Small Current but Highly Productive Synthesis of 1,3-Propanediol from Glycerol by an Electrode-Driven Metabolic Shift in Klebsiella pneumoniae L17

Article abstract of DOI:10.1002/cssc.201902928

Electrofermentation actively regulates the bacterial redox state, which is essential for bioconversion and has been highlighted as an effective method for further improvements of the productivity of either reduced or oxidized platform chemicals. 1,3-Propanediol (1,3-PDO) is an industrial value-added chemical that can be produced from glycerol fermentation. The bioconversion of 1,3-PDO from glycerol requires additional reducing energy under anoxic conditions. The cathode-based conversion of glycerol to 1,3-PDO with various electron shuttles (2-hydroxy-1,4-naphthoquinone, neutral red, and hydroquinone) using Klebsiella pneumoniae L17 was investigated. The externally poised potential of ?0.9 V vs. Ag/AgCl to the cathode increased 1,3-PDO (35.5±3.1 mm) production if 100 μm neutral red was used compared with non-bioelectrochemical system fermentation (23.7±2.4 mm). Stoichiometric metabolic flux and transcriptional analysis indicated a shift in the carbon flux toward the glycerol reductive pathway. The homologous overexpression of glycerol dehydratase (DhaB) and 1,3-PDO oxidoreductase (DhaT) enzymes synergistically enhanced 1,3-PDO conversion (39.3±0.8 mm) under cathode-driven fermentation. Interestingly, a small current uptake (0.23 mmol of electrons) caused significant metabolic flux changes with a concomitant increase in 1,3-PDO production. This suggests that both an increase in 1,3-PDO production and regulation of the cellular metabolic pathway are feasible by electrode-driven control in cathodic electrofermentation.

Full text of DOI:10.1002/cssc.201902928

Accepted Article
Title: Small current but large production of 1,3-propanediol from
glycerol by an electrode-driven metabolic shift in Klebsiella
pneumoniae L17
Authors: Changman Kim, Jae Hyeon Lee, Jiyun Baek, Da Seul Kong,
Jeong-Geol Na, Jinwon Lee, Eric Sundstrom, Sunghoon Park,
and Jung Rae Kim
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: ChemSusChem 10.1002/cssc.201902928
Link to VoR: http://dx.doi.org/10.1002/cssc.201902928
A Journal of
www.chemsuschem.org

10.1002/cssc.201902928
ChemSusChem
1
2
Date: December 06 2019
For resubmission to ChemSusChem after revision (Manuscript number: cssc.201902928)
3
4
Small current but large production of 1,3-propanediol from glycerol
5
6
by an electrode-driven metabolic shift in Klebsiella pneumoniae L17
†
7
8
Changman Kim1, , Jae Hyeon Lee1, Jiyun Baek1, Da Seul Kong1, Jeong-Geol Na2,
Jinwon Lee2, Eric Sundstrom3, Sunghoon Park4, Jung Rae Kim1,*
9
1 School of Chemical and Biomolecular Engineering, Pusan National University, Busan, 609-735,
Republic of Korea.
10
11
12
2 Department of Chemical and Biomolecular Engineering, Sogang University, 35 Baekbeom-Ro,
Mapo-Gu, Seoul 04107, Republic of Korea
13
14
3 Advanced Biofuel and Bioproducts Process Development Unit, Lawrence Berkeley National
Laboratory, Emeryville, CA, 94608, USA
15
4 School of Energy and Chemical Engineering, UNIST, Ulsan, 689-798, Republic of Korea
16
17
†Present Address: Advanced Biofuel and Bioproducts Process Development Unit, Lawrence
Berkeley National Laboratory, Emeryville, CA, 94608, USA
18
19
20
Running title: Electro-fermentation of 1,3-PDO
21
22
*Corresponding author
23
24
25
26
27
28
29
Jung Rae Kim, PhD
Address: School of Chemical and Biomolecular Engineering,
Pusan National University, Busan 609-735, Republic of Korea
E-mail address: j.kim@pusan.ac.kr
Phone: +82.51.510.2393
Fax: +82.51.510.3943
This article is protected by copyright. All rights reserved.

10.1002/cssc.201902928
ChemSusChem
Electro-fermentation of 1,3-PDO
30
31
Graphical Abstract
32
2
This article is protected by copyright. All rights reserved.

10.1002/cssc.201902928
ChemSusChem
Electro-fermentation of 1,3-PDO
33
Abstract
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
Electro-fermentation actively regulates the bacterial redox state, which is essential for
bioconversion, and has been highlighted for further improvements of the productivity of
either reduced or oxidized platform chemicals. 1,3-Propanediol (1,3-PDO) is an industrial
value-added chemical that can be produced from glycerol fermentation. The bioconversion
of 1,3-PDO from glycerol requires additional reducing energy under anoxic conditions. This
study examined the cathode electrode-based conversion of glycerol to 1,3-PDO with various
electron shuttles (2-hydroxy-1,4-naphthoquinone, neutral red, and hydroquinone) using
Klebsiella pneumoniae L17. The externally-poised potential of - 0.9 V vs. Ag/AgCl to the
cathode electrode increased 1,3-PDO (35.5 ± 3.1 mM) production when 100 μM of neutral
red was used compared to non-BES fermentation (23.7 ± 2.4 mM). Stoichiometric metabolic
flux and transcriptional analysis indicated a shift in the carbon flux toward the glycerol
reductive pathway. The homologous overexpression of DhaB and DhaT enzymes enhanced
1,3-PDO conversion (39.3 ± 0.8 mM) synergistically under cathode electrode-driven
fermentation. Interestingly, small current uptake (0.23 mmol of electrons) caused significant
metabolic flux changes with a concomitant increase in 1,3-PDO production. This suggests
that both an increase in 1,3-PDO production and regulation of the cellular metabolic pathway
are feasible by electrode-driven control in cathodic electro-fermentation.
Keywords: 1,3-Propanediol; Electrode-driven conversion; Metabolic shift;
Bioelectrochemical system; Electro-fermentation
3
This article is protected by copyright. All rights reserved.

