Metabolic engineering of Escherichia …homepages.rpi.edu/~koffam/papers/2010_Xu_Koffas.pdf · Metabolic engineering of Escherichia coli for biofuel production Review

future science group 493ISSN 1759-726910.4155/BFS.10.13 © 2010 Future Science Ltd

1Department of Chemical and Biological Engineering, University at Buffalo, The State University of New York, Buffalo, 303 Furnas Hall, NY 14260, USA†Author for correspondence: Tel.: +1 716 645 1198; Fax: +1 716 645 3822; E-mail: [email protected]

Biofuels as an alternative to petroleum fuelsPetroleum-based fuels have provided the means to achieve the economic growth and life improvements experienced since the 19th Century. However, an array of political and environmental concerns have recently arisen, related to energy security and climate change issues. In an effort to address these issues, governments and leading oil companies are now investing in big shifts in the use of energy, with billions in funding aimed towards developing renewable alternative sources that would turn simple and complex sugars into biofuels with the use of engineered microorganisms. Compared with other energy forms (i.e., solar, wind and nuclear energy), biological fuels are more desirable as they have a more favorable carbon footprint and are compatible with cur-rent fuel infrastructure as far as storage, distribution and engine compressibility is concerned [1].

Having evolved for millions of years, micro organisms can produce a variety of chemicals such as alcohols, hydro-carbons and esters, which could serve as an inexhaustible pool for fuel substitutes. However, fuel-like compounds cannot be efficiently synthesized from native organisms for cost-effective application, except for ethanol, which has been produced traditionally from yeast. For this rea-son, a variety of new concepts and approaches must be introduced to make biofuels a sustainable alternative.

Metabolic engineering has emerged as a powerful tech-nological platform for improving the cellular phenotype, such as production ability [2]. Metabolic engineering can be defined as the design of native or entirely new meta-bolic pathways in a cell. It provides an integrative method to manipulate the metabolic pathway within the con-text of the whole metabolism rather than a single reac-tion step (or a single gene) [3]. The user-friendly host, Escherichia coli, has been given preference for metabolic pathway manipulation in the biotechnological realm due to its substrate suitability, extremely fast growth rate, trac-table genetic system, as well as our extensive knowledge of its overall physiology [4]. In this review, we briefly sum-marize some recent advances associated with metabolic pathway modifications in E. coli for production of biofuels and the challenges involved thereafter. Prospective biofu-els such as ethanol, butanol, higher chain alcohols, biodie-sel and isoprenoids and the corresponding biosynthetic pathways are emphasized in this article.

Ethanol production in E. coli using the fermentative pathway Ethanol, currently the most prevalent form of bio-fuel, has gained tremendous success in the last three decades. According to the Renewable Fuels Association, more than 100 biofuel factories in the USA produced

Biofuels (2010) 1(3), 493–504

Metabolic engineering of Escherichia coli for biofuel production

Peng Xu1 & Mattheos AG Koffas†1

Global energy demand and environmental concerns have stimulated increased efforts to develop more sustainable and cost-effective fuels with economical production processes that would make it feasible to replace petroleum-based fuels. The biological synthesis of such fuels relies on the exploitation of the diverse metabolic pathways leading to fuel-like biomolecules and has further opened up the possibility of synthesizing biofuels other than those naturally produced through fermentative pathways. This is because, as a framework, metabolic engineering has made it possible to reconstruct and assemble biosynthetic pathways in user-friendly microorganisms for de novo synthesis of fuel molecules. To highlight the advancements and the tremendous potential that exists, we review the recent progress in engineering of Escherichia coli for biofuel production.

ReviewFor reprint orders, please contact [email protected]

Biofuels (2010) 1(3) future science group494

Review Xu & Koffas

6.5 billion gallons of bioethanol in 2007. To date, the majority of eth-anol is produced from corn starch by yeast. However, corn-derived ethanol has several problems, with the most alarming being the syner-gistic effect of its production with a dramatic increase in food prices. The International Organization for Economic Cooperation and Development (OECD) has warned that the “rapid growth of the bio-fuels industry” could bring about fundamental shifts in agricultural markets and could even “cause food shortages”. Furthermore, corn-derived ethanol may not be as energy efficient as required. This is because a significant amount of energy is required both for growing the corn and running fermentation facili-ties. In fact, estimates suggest that only 25% more energy is obtained from corn-derived ethanol than is used to produce it. Finally, the avail-ability of arable land is another sig-nificant constraint. According to some estimates, even if all the corn planted in the USA were used for ethanol, only approximately 12% of gasoline would be displaced. For all these reasons, the development of cost-efficient microbial systems for conversion of cellulosic material

(wood, agricultural waste and grasses) to ethanol has been under the spotlight around the world [5]. Besides relying on a vast supply of feedstock, the growth of cel-lulosic material requires far less energy than corn. The US Department of Energy has established the goal to replace approximately 30% of petroleum-based fuels with cellulosic ethanol by 2030.

Besides issues related to the carbon source, large-scale replacement of gasoline with ethanol is hampered by other roadblocks; primarily the current producer organism (Saccharomyces cerevisiae) cannot efficiently metabolize pentoses, the main components of ligno-cellulose hydrolysates after pretreatment, to fermentable sugars [6,7]. As such, a significant amount of effort has been devoted towards engineering pentose metabolism in yeast for hemicellulosic ethanol production, which has been covered in other reviews [8,9]. Unlike yeast, E. coli is naturally able to convert a number of pentose and hexose sugars into a mixture of acids (succinic and acetic acid) and ethanol via the heteroethanologenic

pathway. One disadvantage is that this pathway can only produce 1 mole of ethanol from 1 mole of glu-cose, which is relatively inefficient as opposed to the homoethanologenic pathway, the Embden–Meyerhof–Parnas (EMP) pathway by Saccharomyces species and the Entner–Doudoroff (ED) pathway by Zymomonas spe-cies. Consequently, significant effort has been devoted into engineering the existing E. coli pathways for the purpose of efficient production of ethanol.

In heteroethanologenic organisms such as E. coli, ethanol is generated from pyruvate, a reaction catalyzed by pyruvate formate lyase (PFL). At the same time, the cofactor NADPH must be regenerated by reducing the oxidative intermediates (i.e., lactate and fumurate), a pro-cess that is known as mixed acid fermentation. For homo-ethanologenic organisms (e.g., yeast and Zymomonas mobilis), ethanol is generated as the sole product from pyruvate decarboxylase (PDC). Engineering of the homoethanologenic pathway into hexose- and pentose-fermenting E. coli has been reported by Ingram et al. [10]. Briefly, genes encoding pyruvate decarboxylase (pdc) and alcohol dehydrogenase (ADH II [adhII]) from Z. mobilis have been combined into a portable ethanol production cassette under the control of an artificial pet operon and integrated into the chromosome of E. coli B at the pyru-vate formate lyase (pfl) locus to generate E. coli strain KO11. The fumarate reductase gene ( frd) and lactate dehydrogenase gene (ldh) were also knocked out to min-imize succinate and lactate production (Figure 1). The resulting KO11 strain was able to effectively direct carbon flux to higher ethanol production. One explanation for this result is that the introduced PDC from Z. mobilis has much higher affinity to pyruvate (low K

m value), there-

fore eliminating or attenuating the competing pathway for consumption of pyruvate and leading to the exclusive production of ethanol.

