Whole Cell-Based Cascade Biotransformation for the Production of (S)-Mandelic Acid from Styrene, L-Phenylalanine, Glucose, or Glycerol

Article abstract of DOI:10.1002/adsc.201900373

(S)-Mandelic acid is a useful and high value chemical with many synthetic applications, but its synthesis often requires the use of toxic cyanide. Here, we report the development of several cyanide-free methods to prepare (S)-mandelic acid via cascade biotransformation. Enhanced production of (S)-mandelic acid from styrene via 4-step enzyme cascades was achieved with Escherichia coli (A-M1_R-M2) cells, giving 118–144 mM (18–21.9 g/L, 72–79% yield) product. The process was scaled up to 1.5 L to produce 140 mM (21.3 g/L) of (S)-mandelic acid in 70% yield. A strategy with the recycling and reuse of Escherichia coli cells, unreacted substrate, and organic solvent was developed to enhance the productivity of (S)-mandelic acid through repeated batches, affording 190 mM (95% yield) and 328 mM (82% yield) product in two and four batches, respectively. (S)-mandelic acid was also produced from bio-derived L-phenylalanine via 6-step enzyme cascades. Biotransformation of L-phenylalanine with Escherichia coli (LZ37) cells expressing all enzymes for the reactions gave 160 mM (S)-mandelic acid in 80% yield. Moreover, (S)-mandelic acid were synthesized from glycerol or glucose via L-phenylalanine biosynthesis pathway and the 6-step enzyme cascade. Coupling of Escherichia coli (NST74-Phe) with Escherichia coli (LZ37) enabled the sustainable production of 63 mM (10 g/L) or 52 mM (8 g/L) (S)-mandelic acid from renewable feedstocks glycerol and glucose, respectively. (Figure presented.).

Full text of DOI:10.1002/adsc.201900373

Title: Whole Cell-Based Cascade Biotransformation for the
Production of (<i>S</i>)-Mandelic Acid from Styrene, <i>L</i>-
Phenylalanine, Glucose, or Glycerol
Authors: Benedict Ryan Lukito, Balaji Sundara Sekar, Shuke Wu, 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: Adv. Synth. Catal. 10.1002/adsc.201900373
Link to VoR: http://dx.doi.org/10.1002/adsc.201900373

10.1002/adsc.201900373
Advanced Synthesis & Catalysis
FULL PAPER
DOI: 10.1002/adsc.201((will be filled in by the editorial staff))
Whole Cell-Based Cascade Biotransformation for the
Production of (S)-Mandelic Acid from Styrene, L-Phenylalanine,
Glucose, or Glycerol
Benedict Ryan Lukito,a Balaji Sundara Sekar,a Shuke Wu,a and Zhi Lia*
a
Department of Chemical and Biomolecular Engineering, National University of Singapore, 4 Engineering Drive 4,
Singapore 117585
Phone: (+65) 6516-8416; Email: chelz@nus.edu.sg
Received: ((will be filled in by the editorial staff))
Supporting information for this article is available on the WWW under http://dx.doi.org/10.1002/adsc.201######.((Please
delete if not appropriate))
Abstract: (S)-Mandelic acid is a useful and high value (S)-mandelic acid was also produced from bio-derived L-
chemical with many synthetic applications, but its synthesis phenylalanine
via
6-step
enzyme
cascades.
often requires the use of toxic cyanide. Here, we report the Biotransformation of L-phenylalanine with Escherichia coli
development of several cyanide-free methods to prepare (S)- (LZ37) cells expressing all enzymes for the reactions gave
mandelic acid via cascade biotransformation. Enhanced 160 mM (S)-mandelic acid in 80% yield. Moreover, (S)-
production of (S)-mandelic acid from styrene via 4-step mandelic acid were synthesized from glycerol or glucose
enzyme cascades was achieved with Escherichia coli (A- via L-phenylalanine biosynthesis pathway and the 6-step
M1_R-M2) cells, giving 118-144 mM (18-21.9 g/L, 72-79% enzyme cascade. Coupling of Escherichia coli (NST74-
yield) product. The process was scaled up to 1.5 L to Phe) with Escherichia coli (LZ37) enabled the sustainable
produce 140 mM (21.3 g/L) of (S)-mandelic acid in 70% production of 63 mM (10 g/L) or 52 mM (8 g/L) (S)-
yield. A strategy with the recycling and reuse of Escherichia mandelic acid from renewable feedstocks glycerol and
coli cells, unreacted substrate, and organic solvent was glucose, respectively.
developed to enhance the productivity of (S)-mandelic acid
through repeated batches, affording 190 mM (95% yield)
and 328 mM (82% yield) product in two and four batches,
respectively.
Keywords: asymmetric synthesis; biotransformations;
cascade reactions; enzyme catalysis; mandelic acid;
sustainable chemistry
catalyzed kinetic resolution of racemic mandelic acid
via enantioselective transesterification was reported,
but suffering from 50% theoretical yield.[6]
Oxynitrilase was reported to catalyze the addition of
benzaldehyde with HCN to give mandelonitrile,
followed by nitrilase-catalyzed hydrolysis to yield
Introduction
Enantiopure mandelic acid is a useful and valuable
intermediate for chemical and pharmaceutical
synthesis. For instance, (R)-mandelic acid is used for
the production of antibiotics such as cephalosporins
and semi-synthetic penicillins, while (S)-mandelic
acid is used for the production of non-steroidal anti-
inflammatory drugs including deracoxib and
celecoxib.[1] The conventional synthesis of this chiral
chemical is mainly based on the cyanation of
benzaldehyde and subsequent hydrolysis with HCl to
produce racemic mandelic acid, followed by the
resolution with a maximum theoretical yield of
50%.[2] Cyanide-based asymmetric syntheses of (R)-
mandelic acid via cyanohydrin[3] and (S)-mandelic
acid via phenylglyoxylic[4] were reported. However,
all of these chemical methods use highly toxic
cyanide chemistry.