10.1002/cssc.201902928
ChemSusChem
Electro-fermentation of 1,3-PDO
54
1. Introduction
55
56
57
58
59
60
61
62
63
64
Electro-fermentation (EF) has been highlighted as a novel bioconversion strategy
for the production of value-added chemicals and metabolites1-4. The interaction between
microbe and a carbon electrode has been reported to alter the microbial intracellular redox
5, 6
balance, shifting the conversion process towards a more reduced or oxidized state
.
Previous studies have shown that electrode-based regulation triggered a change in the
metabolic (carbon) fluxes, and concurrently produced more reduced/oxidized metabolites,
such as 3-hydroxypropionic acid, L-lysine, and acetoin7-13. Electro-fermentation provides a
potential platform to further improve the titer and yield, particularly for targeted
oxidized/reduced products in biochemical reactions, which are difficult to achieve through
conventional fermentation1, 4.
65
66
67
68
69
70
71
72
73
74
1,3-propanediol (1,3-PDO) is an important C3 platform precursor for further
chemical synthesis, including the production of polytrimethylene terephthalate. Therefore,
the development of 1,3-PDO bio-production from glycerol has been examined previously 14,
15. Biological glycerol conversion to 1,3-PDO consists of two consecutive reactions: glycerol
to 3-hydroxypropionaldehyde (3-HPA) by glycerol dehydratase with coenzyme B12 as a
cofactor, and 3-HPA to 1,3-PDO by 1,3-propanediol oxidoreductase with a reducing
16
equivalent in the form of NADH . Two moles of additional reducing equivalents are
necessary for glycerol-based 1,3-PDO production:
Glycerol (C3O3H8) + 3H2O
1,3-PDO (C3O2H8) + 4H2O
Glycerol (C3O3H8) + 2H+ + 2e-
→
→
→
3CO2 + 14H+ + 14e-
3CO2 + 16H+ + 16e-
1,3-PDO (C3O2H8) + H2O
(1)
(2)
(3)
75
76
4
This article is protected by copyright. All rights reserved.

10.1002/cssc.201902928
ChemSusChem
Electro-fermentation of 1,3-PDO
77
78
79
80
81
82
To resolve the electron imbalance between the substrate (glycerol) and product (1,3-
PDO), and further increase the level of 1,3-PDO productivity, an appropriate pathway for
providing necessary reducing equivalents is essential. Previous studies reported that the
activation of byproduct-formation pathways, such as ethanol, 3-hydroxypropionic acid, and
2,3-butanediol, brought additional NADH synthesis and increased 1,3-PDO production 17-20
.
83
84
85
86
87
88
89
90
91
92
Naturally, some facultative anaerobes, such as Klebsiella, Citrobacter, and
Lactobacillus species, have indigenous 1,3-PDO producing pathways 14, 21, 22. Klebsiella
pneumoniae is one of the best candidates for 1,3-PDO bioproduction, because it possesses
crucial active enzymes for the glycerol reductive pathways, natural coenzyme B12 synthesis,
16
and a well-developed glycerol uptake module . Among the K. pneumoniae strains, K.
13
pneumoniae L17 has the ability to exchange respiratory electrons with the electrode .
Previous studies have reported that the application of an electrical potential through a carbon
electrode in a bioelectrochemical system induced a change in the metabolic flux and gene
expression of L17 12, 13. Such anodic EF increased the productivity of 3-HP 1.8 fold using a
+0.5 V (vs. Ag/AgCl) polarized electrode in a bioelectrochemical system 10.
93
94
95
96
97
98
99
This study evaluated an electrode-based electron supply for glycerol conversion to
1,3-PDO using Klebsiella pneumoniae L17. The current input through the electrode is
believed to alter the level of 1,3-PDO production from glycerol compared to the control with
a non-polarized electrode. Quantitative flux and transcriptional analyses were performed to
determine if the electrode-based electron transference shifts the metabolic state to a more
reduced circumstance in glycerol conversion. To enhance electron transfer between the
bacteria and electrode, three different electron shuttles were examined: 2-hydroxy-1,4-
5
This article is protected by copyright. All rights reserved.

10.1002/cssc.201902928
ChemSusChem
Electro-fermentation of 1,3-PDO
100
101
102
103
104
105
106
107
108
109
110
111
112
113
114
115
116
117
118
119
120
121
napthoquinone, neutral red, and hydroquinone. The recombinant K. pneumoniae L17
overexpressing DhaB and DhaT in the glycerol reductive pathway shows that the
electrochemical interaction can be corroborated by the appropriate strain development. This
is the first K. pneumoniae study to examine glycerol conversion to 1,3-PDO by providing a
reducing equivalent in a bioelectrochemical system, and simultaneously identifying the
change in metabolic flux induced by electrode-driven bioconversion.
2. Materials and methods
2.1. Strains and culture media
Klebsiella pneumoniae L17 (hereinafter L17) was purchased from CCTCC (China
Center for Type Culture Collection) and stored at - 80 °C prior to inoculation. Table S1 lists
the recombinant L17 and plasmids used in this study. Two plasmids harboring glycerol
dehydratase (dhaB) or 1,3-propanediol oxidoreductase (dhaT) were synthesized in the pDK7
or pCU19 vector, respectively 23. The transformations of each plasmid in L17 were
performed using the method reported elsewhere 10. The modified M9 media with the
following composition were used after sterilization by autoclaving or filter-sterilizing: 1 g/L
NaCl, 1 g/L NH4Cl, 0.25 g/L MgSO4∙7H2O, 0.5 g/L yeast extract, 150 mM glycerol, 100 mM
13
potassium phosphate (pH = 8.0), 1.25 ml vitamin solution (10x) , and 10 ml of trace
13
element solution (1.25x) . The medium was adjusted to pH 7.8 using 10 % HCl and 5 N
NaOH solutions. Different concentrations (10 and 100 μM in the final) of electron shuttles
(2-hydroxy-1,4-naphthoquinone, HNQ; neutral red, NR; hydroquinone, HQ) were filter-
sterilized and added to the cathodic chamber of the bioelectrochemical system.
122
123
2.2. Bioelectrochemical system (BES) operation
6
This article is protected by copyright. All rights reserved.

