Can microbially derived advanced biofuels ever compete with conventional bioethanol? A critical review

Abstract

Microbially derived alkanes and their derivatives are recognized as promising alternatives to petroleum-based fuels and chemicals. We review recent developments in their production, assess progress, and their potential against conventional bioethanol fermentation pathways. The success rate of genetic engineering efforts and their commercialization prospects are assessed, as well as challenges for producing fuels and chemicals from lignocellulosic biomass. Although significant progress has been made in the genetic engineering of microbes used in the production of long-chain hydrocarbons and their derivatives, titer and yield of these biomolecules are currently too low to compete with petroleum-derived products. As for microbially derived isoprenoids or fatty acids, the inherent complexity of micro-organism development will continue to present formidable challenges, making it highly unlikely of any short-term commercial take off. Nonetheless, first generation bioethanol (starch/sugar based) production is commercially established and therefore continued advancements in chemical synthesis should enable broad-scale use of bio-ethanol as a chemical feedstock for the production of advanced biofuels including butanol and other long-chain hydrocarbons.

Full Article

Can Microbially Derived Advanced Biofuels Ever Compete with Conventional Bioethanol? A Critical Review

Seena Illath Veettil,^a Linoj Kumar,^b and Alexander A. Koukoulas ^c

Microbially derived alkanes and their derivatives are recognized as promising alternatives to petroleum-based fuels and chemicals. We review recent developments in their production, assess progress, and their potential against conventional bioethanol fermentation pathways. The success rate of genetic engineering efforts and their commercialization prospects are assessed, as well as challenges for producing fuels and chemicals from lignocellulosic biomass. Although significant progress has been made in the genetic engineering of microbes used in the production of long-chain hydrocarbons and their derivatives, titer and yield of these biomolecules are currently too low to compete with petroleum-derived products. As for microbially derived isoprenoids or fatty acids, the inherent complexity of micro-organism development will continue to present formidable challenges, making it highly unlikely of any short-term commercial take off. Nonetheless, first generation bioethanol (starch/sugar based) production is commercially established and therefore continued advancements in chemical synthesis should enable broad-scale use of bio-ethanol as a chemical feedstock for the production of advanced biofuels including butanol and other long-chain hydrocarbons.

Keywords: Biochemicals; Bioethanol; Biobutanol; Biofuels; Fermentation; Lignocellulosics

Contact information: a: Department of Wood Science, Faculty of Forestry, University of British Columbia, Vancouver, BC, Canada; b: Herty Advanced Materials Development Center, Savannah, GA, United States; c: ANL Consultants LLC, P.O. Box 2981, Savannah, GA, 31402 United States; * Corresponding author: aakoukoulas@gmail.com

INTRODUCTION

Climate change and energy security have placed enormous pressure on governments and industry to replace carbon-intensive petroleum with biomass-derived alternatives (Guo et al. 2013; Su et al. 2015). As a result of this global surge, bioethanol production and its use has increased exponentially in the last decade, reaching 25 billion gallons in 2013, which is more than three times higher than 2000 production levels (RFA 2015; Su et al. 2015). Although bioethanol currently dominates the biofuels market, there is increasing recognition that multiple biofuel options are needed to achieve broad-based substitution of petroleum.

Ethanol has several limitations as a “drop-in” replacement fuel (Lee et al. 2008; IEA 2011; Snow 2013). Ethanol can neither be a universal substitute to all types of fuels nor be a complete replacement of gasoline. Despite its high octane number, ethanol has a low energy density, which is 30% lower than gasoline (Atsumi et al. 2008). Ethanol is highly hygroscopic and corrosive, which not only limits its usage in existing engines, but makes it incompatible with the current petroleum storage and distribution infrastructure (Lee et al. 2008; Li et al. 2010b; IEA 2011). Therefore, in order to enable a sustainable transition from a hydrocarbon economy to a carbohydrate-based society, alternative biofuels must be compatible with the existing engine and fungible with respect to the supply chain infrastructure (Fortman et al. 2008).

Both natural and genetically engineered fermentation pathways in microorganisms can produce compounds that not only can substitute for gasoline, but can replace all petroleum-derived fuels (Steen et al. 2008; Yan and Liao 2009). These microbially-derived compounds can include long-chain alcohols, such as isopropanol, butanol, isobutanol, longer chain alcohols, isoprenoids, long-chain fatty acids, and their corresponding alcohols, esters, as well as alkanes (Peralta-Yahya and Keasling 2010; Steen et al. 2010; Carmann 2011). These biofuels are referred to as “advanced biofuels” to distinguish them from first-generation fuels, as these fuels are yet to be produced at commercial scale. Butanol and isopropanol are both naturally produced by the solventogenic Clostridia or can also be derived by the genetic engineering of natural Clostridial pathway or rerouting the amino acid biosynthesis pathways in both bacteria and yeasts (Papoutsakis 2008). Production of isoprenoids and long-chain fatty acids are achieved through the genetic engineering of the respective biosynthesis pathways. All such molecules have both higher energy content and hydrophobicity than ethanol, and most can be transported through existing pipelines (Dürre 2007). Many of these compounds, such as farnesane (isoprenoid based), have energy densities and other fuel properties equivalent to that of petroleum distillates and can be blended with the fossil-based counterparts in any proportions and used as “drop-in” fuels (Peralta-Yahya and Keasling 2010). The present article reviews different types of advanced biofuels that can be produced by microbial fermentation and assesses research progress against the developments in bioethanol production. The major liquid biofuels considered here are biochemically derived long-chain alcohols (> C3), long-chain fatty acids and their esters, and isoprenoid-based fuels. In addition to reviewing the progress made in the natural production pathways, the success rate is considered for the genetic engineering efforts on the Clostridial pathway and three other metabolic pathways: amino acid biosynthesis, fatty acid biosynthesis, and isoprenoid biosynthesis.

The feasibility of producing advanced biofuels from lignocellulosic substrates is a key focus. Much has been written on cellulosic biomass and its enormous potential for being a second generation feedstock (Perlack et al. 2005). Beyond its wide-scale availability, it is the only alternate carbon source to fossil fuels that does not compromise food security (Singhvi et al.2014). Nonetheless, there are numerous challenges that must be addressed before advanced biofuels can be produced from lignocellulosic biomass. In addition to the presence of multiple sugars, lignocellulosic hydrolysate contains a spectrum of compounds, which are potentially toxic to the enzymes and/or microbes used in the bioconversion process (Azhar et al. 1981; Banerjee et al. 1981; Clark and Mackie 1984). These toxic compounds or inhibitors are both naturally present in the lignocellulosic substrates and also ‘process derived,’ typically from conversion processes such as steam pretreatment (Hahn-Hägerdal et al. 2007; Matsushika et al. 2009). However, the extent to which these different inhibitory compounds influence the rate and yield of the products is dependent on the nature of enzymes/microorganisms involved, the mode of fermentation, and the type of final products (Hahn-Hägerdal et al. 2007; Matsushika et al. 2009). The robustness of the genetically engineered strains to consume multiple sugars while being tolerant to a range of inhibitory compounds present in the lignocellulosic hydrolysates is assessed in this article. The additional and unique challenges related to producing each type of advanced fuels from cellulosic biomass are also analyzed. Finally, the status of commercialization of some of the advanced biofuels compared to bioethanol, potential research avenues, and the future perspectives for their commercial success are discussed.

ALTERNATIVE LIQUID BIOFUELS

Three different types of conventional liquid fuels used today include gasoline, middle distillates, and aviation fuels. All have different chemical compositions, primarily with respect to the type of hydrocarbons; consequently, they have different fuel properties (cetane and octane numbers, flash point, freezing point, etc.) and thus distinct applications (Table 1) (Ghosh et al. 2006; Fortman et al. 2008). Each category of petroleum fuels can be replaced by their biobased counterparts including short chain and long-chain alkanes, alcohols, esters, and fatty acids (Table 1). However, industrially relevant microorganisms and efficient bioprocesses are necessary to enable the commercial production of these compounds from biomass.

Table 1. Major Categories of Petroleum-based Conventional Fuels and their Potential Biomass-based Equivalents

Isopropanol, Butanol, and Long-Chain Alcohols

Isopropanol and butanol

Various species of Clostridium naturally produce isopropanol and butanol from acetyl-CoA through the ABE/IBE (acetone/isopropanol, butanol, and ethanol) fermentation route (Fig. 1) (George and Chen 1983; Lin and Blaschek 1983; Chen and Hiu 1986; Ismaiel et al. 1993). Because this fermentation generally produces a mixture of solvents (Lee et al. 2008) (acetone/isopropanol, butanol and ethanol) and the solvent titer is very poor (< 2%) (Lee et al.2008), several efforts have been made to genetically engineer Clostridia to selectively enhance the yield of isopropanol and butanol (Peralta-Yahya et al. 2010; Branduardi et al. 2014). However, with limited genetic tools available to engineer Clostridia, researchers soon realized that even with genetic improvements, Clostridium, a strict anaerobe, has several limitations affecting its potential use as an industrial host (Branduardi et al. 2014). Therefore, subsequent efforts were mainly limited to heterologously express the Clostridial pathway in industrially relevant microorganisms such as E. coli and S. cerevisiae (Branduardi et al. 2014).

Production of ABE/IBE (Acetone/Isopropanol-Butanol-Ethanol) Solvents by Natural Microbial Strains

Several bacterial genera including Clostridium, Butylvibrio, Butyribacterium, Sarcina,Eubacterium, Eusobacterium, and Megasphera generate butanol (Ni and Sun 2009). Despite several of the inherent limitations, Clostridium strains are the most industrially relevant among natural butanol-producing strains (Ni and Sun 2009). Common solventogenic Clostridium species are Clostridium acetobutylicum, Clostridium saccharobutylicum, Clostridium beijerinckii, and Clostridium saccharoperbutyl-acetonicum (George and Chen 1983; Lin and Blaschek 1983; Chen and Hiu 1986; Ismaiel et al. 1993). While most of Clostridia species such as C. acetobutylicumgenerate ABE solvents, some others, such as C. beijerinckii, produce IBE (George and Chen 1983; Chen and Hiu 1986). Some promising butanol-producing strains reported are Clostridium acetobutylicum ATCC 824, C. saccharobutylicum P262, C. beijerinckii P260, and C. beijerinckii BA 101 (Qureshi et al. 2006; Dürre 2007; Ni and Sun 2009; Green 2011).

Fig. 1. Butanol and isopropanol biosynthesis pathway. Blue letters indicate Clostridia genes. Red letters refer to equivalent genes from microorganisms used in the genetic engineering studies.

In ABE/IBE fermentation, after glycolysis, pyruvate is converted to acetyl-CoA by the enzyme pyruvate ferredoxin oxidoreductases (Ezeji et al. 2007; Fischer et al. 2008; Zheng et al. 2009; Jang et al. 2012). Subsequently, two acetyl-CoA molecules undergo Claisen condensation to form acetoacetyl-CoA, which is reduced to 3-hydroxybutyryl-CoA, then dehydrated to form crotonyl-CoA and further reduced to butyryl-CoA, butyraldehyde, and finally butanol (Fig. 1) (Fischer et al. 2008; Nielsen et al. 2009; Zheng et al. 2009; Jang et al. 2012). Some of the acetyl-CoA is converted to acetaldehyde, which in turn is reduced by alcohol dehydrogenase to ethanol (Jones and Wood 1986). A smaller fraction of the acetoacetyl-CoA is converted to acetone, which in some species is further reduced to isopropanol by isopropanol dehydrogenase (Dürre 2007). In the IBE fermentation, depending on the strain and the cultivation conditions, residual acetone may also be an end-product (Ismaiel et al. 1993). During ABE fermentation, the acetone, butanol, and ethanol are typically generated in the ratio of 3:6:1 (Nimcevic and Gapes 2000).

The ABE/IBE fermentation involves a biphasic kinetic pattern comprising a first growth phase with the production of acetic acid and butyric acid (acidogenesis) followed by their re-assimilation into ABE/IBE solvents (solventogenesis) (Jones and Woods 1986; Girbal and Soucaille 1998) (Fig. 1). Production of organic acids in the acidogenic phase of the ABE/IBE fermentation lowers the pH outside the cell, and considerable amounts of undissociated acids can diffuse into the cytoplasm reducing the pH resulting in cell death (Huang et al. 1985; Jones and Wood 1986). Therefore, the cell is forced to re-assimilate the generated acids to alcohol to help maintain the external pH (Huang et al. 1985). This is the reason that subsequent production of acetic and butyric acids in the exponential growth, fermentation switches to the formation of neutral solvents shortly before entering the stationary phase (Dürre 2007). During the solventogenesis, microbial growth slows down and cells form granules and endospores (Paredes et al. 2005; Alsaker et al. 2004). In addition to ABE/IBE solvents, the process also produces carbon dioxide and hydrogen (Jones and Wood 1986).

One of the greatest challenges with using natural Clostridia strains for butanol production has been the low butanol titer (below 12 g/L) (Lee et al. 2008), which is more than ten times lower than the ethanol titer generally obtained in starch- or sugarcane-based ethanol fermentations (Amorim et al. 2011). Therefore, several genetic engineering studies were carried out to enhance the butanol titer, selectivity, and yield (Peralta-Yahya et al. 2010; Branduardi et al. 2014).

Genetic Engineering of Clostridia for Enhanced Butanol and Isopropanol Production

Several workers have investigated specific genetic manipulation, chemical mutagenesis, or their combination (Qureshi and Blaschek 1999; Gao et al. 2012; Branduardi et al. 2014). Successful examples of the specific genetic manipulations include the overexpression of acetone formation pathway by amplifying the adc (acetoacetate decarboxylase) and ctfAB (CoA transferase) genes, overexpression of adhE (aldehyde dehydrogenase), regulation of soIR (regulator of solvent production), reducing the activities of butyrate forming enzymes, and overexpression of heat shock proteins (Mermelstein et al. 1993; Nair et al. 1994; Harris et al. 2001; Lee et al. 2008; Papoutsakis, 2008). Controlling the regulator of solvent producing genes resulted in the best butanol concentration (17.6g/L) in C. acetobutylicum ATCC 824 (Nair et al. 1999; Harris et al.2001; Thormann et al. 2002), which remains the most extensively studied and manipulated strain. Its genome sequence was published in 2001 (Nolling et al. 2001) followed by the genome sequence of C. beijerinckii NCIMB 8052 in 2007 (Ezeji et al. 2007).

The best genetically modified Clostridia—C. beijerinckii BA 101—produces a butanol concentration of 19.6 g/L (Qureshi and Blaschek 1999). This strain was developed by chemical mutagenesis with N-methyl N-9-nitro-N-nitrosoguanidine (Qureshi and Blaschek 1999; Gao et al.2012). C. beijerinckii BA 101 was able to produce butanol and acetone yields two-fold greater than its parent strain (Qureshi and Blaschek 1999). The highest butanol ratio in the solvent mixture (0.7) was exhibited by C. acetobutylicum EA 2018, which was also developed using chemical mutagenesis (Hu et al. 2011). The acetone pathway was further knocked out in this strain increasing the butanol-to-total solvent ratio to 0.8 (Hu et al. 2011).

As isopropanol is commercially a more valuable chemical than acetone (Collas et al. 2012), genetic engineering efforts were also focused on the enhanced production of isopropanol (Lee et al. 2012). One of the earliest reported isopropanol concentrations from Clostridium was 1.8 g/L (Chen et al. 1986). Later, the adh gene from C. beijerinckii was cloned into the ABE-producing strain C. acetobutylicum ATCC 824. The resulting transformant excreted 6.1 g/L isopropanol and a minor amount of acetone (Lee et al. 2012). In C. beijerinckii NRRL B593, the reduction of acetone into isopropanol is catalyzed by a NADPH-dependent secondary-alcohol dehydrogenase (s-adh), which has been extensively characterized (Yan et al. 1988; Ismaiel et al. 1993; Korkhin et al. 1998; Goihberg et al. 2010). Although the s-adh was clearly distinct from Clostridialprimary-alcohol dehydrogenases (Chen 1995) that reduce butyraldehyde into butanol, the s-adhshowed activity on both primary and secondary alcohols, with a preference for secondary alcohols (Ismaiel et al. 1993). Kinetic studies confirmed that the physiological substrate was acetone.

Although Clostridium can naturally produce isopropanol and butanol, it is a strict anaerobe with a spore-forming life cycle characterized by a slow growth rate (Jones and Woods 1986; Zheng et al.2009; Green 2011; Jang et al. 2012). In addition, limited genetic tools were available to engineer solvent production in this organism (Branduardi et al. 2014). These limitations have prompted researchers to attempt the heterologous expression of isopropanol and butanol pathways in E. coliand S. cerevisiae (Hanai et al. 2007; Atsumi et al. 2008a; Jojima et al. 2008; Inui et al. 2008; Steen et al. 2008; Nielsen et al. 2009; Peralta-Yahya et al. 2010; Branduardi et al. 2014).

Heterologous Expression of Clostridial Pathways in E. coli / S. cerevisiae

Metabolic engineering has been employed to create pathways for isopropanol production in E. coli. Introduction of four genes from C. acetobutylicum (ctfA, ctfB, adc, and thiolase (thl) into E. coli generated a strain capable of producing acetone, the precursor for isopropanol (Fig. 1) (Bermejo et al. 1998). Subsequent efforts on overexpressing the C. beijerinckii adh gene in combination with the above genes made it possible to overproduce isopropanol in E. coli. The engineered E. coli strains surpassed the best reported wild-type Clostridial strains, C. beijerinckiiand C. isopropylicum, excreting approximately 4 g/L isopropanol (Groot and Luyben 1986; Matsumura et al. 1992). Later, use of E. coli atoB and atoAD instead of thl and ctfAB resulted in a concentration of 4.9 g/L (Hanai et al. 2007). Jojima and coworkers expressed C. acetobutylicumgenes ctfAB, ctfB, and thl and C. beijerinckii adh genes from a dedicator promoter in a single vector instead of a polycistronic expression of the single operon and a two vector system (Jojima et al. 2008). This approach enabled them to obtain 13.6 g/L of isopropanol concentration, which was greater than what was reported with homologous expression (Jojima et al. 2008). A major advantage of the engineered E. coli strain compared to S. cerevisiae or Clostridium was the lack of important competing pathways for by-products such as ethanol or butanol (Hanai et al. 2007).

