Center
For Energy Sciences Research at University of Nebraska –
Lincoln (Cycle-9 grant to J.G. and W.N). The authors would like
to acknowledge assistance from the following individuals at the
University of Nebraska-Lincoln: Prof. David Berkowitz
(Department of Chemistry, chiral HPLC) and Prof. Martha
Morton (Chemistry Research Instrumentation Facility, NMR).
Fermentations were carried out in a 2.0 L working capacity
UniVessel® controlled by a Biostat® B controller (Sartorius AG).
Temperature and pH were controlled with proportional-integral-
derivative (PID) control loops. The temperature was controlled
at 37 °C. The pH was controlled at 7.0 by the addition of acid (2
N H2SO4) and base (concentrated NH4OH at the first stage or
20% NaHCO3 at the second stage). Dissolved oxygen was
measured using an OxyFerm probe. Fed-batch fermentations
were run in duplicate. Results were reported as the average of
the two runs. A seed culture was started by the introduction of a
single colony into 5 mL of M9 glucose medium. Culture was
grown at 37 °C with shaking for 16 h and subsequently
transferred into 95 mL of M9 glucose medium. The culture was
Conflict of interest
The authors declare no conflict of interest.
grown under the same condition for an additional 10 h.
A
fermentation was initiated by transferring the seed culture (100
mL) into the fermentation vessel, which contained the fed-batch
fermentation medium (850 mL) and 20 g of glucose (50 mL).
Keywords: carboxylic acid reductase • biocatalysis • carboxylic
acid • aldehyde • industrial chemical
The total initial volume of the fermentation culture was 1 L.
A
two-stage cultivation scheme was used. In the first stage, the
D.O. was maintained at 10% air saturation through sequential
ramping of impeller speed and airflow rate to the preset values
of 1100 rpm and 1.0 cubic liter per minute, respectively. At this
point, IPTG was added to a final concentration of 0.25 mM.
Following an additional 1 h of cultivation, the microaerobic stage
was initiated by reducing the airflow rate to 50 cubic centimeter
per minute. At the same time, glucose solution (600 g/L) was
fed into the fermenter at a rate of 4.17 g/h for 24 h. Throughout
the second stage, the D.O. value registered as 0% air saturation.
Samples were taken at indicated time points. Cell densities
were determined by measurement of absorption at 600 nm
(OD600). Glucose concentration was monitored by HPLC. Cell-
free broth was obtained for metabolite analysis. Cell pellets
were lysed for protein expression analysis.
[1]
[2]
[3]
R. A. Sheldon, Green Chem. 2014, 16, 950-963.
A. J. J. Straathof, Chem. Rev. 2014, 114, 1871-1908.
W. Niu, J. Guo, S. Van Dien, in Metabolic Engineering for Bioprocess
Commercialization (Ed.: S. Van Dien), Springer International Publishing,
Cham, 2016, pp. 31-52.
[4]
[5]
A. M. Weeks, M. C. Y. Chang, Biochemistry 2011, 50, 5404-5418.
J. W. Lee, D. Na, J. M. Park, J. Lee, S. Choi, S. Y. Lee, Nat. Chem. Biol.
2012, 8, 536-546.
[6]
[7]
[8]
[9]
C. E. Nakamura, G. M. Whited, Curr. Opin. Biotechnol. 2003, 14, 454-
459.
A. Burgard, M. J. Burk, R. Osterhout, S. Van Dien, H. Yim, Curr. Opin.
Biotechnol. 2016, 42, 118-125.
H. White, G. Strobl, R. Feicht, H. Simon, Eur J Biochem 1989, 184, 89-
96.
A. He, T. Li, L. Daniels, I. Fotheringham, J. P. N. Rosazza, Appl.
Environ. Microbiol. 2004, 70, 1874-1881.
Analytical Methods.
[10] B. Soehling, G. Gottschalk, Eur. J. Biochem. 1993, 212, 121-127.
[11] N. A. Leal, G. D. Havemann, T. A. Bobik, Arch. Microbiol. 2003, 180,
353-361.
An Agilent 1260 Infinity HPLC instrument (Agilent Technologies)
equipped with an Aminex HPX-87H column (300x7.8 mm) (Bio-
Rad Laboratories) was used for metabolite identification and
quantification. To prepare samples for analysis, cell-free broth
was mixed with internal standard (100 mM oxalic acid) at 95:5
(v/v) ratio, then filtered through a 0.2 µm PTFE syringe filter. A
10 µL of the prepared sample was injected. Samples were
eluted at 45 °C with 5 mM sulfuric acid (0.6 mL/min). All the
carboxylic acids were detected at 210 nm using a diode array
detector. Alcohols and glucose were detected by a refractive
index detector. Retention times of all analytes are shown in
Table S3 of Supporting Information. 1H NMR spectra were
recorded on a Bruker Avance III-HD NMR Spectrometer (300
MHz) instrument. Samples were prepared by mixing cell-free
broth with D2O at a 9:1 (v/v) ratio. The sodium salt of 3-
(trimethylsilyl)propionic-2,2,3,3-d4 acid (TSP) was included in the
[12] S. Weinitschke, K. Hollemeyer, B. Kusian, B. Bowien, T. H. M. Smits, A.
M. Cook, J. Biol. Chem. 2010, 285, 35249-35254.
[13] J. I. Harris, M. Waters, in The Enzymes, Vol. 13 (Ed.: P. D. Boyer),
Academic, 1976, pp. 1-49.
[14] H. Yim, R. Haselbeck, W. Niu, C. Pujol-Baxley, A. Burgard, J. Boldt, J.