10.1002/cssc.201902928
ChemSusChem
Electro-fermentation of 1,3-PDO
124
125
126
A H-type two-chamber BES reactor was used as described elsewhere13. Both the
anode and cathode chambers (310 ml each) were connected with a glass tube bridge and a
proton exchange membrane (PEM, Nafion 117, Dupont, USA). Carbon cloth (2.5 cm × 5
127
128
129
130
131
132
133
134
135
136
137
138
cm, Nara Cell-Tech Co., Korea) was used as the anode and cathode electrodes. The Ag/AgCl
reference electrode was placed in the cathode chambers. The BES reactor was prewashed
completely with 10 % HCl and 5 N NaOH solutions and sterilized in an autoclave (121 °C,
15 min). After sterilization, the cathode chamber was filled with the modified M9 media,
while in the anode chamber, 50 mM of ferrous sulfate heptahydrate in 100 mM of potassium
phosphate buffer (pH = 8.0) was added (the working volume of both chambers was 250 ml).
Before starting the experiments, K. pneumoniae L17 was pre-cultured overnight in LB
medium, followed by 12 hours of activation in M9 medium. The cathode chamber was
inoculated with the pre-cultured L17 (initial OD600 = 0.07). Ampicillin was added to the
anode and cathode chamber to prevent contamination (100 μM). The BES fermentation
media for the recombinant strain culture contained 0.1 mM of IPTG and 25 μM of kanamycin
and/or chloramphenicol instead of ampicillin.
139
140
141
142
143
144
Both BES and non-BES (control without externally poised potential) reactors were
placed in an incubator (30 °C) on a magnetic stirrer plate (100 rpm). Both anode and cathode
chambers were purged continuously with nitrogen gas (99.9 %) to maintain anoxic
conditions during the experiment. Chronoamperometry (-0.9 V vs. Ag/AgCl) was performed
using a potentiostat (WBCS3000Lee32, WonA Tech, Korea) with an externally poised
potential applied continuously to the cathode electrode.
145
146
2. 3. Analyses
Samples for the analysis of cell growth, pH, and metabolite concentrations were
7
This article is protected by copyright. All rights reserved.

10.1002/cssc.201902928
ChemSusChem
Electro-fermentation of 1,3-PDO
147
148
149
150
151
taken periodically from the cathode chamber at 0, 6, 12, 24, and 36 h. To analyze the
metabolite concentration, the liquid samples were centrifuged (5,000 rpm for 10 min) and
filtered using a syringe filter (0.22 μm, Nylon membrane, Whatman, UK). The filtered
samples were analyzed by high performance liquid chromatography (HPLC, HP 1160 series,
Agilent Technologies, USA) using an Aminex HPX-87H column (300 × 7.8 mm, Bio-Rad,
152
153
USA) at 65 °C. H2SO4 buffer (2.5 mM) was used as the mobile phase (flow rate = 0.5 ml/s),
and a refractive index (RI) and photodiode array (PDA) detector were used.
154
155
156
157
158
2. 4. Metabolic flux model and analysis
The metabolic network model of glycerol conversion in BES was constructed for
in silico metabolic flux analysis (MFA), as shown in Table S2 in the Supplementary
information. The electron uptake pathways by electron transfer from the electrode were
added to the flux model according to the following equation:
159
2 electron-ex + Q + 2H+ → QH2
(4)
160
161
162
Maximization of the cell growth or 1,3-PDO production pathways were the
objective functions for the simulation in BESs and non-BESs. Linear optimization using the
MetaFluxNet program was implemented to solve the metabolic flux equations 13, 24
.
163
164
165
166
167
168
The expression of essential enzymes for glycerol conversion was investigated by
RT-PCR, as reported elsewhere12. mRNA extraction was performed using a total RNA
isolation kit (Macherey-Nagel, Germany); reverse transcription for cDNA synthesis was
conducted using a RevertAid First Strand cDNA synthesis kit (ThermoFisher, USA). The
SYBR green method was used for quantitative real-time PCR on a Step One Real Time PCR
system (Applied Biosystems, USA). The forward and reverse primers are described
8
This article is protected by copyright. All rights reserved.

10.1002/cssc.201902928
ChemSusChem
Electro-fermentation of 1,3-PDO
169
170
171
172
elsewhere12. All experiments were conducted in quadruplicate.
3. Results and discussion
3.1. Current consumption in BESs with different mediators
173
174
175
176
177
178
179
180
181
182
183
184
Fig 1 shows the current consumption profiles under -0.9V vs. Ag/AgCl during
cathodic glycerol conversion. The effects of different electron shuttles (NR and HNQ) with
low (10 μM) and high (100 μM) concentrations were compared in the BESs. With the low
concentration of electron shuttles, the current consumptions were stabilized at approximately
−20, −30, and −110 μA in BES without a mediator (i.e. control) and with HNQ and NR,
respectively (hereinafter called BES, BES-HNQ-10, and BES-NR-10) after 12 hours. In
contrast, relatively higher current consumption was observed at −100 and −200 μA in BESs
with higher concentrations of HNQ and NR (each 100 μM), respectively (hereinafter called
BES-HNQ-100 and BES-NR-100). Electron transfer from the electrode was estimated to be
0.02, 0.05, and 0.11 mmol based on the Faraday constant (96,485 C/mole of e-) in BES, BES-
HNQ-10, and BES-NR-10, respectively, whereas they were much higher in BES-HNQ-100
(0.11 mmol) and BES-NR-100 (0.23 mmol).
9
This article is protected by copyright. All rights reserved.

10.1002/cssc.201902928
ChemSusChem
Electro-fermentation of 1,3-PDO
(a) 10 M
(b) 100 M
0
-100
-200
-300
-400
BES
BES-HNQ
BES-NR
BES-HNQ
BES-NR
0
10
20
30
0
10
20
30
Time (h)
Time (h)
185
186
Fig 1. Current generation profiles of BESs with (a) 10 μM and (b) 100 μM of each electron
shuttle (HNQ: 2-hydroxy-1,4-naphthoquinone; NR: neutral red)
187
188
189
190
191
192
193
194
195
196
197
198
199
200
K. pneumoniae L17 is an exoelectrogen that can transfer/excrete its respiratory
electron to the electrode using mainly an indirect electron transfer module via
electrochemically active shuttle molecules10, 12, 13. This shuttle-based electron transfer was
reported to induce a change in the intracellular redox balance and consequently a metabolic
shift 12, 13. Under anodic glycerol electro-fermentation conditions, L17 produced 80% higher
3-hydroxypropinate, while the respiratory electrons were discharged to the anode 10. On the
other hand, 1,3-PDO production from glycerol is a representative reductive bioconversion;
the productivity of 1,3-PDO is expected to increase when the reducing equivalent is provided
from the electrode by the negatively poised potential on the cathode electrode. According to
a previous study, mediator-based electron uptake from the electrode to the inner membrane
quinone species could be possible in heterotrophic bacteria of K. pneumoniae species when
the appropriate redox electron shuttle was provided 25. Although the mechanism of an
increased intracellular redox level by the reduced quinone species is unclear, it is
10
This article is protected by copyright. All rights reserved.