Heterologous expression of Clostridial butanol pathway in an industrial host has resulted in far lower butanol titer and yield than those obtained with C. beijerinckii BA 101. Atsumi et al.(2008a) employed a two operon system in E. coli. The first operon encoded thl and adhE2, and the second encoded the rest of the enzymes (hbd, crt, bcd/etfAB) (Atsumi et al. 2008a) (Fig. 1). Despite successful heterologous expression, the butanol concentration was only 13 mg/L, which was substantially lower than that obtained with natural Clostridia (Atsumi et al. 2008a). Subsequently, researchers have tried expressing several permutations and combinations of equivalent genes from different bacteria and yeasts including E. coli, C. acetobutylicum, C. beijerinckii, Megasphaera elsdenii, Streptomyces coelicor, and S. cerevisiae (Atsumi et al. 2008a; Inui et al. 2008). Efforts included deleting some of the host genes, which compete with the butanol production pathways primarily for acetyl-CoA and NADH (Hanai et al. 2007; Atsumi et al. 2008a; Inui et al. 2008). Among these, the most noted progress was made by Hanai and coworkers (2007) in their work on fnr deletion to activate pyruvate dehydrogenase complex. They obtained a butanol concentration of 373 mg/L. Inui and coworkers (2008) expressed the butanol pathway in a single vector with the first promoter expressing crt–bcd-/etfAB/hbd, the second promoter expressing thl, and the third promoter expressing adhE1/adhE2. In this system, the use of adhE2 instead of adhE1 enhanced butanol production from 320 to 540 mg/L respectively, which was further enhanced to 1.2 g/L by extending the fermentation time to 60 hours. Nielsen et al. (2009) compared a polycistronic two-vector system with the individual expression of the genes from dedicated promoters in a four vector system (Nielsen et al. 2009). Individual gene expressions enabled a six-fold improvement in butanol concentration. This approach combined with the replacement of thl with E. coli atoB, incorporation of S. cerevisiae fdh1 (formate dehydrogenase to enhance the NADH concentration) and overexpression of gapA to enhance the carbon flux through glycolytic pathway enhanced the butanol concentration up to 580 mg/L (Nielsen et al. 2009). Although the titers were higher than previous work, the values were considerably lower than the titer obtained with the natural Clostridium (Inui et al. 2008).

Despite many attempts to introduce the Clostridial pathway in S. cerevisiae (Fig. 1), the yields and productivities were poorer than engineered E. coli. Steen and coworkers (2008) have investigated expressing different combinations of acyl transferases (Ralstonia eutropha phaA, E. coli atoB, and S. cerevisiae ERG10), 3-hydro-xybutyryl-CoA dehydrogenase (R. eutropha phaB, C. beijerinckii hbd), crotonase (C. beijerinckii), butyryl-CoA dehydrogenase (Streptomyces collinus ccr), and alcohol dehydrogenase (Steen et al. 2008). The best optimized combination was Erg10–hbd–crt–ccr–adhE2, which resulted in 2.5 mg/L of butanol; 200 times lower than the best values obtained with engineered E. coli (Steen et al. 2008).

Bio-Alcohols via Rerouting of the Amino Acid Biosynthesis Pathway

In addition to the heterologous expression of Clostridial pathways in E. coli or S. cerevisiae, re-routing the amino acid biosynthesis pathway is another approach to produce a range of C4 or higher bioalcohols including n-butanol (Atsumi and Liao 2008a; Lamsen and Atsumi 2012; Bujis et al. 2013). In this approach, ketoacid intermediates generated during amino acid biosynthesis are rerouted via sequential decarboxylation to the aldehyde (by keto-acid decarboxylase) followed by the reduction of the aldehyde to alcohol (by alcohol dehydrogenase) (Lamsen and Atsumi 2012; Bujis et al. 2013) (Table 2). Interestingly, the same pathway—the Ehrlich pathway—naturally exists in yeasts for the production of “fusel oil”, a by-product of fermentation (Fig. 2). There has been extensive research conducted to reconstitute the Erlich pathway in E. coli or S. cerevisiae for the production of different types of alcohols.

Table 2. Alcohols Produced by Re-routing the Amino Acid Biosynthesis Pathway in E. coli / S. cerevisiae and Ketoacid Intermediate Involved in their Production

Fig. 2. Generic Erlich pathway for production of fusel oil (Hazelwood et al. 2008)

Lack of selectivity is one of the challenges when re-routing the amino acid biosynthesis pathway for the production of a given alcohol. Atsumi and coworkers (2008b) successfully introduced alpha-isoketovalerate decarboxylase from Lactococcus lactis (kivd) and alcohol dehydrogenase from S. cerevisiae (adh2) in E. coli with the objective of producing isobutanol. However, due to the broad substrate specificity of the decarboxylase enzymes, a range of keto-acids were converted to corresponding alcohols (Table 2), resulting in a mixture of six different alcohols (1-propanol, 2-methyl 1-butanol, isobutanol, 3-methyl 1-butanol, 2-phenyl ethanol and 1-butanol) with isobutanol produced in the highest concentration (389 mg/L) (Atsumi and Liao 2008b). Subsequent attempts have been made to enhance the selectiveness of isobutanol generated in the alcohol mix (Atsumi et al. 2008b).

The three main strategies that have been adapted to enhance the isobutanol concentration include: a) overexpression of native or heterologous expression of the enzymes of the ketoisovalerate pathway; b) overexpression of ketoacid decarboxylase; and c) overexpression of alcohol dehydrogenase. Selective increase in the concentration of the precursor acid isoketovalerate (by overexpressing of the ilvHCD genes in E. coli) together with deletion of the competing pathways that divert the isoketovalerate and pyruvate pools, resulted in an isobutanol concentration of 22 g/L, which surpassed the best reported n-butanol values (19.6 g/L) with Clostridium (Qureshi and Blaschek 1999; Atsumi et al. 2008b). Similar approaches were also investigated in S. cerevisiae, where the most effective combination was the heterologous expression of Bacillus subtilisacetolactase synthase (AlsS) to enhance the pool of 2-keto isovalerate and L. lactis KivDdecarboxylase together with deletion of the ethanol formation pathways (deletion of pyruvate decarboxylase genes) (Anthony et al. 2010; Festel et al. 2011; Urano et al. 2012; Lies et al.2012). These attempts resulted in titers of isobutanol up to 12 g/L (Urano et al. 2012; Lies et al.2012). Rerouting the isoketovalerate pathway later formed the basis of the industrial initiatives on isobutanol by Gevo and Butamax.

Similar to isobutanol, strategies have been developed to enhance the concentration of 1-butanol by re-routing the amino acid biosynthesis pathway. Some of the successful approaches include: deregulation of the threonine biosynthetic pathway; enhancing the concentration of the 2-ketovalerate by the overexpression of LeuABCD, which converts 2-ketobutyrate derived from the threonine biosynthetic pathway to 2-ketovalerate; and deletion of the interfering pathways that involve the production of 2-ketoisovalerate and pathways for threonine consumption. With all these strategies, the highest isopropanol/butanol concentration obtained was 2 g/L. Atsumi and Liao (2008c) have also constructed a more direct pathway for the production of 2-ketovalerate in E. coli by introducing the citramalate synthase pathway, which bypasses the threonine biosynthetic pathway in the production of 2-ketovalerate. This attempt resulted in a combined isopropanol/butanol production of 4 g/L (Atsumi and Liao 2008c).

Production of 3-methyl-1-butanol was accomplished by the conversion of 2-ketoisovalerate to 2-ketoisocaproate (2-keto-4-methylpentanoate), which was then decarboxylated and reduced (Connor et al. 2010). Enhancing the concentration of 2-ketoisovalerate (heterologous expression of B. subtilis alsS instead of E. coli ilvlHCD and enhancing their subsequent conversion to 2-ketoisocaproate by enhancing the activity of leu ABCD), the heterologous expression of alpha-isoketovalerate decarboxylase from L. lactis kivd and alcohol dehydrogenase from S. cerevisiaeadh2 were some of the genetic engineering tools employed. The best reported concentration has been 4.4 g/L of 3-methyl-1-butanol (Connor et al. 2010).

Similar to the production of 3-methyl-1-butanol, 2-methyl-1-butanol could also be produced from 2-keto-3-methylvalerate (Cann and Liao 2008). Acetohydroxyacid synthase (AHAS) is regarded as an important enzyme in this pathway because it catalyses the condensation of 2-ketobutyrate and pyruvate to form 2-aceto-2-hydroxy butyrate. Cann and Liao constructed an E. coli strain by overexpressing the native thrABC to enhance the pool of threonine and used threonine deaminase (IlvA) from Corynebacterium glutamicum to facilitate the transamination of threonine to 2-ketobutyrate and overexpressed AHAS gene from Salmonella typhimurium to catalyze the condensation of 2-ketobutyrate and pyruvate to form 2-aceto-2-hydroxy butyrate. Expressing the native ilvC and ilvD enabled the conversion of 2-aceto-2-hydroxy butyrate to 2-keto-3-methylvalerate, which is then decarboxylated by using L. lactis kivd and dehydrogenated using S. cerevisiae adh2 to form 2-methyl-1-butanol. This optimization including the knocking out of competing pathways resulted in a production titer of 1.3 g/L.

Researchers were able to produce long chain (C5-C8) alcohols in E. coli by expanding the natural metabolic network by producing longer chain ketoacids and respective alcohols (Zhang et al.2008). Because 2-ketoisovalerate can be naturally converted to 2-ketoisocaproate (leucine biosynthesis) by a three step chain elongation using LeuABCD, Zhang and coworkers (2008) explored the potential of LeuA, LeuB, LeuC, and LeuD to extend the elongation in the similar fashion for 2-keto-3-methylvalerate to 2-keto-4-methylhexanoate. Subsequently, L. lactisdecarboxylase (kivd) and S. cerevisiae dehydrogenase (adh6) was used to generate corresponding six-carbon alcohols, 3-methyl-1-pentanol albeit a low concentration (794 mg/L) (Zhang et al.2008; Lamsen and Atsumi 2012). Similar to 2-ketoisovaerate, 2-keto-valerate is elongated to 2-ketocaproate during isoleucine biosynthesis. The 2-ketocaproate was also shown to be further elongated by LeuABCD to 2-ketoheptanoate, which is subsequently decarboxylated and reduced to n-hexanol (Zhang et al. 2008; Lamsen and Atsumi 2012).

Similar to the heterologous expression of Clostridium pathway, re-routing amino acid biosynthesis pathways were relatively successful in E. coli compared to S. cerevisiae (Lamsen and Atsumi 2012; Buijs et al. 2013). One of the challenges with the yeasts has been the co-factor imbalances and the difficulty in dealing with a more complex system compared to bacteria (Buijs et al. 2013). Although researchers attempted to engineer for cofactor specificities of the involved oxidoreductases, the yields were substantially low (Buijs et al. 2013).

Despite a relatively higher success rate in the engineering of Clostridia pathway and amino acid biosynthesis pathway in E. coli as compared to S. cerevisiae (Rabinovitch-Deere et al. 2013), overcoming E. coli solvent toxicity presents a continued challenge (Table 3). It is apparent that E. coli can tolerate solvent concentrations in the vicinity of 1 to 2%. Increasing chain length increases the hydrophobicity of alcohol and its toxicity (Dunlop 2011). Increased hydrophobicity results in the disruption of the phospholipid component of the cell membrane (Ingram and Buttke 1984; Ingram 1986; Chin et al. 2013; Kolek et al. 2015). For example, increasing the butanol concentration from 0.5 to 1% can cause a 30% increase in the fluidity of lipid dispersal (Vollherbst-Schneck et al. 1984). The increase in the ratio of saturated to unsaturated fatty acids is a consequence of cell growth in the presence of long chain alcohols (Chin et al. 2013). Other consequences of solvent toxicity include inhibition of membrane bound ATPase activity, collapse of the membrane pH gradient, decreased intracellular ATP levels, and inhibition of the uptake of sugars and amino acids (Moreira et al. 1981; Ounine et al. 1985; Bowles and Ellefson 1985; Gottwald and Gottschalk 1985). Efforts to optimize the process strategies and decrease the butanol toxicity of Clostridia include the addition of saturated fatty acids, decreasing in temperature from 30 °C to 24 °C in the solventogenic phase, and developing mutants with increased butanol tolerance (Liu and Qureshi 2009; Dunlop 2011). Although S. cerevisiae may be a more robust industrial host that can tolerate higher product concentrations, a multi-fold increase in the product yield is necessary to achieve yields that are industrially relevant. Interestingly, isobutanol is comparatively less toxic than other alcohols, which largely enabled the researchers to obtain isobutanol titers surpassing the highest titer reported for n-butanol (Anthony et al. 2010; Festel et al. 2011; Lamsen and Atsumi 2012; Lies et al. 2012; Urano et al. 2012). Isobutanol is produced in wild type S. cerevisiae via valine metabolism. Therefore, if genetic engineering of S. cerevisiae is further optimized, isobutanol production seems promising (Hazelwood et al. 2008). However, it is worth highlighting that the amino acid biosynthesis pathway is tightly regulated in nature via feedback inhibition of the intermediates. Constructing a fast and deregulated pathway with high metabolic flux requires significant efforts on metabolic engineering and optimization (Buijs et al. 2013).

Table 3. Tolerance of Naturally Occurring, Genetically Engineered Microbial Strains to Ethanol, Butanol and Other Advanced Biofuel Molecules (Liu and Qureshi 2009; Dunlop 2011; Amyris 2013)

Fats, Lipids, and Esters

Long-chain fatty acids represent a key precursor for advanced fuels and chemicals because they can be transformed to a range of hydrocarbons and their derivatives. Fatty acids can be reduced to aldehydes and can then be subjected to decarbonylation, sequential reduction, or decarboxylation to produce equivalent alkanes as petroleum distillates (Fig. 3) (Lennen and Pfleger 2012). Fatty acid biosynthesis is one of the anabolic pathways capable of producing large hydrophobic fatty acid molecules and their derivatives (Fig. 4) Therefore, engineering fatty acid synthesis in industrial hosts such as E. coli and S. cerevisiae for the production of fatty acids is a potential route for the production of advanced biofuels and chemicals.

Fig. 3. Examples of target molecules obtained by fatty acid biosynthesis: free fatty acid (FFA), fatty acid methyl ester (FAME), triacylglyceride (TAG), fatty alcohol, and long-chain alkane

Microbial overproduction of fatty acids

Synthesis of fatty acid is an important part of lipogenesis in prokaryotes and eukaryotes. In the fatty acid biosynthesis pathway, fatty acids are generated from acetyl-CoA and malonyl-CoA precursors through action of fatty acid synthases (Steinbuechel 2014). Two carbon units are sequentially condensed or elongated onto growing fatty acyl chains. In the first step, acetyl-CoA carboxylase catalyzes the formation of malonyl-CoA from acetyl-CoA and bicarbonate. Subsequently, acetyl-CoA and malonyl-CoA converts to acetyl ACP and malonyl ACP by their respective transacylase enzymes. The next step is the condensation of aceyl-ACP and malonyl ACP to form acetoacetyl ACP (by the enzyme, ß-ketoacyl ACP synthase) followed by the reduction of acetoacetyl ACP (catalyzed by ß-keto acyl ACP reductase) and dehydration (by ß-hydroxy acyl-ACP dehydratase) and a second reduction (by enoyl-ACP reductase) to form butyryl ACP. This elongation cycle involving condensation, reduction, dehydration, and reduction is repeated (Fig. 4). After reaching the required chain length, the respective fatty acid-ACP is hydrolyzed by thioesterase to generate free fatty acid in eukaryotes, whereas in prokaryotes fatty acid-ACP is converted to glycerol-3-phosphate (Steinbuechel 2014).

Fig. 4. Biosynthesis of fatty acid and its derivatives (Peralta-Yahya and Keasling 2010; Steinbuchel and JanBen 2014)

Thioesterases have been one of the key targets of metabolic engineering to achieve the desired carbon chain length (Davis et al. 2000; Lu et al. 2008). In eukaryotes, the four enzymes necessary for these four reaction steps in the elongation cycle are present in a multi-domain protein called type I fatty acid synthase, whereas in prokaryotes each step of the elongation cycle is performed by mono-functional enzyme called type II fatty acid synthase (Steinbuechel 2014). Two strategies have been adapted to engineer microorganisms for enhanced production of fatty acids in E. coliand S. cerevisiae; one is to minimize the breakdown of fatty acids by targeting the fatty acyl-CoA synthase genes, which are involved in the aerobic and anaerobic oxidation of fatty acid. The second strategy is to overexpress the genes involved in the fatty acid biosynthesis pathway, thus diverting maximum acetyl-CoA towards the production of fatty acids (Davis et al. 2000; Lu et al.2008; Pfromm et al. 2010). Deletion of E. coli fadD and overexpression of the acetyl-CoA carboxylase (acc) and thioesterase enzymes were reported to result in a free fatty acid titer of 2.5 g/L (Davis et al. 2000; Lu et al. 2008). Engineering the reversal of β-oxidation cycle to avoid the degradation of fatty acids has resulted in the best fatty acid titer, 6.6 g/L in engineered E. coli(Dellomonaco et al. 2010). However, it was found that majority of the fatty acids produced were reabsorbed by the cell to be used as a carbon and energy source (Lu et al. 2008).