(S)-mandelic
acid.[7]
Alcohol
dehydrogenase
catalyzed bioreduction of phenylglyoxylic acid[1a, 8] or
ethyl benzyolformate[9] to produce (R)-mandelic acid.
Nevertheless, all of these enzymatic methods require
the use of highly toxic cyanide for the preparation of
the substrate or for the biotransformation.
In order to develop cyanide-free synthesis of chiral
mandelic acid, we have been focusing on enzymatic
cascade reactions which allows for multi-step
reaction in one pot to enable new chemistry and avoid
the costly isolation of intermediates and generation of
waste.[10] By developing cascade biotransformation, it
is possible to synthesize chiral mandelic acid from
easily available substrate without the use of cyanide.
Enzyme catalysis is a green and useful tool for the
synthesis of fine chemicals due to high chem-, regio-,
stereo-selectivity,[5] and it has also been used for the
production of enantiopure mandelic acid. Lipase-
We previously reported
a
novel cascade
biotransformation to convert the cheap and easily
available styrene to (S)-mandelic acid via epoxidation
– hydrolysis – oxidation – oxidation (Scheme 1a).[11]
1
This article is protected by copyright. All rights reserved.

10.1002/adsc.201900373
Advanced Synthesis & Catalysis
Scheme 1. Cascade biotransformation to produce (S)-mandelic acid 5. (a) Cascade biotransformation of styrene 1 to (S)-5
with enzymes cascades containing SMO, SpEH, AlkJ, and EcALDH. (b) Cascade biotransformation of L-phenylalanine 6
to (S)-5 with enzymes cascades containing PAL, PAD, SMO, SpEH, AlkJ, and EcALDH. (c) Cascade biotransformation of
glycerol 8 to (S)-5 by combining shikimate pathway with enzyme cascades containing PAL, PAD, SMO, SpEH, AlkJ, and
EcALDH. (d) Cascade biotransformation of glucose 9 to (S)-5 by combining shikimate pathway with enzyme cascades
containing PAL, PAD, SMO, SpEH, AlkJ, and EcALDH.
buffer (200 mM, pH 8.0, 0.5% glucose) and 10 mL n-
This provides with a novel cyanide-free method for
asymmetric oxy-functionalization of terminal alkenes
to hydroxycarboxylic acids. Recently, we also
developed a novel cascade biotransformation to
convert renewable L-phenylalanine to biobased (S)-
mandelic acid via deamination – decarboxylation –
epoxidation – hydrolysis – oxidation – oxidation
(Scheme 1b).[12] Here we report the enhancement of
the production of (S)-mandelic acid from styrene or
L-phenylalanine by whole cell-based cascade
biotransformation as well as the development of a
new strategy to achieve high productivity via the
recycling and reuse of the cells, unreacted substrate,
and organic solvent. Moreover, we also report the
microbial production of biobased (S)-mandelic acid
from glycerol and glucose with engineered whole-cell
biocatalysts.
hexadecane containing 150, 200, 250, or 300 mM
styrene 1. Reaction was carried out at 30oC for 29 h.
As shown in Table 1, 118 mM (S)-mandelic acid 5
(18 g/L, 79% yield) was obtained from 150 mM
styrene 1 at 20 g cdw/L of cells. The highest
concentration of (S)-5 was 144 mM (72% yield),
which was achieved by using 20 g cdw/L of cells and
200 mM styrene 1. 7 mM of 2-phenylethanol was
detected as the byproduct possibly produced by the
endogenous alcohol dehydrogenase inside the E. coli
cells which has been reported before.[13] The detailed
time course on the production of (S)-5 from styrene 1
using this conditions was shown in Figure S2.
Biotransformation with Resting Cells on 1.5 L-
Scale
Scale-up was performed at the established optimal
condition in 1.5 L-scale bioreactor. Biotransformation
of styrene 1 to (S)-5 was carried out with 20 g cdw/L
of E. coli (A-M1_R-M2) cells in a mixture of 750 mL
KP buffer (200 mM, pH 8.0, 0.5% glucose) and 750
mL n-hexadecane containing 200 mM styrene 1 at
30oC for 29 h. The time course on the production of
(S)-5 was shown in Figure 1a: styrene 1 was rapidly
converted to (S)-5 in the first 12 h. At the end of the
biotransformation, 140 mM (21.3 g/L) of (S)-5 was
obtained in 70% yield. 25% of styrene 1 remained
Results and Discussion
Cascade Biotransformation of Styrene 1 to (S)-
Mandelic Acid 5 with E. coli (A-M1_R-M2) Cells
Optimization of Reaction Conditions
Optimization was performed with different cell
density (10, 15, 20, 25, or 30 g cdw/L) of Escherichia
coli (A-M1_R-M2)[11] in a mixture of 10 mL KP
2
This article is protected by copyright. All rights reserved.