10.1002/cssc.201902928
ChemSusChem
Electro-fermentation of 1,3-PDO
201
202
203
hypothesized that the reduced quinol species not only block NADH oxidation by the
indigenous respiratory system, but also prompt NAD+ reduction via the cellular NADH
dehydrogenases under these cathodic EF conditions 25.
204
205
206
207
208
209
210
211
212
213
214
215
216
217
218
219
220
The three electron mediators, 2-hydroxy-1,4-naphtoquinone (HNQ, Eo = -0.6 V vs.
SHE), neutral red (NR, Eo = -0.5 V vs. SHE), and hydroquinone (HQ, Eo = -0.7 V vs. SHE),
were compared in BES under -0.9V vs. Ag/AgCl (Fig 1). Although electron transfer from
the bacteria to the electrode was reported to be facilitated by HNQ in an anodic EF in a
previous study13, the significantly lower electron consumption in cathodic EF (i.e. reverse
direction of electron transfer from anodic EF) in Fig 1, suggests that HNQ is less efficient
than NR for intracellular electron transfer. In contrast, NR has been used extensively as an
electron shuttle in exoelectrogens, which interact with menaquinone in the inner membrane
of E. coli during electro-fermentation25. In this study, the significantly enhanced current
consumption was obtained from the NR implemented cathodic EF. HQ has a relatively low
molecular weight (MW 110.11) compared to the other two shuttles (HNQ, MW 174.15; NR,
MW 288.78), and is expected to have advantages for cellular membrane penetration. On the
other hand, no apparent cell growth and glycerol consumption were identified under both
BES and non-BES conditions with HQ, probably due to the toxicity of hydroquinone (See
Supplementary Information Fig. S4).
3.2. Glycerol consumption and 1,3-PDO production in BESs and non-BESs
221
222
223
Fig 2 and Table 1 present the glycerol uptake and simultaneous 1,3-PDO and ethanol
production. Most of the glycerol was consumed in the non-BES system within 12 hours,
whereas more than half of the glycerol remained in most of the BESs. The input current
11
This article is protected by copyright. All rights reserved.

10.1002/cssc.201902928
ChemSusChem
Electro-fermentation of 1,3-PDO
224
225
226
227
228
229
230
231
appeared to decrease glycerol consumption, probably due to the more reduced redox
circumstance induced by the negatively poised potential. The bacterial growth profiles of
different conditions showed a similar trend to glycerol consumption (Fig S2). The growth of
non-BES reached the stationary phase within 12 hours, whereas it was obtained at
approximately 24 hours in BES. With NR, cell growth was initially retarded, but the final
cell concentration was slightly higher than that under the other conditions (Fig S2c). These
results show that additional electron transfer to bacteria may decrease the glycerol uptake
and consequently the cell growth rate in a cathodic EF.
120
(a) WO
(d) NR-10
(c) HNQ-100
(b) HNQ-10
(e) NR-100
100
80
60
40
20
0
BES
non-BES
40
30
20
10
BES
non-BES
0
80
60
40
20
0
BES
non-BES
0
10 20 30
0
10 20 30
0
10 20 30
0
10 20 30
0
10 20 30
Time (h)
Time (h)
Time (h)
Time (h)
Time (h)
232
233
234
235
Fig 2. Metabolite profiles (black-glycerol, blue-1,3-PDO, red-ethanol) in BES and non-
BES (a) without a mediator, (b) with 10 μM HNQ, (c) with 100 μM NR, (d) with 10 μM
12
This article is protected by copyright. All rights reserved.

10.1002/cssc.201902928
ChemSusChem
Electro-fermentation of 1,3-PDO
236
237
NR, and (e) with 100 μM NR. The closed symbols indicate BES while the open symbols
represent non-BES operation.
13
This article is protected by copyright. All rights reserved.

10.1002/cssc.201902928
ChemSusChem
238
239
Table 1. Carbon recovery, glycerol consumption, and metabolite production of KpL17 in BESs and non-BESs
Carbon
recovery
(%)
Glycerol consumption
DCW (g/L)
PDO (mM)
Ethanol (mM)
3-HP (mM)
Lactate (mM) Acetate (mM) Succinate (mM)
(mM)
Without mediator
BES
103.7 ± 3.2
101.5 ± 7.1
0.7 ± 0.1
0.8 ± 0.1
31.1 ± 3.0
21.8 ± 3.2
35.7 ± 3.7
64.7 ± 6.5
11.4 ± 0.9
5.4 ± 0.4
5.6 ± 0.9
2.1 ± 0.3
5.9 ± 0.5
3.9 ± 0.3
4.1 ± 1.3
4.3 ± 1.4
94.3
104.6
non-BES
HNQ-10μM
BES
103.6 ± 5.6
100.7 ± 6.6
0.6 ± 0.0
0.7 ± 0.1
28.7 ± 2.1
24.5 ± 2.1
39.9 ± 5.0
62.8 ± 6.0
8.7 ± 0.7
6.7 ± 0.5
5.6 ± 0.9
3.0 ± 0.5
6.9 ± 0.5
4.9 ± 0.4
4.4 ± 1.4
2.1 ± 0.7
95.5
107.8
non-BES
HNQ-100μM
BES
106.2 ± 7.4
106.5 ± 7.5
0.8 ± 0.1
0.8 ± 0.1
25.9 ± 2.7
22.4 ± 2.2
52.0 ± 5.0
69.0 ± 5.9
11.9 ± 0.8
7.2 ± 0.5
1.8 ± 0.1
6.4 ± 0.5
4.7 ± 0.3
5.2 ± 0.4
4.1 ± 0.3
2.6 ± 0.2
90.5
102.6
non-BES
NR-10μM
BES
104.1 ± 8.0
101.7 ± 7.0
0.8 ± 0.1
0.8 ± 0.1
31.1 ± 3.0
21.1 ± 2.1
36.7 ± 3.1
64.2 ± 6.2
10.1 ± 0.8
5.2 ± 0.4
3.9 ± 0.6
1.5 ± 0.2
4.8 ± 0.4
4.4 ± 0.3
4.3 ± 1.4
4.5 ± 1.4
93.2
103.5
non-BES
NR-100μM
BES
105.5 ± 7.4
102.1 ± 7.2
0.8 ± 0.1
0.8 ± 0.1
35.5 ± 3.1
23.7 ± 2.4
25.1 ± 3.7
67.6 ± 4.8
13.5 ± 0.9
6.6 ± 0.5
3.5 ± 0.2
6.0 ± 0.4
5.6 ± 0.4
5.7 ± 0.4
2.4 ± 0.2
2.1 ± 0.1
91.4
105.5
non-BES
240
241
This article is protected by copyright. All rights reserved.

