Production of C3 platform chemicals from CO2 by genetically engineered cyanobacteria

Article abstract of DOI:10.1039/c5gc00129c

Platform chemicals can be readily converted into various value-added chemicals and fuels. Photosynthetic production of platform chemicals directly from CO₂ by cyanobacteria, in the presence of sunlight, holds promise for addressing global energy and environmental concerns. Herein, we report the photosynthetic production of C3 platform chemicals using engineered Synechococcus elongatus PCC7942 as the kernel. The engineered S. elongatus strain YW1 expressing glycerol-3-phosphatase produced a C3 intermediate, glycerol, with a high concentration of 1.17 g L^-1 and a maximum production rate of 7733 μg L^-1 H^-1. Strain YW1 could serve as the kernel for the production of various C3 chemicals. By extending heterologous pathways in the cyanobacterial kernel, the carbon flux was further channelled to produce two platform chemicals: dihydroxyacetone by introducing glycerol dehydrogenase and 3-hydroxypropionic acid by introducing glycerol dehydratase and aldehyde dehydrogenase. Co-cultivation of the cyanobacterial kernel and another microbe, Klebsiella pneumoniae, was also performed to convert the C3 intermediate produced from CO₂ to 1,3-propanediol, an important monomer for biodegradable material production. Besides direct photosynthetic production and co-cultivation, we demonstrated that glycerol produced by the cyanobacterial kernel can be used as a fermentation feedstock after simple concentration. The production processes presented here display great potential for carbon capture and storage and for sustainable production of chemicals and fuels.

Full text of DOI:10.1039/c5gc00129c

Green Chemistry
View Article Online
View Journal
PAPER
Production of C3 platform chemicals from CO2
by genetically engineered cyanobacteria†
Cite this: DOI: 10.1039/c5gc00129c
Yu Wang,‡a,b Fei Tao,‡a,b Jun Ni,a,b Chao Lia,b and Ping Xu*a,b,c
Platform chemicals can be readily converted into various value-added chemicals and fuels. Photosyn-
thetic production of platform chemicals directly from CO2 by cyanobacteria, in the presence of sunlight,
holds promise for addressing global energy and environmental concerns. Herein, we report the photosyn-
thetic production of C3 platform chemicals using engineered Synechococcus elongatus PCC7942 as the
kernel. The engineered S. elongatus strain YW1 expressing glycerol-3-phosphatase produced a C3 inter-
mediate, glycerol, with a high concentration of 1.17 g L−1 and a maximum production rate of 7733 μg L−1
H−1. Strain YW1 could serve as the kernel for the production of various C3 chemicals. By extending
heterologous pathways in the cyanobacterial kernel, the carbon flux was further channelled to produce two
platform chemicals: dihydroxyacetone by introducing glycerol dehydrogenase and 3-hydroxypropionic
acid by introducing glycerol dehydratase and aldehyde dehydrogenase. Co-cultivation of the cyanobac-
terial kernel and another microbe, Klebsiella pneumoniae, was also performed to convert the C3 inter-
mediate produced from CO2 to 1,3-propanediol, an important monomer for biodegradable material
production. Besides direct photosynthetic production and co-cultivation, we demonstrated that glycerol
produced by the cyanobacterial kernel can be used as a fermentation feedstock after simple concen-
tration. The production processes presented here display great potential for carbon capture and storage
and for sustainable production of chemicals and fuels.
Received 20th January 2015,
Accepted 23rd March 2015
DOI: 10.1039/c5gc00129c
www.rsc.org/greenchem
chemicals and fuels,7–19 such as lactate, 1,2-propanediol, iso-
butyraldehyde, and butanol. However, cyanobacteria are more
Introduction
Increasing petroleum consumption and growing environ- difficult to engineer than genetically tractable hosts such as
mental concerns necessitate the sustainable production of Escherichia coli and Saccharomyces cerevisiae.20 In addition, het-
chemicals and fuels from renewable resources.1 Extensive erologous pathways do not always work efficiently in cyanobac-
efforts have been invested in the screening of novel enzymes, teria due to the enzyme oxygen sensitivity and cofactor
development of new pathways, and metabolic engineering of bias.19,21 Therefore, it is wise to construct engineered cyano-
microorganisms for the production of chemicals and fuels.2–4 bacteria to produce intermediates, and convert them into a
Direct conversion of CO2 to valuable products by photoauto- broad array of valuable products through well-established bio-
trophic organisms could potentially reduce CO2 emissions logical and chemical routes. This strategy could potentially
from fossil fuel combustion, and avoid costly harvesting, pro- bypass additional difficult and time-consuming genetic
cessing, and decomposition of recalcitrant biomass.5,6 Cyano- manipulation in cyanobacteria, and take full advantage of the
bacteria have been engineered for the production of several well-characterized biosynthetic pathways and microbial produ-
cers, especially when the heterologous pathways are incompati-
ble with cyanobacterial metabolism or the products are highly
aState Key Laboratory of Microbial Metabolism, and School of Life Sciences &
toxic to cyanobacteria. The engineered cyanobacterium that
Biotechnology, Shanghai Jiao Tong University, Shanghai 200240, People’s Republic
produces intermediates could be considered as a kernel, ready
to be assimilated into highly optimized bio-fermentation and
chemical manufacturing processes.
To select a good intermediate for the cyanobacterial kernel,
the following criteria should be fulfilled. First, branchable
intermediates, which could serve as versatile building blocks
for the synthesis of useful products, are preferred products for
of China. E-mail: pingxu@sjtu.edu.cn
bJoint International Research Laboratory of Metabolic & Developmental Sciences,
Shanghai Jiao Tong University, Shanghai 200240, People’s Republic of China
cShanghai Collaborative Innovation Center for Biomanufacturing, East China
University of Science and Technology, Shanghai 200237, People’s Republic of China
†Electronic supplementary information (ESI) available. See DOI: 10.1039/
c5gc00129c
‡These authors contributed equally to this work.
This journal is © The Royal Society of Chemistry 2015
Green Chem.

