当前位置:首页 >> 电子/电路 >>

Systematic metabolic engineering of Corynebacterium glutamicum for the industrial-level production

Green Chemistry
View Article Online
View Journal | View Issue

Cite this: Green Chem., 2017, 19, 5691

Systematic metabolic engineering of Corynebacterium glutamicum for the industriallevel production of optically pure D-(?)-acetoin?
Yufeng Mao, ?a Jing Fu, ?a Ran Tao, ?a Can Huang, Ya-Jie Tang,b Tao Chen *a,b and Xueming Zhaoa

Zhiwen Wang,


Published on 23 October 2017. Downloaded on 17/04/2018 01:20:46.

Acetoin is a high-value-added industrial product and a promising bio-based platform chemical. The two chiral enantiomers of acetoin are important potential pharmaceutical intermediates. However, at present it is very expensive to produce optically pure acetoin using conventional chemical synthesis methods. While classical biocatalysis ful?ls many key criteria of green chemistry, it also comes with many other disadvantages, e.g. complicated process engineering, the need for expensive substrates and/or inducers, problematic production strains, or unsatisfactory titers. By redirecting metabolic ?uxes, ?ne-tuning the activities of key enzymes and improving the strains’ genetic stability, the systematically engineered strain CGR5 was constructed, which o?ered improved D-(?)-acetoin production with optical purity surpassing 99.9%. Furthermore, acetoin production was increased signi?cantly by the optimization of dissolved oxygen levels and culture medium components. Under aerobic conditions, the best strain CGR7 produced 96.2 g L?1 D-(?)-acetoin, with a productivity of 1.30 g (L h)?1 in a 5 L batch fermenter. To the best of our knowledge, this is the ?rst report on producing optically pure D-(?)-acetoin with the highest titer, without
Received 9th September 2017, Accepted 23rd October 2017 DOI: 10.1039/c7gc02753b rsc.li/greenchem

multi-stage fermentation, concentration of cells by centrifugation or other complicated process steps. The process therefore is e?cient and environmentally friendly. These results furthermore demonstrate that the biosafety level 1 microorganism C. glutamicum is a competitive choice for industrial-level production of optically pure D-(?)-acetoin.



Acetoin (AC) is a high-value-added chemical with a pleasant cheese-like odor and a creamy butter taste, which has been the subject of intense research.1,2 Moreover, AC was classified as one of the 30 platform chemicals whose development and utilization was given priority by the U.S. Department of Energy, since it is a highly promising bio-based platform chemical.3,4 AC has two chiral enantiomers, the D-(?)- and L-(+)-forms,

a Key Laboratory of Systems Bioengineering (Ministry of Education); SynBio Research Platform, Collaborative Innovation Center of Chemical Science and Engineering (Tianjin); School of Chemical Engineering and Technology, Tianjin University, Tianjin, 300072, China. E-mail: chentao@tju.edu.cn, maoyf@tju.edu.cn, jfu@tju.edu.cn, taoran@tju.edu.cn, huangcan@tju.edu.cn, zww@tju.edu.cn, xmzhao@tju.edu.cn; Fax: +86 22 85356617; Tel: +86 22 85356617 b Key Laboratory of Fermentation Engineering (Ministry of Education); Hubei Key Laboratory of Industrial Microbiology, Hubei Provincial Cooperative Innovation Center of Industrial Fermentation, Hubei University of Technology, Wuhan 430068, China. E-mail: yajietang@qq.com ? Electronic supplementary information (ESI) available. See DOI: 10.1039/ c7gc02753b ? These authors contributed equally to this article.

which are important potential pharmaceutical intermediates.1 Optically pure AC has also been widely used to synthesize liquid crystal materials and novel, optically active α-hydroxyketone derivatives.2,5 Chemical synthesis of a single optical isomer of AC, which requires crystallization or chiral separation, comes with many disadvantages such as low yield, instability of product quality, high cost and high pollution.6,7 Consequently, increasing attention has been given to biocatalytic processes, which can be more ecological or sustainable than their chemical counterparts.8 However, some of the manufacturing processes involve complicated steps, such as centrifugation to concentrate the catalyst cells, exceedingly precise control of the fermentation conditions, or cooperation of di?erent microorganisms.9,10 Such processes can also be very expensive due to the need for chiral 2,3-butanediol as the substrate, or the addition of expensive inducers.10–13 Moreover, some of the natural AC producers are opportunistic pathogens,14,15 and some of the substrates are noxious (e.g. diacetyl),16 which brings with it di?cult challenges for industrial-scale fermentation. The highest reported D-(?)-AC titer of 86.7 g L?1 was achieved by converting optically pure meso-(?)-2,3-butanediol, which is

This journal is ? The Royal Society of Chemistry 2017

Green Chem., 2017, 19, 5691–5702 | 5691

View Article Online


Green Chemistry

also a high-value-added platform chemical.10 Considering the high cost of the optically active substrate and the complicated centrifugation process, this approach was not suitable for the industrial-scale production of D-(?)-AC. Instead, a fermentation method that can produce D-(?)-AC directly from glucose is a better choice, and 60.3 g L?1 D-(?)-AC was already produced by engineered E. coli.17 However, the titer was still insu?cient for economical production. It has been widely reported that many microorganisms can produce a mixture of both enantiomers of AC, or AC with unreported optical purity.1 The genus Bacillus is considered to contain the most e?cient AC producers.4 Using conventional enrichment and screening procedures, without mutagenesis and genetic manipulation, native Bacillus strains were found that can already produce acceptable yields and titers of AC in suitable fermentation processes.18–20 However, the yields and titers were not satisfactory in all the discovered strains. Although 75.2 g L?1 AC was produced from sucrose by Serratia marcescens H32 that expresses a water-forming NADH oxidase, this biosafety class II microorganism can be hazardous to humans, which drastically increases the cost of industrialscale production.21 The model strain Bacillus subtilis 168, which cannot use xylose naturally, was engineered to produce 62.6 g L?1 AC by assimilating glucose, xylose and arabinose without preference.22,23 The highest AC titer to date, at 100 g L?1, was produced using Saccharomyces cerevisiae within 60 h.24 However, the enantiomeric configuration of the product was not reported. C. glutamicum, which has been granted the GRAS (generally regarded as safe) status by the US Food and Drug Administration, has been applied as a model system in the biochemistry, genetics and physiology of Gram-positive bacteria.25 It has long been used in food, medical, and industrial applications.26 AC production by C. glutamicum using a simple onestep aerobic process not only lowers the cost of fermentation, but can also bring about high productivity and minimal environmental impact. AC formation from pyruvate is catalyzed in a two-step reaction by acetolactate synthase (ALS) and acetolactate decarboxylase (ALD).1 The downstream enzyme AC reductase (encoded by butA in C. glutamicum), which converts AC into 2,3-butanediol, must be deleted to prevent the further conversion of the generated AC. In this work, a series of mutant strains were constructed to improve the yield and production of AC step by step, including the disruption of all relevant genes ( pta, ack, ldh, nagD and ppc/pyc) in the biosynthesis of competing by-products, i.e. acetate, lactate, glycerin and succinate, as well as the selective introduction of alsS (BG10471) and alsD (BG10472) from the AC synthetic pathway of B. subtilis 168 driven by a strong constitutive promoter (Fig. 1). The resulting strain produced D(?)-AC with an optical purity of more than 99.9%. Furthermore, AC production was increased significantly by the optimization of the dissolved oxygen level and culture medium components. Fed-batch cultures under one-step aerobic conditions were employed to investigate the performances of these genetically modified strains. The best strain CGR9 pro-

Published on 23 October 2017. Downloaded on 17/04/2018 01:20:46.