10.1002/adsc.201900373
Advanced Synthesis & Catalysis
Figure 1. (a) Time course of biotransformation of styrene 1 (200 mM) to (S)-5 with resting cells of E. coli (A-M1_R-M2)
cells (20 g cdw/L) in a 1.5 L mixture of KP buffer (200 mM, pH 8.0, 0.5% glucose) and n-hexadecane (1:1, v/v) in
bioreactor at 30oC. (b) Time course of biotransformation of styrene 1 (200 mM) to (S)-5 with growing cells of E. coli (A-
M1_R-M2) in a bioreactor. The cells were grown in 0.75 L M9 medium containing 2% (w/v) glucose at 22oC. At 12 h,
0.75 L n-hexadecane containing 200 mM styrene were added to start the biotransformation at 30oC. (c) Comparison of
using biodiesel and n-hexadecane as the organic phase in two-phase biotransformation of styrene 1 to (S)-5. The reaction
was performed with the resting cells of E. coli (A-M1_R-M2) cells (20 g cdw/L) in a mixture of 10 mL KP buffer (200
mM, pH 8.0, 0.5% glucose) and 10 mL n-hexadecane or biodiesel containing 200 mM styrene at 30oC for 29 h. (d) Time
course of biotransformation of styrene 1 (200 mM) to (S)-5 with E. coli (A-M1_R-M2) cells (20 g cdw/L) in a 1.5 L
mixture of KP buffer (200 mM, pH 8.0, 0.5% glucose) and biodiesel (1:1, v/v) in bioreactor at 30oC.
unreacted, together with low concentration of the
intermediates (S)-3 and (S)-4. These results in the
bioreactor were similar to those obtained in the
shaking flask under the same conditions, indicating
that our biotransformation could be easily scale up.
1b: At 12 h, 15 g cdw/L of cells were obtained which
was significantly higher than the cell density
achieved in 250-mL shaking flask. This could be due
to the better aeration as well as better control of pH
and temperature in bioreactor. The cells reached the
highest cell density of 17 g cdw/L at 16 h and
maintained the same density during the
biotransformation. After 29 h of biotransformation,
133 mM (20.2 g/L) of (S)-mandelic acid 5 was
obtained together with 52 mM unreacted styrene 1.
The concentration of intermediates (S)-3 and (S)-4
were maintained low. The productivity of this
growing system is similar to that of resting cells.
Biotransformation with Growing Cells on 1.5 L-
Scale
E. coli (A-M1_R-M2) cells were cultured in 750 mL
M9 medium containing 20 g/L glucose, 6 g/L yeast
extract, 50 mg/L chloramphenicol, and 50 mg/L
kanamycin. After 12 hours of cell growth, 750 mL n-
hexadecane containing 200 mM of styrene 1 was
added to perform the biotransformation at 30oC for
another 29 h. The time course of the one-pot cell
growth and biotransformation was shown in Figure
3
This article is protected by copyright. All rights reserved.

10.1002/adsc.201900373
Advanced Synthesis & Catalysis
Table 1. Biotransformation of styrene 1 to (S)-mandelic
acid 5 via four-step enzyme cascades with E. coli (A-
M1_R-M2) cells in KP buffer-n-hexadecane two-phase
system.
biodiesel (B100, pure, Alpha Biofuels). With
biodiesel, 113 mM and 133 of (S)-mandelic acid
5 was produced within 12 and 24 h, corresponding to
57% and 67% yield, respectively. The system was
then scaled up in a 2-Liter fermenter. The time course
of biotransformation using the biodiesel-aqueous
system was shown in Figure 1d: styrene 1 was rapidly
converted to (S)-mandelic acid 5 in the first 12 h,
giving 112 mM product. At the end of the
biotransformation, 136 mM (20.7 g/L) of (S)-
mandelic acid 5 was obtained in 68% yield, while
25% of styrene 1 (52 mM) remained unreacted. The
intermediates (S)-3 and (S)-4 existed in a low
concentration.
Styrene 1
Cell
(S)-5
Conversion
to (S)-5
(%)c)
Concentration Density Concentration
(mM)a)
(g
cdw/L)
10
(mM) b,c)
150
150
150
150
150
200
200
200
200
200
250
250
250
250
250
300
300
300
300
300
104±4
116±5
118±5
113±4
111±4
132±6
135±6
144±5
137±6
134±7
137±6
142±5
144±7
143±8
138±7
131±7
132±5
134±7
133±7
136±7
69±3
77±3
79±3
75±3
74±3
66±2
68±3
72±2
69±2
67±3
55±2
57±2
58±3
58±3
55±3
44±2
44±1
45±2
44±2
45±3
15
20
25
30
10
15
20
25
30
10
15
20
25
30
10
15
Strategy of Achieving High Productivity for the
Cascade Biotransformation of Styrene 1 to (S)-
Mandelic Acid 5 via Recycling and Reuse of
Biocatalyst, Unreacted Substrate, and Organic
Solvent
Product inhibition was observed for the cascade
biotransformation of styrene 1 to (S)-mandelic acid 5
with E. coli (A-M1_R-M2) cells. To investigate the
inhibition, 20 g cdw/L of E. coli (A-M1_R-M2) cells
were pretreated with (S)-mandelic acid 5 (0, 50, 100,
200, 250, 300, 400, or 600 mM), followed by
biotransformation with the pretreated cells. The
results was shown in Figure S3, and the cells started
to lose its productivity after being pretreated with (S)-
mandelic acid 5 at 200 mM. In n-hexadecane-
aqueous system, >99% of the substrate remained in
the organic phase, and >99% of the product still
remained in aqueous phase, which could cause the
toxicity to the cells.[16]
To solve the product inhibition problem, we
developed a new strategy for repeated batch with the
recycling and reuse of biocatalyst, unreacted substrate,
and organic solvent. The principle was demonstrated
as follows: 10 mL KP buffer (200 mM, pH 8.0)
containing 20 g cdw/L of E. coli (A-M1_R-M2) cell
suspension and 0.5% glucose were mixed with 10 mL
n-hexadecane containing different concentration of
styrene 1 (200, 250, 300, 350, 400, or 500 mM) to
perform the biotransformation at 30oC for 12 h,
followed by separation of aqueous and organic phase
(Figure 2a). Cells were further removed from the
aqueous phase and resuspended in fresh buffer, and
the recycled organic phase containing the unreacted
substrate was added. The biotransformation was
carried out under the same conditions. Significant
improvement was observed using the new system.