10.1002/cssc.201902928
ChemSusChem
242
243
244
245
246
247
248
249
250
The glycerol fermentation pathway of K. pneumoniae L17 is well programmed for
the production of reduced metabolites, such as 1,3-PDO and ethanol, under anaerobic
conditions. 1,3-PDO conversion was relatively higher in BESs (31.1 ± 3.0, 28.7 ± 2.1 and
31.1 ± 3.0 mM in BES, BES-HNQ-10 and BES-NR-10, respectively) than in non-BESs (21.8
± 3.2, 24.5 ± 2.1 and 21.1 ± 2.1 mM in non-BES, non-BES-HNQ-10 and non-NR-10,
respectively). In contrast, significantly lower ethanol production was obtained in BESs (35.7
± 3.7, 39.9 ± 4.0 and 36.7 ± 3.1 mM in BES, BES-HNQ-10 and BES-NR-10, respectively)
than those in non-BESs (64.7 ± 6.5, 62.8 ± 6.0 and 64.2 ± 6.2 mM in non-BES, non-BES-
HNQ-10 and non-BES-NR-10, respectively).
251
252
253
254
255
256
257
258
259
260
The effects of the addition of electron shuttles were obscure with 10 μM, probably
due to the low concentration. At higher concentrations (100 μM), however, metabolite
production was different (Fig 2c & e). Interestingly, 1,3-PDO production of BES-HNQ-100
(25.9 ± 2.7 mM) was lower than that of BES-HNQ-10 (28.7 ± 2.1 mM). In contrast, the use
of NR increased the level of 1,3-PDO production (35.5 ± 3.1 mM of BES-NR-100)
significantly compared to BES-NR-10 (31.1 ± 3.0 mM), non-BES-NR-10 (21.1 ± 2.2 mM),
and non-BES-NR-100 (23.7 ± 2.4 mM). Ethanol production was higher in BES-HNQ-100
(52.0 ± 5.0 mM) compared to BES-HNQ-10 (39.9 ± 4.0 mM), but a significant decrease in
ethanol production was observed in BES-NR-100 (25.1 ± 3.7 mM) compared to BES-NR-
10 (36.7 ± 3.1 mM).
261
262
263
264
Both 1,3-PDO and ethanol are reduced metabolites, which are synthesized by
NADH-dependent aldehyde reductases (acetaldehyde reductase and 1,3-propanediol
oxidoreductase for ethanol and 1,3-PDO conversion, respectively). On the other hand, the
net NADH balances for 1,3-PDO and ethanol conversions are -1 and 0, respectively, based
This article is protected by copyright. All rights reserved.

10.1002/cssc.201902928
ChemSusChem
Electro-fermentation of 1,3-PDO
265
266
267
268
269
270
271
272
273
274
275
on the stoichiometric calculation of the glycerol metabolic fluxes (all pyruvate is believed to
be converted to acetyl-CoAby pyruvate formate lyase). Therefore, the 1,3-PDO/ethanol ratio
can reflect the intracellular redox balance; a higher ratio indicates a more reduced condition.
The 1,3-PDO/ethanol ratios from non-BESs were estimated to be 0.32 - 0.39. On the other
hand, a significant increase in the ratio was identified in BES without a mediator (BES, 0.87).
Although the BES-NR-10 (0.82) was similar to the BES without a mediator, the clear
increase in 1,3-PDO/ethanol ratio (1.41) indicates that the intracellular redox state was
shifted to a more reduced state when an appropriate shuttle concentration was provided
(BES-NR-100). In contrast, the addition of HNQ and the poised potential appeared to have
a negative effect on 1,3-PDO production, as illustrated by the decrease in 1,3-PDO/ethanol
ratio (0.72 and 0.50 with BES-HNQ-10 and BES-HNQ-100, respectively).
276
277
278
279
280
281
282
283
284
Cathodic EF also produced several acidic byproducts from glycerol (Fig S3). No
significant differences in acetate, succinate, and lactate were identified between BES and
non-BES. Only 3-HP production was increased under BES conditions (11.4 ± 0.9, 8.7 ± 0.7,
10.1 ± 0.8, 11.9 ± 0.8 and 13.5 ± 0.9 in BES, BES-HNQ-10, BES-NR-10, BES-HNQ-100,
and BES-NR-100, respectively) compared to non-BES conditions (5.4 ± 0.4, 6.7 ± 0.5, 5.2
± 0.4, 7.2 ± 0.5 and 6.6 ± 0.5; non-BES, non-BES-HNQ-10, non-BES-NR-10, non-BES-
HNQ-100, and non-BES-NR-100, respectively)
3.3. In silico metabolic flux analysis
285
286
287
The accumulation of metabolites provides information on the metabolic flux during
glycerol fermentation in both BES and non-BES. This metabolic history can be interpreted
by stoichiometric metabolic flux analysis (MFA) to understand the overall metabolic and
16
This article is protected by copyright. All rights reserved.

10.1002/cssc.201902928
ChemSusChem
Electro-fermentation of 1,3-PDO
288
289
290
291
292
293
294
295
296
297
298
299
carbon flux. In particular, the identification of a balance between the oxidative and reductive
pathway is crucial not only to examine the bacterial metabolism, but also to estimate the
productivity of 1,3-PDO in glycerol conversion, and provide a further strategy for an
improvement of yield and titer. Fig 3 shows the results of MFA under BES and non-BES
conditions. Non-BES could be considered for conventional anaerobic glycerol fermentation
because the electrode does not act as an electron acceptor or donor under open circuit
conditions. The ratios of glycerol oxidative to reductive pathway were estimated to be
approximately 70:30 under non-BES conditions (73:27, 69:31, 74:26, 72:28, and 72:28 in
non-BES, non-BES-HNQ-10, non-BES-NR-10, non-BES-HNQ-100, and non-BES-NR-100,
respectively). In contrast, relatively higher ratios of reductive pathways were obtained under
BES conditions (59:41, 64:36. 61:39, 60:40, and 54:46 in BES, BES-HNQ-10, BES-NR-10,
BES-HNQ-100, and BES-NR-100, respectively).
17
This article is protected by copyright. All rights reserved.

10.1002/cssc.201902928
ChemSusChem
300
This article is protected by copyright. All rights reserved.

10.1002/cssc.201902928
ChemSusChem
Electro-fermentation of 1,3-PDO
301
302
Fig 3. In silico metabolic flux balance analysis of BESs (a) and non-BESs (b) with and without electron mediators
19
This article is protected by copyright. All rights reserved.