Fig. 1 The D-(?)-acetoin biosynthetic pathway of C. glutamicum. Genes that were overexpressed or introduced are indicated by bold and red arrows, and those that were interrupted are shown with overlaid red crosses. The enzymes and their encoding genes are: acetolactate synthase, alsS; acetolactate decarboxylase, alsD; phosphotransacetylase, pta; acetate kinase, ack; lactate dehydrogenase, ldh; pyruvate dehydrogenase E1 component, aceE; dihydrolipoamide dehydrogenase, lpd; succinyl-CoA:acetate CoA-transferase, actA; putative phosphatase, nagD; phosphoenolpyruvate carboxylase, ppc; pyruvate carboxylase, pyc; meso-butanediol dehydrogenase, butA.

duced 96.2 g L?1 D-(?)-AC with a chemical purity of more than 95% and a yield of 0.36 g g?1 glucose in fed batch experiments.

2. Experimental
2.1 Bacterial strains, media and growth conditions All C. glutamicum strains in this study were derived from the wild-type C. glutamicum ATCC 13032. The strains and plasmids used in this study are listed in Table 1. The methods used to construct the strains and plasmids are described in the ESI.? The lysogeny broth (LB) medium, containing tryptone (10 g L?1), yeast extract (5 g L?1) and NaCl (10 g L?1), was used for plasmid construction in E. coli. CGXIIY medium, containing 10 g L?1 (NH4)2SO4, 5 g L?1 urea, 1 g L?1 KH2PO4, 1 g L?1 K2HPO4, 0.25 g L?1 MgSO4·7H2O, 10 mg L?1 CaCl2, 10 mg L?1 FeSO4·7H2O, 0.1 mg L?1 MnSO4·H2O, 1 mg L?1 ZnSO4·7H2O, 0.2 mg L?1 CuSO4·5H2O, 20 μg L?1 NiCl2·6H2O, 0.5 g L?1 yeast extract (Sangon, China), with the indicated amounts of glucose was used as the minimal medium for flask fermentations to test the physiological characteristics of the engineered strains. Biotin (Sangon, China) was added to the CGXIIY medium for all C. glutamicum strains to a final concentration of 0.2 mg L?1. Genetic modifications in C. glutamicum were carried out by electroporation, and the recombinant strains were cultured on brain heart infusion-sorbitol (BHIS) agar plates containing appropriate antibiotics.27 LB-sucrose medium (LB medium with 10% sucrose and 1.5% agar) was used for the selection of marker-free engineered colonies of C. glutamicum. 1000 g L?1 glucose, which was used for fedbatch fermentation, was prepared as follows. 200 g glucose and 72 mL ddH2O were added into a Schott–Duran bottle, and then sterilized at 110 °C for 10 min. A series of optimized

5692 | Green Chem., 2017, 19, 5691–5702

This journal is ? The Royal Society of Chemistry 2017

View Article Online

Green Chemistry
Table 1 Strains and plasmids


Strain or plasmid Strain B. subtilis 168 C. glutamicum ATCC 13032 CGF0 CGF1 CGF2 CGF3 CGF4 CGF5 CGF6 CGR3 CGR4 CGR5 CGR6 CGR7 CGR9 CGR10 Plasmids pUC18 pEC-XK99E pEC-XK99E-alsSD pHY300 pD-sacB pD-sacB-butA pD-sacB-ldh pD-sacB-pta-ackA pD-sacB-nagD pD-sacB-pta-ackA-alsSD pHY300-ppc pHY300-pyc pEC-XK99E-alsSD-ΔlacIq

Description The alsS and alsD gene donor. Parent strain ATCC13032ΔptaΔackΔldh ATCC13032ΔbutA ATCC13032ΔptaΔackΔldhΔbutA ATCC13032 pEC-XK99E-alsSD ATCC13032ΔptaΔackΔldh; pEC-XK99E-alsSD ATCC13032ΔbutA; pEC-XK99E-alsSD ATCC13032ΔptaΔackΔldhΔbutA; pEC-XK99E-alsSD ATCC13032ΔptaΔackΔldhΔbutA; pEC-XK99E-alsSD-ΔlacIq ATCC13032ΔptaΔackΔldhΔbutA; ackA::alsSD ATCC13032ΔptaΔackΔldhΔbutA; ackA::alsSD; pEC-XK99E-alsSD-ΔlacIq ATCC13032ΔptaΔackΔldhΔbutAΔnagD; ackA::alsSD ATCC13032ΔptaΔackΔldhΔbutAΔnagD; ackA::alsSD;pEC-XK99E-alsSD-ΔlacIq ATCC13032ΔptaΔackΔldhΔbutAΔnagDΔppc; ackA::alsSD;pEC-XK99E-alsSD-ΔlacIq ATCC13032ΔptaΔackΔldhΔbutAΔnagDΔpyc; ackA::alsSD; pEC-XK99E-alsSD-ΔlacIq Ampr Kanr Kanr Tetr Kanr pD-sacB containing butA flanks pD-sacB containing ldh flanks pD-sacB containing pta-ackA flanks pD-sacB containing nagD flanks pD-sacB-pta-ack containing Tuf-alsSD flanks pHY300 containing ppc flanks pHY300 containing pyc flanks pEC-XK99E containing alsSD with lacIq deleted Lab stock Lab stock This study This study This study This study This study This study This study This study This study This study This study This study This study This study Lab Stock Lab Stock 36 Lab Stock Lab Stock 36 This study This study This study This study This study This study This study

Published on 23 October 2017. Downloaded on 17/04/2018 01:20:46.

media with glucose were used to enhance AC production, and their compositions are listed in the ESI tables.? Antibiotics were added where appropriate as follows: for C. glutamicum, 25 μg mL?1 kanamycin, 5 μg mL?1 tetracycline; for E. coli, 100 μg mL?1 ampicillin, 25 μg mL?1 kanamycin and 15 μg mL?1 tetracycline. 2.2 Culture conditions