Fatty acid has been esterified in situ in E. coli by introducing pyruvate decarboxylase and alcohol dehydrogenase from Z. mobilis and overexpressing the diacylglycerol acetyl transferase from Acinetobacter baylyi, resulting in 1.3 g/L of fatty acid ethyl ester (Kalscheuer et al. 2006). Shi and coworkers (2012) heterologously expressed five different wax ester synthases in S. cerevisiae, and the best results (6.3 mg/L) were obtained with the ester synthase from Marinobacter hydrocarbonoclasticus.

Fig. 5. Examples of isoprene derivatives (hemi-, mono-, and sesquiterpenoids) obtained via the isoprenoid biosynthesis pathway

Isoprenoids and Their Derivatives

Different types of isoprenoids

Terpenes belong to a large and diverse isoprenoid family with important medical and industrial applications (Chandran et al. 2011; Bujis et al. 2013; Schwab et al. 2013) (Fig. 5). As molecules in this category are diverse, it is uncertain as to what terpene would fit best as a gasoline, diesel, or jet fuel substitute (Chandran et al. 2011). While branched chain and short terpenes seem suitable as a gasoline substitute, longer cyclic or branched chain terpenes are appropriate as a diesel/jet fuel substitute (Fig. 5). Some monoterpenes currently explored for jet fuels include pinene, terpinene, and sabinene. A sesquiterpene, α-farnesane, which is reduced from α-farnesene, has also been proposed as a jet fuel precursor due to its low hygroscopicity and high energy density (Bujis et al. 2013).

Metabolic engineering for isoprenoid production

Terpenes are microbially produced by the metabolic engineering of the isoprenoid pathway (Chandran et al. 2011) (Fig. 6), which is a cellular metabolic pathway critical to the biosynthesis of many molecules including terpenoids and steroids (Wallaart et al. 2000). As the sesquiterpenes, including α-farnesene, are naturally produced in limited quantities (Wallaart et al. 2000), metabolic engineering is essential to overproduce these compounds (Peralta-Yahya et al. 2011; Westfall et al. 2012). The basic precursor monomers for isoprenoids are isoprenyl pyrophosphate (IPP) and its isomer dimethyl allyl pyrophosphate (DMAP) (Chandran et al. 2011).

These monomers are naturally produced from the 2-C-methyl-D-erythritol-4-phosphate (MEP) pathway in E. coli (also known as deoxyxylulose pathway – DXP pathway) and the mevalonate (MVA) pathway in eukaryotes (Fig. 6). The monomers are condensed by IPP synthases to their corresponding monoterpenes, sesquiterpene, diterpenes, etc. (Peralta-Yahya et al. 2011) (Fig. 6). Resulting long-chain pyrophosphate molecules are further transformed into branched chain or cyclic alkenes by terpene synthases (Chandran et al. 2011) (Fig. 6). Finally, isoprenoid tailoring enzymes oxidize alkenes to alcohols or oxidize them to corresponding alcohols (Chandran et al.2011).

Current metabolic engineering for isoprene biofuels is based on previous efforts to genetically engineer E. coli to produce two sesquiterpene derivatives: amorphadiene, a precursor for therapeutics/nutraceuticals, and artemisinic acid, a precursor for antimalarial drug (Peralta-Yahya and Keasling, 2010; Buijs et al. 2013). Since the production of IPPs and DMAPs are critical to the synthesis of isoprenoids, many researchers have attempted to optimize the deoxyxylulose pathway (DXP) (Farmer and Liao 2000; Alper et al. 2005; Alper et al. 2006; Yuan et al. 2006) and the heterologous expression of S. cerevisiae mevalonate pathway in E. coli (Martin et al.2003) to enhance the production of these monomers (Fig. 6). Despite extensive efforts on optimizing DXP pathway, titers of amorphadiene and artemisinic acid were less than 500 mg/L (Martin et al. 2003). Much better results were obtained when the S. cerevisiae mevalonate pathway was expressed in E. coli, but it required extensive optimization efforts, as the native regulation of the pathway gets substantially altered resulting in an imbalance in gene expression and enzyme activity (Martin et al. 2003). Consequences faced were many including the depletion of the precursor compounds, accumulation of toxic intermediates, and changes in enzyme activity. For example, Pitera and coworkers (2007) found that enhanced concentrations of mevalonate led to the accumulation of the co-enzyme HMG-CoA. After extensive optimization efforts, the best titer was obtained with the combination of the heterologous expression of Staphylococcus aureus mvaS, mvaA, and E. coli atoB resulting in an amorphadiene concentration of 27 g/L. Over expression of the mevalonate pathway also enabled high titer amorphadiene production in S. cerevisiae (37 g/L) (Westfall et al. 2012).

In addition to amorphadiene and artemisinic acid, there were efforts to genetically engineer the isoprenoid pathway for biofuels. Yoshikuni and coworkers (2006) first introduced terpene synthase mutants into E. coli for the production of IPP and DMAP monomers. Eight different terpenes were obtained (α-farnesene, sabinene, γ-humulene, sibirene, longifolene, α-longpinene, α-ylange, and β-bisabolen), most of which have potential as biofuels.

Fig. 6. The 2-C-methyl-d-erythritol-4-phosphate (MEP) and mevalonate (MVA) pathways for the biosynthesis of isoprenoids and their derivatives (Chandran et al. 2011)

A U.S. company, Amyris, has been working on developing microbes to overproduce farnesene, which can be further hydrogenated to farnesane. They used S. cerevisiae as the host and genetic engineering strategies focused on enhancing the acetyl-CoA pool, precursors for the MVA pathway. Overexpression of the enzymes ald6 and Salmonella enterica acs1, adh2, and repression of ERG9, which encodes squalene synthase, and over expression of HMG1 were some of the strategies used to enhance farnesene yield (Shiba et al. 2007; Ohto et al. 2009; Asadollahi et al.2008; Asadollahi et al. 2010; Chen et al. 2012). The titer of farnesane obtained was ~14 g/L, which still far lower than what is obtained in industrial ethanol fermentation (Amyris 2013).

TECHNOLOGICAL CHALLENGES FACING ADVANCED BIOFUEL PRODUCTION

Low Product Titer and Yield

Challenges with strain development

Robust microbial strains with high product yields, titers, and tolerance to products and other inhibitory compounds are prerequisites to the economic competitiveness of advanced biofuels (Connor and Liao 2008; Nicolaou et al. 2010; Dunlop 2011; Lamsen and Atsumi 2012; Bujis et al. 2013; Dellomonaco et al. 2010; Ezeji et al. 2010). However, the current titer and yield of advanced biofuel molecules is substantially lower than first-generation ethanol (Table 3). Despite extensive efforts on the screening of natural isolates together with genetic engineering, the highest butanol tolerance so far reported is 19.6 g/L (Qureshi and Blaschek 1999) with a total solvent production of 29 g/L (Qureshi and Blaschek 1999). Similarly, tolerance limits of most isoprenoids are < 40 g/L (see Table 3) and the highest titer obtained for amorphadiene at the lab scale (in engineered S. cerevisiae) is 37 g/L (Pitera et al. 2007). Achieving high titers with free fatty acids (maximum yield ~7 g/L) is even more challenging, with significant reduction in the colony-forming unit (CFU) occurring at concentrations as low as ~1 g/L. Economic viability requires a titer of at least 100 g/L (10%) with a productivity of 2 g/L/h and a theoretical yield of > 90% (Sheridan 2009). Although first-generation bioethanol production has reached these performance levels (Amorim et al. 2011), titer, productivity, and yields of advanced biofuel systems lag considerably, even when simple substrates such as glucose are used (Peralta-Yahya and Keasling 2010; Lamsen and Atsumi 2012; Bujis et al. 2013; Branduardi et al. 2014).

In spite of recent developments of a universal gene knock-out system for Clostridia and the expected advancements in metabolic engineering (Heap et al. 2007; Papoutsakis, 2008; Jang et al.2012), Clostridia have several disadvantages in becoming an industrial host for the bulk production of fuel and chemicals (Bujis et al. 2013; Branduardi et al. 2014). Clostridia are strict anaerobes and have a slow growth rate with rapid degeneration (Jones and Wood 1986; Nimcevic and Gapes 2000). They are highly susceptible to bacterial contamination, which was one of the reasons for the closure of many ABE fermentation plants in the 19^th century (Nimcevic and Gapes 2000). Therefore, due to the familiarity of the genetics, molecular biology, and fermentation characteristics of S. cerevisiae or E. coli, constructing the Clostridia pathway in these industrial hosts seems to be a relevant strategy (Lamsen and Atsumi 2012; Bujis et al. 2013; Branduardi et al. 2014). However, both S. cerevisiae and E. coli also exhibit limited growth in > 2% butanol (Table 3). Although Gram-positive Lactobacillus strains, L. delbrueckii and L. brevis, can tolerate up to 3% butanol (Table 3), they do not naturally produce butanol (Liu and Qureshi 2009). Heterologous expression of their genetic traits in E. coli or S. cerevisiae or introducing the Clostridia pathway in these Lactobacillus strains has not been explored to a large extent. Similarly, Hypothermus butylicus produces a considerable amount of butanol, and limited genetic engineering has been carried out on this organism (Zillig et al. 1990). Thus, these areas may present opportunities for future research.

Although genetically tractable, E. coli has several limitations as an industrial organism for the bulk production of biofuels. E. coli is highly sensitive to changes in pH, temperature, salt concentration, and product concentration. In addition, most current processes based on genetically engineered E. coli or S. cerevisiae require glucose as a substrate. Utilization of sucrose present in sugarcane feedstocks presents a formidable challenge to these systems. It should be noted that most E. coli cannot directly utilize sucrose and need to be engineered for sucrose utilization (Sabri et al. 2013). Furthermore, if lignocellulosic substrates are used, in addition to exhibiting product tolerance, microbial strains will need to be tolerant to a range of inhibitors while consuming C5 and C6 sugars present in feedstocks (Dien et al. 2003; Vickers et al. 2012; Ling et al. 2014). It is apparent that to be an industrially relevant biofuel producer, the strain will require substantial metabolic engineering and optimization (Yomano et al. 1998; Alper and Stephanopoulos 2007; Miller and Ingram 2007; Knoshaug and Zhang 2009).

No known microorganism has been isolated or engineered with all the necessary traits for the bulk production of advanced biofuels. The most relevant microorganism that has been extensively used in the food, beverage, and fuel ethanol industries is the yeast, S. cerevisiae. Yeasts have many promising features such as their shear stress tolerance, reasonably stable genetic composition, short generation times, intrinsically high sugar tolerances, minimal generation of undesired metabolic by-products (due to well-controlled glycolysis), and a unique extracellular growth-permitting pH range (Amorim et al. 2011; Kitagaki and Kitamoto 2013). These factors make them favorable for alcoholic fermentation. In addition, some yeasts are able to flocculate and therefore can be easily filtered out of culture broth, a useful feature for cell recycle fermentations and product recovery (Amorim et al. 2011). These qualities make S. cerevisiae a preferred organism for industrial processes (Bujis et al. 2013). However, substantial reengineering of cellular metabolism in S. cerevisiae is required to obtain desirable yield and titers to enable industrial production of advanced biofuels (Peralta-Yahya and Keasling 2010; Lamsen and Atsumi 2012; Bujis et al. 2013; Branduardi et al. 2014).

Viability of in situ product removal techniques

While distillation process is industrially established, it would be highly energy-intensive to recover advanced biofuels from a low titer fermentation broth using distillation alone. For every ton of butanol recovered from a 2% broth, 6 tons of steam are required (Abdehagh et al. 2013) (In this article, metric units are used throughout, and the term “tons” indicates metric tons). This corresponds to a distillation cost of $54 per ton of butanol assuming the use of a natural gas boiler for steam generation and a natural gas cost of $2.95 per 1000 ft³ ($2.78 per GJ). Therefore, alternative techniques were developed that enable instantaneous product removal from the broth, which not only enriches the concentration of solvent prior to distillation, but allows continuous fermentation by maintaining a low enough titer to reduce the microbial toxicity. Some prominent strategies used for in situ product removal include gas stripping, solvent extraction, and pervaporation (Table 4) (Ezeji et al. 2003; Ezeji et al. 2010; Dürre 2007; Abdehagh et al. 2013).

Table 4. Typical in situ Product Removal Strategies for Enhancing Product Titer in ABE Fermentation (Ezeji et al. 2004; Dürre 2007; Inokuma et al. 2010; Baez et al. 2011; Abdehagh et al. 2013; Xue et al. 2014)

In gas stripping, nitrogen or fermentation gases are bubbled through the fermentation broth to strip away the products (de Vrije et al. 2013; Xue et al. 2013b; 2014a; 2014b; Ezeji et al. 2003). The stripped gas is then passed through a condenser to recover solvent vapors. After the solvents are condensed, inert gas is recycled back to the fermenter (Xue et al. 2014a). The efficiency of gas stripping depends on the flow rate, level of foam, and the presence of other components in fermentation broth (Xue et al. 2014). The highest butanol selectivity was obtained when gas stripping was performed at 67 °C (Abdehagh et al. 2013). Although batch, fed batch, and continuous fermentation can be coupled with in situ product removal, continuous fermentation had the highest beneficial effect. The highest recovered ABE concentration by this method was 420 to 460 g/L (Xue et al. 2013; Ezeji et al. 2013) and 143 g/L and 56 g/L for isopropanol and isobutanol, respectively (Table 4) (Inokuma et al. 2010). Gas stripping can be more effective when combined with adsorption, such that gas-phase molecules are adsorbed on a solid surface to recover products (Oudshoorn et al. 2009). Adsorption using hydrophobic columns is effective for butanol or higher hydrocarbon derivatives such as fatty acids or isoprenoids (Ezeji et al. 2004). Despite its low selectivity and poor removal efficiency, gas stripping does not incur the clogging or fouling associated with membrane separation (Xue et al. 2013b). Although for the reasons discussed, gas stripping is a promising alternative to distillation, it remains capital and energy intensive and therefore expensive (Xue et al. 2014b; Ezeji et al. 2010). Nonetheless, this integrated approach still reduces energy requirements 5 to 10 times over conventional distillation and costs less than directly distilling a 2% broth (Matsumura et al. 1988; Dürre 2007; Oudshoorn et al. 2009).

Efficiency of solvent extraction depends on the miscibility and affinity properties of the liquids (Matsumura et al. 1988). For butanol recovery, a non-polar solvent such as oleyl alcohol is mixed with the fermentation broth, and butanol selectively concentrates in the organic phase (Heipieper et al. 1994). An ideal solvent will have high partition coefficient for the products, be immiscible with the fermentation broth, be inexpensive and easily recoverable from the solvent, and be nontoxic to the microbial strain (Gyamerah and Glover 1996; Ezeji et al. 2004; Oudshoorn et al.2009; Dellomonaca et al. 2010). Solvent toxicity can be minimized by using a membrane between the solvent phase and fermentation broth to restrict the direct contact of the solvent. In this technique, known as perstraction, products diffuse preferentially across the membrane and dissolve in the solvent; other components and fermentation intermediates (e.g., acetic and butyric acids) are retained in the aqueous phase (Qureshi and Ezeji 2007). The main limitations with perstraction are membrane fouling and reduction in extraction efficiency (Ezeji et al. 2007).

Pervaporation combines membrane permeation and evaporation and has been recommended to enhance the recovery of ethanol and butanol from fermentation broths. The process uses a selective non-porous membrane. Target molecules diffuse through the membrane, leaving behind nutrients, sugar, and microbial cells. The permeate is removed in a vapor state from the other side of the membrane using a vacuum or sweeping gas such as N₂ (Matsumura et al. 1988; Ezeji et al.2004; Vane 2008; Fouad and Feng 2008; Oudshoorn et al. 2009; Ezeji et al. 2010). Despite the advantages over distillation, the process has drawbacks such as fouling of the membrane (Vane 2008). However, the method is highly flexible. Higher efficiencies can be achieved by modifying the membrane chemistry by changing its charge or hydrophobicity (Nguyen 1999).

Although several product removal strategies have been developed to reduce the microbial toxicity in the fermentation broth, none have proven to be efficient in recovering absolute/near absolute concentrations of the targeted product. The best recovery reported for butanol was a titer of 46% (Table 4), which needs to be further distilled. Despite the development of numerous separation methods, very few have been proven on an industrial scale (Bujis et al. 2013). For the recovery of isobutanol, DuPont uses a combination of oleyl alcohol extraction with gas stripping, whereas Gevo uses a customized process involving flash distillation followed by phase separation (Bujis et al. 2013). The feasibility of these product removal strategies needs to be investigated on other advanced biofuel molecules such as fatty acids or isoprene derivatives.

Risk of Contamination

The characteristic low product titers, low cell densities, and prolonged incubation times associated with advanced biofuel fermentation indicate that they will be highly prone to contamination, especially in anaerobic environments. One of the challenges with ABE fermentation in the 19^th century was the risk of contamination due to the low product titer (Jones and Wood 1986; Nimcevic and Gapes 2000). Notably, high gravity substrates and its resulting high bioethanol titer minimizes the risk of contamination in bioethanol fermentation (Puligundla et al. 2011). In contrast, an advanced biofuel system marked with low product tolerance makes it difficult to employ the high gravity approach (Dunlop 2011). Again, developing a microbial strain with increased substrate and product tolerance will be crucial to minimize the risk of contamination.

Additional Challenges Using Lignocellulosic Substrates

Carbon sourcing is the primary cost component of any biofuel system. Cheap sources of sugar are therefore a pre-requisite for the economic viability of any process (Singhvi et al. 2014). Sustainable production of different biofuel equivalents on a large scale invariably requires abundantly available and inexpensive lignocellulosic feedstocks, which have the added benefit that they do not compromise food security (Singhvi et al. 2014). However, there are numerous technical challenges for obtaining commercially-relevant quantities of biomass and in achieving sufficient conversion efficiencies for lignocellulosic biomass to advanced biofuels (Galbe and Zacchi 2012; Huffer et al. 2012; Singhvi et al. 2014).