10.1002/cssc.201902928
ChemSusChem
303
304
305
306
307
308
309
310
311
312
313
314
315
316
317
318
319
320
321
322
323
324
325
326
Several other metabolites (ethanol, acetate, 3-HP, and succinate) as well as 1,3-PDO
were also produced during anaerobic glycerol fermentation. Among them, the 1,3-PDO is
the only NADH-consuming product in terms of the net balance. In the metabolic flux
analysis of L17 under anaerobic circumstances (including non-BES in this study), the
glycerol oxidative to reductive pathway might be programmed naturally as 70:30 based on
the estimated 1,3-PDO/ethanol ratio of 0.35, as described above. The externally poised
potential shifted the ratio to a more reductive pathway of 60:40 with a higher 1,3-
PDO/ethanol ratio of 0.87 under BES conditions. On the other hand, 6.1 mmol of electron
recovery is required for 11.8 mM of 1,3-PDO production in BES-HNQ-100 (estimated by
NADH consumption) compared to that under non-BES conditions. Nevertheless, it was
estimated that only 0.23 mmol of electrons were actually transferred from the electrode in
BES-NR-100 (even under the assumption that all the electrons were transferred to the
bacteria and converted to NADH) (Fig. 1), which is significantly lower than the amount of
1,3-PDO produced. In addition, the ‘rebalance’ of reducing equivalent under BES condition
(i.e. additional NAD(P)H from the reduced metabolites production such as ethanol and
lactate) may influence 1,3-PDO conversion. A similar electricity-driven metabolic shift was
reported in an electroactive heterotroph Clostridium pasteurianum for 1,3-PDO production
from glycerol7. In this study, a small amount of electron uptake induces the NADH-
consuming pathways over the stoichiometric contribution of the electrons. These results
suggest that a small current uptake might result in a dramatic change in the metabolic flux
under the unnatural conditions of cathodic EF for 1,3-PDO production. The triggering effect
by the poised potential and current in EF requires further study for engineering applications
to enhance the yield and titer of the target products.
This article is protected by copyright. All rights reserved.

10.1002/cssc.201902928
ChemSusChem
Electro-fermentation of 1,3-PDO
327
3.4. Transcriptional analysis of glycerol conversion under cathodic EF
328
329
330
331
332
333
334
335
To determine if the poised potential shifts the transcription of the gene, the
expression levels of the three major glycerol assimilating enzymes, glycerol kinase (GlpK),
glycerol dehydrogenase (DhaD) and glycerol dehydratase (DhaB), were investigated by RT-
PCR (Fig 4). Significantly lower (0.7 fold, dhaD) and higher (4.8 fold, dhaB) expression
levels were analyzed from BES than those of non-BES, respectively, whereas the difference
in glpK was negligible in both conditions. These results show that the electrode-based
reducing equivalent transfer affects the expression of enzymes involved in glycerol
conversion, and activates the reductive pathway rather than the oxidative pathway.
336
337
338
Fig 4. mRNA expression profiles for glycerol-converting enzymes (glpK, dhaD and dhaB).
Both BES and non-BES used NR (100 mM) as an electron shuttle
339
21
This article is protected by copyright. All rights reserved.

10.1002/cssc.201902928
ChemSusChem
Electro-fermentation of 1,3-PDO
340
341
342
343
344
345
346
347
348
349
350
351
352
353
354
355
K. pneumoniae has three glycerol utilization pathways: glp and dhaD module for the
oxidative pathway, and the reductive pathway via dhaB. Under aerobic conditions, glycerol
would be phosphorylated into glycerol-3-phosphate (G-3-P) by glycerol kinase (GlpK). G-
3-P is oxidized further by GlpD (glycerol-3-phosphate dehydrogenase) to synthesize
dehydroxyactone-phosphate (DHAP) for the central carbon assimilation pathways (see
Supplementary Information Fig. S1). Under anaerobic conditions, two different cascade
modules and reactions are required for glycerol oxidation or reduction: glycerol to
dihydroxyacetone (by DhaD, glycerol dehydrogenase), then to dihydroxyacetone phosphate
(by DhaK, dihydroxyacetone kinase). Through both aerobic and anaerobic glycerol
oxidation, one DHAP is produced from one mole of glycerol with equivalent NADH
production and ATP consumption. In contrast, glycerol reduction is conducted in two steps
in the reductive pathway: glycerol to 3-HPA and 3-HPA (by DhaB, glycerol dehydratase) to
1,3-PDO (by DhaT, 1,3-propanediol oxidoreductase). The net reaction of glycerol reduction
is as follows:
Glycerol + NADH
→
1,3-PDO + NAD+
(5)
356
357
358
359
360
361
362
363
GlpR regulates the expression of the glp regulon, which consists of glpK and glpD
16
as a repressor . G-3-P has high affinity to GlpR and the formation of G-3-P and GlpR
complex then renders less repression of glpK and glpD as well as other glp module enzymes
(e.g. glpF). In this study, L17 in both BES and non-BES have much less activity for
extracellular electron transfer; GlpD has difficulty in catalyzing G-3-P to DHAP. Therefore,
the expression of glpK should not change according to externally provided electrons. On the
other oxidative pathway regulator, the dha regulon is controlled by DhaR as an activator 26,
22
This article is protected by copyright. All rights reserved.

10.1002/cssc.201902928
ChemSusChem
Electro-fermentation of 1,3-PDO
364
365
366
367
368
369
370
371
372
373
374
27. One subunit of DhaK (DhaKII::ADP) is an activator of DhaR; another subunit of DhaK
(DhaKI) is an antagonist for DhaKII. DhaD oxidizes glycerol to DHA using NAD+ as a
cofactor. DHA decreases the DhaKI affinity to bind DhaR. In other words, the availability
of NAD+ can determine not only the level of DHA production by DhaD, but also dha regulon
expression by the interruption of DhaKI binding to DhaR. Under BES conditions, additional
reducing energy from the electrode might increase the NADH/NAD+ ratio and decrease the
expression of the dha regulon, including DhaD, as shown in Fig 4. Although the
transcriptional regulation of dhaB of K. pneumoniae has not been identified clearly, the high
level of dhaB expression in BES might indicate that one of the bacterial redox regulators
(probably not an oxygen sensing regulator), would control dhaB expression under cathodic
EF condition 28.
23
This article is protected by copyright. All rights reserved.