Cumulatively, 190 mM of (S)-mandelic acid 5 (95%
yield) was obtained from 200 mM styrene 1 with two
repeated batch experiments. This result is better than
the 144 mM (S)-5 obtained from 200 mM styrene 1
without the use of this recycling system. With three
batches biotransformation, 220 mM (S)-mandelic
acid 5 was obtained from 250 mM styrene 1 in 88%
yield. The highest product concentration reached 328
mM from 400 mM styrene 1 in 82% yield. The new
approach successfully enhanced the overall
20
25
30
a)
The concentration refers to the organic phase, and the
reaction was performed in a mixture of KP buffer (200
mM, pH 8.0, 0.5% glucose) and n-hexadecane (1:1, v/v) at
30oC for 29 h. b) The concentration refers to the aqueous
phase and determined by HPLC analysis. No (S)-5 was
c)
detected in the organic phase. The data are the mean
value with standard deviation of triplicate experiments.
Biotransformation with Resting Cells in Biodiesel-
Aqueous Two-Phase System
Selection of potential biocompatible organic solvent
for the two-phase biotransformations is very
important for developing an effective and
environmentally-friendly
biotransformation
system.[14] While n-hexadecane was a good solvent
for the production of (S)-mandelic acid 5, there are
still some drawbacks of utilizing the conventional
organic solvents. Therefore, we explored biodiesel as
an alternative, cheap, and greener organic solvent in
two-phase biotransformation.[15] Reaction was
performed with 20 g cdw/L of E. coli (A-M1_R-M2)
cells in a mixture of 10 mL KP buffer (200 mM, pH
8.0) containing 0.5% glucose and 10 mL n-
hexadecane or biodiesel containing 200 mM styrene 1
at 30oC for 29 h. Figure 1c gave the similar trends for
the two cases where the concentration of (S)-
mandelic acid 5 reached 22 mM at the first 2 h of the
biotransformation using either n-hexadecane or
4
This article is protected by copyright. All rights reserved.

10.1002/adsc.201900373
Advanced Synthesis & Catalysis
con
vers
ion
of
sub
stra
te
to
pro
duc
t
and
red
uce
d
the
pro
duc
tion
cost
of
enz
ym
es
and
org
anic
solv
ent
Fig
ure
2.
(a) Flowchart of two-phase biotransformation of styrene 1 to (S)-5 by using resting cells of E. coli (A-M1_R-M2) with the
recycling and reuse of the cells, unreacted substrate, and organic phase. (b) Production of (S)-5 using multi-batch
biotransformation of styrene 1 (200-500 mM) in two-phase system with the recycling and reuse of E. coli (A-M1_R-M2)
cells, unreacted substrate, and organic solvent. Reaction was carried out with E. coli (A-M1_R-M2) cells (20 g cdw/L) in a
mixture of KP buffer (200 mM, pH 8.0, 0.5% glucose) and n-hexadecane (1:1, v/v) at 30oC for 12 h, followed by the
separation of organic and aqueous phase. Cells were further removed from the aqueous phase, resuspended in fresh buffer,
and mixed with the recycled organic solvent containing unreacted substrate to carry out the next batch of
biotransformation under the same conditions.
through recycling. To our best knowledge, this is the
highest product concentration in bioproduction of (S)-
mandelic acid 5 reported thus far (See Table S2 in SI).
based cascade biotransformation was further
optimized. Different concentration of E. coli (LZ37)
cells (10-25 g cdw/L), L-phenylalanine 6 (120-200
mM), and glucose (0.5-1%) were examined, and the
reaction was performed in a mixture of KP buffer
(200 mM, pH 8.0) and n-hexadecane (1:1, v/v) at
30oC for 27 h. As shown in Table 2, 160 mM (24
g/L) (S)-mandelic acid 5 was achieved in 80% yield
from 200 mM L-phenylalanine 6 with 20 g cdw/L
cells. In comparison, this product concentration from
L-phenylalanine 6 is even higher than that (144 mM)
from styrene 1 in a single batch reaction.
Cascade Biotransformation of L-phenylalanine 6
to (S)-Mandelic Acid 5 with Resting Cells of E. coli
(LZ37)
Development of green and sustainable process has
attracted increasing attention towards the synthesis of
(S)-mandelic acid 5 from bio-derived L-phenylalanine
6
via
biotransformation.