Inhibitor tolerance of genetically engineered microbial strains

When using lignocellulosic substrates, in addition to exhibiting product tolerance, the microbial strains used to produce biofuels should be tolerant to a range of inhibitors derived from the lignocellulosic substrates. Several inhibitors including organic acids, phenolics, and furan derivatives can both affect cell growth and interfere with metabolic pathways for the production of advanced biofuels (Ezeji et al. 2007; Nicolaou et al. 2010; Gao et al. 2012).

After ethanol, the largest body of work has been directed to butanol production from lignocellulosic feedstocks. Researchers have examined a range of biomass sources including corn stover, corn fiber, barley straw, and switch grass, as well as a variety of pretreatment conditions to assess their influence on n-butanol production (Parekh and Blaschek 1999; Parekh and Formanek 1999; Qureshi et al. 1999; Qureshi et al. 2006; Ezeji et al. 2007; Zhang et al. 2009; Qureshi et al.2010; Lin et al. 2011; Mu et al. 2011; Gao et al. 2012; Guo et al. 2012, 2013). Sugar concentrations ranged from 31 to 58 g/L with solvent titers ranging from 2 to 26 g/L (Parekh and Blaschek 1999; Parekh and Formanek 1999; Qureshi et al. 1999, 2006; Ezeji et al. 2007; Zhang et al. 2009; Qureshi et al. 2010; Lin et al. 2011; Mu et al. 2011; Gao et al. 2012; Guo et al. 2012, 2013). The highest yield reported was 0.44 g/g (Qureshi et al. 2006, 2010), and the highest productivity was 5 g/L/h (Zhang et al. 2009). The lignocellulose-derived inhibitors consisted primarily of soluble phenolics that severely restrict the efficiency of solvent production (Guo et al. 2012).

For the production of advanced biofuels, fermentation can be carried out under aerobic/semi-aerobic conditions by re-routing the biosynthesis pathway, which could be advantageous to overcome some inhibitors (Amyris 2013). The robustness of genetically engineered strains for advanced biofuel production in surviving an inhibitor-rich lignocellulosic substrate is not yet known. However, Amyris, a company which produces farnesane using engineered S. cerevisiae, claims that their aerobic conditions helps the microbe effectively assimilate acetic acid, thus minimizing its inhibitory effect on fermentation (Amyris 2013); the influence of inhibitors such as phenolics and furan derivatives was not mentioned in their work. This is expected to present serious hurdles to commercialization, as the titer of terpenes produced from ionic liquid pretreated substrates was 12 times lower than the theoretical yield (Bokinsky et al. 2011). It appears that substantial metabolic engineering will be required to circumvent yield reduction from inhibition issues along with the ability to consume multiple sugars if reasonable production titers from lignocellulosic substrates are expected.

Previous work to increase the tolerance of micro-organisms to lignocellulosic inhibitors includes: rational engineering, evolutionary engineering, and inverse metabolic engineering (Nevoigt 2008; Jin et al. 2005). Some examples of successful rational engineering efforts include: engineering of the redox state by enhancing the intracellular glutathione by GSH1 overexpression; co-expression of transaldolase and aldehyde dehydrogenase in the presence of furfural; over expression of aldehyde dehydrogenase to regenerate NAD⁺ and relieve redox imbalance; over expression of RNA binding protein lsm6 for improved tolerance to furfural acetic acid and sulfur compounds; and combining the NADH consuming acetic acid consumption pathway and NADH producing xylose utilization pathway (Fujitomi et al. 2012; Gao and Xia 2012; Tanaka et al. 2012; Kim and Hahn 2013). Evolutionary engineering was carried out both by mutagenesis or adaptation to lignocellulose derived inhibitors. Examples include evolutionary engineering with 5HMF, which yielded a yeast mutant with high levels of ari1 and adh7, higher ethanol productivity, higher growth and higher alcohol dehydrogenase activity (Nevoigt 2008; Ling et al. 2014). To produce advanced biofuels from lignocellulosic biomass, genetic modifications used to achieve enhanced inhibitor tolerance will have to be synergistically combined with the genetic makeup required for the producing the advanced biofuels (Ling et al. 2014). Despite extensive research efforts, there has been limited success in developing a commercial microbial strain for producing ethanol that is both multiple-sugar consuming and inhibitor tolerant. As a result, substantial R&D would be required for metabolic engineering and optimization to develop a suitable microbial strain capable of producing advanced biofuels from lignocellulose (Ling et al. 2014).

High feedstock requirement

Even with a complete conversion, the typical requirement of monomeric sugars for producing most advanced biofuels would be higher than that of ethanol. As shown in Table 5, the maximum theoretical yield for isoprene is roughly 0.3 g/g sugar, whereas with ethanol it is 0.51 g/g.

This inherent limitation together with the low titer and low conversion efficiency of advanced biofuels places additional demands on feedstock requirements. For example, in ABE fermentation, low yield, together with the formation of acetone and ethanol as byproducts demands 6 to 7 tons of corn for every ton of butanol, whereas only 3 tons of corn is needed for the same amount of fuel ethanol production (Xue et al. 2013). Due to the presence of non-carbohydrate components and lower conversion efficiency, the amount of lignocellulosic substrate required would be up to 4 to 5 times higher than these values resulting in an increase in the feedstock cost. This, in turn, places additional demands on plant design as the capacity of the reactor systems will generally need to be larger, which escalates capital and operating costs.

Consumption of multiple sugars

In order to maximize the conversion efficiency of lignocellulosic biomass, both cellulose- and hemicellulose-derived sugars should be converted to products in high yield and titer. This means simultaneous or at least sequential consumption of multiple sugars (both C6 and C5 sugars) is a pre-requisite (Hahn-Hägerdal et al. 2007; Matsushika et al. 2009). Many of the biofuel producing microbial strains prefer glucose over other carbon sources due to carbon catabolite repression (Matsushika et al. 2009). Although considerable work has been done to genetically engineer yeast and bacteria to consume multiple sugars, most work was focused on the production of ethanol (Hahn-Hägerdal et al. 2007; Matsushika et al. 2009) and not on the production of advanced biofuels. Despite extensive efforts, construction of an efficient strain that can simultaneously metabolize C5 and C6 sugars while maintaining all of the required traits, (high sugar and ethanol tolerance, inhibitor tolerance etc.) for industrial scale ethanol production has been unsuccessful (Madhavan et al. 2012). Therefore, it is apparent that extensive research is required to convert both C6 and C5 sugars to advanced biofuels in reasonable titer and yield.

Table 5. Theoretical Yield of Microbially-Derived Fuel Precursors (Nakamura and Whited 2003; Atsumi et al. 2008; Huerta-Beristain et al. 2008; Whited et al. 2010; Wang et al. 2011; Huffer et al. 2012)

Despite the high robustness and industrial relevance of S. cerevisiae, the strain naturally does not ferment pentose sugars (Palmqvist and Hahn-Hagerdal 2000). In addition to integrating xylose and arabinose utilization pathways (e.g., expression of XYL1, XYL2, and XYL3 genes), there have been several efforts focusing on the over expression of specific pentose transporters and the genes of the non-oxidative part of pentose phosphate pathways (Kuyper et al. 2005; Hahn-Hagerdal et al. 2007). One of the challenges in synergistically expressing xylitol dehydrogenase and xylose reductase was their co-enzyme specificities (NADPH vs. NADH). There have been several works to solve this redox imbalance including the heterologous expression of transhydrogenase enzymes (Kuyper et al. 2005). Later efforts in expressing xylose isomerase in S. cerevisiae partially resolved this problem, as the enzyme can directly convert xylose to xylulose, which can directly enter the pentose phosphate pathway (Brat et al. 2009; Matsushika et al. 2009). This approach was adapted in genetically engineered S. cerevisiae for the production of farnesane from a mixture of glucose and xylose (Amyris 2013).

Construction of a genetically engineered strain to consume multiple sugars has been shown to compromise product tolerance and yield as well as the ability to tolerate inhibitors (Bellissimi et al. 2009; Clomburg and Gonzalez 2010; Vickers et al. 2012). Wisselink and coworkers (2009) attempted to evolve the engineered S. cerevisiae on the pentose sugars without losing their performance on hexose sugars. However, researchers found that the strain was highly sensitive to inhibitors (Bellisimi et al. 2009). Although E. coli is well known for its wide substrate utilization range (it can metabolize both C5 and C6 sugars), genetically engineered E. coli’s productivities and yields are not yet high enough for industrial use (Dien et al. 2003; Vickers et al. 2012). In addition to the recent work on advanced biofuel producing E. coli, research has been carried out for the construction of ethanologenic E. coli, which can consume multiple sugars from lignocellulose (Flores et al. 1996; Hernandez-Montalvo et al. 2001; Nichols et al. 2001; Yomano et al. 2009). Although researchers could successfully express pyruvate decarboxylase and alcohol dehydrogenase genes of Z. mobilis, the carbon catabolite expression still existed in E. coli in which glucose repressed the expression of sugar specific transporters and the key enzymes needed for the metabolism of alternative sugars (Dien et al. 2003). A more successful attempt was done by Yomino et al. (2009) in developing an E. coli ethanologenic strain by deleting the methyl glyoxal synthase gene which enabled the co-utilization of a 10% mixture of mannose, glucose, arabinose, xylose, and galactose (2% each). However, this sugar concentration is still considerably lower than what would be expected in an industrial process. The performance of this strain therefore needs to be assessed at high sugar and inhibitor concentrations.

Z. mobilis is not an ideal candidate for converting lignocellulosic sugars because it can only ferment glucose, fructose, and sucrose (Behera et al.2010). The strain lacks some of the glycolytic and pentose phosphate pathway enzymes necessary to ferment pentose sugars. Genetic engineering of Z. mobilis to express these enzymes had some successes (Zhang et al. 1995; Deanda et al. 1996), but the resulting strain had a much lower tolerance to inhibitors such as acetic acid (Lawford and Rousseau 2002; Mohagheghi et al. 2002). Through random mutagenesis and partial deletions, Ren and coworkers (2009) were able to evolve a glucose facilitator from Z. mobilis that transports xylose. However, the authors found that the intracellular xylose metabolism was inhibited by the presence of glucose (Ren et al. 2009).

Butanol-producing Clostridial strains are known for utilizing multiple sugars present in the lignocellulosic substrates including glucose, xylose, arabinose, and galactose (Ounine et al. 1983; Qureshi et al. 2007; Mu et al. 2011). When fermenting glucose, xylose, and galactose, the highest butanol titer (21.4 g/L) was obtained from glucose followed by arabinose (15.2 g/L), xylose (11.1 g/L), and galactose (10.1 g/L). Although C. acetobutylicum is recognized in utilizing both glucose and xylose, utilization of xylose is slower (Lin and Blaschek 1983; Ni and Sun 2009). The hyper-butanol producing strain C. beijerinckii BA101 can consume xylose together with glucose but exhibited poor performance in the presence of inhibitors (Ezeji et al. 2007a; b).

The Gram-negative bacterium Klebsiella oxytoca is well known for its ability to metabolize multiple sugars including some glucose oligomers: cellobiose and cellotriose (Wood and Ingram 1992). Genetic engineering of K. oxytoca to incorporate the Z. mobilis PET operon enabled the production of ethanol from a range of substrates (Wood and Ingram 1992; Wood et al. 1992). However, its ethanol tolerance was lower than that of even E. coli. Scheffersomyces stipitis (Pichia stipitis) is one of the naturally xylose fermenting strains, which can also ferment galactose, glucose, and cellobiose (Parekh et al. 1986). The redox balance between xylose reductase and xylitol dehydrogenase is less drastic in this yeast because the respective xylose reductase can use both NADH and NADPH (Verduyn et al. 1985). In addition, P. stipitis possesses a non-cytochrome electron transport chain that can resolve cofactor imbalances (Jeppsson et al. 1995). Despite these advantages, there are several limitations with P. stipitiscompared to S. cerevisiae including lower sugar consumption rates and considerably lower ethanol and inhibitor tolerance (Bellido et al. 2011). Importantly, in glucose/xylose mixtures, glucose is still the preferred, while sugar and ethanol production is optimal only in micro-aerophilic conditions (Papini et al. 2012). Being Crabtree negative yeast, no ethanol production was observed under aerobic conditions (Papini et al. 2012). All of these reasons explain why P. stipitis has not yet been considered for advanced biofuel production from lignocellulosic sugars despite its multiple substrate utilization potential.

Some thermotolerant yeast strains such as Kluveromyces and Hansenula polymorpha can ferment multiple sugars including pentoses (Lin et al. 2013; Dmytruk and Sibirny, 2013). Kluveromyces is highly Crabtree negative yeast, and its performance would heavily depend on the oxygen levels in the medium. Xylose fermentation is poor compared to that of glucose, which has a yield of up to 98% (Lin et al. 2013). Although Hansenula polymorpha was genetically modified to enhance the production of ethanol from multiple sugars, the ethanol production rate on xylose was found to be poor due to its lower uptake. No attempts have been reported to engineer these strains for the efficient production of advanced biofuels.

Most genetic engineering studies for re-routing the anabolic pathways for the production of advanced biofuels are carried out with pure glucose solutions. As a result, the ability of these strains to consume multiple sugars present in the lignocellulosic substrate and direct the metabolic flux to the desired product has yet to be engineered and optimized. The degree of carbon catabolite repression on the genetically engineered strains for the production of advanced biofuels is not yet known. With the limited success in developing E. coli / S. cerevisiae strains for the production of ethanol from the multiple sugars present in lignocellulosic substrates, construction of a microbe for lignocellulose-based advanced biofuel production would be extremely challenging.

Lack of a clean fractionation process and low-cost cellulase enzymes

The first step in the conversion of lignocellulose to advanced biofuel is the recovery of monomeric sugars in high yield and concentration. This generally requires a two-stage process involving both pretreatment and enzymatic hydrolysis to release sugars in hemicellulose and cellulose, while minimizing the formation of inhibitors (Bhutto et al. 2014). However, despite extensive efforts no such ideal fractionation process, which gives rise to high sugar yield in high concentration with a minimum amount of inhibitors, has yet to be developed. Compared to numerous pretreatment options, dilute acid/steam pretreatment is closest to commercialization (Galbe and Zacchi 2012). The cellulase enzyme loading used for hydrolysis of pretreated substrate is high and is a significant contributor to the overall production cost (Galbe and Zacchi 2012; Sorek et al.2014). Therefore, both pretreatment and enzymatic hydrolysis continue to be optimized to obtain overall high sugar yield and concentration from minimum enzyme and chemical/energy input (Hasunuma et al.2013). Above all, the presence of lignin presents a major hurdle that restricts the efficiency of pretreatment enzymatic hydrolysis processes. Incorporating a step for the valorization of lignin would potentially improve the economic viability of the overall process.

STATE OF COMMERCIALIZATION

Progress in the Commercialization of Bioethanol

Among different types of biofuel molecules that can be biochemically derived, the bioethanol industry, which has been commercialized for decades, is clearly dominant. Presently, bioethanol produced from sugar cane and starch feedstocks is the leading alternate fuel industry used throughout the world (Balat and Balat 2009; Balan et al. 2013). Although not completely optimized and economically competitive, many lignocellulose-based bioethanol plants started pre-commercial demonstration facilities (Balan et al. 2013). It should be noted that despite significant investments, current large scale cellulosic ethanol facilities deliver less than six percent of their nameplate capacity (Rapier 2015). The major companies involved in the demonstration of lignocellulosic ethanol include: Abengoa Bioenergy, Chemtex/Beta Renewable, Poet-DSM, and DowDuPont (Balan et al. 2013). The success of these emerging production platforms will depend on a variety of factors including market conditions for liquid fuels, government and social support, and technical success factors.

Early Commercialization Efforts on ABE Fermentation and Lessons Learned

ABE fermentation was industrially established prior to the emergence of fossil based butanol. Beginning in the 1920s and through the 1950s, the ABE fermentation process ranked second to ethanol fermentation (Green 2011). Many ABE fermentation plants were built during the First and Second World Wars largely due to the demand for acetone used in the manufacture of cordite (Table 6) (Dürre 1998).

The bioacetone plant at King’s Lynn, UK produced 440 kg of acetone per week using potato as the feedstock and C. acetobutylicum as a microbial strain, which was also called the Weizmann strain named after Chaim Weizmann, who first isolated this anaerobic bacterium (Gabriel 1928). The progress made by Weizmann’s team further enabled the use of corn as a feedstock, which increased the production of acetone to 2000 tons a week. Subsequently, another plant was commissioned at Dorset and six additional distilleries were retrofitted for the production of acetone. Grain shortages prompted the use of horse chestnuts as a fermentation feedstock. However, the eventual shortage of chestnuts and the challenges with foaming made the British stop domestic production, which was transferred to Canada. In 1916, the Gooderham and Worts distillery in Toronto was retrofitted to adapt the Weizmann process to produce acetone, achieving an output of 3000 tons per week. As the United States joined the war in 1917, two plants were built in Terre Haute, Indiana located in the US corn belt. Some of the challenges faced in the bulk production of acetone were the poor titer, microbial contamination, and slow growth rate of Clostridia, which resulted in long incubation times. Bacterial contamination and consequent impact on the solvent production prompted the closure of many ABE fermentation plants in 1920s (Ross 1961). However, at the end of World War II, two-thirds of butanol and one-tenth of acetone production in the Unites States were from fermentation processes (Ross 1961; Ranjan and Moholkar 2012).