10.1002/cssc.201902928
ChemSusChem
Electro-fermentation of 1,3-PDO
120
100
80
60
40
20
0
(c) L17BT
(a) L17B
(b) L17T
BES
non-BES
40
30
20
10
BES
non-BES
0
80
60
40
20
0
BES
non-BES
0
10 20 30 40 50
0
10 20 30 40 50
0
10 20 30 40 50
Time (h)
Time (h)
Time (h)
375
376
377
378
379
380
381
382
Fig 5. Glycerol, PDO, and EtOH metabolites profiles of BES and non-BES using
recombinant L17 strains of L17B (overexpression of DhaB only), L17T (DhaT only) and
L17BT (both DhaB and DhaT). The closed symbols indicate BES while the open symbols
represent the non-BES operation.
3.5. 1,3-PDO production using recombinant L17
24
This article is protected by copyright. All rights reserved.

10.1002/cssc.201902928
ChemSusChem
Electro-fermentation of 1,3-PDO
383
384
385
386
387
388
389
390
391
392
393
To improve the 1,3-PDO productivity further, the glycerol reductive pathway was
overexpressed in L17. Three strains harboring the plasmid for the overexpression of dhaB
and/or dhaT were developed: L17B (overexpression of DhaB only), L17T (DhaT only), and
L17BT (both DhaB and DhaT) (See Supplementary Information Table S1). L17B showed
significantly less glycerol consumption without PDO production under non-BES conditions,
whereas the poised potentials enabled the utilization of glycerol and produced 1,3-PDO
simultaneously under BES conditions (Fig 5a). The L17T showed similar 1,3-PDO
production (34.2 ± 0.8 and 34.0 ± 1.6 mM, respectively) in both BES and non-BES (Fig. 5b).
The improvement of 1,3-PDO production was identified in L17BT in BES (39.3 ± 0.8 mM)
compared to L17BT in non-BES (30.5 ± 0.5 mM) (Fig. 5c), and wild type L17 in the same
BES (35.5 ± 3.1 mM of BES-NR-100) (Fig. 2e).
394
395
396
397
398
399
400
401
402
403
Generally, glycerol dehydration by DhaB is considered a rate-limiting step for PDO
production27; hence, the transcription level of DhaB appears to have increased under BES
conditions, as shown in Fig. 4. The Lac promoter-based homologous overexpression of
DhaB only in the K. pneumoniae strain (including L17B results) usually causes the
accumulation of the toxic intermediate (3-HPA). Accordingly, glycerol fermentation in non-
BES was interrupted significantly, resulting in negligible cell growth 27. In contrast, the same
strain under BES conditions exhibited similar behavior to the wild type and other robust
recombinant strains. This suggests that cathodic electron transfer might lead to the
expression and/or activation of unknown reductases, which are isoenzymes for DhaT, and
concurrently alleviates the accumulation of toxic intermediate under cathodic EF conditions.
404
405
Based on these results, intracellular electron transfer from the electrode may
function as an activator for the expression of the reductive pathway. Although the
25
This article is protected by copyright. All rights reserved.

10.1002/cssc.201902928
ChemSusChem
Electro-fermentation of 1,3-PDO
406
407
408
409
410
411
412
413
414
overexpression of dhaT increased 1,3-PDO production, regardless of BES and non-BES
conditions, as shown in Fig. 5b 29, significantly less ethanol production was also obtained in
non-BES. These results suggest that the poised potential efficiently drives the pathway to a
more reductive direction in both L17T and L17BT compared to L17B. The simultaneous
overexpression of both dhaB and dhaT facilitates L17BT to produce higher 1,3-PDO levels
by providing reducing energy from the poised electrode and the homologous expression of
essential enzymes for glycerol conversion.
3.4. Implications
415
416
Glycerol is a relatively reduced feedstock for bioconversion. Anoxic glycerol
utilization generally increases the intracellular NADH/NAD+ level because two moles of
417
418
419
420
421
422
423
NADH are synthesized to produce one mole of pyruvate. Accordingly, the reductive pathway
is activated in glycerol fermentation (e.g. for the production of 1,3-PDO). The natural
glycerol metabolic flux of Klebsiella sp. might be well programmed for the maintenance of
redox homeostasis by auxiliary reduction/oxidation pathways. Nevertheless, providing an
additional electrode-based reducing equivalent, which was attempted in this study, alters the
homeostasis to drive a more reduced circumstance, and then activates the reductive modules
of glycerol conversion simultaneously.
424
425
426
427
428
Electro-fermentation allows electrode-based microbial redox control for
bioconversion. Recently, the application of EF to various bioprocesses has been highlighted
because the electrode acts as an electron donor/acceptor that can be regulated by electrical
energy. EF is expected to replace the use of conventional oxidizing/reducing reagents, which
not only increases the operating cost, but also accumulates undesirable toxic byproducts. In
26
This article is protected by copyright. All rights reserved.

10.1002/cssc.201902928
ChemSusChem
Electro-fermentation of 1,3-PDO
429
430
431
432
433
434
435
436
particular, the electricity from future energy infrastructure, including renewable energy
systems, can be utilized in the bioprocess through an EF platform. In-situ monitoring and
control by an electrode-based bioprocess are other advantages with the efficient management
of the product yield and titer. On the other hand, one of the main challenges of EF is that
most industrially available strains (e.g. E. coli) do not have sufficient electro-chemical
activity; hence, the intra- and extra- cellular electron transfer rate in those strains is limited.
This drawback might be overcome to some extent using redox shuttles and electrochemically
reversible reagents, such as neutral red and designed shuttle molecules.
437
438
439
440
441
442
443
444
445
446
447
448
449
450
451
452
These results also suggest that the small electron input shifts the entire metabolic
fluxes and simultaneously produces higher yields of 1,3-PDO in the cathodic EF of glycerol.
The coulombic efficiencies of electro-fermentation are generally low (1 to 38%) using
1
fermentable substrates, such as glycerol, glucose, sucrose and lactose .This result can be
compared with another representative cathodic conversion of CO2, microbial
electrosynthesis, with 70-100 % coulombic efficiency in BES, which suggests that most of
the electrons transferred from the cathode are delivered to CO2 reduction by bacteria 30. The
significant difference in electron recovery is due to the different redox levels of the starting
feedstock (glycerol vs. CO2). In the glycerol EF, electron transfer by the electrode affects
the intracellular redox balance, and triggers a rebalance of the metabolic flux by
expression/depression of the involved enzymes. This metabolic ‘butterfly effect’ could be
applied to various bioconversions for the reduction/oxidation of feedstock using not only the
electro-chemically active strain in EF, but also conventional fermentation using a range of
industrially available strains.
4. Conclusion
27
This article is protected by copyright. All rights reserved.