View Article Online
Paper
Green Chemistry
Fig. 1 Schematic illustration of the production of C3 platform chemicals from CO2 with a genetically engineered cyanobacterium as the kernel. The
cyanobacterial kernel that produced the C3 intermediate, glycerol, from CO2 and sunlight is shaded in green. The kernel can be applied to produce
various chemicals and fuels through extending pathways in cyanobacteria (green box), co-cultivation of cyanobacteria and glycerol utilizers (purple
box), glycerol fermentation (blue box), or chemical synthesis (orange box). The major products studied in this work, namely, glycerol, DHA, 3-HP,
and 1,3-PD, are in blue font. F6P, fructose-6-phosphate; GAP, glyceraldehyde-3-phosphate; DHAP, dihydroxyacetone phosphate; G3P, glycerol-3-
phosphate; DHA, dihydroxyacetone; 3-HPA, 3-hydroxypropionaldehyde; 3-HP, 3-hydroxypropionic acid; 1,3-PD, 1,3-propanediol; G3PDH, glycerol-
3-phosphate dehydrogenase; GPP, glycerol-3-phosphatase; GDH, glycerol dehydrogenase; GDHt, glycerol dehydratase; ALDH, aldehyde
dehydrogenase.
the cyanobacterial kernel. Second, high-titer production of the form chemicals including dihydroxyacetone (DHA), 3-hydroxy-
target intermediate is required for the construction of an econo- propionic acid (3-HP), and 1,3-propanediol (1,3-PD) by
mically competitive production process. Third, the biosyn- extending heterologous pathways in S. elongatus, co-culturing
thetic pathway should minimize the loss of fixed carbon, for S. elongatus with another microbe, or fermentation (Fig. 1).
the chemical production is fundamentally limited by carbon
conservation.22 C3 platform chemicals, such as glycerol and
lactate, possess great potential for use as building blocks for
Results
Construction of the cyanobacterial kernel for C3 intermediate
production
the synthesis of various products including chemicals, fuels,
and materials.23–25 In the active Calvin cycle of cyanobacteria,
there are several precursors for C3 chemical synthesis.19 There- Production of valuable C3 chemicals from CO2 is premised on
fore, photosynthetic production of C3 platform chemicals the construction of a cyanobacterial kernel that can produce a
could withdraw carbon close to fixation, which is beneficial for C3 intermediate efficiently from CO2. Glycerol can be syn-
maximizing the pool of available carbon, and achieving strong thesized via the glycerol-3-phosphate (G3P) pathway in micro-
production rates.22 In addition, without decarboxylation steps, organisms. The glycolytic intermediate dihydroxyacetone
perfect conservation of fixed carbon could be obtained in the phosphate (DHAP) is converted to glycerol in two steps that are
photosynthetic production of C3 platform chemicals.
catalyzed by G3P dehydrogenase (G3PDH) and glycerol-3-phos-
In this study, a C3 intermediate with a highly functiona- phatase (GPP), respectively.26 Glycerol is not a natural product
lized nature, glycerol, was chosen as the target intermediate of S. elongatus PCC7942; however, the active Calvin cycle in
for the construction of the cyanobacterial kernel and the pro- S. elongatus provides sufficient precursors for glycerol synthesis.
duction of value-added chemicals from CO2. First, the glycerol Glycerol synthesis can start from the Calvin cycle intermediate
biosynthetic pathway was developed in Synechococcus elongatus DHAP. Introduction of the exogenous glycerol biosynthetic
PCC7942 (ATCC 33912), obtaining a cyanobacterial kernel that pathway can potentially increase the carbon flux output from
could efficiently produce glycerol from CO2. The cyanobacterial the Calvin cycle, which may be beneficial for increasing the
kernel was then applied to produce several important C3 plat- photosynthesis efficiency.
Green Chem.
This journal is © The Royal Society of Chemistry 2015

View Article Online
Green Chemistry
Paper
Fig. 2 Glycerol production by the engineered cyanobacterial kernel. (A) Schematic representation of gpp1 gene integration into NSI of S. elongatus
PCC7942. gpp1, encoding gene of glycerol-3-phosphatase from S. cerevisiae. (B) Effects of IPTG concentration on the cell growth and glycerol pro-
duction of strain YW1. (C) Cell density of S. elongatus wild-type (WT) and YW1 strains incubated in stationary culture, with bubbling ordinary air or
bubbling air containing 5% CO2. (D) Glycerol accumulation of S. elongatus WT and YW1 strains incubated in stationary culture, with bubbling ordi-
nary air or bubbling air containing 5% CO2. (E) Total biomass production per day of S. elongatus WT (open circle) and YW1 (closed square) strains
incubated with bubbling ordinary air. Cell biomass was determined as cell dry weight. Total biomass was cell biomass plus glycerol produced. Error
bars indicate SD (n = 3).
S. elongatus PCC7942 has a native G3PDH. To construct the trations of IPTG were added to the YW1 cultures. By the
glycerol biosynthesis pathway, the introduction of a GPP was addition of IPTG, cell growth was hindered, whereas glycerol
required (Fig. 1). Therefore, gpp1, which encodes GPP1 from synthesis was enhanced (Fig. 2B). As the concentration of IPTG
S. cerevisiae,27 was cloned into plasmid pAM-YW1 using an iso- rose from 0.2 to 1.0 mM, glycerol synthesis increased. The
propyl β-D-1-thiogalactopyranoside (IPTG) inducible promoter, addition of 2.0 mM IPTG slightly reduced glycerol production,
Ptrc. Plasmid pAM-YW1 was designed for recombination at which was reasonable, because high expression of hetero-
Neutral Site I (NSI) of the S. elongatus genome (Fig. 2A). Gene logous genes might cause decreased product synthesis.28 Gly-
insertion was verified by whole cell colony polymerase chain cerol was synthesized in a small amount even without
reaction (PCR) of the transformants (Fig. S1†). The resulting induction, indicating the leaky expression of gpp1. Leaky
recombinant strain YW1 was cultured in the BG-11 medium expression of IPTG-inducible promoters, such as Ptrc, has been
for glycerol production.
previously reported.29 The optimal IPTG concentration
(1.0 mM) was used in subsequent experiments.
C3 intermediate production from CO2
Next, we tested the glycerol production by strain YW1 cul-
To identify the optimal IPTG concentration required for the tured stationarily with constant light exposure. The wild-type
synthesis of the C3 intermediate, glycerol, various concen- strain (WT) was used as a negative control. As shown in Fig. 2C
This journal is © The Royal Society of Chemistry 2015
Green Chem.

View Article Online
Paper
Green Chemistry
and Fig. 2D, glycerol production was sustained for 20 days,
and the final titer reached 500.8 mg L−1. No glycerol was
detected in the culture of wild-type strain. It has been reported
that glycerol production in green algae was significantly
enhanced by high CO2 concentrations.30 To increase the CO2
supply for glycerol biosynthesis, strain YW1 was cultured with
bubbling ordinary air. As expected, an increased glycerol pro-
duction rate was achieved, especially in the first 10 days of cul-
turing. Finally, a high concentration of 1.17 g L−1 glycerol with
a maximum production rate of 7733 μg L−1 H−1 (2 days
average) was obtained in 20 days (Fig. 2D and Table S1†). To
further increase the CO2 supply, air containing 5% CO2 was
used. Similarly, the glycerol production rate obtained by using
air containing 5% CO2 decreased obviously after 10 days of
culturing (Fig. 2D and Table S1†). Glycerol production lasted
for 20 days and the final titer was only slightly improved to
1.24 g L−1. Therefore, bubbling ordinary air is more commer-
cially viable.