Initially, butanol was considered to be a waste by-product of ABE fermentation (Nimcevic and Gapes 2000). However, rapid expansion of the automobile industry created new markets for butanol as a solvent in quick-drying lacquers, leading to its re-birth in ABE fermentation prompting investment in new plants with greater capacity despite their inherently lower titer and yield (Table 6). Ultimately, fossil-based butanol production proved much cheaper. Together with the competition for molasses as cattle feed and the consequent rise in the feedstock prices, it outcompeted bio-butanol production. This resulted in a rapid decline of the butanol fermentation plants in 1950s and their total shut down in the 1960s in both United States and Britain (Jones and Wood 1986; Nimcevic and Gapes 2000).

Table 6. Examples of Bio-butanol Plants and Major Technical Challenges Faced (Jones and Wood 1986; Nimcevic and Gapes 2000; Green 2011)

While plants were being shut down in the United States and Britain, butanol fermentation plants were still operating in China, South Africa and Russia until the early 1980s (Nimcevic and Gapes 2000; Chiao and Sun 2007). Availability of coal for energy requirements and less expensive molasses largely enabled their operation. However, the shortage of molasses in the early 1980s forced their closure as well (Nimcevic and Gapes 2000). Although butanol research has progressed considerably because its early development, there are pertinent lessons to be learned from early commercialization efforts, especially with regards to key challenges, namely: a) lack of a sustainable feedstock supply; b) the risk associated with purely “market driven” production; c) the limitation of strictly anaerobic Clostridium as an industrial microorganisms; d) the resulting low titer even with sugar cane/starch based feedstocks. It is also important to note that butanol was never used as a fuel in this early commercial era and that the applications were primarily in the production of high-value chemicals and materials (Green 2011).

Table 7. Recent Demonstration and Commercialization Activities of Bio-butanol and Isobutanol (Ni and Sun 2009; Balan et al. 2013; Bujis et al. 2013)

Current Commercialization Initiatives for n-Butanol and Isobutanol

Recent developments in butanol fermentation and an increasing surge for alternative fuel/chemicals have catalyzed increased interest in the fermentative butanol production. China is currently one of the leading countries spearheading the commercialization of the ABE fermentation process (Ni and Sun 2009). They have recently invested $200 million to install 200,000 tons of new annual production capacity with six large scale plants each producing approximately 30,000 t/a. These technologies use wild type C. acetobutylicum as the microbial strain. Fermentation is conducted in a semi-continuous mode with a residence time of 21 days. Each plant consists of a battery of approximately eight 300 to 400 m³ fermentation reactors connected in series (Chiao and Sun 2007). Most of these plants use conventional distillation techniques to recover the solvents (Chiao and Sun 2007). In Brazil, an 8000 t/a bio-butanol plant was also constructed by HC Sucroquimica to produce butanol from sugarcane juice (Ni and Sun 2009).

Extensive research advancements in the last two decades to enhance butanol/isobutanol titer by the genetic engineering of E. coli / S. cerevisiae has resulted in recent demonstration and commercialization activities by the EU and U.S.-based companies (Table 7).

For example, Cobalt Technologies (now SGBio) targets production of n-butanol from cellulosic biomass using a proprietary pretreatment and fractionation process and a wild type Clostridium strain. The company entered into a joint venture with Solvay-Rhodia, the tenth largest global chemical company, to commercialize the technology in South America, primarily using sugarcane bagasse as feedstock (Cobalt Technologies 2013). Gevo acquired a commercial scale ethanol plant to retrofit to produce isobutanol using engineered E. coli. DuPont (now DowDupont) also engineered several microbial strains for isobutanol and assigned the technology to Butamax^TM, which is a joint venture between BP and DuPont (Gevo 2014). The approach of both Butamax™ and Gevo used efficient heterologous enzymes (e.g., Bacillus subtilis AlsS, Lactococcus lactis Ilvd) in S. cerevisiae and enhance the metabolic flux towards butanol by enhancing the pyruvate pool while minimizing by-product formation (Bujis et al. 2013).

Current commercialization efforts on bio-butanol and bio-isobutanol are targeted towards high value applications. Butanol and isobutanol are currently high value chemical commodities with global markets worth $6 billion and $560 million, respectively (Green 2011; Mascal 2012). Approximately 50% of the current butanol production is used for manufacturing acrylate esters, which are polymerized for applications in surface coatings, adhesives, elastomers, textiles, fibers, and plastics (Mascal 2012). Glycol ethers are important butanol derivatives used in industrial coatings and hard surface cleaners. Butyl acetate is used in paints, leather, ink, and coatings. Other applications are in the production of butyl amines, which have applications ranging from pharmaceuticals to pesticides to nylon manufacturing (Mascal 2012). Although not as large as n-butanol market, isobutanol is a valuable commodity with applications as a solvent in adhesives and surface coatings, dispersion and flotation agents, chemical synthesis intermediates, plastics, fibers, and bio-based jet fuel blend stocks (Green 2011).

Current Commercialization Efforts for Biochemically-derived Isoprenoids and Fatty Acid Esters

Microbial production of advanced biomolecules such as long-chain alcohols, fatty acids, and isoprenoids are mostly carried out at the lab scale. From the discussion in the earlier sections, it is apparent that substantial improvement in genetic engineering and process chemistry will be needed to enable their eventual commercialization. Because genetic engineering for the production of these molecules has been done mostly in two industrial hosts (namely, E. coli and S. cerevisiae), it may be important to analyze how such engineered microbes have been used in the bulk production of commodities. However, there are but few examples where research efforts with engineered E. coli resulted in industrial scale production of biobased bulk chemicals. One excellent example is the production of 1,3-propanediol, which is used in a variety of industrial applications including adhesives and coatings. Based on E. coli fermentation, the biochemical process for 1,3-propanediol has proven commercially viable and provides a product quality advantage compared to the petrochemical route (Vickers et al. 2012). While the scale of production of isoprenoids or fatty acids for fuel applications would be substantially higher than a “niche” commodity such as 1,3-propanediol, the feasibility of using engineered E. coli for such a large scale production is yet to be proven.

Amyris is working to commercialize the bulk production of sesquiterpenes (namely, farnesene under their trade name of Biofene®) using engineered S. cerevisiae (Balan et al. 2013; Bujis et al.2013). These molecules are currently produced in their pilot-scale plants. Amyris’s farnesene strains seem to consume both hexose and pentose sugars. They have heterologously expressed E. coli’s xylose isomerase in their yeast strain, which enabled them to bypass the native route of xylose consumption in yeast (Amyris 2013). The company claims that they can produce farnesene from cellulosic feedstocks, which can then be chemically hydrogenated downstream to farnesane. As the microbial process is aerobic, the yeast cells can effectively assimilate the acetic acid, thus minimizing its inhibitory effects (Amyris 2013). However, the company has not discussed the influence of a range of other inhibitory compounds that are produced during the pretreatment of lignocellulose. Despite these achievements, the maximum reported farnesane titer is 14 g/L (Amyris 2013), which indicates that the downstream processing needed to recover the product would be as expensive as in the case of butanol.

The biochemically-derived isoprenoids are also expected to have a number of high-value applications in the materials, specialty chemical, and pharmaceutical/nutraceutical markets due to their higher selectivity and purity (Adlington et al. 2015; Hernandez 2015). It is worth noting that biochemical pathways for isoprenoids are currently used for the production of artemisimic acid and amorphadiene, which have pharmaceutical applications. Likewise, oleochemicals are sold at much higher prices than fuels. Fatty alcohols, aldehydes/esters have large established markets ($3 billion) and used in the production of soaps, detergents, cosmetic additives, pheromones, and flavoring compounds (Rupilius and Ahamad 2006; Guzialowska et al. 2012). It is apparent that bulk production of isoprenoid precursors or fatty acid derivatives that are competitive with petroleum fuels will require extensive efforts on microbial development, process optimization, and engineering. Therefore, commercial processes that leverage microbial catalysts for the production of these molecules will likely target non-fuel applications.

SUMMARY AND FUTURE PROSPECTS

In this review, developments in the microbial production of advanced biofuels were examined and compared with progress in bioethanol fermentation. In addition to looking at the natural conversion pathways such as ABE fermentation, research achievements were analyzed in the genetic engineering of microbes for enhanced production of alcohols, isoprenoids, and fatty acids. Genetic engineering efforts were focused on either overexpressing the native production pathways to maximize the innate microbial capacity or heterologously expressing suitable genes in a genetically tractable and industrially relevant hosts such as E. coli and S. cerevisiae. Four metabolic pathways were primarily focused for the production of advanced biofuels: a) heterologous expression of ABE fermentation pathway; b) re-routing the amino acid biosynthesis pathways for the production of short and long-chain alcohols; c) the fatty acid biosynthesis pathway for the production of fatty acids and their corresponding alcohols and alkanes; and d) engineering of the isoprenoid pathway for the production of isoprene derivatives.

Industrial ABE/IBE fermentation for the production of butanol and isopropanol has a proven track record. Numerous commercial plants existed prior to 1980s. However, the bio-butanol industry met its demise once low-cost fossil-based butanol entered the market. Despite its ability to consume multiple sugars, Clostridia-based bio-butanol production proved to be uncompetitive with fossil-based butanol due to its extremely low butanol tolerance (< 12 g/L), susceptibility to microbial contamination, and the slow growth rate of spore-forming Clostridia. Regardless of whether it is biomass or fossil derived, butanol was never used as a fuel, but rather used in high value applications, mostly as an industrial solvent.

Lessons learned from the early commercialization efforts for bio-butanol contributed much to the development of antibiotics and the production of recombinant proteins. Later research efforts on genetic engineering of Clostridia to enhance butanol titer and yield resulted in improved butanol producing traits. The best organism for butanol production thus far is C. beijerinckii BA 101developed by chemical mutagenesis. Using pure glucose as a feedstock, this strain has achieved the highest butanol titer of 19.6 g/L. In stark contrast, the highest ethanol titer obtained by S. cerevisiae from the industrial fermentation of starch and sugarcane-based feedstocks is 120-150 g/L, which is 6 to 8 times higher than the butanol titer. Therefore, to be an economically competitive biofuel will require substantially enhanced titers of bio-butanol. Neither heterologous expression of the Clostridial pathway nor re-routing of the amino acid biosynthesis pathway have resulted in commercially-relevant yields of biobutanol. Re-routing the amino acid biosynthesis pathway was apparently more successful for isobutanol with a production titer of 22 g/L. Development of instantaneous product removal (e.g., gas stripping) together with continuous fermentation enabled researchers to raise the butanol, isopropanol, and isobutanol concentrations to 460, 143 and 50 g/L respectively, which are acceptable titer values for further recovery by distillation. However, such in situ product removal techniques combined with distillation make the overall process highly capital intensive. Alternatively, bioethanol, as a commercially established bulk chemical, could be used as a feedstock in the chemical synthesis of long-chain alcohols and other value-added chemicals using one of the many established chemical conversion pathways.

Microbial bulk production of isoprenoids or fatty acids is at a very early stage of development and will require significant research, development, and demonstration activities to validate their potential. As such, these products will require a much longer time horizon for commercialization. Although long-chain alcohols, fatty acids, and isoprenoids were overproduced by the genetically engineered industrial hosts, E. coli and S. cerevisiae, their production has been limited to pure glucose, conducted mostly at the lab scale. The maximum titers obtained with isoprenoids (14 g/L farnesene) and fatty acid derivatives (6.6 g/L) are still too low to be commercially viable, indicating that further improvements and optimization in metabolic engineering will be necessary. Performance of these strains on sugarcane, starch-based, or lignocellulosic feedstocks are not yet fully assessed even at the lab scale, whereas bioethanol production from sugarcane and starch-based feedstocks has been commercial for years. In addition, many lignocellulosic ethanol plants have started operation at a near-commercial scale. A few studies conducted on the production of isoprene derivatives from lignocellulose indicated extremely low titers and yields in the vicinity of mg/L.

The ideal microbe, one that exhibits all of the required traits for the bulk production of advanced biofuels, does not exist. The combination of features required in an industrial biocatalyst is far from the optimal natural conditions at which these molecules are produced in a microbial cell. Achieving all of the desired traits will require the continued development of advanced genetically-engineered microbes. Although it is difficult to judge which microorganism would best suited for this purpose, our familiarity and experience in the use of yeasts for bulk industrial production of chemicals suggest that genetically engineered S. cerevisiae has the best chance to fulfill this function. S. cerevisiae has been industrially proven for bulk production in both the ethanol and food industries. While it can be argued that genetic engineering efforts with E. colihave been relatively more successful than S. cerevisiae (Rabinovitch-Deere et al. 2013), E. colihas yet to be used at a large scale. A successful example of smaller scale industrial applications of genetically engineered E. coli is the production of 1,3 propanediol and polyhydroxybutyrate (PHA), under development by Metabolix and ADM. High value “niche” industrial applications of genetically engineered E. coli include glycosylated pharmaceuticals, antibiotics, and recombinant protein production. However, the scale of production for advanced biofuels is expected to be orders of magnitude larger than these niche industrial applications and thus present a significant barrier to commercialization. In addition, E. coli’s tolerances to long-chain alcohols, fatty acids, and isoprenoids are much lower than S. cerevisiae, indicating that without enhancing the product tolerance of E. coli, the genetic modification efforts will not be effective. Either way, to be successfully applied in the production of advanced biofuels, the metabolic engineering in E. coli / S. cerevisiae must be substantially optimized.

Given the challenges for using lignocellulose, the production of advanced biofuels will require significant advances in metabolic and process engineering. Although some lignocellulosic ethanol/butanol plants are in the pre-commercial demonstration scale, these plants are not based on completely optimized processes. Continuous investments in R&D will be necessary to make advanced biofuels including lignocellulosic ethanol more competitive with fossil fuels. The upstream pretreatment and enzymatic hydrolysis steps need further research to enhance sugar yield and concentration with minimum formation of inhibitors. This will have to be done with minimum use of enzymes and chemicals. Although there have been several attempts to develop a single strain to carry out consolidated bioprocessing, the future of such an approach seems to be obscure as it requires the microbial cell to accommodate numerous and often competing functions without interfering with its basic physiological characteristics, growth rate, and tolerance. Therefore, it is highly likely that a commercial bioconversion scheme for lignocellulosic advanced biofuels will be based on a minimum of three process steps: pretreatment, enzymatic hydrolysis, and fermentation. To obtain reasonable performance in the fermentation step alone, microbial strains need to be constructed with substantial optimization to achieve fast and deregulated pathways for multiple sugar transport, good tolerance to inhibitors, high level of product tolerance, and high metabolic flux.

Due to the challenges discussed in this review and the relatively low cost of petroleum ($45 per barrel at the time of this writing), advanced biofuels are not attractive in the near to midterm. Therefore, these processes will first find commercial application in the production of chemical feedstocks, such as 1,3-propanediol or butanol. Among advanced biomolecules, butanol and isobutanol are closest to commercialization and will likely enter the chemical markets within the next few years as demonstration plants come on line. Global butanol capacity is 5 million tons with a market value over $6 billion. However, the justification for the commercial investment in biobutanol was made when the average price of industrial butanol was approximately $8/gal and pricing will need to recover before investments in additional capacity are made. Nonetheless, butanol remains an important bulk chemical with numerous applications in high-value industrial chemicals/materials such as acrylate esters, glycol ethers, butyl acetate, and butyl amines. While it may take a longer time for large scale production of microbe-derived isoprene and fatty acids to emerge, their first market applications will likely be high-value “niche” pharmaceuticals and/or nutraceutical applications.

Ethanol will continue to remain a major component of the global renewable energy mix due to its use as a fuel, oxygenate, and chemical feedstock. Moreover, its potential for growth is enormous, as bioethanol represents less than two percent of the global petroleum market, which is far below the blend wall. While starch and sugarcane will remain the primary feedstock supply for first generation bioethanol, further expansion of ethanol production will most likely rely on lignocellulosic biomass (second generation bioethanol) due to social opposition (“food versusfuel”) and subsidies that require compliance with strict carbon dioxide reduction targets. Despite the significant research advancements required, the diversity of molecules that can be produced by microbial pathways and their suitability to replace the entire spectrum of fossil-derived molecules does provide great hope for the gradual transition of our hydrocarbon-based economy to one based on carbohydrates.

REFERENCES CITED

Abdehagh, N., Tezel, F. H., and Thibault, J. (2013). “Separation techniques in butanol production: Challenges and developments,” Biomass Bioenerg. 60, 222-246. DOI: 10.1016/j.biombioe.2013.10.003

Adlington, K., El Harfi, J., Li, J., Carmichael, K., Guderian, J. A., Fox, C. B., and Irvine, D. J. (2015). “Molecular design of squalene/squalane countertypes via the controlled oligomerization of isoprene and evaluation of vaccine adjuvant applications,” Biomacromolecules 17, 165-172. DOI: 10.1021/acs.biomac.5b01285

Alsaker, K. V., Spitzer, T. R., and Papoutsakis, E. T. (2004). “Transcriptional analysis of spo0A overexpression in Clostridium acetobutylicum sporulation and its effects on the cell’s response to butanol stress,” J. Bacteriol. 186, 1959-1971. DOI: 10.1128/JB.186.7.1959–1971.2004

Alper, H., Miyaoku, K., and Stephanopoulos, G. (2005). “Construction of lycopene-overproducing E. coli strains by combining systematic and combinatorial gene knockout targets,” Nat. Biotechnol. 23, 612-616.DOI: 10.1038/nbt1083

Alper, H., Miyaoku, K., and Stephanopoulos, G. (2006). “Characterization of lycopene-overproducing E-coli strains in high cell density fermentations,” Appl. Microbiol. Biot. 72, 968-974. DOI: 10.1007/s00253-006-0357-y

Alper, H., and Stephanopoulos, G. (2007). “Global transcription machinery engineering: A new approach for improving cellular phenotype,” Metab. Eng. 9, 258-267. DOI: 10.1016/j.ymben.2006.12.002

Amorim, H. V., Lopes, M. L., de Castro Oliveira, J. V., Buckeridge, M. S., and Goldman, G. H. (2011). “Scientific challenges of bioethanol production in Brazil,” Appl. Microbiol. Biot. 91, 1267-1275. DOI: 10.1007/s00253-011-3437-6

Amyris Inc. (2013). “Scale-up and mobilization of renewable diesel and chemical production from common intermediate using US based fermentable sugar feedstocks,” Grant report. Award No: EE0002869 http://www.osti.gov/scitech/servlets/purl/1122942/

Anthony, L. C., Huang, L. L., and Ye, R. W. (2010). “Production of isobutanol in yeast mitochondria,” U.S. Patent No. 2010/0129886 A1.