Based on transcriptional profiling ana lysis, the same authors investigated the expression levels of 30 genes involved in xylose catabolism in the parental (B) and engineered (KO11) strains; increased glycolytic flux and expression levels of glycolytic genes were dem-onstrated in the strain KO11 during xylose fermen-tation [11]. Recently, a novel mutant of E. coli KO11 (E. coli SE2378) that relies only on native enzymes has been constructed for homoethanologenic fermentation, which could effectively ferment both hexose and xylose to ethanol. This strain has a mutation within the pdh operon (pdhR aceEF lpd), which encodes components of the pyruvate dehydrogenase complex. A novel anaerobic pathway that involved pdh was also proposed to explain the homoethanologenic mechanism [12].

E. coli strain KO11 can achieve high ethanol titer in rich media; however, this strain performs poorly in minimal media. The low ethanol production by KO11

Key terms

Metabolic engineering: Practice of optimizing metabolic pathways and regulatory networks within cells by recombinant DNA technology in order to increase the cells’ production capability

Escherichia coli: Facultative anaerobic bacterium, that has been long used as a model organism for genetic modification and industrial fermentation

Homoethanologenic fermentation: Anaerobic conversion of pyruvic acid into ethanol and carbon dioxide through the glycolysis pathway with concomitant oxidation of NADH to NAD+

Catabolite repression: Repression (inactivation) of certain sugar-metabolizing operons (i.e., the lac operon) in favor of glucose utilization when glucose is the predominant carbon source in the environment of the bacterial culture

CoA-dependent pathway: Pathway that requires the involvement of coenzyme A as an acyl carrier protein, plays a key role in the synthesis and oxidation of fatty acids, as well as the oxidation of pyruvate in the citric acid cycle

Synthetic network: Non-native or artificial metabolic network, which is designed or constructed for novel biological functions and systems

Systems biology: Integration of genome-wide biological approach and small-scale computational approach to understand the dynamics of the system, global properties (regulation and robustness) and other complex properties

O

HOOH

OH

CH2OH

OH

O

OH

OH

OH

HOCH2

D-Xylose D-Glucose

O

OH

OH

OH

O32 POCH2 O

HOOH

OH

CH2OPO32

OH

ATP

ADP

ATP

ADP

Xylulose-5-P

OOH

OPO32

Glyceraldehyde 3-P

Pentose phosphate pathway Glycolysis (EMP) pathway

NAD+

NADHO

HOOPO3

2

PEPO

O

OH

Pyruvate O

OHO

OHO

Oxaloacetate

NADH

NAD+

HO

OH O

OHO

Malate

O

OH

OH

Lactate

Glucose-6-P

O OH

H

Formate

SCoA

O

Acetyl-CoA

OPO32

O

Acetyl phosphate H

O

AcetaldehydeNADH

NAD+

OH

Ethanol

ADP

ATP

OHOAcetate

H

O

AcetaldehydeNADH

NAD+

OH

Ethanol

HOO

OHO

Succinate

HOO

OH

Fumarate

O H

H

CO2

NADH

NAD+

CO2pdc

adhII

ldh

pta

PFL

frd

PEPC

AK

GAPD

PK

HKXK

TCA cycle

Metabolic engineering of Escherichia coli for biofuel production Review

future science group www.future-science.com 495

in minimal media was identified to be due to the suboptimal partition-ing of pyruvate for biosynthesis and the corresponding inhibitory effect of NADH on citrate synthase [13,14]; therefore, the biomass yield and etha-nol titer could be partially restored by addition of pyruvate or keto glutarate to oxidize the excess NADH. This hypothesis has been further verified by expression of a NADH-insensitive citrate synthase (NADH feedback-resistant mutant of citrate synthase) from Bacillus in E. coli.

For mixed-substrate fermenta-tion, organisms will utilize glucose in preference to other sugars, which is a global regulatory mechanism and well known as the catabolite repressor. Relieved or eliminated catabolite repression would be bene-ficial for primary-product formation in mixed-substrate fermentation. Accordingly, a catabolite repression-deficient strain that can simultane-ously utilize glucose, arabinose and xylose was constructed by introduc-ing a mutation to the phosphoenol pyruvate–glucose phosphotrans-ferase system (ptsG -) in E. coli [15]. This 6-pyruvoyl-tetrahydropterin synthase-deficient strain was able to efficiently convert complex sub-strate mixtures to ethanol. The relaxed control of the cAMP-CRP regulatory system over transcrip-tion was also confirmed by global gene expression (DNA array) ana-lysis in the engineered xylose-grown strain [16]; and the measured mRNA levels of many genes showed good consistency with the calculated flux distribution.

Metabolic network redundancy evolved by natural selection has allowed organisms to survive in diverse environments. The resulting biological complexity is manifested (among others) by an intricate metabolic network comprised of highly interconnected reac-tions; however, it was recently shown that this network can be reduced to uniquely organized pathways with minimal metabolic functionality, through which a minimal cell that is committed to produce a certain compound, such as ethanol, in the most efficient way can be constructed [17]. More specifically, based on the

nongrowth-associated production and balanced redox state assumption, the authors employed an iterative algorithm to optimize the functionality of the central metabolic network, resulting in a reduced network that consists of six pathways (four for substrate decompo-sition and two for cell anaerobic growth) from over 15,000 possible combinations. The constructed strain is free from catabolite repression and able to simultane-ously convert pentoses and hexoses to ethanol at the theoretical yield. Based on elementary mode ana lysis, an E. coli strain with nine gene knockouts for efficient

Figure 1. Homoethanologenic pathway in engineered Escherichia coli. The black arrows represent the native pathway; dash lines indicate multienzymatic steps; the underlined italicized gene names represent the pathway that has been overexpressed and gene names with strikethrough lines represent pathways that have been knocked out. AK: Acetylphosphate kinase; EMP: Embden–Meyerhof–Parnas; GAPD: Glyceraldehyde-3-phosphate dehydrogenase; HK: Hexose kinase; PEP: Phosphoenolpyruvate; PEPC: Phosphoenolpyruvate carboxylase; PFL: Pyruvate formate lyase; XK: Xylulose kinase.


Review Xu & Koffas

conversion of glycerol to ethanol was designed [18]. In defined medium, the evolved strain could convert 40 g/l of glycerol to ethanol in 48 h with 90% of the theoreti-cal ethanol yield, which demonstrated the usefulness of this analytical tool.