Glycerol accumulation suggests that the carbon flux of
strain YW1 was successfully redirected to target product for-
mation. Considering that YW1 cultures showed a decrease of
approximately 30% in cell density compared with the wild-type
strain (Fig. 2C), the fixed carbon for glycerol production might
be diverted from cell growth processes. Therefore, we moni-
tored the production of cell biomass (dry weight) and glycerol
of wild-type and YW1 strains cultured with bubbling ordinary
air. Strain YW1 accumulated less cell biomass than the wild-
type strain, whereas the production of glycerol biomass in
strain YW1 resulted in increased total biomass production.
The glycerol produced by strain YW1 amounted to 21.3% of its
total biomass (Fig. 2E and Fig. S2†). These results suggest that
glycerol production could increase the carbon flux output from
the central metabolism and stimulate a positive effect on the
efficiency of CO2 utilization. The high-yield glycerol biosyn-
thesis suggests that strain YW1 is a potential glycerol producer
and could act as a good cyanobacterial kernel for the pro-
duction of C3 chemicals.
Fig. 3 DHA production by extending pathways in the cyanobacterial
kernel. (A) Schematic representation of gpp1 and dhaD gene integration
into the NSI of S. elongatus PCC7942. gpp1, encoding gene of glycerol-
3-phosphatase from S. cerevisiae; dhaD, encoding gene of glycerol
dehydrogenase from K. pneumoniae. (B) DHA and glycerol accumulation
of S. elongatus strain YW2 incubated with bubbling ordinary air. Error
bars indicate SD (n = 3).
The final titer of DHA reached 78.6 mg L−1 in 16 days (Fig. 3B).
Meanwhile, over 700 mg L−1 of glycerol accumulated in the
culture broth of strain YW2, suggesting that only part of the
carbon flux was channelled to DHA. Considering that the
NADPH/NADP+ pool was much larger than the NADH/NAD+
pool in S. elongatus PCC7942,35 the NAD+-dependent oxidation
of glycerol catalyzed by dhaD might be the limiting step in the
DHA production pathway. To further optimize DHA production
in engineered cyanobacteria, it could be beneficial to find the
NADP+-dependent counterpart of dhaD.
Extending pathways in the cyanobacterial kernel for DHA
production
DHA is frequently used in the organic synthesis and cosmetic
industries.31 The acetic acid bacterium Gluconobacter oxydans
is a good DHA producer and oxidizes glycerol to DHA using
Extending pathways in the cyanobacterial kernel for 3-HP
production
the membrane-bound pyrroloquinoline quinone-dependent 3-HP is a platform chemical that is ranked in the top three
glycerol dehydrogenase (GDH).32 However, this GDH contains among the twelve industrial building blocks selected by the US
two subunits (SldA and SldB), and its membrane-bound Department of Energy.36 It has recently received significant
feature may cause problems for heterologous expression.33 The attention due to its numerous applications.37 Several biosyn-
NAD+-dependent GDH from Klebsiella pneumoniae encoded by thetic pathways have been suggested for 3-HP production from
dhaD was heterologously expressed and well characterized.34 biomass,38 and a number of recombinant strains have been
Therefore, the dhaD gene was selected for introduction into developed.39–41 However, to the best of our knowledge, no
S. elongatus YW1 to produce DHA from CO2 (Fig. 3A). Whole- reports have thus far described 3-HP production by engineered
cell colony PCR of the transformants was used for verification autotrophic organisms. 3-HP production from glycerol consists
of gene insertion (Fig. S1†).The recombinant strain harbouring of two reactions: (i) the dehydration of glycerol to 3-hydroxy-
both gpp1 and dhaD genes was designated as YW2.
propionaldehyde (3-HPA), catalyzed by glycerol dehydratase
By developing the DHA synthetic pathway in the cyanobac- (GDHt); and (ii) the oxidation of 3-HPA to 3-HP, catalyzed by
terial kernel, direct DHA production from CO2 was obtained. aldehyde dehydrogenase (ALDH).38 Based on the cyanobacterial
Green Chem.
This journal is © The Royal Society of Chemistry 2015

View Article Online
Green Chemistry
Paper
kernel (strain YW1) that produces glycerol from CO2, introduc- the S. elongatus chromosome (Fig. 4A). Colony PCR of transfor-
tion of a GDHt and an ALDH was required for 3-HP biosyn- mants was then performed to confirm insertion of the genes
thesis from CO2.38
(Fig. S1†). The resulting strains were designated as YW3 and
The coenzyme B12-independent GDHt from Clostridium YW4, respectively. To check whether all the heterologous genes
butyricum (encoded by dhaB(CB)) and the coenzyme B12-depen- were successfully introduced and transcribed, quantitative
dent GDHt from K. pneumoniae (encoded by dhaB(KP))42 were reverse-transcription PCR was performed. The housekeeping
both tested, since their performance in cyanobacteria has not gene rnpB was used as a control,19 and no-reverse transcriptase
been characterized. Among the various ALDHs, an NAD(P)+- controls were also carried out to confirm that genomic DNA
dependent ALDH from E. coli (encoded by puuC) was selected contamination during RNA extraction was minimal. As shown
for its high activity toward 3-HPA and its ability to utilize both in Fig. 4B, gpp1, dhaB(CB), dhaB(KP), and puuC were success-
NADP+ and NAD+.43 To introduce the 3-HP biosynthesis fully transcribed under the induction of IPTG. Different DNA
pathway into S. elongatus, dhaB and puuC genes were cloned transcription levels of the heterologous genes were observed,
into an artificial operon driven by Ptrc. Plasmids pAM-YW3 and which may have been caused by different arrangements and
pAM-YW4 were constructed to insert the operon and gpp1 into sources of the genes.44
Fig. 4 3-HP production by extending pathways in the cyanobacterial kernel. (A) Schematic representation of gpp1, dhaB, and puuC gene integration
into the NSI of S. elongatus PCC7942. gpp1, encoding gene of glycerol-3-phosphatase from S. cerevisiae; dhaB(CB), encoding gene of B12-indepen-
dent glycerol dehydratase from C. butyricum; dhaB(KP), encoding gene of B12-dependent glycerol dehydratase from K. pneumoniae; puuC, encod-
ing gene of NAD(P)+-dependent ALDH from E. coli. (B) Quantitative reverse-transcription PCR analysis of relative transcriptional levels of foreign
genes in strains YW3 and YW4. (C) Comparison of maximum 3-HP production of strain YW3 under combined conditions. Strain YW3 was incubated
in test tubes under various conditions for 10 days. Dark conditions indicate that the test tubes were wrapped with aluminum foil, while light con-
ditions indicate that the test tubes were not wrapped and were cultured with an illumination intensity of 100 μE s−1 m−2. Anaerobic conditions indi-
cate that the headspace of test tubes was purged with sterilized ultra-pure nitrogen gas several times, while aerobic conditions indicate that the
headspace was not purged and the atmospheric oxygen was retained. N/P source (−) indicates that NaNO3 and K2HPO4 were not added to the
BG-11 medium. Error bars indicate SD (n = 3).