Asadollahi, M. A., Maury, J., Møller, K., Nielsen, K. F., Schalk, M., Clark, A., and Nielsen, J. (2008). “Production of plant sesquiterpenes in Saccharomyces cerevisiae: effect of ERG9 repression on sesquiterpene biosynthesis,” Biotechnol. Bioeng. 99:666-677. DOI: 10.1002/bit.21581

Asadollahi, M. A., Maury, J., Schalk, M., Clark, A., and Nielsen, J. (2010). “Enhancement of farnesyl diphosphate pool as direct precursor of sesquiterpenes through metabolic engineering of the mevalonate pathway in Saccharomyces cerevisiae,” Biotechnol. Bioeng. 106, 86-96. DOI: 10.1002/bit.22668

Atsumi, S., Cann, A. F., Connor, M. R., Shen, C. R., Smith, K. M., Brynildsen, M. P., Chou, K. J. Y., Hanai, T., and Liao, J. C. (2008a). “Metabolic engineering of Escherichia coli for 1-butanol production,” Metab. Eng. 10, 305-311. DOI: 10.1016/j.ymben.2007.08.003

Atsumi, S., Hanai, T., and Liao, J. C. (2008b). “Non-fermentative pathways for synthesis of branched-chain higher alcohols as biofuels,” Nature 451, 86-U13. DOI: 10.1038/nature06450

Atsumi, S., and Liao, J. C. (2008c). “Directed evolution of Methanococcus jannaschii citramalate synthase for biosynthesis of 1-propanol and 1-butanol by Escherichia coli,” Appl. Environ. Microbiol. 74, 7802-7808. DOI: 10.1128/AEM.02046-08

Azhar, A., Bery, M., Colcord, A., Roberts, R., and Corbitt, G. (1981). “Factors affecting alcohol fermentation of wood acid hydrolysate,” Biotechnol. Bioeng. 293-300. DOI: 10.1002/bit.260230417

Banerjee, N., Bhatnagar, R., and Viswanathan, L. (1981). “Inhibition of glycolysis by furfural in Saccharomyces-Cerevisiae,” Eur. J. Appl. Microb. Biotechnol. 11, 226-228. DOI: 10.1007/BF00505872

Baez, A., Cho, K., and Liao, J. C. (2011). “High-flux isobutanol production using engineered Escherichia coli: A bioreactor study with in situ product removal,” Appl. Microbiol. Biotechnol.90, 1681-1690. DOI: 10.1007/s00253-011-3173-y

Balan, V., Chiaramonti, D., and Kumar, S. (2013). “Review of US and EU initiatives toward development, demonstration, and commercialization of lignocellulosic biofuels,” Biofuels Bioproducts & Biorefining-Biofpr 7, 732-759. DOI: 10.1002/bbb.1436

Balat, M., and Balat, H. (2009). “Recent trends in global production and utilization of bio-ethanol fuel,” Appl. Energy 86, 2273-2282. DOI: 10.1016/j.apenergy.2009.03.015

Behera, S., Mohanty, R. C., and Ray, R. C. (2010). “Comparative study of bio-ethanol production from mahula (Madhuca latifolia L.) flowers by Saccharomyces cerevisiae and Zymomonas mobilis,” Appl. Energy 87, 2352-2355. DOI: 10.1016/j.apenergy.2009.11.018

Bellido, C., Bolado, S., Coca, M., Lucas, S., Gonzalez-Benito, G., and Teresa Garcia-Cubero, M. (2011). “Effect of inhibitors formed during wheat straw pretreatment on ethanol fermentation by Pichia stipitis,” Bioresource Technol. 102, 10868-10874. DOI: 10.1016/j.biortech.2011.08.128

Bellissimi, E., van Dijken, J. P., Pronk, J. T., and van Maris, A. J. A. (2009). “Effects of acetic acid on the kinetics of xylose fermentation by an engineered, xylose-isomerase-based Saccharomyces cerevisiae strain,” Fems. Yeast Res. 9, 358-364. DOI: 10.1111/j.1567-1364.2009.00487.x

Bermejo, L. L., Welker, N. E., and Papoutsakis, E. T. (1998). “Expression of Clostridium acetobutylicum ATCC 824 genes in Escherichia coli for acetone production and acetate detoxification,” Appl. Environ. Microbiol. 64, 1079-1085.

Bhutto, A. W., Qureshi, K., Harijan, K., Zahedi, G., and Bahadori, A. (2014). “Strategies for the consolidation of biologically mediated events in the conversion of pre-treated lignocellulose into ethanol,” RSC Advances 4, 3392-3412. DOI: 10.1039/C3RA44020F

Bokinsky, G., Peralta-Yahya, P. P., George, A., Holmes, B. M., Steen, E. J., Dietrich, J., Lee, T. S., Tullman-Ercek, D., Voigt, C. A., Simmons, B. A., and Keasling, J. D. (2011). “Synthesis of three advanced biofuels from ionic liquid-pretreated switch grass using engineered Escherichia coli,” Proc. Natl. Acad. Sci. USA 108(50), 19949-19954. DOI: 10.1073/pnas.1106958108

Bowles, L. K., and Ellefson, W. L. (1985). “Effects of butanol on Clostridium-Acetobutylicum,” Appl. Environ. Microbiol. 50, 1165-1170.

Branduardi, P., de Ferra, F., Longo, V., and Porro, D. (2014). “Microbial n-butanol production from Clostridia to non-Clostridial hosts,” Eng. Life Sci. 14, 16-26. DOI: 10.1002/elsc.201200146

Brat, D., Boles, E., and Wiedemann, B. (2009). “Functional expression of a bacterial xylose isomerase in Saccharomyces cerevisiae,” Appl. Environ. Microbiol. 75, 2304-2311. DOI: 10.1128/AEM.02522-08

Buijs, N. A., Siewers, V., and Nielsen, J. (2013). “Advanced biofuel production by the yeast Saccharomyces cerevisiae,” Curr. Opin. Chem. Biol. 17, 480-488. DOI: 10.1016/j.cbpa.2013.03.036

Cann, A. F, and Liao, J. C. (2008). “Production of 2-methyl-1-butanol in engineered Es-cherichia coli,” Appl. Microbiol. Biot. 81, 89-98. DOI: 10.1007/s00253-008-1631-y

Carmann, T. M. (2011). “Bio-butanol: Profile of an advanced biofuel and its path to market,” Int. Sugar J. 113, 180-187.

Chandran, S. S., Kealey, J. T., and Reeves, C. D. (2011). “Microbial production of iso-prenoids,” Process Biochem. 46(9), 1703-1710. DOI: 10.1016/j.procbio.2011.05.012

Chen, J. S. (1995). “Alcohol-dehydrogenase – Multiplicity and relatedness in the solvent-producing Clostridia,” FEMS Microbiol. Rev. 17, 263-273. DOI: 10.1111/j.1574-6976.1995.tb00210.x

Chen, J. S., and Hiu, S. F. (1986). “Acetone butanol isopropanol production by Clostridium beijerinckii (Synonym, Clostridium butylicum),” Biotechnol. Lett. 8, 371-376. DOI: 10.1007/BF01040869

Chen, S., Chin, W., Tsuge, K., Huang, C., and Li, S. (2013). “Fermentation approach for enhancing 1-butanol production using engineered butanologenic Escherichia coli,” Biores. Tech.145, 204-209. DOI: 10.1016/j.biortech.2013.01.115

Chen, Y., Daviet, L., Schalk, M., Siewers, V., and Nielsen, J. (2012). “Establishing a platform cell factory through engineering of acetyl-CoA metabolism,” Metab. Eng. 15, 48-54 DOI: 10.1016/j.ymben.2012.11.002

Chiao, J., and Sun, Z. (2007). “History of the acetone-butanol-ethanol fermentation industry in China: Development of continuous production technology,” J. Mol. Microbiol. Biotechnol. 13, 12-14. DOI: 10.1159/000103592

Chin, W., Lin, K., Chang, J., and Huang, C. (2013). “Improvement of n-butanol tolerance in Escherichia coli by membrane-targeted tilapia metallothionein,” Biotechnol. Biofuels 6(130), 1-9. DOI: 10.1186/1754-6834-6-130

Collas, F., Kuit, W., Clément, B., Marchal, R., López-Contreras, A. M., and Monot, F. (2012). “Simultaneous production of isopropanol, butanol, ethanol and 2,3-butanediol by Clostridium acetobutylicum ATCC 824 engineered strains,” AMB Express 2(45), 1-10. DOI: 10.1186/2191-0855-2-45

Clark, T., and Mackie, K. (1984). “Fermentation inhibitors in wood hydrolysates derived from the softwood Pinus-Radiata,” J. Chem. Technol. Biotech. B-Biotechnology 34, 101-110. DOI: 10.1002/jctb.280340206

Clomburg, J. M., and Gonzalez, R. (2010). “Biofuel production in Escherichia coli: The role of metabolic engineering and synthetic biology,” Appl. Microbiol. Biot. 86, 419-434. DOI: 10.1007/s00253-010-2446-1

Cobalt Technologies Inc. (2014). “http://www.cobalttech.com/executive-summary.html Last accessed on 12 March (2014).

Connor, M. R., Cann, A. F., and Liao, J. C. (2010). “3-Methyl-1-butanol production in Escherichia coli: Random mutagenesis and two-phase fermentation,” Appl. Microbiol. Biot. 86, 1155-1164. DOI: 10.1007/s00253-009-2401-1

Connor, M. R., and Liao, J. C. (2008). “Engineering of an Escherichia coli strain for the production of 3-methyl-1-butanol,” Appl. Environ. Microbiol. 74, 5769-5775.DOI: 10.1128/AEM.00468-08

Davis, M. S., Solbiati, J., and Cronan, J. E. (2000). “Overproduction of acetyl-CoA carboxylase activity increases the rate of fatty acid biosynthesis in Escherichia coli,” J. Biol. Chem. 275, 28593-28598. DOI: 10.1074/jbc.M004756200

Deanda, K., Zhang, M., Eddy, C., and Picataggio, S. (1996). “Development of an arabinose-fermenting Zymomonas mobilis strain by metabolic pathway engineering,” Appl. Environ. Microbiol. 62:4465-4470.

Dellomonaco, C., Fava, F., and Gonzalez, R. (2010). “The path to next generation biofuels: successes and challenges in the era of synthetic biology,” Microbial Cell Factories 9(3), 1-15. DOI: 10.1186/1475-2859-9-3

de Vrije, T., Budde, M., van der Wal, H., Claassen, P. A. M., and Lopez-Contreras, A. M. (2013). “In situ removal of isopropanol, butanol and ethanol from fermentation broth by gas stripping,” Bioresource Technol. 137, 153-159. DOI: 10.1016/j.biortech.2013.03.098

Dien, B., Cotta, M., and Jeffries, T. (2003). “Bacteria engineered for fuel ethanol production: Current status,” Appl. Microbiol. Biot. 63, 258-266. DOI: 10.1007/s00253-003-1444-y

Dmytruk, K. V., and Sibirny, A. A. (2013). “Metabolic engineering of the yeast Hansenula polymorpha for the construction of efficient ethanol producers,” Cytology and Genetics 47, 329-342. DOI: 10.3103/S0095452713060029

Dunlop, M. J. (2011). “Engineering microbes for tolerance to next-generation biofuels,” Biotechnol. Biofuels 4(32), 1-9. DOI: 10.1186/1754-6834-4-32

Dürre, P. (2007). “Bio-butanol: An attractive biofuel,” Biotechnol. J. 2(12), 1525-1534. DOI: 10.1002/biot.200700168

Ezeji, T. C., Qureshi, N., and Blaschek, H. P. (2003). “Production of acetone, butanol and ethanol by Clostridium beijerinckii BA101 and in situ recovery by gas stripping,” World J. Microbiol. Biotechnol. 19, 595-603. DOI: 10.1023/A:1025103011923

Ezeji, T. C., Qureshi, N., and Blaschek, H. P. (2004). “Acetone butanol ethanol (ABE) production from concentrated substrate: Reduction in substrate inhibition by fed-batch technique and product inhibition by gas stripping,” Appl. Microbiol. Biot. 63, 653-658. DOI: 10.1007/s00253-003-1400-x

Ezeji, T. C., Qureshi, N., and Blaschek, H. P. (2007). “Bioproduction of butanol from biomass: From genes to bioreactors,” Curr. Opin. Biotechnol. 18(3), 220-227. DOI: 10.1016/j.copbio.2007.04.002

Ezeji, T., Milne, C., Price, N. D., and Blaschek, H. P. (2010). “Achievements and perspectives to overcome the poor solvent resistance in acetone and butanol-producing microorganisms,” Appl. Microbiol. Biot. 85, 1697-1712. DOI: 10.1007/s00253-009-2390-0

Ezeji, T. C., Qureshi, N., and Blaschek, H. P. (2013). “Microbial production of a biofuel (acetone-butanol-ethanol) in a continuous bioreactor: impact of bleed and simultaneous product removal,” Bioprocess Biosyst. Eng. 36, 109-116. DOI: 10.1007/s00449-012-0766-5

Farmer, W. R., and Liao, J. C. (2000). “Improving lycopene production in Escherichia coli by engineering metabolic control,” Nat. Biotechnol. 18, 533-537. DOI: 10.1038/75398

Festel, G., Boles, E., Weber, C., and Brat, D. (2011). “Fermentative production of isobutanol with yeast,” U.S. Patent No. 2011/053235 A1.

Fischer, C. R., Klein-Marcuschamer, D., and Stephanopoulos, G. (2008). “Selection and optimization of microbial hosts for biofuels production,” Metab. Eng. 10, 295-304. DOI: 10.1016/j.ymben.2008.06.009

Fortman, J. L., Chhabra, S., Mukhopadhyay, A., Chou, H., Lee, T.S., Steen, E., and Keasling, J. D. (2008). “Biofuel alternatives to ethanol: pumping the microbial well,” Trends Biotechnol. 26, 375-381. DOI: http://dx.doi.org/10.1016/j.tibtech.2008.03.008

Fouad, E. A., and Feng, X. (2008). “Use of pervaporation to separate butanol from dilute aqueous solutions: Effects of operating conditions and concentration polarization,” J. Membr. Sci. 323(2), 428-435. DOI: 10.1016/j.memsci.2008.06.054

Fujitomi, K., Sanda, T., Hasunuma, T., and Kondo, A. (2012). “Deletion of the PHO13 gene in Saccharomyces cerevisiae improves ethanol production from lignocellulosic hydrolysate in the presence of acetic and formic acids, and furfural,” Bioresource Technol. 111, 161-166 DOI: 10.1016/j.biortech.2012.01.161

Galbe, M., and Zacchi, G. (2012). “Pretreatment: The key to efficient utilization of lignocellulosic materials,” Biomass Bioenerg. 46, 70-78.DOI: 10.1016/j.biombioe.2012.03.026

Gao, K., Li, Y., Tian, S., and Yang, X. (2012). “Screening and characteristics of a butanol-tolerant strain and butanol production from enzymatic hydrolysate of NaOH-pretreated corn stover,” World J. Microbiol. Biotechnol. 28, 2963-2971. DOI: 10.1007/s11274-012-1107-1

Gao, L., and Xia, L. (2012). “Sm-like protein enhanced tolerance of recombinant Saccharomyces cerevisiae to inhibitors in hemicellulosic hydrolysate,” Bioresource Technol. 124, 504-507. DOI: 10.1016/j.biortech.2012.08.104

George, H. A., and Chen, J. S. (1983). “Acidic conditions are not obligatory for onset of butanol formation by Clostridium beijerinckii (Synonym, Clostridium butylicum),” Appl. Environ. Microbiol. 46, 321-327.

Gevo. (2014). (http://www.gevo.com/about/our-business/our-science-and-technology/), Accessed on 14 March 2014.

Ghosh, P., Hickey, K. J., and Jaffe, S. B. (2006). “Development of a detailed gasoline composition-based octane model,” Ind. Eng. Chem. Res. 45, 337-345. DOI: 10.1021/ie050811h

Girbal, L., and Soucaille, P. (1998). “Regulation of solvent production in Clostridium acetobutylicum,” Trends Biotechnol. 16, 11-16. DOI: http://dx.doi.org/10.1016/S0167-7799(97)01141-4

Goihberg, E., Peretz, M., Tel-Or, S., Dym, O., Shimon, L., Frolow, F., and Burstein, Y. (2010). “Biochemical and structural properties of chimeras constructed by exchange of cofactor-binding domains in alcohol dehydrogenases from thermophilic and mesophilic microorganisms,” Biochemistry (NY) 49, 1943-1953. DOI: 10.1021/bi901730x

Gottwald, M., and Gottschalk, G. (1985). “The internal pH of Clostridium acetobutylicum and its effect on the shift from acid to solvent formation,” Arch. Microbiol. 143, 42-46. DOI: 10.1007/BF00414766

Green, E. M. (2011). “Fermentative production of butanol—The industrial perspective,” Curr. Opin. Biotechnol. 22, 337-343. DOI: 10.1016/j.copbio.2011.02.004

Groot, W. J., and Luyben, K. C. A. M. (1986). “In situ product recovery by adsorption in the butanol isopropanol batch fermentation,” Appl. Microbiol. Biotechnol. 25, 29-31. DOI: 10.1007/BF00252508

Guo, T., Tang, Y., and Zhang, Q.-Y. (2012). Clostridium beijerinckii mutant with high inhibitor tolerance obtained by low-energy ion implantation,” J. Ind. Microbiol. Biotechnol. 39, 401-407. DOI: 10.1007/s10295-011-1017-5

Guo, T., He, A., and Du, T. (2013). Butanol production from hemicellulosic hydrolysate of corn fiber by a Clostridium beijerinckii mutant with high inhibitor-tolerance,” Bioresource Technol.135, 379-385. DOI: 10.1016/j.biortech.2012.08.029

Guzialowska-Tic, J., Hreczuch, W., and Tic, W. J. (2012). “Market search of oils and fats as raw materials for the oleochemical industry,” Przemysl Chemiczny 91, 1185-1190.