Final ethanol titer is an important determinant of the overall cost because it significantly affects the down-stream process (distillation), another factor for competi-tive production of ethanol [2]. Low ethanol titer could be caused by many factors, such as limited ethanol-tol-erance capability and the inhibitory effect of feedstock hydrolysates (i.e., aromatics, furfurals, furan derivatives and phenolics) on cell growth. To address this issue, several attempts have been made to develop ethanol-tol-erant and byproduct-resistant strains. The ethanol-toler-ant strain E. coli LY01, which was obtained by directed evolution of homoethanologenic E. coli KO11, could produce up to 60 g/l ethanol [19]. Gene arrays were used to identify the expression differences between strain LY01 and KO11 and indicated that increased tolerance was related to higher expression of transcriptional regu-lator gene ( fnr) and osmolyte synthesis gene (gcv and betIBA) [20]. Accumulating evidence has suggested that a complex phenotype such as the ethanol-tolerant trait could be elicited by a handful of genes rather than a single gene. As such, Alper and Stephanopoulos dem-onstrated the feasibility of using global transcription machinery engineering (gTME) to optimize the com-plex phenotype of producer organism [21,22]; specifically, random mutations were introduced to the components of transcriptional regulatory machinery (i.e., s70, a 70-KDa transcription initiation factor), mutations that would not have been attainable through rational design and traditional strain improvement. Essentially, this gTME paradigm allows for a combination of meta-bolic engineering and protein evolution approaches to improve the cellular phenotypes such as ethanol toler-ance and metabolite overproduction. Other studies that will not be covered in this review have demonstrated that immobilization of recombinant cells in continu-ous fluidized bed culture can increase the ethanol tol-erance and phenotypic stability [23]. Inhibitory effects of toxic byproducts (i.e., furfural) in the hemicellulose hydrolysates on cell growth are another barrier for high ethanol production. Recently, Miller et al. demonstrated that elimination of the competing pathway for the con-sumption of NADPH could endow the E. coli strain EMFR 9 with increased furfural-resistant capability, simply by silencing the genes yqhD and dkgA, involved in the reduction step of furmural degradation [24].

Despite its great promise, the exuberance over ethanol has significantly diminished in the past few years due to its several disadvantages [25]. To start with, ethanol was never designed to be a biofuel: a two-carbon molecule,

ethanol has only two-thirds the energy content of gaso-line. Since ethanol mixes with water, a costly distillation step is required at the end of the fermentation process. Finally, because ethanol is more easily contaminated with water than regular hydrocarbons, it cannot be transferred using the current petroleum pipelines but instead must be shipped in specialized trucks. For these reasons, the production of longer chain molecules has been pursued, which could perhaps circumvent some of ethanol’s drawbacks.

Butanol & isopropanol production in E. coli using the CoA-dependent pathwayButanol is an alcohol with thermodynamic and physi-cochemical properties strongly resembling those of gasoline. It has more energy than ethanol, with a gal-lon of butanol containing approximately 90% as much energy as a gallon of gasoline. Butanol can be shipped in unmodified gasoline pipelines and it can be blended with gasoline in higher percentages than ethanol with-out requiring modifications to engines. Biobutanol production studied to date is carried out exclusively by a number of Clostridia species through acetone–buta-nol–ethanol (ABE) pathway. ABE fermentation was originally developed for solvent production in the 1920s and currently has attracted particular interest in the sci-entific community and industrial field for its biobutanol production potential. Most notably, the Clostridium spe-cies can produce a variety of cellulases that can be used directly for conversion of cellulosic feedstocks to buta-nol and ethanol. Recently, Cobalt Biofuels has invented a high-throughput screening technique for identifying high-butanol producers by expressing a luminescent protein in Clostridium bacteria [101]; a startup company, Qteros, has claimed the discovery of a Clostridrium bac-terium capable of converting plant biomass to ethanol at 7%, which is “more economical than any other pro-cess to date” [102]. Such recent efforts in engineering Clostridium species for butanol production have been covered in other reviews [26].

Nevertheless, Clostridia are not ideal produc-ers, due to relatively poor characterization of their genetic system, lack of suitable tools to manipulate their metabolism and their sensitivity to butanol tox-icity [27]. Consequently, it is possible to reconstruct the butanologenic pathway in a user-friendly host and facilitate the homo fermentative production of butanol [28]. For example, Inui et al. have introduced a synthetic butanol pathway into the E. coli system, composed of Clostridium acetyl-CoA acetyltransferase, b-hydroxybutyryl-CoA dehydrogenase, 3-hydroxybu-tyryl-CoA dehydratase, butyryl-CoA dehydrogenase, butyraldehyde dehydrogenase and butanol dehydro-genase [29]. To further improve the butanol yield, the

Glycolysis (EMP) pathwayNAD+

NADH

O

OHO

OHO

Oxaloacetate

O

OH

OH

Lactate

NADPH

NADP+

OHOAcetate

HOO

OHO

Succinate

HOO

OH

Fumarate

O H

H

CO2

ldh

frdABCD

ATP

ADP

O

HOOH

OH

CH2OH

OH

D-Glucose

O

O

OH

PyruvateO OH

H

FormateCO2 + H2 NAD+

NADHCO2

CoA

SCoA

O

Acetyl-CoA

OPO32

O

Acetyl phosphateO O

SCoA

Acetoacetyl-CoA

SCoA

O

Acetyl-CoAO O

OHAcetoacetate

O

Acetone

OHIsopropanol

TCA cycle

NADH

NAD+

OH O

SCoA

3-Hydroxybutyl-CoAO

SCoA

Crotonyl-CoAO

SCoA

Butyryl-CoA

O

ButyraldehydeH

n-ButanolOH

NADH

NAD+

NADH

NAD+

NADH

NAD+

fdhF

sadh

thl

hbd

crt

bcd

adhE

adhE

atoAD

adc



author proposed to delete the native pathway competing for both carbon flux and reducing power, including ldhA, fdhF and frdABCD genes (Figure 2). Finally, the engineered E. coli BUT 2 strain could pro-duce up to 16.2 mM (1200 mg/l) of butanol using glucose as the sole carbon source.

In Clostridium acetobutylium, the genes (crt, bcd, etfAB and hbd) encoding for enzymatic reactions for sequential conversion of aceto-acetyl-CoA to butyryl-CoA (cata-lyzed by crotonase, butyryl-CoA dehydrogenase, electron transfer proteins and 3-hydroxybutyryl-CoA dehydrogenase, respectively) have been assembled together in the polycistronic BCS operon [30]. Recently, Nielsen et al. reported that the polycistronic expression of butanol pathway genes in E. coli could lead to the production of 34 mg/l butanol, whereas indi-vidual expression of the pathway genes improved titers to 200 mg/l. It is possible that this improvement could be attributed to the coor-dinated functionality of the indi-vidual enzymes. To further improve butanol titers, formate dehydroge-nase ( fdh1) from Saccharomyces cerevisiae and the native glyceral-dehyde 3-phosphate dehydroge-nase (gapA) were over-expressed to regenerate NADH and enhance the glycolytic flux, which could elevate final titers to 580 mg/l [31]. They also demonstrated the applicability of engineering butanol pathway in microorganisms other than E. coli and yeast, by engineering the pathway in Pseudomonas putida and Bacillus subtilis.

Isopropanol is a secondary alcohol that can be natu-rally synthesized in large quantities by some microbes via threonine catabolism. Isopropanol is also used instead of methanol or ethanol to esterify various fatty acids to produce a diesel with a higher freezing point [32]. E. coli expressing the combination of C. ace-tobutylicum thl, E. coli atoAD, C. acetobutylicum adc and C. beijerinckii adh resulted in an isopropanol produc-tion of 81.6 mM [33]. Jojima reported that genetically engineered E. coli harboring the isopropanol-produc-ing pathway consisting of four genes from Clostridium

acetobutylicum and one primary–secondary alcohol dehydrogenase from C. beijerinckii, could produce up to 227 mM of isopropanol from glucose under aerobic-fed batch fermentation, which is the highest isopropa-nol titer ever reported [33]. Nevertheless, this does not exclude the possibility of other gene combinations that might improve isopropanol production further.