This
type
of
biotransformation is catalyzed via a six-step cascade
reactions (Scheme 1b), and E. coli (LZ37) had been
previously engineered to express all the necessary
enzymes for the six-step reactions.[12] The whole-cell
5
This article is protected by copyright. All rights reserved.

10.1002/adsc.201900373
Advanced Synthesis & Catalysis
Table 2. Biotransformation of L-phenylalanine 6 to (S)-mandelic acid 5 via six-step enzyme cascades with E. coli (LZ37)
cells in KP buffer-n-hexadecane two-phase system.
L-6 concentration
Cell Density
( g cdw/L)
Glucose
Concentration (%)
(S)-5 Concentration
(mM) b,c)
98±5
Conversion to (S)-5 (%)c)
(mM)a)
120
120
150
150
200
200
200
200
200
200
200
10
10
10
10
10
10
15
15
20
20
25
0.5
1
0.5
1
0.5
1
0.5
1
82±4
86±5
73±5
75±5
40±3
48±3
52±3
61±3
62±4
80±4
76±4
103±6
110±7
113±7
79±6
97±5
103±6
121±6
124±8
0.5
1
1
160±8
153±7
a) The concentration refers to the organic phase, and the reaction was performed in a mixture of KP buffer (200 mM, pH
8.0) and n-hexadecane (1:1, v/v) at 30oC for 27 h. The concentration refers to the aqueous phase and determined by
b)
c)
HPLC analysis. No (S)-5 was detected in the organic phase. The data are the mean value with standard deviation of
triplicate
experiment.
Figur
e
3.
(a)
Flowc
hart of
the
produ
ction
of (S)-
5
from
glycer
ol 8 or
glucos
e
9
via
coupl
ed
E.coli
(NST
74-
Phe)
cells
and E.
coli
(LZ37
) cells.
(b)
Time
course of biotransformation of glycerol 8 to (S)-5 via coupling E.coli (NST74-Phe) cells, which produced 72 mM L-6, with
E. coli (LZ37) cells which catalyzed the conversion of L-6 to (S)-5. E. coli (LZ37) cell pellet (15 g cdw/L) was added into
the reaction mixture containing L-6, E. coli (NST74-Phe), KP buffer (200 mM, pH 8.0), and glucose (1%). n-Hexadecane
(1:1, v/v) was added to start the biotransformation at 30oC. (c) Time course of biotransformation of glucose 8 to (S)-5 via
coupling E.coli (NST74-Phe) cells, which produced 60 mM L-6, with E. coli (LZ37) cells (15 g cdw/L) which catalyzed
the conversion of L-6 to (S)-5. E. coli (LZ37) cell pellet (15 g cdw/L) was added into the reaction mixture containing L-6,
E. coli (NST74-Phe), KP buffer (200 mM, pH 8.0), and glucose (1%). n-Hexadecane (1:1, v/v) was added to start the
biotransformation
at
30oC.
6
This article is protected by copyright. All rights reserved.

10.1002/adsc.201900373
Advanced Synthesis & Catalysis
Bioproduction of (S)-Mandelic Acid 5 from
Glycerol 8 or Glucose 9 via L-phenylalanine 6
Biosynthesis Pathway and 6-Step Enzyme
Cascades by Coupling of Two Escherichia coli
Strains.
biosynthesis pathway and six-step enzyme cascades
by coupling two recombinant E. coli strains. 63 mM
(10 g/L) and 52 mM (8 g/L) (S)-mandelic acid 5 were
obtained from glycerol and glucose, respectively.
These results provide with simple and high-yielding
synthesis of enantiopure (S)-mandelic acid 5 as
biobased chemicals from the easily renewable
feedstocks, giving much higher product concentration
than the best results (0.24 g/L of mandelic acid) of
the reported synthesis from glucose. The developed
four cyanide-free methods enable green synthesis of
(S)-mandelic acid with the potential of synthetic
application
Sustainable synthesis of (S)-mandelic acid 5 from
glycerol or glucose was examined by coupling E. coli
(NST74-Phe) to convert glycerol or glucose to L-
phenylalanine 6 via Shikimate pathway[17] and E. coli
(LZ37) to further convert L-phenylalanine 6 to (S)-
mandelic acid 5 via six-step enzyme cascades (Figure
3a). E. coli (NST74-Phe) was first grown on glycerol
or glucose to give 78 mM and 67 mM L-
phenylalanine 6 at 28 h, respectively. E. coli (LZ37)
cells were then added to the fermentation mixture of
E. coli (NST74-Phe) to catalyze the conversion of L-
phenylalanine 6 to (S)-mandelic acid 5 at 30oC for 24
h. As shown in Figure 3b and 3c, 62 and 52 mM (S)-
Experimental Section
Biotransformation of Styrene 1 to (S)-Mandelic Acid 5
with Resting Cells of E. coli (A-M1_R-M2) in Two-
Phase System Involving n-Hexadecane or Biodiesel on
1.5 L-Scale
mandelic acid
5
were obtained from the
biotransformation of glycerol or glucose, respectively.