The strains were stored at ?80 °C and revived by growing on LB agar slants/plates at 30 °C. Single colonies were transferred into 5 mL LB medium containing the appropriate antibiotics when required. All cultivations were performed at 30 °C. After 12 h of incubation at 220 rpm, a 2% (v/v) inoculum was added to 50 mL CGXIIY medium with 20 g L?1 glucose in a 250 mL shake flask and cultivated for 12 h at 220 rpm to prepare the seed culture. Cell growth was determined by measuring the optical density at 600 nm (OD600) using a UV-Vis spectrophotometer. The main culture was inoculated at an initial OD600 of 1 under culture conditions for flask fermentation, and 50 mL CGXIIY in a 250 mL flask with silica gel plug, under shaking at 180 rpm, was used for the fermentation tests. 2.3 Enzyme assays

the amount of acetolactate formed as described previously,28,29 except that the reaction temperature was set at 30 °C. The ALD activity was also assayed at 30 °C by measuring the production of AC as described previously.29,30 Cells were harvested from 1 mL culture samples by centrifugation for 1 min, and the supernatant was removed. The pellet was then lysed in 400 μL of 0.1 M phosphate bu?er ( pH = 7.0) by sonication (130 W, 20 kHz, pulse: 5 s on; 5 s o?; total: 20 min) on ice, followed by centrifugation at 4 °C and 10 000g for 5 min to remove cell debris. Total protein concentrations were determined according to the Bradford method.31


Detection of metabolites

To determine ALS and ALD activities, bacteria were cultured in CGXIIY medium at 220 rpm until reaching the exponential growth phase. The ALS activity was determined by measuring

Glucose was monitored by using an SBA sensor machine (Institute of Microbiology, Shandong, China). Extracellular metabolite concentrations were determined by HPLC as described previously.32 The mobile phase consisted of 5 mM H2SO4 with a flow rate of 0.4 ml min?1, and the column temperature was set at 65 °C. To determine the enantiomeric distribution of AC and 2,3-butanediol, the fermentation samples were extracted with ethyl acetate, and the optical isomers in the extracts were analyzed by GC-FID (PERSEE, Beijing, China) equipped with a HP-chiral 20b column (Agilent, USA), as described previously.33–35 The oven temperature program was as follows: 40 °C (2 min), increased to 75 °C (4 min) at

This journal is ? The Royal Society of Chemistry 2017

Green Chem., 2017, 19, 5691–5702 | 5693

View Article Online


Green Chemistry

5 °C min?1, followed by a ramp of 1 °C min?1 to 80 °C (2 min), and finally 15 °C min?1 to 230 °C (4 min). 2.5. Scale-up of the biosynthesis of D-(?)-AC in fed-batch fermentations For fed-batch fermentations, 100 mL seed culture was inoculated in a 5 L fermenter (Bailun, Shanghai, China) with an operating volume of about 2 L using LBRC medium. The inoculation procedure was identical to that described for flask experiments. All cultivations were carried out at 30 °C at an aeration rate of 1 vvm. The agitation speed was maintained at 400 rpm. The initial pH of the medium was 7.0. During the fermentation process, the pH value was not controlled unless it fell below 6.0. The initial glucose concentration was 55 g L?1, and about 50 mL of 1000 g L?1 glucose was added to the medium when the residual glucose concentration dropped to about 5 g L?1, to keep the glucose concentration between 30 and 70 g L?1.

Published on 23 October 2017. Downloaded on 17/04/2018 01:20:46.

3. Results and discussion
3.1 Disruption of the synthesis pathways of the by-products lactate and acetate, and the downstream product 2,3butanediol, to enhance AC production In our previous preliminary study, engineered C. glutamicum CGL3 (C. glutamicum ΔaceEΔldhΔbutA; pEC-XK99E-alsSD) produced 5.09 g L?1 AC under micro-aerobic conditions.36 However, an additional acetate was a requisite in the culture due to the deletion of aceE, which resulted in a lower growth rate and consequently low AC productivity. Therefore, the deletion of aceE was not optimal for the elimination of the byproduct pathway of acetate. Instead, the other acetate synthetic pathway, involving phosphotransacetylase and acetate kinase, was taken into consideration as the candidate (Fig. 1). In order to decrease the by-products of AC production, the relevant genes were deleted in the lactate, acetate and 2,3-butanediol synthetic pathways. To reduce lactate and acetate production, ldh, pta and ack were deleted together in C. glutamicum ATCC 13032, yielding CGF0. To abolish 2,3-butanediol production, the native butA gene was deleted in ATCC 13032 and CGF0, yielding CGF1 and CGF2, respectively (Fig. 2). As shown in Table 2, lactate production was completely abolished in CGF0 and CGF2. However, the growth rate and glucose consumption rate were decreased drastically by the interruption of lactate production, and the maximum biomass significantly declined, probably due to an excess of reducing power.35 In addition, acetate production was partially increased, probably due to the imbalance of reducing power caused by the deletion of ldh (Fig. 1). Due to the deletion of acetolactate decarboxylase, AC was not produced in these strains; the metabolic profiles were almost the same in ATCC 13032 and CGF1. To further investigate the e?ects of deleting the relevant genes on AC production, the C. glutamicum expression vector pEC-XK99E-alsSD constructed in our previous study36 was introduced into the strains. This vector overexpresses the alsS

Fig. 2 Metabolic pro?les of the strains CGF0 and CGF2. The strains were cultured in CGXIIY medium with glucose at 30 °C and 180 rpm in 250 mL ?asks with a ?nal culture volume of 50 mL. A: The fermentation characteristics of CGF0; B: the fermentation characteristics of CGF2.

and alsD genes from the trc promoter induced by isopropyl β-D-1-thiogalactopyranoside (IPTG). As shown in Fig. 9, all the strains carrying pEC-XK99E-alsSD produced D-(?)-AC in CGXII medium with an optical purity of more than 99%. Moreover, since they contain butA, CGF3 and CGF4 also respectively produced 4.58 and 5.01 g L?1 of meso-2,3-butanediol. As expected, the production of meso-2,3-butanediol was completely abolished by the inactivation of AC reductase in CGF5 and CGF6, which led to a dramatic increase of AC production. Notably, after the blocking of lactate production, the by-products acetate and glycerin increased significantly, most likely due to the surplus of NADH which would otherwise be consumed in the lactate synthetic pathway (Fig. 1). However, the AC titer, yield and productivity were markedly improved in CGF6, with an increase of more than 30% compared to CGF5 at 36 h, with only a small amount of residual glucose (around 2–5 g L?1). By the time the glucose was depleted, the final D-(?)-AC pro-

5694 | Green Chem., 2017, 19, 5691–5702

This journal is ? The Royal Society of Chemistry 2017

View Article Online

Green Chemistry AC yieldc (mol mol?1) (at 48 h) Bacteria were cultivated with a final volume of 50 mL CGXIIY in a 250 mL flask shaking at 180 rpm and 30 °C; data are average values and standard deviations of triplicate experiments. WT, wild type, ATCC13032; 2,3-BD 2,3-butanediol, AC acetoin; ND not detected or lower than 0.01 g L?1. a The biomass was measured at 30 h. b These metabolites were measured at 36 h. c These metabolites were measured at 48 h (the residual glucose had reached zero between 36 and 48 h).


