J Ind Microbiol Biotechnol (2009) 36:471–479 DOI 10.1007/s10295-009-0532-0 REVIEW Engineering metabolic systems for production of advanced fuels Yajun Yan · James C. Liao Received: 7 December 2008 / Accepted: 14 January 2009 / Published online: 7 February 2009 © The Author(s) 2009. This article is published with open access at Springerlink.com Abstract The depleting petroleum storage and increasing limitations, such as low energy density, high vapor environmental deterioration are threatening the sustainable pressure, and corrosiveness, which prevent its widespread development of human societies. As such, biofuels and utilization given the existing infrastructure. chemical feedstocks generated from renewable sources are Higher alcohols (with more than two carbons), biodie- becoming increasingly important. Although previous eVorts sels, and fatty acid derivatives are thought to be more suit- led to great success in bio-ethanol production, higher alco- able fuels. Their physicochemical properties are more hols, fatty acid derivatives including biodiesels, alkanes, compatible with gasoline-based fuels and allow direct utili- and alkenes oVer additional advantages because of their zation of existing infrastructure for storage and distribution. compatibility with existing infrastructure. In addition, some Furthermore, some of these fuel molecules also serve as of these compounds are useful chemical feedstocks. Since important chemical feedstocks. Although the individual native organisms do not naturally produce these com- biochemical steps for synthesizing these compounds in pounds in high quantities, metabolic engineering becomes microbes have been described previously, eVorts in putting essential in constructing producing organisms. In this arti- together highly productive metabolic systems have only cle, we brieXy review the four major metabolic systems, the begun recently. In this article, we Wrst summarize the meta- coenzyme-A mediated pathways, the keto acid pathways, bolic networks for producing these compounds and then the fatty acid pathway, and the isoprenoid pathways, that review eVorts in engineering the non-native producing allow production of these fuel-grade chemicals. organism, Escherichia coli. The metabolic networks dis- cussed include the traditional butanol pathway in Clostrid- ium species, the keto acid pathways for higher alcohols, the Introduction isoprenoid pathways, and the fatty acid biosynthesis. The depleting petroleum reserve, recurring energy crisis, and global climate change are reigniting the enthusiasm for The coenzyme-A-dependent fermentative pathways seeking sustainable technologies for replacing petroleum as a source of fuel and chemicals. In the past few decades, Among the higher alcohols, n-butanol and isopropanol are eVorts in the development of bio-ethanol as an alternative the only two that are overproduced in nature by Clostrid- fuel have led to signiWcant success [14–16, 19]. In 2007, ium species. n-Butanol has been produced by Clostridium 6.5 billion gallons of bio-ethanol was produced in the in acetone–butanol–ethanol (ABE) fermentation. The fer- United State [5]. However, bio-ethanol exhibits some mentative pathway (Fig. 1) in this organism starts from acetyl-CoA. The enzyme acetyl-CoA acetyltransferase, also known as thiolase, condenses two molecules of Y. Yan · J. C. Liao (&) acetyl-CoA to one molecule of acetoacetyl-CoA. From Department of Chemical and Biomolecular Engineering, University of California at Los Angeles, 5531 Boelter Hall, this molecule, the pathway branches into isopropanol and 420 Westwood Plaza, Los Angeles, CA 90095, USA n-butanol. For the isopropanol biosynthesis, an acetoace- e-mail: [email protected] tyl-CoA transferase (ACoAT) transfers the CoA group 123 472 J Ind Microbiol Biotechnol (2009) 36:471–479 Glucose step of dehydration. Acetoacetate is Wrst reduced to 2NAD+ 3-hydroxybutyryl-CoA by 3-hydroxybutyryl-CoA dehydro- genase (HBD). Then, 3-hydroxybutyryl-CoA is dehydrated 2NADH to crotonyl-CoA by a crotonase (CRT). Third, a butyryl- 2 Pyruvate CoA dehydrogenase (BCD) catalyzes the reduction of 2NAD+ aceEF crotonyl-CoA to butyryl-CoA. Finally, an aldehyde/alcohol lpd 2NADH dehydrogenase (AADH) converts butyryl-CoA to n-butanol 2CO 2 through two consecutive reduction reactions. 