This journal is © The Royal Society of Chemistry 2015
Green Chem.

View Article Online
Paper
Green Chemistry
Optimization of 3-HP production conditions
3-HP titer obtained was 31.7 mg L−1 using dark, anaerobic and
nitrate/phosphate limited conditions in 10 days. The 3-HP
secreted by strain YW3 was confirmed by liquid chromato-
graphy-mass spectrometry (LC-MS) (Fig. S4†).
3-HP production by strains YW3 and YW4 was then performed
under standard oxygenic incubation with constant light
exposure. Strain YW3 exhibited a 50% decrease in cell density
relative to that of the wild-type strain. Unexpectedly, strain
YW4 stopped growing after induction (Fig. S3†). The C3 inter-
mediate, glycerol, accumulated in the cultures of YW3 and
YW4, whereas no 3-HP was detected. (Fig. S3†). We speculated
that the 3-HP titer might be below the detection limit of the
UV/Vis detector of HPLC (approximately 5 mg L−1). Consider-
ing the accumulation of glycerol, the dehydration of glycerol
catalyzed by the oxygen-sensitive GDHt might be a key limiting
factor. It has been suggested that oxygen could cause partial or
even complete loss of product synthesis in engineered strains
containing oxygen-sensitive enzymes, and dark, anaerobic,
and nitrate/phosphate limited conditions may aid in the
accumulation of products in this situation.13,45 To bypass the
growth inhibition of strain YW4, strain YW3 was selected for
further studies.
First, strain YW3 was incubated under standard aerobic
and light conditions until the OD730 nm reached 1.0. Then the
culture was concentrated to an OD730 nm of 5.0 to increase the
potential 3-HP titer, and switched to different conditions as
shown in Fig. 4C with 1.0 mM IPTG. As expected, oxygen
severely hindered 3-HP production because aerobic conditions
caused the complete loss of 3-HP biosynthesis in strain YW3
cultured in the BG-11 medium. However, about 10 mg L−1
3-HP accumulated under light and anaerobic conditions,
suggesting that 3-HP production in strain YW3 is oxygen-sensi-
tive. A higher 3-HP titer (approximately 20 mg L−1) was
achieved under dark and anaerobic conditions. Strain YW3
incubated in nitrate/phosphate-limited medium produced
more 3-HP (Fig. 4C). The removal of nitrogen and phosphate
enabled 3-HP production even under dark and aerobic con-
ditions, which was observed in a previous study.13 The highest
Co-culturing the cyanobacterial kernel with another microbe
for 1,3-PD production
1,3-PD is another important C3 platform chemical due to its
use as a monomer for the production of polyesters, polyethers,
and polyurethanes, such as polytrimethylene terephthalate
(PTT).42 Synthesis of 1,3-PD also requires the dehydration of
glycerol catalyzed by the oxygen-sensitive GDHt, which is
similar to the conversion of glycerol to 3-HP.43 Co-cultivation
of a native 1,3-PD producer, such as K. pneumoniae, and the
glycerol-producing cyanobacterial kernel (strain YW1) could
bypass the expression of oxygen-sensitive pathways in S. elon-
gatus and facilitate 1,3-PD production from CO2. To maintain a
suitable pH range for the growth of K. pneumoniae, HEPES
(10 mM, pH 7.5), which slightly affected the growth and gly-
cerol production of strain YW1 (Fig. S5†), was added to the
BG-11 medium. When co-cultured with the cyanobacterial
kernel, K. pneumoniae was able to grow without the addition of
an external carbon source, and produce approximately 40 mg
L−1 1,3-PD in 5 days (Fig. 5A and Fig. S6†). A number of micro-
organisms are able to grow on glycerol as the sole carbon and
energy source, including several genetically tractable hosts,
such as E. coli and S. cerevisiae.31 Therefore, the co-cultivation
strategy could be a promising route for the production of
chemicals and fuels from CO2.
Fermentation using the product of the cyanobacterial kernel
In addition to co-cultivation, the glycerol produced by strain
YW1 was also concentrated to about 10 g L−1 and used as the
feedstock for 1,3-PD fermentation by K. pneumoniae. Com-
pared with pure glycerol, the glycerol produced by the cyano-
Fig. 5 Biosynthesis of 1,3-PD from glycerol produced by the engineered cyanobacterial kernel. (A) Cell growth and 1,3-PD production of K. pneu-
moniae supported by the glycerol produced by strain YW1 in co-culture. K. pneumoniae was diluted to obtain a concentration of approximately 106
cells mL−1 in the cultures of wild-type (WT) and YW1 strains in BG-11 medium supplemented with 1.0 mM IPTG, 10 mM HEPES (pH 7.5), and 1 g L−1
NH4Cl. The co-culture was diluted and plated on Luria-Bertani (LB) agar supplemented with 100 μg mL−1 ampicillin. Growth of K. pneumoniae was
determined by the number of CFU. (B) Cell growth and 1,3-PD production of K. pneumoniae using pure glycerol (PG) and the concentrated glycerol
produced by strain YW1 (CG). The supernatant from culture of strain YW1 was concentrated about 10-fold by rotary evaporation to increase the con-
centration of glycerol. The concentrated glycerol was then used as the sole carbon source for K. pneumoniae. Growth of K. pneumoniae was moni-
tored by measuring the optical density at 620 nm.
Green Chem.
This journal is © The Royal Society of Chemistry 2015

View Article Online
Green Chemistry
Paper
bacterial kernel exhibited slight inhibition on the growth of to maximize the pool of available carbon, the designed biosyn-
K. pneumoniae (Fig. 5B). However, the glycerol consumption and thetic pathways should withdraw carbon close to carbon fix-
1,3-PD production rates were almost identical. Glycerol (10 g L−1
)
ation.22 In the present study, DHAP, which is only four steps
was consumed in 8 h and 4.65 g L−1 1,3-PD was produced. The from carbon fixation (D-ribulose 1,5-bisphosphate + CO2
→
two-step strategy demonstrated here can be used to produce G3P → 1,3-bisphosphoglycerate → GAP → DHAP), was selected
desired chemicals and fuels from CO2 without the addition of as the starting point for the production of the C3 intermediate,
an external carbon feedstock in situations where synthesis of glycerol, and C3 platform chemicals. Depletion of the DHAP
the products is incompatible with cyanobacterial metabolism.
pool through the artificial pathway may help create a selection
force driving carbon flux to DHAP from fixation. As expected,
the high-yield production of glycerol in S. elongatus strain YW1
was achieved with a final titer of 1.17 g L−1 and a maximum
production rate of 7733 μg L−1 H−1. The total biomass pro-
duced by strain YW1 was higher than that by the wild-type
strain. The percentage of biomass produced as glycerol in
stain YW1 ranged from 19% to 36% of total biomass (Fig. 2E).