duction by CGF6 had reached 10.76 g L?1, with a yield of 0.579 mol mol?1 glucose, and a productivity of 0.224 g (L h)?1. Therefore, CGF6, which carries deletions of butA, ldh, pta and ack, was chosen for further improvement. 3.2 Increasing the metabolic flux from pyruvate to AC by constitutively overexpressing alsS and alsD Aiming to increase the metabolic flux in the AC synthetic pathway, the AC pathway genes, alsS and alsD from B. subtilis, were introduced into C. glutamicum in conjunction with di?erent promoters and vectors, and the resulting constructs’ e?ects on AC production were compared. The trc promoter in the expression vector pEC-XK99E-alsSD which was used in CGF6 requires induction with IPTG, which was not suitable for industrial-scale AC production due to its high price as an inducer. Therefore, the lac repressor lacIq was removed from pEC-XK99E-alsSD to yield the vector pEC-XK99E-alsSD-ΔlacIq, which was then introduced into CGF2, yielding CGR3. As shown in Fig. 3A and B, the final AC production of CGR3, which carries the de-repressed trcΔlacIq construct, reached 10.98 g L?1 at 48 h, with a yield of 0.576 mol mol?1 glucose, and productivity of 0.229 g (L h)?1 almost unchanged compared with the 10.76 g L?1 produced by CGF6. Thus, this construct has the potential to drastically lower the cost of manufacturing by abrogating the need for the addition of IPTG, which would also simplify the fermentation process. In order to improve the genetic stability of the engineered strains, a single copy of the alsSD operon driven by the native constitutive tuf promoter was introduced into the chromosome of CGF2 at the pta-ack locus to generate CGR4. Interestingly, a significant increase of D-(?)-AC production was achieved as shown in Fig. 3C, and 12.16 g L?1 of D-(?)-AC was produced by CGR4 with a yield of 0.655 mol mol?1 glucose and a productivity of 0.337 g (L h)?1 at 36 h, which were 48.7, 28.5 and 48.6% higher than those of CGR3, respectively. Moreover, the cell growth rate in the exponential phase was drastically increased, and the glucose consumption rate was much faster during the entire fermentation. The activities of ALS and ALD in the recombinant strains were also assayed. As shown in Table 3, the activity ALS in CGF6 was about 41-fold higher than in ATCC 13032 at 10 h (exponential phase) and 36-fold higher at 20 h (stable phase). As expected, there was no significant di?erence in the activities of ALS and ALD between CGF6 and CGR3, which yielded similar D-(?)-AC titers. Thus, a constitutive promoter appears to be a superior candidate for AC production on the industrial scale, which could provide a simplified fermentation process and lower the cost for the same level of AC production. Noticeably, as shown in Table 3, the ALS and ALD activities of CGR4 were much lower than those of CGR3, most likely due to the di?erent copy numbers of alsSD in these strains. However, the AC titer of CGR4 was much higher than that of CGR3. The vector pEC-XK99E, which is a widely used expression vector for C. glutamicum, has a medium copy number,37 and thus the copy number of the alsSD operon was expected to be much higher than one. This result suggested

Glucose AC yield /proportion AC α-Ketoglutarateb consumed productivityb of theoretical yield ACc (g L?1) ?1 ?1 b ?1 (g L ) (g L ) (g (L h) ) (mol mol?1)b (at 48 h) Succinateb Glycerinb (g L?1) (g L?1) Lactateb (g L?1) Acetateb (g L?1) Metabolic profile at 36 h meso-2,3-BDb D-(?)-ACb (g L?1) (g L?1) Strain OD600 a

Published on 23 October 2017. Downloaded on 17/04/2018 01:20:46.

Metabolic characterization of C. glutamicum strains cultivated in CGXIIY medium supplemented with 40 g L?1 glucose at 180 rpm

Table 2

This journal is ? The Royal Society of Chemistry 2017


23.41 ± 0.52 15.16 ± 0.73 19.14 ± 0.14 12.59 ± 0.37 20.27 ± 0.39 17.24 ± 0.56 21.83 ± 0.30 19.44 ± 0.25 18.90 ± 0.34 22.56 ± 1.10 25.10 ± 0.37 24.25 ± 0.49 25.65 ± 0.01

ND ND ND ND 4.58 ± 0.91 5.01 ± 0.71 ND ND ND ND ND ND ND

ND ND ND ND 2.37 ± 0.14 4.09 ± 0.07 6.33 ± 0.24 8.29 ± 0.13 8.18 ± 0.17 12.16 ± 0.34 12.93 ± 0.08 13.34 ± 0.40 13.71 ± 0.45

4.23 ± 0.34 3.87 ± 0.56 2.01 ± 0.04 5.25 ± 0.12 0.29 ± 0.08 2.95 ± 0.13 0.55 ± 0.15 2.51 ± 0.05 2.75 ± 0.20 2.37 ± 0.27 1.58 ± 0.12 2.38 ± 0.28 1.26 ± 0.05

12.38 ± 1.43 ND 8.54 ± 0.32 ND 12.65 ± 0.83 ND 10.14 ± 0.13 ND ND ND ND ND ND

0.11 ± 0.05 2.87 ± 0.82 0.19 ± 0.02 1.02 ± 0.14 0.02 ± 0.01 3.22 ± 0.03 0.03 ± 0.01 2.45 ± 0.33 2.13 ± 0.46 1.56 ± 0.68 3.05 ± 0.16 0.71 ± 0.08 2.94 ± 0.02

0.74 ± 0.39 1.69 ± 0.51 0.56 ± 0.05 0.94 ± 0.01 0.95 ± 0.04 2.45 ± 0.28 1.39 ± 0.07 2.45 ± 0.17 2.41 ± 0.20 2.79 ± 0.03 1.59 ± 0.15 0.96 ± 0.03 0.85 ± 0.01

0.33 ± 0.14 1.17 ± 0.61 0.46 ± 0.19 1.73 ± 0.21 0.42 ± 0.05 0.37 ± 0.03 0.65 ± 0.05 0.45 ± 0.01 0.67 ± 0.09 0.33 ± 0.09 0.23 ± 0.02 0.52 ± 0.08 0.23 ± 0.00

29.13 26.95 27.74 25.30 31.63 34.35 34.20 33.65 32.90 38.00 39.50 40.50 38.50

0 0 0 0 0.066 0.114 0.176 0.230 0.227 0.338 0.359 0.371 0.381

0 0 0 0 0.153 0.244 0.379 0.504 0.509 0.655 0.670 0.673 0.728

0 0 0 0 3.79 ± 0.42 5.78 ± 0.33 7.94 ± 0.29 10.76 ± 0.29 10.98 ± 0.32 12.11 ± 0.31 12.72 ± 0.10 12.91 ± 0.45 13.04 ± 0.22

0 0 0 0 0.210 0.320 0.439 0.579 0.576 0.652 0.659 0.652 0.693

Green Chem., 2017, 19, 5691–5702 | 5695

View Article Online


Green Chemistry

Published on 23 October 2017. Downloaded on 17/04/2018 01:20:46.

Fig. 3 Metabolic pro?les of the strains CGF6, CGR3, CGR4 and CGR5. The strains were cultured in CGXIIY medium with glucose at 30 °C and 180 rpm in 250 mL ?asks with a ?nal culture volume of 50 mL. A: The fermentation characteristics of CGF6; B: the fermentation characteristics of CGR3; C: the fermentation characteristics of CGR4; D: the fermentation characteristics of CGR5.