Gyamerah, M., and Glover, J. (1996). “Production of ethanol by continuous fermentation and liquid-liquid extraction,” J. Chem. Technol. Biotech. 66, 145-152. DOI: 10.1002/(SICI)1097-4660(199606)66:2<145

Hahn-Hägerdal, B., Karhumaa, K., Jeppsson, M., and Gorwa-Grauslund, M. F. (2007). “Metabolic engineering for pentose utilization in Saccharomyces cerevisiae,” Biofuels 108, 147-177. DOI: 10.1007/10_2007_062

Hanai, T., Atsumi, S., and Liao, J. C. (2007). “Engineered synthetic pathway for isopropanol production in Escherichia coli,” Appl. Environ. Microbiol. 73, 7814-7818. DOI: 10.1128/AEM.01140-07

Harris, L. M., Blank, L., Desai, R. P., Welker, N. E., and Papoutsakis, E. T. (2001). “Fermentation characterization and flux analysis of recombinant strains of Clostridium acetobutylicum with an inactivated solR gene,” J. Ind. Microbiol. Biotechnol. 27, 322-328. DOI: 10.1038/sj/jim/7000191

Hasunuma, T., Okazaki, F., Okai, N., Hara, K. Y., Ishii, J., and Kondo, A. (2013). “A review of enzymes and microbes for lignocellulosic biorefinery and the possibility of their application to consolidated bioprocessing technology,” Bioresource Technol. 135, 513-522. DOI: 10.1016/j.biortech.2012.10.047

Hazelwood, L. A., Daran, J., van Maris, A. J. A., Pronk, J. T., and Dickinson, J. R. (2008). “The Erlich pathway for fusel alcohol production: A century of research on Saccharomyces cerevisiaemetabolism,” Appl. Environ. Microbiol. 74, 3920-3920. DOI: 10.1128/AEM.02625-07

Heipieper, H. J., Weber, F. J., Sikkema, J., Keweloh, H., and Debont, J. A. M. (1994). “Mechanisms of resistance of whole cells to toxic organic solvents,” Trends Biotechnol. 12(10), 409-415. DOI: http://dx.doi.org/10.1016/0167-7799(94)90029-9

Hernandez, E.M. (2015). “Specialty Oils: Functional and Nutraceutical Properties,” in: Functional Dietary Lipids: Food Formulation, Consumer Issues and Innovation for Health,T.A.B. Sanders, (ed.), Elsevier, New York, pp. 69-102.

Hernandez-Montalvo, V., Valle, F., Bolivar, F., and Gosset, G. (2001). “Characterization of sugar mixtures utilization by an Escherichia coli mutant devoid of the phosphotransferase system,” Appl. Microbiol. Biot. 57, 186-191. DOI: 10.1007/s002530100752

Hu, S., Zheng, H., Gu, Y., Zhao, J., Zhang, W., Yang, Y., Wang, S., Zhao, G., Yang, S., and Jiang, W. (2011). “Comparative genomic and transcriptomic analysis revealed genetic characteristics related to solvent formation and xylose utilization in Clostridium acetobutylicum EA 2018,” BMC Genomics 12(93), 1-18. DOI: 10.1186/1471-2164-12-93

Huang, L., Gibbins, L. N., and Forsberg, C. W. (1985). “Transmembrane pH gradient and membrane-potential in Clostridium acetobutylicum during growth under acetogenic and solventogenic conditions,” Appl. Environ. Microbiol. 50, 1043-1047.

Huerta-Beristain, G., Utrilla, J., Hernandez-Chavez, G., Bolivar F, Gosset, G., and Martinez, A. (2008). “Specific ethanol production rate in ethanologenic Escherichia coli strain KO11 is limited by pyruvate decarboxylase,” J. Mol. Microbiol. Biotechnol. 15, 55-64. DOI: 10.1159/000111993

Ingram, L., and Buttke, T. (1984). “Effects of alcohols on microorganisms,” Adv. Microb. Physiol.25, 253-300.

Ingram, L. O. (1986). “Microbial tolerance to alcohols – Role of the cell-membrane,” Trends Biotechnol. 4, 40-44. DOI: http://dx.doi.org/10.1016/0167-7799(86)90152-6

Inokuma, K., Liao, J.C., Okamoto, M., and Hanai, T. (2010). “Improvement of isopropanol production by metabolically engineered Escherichia coli using gas stripping,” J. Biosci. Bioeng.110, 696-701. DOI: 10.1016/j.jbiosc.2010.07.010

International Energy Agency (IEA). (2011). Technology Roadmap: Biofuels for Transport, (http://www.iea.org/publications/freepublications/publication/biofuels_roadmap.pdf), Accessed on 12 February 2014. DOI: 10.1787/9789264118461-en

International Energy Agency (IEA). (2013). (http://www.worldenergyoutlook.org/), Accessed on 17 April 2014.

Inui, M., Suda, M., Kimura, S., Yasuda, K., Suzuki, H., Toda, H., Yamamoto, S., Okino, S., Suzuki, N., and Yukawa, H. (2008). “Expression of Clostridium acetobutylicum butanol synthetic genes in Escherichia coli,” Appl. Microbiol. Biot. 77, 1305-1316.

Ismaiel, A. A, Zhu, C. X, Colby, G. D, and Chen, J. S. (1993). “Purification and characterization of a primary-secondary alcohol-dehydrogenase from 2 strains of Clostridium beijerinckii,” J. Bacteriol. 175:5097-5105.

Jang, Y. S., Malaviya, A., Cho, C., and Lee J. (2012). “Butanol production from renewable biomass by Clostridia,” Bioresource Technol. 123, 653-663.DOI: 10.1016/j.biortech.2012.07.104

Jeppsson, H., Alexander, N. J., and Hahnhagerdal, B. (1995). “Existence of cyanide-insensitive respiration in the yeast Pichia-stipitis and its possible influence on product formation during xylose utilization,” Appl. Environ. Microbiol. 61, 2596-2600.

Jin, Y., Alper, H., Yang, Y., and Stephanopoulos, G. (2005). “Improvement of xylose uptake and ethanol production in recombinant Saccharomyces cerevisiae through an inverse metabolic engineering approach,” Appl. Environ. Microbiol. 71, 8249-8256. DOI: 10.1128/AEM.71.12.8249-8256.2005

Jojima, T., Inui, M., and Yukawa, H. (2008). “Production of isopropanol by metabolically engineered Escherichia coli,” Appl. Microbiol. Biot. 77, 1219-1224. DOI: 10.1007/s00253-007-1246-8

Jones, D. T., and Woods, D. R. (1986). “Acetone-butanol fermentation revisited,” Microbiol. Rev.50, 484-524.

Kalscheuer, R., Stolting. T., and Steinbuchel, A. (2006). “Microdiesel: Escherichia coliengineered for fuel production,” Microbiol. 152, 2529-2536. DOI: 10.1099/mic.0.29028-0

Kim, D., and Hahn, J. S. (2013). “Roles of the Yap1 transcription factor and antioxidants in Saccharomyces cerevisiae tolerance to furfural and 5-hydroxymethylfurfural, which function as thiol-reactive electrophiles generating oxidative stress,” Appl. Environ. Microbiol. 79, 5069-5077. DOI: 10.1128/AEM.00643-13

Kitagaki, H., and Kitamoto, K. (2013). “Breeding research on sake yeasts in Japan: History, recent technological advances, and future perspectives,” Annu. Rev. Food Sci. Technol. 4, 215-235. DOI: 10.1146/annurev-food-030212-182545

Knoshaug, E. P., and Zhang, M. (2009). “Butanol tolerance in a selection of microorganisms,” Appl. Biochem. Biotechnol. 153, 13-20. DOI: 10.1007/s12010-008-8460-4

Kolek, J., Patáková, P., Melzoch, K., Sigler, K., and Řezanka, Y. (2015). “Changes in membrane plasmalogens of Clostridium pasteurianum during butanol fermentation as determined by lipidomic analysis,” PLoS One 10(3), e0122058. DOI: 10.1371/journal.pone.0122058

Korkhin, Y., Kalb, A. J., Peretz, M., Bogin, O., Burstein, Y., and Frolow, F. (1998). “NADP-dependent bacterial alcohol dehydrogenases: Crystal structure, cofactor-binding and cofactor specificity of the ADHs of Clostridium beijerinckii and Thermoanaerobacter brockii,” J. Mol. Biol. 278, 967-981. DOI: 10.1006/jmbi.1998.1750

Kuyper, M., Hartog, M. M. P., Toirkens, M. J., Almering, M. J. H., Winkler, A. A., van Dijken, J. P., and Pronk, J. T. (2005). “Metabolic engineering of a xylose-isomerase-expressing Saccharomyces cerevisiae strain for rapid anaerobic xylose fermentation,” Fems. Yeast Res. 5, 399-409. DOI: 10.1016/j.femsyr.2004.09.010

Lamsen, E. N., and Atsumi, S. (2012). “Recent progress in synthetic biology for microbial production of C3-C10 alcohols,” Frontiers Microbiol. 3(196), 1-13. DOI: 10.3389/fmicb.2012.00196

Lawford, H. G., and Rousseau, J. D. (2002). “Performance testing of Zymomonas mobilismetabolically engineered for cofermentation of glucose, xylose, and arabinose,” Appl. Biochem. Biotechnol. 98, 429-448. DOI: 10.1385/ABAB:98-100:1-9:429

Lee, S. K., Chou, H., Ham, T. S., Lee, T. S., and Keasling, J. D. (2008). “Metabolic engineering of microorganisms for biofuels production: From bugs to synthetic biology to fuels,” RID E-6525-(2010), Curr. Opin. Biotechnol. 19, 556-563. DOI: 10.1016/j.copbio.2008.10.014

Lee, J., Jang, Y., Choi, S. J., Im, J. A., Song, H., Cho, J. H., Seung, D. Y., Papoutsakis, E. T., Bennett, G. N., and Lee, S. Y. (2012). “Metabolic engineering of Clostridium acetobutylicum ATCC 824 for isopropanol-butanol-ethanol fermentation,” Appl. Environ. Microbiol. 78, 1416-1423. DOI: 10.1128/AEM.06382-11

Lennen, R. M., and Pfleger, B. F. (2012). “Engineering Escherichia coli to synthesize free fatty acids,” Trends Biotechnol. 30, 659-667. DOI: http://dx.doi.org/10.1016/j.tibtech.2012.09.006

Li, H., Cann, A. F., and Liao, J. C. (2010). “Biofuels: Biomolecular engineering fundamentals and advances,” Annu. Rev. Chem. Biomol. Eng. 119-361. DOI: 10.1146/annurev-chembioeng-073009-100938

Lies, D., Porter-Scheinman, S., Kelly, J., Dundon, C. A., Aristidou, A., and Hawkins, A. (2012). “Reduced by-product accumulation for improved production of isobutanol,” U.S. Patent No. 8158404 B2.

Lin, Y. L., and Blaschek, H. P. (1983). “Butanol production by a butanol-tolerant strain of Clostridium-Acetobutylicum in extruded corn broth,” Appl. Environ. Microbiol. 45, 966-973.

Lin, Y., Wang, J., Wang, X., and Sun, X. (2011). “Optimization of butanol production from corn straw hydrolysate by Clostridium acetobutylicum using response surface method,” China Sci. Bull. 56, 1422-1428. DOI: 10.1007/s11434-010-4186-0

Lin, Y., Lee, W., Duan, K., and Lin, Y. (2013). “Ethanol production by simultaneous saccharification and fermentation in rotary drum reactor using thermotolerant Kluveromyces marxianus,” Appl. Energy 105, 389-394. DOI: 10.1016/j.apenergy.2012.12.020

Liu, S., and Qureshi, N. (2009). “How microbes tolerate ethanol and butanol,” New Biotechnol.26, 117-121. DOI: 10.1016/j.nbt.2009.06.984

Lu, X., Vora, H., and Khosla, C. (2008). “Overproduction of free fatty acids in E. coli: Implications for biodiesel production,” Metab. Eng. 10, 333-339. DOI: 10.1016/j.ymben.2008.08.006

Madhavan, A., Srivastava, A., Kondo, A., and Bisaria, V. S. (2012). “Bioconversion of lignocellulose-derived sugars to ethanol by engineered Saccharomyces cerevisiae,” Crit. Rev. Biotechnol. 32, 22-48. DOI: 10.3109/07388551.2010.539551

Martin, V. J. J., Pitera, D. J., Withers, S. T., Newman, J. D., and Keasling, J. D. (2003). “Engineering a mevalonate pathway in Escherichia coli for production of terpenoids,” Nat. Biotechnol. 21, 796-802. DOI: 10.1038/nbt833

Mascal, M. (2012). “Chemicals from bio-butanol: Technologies and markets,” Biofuels Bioprod. Bioref. 6, 483-493. DOI: 10.1002/bbb.1328

Matsumura, M., Kataoka, H., Sueki, M., and Araki, K. (1988). “Energy saving effect of pervaporation using oleyl alcohol liquid membrane in butanol purification,” Bioprocess Eng. 3(2), 93-100. DOI: 10.1007/BF00369334

Matsumura, M., Takehara, S., and Kataoka, H. (1992). “Continuous butanol isopropanol fermentation in down-flow column reactor coupled with pervaporation using supported liquid membrane,” Biotechnol. Bioeng. 39, 148-156. DOI: 10.1002/bit.260390205

Matsushika, A., Inoue, H., Kodaki, T., and Sawayama, S. (2009). “Ethanol production from xylose in engineered Saccharomyces cerevisiae strains: Current state and perspectives,” Appl. Microbiol. Biot. 84, 37-53. DOI: 10.1007/s00253-009-2101-x

Mermelstein, L. D, Papoutsakis, E. T., Petersen, D. J., and Bennett, G. N. (1993). “Metabolic engineering of Clostridium acetobutylicum ATCC 824 for increased solvent production by enhancement of acetone formation enzyme-activities using a synthetic acetone operon,” Biotechnol. Bioeng. 42, 1053-1060. DOI: 10.1002/bit.260420906

Miller, E. N., and Ingram, L. O. (2007). “Combined effect of betaine and trehalose on osmotic tolerance of Escherichia coli in mineral salts medium,” Biotechnol. Lett. 29, 213-217. DOI: 10.1007/s10529-006-9226-0

Mohagheghi, A., Evans, K., Chou, Y.C., and Zhang, M. (2002). “Cofermentation of glucose, xylose, and arabinose by genomic DNA-integrated xylose/arabinose fermenting strain of Zymomonas mobilis AX101,” Appl. Biochem. Biotechnol. 98, 885-898. DOI: 10.1007/BF02920145

Moreira, A. R., Ulmer, D. C., and Linden, J. C. (1981). “Butanol toxicity in the butylic fermentation,” Biotechnol. Bioeng. 11, 567-579.

Mu, X., Sun, W., Liu, C., and Wang, H. (2011). “Improved efficiency of separate hexose and pentose fermentation from steam-exploded corn stalk for butanol production using Clostridium beijerinckii,” Biotechnol. Lett. 33, 1587-1591. DOI: 10.1007/s10529-011-0598-4

Nair, R. V., Bennett, G. N., and Papoutsakis, E. T. (1994). “Molecular characterization of an aldehyde/alcohol dehydrogenase gene from Clostridium acetobutylicum ATCC 824,” J. Bacteriol.176, 871-885. DOI: 10.1128/JB.184.3.821-830.2002 J

Nair, R. V., Green, E. M., Watson, D. E., Bennett, G. N., and Papoutsakis, E. T. (1999). “Regulation of the sol locus genes for butanol and acetone formation in Clostridium acetobutylicum ATCC 824 by a putative transcriptional repressor,” J. Bacteriol. 181, 319-330.

Nakamura, C. E., and Whited, G. M. (2003). “Metabolic engineering for the microbial production of 1,3-propanediol,” Curr. Opin. Biotechnol. 14, 454-459. DOI: 10.1016/j.copbio.2003.08.005

Nevoigt, E. (2008). “Progress in metabolic engineering of Saccharomyces cerevisiae,” Microbiol. Mol. Bio. Rev. 72, 379-412. DOI: 10.1128/MMBR.00025-07

Nguyen, Q. T. (1999). “Organic membrane preparation techniques for specific separations,” Actualite Chimique (2), 6-13.