These results indicate that withdrawing carbon close to
fixation may be an effective strategy to increase production
yields.
Selection of a chemical target with good biocompatibility
was also important for high-yield and long-term cyanobacterial
production. Several C3 intermediates, such as glycerol and
lactate, are biocompatible chemicals and can be easily pro-
duced at high concentrations by microorganisms.47 Glycerol
can also act as an osmoregulator and an osmoprotectant in
some yeasts and green algae under salt-stress conditions.30
Toxicity tests also suggested that 10 g L−1 of glycerol was not
toxic to S. elongatus (Fig. 5A). With further improvements, i.e.,
enhancing the reaction catalyzed by native G3PDH, cyanobac-
terial glycerol production might be increased and become
more commercially viable. Another advantage of selecting gly-
Tolerance of S. elongatus to C3 platform chemicals
The toxicity of products may retard cell growth and prevent
higher production of these chemicals.7,9 Therefore, we
measured the tolerance level of S. elongatus to potential pro-
ducts including glycerol, DHA, 3-HP, 1,3-PD, and glyceric acid
(GA), and the pH changes of the cultures (Fig. S7†). Minimal
decrease in cell growth was observed in the presence of 10 g
L−1 glycerol or 1,3-PD, suggesting that glycerol and 1,3-PD are
suitable targets for high titer cyanobacterial production. Con-
versely, DHA above 0.5 g L−1 caused severe growth inhibition.
The other two C3 acids, 3-HP and GA, are more toxic. Exposing
S. elongatus cells to over 200 mg L−1 of extracellular 3-HP or GA
seriously inhibited cell growth, possibly owing to the acidifica-
tion of the cultures (Fig. S7†). The highest titers of products
obtained in the present study did not approach lethal levels.
Thus, C3 chemical production in cyanobacteria may be further
improved.
Discussion
Cyanobacterial production of many chemicals and fuels has cerol as the target product of the cyanobacterial kernel is that
been reported.22 To synthesize a specific product, a specific it makes the best use of fixed CO2 to generate the carbon skel-
heterologous pathway needs to be developed in cyanobacteria. eton in the product molecule, since the enzyme responsible
In the present study, we constructed a cyanobacterial kernel for CO2 fixation is notoriously inefficient.48 Although decarb-
that efficiently produces the C3 intermediate, glycerol, by oxylation steps are good driving forces of heterologous
genetically engineering S. elongatus PCC7942. Taking advan- pathways,7–9 they came at the cost of lower carbon conserva-
tage of the cyanobacterial kernel, difficult and time-consuming tion ratios, because a great number of fixed CO2 was released.
genetic manipulation and pathway construction could be sim- Perfect carbon conservation is achieved only in the production
plified, because the cyanobacterial kernel could be used to syn- of 1,2-propanediol19 and lactate.10 The biosynthetic pathways
thesize various products through extending heterologous of C3 platform chemicals constructed here took full advantage
pathways, co-cultivation, fermentation, or chemical synthesis. of the carbon fixed and achieved a 100% carbon conversion
This strategy would be especially useful when the heterologous ratio. It has been suggested that chemical production is funda-
pathways are incompatible with cyanobacterial metabolism or mentally limited by carbon conservation.22 Thus, the high
the target products are highly toxic to cyanobacteria. By using carbon conservation ratio may also have contributed to the
the photosynthetic platform based on the cyanobacterial high-yield glycerol production observed in this study.
kernel, several important C3 platform chemicals including
DHA, 3-HP, and 1,3-PD were produced from CO2 in this study.
Taking advantage of the cyanobacterial kernel, photosyn-
thetic production of valuable C3 chemicals from CO2 is attain-
To better serve as a versatile cyanobacterial kernel, high- able. As shown in Fig. 6, glycerol could be converted to DHA,
yield production of the intermediate from CO2 was required. A 3-HP, 1,3-PD, GA, malonic acid, hydroxypyruvic acid, and 1,2-
technoeconomic analysis suggested that a product-to-biomass propanediol by extending heterologous pathways in the cyano-
carbon partitioning ratio of above 20% is required for commer- bacterial kernel. In this study, the DHA and 3-HP biosynthesis
cially viable photosynthetic production.46 However, due to the pathways were developed in S. elongatus, demonstrating the
stringent control of carbon partitioning in cyanobacteria, first cyanobacterial production of these chemicals. The highest
80%–85% of carbon is directed toward biomass accumulation, titers of DHA and 3-HP obtained by engineered S. elongatus
and the cells compete for carbon partitioning at every step strains YW2 and YW3 reached 78.6 mg L−1 and 31.7 mg L−1
from carbon fixation to endpoints in metabolism.46 Therefore, respectively. Nevertheless, the photosynthetic production
,
This journal is © The Royal Society of Chemistry 2015
Green Chem.

View Article Online
Paper
Green Chemistry
Fig. 6 Chemicals and fuels produced using glycerol as the feedstock through biological routes and conventional catalytic conversions. The biologi-
cal and chemical routes are shaded in blue and yellow, respectively. Products in gray shaded boxes represent those whose synthesis from glycerol
has been demonstrated in previous studies. The reactions involved in the conversion of 3-HP to malonic acid are catalyzed by alcohol dehydrogen-
ase and aldehyde dehydrogenase, which were predicted by Pathway Prediction server Ver. 1.13 (http://www.genome.jp/tools/pathpred). Broken lines
represent multiple reaction steps. 3-HP, 3-hydroxypropionic acid; 3-HPA, 3-hydroxypropionaldehyde; 1,3-PD, 1,3-propanediol; 1,2-PD, 1,2-propane-
diol; GA, glyceric acid; HPA, hydroxypyruvic acid; DHA, dihydroxyacetone; DHAP, dihydroxyacetone phosphate; PEP, phosphoenolpyruvate; 2,3-BD,
2,3-butanediol; PHB, poly(3-hydroxybutyrate).
processes were bottlenecked by the cofactor bias or by inacti- developed in this study hold great potential for assimilation
vation of the oxygen-sensitive GDHt in cyanobacteria. Discov- into the chemical and energy industries.