Table 3 Enzyme activities of C. glutamicum strains harboring the acetoin pathway gene cluster with di?erent promoters or copy numbers in CGXIIY medium

Exponential phase (10 h) Strains ATCC 13032 CGF2 CGF6 CGR3 CGR3 CGR4 CGR5 CGR7 IPTG added (yes/no) No No Yes No Yes No No No ALS activity (μmol mg?1 min?1) 0.111 ± 0.027 0.109 ± 0.007 4.614 ± 0.764 5.379 ± 0.383 5.437 ± 0.483 0.682 ± 0.083 8.987 ± 0.706 8.500 ± 0.546 ALD activity (μmol mg?1 min?1) 0.005 ± 0.001 0.002 ± 0.001 1.219 ± 0.068 1.158 ± 0.095 1.018 ± 0.074 0.655 ± 0.054 2.034 ± 0.087 2.008 ± 0.042

Stable phase (20 h) ALS activity (μmol mg?1 min?1) 0.069 ± 0.023 0.069 ± 0.009 2.547 ± 0.477 2.254 ± 0.085 2.566 ± 0.074 0.707 ± 0.045 2.421 ± 0.103 2.864 ± 0.051 ALD activity (μmol mg?1 min?1) ND ND 0.902 ± 0.11 0.881 ± 0.076 0.938 ± 0.085 0.584 ± 0.022 0.637 ± 0.017 0.663 ± 0.016

ALS: α-Acetolactate synthase; ALD: α-Acetolactate decarboxylase; ND, not detected. Data are average values and standard deviations of triplicate experiments.

that in the shake flask experiment, the activities of ALS and ALD were adequate to enable enhanced AC production and improved cell growth rate.

To further investigate the e?ect of expressing the alsSD operon on AC production, the plasmid pEC-XK99E-alsSD-ΔlacIq was also introduced into CGR4, yielding CGR5. As shown in

5696 | Green Chem., 2017, 19, 5691–5702

This journal is ? The Royal Society of Chemistry 2017

View Article Online

Green Chemistry


Fig. 3D, the AC production of CGR5 was slightly increased with a final titer of 12.93 g L?1 and a yield of 0.670 mol mol?1 glucose. In addition, high activities of ALS and ALD could be a workable strategy to obtain a reliable conversion from pyruvate to D-(?)-AC. Therefore, CGR5 was chosen for further improvement. According to the KEGG database (http://www.kegg.jp), the native acetolactate synthase of C. glutamicum is encoded by ilvB, while acetolactate decarboxylase is absent. This was in agreement with the activity measurements shown in Table 3. This is also why the wild-type strain ATCC 13032 could not naturally produce AC. 3.3
Published on 23 October 2017. Downloaded on 17/04/2018 01:20:46.

Reduction of the by-product glycerin by deleting nagD

Even though the AC production was significantly improved, there were still considerable amounts of the by-products; acetate, glycerin and succinate. Acetate production was reduced by the deletion of the pta and ack genes, but it increased unexpectedly after the elimination of the production

of lactate by the deletion of ldh, and of meso-2,3-butanediol by the deletion of butA. This result may be caused by the surplus of reducing power in the form of NADH and the consequent unmet requirement for the regeneration of NAD+, and the presence of multiple synthetic pathways for acetate. In addition, the by-products glycerin and succinate increased in CGF5 when other by-products, such as lactate, acetate and meso-butanediol, which also consume certain amounts of NADH (Fig. 1), were restricted. Therefore, the relevant genes in the glycerin and succinate synthetic pathways should also be disrupted. Consequently, the gene responsible for glycerin production was identified and deleted. The coding gene of the enzyme responsible for the production of 1,3-dihydroxyacetone (DHA, the direct biosynthetic precursor of glycerin) was first reported to be hdpA (cgR_2128) in the wild-type C. glutamicum strain R (JCM 18229),29 and thereafter enzymatic and genetic investigations indicated that glycerin was largely produced from DHA in a reaction catalyzed by meso-butanediol dehydrogenase encoded by butA, which inherently catalyzes the interconver-

Fig. 4 Metabolic pro?les of the strains CGR6 and CGR7. The strains were cultured in CGXIIY medium with glucose at 30 °C and 180 rpm in 250 mL-?asks with a ?nal culture volume of 50 mL. A: The fermentation characteristics of CGR6; B: the fermentation characteristics of CGR7.

Fig. 5 Metabolic pro?les of the strains CGR9 and CGR10. The strains were cultured in CGXIIY medium with glucose at 30 °C and 180 rpm in 250 mL ?asks with a ?nal culture volume of 50 mL. A: The fermentation characteristics of CGR9; B: the fermentation characteristics of CGR10.

This journal is ? The Royal Society of Chemistry 2017

Green Chem., 2017, 19, 5691–5702 | 5697

View Article Online

Table 4 Metabolic characterization of C. glutamicum CGR7 cultivated in CGXIIY medium at 140, 180, and 220 rpm

Green Chemistry

Speed 140 180 220

Maximum OD600 21.9 ± 1.27 26.3 ± 0.86 31.4 ± 1.59

a D-(?)-AC

(g L?1)

AC productivityb (g (L h?1)) 0.318 0.413 0.469

AC yielda (mol mol?1) 0.640 0.719 0.723

Glucose consumed (g L?1) 37 38 38

11.58 ± 0.51 13.36 ± 0.48 13.43 ± 0.75

Bacteria were cultivated with a final volume of 50 mL CGXIIY with around 40 g L?1 glucose in a 250 mL flask shaking at 140, 180, and 220 rpm and 30 °C; data are average values and standard deviations of triplicate experiments. AC: acetoin. a These metabolites were measured when glucose was depleted. b The AC productivity was measured at 30 h.

Published on 23 October 2017. Downloaded on 17/04/2018 01:20:46.

sion between AC and 2,3-butanediol, and a putative phosphatase in n-acetylglucosamine metabolism encoded by hdpA.30 In this study, the corresponding gene in C. glutamicum ATCC 13032 was confirmed to be nagD (Fig. 1) via gene blast. Therefore, nagD was deleted in CGR4, yielding CGR6, after which pEC-XK99E-alsSD-ΔlacIq was introduced into it to yield CGR7. As shown in Fig. 4A and B, at 36 h the maximum AC titer of CGR6 was 13.34 g L?1, with a yield of 0.674 mol mol?1 glucose and a productivity of 0.371 g (L h)?1, while the titer of CGR7 was 13.71 g L?1 with a yield of 0.728 mol mol?1 glucose and a productivity of 0.381 g (L h)?1. When glucose was depleted and the maximum AC titer was reached at 36 h, the glycerin production of CGR7 was reduced from 1.59 g L?1 to 0.81 g L?1 compared with CGR5. There was also a similar decrease of glycerin production in CGR6 compared with CGR4, from 2.79 g L?1 to 0.96 g L?1. Therefore, the interruption of nagD was demonstrated to reduce glycerin production, and CGR7 was selected to further enhance AC production. Notably, the succinate production of CGR7 reached its maximum of 4.31 g L?1, and then reduced to 2.94 g L?1 at 36 h. The succinate production was further investigated for the enhancement of AC production. 3.4 Reduction of the by-product succinate by deleting ppc/pyc

phosphoenolpyruvate (PEP) carboxylase encoded by ppc and pyruvate carboxylase encoded by pyc, respectively. Therefore, ppc and pyc were individually deleted in CGR7, yielding CGR9 and CGR10, respectively. As shown in Fig. 5A, succinate production was decreased by 81.0% in CGR9 compared to CGR7, with a maximum titer of 0.56 g L?1 at 36 h. In addition, the glucose consumption rate