Higher carbon chain alcohol production in E. coli using the keto acid pathwayIn contrast to the fermentative pathways, the non-fermentative keto acid pathways for production of higher chain alcohols has attracted particular atten-tion recently. Higher chain alcohols, however, are not

Figure 2. Butanol and isopropanol production in engineered Escherichia coli. The black arrows represent the native pathway; dash lines indicate multienzymatic steps; the underlined italicized gene names represent the pathway that has been overexpressed and gene names with strikethrough lines represent pathways that have been knocked out. EMP: Embden–Meyerhof–Parnas.


Review Xu & Koffas

commonly synthesized in large quantities by naturally occurring organisms, with the exception of n-butanol. To address this problem, Liao and co-workers at the University of California, CA, USA have successfully demonstrated the feasibility of constructing synthetic pathways that allow production of an array of higher alcohols with carbon number ranging from four to six (i.e., isobutanol, 1-butanol, 2-methyl-1-butanol, 3-methyl-1-butanol, 2-phenylethanol and 3-methyl-1-pentanol) [34,35], which could be used as fuel alternatives. In this work, non-natural metabolism for conversion of keto acids to higher chain alcohols has been achieved by heterologous expression of keto acid decarboxylase (KDC) and ADH in E. coli [34]. In the branched-chain amino acid (i.e., threonine, valine, leucine and isoleu-cine) pathway (also known as the Ehrlich pathway), various keto acid intermediates (i.e., 2-ketobutyrate, 2-keto-3-methyl-valerate, 2-keto-isovalerate, 2-keto-4-methyl-pentanoate and phenylpyruvate) are generated within the cell. These could be exploited as a direct precursor for fuel bio synthesis. The foreign pathway, oxidative decarboxylation of keto acids to the corre-sponding carboxylic acids by KDC and the sequen-tial reduction of carboxylic acids to various alcohols by ADH, was successfully constructed to accomplish this goal (Figure 3). Most distinctly, this engineered bio-synthetic route for biofuel production circumvents the need to involve CoA-dependent intermediates, which could avoid withdrawal of CoA that is essential for pro-tein synthesis and cell metabolism [36]. To select the most powerful and versatile KDC, a range of KDCs from different organisms were tested for their capabil-ity to metabolize the endogeneous keto acids in E. coli. Production levels of isobutanol that could be achieved by this strategy were as high as 5.3 mM when Lactoccus Kivd was employed [34].

One possible strategy in metabolic engineering is to increase the precursor availability through overexpres-sion of the upstream genes and/or deletion of the com-petitive pathways. Enhanced precursor availability will make the reaction kinetically favorable and could pos-sibly lead to large amounts of desired products. Indeed, the wild-type E. coli only produces trace amounts of keto acid intermediates. For efficient production of higher chain alcohols, the existing E. coli metabolic pathways were genetically modified to increase the pool of the specific keto acids. For example, the authors overex-pressed an array of genes, including alsS from Bacillus, the native genes ilvA and leuABCD, which could lead to an approximately fivefold production increase over the strain without gene expression. On the other hand, genes responsible for competitive pathways were knocked out to eliminate byproduct formation, including ldhA, frdAB, pta and pflB (Figure 3), which could further

increase the isobutanol titer to 300 mM, approximately 86% of the theoretical maximum. In this synthetic approach, most of the carbon flux is diverted into the final product of interest, which would be efficient and economic for industrial applications [34].

Directed evolution, together with metabolic engi-neering, has created a new frontier for the discovery and production of useful compounds, which is far beyond the conventional concept of natural selection. As evi-dent in a recent article by Zhang et al., the authors were able to extend the branched-chain amino acid pathways to produce longer chain keto acids and alco-hols by using a non-natural metabolic approach. By comparing the conformation of the substrate-binding region of Lactococcus lactis ketoisovalerate decarboxylase (KVID), Zymomonas mobilis pyruvate decarboxylase (ZmPDC) and Enterobacter cloacae indolepyruvate decarboxylase (IPDC) using a homology model, the authors concluded that the bulkier size of a side chain can account for the wide substrate spectrum of these KDCs and suggested that substitution of related amino acids with smaller hydrophobic side chains might be able to result in higher substrate specificity. Using this rational design, they successfully obtained a KVID variant, with a specificity constant 40-fold higher than that of the wild-type KVID. Analogous to this strat-egy, they also engineered the chain elongation activity of 2-isopropylmalate synthase by enlarging the sub-strate binding pocket of LeuA to relieve the steric hin-drance. The resulting mutant was shown to be more active on longer-chain substrates. By this combinato-rial approach, non-natural alcohols such as 3-methyl-1-pentanol could be efficiently produced for the first time at 794 mg/l [35].

Instead of using the naturally evolved fermentative pathways for alcohol production in microbes, n-butanol and isopropanol could also be synthesized by expanding the keto acid pathway in E.coli [37]. Upon overexpression of kvid (L. lactis) and adh2 (Saccharomyces cerevisiae) and the E. coli ilvA, leuABCD, thrAfbrBC, the engi-neered E. coli strain achieved co-production of butanol and propanol. Downregulation of the threonine bio-synthesis and removal of the diverging pathways cata-lyzed by MetA (homoserine O-succinyltransferase) and threonine dehydrogenase (Tdh) could further improve isopropanol production up to 2 g/l. However, elimina-tion of feedback inhibition of end products on LeuA failed to increase the production, presumably because the intracellular leucine level is insufficient to cause an adverse effect on LeuA [37]. An alternative enzyme, citra-malate synthase (CimA) from Methanococcus jannaschii, which can directly convert pyruvate to 2-ketobutyrate, has recently been engineered for producing propanol and butanol [38]. Using a growth-based evolutionary


NADH

O

OHO

OHO Oxaloacetate

O

OH

OH

Lactate

HOO

OHO

Succinate

ldh

frdAB

ATP

ADP

O

HOOH

OH

CH2OH

OH

D-Glucose

O

O

OH

PyruvateO OH

H

Formate

SCoA

O

Acetyl-CoA

OPO32

O

Acetyl phosphate

NADHNAD+

NH2

O

OHHO

HomoserinemetA

O-succinyl-homoserineOH

NH2

O

OHL-Threonine

tdh

2-amino-oxobutanoateNH2

O

OH

O

O

O

OH

2-ketobutyratekivd

adh2OH

Propanol

O

OH

O

2-keto-3-methyl-valerate

O

OH

O

2-keto-4-methyl-hexanoate

kivdadh2

kivdadh2

OH

2-methyl-1-butanol

OH

3-methyl-1-pentanol

O

OH

O

kivd

adh2

OH

1-butanol

O O

OHOH

2-acetolactate

O

O

OH

2-ketoisovalerate

kivd

adh2

OH

Isobutanol

NADHNAD+

NADHNAD+

2-Ketovalerate

O

OH

Okivd

adh2

3-methyl-1-butanol

OH

2-keto-4-methyl-pentanoate

pflB

pta thrA fbr

thrBC

ilvA

ilvHCD

leuABCD

leuABCD

leuABCD

alsS

ilvCD



strategy, the best CimA variant, with a truncated C-terminal domain, exhibited both improved catalytic activity and insensitivity to feedback inhibition by isoleucine, which enabled 9.2- and 21.9-fold higher production of propanol and butanol compared with the strain expressing the wild-type CimA. With this effort, a production level of 3.5 g/l isopropanol and 524 mg/l butanol was achieved.