2 Acetyl-CoA Isopropanol production in Escherichia coli atoB/thl CoA The secondary alcohol, isopropanol, is both a desirable fuel Acetoacetyl-CoA and an important chemical feedstock in the petrochemical industry. Its dehydrated product, propylene, serves as the Acetate NADH ctfAB/atoAD hbd monomer for making polypropylene. In addition, it can be Acetyl-CoA NAD+ used as an additive to petroleum-based fuels. Replacing meth- W Acetoacetate 3-Hydroxybutyryl-CoA anol with isopropanol in the esteri cation process of fat and oil could generate crystallization-resistant biodiesels [12]. adc crt As described above, isopropanol is produced by Clos- CO H2O 2 tridium species in nature. However, as a native metabolite, it can only be produced in a limited amount for the hosts’ Acetone Crotonyl-CoA own beneWts as a detoxiWcation response to low pH condi- NADPH NADH tions. The maximum titer reported in its native producer, sadh bcd etf NADP+ NAD+ Clostridium, was 1.8 g/l [9]. To improve the production of isopropanol, the fully characterized isopropanol biosyn- Isopropanol Butyryl-CoA thetic pathway (Fig. 1) was reconstructed in the genetic NADH tractable host E. coli [12]. adhE2 NAD+ Escherichia coli has been reported to produce acetone [6], the immediate precursor of isopropanol, by expressing Butyraldehyde the intact pathway from Clostridium acetobutylicum NADH ATCC824 consisting of the acetyl-CoA acyltransferase, adhE2 ACoAT, ADC encoded by the thl, ctfAB, and adc genes, NAD+ respectively. The reported titer was around 5.4 g/l, similar n-Butanol to the yield of native host for acetone. Furthermore, with a Fig. 1 Metabolic pathways for isopropanol and 1-butanol production SADH co-expressed with the acetone pathway in E. coli, in engineered E. coli. The dashed line indicates omitted steps. the isopropanol production was achieved [12]. The pathway Isopropanol pathway consists of four enzymatic steps from acetyl- eYciency was tuned by using genes from diVerent organ- CoA. 1-Butanol pathway consists of six enzymatic steps. aceEF and lpd encode pyruvate dehydrogenase; atoB/thl encodes acetyl-CoA isms, a bio-prospecting approach. Since the genes from acetyltransferase; ctfAB/atoAD encodes acetoacetyl-CoA transferase; Clostridium usually have a low GC content, which may adc, acetoacetate decarboxylase; sadh encodes secondary alcohol lead to poor expression, the E. coli native genes atoB and dehydrogenase; hbd encodes 3-hydroxybutyryl-CoA dehydrogenase; atoAD, encoding acetyl-CoA acyltransferase and ACoAT, crt encodes crotonase; bcd encodes butyryl-CoA dehydrogenase; etf encodes electron transfer Xavoprotein; adhE2 encodes aldehyde/ were also tested as pathway components. Additionally, two alcohol dehydrogenase genes from C. beijerinckii NRRL B593 and Thermoanae- robacter brockii HTD4, encoding SADHs, were totally away from acetoacetyl-CoA to acetate or butyrate, form- synthesized with codon optimization and installed into the ing acetoacetate. The acetyl-CoA is recycled back to ace- pathway to test for production. With these eVorts, the strain tate by the combined phosphotransacetylase and acetate with a combination of C. acetobutylicum thl, E. coli atoAD, kinase reaction. Further, acetoacetate is decarboxylated to C. acetobutylicum adc, and C. beijerinckii adh achieved the acetone by an acetoacetate decarboxylase (ADC). Then highest titer (»5.0 g/l). The result is promising, since it acetone is reduced to isopropanol by a NADPH-depen- demonstrates 43.5% (mol/mol) conversion ratio. The theo- dent secondary alcohol dehydrogenase (SADH) [12]. retical yield is 1 mol isopropanol per mole glucose. For n-butanol biosynthesis, acetoacetate has to go The production of isopropanol from glucose is not through four steps of NADH-dependent reduction and one redox-balanced. Four moles of NADH is produced, while 123 J Ind Microbiol Biotechnol (2009) 36:471–479 473 1 mol of NADPH is consumed per mole of isopropanol. acetate, ethanol, and succinate. The highest titer of 552 mg/l Therefore, an external electron acceptor is required or a was reported with optimized pathway and improved strain. byproduct is served as an electron acceptor. Although the yield was still low, this work demonstrated the feasibility of heterologous n-butanol production and n-Butanol production in E. coli proposed the principles for further optimization. n-Butanol was proposed to be one of the better substitutes for gasoline-based transportation fuel, because of its high The keto acid pathways energy density and hydrophobicity. Its energy content (27 MJ/l) is similar to that of gasoline (32 MJ/l). The high Importing a non-native pathway in a heterologous host such hydrophobicity enables its transportation and storage using as E. coli unavoidably introduces non-native metabolites existing petrochemical infrastructure with minimal modiW- and potential toxicity, in addition to diYculties in express- cation. In addition, n-butanol has a low vapor pressure of ing heterologous enzymes. The resulting metabolic imbal- 4 mmHg at 20°C, which allows its mixing with gasoline at ance and cytotoxicity pose a barrier for large quantity any ratio without exceeding air quality speciWcations. production. In this context, it is desirable to seek for the The microbial