Two major succinate synthetic pathways in C. glutamicum were taken into consideration (Fig. 1). Their crucial enzymes are

Fig. 6 Optimization of media with 20 g L?1 glucose as the carbon source. CGR7 was cultured in di?erent media at 30 °C and 180 rpm in 250 mL ?asks with a ?nal culture volume of 50 mL.

Fig. 7 D-(?)-Acetoin production from glucose using CGR9 and CGR10 in fed-batch fermentations. The strains were cultured in LBRC medium with around 50 g L?1 glucose at 30 °C and 500 rpm in a 5 L fermenter under aeration of 1 vvm. 1000 g L?1 glucose was added when the glucose concentration dropped below 20 g L?1. A: The fermentation characteristics of CGR9; B: the fermentation characteristics of CGR10.

5698 | Green Chem., 2017, 19, 5691–5702

This journal is ? The Royal Society of Chemistry 2017

View Article Online

Green Chemistry


also much higher than those of CGR7, but were similar to those of CGR9. Notably, the succinate production of CGR10, which carries a deletion of pyc, was almost unchanged compared with CGR7, suggesting that succinate was largely produced from phosphoenolpyruvate (PEP) under aerobic conditions, in a reaction catalyzed by PEP carboxylase encoded by ppc (Fig. 1). Among the secondary by-products, acetate reached 1.26 g L?1 at 36 h. However, due to the complicated synthetic pathway and the relatively low concentration under the conditions of surplus reducing power in AC production, a fermentation optimization strategy was preferred to restrict its amount, rather than further genetic manipulation. Overall, CGR7, CGR9 and CGR10 could be the candidates for further improvement.
Published on 23 October 2017. Downloaded on 17/04/2018 01:20:46.

Fig. 8 D-(?)-Acetoin production from glucose using CGR7 in fed-batch fermentations. CGR7 was cultured in LBRC medium with around 50 g L?1 glucose at 30 °C and 500 rpm in a 5 L fermenter under aeration of 1 vvm. About 50 mL of 1000 g L?1 glucose was added to the medium when the residual glucose concentration dropped to about 5 g L?1, to maintain the glucose concentration between 30 and 70 g L?1.

3.5 Optimization of dissolved oxygen levels and culture medium to achieve industrial-level AC production As mentioned before, CGR7 was suggested to o?er the best performance with regard to D-(?)-AC production. The constitutive promoter used in its construction made it convenient to operate during the entire fermentation process, and economically advantageous due to no requirement for the expensive inducer IPTG. The biosynthesis of AC is regulated by the NADH/NAD+ ratio, and the oxygen supply can influence the intracellular NAD+ level. Because the dissolved oxygen (DO) is associated with agitation speed, the e?ects of this factor on D(?)-AC production were studied in detail. Aiming at selecting a suitable DO level, shaking at 140, 180 and 220 rpm was tested with cultures in CGXIIY medium. As shown in Table 4, at 220 rpm, the productivity of D-(?)-AC was much higher than that at 140 or 180 rpm, indicating that a high DO level was indeed beneficial for e?cient AC production. Therefore, a relatively high DO should be preferred in further fed-batch fermentations. In a previous study, high DO

and growth rate of CGR9 were much higher than those of CGR7 within 24 h, while the AC productivity (0.490 g (L h)?1) was much higher than that of CGR7 at 24 h. However, when glucose was depleted, the D-(?)-AC produced by CGR9 was almost unchanged, even with a slight decrease. As shown in Fig. 5B, the AC titer of CGR10 was 13.43 g L?1 at 36 h when glucose was depleted. Unexpectedly, it then increased to 14.05 g L?1 at 48 h with a yield of 0.777 mol mol?1 glucose, but the productivity fell to 0.293 g (L h)?1. Within 24 h, the glucose consumption rate and growth rate of CGR10 were

Fig. 9 Identi?cation of acetoin and 2,3-butanediol enantiomers by GC-FID. A: D-(?)-Acetoin and L-(+)-acetoin enantiomeric standards had retention times of 10.892 and 11.190 min, respectively; B: fermentation products of CGR7 in CGXII medium; C: fermentation products of CGR7 in CGXIIY medium; D: fermentation products of CGR7 in LBRC medium.

This journal is ? The Royal Society of Chemistry 2017

Green Chem., 2017, 19, 5691–5702 | 5699

View Article Online


Green Chemistry

levels had been shown to have a profound e?ect on the product distribution between 2,3-BD and AC, with AC being preferred under DO above 100 parts per billion ( ppb).38

Published on 23 October 2017. Downloaded on 17/04/2018 01:20:46.

In order to develop a superior medium for the e?cient production of AC, 9 kinds of media were tested in conjunction with CGR7. As shown in Fig. 6, LBRC medium with the addition of urea, acetate and a cheap nitrogen source (corn steep liquor: yeast extract = 5 : 1) o?ered the best AC titer, yield and productivity. This was in accordance with a recent study, which found that the initial concentration of corn steep liquor had remarkable e?ects on both 2,3-BD production and the ratio of 2,3-BD to AC.39 With the improved fermentation conditions, in a flask experiment using LBRC medium at 220 rpm with 40 g L?1 glucose, CGR7 could produce 17.10 g L?1 D-(?)-AC with a yield of 0.875 mol mol?1 glucose. Thus, fed-batch fermentation with a relatively high OD level using LBRC medium for the improved fermentation conditions should be conducted for enhanced D-(?)-AC production. 3.6 Fed-batch fermentations with optimized oxygen supply for high D-(?)-AC production

Fig. 10 Diagram summarizing the systematic approach for the metabolic engineering of C. glutamicum to produce D-(?)-acetoin.