ery of NADP+/NADPH-preferred and oxygen-tolerant enzymes
would be helpful in this regard.19,21 Alternatively, co-cultiva-
tion of the cyanobacterial kernel with a glycerol utilizer could
also make full use of the glycerol produced from CO2 to
Conclusions
produce target products, for which synthesis pathways are This study describes a new strategy for the photosynthetic pro-
incompatible with cyanobacterial metabolism.49 Co-cultivation duction of C3 platform chemicals from CO2 using an engine-
of S. elongatus YW1 with K. pneumoniae allowed the growth of ered S. elongatus PCC7942 as the kernel. The cyanobacterial
K. pneumoniae and 1,3-PD accumulation. Considering that kernel could be applied in the production of chemicals and
several native or engineered microbes can utilize glycerol as fuels through extending heterologous pathways, co-cultivation
their sole carbon and energy source, this approach could also of cyanobacteria and glycerol utilizers, glycerol fermentation,
be used for the production of various chemicals and fuels or chemical synthesis. Several important C3 platform chemi-
from CO2.
cals including DHA, 3-HP, and 1,3-PD were successfully syn-
To reduce carbon dioxide emissions from fossil fuel com- thesized from CO2 using the kernel. This demonstrated the
bustion, carbon capture and storage technologies have been potential of the new strategy for sustainably producing chemi-
developed.50 The photosynthetic production processes pre- cals and fuels from CO2.
sented here could capture CO2 and store it in the form of gly-
cerol. The produced glycerol can be used as a feedstock to
produce building blocks, fuels, biodegradable materials, food
additives, and other value-added products through both bio-
logical routes and conventional catalytic conversions
Experimental procedures
Chemicals and reagents
(Fig. 6).23,24 We have demonstrated that glycerol produced by All chemicals were purchased from Sigma-Aldrich (St. Louis,
cyanobacteria can be used as the fermentation substrate after MO) or ACROS (Geel, Belgium), unless otherwise specified.
simple concentration. Although technological and commercial The FastPfu DNA polymerase was acquired from TransGen
challenges remain to be overcome, the production processes Biotech (Beijing, China). The restriction enzymes and T4 DNA
Green Chem.
This journal is © The Royal Society of Chemistry 2015

View Article Online
Green Chemistry
Paper
ligase were purchased from New England Bio-Labs (Beverly, YW3 and YW4 were constructed by recombination of plasmids
MA, USA). Oligonucleotide synthesis was carried out by pAM-YW1, pAM-YW2, pAM-YW3 and pAM-YW4 into S. elonga-
Generay Biotech (China). Genomic DNA was extracted using tus PCC7942, respectively. S. elongatus PCC7942 was trans-
the Wizard Genomic DNA Purification kit (Promega, Madison, formed using previously described protocols.52 All
WI, USA).
transformants were segregated on the BG-11 plates with 20 μg
mL−1 spectinomycin for the selection of a successful recombi-
nation. The correct recombinants were confirmed by PCR
Culture medium and conditions
BG-11 medium (1.5 g L−1 NaNO3, 0.027 g L−1 CaCl2, 0.006 g using primers listed in Table S3† to verify the integration of
L−1 ferric ammonium citrate, 0.001 g L−1 Na2EDTA·2H2O, the target genes into the chromosome.
0.039 g L−1 K2HPO4, 0.075 g L−1 MgSO4·7H2O, 0.020 g L−1
Quantitation of gpp1, dhaB, and puuC mRNAs
Na2CO3, 0.006 g L−1 citrate, 2.860 mg L−1 H3BO3, 1.810 mg L−1
MnCl2·4H2O, 0.220 mg L−1 ZnSO4·7H2O, 0.390 mg L−1 When the cell density of the S. elongatus cultures reached an
Na2MoO4·2H2O, 0.080 mg L−1 CuSO4·5H2O, and 0.050 mg L−1 OD730 nm of 0.4 to 0.6, 1.0 mM IPTG was added to induce the
Co(NO3)2·6H2O) was used as a basic medium for S. elongatus expression of the target genes. Total RNA was prepared using
PCC7942 strains. The BG-11−N,P medium was modified from the RNAprep Pure Cell/Bacteria kit (Tiangen Biotech, China).
the BG-11 medium for 3-HP production by excluding NaNO3 Total RNA was treated with DNase I (Fermentas) and then used
and K2HPO4. Spectinomycin (20 μg mL−1) was added as as a template for cDNA synthesis with random primers and
required. For stationary culture, S. elongatus was grown statio- SuperScript III Reverse Transcriptase (Invitrogen, Carlsbad,
narily at 30 °C with an illumination intensity of 100 μE s−1 CA, USA). The resulting cDNA was used for quantitative PCR
m−2, using 50 mL BG-11 medium in a 250 mL Erlenmeyer analysis. Primers for gpp1, dhaB(CB), dhaB(KP), and puuC were
flask. For culture with bubbling ordinary air or air containing designed using Beacon Designer software (Biosoft Inter-
5% CO2, gases were bubbled at a flow rate of 50 mL min−1
.
national, Palo Alto, California, USA) (Table S3†). Expression of
Cell growth was monitored by measuring the optical density at the rnpB gene was used as a control. Quantitative PCR was per-
730 nm. Cell biomass was monitored by measuring the dry cell formed using the SuperReal PreMix SYBR Green kit (Tiangen
weight of S. elongatus cells.
Biotech, China) and the CFX96 Real-Time system (Bio-Rad,
Hercules, CA, USA), according to the manufacturer’s
instructions.
DNA manipulations
Plasmids were constructed in E. coli DH5α strains for propa-
gation and storage (Table S2†). Plasmid pAM-YW1 was con-
Optimization of 3-HP production conditions
structed by the insertion of gpp1 into pAM-MCS12 between the For the optimization of 3-HP production conditions, protocols
EcoRI and BglII sites under the control of the IPTG-inducible described previously were used with modifications.13 Strain
promoter Ptrc. To construct plasmid pAM-P2 harbouring two YW3 was grown stationarily at 30 °C with an illumination
Ptrc promoters, Ptrc was amplified from plasmid pAM2991 and intensity of 100 μE s−1
m
−2, using the BG-11 medium, until
ligated into pAM-MCS12 using the BglII and XhoI sites. the OD730 nm reached 1.0. The grown cells were then collected
Plasmid pAM-YW2 was constructed by the insertion of gpp1 by centrifugation at 5000g. The cells were resuspended in one-
into pAM-P2 between the EcoRI and BglII sites under the fifth the original volume of the appropriate media (BG-11 and
control of the first Ptrc, and by the insertion of dhaD between BG-11−N,P) with 1.0 mM IPTG, and separated into different test
the XhoI and BamHI sites under the control of the second Ptrc
.