Other alcohol derivatives such as 2-methyl-1-buta-nol (2MB) and 3-methyl-1-butanol (3MB) could also be synthesized by harnessing the keto acid biosynthetic pathway [39,40]. Cann and Liao investigated the speci-ficity and diversity of enzymes catalyzing key parts of

the isoleucine biosynthetic pathway, indicating that ilvGM from Salmonella typhimurium and ilvA from Corynebacterium glutamicum improved 2MB pro-duction, which is the highest production reported in E. coli. Further strain improvement included overex-pression of the native threonine biosynthetic operon (thrABC) without the transcription regulatory element and knockout of the competing pathways upstream of threonine production (DmetA and Dtdh), which led to an ultimate titer of 1.25 g/l 2MB in 24 h [39]. For 3MB, plasmid-based expression of valine and leucine biosynthetic pathway genes could lead to a produc-tion of 67 mg/l. It has also been demonstrated that

Figure 3. Synthetic networks for the nonfermentative alcohol production in engineered Escherichia coli. The black arrows represent the native pathway; dash lines indicate multienzymatic steps; the underlined italicized gene names represent the pathway that has been overexpressed and gene names with strikethrough lines represent pathways that have been knocked out.


Review Xu & Koffas

the main bottleneck to 3MB production in E. coli is due to feedback inhibition of leuA gene, caused by free leucine. In addition, 3MB and isobutanol compete for the same precursor, 2-keto-isovalerate (KIV). In this regard, it is necessary to eliminate the feedback inhibition and competing pathways and increase the specificity of LeuA on KIV. When these approaches were combined, the engineered E. coli could achieve a final titer of 1.28 g/l 3MB in 28 h [40].

Through a combination of synthetic pathway and directed evolution, the previously reviewed work has opened up an unexplored frontier for production of an array of higher chain alcohols, which could hold a greater promise for gasoline alternatives than ethanol. One of the hallmarks of this research is that the inter-mediary metabolism of user-friendly strain E. coli was expanded and high production of this fuel-like alcohol was achieved for the first time.

Biodiesel production in E.coli using the fatty acids biosynthesis pathway Biodiesel is defined as a fuel comprised of mono-alkyl esters of long-chain fatty acids derived from vegeta-ble oils or animal fats. It is a possible substitute for petroleum-based fuel and is produced from plant oils through transesterification of short-chain alcohols, primarily methanol and ethanol. The resulting prod-ucts are also known as fatty acid methyl esters and fatty acid ethyl esters (FAEEs) [41]. Despite its several advantages, including biodegradability, low toxicity, reduced emission of carbon monoxide and attractive fuel-like properties, large-scale application of biodiesel is limited by the availability of vegetable oil feedstocks and geographical and seasonal restrictions, resulting in a conflict between its use as a food component and its use as a biofuel, similar to the conflict previously described for corn-derived ethanol. Biotechnological approaches for the production of biodiesel, in addi-tion to chemical and biochemical methodologies, can potentially ease this conflict. Many autotrophic micro-algaes, microorganisms such as Cryptococcus albidus, Lipomyces lipofera, Lipomyces starkeyi, Rhodosporidium toruloides, Rhodotorula glutinis, Trichosporon pullu-lan, Yarrowia lipolytica and bacteria belonging to the Gordonia, Rhodococcus and Nocardia sp. are able to accumulate oils [42,43], with a wide range of molecules utilized by these microorganisms as carbon sources. The utilization of such substrates makes the process very attractive and possibly economically acceptable.

In addition, extensive efforts have been made in order to engineer the efficient production of bio-diesel in genetically tractable hosts such as E. coli. Kalscheuer et al. achieved the biosynthesis of FAEEs in metabolically engineered E. coli. For achieving this,

the homoethanologenic pathway of Zymomonas mobi-lis, consisting of pyruvate decarboxylase and alcohol dehydrogenase, was introduced into E. coli to produce ethanol. This resulting ethanol was further esterified to FAEEs by heterologous expression of unspecific acyltransferase from Acinetobacter baylyi strain ADP1 (Figure 4). E. coli expressing these enzymes reached a production titer of 1.28 g/l, representing 26% of cellular dry mass by using glucose and oleic acid as substrates. By co-expression of an alcohol-producing bifunctional acyl-CoA reductase from the jojoba plant and a bacterial wax ester synthase from A. baylyi strain ADP1, the engineered E. coli was unexpectedly found to produce fatty acids butyl esters in the presence of oleate [44].

An enhanced fatty acid pool would be beneficial for biodiesel production. Acetyl-CoA carboxylase (ACC), which catalyzes the reaction from acetyl-CoA to malonyl-CoA (the main fatty acid precursor), has been assumed to be the major limiting step for the synthesis of an array of economically important com-pounds such as fatty acids, polyketides and flavonoids. For example, overexpresion of ACC has proven effec-tive for increasing the fatty acid biosynthesis rate in E. coli [45]. Augmentation of the intracellular malonyl-CoA pool through the coordinated overexpression of four ACC subunits (accBCAD) from Photorhabdus luminescens could result in an increase in flavanone production up to 576% [46]. By knocking out fabD and over expression of acc and a plant-derived thioesterase, Lu et al. have achieved efficient production of fatty acids (2.5 g/l) in an engineered E. coli [47], which could be readily used for biodiesel production. Recently, a recombinant E. coli expressing the Proteus sp. lipK (a novel alkaline lipase) has been constructed as a whole-cell biocatalyst for conversion of methanol and olive oil to biodiesel [48].

Isoprenoid production in E. coli using the mevalonate pathway Isoprenoids are a diverse family of metabolites syn-thesized from the C5 units isopentenyl pyrophos-phate (IPP) and dimethylallyl pyrophosphate [49]. An array of natural products such as monoterpene (C10), sesquiterpene (C15), diterpene (C20) and tri-terpene (C30) are derived from this pathway. Most notably, these compounds have been known for their pharmaceutical or nutritional value. Engineered microbes have been developed for overproduction of artemisinic acid [49], b-carotene [50], lycopene [51] and terpenoids [53]. These strains are now being adapted for biosynthesis of isoprenoid-based fuels, such as cyclic alkanes, alkenes and alcohols (Figure 5). For example, introduction of nudF into engineered E. coli with


NADHATP

ADP

O

HOOH

OH

CH2OH

OH

D-Glucose

O

O

OH

Pyruvate

NAD+

NADHCO2

CoA

SCoA

O

Acetyl-CoA

OHEthanol

Malonyl-CoA

Fatty acid synthesis

O

OHR Fatty acid

CO2

accACP

CoA

ACP

CoA

O O

SCoA

HO

O

SACP

O O

SACP

HO

CoA

R

O

SCoAFatty acyl-CoA

Acetyl-ACPMalonyl-ACP

OHEthanol

OR

O

OR

O

acetyltransferase

Fatty acid ethyl esterFatty acid butyl ester

Acyl-CoA synthetase

pdc

adhfabH fabD

OH

Butanol

Acyl-CoA reductase

Ester synthase



mevalonate-based isopentenyl pyrophosphate biosyn-thetic pathway resulted in the production of isopen-tenol at 110 mg/l [53]. Through the upregulation of HMG-CoA reductase activity, Pitera et al. successfully rebalanced the carbon flux through the heterologous pathway and, at the same time, achieved increased mevalonate production [54]. Although the production level is low, the longer cyclic or branched-chain iso-prenoids may be suitable diesel and jet fuel substitutes in the long run, due to their high energy density and low freezing point.