Ni, Y., and Sun, Z. (2009). “Recent progress on industrial fermentative production of acetone-butanol-ethanol by Clostridium acetobutylicum in China,” Appl. Microbiol. Biot. 83, 415-423. DOI: 10.1007/s00253-009-2003-y

Nicolaou, S. A., Gaida, S. M., and Papoutsakis, E. T. (2010). “A comparative view of metabolite and substrate stress and tolerance in microbial bioprocessing: From biofuels and chemicals, to biocatalysis and bioremediation,” Metab. Eng. 12, 307-331. DOI: 10.1016/j.ymben.2010.03.004

Nichols, N. N., Dien, B. S., and Bothast, R. J. (2001). “Use of catabolite repression mutants for fermentation of sugar mixtures to ethanol,” Appl. Microbiol. Biot. 56, 120-125. DOI: 10.1007/5002530100628

Nielsen, D. R., Leonard, E., Yoon, S., Tseng, H., Yuan, C., and Prather, K. L. J. (2009). “Engineering alternative butanol production platforms in heterologous bacteria,” Metab. Eng. 11, 262-273. DOI: 10.1016/j.ymben.2009.05.003

Nimcevic, D., and Gapes, J. (2000). “The acetone-butanol fermentation in pilot plant and pre-industrial scale,” J. Mol. Microbiol. Biotechnol. 2, 15-20.

Nolling, J., Breton, G., Omelchenko, M. V., Makarova, K. S., Zeng, Q., Gibson, R., Lee, H. M., Dubois, J., Qiu, D., Hitti, J., Wolf, Y. I., Tatusov, R. L., Sabathe, F., Doucette-Stamm, L., Soucaille, P., Daly, M. J., Bennett, G. N., Koonin, E. V., and Smith, D. R. (2001). “Genome sequence and comparative analysis of the solvent-producing bacterium Clostridium acetobutylicum,” J. Bacteriol. 183, 4823-4838. DOI: 10.1128/JB.183.16.4823-4838.2001

Ohto, C., Muramatsu, M., Obata, S., Sakuradani, E., and Shimizu, S. (2009). “Overexpression of the gene encoding HMG-CoA reductase in Saccharomyces cerevisiae for production of prenyl alcohols,” Appl. Microbiol. Biot. 82, 837-845. DOI: 10.1007/s00253-008-1807-5

Oudshoorn, A., van der Wielen, L. A. M., and Straathof, A. J. J. (2009). “Adsorption equilibria of bio-based butanol solutions using zeolite,” Biochem. Eng. J. 48, 99-103. DOI: 10.1016/j.bej.2009.08.014

Ounine, K., Petitdemange, H., Raval, G., and Gay, R. (1985). “Regulation and butanol inhibition of D-xylose and D-glucose uptake in Clostridium acetobutylicum,” Appl. Environ. Microbiol. 49, 874-878.

Ounine, K., Petitdemange, H., Raval, G., and Gay, R. (1983). “Acetone-butanol production from pentoses by Clostridium acetobutylicum,” Biotechnol. Lett. 5, 605-610. DOI: 10.1007/BF00130841

Palmqvist, E., and Hahn-Hagerdal, B. (2000). “Fermentation of lignocellulosic hydrolysates. II: Inhibitors and mechanisms of inhibition,” Bioresource Technol. 74, 25-33. DOI: 10.1016/S0960-8524(99)00161-3

Papini, M., Nookaew, I., Uhlen, M., and Nielsen, J. (2012). “Scheffersomyces stipitis: A comparative systems biology study with the Crabtree positive yeast Saccharomyces cerevisiae,” Microbial Cell Factories 11, 136, 1-16. DOI: 10.1186/1475-2859-11-136

Papoutsakis, E. T. (2008). “Engineering solventogenic Clostridia,” Curr. Opin. Biotechnol. 19, 420-429. DOI: 10.1016/j.copbio.2008.08.003

Perlack, R. D., Wright, L. L., Turhollow, A. F., Graham, R. L., Stokes, B. J., and Erbach, D. C.(2005). “Biomass as feedstock for a bioenergy and bioproducts industry,” DOE/GO-102995-2135.

Parekh, S. R., Yu, S., and Wayman, M. (1986). “Adaptation of Candida-Shehatae and Pichia-Stipitis to wood hydrolysates for increased ethanol-production,” Appl. Microbiol. Biot. 25, 300-304. DOI: 10.1007/BF00253667

Parekh, M., and Blaschek, H. P. (1999). “Butanol production by hyper solvent producing mutant Clostridium beijerinckii BA101 in corn steep water medium containing maltodextrin,” Biotechnol. Lett. 21, 45-48. DOI: 10.1023/A:1005406120874

Parekh, M., and Formanek, J. (1999). “Pilot-scale production of butanol by Clostridium beijerinckii BA101 using a low-cost fermentation medium based on corn steep water,” Appl. Microbiol. Biot. 51, 152-157. DOI: 10.1007/s002530051375

Peralta-Yahya, P. P., and Keasling, J. D. (2010). “Advanced biofuel production in microbes,” Biotechnol. J. 5, 147-162. DOI: 10.1002/biot.200900220

Peralta-Yahya, P. P., Ouellet, M., Chan, R., Mukhopadhyay, A., Keasling, J. D., and Lee, T. S. (2011). “Identification and microbial production of a terpene-based advanced biofuel,” Nat. Commun. 2(483), 1-8. DOI: 10.1038/ncomms1494

Pfromm, P. H., Amanor-Boadu, V., Nelson, R., Vadlani, P., and Madl, R. (2010). “Bio-butanol vs. bio-ethanol: A technical and economic assessment for corn and switch grass fermented by yeast or Clostridium acetobutylicum,” Biomass Bioenerg. 34, 515-524. DOI: 10.1016/j.biombioe.2009.12.017

Pitera, D. J., Paddon, C. J., Newman, J. D., and Keasling, J. D. (2007). “Balancing a heterologous mevalonate pathway for improved isoprenoid production in Escherichia coli,” Metab. Eng. 9, 193-207. DOI: 10.1016/j.ymben.2006.11.002

Qureshi, N., and Blaschek, H. (1999). “Production of acetone butanol ethanol (ABE) by a hyper-producing mutant strain of Clostridium beijerinckii BA101 and recovery by pervaporation,” Biotechnol. Prog. 15, 594-602. DOI: 10.1021/bp990080e

Qureshi, N., Li, X.-L., and Hughes, S. (2006). “Butanol production from corn fiber xylan using Clostridium acetobutylicum,” Biotechnol. Prog. 22, 673-680. DOI: 10.1021/bp050360w

Qureshi, N., Saha, B. C., and Dien, B. (2010). “Production of butanol (a biofuel) from agricultural residues: Part I—Use of barley straw hydrolysate,” Biomass Bioenerg. 34, 559-565. DOI: 10.1016/j.biombioe.2009.12.023

Rabinovitch-Deere, C. A., Oliver, J. W. K., Rodriguez, G. M., and Atsumi, S. (2013). “Synthetic biology and metabolic engineering approaches to produce biofuels,” Chem. Rev. 113, 4611-4632. DOI: 10.1021/cr300361t

Ranjan, A., and Moholkar, V. S. (2012). “Bio-butanol: Science, engineering, and economics,” Int. J. Energy Res. 36, 277-323. DOI: 10.1002/er.1948

Renewable Fuel Association (RFA). “(2014). “Statistics on world fuel ethanol production,” (http://ethanolrfa.org/pages/World-Fuel-Ethanol-Production), Accessed on 23 April 2014.

Ren, C., Chen, T., Zhang, J., Liang, L., and Lin, Z. (2009). “An evolved xylose transporter from Zymomonas mobilis enhances sugar transport in Escherichia coli,” Microbial Cell Factories8(66), 1-9. DOI: 10.1186/1475-2859-8-66

Ross, D. (1961). “The acetone-butanol fermentation,” Prog. Ind. Microbiol. 3, 71-90.

Rupilius, W., and Ahmad, S. (2006). “The changing world of oleochemicals,” Palm Oil Develop.44, 15-28.

Sabri, S., Nielsen, L. K., and Vickers, C. E. (2013). “Molecular control of sucrose utilization in Escherichia coli W, an efficient sucrose-utilizing strain,” Appl. Environ. Microbiol. 79, 478-487. DOI: 10.1128/AEM.02544-12. 478

Schwab, W., Fuchs, C., and Huang, F. (2013). “Transformation of terpenes into fine chemicals,” Eur. J. Lipid Sci. Technol. 115, 3-8. DOI: 10.1002/ejlt.201200157

Shiba, Y., Paradise, E. M., Kirby, J., Ro, D.-K., and Keasling, J. D. (2007). “Engineering of the pyruvate dehydrogenase bypass in Saccharomyces cerevisiae for high-level production of isoprenoids,” Metab. Eng. 9, 160-168. DOI: 10.1016/j.ymben.2006.10.005

Singhvi, M. S., Chaudhari, S., and Gokhale, D. V. (2014). “Lignocellulose processing: A current challenge,” RSC Adv. 4, 8271-8277. DOI: 10.1039/c3ra46112b

Snow, N. (2013). “Advanced biofuels could remove blend wall, Senate panel told,” Oil & Gas J. 111, 18-19.

Sorek, N., Yeats, T. H., Szemenyei, H., Youngs, H., Somerville, C. R. (2014). “The implications of lignocellulosic biomass chemical composition for the production of advanced biofuels,” Biosci.64, 192-201. DOI: 10.1093/biosci/bit037

Steen, E. J., Chan, R., Prasad, N., Myers, S., Petzold, C. J., Redding, A., Ouellet, M., and Keasling, J. D. (2008). “Metabolic engineering of Saccharomyces cerevisiae for the production of n-butanol,” Microbial Cell Factories 7(36), 1-8. DOI: 10.1186/1475-2859-7-36

Steen, E. J., Kang, Y., Bokinsky, G., Hu, Z., Schirmer, A., McClure, A., del Cardayre, S. B., and Keasling, J. D. (2010). “Microbial production of fatty-acid-derived fuels and chemicals from plant biomass,” Nature 463, 559-U182. DOI: 10.1038/nature08721

Steinbuechel, A. (2014). “Fatty acid synthesis in Escherichia coli and its applications towards the production of fatty acid based biofuels,” Biotechnol. Biofuels 7, 1-26. DOI: 10.1186/1754-6834-7-7

Tanaka, K., Ishii, Y., Ogawa, J., and Shima, J. (2012). “Enhancement of acetic acid tolerance in Saccharomyces cerevisiae by overexpression of the HAA1 gene, encoding a transcriptional activator,” Appl. Environ. Microbiol. 78, 8161-8163. DOI:10.1128/AEM.02356-12

Thormann, K., Feustel, L., Lorenz, K., Nakotte, S., and Dürre, P. (2002). “Control of butanol formation in Clostridium acetobutylicum by transcriptional activation,” J. Bacteriol. 184, 1966-1973. DOI: 10.1128/JB.184.7.1966-1973.2002

Urano, J., Dundon Asleson, C., Meinhold, P., Feldman, R., Aristidou, A., Hawkins, A., Buelter, T., Peters, M., Lies, D., and Porter-Scheinman, S. (2012). “Cytosolic isobutanol pathway localization for the production of isobutanol,” U.S. Patent No. 8232089 B2.

Vane, L. M. (2008). “Separation technologies for the recovery and dehydration of alcohols from fermentation broths,” Biofuels Bioproducts & Biorefining-Biofpr 2(6):553-588. DOI: 10.1002/bbb.108

Verduyn, C., Vankleef, R., Jzn, J. F., Schreuder, H., Vandijken, J. P., and Scheffers, W. A. (1985). “Nad(p)h-dependent aldose reductase from the xylose-fermenting yeast Pichia-Stipitis,” Antonie Van Leeuwenhoek J. Microbiol. 51, 562-562. DOI: 10.1007/BF00404545

Vickers, C. E., Klein-Marcuschamer, D., and Krömer, J. O. (2012). “Examining the feasibility of bulk commodity production in Escherichia coli,” Biotechnol. Lett. 34, 585-596. DOI: 10.1007/s10529-011-0821-3

Wallaart, T. E., Pras, N., Beekman, A. C., and Quax, W. J. (2000). “Seasonal variation of artemisinin and its biosynthetic precursors in plants of Artemisia annua of different geographical origin: Proof for the existence of chemotypes,” Planta Med. 66, 57-62. DOI: 10.1055/s-2000-11115

Wang, C., Yoon, S., Jang, H., Chung, Y., Kim, J., Choi, E., and Kim, S. (2011). “Metabolic engineering of Escherichia coli for alpha-farnesene production,” Metab. Eng. 13, 648-655. DOI: 10.1016/j.ymben.2011.08.001

Westfall, P. J., Pitera, D. J., Lenihan, J. R., Eng, D., Woolard, F. X., Regentin, R., Horning, T., Tsuruta, H., Melis, D. J., and Owens, A. (2012). “Production of amorphadiene in yeast, and its conversion to dihydroartemisinic acid, precursor to the antimalarial agent, artemisinin,” Proc. Natl. Acad. Sci. USA 109, E111-E118. DOI: 10.1073/pnas.1110740109

Wisselink, H. W., Toirkens, M. J., Wu, Q., Pronk, J. T., and van Maris, A. J. A. (2009). “Novel evolutionary engineering approach for accelerated utilization of glucose, xylose, and arabinose mixtures by engineered Saccharomyces cerevisiae strains,” Appl. Environ. Microbiol. 75, 907-914. DOI: 10.1128/AEM.02268-08

Wood, B. E., and Ingram, L. O. (1992). “Ethanol – production from cellobiose, amorphous cellulose, and crystalline cellulose by recombinant Klebsiella oxytoca containing chromosomally integrated Zymomonas-mobilis genes for ethanol-production and plasmids expressing thermostable cellulase genes from Clostridium thermocellum,” Appl. Environ. Microbiol. 58, 2103-2110.

Xue, C., Zhao, J., Chen, L., Bai, F., Yang, S., and Sun, J. (2014a). “Integrated butanol recovery for an advanced biofuel: Current state and prospects,” Appl. Microbiol. Biot. 98, 3463-3474. DOI: 10.1007/s00253-014-5561-6

Xue, C., Du, G., Sun, J., Chen, L., Gao, S., Yu, M., Yang, S., and Bai, F. (2014b). “Characterization of gas stripping and its integration with acetone-butanol-ethanol fermentation for high-efficient butanol production and recovery,” Biochem. Eng. J. 83, 55-61. DOI: 10.1016/j.bej.2013.12.003

Xue, C., Zhao, X., Liu, C., Chen, L., and Bai, F. (2013a). “Prospective and development of butanol as an advanced biofuel,” Biotechnol Adv 31, 1575-1584. DOI: 10.1016/j.biotechadv.2013.08.004

Xue, C., Zhao, J., Liu, F., Lu, C., Yang, S. T., and Bai, F. W. (2013b). “Two-stage in situ gas stripping for enhanced butanol fermentation and energy-saving product recovery,” Bioresource Technol. 135, 396-402. DOI: 10.1016/j.biortech.2012.07.062

Yan, R. T., Zhu, C. X., Golemboski, C., and Chen, J. S. (1988). “Expression of solvent-forming enzymes and onset of solvent production in batch cultures of Clostridium bijerinckii(Clostridium-Butylicum),” Appl. Environ. Microbiol. 54, 642-648.

Yan, Y., and Liao, J. C. (2009). “Engineering metabolic systems for production of advanced fuels,” J. Ind. Microbiol. Biotechnol. 36,471-479. DOI: 10.1007/s10295-009-0532-0

Yoshikuni, Y., Ferrin, T., and Keasling, J. (2006). “Designed divergent evolution of enzyme function,” Nature 440, 1078-1082. DOI: 10.1038/nature04607

Yomano, L. P., York, S. W., and Ingram, L. O. (1998). “Isolation and characterization of ethanol-tolerant mutants of Escherichia coli KO11 for fuel ethanol production,” J. Ind. Microbiol. Biotechnol. 20, 132-138. DOI: 10.1038/sj.jim.2900496

Yomano, L. P., York, S. W., Shanmugam, K. T., and Ingram, L. O. (2009). “Deletion of methylglyoxal synthase gene (mgsA) increased sugar co-metabolism in ethanol-producing Escherichia coli,” Biotechnol. Lett. 31, 1389-1398. DOI: 10.1007/s10529-009-0011-8

Yuan, L. Z., Rouviere, P. E., LaRossa, R. A., and Suh, W. (2006). “Chromosomal promoter replacement of the isoprenoid pathway for enhancing carotenoid production in E. coli,” Metab. Eng. 8, 79-90. DOI: 10.1016/j.ymben.2005.08.005

Zhang, M., Eddy, C., Deanda, K., Finkestein, M., and Picataggio, S. (1995). “Metabolic engineering of a pentose metabolism pathway in ethanologenic Zymomonas mobilis,” Science 267, 240-243. DOI: 10.1126/science.267.5195.240

Zhang, K., Sawaya, M. R., Eisenberg, D. S., and Liao, J. C. (2008). “Expanding metabolism for biosynthesis of non-natural alcohols,” Proc. Natl. Acad. Sci. USA 105, 20653-20658. DOI: 10.1073/pnas.0807157106

Zhang, Y., Ma, Y., Yang, F. and Zhang, C. (2009). “Continuous acetone–butanol–ethanol production by corn stalk immobilized cells,” J. Ind. Microbiol. Biotechnol. 36, 1117-1121. DOI: 10.1007/s10295-009-0582-3

Zheng, Y. N., Li, L. Z., Xian, M., and Ma, Y. J. (2009). “Problems with the microbial production of butanol,” J. Ind. Microbiol. Biotechnol. 36, 1127-1138. DOI: 10.1007/s10295-009-0609-9

Zillig, W., Holz, I., Janekovic, D., Klenk, H. P., Imsel, E., Trent, J., Wunderl, S., Forjaz, V. H., Coutinho, R., and Ferreira, T. (1990). “Hyperthermus-Butylicus, hyperthermophilic sulfur-reducing archaebacterium that ferments peptides,” J. Bacteriol. 172, 3959-3965.

Article submitted: May 26, 2016; Peer review completed: July 22, 2016; Revised version received and accepted: August 22, 2016; Published: September 28, 2016.

DOI: 10.15376/biores.11.4.Veettil