tubes. Dark culture conditions were maintained by wrapping
Plasmid pAM-YW3 was constructed by the insertion of gpp1 the test tubes with aluminum foil, while light culture con-
into pAM-P2 between the EcoRI and BglII sites, and by the ditions were maintained by incubating the cultures in test
insertion of dhaB(CB) and puuC between the XhoI and BamHI tubes that were not wrapped. Anaerobic culture conditions
sites. Plasmid pAM-YW4 was constructed by the insertion of were maintained by purging the headspace of test tubes with
gpp1 into pAM-P2 between the XhoI and BamHI sites and by sterilized ultra-pure nitrogen gas several times. Aerobic culture
the insertion of dhaB(KP) and puuC between the EcoRI and conditions were maintained by not purging the head-space,
BglII sites. The gpp1, dhaD, dhaB(KP), dhaB(CB) and puuC and by retaining the atmospheric oxygen present in the head-
genes were amplified by PCR using the primers listed in space. Samples were collected periodically to determine the
Table S3† with the genomic DNA of S. cerevisiae, K. pneumo- concentration of products.
niae, C. butyricum, and E. coli BL21 (DE3) as templates, respect-
ively. The genes were subcloned into the pEASY-Blunt vector
Co-cultivation of S. elongatus with K. pneumoniae
and sequenced.
S. elongatus wild-type and YW1 strains were firstly cultured sta-
tionarily at 30 °C with an illumination intensity of 100 μE s−1
m−2 in BG-11 medium supplemented with 10 mM HEPES (pH
Strain construction and transformation
The strains used and constructed are listed in Table S2.† All 7.5) and 1 g L−1 NH4Cl. When the OD730 nm of the cultures
chromosomal manipulations were carried out by homologous reached 0.4–0.6, 1.0 mM IPTG was added to induce the gly-
recombination of plasmid DNA into the S. elongatus cerol production. Then K. pneumoniae cells were washed with
PCC7942 genome at neutral sites I (NSI).51 Strains YW1, YW2, phosphate-buffered saline (PBS) buffer three times, and 106
This journal is © The Royal Society of Chemistry 2015
Green Chem.

View Article Online
Paper
Green Chemistry
cells mL−1 were added into the S. elongatus cultures. HEPES The cell density and pH of the cultures were monitored for the
(10 mM, pH 7.5) was added to maintain a suitable pH for the next few days.
growth of K. pneumoniae. NH4Cl was used as the nitrogen
source for K. pneumoniae. Growth of K. pneumoniae was moni-
tored by plating serial dilutions on Luria-Bertani (LB) agar
Acknowledgements
supplemented with 100 μg mL−1 ampicillin, which inhibits
the growth of S. elongatus but does not inhibit the growth of This work was partially supported by the Chinese National
K. pneumoniae.
Program for High Technology Research and Development
(2011AA02A207). The authors acknowledge the National Basic
Research Program of China (2011CBA00800) from the Ministry
of Science and Technology of China and a grant from the
1,3-PD fermentation by K. pneumoniae using glycerol
produced by S. elongatus
The cultures of strain YW1 were centrifuged for 5 min at National Natural Science Foundation of China (31100078).
12 000g. The supernatant was collected and concentrated 10-
fold by rotary evaporation to increase the concentration of gly-
cerol to about 10 g L−1. The concentrated glycerol produced by
S. elongatus was then used as the solo carbon source for
Notes and references
K. pneumoniae. Pure glycerol was used as a control. The medium
used for glycerol fermentation by K. pneumoniae contained 5 g
L−1 yeast extract, 10 g L−1 K2HPO4·3H2O, 2 g L−1 KH2PO4, 1 g
L−1 NH4Cl, 0.1 g L−1 MgSO4·7H2O, 30 mg L−1 FeCl3·6H2O,
5 mg L−1 CoCl2·6H2O, and 10 g L−1 pure glycerol or concen-
trated glycerol produced by S. elongatus. K. pneumoniae was
incubated at a shaker speed of 200 rpm at 37 °C. Growth of
K. pneumoniae was monitored by measuring the optical density
at 620 nm.
1 P. P. Peralta-Yahya, F. Zhang, S. B. Del Cardayre and
J. D. Keasling, Nature, 2012, 488, 320–328.
2 J. W. Lee, D. Na, J. M. Park, J. Lee, S. Choi and S. Y. Lee,
Nat. Chem. Biol., 2012, 8, 536–546.
3 A. J. Straathof, Chem. Rev., 2014, 114, 1871–1908.
4 U. T. Bornscheuer, G. W. Huisman, R. J. Kazlauskas,
S. Lutz, J. C. Moore and K. Robins, Nature, 2012, 485, 185–194.
5 K. Sanderson, Nature, 2011, 474, S12–S14.
6 H. Li and J. C. Liao, Energy Environ. Sci., 2013, 6, 2892–
2899.
7 S. Atsumi, W. Higashide and J. C. Liao, Nat. Biotechnol.,
2009, 27, 1177–1180.
8 Z. Gao, H. Zhao, Z. Li, X. Tan and X. Lu, Energy Environ.
Sci., 2012, 5, 9857–9865.
9 J. W. Oliver, I. M. Machado, H. Yoneda and S. Atsumi, Proc.
Natl. Acad. Sci. U. S. A., 2013, 110, 1249–1254.
Product analysis
Culture samples (1 mL) were centrifuged for 5 min at 12 000g.
The supernatant was used for product analysis. The concen-
trations of glycerol, DHA, 3-HP and 1,3-PD were determined by
high-performance liquid chromatography (HPLC, Agilent 1200
series). The system was equipped with a Bio-Rad Aminex
HPX-87H column (300 × 7.8 mm) and refractive index and UV/ 10 A. M. Varman, Y. Yu, L. You and Y. J. Tang, Microb. Cell
Vis detectors. The analysis was performed at 55 °C with a
mobile phase of 5 mM H2SO4 at a flow rate of 0.4 mL min−1
However, under these conditions, both glycerol and 3-HP had
Fact., 2013, 12, 117.
11 C. R. Shen and J. C. Liao, Energy Environ. Sci., 2012, 5,
9574–9583.
.
the same retention time and similar response values with the 12 J. Ungerer, L. Tao, M. Davis, M. Ghirardi, P. C. Maness and
refractive index detector. Therefore, the 3-HP concentration J. Yu, Energy Environ. Sci., 2012, 5, 8998–9006.
was determined using the UV/Vis detector because glycerol is 13 T. Kusakabe, T. Tatsuke, K. Tsuruno, Y. Hirokawa,
not detected by the UV/Vis detector at 210 nm. The glycerol
concentration was estimated as the difference between the
S. Atsumi, J. C. Liao and T. Hanai, Metab. Eng., 2013, 20,
101–108.
concentrations obtained using the refractive index detector 14 P. Lindberg, S. Park and A. Melis, Metab. Eng., 2010, 12,
(glycerol + 3-HP) and those obtained using the UV/Vis detector 70–79.