Future perspective in engineering E. coli for biofuel productionThe targeted manipulation of metabolic pathways has made it possible for genetically tractable micro-organisms, such as E. coli, to become promising pro-ducers of desired fuel alternatives. However, several challenges stand in the way of making such biofuel production by recombinant E. coli economically feasible, the most important of which is achieving high production titers. Great strides have been made towards this direction, with strategies that involved both the manipulation of the native regulatory net-work of intermediary metabolism as well as the pro-duction pathways of interest themselves. In light of the complexities of metabolism, it is becoming increas-ingly important to use such hybrid approaches in order to not only improve productivity but also to produce novel compounds of potentially superior value as biofuel molecules [55,56]. It is important, however, to note that the use of more global approaches, such as systems biology, has hardly been demonstrated in the case of biofuel production. Such approaches could potentially lead to the development of strains with higher titers by targeting enzymes and pathways that are seemingly unrelated to the metabolic pathways of interest. It has recently been demonstrated, for exam-ple, how stoichiometric modeling of the genome-wide E.coli metabolism and flux balance analysis resulted in significant increases in the intracellular flux towards malonyl-CoA, a key precursor for biodiesel produc-tion [57]. Such systems approaches could also help develop strategies for satisfying cellular energetic and cofactor requirements in the case of biofuel produc-tion [58]. Such requirements arise from the fact that cells will not waste energy and resources to overpro-duce a certain substance; in addition, introduction and expression of non-native pathways will result in meta-bolic imbalance. For these reasons, regeneration of reducing power (NADH or NADPH) and coenzymes is crucial for product formation [25,59]. In this regard, constraint-based flux balance ana lysis can result in a genome-wide rewiring of cellular metabolism, in order

to construct the most efficient cellular phenotypes. Such rewiring can be coupled with recently devel-oped metabolic engineering tools, such as promoter evolution [60], which would allow the fine-tuning of mRNA levels as well as tools that allow better post-transcriptional control (synthetic red blood cells and tunable intergenic regions) in order to finally achieve even better biofuel titers [61].

ConclusionMetabolic engineering provides an important plat-form for improving the cell’s capability to produce fuel alternatives and several examples have demon-strated this fact in the case of E. coli. Despite these

Figure 4. Biodiesel production in engineered Escherichia coli. The black arrows represent the native pathway; dash lines indicate multienzymatic steps and the underlined italicized gene names represent the pathway that has been overexpressed. EMP: Embden–Meyerhof–Parnas.

Glycolysis (EMP) pathway

O

HOOH

OH

CH2OH

OH

D-Glucose

SCoA

OAcetyl-CoA

NADH

NAD+

SCoA

O

O O

SCoA

Thiolase

SCoA

OO

HO

OH 3-hydroxy-3-methylglutaryl-CoA

HMG-CoA synthase

HMG-CoA reductase

O

HO

OH

OHMevalonate

ATP

ADPO

HO

OH

OPO32? Mevalonate-5-P

Mevalonate kinase

ATP

ADPO

HO

OH

OPO2OPO33?

Phosphomevalonate kinase

Mevalonate-5-pyrophosphate

OPO2OPO33?

Isopentenyl-5-pyrophosphate (IPP)

Mevalonate-5-PP decarboxylase

OPO2OPO33?

Isopentenyl-PP isomerase

OH

OHDimethylallyl-PP (DMPP)

3-methyl-2-buten-1-ol

3-methyl-3-buten-1-ol

nudF

nudFIsopentenol

Isopentenol


Review Xu & Koffas

accomplishments, substantial challenges remain in order to make E.coli-based processes economically feasible. It is expected that most of these challenges will be addressed in the future through a combina-tion of system biology and metabolic pathway evo-lution strategies. However, the reliance on either corn- or cellulose-derived sugars as carbon sources for E.coli growth and production could potentially be the one key limitation in the industrial applica-tion of such technologies. In addition, the sharp increase in global temperatures that has been linked to greenhouse gases has generated an urgent need to use processes that rely on CO

2 sequestration. To date,

biotic solutions for CO2 sequestration have empha-

sized photosynthetic systems, using algae or photo-synthetic microorganisms, but nonphotosynthetic, microbial CO

2 fixation from recombinant E. coli

could also be a possibility in the immediate future. Such nonphotosynthetic CO

2 sequestration can uti-

lize autotrophic carbon fixation pathways such as the 3-hydoxypropionate/4-hydroxybutyrate cycle and the propionyl-CoA carboxylase cycle that have been iden-tified in archaea [62–64]. Because of the great limita-tions in employing the native archaea organisms, we expect that metabolic engineering will play a key role in the near future towards the development of recom-binant E. coli and even other tractable microorganisms that would be able to functionally expresses the PCC and 4-hydroxybutyrate/4-hydroxybutyrate cycles and as such perform such conversions.

Financial & competing interests disclosureThe authors have no relevant affiliations or financial involvement with any organization or entity with a financial interest in or finan-cial conflict with the subject matter or materials discussed in the manuscript. This includes employment, consultancies, honoraria, stock ownership or options, expert t estimony, grants or patents received or pending, or royalties.

No writing assistance was utilized in the production of this manuscript.

Executive summary

� Biofuels are becoming increasingly attractive as an alternative to petroleum-based fuels. Breakthroughs in metabolic engineering have made it possible to manipulate the diverse metabolic pathway and develop more efficient processes for biofuel production.

� Tremendous work has been made to engineer the fermentative pathway of Escherichia coli for ethanol production. Although ethanol appears to be promising, there are some inherent problems with its use owing to its physicochemical properties (low energy content and high hygroscopicity).

� Biofuels with higher carbon chains are emerging as the next generation molecules of energy, which may find broader application compared with ethanol. Pathway engineering of E. coli to synthesize butanol, isopropanol, biodiesel and isoprenoids has demonstrated this potential.

� Expanding the keto acids pathway to produce non-natural higher chain alcohols in E. coli represents a remarkable success of metabolic pathway evolution. Through a combination of protein rational design and metabolic engineering, the authors achieved the efficient production of an array of novel alcohols, which holds great promise for future applications.

� Most of the challenges that stand in way to make biofuel economically feasible will be addressed through a combination of systems biology and directed evolution.

Figure 5. Representation of isopentenol production in engineered Escherichia coli. The black arrows represent the native pathway; dash lines indicate multienzymatic steps and the underlined italicized gene names represent the pathway that has been overexpressed. EMP: Embden–Meyerhof–Parnas.



BibliographyPapers of special note have been highlighted as:n of interestnn of considerable interest

1 Rude MA, Schirmer A. New microbial fuels: a biotech perspective. Curr. Opin. Microbiol. 12, 274–281 (2009).

2 Stephanopoulos G. Challenges in engineering microbes for biofuels production. Science 315, 801–804 (2007).

3 Stephanopoulos G, Gill RT. After a decade of progress, an expanded role for metabolic engineering. Adv. Biochem. Eng. Biotechnol. 73, 1–8 (2001).

4 Chemler JA, Koffas MA. Metabolic engineering for plant natural product biosynthesis in microbes. Curr. Opin. Biotechnol. 19, 597–605 (2008).