To further improve the AC titer and productivity, fed-batch fermentations were conducted under the optimized fermentation conditions. CGR9 and CGR10 were also tested as candidates in

Table 5

Comparing microbial production of acetoin using di?erent fermentative strains or biocatalysts

Strain Production of optically-pure acetoin C. glutamicum CGR7 K. pneumoniae CGMCC 1.6366 E. coli (pAC-NOX) S. marcescens MG1 E. coli JM109/pAB118 Lactococcus lactis CS4701m Whole cell biocatalysis process E. coli pET-mbdh-nox-vgb Resting cells of B. subtilis E. coli Rosetta/pET28a-dar E. coli BL21(DE3) (pETDuet-ydjLnox) E. coli HB101

Substrate Glucose Glucose Glucose Glucose Glucose Glucose

Titer (g L?1) 96.2 62.3 60.3 21.8 17.5 5.8 86.7 56.7 39.4 41.8 36.7 9.8 9.8 100.1 75.2 71.5 63.0 58.1 62.6 60.5 56.9 55.3 55.2 45.6 53.9 51.2 41.3 25.9 10.8 7.3 5.5 4.3


Yield (g g?1) 0.360 0.140 0.422 0.242 0.438 0.346 0.925 NG 1.970 0.860 0.853 0.968 0.968 0.440 0.360 0.404 NG NG 0.288 NG NG 0.370 0.373 NG 0.359 0.430 0.413 NG 0.282 0.070 0.367 0.072

Productivity (g (L h)?1) 1.30 1.09 1.55 0.73 NG 0.19 0.361 NG NG 3.48 3.05 NG NG 1.67 1.88 1.63 1.05 0.96 0.864 1.44 0.89 1.32 2.69 1.52 0.37 1.42 1.15 0.32 0.90 0.11 0.09 NG

Reference This study 14 17 15 42 43 10 9 16 12 13


meso-2,3-BD Butanedione D-(?)-2,3-BD meso-2,3-BD D-(?)-2,3-BD meso-2,3-BD

Production of racemic acetoin (or optical purity not stated) S. cerevisiae JHY617-SDN Glucose S. marcescens H32-nox Sucrose B. amyloliquefaciens E-11 Glucose B. pumilus XH195 Glucose Sucrose B. subtilis ZB02 Glucose, xylose, and arabinose S. marcescens H32 Sucrose B. subtilis TH-49 Glucose P. polymyxa CS107 Glucose Enterobacter cloacae SDM 45 Glucose Lignocellulosic hydrolysate B. subtilis JNA-UD-6 Glucose B. amyloliquefaciens FMME044 Glucose B. licheniformis MEL09 Glucose K pneumoniae XZF-308 Glucose Enterobacter cloacae DSM01 Glucose Candida glabrata Δadh-Δald-Δbdh-ScPDC1-NOX Glucose B. subtilis BSUL13 Glucose and xylose Clostridium acetobutylicum adc::int (180)(pSY8-alsD) Glucose


24 21 18 19 23 44 45 5 7 46 47 48 49 50 51 22 41

5700 | Green Chem., 2017, 19, 5691–5702

This journal is ? The Royal Society of Chemistry 2017

View Article Online

Green Chemistry


fed-batch fermentations in LBRC medium. As shown in Fig. 7A and B, succinate was almost completely eliminated in CGR9, and the product titer was almost the same as that in CGR10. However, the two strains did not keep producing AC after their titer reached 60 g L?1, when the glucose consumption rate sharply decreased. Therefore, taking the titer, yield, productivity and stability into consideration, CGR7 was selected as the best strain for high-level D-(?)-AC production in fed batch fermentations. With the optimized fermentation conditions, a final titer of 96.2 g L?1 D-(?)-AC with a productivity of 1.30 g (L h)?1 and a yield of 0.736 mol mol?1 glucose was produced by CGR7 (Fig. 8), and its chemical purity surpassed 95% (Fig. 9). Notably, the purity of D-(?)-AC could be more than 99% in CGXII medium. However, when the nitrogen concentration increased, the purity of D-(?)-AC decreased, probably due to the complicated metabolic pathways in C. glutamicum, where the metabolites derived from nitrogen could be transformed into L-(+)-AC. The seed culture was prepared in a 250 mL shake flask with 20 g L?1 glucose and 50 mL CGXIIY medium as described in the Experimental section, and the initial concentration of AC at 0 h was only about 0.2 g L?1 by HPLC (data not shown). The concentrations of the by-products succinate, α-ketoglutarate, acetate and glycerin were 0.73, 1.03, 0.27 and 1.27 g L?1, respectively. No 2,3-butanediol and lactate were detected. During the fermentation, the biomass increased significantly in the first 13 h, with an OD600 of 71.3, which slowly grew to 108 at 53 h, and then fluctuated until the end of fermentation. This fast growth rate and high biomass resulted in a high D-(?)-AC titer and productivity. On the other hand, the yield was not satisfactory due to the high biomass and intense respiration under aerobic conditions. However, AC, and especially its D-(?)-isomer with high optical purity, is a highvalue-added product,11,40 which can make up for the relatively low yield in terms of process economics. Therefore, the strategy used in this study is feasible for industrial-scale D-(?)-AC production (Fig. 10). As far as we know, this is the first report on the production of optically pure D-(?)-acetoin with the highest titer in an economical and environmentally friendly approach, achieved without complicated process steps (Table 5). The pathway was adapted from B. subtilis, while the titers and productivities of AC produced by B. subtilis were much lower than those of CGR7. This study provides a deep understanding of AC metabolism in C. glutamicum, and a strategy for enhancing the D-(?)-AC production through genetic modifications and fermentation optimization in a green chemical process.

potential for industrial-level production of this high-valueadded chemical in the very near future.

Con?icts of interest
The authors declare that they have no competing interests.

This work was supported by the National Natural Science Foundation of China (NSFC-21576191, NSFC-21390201, NSFC21576200 and NSFC-21621004).

Published on 23 October 2017. Downloaded on 17/04/2018 01:20:46.

1 Z. Xiao and P. Xu, Crit. Rev. Microbiol., 2007, 33, 127–140. 2 C. Zhu, T. Shen, D. Liu, J. Wu, Y. Chen, L. Wang, K. Guo, H. Ying and P. Ouyang, Green Chem., 2016, 18, 2165–2174. 3 T. Werpy, G. Petersen, A. Aden, J. Bozell, J. Holladay, J. White, A. Manheim, D. Eliot, L. Lasure and S. Jones, Top value added chemicals from biomass. Volume 1-Results of screening for potential candidates from sugars and synthesis gas, Department of Energy, Washington DC, 2004. 4 Z. Xiao and J. R. Lu, Biotechnol. Adv., 2014, 32, 492–503. 5 L. Zhang, S. Chen, H. Xie, Y. Tian and K. Hu, J. Chem. Technol. Biotechnol., 2012, 87, 1551–1557. 6 L. Gu, T. Lu, X. Li and Y. Zhang, Chem. Commun., 2014, 50, 12308–12310. 7 L. Zhang, Q. Liu, Y. Ge, L. Li, C. Gao, P. Xu and C. Ma, Green Chem., 2016, 18, 1560–1570. 8 S. Wenda, S. Illner, A. Mell and U. Kragl, Green Chem., 2011, 13, 3007. 9 Z. Liu, J. Qin, C. Gao, D. Hua, C. Ma, L. Li, Y. Wang and P. Xu, Bioresour. Technol., 2011, 102, 10741–10744. 10 Z. Guo, X. Zhao, Y. He, T. Yang, H. Gao, G. Li, F. Chen, M. Sun, J.-K. Lee and L. Zhang, J. Microbiol. Biotechnol., 2017, 27, 92–100. 11 T. Bao, X. Zhang, Z. Rao, X. Zhao, R. Zhang, T. Yang, Z. Xu and S. Yang, PloS one, 2014, 9, e102951. 12 Z. Xiao, C. Lv, C. Gao, J. Qin, C. Ma, Z. Liu, P. Liu, L. Li and P. Xu, PLoS One, 2010, 5, e8860. 13 K. Yamada-Onodera, H. Yamamoto, N. Kawahara and Y. Tani, Eng. Life Sci., 2002, 22, 355–362. 14 D. Wang, J. Zhou, C. Chen, D. Wei, J. Shi, B. Jiang, P. Liu and J. Hao, J. Ind. Microbiol. Biotechnol., 2015, 42, 1105– 1115. 15 F. Bai, L. Dai, J. Fan, N. Truong, B. Rao, L. Zhang and Y. Shen, J. Ind. Microbiol. Biotechnol., 2015, 42(5), 779–786. 16 J. Gao, Y. Y. Xu, F. W. Li and G. Ding, Lett. Appl. Microbiol., 2013, 57, 274–281. 17 Q. Xu, L. Xie, Y. Li, H. Lin, S. Sun, X. Guan, K. Hu, Y. Shen and L. Zhang, J. Chem. Technol. Biotechnol., 2015, 90, 93–100. 18 Q. Luo, J. Wu and M. Wu, Process Biochem., 2014, 49, 1223– 1230.