(3-HP alone). Liquid chromatography-mass spectrometry 15 J. Zhou, H. Zhang, Y. Zhang, Y. Li and Y. Ma, Metab. Eng.,
(LC-MS) was performed on an Agilent 6230 TOF-MS equipped 2012, 14, 394–400.
with electrospray ionization sources to identify 3-HP produced 16 Y. Xue, Y. Zhang, D. Cheng, S. Daddy and Q. He, Proc. Natl.
by engineered cyanobacteria. 3-HP in the broth was identified Acad. Sci. U. S. A., 2014, 111, 9449–9954.
by comparing the retention time and the observed mass chro- 17 E. I. Lan and J. C. Liao, Proc. Natl. Acad. Sci. U. S. A., 2012,
matogram with the 3-HP standard.
109, 6018–6023.
18 X. Liu, J. Sheng and R. Curtiss III, Proc. Natl. Acad. Sci. U. S.
A., 2011, 108, 6899–6904.
Toxicity test of C3 platform chemicals
S. elongatus PCC7942 cells were grown according to the pro- 19 H. Li and J. C. Liao, Microb. Cell Fact., 2013, 12, 4.
cedure described above. When the cell density of the cultures 20 L. S. Gronenberg, R. J. Marcheschi and J. C. Liao, Curr.
reached an OD730 nm of 0.25, various concentrations of glycerol,
Opin. Chem. Biol., 2013, 17, 462–471.
DHA, 3-HP, 1,3-PD, and GA were added into the culture 21 E. I. Lan, S. Y. Ro and J. C. Liao, Energy Environ. Sci., 2013,
medium as indicated to achieve the desired concentration.
6, 2672–2681.
Green Chem.
This journal is © The Royal Society of Chemistry 2015

View Article Online
Green Chemistry
Paper
22 J. W. Oliver and S. Atsumi, Photosynth. Res., 2014, 120, 249– 37 Y. Chen and J. Nielsen, Curr. Opin. Biotechnol., 2013, 24,
261. 965–972.
23 F. Yang, M. A. Hanna and R. Sun, Biotechnol. Biofuels, 2012, 38 V. Kumar, S. Ashok and S. Park, Biotechnol. Adv., 2013, 31,
5, 13. 945–961.
24 J. M. Clomburg and R. Gonzalez, Trends Biotechnol., 2013, 39 S. Ashok, S. Mohan Raj, Y. Ko, M. Sankaranarayanan,
31, 20–28.
S. Zhou, V. Kumar and S. Park, Metab. Eng., 2013, 15, 10–
24.
40 M. W. Keller, G. J. Schut, G. L. Lipscomb, A. L. Menon,
I. J. Iwuchukwu, T. T. Leuko, M. P. Thorgersen, W. J. Nixon,
A. S. Hawkins and R. M. Kelly, Proc. Natl. Acad. Sci. U. S. A.,
2013, 110, 5840–5845.
41 W. S. Jung, J. H. Kang, H. S. Chu, I. S. Choi and K. M. Cho,
Metab. Eng., 2014, 23, 116–122.
42 R. Saxena, P. Anand, S. Saran and J. Isar, Biotechnol. Adv.,
2009, 27, 895–913.
25 J. W. Lee, H. U. Kim, S. Choi, J. Yi and S. Y. Lee, Curr. Opin.
Biotechnol., 2011, 22, 758–767.
26 Z. Wang, J. Zhuge, H. Fang and B. A. Prior, Biotechnol. Adv.,
2001, 19, 201–223.
27 F. Remize, L. Barnavon and S. Dequin, Metab. Eng., 2001, 3,
301–312.
28 A. M. Varman, Y. Xiao, H. B. Pakrasi and Y. J. Tang, Appl.
Environ. Microbiol., 2013, 79, 908–914.
29 H. H. Huang, D. Camsund, P. Lindblad and T. Heidorn,
Nucleic Acids Res., 2010, 38, 2577–2593.
30 H. Miyasaka, Y. Ohnishi, T. Akano, K. Fukatsu,
43 Y. Huang, Z. Li, K. Shimizu and Q. Ye, Bioresour. Technol.,
2012, 103, 351–359.
T. Mizoguchi, K. Yagi, I. Maeda, Y. Ikuta, H. Matsumoto 44 Y. Xu, H. Chu, C. Gao, F. Tao, Z. Zhou, K. Li, L. Li, C. Ma
and N. Shioji, J. Ferment. Bioeng., 1998, 85, 122–124. and P. Xu, Metab. Eng., 2014, 23, 22–33.
31 G. P. da Silva, M. Mack and J. Contiero, Biotechnol. Adv., 45 E. I. Lan and J. C. Liao, Metab. Eng., 2011, 13, 353–363.
2009, 27, 30–39. 46 A. Melis, Curr. Opin. Chem. Biol., 2013, 17, 453–456.
32 L. Lu, L. Wei, K. Zhu, D. Wei and Q. Hua, Bioresour. 47 W. C. Ray and M. W. Grinstaff, Macromolecules, 2003, 36,
Technol., 2012, 117, 317–324. 3557–3562.
33 Y. J. Zhou, W. Yang, L. Wang, Z. Zhu, S. Zhang and 48 D. C. Ducat, J. C. Way and P. A. Silver, Trends Biotechnol.,
Z. K. Zhao, Microb. Cell Fact., 2013, 12, 103. 2011, 29, 95–103.
34 Y. Wang, F. Tao and P. Xu, J. Biol. Chem., 2014, 289, 6080– 49 H. Niederholtmeyer, B. T. Wolfstädter, D. F. Savage,
6090.
P. A. Silver and J. C. Way, Appl. Environ. Microbiol., 2010, 76,
3462–3466.
50 R. S. Haszeldine, Science, 2009, 325, 1647–1652.
35 M. Tamoi, T. Miyazaki, T. Fukamizo and S. Shigeoka, Plant
J., 2005, 42, 504–513.
36 T. Werpy and G. Petersen, Top value added chemicals from 51 S. A. Bustos and S. S. Golden, J. Bacteriol., 1991, 173, 7525–
biomass volume I – Results of screening for potential candi- 7533.
dates from sugars and synthesis gas, US Department of 52 S. S. Golden, J. Brusslan and R. Haselkorn, Methods
Energy, 2004, DOI: 10.2172/15008859.
Enzymol., 1987, 153, 215–231.
This journal is © The Royal Society of Chemistry 2015
Green Chem.