Khandurina, J. D. Trawick, R. E. Osterhout, R. Stephen, J. Estadilla, S.
Teisan, H. B. Schreyer, S. Andrae, T. H. Yang, S. Y. Lee, M. J. Burk, S.
Van Dien, Nat. Chem. Biol. 2011, 7, 445-452.
[15] S. Atsumi, A. F. Cann, M. R. Connor, C. R. Shen, K. M. Smith, M. P.
Brynildsen, K. J. Y. Chou, T. Hanai, J. C. Liao, Metab. Eng. 2008, 10,
305-311.
[16] P. Venkitasubramanian, L. Daniels, J. P. N. Rosazza, J. Biol. Chem.
2007, 282, 478-485.
[17] D. Gahloth, M. S. Dunstan, D. Quaglia, E. Klumbys, M. P. Lockhart-
Cairns, A. M. Hill, S. R. Derrington, N. S. Scrutton, N. J. Turner, D. Leys,
Nat. Chem. Biol. 2017, 13, 975-981.
D2O as the internal standard for calibration.
A solvent
suppression program was applied to suppress the signal of
water. The optical purity analysis of 1,2-PDO from the de novo
biosynthesis was carried out following a previously reported
procedure.[26]
[18] W. Finnigan, A. Thomas, H. Cromar, B. Gough, R. Snajdrova, J. P.
Adams, J. A. Littlechild, N. J. Harmer, ChemCatChem 2017, 9, 1005-
1017.
[19] P. Venkitasubramanian, L. Daniels, J. P. N. Rosazza, in Biocatalysis in
the pharmaceutical and biotechnology industries (Ed.: R. N. Patel),
CRC Press LLC, 2007, pp. 425-440.
Acknowledgements
8
This article is protected by copyright. All rights reserved.

10.1002/cbic.201800157
ChemBioChem
FULL PAPER
[20] M. Moura, D. Pertusi, S. Lenzini, N. Bhan, L. J. Broadbelt, K. E. J. Tyo,
Biotechnol. Bioeng. 2016, 113, 944-952.
[31] W. Niu, L. Kramer, K. Liu, J. Thomsen, J. Guo, p. In preparation.
[32] S. Zhou, T. B. Causey, A. Hasona, K. T. Shanmugam, L. O. Ingram,
Appl. Environ. Microbiol. 2003, 69, 399-407.
[21] M. K. Akhtar, N. J. Turner, P. R. Jones, Proc. Natl. Acad. Sci. U. S. A.
2013, 110, 87-92.
[33] T. B. Grabar, S. Zhou, K. T. Shanmugam, L. P. Yomano, L. O. Ingram,
Biotechnol. Lett. 2006, 28, 1527-1535.
[22] Z. Zhu, Y. J. Zhou, M.-K. Kang, A. Krivoruchko, N. A. Buijs, J. Nielsen,
Metab. Eng. 2017, 44, 81-88.
[34] S. Zhou, K. T. Shanmugam, L. O. Ingram, Appl. Environ. Microbiol.
2003, 69, 2237-2244.
[23] M. Winkler, Current opinion in chemical biology 2018, 43, 23-29.
[24] G. Qu, J. Guo, D. Yang, Z. Sun, Green Chemistry 2018, 20, 777-792.
[25] S. Shelley, Chem. Eng. Prog. 2007, 103, 6-9.
[35] J. H. Miller, Experiments in Molecular Genetics, Cold Spring Harbor
Lab., 1972.
[26] W. Niu, J. Guo, ACS Synth. Biol. 2015, 4, 378-382.
[27] A. E. Martin, F. H. Murphy, in Kirk-Othmer Encyclopedia of Chemical
Technology, John Wiley & Sons, Inc., 2000.
[36] J. F. Sambrook, D. W. Russell, Editors, Molecular cloning: A laboratory
manual, third edition, Cold Spring Harbor Laboratory Press, 2000.
[37] R. Lutz, H. Bujard, Nucleic Acids Res. 1997, 25, 1203-1210.
[38] K. A. Datsenko, B. L. Wanner, Proc. Natl. Acad. Sci. U. S. A. 2000, 97,
6640-6645.
[28] S. Kumar, G. Stecher, K. Tamura, Mol. Biol. Evol. 2016, 33, 1870-1874.
[29] A. N. Khusnutdinova, R. Flick, A. Popovic, G. Brown, A. Tchigvintsev, B.
Nocek, K. Correia, J. C. Joo, R. Mahadevan, A. F. Yakunin, Biotechnol.
J. 2017, 12, n/a.
[30] A. Pick, B. Ruhmann, J. Schmid, V. Sieber, Appl. Microbiol. Biotechnol.
2013, 97, 5815-5824.
9
This article is protected by copyright. All rights reserved.

10.1002/cbic.201800157
ChemBioChem
FULL PAPER
Entry for the Table of Contents (Please choose one layout)
Layout 1:
FULL PAPER
Find a CAR: Carboxylic acid
Levi Kramer, Erome Hankore, Yilan Liu,
Kun Liu, Esteban Jimenez, Jiantao Guo,
and Wei Niu
reductases (CARs) from different
sources were fully characterized using
a panel of short-chain dicarboxylic
acids and hydroxyacids. Whole-cell
bioconversions of twelve carboxylic
acid substrates into corresponding
alcohols were investigated by coupling
the CAR activity with an aldehyde
reductase in E. coli hosts. The de
novo stereospecific biosynthesis of
1,2-propanediol isomers was
Page No. – Page No.
Characterization of Carboxylic Acid
Reductases for Biocatalytic
Synthesis of Industrial Chemicals
successfully demonstrated.
10
This article is protected by copyright. All rights reserved.