The best engineered strain and fermentation process optimized in this study exhibited high titer, yield and productivity of AC, highlighting the biosafety level I microorganism C. glutamicum as a competitive AC producer with a fantastic

This journal is ? The Royal Society of Chemistry 2017

Green Chem., 2017, 19, 5691–5702 | 5701

View Article Online


Green Chemistry

Published on 23 October 2017. Downloaded on 17/04/2018 01:20:46.

19 P. Xu, Z. Xiao, Y. Du and Z. Wei, US Pat., 8158402, 2012. 20 Y. Liu, S. Zhang, Y.-C. Yong, Z. Ji, X. Ma, Z. Xu and S. Chen, Process Biochem., 2011, 46, 390–394. 21 J. A. Sun, L. Y. Zhang, B. Rao, Y. L. Shen and D. Z. Wei, Bioresour. Technol., 2012, 119, 94–98. 22 T. Chen, W. Liu, J. Fu, B. Zhang and Y. J. Tang, J. Biotechnol., 2013, 168, 499–505. 23 B. Zhang, X. L. Li, J. Fu, N. Li, Z. Wang, Y. J. Tang and T. Chen, PLoS One, 2016, 11, e0159298. 24 S. J. Bae, S. Kim and J. S. Hahn, Sci. Rep., 2016, 6, 27667. 25 L. Feng, Y. Zhang, J. Fu, Y. Mao, T. Chen, X. Zhao and Z. Wang, Biotechnol. Bioeng., 2016, 113, 1284–1293. 26 O. Kirchner and A. Tauch, J. Biotechnol., 2003, 104, 287– 299. 27 T. Chen, N. Zhu and H. Xia, Bioresour. Technol., 2014, 151, 411–414. 28 J. P. L?ken and F. C. St?rmer, Eur. J. Biochem., 1970, 14, 133–137. 29 F. C. Stormer, Methods Enzymol., 1975, 41, 518–533. 30 Y. T. Yang, M. Peredelchuk, G. N. Bennett and K. Y. San, Biotechnol. Bioeng., 2000, 69, 150–159. 31 M. M. Bradford, Anal. Biochem., 1976, 72, 248–254. 32 M. Wang, J. Fu, X. Zhang and T. Chen, Biotechnol. Lett., 2012, 34, 1877–1885. 33 Y. Yan, C. C. Lee and J. C. Liao, Org. Biomol. Chem., 2009, 7, 3914–3917. 34 J. Fu, G. Huo, L. Feng, Y. Mao, Z. Wang, H. Ma, T. Chen and X. Zhao, Biotechnol. Biofuels, 2016, 9, 90. 35 J. Fu, Z. Wang, T. Chen, W. Liu, T. Shi, G. Wang, Y. J. Tang and X. Zhao, Biotechnol. Bioeng., 2014, 111, 2126–2131. 36 H. Ma, H. Liu, N. Zhu and C. Tao, J. Tianjin Univ., 2014, 47, 967–972.

37 L. Eggeling and M. Bott, Handbook of Corynebacterium glutamicum, CRC press, 2005. 38 J. Moes, M. Griot, J. Keller, E. Heinzle, I. J. Dunn and J. R. Bourne, Biotechnol. Bioeng., 1985, 27, 482–489. 39 T. W. Yang, Z. M. Rao, X. Zhang, M. J. Xu, Z. H. Xu and S. T. Yang, Process Biochem., 2013, 48, 1610–1617. 40 F. Bai, L. Dai, J. Fan, N. Truong, B. Rao, L. Zhang and Y. Shen, J. Ind. Microbiol. Biotechnol., 2015, 42, 779. 41 D. Liu, Y. Chen, F. Ding, T. Guo, J. Xie, W. Zhuang, H. Niu, X. Shi, C. Zhu and H. Ying, Metab. Eng., 2015, 27, 107–114. 42 S. Ui, A. Mimura, M. Okuma and T. Kudo, Lett. Appl. Microbiol., 1998, 26, 275–278. 43 J. Liu, C. Solem and P. R. Jensen, Biotechnol. Bioeng., 2016, 113, 2744–2748. 44 J. Sun, L. Zhang, B. Rao, Y. Han, J. Chu, J. Zhu, Y. Shen and D. Wei, Biotechnol. Bioprocess Eng., 2012, 17, 598–605. 45 H. Xu, S. Jia and J. Liu, 2011 International Conference on Consumer Electronics, Communications and Networks (CECNet), IEEE, 2011, 1524–1527. 46 X. Zhang, R. Zhang, T. Yang, J. Zhang, M. Xu, H. Li, Z. Xu and Z. Rao, World J. Microbiol. Biotechnol., 2013, 29, 1783– 1789. 47 Y. Zhang, S. Li, L. Liu and J. Wu, Bioresour. Technol., 2013, 130, 256–260. 48 L. Li, L. Zhang, K. Li, Y. Wang, C. Gao, B. Han, C. Ma and P. Xu, Biotechnol. Biofuels, 2013, 6, 123. 49 X. J. Ji, Z. F. Xia, N. H. Fu, Z. K. Nie, M. Q. Shen, Q. Q. Tian and H. Huang, Biotechnol. Biofuels, 2013, 6, 7. 50 L. Li, K. Li, Y. Wang, C. Chen, Y. Xu, L. Zhang, B. Han, C. Gao, F. Tao and C. Ma, Metab. Eng., 2015, 28, 19–27. 51 S. Li, N. Xu, L. Liu and J. Chen, Metab. Eng., 2014, 22, 32–39.

5702 | Green Chem., 2017, 19, 5691–5702

This journal is ? The Royal Society of Chemistry 2017