The remaining substrate and intermediates
maintained in low concentration. To our best
knowledge, this is by far the highest concentration of
(S)-mandelic acid 5 produced from glycerol or
glucose. The best results reported thus far was the de
novo biosynthesis using aromatic amino acid pathway
with hydroxymandelate synthases which produced
0.24 g/L of mandelic acid.[18]
750 mL of potassium phosphate buffer (200 mM, pH 8.0)
containing 20 g cdw/L of E. coli (A-M1_R-M2) and 0.5%
glucose were mixed with 750 mL n-hexadecane or
biodiesel containing 200 mM styrene 1 in a 2 L-bioreactor
at 30oC. The mixture was stirred at 700 rpm for 29 h. For
product analysis, 50 µL organic phase was collected and
diluted with 950 µL ethyl acetate containing 2 mM benzyl
alcohol. The sample was analyzed by using Gas
Chromatography (GC) to determine the concentration of
styrene 1 and (S)-2. On the other hand, 50 µL aqueous
phase was also taken from the biotransformation mixture
and diluted with 450 µL ultrapure water containing 0.5%
TFA and 500 µL ACN containing 2 mM benzyl alcohol.
The sample was analyzed by using reverse phase HPLC to
determine the concentration of (S)-3, (S)-4, and (S)-5. 140
mM of (S)-5 was produced after 29 h of biotransformation.
Conclusion
Enhanced production of (S)-mandelic acid 5 from
the easily available styrene 1 was achieved by
Biotransformation of Styrene to (S)-Mandelic Acid
with Growing Cells of E. coli (A-M1_R-M2) on 1.5 L-
Scale
optimization
of
whole
cell-based
cascade
biotransformation. 118 – 144 mM (18 – 21.9 g/L) of
(S)-mandelic acid 5 was obtained from 150 – 200
20 mL of E. coli (A-M1_R-M2) cells were inoculated with
LB medium containing 50 mg/L kanamycin and 50 mg/L
chloramphenicol at 37oC and 250 rpm for 8-10 hours. The
inoculated strain was subsequently transferred and grown
into 2-Liter bioreactor containing M9 medium
supplemented with 20 g/L glucose, 6 g/L yeast extract, 50
mg/L kanamycin, and 50 mg/L chloramphenicol in a total
volume of 750 mL at 22oC and 20% of OD level. IPTG
(0.5 mM) was added at 2 h to induce the protein expression.
750 mL of n-hexadecane containing 200 mM styrene 1
were added during the 12 h of the cell growth into the
bioreactor to perform the reaction at 30oC and 700 rpm for
another 29 h. Sample preparation and product analysis
were performed as described in the previous section. 133
mM of (S)-5 was produced after 29 h of biotransformation.
mM styrene
1
in 72
–
79% yield. The
biotransformation was successfully scaled up to 1.5 L,
producing 140 mM (21.3 g/L) in 70% yield of (S)-
mandelic acid 5 from 200 mM styrene 1. A novel
strategy via the recycling and reuse of E. coli cells,
unreacted substrate, and organic solvent to further
improve
productivity
was
developed.
Biotransformation with this approach cumulatively
produced 190 mM of (S)-mandelic acid 5 in 95%
yield with 2 batches and 328 mM of (S)-mandelic
acid 5 in 82% yield with 4 batches, respectively. The
approach significantly enhanced the overall
conversion, save the cost of enzymes and organic
solvent through the recycling, and provide with
practical synthesis of (S)-mandelic acid 5 from the
easily available starting material. Enhanced
productivity for the production of (S)-mandelic acid 5
from biobased L-phenylalanine 6 was also achieved
via whole cell-based cascade biotransformation,
producing 160 mM (24 g/L) (S)-mandelic acid 5 from
L-phenylalanine in 80% yield. Last but not least, (S)-
mandelic acid 5 was also successfully produced from
General Procedure of Biotransformation of Styrene 1
to (S)-Mandelic Acid 5 with the Recycling and Reused
of E. coli (A-M1_R-M2), Unreacted Substrate, and
Organic Solvent
10 mL aqueous phase containing 20
g cdw/l of
resuspended E. coli (A-M1_R-M2) cells and 0.5% glucose
inside the KP Buffer (200 mM, pH 8) were mixed with 10
mL n-hexadecane containing 200-500 mM styrene 1 to
perform the biotransformation at 30oC and 250 rpm for 12
h. Both organic and aqueous phase were separated, and
cells were subsequently removed from the aqueous phase,
resuspended in fresh buffer, and mixed with original n-
hexadecane phase containing unreacted substrate to carry
glycerol
and
glucose
via
L-phenylalanine
7
This article is protected by copyright. All rights reserved.

10.1002/adsc.201900373
Advanced Synthesis & Catalysis
on the biotransformation under the same conditions.
Sample preparation and product analysis were performed
as described in the previous section. Biotransformation of
200 mM and 400 mM styrene 1 gave cumulatively 190
mM (95% yield) in 2 batches and 328 mM of (S)-mandelic
acid 5 (82% yield) in 4 batches, respectively.
Macromol. 2016, 88, 9-17; b) Z.-J. Zhang, J.-H. Xu,
Y.-C. He, L.-M. Ouyang, Y.-Y. Liu, T. Imanaka,
Process Biochem. 2010, 45, 887-891; c) S. K. Bhatia,
P. K. Mehta, R. K. Bhatia, T. C. Bhalla, Appl.
Microbiol. Biotechnol. 2014, 98, 83-94.
[2] A. J. Blacker, I. N. Houson, WO Patent
2002066410A1 2002.