5 Dien BS, Cotta MA, Jeffries TW. Bacteria engineered for fuel ethanol production: current status. Appl. Microbiol. Biotechnol. 63, 258–266 (2003).

6 Himmel ME, Ding SY, Johnson DK et al. Biomass recalcitrance: engineering plants and enzymes for biofuels production. Science 315, 804–807 (2007).

7 Jarboe LR, Grabar TB, Yomano LP, Shanmugan KT, Ingram LO. Development of ethanologenic bacteria. Adv. Biochem. Eng. Biotechnol. 108, 237–261 (2007).

n Comprehensive review of the efforts in engineering bacteria for ethanol production. Strategies such as expression of different heterologous gene combinations, deletion of competing pathway and increased NADH availability in host cell are emphasized.

8 Hahn-Hägerdal B, Karhumaa K, Jeppsson M, Gorwa-Grauslund MF. Metabolic engineering for pentose utilization in Saccharomyces cerevisiae. Adv. Biochem. Eng. Biotechnol. 108, 147–177 (2007).

9 Van Vleet JH, Jeffries TW. Yeast metabolic engineering for hemicellulosic ethanol production. Curr. Opin. Biotechnol. 20, 300–306 (2009).

10 Ingram LO, Aldrich HC, Borges AC et al. Enteric bacterial catalysts for fuel ethanol production. Biotechnol. Prog. 15, 855–866 (1999).

11 Tao H, Gonzalez R, Martinez A et al. Engineering a homoethanol pathway in Escherichia coli: increased glycolytic flux and levels of expression of glycolytic genes during xylose fermentation. J. Bacteriol. 183, 2979–2988 (2001).

12 Kim Y, Ingram LO, Shanmugam KT. Construction of an Escherichia coli K-12 mutant for homoethanologenic fermentation of glucose or xylose without foreign genes. Appl. Environ. Microbiol. 73, 1766–1771 (2007).

13 Underwood SA, Buszko ML, Shanmugam KT, Ingram LO. Flux through citrate synthase limits the growth of ethanologenic Escherichia coli KO11 during xylose fermentation. Appl. Environ. Microbiol. 68, 1071–1081 (2002).

14 Underwood SA, Zhou S, Causey TB, Yomano LP, Shanmugam KT, Ingram LO. Genetic changes to optimize carbon partitioning between ethanol and biosynthesis in ethanologenic Escherichia coli. Appl. Environ. Microbiol. 68, 6263–6272 (2002).

15 Nichols NN, Dien BS, Bothast RJ. Use of catabolite repression mutants for fermentation of sugar mixtures to ethanol. Appl. Microbiol. Biotechnol. 56, 120–125 (2001).

16 Gonzalez R, Tao H, Shanmugam KT, York SW, Ingram LO. Global gene expression differences associated with changes in glycolytic flux and growth rate in Escherichia coli during the fermentation of glucose and xylose. Biotechnol. Prog. 18, 6–20 (2002).

17 Trinh CT, Unrean P, Srienc F. A minimal Escherichia coli cell for the most efficient production of ethanol from hexoses and pentoses. Appl. Environ. Microbiol. 74, 3634–3643 (2008).

n Based on elementary mode analysis, the minimal metabolic network that links the desired phenotype and the corresponding genotype was identified for efficient production of ethanol.

18 Trinh CT, Srienc F. Metabolic engineering of Escherichia coli for efficient conversion of glycerol to ethanol. Appl. Environ. Microbiol. 75, 6696–6705 (2009).

19 Yomano LP, York SW, Ingram LO. Isolation and characterization of ethanol-tolerant mutants of Escherichia coli KO11 for fuel ethanol production. J. Ind. Microbiol. Biotechnol. 20, 132–138 (1998).

20 Gonzalez R, Tao H, Purvis JE, York SW, Shanmugam KT, Ingram LO. Gene array-based identification of changes that contribute to ethanol tolerance in ethanologenic Escherichia coli: comparison of KO11 (parent) to LY01 (resistant mutant). Biotechnol. Prog. 19, 612–623 (2003).

21 Alper H, Moxley J, Nevoigt E, Fink GR, Stephanopoulos G. Engineering yeast transcription machinery for improved ethanol tolerance and production. Science 314, 1565–1568 (2006).

22 Alper H, Stephanopoulos G. Global transcription machinery engineering: a new approach for improving cellular phenotype. Metab. Eng. 9, 258–267 (2007).

23 Zhou B, Martin GJ, Pamment NB. Increased phenotypic stability and ethanol tolerance of recombinant Escherichia coli KO11 when immobilized in continuous fluidized bed culture. Biotechnol. Bioeng. 100, 627–633 (2008).

24 Miller EN, Jarboe LR, Yomano LP, York SW, Shanmugam KT, Ingram LO. Silencing of NADPH-dependent oxidoreductases (yqhD and dkgA) in furfural-resistant ethanologenic Escherichia coli. Appl. Environ. Microbiol. 75, 4315–4323 (2009).

25 Atsumi S, Liao JC. Metabolic engineering for advanced biofuels production from Escherichia coli. Curr. Opin. Biotechnol. 19, 414–419 (2008).

26 Papoutsakis ET. Engineering solventogenic clostridia. Curr. Opin. Biotechnol. 19, 420–429 (2008).

27 Steen EJ, Chan R, Prasad N et al. Metabolic engineering of Saccharomyces cerevisiae for the production of n-butanol. Microb. Cell Fact. 7, 36 (2008).

28 Atsumi S, Cann AF, Connor MR et al. Metabolic engineering of Escherichia coli for 1-butanol production. Metab. Eng. 10, 305–311 (2008).

29 Inui M, Suda M, Kimura S et al. Expression of Clostridium acetobutylicum butanol synthetic genes in Escherichia coli. Appl. Microbiol. Biotechnol. 77, 1305–1316 (2008).

30 Cornillot E, Croux C, Soucaille P. Physical and genetic map of the Clostridium acetobutylicum ATCC 824 chromosome. J. Bacteriol. 179, 7426–7434 (1997).

31 Nielsen DR, Leonard E, Yoon SH, Tseng HC, Yuan C, Prather KJ. Engineering alternative butanol production platforms in heterologous bacteria. Metab. Eng. 11, 262–273 (2009).

32 Hanai T, Atsumi S, Liao JC. Engineered synthetic pathway for isopropanol production in Escherichia coli. Appl. Environ. Microbiol. 73, 7814–7818 (2007).

33 Jojima T, Inui M, Yukawa H. Production of isopropanol by metabolically engineered Escherichia coli. Appl. Microbiol. Biotechnol. 77, 1219–1924 (2008).

34 Atsumi S, Hanai T, Liao JC. Non-fermentative pathways for synthesis of branched-chain higher alcohols as biofuels. Nature 451, 86–89 (2008).

nn Shows a synthetic pathway for production of an array of higher chain alcohols in Escherichia coli. Through the expanded


Review Xu & Koffas

pathway, the endogenous 2-keto acid intermediates were successfully diverted to alcohol biosynthesis, which represents a new route for large-scale production of biofuels.

35 Zhang K, Sawaya MR, Eisenberg DS, Liao JC. Expanding metabolism for biosynthesis of non-natural alcohols. Proc. Natl Acad. Sci. USA 105, 20653–20658 (2008).

nn Demonstrated the construction of an artificial metabolic network by expanding the branched-chain amino acid pathway. Through a combination of protein rational design and metabolic engineering, they successfully achieved the novel alcohol (3-methyl-1-pentanol) production.