10.1002/cssc.201902928
ChemSusChem
Electro-fermentation of 1,3-PDO
453
454
455
456
457
458
459
460
461
462
463
464
465
466
467
468
469
The cathodic electro-fermentation of glycerol to 1,3-PDO was examined with
different electron shuttles, HNQ, NR, and HQ using a Klebsiella pneumoniae L17. The
externally poised potential of - 0.9 V vs. Ag/AgCl clearly increased the level of 1,3-PDO
conversion 1.5 fold (6.1 mmol increase) with 100 μM of neutral red compared to the control,
even with a small amount of electron transfer (0.23 mmol). Stoichiometric and transcription
analyses showed that supplying reducing energy by the electrode shifts the glycerol
metabolic flux to the reductive pathway. The recombinant strain of L17 with the homologous
overexpression of DhaB and DhaT enzymes enhanced 1,3-PDO production further (39.3 ±
0.8 mM) under EF conditions. This suggests that even a small amount of electron transfer
can induce an intracellular metabolic shift, leading to a significant improvement in
productivity.
Acknowledgements
This study was supported by the Mid-Career Researcher Program (NRF-
2018R1A2B6005460) and C1 Gas Refinery Programs (NRF-2018M3D3A1A01055756 &
NRF-2015M3D3A1A01064929) through the National Research Foundation of Korea (NRF)
funded by the Ministry of Science, ICT & Future Planning, Korea.
28
This article is protected by copyright. All rights reserved.

10.1002/cssc.201902928
ChemSusChem
Electro-fermentation of 1,3-PDO
470
References
471
472
473
474
475
476
477
478
479
480
481
482
483
484
485
486
487
488
489
490
491
492
493
494
495
496
497
498
499
500
501
502
503
504
505
506
507
508
509
[1] R. Moscoviz, J. Toledo-Alarcón, E. Trably, N. Bernet Trends in biotechnology. 2016, 34, 856-865.
[2] R. Moscoviz, E. Trably, N. Bernet Microbial Biotechnology. 2018, 11, 74-83.
[3] K. Rabaey, R. A. Rozendal Nature Reviews Microbiology. 2010, 8, 706.
[4] A. Schievano, T. P. Sciarria, K. Vanbroekhoven, H. De Wever, S. Puig, S. J. Andersen, K. Rabaey, D. Pant Trends in
biotechnology. 2016, 34, 866-878.
[5] M. Ishikawa, Y. Tanaka, R. Suzuki, K. Kimura, K. Tanaka, K. Kamiya, H. Ito, S. Kato, T. Kamachi, K. Hori Bioresource
Technology. 2017, 241, 1157-1161.
[6] V. J. Watson, B. E. Logan Biotechnology & Bioengineering. 2010, 105, 489-498.
[7] O. Choi, T. Kim, H. M. Woo, Y. Um Scientific Reports. 2014, 4, 6961.
[8] T. Bursac, J. A. Gralnick, J. Gescher Biotechnology and Bioengineering. 2017, 114, 1283-1289.
[9] A. H. Förster, S. Beblawy, F. Golitsch, J. Gescher Biotechnology for Biofuels. 2017, 10, 65.
[10] C. Kim, M. Y. Kim, I. Michie, B.-H. Jeon, G. C. Premier, S. Park, J. R. Kim biotechnology for Biofuels. 2017, 10, 199.
[11] I. Vassilev, G. Gießelmann, S. K. Schwechheimer, C. Wittmann, B. Virdis, J. O. Krömer Biotechnology and
Bioengineering. 2018, 115, 1499-1508.
[12] M. Y. Kim, C. Kim, S. K. Ainala, H. Bae, B.-H. Jeon, S. Park, J. R. Kim Bioelectrochemistry. 2019, 125, 1-7.
[13] C. Kim, S. K. Ainala, Y.-K. Oh, B.-H. Jeon, S. Park, J. R. Kim Biotechnology & Bioprocess Engineering. 2016, 21,
250-260.
[14] C. Lee, M. Aroua, W. Daud, P. Cognet, Y. Pérès-Lucchese, P. Fabre, O. Reynes, L. Latapie Renewable & Sustainable
Energy Reviews. 2015, 42, 963-972.
[15] S. Lama, E. Seol, S. Park Bioresource Technology. 2017, 245, 1542-1550.
[16] V. Kumar, S. Park Biotechnology Advances. 2018, 36, 150-167.
[17] V. Kumar, M. Sankaranarayanan, M. Durgapal, S. Zhou, Y. Ko, S. Ashok, R. Sarkar, S. Park Bioresource Technology.
2013, 135, 555-563.
[18] V. Kumar, M. Sankaranarayanan, K.-e. Jae, M. Durgapal, S. Ashok, Y. Ko, R. Sarkar, S. Park Applied Microbiology &
Biotechnology. 2012, 96, 373-383.
[19] S. Ashok, S. M. Raj, C. Rathnasingh, S. Park Applied Microbiology & Biotechnology. 2011, 90, 1253-1265.
[20] M. Metsoviti, K. Paraskevaidi, A. Koutinas, A.-P. Zeng, S. Papanikolaou Process Biochemistry. 2012, 47, 1872-1882.
[21] R. Saxena, P. Anand, S. Saran, J. Isar Biotechnology Advances. 2009, 27, 895-913.
[22] S. K. Ainala, S. Ashok, Y. Ko, S. Park Applied Microbiology & Biotechnology. 2013, 97, 5001-5011.
[23] Y. Ko, S. Ashok, S. Zhou, V. Kumar, S. Park Process biochemistry. 2012, 47, 1135-1143.
[24] D.-Y. Lee, H. Yun, S. Park, S. Y. Lee Bioinformatics. 2003, 19, 2144-2146.
[25] T. D. Harrington, A. Mohamed, V. N. Tran, S. Biria, M. Gargouri, J.-J. Park, D. R. Gang, H. Beyenal Bioresource
Technology. 2015, 195, 57-65.
[26] D. Wei, M. Wang, B. Jiang, J. Shi, J. Hao Journal of Biotechnology. 2014, 177, 13-19.
[27] P. Zheng, K. Wereath, J. Sun, J. van den Heuvel, A.-P. Zeng Process Biochemistry. 2006, 41, 2160-2169.
[28] Y. Huang, Z. Li, Q. Ye Applied Biochemistry & Biotechnology. 2016, 178, 1129-1140.
[29] Z. Chen, H. Liu, D. Liu Enzyme and Microbial Technology. 2009, 45, 305-309.
[30] C. H. Im, Y. E. Song, B.-H. Jeon, J. R. Kim Carbon Letters. 2016, 20.
510
29
This article is protected by copyright. All rights reserved.