[3] J.-M. Brunel, I. P. Holmes, Angew. Chem., Int. Ed.
2004, 43, 2752-2778.
[4] a) P.-C. Yan, J.-H. Xie, X.-D. Zhang, K. Chen, Y.-Q.
Li, Q.-L. Zhou, D.-Q. Che, Chem. Commun. 2014, 50,
15987-15990; b) L. Martínková, V. Křen, Appl.
Microbiol. Biotechnol. 2018, 102, 3893-3900.
[5] a) Y. Zhou, S. Wu, Z. Li, Adv. Synth. Catal. 2017,
359, 4305-4316; b) D.-S. Lee, J.-W. Song, M. Voß, E.
Schuiten, R. K. Akula, Y.-U. Kwon, U. Bornscheuer,
J.-B. Park, Adv. Synth. Catal. 2019, 361, 1359-1367.
[6] a) T. Miyazawa, S. Kurita, S. Ueji, T. Yamada,
Biocatal. Biotransfor. 2000, 17, 459-473; b) B.-D.
Ma, H.-L. Yu, J. Pan, J.-Y. Liu, X. Ju, J.-H. Xu,
Bioresour. Technol. 2013, 133, 354-360; c) G. D.
Yadav, P. Sivakumar, Biochem. Eng. J. 2004, 19,
101-107; d) C. Yao, Y. Cao, S. Wu, S. Li, B. He, J.
Mol. Catal. B Enzym. 2013, 85-86, 105-110.
Cascade Biotransformation of L-phenylalanine 6 to (S)-
Mandelic Acid 5 with Resting Cells of E. coli (LZ37)
10 mL aqueous phase containing E. coli (LZ37) cells[12]
(10, 15, 20, or 25 g cdw/L), L-phenylalanine 6 (120, 150,
or 200 mM), and 0.5-1% glucose in the KP buffer (200
mM, pH 8.0) were mixed with 10 mL n-hexadecane to
perform the biotransformation at 30oC for 27 h. 50 µL of
aqueous sample was collected and diluted with 500 µL
ACN containing 2 mM benzyl alcohol and 450 µL
ultrapure water containing 0.5% TFA for HPLC analysis to
quantify the concentration of (S)-3, (S)-4, (S)-5, and L-6.
50 µL of organic phase was also collected and diluted with
950 µL ethyl acetate containing 2 mM benzyl alcohol for
GC analysis to quantify the concentration of styrene 1 and
(S)-2.
Bioproduction of (S)-Mandelic Acid 5 from Glycerol 8
or Glucose 9 via L-phenylalanine 6 Biosynthesis
Pathway and 6-Step Enzyme Cascades by Coupling of
Two Escherichia coli Strains
E. coli (NST74-Phe) was inoculated in 100 mL LB
medium containing 50 µg/mL streptomycin at 37oC and
250 rpm for 12 h. The inoculated strain was subsequently
transferred and cultured in 1.5-liter bioreactor with the
fermentation medium[19] containing 10 g/L (NH4)2SO4, 5
g/L KH2PO4, 5 g/L MgSO4, 5 g/L yeast extract, 50 µg/mL
streptomycin, and 10 g/L carbon source (glycerol or
glucose) for 28 h. IPTG was added to a final concentration
of 0.1 mM at 8 h to induce the protein expression. 30%
NH4OH and 80% carbon sources were used as feed to
control the carbon source at pH 6.8 and 5-10 g/L,
respectively, without acetate accumulation, to give the
final L-phenylalanine 6 concentration of 78 mM or 67 mM
from glycerol or glucose, respectively.
15 g cdw/L of E. coli (LZ37) cell pellet, 250 µL KP
buffer (4M, pH 8.0), and 125 µL 40% glucose solution
were added into the fermented media containing L-6 (78
mM or 67 mM) and E. coli (NST74-Phe) cells to give 10
mL reaction mixture containing 72 or 60 mM of L-6, KP
buffer (200 mM, pH 8.0), 1% glucose, and 15 g cdw/L of
E. coli (LZ37) cell suspension for the bioproduction of (S)-
5. 10 mL n-hexadecane was added to perform the
biotransformation at 30oC and 250 rpm for 24 h. For
product analysis, 50 µL organic phase was collected and
diluted with 950 µL ethyl acetate containing 2 mM benzyl
alcohol. The sample was analyzed by using Gas
Chromatography (GC) to determine the concentration of
styrene 1 and (S)-2. On the other hand, 50 µL aqueous
phase was also taken from the biotransformation mixture
and diluted with 450 µL ultrapure water containing 0.5%
TFA and 500 µL ACN containing 2 mM benzyl alcohol.
The sample was analyzed by using reverse phase HPLC to
determine the concentration of (S)-3, (S)-4, (S)-5 and L-6.
[7] a) C. Mateo, A. Chmura, S. Rustler, F. van Rantwijk,
A. Stolz, R. A. Sheldon, Tetrahedron: Asymmetry
2006, 17, 320-323; b) O. Sosedov, K. Matzer, S.
Bürger, C. Kiziak, S. Baum, J. Altenbuchner, A.
Chmura, F. van Rantwijk, A. Stolz, Adv. Synth. Catal.
2009, 351, 1531-1538; c) A. Chmura, S. Rustler, M.
Paravidino, F. van Rantwijk, A. Stolz, R. A. Sheldon,
Tetrahedron: Asymmetry 2013, 24, 1225-1232; d) S.