36 Yan Y, Liao JC. Engineering metabolic systems for production of advanced fuels. J. Ind. Microbiol. Biotechnol. 36, 471–479 (2009).

n Provides a comprehensive review of the current state of biofuels and the involved pathways.

37 Shen CR, Liao JC. Metabolic engineering of Escherichia coli for 1-butanol and 1-propanol production via the keto-acid pathways. Metab. Eng. 10, 312–320 (2008).

38 Atsumi S, Liao JC. Directed evolution of Methanococcus jannaschii citramalate synthase for biosynthesis of 1-propanol and 1-butanol by Escherichia coli. Appl. Environ. Microbiol. 74, 7802–7808 (2008).

39 Cann AF, Liao JC. Production of 2-methyl-1-butanol in engineered Escherichia coli. Appl. Microbiol. Biotechnol. 81, 89–98 (2008).

40 Connor MR, Liao JC. Engineering of an Escherichia coli strain for the production of 3-methyl-1-butanol. Appl. Environ. Microbiol. 74, 5769–5775 (2008).

41 Kalscheuer R, Stölting T, Steinbüchel A. Microdiesel: Escherichia coli engineered for fuel production. Microbiol. 152, 2529–2536 (2006).

42 Adamczak M, Bornscheuer U, Bednarski W. The application of biotechnological methods for the synthesis of biodiesel. Euro. J. Lipid Sci. and Technol. 111, 800–813 (2009).

43 Li Q, Du W, Liu D. Perspectives of microbial oils for biodiesel production. Appl. Microbiol. Biotechnol. 80, 749–756 (2008).

44 Kalscheuer R, Stöveken T, Luftmann H, Malkus U, Reichelt R, Steinbüchel A. Neutral lipid biosynthesis in engineered Escherichia coli: jojoba oil-like wax esters and fatty acid butyl esters. Appl. Environ. Microbiol. 72, 1373–1379 (2006).

45 Davis MS, Solbiati J, Cronan JE Jr. Overproduction of acetyl-CoA carboxylase activity increases the rate of fatty acid biosynthesis in Escherichia coli. J. Biol. Chem. 275, 28593–28598 (2000).

46 Leonard E, Lim KH, Saw PN, Koffas MA. Engineering central metabolic pathways for high-level flavonoid production in Escherichia coli. Appl. Environ. Microbiol. 73, 3877–3886 (2007).

47 Lu X, Vora H, Khosla C. Overproduction of free fatty acids in E. coli: implications for biodiesel production. Metab. Eng. 10, 333–339 (2008).

48 Gao B, Su E, Lin J, Jiang Z, Ma Y, Wei D. Development of recombinant Escherichia coli whole-cell biocatalyst expressing a novel alkaline lipase-coding gene from Proteus sp. for biodiesel production. J. Biotechnol. 139, 169–175 (2009).

49 Anthony JR, Anthony LC, Nowroozi F, Kwon G, Newman JD, Keasling JD. Optimization of the mevalonate-based isoprenoid biosynthetic pathway in Escherichia coli for production of the anti-malarial drug precursor amorpha-4,11-diene. Metab. Eng. 11, 13–19 (2009).

50 Yoon SH, Lee SH, Das A et al. Combinatorial expression of bacterial whole mevalonate pathway for the production of b-carotene in E. coli. J. Biotechnol. 140, 218–226 (2009).

51 Yoon SH, Lee YM, Kim JE et al. Enhanced lycopene production in Escherichia coli engineered to synthesize isopentenyl diphosphate and dimethylallyl diphosphate from mevalonate. Biotechnol. Bioeng. 94, 1025–1032 (2006).

52 Martin VJ, Pitera DJ, Withers ST, Newman JD, Keasling JD. Engineering a mevalonate pathway in Escherichia coli for production of terpenoids. Nat. Biotechnol. 21, 796–802 (2003).

53 Withers ST, Gottlieb SS, Lieu B, Newman JD, Keasling JD. Identification of isopentenol biosynthetic genes from Bacillus subtilis by a screening method based on isoprenoid precursor toxicity. Appl. Environ. Microbiol. 73, 6277–6283 (2007).

54 Pitera DJ, Paddon CJ, Newman JD, Keasling JD. Balancing a heterologous mevalonate pathway for improved isoprenoid production in Escherichia coli. Metab. Eng. 9, 193–207 (2007).

55 Mukhopadhyay A, Redding AM, Rutherford BJ, Keasling JD. Importance of systems biology in engineering microbes for biofuel production. Curr. Opin. Biotechnol. 19, 228–234 (2008).

56 Koffas MA. Expanding the repertoire of biofuel alternatives through metabolic pathway evolution. Proc. Natl Acad. Sci. USA 106, 965–966 (2009).

57 Fowler ZL, Gikandi WW, Koffas MA. Increasing malonyl-CoA biosynthesis by tuning the Escherichia coli metabolic network and its application to flavanone production. Appl. Environ. Microbiol. DOI: 10.1128/AEM.00270-00209 (2010) (Epub ahead of print).

58 Fortman JL, Chhabra S, Mukhopadhyay A et al. Biofuel alternatives to ethanol: pumping the microbial well. Trends Biotechnol. 26, 375–381 (2008).

59 Lee SK, Chou H, Ham TS, Lee TS, Keasling JD. Metabolic engineering of microorganisms for biofuels production: from bugs to synthetic biology to fuels. Curr. Opin. Biotechnol. 19, 556–563 (2008).

n Discusses different types of biofuels and how synthetic biology will affect metabolic engineering.

60 Alper H, Fischer C, Nevoigt E, Stephanopoulos G. Tuning genetic control through promoter engineering. Proc. Natl Acad. Sci. USA 102, 12678–12683 (2005).

61 Pfleger BF, Pitera DJ, Smolke CD, Keasling JD. Combinatorial engineering of intergenic regions in operons tunes expression of multiple genes. Nat. Biotechnol. 24, 1027–1032 (2006).

62 Hallam SJ, Mincer TJ, Schleper C et al. Pathways of carbon assimilation and ammonia oxidation suggested by environmental genomic analyses of marine Crenarchaeota. PLoS Biol. 4, e95 (2006).

63 Berg IA, Kockelkorn D, Buckel W, Fuchs G. A 3-hydroxypropionate/4-hydroxybutyrate autotrophic carbon dioxide assimilation pathway in Archaea. Science 318, 1782–1786 (2007).

64 Kockelkorn D, Fuchs G. Malonic semialdehyde reductase, succinic semialdehyde reductase, and succinyl-Coenzyme A reductase from Metallosphaera sedula: enzymes of the autotrophic 3-hydroxypropionate/4-hydroxybutyrate cycle in sulfolobales. J. Bacteriol. 191, 6352–6362 (2009).

� Websites101 Butanol reality check – Cobalt Biofuels. New

Energy and Fuel http://newenergyandfuel.com/http:/newenergyandfuel/com/2008/10/31/butanol-reality-check-cobalt-biofuels

102 Technology review: potential energy www.technologyreview.com/blog/energy/23892

Metabolic engineering of Escherichia …homepages.rpi.edu/~koffam/papers/2010_Xu_Koffas.pdf · Metabolic engineering of Escherichia coli for biofuel production Review

Documents