Baum, F. van Rantwijk, A. Stolz, Adv. Synth. Catal.
2012, 354, 113-122; e) Z.-J. Zhang, H.-L. Yu, T.
Imanaka, J.-H. Xu, Biochem. Eng. J. 2015, 95, 71-77;
f) Y. C. He, J. H. Xu, Y. Xu, L. M. Ouyang, J. Pan,
Chin. Chem. Lett. 2007, 18, 677-680.
[8] M.-T. Xiao, Y.-Y. Huang, X.-A. Shi, Y.-H. Guo,
Enzyme Microb. Technol. 2005, 37, 589-596.
[9] O. Zhimin, L. Ma, Y. Niu, J. Cui, Biocatal.
Biotransfor. 2018, 36, 409-416.
[10] a) A. Schmid, J. S. Dordick, B. Hauer, A. Kiener, M.
Wubbolts, B. Witholt, Nature 2001, 409, 258-268; b)
S. G. Burton, D. A. Cowan, J. M. Woodley, Nat.
Biotechnol. 2002, 20, 37-45; c) P. A. Santacoloma, G.
Sin, K. V. Gernaey, J. M. Woodley, Org. Process Res.
Dev. 2011, 15, 203-212; d) R. N. Patel, ACS Catal.
2011, 1, 1056-1074; e) E. Ricca, B. Brucher, J. H.
Schrittwieser, Adv. Synth. Catal. 2011, 353, 2239-
2262; f) A. Bruggink, R. Schoevaart, T. Kieboom,
Org. Process Res. Dev. 2003, 7, 622-640; g) U. T.
Bornscheuer, G. W. Huisman, R. J. Kazlauskas, S.
Lutz, J. C. Moore, K. Robins, Nature 2012, 485, 185-
194; h) M. T. Reetz, J. Am. Chem. Soc 2013, 135,
12480-12496; i) J. Dong, E. Fernández-Fueyo, F.
Hollmann, E. Paul Caroline, M. Pesic, S. Schmidt, Y.
Wang, S. Younes, W. Zhang, Angew. Chem., Int. Ed.
2018, 57, 9238-9261; j) J. H. Schrittwieser, S.
Velikogne, M. Hall, W. Kroutil, Chem. Rev. 2018,
118, 270-348; k) M. Winkler, M. Geier, P. Hanlon
Steven, B. Nidetzky, A. Glieder, Angew. Chem., Int.
Acknowledgements
This research was financially supported by Ministry of Education
of Singapore through an AcRF Tier 1 Grant (Project No. 279-
000-477-112) and National Research Foundation (NRF) through
an AME-IRG grant (Project No. 279-000-511-305.
References
[1] a) X.-P. Jiang, T.-T. Lu, C.-H. Liu, X.-M. Ling, M.-Y.
Zhuang, J.-X. Zhang, Y.-W. Zhang, Int. J. Biol.
8
This article is protected by copyright. All rights reserved.

10.1002/adsc.201900373
Advanced Synthesis & Catalysis
Ed. 2018, 57, 13406-13423; l) G. Hughes, J. C. Lewis,
Chem. Rev. 2018, 118, 831-840.
Holtmann, J. Schrader, Appl. Microbiol. Biotechnol.
2009, 83, 849-857.
[11] S. Wu, Y. Zhou, T. Wang, H.-P. Too, D. I. C. Wang,
Z. Li, Nat. Commun. 2016, 7, 11917-11923.
[12] Y. Zhou, S. Wu, Z. Li, Angew. Chem., Int. Ed. 2016,
55, 11647-11650.
[13] a) J. Ni, Y.-T. Wu, F. Tao, Y. Peng, P. Xu, J. Am.
Chem. Soc. 2018, 140, 16001-16005; b) J. Ni, Y. -Y.
Gao, F. Tao, H. -Y. Liu, P. Xu, Angew. Chem., Int.
Ed. 2018, 57, 1214-1217.
[15] B. R. Lukito, S. Wu, H. J. J. Saw, Z. Li,
ChemCatChem 2019, 11, 831-840.
[16] P. Gao, S. Wu, P. Praveen, K.-C. Loh, Z. Li, Appl.
Microbiol. Biotechnol. 2017, 101, 1857-1868.
[17] Y. Zhou, S. Wu, J. Mao, Z. Li, ChemSusChem 2018,
11, 2221-2228.
[18] M. Reifenrath, E. Boles, Metab. Eng. 2018, 45, 246-
254.
[14] a) C. Laane, S. Boeren, K. Vos, C. Veeger,
Biotechnol. Bioeng. 1987, 30, 81-87; b) H. Schewe, D.
[19] D. Ding, Y. Liu, Y. Xu, P. Zheng, H. Li, D. Zhang, J.
Sun, Sci. Rep. 2016, 6, 32208.
9
This article is protected by copyright. All rights reserved.

10.1002/adsc.201900373
Advanced Synthesis & Catalysis
FULL PAPER
Whole Cell-Based Cascade Biotransformation for
the Production of (S)-Mandelic Acid from Styrene,
L-Phenylalanine, Glucose, or Glycerol
Adv. Synth. Catal. Year, Volume, Page – Page
Benedict Ryan Lukito, Balaji Sundara Sekar,
Shuke Wu, and Zhi Li*
10
This article is protected by copyright. All rights reserved.